An Invasive Ectoparasite of Cervids, the Deer Ked: Dispersion, Cold Tolerance and Predation

dissertations

Sirpa Kaunisto An invasive ectoparasite of

cervids, the deer ked: predation and tolerance cold dispersion, ked: deer the of cervids, ectoparasite invasive | An Kaunisto | No 87 | Sirpa dispersion, cold tolerance and predation

Ectoparasites inhabit the outer surface of their hosts and consume host resources. In general, host-related factors Sirpa Kaunisto (e.g. density) are thought to be the main contributors to geographical distribution of arthropod ectoparasites. How- An invasive ectoparasite ever, temperature is known to strongly regulate survival and distribution in of cervids, the deer ked: many species. Also predation may have ecological significance on parasites, dispersion, cold tolerance although it has rarely been acknowledged. This thesis provides insights and predation into temperature-related survival, life-history characteristics and biological relationships (e.g. predation) of an insect ectoparasite that undergoes a range expansion.

Publications of the University of Eastern Finland Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences No 87 Dissertations in Forestry and Natural Sciences

isbn 978-952-61-0947-3 (printed) issnl 1798-5668 issn 1798-5668 isbn 978-952-61-0948-0 (pdf) issn 1798-5676 (pdf) SIRPA KAUNISTO

An invasive ectoparasite of cervids, the deer ked: dispersion, cold tolerance and predation

Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences No 87

Academic Dissertation To be presented by permission of the Faculty of Science and Forestry for public examination in the Auditorium N100 in Natura Building at the University of Eastern Finland, Joensuu, on November, 07, 2012, at 12 o’clock noon.

Department of Biology Author’s address: University of Eastern Finland Department of Biology P.O.Box 111 80101 JOENSUU FINLAND email: [email protected]

Supervisors: Professor Hannu Ylönen, Ph.D. University of Jyväskylä Department of Biological and Environmental Science P.O.Box 35 40014 JYVÄSKYLÄ FINLAND email: [email protected]

Professor Raine Kortet, Ph.D. University of Eastern Finland Department of Biology P.O.Box 111 80101 JOENSUU FINLAND email: [email protected]

Docent Sauli Härkönen, D.Sc. Kopijyvä Finnish Wildlife Agency Joensuu, 2012 Fantsintie 13–14 Editors: Profs. Pertti Pasanen, 00890 HELSINKI Pekka Kilpeläinen, and Matti Vornanen FINLAND email: [email protected] Distribution:

Supervisors: Professor Hannu Ylönen, Ph.D. University of Jyväskylä Department of Biological and Environmental Science P.O.Box 35 40014 JYVÄSKYLÄ FINLAND email: [email protected]

Professor Raine Kortet, Ph.D. University of Eastern Finland Department of Biology P.O.Box 111 80101 JOENSUU FINLAND email: [email protected]

Eastern Finland University Library / Sales of publications Reviewers: Docent Marko Mutanen, Ph.D. [email protected] University of Oulu www.uef.fi/kirjasto Department of Biology P.O.Box 3000 90014 OULU ISBN: 978-952-61-0947-3 (printed) FINLAND ISSNL: 1798-5668 email: [email protected] ISSN: 1798-5668 ISBN: 978-952-61-0948-0 (PDF) Researcher Bjørnar Ytrehus, Ph.D. Norwegian Veterinary Institute ISSN: 1798-5676 (PDF) Section for Pathology Pb 750 Sentrum N-0106 OSLO NORWAY email: [email protected] Opponent: Professor Manfred Milinski, Ph.D. ABSTRACT Max Planck Institute for Evolutionary Biology August-Thienemann-Str. 2 24306 PLÖN Species distributions and invasion success are affected by a GERMANY range of factors, including various environmental conditions email: [email protected] and biological interactions with other organisms. Ectoparasites are parasites that inhabit surface of the host and consume resources of the host from the outside. Host-related factors, such as host availability and host specificity, are thought to be the main contributors to geographical distribution of terrestrial arthropod ectoparasites. In general, temperature is known to strongly regulate species distributions. However, direct temperature effects on the capacity of ectoparasites to disperse have received less attention. Also other external environmental factors and for example predators may play greater roles than commonly thought in the distribution of terrestrial ectoparasite fauna. In this thesis, I study the life-history characteristics and ecological relationships of an invasive insect ectoparasite, the deer ked (Lipoptena cervi L. 1758, Diptera; Hippoboscidae), to understand if these variables could affect distribution capacity of the species. I focus on temperature and predation as potential constraints for the deer ked’s distribution. The deer ked is a blood-feeding ectoparasite infesting several cervid species. This louse fly can be categorized as an invasive species in Finland, owing to its relatively rapid range expansion towards west and north during the previous five decades. The life cycle of the deer ked can be divided into on-host and off-host stages. In this thesis, my primary focus is on the off-host strategies such as free-living pupal stage and winged adults that have recently emerged from the pupariae and not been contact with host. Pupal and adult characteristics of the deer ked were estimated along species’ distribution range in Finland. Results show that diapausing pupae were smaller in the northern Central Finland than in the southern Finland. This may suggests plastic changes in the life-history characteristics of the deer ked. Small adults emerged earlier from smaller pupae, suggesting lower metabolic reserves of small individuals to sustain long Opponent: Professor Manfred Milinski, Ph.D. ABSTRACT Max Planck Institute for Evolutionary Biology August-Thienemann-Str. 2 24306 PLÖN Species distributions and invasion success are affected by a GERMANY range of factors, including various environmental conditions email: [email protected] and biological interactions with other organisms. Ectoparasites are parasites that inhabit surface of the host and consume resources of the host from the outside. Host-related factors, such as host availability and host specificity, are thought to be the main contributors to geographical distribution of terrestrial arthropod ectoparasites. In general, temperature is known to strongly regulate species distributions. However, direct temperature effects on the capacity of ectoparasites to disperse have received less attention. Also other external environmental factors and for example predators may play greater roles than commonly thought in the distribution of terrestrial ectoparasite fauna. In this thesis, I study the life-history characteristics and ecological relationships of an invasive insect ectoparasite, the deer ked (Lipoptena cervi L. 1758, Diptera; Hippoboscidae), to understand if these variables could affect distribution capacity of the species. I focus on temperature and predation as potential constraints for the deer ked’s distribution. The deer ked is a blood-feeding ectoparasite infesting several cervid species. This louse fly can be categorized as an invasive species in Finland, owing to its relatively rapid range expansion towards west and north during the previous five decades. The life cycle of the deer ked can be divided into on-host and off-host stages. In this thesis, my primary focus is on the off-host strategies such as free-living pupal stage and winged adults that have recently emerged from the pupariae and not been contact with host. Pupal and adult characteristics of the deer ked were estimated along species’ distribution range in Finland. Results show that diapausing pupae were smaller in the northern Central Finland than in the southern Finland. This may suggests plastic changes in the life-history characteristics of the deer ked. Small adults emerged earlier from smaller pupae, suggesting lower metabolic reserves of small individuals to sustain long non-feeding pupal stage. Despite of the different temperature bedding sites increased the predation risk of deer ked pupae by origins and different invasion history of the individuals, the local tit (Paridae) species. The predation pressure on deer ked adult emergence period was relatively synchronized, when pupae can be notable during winter because almost 35% of all pupae were reared under identical temperature and light pupae used in the study were predated upon, even within a conditions. The deer ked had relatively high cold tolerance in all relatively short time interval (5 days). The ability of tits to use free-living stages. The supercooling point, SCP (i.e. temperature the snow discolouration cue was likely dependent on infestation where the spontaneous freezing of body fluids begins) of the intensity of the deer ked on the local moose population and the diapausing pupae was -26°C. Similar values were observed period of invasion history of the deer ked. This suggests that without cold acclimation all year round despite of age (i.e. the predation of deer ked pupae by tits may be a learned birth month of pupae). Diapausing pupae of deer ked survived behavioural response. harsh frosts (-15°C to -20°C) lasting for 3–4 days. Thus, pupae Laboratory feeding experiments indicated that also small- died at several degrees above their supercooling points, mammal species prey upon pupae of deer ked. Out of three suggesting certain pre-freeze mortality during long-term frost tested species this was especially true for the granivorous- exposure, and freezing–intolerant strategy of the species at omnivorous bank vole (Myodes glareolus). Additionally, the pupal stage. Larger diapausing pupae were produced in spring herbivorous field vole (Migrotus agrestis) consumed pupae and they survived better the severe 3-day frosts (-20°C) than the under food limitation, but not if an alternative food was smaller winter pupae. Also developing pupae and adults were available. Surprisingly, the insectivorous and commonly food- able to supercool to relatively low temperatures (-20°C and - constrained common shrew (Sorex araneus) consumed fewer 21°C, respectively), although they hardly ever experience such pupae than the granivorous-herbivorous voles in the low temperatures in nature. experiment. In Finland, the main host of the deer ked is the moose (Alces Owing to the high cold tolerance of the deer ked in all off- alces). Results of my thesis show that the wild forest reindeer host stages and during all seasons, the species has not likely yet (Rangifer tarandus fennicus) and the semi-domestic reindeer encountered the total limits of its distribution and may continue (Rangifer tarandus tarandus) can also serve as hosts, although its range expansion towards north in despite of lower with notably lower infestation intensity of the deer ked temperatures. Predators may decrease population densities of compared to the moose. The host – deer ked interaction can the deer ked locally. One critical aspect for the survival of deer cause observable cues on the bedding sites of the moose. Snow, ked pupae during winter is the timing of new snowfall, which stained with blood and tissue fluid of a host, as well as faeces of likely protects against predators moving above snow level as deer keds, on bedding sites indicates deer ked infestation on a well as insulates pupae against extreme temperatures. host. Most of this reddish-brown snow discolouration was detected on the bedding site segments that had been in touch Universal Decimal Classification: 576.89, 591.543, 591.557.8, 591.69, with the neck and back region of the moose, indicating the most 595.773, 599.735.31 probable predilection sites of adult deer keds on the host. Experimental manipulations in the field revealed that predators CAB Thesaurus: Alces alces; cold tolerance; ectoparasites; Finland; freezing; can exploit this snow-discolouration cue, which indicates heavy Hippoboscidae; hosts; host parasite relationships; insects; invasive species; life deer ked infestation of the host, when searching for a prey. history; Lipoptena cervi; overwintering; predation; pupae; resting places; Experimental manipulation of snow discoloration at artificial temperature non-feeding pupal stage. Despite of the different temperature bedding sites increased the predation risk of deer ked pupae by origins and different invasion history of the individuals, the local tit (Paridae) species. The predation pressure on deer ked adult emergence period was relatively synchronized, when pupae can be notable during winter because almost 35% of all pupae were reared under identical temperature and light pupae used in the study were predated upon, even within a conditions. The deer ked had relatively high cold tolerance in all relatively short time interval (5 days). The ability of tits to use free-living stages. The supercooling point, SCP (i.e. temperature the snow discolouration cue was likely dependent on infestation where the spontaneous freezing of body fluids begins) of the intensity of the deer ked on the local moose population and the diapausing pupae was -26°C. Similar values were observed period of invasion history of the deer ked. This suggests that without cold acclimation all year round despite of age (i.e. the predation of deer ked pupae by tits may be a learned birth month of pupae). Diapausing pupae of deer ked survived behavioural response. harsh frosts (-15°C to -20°C) lasting for 3–4 days. Thus, pupae Laboratory feeding experiments indicated that also small- died at several degrees above their supercooling points, mammal species prey upon pupae of deer ked. Out of three suggesting certain pre-freeze mortality during long-term frost tested species this was especially true for the granivorous- exposure, and freezing–intolerant strategy of the species at omnivorous bank vole (Myodes glareolus). Additionally, the pupal stage. Larger diapausing pupae were produced in spring herbivorous field vole (Migrotus agrestis) consumed pupae and they survived better the severe 3-day frosts (-20°C) than the under food limitation, but not if an alternative food was smaller winter pupae. Also developing pupae and adults were available. Surprisingly, the insectivorous and commonly food- able to supercool to relatively low temperatures (-20°C and - constrained common shrew (Sorex araneus) consumed fewer 21°C, respectively), although they hardly ever experience such pupae than the granivorous-herbivorous voles in the low temperatures in nature. experiment. In Finland, the main host of the deer ked is the moose (Alces Owing to the high cold tolerance of the deer ked in all off- alces). Results of my thesis show that the wild forest reindeer host stages and during all seasons, the species has not likely yet (Rangifer tarandus fennicus) and the semi-domestic reindeer encountered the total limits of its distribution and may continue (Rangifer tarandus tarandus) can also serve as hosts, although its range expansion towards north in despite of lower with notably lower infestation intensity of the deer ked temperatures. Predators may decrease population densities of compared to the moose. The host – deer ked interaction can the deer ked locally. One critical aspect for the survival of deer cause observable cues on the bedding sites of the moose. Snow, ked pupae during winter is the timing of new snowfall, which stained with blood and tissue fluid of a host, as well as faeces of likely protects against predators moving above snow level as deer keds, on bedding sites indicates deer ked infestation on a well as insulates pupae against extreme temperatures. host. Most of this reddish-brown snow discolouration was detected on the bedding site segments that had been in touch Universal Decimal Classification: 576.89, 591.543, 591.557.8, 591.69, with the neck and back region of the moose, indicating the most 595.773, 599.735.31 probable predilection sites of adult deer keds on the host. Experimental manipulations in the field revealed that predators CAB Thesaurus: Alces alces; cold tolerance; ectoparasites; Finland; freezing; can exploit this snow-discolouration cue, which indicates heavy Hippoboscidae; hosts; host parasite relationships; insects; invasive species; life deer ked infestation of the host, when searching for a prey. history; Lipoptena cervi; overwintering; predation; pupae; resting places; Experimental manipulation of snow discoloration at artificial temperature Yleinen suomalainen asiasanasto: Hirvi; hirvikärpänen; hyönteiset; isäntäeläimet; jäätyminen; kotelot; kylmänkestävyys; loiset; lämpötila; parasitismi; saalistus; Suomi; talvehtiminen; tulokaslajit Acknowledgements

There are several people who I want to thank for the assistance during execution of this work. Foremost I want to thank my supervisors Hannu Ylönen (who has been my supervisor since my MSc thesis, 2007), Raine Kortet and Sauli Härkönen. I greatly appreciate your encouragement, support, constructive criticism and valuable comments. I have been co-operating with all of you many years. All this time has been rewarding with respect to my scientific learning process. I am also very thankful to my co-author and friend Laura Härkönen with whom I have been co-operating intensively ever since we met in March 2007. In addition, Heikki Roininen, Panu Välimäki, Arja Kaitala, Sauli Laaksonen, Jani Koskimäki, Pekka Niemelä and Tapani Repo deserve many thanks for significantly contributing to my work. I also thank the members of my thesis follow-up group: Jouni Laakso and Tarmo Ketola. During my several data collection and accommodation periods at Konnevesi Research Station, I often asked for help from the laboratory technicians with different mechanical, logistical, and technical problems. Thus, I am very thankful to the laboratory technicians Janne Koskinen, Jyrki Raatikainen and Risto Latvanen. I want also to thank all the other people working at the Konnevesi Research Station for their helpfulness and kindness. I also thank the School of Forest Sciences for the ability to use the equipments in the winter ecological laboratory, on the Joensuu campus. I would also like to thank Kauko Kilpeläinen, Antti Leivo, Ari Junttila, Reijo Kotilainen, Pertti Rautio and Juho-Antti Junno for their contribution to some of the data. In addition, I thank all the hundreds of Finnish volunteers who helped in the field with pupal collection during 2007 and 2009. I gratefully acknowledge the financial support from the Ella and Georg Ehrnrooth Foundation, the Alfred Kordelin Foundation, The Biological Interactions Graduate School LIST OF ORIGINAL PUBLICATIONS (BIOINT), the Finnish Ministry of Agriculture and Forestry, Societas pro Fauna et Flora Fennica, the Biological Society of This thesis is based on data presented in the following articles, Finland Vanamo, the Finnish Konkordia Foundation, and the which are referred to by the Roman numerals I–V in the text. Oskar Öflund Stiftelse. I sincerely thank the two official pre-examiners of this thesis, Marko Mutanen and Bjørnar Ytrehus, for their critical comments and statements. I also thank Nick DiRienzo who kindly revised I Kaunisto S, Härkönen L, Niemelä P, Roininen H and Ylönen the language of the dissertation. H. Northward invasion of the parasitic deer ked (Lipoptena I am deeply grateful to my parents Sauli and Aino for all the cervi), is there geographical variation in pupal size and support they have given me all these years. Furthermore, I development duration? Parasitology 138: 354–363, 2011. thank my sister Kaija and her family for brightening up my life in many ways. I thank my friends here in Joensuu and II Härkönen L, Kaitala A, Kaunisto S and Repo T. High cold elsewhere for providing invaluable moments and tolerance through four seasons and all free-living stages in an counterbalance to the studies and work. Special thanks belong ectoparasite. Parasitology 139: 926–933, 2012. to my friend and cousin Tuuli who has always shared joys and sorrows of life with me. III Kaunisto S, Kortet R, Härkönen L, Härkönen S, Ylönen H and Laaksonen S. New bedding site examination-based method to analyse deer ked (Lipoptena cervi) infection in cervids. Parasitology Research 104: 919–925, 2009.

IV Kaunisto S, Välimäki P, Kortet R, Koskimäki J, Härkönen S, Kaitala A, Laaksonen S, Härkönen L and Ylönen H. Avian predation on a parasitic fly of cervids during winter: can host-related cues increase the predation risk? Biological Journal of the Linnean Society 106: 275–286, 2012.

V Kaunisto S, Kortet R, Härkönen S, Kaitala A, Laaksonen S and Ylönen H. Do small mammals prey upon an invasive ectoparasite of cervids? Canadian Journal of Zoology 90: 1044– 1050, 2012.

The publications are reprinted with permissions from Cambridge University Press (papers I and II), Springer-Verlag (III), Wiley-Blackwell (IV) and NRC Research Press (V). Foundation, The Biological Interactions Graduate School LIST OF ORIGINAL PUBLICATIONS (BIOINT), the Finnish Ministry of Agriculture and Forestry, Societas pro Fauna et Flora Fennica, the Biological Society of This thesis is based on data presented in the following articles, Finland Vanamo, the Finnish Konkordia Foundation, and the which are referred to by the Roman numerals I–V in the text. Oskar Öflund Stiftelse. I sincerely thank the two official pre-examiners of this thesis, Marko Mutanen and Bjørnar Ytrehus, for their critical comments and statements. I also thank Nick DiRienzo who kindly revised I Kaunisto S, Härkönen L, Niemelä P, Roininen H and Ylönen the language of the dissertation. H. Northward invasion of the parasitic deer ked (Lipoptena I am deeply grateful to my parents Sauli and Aino for all the cervi), is there geographical variation in pupal size and support they have given me all these years. Furthermore, I development duration? Parasitology 138: 354–363, 2011. thank my sister Kaija and her family for brightening up my life in many ways. I thank my friends here in Joensuu and II Härkönen L, Kaitala A, Kaunisto S and Repo T. High cold elsewhere for providing invaluable moments and tolerance through four seasons and all free-living stages in an counterbalance to the studies and work. Special thanks belong ectoparasite. Parasitology 139: 926–933, 2012. to my friend and cousin Tuuli who has always shared joys and sorrows of life with me. III Kaunisto S, Kortet R, Härkönen L, Härkönen S, Ylönen H and Laaksonen S. New bedding site examination-based method to analyse deer ked (Lipoptena cervi) infection in cervids. Parasitology Research 104: 919–925, 2009.

V Kaunisto S, Kortet R, Härkönen S, Kaitala A, Laaksonen S and Ylönen H. Do small mammals prey upon an invasive ectoparasite of cervids? Canadian Journal of Zoology 90: 1044– 1050, 2012.

The publications are reprinted with permissions from Cambridge University Press (papers I and II), Springer-Verlag (III), Wiley-Blackwell (IV) and NRC Research Press (V). AUTHOR’S CONTRIBUTION

In the paper (I), I was mainly responsible for data collection, Contents handling, analysis, and writing. In the paper (II), I took part in planning, data collection and writing. In the paper (III), I was mainly responsible for data handling, analysis, and writing. For 1Introduction ...... 15 the papers (IV) and (V), I collaborated intensively with co- 1.1 Invasion ...... 15 authors in planning and data collection, and was mainly 1.1.1 Invasion process ...... 15 responsible for the data analysis and writing. 1.1.2 Ecological factors affecting invasion success ...... 16 1.1.3 Characteristics of invasive species ...... 17 1.2 Ectoparasites ...... 19 1.2.1 Arthropod ectoparasites ...... 19 1.2.2 Factors affecting distribution of arthropod ectoparasites ...... 20 1.3 Low temperatures and survival of insects and arthropod ectoparasites ...... 21 1.3.1 Cold-hardiness and other factors affecting winter survival .... 22 1.3.2 Life-history adjustments towards higher latitudes and factors affecting offspring’s size...... 25 1.4 Predation as a biotic factor regulating population distribution ...... 28 1.4.1 Predation on insects ...... 29 1.4.2 Predation on arthropod ectoparasites ...... 29 1.4.3 Predation cues arising from the host-ectoparasite interaction 30 1.5 Objectives of the study ...... 31

2Materials and methods ...... 33 2.1 Study model: the Deer ked ...... 33 2.2 General study designs...... 37 2.2.1 Northward invasion, low temperatures and life-history characteristics of the deer ked (I, II) ...... 40 2.2.2 Cues revealing the host-deer ked interaction in winter (III) ... 43 2.2.3 Predation on pupal stage of the deer ked (IV, V) ...... 44

3Results and discussion ...... 47 3.1 Cold tolerance of the deer ked and associated life-history characteristics (I, II) ...... 47 3.2 Cues arising from the host – deer ked interaction (III) ...... 52 AUTHOR’S CONTRIBUTION

In the paper (I), I was mainly responsible for data collection, Contents handling, analysis, and writing. In the paper (II), I took part in planning, data collection and writing. In the paper (III), I was mainly responsible for data handling, analysis, and writing. For 1 Introduction ...... 15 the papers (IV) and (V), I collaborated intensively with co- 1.1 Invasion ...... 15 authors in planning and data collection, and was mainly 1.1.1 Invasion process ...... 15 responsible for the data analysis and writing. 1.1.2 Ecological factors affecting invasion success ...... 16 1.1.3 Characteristics of invasive species ...... 17 1.2 Ectoparasites ...... 19 1.2.1 Arthropod ectoparasites ...... 19 1.2.2 Factors affecting distribution of arthropod ectoparasites ...... 20 1.3 Low temperatures and survival of insects and arthropod ectoparasites ...... 21 1.3.1 Cold-hardiness and other factors affecting winter survival .... 22 1.3.2 Life-history adjustments towards higher latitudes and factors affecting offspring’s size...... 25 1.4 Predation as a biotic factor regulating population distribution ...... 28 1.4.1 Predation on insects ...... 29 1.4.2 Predation on arthropod ectoparasites ...... 29 1.4.3 Predation cues arising from the host-ectoparasite interaction 30 1.5 Objectives of the study ...... 31

2 Materials and methods ...... 33 2.1 Study model: the Deer ked ...... 33 2.2 General study designs...... 37 2.2.1 Northward invasion, low temperatures and life-history characteristics of the deer ked (I, II) ...... 40 2.2.2 Cues revealing the host-deer ked interaction in winter (III) ... 43 2.2.3 Predation on pupal stage of the deer ked (IV, V) ...... 44

3 Results and discussion ...... 47 3.1 Cold tolerance of the deer ked and associated life-history characteristics (I, II) ...... 47 3.2 Cues arising from the host – deer ked interaction (III) ...... 52 3.3 Predation on pupal stage of the deer ked (IV, V) ...... 54

4 Conclusions and future prospects ...... 57 1 Introduction

References ...... 61 1.1 INVASION

1.1.1 Invasion process There has been a considerable and growing interest in invasive species among ecologists during previous decades (e.g. Elton 1958; Williamson 1996; Cox 1999; Mooney et al. 2005; Lockwood et al. 2007; Menke et al. 2007; Carlsson et al. 2009; Li et al. 2011). The terminology of invasion biology has been varying, and thus a subject of debate and confusion (Carlton 2002; Davis 2009). A broader definition for the biological invasion applies all species which arrive beyond their previous ranges and occupy a new region (Williamson 1996; Cox 1999; Reise et al. 2006). In my dissertation, I use this broader definition for invasion which does not distinguish between human-mediated invasions and naturally occurred invasions. The line between human-aided and natural invasions is not clear. Most invasions and range expansions of species are mediated by human actions, intentionally or accidentally (e.g. Williamson 1996; Vermeij 2005). However, some of successful invasions are not associated to human activity and these natural range expansions can appear on small or larger scale, sometimes even between continents (Williamson 1996; reviewed in Davis 2009). The invasion process can be divided into three main phases; 1) dispersal (or introduction) into a new area; 2) establishment and persistence; and 3) further spread to nearby areas (Vermeij 1996; Kolar and Lodge 2001; Sakai et al. 2001; Davis 2009). However, the usage of these terms is nonuniform. Davis (2009) defines that dispersal into a new area and establishment (i.e. surviving until reproduction) are two consecutive and fundamental processes which function at individual level. Many successful dispersal and establish events of individuals leads possibly to the population persistence in a new area, further

15 3.3 Predation on pupal stage of the deer ked (IV, V) ...... 54

4Conclusions and future prospects ...... 57 1 Introduction

References ...... 61 1.1 INVASION

15 spreading to new areas and even to metapopulation persistence species invasions (ecosystem invasibility) (e.g. Lonsdale 1999; (i.e. persistence of many populations) throughout larger region. Davis et al. 2000; Schoolmaster and Snyder 2007). It is often Between the dispersal into a new area and population difficult to identify characteristics affecting invasibility. persistence there is usually a time lag, referring to the time Temporal and spatial heterogeneity in resource availability (e.g. needed for achieving sufficient genetic diversity and host availability) is among the most important factors affecting adaptations to local conditions (reviewed in Sakai et al. 2001). invasibility and thus invasion success of species (Davis et al. Founding population (i.e. first arrived individuals) may achieve 2000; Schoolmaster and Snyder 2007; Davis 2009). Controversial persistence independently or by relying on (totally or partly) diversity-invasibility hypothesis argues that higher species further supplementary propagules from external populations diversity of community makes it more resistant to invaders (Davis 2009). According to Davis (2009), the term, spread, can because fewer empty niches available (Darwin 1859; Elton 1958; refer to the spread of a one single population through large area Fargione and Tilman 2005; Davis 2009). The enemy release or founding of new populations in new areas from the first or hypothesis (ERH), instead, suggests that invader encounter previous persisting populations. fewer natural enemies (e.g. predators, parasites) in a new area, and therefore is able to establish successfully (e.g. Schoener and 1.1.2 Ecological factors affecting invasion success Spiller 1995; Williamson 1996; Kotiaho and Sulkava 2007; Several factors determine invasion success of the species. Shwartz et al. 2009). Species interactions are usually very During the first phase of invasion, dispersion capacity into a complex. For example, positive prey-prey interaction can reduce new area is likely affected by the extent of propagule pressure predation pressure by generalist predators on an invasive prey (i.e. the number of new individuals arriving per dispersal event (Koss and Snyder 2005). Although the ERH has often been and the number of discrete dispersal events) and possible vector ignored and rarely been the focus of research, it is thought that availability/frequency (Johnson and Starks 2004; Carlton and enemy-related processes can be of importance in some Ruiz 2005; Lockwood et al. 2005, 2007). Invasive organisms may distributions (Davis 2009). disperse into a new area independently or by exploiting vectors (e.g. humans, vehicles, water currents, host animals) at different 1.1.3 Characteristics of invasive species life-stages. The relationship between different life-history traits and During the establishment phase, the invader needs to tolerate invasion success is not straightforward, as successful traits vary various new ecological conditions and to be capable to utilize across space and time (Lodge 1993; Williamson and Fitter 1996; new resources in order to secure its survival until reproduction Davis 2009). Invasive species constitute a very heterogenous (Cox 1999; Lockwood et al. 2007; Davis 2009). Abiotic factors group representing a wide range of taxa (Pyšek et al. 2008). such as climate (e.g. photoperiod, temperature, humidity), Despite of many studies on invasive insects, there is no edaphic factors (e.g. physical barriers) and human related agreement about the most significant traits in their factors (e.g. land use) can solely, or together with other factors, establishment success (Peacock and Worner 2008; Pyšek et al. prevent or slow the invasion process. Biotic factors, instead, 2008). One of the main factors regulating insect distribution is include different forms of biological interactions between a new temperature (Gaston 2003; Denlinger and Lee 2010) and arrival species and native species (e.g. competition, predation, therefore advantageous characteristics of invasive insects may parasitism, mutualism). Several abiotic and biotic factors include e.g. broad thermal tolerance at different life-stages together determine how prone or resistant the ecosystem is to (Addo-Bediako et al. 2000; Kingsolver et al. 2007; Calosi et al.

16 17 spreading to new areas and even to metapopulation persistence species invasions (ecosystem invasibility) (e.g. Lonsdale 1999; (i.e. persistence of many populations) throughout larger region. Davis et al. 2000; Schoolmaster and Snyder 2007). It is often Between the dispersal into a new area and population difficult to identify characteristics affecting invasibility. persistence there is usually a time lag, referring to the time Temporal and spatial heterogeneity in resource availability (e.g. needed for achieving sufficient genetic diversity and host availability) is among the most important factors affecting adaptations to local conditions (reviewed in Sakai et al. 2001). invasibility and thus invasion success of species (Davis et al. Founding population (i.e. first arrived individuals) may achieve 2000; Schoolmaster and Snyder 2007; Davis 2009). Controversial persistence independently or by relying on (totally or partly) diversity-invasibility hypothesis argues that higher species further supplementary propagules from external populations diversity of community makes it more resistant to invaders (Davis 2009). According to Davis (2009), the term, spread, can because fewer empty niches available (Darwin 1859; Elton 1958; refer to the spread of a one single population through large area Fargione and Tilman 2005; Davis 2009). The enemy release or founding of new populations in new areas from the first or hypothesis (ERH), instead, suggests that invader encounter previous persisting populations. fewer natural enemies (e.g. predators, parasites) in a new area, and therefore is able to establish successfully (e.g. Schoener and 1.1.2 Ecological factors affecting invasion success Spiller 1995; Williamson 1996; Kotiaho and Sulkava 2007; Several factors determine invasion success of the species. Shwartz et al. 2009). Species interactions are usually very During the first phase of invasion, dispersion capacity into a complex. For example, positive prey-prey interaction can reduce new area is likely affected by the extent of propagule pressure predation pressure by generalist predators on an invasive prey (i.e. the number of new individuals arriving per dispersal event (Koss and Snyder 2005). Although the ERH has often been and the number of discrete dispersal events) and possible vector ignored and rarely been the focus of research, it is thought that availability/frequency (Johnson and Starks 2004; Carlton and enemy-related processes can be of importance in some Ruiz 2005; Lockwood et al. 2005, 2007). Invasive organisms may distributions (Davis 2009). disperse into a new area independently or by exploiting vectors (e.g. humans, vehicles, water currents, host animals) at different 1.1.3 Characteristics of invasive species life-stages. The relationship between different life-history traits and During the establishment phase, the invader needs to tolerate invasion success is not straightforward, as successful traits vary various new ecological conditions and to be capable to utilize across space and time (Lodge 1993; Williamson and Fitter 1996; new resources in order to secure its survival until reproduction Davis 2009). Invasive species constitute a very heterogenous (Cox 1999; Lockwood et al. 2007; Davis 2009). Abiotic factors group representing a wide range of taxa (Pyšek et al. 2008). such as climate (e.g. photoperiod, temperature, humidity), Despite of many studies on invasive insects, there is no edaphic factors (e.g. physical barriers) and human related agreement about the most significant traits in their factors (e.g. land use) can solely, or together with other factors, establishment success (Peacock and Worner 2008; Pyšek et al. prevent or slow the invasion process. Biotic factors, instead, 2008). One of the main factors regulating insect distribution is include different forms of biological interactions between a new temperature (Gaston 2003; Denlinger and Lee 2010) and arrival species and native species (e.g. competition, predation, therefore advantageous characteristics of invasive insects may parasitism, mutualism). Several abiotic and biotic factors include e.g. broad thermal tolerance at different life-stages together determine how prone or resistant the ecosystem is to (Addo-Bediako et al. 2000; Kingsolver et al. 2007; Calosi et al.

16 17 2008; Preisser et al. 2008; Nyamukondiwa et al. 2010). Although different traits along latitudinal clines may have an adaptive little studied, another example of useful traits, possibly value (e.g. Allemand and David 1976; Karan et al. 2000; Demont facilitating insect invasion, are defence mechanisms against and Blanckenhorn 2008). Some insects with several generations indigenous enemies. Few studies have proposed that, for within a short time interval may show rapid adaptive genetic example, aposematic warning colouration, unpalatability, or evolution along geographical cline (Gilchrist et al. 2001). chemical weapons (e.g. secretions from salivary glands) as anti- However, phenotypic plasticity has been suggested to be the predatory defences may decrease the number of natural primary strategy in many insect range expansions, allowing enemies in new ecosystems (Hough-Goldstein et al. 1993a,b; insect populations to tolerate and adapt to new conditions via Wells and Henderson 1993; Lundgren et al. 2010). flexibility in life-history, morphology, physiology or behaviour The ultimate mechanisms behind successful traits of invasive (e.g. Moller 1996; Fischer et al. 2003; Sagata and Lester 2009; species may often remain unclear. The relative importance of Wilson et al. 2009; Michie et al. 2010; Moczek 2010). plasticity and genetic adaptation in the invasion process has been a controversial topic (DeWitt and Scheiner 2004). In phenotypic plasticity, an organism with a certain genotype can 1.2 ECTOPARASITES rapidly fine-tune its phenotype (e.g. life-history traits, development, behaviour, physiology or morphology) to match 1.2.1 Arthropod ectoparasites new abiotic and biotic conditions (Agrawal 2001; West-Eberhard Parasites are organisms that follow a strategy at which they 2003; DeWitt and Scheiner 2004). Also genetic factors and benefit at the expense of other living organisms, called hosts evolutionary processes may have a considerable role in the (e.g. Milinski 1990; Clayton and Moore 1997). The ecological invasion success (reviewed in Sakai et al. 2001). Local genetic research of parasites is a relatively young sub-discipline, but it adaptation means a genetic change in an introduced population has been rapidly developing (e.g. Combes 2001; Poulin 2007). as a response to a new local selection pressure (Kawecki and Parasites constitute a very multiform and diverse group, with Ebert 2004). Both phenotypic plasticity and genetic adaptation parasites belonging to a number of different phyla. There are can contribute to the invasion ability of the species (Sexton et al. several classification frames for parasites, but the line between 2002). Phenotypic plasticity likely plays an important role at the the divisions is usually hard to draw because parasites in early stages of invasion by giving population time to acclimate different classification groups often share similar characteristics to local conditions and enhancing possible further genetic (Clayton and Moore 1997). Parasites can be divided, for divergence (Sexton et al. 2002; West-Eberhard 2003; Richards et example, into 1) endoparasites living and foraging inside the al. 2006; Ghalambor et al. 2007). In certain cases, phenotypic tissues of the host (intracellular or intercellular parasites), and 2) plasticity itself can be seen as an adaptive genetic trait having ectoparasites inhabiting surface of the hosts (e.g. on skin) and possible costs (Moran 1992; Sexton et al. 2002; Pigliucci 2005; thus consuming host resources (e.g. blood, secretions, keratin, Richards et al. 2006; Ghalambor et al. 2007; Lardies and skin cells) from the outside (e.g. Anderson and May 1978; Bozinovic 2008). However, it is possible that adaptive Clayton and Moore 1997; Bush et al. 2001; Grimaldi and Engel phenotypic plasticity can occur without underlying genetic 2005). Ectoparasites are likely more vulnerable to external differentiation (Kawecki and Ebert 2004). physical environmental factors than endoparasites (Clayton and In insect invasion, both phenotypic plasticity and adaptive Moore 1997; Felsõ and Rózsa 2006). In the life-cycle of many genetic responses are shown to occur. Genetic variation in ectoparasites, certain off-host stages alternate with on-host

18 19 2008; Preisser et al. 2008; Nyamukondiwa et al. 2010). Although different traits along latitudinal clines may have an adaptive little studied, another example of useful traits, possibly value (e.g. Allemand and David 1976; Karan et al. 2000; Demont facilitating insect invasion, are defence mechanisms against and Blanckenhorn 2008). Some insects with several generations indigenous enemies. Few studies have proposed that, for within a short time interval may show rapid adaptive genetic example, aposematic warning colouration, unpalatability, or evolution along geographical cline (Gilchrist et al. 2001). chemical weapons (e.g. secretions from salivary glands) as anti- However, phenotypic plasticity has been suggested to be the predatory defences may decrease the number of natural primary strategy in many insect range expansions, allowing enemies in new ecosystems (Hough-Goldstein et al. 1993a,b; insect populations to tolerate and adapt to new conditions via Wells and Henderson 1993; Lundgren et al. 2010). flexibility in life-history, morphology, physiology or behaviour The ultimate mechanisms behind successful traits of invasive (e.g. Moller 1996; Fischer et al. 2003; Sagata and Lester 2009; species may often remain unclear. The relative importance of Wilson et al. 2009; Michie et al. 2010; Moczek 2010). plasticity and genetic adaptation in the invasion process has been a controversial topic (DeWitt and Scheiner 2004). In phenotypic plasticity, an organism with a certain genotype can 1.2 ECTOPARASITES rapidly fine-tune its phenotype (e.g. life-history traits, development, behaviour, physiology or morphology) to match 1.2.1 Arthropod ectoparasites new abiotic and biotic conditions (Agrawal 2001; West-Eberhard Parasites are organisms that follow a strategy at which they 2003; DeWitt and Scheiner 2004). Also genetic factors and benefit at the expense of other living organisms, called hosts evolutionary processes may have a considerable role in the (e.g. Milinski 1990; Clayton and Moore 1997). The ecological invasion success (reviewed in Sakai et al. 2001). Local genetic research of parasites is a relatively young sub-discipline, but it adaptation means a genetic change in an introduced population has been rapidly developing (e.g. Combes 2001; Poulin 2007). as a response to a new local selection pressure (Kawecki and Parasites constitute a very multiform and diverse group, with Ebert 2004). Both phenotypic plasticity and genetic adaptation parasites belonging to a number of different phyla. There are can contribute to the invasion ability of the species (Sexton et al. several classification frames for parasites, but the line between 2002). Phenotypic plasticity likely plays an important role at the the divisions is usually hard to draw because parasites in early stages of invasion by giving population time to acclimate different classification groups often share similar characteristics to local conditions and enhancing possible further genetic (Clayton and Moore 1997). Parasites can be divided, for divergence (Sexton et al. 2002; West-Eberhard 2003; Richards et example, into 1) endoparasites living and foraging inside the al. 2006; Ghalambor et al. 2007). In certain cases, phenotypic tissues of the host (intracellular or intercellular parasites), and 2) plasticity itself can be seen as an adaptive genetic trait having ectoparasites inhabiting surface of the hosts (e.g. on skin) and possible costs (Moran 1992; Sexton et al. 2002; Pigliucci 2005; thus consuming host resources (e.g. blood, secretions, keratin, Richards et al. 2006; Ghalambor et al. 2007; Lardies and skin cells) from the outside (e.g. Anderson and May 1978; Bozinovic 2008). However, it is possible that adaptive Clayton and Moore 1997; Bush et al. 2001; Grimaldi and Engel phenotypic plasticity can occur without underlying genetic 2005). Ectoparasites are likely more vulnerable to external differentiation (Kawecki and Ebert 2004). physical environmental factors than endoparasites (Clayton and In insect invasion, both phenotypic plasticity and adaptive Moore 1997; Felsõ and Rózsa 2006). In the life-cycle of many genetic responses are shown to occur. Genetic variation in ectoparasites, certain off-host stages alternate with on-host

18 19 stages of variable lengths (Hopla et al. 1994; Grimaldi and Engel arachnids and insects to a certain degree (e.g. Krasnov et al. 2005). 2005b; Cumming and Van Vuuren 2006; Shenbrot et al. 2007; Most invertebrate ectoparasites are arthropods (Hopla et al. Härkönen et al. 2010; Bermúdez and Miranda 2011; Mize et al. 1994). Insects and arachnids typically parasitize terrestrial 2011). Many of these ectoparasite studies are case surveys animals while crustaceans inhabit fish. Insect ectoparasites feed and/or cross-sectional studies (e.g. medical, economical, on warm-blooded animals (the mammals and birds), since no agricultural and ecological views combined) investigating insect ectoparasites are known to infest amphibians or reptiles infestations on wild or domesticated animals and geographical unlike diverse mites and ticks (Grimaldi and Engel 2005). In distribution of ectoparasites (e.g. Heath 1994; Sréter et al. 2003; general, blood-feeding insects are not considered as Mamun et al. 2010; Välimäki et al. 2010; Abebe et al. 2011; ectoparasites unless they have not specialized for living on the Bermúdez and Miranda 2011). Since ectoparasites are host for at least part of their life (Marshall 1981; Grimaldi and dependent on their host, the resource availability hypothesis has Engel 2005). been studied to some extent. Host-related factors (e.g. host Ectoparasitism on vertebrates may have evolved in species availability, density and specificity) are generally thought to be having morphological pre-adaptations to blood-feeding (e.g. among the main factors determining geographical distribution mouth parts of predatory arthropods), or in species that have of terrestrial arthropod ectoparasites (e.g. Krasnov et al. 2005b; been in a close contact with nest-breeding vertebrates / with Shenbrot et al. 2007; Välimäki et al. 2011). Temperature is known large colonies of vertebrates (Grimaldi and Engel 2005). to strongly regulate distribution of insect species (Gaston 2003; Coevolution between ectoparasites and their vertebrate hosts Denlinger and Lee 2010) and, therefore, it could be assumed that has lasted 70-100 million years (Balashov 2005). However, in the free-living stages of arthropod ectoparasites do not provide many arthropod groups paleontological records are very scarce exception to this. However, direct temperature effects on and incomplete to trace the accurate era when the ectoparasitic characteristics and distribution of ectoparasites have received life-style begun (Rasnitsyn and Quicke 2002). Nowadays, little attention from the research community (Cumming and arthropod ectoparasites are a diverse and highly adapted group Van Vuuren 2006; Härkönen et al. 2010). Other than (Hopla et al. 1994). As an example, the true flies (order Diptera) temperature, external factors such as vegetation and predators, show a diverse array of ectoparasitic species. Some families of may also play greater roles than commonly thought in the the order, such as the louse-flies (Hippoboscidae), spend most distribution of terrestrial ectoparasite fauna (e.g. Samish and of their adult life on the warm-blooded host and are highly Rehacek 1999; Johnson et al. 2010; Krasnov et al. 2010; Mize et al. adapted morphologically (e.g. blood-feeding and attachment 2011). apparatus) and physiologically to an ectoparasitic life-style (Bequaert 1953; Hopla et al. 1994). 1.3 LOW TEMPERATURES AND SURVIVAL OF INSECTS AND 1.2.2 Factors affecting distribution of arthropod ectoparasites ARTHROPOD ECTOPARASITES Despite of numerous studies on invasive insects, range expansions of ectoparasitic insects have received less attention Climate determines geographical range limits of several species. (e.g. Pyšek et al. 2008; Haas et al. 2010; Härkönen et al. 2010; Low temperatures in winter and sudden temperature extremes Välimäki et al. 2010, 2011). Existing research has focused on during growing season are among the most important geographical distribution and range size of ectoparasitic environmental constraints for the insect distribution towards

20 21 stages of variable lengths (Hopla et al. 1994; Grimaldi and Engel arachnids and insects to a certain degree (e.g. Krasnov et al. 2005). 2005b; Cumming and Van Vuuren 2006; Shenbrot et al. 2007; Most invertebrate ectoparasites are arthropods (Hopla et al. Härkönen et al. 2010; Bermúdez and Miranda 2011; Mize et al. 1994). Insects and arachnids typically parasitize terrestrial 2011). Many of these ectoparasite studies are case surveys animals while crustaceans inhabit fish. Insect ectoparasites feed and/or cross-sectional studies (e.g. medical, economical, on warm-blooded animals (the mammals and birds), since no agricultural and ecological views combined) investigating insect ectoparasites are known to infest amphibians or reptiles infestations on wild or domesticated animals and geographical unlike diverse mites and ticks (Grimaldi and Engel 2005). In distribution of ectoparasites (e.g. Heath 1994; Sréter et al. 2003; general, blood-feeding insects are not considered as Mamun et al. 2010; Välimäki et al. 2010; Abebe et al. 2011; ectoparasites unless they have not specialized for living on the Bermúdez and Miranda 2011). Since ectoparasites are host for at least part of their life (Marshall 1981; Grimaldi and dependent on their host, the resource availability hypothesis has Engel 2005). been studied to some extent. Host-related factors (e.g. host Ectoparasitism on vertebrates may have evolved in species availability, density and specificity) are generally thought to be having morphological pre-adaptations to blood-feeding (e.g. among the main factors determining geographical distribution mouth parts of predatory arthropods), or in species that have of terrestrial arthropod ectoparasites (e.g. Krasnov et al. 2005b; been in a close contact with nest-breeding vertebrates / with Shenbrot et al. 2007; Välimäki et al. 2011). Temperature is known large colonies of vertebrates (Grimaldi and Engel 2005). to strongly regulate distribution of insect species (Gaston 2003; Coevolution between ectoparasites and their vertebrate hosts Denlinger and Lee 2010) and, therefore, it could be assumed that has lasted 70-100 million years (Balashov 2005). However, in the free-living stages of arthropod ectoparasites do not provide many arthropod groups paleontological records are very scarce exception to this. However, direct temperature effects on and incomplete to trace the accurate era when the ectoparasitic characteristics and distribution of ectoparasites have received life-style begun (Rasnitsyn and Quicke 2002). Nowadays, little attention from the research community (Cumming and arthropod ectoparasites are a diverse and highly adapted group Van Vuuren 2006; Härkönen et al. 2010). Other than (Hopla et al. 1994). As an example, the true flies (order Diptera) temperature, external factors such as vegetation and predators, show a diverse array of ectoparasitic species. Some families of may also play greater roles than commonly thought in the the order, such as the louse-flies (Hippoboscidae), spend most distribution of terrestrial ectoparasite fauna (e.g. Samish and of their adult life on the warm-blooded host and are highly Rehacek 1999; Johnson et al. 2010; Krasnov et al. 2010; Mize et al. adapted morphologically (e.g. blood-feeding and attachment 2011). apparatus) and physiologically to an ectoparasitic life-style (Bequaert 1953; Hopla et al. 1994). 1.3 LOW TEMPERATURES AND SURVIVAL OF INSECTS AND 1.2.2 Factors affecting distribution of arthropod ectoparasites ARTHROPOD ECTOPARASITES Despite of numerous studies on invasive insects, range expansions of ectoparasitic insects have received less attention Climate determines geographical range limits of several species. (e.g. Pyšek et al. 2008; Haas et al. 2010; Härkönen et al. 2010; Low temperatures in winter and sudden temperature extremes Välimäki et al. 2010, 2011). Existing research has focused on during growing season are among the most important geographical distribution and range size of ectoparasitic environmental constraints for the insect distribution towards

20 21 higher latitudes and altitudes (Messenger 1959; Gaston 2003; spontaneous freezing of body water occurs. In contrast, the Denlinger and Lee 2010). Temperature affects most freezing-tolerant species have characteristics enabling them to physiological processes and has a strong influence on the survive ice formation and freezing of the extracellular fluids (in development, growth, reproduction, and survival of insects few cases also freezing of a fat body) to a variable degree (Lee (Bale et al. 2002; Régnière et al. 2012). In seasonal temperate 1991; Sinclair 1999; Chown and Nicolson 2009). environments, insects and free-living stages of parasites need to Cold-hardiness is affected by a number of factors, and cope with adverse climatic conditions in order to survive mechanisms of cold-hardiness are often species-specific and (Tinsley 1999; Wharton 1999; Addo-Bediako et al. 2000; working in interaction (Lee and Denlinger 1991; Danks 1996; Denlinger and Lee 2010). Denlinger and Lee 2010). Many insects have been observed to increase cold tolerance and survival at low temperatures by 1.3.1 Cold-hardiness and other factors affecting winter becoming physiologically acclimated (e.g. Lee 1991; Huey et al. survival 1999; Koveos 2001; Morey et al. 2012). Rapid cold hardening Insect cold-hardiness has been a widely studied topic in recent (RCH) refers to the cold acclimation response that occurs within decades (reviewed in Danks 1996; Denlinger and Lee 2010). minutes to hours in many insects when they encounter non- Cold-hardiness can be referred to as the physical and metabolic lethal low temperatures, resulting in an increase in survival adjustments (or capacity) of an insect that increase its cold probability during further colder period (Lee et al. 1987; tolerance and thus survival at low temperatures (Denlinger Denlinger and Lee 2010). The RCH is reported to occur more 1991). There have been numerous categorizing systems of insect often in freeze-intolerant than in freeze-tolerant insects (Koveos cold tolerance (e.g. Lee 1991; Bale 1993, 1996; Sinclair 1999; 2001; Lee et al. 2006; Teets et al. 2008; Denlinger and Lee 2010). Zachariassen and Kristiansen 2003; Denlinger and Lee 2010). The seasonal (or programmed) long-term cold acclimation Cold-tolerance strategies of insects have recently been classified process, instead, can take several weeks or months and works in as: 1) chilling intolerant, 2) freeze-intolerant, or 3) freeze-tolerant preparation for winter (Lee 1991; Šlachta et al. 2002b; Chown species (Chown and Nicolson 2009; Denlinger and Lee 2010). and Nicolson 2009). Both forms of cold acclimation can occur Chilling intolerant species die because of direct or indirect and improve cold-hardiness, and thus survival, at different life- chilling injury in the absence of extracellular ice formation stages, as well as during active or diapausal (i.e. dormant) (above or below 0°C). Direct chilling injury i.e. cold shock (e.g. phases (e.g. Lee et al. 1987; Burks and Hagstrum 1999; Goto et al. changes in cell membranes, protein denaturation) results from a 2001; Šlachta et al. 2002a,b; Andreadis et al. 2005; Miyazaki et al. rapid cooling during a brief low-temperature exposure. Indirect 2006; Koštál et al. 2011). chilling injury, instead, is the consequence of a long-term In addition to cold-hardiness, also diapause is an important exposure (takes days to weeks) to low temperatures likely component of winter survival in many insects and related causing developmental abnormalities. Most insects that arthropod groups (Denlinger 1991). Diapause can be defined as encounter sub-zero temperatures in nature are freeze-intolerant. an endocrine-mediated dormancy, occurring at certain life- Freeze-intolerant species cannot survive the ice formation of stages (Tauber et al. 1986; Denlinger 1991; Madder et al. 1999; body fluids, and have thus evolved various strategies to avoid Belozerov and Naumov 2002; Saunders et al. 2002; Koštál 2006). freezing (see Bale 1993, 1996). In general, freeze-intolerant According to Koštál (2006) diapause syndrome can be divided insects avoid lethal freezing by cooling small volumes of body into several following phases, although some of them are still water, and thus by lowering the temperature at which the poorly understood: 1) pre-diapause (diapause induction and

22 23 higher latitudes and altitudes (Messenger 1959; Gaston 2003; spontaneous freezing of body water occurs. In contrast, the Denlinger and Lee 2010). Temperature affects most freezing-tolerant species have characteristics enabling them to physiological processes and has a strong influence on the survive ice formation and freezing of the extracellular fluids (in development, growth, reproduction, and survival of insects few cases also freezing of a fat body) to a variable degree (Lee (Bale et al. 2002; Régnière et al. 2012). In seasonal temperate 1991; Sinclair 1999; Chown and Nicolson 2009). environments, insects and free-living stages of parasites need to Cold-hardiness is affected by a number of factors, and cope with adverse climatic conditions in order to survive mechanisms of cold-hardiness are often species-specific and (Tinsley 1999; Wharton 1999; Addo-Bediako et al. 2000; working in interaction (Lee and Denlinger 1991; Danks 1996; Denlinger and Lee 2010). Denlinger and Lee 2010). Many insects have been observed to increase cold tolerance and survival at low temperatures by 1.3.1 Cold-hardiness and other factors affecting winter becoming physiologically acclimated (e.g. Lee 1991; Huey et al. survival 1999; Koveos 2001; Morey et al. 2012). Rapid cold hardening Insect cold-hardiness has been a widely studied topic in recent (RCH) refers to the cold acclimation response that occurs within decades (reviewed in Danks 1996; Denlinger and Lee 2010). minutes to hours in many insects when they encounter non- Cold-hardiness can be referred to as the physical and metabolic lethal low temperatures, resulting in an increase in survival adjustments (or capacity) of an insect that increase its cold probability during further colder period (Lee et al. 1987; tolerance and thus survival at low temperatures (Denlinger Denlinger and Lee 2010). The RCH is reported to occur more 1991). There have been numerous categorizing systems of insect often in freeze-intolerant than in freeze-tolerant insects (Koveos cold tolerance (e.g. Lee 1991; Bale 1993, 1996; Sinclair 1999; 2001; Lee et al. 2006; Teets et al. 2008; Denlinger and Lee 2010). Zachariassen and Kristiansen 2003; Denlinger and Lee 2010). The seasonal (or programmed) long-term cold acclimation Cold-tolerance strategies of insects have recently been classified process, instead, can take several weeks or months and works in as: 1) chilling intolerant, 2) freeze-intolerant, or 3) freeze-tolerant preparation for winter (Lee 1991; Šlachta et al. 2002b; Chown species (Chown and Nicolson 2009; Denlinger and Lee 2010). and Nicolson 2009). Both forms of cold acclimation can occur Chilling intolerant species die because of direct or indirect and improve cold-hardiness, and thus survival, at different life- chilling injury in the absence of extracellular ice formation stages, as well as during active or diapausal (i.e. dormant) (above or below 0°C). Direct chilling injury i.e. cold shock (e.g. phases (e.g. Lee et al. 1987; Burks and Hagstrum 1999; Goto et al. changes in cell membranes, protein denaturation) results from a 2001; Šlachta et al. 2002a,b; Andreadis et al. 2005; Miyazaki et al. rapid cooling during a brief low-temperature exposure. Indirect 2006; Koštál et al. 2011). chilling injury, instead, is the consequence of a long-term In addition to cold-hardiness, also diapause is an important exposure (takes days to weeks) to low temperatures likely component of winter survival in many insects and related causing developmental abnormalities. Most insects that arthropod groups (Denlinger 1991). Diapause can be defined as encounter sub-zero temperatures in nature are freeze-intolerant. an endocrine-mediated dormancy, occurring at certain life- Freeze-intolerant species cannot survive the ice formation of stages (Tauber et al. 1986; Denlinger 1991; Madder et al. 1999; body fluids, and have thus evolved various strategies to avoid Belozerov and Naumov 2002; Saunders et al. 2002; Koštál 2006). freezing (see Bale 1993, 1996). In general, freeze-intolerant According to Koštál (2006) diapause syndrome can be divided insects avoid lethal freezing by cooling small volumes of body into several following phases, although some of them are still water, and thus by lowering the temperature at which the poorly understood: 1) pre-diapause (diapause induction and

22 23 preparation phase); 2) diapause (initiation, maintenance and RCH may increase survival also in some arthropod termination); and 3) post-diapause (via post-diapausal ectoparasites that experience sudden changes in their thermal quiescence to direct development depending on changes in environment, as found in certain Dipteran species (see Chen et environmental factors). Diapause, per se, increases tolerance to al. 1987; Czajka and Lee 1990). Ectoparasites leaving the host environmental extremes during winter, enabling insects to during unfavourable seasons may be exposed to a high risk of survive adverse conditions (Lee 1991; Koštál 2006). It is not clear freezing. For example, if an ectoparasitic winter tick how diapause and cold-hardiness are related, although both are (Dermacentor albipictus) drop off the warm cervid host too early usually induced by the same environmental cues such as in spring, low temperatures are likely to kill it (Samuel 2007). In temperature (Denlinger 1991; Koštál 2006). Cold-hardiness can this kind of situations rapid cold-hardening could be of occur simultaneously with diapause (as component of the importance in arthropod ectoparasites. Also diapausing and diapause syndrome or coincidentally) or completely prolonged duration of diapause at certain life-stages can independently (e.g. in active stages of diapausing species) increase survival of ectoparasites under adverse environmental (Denlinger 1991). Diapause-mediated cold-hardening often conditions (Valera et al. 2006). requires weeks of acclimation (Tauber et al. 1986; Bale and Hayward 2010). During this time, decreasing temperatures 1.3.2 Life-history adjustments towards higher latitudes and stimulate the production of cryoprotectants (i.e. substances factors affecting offspring’s size protecting tissues from freezing injury) and thus, tolerance often Insects may also express conspicuous specializations which can is at its highest in mid-winter, decreasing towards spring in enhance their survival and potential for geographical invasion temperate areas. at low temperatures such as reduced/increased body size, faster During off-host stages, survival of temperate parasites also development, wing morphology, and melanistic colors (Nylin depends on their abilities to tolerate low seasonal temperatures and Svärd 1991; Roff 1992; Gilchrist et al. 2001; Chown and (Tinsley 1999; Wharton 1999). In the literature, there have been Nicolson 2009; Michie et al. 2010). Studies, concerning variation studies on cold-hardiness of arthropod ectoparasites, mainly in in insect body size and development, often report the Acari (Lee and Baust 1987; Burks et al. 1996a,b; Dautel and contradictory intra-specific trends towards higher latitudes. Knülle 1997). According to the references available, ectoparasitic The Bergmann’s rule states that animal body size increases mites and ticks are thought to be freezing-intolerant (e.g. toward higher latitudes because of decreasing temperatures Sømme 1981; Lee and Baust 1987; Dörr and Gothe 2001). Despite (e.g. Bergmann 1847; Atkinson and Sibly 1997; Blanckenhorn of the various studies on insect cold-hardiness, the cold- and Demont 2004; Chown and Gaston 2010). In most organisms, hardiness of free-living stages of insect ectoparasites has also in many insects, larger individuals express greater survival, received relatively little attention (e.g Schelhaas and Larson mating success, fecundity, and fertility (Roff 1992; Stearns 1992; 1989; Härkönen et al. 2010; Li et al. 2010; Nieminen et al. 2012; Nylin and Gotthard 1998; Blanckenhorn 2000). It has been paper II). Most of these existing studies concern physical or reported that large body size can increase resistance to low metabolic changes during different cold exposures. Long-term temperatures and decrease associated mortality (e.g. Smith 2002; seasonal acclimation to low temperature may occur in some Fischer et al. 2003; Zhao et al. 2010). One factor relatively often ectoparasitic ticks (Burks et al. 1996b). To my knowledge, the linked with better cold resistance is greater nutrient reserves of rapid cold-hardening process has not been reported in blood- the individual insects (see Bennett et al. 1997; Ohtsu et al. 1998). feeding ectoparasites. However, it could be assumed that the Larger body size may be associated with greater nutrient

24 25 preparation phase); 2) diapause (initiation, maintenance and RCH may increase survival also in some arthropod termination); and 3) post-diapause (via post-diapausal ectoparasites that experience sudden changes in their thermal quiescence to direct development depending on changes in environment, as found in certain Dipteran species (see Chen et environmental factors). Diapause, per se, increases tolerance to al. 1987; Czajka and Lee 1990). Ectoparasites leaving the host environmental extremes during winter, enabling insects to during unfavourable seasons may be exposed to a high risk of survive adverse conditions (Lee 1991; Koštál 2006). It is not clear freezing. For example, if an ectoparasitic winter tick how diapause and cold-hardiness are related, although both are (Dermacentor albipictus) drop off the warm cervid host too early usually induced by the same environmental cues such as in spring, low temperatures are likely to kill it (Samuel 2007). In temperature (Denlinger 1991; Koštál 2006). Cold-hardiness can this kind of situations rapid cold-hardening could be of occur simultaneously with diapause (as component of the importance in arthropod ectoparasites. Also diapausing and diapause syndrome or coincidentally) or completely prolonged duration of diapause at certain life-stages can independently (e.g. in active stages of diapausing species) increase survival of ectoparasites under adverse environmental (Denlinger 1991). Diapause-mediated cold-hardening often conditions (Valera et al. 2006). requires weeks of acclimation (Tauber et al. 1986; Bale and Hayward 2010). During this time, decreasing temperatures 1.3.2 Life-history adjustments towards higher latitudes and stimulate the production of cryoprotectants (i.e. substances factors affecting offspring’s size protecting tissues from freezing injury) and thus, tolerance often Insects may also express conspicuous specializations which can is at its highest in mid-winter, decreasing towards spring in enhance their survival and potential for geographical invasion temperate areas. at low temperatures such as reduced/increased body size, faster During off-host stages, survival of temperate parasites also development, wing morphology, and melanistic colors (Nylin depends on their abilities to tolerate low seasonal temperatures and Svärd 1991; Roff 1992; Gilchrist et al. 2001; Chown and (Tinsley 1999; Wharton 1999). In the literature, there have been Nicolson 2009; Michie et al. 2010). Studies, concerning variation studies on cold-hardiness of arthropod ectoparasites, mainly in in insect body size and development, often report the Acari (Lee and Baust 1987; Burks et al. 1996a,b; Dautel and contradictory intra-specific trends towards higher latitudes. Knülle 1997). According to the references available, ectoparasitic The Bergmann’s rule states that animal body size increases mites and ticks are thought to be freezing-intolerant (e.g. toward higher latitudes because of decreasing temperatures Sømme 1981; Lee and Baust 1987; Dörr and Gothe 2001). Despite (e.g. Bergmann 1847; Atkinson and Sibly 1997; Blanckenhorn of the various studies on insect cold-hardiness, the cold- and Demont 2004; Chown and Gaston 2010). In most organisms, hardiness of free-living stages of insect ectoparasites has also in many insects, larger individuals express greater survival, received relatively little attention (e.g Schelhaas and Larson mating success, fecundity, and fertility (Roff 1992; Stearns 1992; 1989; Härkönen et al. 2010; Li et al. 2010; Nieminen et al. 2012; Nylin and Gotthard 1998; Blanckenhorn 2000). It has been paper II). Most of these existing studies concern physical or reported that large body size can increase resistance to low metabolic changes during different cold exposures. Long-term temperatures and decrease associated mortality (e.g. Smith 2002; seasonal acclimation to low temperature may occur in some Fischer et al. 2003; Zhao et al. 2010). One factor relatively often ectoparasitic ticks (Burks et al. 1996b). To my knowledge, the linked with better cold resistance is greater nutrient reserves of rapid cold-hardening process has not been reported in blood- the individual insects (see Bennett et al. 1997; Ohtsu et al. 1998). feeding ectoparasites. However, it could be assumed that the Larger body size may be associated with greater nutrient

24 25 reserves and thus better stress resistance to declining size-dependent balancing between heat gain and heat loss performance under adverse conditions, as the relative efficiency (Olalla-Tárraga et al. 2006; Stillwell 2010). hypothesis (RE) proposes (e.g. Cushman et al. 1993; Ohgushi The third proposed pattern, countergradient variation (or 1996; Blackburn et al. 1999; Reim et al. 2006a). However, the latitudinal compensation hypothesis) has the same underlying situation is not unambiguous. The absolute energy demand mechanism as the Converse Bergmann’s rule, populations hypothesis (AED), instead, predicts that larger individuals developing faster at higher latitudes as compensation to the require more energy to be sustained and thus are at a shorter growing season (Conover and Schultz 1995; Arnett and disadvantage under demanding conditions (e.g. at low Gotelli 1999; Blanckenhorn and Demont 2004). The temperatures or under starvation) (Reim et al. 2006b; Ismail et al. countergradient variation is thought to have a genetic basis. The 2012). growth rate can be optimally compensated in relation to season Shorter season at higher latitudes may limit the time length when no size clines exist between latitudinal populations available for the utilization of resources (i.e. time for foraging, (reviewed in Blanckenhorn and Demont 2004). In addition, the growth, development and reproduction) (Blanckenhorn and counter gradient variation can be over- or undercompensated Demont 2004). The Converse Bergmann’s rule, mediated by leading to clines of Bergmann or Converse Bergmann, season length, predicts that development time and thus body respectively. size of insects decrease towards the poles. There is likely a Although intraspecific phenotypic variation between trade-off between adult body size and survival during geographical populations of many species is common, often the development until adulthood (Masaki 1978; Roff 1980; Nylin ultimate and proximate determinants behind this differentiation and Svärd 1991; Mousseau 1997; Blanckenhorn and Demont remain unclear (Stillwell 2010). Body size is affected by a range 2004). This means that shorter growth season would result in of factors. For example, maternal effects can result in phenotypic smaller individuals by constraining developmental time. At the similarities between mother and offspring via non-genetic same time, survival probability until adulthood may increase transmission of traits or maternal-effect genes (reviewed in with decreasing developmental time. In insects this `smaller size Mousseau and Fox 1998; Berloco et al. 2001). Maternal related – higher latitude´ –phenomenon has been widely observed. This factors (e.g. age, condition, physical trait) and experienced pattern may be a result of adaptive phenotypic plasticity in the environment of the mother can affect phenotype (e.g. size) of high latitude -related developmental processes and/or local individual offspring (e.g. Fox et al. 1995; Mousseau and Fox genetic adaptation (e.g. Nylin and Svärd 1991; Nylin and 1998). Mothers can also match the phenotype of their offspring Gotthard 1998; Blanckenhorn and Demont 2004; Nygren et al. to changes in the local environment and increase fitness of 2008). offspring, although sometimes maternal effects have been The categorization of the size-latitude relationships between reported to decrease offspring fitness (Marshall and Uller 2007). the Bergmann’s rule (temperature-determined large size) and To my knowledge, the role of maternal effects in the formation the Converse Bergmann’s rule (season length-mediated small of latitudinal size clines has not been studied in arthropods. size) can be an over-simplification in certain cases (Stillwell However, some information is available in other ectotherms, 2010). The situation is likely a far more complex than the such as in amphibians (Laugen et al. 2002). It is possible that contradictory rules have originally proposed. For example, maternal effects can regulate offspring’s size and, thus, function temperature-related trade-offs can result in small size instead of as a contributing factor to the latitudinal variation between large size at high latitudes in some ectotherms, owing to the populations.

26 27 reserves and thus better stress resistance to declining size-dependent balancing between heat gain and heat loss performance under adverse conditions, as the relative efficiency (Olalla-Tárraga et al. 2006; Stillwell 2010). hypothesis (RE) proposes (e.g. Cushman et al. 1993; Ohgushi The third proposed pattern, countergradient variation (or 1996; Blackburn et al. 1999; Reim et al. 2006a). However, the latitudinal compensation hypothesis) has the same underlying situation is not unambiguous. The absolute energy demand mechanism as the Converse Bergmann’s rule, populations hypothesis (AED), instead, predicts that larger individuals developing faster at higher latitudes as compensation to the require more energy to be sustained and thus are at a shorter growing season (Conover and Schultz 1995; Arnett and disadvantage under demanding conditions (e.g. at low Gotelli 1999; Blanckenhorn and Demont 2004). The temperatures or under starvation) (Reim et al. 2006b; Ismail et al. countergradient variation is thought to have a genetic basis. The 2012). growth rate can be optimally compensated in relation to season Shorter season at higher latitudes may limit the time length when no size clines exist between latitudinal populations available for the utilization of resources (i.e. time for foraging, (reviewed in Blanckenhorn and Demont 2004). In addition, the growth, development and reproduction) (Blanckenhorn and counter gradient variation can be over- or undercompensated Demont 2004). The Converse Bergmann’s rule, mediated by leading to clines of Bergmann or Converse Bergmann, season length, predicts that development time and thus body respectively. size of insects decrease towards the poles. There is likely a Although intraspecific phenotypic variation between trade-off between adult body size and survival during geographical populations of many species is common, often the development until adulthood (Masaki 1978; Roff 1980; Nylin ultimate and proximate determinants behind this differentiation and Svärd 1991; Mousseau 1997; Blanckenhorn and Demont remain unclear (Stillwell 2010). Body size is affected by a range 2004). This means that shorter growth season would result in of factors. For example, maternal effects can result in phenotypic smaller individuals by constraining developmental time. At the similarities between mother and offspring via non-genetic same time, survival probability until adulthood may increase transmission of traits or maternal-effect genes (reviewed in with decreasing developmental time. In insects this `smaller size Mousseau and Fox 1998; Berloco et al. 2001). Maternal related – higher latitude´ –phenomenon has been widely observed. This factors (e.g. age, condition, physical trait) and experienced pattern may be a result of adaptive phenotypic plasticity in the environment of the mother can affect phenotype (e.g. size) of high latitude -related developmental processes and/or local individual offspring (e.g. Fox et al. 1995; Mousseau and Fox genetic adaptation (e.g. Nylin and Svärd 1991; Nylin and 1998). Mothers can also match the phenotype of their offspring Gotthard 1998; Blanckenhorn and Demont 2004; Nygren et al. to changes in the local environment and increase fitness of 2008). offspring, although sometimes maternal effects have been The categorization of the size-latitude relationships between reported to decrease offspring fitness (Marshall and Uller 2007). the Bergmann’s rule (temperature-determined large size) and To my knowledge, the role of maternal effects in the formation the Converse Bergmann’s rule (season length-mediated small of latitudinal size clines has not been studied in arthropods. size) can be an over-simplification in certain cases (Stillwell However, some information is available in other ectotherms, 2010). The situation is likely a far more complex than the such as in amphibians (Laugen et al. 2002). It is possible that contradictory rules have originally proposed. For example, maternal effects can regulate offspring’s size and, thus, function temperature-related trade-offs can result in small size instead of as a contributing factor to the latitudinal variation between large size at high latitudes in some ectotherms, owing to the populations.

26 27 In the case of parasites, also condition of a host may affect be negatively affected and, thus, declined by the density of the reproduction performance of parasites and, thus, their alternatively prey (Snyder et al. 2005). offspring’s size. Food-limited hosts in a poor condition are likely less able to develop and maintain costly immunological or 1.4.1 Predation on insects physiological defence mechanisms, which in turn may increase Predation has often been demonstrated to regulate insect offspring quality in parasites, as reported also in some populations. For example, arthropods (e.g. Lang et al. 1999; arthropod ectoparasites (Roulin et al. 2003; Krasnov et al. 2005a; Turchin et al. 1999; Meihls et al. 2010), birds (e.g. Schultz 1983; Poulin 2007; Tschirren et al. 2007). Roland et al. 1986; Pimentel and Nilsson 2007; Van Bael et al. 2007; Sinu 2011) and small mammals (e.g. Holling 1959; Hanski and Parviainen 1985; Churchfield et al. 1991) have been shown 1.4 PREDATION AS A BIOTIC FACTOR REGULATING to cause notable mortality on their insect prey. Commonly, the POPULATION DISTRIBUTION early larval and pupal stages of insects are vulnerable to predation (e.g. Frank 1967; Hanski and Parviainen 1985; In a very general sense, predation includes carnivory (animal Tanhuanpää et al. 1999; Hastings et al. 2002; Barbaro and Battisti preys upon other living animals) and herbivory (organism eats 2011). autotrophs e.g. plants) (Price et al. 2011). Predation is often considered as an important factor for the population regulation 1.4.2 Predation on arthropod ectoparasites of prey species and also as a force driving evolution of The ecological significance of predation on parasites has rarely interacting species. Although previously a neglected topic in been acknowledged, especially when predation does not lead to ecological studies, invasive species provide an interesting transmission of parasites, or the transmission of parasite-related opportunity to study predator-prey interactions in ecosystems bacteria or viruses between individuals and species (Johnson et that have been recently occupied by a new species. Predation al. 2010). Parasites are often excluded from food-web related may have a considerable role in the invasion success of prey in research because they are difficult to measure and assumed to some cases (Pimm 1989; Lodge 1993; Vermeij 1996). There are have low total biomass (Lafferty et al. 2006; Lafferty et al. 2008; examples in the literature that native predator species switch to Johnson et al. 2010). Although previously neglected in the consume novel species possibly impeding invasion (see review ecological research, predation on parasites is widespread in in Carlsson et al. 2009; Li et al. 2011). On the contrary, according nature and found in a variety of forms (Lafferty et al. 2006). to the enemy release hypothesis (ERH), invasive species may According to Johnson et al. (2010) these forms include; 1) encounter fewer natural enemies, like predators, in novel predation on the parasite together with its host (i.e. concomitant environments resulting in rapid increase in population size (e.g. predation), 2) grooming by host itself or predation occurring on Schoener and Spiller 1995; Kotiaho and Sulkava 2007; reviewed a host by mutualistic conspecifics/other species, 3) predation on in Davis 2009). These are the two opposite ends of the free-living stages outside the host, or 4) predation within the continuum and species interactions are usually very complex in body of the host (i.e. intraguild predation). nature. For example, predation pressure may not be particularly Other biotic factors than host-related resource factors, high against a certain prey species by generalist predators, possibly regulating arthropod ectoparasite populations, have which can enable the establishment and even invasion of novel been studied to some extent. For example, natural enemies like species (Koss and Snyder 2005). In contrast, a new species can predators and pathogens can have potential to regulate

28 29 In the case of parasites, also condition of a host may affect be negatively affected and, thus, declined by the density of the reproduction performance of parasites and, thus, their alternatively prey (Snyder et al. 2005). offspring’s size. Food-limited hosts in a poor condition are likely less able to develop and maintain costly immunological or 1.4.1 Predation on insects physiological defence mechanisms, which in turn may increase Predation has often been demonstrated to regulate insect offspring quality in parasites, as reported also in some populations. For example, arthropods (e.g. Lang et al. 1999; arthropod ectoparasites (Roulin et al. 2003; Krasnov et al. 2005a; Turchin et al. 1999; Meihls et al. 2010), birds (e.g. Schultz 1983; Poulin 2007; Tschirren et al. 2007). Roland et al. 1986; Pimentel and Nilsson 2007; Van Bael et al. 2007; Sinu 2011) and small mammals (e.g. Holling 1959; Hanski and Parviainen 1985; Churchfield et al. 1991) have been shown 1.4 PREDATION AS A BIOTIC FACTOR REGULATING to cause notable mortality on their insect prey. Commonly, the POPULATION DISTRIBUTION early larval and pupal stages of insects are vulnerable to predation (e.g. Frank 1967; Hanski and Parviainen 1985; In a very general sense, predation includes carnivory (animal Tanhuanpää et al. 1999; Hastings et al. 2002; Barbaro and Battisti preys upon other living animals) and herbivory (organism eats 2011). autotrophs e.g. plants) (Price et al. 2011). Predation is often considered as an important factor for the population regulation 1.4.2 Predation on arthropod ectoparasites of prey species and also as a force driving evolution of The ecological significance of predation on parasites has rarely interacting species. Although previously a neglected topic in been acknowledged, especially when predation does not lead to ecological studies, invasive species provide an interesting transmission of parasites, or the transmission of parasite-related opportunity to study predator-prey interactions in ecosystems bacteria or viruses between individuals and species (Johnson et that have been recently occupied by a new species. Predation al. 2010). Parasites are often excluded from food-web related may have a considerable role in the invasion success of prey in research because they are difficult to measure and assumed to some cases (Pimm 1989; Lodge 1993; Vermeij 1996). There are have low total biomass (Lafferty et al. 2006; Lafferty et al. 2008; examples in the literature that native predator species switch to Johnson et al. 2010). Although previously neglected in the consume novel species possibly impeding invasion (see review ecological research, predation on parasites is widespread in in Carlsson et al. 2009; Li et al. 2011). On the contrary, according nature and found in a variety of forms (Lafferty et al. 2006). to the enemy release hypothesis (ERH), invasive species may According to Johnson et al. (2010) these forms include; 1) encounter fewer natural enemies, like predators, in novel predation on the parasite together with its host (i.e. concomitant environments resulting in rapid increase in population size (e.g. predation), 2) grooming by host itself or predation occurring on Schoener and Spiller 1995; Kotiaho and Sulkava 2007; reviewed a host by mutualistic conspecifics/other species, 3) predation on in Davis 2009). These are the two opposite ends of the free-living stages outside the host, or 4) predation within the continuum and species interactions are usually very complex in body of the host (i.e. intraguild predation). nature. For example, predation pressure may not be particularly Other biotic factors than host-related resource factors, high against a certain prey species by generalist predators, possibly regulating arthropod ectoparasite populations, have which can enable the establishment and even invasion of novel been studied to some extent. For example, natural enemies like species (Koss and Snyder 2005). In contrast, a new species can predators and pathogens can have potential to regulate

28 29 populations of arthropod ectoparasites (Harris and Burns 1972; 1.5 OBJECTIVES OF THE STUDY Samish and Rehacek 1999). In general, interspecific interactions where an organism preys upon ectoparasites of another, The present work examines temperature- and predation- related represent a relatively common form of mutualism in nature (i.e. factors, which may regulate survival and distribution of cleaning symbioses or on-host predation) (Wittenberger 1981; invasive insect ectoparasites. I have used the ectoparasitic deer Peres 1996). For example, avian predators are observed to ked (Lipoptena cervi L. 1758; Diptera; Hippoboscidae) as a model remove ectoparasites directly on the ungulate hosts (e.g. Massei species in all the studies of this thesis. This species has invaded and Genov 1995; Peres 1996; Fry and Keith 2000; Samuel et al. approximately two-thirds of Finland during the previous five 2000; Sazima 2007). There is also some documentation that decades (Välimäki et al. 2010). vertebrate and invertebrate predators can consume free-living stages of ectoparasites outside the host (Wilkinson 1970; The specific aims of the thesis were: Mwangi et al. 1991; Samish and Rehacek 1999; Sutherst et al. 2000). x to study, whether the invasive deer ked currently shows plastic and/or rapid evolutionary changes in 1.4.3 Predation cues arising from the host-ectoparasite pupal and adult characteristics between northern and interaction southern populations along temperature and historical Birds and small mammals may mainly use olfactory, auditory, invasion gradients (I). and visual cues when searching for a prey (Bennett and Cuthill 1994; Montgomerie and Weatherhead 1997; Mennerat et al. 2005; x to examine cold tolerance of free-living stages of the Nevitt et al. 2008; Vaughan et al. 2011). Direct endothermic host – deer ked: whether the off-host pupal stage (diapausing ectoparasite interaction may provide visual/olfactory cues for pupae and developing pupae) of the deer ked survive possible predators to locate parasitized host and therefore frosts occurring during winter and spring, and how exploit ectoparasites as a food source. However, so far this emerged adults tolerate autumnal frosts. The cold- aspect has not been studied and the mechanisms behind host- tolerance strategy of pupae (freezing-tolerant vs. ectoparasite-predator-interactions remain unclear. It has been freezing-intolerant) was explored. Also, possible observed, though, that avian predators can associate certain seasonal variation in cold hardiness of diapausing cues arising from plant-herbivore interactions to herbivore prey. pupae was explored, as well as possible differences in Olfactory cues, such as volatile organic compound (VOC) provisioning of mothers for pupae between winter and emissions by damaged plants, may enhance the herbivore prey spring (II). location of insectivorous birds (Mäntylä et al. 2008). Many birds can also detect ultraviolet light and/or chromatic cues when x to investigate whether there are detectable cues arising searching for a prey (Bennett and Cuthill 1994; Viitala et al. 1995; from the deer ked - host interaction on bedding sites of Church et al. 1998; Mäntylä et al. 2008; Rajchard 2009). hosts during winter. An additional aim was to develop Indicators, like leaf damage or the specific feeding habits of leaf a method to diagnose deer ked infestation on hosts that feeding caterpillars, can serve as visual cues for insectivorous would also help in the monitoring of dispersion of the birds to locate herbivore prey (Heinrich and Collins 1983; deer ked in future (III). Murakami 1999).

30 31 populations of arthropod ectoparasites (Harris and Burns 1972; 1.5 OBJECTIVES OF THE STUDY Samish and Rehacek 1999). In general, interspecific interactions where an organism preys upon ectoparasites of another, The present work examines temperature- and predation- related represent a relatively common form of mutualism in nature (i.e. factors, which may regulate survival and distribution of cleaning symbioses or on-host predation) (Wittenberger 1981; invasive insect ectoparasites. I have used the ectoparasitic deer Peres 1996). For example, avian predators are observed to ked (Lipoptena cervi L. 1758; Diptera; Hippoboscidae) as a model remove ectoparasites directly on the ungulate hosts (e.g. Massei species in all the studies of this thesis. This species has invaded and Genov 1995; Peres 1996; Fry and Keith 2000; Samuel et al. approximately two-thirds of Finland during the previous five 2000; Sazima 2007). There is also some documentation that decades (Välimäki et al. 2010). vertebrate and invertebrate predators can consume free-living stages of ectoparasites outside the host (Wilkinson 1970; The specific aims of the thesis were: Mwangi et al. 1991; Samish and Rehacek 1999; Sutherst et al. 2000). x to study, whether the invasive deer ked currently shows plastic and/or rapid evolutionary changes in 1.4.3 Predation cues arising from the host-ectoparasite pupal and adult characteristics between northern and interaction southern populations along temperature and historical Birds and small mammals may mainly use olfactory, auditory, invasion gradients (I). and visual cues when searching for a prey (Bennett and Cuthill 1994; Montgomerie and Weatherhead 1997; Mennerat et al. 2005; x to examine cold tolerance of free-living stages of the Nevitt et al. 2008; Vaughan et al. 2011). Direct endothermic host – deer ked: whether the off-host pupal stage (diapausing ectoparasite interaction may provide visual/olfactory cues for pupae and developing pupae) of the deer ked survive possible predators to locate parasitized host and therefore frosts occurring during winter and spring, and how exploit ectoparasites as a food source. However, so far this emerged adults tolerate autumnal frosts. The cold- aspect has not been studied and the mechanisms behind host- tolerance strategy of pupae (freezing-tolerant vs. ectoparasite-predator-interactions remain unclear. It has been freezing-intolerant) was explored. Also, possible observed, though, that avian predators can associate certain seasonal variation in cold hardiness of diapausing cues arising from plant-herbivore interactions to herbivore prey. pupae was explored, as well as possible differences in Olfactory cues, such as volatile organic compound (VOC) provisioning of mothers for pupae between winter and emissions by damaged plants, may enhance the herbivore prey spring (II). location of insectivorous birds (Mäntylä et al. 2008). Many birds can also detect ultraviolet light and/or chromatic cues when x to investigate whether there are detectable cues arising searching for a prey (Bennett and Cuthill 1994; Viitala et al. 1995; from the deer ked - host interaction on bedding sites of Church et al. 1998; Mäntylä et al. 2008; Rajchard 2009). hosts during winter. An additional aim was to develop Indicators, like leaf damage or the specific feeding habits of leaf a method to diagnose deer ked infestation on hosts that feeding caterpillars, can serve as visual cues for insectivorous would also help in the monitoring of dispersion of the birds to locate herbivore prey (Heinrich and Collins 1983; deer ked in future (III). Murakami 1999).

30 31 x to study whether the deer ked is predated upon during winter and to explore magnitude of possible pupal predation on bedding sites of hosts. An additional aim 2 Materials and methods was to study whether cues arising from the host-deer ked interaction increase the risk of pupal predation on host bedding sites. Finally, to evaluate if the deer ked 2.1 STUDY MODEL: THE DEER KED infestation intensity on moose population affects predation rate of pupae locally (IV). The deer ked is a haematophagous (i.e. blood feeding) louse fly (Bequaert 1953; Haarløv 1964). This ectoparasite infest on x to examine whether the deer ked has possible small several species belonging to the family Cervidae. As obligatory mammalian predators (V). parasites in general, the deer ked can’t complete its life cycle without a host (see Clayton and Moore 1997). The species can be regarded as a temporary ectoparasite, because it does not spend the entire life cycle as a parasite as permanent parasites do (see Smit and Basson 2002; Robinson 2005). The deer ked is dependent on host resources (e.g. regular blood meals and warm habitat) at certain parts of its life-cycle (Haarløv 1964). In northern Europe, adults emerge from pupae during late summer or early autumn, depending on latitude (Hackman et al. 1983; Härkönen et al. 2010). Adult emergence takes place within the forest floor litter. After emergence, adults try to find a suitable host until the first cold autumn days appear, which prevent flying (Nieminen et al. 2012). Recently emerged adults must survive without foraging until they find a suitable host. The deer ked has a direct life cycle, meaning that individual deer ked infests only one principal host during its life time. After settling on a host, males and females lose their wings (Bequaert 1953; Haarløv 1964). The on-host period of adults last for several months, to almost one year (Haarløv 1964; Härkönen 2012). During this period, adults feed repetitively on blood of the host while also reproducing. The deer ked has univoltine life cycle in the northern Europe, meaning that the species produces only one new generation in a year, adults emerging at yearly intervals (Hackman et al. 1983). The deer ked, like other louse flies (Hippobocidae), follows a reproductive strategy called adenotrophic viviparity (Bequaert 1953; Haarløv 1964; Meier et al. 1999). This means that a poorly

32 33 x to study whether the deer ked is predated upon during winter and to explore magnitude of possible pupal predation on bedding sites of hosts. An additional aim 2 Materials and methods was to study whether cues arising from the host-deer ked interaction increase the risk of pupal predation on host bedding sites. Finally, to evaluate if the deer ked 2.1 STUDY MODEL: THE DEER KED infestation intensity on moose population affects predation rate of pupae locally (IV). The deer ked is a haematophagous (i.e. blood feeding) louse fly (Bequaert 1953; Haarløv 1964). This ectoparasite infest on x to examine whether the deer ked has possible small several species belonging to the family Cervidae. As obligatory mammalian predators (V). parasites in general, the deer ked can’t complete its life cycle without a host (see Clayton and Moore 1997). The species can be regarded as a temporary ectoparasite, because it does not spend the entire life cycle as a parasite as permanent parasites do (see Smit and Basson 2002; Robinson 2005). The deer ked is dependent on host resources (e.g. regular blood meals and warm habitat) at certain parts of its life-cycle (Haarløv 1964). In northern Europe, adults emerge from pupae during late summer or early autumn, depending on latitude (Hackman et al. 1983; Härkönen et al. 2010). Adult emergence takes place within the forest floor litter. After emergence, adults try to find a suitable host until the first cold autumn days appear, which prevent flying (Nieminen et al. 2012). Recently emerged adults must survive without foraging until they find a suitable host. The deer ked has a direct life cycle, meaning that individual deer ked infests only one principal host during its life time. After settling on a host, males and females lose their wings (Bequaert 1953; Haarløv 1964). The on-host period of adults last for several months, to almost one year (Haarløv 1964; Härkönen 2012). During this period, adults feed repetitively on blood of the host while also reproducing. The deer ked has univoltine life cycle in the northern Europe, meaning that the species produces only one new generation in a year, adults emerging at yearly intervals (Hackman et al. 1983). The deer ked, like other louse flies (Hippobocidae), follows a reproductive strategy called adenotrophic viviparity (Bequaert 1953; Haarløv 1964; Meier et al. 1999). This means that a poorly

32 33 developed larva hatches from an egg already within (2010), the Finnish deer ked population has expanded its range reproductive system of a female and feeds orally secretions from at an average rate of 11 km/year towards the north, from the accessory gland of the uterus until the developed larva is ready latitude of 60°N to 65°N during the period 1960–2009. to pupate. Females deposit only one fully grown prepupated There are interesting differences between the expansion rates larva at a time. The total number of pupae produced during the of the Finnish and other Fennoscandian deer ked populations reproductive period (between the previous autumn and likely (Välimäki et al. 2010). According to Välimäki et al. (2010) the the next summer) per one female is not known, but in the other expansion rate of the eastern population in Finland has been pupiparous louse fly, the sheep ked (Melophagus ovinus), the almost twice as high as the northwards expansion rate of the offspring number has been estimated as a few dozen pupae western population in Sweden and Norway (to the latitude of (reviewed in Small 2005; Härkönen 2012). 62°N and 61°N, respectively). Potentially, the most significant After emerging from the female, pupae drop off the host at factor underlying the differences in the expansion rates of the random sites, depending on the movements of the host eastern and western Fennoscandian populations is the different (Hackman et al. 1983; Härkönen et al. 2010). Many of the pupae geographical origin. Swedish and Norwegian deer ked drop onto the bedding sites of the hosts (paper III; Välimäki et populations likely originated from Central Europe. The Swedish al. 2011). Duration of the off-host pupal stage (up to ten months) population has existed at least since the late eighteenth century, depends on birth time of pupa (paper II). Pupal stage lasts until while the Norwegian population has expanded from Sweden in the following autumn when emergence of new generation of the 1980’s (Andersson 1985; Välimäki et al. 2010). In Finland, the adults takes place. The oldest pupae, produced in autumn and increased density of the moose (Alces alces) is thought to be the early winter, overwinter in the subnivean space or within the main reason underlying the fast spread and increase of the deer different layers of snow blanket (S. Kaunisto, personal ked population during the previous 40 years (Hackman et al. observations). According to Härkönen (2012), photoperiod has 1983; Välimäki et al. 2010, 2011). However, in Sweden, moose no role in diapause regulation of the deer ked. Instead, pupae of densities are currently three times that of Finland (Lavsund et al. deer ked overwinters at a thermally opportunistic diapause, 2003), suggesting the importance of environmental factors and meaning that pupae can start direct development without differential host use in the distribution of the western deer ked obligatory diapausing when thermal conditions are suitable (see population (Välimäki et al. 2010, 2011). Tauber et al. 1986; Denlinger 1991; Saunders et al. 2002; Koštál Parasites exhibit differences in their host specificity, meaning 2006; Härkönen 2012). that some prefer only a certain host, whereas other parasite The deer ked is originally a palearctic species, but has also species are opportunistic generalists in variable degrees been introduced to Nearctic region (Maa 1969; Hackman et al. (Krasnov et al. 2004). The deer ked can be seen more as an 1983; Matsumoto et al. 2008; Välimäki et al. 2010). In Finland, the ectoparasite following a generalist strategy than as a specialist, deer ked can be classified as an invasive species owing to the because the deer ked can infest several species on family relatively rapid range expansion towards west and north during Cervidae. The main breeding host (i.e. supporting reproduction) the previous five decades (Välimäki et al. 2010). The species of the deer ked in Fennoscandia is the moose (paper III; arrived to the southeastern corner of Finland in 1960, Välimäki et al. 2011). It has been reported that one moose bull assumingly from Russia (Hackman et al. 1983). During 1970’s can host as many as 17 500 adult deer keds (Paakkonen et al. and 1980’s the deer ked spread rapidly westwards and 2010). In Finland there is also evidence of deer ked infestation simultaneously towards the north. According to Välimäki et al. on the wild forest reindeer (Rangifer tarandus fennicus) and the

34 35 developed larva hatches from an egg already within (2010), the Finnish deer ked population has expanded its range reproductive system of a female and feeds orally secretions from at an average rate of 11 km/year towards the north, from the accessory gland of the uterus until the developed larva is ready latitude of 60°N to 65°N during the period 1960–2009. to pupate. Females deposit only one fully grown prepupated There are interesting differences between the expansion rates larva at a time. The total number of pupae produced during the of the Finnish and other Fennoscandian deer ked populations reproductive period (between the previous autumn and likely (Välimäki et al. 2010). According to Välimäki et al. (2010) the the next summer) per one female is not known, but in the other expansion rate of the eastern population in Finland has been pupiparous louse fly, the sheep ked (Melophagus ovinus), the almost twice as high as the northwards expansion rate of the offspring number has been estimated as a few dozen pupae western population in Sweden and Norway (to the latitude of (reviewed in Small 2005; Härkönen 2012). 62°N and 61°N, respectively). Potentially, the most significant After emerging from the female, pupae drop off the host at factor underlying the differences in the expansion rates of the random sites, depending on the movements of the host eastern and western Fennoscandian populations is the different (Hackman et al. 1983; Härkönen et al. 2010). Many of the pupae geographical origin. Swedish and Norwegian deer ked drop onto the bedding sites of the hosts (paper III; Välimäki et populations likely originated from Central Europe. The Swedish al. 2011). Duration of the off-host pupal stage (up to ten months) population has existed at least since the late eighteenth century, depends on birth time of pupa (paper II). Pupal stage lasts until while the Norwegian population has expanded from Sweden in the following autumn when emergence of new generation of the 1980’s (Andersson 1985; Välimäki et al. 2010). In Finland, the adults takes place. The oldest pupae, produced in autumn and increased density of the moose (Alces alces) is thought to be the early winter, overwinter in the subnivean space or within the main reason underlying the fast spread and increase of the deer different layers of snow blanket (S. Kaunisto, personal ked population during the previous 40 years (Hackman et al. observations). According to Härkönen (2012), photoperiod has 1983; Välimäki et al. 2010, 2011). However, in Sweden, moose no role in diapause regulation of the deer ked. Instead, pupae of densities are currently three times that of Finland (Lavsund et al. deer ked overwinters at a thermally opportunistic diapause, 2003), suggesting the importance of environmental factors and meaning that pupae can start direct development without differential host use in the distribution of the western deer ked obligatory diapausing when thermal conditions are suitable (see population (Välimäki et al. 2010, 2011). Tauber et al. 1986; Denlinger 1991; Saunders et al. 2002; Koštál Parasites exhibit differences in their host specificity, meaning 2006; Härkönen 2012). that some prefer only a certain host, whereas other parasite The deer ked is originally a palearctic species, but has also species are opportunistic generalists in variable degrees been introduced to Nearctic region (Maa 1969; Hackman et al. (Krasnov et al. 2004). The deer ked can be seen more as an 1983; Matsumoto et al. 2008; Välimäki et al. 2010). In Finland, the ectoparasite following a generalist strategy than as a specialist, deer ked can be classified as an invasive species owing to the because the deer ked can infest several species on family relatively rapid range expansion towards west and north during Cervidae. The main breeding host (i.e. supporting reproduction) the previous five decades (Välimäki et al. 2010). The species of the deer ked in Fennoscandia is the moose (paper III; arrived to the southeastern corner of Finland in 1960, Välimäki et al. 2011). It has been reported that one moose bull assumingly from Russia (Hackman et al. 1983). During 1970’s can host as many as 17 500 adult deer keds (Paakkonen et al. and 1980’s the deer ked spread rapidly westwards and 2010). In Finland there is also evidence of deer ked infestation simultaneously towards the north. According to Välimäki et al. on the wild forest reindeer (Rangifer tarandus fennicus) and the

34 35 semi-domestic reindeer (Rangifer tarandus tarandus), although Vuuren 2006; Härkönen et al. 2010; Krasnov et al. 2010). During with much lower infestation intensity compared to the moose the long, free-living pupal stage, the deer ked can be exposed to (paper III; Kynkäänniemi et al. 2010; Välimäki et al. 2011). The various external mortality factors such as adverse winter and western Fennoscandian deer ked population can infest the roe summer temperatures, as well as predators. deer (Capreolus capreolus) more intensively than Finnish deer ked population (Välimäki et al. 2011). In Central Europe the original breeding hosts of the deer ked are likely the red deer (Cervus 2.2 GENERAL STUDY DESIGNS elaphus), the roe deer, and to the lesser extent the fallow deer (Dama dama) (Haarløv 1964; Szczurek and Kadulski 2001; Dehio The papers (I) and (III) represent observational case studies. The et al. 2004). The white-tailed deer (Odocoileus virginianus) has papers (II), (IV) and (V) are experimental in nature. The studies served as a breeding host for the deer ked in the northeastern were conducted in the laboratory or in the field. The studies regions of the USA (Matsumoto et al. 2008). were carried out in accordance with Finnish legislation during There is some evidence that the deer ked may cause direct the years 2007-2011. The main research questions, hypotheses, and indirect negative effects on cervid hosts such as stress materials and methods are briefly summarized below (see also related behavioural changes and hair loss (Kynkäänniemi et al. Table 1 a & b). More detailed descriptions are presented in the 2010; Madslien et al. 2011). The current distribution area of the original articles. deer ked in Finland partly covers the southernmost reindeer husbandry region (Välimäki et al. 2010) and thus there has been a major concern whether the deer ked can adapt to use the semi- domestic reindeer as host. During the host searching period in autumn, adult deer keds may attack humans and other unsuitable or maladaptive host objects (Kortet et al. 2010). Clinical signs of disease, such as allergic reactions on humans resulting from the deer ked bites, have been reported (Rantanen et al. 1982; Laukkanen et al. 2005). The deer ked has been suspected to function as a vector for certain bacterium such as Bartonella schoenbuchensis (Dehio et al. 2004). On the whole, the deer ked provides an excellent opportunity to study factors affecting the geographical distribution of invasive species, and particularly that of insect ectoparasites. The host-related factors (availability, density and common co- evolutive history between host and parasite) are thought to be the main cause underlying the range expansion of the deer ked and ectoparasites in general (Krasnov et al. 2004, 2010; Välimäki et al. 2010, 2011). Despite of that, external factors such as temperature, predators, and vegetation may directly be of importance (e.g. Samish and Rehacek 1999; Cumming and Van

36 37 semi-domestic reindeer (Rangifer tarandus tarandus), although Vuuren 2006; Härkönen et al. 2010; Krasnov et al. 2010). During with much lower infestation intensity compared to the moose the long, free-living pupal stage, the deer ked can be exposed to (paper III; Kynkäänniemi et al. 2010; Välimäki et al. 2011). The various external mortality factors such as adverse winter and western Fennoscandian deer ked population can infest the roe summer temperatures, as well as predators. deer (Capreolus capreolus) more intensively than Finnish deer ked population (Välimäki et al. 2011). In Central Europe the original breeding hosts of the deer ked are likely the red deer (Cervus 2.2 GENERAL STUDY DESIGNS elaphus), the roe deer, and to the lesser extent the fallow deer (Dama dama) (Haarløv 1964; Szczurek and Kadulski 2001; Dehio The papers (I) and (III) represent observational case studies. The et al. 2004). The white-tailed deer (Odocoileus virginianus) has papers (II), (IV) and (V) are experimental in nature. The studies served as a breeding host for the deer ked in the northeastern were conducted in the laboratory or in the field. The studies regions of the USA (Matsumoto et al. 2008). were carried out in accordance with Finnish legislation during There is some evidence that the deer ked may cause direct the years 2007-2011. The main research questions, hypotheses, and indirect negative effects on cervid hosts such as stress materials and methods are briefly summarized below (see also related behavioural changes and hair loss (Kynkäänniemi et al. Table 1 a & b). More detailed descriptions are presented in the 2010; Madslien et al. 2011). The current distribution area of the original articles. deer ked in Finland partly covers the southernmost reindeer husbandry region (Välimäki et al. 2010) and thus there has been a major concern whether the deer ked can adapt to use the semi- domestic reindeer as host. During the host searching period in autumn, adult deer keds may attack humans and other unsuitable or maladaptive host objects (Kortet et al. 2010). Clinical signs of disease, such as allergic reactions on humans resulting from the deer ked bites, have been reported (Rantanen et al. 1982; Laukkanen et al. 2005). The deer ked has been suspected to function as a vector for certain bacterium such as Bartonella schoenbuchensis (Dehio et al. 2004). On the whole, the deer ked provides an excellent opportunity to study factors affecting the geographical distribution of invasive species, and particularly that of insect ectoparasites. The host-related factors (availability, density and common co- evolutive history between host and parasite) are thought to be the main cause underlying the range expansion of the deer ked and ectoparasites in general (Krasnov et al. 2004, 2010; Välimäki et al. 2010, 2011). Despite of that, external factors such as temperature, predators, and vegetation may directly be of importance (e.g. Samish and Rehacek 1999; Cumming and Van

36 37 Table 1 a. Main research topics and hypotheses of the original papers I – V Table 1 b. Summary of the main study designs used in the original papers I – V

Main research topic Main research hypotheses Description of the study Measured variables

Paper I: Paper I: Paper I: Paper I: 1) Do size and development duration of - The pupae from northern Central Finland - An observational case study - Pupal size (i.e. pupal mass) pupae vary along historical invasion would be smaller than the southernmost - Measurements of life-history traits (pupae, - Pupal developmental duration until and temperature zones towards North pupae (see the Converse Bergmann’s rule). emerged adults) adult emergence Finland? - Diapausing pupae were collected from different - Body size of emerged adults (width 2) Is there size-, sex- and population- geographical areas of Finland of the head, body length) dependent variation in pupal - All pupae were reared in identical temperature - Sex development duration? and light conditions until adult emergence

Paper II: Paper II: Paper II: Paper II: 1) How off-host stages tolerate frost? - Cold tolerance would be highest in the older - Measurements of life-history traits (pupae, adults) - Pupal size (i.e. pupal mass) 2) Is there seasonal variation in cold- diapausing pupae that experience harsh frosts - Morphological and physical measurements of the - Sex hardiness of diapausing pupae of soon after birth. pupae collected from all over Finland (I) and from - The supercooling point (SCP) of different age? - Females would produce larger and more cold- northern Central Finland (II) diapausing pupae, developing pupae 3) Do maternal provisioning to offspring tolerant pupae in winter than in spring. - Short- and long-term frost exposure treatments of and adults (i.e. pupal size and cold tolerance) diapausing and developing pupae - Survival rate of pupae after long- change towards spring? - Short-term frost exposure treatment of adults term frost exposure treatments (as Paper III: Paper III: (adults emerged and they were reared in proportion of emerged adults) 1) Does the host-deer ked interaction - The commonly observed reddish-brown controlled temperature conditions) cause detectable cues on bedding sites snow discolouration on bedding sites of of cervids? moose in Finland would correlate with the Paper III: Paper III: occurrence of deer ked pupae on the - The observational data (6 Finnish localities) - Number of deer ked pupae on bedding sites (i.e. with the deer ked - 3 host types (moose; wild forest reindeer; semi- bedding sites infestation on a host). domesticated reindeer) - Occurrence of snow discolouration - The snow discolouration would result - Snow samples were collected from bedding sites - Locations of snow discolouration on mainly from tissue fluids (including blood) for detection of blood and of deer ked faeces the bedding sites of a host (due to the deer ked -related - Counting the pupal number (indicating deer ked irritation of skin) and deer ked faeces. infestation) on bedding sites

Paper IV: Paper IV: Paper IV: Paper IV: - Experimental field manipulations (5 Finnish - Pupal survival in relation to time in 1) Are deer ked pupae predated on winter - Tits (Paridae) would be the most likely localities) different treatments bedding sites of moose? predators on deer ked pupae on the bedding - Artificial bedding sites (imitating moose beddings) - Occurrence of avian predators in 2) Do cues arising from the host-deer ked sites. were provided with deer ked pupae and potential different localities interaction expose pupae to predation? - Deer ked-induced snow discoloration on host-derived cues (moose faecal pellets and snow Deer ked infestation intensity on 3) Does infestation intensity of the deer bedding sites of hosts may serve as a cue of - discoloration) for possible predators local moose population (estimations) ked on local moose population affect the deer ked infestation, and thus increase - Pupal predation monitored (5-day experiments) predation risk of pupae? pupal predation on beddings.

Paper V: Paper V: Paper V: Paper V: 1) Do small mammals prey upon deer ked - Insectivorous common shrews (Sorex - Laboratory feeding experiments with small - Pupal survival in relation to time (in pupae? araneus) would consume deer ked pupae. mammals 2 treatments, in all small mammal 2) Do small mammals prefer deer ked - Granivorous–omnivorous bank voles - Deer ked pupae and alternative food were served species) pupae in their diet? (Myodes glareolus) may consume pupae. for wild-captured common shrews, bank voles, - Consumption of alternative food - Herbivorous field voles (Microtus agrestis) field voles and semi-wild bank voles compared to pupal loss would prefer the alternative plant food provided.

38 39 Table 1 a. Main research topics and hypotheses of the original papers I – V Table 1 b. Summary of the main study designs used in the original papers I – V

Main research topic Main research hypotheses Description of the study Measured variables

38 39 2.2.1 Northward invasion, low temperatures and life-history life-stages tolerate low seasonal temperatures outside the host characteristics of the deer ked (I, II) (i.e. developing pupae in spring and emerged adults in In this part of the thesis, certain life-history characteristics of the autumn). See Table 1a for the main hypotheses of the paper (II). deer ked were estimated along species’ range in Finland (I) and The cold-tolerance strategy of diapausing and developing in relation to different temperatures (II). In the paper (I) possible pupae of the deer ked (freezing-tolerant or freezing-intolerant) variation in pupal size and development duration between was explored, because there was no previous knowledge about individuals, originating from different geographical areas, was the limits of cold tolerance in this species (II). Pupae for the first explored (see Table 1a). Pupae from moose bedding sites were experiment of the paper (II) were collected from all over Finland collected from several Finnish communes between February and (bedding sites of the moose). In further experiments of the paper March 2007 (I). After collection, pupae were weighed (altogether (II), newborn pupae were collected from fresh bedding sites of 395 pupae) and stored in a cold room (+5°C) to continue their the moose (age of <24 h) in the commune of Siikalatva (northern diapause before further laboratory experiment. At the end of Central Finland) during January, February and April. In autumn May, after two months of diapausing in a cold room, the pupae (October) pupae were collected from three moose pelts were transferred to the climate chamber (+17°C) to start their immediately after the moose were shot during annual hunting. post-diapause development, which roughly mimics This was done because moose bedding sites and pupae on them developmental conditions in nature. All the pupae, despite of are difficult to find without snow. The wild-collected pupae (4 their different geographical origins, were reared in controlled seasons) had not experienced low sub-zero temperatures in laboratory conditions with identical temperature and light nature before collections (see more detailed information in the conditions (16 h of light per day) and adult emergence period original paper). All of the pupal samples (in further experiments was observed. To study possible differences in pupal size of the paper II) were collected from the same area and the same (measured as pupal mass) and in pupal development duration over-wintering moose population during different seasons to (measured as days since raising the temperature until adult minimize the other confounding factors. Immediately after emerge) between individuals, wild-collected pupae were collections, pupal mass was measured for all individuals. divided for statistical analyses in respect to their geographical In the paper (II), possible seasonal variation in cold tolerances origin in two ways: (1) temperature zones (from south-west to of newborn diapausing pupae (born in October, January, colder north-east) and (2) invasion history (from early to late February and April), as well as cold tolerance of different off- establishment). Also, possible size- and sex-dependent variation host stages (newborn diapausing pupae, developing pupae in in pupal development duration, as well as the possible effect of the middle of metamorphosis, and emerged adults) were pupal size and pupal developmental duration on adult size was examined by using the SCP–protocol. The supercooling point explored (see Table 1a). (SCP) illustrates the lowest body temperature reached before the In the paper (II), the principal purpose was to explore cold spontaneous freezing of body fluids begins (Sømme 1982; Lee hardiness and the associated characteristics of the deer ked and Costanzo 1998; Wilson et al. 2003). Thus, the SCP has been through four seasons and the free-living off-host stages (i.e. considered as an indicator of lowest possible lethal temperature newborn diapausing pupae, developing pupae at post-diapause in freeze-intolerant species (Lee 1991). The SCP has been widely and emerging adults). The main focus of this work was to used as a comparable measure of cold-hardiness between life- investigate if the diapausing deer ked pupae survive frosts on history stages, populations, or species during short-term cold winter bedding sites of the moose, as well as, if other free-living exposure (Sømme 1982; Tauber et al. 1986; Leather et al. 1993).

40 41 2.2.1 Northward invasion, low temperatures and life-history life-stages tolerate low seasonal temperatures outside the host characteristics of the deer ked (I, II) (i.e. developing pupae in spring and emerged adults in In this part of the thesis, certain life-history characteristics of the autumn). See Table 1a for the main hypotheses of the paper (II). deer ked were estimated along species’ range in Finland (I) and The cold-tolerance strategy of diapausing and developing in relation to different temperatures (II). In the paper (I) possible pupae of the deer ked (freezing-tolerant or freezing-intolerant) variation in pupal size and development duration between was explored, because there was no previous knowledge about individuals, originating from different geographical areas, was the limits of cold tolerance in this species (II). Pupae for the first explored (see Table 1a). Pupae from moose bedding sites were experiment of the paper (II) were collected from all over Finland collected from several Finnish communes between February and (bedding sites of the moose). In further experiments of the paper March 2007 (I). After collection, pupae were weighed (altogether (II), newborn pupae were collected from fresh bedding sites of 395 pupae) and stored in a cold room (+5°C) to continue their the moose (age of <24 h) in the commune of Siikalatva (northern diapause before further laboratory experiment. At the end of Central Finland) during January, February and April. In autumn May, after two months of diapausing in a cold room, the pupae (October) pupae were collected from three moose pelts were transferred to the climate chamber (+17°C) to start their immediately after the moose were shot during annual hunting. post-diapause development, which roughly mimics This was done because moose bedding sites and pupae on them developmental conditions in nature. All the pupae, despite of are difficult to find without snow. The wild-collected pupae (4 their different geographical origins, were reared in controlled seasons) had not experienced low sub-zero temperatures in laboratory conditions with identical temperature and light nature before collections (see more detailed information in the conditions (16 h of light per day) and adult emergence period original paper). All of the pupal samples (in further experiments was observed. To study possible differences in pupal size of the paper II) were collected from the same area and the same (measured as pupal mass) and in pupal development duration over-wintering moose population during different seasons to (measured as days since raising the temperature until adult minimize the other confounding factors. Immediately after emerge) between individuals, wild-collected pupae were collections, pupal mass was measured for all individuals. divided for statistical analyses in respect to their geographical In the paper (II), possible seasonal variation in cold tolerances origin in two ways: (1) temperature zones (from south-west to of newborn diapausing pupae (born in October, January, colder north-east) and (2) invasion history (from early to late February and April), as well as cold tolerance of different off- establishment). Also, possible size- and sex-dependent variation host stages (newborn diapausing pupae, developing pupae in in pupal development duration, as well as the possible effect of the middle of metamorphosis, and emerged adults) were pupal size and pupal developmental duration on adult size was examined by using the SCP–protocol. The supercooling point explored (see Table 1a). (SCP) illustrates the lowest body temperature reached before the In the paper (II), the principal purpose was to explore cold spontaneous freezing of body fluids begins (Sømme 1982; Lee hardiness and the associated characteristics of the deer ked and Costanzo 1998; Wilson et al. 2003). Thus, the SCP has been through four seasons and the free-living off-host stages (i.e. considered as an indicator of lowest possible lethal temperature newborn diapausing pupae, developing pupae at post-diapause in freeze-intolerant species (Lee 1991). The SCP has been widely and emerging adults). The main focus of this work was to used as a comparable measure of cold-hardiness between life- investigate if the diapausing deer ked pupae survive frosts on history stages, populations, or species during short-term cold winter bedding sites of the moose, as well as, if other free-living exposure (Sømme 1982; Tauber et al. 1986; Leather et al. 1993).

40 41 The SCP measured for a taxon, individual, or tissue can vary temperature was defined as the temperature at which no considerably depending on several regulating factors. These individuals (newborn diapausing pupae) survived a four-day regulating factors include, for example, the level of ice exposure (first experiment of paper II). The time and severity of nucleating agents (hemolymph proteins and lipoproteins) frosts that the diapausing pupae survive were estimated using a present in the body fluids and tissues, anti-freeze proteins, total of 450 diapausing pupae collected from all over Finland incidental nucleators, ice–nucleating bacteria, inoculative (from bedding sites of moose). The diapausing pupae were freezing, the accumulation of low-molecular-weight exposed to frosts of -5°C, -15°C, -20°C or -25°C in addition to the cryoprotectants (e.g. glycerol, trehalose and proline), and control treatment (+5°C) for four days. In order to examine the amount of freezable water (Block 1990; Lee 1991; reviewed in tolerance of newborn pupae to prolonged frosts in more detail Lee and Costanzo 1998; Chown and Nicolson 2009). In the paper (i.e. according to pupal birth time and pupal size), an additional (II), the possible effect of rapid cold hardening on the SCP of 216 diapausing newborn pupae were collected in winter diapausing pupae was also tested (cold acclimation temperature (February) and spring (April) from Siikalatva commune and of -5°C and control temperature of + 5°C for 24 h). pupal mass was measured (further experiments II). The wild- Since the ability of the species to survive prolonged periods collected pupae (in winter and spring) had not experienced low of cold must also be evaluated, the supercooling capacity should sub-zero temperatures in nature before collections (see more not be considered as the only measure of cold-hardiness (Bale et detailed information in the original paper). Because the first al. 1988; Sømme 1996; Bale and Hayward 2010). Beside of the experiment showed that a four-day exposure below -15°C killed frequently used SCP in the past, e.g. lower lethal temperature a large proportion of the pupae, in all future long-term frost (LLT) has been used as a measure of cold-hardiness (e.g. Burks exposure treatments the diapausing winter and spring pupae, as et al. 1996a; Watanabe 2002; Bale and Hayward 2010; Khani and well as developing pupae were exposed to -15°C and -20°C for Moharramipour 2010). The LLT can be determined 1) by only three days. A control treatment was +5°C for diapausing keeping organisms at constant low temperature for varying pupae and +20°C for developing pupae. After the treatments, times to assess the longest time interval at which a significant both diapausing and developing pupae were transferred into proportion of individual survived, or 2) by keeping the time the climate chamber to start/continue development. The survival variable constant and expose organism to a range of of individuals was determined afterwards, as indicated by the temperatures (Burks et al. 1996a; Chown and Nicolson 2009). proportion of emerged adults. Since regulation of the SCP is of importance to both freeze- intolerant and freeze-tolerant insect species (Lee 1991) the 2.2.2 Cues revealing the host-deer ked interaction in winter relationship of SCP and LLT has been used as basis for (III) categorizations of cold tolerance-strategies (Bale 1993, 1996; The principal aim of the paper (III) was to explore whether the Sinclair 1999). If the species exhibit low to high mortality above deer ked infestation causes noticeable cues on the bedding sites the SCP, species can be categorized as a freeze-intolerant. of cervids. If there are perceivable cues, the bedding site Freeze-tolerant species, instead, survive at the temperatures examination-method could be a reliable tool for humans to below their SCP. detect the deer ked parasitism on cervids and even monitor In the paper (II), the methods of lower lethal temperature, dispersion of the deer ked. In addition, possible pupal predators LLT (see Czajka and Lee 1990; Burks et al. 1996a; Watanabe can utilize the cues to locate bedding sites of infested hosts, and 2002) were roughly simulated. At first, the lower lethal thus, the pupae of deer ked. The main study question was,

42 43 The SCP measured for a taxon, individual, or tissue can vary temperature was defined as the temperature at which no considerably depending on several regulating factors. These individuals (newborn diapausing pupae) survived a four-day regulating factors include, for example, the level of ice exposure (first experiment of paper II). The time and severity of nucleating agents (hemolymph proteins and lipoproteins) frosts that the diapausing pupae survive were estimated using a present in the body fluids and tissues, anti-freeze proteins, total of 450 diapausing pupae collected from all over Finland incidental nucleators, ice–nucleating bacteria, inoculative (from bedding sites of moose). The diapausing pupae were freezing, the accumulation of low-molecular-weight exposed to frosts of -5°C, -15°C, -20°C or -25°C in addition to the cryoprotectants (e.g. glycerol, trehalose and proline), and control treatment (+5°C) for four days. In order to examine the amount of freezable water (Block 1990; Lee 1991; reviewed in tolerance of newborn pupae to prolonged frosts in more detail Lee and Costanzo 1998; Chown and Nicolson 2009). In the paper (i.e. according to pupal birth time and pupal size), an additional (II), the possible effect of rapid cold hardening on the SCP of 216 diapausing newborn pupae were collected in winter diapausing pupae was also tested (cold acclimation temperature (February) and spring (April) from Siikalatva commune and of -5°C and control temperature of + 5°C for 24 h). pupal mass was measured (further experiments II). The wild- Since the ability of the species to survive prolonged periods collected pupae (in winter and spring) had not experienced low of cold must also be evaluated, the supercooling capacity should sub-zero temperatures in nature before collections (see more not be considered as the only measure of cold-hardiness (Bale et detailed information in the original paper). Because the first al. 1988; Sømme 1996; Bale and Hayward 2010). Beside of the experiment showed that a four-day exposure below -15°C killed frequently used SCP in the past, e.g. lower lethal temperature a large proportion of the pupae, in all future long-term frost (LLT) has been used as a measure of cold-hardiness (e.g. Burks exposure treatments the diapausing winter and spring pupae, as et al. 1996a; Watanabe 2002; Bale and Hayward 2010; Khani and well as developing pupae were exposed to -15°C and -20°C for Moharramipour 2010). The LLT can be determined 1) by only three days. A control treatment was +5°C for diapausing keeping organisms at constant low temperature for varying pupae and +20°C for developing pupae. After the treatments, times to assess the longest time interval at which a significant both diapausing and developing pupae were transferred into proportion of individual survived, or 2) by keeping the time the climate chamber to start/continue development. The survival variable constant and expose organism to a range of of individuals was determined afterwards, as indicated by the temperatures (Burks et al. 1996a; Chown and Nicolson 2009). proportion of emerged adults. Since regulation of the SCP is of importance to both freeze- intolerant and freeze-tolerant insect species (Lee 1991) the 2.2.2 Cues revealing the host-deer ked interaction in winter relationship of SCP and LLT has been used as basis for (III) categorizations of cold tolerance-strategies (Bale 1993, 1996; The principal aim of the paper (III) was to explore whether the Sinclair 1999). If the species exhibit low to high mortality above deer ked infestation causes noticeable cues on the bedding sites the SCP, species can be categorized as a freeze-intolerant. of cervids. If there are perceivable cues, the bedding site Freeze-tolerant species, instead, survive at the temperatures examination-method could be a reliable tool for humans to below their SCP. detect the deer ked parasitism on cervids and even monitor In the paper (II), the methods of lower lethal temperature, dispersion of the deer ked. In addition, possible pupal predators LLT (see Czajka and Lee 1990; Burks et al. 1996a; Watanabe can utilize the cues to locate bedding sites of infested hosts, and 2002) were roughly simulated. At first, the lower lethal thus, the pupae of deer ked. The main study question was,

42 43 whether the commonly observed reddish-brown snow al. 2008). The paper (IV) represents experimental manipulations discolouration found on bedding sites of the moose in Finland conducted in the field and the paper (V) laboratory feeding correlates with occurrence of reproducing deer ked on the host studies. In this section of the thesis, the possible predators on (i.e. presence of pupae on bedding) (see Table 1a). pupae of the deer ked, magnitude of pupal predation, and cues The research was conducted in six Finnish localities and with arising from the host-deer ked interaction likely increasing the three possible hosts (the moose, the wild forest reindeer and predation rate were estimated. I concentrate especially 1) on the semi-domesticated reindeer) in April–May 2008. The predators which may prey upon pupae on the snow, covering bedding sites of moose were examined in Rantsila (within a new winter bedding sites of hosts (mostly avian predators, paper IV) core area of the deer ked in 2008), Kuusamo (outside the deer and 2) on predators possibly preying upon pupae on the ground ked’s range in 2008) and Nurmes (older distribution area of the layer throughout the year, including winter period in the deer ked). The bedding sites of the semi-domesticated reindeer subnivean space (small mammals, paper V) (see Hansson and were investigated in Suomussalmi and in Hyrynsalmi (both Henttonen 1985; Churchfield 2002). localities were situated near the recent expansion front of the In the paper (IV), the principal aim was to explore whether deer ked in 2008 and thus had lower deer ked density). The the deer ked has predators, and pupae are predated on bedding bedding sites of the wild forest reindeer were analysed in sites of the moose in winter (see Table 1a). Second, it was Ristijärvi (within very recent core area of the deer ked in 2008). studied whether host-related cues arising from the host-deer The study areas were selected because they had presumably ked interaction expose pupae to predation. Third, the high density of the potential cervid hosts. The number of pupae opportunity that natural variation in the deer ked infestation was counted on bedding sites (indicating deer ked infestation on intensity between moose populations affects predation risk of host). Snow samples for blood detection were taken from the pupae was investigated. bedding sites to confirm that the snow discolouration was Field experiments were conducted in five Finnish localities in resulting from blood cells or hemoglobin of the host (in four March-April 2009 (paper IV). A rough estimation of occurrence study areas). Some of the samples were studied under the of potential avian predators on each study site was conducted microscope. Moreover, tissue samples and faeces of adult deer before the onset of the experiments, revealing occurrence of tits keds (collected from a dead moose) were compared with snow in every five study locations. Of the avian species, I focused on samples to detect, whether the snow discolouration results also tits (Paridae), because they are known predators of insects and partly from the deer ked faeces. To identify the segments of the seeds and they form feeding flocks in the northern Boreal forests bedding sites where the snow discolouration was most during winter time (see Alatalo 1980, 1982; Suhonen 1993; abundant (i.e. reflecting preferred body parts of host by adult Pimentel and Nilsson 2007). Artificial bedding sites, which deer keds), 13 bedding sites of moose were photographed in imitated natural bedding sites of the moose and contained the Rantsila and later analyzed. pupae of deer ked, were provided. The bedding sites contained also potential cues of host presence, as well as cues of deer ked 2.2.3 Predation on pupal stage of the deer ked (IV, V) infestation (moose faecal pellets and snow discoloration) to There are several ways to obtain predation data such as 1) possible predators. Artificial bedding sites were manipulated in laboratory feeding studies, 2) direct field observations, 3) increasing order of visual and olfactory cues. Pupal predation experimental manipulations in the field and 4) gut or faeces was monitored at regular intervals after the onset of 5-day analysis using e.g. molecular methods (Greenstone 1999; King et experiments. Visual and auditory observations of potential

44 45 whether the commonly observed reddish-brown snow al. 2008). The paper (IV) represents experimental manipulations discolouration found on bedding sites of the moose in Finland conducted in the field and the paper (V) laboratory feeding correlates with occurrence of reproducing deer ked on the host studies. In this section of the thesis, the possible predators on (i.e. presence of pupae on bedding) (see Table 1a). pupae of the deer ked, magnitude of pupal predation, and cues The research was conducted in six Finnish localities and with arising from the host-deer ked interaction likely increasing the three possible hosts (the moose, the wild forest reindeer and predation rate were estimated. I concentrate especially 1) on the semi-domesticated reindeer) in April–May 2008. The predators which may prey upon pupae on the snow, covering bedding sites of moose were examined in Rantsila (within a new winter bedding sites of hosts (mostly avian predators, paper IV) core area of the deer ked in 2008), Kuusamo (outside the deer and 2) on predators possibly preying upon pupae on the ground ked’s range in 2008) and Nurmes (older distribution area of the layer throughout the year, including winter period in the deer ked). The bedding sites of the semi-domesticated reindeer subnivean space (small mammals, paper V) (see Hansson and were investigated in Suomussalmi and in Hyrynsalmi (both Henttonen 1985; Churchfield 2002). localities were situated near the recent expansion front of the In the paper (IV), the principal aim was to explore whether deer ked in 2008 and thus had lower deer ked density). The the deer ked has predators, and pupae are predated on bedding bedding sites of the wild forest reindeer were analysed in sites of the moose in winter (see Table 1a). Second, it was Ristijärvi (within very recent core area of the deer ked in 2008). studied whether host-related cues arising from the host-deer The study areas were selected because they had presumably ked interaction expose pupae to predation. Third, the high density of the potential cervid hosts. The number of pupae opportunity that natural variation in the deer ked infestation was counted on bedding sites (indicating deer ked infestation on intensity between moose populations affects predation risk of host). Snow samples for blood detection were taken from the pupae was investigated. bedding sites to confirm that the snow discolouration was Field experiments were conducted in five Finnish localities in resulting from blood cells or hemoglobin of the host (in four March-April 2009 (paper IV). A rough estimation of occurrence study areas). Some of the samples were studied under the of potential avian predators on each study site was conducted microscope. Moreover, tissue samples and faeces of adult deer before the onset of the experiments, revealing occurrence of tits keds (collected from a dead moose) were compared with snow in every five study locations. Of the avian species, I focused on samples to detect, whether the snow discolouration results also tits (Paridae), because they are known predators of insects and partly from the deer ked faeces. To identify the segments of the seeds and they form feeding flocks in the northern Boreal forests bedding sites where the snow discolouration was most during winter time (see Alatalo 1980, 1982; Suhonen 1993; abundant (i.e. reflecting preferred body parts of host by adult Pimentel and Nilsson 2007). Artificial bedding sites, which deer keds), 13 bedding sites of moose were photographed in imitated natural bedding sites of the moose and contained the Rantsila and later analyzed. pupae of deer ked, were provided. The bedding sites contained also potential cues of host presence, as well as cues of deer ked 2.2.3 Predation on pupal stage of the deer ked (IV, V) infestation (moose faecal pellets and snow discoloration) to There are several ways to obtain predation data such as 1) possible predators. Artificial bedding sites were manipulated in laboratory feeding studies, 2) direct field observations, 3) increasing order of visual and olfactory cues. Pupal predation experimental manipulations in the field and 4) gut or faeces was monitored at regular intervals after the onset of 5-day analysis using e.g. molecular methods (Greenstone 1999; King et experiments. Visual and auditory observations of potential

44 45 avian predators were conducted. Signs, such as subnivean tunnels, tracks, and faeces of possible small mammalian predators (like voles, shrews, and mice), were also recorded if 3 Results and discussion noted in the immediate vicinity of an artificial bedding site. In the paper (V), the main aim was to explore whether small mammals prey upon pupae of deer ked and if small mammals 3.1 COLD TOLERANCE OF THE DEER KED AND ASSOCIATED prefer pupae over other prey items in their diet (see Table 1a). LIFE-HISTORY CHARACTERISTICS (I, II) To answer these questions, only pupae or pupae and alternative food (i.e. two treatments) were served for wild captured The contradictory trends in body size towards higher latitudes common shrews (Sorex araneus), bank voles (Myodes glareolus), have been observed among many insects and other ectotherms field voles (Microtus agrestis) and semi-wild bank voles (were (Masaki 1978; Nylin and Svärd 1991; Conover and Schultz 1995; kept in laboratory for 4-6 weeks) during summer 2008 and 2009. Atkinson and Sibly 1997; Mousseau 1997; Blanckenhorn and All the small-mammal species were captured in areas of forest Demont 2004). In the paper (I), the possible variation in the size edges, forest clear-cuts, or forest regenerations areas. These of diapausing pupae of deer ked was estimated along three habitats likely have relative high prevalence of pupae of deer different temperature zones and also along historical invasion ked owing to frequent visit rates of the moose (Heikkinen 2000). zones. Only the temperature zone of deer ked pupae, and not Observations on possible consumption of pupae and alternative the invasion history (i.e. the time of persistence of the deer ked), food were conducted at regular intervals during the laboratory seem to affect pupal size (I). This may suggest that the experiments. expansion of the deer ked towards north has caused rapid plastic changes in the pupal characteristics. However, without common-garden experiments, reciprocal transplant experiments or other alternative methods it is difficult to distinguish between genetic and environmental contributions to latitudinal size clines (Stillwell 2010). The diapausing pupae from the southernmost temperature zone with longer growing season were heavier compared to pupae collected from the northernmost temperature zone (northern Central Finland), as was expected. This result is in the line with the Converse Bergmann’s rule (Masaki 1978; Nylin and Svärd 1991; Mousseau 1997). The Converse Bergmann’s rule highlights that the season length would be the most prominent reason underlying faster development and smaller size in the north. The Converse Bergmann’s rule is often reported in free- living insects, whose early developmental stages encounter directly seasonal constraints (e.g. Nylin and Svärd 1991; Mousseau 1997; Nygren et al. 2008). Deer ked larva, on the other hand, develops until the fourth instar larval stage or prepupa

46 47 avian predators were conducted. Signs, such as subnivean tunnels, tracks, and faeces of possible small mammalian predators (like voles, shrews, and mice), were also recorded if 3 Results and discussion noted in the immediate vicinity of an artificial bedding site. In the paper (V), the main aim was to explore whether small mammals prey upon pupae of deer ked and if small mammals 3.1 COLD TOLERANCE OF THE DEER KED AND ASSOCIATED prefer pupae over other prey items in their diet (see Table 1a). LIFE-HISTORY CHARACTERISTICS (I, II) To answer these questions, only pupae or pupae and alternative food (i.e. two treatments) were served for wild captured The contradictory trends in body size towards higher latitudes common shrews (Sorex araneus), bank voles (Myodes glareolus), have been observed among many insects and other ectotherms field voles (Microtus agrestis) and semi-wild bank voles (were (Masaki 1978; Nylin and Svärd 1991; Conover and Schultz 1995; kept in laboratory for 4-6 weeks) during summer 2008 and 2009. Atkinson and Sibly 1997; Mousseau 1997; Blanckenhorn and All the small-mammal species were captured in areas of forest Demont 2004). In the paper (I), the possible variation in the size edges, forest clear-cuts, or forest regenerations areas. These of diapausing pupae of deer ked was estimated along three habitats likely have relative high prevalence of pupae of deer different temperature zones and also along historical invasion ked owing to frequent visit rates of the moose (Heikkinen 2000). zones. Only the temperature zone of deer ked pupae, and not Observations on possible consumption of pupae and alternative the invasion history (i.e. the time of persistence of the deer ked), food were conducted at regular intervals during the laboratory seem to affect pupal size (I). This may suggest that the experiments. expansion of the deer ked towards north has caused rapid plastic changes in the pupal characteristics. However, without common-garden experiments, reciprocal transplant experiments or other alternative methods it is difficult to distinguish between genetic and environmental contributions to latitudinal size clines (Stillwell 2010). The diapausing pupae from the southernmost temperature zone with longer growing season were heavier compared to pupae collected from the northernmost temperature zone (northern Central Finland), as was expected. This result is in the line with the Converse Bergmann’s rule (Masaki 1978; Nylin and Svärd 1991; Mousseau 1997). The Converse Bergmann’s rule highlights that the season length would be the most prominent reason underlying faster development and smaller size in the north. The Converse Bergmann’s rule is often reported in free- living insects, whose early developmental stages encounter directly seasonal constraints (e.g. Nylin and Svärd 1991; Mousseau 1997; Nygren et al. 2008). Deer ked larva, on the other hand, develops until the fourth instar larval stage or prepupa

46 47 within female, who is attached on an endothermic host. Thus, greater energy storages and lower vulnerability to chill-injury conditions for the larva are likely relatively stable and resources (e.g. Smith 2002; Fischer et al. 2003; Zhao et al. 2010). However, constantly available until pupal formation (Haarløv 1964; the underlying mechanisms for the relationship between better Hackman et al. 1983; Lehane 2005). However, the season length cold resistance and large size remain largely unsolved. or low temperatures in north could affect pupal size of the deer The pupal size showed a positive association with pupal ked for example via maternal effects. Studies report that development duration and adult size (despite of different temperature during pupal development until adult emergence geographical origin of pupae), indicating that small adults may affect adult phenotype and size in some species (e.g. emerged earlier from smaller pupae compared to larger Stevens 2004), which in turn may have cross-generational effects individuals (paper I). This supports the idea that smaller on the phenotype of offspring (see Mousseau and Fox 1998; individuals contain less metabolic reserves, and thus exhibit a Marshall and Uller 2007). decreased ability to sustain a long non-feeding pupal stage In northern cold environments, smaller size could enhance (Matsuo 2006; Hahn and Denlinger 2007). Larger body size resistance to freezing by being associated with better could be helpful during diapause period when individual supercooling capacity (Lee 1991). Supercooling capacity has consumes the limited amount of resources stored before been found to correlate with water volume of body and body entering the dormancy. Large individuals can cover the costs weight among certain insect and arachnid species, thus the (e.g. body maintaining and possible cold-hardening) that capacity to supercool to lower temperatures decreases with accumulate with prolonged diapause at higher latitudes. increasing body mass (Sømme 1982; Lee 1991; Lee and Costanzo However, length of the favorable growth season can have 1998). This “smaller size – lower SCP” trend can hold true also complex effects on the development time and size (see Nylin intra-specifically (e.g. Johnston and Lee 1990; Pugh 1994; Sinclair and Svärd 1991; Conover and Schultz 1995; Nylin and Gotthard and Chown 2005; Hahn et al. 2008). However, in the paper (II), 1998; Blanckenhorn and Demont 2004). In the north, smaller pupal mass did not significantly affect the supercooling points deer ked pupae with earlier adult emergence (paper I) could of diapausing pupae born in different months when all pupae have an advantage during host searching period which is were collected from the same overwintering area in the northern shortened due to earlier autumnal frosts and winter. Central Finland. Alternatively, small size towards north can be a It was expected that the pupal size would be largest among result of optimal thermal balancing between heat gain and heat older diapausing pupae having longer non-feeding period and loss (Olalla-Tárraga et al. 2006; Stillwell 2010). facing harsh winter temperatures (paper II, see Table 1a). In In the paper (I), the smallest individuals among the northern general, females are expected to invest more in offspring if a pupae seemed unsuccessful because no adults were observed to small increase in provisioning increases survival probability emerge. In the paper (II), it was observed that the mortality of (Smith and Fretwell 1974; Parker and Begon 1986; Plaistow et al. small pupae increased with the severity of long-term frost 2007). However, in the paper (II), seasonal variation in size of exposure (3 days for -15°C or -20°C), whereas large pupae diapausing pupae did not correlate with the occurrence of harsh survived well. These results suggest that small size can be a frosts or birth month. Instead, pupal size increased towards disadvantage, while large pupal size can reduce pre-freeze spring. Additionally, the larger pupal size fully explained the mortality during prolonged frost periods. In certain insect higher cold tolerance of pupae in spring than in winter. In some species, large body size is reported to increase low-temperature insects, the final reproductive investment of old females to their resistance and associated survival. This is potentially a result of last remaining offspring has been reported (c.f. Stearns 1992;

48 49 within female, who is attached on an endothermic host. Thus, greater energy storages and lower vulnerability to chill-injury conditions for the larva are likely relatively stable and resources (e.g. Smith 2002; Fischer et al. 2003; Zhao et al. 2010). However, constantly available until pupal formation (Haarløv 1964; the underlying mechanisms for the relationship between better Hackman et al. 1983; Lehane 2005). However, the season length cold resistance and large size remain largely unsolved. or low temperatures in north could affect pupal size of the deer The pupal size showed a positive association with pupal ked for example via maternal effects. Studies report that development duration and adult size (despite of different temperature during pupal development until adult emergence geographical origin of pupae), indicating that small adults may affect adult phenotype and size in some species (e.g. emerged earlier from smaller pupae compared to larger Stevens 2004), which in turn may have cross-generational effects individuals (paper I). This supports the idea that smaller on the phenotype of offspring (see Mousseau and Fox 1998; individuals contain less metabolic reserves, and thus exhibit a Marshall and Uller 2007). decreased ability to sustain a long non-feeding pupal stage In northern cold environments, smaller size could enhance (Matsuo 2006; Hahn and Denlinger 2007). Larger body size resistance to freezing by being associated with better could be helpful during diapause period when individual supercooling capacity (Lee 1991). Supercooling capacity has consumes the limited amount of resources stored before been found to correlate with water volume of body and body entering the dormancy. Large individuals can cover the costs weight among certain insect and arachnid species, thus the (e.g. body maintaining and possible cold-hardening) that capacity to supercool to lower temperatures decreases with accumulate with prolonged diapause at higher latitudes. increasing body mass (Sømme 1982; Lee 1991; Lee and Costanzo However, length of the favorable growth season can have 1998). This “smaller size – lower SCP” trend can hold true also complex effects on the development time and size (see Nylin intra-specifically (e.g. Johnston and Lee 1990; Pugh 1994; Sinclair and Svärd 1991; Conover and Schultz 1995; Nylin and Gotthard and Chown 2005; Hahn et al. 2008). However, in the paper (II), 1998; Blanckenhorn and Demont 2004). In the north, smaller pupal mass did not significantly affect the supercooling points deer ked pupae with earlier adult emergence (paper I) could of diapausing pupae born in different months when all pupae have an advantage during host searching period which is were collected from the same overwintering area in the northern shortened due to earlier autumnal frosts and winter. Central Finland. Alternatively, small size towards north can be a It was expected that the pupal size would be largest among result of optimal thermal balancing between heat gain and heat older diapausing pupae having longer non-feeding period and loss (Olalla-Tárraga et al. 2006; Stillwell 2010). facing harsh winter temperatures (paper II, see Table 1a). In In the paper (I), the smallest individuals among the northern general, females are expected to invest more in offspring if a pupae seemed unsuccessful because no adults were observed to small increase in provisioning increases survival probability emerge. In the paper (II), it was observed that the mortality of (Smith and Fretwell 1974; Parker and Begon 1986; Plaistow et al. small pupae increased with the severity of long-term frost 2007). However, in the paper (II), seasonal variation in size of exposure (3 days for -15°C or -20°C), whereas large pupae diapausing pupae did not correlate with the occurrence of harsh survived well. These results suggest that small size can be a frosts or birth month. Instead, pupal size increased towards disadvantage, while large pupal size can reduce pre-freeze spring. Additionally, the larger pupal size fully explained the mortality during prolonged frost periods. In certain insect higher cold tolerance of pupae in spring than in winter. In some species, large body size is reported to increase low-temperature insects, the final reproductive investment of old females to their resistance and associated survival. This is potentially a result of last remaining offspring has been reported (c.f. Stearns 1992;

48 49 Plaistow et al. 2007; Kindsvater et al. 2010; Marshall et al. 2010). the experiment. Rapid cold-hardening process has been Old females may invest in large offspring because their own reported to increase survival in species that experience sudden survival probability is low and future reproductive events are changes in their thermal environment (Chen et al. 1987; Czajka unlikely. On the other hand, production of smaller offspring in and Lee 1990). Ectoparasites, such as the deer ked, that drop off the beginning of the reproductive period, when the survival the host during unfavourable seasons may be exposed to a high probability of offspring is lower, may increase the female fitness risk of freezing (see Samuel 2007). Thus, the low supercooling by increasing lifetime fecundity of the mother (Mousseau and points of deer ked pupae, even without possible cold Dingle 1991; Roff 1992; Marshall et al. 2010). In addition, a host acclimation, may increase the survival probability of newborn individual may not be able to maintain costly immunological or pupae when they drop off the host. physiological defence mechanisms during long and food-limited In developing pupae the SCP was -20°C. The developing winter period, which in turn may increase pupal size of the deer pupae had also the potential to survive prolonged harsh frosts. ked towards spring (see Roulin et al. 2003; Krasnov et al. 2005a; However, the mortality was very high during 3-day frost Poulin 2007; Tschirren et al. 2007). exposure above their SCP, suggesting freezing-intolerant The results of the paper (II) propose that the deer ked follows strategy (paper II). Also, at the adult stage, the supercooling a freezing-intolerant strategy during pupal stage, both in point was surprisingly low (ƺ21°C). The life-history stages that diapausing and developing pupae. To be specific, the deer ked do not experience such low temperatures in nature (developing could be classified as following a moderately chill tolerant pupae in spring and summer, and adults in autumn) were still strategy (see Bale 1993, 1996). All diapausing pupae, regardless able to cope with serious frosts. Due to the high cold tolerance of of their birth month, seem to have quite low SCP (-26°C), but all free-living stages of deer ked, the further northward they show relatively high pre-freeze mortality well above the expansion is likely. However, at higher latitudes the shorter SCP after prolonged chilling period. The duration and severity growing season and early autumnal frosts may shorten the host of longer frost exposure had a great impact on pupal survival. search period of emerged deer keds by preventing flying (see Being based on the results of the long-term frost exposure Nieminen et al. 2012). On the other hand, the deer ked seems to treatments, it could be roughly stated that diapausing pupae can show plasticity in the regulation of adult emerging time (paper survive 3 to 4 days on the bedding sites of moose when I). Despite of the different temperature origin and different temperature is ca. -15°C to -20°C, while frosts of ƺ20°C or ƺ25°C invasion history of the pupae, the adult emergence period was for 4 days will kill all diapausing pupae. However, even in relatively synchronized when reared under identical harsh winter conditions deer ked pupae may survive by temperature and light conditions. burrowing under the snow and using it as insulation. Survival The cold tolerance of the deer ked was reduced during the of diapausing deer ked pupae is thus partly dependent on the end of the free-living life-span (paper II), which may be related timing of snow falls. to starvation, because the supercooling capacity can be a Newborn diapausing pupae were able to supercool to consequence of ability to survive prolonged periods of relatively low temperatures, even without rapid cold hardening starvation and desiccation when outside the host (Dautel and (i.e. short-term cold acclimation experiment did not have effect Knülle 1997; Lavy et al. 1997; Chown and Nicolson 2009). on the SCP) (paper II). It is possible, that already the rapid Presumably, the deer ked can rely on its maternally derived temperature change (decrease of 30°C or more) when dropping resources such as body fat composition during long non-feeding off a host had been enough to acclimate deer ked pupae prior to life-stages and possibly also during the cold-hardening. Body fat

50 51 Plaistow et al. 2007; Kindsvater et al. 2010; Marshall et al. 2010). the experiment. Rapid cold-hardening process has been Old females may invest in large offspring because their own reported to increase survival in species that experience sudden survival probability is low and future reproductive events are changes in their thermal environment (Chen et al. 1987; Czajka unlikely. On the other hand, production of smaller offspring in and Lee 1990). Ectoparasites, such as the deer ked, that drop off the beginning of the reproductive period, when the survival the host during unfavourable seasons may be exposed to a high probability of offspring is lower, may increase the female fitness risk of freezing (see Samuel 2007). Thus, the low supercooling by increasing lifetime fecundity of the mother (Mousseau and points of deer ked pupae, even without possible cold Dingle 1991; Roff 1992; Marshall et al. 2010). In addition, a host acclimation, may increase the survival probability of newborn individual may not be able to maintain costly immunological or pupae when they drop off the host. physiological defence mechanisms during long and food-limited In developing pupae the SCP was -20°C. The developing winter period, which in turn may increase pupal size of the deer pupae had also the potential to survive prolonged harsh frosts. ked towards spring (see Roulin et al. 2003; Krasnov et al. 2005a; However, the mortality was very high during 3-day frost Poulin 2007; Tschirren et al. 2007). exposure above their SCP, suggesting freezing-intolerant The results of the paper (II) propose that the deer ked follows strategy (paper II). Also, at the adult stage, the supercooling a freezing-intolerant strategy during pupal stage, both in point was surprisingly low (ƺ21°C). The life-history stages that diapausing and developing pupae. To be specific, the deer ked do not experience such low temperatures in nature (developing could be classified as following a moderately chill tolerant pupae in spring and summer, and adults in autumn) were still strategy (see Bale 1993, 1996). All diapausing pupae, regardless able to cope with serious frosts. Due to the high cold tolerance of of their birth month, seem to have quite low SCP (-26°C), but all free-living stages of deer ked, the further northward they show relatively high pre-freeze mortality well above the expansion is likely. However, at higher latitudes the shorter SCP after prolonged chilling period. The duration and severity growing season and early autumnal frosts may shorten the host of longer frost exposure had a great impact on pupal survival. search period of emerged deer keds by preventing flying (see Being based on the results of the long-term frost exposure Nieminen et al. 2012). On the other hand, the deer ked seems to treatments, it could be roughly stated that diapausing pupae can show plasticity in the regulation of adult emerging time (paper survive 3 to 4 days on the bedding sites of moose when I). Despite of the different temperature origin and different temperature is ca. -15°C to -20°C, while frosts of ƺ20°C or ƺ25°C invasion history of the pupae, the adult emergence period was for 4 days will kill all diapausing pupae. However, even in relatively synchronized when reared under identical harsh winter conditions deer ked pupae may survive by temperature and light conditions. burrowing under the snow and using it as insulation. Survival The cold tolerance of the deer ked was reduced during the of diapausing deer ked pupae is thus partly dependent on the end of the free-living life-span (paper II), which may be related timing of snow falls. to starvation, because the supercooling capacity can be a Newborn diapausing pupae were able to supercool to consequence of ability to survive prolonged periods of relatively low temperatures, even without rapid cold hardening starvation and desiccation when outside the host (Dautel and (i.e. short-term cold acclimation experiment did not have effect Knülle 1997; Lavy et al. 1997; Chown and Nicolson 2009). on the SCP) (paper II). It is possible, that already the rapid Presumably, the deer ked can rely on its maternally derived temperature change (decrease of 30°C or more) when dropping resources such as body fat composition during long non-feeding off a host had been enough to acclimate deer ked pupae prior to life-stages and possibly also during the cold-hardening. Body fat

50 51 is an important energy source during starvation (Colinet et al. with the predictions, host-deer ked interaction -related snow 2006), and fatty acid composition has been found to be involved discolouration is caused by tissue fluid and blood of the host, in the cold-hardening of dipterans (e.g. Bennett et al. 1997; Ohtsu which are likely resulting from bites of the deer ked and the et al. 1998). In addition to this, Nieminen et al. (2012) reported associated irritation of skin. In addition, snow of the bedding that the free-living adult deer keds do not accumulate large sites was found to be stained with deer ked faeces. The concentrations of low molecular weight cryoprotectants (except occurrence of deer ked pupae on bedding sites (i.e. deer ked for certain amino acids such as proline) to increase their cold infestation on a host) was associated with snow discolouration tolerance. This may suggest the possibility of other secondary in the localities having higher deer ked densities. In the two mechanisms behind the cold hardening ability. locations, lacking deer ked or having notably lower deer ked In the paper (II), the survival of adults during prolonged frost infestation intensity (Kuusamo and Suomussalmi), no snow exposure was not studied, but Nieminen et al. (2012) discolouration or blood indicators were detected. demonstrated that adults might survive at temperatures above In Rantsila, where both pupal and snow discolouration ƺ16°C after prolonged frost exposure. However they reported a prevalence were high on the bedding sites of the moose, the significantly higher supercooling point of adults (ƺ7,8°C, n = 6) commonly observed reddish-brown snow discolouration than found here (II) (-21°C, n = 34). According to the high SCP, correlated with the occurrence of reproducing deer ked on hosts Nieminen et al. (2012) concluded that the deer ked adults exploit (paper III). Most of the moose bedding sites in Rantsila were a freeze-tolerant strategy. It has been reported that some insect discoloured (indicated by haemoglobin), and some of the species can switch their cold-hardiness strategy between life- bedding sites contained detectable blood (blood cells). In stages or between years under varying environmental contrast, in Kuusamo (outside the deer ked’s range in 2008), conditions (Kukal and Duman 1989; Vernon and Vannier 2001; bedding sites of moose were clean, and no blood indicators or Bouchard et al. 2006). Taken together, the present results may deer ked pupae were found. still suggest the freeze-intolerant strategy also for adult stage of My results are the first to show that the wild forest reindeer the deer ked (SCP = -21°C, paper II; and LLT = -16°C, Nieminen (Ristijärvi location) and the semi-domesticated reindeer et al. 2012). (Hyrynsalmi) can serve as hosts for the deer ked, although with notably lower infestation intensity compared to the moose (paper III). Occurrence of deer ked pupae associated with the 3.2 CUES ARISING FROM THE HOST-DEER KED INTERACTION snow discolouration (hemoglobin was found) on the bedding (III) sites of the wild forest reindeer and of the semi-domesticated reindeer. To support my findings, Välimäki et al. (2011) has Insect ectoparasites exhibit a wide range of forms of association reported that the wild forest reindeer can be considered a with their mammalian and bird hosts (Wall 2007). These potential, but low-quality auxiliary host for the deer ked ectoparasites are reported to cause direct and indirect harm to population in Finland. Results of experimental infections by their hosts (Balashov 2007; Wall 2007; Kynkäänniemi et al. 2010; Kynkäänniemi et al. (2010) showed that the deer ked can use the Madslien et al. 2011). The paper (III) reports, as expected, that semi-domestic reindeer as a host, but with low reproductive the commonly observed reddish-brown snow discolouration on success. cervid (mostly the moose) bedding sites in Finland results from The degree of snow discolouration was highest on the the deer ked infestation on the host (see table 1a). In the line segments of the bedding sites that had been in touch with the

52 53 is an important energy source during starvation (Colinet et al. with the predictions, host-deer ked interaction -related snow 2006), and fatty acid composition has been found to be involved discolouration is caused by tissue fluid and blood of the host, in the cold-hardening of dipterans (e.g. Bennett et al. 1997; Ohtsu which are likely resulting from bites of the deer ked and the et al. 1998). In addition to this, Nieminen et al. (2012) reported associated irritation of skin. In addition, snow of the bedding that the free-living adult deer keds do not accumulate large sites was found to be stained with deer ked faeces. The concentrations of low molecular weight cryoprotectants (except occurrence of deer ked pupae on bedding sites (i.e. deer ked for certain amino acids such as proline) to increase their cold infestation on a host) was associated with snow discolouration tolerance. This may suggest the possibility of other secondary in the localities having higher deer ked densities. In the two mechanisms behind the cold hardening ability. locations, lacking deer ked or having notably lower deer ked In the paper (II), the survival of adults during prolonged frost infestation intensity (Kuusamo and Suomussalmi), no snow exposure was not studied, but Nieminen et al. (2012) discolouration or blood indicators were detected. demonstrated that adults might survive at temperatures above In Rantsila, where both pupal and snow discolouration ƺ16°C after prolonged frost exposure. However they reported a prevalence were high on the bedding sites of the moose, the significantly higher supercooling point of adults (ƺ7,8°C, n = 6) commonly observed reddish-brown snow discolouration than found here (II) (-21°C, n = 34). According to the high SCP, correlated with the occurrence of reproducing deer ked on hosts Nieminen et al. (2012) concluded that the deer ked adults exploit (paper III). Most of the moose bedding sites in Rantsila were a freeze-tolerant strategy. It has been reported that some insect discoloured (indicated by haemoglobin), and some of the species can switch their cold-hardiness strategy between life- bedding sites contained detectable blood (blood cells). In stages or between years under varying environmental contrast, in Kuusamo (outside the deer ked’s range in 2008), conditions (Kukal and Duman 1989; Vernon and Vannier 2001; bedding sites of moose were clean, and no blood indicators or Bouchard et al. 2006). Taken together, the present results may deer ked pupae were found. still suggest the freeze-intolerant strategy also for adult stage of My results are the first to show that the wild forest reindeer the deer ked (SCP = -21°C, paper II; and LLT = -16°C, Nieminen (Ristijärvi location) and the semi-domesticated reindeer et al. 2012). (Hyrynsalmi) can serve as hosts for the deer ked, although with notably lower infestation intensity compared to the moose (paper III). Occurrence of deer ked pupae associated with the 3.2 CUES ARISING FROM THE HOST-DEER KED INTERACTION snow discolouration (hemoglobin was found) on the bedding (III) sites of the wild forest reindeer and of the semi-domesticated reindeer. To support my findings, Välimäki et al. (2011) has Insect ectoparasites exhibit a wide range of forms of association reported that the wild forest reindeer can be considered a with their mammalian and bird hosts (Wall 2007). These potential, but low-quality auxiliary host for the deer ked ectoparasites are reported to cause direct and indirect harm to population in Finland. Results of experimental infections by their hosts (Balashov 2007; Wall 2007; Kynkäänniemi et al. 2010; Kynkäänniemi et al. (2010) showed that the deer ked can use the Madslien et al. 2011). The paper (III) reports, as expected, that semi-domestic reindeer as a host, but with low reproductive the commonly observed reddish-brown snow discolouration on success. cervid (mostly the moose) bedding sites in Finland results from The degree of snow discolouration was highest on the the deer ked infestation on the host (see table 1a). In the line segments of the bedding sites that had been in touch with the

52 53 neck and back regions of the moose, indicating the likely montanus), the crested tit (Lophophanes cristatus), the great tit preferred attachment sites of adult deer keds (III). Paakkonen et (Parus major), the blue tit (Cyanistes caeruleus), and the coal tit al. (2010) also found the highest densities of adult deer keds to (Periparus ater) were all observed. The predation pressure by tits occur on the scapula and the anterior dorsal region of the on pupae of deer ked was notable during winter because 34.7% moose. of all pupae used in the study (all treatments together) were In addition to possible pupal predators and prey detection, predated upon, even within a relatively short time interval (5 this perceivable cue arising from the host – deer ked interaction days). No signs of small mammalian predators were detected in can be used by humans as a tool to detect deer ked infestation the immediate vicinity of the artificial bedding sites, which was on host individuals and even monitor dispersion of the deer expected because small mammals (voles, shrews, mice) spend ked. Indeed, Välimäki et al. (2011) has further developed this most of the winter foraging in the subnivean space to minimize bedding site examination method within and between possible their predation risk (e.g. Hansson and Henttonen 1985; host species in Finland (the moose, the roe deer, the white-tailed Aitchison 1987). deer, the wild forest reindeer) and Norway (the moose, the roe The reported reddish-brown snow discoloration of moose deer). Välimäki et al. (2011) found that the reproductive output bedding sites, which indicated high deer ked infestation of adult deer ked (i.e. the number of pupae) correlated intensity on a host (paper III; Välimäki et al. 2011), significantly positively with the number of discoloured sections on moose increased the probability of pupae being predated upon on bedding sites both in Finland and in Norway. There was a experimental bedding sites (paper IV, see table 1a). The moose similar relationship between these two variables also in the faecal pellets studied as a possible cue for predators did not Norwegian roe deer population (Välimäki et al. 2011). increase the risk of pupal predation (IV). The lack of response is understandable, as host faeces may not reliably indicate the presence of prey items on a bedding site, unlike the commonly 3.3 PREDATION ON PUPAL STAGE OF THE DEER KED (IV, V) observed snow discoloration that derives from the direct host– parasite interaction. Most of the artificial bedding sites without Prior to my work, it was unknown, whether the host - the snow discoloration cue remained undisturbed during a 5 ectoparasite interaction could cause perceivable and fast day monitoring period. This also held true in the study areas exploitable cues for predators searching for small prey items where the infestation intensity of deer ked within the natural such as ectoparasitic insects and arachnids. However, there moose population was relatively high. Perhaps black pupae were few references on predators consuming free-living stages against a white background do not provide enough information of ectoparasites (Acari) outside the host (Wilkinson 1970; for tits, and thus clean, non-coloured bedding sites are merely Mwangi et al. 1991; Samish and Rehacek 1999; Sutherst et al. foraged upon incidentally. Alternatively, tits can discover non- 2000). In the northern boreal region, endothermic vertebrates are coloured bedding sites, but the lack of information on the potential predators of arthropods and thus free-living stages of profitability of the site leads the predators to reject it. ectoparasites during the winter (see e.g. Alatalo 1980; Jansson Potentially, tits may use discoloration to evaluate expected and von Brömssen 1981). nutritional gain and adjust their time expenditure within a Several tit species were observed in the vicinity of the particular bedding site adaptively (e.g. Naef-Daenzer et al. 2000), artificial bedding sites in all five localities (Konnevesi, Pulkkila, because in the heavily infested study areas many of the Rantsila, Utajärvi and Yli-Ii) (paper IV). The willow tit (Poecile discoloured bedding sites were entirely emptied by predators

54 55 neck and back regions of the moose, indicating the likely montanus), the crested tit (Lophophanes cristatus), the great tit preferred attachment sites of adult deer keds (III). Paakkonen et (Parus major), the blue tit (Cyanistes caeruleus), and the coal tit al. (2010) also found the highest densities of adult deer keds to (Periparus ater) were all observed. The predation pressure by tits occur on the scapula and the anterior dorsal region of the on pupae of deer ked was notable during winter because 34.7% moose. of all pupae used in the study (all treatments together) were In addition to possible pupal predators and prey detection, predated upon, even within a relatively short time interval (5 this perceivable cue arising from the host – deer ked interaction days). No signs of small mammalian predators were detected in can be used by humans as a tool to detect deer ked infestation the immediate vicinity of the artificial bedding sites, which was on host individuals and even monitor dispersion of the deer expected because small mammals (voles, shrews, mice) spend ked. Indeed, Välimäki et al. (2011) has further developed this most of the winter foraging in the subnivean space to minimize bedding site examination method within and between possible their predation risk (e.g. Hansson and Henttonen 1985; host species in Finland (the moose, the roe deer, the white-tailed Aitchison 1987). deer, the wild forest reindeer) and Norway (the moose, the roe The reported reddish-brown snow discoloration of moose deer). Välimäki et al. (2011) found that the reproductive output bedding sites, which indicated high deer ked infestation of adult deer ked (i.e. the number of pupae) correlated intensity on a host (paper III; Välimäki et al. 2011), significantly positively with the number of discoloured sections on moose increased the probability of pupae being predated upon on bedding sites both in Finland and in Norway. There was a experimental bedding sites (paper IV, see table 1a). The moose similar relationship between these two variables also in the faecal pellets studied as a possible cue for predators did not Norwegian roe deer population (Välimäki et al. 2011). increase the risk of pupal predation (IV). The lack of response is understandable, as host faeces may not reliably indicate the presence of prey items on a bedding site, unlike the commonly 3.3 PREDATION ON PUPAL STAGE OF THE DEER KED (IV, V) observed snow discoloration that derives from the direct host– parasite interaction. Most of the artificial bedding sites without Prior to my work, it was unknown, whether the host - the snow discoloration cue remained undisturbed during a 5 ectoparasite interaction could cause perceivable and fast day monitoring period. This also held true in the study areas exploitable cues for predators searching for small prey items where the infestation intensity of deer ked within the natural such as ectoparasitic insects and arachnids. However, there moose population was relatively high. Perhaps black pupae were few references on predators consuming free-living stages against a white background do not provide enough information of ectoparasites (Acari) outside the host (Wilkinson 1970; for tits, and thus clean, non-coloured bedding sites are merely Mwangi et al. 1991; Samish and Rehacek 1999; Sutherst et al. foraged upon incidentally. Alternatively, tits can discover non- 2000). In the northern boreal region, endothermic vertebrates are coloured bedding sites, but the lack of information on the potential predators of arthropods and thus free-living stages of profitability of the site leads the predators to reject it. ectoparasites during the winter (see e.g. Alatalo 1980; Jansson Potentially, tits may use discoloration to evaluate expected and von Brömssen 1981). nutritional gain and adjust their time expenditure within a Several tit species were observed in the vicinity of the particular bedding site adaptively (e.g. Naef-Daenzer et al. 2000), artificial bedding sites in all five localities (Konnevesi, Pulkkila, because in the heavily infested study areas many of the Rantsila, Utajärvi and Yli-Ii) (paper IV). The willow tit (Poecile discoloured bedding sites were entirely emptied by predators

54 55 already within 8 and 24 h. The avian predators may have learned to associate snow discoloration with a potential food source in areas where overwintering moose and deer ked are 4 Conclusions and future abundant, because snow discoloration did not have a large effect on pupal predation in study locations, where the prospects overwintering moose populations were not heavily infested. The laboratory feeding experiment with the small mammals indicated that all three studied species can prey upon pupae of deer ked (paper V). This was especially true for the bank voles. This thesis provides new scientific information on life-history A relative high proportion of bank voles ate pupae even when characteristics and ecological relationships of invasive insect alternative food was available. This reflects the fact that the ectoparasites in general, and specifically, on the deer ked. Pupal bank vole is a granivore-omnivore species, and its food consists and adult characteristics such as size and cold tolerance of the mainly of seeds, berries and green vegetation, but also insects, deer ked were assessed (I and II). Size of diapausing pupae was earthworms and other invertebrates (von Blanckenhagen et al. smaller in the northern Central Finland than in southern Finland 2007). The highest peak in the bank vole’s food consumption in (I). Further studies are needed to confirm the underlying reason the wild (see e.g. Eccard and Ylönen 2001; von Blanckenhagen et for this observed phenomenon (e.g. adaptive/maladaptive al. 2007) overlaps with the highest occurrence of deer ked pupae plasticity in maternal provisioning to offspring). Small size during spring. Then the principal nutrients of the bank voles can might be a disadvantage at low temperatures because pupal size be scarce and predation pressure on deer ked pupae may be of the deer ked did not enhance the SCP capacity and survival of higher. Indeed, the marginally statistically significant pattern (p the larger pupae was better during prolonged frost period (II). = 0.051) indicated that the bank voles consumed all the pupae The deer ked has high cold tolerance through four seasons provided if there was no access to alternative food. and also in all free-living stages (II). The deer ked follows Feeding experiments on the field voles suggest that field freezing-intolerant strategy, possibly being moderately chill voles consume pupae of deer ked under food limitation but not tolerant. The diapausing pupae were able to supercool to if alternative food is available (paper V). This result was relatively low temperatures (-26°C), without cold acclimation expected because, in the wild, main diet of the field vole consists and in any season, which in turn may increase the survival of the green parts of vegetation, but when the principal probability of newborn pupae when they face a sudden fall in nutrients are scarce, as in late winter, field voles might include temperature during dropping of the host. However, in the animal protein (e.g. insects) in their diet (Hanski and Parviainen northern Boreal environments, temperatures on the snow 1985). According to the results of both vole species, it could be surface may remain below ƺ20 °C for several days, and thus assumed that bank voles and field voles under food limitation, tolerance to long frost periods is also crucial for survival. The i.e. in late winter and early spring when no new vegetative diapausing pupae of deer ked are able to survive harsh frosts, growth has occurred yet, can be notable predators on pupae of ƺ20 °C, lasting for 3 to 4 days on the moose bedding sites. deer ked. Surprisingly, insectivorous and most of the time food- Also the life-history stages that do not experience very low constrained shrews consumed less pupae than the granivorous- temperatures in nature (developing pupae and adults) were still herbivorous voles in the experiment. Shrews preferred the red able to cope with serious frosts (II). In northern environments, worm (Eisenia fetida) over pupae in the experiment. mild nocturnal frosts may occur even during summer, thus the high cold tolerance through all life-history stages of the deer ked

56 57 already within 8 and 24 h. The avian predators may have learned to associate snow discoloration with a potential food source in areas where overwintering moose and deer ked are 4 Conclusions and future abundant, because snow discoloration did not have a large effect on pupal predation in study locations, where the prospects overwintering moose populations were not heavily infested. The laboratory feeding experiment with the small mammals indicated that all three studied species can prey upon pupae of deer ked (paper V). This was especially true for the bank voles. This thesis provides new scientific information on life-history A relative high proportion of bank voles ate pupae even when characteristics and ecological relationships of invasive insect alternative food was available. This reflects the fact that the ectoparasites in general, and specifically, on the deer ked. Pupal bank vole is a granivore-omnivore species, and its food consists and adult characteristics such as size and cold tolerance of the mainly of seeds, berries and green vegetation, but also insects, deer ked were assessed (I and II). Size of diapausing pupae was earthworms and other invertebrates (von Blanckenhagen et al. smaller in the northern Central Finland than in southern Finland 2007). The highest peak in the bank vole’s food consumption in (I). Further studies are needed to confirm the underlying reason the wild (see e.g. Eccard and Ylönen 2001; von Blanckenhagen et for this observed phenomenon (e.g. adaptive/maladaptive al. 2007) overlaps with the highest occurrence of deer ked pupae plasticity in maternal provisioning to offspring). Small size during spring. Then the principal nutrients of the bank voles can might be a disadvantage at low temperatures because pupal size be scarce and predation pressure on deer ked pupae may be of the deer ked did not enhance the SCP capacity and survival of higher. Indeed, the marginally statistically significant pattern (p the larger pupae was better during prolonged frost period (II). = 0.051) indicated that the bank voles consumed all the pupae The deer ked has high cold tolerance through four seasons provided if there was no access to alternative food. and also in all free-living stages (II). The deer ked follows Feeding experiments on the field voles suggest that field freezing-intolerant strategy, possibly being moderately chill voles consume pupae of deer ked under food limitation but not tolerant. The diapausing pupae were able to supercool to if alternative food is available (paper V). This result was relatively low temperatures (-26°C), without cold acclimation expected because, in the wild, main diet of the field vole consists and in any season, which in turn may increase the survival of the green parts of vegetation, but when the principal probability of newborn pupae when they face a sudden fall in nutrients are scarce, as in late winter, field voles might include temperature during dropping of the host. However, in the animal protein (e.g. insects) in their diet (Hanski and Parviainen northern Boreal environments, temperatures on the snow 1985). According to the results of both vole species, it could be surface may remain below ƺ20 °C for several days, and thus assumed that bank voles and field voles under food limitation, tolerance to long frost periods is also crucial for survival. The i.e. in late winter and early spring when no new vegetative diapausing pupae of deer ked are able to survive harsh frosts, growth has occurred yet, can be notable predators on pupae of ƺ20 °C, lasting for 3 to 4 days on the moose bedding sites. deer ked. Surprisingly, insectivorous and most of the time food- Also the life-history stages that do not experience very low constrained shrews consumed less pupae than the granivorous- temperatures in nature (developing pupae and adults) were still herbivorous voles in the experiment. Shrews preferred the red able to cope with serious frosts (II). In northern environments, worm (Eisenia fetida) over pupae in the experiment. mild nocturnal frosts may occur even during summer, thus the high cold tolerance through all life-history stages of the deer ked

56 57 can be of importance. Small adults emerged earlier from smaller Finland. However, other factors such as year-round cold- pupae compared to larger individuals, suggesting lower hardiness may have facilitated the rapid northward invasion of metabolic reserves of small individuals to sustain non-feeding the deer ked. In the future, it is possible that the lower density of pupal stage (I). Presumably, the deer ked can rely on maternally the main host (i.e. moose) together with lower temperatures in derived resources such as body fat composition during long north Finland will prevent or slow the northward expansion of non-feeding life-stages and possibly in the cold-hardening the deer ked, assuming the deer ked does not adapt to utilize the process. semi-domesticated reindeer more effectively as its breeding The host – deer ked interaction causes observable cues (i.e. host. Also, predators may decrease population densities of the snow stained with blood and tissue fluid of host and with deer deer ked locally. One critical aspect for the survival of deer ked ked faeces) on the bedding sites of hosts (III). The experimental pupae at low temperatures and against predators above snow manipulations in the field revealed that predators can exploit cover is likely the timing when the new snow falls, which these cues when searching for a prey (IV). Pupae of deer ked are protects offspring by hiding and insulating them. In future, one likely predated upon by a number of tit species, and snow interesting study question concerns possible differences in cold- discolouration, which indicated high deer ked infestation hardiness between the Finnish and Swedish deer ked intensity, increased the risk of pupal predation on bedding sites. populations and whether differential cold-hardiness could The ability of tits to use this cue seems to be dependent on the explain why the Swedish deer ked population has not expanded deer ked prevalence on the local moose population and the its range farther north (L. Härkönen, S. Kaunisto et al., period of invasion history, which suggests that predation may unpublished data). be a learned behavioural response. The laboratory feeding experiments (paper V) indicated that all three studied small- mammal species (the bank vole, the field vole and the common shrew) prey upon pupae of deer ked. Additional studies are needed to explore ability of voles and shrews to find pupae in their natural ground layer habitats. Furthermore, the impact of one group of potential pupal predators, ground-dwelling invertebrates like ants and carabids, needs to be studied. As the next step, an extensive field study will focus on this question in detail (S. Kaunisto et al., unpublished data). Before firm conclusions on the importance of pupal predation can be drawn, direct cut and faeces analyses of predators, and also evaluation of on-host mortality of the deer ked are needed. Nevertheless, pupal predation should not be ignored when evaluating the two- or three-way interactions among the deer ked, host and the predatory species. The increased density in moose population and its migration routes is thought to be the main reason explaining the rapid range expansion of the deer ked during the previous 40 years in

58 59 can be of importance. Small adults emerged earlier from smaller Finland. However, other factors such as year-round cold- pupae compared to larger individuals, suggesting lower hardiness may have facilitated the rapid northward invasion of metabolic reserves of small individuals to sustain non-feeding the deer ked. In the future, it is possible that the lower density of pupal stage (I). Presumably, the deer ked can rely on maternally the main host (i.e. moose) together with lower temperatures in derived resources such as body fat composition during long north Finland will prevent or slow the northward expansion of non-feeding life-stages and possibly in the cold-hardening the deer ked, assuming the deer ked does not adapt to utilize the process. semi-domesticated reindeer more effectively as its breeding The host – deer ked interaction causes observable cues (i.e. host. Also, predators may decrease population densities of the snow stained with blood and tissue fluid of host and with deer deer ked locally. One critical aspect for the survival of deer ked ked faeces) on the bedding sites of hosts (III). The experimental pupae at low temperatures and against predators above snow manipulations in the field revealed that predators can exploit cover is likely the timing when the new snow falls, which these cues when searching for a prey (IV). Pupae of deer ked are protects offspring by hiding and insulating them. In future, one likely predated upon by a number of tit species, and snow interesting study question concerns possible differences in cold- discolouration, which indicated high deer ked infestation hardiness between the Finnish and Swedish deer ked intensity, increased the risk of pupal predation on bedding sites. populations and whether differential cold-hardiness could The ability of tits to use this cue seems to be dependent on the explain why the Swedish deer ked population has not expanded deer ked prevalence on the local moose population and the its range farther north (L. Härkönen, S. Kaunisto et al., period of invasion history, which suggests that predation may unpublished data). be a learned behavioural response. The laboratory feeding experiments (paper V) indicated that all three studied small- mammal species (the bank vole, the field vole and the common shrew) prey upon pupae of deer ked. Additional studies are needed to explore ability of voles and shrews to find pupae in their natural ground layer habitats. Furthermore, the impact of one group of potential pupal predators, ground-dwelling invertebrates like ants and carabids, needs to be studied. As the next step, an extensive field study will focus on this question in detail (S. Kaunisto et al., unpublished data). Before firm conclusions on the importance of pupal predation can be drawn, direct cut and faeces analyses of predators, and also evaluation of on-host mortality of the deer ked are needed. Nevertheless, pupal predation should not be ignored when evaluating the two- or three-way interactions among the deer ked, host and the predatory species. The increased density in moose population and its migration routes is thought to be the main reason explaining the rapid range expansion of the deer ked during the previous 40 years in

58 59 References

Abebe R, Tatek M, Megersa B & Sheferaw D (2011) Prevalence of small ruminant ectoparasites and associated risk factors in selected districts of Tigray region, Ethiopia. Global Veterinaria 7:433–437. Addo-Bediako A, Chown SL & Gaston KJ (2000) Thermal tolerance, climatic variability and latitude. Proceedings of the Royal Society B 267:739–745. Agrawal AA (2001) Phenotypic plasticity in the interactions and evolution of species. Science 294:321–326. Aitchison CW (1987) Review of winter trophic relations of soricine shrews. Mammal Review 17:1–24. Alatalo RV (1980) Seasonal dynamics of resource partitioning among foliage-gleaning passerines in northern Finland. Oecologia 45:190–196. Alatalo RV (1982) Multidimensional foraging niche organization of foliage-gleaning birds in northern Finland. Ornis Scandinavica 13:56–71. Allemand R & David JR (1976) The circadian rhythm of oviposition in Drosophila melanogaster: a genetic latitudinal cline in wild populations. Experientia 32:1403–1405. Anderson RM & May RM (1978) Regulation and stability of host–parasite population interactions. I. Regulatory processes. Journal of Animal Ecology 47:219–247. Andersson H (1985) De Svenska lusflugorna. Entomologisk Tidskrift 106:15–25. Andreadis SS, Milonas PG & Savopoulou-Soultani M (2005) Cold hardiness of diapausing and non-diapausing pupae of the European grapevine moth, Lobesia botrana. Entomologia Experimentalis et Applicata 117:113–118. Arnett AE & Gotelli NJ (1999) Geographic variation in life- history traits of the ant lion, Myrmeleon immaculatus:

60 61 References

60 61 evolutionary implications of Bergmann’s rule. Evolution instar larvae of Eurosta solidaginis. Journal of Comparative 53:1180–1188. Physiology B 167:249–255. Atkinson D & Sibly RM (1997) Why are organisms usually Bequaert JC (1953) The Hippoboscidae or louse-flies (Diptera) of bigger in colder environments? Making sense of a life history mammals and birds. Part I. Structure, physiology and natural puzzle. Trends in Ecology & Evolution 12:235–239. history. Entomologica Americana 32/33:1–442. Balashov YuS (2005) Ecological niches of ectoparasites. Bergmann C (1847) Über die Verhältnisse der Wärmeökonomie Parazitologiia 39:441–456. [in Russian]. der Thiere zu ihrer Grösse. Göttinger Studien 3:595–708. Balashov YuS (2007) Harmfulness of parasitic insects and Berloco M, Fanti L, Breiling A, Orlando V & Pimpinelli S (2001) acarines to mammals and birds. Entomological Review 87:1300– The maternal effect gene, abnormal oocyte (abo), of 1316. Drosophila melanogaster encodes a specific negative regulator Bale JS (1993) Classes of insect cold hardiness. Functional Ecology of histones. Proceedings of the National Academy of Sciences USA 7:751–753. 98:12126–12131. Bale JS (1996) Insect cold hardiness: a matter of life and death. Bermúdez S & Miranda R (2011) Distribution of ectoparasites of European Journal of Entomology 93:369–382. Canis lupus familiaris L. (Carnivora: Canidae) from Panama. Bale JS & Hayward SAL (2010) Insect overwintering in a Revista MVZ Córdoba 16:2274–2282. changing climate. Journal of Experimental Biology 213:980–994. Blackburn TM, Gaston KJ & Loder N (1999) Geographic Bale JS, Harrington R & Clough MS (1988) Low temperature gradients in body size: a clarification of Bergmann’s rule. mortality of the peach-potato aphid Myzus persicae. Ecological Diversity and Distributions 5:165–174. Entomology 13:121–129. Blanckenhorn WU (2000) The evolution of body size: what Bale JS, Masters GJ, Hodkinson ID, Awmack C, Bezemer TM, keeps organisms small? Quarterly Review of Biology 75:385– Brown VK, Butterfield J, Buse A, Coulson JC, Farrar J, Good 407. JEG, Harrington R, Hartley S, Jones TH, Lindroth RL, Press Blanckenhorn WU & Demont M (2004) Bergmann and Converse MC, Symrnioudis I, Watt AD & Whittaker JB (2002) Bergmann latitudinal clines in arthropods: two ends of a Herbivory in global climate change research: direct effects of continuum? Integrative and Comparative Biology 44:413–424. rising temperature on insect herbivores. Global Change Biology Block W (1990) Cold tolerance of insects and other arthropods. 8:1–16. Philosophical Transactions of the Royal Society B: Biological Barbaro L & Battisti A (2011) Birds as predators of the pine Sciences 326:613–633. processionary moth (Lepidoptera: Notodontidae). Biological Bouchard RW, Carrillo MA, Kells SA & Ferrington LC (2006) Control 56:107–114. Freeze tolerance in larvae of the winter-active Diamesa Belozerov VN & Naumov RL (2002) Nymphal diapause and its mendotae Muttkowski (Diptera: Chironomidae): a contrast to photoperiodic control in the tick Ixodes scapularis (Acari: adult strategy for survival at low temperatures. Hydrobiologia Ixodidae). Folia Parasitologica 49:314–318. 568:403–416. Bennett ATD & Cuthill IC (1994) Ultraviolet vision in birds: Burks CS & Hagstrum DV (1999) Rapid cold hardening capacity what is its function? Vision Research 34:1471–1478. in five species of coleopteran pests of stored grain. Journal of Bennett VA, Pruitt NL & Lee RE (1997) Seasonal changes in fatty Stored Products Research 35:65–75. acid composition associated with cold-hardening in third Burks CS, Stewart RL, Needham GR & Lee RE (1996a) Cold hardiness in the ixodid ticks (Ixodidae). In Mitchell R, Horn

62 63 evolutionary implications of Bergmann’s rule. Evolution instar larvae of Eurosta solidaginis. Journal of Comparative 53:1180–1188. Physiology B 167:249–255. Atkinson D & Sibly RM (1997) Why are organisms usually Bequaert JC (1953) The Hippoboscidae or louse-flies (Diptera) of bigger in colder environments? Making sense of a life history mammals and birds. Part I. Structure, physiology and natural puzzle. Trends in Ecology & Evolution 12:235–239. history. Entomologica Americana 32/33:1–442. Balashov YuS (2005) Ecological niches of ectoparasites. Bergmann C (1847) Über die Verhältnisse der Wärmeökonomie Parazitologiia 39:441–456. [in Russian]. der Thiere zu ihrer Grösse. Göttinger Studien 3:595–708. Balashov YuS (2007) Harmfulness of parasitic insects and Berloco M, Fanti L, Breiling A, Orlando V & Pimpinelli S (2001) acarines to mammals and birds. Entomological Review 87:1300– The maternal effect gene, abnormal oocyte (abo), of 1316. Drosophila melanogaster encodes a specific negative regulator Bale JS (1993) Classes of insect cold hardiness. Functional Ecology of histones. Proceedings of the National Academy of Sciences USA 7:751–753. 98:12126–12131. Bale JS (1996) Insect cold hardiness: a matter of life and death. Bermúdez S & Miranda R (2011) Distribution of ectoparasites of European Journal of Entomology 93:369–382. Canis lupus familiaris L. (Carnivora: Canidae) from Panama. Bale JS & Hayward SAL (2010) Insect overwintering in a Revista MVZ Córdoba 16:2274–2282. changing climate. Journal of Experimental Biology 213:980–994. Blackburn TM, Gaston KJ & Loder N (1999) Geographic Bale JS, Harrington R & Clough MS (1988) Low temperature gradients in body size: a clarification of Bergmann’s rule. mortality of the peach-potato aphid Myzus persicae. Ecological Diversity and Distributions 5:165–174. Entomology 13:121–129. Blanckenhorn WU (2000) The evolution of body size: what Bale JS, Masters GJ, Hodkinson ID, Awmack C, Bezemer TM, keeps organisms small? Quarterly Review of Biology 75:385– Brown VK, Butterfield J, Buse A, Coulson JC, Farrar J, Good 407. JEG, Harrington R, Hartley S, Jones TH, Lindroth RL, Press Blanckenhorn WU & Demont M (2004) Bergmann and Converse MC, Symrnioudis I, Watt AD & Whittaker JB (2002) Bergmann latitudinal clines in arthropods: two ends of a Herbivory in global climate change research: direct effects of continuum? Integrative and Comparative Biology 44:413–424. rising temperature on insect herbivores. Global Change Biology Block W (1990) Cold tolerance of insects and other arthropods. 8:1–16. Philosophical Transactions of the Royal Society B: Biological Barbaro L & Battisti A (2011) Birds as predators of the pine Sciences 326:613–633. processionary moth (Lepidoptera: Notodontidae). Biological Bouchard RW, Carrillo MA, Kells SA & Ferrington LC (2006) Control 56:107–114. Freeze tolerance in larvae of the winter-active Diamesa Belozerov VN & Naumov RL (2002) Nymphal diapause and its mendotae Muttkowski (Diptera: Chironomidae): a contrast to photoperiodic control in the tick Ixodes scapularis (Acari: adult strategy for survival at low temperatures. Hydrobiologia Ixodidae). Folia Parasitologica 49:314–318. 568:403–416. Bennett ATD & Cuthill IC (1994) Ultraviolet vision in birds: Burks CS & Hagstrum DV (1999) Rapid cold hardening capacity what is its function? Vision Research 34:1471–1478. in five species of coleopteran pests of stored grain. Journal of Bennett VA, Pruitt NL & Lee RE (1997) Seasonal changes in fatty Stored Products Research 35:65–75. acid composition associated with cold-hardening in third Burks CS, Stewart RL, Needham GR & Lee RE (1996a) Cold hardiness in the ixodid ticks (Ixodidae). In Mitchell R, Horn

62 63 DJ, Needham GR & Welbourn WC (Eds) Acarology IX: Vol. 1, Proceedings of the Royal Society B: Biological Sciences 265:1509– Proceedings. Ohio Biological Survey, Ohio, pp. 85–87. 1514. Burks CS, Stewart RL, Needham GR & Lee RE (1996b) The role Churchfield S (2002) Why are shrews so small? The costs and of direct chilling injury and inoculative freezing in cold benefits of small size in northern temperate Sorex species in tolerance of Amblyomma americanum, Dermacentor variabilis the context of foraging habits and prey supply. Acta and Ixodes scapularis. Physiological Entomology 21:44–50. Theriologica 47:169–184. Bush AO, Fernández JC, Esch GW & Seed JR (2001) Parasitism: Churchfield S, Hollier J & Brown VK (1991) The effects of small the diversity and ecology of animal parasites. Cambridge mammal predators on grassland invertebrates, investigated University Press, Cambridge. by field exclosure experiment. Oikos 60:283–290. Calosi P, Bilton DT, Spicer JI & Atfield A (2008) Thermal Clayton DH & Moore J (Eds) (1997) Host-parasite evolution: tolerance and geographical range size in the Agabus brunneus general principles and avian models. Oxford University Press, group of European diving beetles (Coleoptera: Dytiscidae). New York. Journal of Biogeography 35:295–305. Colinet H, Hance T & Vernon P (2006) Water relations, fat Carlsson NOL, Sarnelle O & Strayer DL (2009) Native predators reserves, survival, and longevity of a cold-exposed parasitic and exotic prey – an acquired taste? Frontiers in Ecology and wasp Aphidius colemani (Hymenoptera: Aphidiinae). the Environment 7:525–532. Environmental Entomology 35:228–236. Carlton JT (2002) Bioinvasion ecology: assessing invasion impact Combes C (2001) Parasitism: the ecology and evolution of intimate and scale. In Leppäkoski E, Gollasch S & Olenin S (Eds) interactions. The University of Chicago Press Ltd., London. Invasive Aquatic Species of Europe: Distribution, Impacts and Conover DO & Schultz ET (1995) Phenotypic similarity and the Management. Kluwer Academic Publishers, Dordrecht, pp. 7– evolutionary significance of counter gradient variation. 19. Trends in Ecology & Evolution 10:248–252. Carlton JT & Ruiz GM (2005) Vector science and integrated Cox GW (1999) Alien species in North America and Hawaii: impacts vector management in bioinvasion ecology: conceptual on natural ecosystems. Island Press, Washington DC. frameworks. In Mooney HA, Mack RN, McNeely JA, Neville Cumming GS & Van Vuuren DP (2006) Will climate change LE, Schei PJ & Waage JK (Eds) Invasive Alien Species: A New affect ectoparasite species ranges? Global Ecology and Synthesis. Island Press, Washington DC, pp. 36–58. Biogeography 15:486–497. Chen C, Denlinger DL & Lee RE (1987) Cold shock injury and Cushman JH, Lawton JH & Manly BFJ (1993) Latitudinal rapid cold hardening in the flesh fly, Sarcophaga crassipalpis. patterns in European ant assemblages: variation in species Physiological Zoology 60:297–304. richness and body size. Oecologia 95:30–37. Chown SL & Nicolson SW (2009) Insect physiological ecology: Czajka MC & Lee RE (1990) A rapid cold-hardening response mechanisms and patterns. Oxford University Press Inc., New protecting against cold shock injury in Drosophila York. melanogaster. Journal of Experimental Biology 148:245–254. Chown SL & Gaston KJ (2010) Body size variation in insects: a Danks HV (1996) The wider integration of studies on insect macroecological perspective. Biological Reviews 85:139–169. cold-hardiness. European Journal of Entomology 93:383–403. Church SC, Bennett ATD, Cuthill IC & Partridge JC (1998) Darwin C (1859) On the origin of species. Murray, London. Ultraviolet cues affect the foraging behaviour of blue tits. Dautel H & Knülle W (1997) Cold-hardiness, supercooling ability and causes of low-temperature mortality in the soft

64 65 DJ, Needham GR & Welbourn WC (Eds) Acarology IX: Vol. 1, Proceedings of the Royal Society B: Biological Sciences 265:1509– Proceedings. Ohio Biological Survey, Ohio, pp. 85–87. 1514. Burks CS, Stewart RL, Needham GR & Lee RE (1996b) The role Churchfield S (2002) Why are shrews so small? The costs and of direct chilling injury and inoculative freezing in cold benefits of small size in northern temperate Sorex species in tolerance of Amblyomma americanum, Dermacentor variabilis the context of foraging habits and prey supply. Acta and Ixodes scapularis. Physiological Entomology 21:44–50. Theriologica 47:169–184. Bush AO, Fernández JC, Esch GW & Seed JR (2001) Parasitism: Churchfield S, Hollier J & Brown VK (1991) The effects of small the diversity and ecology of animal parasites. Cambridge mammal predators on grassland invertebrates, investigated University Press, Cambridge. by field exclosure experiment. Oikos 60:283–290. Calosi P, Bilton DT, Spicer JI & Atfield A (2008) Thermal Clayton DH & Moore J (Eds) (1997) Host-parasite evolution: tolerance and geographical range size in the Agabus brunneus general principles and avian models. Oxford University Press, group of European diving beetles (Coleoptera: Dytiscidae). New York. Journal of Biogeography 35:295–305. Colinet H, Hance T & Vernon P (2006) Water relations, fat Carlsson NOL, Sarnelle O & Strayer DL (2009) Native predators reserves, survival, and longevity of a cold-exposed parasitic and exotic prey – an acquired taste? Frontiers in Ecology and wasp Aphidius colemani (Hymenoptera: Aphidiinae). the Environment 7:525–532. Environmental Entomology 35:228–236. Carlton JT (2002) Bioinvasion ecology: assessing invasion impact Combes C (2001) Parasitism: the ecology and evolution of intimate and scale. In Leppäkoski E, Gollasch S & Olenin S (Eds) interactions. The University of Chicago Press Ltd., London. Invasive Aquatic Species of Europe: Distribution, Impacts and Conover DO & Schultz ET (1995) Phenotypic similarity and the Management. Kluwer Academic Publishers, Dordrecht, pp. 7– evolutionary significance of counter gradient variation. 19. Trends in Ecology & Evolution 10:248–252. Carlton JT & Ruiz GM (2005) Vector science and integrated Cox GW (1999) Alien species in North America and Hawaii: impacts vector management in bioinvasion ecology: conceptual on natural ecosystems. Island Press, Washington DC. frameworks. In Mooney HA, Mack RN, McNeely JA, Neville Cumming GS & Van Vuuren DP (2006) Will climate change LE, Schei PJ & Waage JK (Eds) Invasive Alien Species: A New affect ectoparasite species ranges? Global Ecology and Synthesis. Island Press, Washington DC, pp. 36–58. Biogeography 15:486–497. Chen C, Denlinger DL & Lee RE (1987) Cold shock injury and Cushman JH, Lawton JH & Manly BFJ (1993) Latitudinal rapid cold hardening in the flesh fly, Sarcophaga crassipalpis. patterns in European ant assemblages: variation in species Physiological Zoology 60:297–304. richness and body size. Oecologia 95:30–37. Chown SL & Nicolson SW (2009) Insect physiological ecology: Czajka MC & Lee RE (1990) A rapid cold-hardening response mechanisms and patterns. Oxford University Press Inc., New protecting against cold shock injury in Drosophila York. melanogaster. Journal of Experimental Biology 148:245–254. Chown SL & Gaston KJ (2010) Body size variation in insects: a Danks HV (1996) The wider integration of studies on insect macroecological perspective. Biological Reviews 85:139–169. cold-hardiness. European Journal of Entomology 93:383–403. Church SC, Bennett ATD, Cuthill IC & Partridge JC (1998) Darwin C (1859) On the origin of species. Murray, London. Ultraviolet cues affect the foraging behaviour of blue tits. Dautel H & Knülle W (1997) Cold-hardiness, supercooling ability and causes of low-temperature mortality in the soft

64 65 tick, Argas reflexus, and the hard tick, Ixodes ricinus (Acari: Felsõ B & Rózsa L (2006) Reduced taxonomic richness of lice Ixodoidea) from Central Europe. Journal of Insect Physiology (Insecta: Phthiraptera) in diving birds. Journal of Parasitology 43:843–854. 92:867–869. Davis MA (2009) Invasion biology. Oxford University Press Inc., Fischer K, Brakefield PM & Zwaan BJ (2003) Plasticity in New York. butterfly egg size: why larger offspring at lower Davis MA, Grime JP & Thompson K (2000) Fluctuating temperatures? Ecology 84:3138–3147. resources in plant communities: a general theory of Fox CW, Waddell KJ & Mousseau TA (1995) Parental host plant invasibility. Journal of Ecology 88:528–534. affects offspring life histories in a seed beetle. Ecology 76:402– Dehio C, Sauder U & Hiestand R (2004) Isolation of Bartonella 411. schoenbuchensis from Lipoptena cervi, a blood-sucking Frank JH (1967) The effect of pupal predators on a population of arthropod causing deer ked dermatitis. Journal of Clinical winter moth, Operophtera brumata (L.) (Hydriomenidae). Microbiology 42:5320–5323. Journal of Animal Ecology 36:611–621. Demont M & Blanckenhorn WU (2008) Genetic differentiation in Fry CH & Keith S (Eds) (2000). The birds of Africa, Vol. VI. diapause response along a latitudinal cline in European Academic Press, London. yellow dung fly populations. Ecological Entomology 33:197– Gaston KJ (2003) The structure and dynamics of geographic ranges. 201. Oxford University Press, New York. Denlinger DL (1991) Relationship between cold hardiness and Ghalambor CK, McKay JK, Carroll SP & Reznick DN (2007) diapause. In Lee RE & Denlinger DL (Eds) Insects at Low Adaptive versus non-adaptive phenotypic plasticity and the Temperature. Chapman and Hall, New York, pp. 174–198. potential for contemporary adaptation in new environments. Denlinger DL & Lee RE (Eds) (2010) Low temperature biology of Functinal Ecology 21:394–407. insects. Cambridge University Press, Cambridge. Gilchrist GW, Huey RB & Serra L (2001) Rapid evolution of DeWitt TJ & Scheiner SM (Eds) (2004) Phenotypic plasticity: wing size clines in Drosophila subobscura. Genetica functional and conceptual approaches. Oxford University Press 112/113:273–286. Inc., New York. Goto M, Li Y, Kayaba S, Outani S & Suzuki K (2001) Cold Dörr B & Gothe R (2001) Cold-hardiness of Dermacentor hardiness in summer and winter diapause and post-diapause marginatus (Acari: Ixodidae). Experimental and Applied pupae of the cabbage armyworm, Mamestra brassicae L. under Acarology 25:151–169. temperature acclimation. Journal of Insect Physiology 47:709– Eccard JA & Ylönen H (2001) Initiation of breeding after winter 714. in bank voles: effects of food and population density. Greenstone MH (1999) Spider predation: how and why we Canadian Journal of Zoology 79:1743–1753. study it. Journal of Arachnology 27:333–342. Elton CS (1958) The ecology of invasions by animals and plants. Grimaldi D & Engel MS (2005) Evolution of the insects. Methuen, London. Cambridge University Press, New York. Fargione JE & Tilman D (2005) Diversity decreases invasion via Haarløv N (1964) Life cycle and distribution pattern of Lipoptena both sampling and complementarity effects. Ecology Letters cervi (L.) (Dipt., Hippobosc.) on Danish deer. Oikos 15:93–129. 8:604–611. Haas GE, Wilson N, Kucera JR, Osborne TO, Whitman JS & Johnson WN (2010) Range expansion and hosts of Ctenophthalmus pseudagyrtes Baker (Siphonaptera:

66 67 tick, Argas reflexus, and the hard tick, Ixodes ricinus (Acari: Felsõ B & Rózsa L (2006) Reduced taxonomic richness of lice Ixodoidea) from Central Europe. Journal of Insect Physiology (Insecta: Phthiraptera) in diving birds. Journal of Parasitology 43:843–854. 92:867–869. Davis MA (2009) Invasion biology. Oxford University Press Inc., Fischer K, Brakefield PM & Zwaan BJ (2003) Plasticity in New York. butterfly egg size: why larger offspring at lower Davis MA, Grime JP & Thompson K (2000) Fluctuating temperatures? Ecology 84:3138–3147. resources in plant communities: a general theory of Fox CW, Waddell KJ & Mousseau TA (1995) Parental host plant invasibility. Journal of Ecology 88:528–534. affects offspring life histories in a seed beetle. Ecology 76:402– Dehio C, Sauder U & Hiestand R (2004) Isolation of Bartonella 411. schoenbuchensis from Lipoptena cervi, a blood-sucking Frank JH (1967) The effect of pupal predators on a population of arthropod causing deer ked dermatitis. Journal of Clinical winter moth, Operophtera brumata (L.) (Hydriomenidae). Microbiology 42:5320–5323. Journal of Animal Ecology 36:611–621. Demont M & Blanckenhorn WU (2008) Genetic differentiation in Fry CH & Keith S (Eds) (2000). The birds of Africa, Vol. VI. diapause response along a latitudinal cline in European Academic Press, London. yellow dung fly populations. Ecological Entomology 33:197– Gaston KJ (2003) The structure and dynamics of geographic ranges. 201. Oxford University Press, New York. Denlinger DL (1991) Relationship between cold hardiness and Ghalambor CK, McKay JK, Carroll SP & Reznick DN (2007) diapause. In Lee RE & Denlinger DL (Eds) Insects at Low Adaptive versus non-adaptive phenotypic plasticity and the Temperature. Chapman and Hall, New York, pp. 174–198. potential for contemporary adaptation in new environments. Denlinger DL & Lee RE (Eds) (2010) Low temperature biology of Functinal Ecology 21:394–407. insects. Cambridge University Press, Cambridge. Gilchrist GW, Huey RB & Serra L (2001) Rapid evolution of DeWitt TJ & Scheiner SM (Eds) (2004) Phenotypic plasticity: wing size clines in Drosophila subobscura. Genetica functional and conceptual approaches. Oxford University Press 112/113:273–286. Inc., New York. Goto M, Li Y, Kayaba S, Outani S & Suzuki K (2001) Cold Dörr B & Gothe R (2001) Cold-hardiness of Dermacentor hardiness in summer and winter diapause and post-diapause marginatus (Acari: Ixodidae). Experimental and Applied pupae of the cabbage armyworm, Mamestra brassicae L. under Acarology 25:151–169. temperature acclimation. Journal of Insect Physiology 47:709– Eccard JA & Ylönen H (2001) Initiation of breeding after winter 714. in bank voles: effects of food and population density. Greenstone MH (1999) Spider predation: how and why we Canadian Journal of Zoology 79:1743–1753. study it. Journal of Arachnology 27:333–342. Elton CS (1958) The ecology of invasions by animals and plants. Grimaldi D & Engel MS (2005) Evolution of the insects. Methuen, London. Cambridge University Press, New York. Fargione JE & Tilman D (2005) Diversity decreases invasion via Haarløv N (1964) Life cycle and distribution pattern of Lipoptena both sampling and complementarity effects. Ecology Letters cervi (L.) (Dipt., Hippobosc.) on Danish deer. Oikos 15:93–129. 8:604–611. Haas GE, Wilson N, Kucera JR, Osborne TO, Whitman JS & Johnson WN (2010) Range expansion and hosts of Ctenophthalmus pseudagyrtes Baker (Siphonaptera:

66 67 Ctenophthalmidae) in central Alaska. Journal of the Hough-Goldstein JA, Geiger J, Chang D & Saylor W (1993a.) Entomological Society of British Columbia 107:87–88. Palatability and toxicity of the Colorado potato beetle Hackman W, Rantanen T & Vuojolahti P (1983) Immigration of (Coleoptera: Chrysomelidae) to domestic chickens. Annals of Lipoptena cervi (Diptera, Hippoboscidae) in Finland, with the Entomological Society of America 86:158–164. notes on its biology and medical significance. Notulae Hough-Goldstein JA, Heimpel GE, Bechman HE & Mason CE Entomologicae 63:53–59. (1993b) Arthropod natural enemies of the Colorado potato Hahn DA & Denlinger DL (2007) Meeting the energetic beetle. Crop Protection 12:324–334. demands of insect diapause: nutrient storage and utilization. Huey RB, Berrigan D, Gilchrist GW & Herron JC (1999) Testing Journal of Insect Physiology 53:760–773. the adaptive significance of acclimation: a strong inference Hahn DA, Martin AR & Porter SD (2008) Body size, but not approach. American Zoologist 39:323–336. cooling rate, affects supercooling points in the red imported Härkönen L (2012) Seasonal variation in the life histories of a fire ant, Solenopsis invicta. Environmental Entomology 37:1074– viviparous ectoparasite, the deer ked. PhD Thesis, Acta 1080. Universitatis Ouluensis, Oulu. Hanski I & Parviainen P (1985) Cocoon predation by small Härkönen L, Härkönen S, Kaitala A, Kaunisto S, Kortet R, mammals, and pine sawfly population dynamics. Oikos Laaksonen S & Ylönen H (2010) Predicting range expansion 45:125–136. of an ectoparasite – the effect of spring and summer Hansson L & Henttonen H (1985) Gradients in density temperatures on deer ked Lipoptena cervi (Diptera: variations of small rodents: the importance of latitude and Hippoboscidae) performance along a latitudinal gradient. snow cover. Oecologia 67:394–402. Ecography 33:906–912. Harris WG & Burns EC (1972) Predation on the lone star tick by Ismail M, Vernon P, Hance T, Pierre JS & van Baaren J (2012) the imported fire ant. Environmental Entomology 1:362–365. What are the possible benefits of small size for energy- Hastings FL, Hain FP, Smith HR, Cook SP & Monahan JF (2002) constrained ectotherms in cold stress conditions? Oikos (in Predation of gypsy moth (Lepidoptera: Lymantriidae) pupae press). in three ecosystems along the southern edge of infestation. Jansson C & von Brömssen A (1981) Winter decline of spiders Environmental Entomology 31:668–675. and insects in spruce Picea abies and its relation to predation Heath ACG (1994) Ectoparasites of livestock in New Zealand. by birds. Ecography (Holarctic Ecology) 4:82–93. New Zealand Journal of Zoology 21:23–38. Johnson PTJ, Dobson A, Lafferty KD, Marcogliese DJ, Memmott Heikkinen S (2000) Hirven vuosi. Suomen Riista 46:82–91. [in J, Orlofske SA, Poulin R & Thieltges DW (2010) When Finnish]. parasites become prey: ecological and epidemiological Heinrich B & Collins SL (1983) Caterpillar leaf damage, and the significance of eating parasites. Trends in Ecology & Evolution game of hide-and-seek with birds. Ecology 64:592–602. 25:362–371. Holling CS (1959) The components of predation as revealed by a Johnson RN & Starks PT (2004) A surprising level of genetic study of small mammal predation of the European pine diversity in an invasive wasp: Polistes dominulus in the sawfly. Canadian Entomologist 91:293–320. northeastern United States. Annals of the Entomological Society Hopla CE, Durden LA & Keirans JE (1994) Ectoparasites and of America 97:732–737. classification. Revue Scientifique et Technique (International Office of Epizootics) 13:985–1017.

68 69 Ctenophthalmidae) in central Alaska. Journal of the Hough-Goldstein JA, Geiger J, Chang D & Saylor W (1993a.) Entomological Society of British Columbia 107:87–88. Palatability and toxicity of the Colorado potato beetle Hackman W, Rantanen T & Vuojolahti P (1983) Immigration of (Coleoptera: Chrysomelidae) to domestic chickens. Annals of Lipoptena cervi (Diptera, Hippoboscidae) in Finland, with the Entomological Society of America 86:158–164. notes on its biology and medical significance. Notulae Hough-Goldstein JA, Heimpel GE, Bechman HE & Mason CE Entomologicae 63:53–59. (1993b) Arthropod natural enemies of the Colorado potato Hahn DA & Denlinger DL (2007) Meeting the energetic beetle. Crop Protection 12:324–334. demands of insect diapause: nutrient storage and utilization. Huey RB, Berrigan D, Gilchrist GW & Herron JC (1999) Testing Journal of Insect Physiology 53:760–773. the adaptive significance of acclimation: a strong inference Hahn DA, Martin AR & Porter SD (2008) Body size, but not approach. American Zoologist 39:323–336. cooling rate, affects supercooling points in the red imported Härkönen L (2012) Seasonal variation in the life histories of a fire ant, Solenopsis invicta. Environmental Entomology 37:1074– viviparous ectoparasite, the deer ked. PhD Thesis, Acta 1080. Universitatis Ouluensis, Oulu. Hanski I & Parviainen P (1985) Cocoon predation by small Härkönen L, Härkönen S, Kaitala A, Kaunisto S, Kortet R, mammals, and pine sawfly population dynamics. Oikos Laaksonen S & Ylönen H (2010) Predicting range expansion 45:125–136. of an ectoparasite – the effect of spring and summer Hansson L & Henttonen H (1985) Gradients in density temperatures on deer ked Lipoptena cervi (Diptera: variations of small rodents: the importance of latitude and Hippoboscidae) performance along a latitudinal gradient. snow cover. Oecologia 67:394–402. Ecography 33:906–912. Harris WG & Burns EC (1972) Predation on the lone star tick by Ismail M, Vernon P, Hance T, Pierre JS & van Baaren J (2012) the imported fire ant. Environmental Entomology 1:362–365. What are the possible benefits of small size for energy- Hastings FL, Hain FP, Smith HR, Cook SP & Monahan JF (2002) constrained ectotherms in cold stress conditions? Oikos (in Predation of gypsy moth (Lepidoptera: Lymantriidae) pupae press). in three ecosystems along the southern edge of infestation. Jansson C & von Brömssen A (1981) Winter decline of spiders Environmental Entomology 31:668–675. and insects in spruce Picea abies and its relation to predation Heath ACG (1994) Ectoparasites of livestock in New Zealand. by birds. Ecography (Holarctic Ecology) 4:82–93. New Zealand Journal of Zoology 21:23–38. Johnson PTJ, Dobson A, Lafferty KD, Marcogliese DJ, Memmott Heikkinen S (2000) Hirven vuosi. Suomen Riista 46:82–91. [in J, Orlofske SA, Poulin R & Thieltges DW (2010) When Finnish]. parasites become prey: ecological and epidemiological Heinrich B & Collins SL (1983) Caterpillar leaf damage, and the significance of eating parasites. Trends in Ecology & Evolution game of hide-and-seek with birds. Ecology 64:592–602. 25:362–371. Holling CS (1959) The components of predation as revealed by a Johnson RN & Starks PT (2004) A surprising level of genetic study of small mammal predation of the European pine diversity in an invasive wasp: Polistes dominulus in the sawfly. Canadian Entomologist 91:293–320. northeastern United States. Annals of the Entomological Society Hopla CE, Durden LA & Keirans JE (1994) Ectoparasites and of America 97:732–737. classification. Revue Scientifique et Technique (International Office of Epizootics) 13:985–1017.

68 69 Johnston SL & Lee RE (1990) Regulation of supercooling and profiles of the larvae of the fruit fly Drosophila melanogaster. nucleation in a freeze intolerant beetle (Tenebrio molitor). Public Library of Science ONE 6(9):e25025. Cryobiology 27:562–568. Kotiaho JS & Sulkava P (2007) Effects of isolation, area and Karan D, Dubey S, Moreteau B, Parkash R & David JR (2000) predators on invasion: a field experiment with artificial Geographical clines for quantitative traits in natural islands. Applied Soil Ecology 35:256–259. populations of a tropical drosophilid: Zaprionus indianus. Koveos DS (2001) Rapid cold hardening in the olive fruit fly Genetica 108:91–100. Bactrocera oleae under laboratory and field conditions. Kawecki TJ & Ebert D (2004) Conceptual issues in local Entomologia Experimentalis et Applicata 101:257–263. adaptation. Ecology Letters 7:1225–1241. Krasnov BR, Mouillot D, Shenbrot GI, Khokhlova IS & Poulin R Khani A & Moharramipour S (2010) Cold hardiness and (2004) Geographical variation in host specificity of fleas supercooling capacity in the overwintering larvae of the (Siphonaptera) parasitic on small mammals: the influence of codling moth, Cydia pomonella. Journal of Insect Science 10:83. phylogeny and local environmental conditions. Ecography Kindsvater HK, Alonzo SH, Mangel M & Bonsall MB (2010) 27:787–797. Effects of age- and state-dependent allocation on offspring Krasnov BR, Khokhlova IS, Arakelyan MS & Degen AA (2005a) size and number. Evolutionary Ecology Research 12:327–346. Is a starving host tastier? Reproduction in fleas parasitizing King RA, Read DS, Traugott M & Symondson WOC (2008) food-limited rodents. Functional Ecology 19:625–631. Molecular analysis of predation: a review of best practice for Krasnov BR, Poulin R, Shenbrot GI, Mouillot D & Khokhlova IS DNA-based approaches. Molecular Ecology 17:947–963. (2005b) Host specificity and geographic range in Kingsolver JG, Massie KR, Ragland GJ & Smith MH (2007) haematophagous ectoparasites. Oikos 108:449–456. Rapid population divergence in thermal reaction norms for Krasnov BR, Mouillot D, Shenbrot GI, Khokhlova IS, Vinarski an invading species: breaking the temperature–size rule. MV, Korallo-Vinarskaya NP & Poulin R (2010) Similarity in Journal of Evolutionary Biology 20:892–900. ectoparasite faunas of Palaearctic rodents as a function of Kolar CS & Lodge DM (2001) Progress in invasion biology: host phylogenetic, geographic or environmental distances: predicting invaders. Trends in Ecology & Evolution 16:199–204. which matters the most? International Journal for Parasitology Kortet R, Härkönen L, Hokkanen P, Härkönen S, Kaitala A, 40:807–817. Kaunisto S, Laaksonen S, Kekäläinen J & Ylönen H (2010) Kukal O & Duman JG (1989) Switch in the overwintering Experiments on the ectoparasitic deer ked that often attacks strategy of two insect species and latitudinal differences in humans: preferences for body parts, colour and temperature. cold hardiness. Canadian Journal of Zoology 67:825–827. Bulletin of Entomological Research 100:279–285. Kynkäänniemi SM, Kortet R, Härkönen L, Kaitala A, Paakkonen Koss AM & Snyder WE (2005) Alternative prey disrupts T, Mustonen AM, Nieminen P, Härkönen S, Ylönen H & biocontrol by a guild of generalist predators. Biological Control Laaksonen S (2010) Threat of an invasive parasitic fly, the 32:243–251. deer ked (Lipoptena cervi), to the reindeer (Rangifer tarandus Koštál V (2006) Eco-physiological phases of insect diapause. tarandus): experimental infection and treatment. Annales Journal of Insect Physiology 52:113–127. Zoologici Fennici 47:28–36. Koštál V, Korbelová J, Rozsypal J, Zahradní²ková H, Cimlová J, Lafferty KD, Dobson AP & Kuris AM (2006) Parasites dominate Tom²ala A & Šimek P (2011) Long-term cold acclimation food web links. Proceedings of the National Academy of Sciences extends survival time at 0°C and modifies the metabolomic USA 103:11211–11216.

70 71 Johnston SL & Lee RE (1990) Regulation of supercooling and profiles of the larvae of the fruit fly Drosophila melanogaster. nucleation in a freeze intolerant beetle (Tenebrio molitor). Public Library of Science ONE 6(9):e25025. Cryobiology 27:562–568. Kotiaho JS & Sulkava P (2007) Effects of isolation, area and Karan D, Dubey S, Moreteau B, Parkash R & David JR (2000) predators on invasion: a field experiment with artificial Geographical clines for quantitative traits in natural islands. Applied Soil Ecology 35:256–259. populations of a tropical drosophilid: Zaprionus indianus. Koveos DS (2001) Rapid cold hardening in the olive fruit fly Genetica 108:91–100. Bactrocera oleae under laboratory and field conditions. Kawecki TJ & Ebert D (2004) Conceptual issues in local Entomologia Experimentalis et Applicata 101:257–263. adaptation. Ecology Letters 7:1225–1241. Krasnov BR, Mouillot D, Shenbrot GI, Khokhlova IS & Poulin R Khani A & Moharramipour S (2010) Cold hardiness and (2004) Geographical variation in host specificity of fleas supercooling capacity in the overwintering larvae of the (Siphonaptera) parasitic on small mammals: the influence of codling moth, Cydia pomonella. Journal of Insect Science 10:83. phylogeny and local environmental conditions. Ecography Kindsvater HK, Alonzo SH, Mangel M & Bonsall MB (2010) 27:787–797. Effects of age- and state-dependent allocation on offspring Krasnov BR, Khokhlova IS, Arakelyan MS & Degen AA (2005a) size and number. Evolutionary Ecology Research 12:327–346. Is a starving host tastier? Reproduction in fleas parasitizing King RA, Read DS, Traugott M & Symondson WOC (2008) food-limited rodents. Functional Ecology 19:625–631. Molecular analysis of predation: a review of best practice for Krasnov BR, Poulin R, Shenbrot GI, Mouillot D & Khokhlova IS DNA-based approaches. Molecular Ecology 17:947–963. (2005b) Host specificity and geographic range in Kingsolver JG, Massie KR, Ragland GJ & Smith MH (2007) haematophagous ectoparasites. Oikos 108:449–456. Rapid population divergence in thermal reaction norms for Krasnov BR, Mouillot D, Shenbrot GI, Khokhlova IS, Vinarski an invading species: breaking the temperature–size rule. MV, Korallo-Vinarskaya NP & Poulin R (2010) Similarity in Journal of Evolutionary Biology 20:892–900. ectoparasite faunas of Palaearctic rodents as a function of Kolar CS & Lodge DM (2001) Progress in invasion biology: host phylogenetic, geographic or environmental distances: predicting invaders. Trends in Ecology & Evolution 16:199–204. which matters the most? International Journal for Parasitology Kortet R, Härkönen L, Hokkanen P, Härkönen S, Kaitala A, 40:807–817. Kaunisto S, Laaksonen S, Kekäläinen J & Ylönen H (2010) Kukal O & Duman JG (1989) Switch in the overwintering Experiments on the ectoparasitic deer ked that often attacks strategy of two insect species and latitudinal differences in humans: preferences for body parts, colour and temperature. cold hardiness. Canadian Journal of Zoology 67:825–827. Bulletin of Entomological Research 100:279–285. Kynkäänniemi SM, Kortet R, Härkönen L, Kaitala A, Paakkonen Koss AM & Snyder WE (2005) Alternative prey disrupts T, Mustonen AM, Nieminen P, Härkönen S, Ylönen H & biocontrol by a guild of generalist predators. Biological Control Laaksonen S (2010) Threat of an invasive parasitic fly, the 32:243–251. deer ked (Lipoptena cervi), to the reindeer (Rangifer tarandus Koštál V (2006) Eco-physiological phases of insect diapause. tarandus): experimental infection and treatment. Annales Journal of Insect Physiology 52:113–127. Zoologici Fennici 47:28–36. Koštál V, Korbelová J, Rozsypal J, Zahradní²ková H, Cimlová J, Lafferty KD, Dobson AP & Kuris AM (2006) Parasites dominate Tom²ala A & Šimek P (2011) Long-term cold acclimation food web links. Proceedings of the National Academy of Sciences extends survival time at 0°C and modifies the metabolomic USA 103:11211–11216.

70 71 Lafferty KD, Allesina S, Arim M, Briggs CJ, De Leo G, Dobson Lee RE & Costanzo JP (1998) Biological ice nucleation and ice AP, Dunne JA, Johnson PTJ, Kuris AM, Marcogliese DJ, distribution in cold-hardy ectothermic animals. Annual Martinez ND, Memmott J, Marquet PA, McLaughlin JP, Review of Physiology 60:55–72. Mordecai EA, Pascual M, Poulin R & Thieltges DW (2008) Lee RE, Chen C & Denlinger DL (1987) A rapid cold-hardening Parasites in food webs: the ultimate missing links. Ecology process in insects. Science 238:1415–1417. Letters 11:533–546. Lee RE, Damodaran K, Yi SX & Lorigan GA (2006) Rapid cold- Lang A, Filser J & Henschel JR (1999) Predation by ground hardening increases membrane fluidity and cold tolerance of beetles and wolf spiders on herbivorous insects in a maize insect cells. Cryobiology 52:459–463. crop. Agriculture, Ecosystems & Environment 72:189–199. Lehane MJ (2005) The biology of blood-sucking in insects. Lardies MA & Bozinovic F (2008) Genetic variation for plasticity Cambridge University Press, Cambridge. in physiological and life-history traits among populations of Li F, Meng F & Lin J (2010) Effect of sex and development stage an invasive species, the terrestrial isopod Porcellio laevis. on the supercooling point of cat flea Ctenocephalides felis felis Evolutionary Ecology Research 10:747–762. (Siphonaptera: Pulicidae). Chines Journal of Vector Biology and Laugen AT, Laurila A & Merilä J (2002) Maternal and genetic Control 21:531–533. contributions to geographical variation in Rana temporaria Li Y, Ke Z, Wang S, Smith GR & Liu X (2011) An exotic species is larval life-history traits. Biological Journal of the Linnean Society the favorite prey of a native enemy. Public Library of Science 76:61–70. ONE 6(9):e24299. Laukkanen A, Ruoppi P & Mäkinen-Kiljunen S (2005) Deer ked- Lockwood JL, Cassey P & Blackburn T (2005) The role of induced occupational allergic rhinoconjunctivitis. Annals of propagule pressure in explaining species invasions. Trends in Allergy, Asthma & Immunology 94:604–608. Ecology & Evolution 20:223–228. Lavsund S, Nygrén T & Solberg EJ (2003) Status of moose Lockwood JL, Hoopes MF & Marchetti MP (2007) Invasion populations and challenges to moose management in ecology. Blackwell Publishing, Oxford. Fennoscandia. Alces 39:109–130. Lodge DM (1993) Biological invasions: lessons for ecology. Lavy D, Nedved O & Verhoef HA (1997) Effects of starvation on Trends in Ecology & Evolution 8:133–137. body composition and cold tolerance in the collembolan Lonsdale WM (1999) Global patterns of plant invasions and the Orchesella cincta and the isopod Porcellio scaber. Journal of concept of invasibility. Ecology 80:1522–1536. Insect Physiology 43:973–978. Lundgren JG, Toepfer S, Haye T & Kuhlmann U (2010) Leather SR, Walters KFA & Bale JS (1993) The ecology of insect Haemolymph defence of an invasive herbivore: its breadth of overwintering. Cambridge University Press, Cambridge. effectiveness against predators. Journal of Applied Entomology Lee RE (1991) Principles of insect low temperature tolerance. In 134:439–448. Lee RE & Denlinger DL (Eds) Insects at Low Temperature. Maa TC (1969) A revised checklist and concise host index of Chapman and Hall, New York, pp. 17–46. Hippoboscidae (Diptera). Pacific Insects Monograph 20:261– Lee RE & Baust JG (1987) Cold-hardiness in the Antarctic tick, 299. Ixodes uriae. Physiological Zoology 60:499–506. Madder M, Speybroeck N, Brandt J & Berkvens D (1999) Lee RE & Denlinger DL (Eds) (1991) Insects at low temperature. Diapause induction in adults of three Rhipicephalus Chapman and Hall, New York. appendiculatus stocks. Experimental and Applied Acarology 23:961–968.

72 73 Lafferty KD, Allesina S, Arim M, Briggs CJ, De Leo G, Dobson Lee RE & Costanzo JP (1998) Biological ice nucleation and ice AP, Dunne JA, Johnson PTJ, Kuris AM, Marcogliese DJ, distribution in cold-hardy ectothermic animals. Annual Martinez ND, Memmott J, Marquet PA, McLaughlin JP, Review of Physiology 60:55–72. Mordecai EA, Pascual M, Poulin R & Thieltges DW (2008) Lee RE, Chen C & Denlinger DL (1987) A rapid cold-hardening Parasites in food webs: the ultimate missing links. Ecology process in insects. Science 238:1415–1417. Letters 11:533–546. Lee RE, Damodaran K, Yi SX & Lorigan GA (2006) Rapid cold- Lang A, Filser J & Henschel JR (1999) Predation by ground hardening increases membrane fluidity and cold tolerance of beetles and wolf spiders on herbivorous insects in a maize insect cells. Cryobiology 52:459–463. crop. Agriculture, Ecosystems & Environment 72:189–199. Lehane MJ (2005) The biology of blood-sucking in insects. Lardies MA & Bozinovic F (2008) Genetic variation for plasticity Cambridge University Press, Cambridge. in physiological and life-history traits among populations of Li F, Meng F & Lin J (2010) Effect of sex and development stage an invasive species, the terrestrial isopod Porcellio laevis. on the supercooling point of cat flea Ctenocephalides felis felis Evolutionary Ecology Research 10:747–762. (Siphonaptera: Pulicidae). Chines Journal of Vector Biology and Laugen AT, Laurila A & Merilä J (2002) Maternal and genetic Control 21:531–533. contributions to geographical variation in Rana temporaria Li Y, Ke Z, Wang S, Smith GR & Liu X (2011) An exotic species is larval life-history traits. Biological Journal of the Linnean Society the favorite prey of a native enemy. Public Library of Science 76:61–70. ONE 6(9):e24299. Laukkanen A, Ruoppi P & Mäkinen-Kiljunen S (2005) Deer ked- Lockwood JL, Cassey P & Blackburn T (2005) The role of induced occupational allergic rhinoconjunctivitis. Annals of propagule pressure in explaining species invasions. Trends in Allergy, Asthma & Immunology 94:604–608. Ecology & Evolution 20:223–228. Lavsund S, Nygrén T & Solberg EJ (2003) Status of moose Lockwood JL, Hoopes MF & Marchetti MP (2007) Invasion populations and challenges to moose management in ecology. Blackwell Publishing, Oxford. Fennoscandia. Alces 39:109–130. Lodge DM (1993) Biological invasions: lessons for ecology. Lavy D, Nedved O & Verhoef HA (1997) Effects of starvation on Trends in Ecology & Evolution 8:133–137. body composition and cold tolerance in the collembolan Lonsdale WM (1999) Global patterns of plant invasions and the Orchesella cincta and the isopod Porcellio scaber. Journal of concept of invasibility. Ecology 80:1522–1536. Insect Physiology 43:973–978. Lundgren JG, Toepfer S, Haye T & Kuhlmann U (2010) Leather SR, Walters KFA & Bale JS (1993) The ecology of insect Haemolymph defence of an invasive herbivore: its breadth of overwintering. Cambridge University Press, Cambridge. effectiveness against predators. Journal of Applied Entomology Lee RE (1991) Principles of insect low temperature tolerance. In 134:439–448. Lee RE & Denlinger DL (Eds) Insects at Low Temperature. Maa TC (1969) A revised checklist and concise host index of Chapman and Hall, New York, pp. 17–46. Hippoboscidae (Diptera). Pacific Insects Monograph 20:261– Lee RE & Baust JG (1987) Cold-hardiness in the Antarctic tick, 299. Ixodes uriae. Physiological Zoology 60:499–506. Madder M, Speybroeck N, Brandt J & Berkvens D (1999) Lee RE & Denlinger DL (Eds) (1991) Insects at low temperature. Diapause induction in adults of three Rhipicephalus Chapman and Hall, New York. appendiculatus stocks. Experimental and Applied Acarology 23:961–968.

72 73 Madslien K, Ytrehus B, Vikøren T, Malmsten J, Isaksen K, Menke SB, Fisher RN, Jetz W & Holway DA (2007) Biotic and Hygen HO & Solberg EJ (2011) Hair-loss epizootic in moose abiotic controls of Argentine ant invasion success at local and (Alces alces) associated with massive deer ked (Lipoptena landscape scales. Ecology 88:3164–3173. cervi) infestation. Journal of Wildlife Diseases 47:893–906. Mennerat A, Bonadonna F, Perret P & Lambrechts MM (2005) Mamun MAA, Begum N, Shahadat HM & Mondal MMH (2010) Olfactory conditioning experiments in a food-searching Ectoparasites of buffaloes (Bubalus bubalis) in Kurigram passerine bird in semi-natural conditions. Behavioural district of Bangladesh. Journal of the Bangladesh Agricultural Processes 70:264–270. University 8:61–66. Messenger PS (1959) Bioclimatic studies with insects. Annual Marshall AG (1981) The ecology of ectoparasitic insects. Academic Review of Entomology 4:183–206. Press, London. Michie LJ, Mallard F, Majerus MEN & Jiggins FM (2010) Melanic Marshall DJ & Uller T (2007) When is a maternal effect adaptive? through nature or nurture: genetic polymorphism and Oikos 116:1957–1963. phenotypic plasticity in Harmonia axyridis. Journal of Marshall DJ, Heppell SS, Munch SB & Warner RR (2010) The Evolutionary Biology 23:1699–1707. relationship between maternal phenotype and offspring Milinski M (1990) Parasites and host decision-making. In quality: do older mothers really produce the best offspring? Barnard CJ & Behnke JM (Eds) Parasitism and Host Behavior. Ecology 91:2862–2873. Taylor & Francis, London, pp. 95–116. Masaki S (1978) Seasonal and latitudinal adaptations in the life Miyazaki S, Kayukawa T, Chen B, Nomura M & Ishikawa Y cycles of crickets. In Dingle H (Ed) Evolution of Insect (2006) Enhancement of cold hardiness by acclimation is Migration and Diapause. Springer-Verlag, New York, pp. 72– stage-specific in the non-diapausing pupae of onion maggot 100. Delia antiqua (Diptera: Anthomyiidae). European Journal of Massei G & Genov P (1995) Observations of black-billed magpie Entomology 103:691–694. (Pica pica) and carrion crow (Corvus corone comix) grooming Mize EL, Tsao JI & Maurer BA (2011) Habitat correlates with the wild boar (Sus scrofa). Journal of Zoology 236:338–341. spatial distribution of ectoparasites on Peromyscus leucopus in Matsumoto K, Berrada ZL, Klinger E, Goethert HK & Telford SR southern Michigan. Journal of Vector Ecology 36:308–320. (2008) Molecular detection of Bartonella schoenbuchensis from Moczek AP (2010) Phenotypic plasticity and diversity in insects. ectoparasites of deer in Massachusetts. Vector-Borne and Philosophical Transactions of the Royal Society B: Biological Zoonotic Diseases 8:549–554. Sciences 365:593–603. Matsuo Y (2006) Cost of prolonged diapause and its relationship Moller H (1996) Lessons for invasion theory from social insects. to body size in a seed predator. Functional Ecology 20:300–306. Biological Conservation 78:125–142. Meier R, Kotrba M & Ferrar P (1999) Ovoviviparity and Montgomerie R & Weatherhead PJ (1997) How robins find viviparity in the Diptera. Biological Reviews 74:199–258. worms? Animal Behaviour 54:143–151. Meihls LN, Clark TL, Bailey WC & Ellersieck MR (2010) Mooney HA, Mack RN, McNeely JA, Neville LE, Schei PJ & Population growth of soybean aphid, Aphis glycines, under Waage JK (Eds) (2005) Invasive alien species: a new synthesis. varying levels of predator exclusion. Journal of Insect Science Island Press, Washington DC. 10:144. Moran NA (1992) The evolutionary maintenance of alternative phenotypes. American Naturalist 139:971–989.

74 75 Madslien K, Ytrehus B, Vikøren T, Malmsten J, Isaksen K, Menke SB, Fisher RN, Jetz W & Holway DA (2007) Biotic and Hygen HO & Solberg EJ (2011) Hair-loss epizootic in moose abiotic controls of Argentine ant invasion success at local and (Alces alces) associated with massive deer ked (Lipoptena landscape scales. Ecology 88:3164–3173. cervi) infestation. Journal of Wildlife Diseases 47:893–906. Mennerat A, Bonadonna F, Perret P & Lambrechts MM (2005) Mamun MAA, Begum N, Shahadat HM & Mondal MMH (2010) Olfactory conditioning experiments in a food-searching Ectoparasites of buffaloes (Bubalus bubalis) in Kurigram passerine bird in semi-natural conditions. Behavioural district of Bangladesh. Journal of the Bangladesh Agricultural Processes 70:264–270. University 8:61–66. Messenger PS (1959) Bioclimatic studies with insects. Annual Marshall AG (1981) The ecology of ectoparasitic insects. Academic Review of Entomology 4:183–206. Press, London. Michie LJ, Mallard F, Majerus MEN & Jiggins FM (2010) Melanic Marshall DJ & Uller T (2007) When is a maternal effect adaptive? through nature or nurture: genetic polymorphism and Oikos 116:1957–1963. phenotypic plasticity in Harmonia axyridis. Journal of Marshall DJ, Heppell SS, Munch SB & Warner RR (2010) The Evolutionary Biology 23:1699–1707. relationship between maternal phenotype and offspring Milinski M (1990) Parasites and host decision-making. In quality: do older mothers really produce the best offspring? Barnard CJ & Behnke JM (Eds) Parasitism and Host Behavior. Ecology 91:2862–2873. Taylor & Francis, London, pp. 95–116. Masaki S (1978) Seasonal and latitudinal adaptations in the life Miyazaki S, Kayukawa T, Chen B, Nomura M & Ishikawa Y cycles of crickets. In Dingle H (Ed) Evolution of Insect (2006) Enhancement of cold hardiness by acclimation is Migration and Diapause. Springer-Verlag, New York, pp. 72– stage-specific in the non-diapausing pupae of onion maggot 100. Delia antiqua (Diptera: Anthomyiidae). European Journal of Massei G & Genov P (1995) Observations of black-billed magpie Entomology 103:691–694. (Pica pica) and carrion crow (Corvus corone comix) grooming Mize EL, Tsao JI & Maurer BA (2011) Habitat correlates with the wild boar (Sus scrofa). Journal of Zoology 236:338–341. spatial distribution of ectoparasites on Peromyscus leucopus in Matsumoto K, Berrada ZL, Klinger E, Goethert HK & Telford SR southern Michigan. Journal of Vector Ecology 36:308–320. (2008) Molecular detection of Bartonella schoenbuchensis from Moczek AP (2010) Phenotypic plasticity and diversity in insects. ectoparasites of deer in Massachusetts. Vector-Borne and Philosophical Transactions of the Royal Society B: Biological Zoonotic Diseases 8:549–554. Sciences 365:593–603. Matsuo Y (2006) Cost of prolonged diapause and its relationship Moller H (1996) Lessons for invasion theory from social insects. to body size in a seed predator. Functional Ecology 20:300–306. Biological Conservation 78:125–142. Meier R, Kotrba M & Ferrar P (1999) Ovoviviparity and Montgomerie R & Weatherhead PJ (1997) How robins find viviparity in the Diptera. Biological Reviews 74:199–258. worms? Animal Behaviour 54:143–151. Meihls LN, Clark TL, Bailey WC & Ellersieck MR (2010) Mooney HA, Mack RN, McNeely JA, Neville LE, Schei PJ & Population growth of soybean aphid, Aphis glycines, under Waage JK (Eds) (2005) Invasive alien species: a new synthesis. varying levels of predator exclusion. Journal of Insect Science Island Press, Washington DC. 10:144. Moran NA (1992) The evolutionary maintenance of alternative phenotypes. American Naturalist 139:971–989.

74 75 Morey AC, Hutchison WD, Venette RC & Burkness EC (2012) Nygren GH, Bergström A & Nylin S (2008) Latitudinal body size Cold hardiness of Helicoverpa zea (Lepidoptera: Noctuidae) clines in the butterfly Polyommatus icarus are shaped by gene- pupae. Environmental Entomology 41:172–179. environment interactions. Journal of Insect Science 8:47. Mousseau TA (1997) Ectotherms follow the converse to Nylin S & Svärd L (1991) Latitudinal patterns in the size of Bergmann’s Rule. Evolution 51:630–632. European butterflies. Ecography (Holarctic Ecology) 14:192–202. Mousseau TA & Dingle H (1991) Maternal effects in insect life Nylin S & Gotthard K (1998) Plasticity in life-history traits. histories. Annual Review of Entomology 36:511–534. Annual Review of Entomology 43:63–83. Mousseau TA & Fox CW (1998) The adaptive significance of Ohgushi T (1996) Consequences of adult size for survival and maternal effects. Trends in Ecology & Evolution 13:403–407. reproductive performance in a herbivorous ladybird beetle. Murakami M (1999) Effect of avian predation on survival of leaf- Ecological Entomology 21:47–55. rolling lepidopterous larvae. Researches on Population Ecology Ohtsu T, Kimura MT & Katagiri C (1998) How Drosophila 41:135–138. species acquire cold tolerance: qualitative changes of Mwangi EN, Newson RM & Kaaya GP (1991) Predation of free- phospholipids. European Journal of Biochemistry 252:608–611. living engorged female Rhipicephalus appendiculatus. Olalla-Tárraga MÁ, Rodríguez MÁ & Hawkins BA (2006) Experimental and Applied Acarology 12:153–162. Broad-scale patterns of body size in squamate reptiles of Mäntylä E, Alessio GA, Blande JD, Heijari J, Holopainen JK, Europe and North America. Journal of Biogeography 33:781– Laaksonen T, Piirtola P & Klemola T (2008) From plants to 793. birds: higher avian predation rates in trees responding to Paakkonen T, Mustonen AM, Roininen H, Niemelä P, Ruusila V insect herbivory. Public Library of Science ONE 3(7):e2832. & Nieminen P (2010) Parasitism of the deer ked, Lipoptena Naef-Daenzer L, Naef-Daenzer B & Nager RG (2000) Prey cervi, on the moose, Alces alces, in eastern Finland. Medical and selection and foraging performance of breeding great tits Veterinary Entomology 24:411–417. Parus major in relation to food availability. Journal of Avian Parker GA & Begon M (1986) Optimal egg size and clutch size: Biology 31:206–214. effects of environment and maternal phenotype. American Nevitt GA, Losekoot M & Weimerskirch H (2008) Evidence for Naturalist 128:573–592. olfactory search in wandering albatross, Diomedea exulans. Peacock L & Worner SP (2008) Biological and ecological traits Proceedings of the National Academy of Sciences USA 105:4576– that assist establishment of alien invasive insects. New 4581. Zealand Plant Protection Society 61:1–7. Nieminen P, Paakkonen T, Eerilä H, Puukka K, Riikonen J, Peres CA (1996) Ungulate ectoparasite removal by black Lehto VP & Mustonen AM (2012) Freezing tolerance and low caracaras and pale-winged trumpeters in Amazonian forests. molecular weight cryoprotectants in an invasive parasitic fly, Wilson Bulletin 108:170–175. the deer ked (Lipoptena cervi). Journal of Experimental Zoology Pigliucci M (2005) Evolution of phenotypic plasticity: where are Part A: Ecological Genetics and Physiology 317:1–8. we going now? Trends in Ecology & Evolution 20:481–486. Nyamukondiwa C, Kleynhans E & Terblanche JS (2010) Pimentel C & Nilsson JÅ (2007) Response of great tits Parus Phenotypic plasticity of thermal tolerance contributes to the major to an irruption of a pine processionary moth invasion potential of Mediterranean fruit flies (Ceratitis Thaumetopoea pityocampa population with a shifted capitata). Ecological Entomology 35:565–575. phenology. Ardea 95:191–199.

76 77 Morey AC, Hutchison WD, Venette RC & Burkness EC (2012) Nygren GH, Bergström A & Nylin S (2008) Latitudinal body size Cold hardiness of Helicoverpa zea (Lepidoptera: Noctuidae) clines in the butterfly Polyommatus icarus are shaped by gene- pupae. Environmental Entomology 41:172–179. environment interactions. Journal of Insect Science 8:47. Mousseau TA (1997) Ectotherms follow the converse to Nylin S & Svärd L (1991) Latitudinal patterns in the size of Bergmann’s Rule. Evolution 51:630–632. European butterflies. Ecography (Holarctic Ecology) 14:192–202. Mousseau TA & Dingle H (1991) Maternal effects in insect life Nylin S & Gotthard K (1998) Plasticity in life-history traits. histories. Annual Review of Entomology 36:511–534. Annual Review of Entomology 43:63–83. Mousseau TA & Fox CW (1998) The adaptive significance of Ohgushi T (1996) Consequences of adult size for survival and maternal effects. Trends in Ecology & Evolution 13:403–407. reproductive performance in a herbivorous ladybird beetle. Murakami M (1999) Effect of avian predation on survival of leaf- Ecological Entomology 21:47–55. rolling lepidopterous larvae. Researches on Population Ecology Ohtsu T, Kimura MT & Katagiri C (1998) How Drosophila 41:135–138. species acquire cold tolerance: qualitative changes of Mwangi EN, Newson RM & Kaaya GP (1991) Predation of free- phospholipids. European Journal of Biochemistry 252:608–611. living engorged female Rhipicephalus appendiculatus. Olalla-Tárraga MÁ, Rodríguez MÁ & Hawkins BA (2006) Experimental and Applied Acarology 12:153–162. Broad-scale patterns of body size in squamate reptiles of Mäntylä E, Alessio GA, Blande JD, Heijari J, Holopainen JK, Europe and North America. Journal of Biogeography 33:781– Laaksonen T, Piirtola P & Klemola T (2008) From plants to 793. birds: higher avian predation rates in trees responding to Paakkonen T, Mustonen AM, Roininen H, Niemelä P, Ruusila V insect herbivory. Public Library of Science ONE 3(7):e2832. & Nieminen P (2010) Parasitism of the deer ked, Lipoptena Naef-Daenzer L, Naef-Daenzer B & Nager RG (2000) Prey cervi, on the moose, Alces alces, in eastern Finland. Medical and selection and foraging performance of breeding great tits Veterinary Entomology 24:411–417. Parus major in relation to food availability. Journal of Avian Parker GA & Begon M (1986) Optimal egg size and clutch size: Biology 31:206–214. effects of environment and maternal phenotype. American Nevitt GA, Losekoot M & Weimerskirch H (2008) Evidence for Naturalist 128:573–592. olfactory search in wandering albatross, Diomedea exulans. Peacock L & Worner SP (2008) Biological and ecological traits Proceedings of the National Academy of Sciences USA 105:4576– that assist establishment of alien invasive insects. New 4581. Zealand Plant Protection Society 61:1–7. Nieminen P, Paakkonen T, Eerilä H, Puukka K, Riikonen J, Peres CA (1996) Ungulate ectoparasite removal by black Lehto VP & Mustonen AM (2012) Freezing tolerance and low caracaras and pale-winged trumpeters in Amazonian forests. molecular weight cryoprotectants in an invasive parasitic fly, Wilson Bulletin 108:170–175. the deer ked (Lipoptena cervi). Journal of Experimental Zoology Pigliucci M (2005) Evolution of phenotypic plasticity: where are Part A: Ecological Genetics and Physiology 317:1–8. we going now? Trends in Ecology & Evolution 20:481–486. Nyamukondiwa C, Kleynhans E & Terblanche JS (2010) Pimentel C & Nilsson JÅ (2007) Response of great tits Parus Phenotypic plasticity of thermal tolerance contributes to the major to an irruption of a pine processionary moth invasion potential of Mediterranean fruit flies (Ceratitis Thaumetopoea pityocampa population with a shifted capitata). Ecological Entomology 35:565–575. phenology. Ardea 95:191–199.

76 77 Pimm SL (1989) Theories of predicting success and impact of starvation resistance in yellow dung flies. Evolutionary introduced species. In Drake JA, Mooney HA, di Castri F, Ecology Research 8:1215–1234. Groves RH, Kruger FJ, Rejmánek M & Williamson M (Eds) Reim C, Teuschl Y & Blanckenhorn WU (2006b) Size-dependent Biological Invasions: a Global Perspective. John Wiley and Sons effects of larval and adult food availability on reproductive Ltd., New York, pp. 351–367. energy allocation in the yellow dung fly. Functional Ecology Plaistow SJ, St. Clair JJH, Grant J & Benton TG (2007) How to 20:1012–1021. put all your eggs in one basket: empirical patterns of Reise K, Olenin S & Thieltges DW (2006) Are aliens threatening offspring provisioning throughout a mother’s lifetime. European aquatic coastal ecosystems? Helgoland Marine American Naturalist 170:520–529. Research 60:77–83. Poulin R (2007) Evolutionary ecology of parasites. Princeton Richards CL, Bossdorf O, Muth NZ, Gurevitch J & Pigliucci M University Press, Princeton. (2006) Jack of all trades, master of some? On the role of Preisser EL, Elkinton JS & Abell K (2008) Evolution of increased phenotypic plasticity in plant invasions. Ecology Letters 9:981– cold tolerance during range expansion of the elongate 993. hemlock scale Fiorinia externa Ferris (Hemiptera: Robinson WH (2005) Handbook of urban insects and arachnids. Diaspididae). Ecological Entomology 33:709–715. Cambridge University Press, Cambridge. Price PW, Denno RF, Eubanks MD, Finke DL & Kaplan I (2011) Roff DA (1980) Optimizing development time in a seasonal Insect ecology: behavior, populations and communities. environment: the ‘ups and downs’ of clinal variation. Cambridge University Press, Cambridge. Oecologia 45:202–208. Pugh PJA (1994) Supercooling points and water contents in Roff DA (1992) The evolution of life histories: theory and analysis. Acari. Acta Oecologica 15:71–77. Chapman & Hall, New York. Pyšek P, Richardson DM, Pergl J, Jarosík V, Sixtová Z & Weber Roland J, Hannon SJ & Smith MA (1986) Foraging pattern of E (2008) Geographical and taxonomical biases in invasion pine siskins and its influence on winter moth survival in an ecology. Trends in Ecology & Evolution 23:237–244. apple orchard. Oecologia 69:47–52. Rajchard J (2009) Ultraviolet (UV) light perception by birds: a Roulin A, Brinkhof MWG, Bize P, Richner H, Jungi TW, Bavoux review. Veterinární medicína 54:351–359. C, Boileau N & Burneleau G (2003) Which chick is tasty to Rantanen T, Reunala T, Vuojolahti P & Hackman W (1982) parasites? The importance of host immunology vs. parasite Persistent pruritic papules from deer ked bites. Acta Dermato- life history. Journal of Animal Ecology 72:75–81. Venereologica 62:307–311. Sagata K & Lester PJ (2009) Behavioural plasticity associated Rasnitsyn AP & Quicke DL (Eds) (2002) History of insects. with propagule size, resources, and the invasion success of Kluwer Academic Publishers, Dordrecht. the Argentine ant Linepithema humile. Journal of Applied Régnière J, Powell J, Bentz B & Nealis V (2012) Effects of Ecology 46:19–27. temperature on development, survival and reproduction of Sakai AK, Allendorf FW, Holt JS, Lodge DM, Molofsky J,With insects: experimental design, data analysis and modeling. KA, Baughman S, Cabin RJ, Cohen JE, Ellstrand NC, Journal of Insect Physiology 58:634–647. McCauley DE, O’Neil P, Parker IM, Thompson JN & Weller Reim C, Teuschl Y & Blanckenhorn WU (2006a) Size-dependent SG (2001) The population biology of invasive species. Annual effects of temperature and food stress on energy reserves and Review of Ecology, Evolution, and Systematics 32:305–332.

78 79 Pimm SL (1989) Theories of predicting success and impact of starvation resistance in yellow dung flies. Evolutionary introduced species. In Drake JA, Mooney HA, di Castri F, Ecology Research 8:1215–1234. Groves RH, Kruger FJ, Rejmánek M & Williamson M (Eds) Reim C, Teuschl Y & Blanckenhorn WU (2006b) Size-dependent Biological Invasions: a Global Perspective. John Wiley and Sons effects of larval and adult food availability on reproductive Ltd., New York, pp. 351–367. energy allocation in the yellow dung fly. Functional Ecology Plaistow SJ, St. Clair JJH, Grant J & Benton TG (2007) How to 20:1012–1021. put all your eggs in one basket: empirical patterns of Reise K, Olenin S & Thieltges DW (2006) Are aliens threatening offspring provisioning throughout a mother’s lifetime. European aquatic coastal ecosystems? Helgoland Marine American Naturalist 170:520–529. Research 60:77–83. Poulin R (2007) Evolutionary ecology of parasites. Princeton Richards CL, Bossdorf O, Muth NZ, Gurevitch J & Pigliucci M University Press, Princeton. (2006) Jack of all trades, master of some? On the role of Preisser EL, Elkinton JS & Abell K (2008) Evolution of increased phenotypic plasticity in plant invasions. Ecology Letters 9:981– cold tolerance during range expansion of the elongate 993. hemlock scale Fiorinia externa Ferris (Hemiptera: Robinson WH (2005) Handbook of urban insects and arachnids. Diaspididae). Ecological Entomology 33:709–715. Cambridge University Press, Cambridge. Price PW, Denno RF, Eubanks MD, Finke DL & Kaplan I (2011) Roff DA (1980) Optimizing development time in a seasonal Insect ecology: behavior, populations and communities. environment: the ‘ups and downs’ of clinal variation. Cambridge University Press, Cambridge. Oecologia 45:202–208. Pugh PJA (1994) Supercooling points and water contents in Roff DA (1992) The evolution of life histories: theory and analysis. Acari. Acta Oecologica 15:71–77. Chapman & Hall, New York. Pyšek P, Richardson DM, Pergl J, Jarosík V, Sixtová Z & Weber Roland J, Hannon SJ & Smith MA (1986) Foraging pattern of E (2008) Geographical and taxonomical biases in invasion pine siskins and its influence on winter moth survival in an ecology. Trends in Ecology & Evolution 23:237–244. apple orchard. Oecologia 69:47–52. Rajchard J (2009) Ultraviolet (UV) light perception by birds: a Roulin A, Brinkhof MWG, Bize P, Richner H, Jungi TW, Bavoux review. Veterinární medicína 54:351–359. C, Boileau N & Burneleau G (2003) Which chick is tasty to Rantanen T, Reunala T, Vuojolahti P & Hackman W (1982) parasites? The importance of host immunology vs. parasite Persistent pruritic papules from deer ked bites. Acta Dermato- life history. Journal of Animal Ecology 72:75–81. Venereologica 62:307–311. Sagata K & Lester PJ (2009) Behavioural plasticity associated Rasnitsyn AP & Quicke DL (Eds) (2002) History of insects. with propagule size, resources, and the invasion success of Kluwer Academic Publishers, Dordrecht. the Argentine ant Linepithema humile. Journal of Applied Régnière J, Powell J, Bentz B & Nealis V (2012) Effects of Ecology 46:19–27. temperature on development, survival and reproduction of Sakai AK, Allendorf FW, Holt JS, Lodge DM, Molofsky J,With insects: experimental design, data analysis and modeling. KA, Baughman S, Cabin RJ, Cohen JE, Ellstrand NC, Journal of Insect Physiology 58:634–647. McCauley DE, O’Neil P, Parker IM, Thompson JN & Weller Reim C, Teuschl Y & Blanckenhorn WU (2006a) Size-dependent SG (2001) The population biology of invasive species. Annual effects of temperature and food stress on energy reserves and Review of Ecology, Evolution, and Systematics 32:305–332.

78 79 Samish M & Rehacek J (1999) Pathogens and predators of ticks invasive birds: a regional test using the rose-ringed parakeet and their potential in biological control. Annual Review of (Psittacula krameri) as a case study. Diversity and Distributions Entomology 44:159–182. 15:310–318. Samuel WM (2007) Factors affecting epizootics of winter ticks Sinclair BJ (1999) Insect cold tolerance: how many kinds of and mortality of moose. Alces 43:39–48. frozen? European Journal of Entomology 96:157–164. Samuel WM, Mooring MS & Aalangdong OI (2000) Adaptations Sinclair BJ & Chown SL (2005) Deleterious effects of repeated of winter ticks (Dermacentor albipictus) to invade moose and cold exposure in a freeze-tolerant sub-Antarctic caterpillar. moose to evade ticks. Alces 36:183–195. Journal of Experimental Biology 208:869–879. Saunders DS, Steel CGH, Vafopoulou X & Lewis RD (2002) Sinu PA (2011) Avian pest control in tea plantations of sub- Insect clocks, (3rd edn). Elsevier Science B.V., Amsterdam. Himalayan plains of Northeast India: mixed-species foraging Sazima I (2007) Unexpected cleaners: black vultures (Coragyps flock matters. Biological Control 58:362–366. atratus) remove debris, ticks, and peck at sores of capybaras Šlachta M, Berková P, Vambera J & Koštál V (2002a) Physiology (Hydrochoerus hydrochaeris), with an overview of tick- of cold-acclimation in non-diapausing adults of Pyrrhocoris removing birds in Brazil. Revista Brasileira de Ornitologia apterus (Heteroptera). European Journal of Entomology 99:181– 15:417–426. 187. Schelhaas DP & Larson OR (1989) Cold hardiness and winter Šlachta M, Vambera J, Zahradnícková H & Koštál V (2002b) survival in the bird flea, Ceratophyllus idius. Journal of Insect Entering diapause is a prerequisite for successful cold- Physiology 35:149–153. acclimation in adult Graphosoma lineatum (Heteroptera: Schoener TW & Spiller DA (1995) Effect of predators and area Pentatomidae). Journal of Insect Physiology 48:1031–1039. on invasion: an experiment with island spiders. Science Small RW (2005) A review of Melophagus ovinus (L.), the sheep 267:1811–1813. ked. Veterinary Parasitology 130:141–155. Schoolmaster DR & Snyder RE (2007) Invasibility in a Smit NJ & Basson L (2002) Gnathia pantherina sp. n. (Crustacea: spatiotemporally fluctuating environment is determined by Isopoda: Gnathiidae), a temporary ectoparasite of some the periodicity of fluctuations and resident turnover rates. elasmobranch species from southern Africa. Folia Proceedings of the Royal Society B: Biological Sciences 274:1429– Parasitologica (Praha) 49:137–151. 1435. Smith CC & Fretwell SD (1974) The optimal balance between Schultz TD (1983) Opportunistic foraging of Western kingbirds size and number of offspring. American Naturalist 108:499– on aggregations of tiger beetles. Auk 100:496–497. 506. Sexton JP, Mckay JK & Sala A (2002) Plasticity and genetic Smith RJ (2002) Effect of larval body size on overwinter survival diversity may allow saltcedar to invade cold climates in and emerging adult size in the burying beetle, Nicrophorus North America. Ecological Applications 12:1652–1660. investigator. Canadian Journal of Zoology 80:1588–1593. Shenbrot G, Krasnov B & Lu L (2007) Geographical range size Snyder WE, Chang GC & Prasad RP (2005) Conservation and host specificity in ectoparasites: a case study with biological control: biodiversity influences the effectiveness of Amphipsylla fleas and rodent hosts. Journal of Biogeography predators. In Barbosa P & Castellanos I (Eds) Ecology of 34:1679–1690. Predator-Prey Interactions. Oxford University Press, London, Shwartz A, Strubbe D, Butler CJ, Matthysen E & Kark S (2009) pp. 324–343. The effect of enemy-release and climate conditions on

80 81 Samish M & Rehacek J (1999) Pathogens and predators of ticks invasive birds: a regional test using the rose-ringed parakeet and their potential in biological control. Annual Review of (Psittacula krameri) as a case study. Diversity and Distributions Entomology 44:159–182. 15:310–318. Samuel WM (2007) Factors affecting epizootics of winter ticks Sinclair BJ (1999) Insect cold tolerance: how many kinds of and mortality of moose. Alces 43:39–48. frozen? European Journal of Entomology 96:157–164. Samuel WM, Mooring MS & Aalangdong OI (2000) Adaptations Sinclair BJ & Chown SL (2005) Deleterious effects of repeated of winter ticks (Dermacentor albipictus) to invade moose and cold exposure in a freeze-tolerant sub-Antarctic caterpillar. moose to evade ticks. Alces 36:183–195. Journal of Experimental Biology 208:869–879. Saunders DS, Steel CGH, Vafopoulou X & Lewis RD (2002) Sinu PA (2011) Avian pest control in tea plantations of sub- Insect clocks, (3rd edn). Elsevier Science B.V., Amsterdam. Himalayan plains of Northeast India: mixed-species foraging Sazima I (2007) Unexpected cleaners: black vultures (Coragyps flock matters. Biological Control 58:362–366. atratus) remove debris, ticks, and peck at sores of capybaras Šlachta M, Berková P, Vambera J & Koštál V (2002a) Physiology (Hydrochoerus hydrochaeris), with an overview of tick- of cold-acclimation in non-diapausing adults of Pyrrhocoris removing birds in Brazil. Revista Brasileira de Ornitologia apterus (Heteroptera). European Journal of Entomology 99:181– 15:417–426. 187. Schelhaas DP & Larson OR (1989) Cold hardiness and winter Šlachta M, Vambera J, Zahradnícková H & Koštál V (2002b) survival in the bird flea, Ceratophyllus idius. Journal of Insect Entering diapause is a prerequisite for successful cold- Physiology 35:149–153. acclimation in adult Graphosoma lineatum (Heteroptera: Schoener TW & Spiller DA (1995) Effect of predators and area Pentatomidae). Journal of Insect Physiology 48:1031–1039. on invasion: an experiment with island spiders. Science Small RW (2005) A review of Melophagus ovinus (L.), the sheep 267:1811–1813. ked. Veterinary Parasitology 130:141–155. Schoolmaster DR & Snyder RE (2007) Invasibility in a Smit NJ & Basson L (2002) Gnathia pantherina sp. n. (Crustacea: spatiotemporally fluctuating environment is determined by Isopoda: Gnathiidae), a temporary ectoparasite of some the periodicity of fluctuations and resident turnover rates. elasmobranch species from southern Africa. Folia Proceedings of the Royal Society B: Biological Sciences 274:1429– Parasitologica (Praha) 49:137–151. 1435. Smith CC & Fretwell SD (1974) The optimal balance between Schultz TD (1983) Opportunistic foraging of Western kingbirds size and number of offspring. American Naturalist 108:499– on aggregations of tiger beetles. Auk 100:496–497. 506. Sexton JP, Mckay JK & Sala A (2002) Plasticity and genetic Smith RJ (2002) Effect of larval body size on overwinter survival diversity may allow saltcedar to invade cold climates in and emerging adult size in the burying beetle, Nicrophorus North America. Ecological Applications 12:1652–1660. investigator. Canadian Journal of Zoology 80:1588–1593. Shenbrot G, Krasnov B & Lu L (2007) Geographical range size Snyder WE, Chang GC & Prasad RP (2005) Conservation and host specificity in ectoparasites: a case study with biological control: biodiversity influences the effectiveness of Amphipsylla fleas and rodent hosts. Journal of Biogeography predators. In Barbosa P & Castellanos I (Eds) Ecology of 34:1679–1690. Predator-Prey Interactions. Oxford University Press, London, Shwartz A, Strubbe D, Butler CJ, Matthysen E & Kark S (2009) pp. 324–343. The effect of enemy-release and climate conditions on

80 81 Sømme L (1981) Cold tolerance of alpine, Arctic, and Antarctic role for calcium. American Journal of Physiology: Regulatory, Collembola and mites. Cryobiology 18:212–220. Integrative and Comparative Physiology 294: R1938–1946. Sømme L (1982) Supercooling and winter survival in terrestrial Tinsley RC (1999) Overview: extreme environments. Parasitology arthropods. Comparative Biochemistry and Physiology: Part A 119 (Suppl.):S1–S6. 73:519–543. Tschirren B, Bischoff LL, Saladin V & Richner H (2007) Host Sømme L (1996) The effect of prolonged exposures at low condition and host immunity affect parasite ętness in a bird– temperatures in insects. Cryo-Letters 17:341–346. ectoparasite system. Functional Ecology 21:372–378. Sréter T, Széll Z & Varga I (2003) Ectoparasite infestations of red Turchin P, Taylor AD & Reeve JD (1999) Dynamical role of foxes (Vulpes vulpes) in Hungary. Veterinary Parasitology predators in population cycles of a forest insect: an 115:349–354. experimental test. Science 285:1068–1071. Stearns SC (1992) The evolution of life histories. Oxford University Valera F, Casas-Crivillé A & Calero-Torralbo MA (2006) Press Inc., New York. Prolonged diapause in the ectoparasite Carnus hemapterus Stevens DJ (2004) Pupal development temperature alters adult (Diptera: Cyclorrhapha, Acalyptratae) – how frequent is it in phenotype in the speckled wood butterfly, Pararge aegeria. parasites? Parasitology 133:179–186. Journal of Thermal Biology 29:205–210. Van Bael SA, Bichier P & Greenberg R (2007) Bird predation on Stillwell RC (2010) Are latitudinal clines in body size adaptive? insects reduces damage to the foliage of cocoa trees Oikos 119:1387–1390. (Theobroma cacao) in western Panama. Journal of Tropical Suhonen J (1993) Predation risk influences the use of foraging Ecology 23:715–719. sites by tits. Ecology 74:1197–1203. Vaughan TA, Ryan JM & Czaplewski NJ (2011) Mammalogy, (5th Sutherst RW, Wilson LJ & Cook IM (2000) Predation of the cattle edn). Jones and Bartlett Publishers, Massachusetts. tick, Boophilus microplus (Canestrini) (Acarina: Ixodidae), in Vermeij GJ (1996) An agenda for invasion biology. Biological three Australian pastures. Australian Journal of Entomology Conservation 78:3–9. 39:70–77. Vermeij GJ (2005) Invasion as expectation: a historical fact of life. Szczurek B & Kadulski S (2001) Dynamics of infestation of the In Sax DF, Stachowicz JJ & Gaines SD (Eds) Species Invasions: Lipoptena cervi (L.) (Diptera, Hippoboscidae) with the fallow Insights Into Ecology, Evolution and Biogeography. Sinauer deer from Pomerania. Wiadomoïci Parazytologiczne 47:67–72. Associates Inc., Sunderland, pp. 315–339. [in Polish]. Vernon P & Vannier G (2001) Freezing susceptibility and Tanhuanpää M, Ruohomäki K, Kaitaniemi P & Klemola T (1999) freezing tolerance in Palaearctic Cetoniidae (Coleoptera). Different impact of pupal predation on populations of Canadian Journal of Zoology 79:67–74. Epirrita autumnata (Lepidoptera; Geometridae) within and Viitala J, Korpimäki E, Palokangas P & Koivula M (1995) outside the outbreak range. Journal of Animal Ecology 68:562– Attraction of kestrels to vole scent marks visible in ultraviolet 570. light. Nature 373:425–427. Tauber MJ, Tauber CA & Masaki S (1986) Seasonal adaptations of Von Blanckenhagen F, Eccard JA & Ylönen H (2007) Animal insects. Oxford University Press Inc., New York. protein as a reproductive constraint in spring reproduction of Teets NM, Elnitsky MA, Benoit JB, Lopez-Martinez G, Denlinger the bank vole. Ecoscience 14:323–329. DL & Lee RE (2008) Rapid cold-hardening in larvae of the Välimäki P, Madslien K, Malmsten J, Härkönen L, Härkönen S, Antarctic midge Belgica antarctica: cellular cold-sensing and a Kaitala A, Kortet R, Laaksonen S, Mehl R, Redford L, Ylönen

82 83 Sømme L (1981) Cold tolerance of alpine, Arctic, and Antarctic role for calcium. American Journal of Physiology: Regulatory, Collembola and mites. Cryobiology 18:212–220. Integrative and Comparative Physiology 294: R1938–1946. Sømme L (1982) Supercooling and winter survival in terrestrial Tinsley RC (1999) Overview: extreme environments. Parasitology arthropods. Comparative Biochemistry and Physiology: Part A 119 (Suppl.):S1–S6. 73:519–543. Tschirren B, Bischoff LL, Saladin V & Richner H (2007) Host Sømme L (1996) The effect of prolonged exposures at low condition and host immunity affect parasite ętness in a bird– temperatures in insects. Cryo-Letters 17:341–346. ectoparasite system. Functional Ecology 21:372–378. Sréter T, Széll Z & Varga I (2003) Ectoparasite infestations of red Turchin P, Taylor AD & Reeve JD (1999) Dynamical role of foxes (Vulpes vulpes) in Hungary. Veterinary Parasitology predators in population cycles of a forest insect: an 115:349–354. experimental test. Science 285:1068–1071. Stearns SC (1992) The evolution of life histories. Oxford University Valera F, Casas-Crivillé A & Calero-Torralbo MA (2006) Press Inc., New York. Prolonged diapause in the ectoparasite Carnus hemapterus Stevens DJ (2004) Pupal development temperature alters adult (Diptera: Cyclorrhapha, Acalyptratae) – how frequent is it in phenotype in the speckled wood butterfly, Pararge aegeria. parasites? Parasitology 133:179–186. Journal of Thermal Biology 29:205–210. Van Bael SA, Bichier P & Greenberg R (2007) Bird predation on Stillwell RC (2010) Are latitudinal clines in body size adaptive? insects reduces damage to the foliage of cocoa trees Oikos 119:1387–1390. (Theobroma cacao) in western Panama. Journal of Tropical Suhonen J (1993) Predation risk influences the use of foraging Ecology 23:715–719. sites by tits. Ecology 74:1197–1203. Vaughan TA, Ryan JM & Czaplewski NJ (2011) Mammalogy, (5th Sutherst RW, Wilson LJ & Cook IM (2000) Predation of the cattle edn). Jones and Bartlett Publishers, Massachusetts. tick, Boophilus microplus (Canestrini) (Acarina: Ixodidae), in Vermeij GJ (1996) An agenda for invasion biology. Biological three Australian pastures. Australian Journal of Entomology Conservation 78:3–9. 39:70–77. Vermeij GJ (2005) Invasion as expectation: a historical fact of life. Szczurek B & Kadulski S (2001) Dynamics of infestation of the In Sax DF, Stachowicz JJ & Gaines SD (Eds) Species Invasions: Lipoptena cervi (L.) (Diptera, Hippoboscidae) with the fallow Insights Into Ecology, Evolution and Biogeography. Sinauer deer from Pomerania. Wiadomoïci Parazytologiczne 47:67–72. Associates Inc., Sunderland, pp. 315–339. [in Polish]. Vernon P & Vannier G (2001) Freezing susceptibility and Tanhuanpää M, Ruohomäki K, Kaitaniemi P & Klemola T (1999) freezing tolerance in Palaearctic Cetoniidae (Coleoptera). Different impact of pupal predation on populations of Canadian Journal of Zoology 79:67–74. Epirrita autumnata (Lepidoptera; Geometridae) within and Viitala J, Korpimäki E, Palokangas P & Koivula M (1995) outside the outbreak range. Journal of Animal Ecology 68:562– Attraction of kestrels to vole scent marks visible in ultraviolet 570. light. Nature 373:425–427. Tauber MJ, Tauber CA & Masaki S (1986) Seasonal adaptations of Von Blanckenhagen F, Eccard JA & Ylönen H (2007) Animal insects. Oxford University Press Inc., New York. protein as a reproductive constraint in spring reproduction of Teets NM, Elnitsky MA, Benoit JB, Lopez-Martinez G, Denlinger the bank vole. Ecoscience 14:323–329. DL & Lee RE (2008) Rapid cold-hardening in larvae of the Välimäki P, Madslien K, Malmsten J, Härkönen L, Härkönen S, Antarctic midge Belgica antarctica: cellular cold-sensing and a Kaitala A, Kortet R, Laaksonen S, Mehl R, Redford L, Ylönen

82 83 H & Ytrehus B (2010) Fennoscandian distribution of an Wittenberger JF (1981) Animal social behavior. Duxbury Press, important parasite of cervids, the deer ked (Lipoptena cervi), Boston. revisited. Parasitology Research 107:117–125. Zachariassen KE & Kristiansen E (2003) What determines the Välimäki P, Kaitala A, Madslien K, Härkönen L, Várkonyi G, strategy of cold-hardiness? Acta Societatis Zoologicae Bohemicae Heikkilä J, Jaakola M, Ylönen H, Kortet R & Ytrehus B (2011) 67:51–58. Geographical variation in host use of a blood-feeding Zhao J, Cui N, Zhang F, Yin X & Xu Y (2010) Effects of body size ectoparasitic fly: implications for population invasiveness. and fat content on cold tolerance in adults of Harmonia Oecologia 166:985–995. axyridis (Pallas) (Coleoptera: Coccinellidae). Acta Entomologica Wall R (2007) Ectoparasites: future challenges in a changing Sinica 53:1213–1219. world. Veterinary Parasitology 148:62–74. Watanabe M (2002) Cold tolerance and myo-inositol accumulation in overwintering adults of a lady beetle, Harmonia axyridis (Coleoptera: Coccinellidae). European Journal of Entomology 99:5–9. Wells JD & Henderson G (1993) Fire ant predation on native and introduced subterranean termites in the laboratory: effect of high soldier number in Coptotermes formosanus. Ecological Entomology 18:270–274. West-Eberhard MJ (2003) Developmental plasticity and evolution. Oxford University Press Inc., New York. Wharton DA (1999) Parasites and low temperatures. Parasitology 119(Suppl):S7–S17. Wilkinson PR (1970) A preliminary note on predation on free- living engorged female Rocky Mountain wood ticks. Journal of Medical Entomology 7:493–496. Williamson M (1996) Biological invasions. Chapman & Hall, London. Williamson M & Fitter A (1996) The varying success of invaders. Ecology 77:1661–1666. Wilson EE, Mullen LM & Holway DA (2009) Life history plasticity magnifies the ecological effects of a social wasp invasion. Proceedings of the National Academy of Sciences USA 106:12809–12813. Wilson PW, Heneghan AF & Haymet ADJ (2003) Ice nucleation in nature: supercooling point (SCP) measurements and the role of heterogeneous nucleation. Cryobiology 46:88–98.

84 85 H & Ytrehus B (2010) Fennoscandian distribution of an Wittenberger JF (1981) Animal social behavior. Duxbury Press, important parasite of cervids, the deer ked (Lipoptena cervi), Boston. revisited. Parasitology Research 107:117–125. Zachariassen KE & Kristiansen E (2003) What determines the Välimäki P, Kaitala A, Madslien K, Härkönen L, Várkonyi G, strategy of cold-hardiness? Acta Societatis Zoologicae Bohemicae Heikkilä J, Jaakola M, Ylönen H, Kortet R & Ytrehus B (2011) 67:51–58. Geographical variation in host use of a blood-feeding Zhao J, Cui N, Zhang F, Yin X & Xu Y (2010) Effects of body size ectoparasitic fly: implications for population invasiveness. and fat content on cold tolerance in adults of Harmonia Oecologia 166:985–995. axyridis (Pallas) (Coleoptera: Coccinellidae). Acta Entomologica Wall R (2007) Ectoparasites: future challenges in a changing Sinica 53:1213–1219. world. Veterinary Parasitology 148:62–74. Watanabe M (2002) Cold tolerance and myo-inositol accumulation in overwintering adults of a lady beetle, Harmonia axyridis (Coleoptera: Coccinellidae). European Journal of Entomology 99:5–9. Wells JD & Henderson G (1993) Fire ant predation on native and introduced subterranean termites in the laboratory: effect of high soldier number in Coptotermes formosanus. Ecological Entomology 18:270–274. West-Eberhard MJ (2003) Developmental plasticity and evolution. Oxford University Press Inc., New York. Wharton DA (1999) Parasites and low temperatures. Parasitology 119(Suppl):S7–S17. Wilkinson PR (1970) A preliminary note on predation on free- living engorged female Rocky Mountain wood ticks. Journal of Medical Entomology 7:493–496. Williamson M (1996) Biological invasions. Chapman & Hall, London. Williamson M & Fitter A (1996) The varying success of invaders. Ecology 77:1661–1666. Wilson EE, Mullen LM & Holway DA (2009) Life history plasticity magnifies the ecological effects of a social wasp invasion. Proceedings of the National Academy of Sciences USA 106:12809–12813. Wilson PW, Heneghan AF & Haymet ADJ (2003) Ice nucleation in nature: supercooling point (SCP) measurements and the role of heterogeneous nucleation. Cryobiology 46:88–98.

84 85

ORIGINAL PAPERS

Northward invasion of the parasitic deer ked (Lipoptena cervi), is there geographical variation in pupal size and development duration?

Sirpa Kaunisto, Laura Härkönen, Pekka Niemelä, Heikki Roininen and Hannu Ylönen

Parasitology 138:354–363, 2011

Reprinted with the kind permission of the Cambridge University Press

ORIGINAL PAPERS

Northward invasion of the parasitic deer ked (Lipoptena cervi), is there geographical variation in pupal size and development duration?

Sirpa Kaunisto, Laura Härkönen, Pekka Niemelä, Heikki Roininen and Hannu Ylönen

Parasitology 138:354–363, 2011

Reprinted with the kind permission of the Cambridge University Press

354 Northward invasion of the parasitic deer ked (Lipoptena cervi), is there geographical variation in pupal size and development duration?

SIRPA KAUNISTO1*, LAURA HÄRKÖNEN2, PEKKA NIEMELÄ1,3, HEIKKI ROININEN1 and HANNU YLÖNEN4 1 University of Eastern Finland, Faculty of Science and Forestry, Department of Biology, P.O. Box 111, FI-80101 Joensuu, Finland 2 University of Oulu, Department of Biology, P.O. Box 3000, FI-90014 Oulu, Finland 3 University of Turku, Department of Biology, FI-20014 Turku, Finland 4 University of Jyväskylä, Department of Biological and Environmental Science, Konnevesi Research Station, P.O. Box 35, FI-40014 Jyväskylä, Finland

(Received 13 March 2010; revised 21 June and 26 August 2010; accepted 27 August 2010; first published online 30 September 2010)

SUMMARY

The deer ked (Lipoptena cervi) is a common ectoparasite of cervids. During the last decades the species has rapidly invaded in northern Europe, especially in Finland, towards the north and increased its prevalence on the moose population. Consequently, during this rapid invasion the deer ked has faced more severe climatic conditions. We studied whether pupal size (measured as pupal weight) and pupal development duration of the deer ked varies along historical invasion zones and temperature zones towards north in Finland. Moreover, we explored possible size- and gender-dependent variation in pupal development duration. We divided wild-collected pupae in respect to their origin in two ways: (1) temperature zones (from south-west to colder north-east) and (2) invasion history (from early to late establishment). We reared pupae in the controlled laboratory conditions in identical temperature and light conditions. Pupal size decreased towards north and the smaller pupae developed faster. However, the results do not show diﬀerences in pupal size or developmental characteristics between the invasion zones. This supports the idea of rapid developmental plasticity of the deer ked and that not the invasion history but the current temperature regime determines the life history of the deer ked when invading towards a colder environment.

Key words: ectoparasite, cervid, invasive species, Hippoboscidae, pupa, adult emergence, the converse Bergmann’s rule.

INTRODUCTION Present, 1990). This pattern has the same underlying mechanism than converse Bergmann’s rule, but Variable patterns of body size in relation to latitude specifically refers to the genetic response involved are known to occur in insects. Opposing intraspecific in compensation to seasonal limitations at higher clines in body size and development time have latitudes that results in faster growth compared been documented in ectothermic insects. First, the to their low-latitude conspecifics. Observed field Bergmann’s rule, states that body size increases with patterns can be also due to complex interactions, latitude (Blanckenhorn and Demont, 2004; Chown e.g. between Bergmann and converse Bergmann and Gaston, 2010). The mechanism behind the trends (Blanckenhorn and Demont, 2004). However, Bergmann’s rule remains unclear, although it is geographical patterns related to phenotypic charac- commonly agreed that Bergmann’s rule seems to be teristics may be essential for invasion capacity effected by temperature per se (Atkinson and Sibly, and dispersion success of various species, including 1997). On the contrary, the converse Bergmann’s parasitic insects. The present work provides new rule, mediated by season length, predicts that information about developmental and emergence developmental rate and body size decreases towards characteristics in relation to latitude and body size the poles (Masaki, 1978; Nylin and Svärd, 1991; in an invasive ectoparasite. Telfer and Hassall, 1999). In addition, a third Insects that have short generation times in relation observed pattern, called counter-gradient seems to to the length of the growing season (at least more exist (Levinton and Monahan, 1983; Conover and than 2–3 generations per season), but which inhabit temporal habitats like rotting fruit or fungi, could be expected to increase their body size and decrease * Corresponding author: University of Eastern Finland, Faculty of Science and Forestry, Department of Biology, development time with latitude (Chown and Gaston, P.O. Box 111, FI-80101 Joensuu, Finland. Fax: 1999). In contrast, insects with a generation length +358 13 251 3590. E-mail: [email protected].fi more similar to the length of the season can instead

Parasitology (2011), 138, 354–363. © Cambridge University Press 2010 doi:10.1017/S0031182010001332 Geographical variation in the deer ked 355

(a) (b)

Fig. 1. (a) Invasion zones (=our study zones): (1) the south-eastern invasion zone (invasion years 1964–1975), (2) the middle invasion zone (years 1980–1990) and (3) the northernmost invasion zone (years 1990–2007), collection localities and the rough estimation of the northernmost limit of the current distribution area of the deer ked (year 2009) (according to Kaitala et al. 2009 and Välimäki et al. 2010). Collection localities are marked by symbols: triangles (northernmost localities, n=14), circles (middle localities, n=18) and stars (south-eastern localities, n=14). (b) Temperature zones: (1) southern (hemiboreal), (2) central (southern boreal) and (3) northern (middle boreal) and the rough estimation of the northernmost limit (dashed line) of the current distribution area of the deer ked (Kaitala et al. 2009). Collection localities are marked using the symbols: triangles (northern localities, n=15), circles (middle localities, n=14) and stars (south-eastern localities, n=17). Values are annual average temperatures of 30-year average. (Figure has been modified according to the Finnish Meteorological Institute.) be expected to be constrained by season length and autumn and winter. After overwintering in the pupal decrease their mass at high latitudes and altitudes stage, development takes several months and adults (Blanckenhorn and Demont, 2004). emerge in the following late summer and autumn The deer ked, Lipoptena cervi (Diptera; (review by Haarløv, 1964). Thus, the pupal stage Hippoboscidae), is a haematophagous ectoparasite is one of the key developmental phases of the life which exploits several host species from the cervids cycle of the deer ked and likely the most sensitive to (Hackman et al. 1983; Kadulski, 1996). The deer ked extrinsic mortality factors (Bequaert, 1954; Haarløv, occasionally attacks on humans and consequently the 1964; Lehane, 2005). deer ked has drawn strong public attention during In this study, we explore possible differences in recent years in Finland because of relatively rapid pupal characteristics along (1) geographical tempera- increase in abundance and dispersion into new areas ture zones and (2) historical invasion zones of the deer towards the north (Fig. 1a). During the last 5 decades ked. We wanted to study whether the northward the deer ked expanded its range almost 1000 km spreading deer ked already shows plastic and/or rapid northward (Kaitala et al. 2009; Välimäki et al. 2010). evolutionary changes in pupal characteristics be- Until today there is no detailed knowledge on mech- tween populations after facing the new environment. anisms that have enabled the rapid invasion of the The principal question is whether the size of the pupa deer ked into the northern conditions and almost (measured as pupal weight), varies between different nothing is known about possible plastic life-history temperature or invasion zones towards north in traits of the deer ked (see Härkönen et al. 2010). Finland. To study pupal development characteristics Pupae of the deer ked drop passively from the host and variables associated with the development suc- onto the ground or into the snow during the late cess (e.g. pupal size, completion of metamorphosis, Sirpa Kaunisto and others 356 proportion of eclosed pupae and development dura- (Capreolus capreolus), and, to a lesser extent, the tion) in relation to the temperature gradient and fallow deer (Dama dama) (Haarløv, 1964). The main invasion zones we measured wild-collected deer ked host of the deer ked in Fennoscandia and in Finland is pupae divided in respect to their origin and reared the moose (Alces alces) (Paakkonen et al. 2010). The them in identical temperature and light conditions. deer ked is also able to use the wild forest reindeer We measured also the size of the emerged adults. (Rangifer tarandus fennicus) (Kaunisto et al. 2009) Throughout this article, we use also the comparable with lower prevalence, and occasionally the semi- term, eclosion, which refers to adult emergence from domestic reindeer as breeding hosts (Kynkäänniemi the pupal case (see e.g. Qiu and Hardin, 1996; Watari, et al. 2010). The white-tailed deer (Odocoileus 2002). virginianus) is also among the breeding hosts of the We hypothesized that the most northern pupae are deer ked in North America, but the prevalence of smaller than the southern pupae regardless of a infection seems relatively low (Matsumoto et al. relatively small age difference between populations. 2008). Evidence of the white-tailed deer as a breeding We based this assumption, following the converse host in Finland is inadequate so far (Välimäki et al. Bergmann’s rule, on the relative long generation time 2010). Some individuals of this invasive ectoparasite of the deer ked which could be constrained by season can make a mistake and attack also on humans when length (see Blanckenhorn and Demont, 2004). The seeking breeding host (Kortet et al. 2010). other main question considered the pupal develop- Deer ked abundance on an individual host may be ment duration and whether it varies between indi- very high, even up to above 16000 flies have been viduals originating from 3 different latitudinal zones, noted (Vikøren et al. 2008; Paakkonen et al. 2010). or whether pupae express synchronized emergence Taking the high infestation intensity into account, characteristics, when reared in similar temperature deer keds may have negative and detrimental effects and light conditions regardless of the different origins on their hosts. As the controlled experimental deer of the pupae. We hypothesized that the pupal de- ked infections demonstrated, deer keds can lower the velopment duration would be shorter among indi- physical condition of semi-domesticated reindeer and viduals originating from the northern latitudes. Since cause short-term histological, physiological, and invasion history is relatively short, we did not have behavioural changes (Kynkäänniemi et al. 2010). In specific predictions for possible variation between Norway and Sweden, an epizootic of alopecia was invasion zones. associated with massive deer ked infestation in moose (Statens Veterinärmedicinska Anstalt, 2007, 2008; Vikøren et al. 2008). In addition, deer ked is a MATERIALS AND METHODS nuisance and an obstacle to traditional human out- door activity (see Kortet et al. 2010). The deer ked Natural history of the deer ked may cause serious health problems and symptoms, The original distribution of the deer ked covers including chronic deer ked dermatitis (Rantanen Central and Eastern Europe, some parts of Northern et al. 1982) and occupational allergic rhinoconjuncti- Europe and Siberia, northern China, northern Africa vitis in humans (Laukkanen et al. 2005). The species (Algeria) but it has also been introduced to North is also a potential vector of various diseases (Ivanov, America (e.g. Maa, 1969; Dehio et al. 2004). In 1974; Rantanen et al. 1982; Dehio et al. 2004; Finland, the deer ked has rapidly spread towards the Matsumoto et al. 2008). north during the last decades (Fig. 1a). The deer ked The annual life cycle of the deer ked includes the population was established in South Eastern Finland 4 following phases; (1) emergence of a new adult in the early 1960s. The Finnish population has generation in autumn, (2) short flying period and host expanded its range at an average rate of 11 km/year colonization, (3) adult life on the host and pupapar- toward the north during the period 1960–2009 and ous pupal production and (4) pupal stage of variable 11 km toward the west during the period 1960–1990, length on the ground before the synchronized when the west coast was colonized and no further emergence of new adults (Haarløv, 1964; Hackman westward expansion was possible (Välimäki et al. et al. 1983). Adult deer keds emerge and seek hosts in 2010). At present, the limit of the distribution area of Finland from late summer to the end of autumn the deer ked in Finland follows and even crosses (Hackman, 1977). Almost immediately after finding the southernmost areas of the reindeer husbandry a host, both males and females drop their wings (Kaitala et al. 2009; Välimäki et al. 2010). Hence the (Bequaert, 1953; Hackman et al. 1983) and start to Finnish south-eastern population is approximately suck blood and interstitial fluid from the suitable 45 years old. The age difference between south- hosts (Ivanov, 1974). The adults spend wintertime eastern and northernmost populations, used in this and the rest of their lives on the same host (Bequaert, study, is roughly estimated as 15 years. 1953; Haarløv, 1964). It is not known for how long Original Central European breeding hosts (sensu adults can live on the host, but deer ked females Bequaert, 1953) supporting reproduction of the deer produce viviparously pupae one by one, at least from ked are the red deer (Cervus elaphus), the roe deer the autumn until following late winter. Geographical variation in the deer ked 357

Pupae collection and storage conditions Pupal size measurements Volunteers collected deer ked pupae from all over its Before measuring the fresh mass of pupae, we kept current distribution area in Finland. They collected them on a tray in a dry place for 2 h so that any pupae between 9 February and 19 March 2007. possible additional external moisture evaporated. During winter, the deer ked pupae drop off the We measured the pupal mass in March when we parasitized hosts onto the snow cover and pupae can supposed all pupae to be in the same overwintering most probably be seen on bedding sites of the moose stage. We weighed altogether 395 pupae by steelyard (see Kaunisto et al. 2009). We asked the volunteers (Sauter AR 1014). In the temperature zone division (mostly hunters) to fill out the data forms concerning we explored 142 northern origin pupae, 121 pupae exact information about collection dates and places, from the central zone and 132 south origin pupae, as well as host species. According to filled forms the while in the invasion zone analyses 136 pupae came pupae used in this research were collected from the from the northernmost invasion zone, 145 pupae bedding sites of moose. from the middle zone and 114 pupae from the south- After collection, we asked voluntary collectors to eastern invasion zone (Table 2). keep pupae at a low temperature (ca. +3 °C, e.g. in a fridge) and in adequate moisture in a tube filled with a piece of cotton. We asked volunteers to send pupal Rearing conditions of the pupae samples via the post to the Konnevesi Research We placed all the individually weighed pupae into station as soon as possible after collection. After coded rearing containers for later identification of arrival at the research station, we stored the samples their geographical origin. Rearing containers were under identical conditions. We stored pupae in a cold translucent plastic tubes (3 cm in diameter and 5 cm room (+3 °C) until the experiment began and we long), which we covered with a small-gilled net, handled pupal samples with particular care to avoid preventing adult deer keds escaping after their emerg- warming. ences. In addition, we placed a piece of Sphagnum moss into each rearing container with one pupa. The purpose of the moss was to prevent the pupae from Geographical and invasion historical zone division becoming mouldy due to antiseptic effects and drying To study the effect of the temperature zones on pupal of the pupae by balancing moisture conditions. characteristics we divided pupae into 3 blocks Throughout the study we used similar tempera- according to their origin in relation to the temperature, light and moisture conditions in the rearing ture zones (measured as annual average temperatures) environment. We kept pupae for 2 months after and current distribution area of the deer ked in collection in a low temperature (+3 °C) at the Finland: (1) southern (hemiboreal), (2) central laboratory, assuming pupae to continue their over- (southern boreal) and (3) northern (middle boreal) wintering. After that low temperature period, we (Fig. 1b). According to this division the age moved weighed pupal samples to the climate difference between populations is more unsubstan- chamber (+17 °C) to monitor their development at tial, because the temperature and invasion zones the end of May. Hackman (1979) reported that the are not cohesive, but they crisscross (Fig. 1a and b). first adult emergence took place after 2 months when The southern population and the northern popu- brought to room temperature. Those pupae were lation are clearly situated in different temperature collected in Finland in March from the snow and zones. The annual average temperature varies be- reared the whole time in room temperature. It seems tween ca. +2·0 °C and +3·0 °C in the zone where the probable that while staying in the soil the pupae northern population is situated. In the central zone spend the most of their time in diapause (Haarløv, the average temperature varies between +3·0 °C and 1964). Because of asynchrony in dropping time of +4·0 °C and in the southern zone the annual average pupae (may vary several months between some temperature varies between ca. +4·5 °C and +5·5 °C. individuals) from the fur of a host (see Keilbach, We studied pupae from 46 localities with respect to 1966), we expected pupae to terminate diapause the temperature zones and pooled the data within depending on temperature (see e.g. Tauber et al. each temperature zone (Table 1). 1986; Härkönen et al. unpublished observations). When exploring potential differences between the We reared pupae in a long day rhythm (16 h of light invasion zones we categorized the localities above per day) from the end of March. According to again into 3 blocks according to the invasion history unpublished observations made by Härkönen and and current distribution area of the deer ked in coworkers the development of deer ked pupa is not Finland: (1) the south-eastern invasion zone (inva- light dependent. In addition, we kept the relative sion years 1964–1975), (2) the middle invasion zone humidity of the air between 70 and 90% by using an (years 1980–1990) and (3) the northernmost invasion air conditioner, and we watered the pupae once per zone (years 1990–2007) (Fig. 1a). This approach week depending on the appearance of the surround- emphasizes age differences between populations. ing moss. Sirpa Kaunisto and others 358

Table 1. Details of the 46 research localities (all pupae studied, eclosed and uneclosed pupae and number of speciﬁc gender per locality)

Locality Coordinates All pupae (n) Eclosed pupae (%) Ecl. Males (n) Ecl. Females (n) 1 Asikkala 61°13′N, 25°30′E 10 70 2 5 2 Askola 60°32′N, 25°36′E6 0 0 0 3 Elimäki 60°43′N, 26°27′E 9 88·9 3 5 4 Eurajoki 61°12′N, 21°44′E8 0 0 0 5 Halikko 60°24′N, 23°04′E 8 50 3 1 6 Hanko 59°51′N, 23°07′E 8 37·5 1 2 7 Isojoki 62°06′N, 21°57′E 10 80 4 4 8 Jokioinen 60°48′N, 23°30′E 10 90 3 6 9 Joroinen 62°11′N, 27°50′E 6 83·3 3 2 10 Joutseno 61°07′N, 28°30′E 10 50 4 1 11 Kajaani 64°14′N, 27°44′E6 0 0 0 12 Kangasniemi 61°59′N, 26°39′E 9 77·8 3 4 13 Karjaa 60°04′N, 23°40′E 7 14·3 1 0 14 Kemiö 60°10′N, 22°44′E 5 60 0 3 15 Kokkola 63°50′N, 23°08′E 10 100 8 2 16 Korpilahti 62°01′N, 25°34′E 8 50 2 2 17 Kortesjärvi 63°17′N, 23°17′E 20 75 11 4 18 Kärsämäki 63°59′N, 25°46′E 10 90 2 7 19 Köyliö 61°07′N, 22°19′E 8 12·5 0 1 20 Laitila 60°52′N, 21°42′E 8 37·5 0 3 21 Lapinlahti 63°23′N, 27°25′E7 0 0 0 22 Lappajärvi 63°13′N, 23°38′E 10 80 4 4 23 Lappeenranta 61°04′N, 28°12′E 9 55·6 5 0 24 Lappi TL 61°06′N, 21°51′E 5 20 1 0 25 Liminka 64°49′N, 25°26′E 8 62·5 3 2 26 Luhanka 61°48′N, 25°42′E 10 100 7 3 27 Luumäki 60°57′N, 27°46′E 10 90 4 5 28 Maalahti 62°57′N, 21°33′E 7 14·3 1 0 29 Mouhijärvi 61°31′N, 23°00′E 8 25 1 1 30 Muurla 60°21′N, 23°17′E 10 50 2 3 31 Mynämäki 60°41′N, 22°10′E 10 100 9 1 32 Nivala 63°56′N, 24°58′E 10 40 3 1 33 Nurmijärvi 60°28′N, 24°49′E 8 50 2 2 34 Perho 63°13′N, 24°25′E 10 0 0 0 35 Polvijärvi 62°50′N, 29°21′E 9 66·7 5 1 36 Pyhäjoki 61°13′N, 25°30′E 10 100 5 5 37 Pyhäjärvi 63°35′N, 25°58′E 10 90 4 5 38 Rantsila 64°30′N, 25°40′E7 0 0 0 39 Saarijärvi 62°42′N, 25°16′E 7 14·3 0 1 40 Siikainen 61°53′N, 21°49′E 8 25 1 1 41 Somero 60°37′N, 23°33′E 10 40 2 2 42 Sotkamo 64°07′N, 28°24′E 8 62·5 4 1 43 Teuva 62°29′N, 21°45′E 10 90 4 5 44 Tuusula 60°24′N, 25°02′E 3 66·7 1 1 45 Uusikaupunki 60°57′N, 21°21′E 10 80 4 4 46 Viljakkala 61°42′N, 23°16′E7 0 0 0

Observation of pupal development duration we measured the adult size parameters by using the stereomicroscope with a scale. We measured the head When we expected the period of adult emergence to at the widest point and the body length from the tip of begin, we performed the observations of the pupae the abdomen to the base of the haustellum. daily. After the first adult emerged, the pupae were observed a few times per day and the emergence dates of the adults were noted. After the end of the experiment, all the uneclosed pupae were opened Statistical analyses under the microscope. For data analyses we used SPSS for Windows (version 13.0). We used ANOVA for exploring the possible influence of the zone on the pupal size by Adult sex identification and size measurements selecting the zone (temperature or invasion zone) as a We identified genders of the adults by studying the factor and pupal size as a dependent variable. We sex organs under the stereomicroscope. In addition, conducted this for all pupae (uneclosed and eclosed Geographical variation in the deer ked 359

Table 2. Details of the temperature and invasion zone division analyses

Mean pupal mass Localities Pupae Ecl. pupae Males Females Survival (±S.E.) (n) (n) (n) (n) (n) %

Temperature zones South 10·469 (±0·0743) mg 17 132 66 32 34 50 Central 10·374 (±0·0894) mg 14 121 73 41 32 60·3 North 10·127 (±0·0798) mg 15 142 82 49 33 57·8 Invasion zones South-Eastern 10·414 (±0·0825) mg 14 114 69 38 31 60·5 Middle 10·337 (±0·0802) mg 18 145 67 35 32 46·2 Norhernmost 10·215 (±0·0818) mg 14 136 85 49 36 62·5 All 10·317 (±0·0472) mg 46 395 221 122 99 56

zone) and gender on the pupal development duration by selecting pupal size as a covariate, zone and gender as fixed factors and developmental rate as a dependent variable for the model. We conducted the ANCOVA model also to detect possible associations behind the adult size parameters, selecting pupal size as a covariate, zone (temperature or invasion zone) and gender as fixed factors and adult size parameter as a dependent variable. We performed separate ANCOVA models for different adult size parameters (width of the head and adult body length as dependent variables). We used the Bonferroni correction for the two independent comparisons (statistical analyses for the Fig. 2. Mean pupal mass of the deer ked including all temperature and the invasion zone divisions) and weighed pupae (uneclosed and eclosed pupae) in lowered the significance level α from 0·05 to 0·025. 3 temperature zones with a 95% confidence interval.

pupae). We did not include gender for this model, RESULTS because we knew the gender only for eclosed pupae. fl For eclosed pupae, we used the ANOVA model to The geographical pattern in size and possible in uence study the effect of the zone (temperature or invasion of the gender on the pupal size zone) and the gender on pupal size by selecting zone In the temperature zone division, pupae from the and gender as fixed factors and pupal size of eclosed 3 temperature zones differed significantly from each pupae as a dependent factor. For the temperature other in terms of pupal size (ANOVA; F2,392 =4·292, zone division analysis, we used ANOVA to detect P=0·007). The difference in pupal size between the possible differences in pupal size between eclosed and northern and the southern temperature zones was uneclosed pupae in the north and the south by statistically significant when all pupae (eclosed and selecting pupal status (value: 1=uneclosed, 2= uneclosed) were included (Tukey HSD post hoc test; eclosed) as a definitive factor and pupal size as a P=0·007). Southern origin pupae were statistically dependent factor. We conducted this test separately heavier than pupae from the northernmost tempera- among northern and southern pupae. ture zone (Fig. 2, Table 2). In this model, zone was Relating also to the pupal success, we explored a fixed factor and pupal size a dependent factor. potential differences between the zones (temperature However, when only pupae that eclosed were or invasion zone) in the proportion of pupae that included, there was no difference between the tem- eclosed (i.e. survival rate) separately for each zone perature zones in terms of pupal size when the pair by using the Chi-square test. In addition, temperature zone and the gender were fixed factors we studied potential differences between the zones and the pupal size was a dependent factor for the (temperature or invasion zone) in the proportion model (ANOVA; F2,215 =1·138, P=0·32). The gen- of pupae completing metamorphosis by using the der did not affect the pupal size (ANOVA; F1,215 = Chi-square test. 1·406, P=0·24). There was no interaction between We used one single ANCOVA model to test the zone and the gender (ANOVA; F2,215 =0·233, influence of pupal size, zone (temperature or invasion P=0·79). The difference in relation to the pupal size Sirpa Kaunisto and others 360 disappeared among eclosed pupae between the tem- (ANCOVA; F1,209 =0·140, P=0·709). There were no perature zones, likely because eclosed pupae were interactions of any kind in this ANCOVA model. heavier than uneclosed ones in the northernmost In the invasion zone division, there were no temperature zone used in this study when analysed interactions of any kind in the ANCOVA model, separately the northern and the southern populations when pupal size was a covariate, invasion zone and (ANOVA; F1,140 =10·573, P=0·001). gender were fixed factors and pupal development In the invasion zone division, pupae from the duration was a dependent factor for the model. 3 invasion zones did not significantly differ between Invasion zone had no significant influence on the zones in terms of pupal size when all pupae the pupal development duration (ANCOVA; (eclosed and uneclosed) were compared (ANOVA; F2,209 =1·412, P=0·246). Hence, the pupal develop- F2,392 =1·450, P=0·236) (Table 2). In this model, ment duration did not differ between adults from zone was a fixed factor and pupal size a dependent 3 invasion zones with different invasion times when factor. Among only the eclosed pupae, there was no reared in similar conditions. The gender also had no interaction between invasion zone and gender when effect on the pupal development duration (ANCOVA; zone and gender were fixed factors and the pupal size F1,209 =0·331, P=0·566). was a dependent factor (ANOVA; F2,215 =0·377, P=0·69). The invasion zones did not differ from Influence of the pupal size on the pupal development each other in relation to pupal size of eclosed pupae duration and adult morphology (ANOVA; F2,215 =0·124, P=0·88). The gender did not also affect the pupal size (ANOVA; F1,215 =1·623, In the temperature zone division, pupal size had a P=0·20). significant influence on pupal development duration when testing pupal size as covariate, other factors such as temperature zone and gender as fixed factors, Adult emergence and success and pupal developmental rate as a dependent factor for the model (ANCOVA; F =9·052, P=0·003). The adult emergence period began at the end of July. 1,209 The positive relationship between pupal size and The first adult emerged on 30 July 2007 and the development duration was equal, hence there were no emergence period lasted for 31 days, until 30 August. interactions, in all 3 geographical zones (ANCOVA; Only 34 out of 172 uneclosed pupae (19·8%) had F =0·120, P=0·887) and in both genders seemingly completed the metamorphosis from larvae 2,209 (ANCOVA; F =0·001, P=0·971). to adult inside the puparium but the adult emergence 1,209 There were no interactions of any kind in the did not take place for some reason and adults had died ANCOVA model, when testing pupal size as covari- inside the puparium. ate, temperature zone and gender as fixed factors and Between the temperature zones there were no adult size parameter (width of the head or body differences in the proportion of pupae completing length) as a dependent factor for the model. metamorphosis inside the puparium nor in the We found a positive association between pupal size survival rate (i.e. proportion of pupae that eclosed). and adult size parameters like width of the head Between the invasion zones, there were also no (ANCOVA; F =28·03, P<0·001) and body differences in the proportion of pupae completing 1,128 length (ANCOVA; F =23·10, P<0·001). metamorphosis inside the puparium. The survival 1,102 In the invasion zone division, when testing pupal rate was lowest in the middle zone compared to the 2 size as covariate, invasion zone and gender as fixed south-eastern (X =5·248, D.F.=1, P=0·022) and 2 factors and pupal developmental rate as a dependent northernmost (X =7·502, D.F.=1, P=0·006) inva- factor for the ANCOVA model, pupal size had also sion zone. Between the northernmost and the south- a significant influence on pupal development dura- eastern invasion zone there was no difference in terms 2 tion, indicating that smaller pupae eclose earlier of survival rate (X =0·102, D.F.=1, P=0·749). (ANCOVA; F1,209 =10·829, p=0·001). The relationship between pupal size and development duration was equal, hence there were no interactions, in all 3 Influence of the geographical origin and the gender invasion zones (ANCOVA; F =0·899, P=0·409) on the pupal development duration 2,209 nor in both genders (ANCOVA; F1,209 =0·043, Temperature zone had no significant influence on P=0·836). In addition, when pupal size was used as pupal development duration when testing pupal size covariate, invasion zone and gender as fixed factors as covariate, temperature zone and gender as fixed and adult size parameter (width of the head or body factors and pupal development duration as a depen- length) as dependent factor for the model, there was a dent factor for the model (ANCOVA; F2,209 =0·255, positive association between pupal size and adult P=0·775). Hence, the pupal development duration size parameters like width of the head (ANCOVA; did not differ between adults from the 3 temperature F1,128 =24·92, P<0·001) and body length (ANCOVA; zones when reared in similar conditions. The gender F1,102 =21·71, P<0·001). There were no interactions also had no effect on the pupal development duration of any kind in this ANCOVA model. Geographical variation in the deer ked 361

DISCUSSION real differences in season length and associated temperatures are small or missing. Our data indicate that the expansion of the deer ked’s After taking into account only the pupae that distribution area towards the north seems to have eclosed, there was no significant difference between caused plastic changes in the pupal characteristics temperature zones in terms of pupal size. This means along the temperature gradient. The invasion history, that eclosed pupae were heavier than uneclosed ones or the time of establishment of the parasite to the in the northern zone. It may suggest that our selected environment and its host specimen, did not seem to constant rearing temperature potentially interfered affect life-history characteristics of the deer ked. The with the developing process of the smallest pupae deer ked pupae from the southernmost temperature and may have hindered us to observe the adaptation zones were heavier than pupae from the northern- to local environmental conditions more closely. most temperature zones (used in this study) when In both division analyses (the temperature zone compared between temperature regions. However, and invasion zone division) we found a positive the pupal size did not vary between invasion zones in correlation between the pupal size and pupal devel- Finland. Pupal development duration did not vary opment duration. Adults that emerged from smaller between individuals having a different temperature pupae seemed to emerge earlier than adults from origin or invasion history, contrasting our expec- larger pupae under identical conditions. This sup- tations. However, the adult emergence period was ports a general expectation that smaller individuals relatively synchronized regardless of different origins contain less metabolic reserves, which indicates a (temperature and invasion history) when the indi- decreased ability to sustain a long non-feeding pupal viduals were reared under the same temperature and stage (e.g. Feder et al. 2010). There was no difference light conditions. Among the tested variables, only between the male and the female pupae in terms of pupal size was associated positively with pupal de- pupal size. In the northern temperature zone, smaller velopment duration, indicating that smaller pupae pupae with earlier eclosion could have an advantage eclosed earlier in both zone divisions. because lower summer temperatures prolong the The first result concerning the relationship of developmental period and shorten the suitable host pupal size and temperature zone was in accordance search time by several weeks. Pupal size also with our first hypothesis which predicted decreasing correlated positively with adult size parameters size towards the north (see Masaki, 1978; Mousseau, (width of the head and body length) indicating that 1997; Blanckenhorn and Demont, 2004). Thus, one larger pupae produced larger adults in all zones and in would expect a latitudinal pattern of decreasing size, both sexes when reared under identical temperature because of the seasonal constraints (i.e. shorter and conditions. However, it is not known how the short colder growing season in the north) as the converse growing season with lower temperatures in the north Bergmann’s rule states (e.g. Masaki, 1978). That could affect this association. Studies report that pupal rule has often been observed in insects and other development temperature affects adult phenotype arthropods (Mousseau, 1997). In many insects, and size in some species (e.g. Stevens, 2004). This (e.g. crickets and grasshoppers), body size appears could be an important topic in future research also to be largely genetically determined so that northern with the deer ked. According to these associative populations are smaller (e.g. Masaki, 1967). On results pupal size and, for example, adult body length the other hand, deer ked adults live and reproduce and width of the head could be used as a tool for their whole life in constant temperature inside the fur estimating the morphological differences between and, consequently, are relatively unaffected by out- deer ked populations in future investigations. door temperature. Why then should they follow the Geographical origin and sex, instead, did not affect converse Bergmann’s rule? Alternatively, the smaller pupal development duration. Despite the different body size in the north can also be a result from temperature origin and different invasion history of phenotypic plasticity when an organism’s life cycle the pupae, the adult emergence period was relatively is linked to season length (e.g. Leimar, 1996). synchronized when reared under identical tempera- Environmentally induced phenotypic plasticity ture and light conditions, which was also contrary to might explain the observed geographical variation our expectations. This synchronization of eclosion, in the deer ked pupal size between temperature zones despite the different origin, could suggest that the (cf. West-Eberhard, 2003). The mechanism behind northernmost pupae have not yet genetically adapted this could be that females in the northern latitudinal to their new environment or, perhaps the deer ked zones show plasticity in their investment to smaller had not yet faced the total limit of its distribution, pupae. In addition, the age difference between which would force differences to outcome through individuals from northernmost and southernmost adaptations. temperature zones may not be so high for adaptation Among various insects, males usually emerge due to the overlapping invasion and temperature earlier than females. This phenomenon is known zones. One possible explanation for the lack of the as a protandry, which can be understood in terms clear pattern between the invasion zones is that the of sexual selection acting on males to maximize Sirpa Kaunisto and others 362 the number of matings, or on females to increase evolution and revision of American genera and species. reproductive success by minimizing the pre- Entomologica Americana 34,1–232. reproductive period (e.g. Wiklund et al. 1996; Blanckenhorn, W. U. and Demont, M. (2004). Taylor et al. 1998). This may not be so in the case Bergmann and Converse Bergmann latitudinal clines in of the deer ked, perhaps because the host searching arthropods: two ends of a continuum? Integrative and Comparative Biology 44, 413–424. periods of adults is relative short or this type of ff Chown, S. L. and Gaston, K. J. (1999). Exploring links method was not enough to raise possible di erences between physiology and ecology at macro-scales: the role in adult emergence between genders. of respiratory metabolism in insects. Biological Reviews of To conclude, we know currently that adult deer the Cambridge Philosophical Society 74, 87–120. keds face constant and relatively safe conditions Chown, S. L. and Gaston, K. J. (2010). Body size within the host’s fur, but pupae are surrounded by variation in insects: a macroecological perspective. the very variable extrinsic mortality factors (Välimäki Biological Reviews 85, 139–169. et al. 2010). Our results about pupal size between Conover, D. O. and Present, T. M. C. (1990). temperature zones may verify the great plasticity of Countergradient variation in growth rate: Compensation this species. Thus, the deer ked likely has a notable for length of the growing season among Atlantic ff capability to continue its dispersion towards the silversides from di erent latitudes. Oecologia 83, 316–324. north into new areas. This is indeed predictable, since Dehio, C., Sauder, U. and Hiestand, R. (2004). Isolation a colder and shorter growing season in northern of Bartonella schoenbuchensis from Lipoptena cervi,a Finland may not totally constrain deer ked invasion blood-sucking arthropod causing deer ked dermatitis. according to transplant experiments conducted be- Journal of Clinical Microbiology 42, 5320–5323. yond the current range of the deer ked (see Härkönen Feder, J. L., Powell, T. H. Q., Filchak, K. and et al. 2010). In general, documented and forecast Leung, B. (2010). The diapause response of changes in the local species richness due to climate Rhagoletis pomonella to varying environmental change likely signify that geographical ranges of conditions and its significance for geographic and host species will continue to shift substantially polewards plant-related adaptation. Entomologia Experimentalis (e.g. Beaumont and Hughes, 2002; Parmesan, 2006; et Applicata 136, 31–44. Vanhanen et al. 2007). Finnish Meteorological Institute (2010). http://www. fmi.fi/saa/tilastot_99.html#1. Haarløv, N. (1964). Life cycle and distribution pattern of Lipoptena cervi (L.) (Dipt., Hippobosc.) on Danish ACKNOWLEDGEMENTS deer. Oikos 15, 93–129. We want to thank all the hundreds of volunteer pupa Hackman, W. (1977). Hirven täikärpänen ja sen fi ff collectors for their help in the eld and sta of Konnevesi levittäytyminen Suomeen. Luonnon Tutkija 81, 75–77. Research station during the laboratory period of this Hackman, W. (1979). Älglusflugans, Lipoptena cervi, study. We also thank Raine Kortet and two very careful invandringshistoria i Finland. Entomologisk Tidsskrift anonymous referees for providing useful comments to this , 208–210. manuscript. 100 Hackman, W., Rantanen, T. and Vuojolahti, P. (1983). Immigration of Lipoptena cervi (Diptera, Hippoboscidae) in Finland, with notes on its biology FINANCIAL SUPPORT and medical significance. Notulae Entomologicae 63, This study was partly funded by the Vanamo, Finnish 53–59. Biologists’ Association (to S.K), Ella and Georg Ehrnrooth Härkönen, L., Härkönen, S., Kaitala, A., Kaunisto, S., Foundation (to S.K), Pro Societas Flora et Fauna Fennica Kortet, R., Laaksonen, S. and Ylönen, H. (2010). (to S.K.) and the Finnish Ministry of Agriculture and Predicting range expansion of an ectoparasite – the effect Forestry (to S.K.). of summer temperatures on deer ked (Lipoptena cervi, Diptera: Hippoboscidae) performance along a latitudinal gradient. Ecography (in the Press). REFERENCES doi:10.1111/j.1600–0587.2009.05890.x Atkinson, D. and Sibly, R. M. (1997). Why are organisms Ivanov, V. I. (1974). On the damage done by Lipoptena usually bigger in colder environments? Making sense of a cervi L. (Diptera, Hippoboscidae) in Byelorussia. life history puzzle. Trends in Ecology & Evolution 12, Parazitologiya 8, 252–253. 235–239. Kadulski, S. (1996). Ectoparasites of Cervidae in Beaumont, L. J. and Hughes, L. (2002). Potential north-east Poland. Acta Parasitologica 41, 204–210. changes in the distributions of latitudinally restricted Kaitala, A., Kortet, R., Härkönen, S., Laaksonen, S., Australian butterfly species in response to climate Härkönen, L., Kaunisto, S. and Ylönen, H. (2009). change. Global Change Biology 8, 954–971. Deer ked, an ectoparasite of moose in Finland: A brief Bequaert, J. C. (1953). The Hippoboscidae or louse-flies review of its biology and invasion. Alces 45, 85–88. (Diptera) of mammals and birds. Part I. Structure, Kaunisto, S., Kortet, R., Härkönen, L., Härkönen, S., physiology and natural history. Entomologica Americana Ylönen, H. and Laaksonen, S. (2009). New bedding 32,1–209. site examination-based method to analyse deer ked Bequaert, J. C. (1954). The Hippoboscidae or louse-Flies (Lipoptena cervi) infection in cervids. Parasitology (Diptera) of mammals and birds. Part II. Taxonomy, Research 104, 919–925. Geographical variation in the deer ked 363

Keilbach, R. (1966). Die tierischen Schädlinge Parmesan, C. (2006). Ecological and evolutionary Mitteleuropas. Gustav Fischer, Jena, Germany. responses to recent climate change. Annual Review of Kortet, R., Kaunisto, S., Härkönen, L., Hokkanen, P., Ecology and Systematics 37, 637–669. Härkönen, S., Kaitala, A. and Ylönen, H. (2010). Qiu, J. and Hardin, P. E. (1996). Developmental state and Experiments on the ectoparasitic deer ked that often the circadian clock interact to influence the timing of attacks humans; preferences for body parts, colour and eclosion in Drosophila melanogaster. Journal of Biological temperature. Bulletin of Entomological Research 100, Rhythms 11, 75–86. 279–285. doi:10.1017/S0007485309990277. Rantanen, T., Reunala, T., Vuojolahti, P. and Kynkäänniemi, S. M., Kortet, R., Härkönen, L., Hackman, W. (1982). Persistent pruritic papules Kaitala, A., Paakkonen, T., Mustonen, A. M., from deer ked bites. Acta Dermato-Venereologica 62, Nieminen, P., Härkönen, S., Ylönen, H. and 307–311. Laaksonen, S. (2010). Threat of an invasive parasitic fly, Statens Veterinärmedicinska Anstalt (2007). the deer ked (Lipoptena cervi), to the reindeer Sjukdomsläget hos vilt I Sverige 2006. National (Rangifer tarandus tarandus): experimental infection Veterinary Institute’s Report Series No. 1. National and treatment. Annales Zoologici Fennici 47, 28–36. Veterinary Institute, Uppsala, Sweden. Laukkanen, A., Ruoppi, P. and Mäkinen-Kiljunen, S. Statens Veterinärmedicinska Anstalt (2008). (2005). Deer ked induced occupational allergic Sjukdomsläget hos vilt I Sverige 2007. National rhinoconjunctivitis. Annals of Allergy, Asthma and Veterinary Institute’s Report Series No. 6. National Immunology 94, 604–608. Veterinary Institute, Uppsala, Sweden. Lehane, M. (2005). The Biology of Blood-Sucking in Stevens, D. J. (2004). Pupal development temperature Insects. Cambridge University Press, Cambridge, UK. alters adult phenotype in the speckled wood butterfly, II Leimar, O. (1996). Life history plasticity: influence of Pararge aegeria. Journal of Thermal Biology 29, 205–210. photoperiod on growth and development in the common Tauber, M. J., Tauber, C. A. and Masaki, S. (1986). blue butterfly. Oikos 76, 228–234. Seasonal Adaptations of Insects. Oxford University Press, High cold tolerance through four seasons and all free-living Levinton, J. S. and Monahan, R. K. (1983). The New York, USA. stages in an ectoparasite. latitudinal compensation hypothesis: Growth data and a Taylor, B. W., Anderson, C. R. and Peckarsky, B. L. model of latitudinal growth differentiation based upon (1998). Effects of size at metamorphosis on stonefly energy budgets. II. Intraspecific comparisons between fecundity, longevity and reproductive success. Oecologia subspecies of Ophryotrocha puerilis. The Biological 114, 494–502. Laura Härkönen, Arja Kaitala, Sirpa Kaunisto and Bulletin 165, 699–707. Telfer, M. G. and Hassall, M. (1999). Ecotypic Maa, T. C. (1969). A revised checklist and concise host differentiation in the grasshopper Chorthippus brunneus: Tapani Repo index of Hippoboscidae (Diptera). Pacific Insects life history varies in relation to climate. Oecologia 121, Monograph 20, 261–299. 245–254. Masaki, S. (1967). Geographic variation and climatic Vanhanen, H., Veteli, T. O., Päivinen, S., Parasitology 139: 926–933, 2012 adaptation in a field cricket (Orthopetra: Gryllidae). Kellomäki, S. and Niemelä, P. (2007). Climate Evolution 21, 725–741. change and range shifts in two insect defoliators: gypsy Masaki, S. (1978). Seasonal and latitudinal adaptations moth and nun moth – a model study. Silva Fennica 41, Reprinted with the kind permission of the Cambridge in the life cycles of crickets. In Evolution of Insect 621–638. Migration and Diapause (ed. Dingle, H.), pp. 72–100. Vikøren, T., Ytrehus, B. and Handeland, K. (2008). University Press Springer-Verlag, New York, USA. Helseovervåkingsprogrammet for hjortevilt (HOP), Matsumoto, K., Berrada, Z. L., Klinger, E., Årsrapport for 2006 og 2007 – National Veterinary Goethert, H. K. and Telford, S. R. (2008). Institute’s Report Series 19/2008. National Veterinary Molecular detection of Barthonella schoenbuchensis Institute, Oslo, Norway. from ectoparasites of deer in Massachusetts. Vector Välimäki, P., Härkönen, L., Härkönen, S., Kaitala, A., Borne and Zoonotic Diseases 8, 549–554. Kortet, R., Madslien, K., Malmsten, J., Redford, L., Mcinnis, D. O., Wendel, L. E. and Whitten, C. J. (1983). Ylönen, H. and Ytherus, B. (2010). Fennoscandian Directional selection and heritability for pupal weight distribution of an important parasite of cervids, the deer in the screwworm, Cochliomyia hominivorax (Diptera: ked (Lipoptena cervi), revisited. Parasitology Research Calliphoridae). Annals of the Entomological Society of 107, 117–125. doi: 10.1007/s00436-010-1845-7. America 76, 30–36. Watari, Y. (2002). Comparison of the circadian eclosion Mousseau, T. A. (1997). Ectotherms follow the converse rhythm between non-diapause and diapause pupae in the to Bergmann’s Rule. Evolution 51, 630–632. onion fly, Delia antiqua: the effect of thermoperiod. Nylin, S. and Svärd, L. (1991). Latitudinal patterns in Journal of Insect Physiology 48, 881–886. the size of European butterflies. Holarctic Ecology 14, West-Eberhard, M. J. (2003). Developmental Plasticity 192–202. and Evolution. Oxford University Press, New York, Paakkonen, T., Mustonen, A. M., Roininen, H., USA. Niemelä, P., Ruusila, V. and Nieminen, P. (2010). Wiklund, C., Lindfors, V. and Forsberg, J. (1996). Early Parasitism of the deer ked, Lipoptena cervi, on the moose, male emergence and reproductive phenology of the adult Alces alces, in eastern Finland. Medical and Veterinary overwintering butterfly Gonepteryx rhamni in Sweden. Entomology (in the Press). Oikos 75, 227–240.

High cold tolerance through four seasons and all free-living stages in an ectoparasite.

Laura Härkönen, Arja Kaitala, Sirpa Kaunisto and Tapani Repo

Parasitology 139: 926–933, 2012

Reprinted with the kind permission of the Cambridge University Press

926 High cold tolerance through four seasons and all free-living stages in an ectoparasite

LAURA HÄRKÖNEN1*, ARJA KAITALA1, SIRPA KAUNISTO2 and TAPANI REPO3 1 Department of Biology, University of Oulu, PO Box 3000, FI-90014 Oulu, Finland 2 Department of Biology, University of Eastern Finland, PO Box 111, FI-80101 Joensuu, Finland 3 Finnish Forest Research Institute, PO Box 68, FI-80101 Joensuu, Finland

(Received 22 November 2011; revised 14 December 2011; accepted 15 December 2011; first published online 8 February 2012)

SUMMARY

Off-host stages of temperate parasites must cope with low temperatures. Cold tolerance is often highest in winter, as a result of diapause and cold acclimation, and low during the active summer stages. In some blood-feeding ectoparasites, offspring provisioning determines cold tolerance through all the non-feeding, off-host stages. Large size increases survival in the cold, but so far seasonal variation in within-female offspring size has not been associated with offspring cold tolerance. The deer ked (Lipoptena cervi) reproduces on cervids from autumn to spring. Newborn pupae drop off the host, facing frosts without any acclimation. We examined cold tolerance through 4 seasons and from birth to adulthood by means of short- and long- term frost exposure. We expected females to produce more tolerant offspring in winter than in spring. Large spring pupae survived prolonged frosts better than did small winter pupae. Thus more tolerant offspring were not produced when the temperature outside the host is at its lowest. Unexpectedly, the freezing points were 20 °C or below all year round. We − showed that high cold tolerance is possible without acclimation regardless of life stage, which presumably correlates with other survival characteristics, such as the starvation resistance of free-living ectoparasites.

Key words: cold-hardiness, Hippoboscidae, moose louse-ﬂy, oﬀ-host survival, pupal size, seasonality, supercooling, viviparity.

INTRODUCTION Leaving the warm host during the winter may be risky for ectoparasites. Temperate insects enter The survival of temperate ectoparasites depends on diapause in order to escape low temperatures well in their ability to tolerate low seasonal temperatures advance of severe conditions arising (Tauber et al. outside the host (Tinsley, 1999; Wharton, 1999; 1986; Danks, 1987; Leather et al. 1993; Bale and Lehane, 2005). Endothermic hosts guarantee con- Hayward, 2010). Diapause, per se, increases tolerance stant resources and a safe environment, and some to environmental extremes during winter, but dia- blood-feeding ectoparasites therefore remain repro- pause-mediated cold-hardening often requires weeks ductively active throughout the year (Haarløv, 1964; of acclimation: decreasing temperatures stimulate the Krasnov et al. 2002). An extended reproductive production of cryoprotectants and thus, tolerance period means that females will age, and the offspring often is at its highest in mid-winter, decreasing that are shed from the host will encounter very towards spring (Tauber et al. 1986; Bale and different seasonal conditions and a risk of freezing. Hayward, 2010). Freeze-intolerant insects reduce Maternal age often correlates reliably with seasonal the risk of lethal freezing by lowering their freezing environmental changes (Mousseau and Dingle, point, i.e. by supercooling (Leather et al. 1993). The 1991). Mothers may therefore have good ability to period during which an insect can stay in a super- predict the environment that the young will face, and cooled state varies with the time and severity of cold produce offspring with a phenotype that best fits exposure (Knight et al. 1986). Large individuals with those conditions (McGinley et al. 1987; Schultz, larger energy reserves (especially body fat) often have 1991; Roff, 1992; Plaistow et al. 2007). For example, a significant advantage for survival at prolonged low large offspring are often produced in cold or temperatures (Colinet et al. 2006; Matsuo, 2006). For otherwise poor environments (Fischer et al. 2003). example, in large individuals the formation of lethal So far there is no empirical evidence as to whether body ice may be slower than in small individuals ectoparasites produce offspring whose quality is (Ansart and Vernon, 2004). However, the opposite modified according to predictable seasonal variation pattern has also been reported: larger individuals may in temperature outside the host. have higher supercooling points and hence they are less tolerant to cold (e.g. Hahn et al. 2008). * Corresponding author: University of Oulu, Department Resistance to low temperatures varies between life- of Biology, PO Box 3000, FI-90014 Oulu, Finland. Tel: +358 8 553 1221. Fax: +358 8 553 1061. E-mail: laura. history stages and according to season (Tauber et al. harkonen@oulu.fi 1986; Heinrich, 1999). This is because preparing for

Parasitology (2012), 139, 926–933. © Cambridge University Press 2012 doi:10.1017/S0031182012000091 Cold tolerance of an ectoparasite 927 and maintaining high cold tolerance is energetically their wings. The adults stay attached to the same expensive (Leather et al. 1993; Colinet et al. 2006). cervid host where they feed and reproduce. The total Therefore, low tolerance to frost is often found in number of offspring is approximately 20–40 per deer active stages (i.e. during growth, development and ked female (Popov, 1965). Since they are produced flight period), when only favourable seasonal temp- one at a time, the birth time and age of offspring differ eratures are experienced (Wharton, 1999). by several months (Haarløv, 1964). The hosts may Some blood-feeding ectoparasites feed only on move for tens, or even hundreds of kilometres during host blood as adults, whereas offspring outside the the fly’s reproductive life span (Heikkinen, 2000). host depend totally on maternally derived resources Thus, the site for pupal diapause and development is (Langley and Clutton-Brock, 1998). These energy determined by host movements. reserves determine the offspring’s resistance to The non-feeding, off-host period covers diapause starvation during non-feeding periods but also their in winter, pupal development (i.e. adult metamor- resistance to environmental adversities (Roff, 1992; phosis) during summer and the host search period of Colinet et al. 2006; Piiroinen et al. 2011). Thus adults in autumn. The deer ked pupae overwinter in a tolerance to low temperatures is likely to decrease diapause, which is terminated by high temperature with age and decreasing energy reserves as the off- (Härkönen and Kaitala, unpublished observations). host period progresses (Colinet et al. 2006; Bowler They diapause at a stage when adult metamorphosis and Terblanche, 2008). has not yet begun. Pupae overwinter below or within We used the ectoparasitic deer ked (Lipoptena cervi a snow blanket that creates a microclimate, where L., Hippoboscidae) to test whether females would temperatures may remain around zero despite low air produce offspring with higher cold tolerance when temperatures (Leather et al. 1993). Pupal develop- they are most likely to experience a high risk of ment initiates in late spring when temperatures freezing outside the host. The deer ked lives and feeds increase high enough and adults emerge after on temperate cervids, mainly moose (Alces alces), approximately 3 months of development. Adults as adults (Kaunisto et al. 2009; Välimäki et al. 2011). may survive more than 2 months without a host Off-host stages do not feed but rely totally on (Härkönen et al. unpublished observations). maternal provisioning. Viviparous females give birth to immediately pupating larvae, one at a time, during their reproductive period extending from autumn until summer. The newborn pupae are shed Material collection and preparation off the warm host, experiencing frosts of varying degree and duration without cold acclimation. After We had no previous knowledge about the limits of diapause, developing pupae and adults are not cold tolerance in this species. In order to study the normally exposed to harsh frosts. time and severity of frosts that the pupae survive, we This study focuses on factors influencing the cold needed to gather the largest sample sizes for as many tolerance (i.e. lack of long cold acclimation, pupal size treatments as possible. For this part of the study, we and birth time) of deer ked pupae at diapause and used pupae collected all over the current distribution during active off-host stages. We determine the cold area in Finland in March. The more detailed tolerance as cold-hardening capacity (SCP) and as the collection methods and places are described by length and severity of the frost periods that pupae Härkönen et al. (2010) and Kaunisto et al. (2011). survive, using manipulative short- or long-term frost To study cold tolerance in detail, we collected deer exposures. We expect cold tolerance to be highest in ked pupae from moose bedding sites in the com- diapausing pupae that experience harsh frosts soon mune of Siikalatva, Central Finland (64°30′20″N, after birth. We conduct tests to establish whether cold 25°39′00″E; 60 m AMSL). The collection dates were tolerance is higher in newborn pupae produced chosen to be during mild weather and several days before or during winter than in spring, and whether after periods of harsh frosts, and only pupae from it increases with larger size at birth. We also test recently used bedding sites (< 24 h) were used for whether cold tolerance follows seasonal temperatures experiments. Thus the pupae experienced only mild and decreases with the length of the non-feeding temperatures (ca.0–5 °C) before the freezing tests period so that developing pupae and adults have were started. In autumn (October) we collected reduced cold-hardening capacity during summer and pupae from 3 moose pelts immediately after the autumn. moose were killed. This was done because moose bedding sites and the deer ked pupae on them are difficult to find without snow. The moose used in this MATERIALS AND METHODS study originated from the same area and were killed during the moose hunting season. After collection, Study species we measured the pupal mass (used as a measure of Deer keds emerge and fly from late summer to late offspring size) with a precision balance (Mettler autumn. After accepting a suitable host, they cut off Toledo MT 5, accuracy of 0·001 mg). We kept the Laura Härkönen and others 928 pupae in a cold room (+5 °C) until the experiments started within a few days after collection.

Cold tolerance of diapausing pupae We examined ability to tolerate cold in newborn diapausing pupae by determining their cold hardiness in relation to their birth time and size. Cold hardiness is defined as the capacity of a species to survive short- or long-term exposure to low temperatures. The supercooling point is considered to be an indicator of lower lethal temperatures during short-term cold exposure in freeze-avoiding species. By contrast, freeze-tolerant insects can survive in temperatures below their SCP (Salt, 1961). Fig. 1. The supercooling points of acclimated (at 5 °C) − Therefore, supercooling capacity should not be and non-acclimated (at +5 °C) diapausing deer ked pupae. considered the only measure of cold-hardiness in a given species; the ability of the species to survive prolonged periods of cold exposure must be evaluated diapausing pupae in relation to their birth time and (Sømme, 1996; Bale and Hayward, 2010). size, we collected pupae at intervals of approximately 6 weeks from autumn to spring on October 19 (n=40), January 14 (n=37), February 26 (n =30) and Supercooling point measurements April 3 (n=36). We measured their SCPs a few days after collection (the measurements of 1 month’s We tested the freezing point of the newborn diapaus- samples taking 4 days altogether). ing pupae after dropping off the host. Lethal freezing of body fluids takes place at the supercooling point (SCP) (Wilson et al. 2003). The degree of super- Tolerance to prolonged frost cooling, i.e. the temperature just before the onset of ice crystal formation and consequent release of heat, can We evaluated the cold tolerance of diapausing be measured by differential thermal analysis (DTA). pupae in nature by testing their ability to survive The DTA was conducted with a custom-designed prolonged cold exposures. For the frost treatments, device, which consisted of 4 aluminum modules, each we placed the pupae individually in Eppendorf tubes module having 3 spaces for the samples (Räisänen and sank them in boxes filled with sand to stabilize et al. 2006). The differential temperature between the possible temperature fluctuations in the freezer sample and the reference point was measured for 12 rooms. To control natural mortality as compared samples at a time. The differential temperature was to frost kills, an additional control group (without measured using NiCr/Ni thermocouples (diameter frost exposure) was kept in the cold room (at +5 °C). 0·25 mm) and the temperature of the aluminum Four days after the treatments, we moved all the block using a Pt-100 thermistor. The modules pupae to a climate chamber (+20 °C, 60% RH and were in a programmable freezing chamber (ARC a photoperiod of 19 h/5 h for day/night) in order 300/–55/ +20, Arctest, Finland). The temperature to detect survival (assessed as adult emergence changes during gradual cooling of each sample were success). recorded using DASYlab 8.0 software. First, we estimated the time and severity of frosts Cold acclimation at subzero temperatures may that the pupae survive, using a total of 450 (n=90/ affect the SCP (e.g. Koštál and Šimek, 1995). We treatment) pupae collected from all over Finland. In collected 72 diapausing pupae (n=36/treatment) in addition to the control treatment, we exposed them to March and tested the effect of cold acclimation on the frosts of 5 °C, 15 °C, 20 °C or 25 °C for 4 days SCP. The pupae were kept at either +5 °C or 5 °C (96 h). Second,− in− order to− examine− the cold tolerance for 24 h before DTA. Correspondingly, the DTA− was of newborn pupae in detail, i.e. according to birth conducted with 2 initial temperatures for the time and size, we collected altogether 216 pupae exposure programme (+5 °C or 5 °C). The acclim- on February 26 and April 3 (n=36/treatment). ation test was also used as a control− test for the cooling Hereafter, we will refer to these dates as ‘winter’ protocol for further SCP measurements. The SCP of and ‘spring’, respectively. We divided the pupae diapausing pupae was not lowered by acclimation at equally into treatments (incl. control) according to subzero temperature (P>0·05, Fig. 1). Hence, for the size range, so that within either season their mean later tests, we used an initial temperature of +5 °C, a mass or variance did not vary between the treatments 1 cooling rate of 5 °C hour− and a target temperature (ANOVAs: Winter F2,90 =0·160, P=0·852; Spring of 40 °C. To examine the supercooling capacity of F =0·031, P=0·970). Since we found that − 2,106 Cold tolerance of an ectoparasite 929 exposure for 4 days killed a large proportion of the pupae if the temperature was below 15 °C, we exposed these pupae for 3 days (72 h) to either− 15 °C or 20 °C. − −

Cold tolerance of free-living stages Although SCP may not be directly applicable to survival in field conditions, it has been used widely as a comparative measure of cold tolerance between different life-history stages (Leather et al. 1993). To test possible differences in cold-hardening potential between different free-living life-stages (diapause and non-diapause pupae and adult), we measured freezing points (SCP) also during pupal development (representing the summer stage) and after adult Fig. 2. Mean supercooling points (°C± S.E.) through 4 seasons and all free-living stages of the deer ked. emergence (the autumn stage). The developing Supercooling points are measured for newborn stage consisted of pupae (n=31) collected on April diapausing pupae produced between Oct. and April 10 and reared in a climate chamber (+20 °C, 60% RH (A–D), and for the active stage in summer (E: pupae in and a photoperiod of 19 h/5 h for day/night) for 40 the middle of the development) and in autumn days (i.e. at the halfway point of pupal morphogen- (F: 1-week-old adults). P-values for SCP differences esis). To obtain young adults (n =34) we reared pupae according to Tukey post-hoc tests: non-significant within without any cold exposure and measured their SCP a diapause (i.e. between ABCD) but between active stages week after adult emergence. In addition, we exam- (EF) P<0·001. ined tolerance to long frost exposure in developing pupae that we had collected and reared simul- the pair-wise comparisons Tukey HSD post-hoc taneously with the pupae used for SCP measure- tests. ments. We exposed 36 pupae for 3 days to +5 °C, 15 °C or 20 °C, while the control treatment was − − executed at +20 °C. RESULTS The short-term cold-hardening capacity of the deer Data analysis keds (measured as freezing points) was high through We analysed all the data with SPSS for Windows 4 seasons and all free-living life-stages, but the ability (version 15.0). We tested the effect of cold acclim- to survive prolonged frosts was highest in diapausing ation at 5 °C on supercooling capacity at diapause pupae. The survival of diapausing pupae decreased by means− of a t-test. We analysed the differences in with the length and severity of frosts above the the supercooling points of diapausing pupae accord- supercooling point, whereas large size increased the ing to their birth time by Covariance Analysis: the rate of survival. SCP was set as the dependent factor, pupal mass as a covariate and the birth month as a fixed factor. To test Cold tolerance of diapausing pupae the influence of three-day treatments (no frost, 15 °C or 20 °C), the effect of season (winter or The supercooling points of diapausing pupae − − spring) and pupal mass on survival, we used a GLM (mean ± S.E,: 26·1 °C±0·13) were not affected by with binomial error distribution and the logit-link cold acclimation (t=0·187, D.F.=70, P=0·852, function. In the model, survival was set as the Fig. 1) or birth month (F3,134 =1·133, P=0·338, dependent factor, pupal mass as a covariate, and the Fig. 2). Neither pupal mass (F1,134 =0·087, P=0·768) treatment and season as predictors. First, a model nor Month × Mass interaction (F3,134 =0·959, with all interactions and main effects was fitted to the P=0·414) affected the supercooling points of young data. Then we removed unnecessary parameters diapausing pupae. according to a principle of hierarchy in order to All the diapausing pupae died if the temperature gain the definitive model. More detailed comparisons was 20 °C or 25 °C for 4 days, but 33·3% of the for the survival probability within/between seasons pupae− survived− at 15 °C, and 76·7% survived at and frost treatments we conducted by means of Χ2- 5 °C. During the− 3-day exposure, survival prob- tests. We analysed the seasonal variation in mean ability− decreased with severity of frost (Wald 2 pupal mass (i.e. in winter and spring) by a t-test. For X =9·517, D.F.=2, P=0·009, Fig. 3). Survival was the variation in the supercooling points between life- significantly higher in young spring pupae than in 2 history stages (diapause, developmental and adult winter, but only at 20 °C (X =8·172, D.F.=1, stages) we used One-way Analysis of Variance and for P=0·004). However,− survival during 3-day frosts Laura Härkönen and others 930

Fig. 4. Mean pupal masses (mg± S.E.) for all pupae tested Fig. 3. Survival of diapausing pupae produced in winter in winter and spring contrasted with pupae that survived and spring after a 3-day frost periods at 15 °C or 20 °C − − after being exposed to frosts for 3 days. compared with survival of pupae without frost experience. potential to survive prolonged harsh frosts. Thus life- history stages that do not experience very low was not affected by the season as such (Wald 2 temperatures in nature were still able to cope with Χ1 =1·922, P=0·166), although it increased with 2 serious frosts. pupal mass (Wald Χ =9·205, D.F.=1, P=0·004, Fig. 4). In spring, the pupae were on average larger than in winter (t=5·010, D.F.=197, P<0·001). Diapause-mediated cold-hardening The freezing points of newborn diapausing pupae Cold tolerance of free-living stages were low without long cold acclimation. Our results thus contradict the general assumption that diapause- The SCPs of the 3 life stages (diapause, development, mediated cold-hardening requires days or weeks ff fi adult) di ered signi cantly (F2,207 =332·017, before actual severe winter conditions arise (Tauber P<0·001, Fig. 2). The mean SCP (°C± S.E.) was et al. 1986). Ectoparasites dropping off the host lower in diapausing pupae ( 26·1 ±0·13) than in − during unfavourable seasons may be exposed to a active stages, when developing pupae ( 19·9±0·23) high risk of freezing. For example, if an ectoparasitic had higher freezing points than adults (−21·2±0·23) − winter tick (Dermacentor albipictus) leaves the (post-hoc: P=0·001). The developing pupae had the warm cervid host too early in spring, low tempera- potential to survive prolonged frosts although survi- tures are likely to kill it (Samuel et al. 2000). Rapid val probability is very low during 3-day exposure: cold-hardening processes increase survival in species temperatures of 20 °C killed all the developing − that experience sudden changes in their thermal pupae, but one adult emerged after exposure of 3 days environment (Chen et al. 1987; Czajka and Lee, at 15 °C. 63·9% of pupae survived a 3-day period at − 1990). Low freezing points without cold acclimation +5 °C, whereas survival in the control treatment may increase the survival probability of newborn (+20 °C) was 66·7%. deer ked pupae immediately when they drop off the host. However, in Northern Boreal environments, temperatures on the snow surface during midwinter DISCUSSION may remain below 20 °C for several days, and Newborn diapausing pupae were able to cold-harden thus tolerance to long− frost periods is crucial for well without acclimation and they survived harsh survival. frosts lasting from 3–4 days. The freezing points of Cold-hardening capacity measured as supercooling non-acclimated diapausing pupae ( 26 °C) did not points determines only lethal frost during a short cold vary with birth time or size. Tolerance− to prolonged exposure and may not be directly translated into frosts was, however, unexpectedly higher in spring long-term survival in the field (Leather et al. 1993). than in winter, but the observed seasonal differences This is because in freeze-intolerant insects, pre-freeze were explained by pupal size: in spring the pupae mortality may occur well above the supercooling were larger and their tolerance better than in winter. point, and the mortality rate increases with time Cold-hardening capacity was highest in the diapaus- spent in a supercooled state (Knight et al. 1986). Deer ing pupae, yet it was surprisingly high also after ked pupae seemed to be intolerant to freezing and diapause in developing pupae ( 20 °C) and at the suffered from higher mortality when severity and adult stage ( 21 °C). Developing− pupae also had the duration of frost exposure increased: they survived − Cold tolerance of an ectoparasite 931 relatively well for 3 days at 20 °C, but died after 4 seasonally: many insects are sensitive to frosts after days of exposure to those temperatures.− diapause termination, and they die mainly from cold The freezing points of diapausing deer ked pupae shock injuries rather than freezing (Tauber et al. did not vary with size or according to birth month, 1986; Chen et al. 1987). For instance, non-diapausing whereas tolerance to prolonged frost exposure was pupae of Sarcophaga crassipalpis have a supercooling higher in spring than in winter. However, differences point of 23 °C but do not survive to adult emer- in the pupal mass of the deer ked fully explained the gence when− exposed to 17 °C for as little as 20 min unexpectedly higher cold tolerance of young pupae in (Lee and Denlinger, −1985). Surprisingly, non- spring than in winter. The mortality of small pupae diapausing deer ked pupae have the potential to increased with the severity of 3-day frost, whereas survive harsh summer frosts: an adult emerged after large pupae survived well. This suggests that large the 3-day exposure to 15 °C, but a slightly milder pupal size increases resistance to pre-freeze mortality frost or shorter exposure− might have increased their during prolonged frost periods. survival rate. Altogether, the mean supercooling Females are expected to invest more in offspring points of diapausing and non-diapausing deer ked if even a small increase in provisioning increases pupae were a relatively good predictor of cold- survival probability (Smith and Fretwell, 1974; hardiness when estimating survival after a 3-day Parker and Begon, 1986; Plaistow et al. 2007). frost exposure: they had potential to survive if the Pupal size correlated positively with cold tolerance frost was not much harsher than approximately 5 °C as was expected, but seasonal variation in offspring above their supercooling point. size did not correlate with the occurrence of harsh This study and the study by Nieminen et al. (2012) frosts. Offspring size increased towards spring and both report high cold tolerance in adults, but with the explanation for this trend is currently under partly differing results. Nieminen et al. (2012) investigation: old females in spring may be able to demonstrated that adults might survive at tempera- invest more in one offspring, or the deterioration in tures above 16 °C, but they report a significantly host (moose) condition may promote maternal ability higher supercooling− point of adults ( 7·8 °C, n=6) to produce larger offspring (Härkönen et al. unpub- than found in this study ( 21·2 °C, n=34).− Based on lished observations). the freezing point, they concluded− that the deer ked Harsh frosts are presumably a relatively minor exploits freezing tolerance strategy. Conversely, our factor regulating the deer ked population as new data, which also included the other off-host stages, pupae are produced during a 9-month period. The strongly suggest a freeze avoidance strategy to be pupae produced in autumn will be covered by snow utilized and high pre-freeze mortality at temperatures before the onset of harsh winter frosts: snow approaching the freezing point. However, Nieminen significantly buffers both minimum temperatures et al. (2012) used adults collected from the field and cooling rates, thus protecting overwintering during several weeks, and thus the body condition insects (Leather et al. 1993). During winter, it may and age (i.e. length of starvation) differ from those of take days or weeks before an insulating layer of snow our specimens. Moreover, they used significantly provides cover, and thus pupae that are produced older adults for the SCP measurements than for the during extremely long and harsh frost periods may survival tests, and thus their conclusion may be have reduced survival probability. However, frosts incorrect. below 20 °C in Finland occur only during a few High cold tolerance all year round and through months− (Dec–Mar) (Finnish Meteorological Insti- free-living stages may correlate with other factors tute Database). In spring the amount of snow increasing off-host survival in ectoparasites. For diminishes, exposing all pupae to spring frosts, example, it has been suggested that the high super- which are not, however, as severe as in winter. cooling capacity in the parasitic tick Argas reflexus (Argasidae) is a consequence of its ability to survive prolonged periods of starvation and desiccation when outside the host (Dautel and Knülle, 1997). The cold Cold-hardening capacity through the free-living period tolerance of the deer ked was reduced during the end Supercooling points of 30 °C during diapause are of the free-living life-span, which may be related to not uncommon among− hibernating insects in tem- starvation (Verhoef et al. 1997). Nieminen et al. perate regions, but only some Arctic and Antarctic (2012) reported that the free-living adult stage species retain a high degree of cold-hardiness all year does not use low molecular weight cryoprotectants round (Tauber et al. 1986; Bale et al. 2001). The (sugars, polyols or amino acids) to increase their cold freezing points of the deer ked were lowest during tolerance, but the characterization of pupal contents diapause as expected, but they remained at 20 °C or (i.e. maternal-derived resources) is a subject for below even after diapause, although active stages− will further study. We presume that the high tolerance never experience such temperatures. Even if freezing of all stages could be a side effect of offspring points could be kept nearly constant throughout the provisioning for their long non-feeding period. For year, susceptibility to lethal cold injury varies example, body fat is an important energy source Laura Härkönen and others 932 during starvation (Colinet et al. 2006), and fatty acid FINANCIAL SUPPORT composition has been found to be involved in the This study has been funded by the Academy of Finland cold-hardening of dipterans (Bennett et al. 1997; (to A.K., L.H.), University of Oulu (L.H.), EnviroNet Ohtsu et al. 1998). (L.H.) and several foundations: the Ella and Georg The majority of insects enter diapause well in Ehrnrooth Foundation (L.H., S.K.), the Biological Society of Finland Vanamo (L.H., S.K.), Societas Pro advance in order to withstand decreasing autumnal Fauna et Flora Fennica (L.H., S.K.), and the Alfred temperatures. However, the ability to remain active Kordelin Foundation (S.K.). under harsh climatic conditions may be a significant key to fitness (Colinet et al. 2006) and thus increase e.g. range expansion potential (Bale and Hayward, REFERENCES 2010; Nieminen et al. 2012). In northern environments, nocturnal frosts may occur even in summer. Ansart, A. and Vernon, P. (2004). Cold hardiness abilities vary with the size of the land snail Cornu aspersum. Comparative Biochemistry and During autumn subzero temperatures may occur Physiology – Part A: Molecular and Integrative Physiology 39, 205–211. nightly, whereas day temperatures still remain Bale, J. S. and Hayward, S. A. L. (2010). Insect overwintering in a favourable for flying: the deer ked adults have been changing climate. The Journal of Experimental Biology 213, 980–994. doi: fi 10.1242/jeb.037911. found to survive until December in the eld despite Bale, J. S., Worland, M. R. and Block, W. (2001). Effects of summer frequent nocturnal frosts close to 10 °C occurring frost exposures on the cold tolerance strategy of a sub-Antarctic beetle. in late autumn/early winter (Härkönen− et al. unpub- Journal of Insect Physiology 47, 1161–1167. doi: 10.1016/S0022-1910(01) 00097-X. lished observations). For ectoparasites with a long Bennett, V. A., Pruitt, N. L. and Lee, R. E., Jr. (1997). Seasonal changes development time (in relation to season length) in fatty acid composition associated with cold hardening in third instar and late flight time, shorter season length at high larvae of Eurosta solidaginis. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology 167, 249–255. doi: latitudes or altitudes may be compensated by 10.1007/s003600050071. extended flight time. For example, high cold- Bowler, K. and Terblanche, J. S. (2008). Insect thermal tolerance: what is hardening capacity all year around may have facili- the role of ontogeny, ageing and senescence? Biological Reviews 83, 339–355. doi: 10.1111/j.1469-185X.2008.00046.x. tated the rapid northward invasion of the deer ked Chen, C.-P., Denlinger, D. L. and Lee, R. E. (1987). Cold injury and during the past few decades (see Kaitala et al. 2009; rapid cold-hardening in the flesh fly, Sarcophaga crassipalpis. Physiological Härkönen et al. 2010). Zoology 60, 297–304. Czajka, M. C. and Lee, R. E. (1990). A rapid cold-hardening response protecting against cold shock injury in Drosophila malanogaster. Journal of Experimental Biology 148, 245–254. Colinet, H., Hance, T. and Vernon, P. (2006). Water relations, fat Conclusions reserves, survival and longevity of a cold-exposed parasitic wasp Aphidius colemani (Hymenoptera: Aphidiinae). Environmental Entomology 35, In the deer ked cold tolerance was highest during 228–236. doi: 10.1603/0046-225X-35.2.228. diapause and large size increased tolerance to Danks, H. V. (1987). Insect Dormancy: an Ecological Perspective. Entomological Society of Canada, Ottawa, Canada. prolonged frosts. However, the deer ked contradicted Dautel, H. and Knülle, W. (1997). Cold hardiness, supercooling ability a few general findings regarding cold tolerance and causes of low-temperature mortality in Argas reflexus and Ixodes ricinus in temperate insects: (i) diapause-mediated cold- (Acari: Ixodoidea) from Central Europe. Journal of Insect Physiology 43, 843–854. hardening did not require a long period of cold Fischer, K., Brakefield, P. M. and Zwaan, B. J. (2003). Plasticity in acclimation, (ii) tolerance to long cold exposures was butterfly egg size: Why larger offspring at lower temperatures? Ecology 84, highest in spring, not in winter, and (iii) life-history 3138–3147. Haarløv, N. (1964). Life cycle and distribution pattern of Lipoptena cervi stages experiencing only favourable conditions (L.) (Dipt., Hippobosc.) on Danish deer. Oikos 15, 93–129. had high cold-hardening capacity. Our results Hahn, D. A., Martin, A. R. and Porter, S. D. (2008). Body size, but not indicate that in blood-feeding ectoparasites, tolerance cooling rate, affects supercooling points in the red imported fire ant, Solenopsis invicta. Environmental Entomology 37, 1074–1080. to low temperatures through 4 seasons correlates Härkönen, L., Härkönen, S., Kaitala, A., Kaunisto, S., Kortet, R., with other survival characteristics, such as the high Laaksonen, S. and Ylönen, H. (2010). Predicting range expansion of an starvation resistance, of all the free-living stages of ectoparasite – the effect of spring and summer temperatures on deer ked (Lipoptena cervi, Diptera: Hippoboscidae) performance along a latitudinal ectoparasites. gradient. Ecography 33, 906–912. doi: 10.1111/j.1600-0587.2009.05890.x. Heikkinen, S. (2000). Hirven vuosi. Suomen Riista 46, 82–91. [In Finnish.] Heinrich, B. (1999). The Thermal Warriors: Strategies of Insect Survival. Harvard University Press, Cambridge, MA, USA. ACKNOWLEDGEMENTS Kaitala, A., Kortet, R., Härkönen, S., Laaksonen, S., Härkönen, L., Kaunisto, S. and Ylönen, H. (2009). Deer ked, an ectoparasite of moose in We wish to thank all the volunteers who helped us with Finland: A brief review of its biology and invasion. Alces 45, 85–88. material collection, especially Juho-Antti Junno, Panu Kaunisto, S., Härkönen, L., Niemelä, P., Roininen, H. and Ylönen, H. Välimäki, Jani Koskimäki, Mervi Jaakola and the staff at (2011). Northward invasion of the parasitic deer ked (Lipoptena cervi), is Friendship Park Research Centre. We are indebted to four there geographical variation in pupal size and development duration? anonymous reviewers, Eija Hurme, Laura Vesala and the Parasitology 138, 354–363. doi: 10.1017/S0031182010001332. Evolution and Behaviour Research Group at the University Kaunisto, S., Kortet, R., Härkönen, L., Härkönen, S., Ylönen, H. and of Oulu for their very helpful comments on earlier drafts of Laaksonen, S. (2009). New bedding site examination-based method to analyse deer ked (Lipoptena cervi) infection in cervids. Parasitology Research the manuscript. We thank Rosemary Mackenzie and Marko 104, 919–925. Pyhähuhta for checking the English of the earlier drafts and Knight, J. D., Bale, J. S., Franks, F., Mathias, S. F. and Baust, J. G. final version of this manuscript. This study complies with (1986). Insect cold hardiness – supercooling points and prefreeze mortality. the current laws of Finland. Cryo Letters 7, 194–203. Cold tolerance of an ectoparasite 933

Koštál, V. and Šimek, P. (1995). Dynamics of cold hardiness, supercooling Plaistow, S., St. Clair, J., Grant, J. and Benton, T. (2007). How to and cryoprotectants in diapausing and non-diapausing pupae of the cabbage put all your eggs in one basket: empirical patterns of offspring root fly, Delia radicum L. Journal of Insect Physiology 41, 627–634. provisioning throughout a mother’s lifetime. American Naturalist 170, Krasnov, B. R., Burdelova, N. V., Shenbrot, G. I. and Khokhlova, I. S. 520–529. (2002). Annual cycles of four flea species in the central Negev desert. Popov, A. V. [Попов, А.В.] (1965). The life cycle of louse-flies Lipoptena Medical and Veterinary Entomology 16, 266–276. doi: 10.1046/j.1365- cervi L. and Stenepteryx hirundinis L. (Diptera, Hippoboscidae). Revue 2915.2002.00374.x. d’Entomologie de l’URSS XLIV, 573–583. [in Russian]. Langley, P. A. and Clutton-Brock, T. H. (1998). Does reproductive Räisänen, M., Repo, T. and Lehto, T. (2006). Effect of thawing time, investment change with age in tsetse flies, Glossina morsitans morsitans cooling rate and boron nutrition on freezing point of the primordial shoot in (Diptera: Glossinidae)? Functional Ecology 12, 866–870. Norway spruce buds. Annals of Botany 97, 593–599. Leather, S. R., Walters, K. F. A. and Bale, J. S. (1993). The Ecology of Roff, D. A. (1992). The Evolution of Life Histories: Theory and Analysis. Insect Overwintering. Cambridge University Press, Cambridge, UK. Chapman and Hall, New York, USA. Lee, R. E. and Denlinger, D. L. (1985). Cold tolerance in diapausing and Salt, R. W. (1961). Principles of insect cold hardiness. Annual Review of non-diapausing stages of the flesh fly, Sarcophaga crassipalpis. Physiological Entomology 6, 55–74. Entomology 10, 309–315. Samuel, W. M., Mooring, M. S. and Aalangdong, O. I. (2000). Lehane, M. (2005). The Biology of Blood-Sucking in Insects. Cambridge Adaptations of winter ticks to invade moose and moose to evade ticks. University Press, Cambridge, UK. Alces 36, 183–195. Matsuo, Y. (2006). Costs of prolonged diapause and its relationship to body Schultz, D. L. (1991). Parental investment in temporally varying environ- size in a seed predator. Functional Ecology 20, 300–306. ments. Evolutionary Ecology 5, 415–427. doi: 10.1007/BF02214158. McGinley, M. A., Temme, D. H. and Geber, M. A. (1987). Parental Smith, C. C. and Fretwell, S. D. (1974). The optimal balance between size investment in offspring in variable environments: Theoretical and empirical and number of offspring. American Naturalist 108, 499–506. considerations. American Naturalist 130, 370–398. Sømme, L. (1996). The effect of prolonged exposures at low temperatures Mousseau, T. A. and Dingle, H. (1991). Maternal effects in insect life in insects. Cryo Letters 17, 341–346. histories. Annual Review of Entomology 36, 511–534. doi: 10.1146/annurev. Tauber, M. J., Tauber, C. A. and Masaki, S. (1986). Seasonal Adaptations en.36.010191.002455. of Insects. Oxford University Press, New York, USA. Nieminen, P., Paakkonen, T., Eerilä, H., Puukka, K., Riikonen, J., Tinsley, C. R. (1999). Overview: extreme temperatures. Parasitology 119 III Lehto, V.-P. and Mustonen, A.-M. (2012). Freezing tolerance and low (Suppl.), S1–S6. molecular weight cryoprotectants in an invasive parasitic fly, the deer ked Wharton, D. A. (1999). Parasites and low temperatures. Parasitology 119 (Lipoptena cervi). Journal of Experimental Zoology 317A,1–8. [Published (Suppl.), S7–S17. New bedding site examination-based method to analyse deer online 10 NOV 2011]. doi: 10.1002/jez.722. Wilson, P. W., Heneghan, A. F. and Haymet, A. D. J. (2003). Ice Ohtsu, T., Kimura, M. T. and Katagiri, C. (1998). How Drosophila nucleation in nature: supercooling point (SCP) measurements and the role ked (Lipoptena cervi) infection in cervids. species acquire cold tolerance: Qualitative changes of phospholipids. of heterogeneous nucleation. Cryobiology, 46, 88–98. European Journal of Biochemistry 252, 608–611. doi: 10.1046/j.1432- Välimäki, P., Kaitala, A., Madslien, K., Härkönen, L., Várkonyi, G., 1327.1998.2520608.x. Heikkilä, J., Jaakola, M., Ylönen, H., Kortet, R. and Ytrehus, B. (2011). Parker, G. A. and Begon, M. (1986). Optimal egg size and clutch size: Geographical variation in host use of a blood-feeding ectoparasitic fly: Sirpa Kaunisto, Raine Kortet, Laura Härkönen, effects of environment and maternal phenotype. American Naturalist 128, implications for population invasiveness. Oecologia 166, 985–995. doi: 573–592. 10.1007/s00442-011-1951-y. Sauli Härkönen, Hannu Ylönen and Sauli Laaksonen Piiroinen, S. T., Ketola, A., Lyytinen, A. and Lindström, L. (2011). Verfoef, H. A., Nedved, O. and Lavy, D. (1997). Effects of starvation Energy use, diapause behaviour and northern range expansion potential in on body composition and cold tolerance in the collembolan Orchesella the invasive Colorado potato beetle. Functional Ecology 25, 527–536. doi: cincta and the isopod Porcellio scaber. Journal of Insect Physiology 43, 10.1111/j.1365-2435.2010.01804.x. 973–978. Parasitology Research 104: 919–925, 2009

Reprinted with the kind permission of the Springer-Verlag

III

New bedding site examination-based method to analyse deer ked (Lipoptena cervi) infection in cervids.

Sirpa Kaunisto, Raine Kortet, Laura Härkönen, Sauli Härkönen, Hannu Ylönen and Sauli Laaksonen

Parasitology Research 104: 919–925, 2009

Reprinted with the kind permission of the Springer-Verlag

Parasitol Res (2009) 104:919–925 DOI 10.1007/s00436-008-1273-0

ORIGINAL PAPER

New bedding site examination-based method to analyse deer ked (Lipoptena cervi) infection in cervids

Sirpa Kaunisto & Raine Kortet & Laura Härkönen & Sauli Härkönen & Hannu Ylönen & Sauli Laaksonen

Received: 30 August 2008 /Accepted: 7 November 2008 / Published online: 3 December 2008 # Springer-Verlag 2008

Abstract Invasion of the deer ked (Lipoptena cervi), an discolouration (host tissue fluid and deer ked faeces) on the ectoparasitic fly commonly found in cervids, has been rapid bedding sites to the extent that parasitism can be diagnosed. in Finland during the last four decades. As the distribution Hence, we suggest that deer ked infection prevalence and area of this species has expanded from the south towards range expansion could be rapidly monitored using our new the northern latitudes, the associated problems have become practical and reliable method. In the future, bedding site more evident. Various animals such as horses, cattle and analyses will likely be useful in predicting and potentially especially reindeer have been reported to host this parasite. preventing the negative effects of this ecologically and Moreover, in certain areas, the deer ked causes major socio-economically important parasite. inconveniences for humans potentially limiting recreational activities in forests. We wanted to study if deer ked parasitism and intensity of the infection in winter time Introduction could be detected by using visual examination of the snow on cervid bedding sites and by analysing biotic samples Insect ectoparasites display a range of forms of association found from the bedding sites. Our results demonstrate that with their mammalian and bird hosts. These parasites are chronic deer ked infection causes reddish-brown snow reported to cause direct and indirect harm to their hosts (Balashov 2007; Wall 2007). In the most extreme cases, both sexes of the ectoparasite live on the host most of their S. Kaunisto (*) Faculty of Biosciences, University of Joensuu, adult life feeding on blood or other tissues (Yuval 2006). P.O. Box 111, 80101 Joensuu, Finland The deer ked (Lipoptena cervi, Hippoboscidae) is a e-mail: [email protected] haematophagous (blood-feeding) ectoparasite exploiting several host species from the family Cervidae (Hackman R. Kortet : L. Härkönen Department of Biology, University of Oulu, et al. 1983; Kadulski 1996). P.O. Box 3000, 90014 Oulu, Finland In the spite of the wide distribution range of the deer ked (e.g. Maa 1969), its effects on the biology of different S. Härkönen cervid hosts have been studied surprisingly little. Blood- Joensuu Research Unit, Finnish Forest Research Institute, P.O. Box 68, 80101 Joensuu, Finland feeding ectoparasites may cause direct damage to skin tissue, inflammation, hyperemia, and even significant blood H. Ylönen loss on their hosts (Van den Broek et al. 2003; Wall 2007). Department of Biological and Environmental Science, For example, our own unpublished controlled experimental University of Jyväskylä, Konnevesi, Research Station, P.O. Box 35, 40014 Jyväskylä, work on the deer ked infection in semi-domesticated Finland reindeer (Rangifer tarandus tarandus) indicated that deer keds likely lower the physical condition of the host and S. Laaksonen cause eye-visible skin damages, as well as serious histo- Fish and Wildlife Health Research Unit, Finnish Food Safety Authority (Evira), logical, physiological and behavioural changes to the host. P.O. Box 517, 90101 Oulu, Finland Rantanen et al. (1982) reported that deer keds may cause 920 Parasitol Res (2009) 104:919–925 dermatitis and allergic reactions also on human skin. We predicted that, if observed, the reddish-brown snow Moreover, Laukkanen et al. (2005) observed deer ked discolouration on cervid bedding site would strongly associated occupational allergic rhinoconjunctivitis in hu- correlate with occurrence of deer ked pupae verifying the man. Thus, it is likely that the salivary and faecal antigens deer ked infection. We assumed that, if found, the produced by the deer ked stimulate host immune responses, discolouration would result mainly from cervid tissue fluids leading to hypersensitivity, dermatitis and allergic reactions (including blood) due to injured tissue of host skin and deer also in cervid hosts. ked faeces. The signs of deer-ked-associated blood would Deer keds can cause harm for the host also by increasing indicate that the skin tissue could have been damaged by infection-associated rubbing behaviour of the host. In- the deer keds or by the host itself as a response for the creased rubbing behaviour can lead to reduced time spent nuisance inflicted by these ectoparasites. Finally, it is ruminating (Berriatua et al. 1999) and hence weight lost, known that the deer ked prefers the neck and back regions lowered nutrition and viability of host (Broce 2006; of its host (Haarløv 1964). Thus, we predicted that the Balashov 2007). The excessive ectoparasite-associated reddish-brown snow discolouration would mainly locate in rubbing behaviour and other avoidance behaviour can harm those segments of the bedding sites (Fig. 2a, b). the host also per se (Berriatua et al. 1999; Broce 2006). Finally, the deer ked may cause indirect harm as being a potential vector for various pathogenic diseases, transmit- ting microorganisms between the hosts of the same or Materials and methods different species (Ivanov 1974; Rantanen et al. 1982; Dehio et al. 2004). For example, Dehio et al. (2004) managed to Natural history of the deer ked isolate Bartonella schoenbuchensis from the mid-gut of the deer ked. This bacterium may cause bacteremia in The deer ked has originally been found in Europe, some ruminants (Dehio et al. 2001). parts of Siberia, northern China, northern Africa (Algeria), Many of previous bedding site studies focus on the and it has also been introduced to North America (e.g. Maa cervid habitat selection (e.g. Mysterud and Ostbye 1995; 1969; Dehio et al. 2004). In Finland, the deer ked has Ratikainen et al. 2007). In some studies, biotic samples of rapidly spread towards the north during the last decades cervids found from the bedding sites such as faecal and hair (Hackman et al. 1983; Kaitala et al. 2008). At present, the samples have been used in analysing sex ratio and age limit of the distribution area of the deer ked in Finland structure, thus providing information about wild popula- follows and sometimes crosses the southernmost areas of tions (e.g. Huber et al. 2002). However, to our knowledge, the reindeer husbandry (Fig. 1). there are no previous bedding site studies in cervids In Fennoscandia, the main host of the deer ked is the moose focusing directly on ectoparasites. This is somewhat (Alces alces), abundances being up to 17,000 individuals in a surprising when considering the negative impact of the single moose (Paakkonen 2008). Our own monitoring in ectoparasites on vital functions and reproduction of their 2007 revealed the previously unreported finding in Finland hosts (e.g. Møller 1993). Our study introduces a new that the deer ked uses the wild forest reindeer (Rangifer technique to monitor cervids parasitised by the deer ked. tarandus fennicus) and occasionally the semi-domestic This technique is based on visual examination of bedding reindeer as hosts. The deer ked can also parasitise the roe sites and analysis of biotic samples found from bedding deer (Capreolus capreolus) and the Finnish population of the sites. white-tailed deer (Odocoileus virginianus), with a low Precisely, the main aim of our work is to study whether prevalence (Rantanen et al. 1982). An original host in the bedding site examination can be used as a reliable tool Eurasia is likely the red deer (Cervus elaphus; Haarløv for detecting deer ked parasitism in cervids during the 1964). The deer ked often fails in its host search by attacking snowy winter time. This method would be a practical and and accepting also hosts unsuitable for its reproduction (e.g. conceivable tool to monitor deer ked range expansion and humans; Haarløv 1964; Hackman et al. 1983). predicted invasion to the northern latitudes into economi- Adult deer keds emerge and seek hosts in Finland from cally important reindeer husbandry area. Another aim of the late summer to the end of autumn (Hackman 1977). Almost study was to confirm whether deer ked infection causes immediately after finding a host, both males and females bleeding of cervid hosts and to examine which body parts drop their wings (Bequaert 1953; Hackman et al. 1983) and of the cervid host have the highest degree of infection. To start to suck blood and interstitial fluid from the suitable reach our aims, we studied bedding sites from three host recurrently, up to 20 times per day (Ivanov 1974). One different host species. Our study sites were chosen from blood meal is approximately 0.0002–0.0003 g (Ivanov inside and outside of the estimated current distribution area 1974). The adults spend wintertime and the rest of their of the deer ked in Finland (Fig. 1). lives on the same host (Bequaert 1953). Deer ked female Parasitol Res (2009) 104:919–925 921

Fig. 1 Study locations and the rough estimation of the northernmost limit (dash line) of the current distribution area of the deer ked

viviparously produces new pre-pupae throughout the year Bedding sites and occurrence of deer ked pupae after mating. Pre-pupae deposit a fully mature, almost completely immobile third instar larva that quickly pupates The research was conducted in the six localities in Finland: within its last larval skin. Complete pupae then drop Rantsila (64°30′ N, 25°40′ E), Kuusamo (65°58′ N, 29°12′ passively from the host onto the ground or into the E), Nurmes (63°33′ N, 29°08′ E), Ristijärvi (64°30′ N, 28° vegetation, waiting there for the emergence of the latter 13′ E), Suomussalmi (64°53′ N, 29°00′ E) and Hyrynsalmi (review in Haarløv 1964). Complete deer ked pupae are (64°41′ N, 28°29′ E) in April–May 2008 (Fig. 1, Table 1). black, oval- or almost round-shaped and approximately The bedding sites of the moose were examined in Rantsila, 3 mm in diameter. They have hard chitinous cover and Kuusamo and Nurmes. The bedding sites of the semi- liquid interior (Bequaert 1953). domesticated reindeer were investigated in Suomussalmi 922 Parasitol Res (2009) 104:919–925

and in Hyrynsalmi, while the bedding sites of the wild forest reindeer were analysed in Ristijärvi. The study areas were selected because they had presumably high density of the potential cervid hosts. The study localities were explored for the bedding sites within a 20-km square area. The bedding sites were traced by following the recent tracks in snow by foot, skis, snowshoes or snowmobiles, depending on snow depth and the size of the study area examined. Only fresh and undisturbed bedding sites that were estimated to be not older than 24 h after the latest snowfall were included the study. The number of the pupae was counted on bedding sites. Despite their relative small size, black pupae are easy to observe against white snow. Then, bedding sites were photographed, and finally, the presence of the discoloured snow in the different parts of the 13 bedding sites was studied in detail from the pictures.

Samples for blood detection

Bright red discoloured spots on the snow of the bedding site were considered as a mark for direct bleeding. The snow samples taken from the bedding sites were collected from 20 cm2 area (approximately 2 cm deep) and analysed for the presence of blood (from four study areas). Moreover, five additional samples of adult deer ked tissues and faeces (collected from a moose that died in a car accident in April 8 2008) were analysed to explore if the snow discolouration could result also partly from the deer ked faeces after its blood meal. Snow samples were tested after melting by Combur-5 Test D test strips for the semi-quantitative determination of blood (Roche Norge A/S). The melted, heavily Fig. 2 a Resting moose divided into six indicative segments: A neck region, B back region, C rear end, D anterior segment, E front legs and discoloured snow from the bedding site was centrifuged lower abdomen, and F hind legs and lower abdomen. b Bedding site with 2,500 rpm for 10 min and then air-dried and of the moose divided into six indicative segments: A neck region, B stained. Some of the samples were finally studied under back region, C rear end, D anterior segment, E front legs and lower the microscope for the presence of blood or other cells abdomen, and F hind legs and lower abdomen. originating from the host.

Table 1 Details of the studied bedding sites from three cervids (Alces alces, Rangifer tarandus tarandus and Rangifer tarandus fennicus)

Location Number of Total pupae Pupae abundance Pupae prevalence Discolouration beddings count (%) (%) prevalence (%)

A. alces Rantsila 56 400 7.14 85.71 87.5 Kuusamo 23 0 0 0 0 Nurmes 33 78 2.36 N/A 100 R. tarandus tarandus Suomussalmi 150 0 0 0 0 Hyrynsalmi 68 5 0.07 7.35 2.94 R. tarandus fennicus Ristijärvi 67 2 0.03 2.98 2.98 Parasitol Res (2009) 104:919–925 923

The coverage of discoloured snow at the different parts discolouration were detected from the bedding sites, pupae of the bedding site abundances varying from 0.03 to 7.14 (Table 1). However, all the bedding sites in the two study locations outside of We investigated at which part of the bedding site deer ked the current rough estimation of the northernmost limit of faeces and cervid blood were located by analysing the photo- the deer ked (Kuusamo and Suomussalmi) lacked both the graphed bedding sites (Fig. 2b). In total, 13 pictures were pupae and the snow discolouration. In Kuusamo, all the 23 analysed in detail. Each picture represented one bedding site bedding sites of the moose were clean and white without of the moose. All the analysed pictures were taken in Rantsila pupae or discoloured snow. Similarly, all the 150 bedding where occurrence of pupae and snow discolouration in sites of the semi-domesticated reindeer in Suomussalmi bedding sites were most abundant. The bedding sites were lacked pupae, deer ked faeces, and discoloured snow. divided into six indicative and equal-size segments (Fig. 1b). In Rantsila, the presence of the deer ked pupae was Also the fringe area just round the bedding site (approximately strongly associated with snow discolouration in the bedding 20 cm) was included for snow discolouration analyses. sites of the moose [χ2(1)=28.10, p<0.001]. Similar result was observed in bedding sites of the wild forest reindeer in Statistical analyses Ristijärvi [χ2(1)=32.47, p<0.001]. Also in the bedding sites of the semi-domesticated reindeer in Hyrynsalmi, the SPSS for Windows (version 13.0) was used for statistical occurrence of deer ked pupae was significantly related with analyses. The deer ked’s role for the discolouration of the the snow discolouration [χ2(1)=25.96, p<0.001]. Unfortu- bedding sites indicated by possible significant association nately, the exact number of pupae per bedding site is not between the deer ked pupae and discoloured snow on bedding available from the moose bedding sites in Nurmes, but it site was studied using chi-square test. Possible differences in fluctuated from zero to eight pupae per bedding site. the proportion of discoloured snow between different segments Moreover, in all of the 33 moose bedding sites in Nurmes, of the bedding site were tested with Kruskal–Wallis test. the discoloured snow was detected. Results were considered statistically significant at p<0.05. The coverage of the discoloured snow at different parts of bedding site Results The segments of the bedding site differed significantly from Samples for blood detection each other regarding the coverage of the snow discolouration [χ2(5)=51.72, p<0.001] (Fig. 3). Coverage of dis- In Rantsila, most of the moose bedding sites were coloured snow was highest in the segments that had been in discoloured (Table 1) some of them containing also brightly touch with the neck and back regions of the host (1, 2 and visible fresh blood. In that study site, 16 of the 20 samples 4; Fig. 3). Small discoloured spots were also found on the analysed (80%) contained blood, indicated by haemoglobin surrounding snow in the vicinity of the bedding sites, detected by the test strips. Moreover, seven samples in suggesting that the animal had shaken its body and spread Rantsila contained also detectable red blood cells. In the droplets. contrast, in five moose bedding site samples analysed from Kuusamo (outside the deer ked distribution area), no blood indicators were detected. In Ristijärvi, two of the 67 Discussion bedding sites analysed contained haemoglobin but no red blood cells. In Hyrynsalmi, two out of the seven samples Our aim was to answer the question whether the commonly from semi-domesticated reindeer bedding sites contained observed discolouration of cervid bedding site is correlated haemoglobin, but no red blood cells were detected. Finally, with occurrence of reproducing deer ked in the cervid host five additional blood detection tests used in analysing adult and thus the occurrence of pupae in the bedding site. After deer ked tissues and faeces were positive suggesting that we found this correlation, we further asked if the snow discolouration on bedding sites likely results of both discolouration could be linked to direct or indirect skin cervid tissue fluids and deer ked faeces. damage caused by this parasite. Because both these conditions were fulfilled, we managed to develop a new Association between the occurrence of deer ked pupae method for monitoring the deer ked parasitism. The present and snow discolouration on bedding sites results clearly suggest that deer ked infection causes discolouration on the bedding site and also bleeding of In the four study locations (Rantsila, Nurmes, Ristijärvi and the host. This is a prerequisite that new bedding site Hyrynsalmi), deer ked pupae and faeces as well as snow examination-based method can be used. Deer ked pupae 924 Parasitol Res (2009) 104:919–925

Fig. 3 The coverage of discoloured snow at six segments (see Fig. 1b) of bedding sites (n=13). Coverage was classified into five classes varying from 0% to >25%

were associated with discolouration of the bedding sites in mouthparts especially adapted for piercing, cutting or all of the cervid populations where the deer ked was burrowing (e.g. Reid 2002; Kuhn et al. 2008). The deer present, while in both of the two locations without deer ked, as other louse flies, is adapted for piercing skin of the keds (Kuusamo and Suomussalmi), no blood indicators host with feeding apparatus including three sets of tiny were detected. Therefore, we feel that a convincing teeth, one set situated just in the entrance of haustellum association between the occurrence of deer ked pupae and (Haarløv 1964). Haustellum of the deer ked pierces into the tracks of cervid blood at bedding sites is being demonstrat- skin only about the depth of 1 mm. Thus, blood-sucking ed. Our results about bedding site discolouration are also behaviour of an individual deer ked may not be as painful supported by numerous unpublished observations by and cause noticeable direct bleeding than, for example, a Finnish hunters. cut of skin by a micropredatory horse fly (Tabanidae). Low Deer ked faeces, pupae and cervid blood were not found infections by arthropod parasites, in general, have been from the bedding sites in the two northernmost study areas thought to cause relatively little damage and few nuisance (Kuusamo and Suomussalmi). These observations are in because of the relatively small size of the parasite in line with the most recent data about the distribution area of relation to the host size and because of the fact that each the deer ked in Finland (Kaitala et al. 2008). However, it is individual consumes relatively small amount of blood per likely that the distribution area of the deer ked will still day (Poulin 1996). At low densities, the deer ked may be expand towards more northern areas in the future (Kaitala responsible for some injury, but the damage that an et al. 2008). If so, the problems caused by the deer ked for individual deer ked does is difficult to measure. Hence, semi-domesticated reindeer and for other host animals will we suggest that the number of the deer keds per host and become increasingly serious. The deer ked causes likely chronicity of the infection are the main factors affecting to notable harm to reindeers and could be regarded as an the extent of the tissue damage and the amount of the economic threat if it reaches the whole area of the reindeer potential bleeding of the host. In our own unpublished husbandry. Hence, a suitable method for monitoring the controlled experiment, a few hundred deer keds caused dispersion of the deer ked is needed from ecological, nuisance and pain for semi-domesticated reindeer altering veterinary and economical point of views. The ultimate aim the behaviour of the host. Without a doubt, a cumulative of this work was to answer to this need and develop a new effect of long-term parasitism by thousands of individuals and practical method for monitoring deer ked invasion in may cause damage to capillary veins and skin of the host to the northern latitudes. Our present results demonstrate that the extent that the bleeding occurs. A cervid host itself can analysing bedding sites is a reliable method for monitoring make the bleeding and possible tissue damages even worse purposes. There might potentially be other haematophagous by its own behaviour. An excessive scratching or rubbing ectoparasites, like mallophagans and psoroptids, on cervid behaviour, using cloven hooves or teeth, as a response to skin during wintertime, especially in the other geographical allergic reactions, discomfort and itching can lead to areas (e.g. Kadulski 1996). However, these parasites are not scabbing and possibly to secondary infections. Our unpub- notably abundant in the present study areas and have never lished data suggests that beside skin damages and bleeding, been reported to cause snow discolouration. the deer ked also causes hair loss in cervids. In the winter The extent of the tissue damage and the amount of time, extensive coat damages with hairless regions can bleeding of the host depend partly on the feeding cause heat loss, further increasing critically host energy morphology of blood-feeding arthropod species on ques- requirements. In certain conditions, these negative impacts tion. Arthropods that can pierce vertebrate skin have are likely detrimental to the host. Parasitol Res (2009) 104:919–925 925

Most of the reddish-brown snow discolouration was Dehio C, Lanz C, Pohl R, Behrens P, Bermond D, Piemont Y, Pelz K, detected in the bedding site segments that had been in touch Sander A (2001) Bartonella schoenbuchii sp. nov., isolated from the blood of wild roe deer. Int J Syst Evol Microbiol 51:1557– with the neck and back region of the host, indicating the 1565 most probable predilection sites of the deer ked on the host. Dehio C, Sauder U, Hiestand R (2004) Isolation of Bartonella These observations are congruent with our predictions and schoenbuchensis from Lipoptena cervi, a blood-sucking arthro- can be used in future work estimating the impact this pod causing deer ked dermatitis. J Clinic Microbiol 42:5320– 5323 parasite on the host. Haarløv N (1964) Life cycle and distribution pattern of Lipoptena The host’s movements regulate the deer ked’s eclosion cervi (L.) (Dipt., Hippobosc.) on Danish deer. Oikos 15:93–129 habitat and therefore also the ecological dispersion of the Hackman W (1977) Hirven täikärpänen ja sen levittäytyminen deer ked. Adult deer ked itself cannot fly long distances Suomeen. Luonnon Tutkija 81:75–77 Hackman W, Rantanen T, Vuojolahti P (1983) Immigration of (Hackman et al. 1983), but they can disperse with the Lipoptena cervi (Diptera, Hippoboscidae) in Finland, with notes moose or the roe deer easily 50–150 km during annual on its biology and medical significance. Not Entomol 63:53–59 migrations of the host (e.g. Heikkinen 2000). The visual Heikkinen S (2000) Hirven vuosi. Suomen Riista 46:82–91 examination of the bedding sites and biotic samples found Huber S, Bruns U, Arnold W (2002) Sex determination of red deer using polymerase chain reaction of DNA from faeces. Wildl Soc from them provide valuable information for monitoring Bull 30:208–212 dispersion of the deer ked. When combined with the Ivanov VI (1974) On the damage done by Lipoptena cervi L. (Diptera, appearance of cervid host (e.g. hairless areas on body or Hippoboscidae) in Byelorussia. Parazitologiya 8:252–253 coat discolouration), the discolouration of the bedding sites Kadulski S (1996) Ectoparasites of Cervidae in north-east Poland. Acta Parasitol 41:204–210 can be used as a tool to detect parasitised animals. The only Kaitala A, Kortet R, Härkönen S, Laaksonen S, Ylönen H (2008) limitation of the method developed prior to this study is that Invasion of the moose ectoparasite Lipoptena cervi in Finland. it is practical and effective only in wintertime. However, the In: Baskin LM (ed) Proceedings of the VIth International Moose use of snow mobiles enables to comb and explore quickly Symposium. Yakutsk, pp 37–39 Kuhn C, Lucius R, Matthes HF, Meusel G, Reich B, Kalinna BH large areas for the cervid bedding sites. (2008) Characterisation of recombinant immunoreactive antigens To conclude, the reddish-brown snow discolouration at of the scab mite Sarcoptes scabiei. Vet Parasitol 153:329–337 bedding sites is an easily detectable but reliable indicator of Laukkanen A, Ruoppi P, Mäkinen-Kiljunen S (2005) Deer ked- parasitised hosts. This cue can be combined with the induced occupational allergic rhinoconjunctivitis. Ann Allergy Asthma Immunol 94:604–608 occurrence of deer ked pupae and appearance of the host. Maa TC (1969) A revised checklist and concise host index of Thus, in the future, using bedding site analysis method in Hippoboscidae (Diptera). Pacific Ins Monogr 20:261–299 reindeer husbandry areas, parasitised reindeers could be Møller AP (1993) Ectoparasites increase the cost of reproduction in easily recognised and suspended from the herd. Finally, this their hosts. J Anim Ecol 62:309–322 Mysterud A, Ostbye E (1995) Bedding-site selection by European roe new bedding site analysis method can be applied in most deer (Capreolus capreolus) in southern Norway during winter. parts of northern boreal hemisphere that have a long-lasting Can J Zool 73:924–932 permanent snow cover for monitoring and mapping the Paakkonen T (2008) Hirvikärpänen ja sen vaikutukset hirveen ja distribution area of the deer ked. poroon. In: RKTL, Riistapäivät 2008 Oulu 22–23.1., Kooste Riistapäivien esitelmätiivistelmistä. RKTL Poulin R (1996) The evolution of life history strategies in parasitic Acknowledgements We want to thank Kauko Kilpeläinen, Antti animals. Adv Parasitol 37:107–134 Leivo, Ari Junttila, Reijo Kotilainen and Pertti Rautio for their kind Rantanen T, Reunala T, Vuojolahti P, Hackman W (1982) Persistent help during the experiment. This study was partly funded by the pruritic papules from deer ked bites. Acta Dermatol Venereol MAKERA and University of Oulu (RK). Our experiments comply 62:307–311 with the current laws of Finland. Ratikainen II, Panzacchi M, Mysterud A, Odden J, Linnell JDC, Andersen R (2007) Use of winter habitat by roe deer at a northern latitude where Eurasian lynx are present. J Zool References 273:192–199 Reid SA (2002) Trypanosoma evansi control and containment in Balashov YS (2007) Harmfulness of parasitic insects and acarines to Australasia. Trends Parasitol 18:219–224 mammals and birds. Entomol Rev 87:1300–1316 Van den Broek AHM, Huntley JF, Halliwell REW, Machell J, Taylor Bequaert JC (1953) The hippoboscidae or louse-flies (Diptera) of M, Miller HRP (2003) Cutaneous hypersensitivity reactions to mammals and birds. Part 1. Structure, physiology and natural Psoroptes ovis and Der p 1 in sheep previously infested with P. history. Entomol Am 32:1–209 ovis—the sheep scab mite. Vet Immunol Immunopathol 91:105– Berriatua E, French NP, Wall R, Smith KE, Morgan KL (1999) 117 Within-flock transmission of sheep scab in naive sheep housed Wall R (2007) Ectoparasites: Future challenges in a changing world. with single infested sheep. Vet Parasitol 83:277–289 Vet Parasitol 148:62–74 Broce A (2006) Ectoparasite control. Vet Clin North Am Food Anim Yuval B (2006) Mating systems of blood-feeding flies. Annu Rev Pract 22:463–474 Entomol 51:413–440

Avian predation on a parasitic fly of cervids during winter: can host-related cues increase the predation risk?

Sirpa Kaunisto, Panu Välimäki, Raine Kortet, Jani Koskimäki, Sauli Härkönen, Arja Kaitala, Sauli Laaksonen, Laura Härkönen and Hannu Ylönen

Biological Journal of the Linnean Society 106: 275–286, 2012

Reprinted with the kind permission of the Wiley-Blackwell

Avian predation on a parasitic fly of cervids during winter: can host-related cues increase the predation risk?

Sirpa Kaunisto, Panu Välimäki, Raine Kortet, Jani Koskimäki, Sauli Härkönen, Arja Kaitala, Sauli Laaksonen, Laura Härkönen and Hannu Ylönen

Biological Journal of the Linnean Society 106: 275–286, 2012

Reprinted with the kind permission of the Wiley-Blackwell

bs_bs_banner

Biological Journal of the Linnean Society, 2012, 106, 275–286. With 3 ﬁgures

Avian predation on a parasitic ﬂy of cervids during winter: can host-related cues increase the predation risk?

SIRPA KAUNISTO1*, PANU VÄLIMÄKI2, RAINE KORTET1, JANI KOSKIMÄKI2, SAULI HÄRKÖNEN3, ARJA KAITALA2, SAULI LAAKSONEN4, LAURA HÄRKÖNEN2 and HANNU YLÖNEN5

1University of Eastern Finland, Department of Biology, PO Box 111, FI-80101 Joensuu, Finland 2University of Oulu, Department of Biology, PO Box 3000, FI-90014 Oulu, Finland 3Finnish Wildlife Agency, Fantsintie 13–14, FI-00890 Helsinki, Finland 4University of Helsinki, Nurminiementie 2, FI-93600 Kuusamo, Finland 5University of Jyväskylä, Department of Biological and Environmental Science, Konnevesi Research Station, PO Box 35, FI-40014 Jyväskylä, Finland

Received 12 October 2011; revised 27 December 2011; accepted for publication 28 December 2011bij_1869 275..286

The deer ked (Lipoptena cervi) is an ectoparasitic fly on cervids that has expanded its distribution rapidly in Northern Europe. However, the regulating biotic factors such as predation remain unknown. The host- independent pupal stage of the fly lasts for several months. Blackish pupae are visible against snow, especially on the bedding sites of hosts, and are thus exposed to predators. To evaluate the role of predation on the invasion dynamics and evolution of L. cervi, we monitored pupal predation on artificial bedding sites in three geographical areas in Finland during winter. We explored: (1) possible predators; (2) magnitude of predation; and (3) whether predation risk is affected by host-derived cues. We demonstrate that pupae are predated by a number of tit species. Any reddish brown snow discoloration on bedding sites, indicating heavy infestation of the host, serves as an exploitable cue for avian predators, thereby increasing the risk of pupal predation. The ability of tits to use this host-derived cue seems to be dependent on the prevalence of L. cervi and the period of invasion history, which suggests that it may be a learned behavioural response. Predation by tits may potentially affect the L. cervi population dynamics locally. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 106, 275–286.

ADDITIONAL KEYWORDS: bedding sites – blood feeding – ectoparasite – Hippoboscidae – invasive species – moose – parasitism – tits.

INTRODUCTION may show a preference for the invader and thereby even prevent its establishment (e.g. Barber, Marquis The successful establishment of invasive species in & Tori, 2008; Carlsson, Sarnelle & Strayer, 2009). On new environments depends on both abiotic and biotic the other hand, the enemy release hypothesis argues factors, with the latter group including interactions that invasion success of a new arrival results from with the resident biota (Vermeij, 1996; Lockwood, reduced natural enemy attack (e.g. Williamson, 1996; Hoopes & Marchetti, 2007; Menke et al., 2007). Pre- Shwartz et al., 2009). dation is one of the major biological factors that Predation on the pupal stage of insects has been regulates prey populations, and in some cases can reported, especially by small mammals, insectivorous affect their invasion success (e.g. Schoener & Spiller, birds, and arthropods (e.g. Frank, 1967; Tanhuanpää 1995; Kotiaho & Sulkava, 2007). Natural enemies et al., 1999; Hastings et al., 2002; Barbaro & Battisti, 2011). In the northern boreal region, endothermic *Corresponding author. E-mail: sirpa.kaunisto@uef.ﬁ vertebrates are potential predators of insects during

© 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 106, 275–286 275 276 S. KAUNISTO ET AL. the winter (e.g. Alatalo, 1980; Jansson & von Bröms- effects of host-related factors upon the probability of sen, 1981). Birds and small mammals may use olfac- predation. In general, factors like body fluids and tory, auditory, and visual cues when searching for secretions from the prey may increase the predation prey (Bennett & Cuthill, 1994; Montgomerie & risk, especially by avian predators, because of the Weatherhead, 1997; Nevitt, Losekoot & Weimer- UV visibility of secretions (e.g. Viitala et al., 1995; skirch, 2008; Vaughan, Ryan & Czaplewski, 2011). Koivula & Korpimäki, 2001). The snow covering Olfactory cues such as volatile organic compound bedding sites of infested cervids is characterized by (VOC) emissions by damaged plants may enhance the reddish brown discoloration of tissue fluid and the prey location efficiency of insectivorous birds blood from the host (caused by L. cervi bites and (Mäntylä et al., 2008). According to current under- irritation of skin) and L. cervi faeces, with the degree standing, however, the predominant mode of prey of discoloration correlating positively with infestation detection among birds is vision (see Rajchard, 2009 intensity (Kaunisto et al., 2009; Välimäki et al., 2011). for a review). Many birds can detect ultraviolet light Hence we hypothesized that parasite-induced snow and/or chromatic cues when searching for prey discoloration on host bedding sites may serve as a cue (Bennett & Cuthill, 1994; Viitala et al., 1995; Church of the presence of parasites, and thus increase the et al., 1998; Rajchard, 2009). Moreover, indicators like predation risk. Associated with the contour of a leaf damage (e.g. Heinrich & Collins, 1983) or the bedding site, host faecal pellets may provide an addi- specific feeding habits (Murakami, 1999) of leaf- tional cue by revealing the previous presence of the feeding caterpillars can serve as visual cues for insec- host, and thereby indicating a potential foraging site. tivorous birds. Although predation by birds may This is a reasonably assumption, as cervids often potentially regulate invertebrate populations, in feceate onto their bedding site once they leave it, and general with the highest prey mortality at low prey brownish faecal pellets are more conspicuous against density (Fowler et al., 1991), the effects on insect snow than clean bedding sites as such. By using population dynamics are not well understood. Even different combinations of cues, we were able to dis- less is known about predation as a potential regulator entangle whether the host-derived cue (faecal pellets) of the off-host stages of ectoparasites. expose pupae to predation or whether the infestation- The deer ked, Lipoptena cervi (Diptera; Hippobos- induced cue (snow discoloration) triggers the behav- cidae), is an obligate haematophagous ectoparasite of ioural response in possible predators. Moreover, we several species of cervids (Haarløv, 1964). It can affect evaluated whether the ability to exploit certain cues the health of its host (Kynkäänniemi et al., 2010) and is innate in predators or a learned response to reliable is a nuisance to humans: for example, causing allergic cues. This knowledge would also allow us to better reactions (Rantanen et al., 1982; Kortet et al., 2010). understand evolutionary mechanisms of the host– In Northern Europe, L. cervi has rapidly expanded parasite interactions in the present study system. its distribution and increased in abundance during the preceding four decades (Välimäki et al., 2010; Kaunisto et al., 2011). Yet, there is significant spatial MATERIAL AND METHODS variation in the expansion rate of the species (Välimäki et al., 2010). To date the role of potentially NATURAL HISTORY OF LIPOPTENA CERVI limiting biotic factors such as predation has not been Lipoptena cervi is a Palearctic species that has been studied, even though Haarløv (1964) proposed the introduced to the Nearctic region (e.g. Maa, 1969; possibility almost 50 years ago. To understand Dehio, Sauder & Hiestand, 2004). In Fennoscandia, the ecology and evolutionary invasion dynamics of especially in Finland, L. cervi has rapidly spread L. cervi, the interactions between the parasite, its north up to the southernmost part of the region hosts and possible predators need to be studied. of reindeer husbandry (Hackman, Rantanen & We examined pupal predation as a possible factor Vuojolahti, 1983; Välimäki et al., 2010). In Finland, affecting the population dynamics of L. cervi. The the main breeding host (i.e. host supporting reproduc- pupal stage is independent of the mobile warm- tion) of L. cervi is the moose (Alces alces) (Välimäki blooded host, and may therefore be among the most et al., 2011). A single A. alces bull may host up to vulnerable stages to extrinsic mortality agents. First, 17 500 adult L. cervi (Paakkonen et al., 2010). The we explored whether there is predation on L. cervi wild forest reindeer (Rangifer tarandus fennicus) pupae in the Northern Boreal region during winter. (Kaunisto et al., 2009) and occasionally the semi- We assumed that tits (Paridae), which are highly domesticated reindeer (Rangifer tarandus tarandus) adaptable predators (Pimentel & Nilsson, 2007) and (Kynkäänniemi et al., 2010) can also serve as hosts. form feeding flocks during winter (Suhonen, 1993; Lipoptena cervi can also parasitize the roe deer Suhonen, Alatalo & Gustafsson, 1994), would be (Capreolus capreolus), the fallow deer (Dama dama), the most likely predators. Second, we examined the the red deer (Cervus elaphus) (Haarløv, 1964), and the

© 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 106, 275–286 PREDATION ON THE DEER KED 277 white-tailed deer (Odocoileus virginianus) (Matsu- L. cervi, whereas Utajärvi (64°45′N, 26°53′E) and moto et al., 2008), but this does not seem to be the Yli-Ii (65°26′N, 26°10′E) are closer to its recent expan- case in Finland (Välimäki et al., 2011). The increased sion front (Fig. 1). We chose the study sites based on density of A. alces is likely to be the main reason that A. alces and L. cervi densities (see below). Also, the underlies the rapid spread and increase of the ecto- occurrence of the assumed predator guild, the tits, parasite in Finland (Välimäki et al., 2010). In addi- affected the study site selection. A rough estimation of tion to the host availability, ongoing climate warming tit species occurrence on each study site was based on may be a factor that enhances the range expansion visual and auditory observations on tit species made (Härkönen et al., 2010). before the onset of the experiments. Adult L. cervi emerge during late summer until the We produced artificial bedding sites by digging onset of winter (Hackman et al., 1983). Immediately grooves [150 cm ¥ 75 cm ¥ 30 cm (depth)] onto snow to after finding a host, flies shed their wings and start imitate natural A. alces bedding sites. We provided to feed on blood (Bequaert, 1953). The adults spend the artificial bedding sites with L. cervi pupae as well wintertime on the same host, with viviparous females as potential host-derived cues, such as A. alces faecal giving birth to already prepupated larvae. Offspring pellets and snow discoloration (discoloration directly are produced one at a time, with the female’s repro- indicates L. cervi infestation), and monitored the ductive output being dispersed evenly across the pupal loss caused by predators. We manipulated the whole reproductive phase, which lasts from autumn artificial bedding sites in increasing order of visual to the following spring (Haarløv, 1964; Popov, 1965). and olfactory cues, revealing the feeding site for The boreal areas are largely covered by snow at that potential predators. The number of experimental time of year. Pupae fall off the host and are thus manipulations varied among the study questions I– exposed to extrinsic mortality factors for several III (Table 1). We placed L. cervi pupae near an imagi- months until the emergence of the new adult genera- nary neck area of the artificial bedding site to mimic tion in the following autumn. natural infestation (Kaunisto et al., 2009). Because Blackish and oval-shaped (c. 3 mm in diameter) the snow melts underneath a resting large-bodied pupae fall off the host during host activity and rest. endotherm, the upper snow layer of natural bedding The resting bouts of A. alces may last for more than sites becomes very dense, which prevents pupae 18 h a day in the winter (see Van Ballenberghe & from sinking into the snow, leaving them visible for Miquelle, 1990). Assuming a constant reproductive relatively long times. To prevent the pupae from rate of female L. cervi, about three-quarters of their sinking and to estimate the maximal predation risk, reproductive output would thereby be exposed to pre- we placed the pupae on a white piece of gauze dation on host bedding sites. Depending on the infes- (20 cm ¥ 20 cm). We removed fresh snow in the case of tation intensity, the total number of pupae produced snowfall. When applicable, the number of A. alces by a group of females that reproduce simultaneously faecal pellets was ~20. In two experimental manipu- on a particular host individual varies from zero to lations, we created snow discoloration by spraying a dozens on natural bedding sites (Välimäki et al., water solution that consisted of reindeer blood and 2011). We have observed only a few dead adult yellow and brown watercolour pellets onto the imagi- L. cervi on A. alces bedding sites during the winter nary neck and back region of the artificial bedding months, which together with a constant reproductive sites. This was supposed to mimic the natural snow output suggests relatively low on-host mortality of discoloration on bedding sites (Kaunisto et al., 2009; reproductive adults (P. Välimäki, pers. observ.). Thus, Välimäki et al., 2011). The artificial discoloration was the pupa stage is likely to be the most vulnerable composed mainly of reindeer blood, which ensured a developmental stage, and on the other hand the close resemblance to the natural discoloration not bedding sites are locations where predation may be only visually but also by other properties, such as intense enough to affect L. cervi population dynamics. UV-reflectance, which appears to be important for foraging birds (Bennett & Cuthill, 1994; Church et al., 1998; Koivula & Korpimäki, 2001). We adjusted the STUDY AREAS AND EXPERIMENTAL SET-UP magnitude of artificial discoloration to correspond to We conducted the study in five localities, representing the natural bedding site of a highly infested host. three geographical areas (Fig. 1). The study took Depending on the experimental set-up, distances place during late winter, in March and early April between individual bedding sites or groups of bedding 2009. Konnevesi (62°41′N, E 26°16′E) is situated in sites (replicates) were ~0.5 km, so as to avoid pseu- southern Central Finland, whereas the other locali- doreplication. Each replicate was located within pre- ties are situated in northern Central Finland. Of the dominantly coniferous forests. We monitored the latter areas, Rantsila (64°28′N, 25°48′E) and Pulkkila artificial bedding sites at 8, 24, 48, 72 and 116 h after (64°18′N, 25°47′E) fall within the current core area of the onset of experiments. At each visit to the artificial

Figure 1. Locations of the study sites in Finland (black dots). The current geographic range of Lipoptena cervi (grey line) is deﬁned according to Välimäki et al. (2010).

bedding sites, which lasted for c. 10 min, we counted the avian species, we concentrated on tits, which are the number of pupae and recorded all auditory the known wintertime predators in the Northern and visual observations of potential avian predators Boreal forests (see Alatalo, 1982; Suhonen, 1993; within ~50 m radius from a particular replicate. We Pimentel & Nilsson, 2007). During typically cold and also recorded signs such as subnivean tunnels, tracks, short winter days, tits, especially the willow tit and faeces of possible small mammalian predators (Poecile montanus) and the coal tit (Periparus ater) like voles, shrews, and mice in the immediate vicinity are very sedentary as long as food accessibility of each artiﬁcial bedding site (within ~5 m radius). Of remains sustainable (e.g. Brotons, 1997; Lahti et al.,

Table 1. Total number of impacted bedding sites and the percentage of eaten pupae (in parenthesis), in relation to the study questions and experimental manipulations of bedding sites in each locality, with varying Lipoptena cervi abundance. The study questions: I, are pupae predated upon at bedding sites?; II, do host-related cues expose pupae to predation?; III, does natural variation in L. cervi infestation intensity affect predation risk?

Experimental manipulation

L. cervi Study Pupae + Pupae + Pupae + faeces + Location abundance question Pupae faeces colour colour

Konnevesi (62°41′N) high I 4 (16.3%) Pulkkila (64°18′N) high I 4 (10.3%) Rantsila (64°28′N) high II, III 4 (8.5%) 3 (15.0%) 20 (100%) 19 (95.0%) Pulkkila (64°18′N) high III 3 (10.0%) 15 (94%) Utajärvi (64°45′N) low III 2 (5.6%) 9 (50.6%) Yli-Ii (65°26′N) low III 5 (19.3%) 6 (36.9%)

1998). Thus, the distance of 0.5 km decreased the pellets; (3) pupae and snow discoloration; or (4) possibility that the same tit individuals would visit pupae, snow discoloration, and A. alces faecal pellets several replicates during the experiment. We admit (see Appendix S1). We organized the artificial bedding that our experimental design is not totally compre- sites into a 40-m ¥ 40-m quadrat, with each corner hensive because, for example, coal tits may shift their standing for one of the four manipulations. We repli- foraging habitat preference in winter, and increase cated the set-up 20 times so that the exact locations their range in search of new food resources (see, e.g., of artificial bedding sites (i.e. particular manipula- Brotons & Herrando, 2003). However, we stress that tions) were randomized within each quadrat. Hence, the transects of replicate experimental units were we constructed a total of 80 artificial bedding sites. from 8 to 10 km in length, and thus the assumption of The number of pupae on each bedding site was 10. independence was probably not severely violated. Does natural variation in Lipoptena cervi Are pupae predated on bedding sites? infestation intensity affect predation risk? First, we studied if any pupal predation occurs in the Finally, we constructed additional experimental set- absence of other host-related cues, except for the ups to explore whether the patterns observed in Ran- contour of an artificial bedding site. We conducted tsila can be generalized, or whether there are spatial this experiment in Konnevesi and Pulkkila, which are differences in predatory response to the host-derived characterized by abundant A. alces and L. cervi popu- cues (see Appendix S1). We organized 16 pairs of lations (see Välimäki et al., 2010, 2011), with the artificial bedding sites within each locality so that the respective number of replicates (artificial bedding exact locations of two experimental manipulations sites) being 16 and 20 (see Appendix S1). The number (pupae only and pupae with snow discoloration) were of pupae on each bedding site was 15. randomized within each pair. We applied only two of the four possible manipulations because snow discol- Do host-related cues expose pupae to predation? oration appeared to be the most important cue for The aim of this experiment was to study whether predators in the former trial (see Results). The predators can take advantage of cues that derive number of pupae on each bedding site was 10. either from the cervids themselves (faecal pellets) The four study localities (Rantsila, Pulkkila, Uta- or from the interactions between the host and the järvi, and Yli-Ii) varied in relation to A. alces and parasite (snow discoloration), and if the response of L. cervi abundance, but the main avian predators, the predator is cumulative to these cues. We per- tits, were observed near the artificial bedding sites in formed this experiment in Rantsila, which had an each locality. In Utajärvi, the A. alces winter popula- abundant and heavily infested winter population of tion was quite similar in abundance to the one in A. alces at that time [proportion of infested A. alces, Rantsila, but L. cervi was less abundant in the former 0.96 (N = 46); average number of pupae on natu- locality [proportion of infested A. alces, 0.81 (N = 42) ral bedding sites, 7.46 ± 1.58 (95% CI); data from versus 0.96 (N = 46); average number of pupae on Välimäki et al., 2011]. We performed four different natural bedding sites, 1.86 ± 1.13 (95% CI) versus experimental manipulations using artificial bedding 7.46 ± 1.58 (95% CI); data from Välimäki et al., 2011]. sites with: (1) just pupae; (2) pupae and A. alces faecal In Yli-Ii, the A. alces winter population was scarce,

© 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 106, 275–286 280 S. KAUNISTO ET AL. and as an indication of that, we did not detect any of independent cues. We built the model so that it A. alces tracks during the experiment in the area. The allowed random intercepts for each quadrat as well as abundance of L. cervi was also very low, as neither quadrat-specific random slopes in relation to experi- the seven natural bedding sites found ~5 km east of mental manipulations, and interactions between the the study site nor the 12 additional bedding sites two. inspected in 2010 and 2011 within the study site Secondly, we tested possible spatial variation in the showed any signs of L. cervi infestation (only one liability of the most prominent environmental cue pupa was found on a natural bedding site of a semi- (snow discoloration) by fitting a generalized mixed- domesticated reindeer; N = 40). On the contrary, the effect model to the data from the study areas of study area of Pulkkila was very similar to Rantsila, Rantsila (two manipulations: only pupae; snow dis- both in terms of A. alces and L. cervi abundance [pro- coloration), Utajärvi, Yli-Ii, and Pulkkila (additional portion of infested A. alces, 0.90 (N = 10) versus 0.96 set-ups). The response variable was the frequency of (N = 46); average number of pupae on natural eaten and uneaten pupae at a particular moment in bedding sites, 5.60 ± 2.65 (95% CI) versus 7.46 ± 1.58 time on a particular artificial bedding site. The fixed (95% CI); data from Välimäki et al., 2011]. Thus, we factors were presence of snow discoloration and ended up with two independent replicates of areas natural infestation intensity (high/low). In addition, representing either the core (Pulkkila, Rantsila) or we set the continuous explanatory variable, which edge-of-range (Utajärvi, Yli-Ii) areas of L. cervi. These was time elapsed since the onset of the experiment. core or edge-of-range areas were characterized either We defined random effects by nesting random pairs of by high (i.e. high levels of snow discoloration on artificial bedding sites within random study areas so natural bedding sites) or by low (i.e. few or no areas that pair-specific random intercepts and random snow of snow discoloration on natural bedding sites) infes- discoloration ¥ time interactions were allowed. We tation intensity on A. alces, respectively. assessed the goodness-of-fit of the model by visual evaluation of residual plots in each case.

STATISTICAL ANALYSES To evaluate possible differences between the study RESULTS localities in tit occurrence, we conducted a chi-square We observed several tit species on both artificial and test. We reduced observations to presence/absence natural A. alces bedding sites in all five localities. data on tits as a group within a particular replicate The willow tit (Poecile montanus), the crested tit during the 5-day experiment. Secondly, we explored (Lophophanes cristatus), the great tit (Parus major), whether pupae are under predation in the first place, the blue tit (Cyanistes caeruleus), and the coal tit and applied the data from the first experiment con- (Periparus ater) were the species observed. We ducted in Konnevesi and Pulkkila. We tested whether observed tits to occur at roughly the same frequencies the percentages of bedding sites impacted by preda- in each study site (tit observations: Konnevesi 100%, tors vary between the two localities with a Fisher’s Pulkkila 85%, Rantsila 75%, Utajärvi 75%, and Yli-Ii exact test. The analyses were performed in SPSS 15.0 100% of all replicates; c2 = 9.032, d.f. = 4, P = 0.06). for Windows. We did not observe any signs of small mammalian We performed further statistical analyses with predators in the immediate vicinity of the artificial R 2.10.1 (R Development Core Team, 2009). We analy- bedding sites. sed the effects of exploitable cues that derive from the host itself (faeces) or from the host–parasite interaction (snow discoloration) with a generalized linear BEDDING SITES IMPACTED BY PREDATION mixed-effect model (function glmer) fitted with In Konnevesi and Pulkkila, 25 and 20% of the artifi- Laplace approximation, as implemented in the R cial bedding sites, respectively, were impacted by package lme4 (Bates & Maechler, 2009). We used predators during the 116-h trials (Table 1). The per- binomial error distribution with logistic link function. centage of bedding sites impacted by predation did First, we set the frequencies of eaten and uneaten not differ statistically between the two areas (Fisher’s pupae at the lowest level of hierarchy (the unit of 10 exact test, P = 1). All the pupae disappeared from two pupae in a particular quadrat–manipulation combi- bedding sites in Konnevesi and from one bedding site nation, measured 8, 24, 48, 72, and 116 h after the in Pulkkila. onset of the experiment) as the response variable (data from Rantsila). We set the incidence of snow discoloration and faecal pellets as fixed factors. We HOST-RELATED CUES AND PUPAL PREDATION included an interaction term between the fixed factors There was apparent variation in the percentage in the model to test the possible cumulative effect of artificial bedding sites impacted by predators

Table 2. Generalized linear mixed-effect model for the frequency of Lipoptena cervi pupae eaten and uneaten in relation to time on artiﬁcial bedding sites characterized by cues (snow discoloration and host faecal pellets) potentially exploitable by avian predators. The data is from the Rantsila population

Source of variation Parameter Estimate SE ZP

Pupal survival Intercept 10.317 3.021 3.415 0.0006 Discoloration -11.020 3.037 -3.629 0.0002 Faecal pellets 1.365 5.640 -0.214 NS Discoloration ¥ faecal pellets -1.209 5.640 -0.214 NS

The model included experimental quadrat-speciﬁc random intercepts with random snow discoloration ¥ faecal pellets interactions. NS, not signiﬁcant.

Figure 2. Proportion of uneaten Lipoptena cervi pupae (±95% CI) on artiﬁcial bedding sites in Rantsila at 0, 8, 24, 48, 72, and 116 h since the start of the experiment, in relation to an experimental manipulation of bedding sites.

in relation to manipulation [manipulation 1 (only 16.4–45.6 h) and the sites with both discoloration and pupae), 20%; manipulation 2 (pupae + faecal pellets), A. alces faecal pellets (33 h; 95% CI: 18.2–47.8 h). 15%; manipulation 3 (pupae + discoloration), 100%; manipulation 4 (pupae + snow discoloration + faecal pellets), 95%]. Each of the impacted bedding sites LIPOPTENA CERVI INFESTATION INTENSITY AND within manipulations 3 and 4 were totally emptied of PREDATION RATE OF PUPAE pupae, with the numbers of totally emptied bedding The frequency of uneaten pupae changed over time sites being one and three in manipulations 1 and 2, across the study areas of Utajärvi, Yli-Ii, Pulkkila, respectively. Snow discoloration on the artificial and Rantsila (Table 3). In this data set, the effect of bedding sites was the only cue with a significant effect snow discoloration on the pupal predation rate was on pupal predation rate (Table 2). Snow discoloration not straightforward, but depended on the prevailing increased the predation risk of pupae (Fig. 2). There L. cervi infestation intensity in natural A. alces popu- was neither a main effect of faecal pellets on the lations. This was indicated by the third-order inter- pupal predation rate nor significant interaction actions among discoloration, infestation intensity, and between the two factors (faecal pellets and snow dis- time (Table 3). In the study areas where L. cervi infes- coloration) (Table 2), with the latter indicating a lack tation intensity was relatively low (or zero), snow of cumulative effects of various cues. The time until discoloration only moderately increased the pupal all pupae were consumed did not differ between the predation rate on artificial bedding sites (Fig. 3A) bedding sites with discoloration alone (31 h; 95% CI compared with the apparent increase in the highly

Table 3. Generalized linear mixed-effect model for the number of Lipoptena cervi pupae eaten and uneaten in two experimental manipulations, in relation to natural infestation intensity on Alces alces and time elapsed since the beginning of the experiment. The data are from Utajärvi, Yli-Ii, Pulkkila, and Rantsila populations

Source of variation Parameter Estimate SE ZP

Pupal survival Intercept 4.4922 0.334 13.450 < 2.00 ¥ 10-16 Discoloration -3.16 0.2602 -12.146 < 2.00 ¥ 10-16 Infestation intensity 1.0973 0.5471 2.006 0.0449 Time -0.0208 0.005 -4.141 3.45 ¥ 10-5 Discoloration ¥ infestation intensity 0.3778 0.4683 0.807 NS Discoloration ¥ time -0.03559 0.0085 -4.189 2.80 ¥ 10-5 Infestation intensity ¥ time -0.011 0.0075 -1.463 NS Discoloration ¥ infestation intensity ¥ time 0.0418 0.0121 3.452 0.0006

Random effects were modelled so that random pairs of bedding sites were nested within random study areas (1–4), including pair-speciﬁc random intercepts and random snow discoloration ¥ time interactions.

Figure 3. Proportion of uneaten Lipoptena cervi pupae (±95% CI) on artiﬁcial bedding sites either with (solid lines) or without (dashed lines) snow discoloration at 0, 8, 24, 48, 72, and 116 h since the start of the experiment in areas with a low (A, Utajärvi and Yli-Ii) and a high (B, Pulkkila and Rantsila) natural L. cervi infestation intensity on Alces alces. infested areas (Fig. 3B). Moreover, the percentages of region to demonstrate that avian predation may affect impacted bedding sites with no snow discoloration in the population dynamics of ectoparasites feeding on the study areas showing a low infestation intensity large mammals. This result should be acknowledged (Utajärvi, 13%; Yli-Ii, 25%) were of the same magni- in future models when evaluating L. cervi–A. alces tude as in the heavily infested areas (Pulkkila, 19%; interactions in an evolutionary time scale. The Rantsila, 20%), but there was a difference concerning reddish brown snow discoloration on host bedding discoloured bedding sites (Utajärvi, 56%; Yli-Ii, 38%; sites, indicating high L. cervi infestation intensity Pulkkila, 94%; Rantsila, 100%). (see Kaunisto et al., 2009; Välimäki et al., 2011), and consequently a rewarding feeding site for tits, seems to increase the risk of pupal predation. DISCUSSION The assumption that avian species are potential Our results show that the predation pressure from predators of L. cervi holds true, as we witnessed tits on L. cervi pupae is notable during winter. Thus, several tit species (Poecile montanus, Lophophanes our data are among the ﬁrst in the Northern Boreal cristatus, Parus major, Cyanistes caeruleus, and

Periparus ater) foraging on pupae on either natural or regions where they feed on blood and reproduce (see artificial bedding sites. Tits are highly adaptable Paakkonen et al., 2010). The clumped distribution predators that can rapidly respond to the occurrence is probably the most common dispersion pattern of of a new abundant resource, especially invasive forest insects (Coulson & Witter, 1984), although or suddenly increasing insect populations (Pimentel some of L. cervi pupae may drop off a host randomly & Nilsson, 2007). During the winter, the tit guild outside the bedding sites. There is disagreement forages frequently in mixed-species flocks (Suhonen about whether avian predators can limit insect et al., 1994; Dolby & Grubb, 2000), which may explain numbers (Roland, Hannon & Smith, 1986). However, the fast and complete disappearance of the pupae birds can opportunistically exploit aggregations of from discoloured bedding sites once discovered. insects, and even at lower insect densities avian Indeed, interspecific interactions with other insec- predation can inflict notable mortality on an tivorous birds indicating high food availability can insect population (e.g. Schultz, 1983; Roland et al., increase predation on a specific prey (e.g. Forsman, 1986). Hjernquist & Gustafsson, 2009). Native avian preda- Lipoptena cervi pupae that drop off a host with tors can benefit for the expansion of an invasive insect relatively low infestation intensity are likely to escape in some cases (see e.g. Barber et al., 2008). In addition predation. Most of the artificial bedding sites without to the species observed, a few other passerine species snow discoloration (80%) remained undetected during foraging frequently on the ground and on snow in the a monitoring period of 116 h. This also held true in Northern Boreal forests may have been involved in the study areas where the infestation intensity of the predatory guild (see Alatalo, 1980). We stress L. cervi within the natural A. alces population was that in this study other potential predators such as relatively high (Konnevesi, Pulkkila, and Rantsila) voles, shrews, and mice are unlikely to be responsible (N = 92). Hence, it can be assumed that black pupae for pupal disappearance above the snow layer. This against a white background do not provide enough is because we did not observe any signs of those information for tits, and that clean, non-coloured species in the immediate vicinity of the artificial bedding sites are merely foraged upon incidentally. bedding sites, and small mammals spend most of Alternatively, tits can discover non-coloured bedding the winter beneath the snow anyway (e.g. Hansson & sites, but a lack of information on the profitability of Henttonen, 1985; Aitchison, 1987). the site leads the predators to reject it. Even when The A. alces faecal pellets studied as a possible cue clean bedding sites were found, predation on them for predators did not increase the risk of pupal pre- took place at earliest after 48 h. Weather conditions dation. The lack of response is understandable, as may change even within a day in winter, when pupae host faeces may not reliably indicate the presence of are exposed to predation for relatively short times prey items on a bedding site, unlike the commonly until being covered by snow. Consequently, a time observed snow discoloration that derives from the lag of 48 h is enough to decrease predation risk direct host–parasite interaction. Indeed, the reported considerably. reddish brown snow discoloration of A. alces bedding In the heavily infested study areas (Rantsila and sites (see Kaunisto et al., 2009) significantly increased Pulkkila), 39 and 63% of discoloured bedding sites the probability of pupae being predated upon. In (N = 56) were entirely emptied by predators within 8 addition to being a visual cue, discoloured snow may and 24 h, respectively. According to these results, offer olfactory stimuli. Avian olfaction is still poorly L. cervi pupae were soon consumed by predators once understood, but it may be a more important sense they had been discovered. Hence, relatively low pupal than is generally believed (see Mäntylä et al., 2008; loss in the two treatments without snow discoloration Steiger et al., 2008). indicates that potential predators had not discovered An increase in pupal predation with increasing these bedding sites, rather than the L. cervi pupae infestation density and degree of snow discoloration being unpalatable prey. It has been shown that tits may suggest an anti-apostatic selection, meaning forage according to the optimal foraging theory (see over-predation on rare prey types at high densities Krebs et al., 1977; Koivula, Rytkönen & Orell, 1995; rather than an apostatic selection (over-predation on Brotons, 1997). For example, tits can adjust their common prey types) (see Allen & Anderson, 1984). In foraging behaviour on the basis of prey density and the natural bedding sites of A. alces, L. cervi pupae quality (e.g. Naef-Daenzer, Naef-Daenzer & Nager, can most probably be observed on the spots of 2000). Potentially, tits may use discoloration to evalu- discoloured snow, near the neck and back regions ate expected nutritional gain and adjust their time of a bedding site (S. Kaunisto & P. Välimäki, pers. expenditure within a particular bedding site adap- observ.). The clumped distribution of pupae on the tively. In line with this scenario, a group of pupae bedding sites is caused by the aggregation of adult deposited on experimental bedding sites without snow L. cervi on the host, favouring the neck and back discoloration were relatively frequently only partially

© 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 106, 275–286 284 S. KAUNISTO ET AL. consumed (70% of impacted clean bedding sites), even but later on, when the parasite population has estab- in the heavily infested areas where discoloured bed- lished properly, tit predation could strongly affect the dings sites were regularly emptied as a whole. local population dynamics. This would occur if birds The avian predators may have learned to associate learn to take advantage of cues that derive from snow discoloration with a potential food source in the host–parasite interaction. Even then, hosts that areas where overwintering A. alces and L. cervi are harbour parasites with low infestation intensity could abundant. Birds may also have an inherent tendency ensure the sustainability and spread of the parasite to be attracted to signals like blood containing population. Before firm conclusions on the importance UV-fluorescent molecules (see Bennett & Cuthill, of pupal predation could be drawn, analyses of 1994; Viitala et al., 1995; Church et al., 1998). Our tit diets and on-host mortality of L. cervi are war- results, however, do not support this latter possibility, ranted. Nevertheless, pupal predation should not as discoloration did not have such a large effect on be ignored when evaluating the three- or two-way pupal predation in Yli-Ii and Utajärvi, where the interactions among L. cervi, A. alces, and the preda- overwintering A. alces populations were not heavily tory tits. infested, resulting in a relatively low degree of discoloration on natural bedding sites. Our results indicate that predator response to snow discoloration is ACKNOWLEDGEMENTS likely the result of phenotypic plasticity, where individuals fine-tune their behaviour to match the pre- We thank the volunteer collectors of L. cervi pupae vailing environment (e.g. Carlsson et al., 2009). There used in the experiments. We thank Konnevesi are several mechanisms by which native predators Research Station and the staff for help in the field. We may become better in exploiting novel species as prey, also thank nine anonymous referees for substantially including the formation of a search image, associative improving our contribution. This study was partly learning and social transmission (e.g. Krebs & funded by the Jenny and Antti Wihuri Foundation, the Davies, 1993). The observed pattern may also result Ella and Georg Ehrnrooth Foundation, the Alfred from the spatial variation in tit abundance that was Kordelin Foundation, Societas pro Fauna et Flora not rigorously assessed, and thus some caution is Fennica, the Finnish Ministry of Agriculture and warranted. It is worth noting, however, that tits were Forestry, BIOINT Graduate School, and the Finnish observed in the same frequencies within each study Cultural Foundation. We conducted this study in site, and the most parsimonious explanation for equal accordance with Finnish legislation. predation upon pupae deposited on non-coloured bedding sites would be equal predator density. We provide the first estimations about the magni- REFERENCES tude of predation on the novel invading prey, L. cervi, in the Northern Boreal region during winter. Our Aitchison CW. 1987. Review of winter trophic relations of results share a high resemblance to natural condi- soricine shrews. Mammal Review 17: 1–24. tions, as each experiment was conducted in the field Alatalo RV. 1980. Seasonal dynamics of resource partitioning with natural predator guilds and the extent of experi- among foliage-gleaning passerines in northern Finland. mental manipulations fell within the natural varia- Oecologia 45: 190–196. tion of A. alces bedding characteristics in areas with Alatalo RV. 1982. Multidimensional foraging niche organiza- moderate L. cervi infestation intensity. As 34.7% of tion of foliage-gleaning birds in northern Finland. Ornis all pupae used in the experiments were predated Scandinavica 13: 56–71. upon, even within a relatively short time interval, we Allen JA, Anderson KP. 1984. Selection by passerine birds argue that pupal predation during winter can be of is anti-apostatic at high prey density. Biological Journal of the Linnean Society 23: 237–246. importance. However, avian predation is unlikely to Barbaro L, Battisti A. 2011. Birds as predators of the pine prevent the parasite from spreading into new areas, processionary moth (Lepidoptera: Notodontidae). Biological because predation on the bedding sites of hosts Control 56: 107–114. with low infestation intensity, and thus without Barber NA, Marquis RJ, Tori WP. 2008. Invasive prey remarkable snow discoloration, seems coincidental impacts the abundance and distribution of native predators. and ineffective. In addition, the predation risk for an Ecology 89: 2678–2683. individual pupa that drops off of a moving cervid host Bates D, Maechler M. 2009. lme4. Linear mixed-effects is likely to be marginal, unless avian predators learn models using S4 classes. R Package, Version 0.999375-32. to follow the movements of parasitized hosts (see, Available at: http://cran.r-project.org/web/packages/lme4/ e.g., Haarløv, 1964; Samuel, Mooring & Aalangdong, index.html 2000). In general, the parasite infestation intensity is Bennett ATD, Cuthill IC. 1994. Ultraviolet vision in birds: likely to be low during the early phase of invasion, what is its function? Vision Research 34: 1471–1478.

Bequaert JC. 1953. The Hippoboscidae or louse-flies (Diptera) spiders and insects in spruce Picea abies and its relation to of mammals and birds. Part I. Structure, physiology and predation by birds. Ecography 4: 82–93. natural history. Entomologica Americana 32: 1–209. Kaunisto S, Härkönen L, Niemelä P, Roininen H, Brotons L. 1997. Changes in foraging behaviour of the coal Ylönen H. 2011. Northward invasion of the parasitic deer tit Parus ater due to snow cover. Ardea 85: 249–257. ked (Lipoptena cervi), is there geographical variation in Brotons L, Herrando S. 2003. Effect of increased food abun- pupal size and development duration? Parasitology 138: dance near forest edges on flocking patterns of coal tit Parus 354–363. ater winter groups in mountain coniferous forests. Bird Kaunisto S, Kortet R, Härkönen L, Härkönen S, Ylönen Study 50: 106–111. H, Laaksonen S. 2009. New bedding site examination- Carlsson NOL, Sarnelle O, Strayer DL. 2009. Native based method to analyse deer ked (Lipoptena cervi) infection predators and exotic prey – an acquired taste? Frontiers in in cervids. Parasitology Research 104: 919–925. Ecology and the Environment 7: 525–532. Koivula K, Rytkönen S, Orell M. 1995. Hunger-dependency Church SC, Bennett ATD, Cuthill IC, Partridge JC. of hiding behaviour after a predator attack in dominant and 1998. Ultraviolet cues affect the foraging behaviour of blue subordinate willow tits. Ardea 83: 397–404. tits. Proceedings of the Royal Society B: Biological Sciences Koivula M, Korpimäki E. 2001. Do scent marks increase 265: 1509–1514. predation risk of microtine rodents? Oikos 95: 275–281. Coulson RN, Witter JA. 1984. Forest entomology: ecology Kortet R, Härkönen L, Hokkanen P, Härkönen S, and management. New York: John Wiley & Sons, Inc. Kaitala A, Kaunisto S, Laaksonen S, Kekäläinen J, Dehio C, Sauder U, Hiestand R. 2004. Isolation of Ylönen H. 2010. Experiments on the ectoparasitic deer ked Bartonella schoenbuchensis from Lipoptena cervi, a blood- that often attacks humans: preferences for body parts, sucking arthropod causing deer ked dermatitis. Journal of colour and temperature. Bulletin of Entomological Research Clinical Microbiology 42: 5320–5323. 100: 279–285. Dolby AS, Grubb TC. 2000. Social context affects risk taking Kotiaho JS, Sulkava P. 2007. Effects of isolation, area and by a satellite species in a mixed-species foraging group. predators on invasion: a field experiment with artificial Behavioral Ecology 11: 110–114. islands. Applied Soil Ecology 35: 256–259. Forsman JT, Hjernquist MB, Gustafsson L. 2009. Krebs JR, Davies NB. 1993. An introduction to behavioural Experimental evidence for the use of density based inter- ecology, 3rd edn. Oxford: Blackwell Science Ltd. specific social information in forest birds. Ecography 32: Krebs JR, Erichsen JT, Webber MI, Charnov EL. 1977. 539–545. Optimal prey selection in the great tit (Parus major). Fowler AC, Knight RL, George TL, McEwen LC. 1991. Animal Behaviour 25: 30–38. Effects of avian predation on grasshopper populations in Kynkäänniemi SM, Kortet R, Härkönen L, Kaitala A, North Dakota grasslands. Ecology 72: 1775–1781. Paakkonen T, Mustonen AM, Nieminen P, Härkönen Frank JH. 1967. The effect of pupal predators on a popula- S, Ylönen H, Laaksonen S. 2010. Threat of an invasive tion of winter moth, Operophtera brumata (L.) (Hydri- parasitic fly, the deer ked (Lipoptena cervi), to the reindeer omenidae). Journal of Animal Ecology 36: 611–621. (Rangifer tarandus tarandus): experimental infection and Haarløv N. 1964. Life cycle and distribution pattern of treatment. Annales Zoologici Fennici 47: 28–36. Lipoptena cervi (L.) (Dipt., Hippobosc.) on Danish deer. Lahti K, Orell M, Rytkönen S, Koivula K. 1998. Time and Oikos 15: 93–129. food dependence in willow tit winter survival. Ecology 79: Hackman W, Rantanen T, Vuojolahti P. 1983. Immigra- 2904–2916. tion of Lipoptena cervi (Diptera, Hippoboscidae) in Finland, Lockwood JL, Hoopes MF, Marchetti MP. 2007. Invasion with notes on its biology and medical significance. Notulae ecology. Oxford: Blackwell Publishing. Entomologicae 63: 53–59. Maa TC. 1969. A revised checklist and concise host index of Hansson L, Henttonen H. 1985. Gradients in density varia- Hippoboscidae (Diptera). Pacific Insects Monograph 20: tions of small rodents: the importance of latitude and snow 261–299. cover. Oecologia 67: 394–402. Mäntylä E, Alessio GA, Blande JD, Heijari J, Härkönen L, Härkönen S, Kaitala A, Kaunisto S, Kortet Holopainen JK, Laaksonen T, Piirtola P, Klemola T. R, Laaksonen S, Ylönen H. 2010. Predicting range expan- 2008. From plants to birds: higher avian predation rates sion of an ectoparasite – the effect of spring and summer in trees responding to insect herbivory. PLoS ONE 3: temperatures on deer ked Lipoptena cervi (Diptera: Hippo- 2832. boscidae) performance along a latitudinal gradient. Ecogra- Matsumoto K, Berrada ZL, Klinger E, Goethert HK, phy 33: 906–912. Telford SR III. 2008. Molecular detection of Bartonella Hastings FL, Hain FP, Smith HR, Cook SP, Monahan schoenbuchensis from ectoparasites of deer in Massachu- JF. 2002. Predation of gypsy moth (Lepidoptera: Lymantri- setts. Vector-Borne and Zoonotic Diseases 8: 549–554. idae) pupae in three ecosystems along the southern edge of Menke SB, Fisher RN, Jetz W, Holway DA. 2007. Biotic infestation. Environmental Entomology 31: 668–675. and abiotic controls of Argentine ant invasion success at Heinrich B, Collins SL. 1983. Caterpillar leaf damage, and local and landscape scales. Ecology 88: 3164 –3173. the game of hide-and-seek with birds. Ecology 64: 592–602. Montgomerie R, Weatherhead PJ. 1997. How robins find Jansson C, von Brömssen A. 1981. Winter decline of worms? Animal Behaviour 54: 143–151.

Murakami M. 1999. Effect of avian predation on survival of Shwartz A, Strubbe D, Butler CJ, Matthysen E, Kark S. leaf-rolling lepidopterous larvae. Researches on Population 2009. The effect of enemy-release and climate conditions on Ecology 41: 135–138. invasive birds: a regional test using the rose-ringed para- Naef-Daenzer L, Naef-Daenzer B, Nager RG. 2000. Prey keet (Psittacula krameri) as a case study. A Journal of selection and foraging performance of breeding great tits Conservation Biogeography 15: 310–318. Parus major in relation to food availability. Journal of Steiger SS, Fidler AE, Valcu M, Kempenaers B. 2008. Avian Biology 31: 206–214. Avian olfactory receptor gene repertoires: evidence for a Nevitt GA, Losekoot M, Weimerskirch H. 2008. Evidence well-developed sense of smell in birds? Proceedings of the for olfactory search in wandering albatross, Diomedea Royal Society B: Biological Sciences 275: 2309–2317. exulans. Proceedings of the National Academy of Sciences of Suhonen J. 1993. Predation risk influences the use of forag- the United States of America 105: 4576–4581. ing sites by tits. Ecology 74: 1197–1203. Paakkonen T, Mustonen AM, Roininen H, Niemelä P, Suhonen J, Alatalo RV, Gustafsson L. 1994. Evolution of Ruusila V, Nieminen P. 2010. Parasitism of the deer ked, foraging ecology in Fennoscandian tits (Parus spp.). Pro- Lipoptena cervi, on the moose, Alces alces, in eastern ceedings of the Royal Society B: Biological Sciences 258: Finland. Medical and Veterinary Entomology 24: 411–417. 127–131. Pimentel C, Nilsson JÅ. 2007. Response of great tits Parus Tanhuanpää M, Ruohomäki K, Kaitaniemi P, Klemola T. major to an irruption of a pine processionary moth Thau- 1999. Different impact of pupal predation on populations of V metopoea pityocampa population with a shifted phenology. Epirrita autumnata (Lepidoptera; Geometridae) within and Ardea 95: 191–199. outside the outbreak range. Journal of Animal Ecology 68: Popov AV. [????? A?] 1965. The life cycle of louse-flies 562–570. Do small mammals prey upon an invasive ectoparasite of Lipoptena cervi L. and Stenepteryx hirundinis L. (Diptera, Välimäki P, Kaitala A, Madslien K, Härkönen L, Várko- Hippoboscidae). Revue d’Entomologie de l’URSS XLIV: 573– nyi G, Heikkilä J, Jaakola M, Ylönen H, Kortet R, cervids? 583. (in Russian). Ytrehus B. 2011. Geographical variation in host use of a R Development Core Team. 2009. R: a language and envi- blood-feeding ectoparasitic fly: implications for population ronment for statistical computing. R Foundation for Statis- invasiveness. Oecologia 166: 985–995. Sirpa Kaunisto, Raine Kortet, Sauli Härkönen, Arja Kaitala, tical Computing, Vienna, Austria. Available at: http:// Välimäki P, Madslien K, Malmsten J, Härkönen L, www.R-project.org. Härkönen S, Kaitala A, Kortet R, Laaksonen S, Mehl Sauli Laaksonen and Hannu Ylönen Rajchard J. 2009. Ultraviolet (UV) light perception by birds: R, Redford L, Ylönen H, Ytrehus B. 2010. Fennoscan- a review. Veterinarni Medicina 54: 351–359. dian distribution of an important parasite of cervids, the Rantanen T, Reunala T, Vuojolahti P, Hackman W. 1982. deer ked (Lipoptena cervi), revisited. Parasitology Research Canadian Journal of Zoology 90: 1044–1050, 2012 Persistent pruritic papules from deer ked bites. Acta 107: 117–125. Dermato-Venereologica 62: 307–311. Van Ballenberghe V, Miquelle DG. 1990. Activity of moose Roland J, Hannon SJ, Smith MA. 1986. Foraging pattern during spring and summer in interior Alaska. The Journal of pine siskins and its influence on winter moth survival in of Wildlife Management 54: 391–396. Reprinted with the kind permission of the NRC Research Press an apple orchard. Oecologia 69: 47–52. Vaughan TA, Ryan JM, Czaplewski NJ. 2011. Mammal- Samuel WM, Mooring MS, Aalangdong OI. 2000. Adapta- ogy, 5th edn. Ontario: Jones and Bartlett Publishers. tions of winter ticks (Dermacentor albipictus) to invade Vermeij GJ. 1996. An agenda for invasion biology. Biological moose and moose to evade ticks. Alces 36: 183–195. Conservation 78: 3–9. Schoener TW, Spiller DA. 1995. Effect of predators and Viitala J, Korpimäki E, Palokangas P, Koivula M. 1995. area on invasion: an experiment with island spiders. Science Attraction of kestrels to vole scent marks visible in ultra- 267: 1811–1813. violet light. Nature 373: 425–427. Schultz TD. 1983. Opportunistic foraging of Western king- Williamson M. 1996. Biological Invasions. London: Chapman birds on aggregations of tiger beetles. Auk 100: 496–497. & Hall.

SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article: Appendix S1. A schematic view of the experimental designs used at the different study sites. Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

Do small mammals prey upon an invasive ectoparasite of cervids?

Sirpa Kaunisto, Raine Kortet, Sauli Härkönen, Arja Kaitala, Sauli Laaksonen and Hannu Ylönen

Canadian Journal of Zoology 90: 1044–1050, 2012

Reprinted with the kind permission of the NRC Research Press

1044 Do small mammals prey upon an invasive ectoparasite of cervids?

Sirpa Kaunisto, Raine Kortet, Sauli Härkönen, Arja Kaitala, Sauli Laaksonen, and Hannu Ylönen

Abstract: Predation is often considered an important factor for population regulation and in some cases for the invasion success of prey. Small mammalian predation may be a major force in the population regulation of many ground-dwelling invertebrate species. The deer ked (Lipoptena cervi (L., 1758)) is an ectoparasitic fly of cervids. The species has a large distribution area and it has relatively rapidly spread in northern Europe during the previous four decades. The factors possibly regulating the distribution and invasion of this fly are poorly known. During the off-host stage of several months, pupae of deer ked are likely exposed to many ground-dwelling predators. To study whether small mammals would feed on deer keds, we conducted experiments by serving pupae of deer ked to wild-captured common shrews (Sorex araneus L., 1758), bank voles (Myodes glareolus (Schreber, 1780)), field voles (Microtus agrestis (L., 1761)), and semi-wild bank voles, and assessed pupal survival. As a control, we provided alternative food including common nutrients used by small mammals in their natural habitats. The results show that variable amounts of pupae of deer ked are consumed by all small-mammal species studied. Surprisingly, insectivorous and most of the time food-constrained shrews consumed less pupae than granivorous–herbivorous voles. Key words: Cervidae, ectoparasite, foraging, Hippoboscidae, invasion, Lipoptena cervi, deer ked, predator, small mammal. Résumé : La prédation est souvent considérée comme un facteur important pour la régulation des populations et, dans cer- tains cas, pour le succès d’invasion de proies. La prédation par de petits mammifères peut jouer un rôle majeur dans la régu- lation des populations de nombreuses espèces d’invertébrés terricoles. La mouche du cerf (Lipoptena cervi (L., 1758)) est un ectoparasite des cervidés. Son aire de répartition est vaste, et elle a connu une propagation relativement rapide en Europe du Nord au cours des quatre dernières décennies. Les facteurs qui pourraient réguler la répartition et l’invasion de cette mouche demeurent méconnus. Durant le stade hors de l’hôte qui dure plusieurs mois, les pupes de mouche du cerf sont pro- bablement exposées à de nombreux prédateurs terricoles. Afin de déterminer si de petits mammifères se nourrissent de mou- ches du cerf, nous avons fait des expériences dans lesquelles des pupes de mouche du cerf ont été servies à des

For personal use only. musaraignes communes (Sorex araneus L., 1758), des campagnols roussâtres (Myodes glareolus (Schreber, 1780)), des campagnols agrestes (Microtus agrestis (L., 1761)) et des campagnols roussâtres semi-sauvages et nous avons évalué la survie des pupes. À titre d’expérience témoin, nous avons servi d’autres nourritures dont des aliments couramment utilisés par les petits mammifères dans leurs habitats naturels. Les résultats indiquent que toutes les espèces de petits mammifères étudiées consomment des quantités variables de pupes de mouche du cerf. Fait surprenant, les musaraignes, qui sont insectivores et dont la nourriture est, la plupart du temps, limitée, ont consommé moins de pupes que les campagnols granivores– herbivores. Mots‐clés : cervidés, ectoparasite, quête de nourriture, hippoboscidés, invasion, Lipoptena cervi, mouche du cerf, prédateur, petit mammifère. [Traduit par la Rédaction]

Introduction tion in ecosystems that have been recently occupied by a new species. According to the enemy release hypothesis, in- Predation is one of the key factors in the population regu- vasive species may encounter fewer natural enemies, like Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of Eastern Finland on 08/28/12 lation of prey species (Pimm 1989; Lodge 1993; Vermeij predators, in novel environments resulting in rapid increase 1996; Lockwood et al. 2007; Price et al. 2011). Invasive spe- in population size (e.g., Pimm 1991; Schoener and Spiller cies provide an interesting opportunity to study this regula- 1995; Holway et al. 1998; reviewed in Davis 2009). Cogni-

Received 16 March 2012. Accepted 14 June 2012. Published at www.nrcresearchpress.com/cjz on 1 August 2012. S. Kaunisto and R. Kortet. University of Eastern Finland, Department of Biology, P.O. Box 111, FI-80101 Joensuu, Finland. S. Härkönen. Finnish Wildlife Agency, Fantsintie 13-14, FI-00890 Helsinki, Finland. A. Kaitala. University of Oulu, Department of Biology, P.O. Box 3000, FI-90014 Oulu, Finland. S. Laaksonen. University of Helsinki, Department of Veterinary Biosciences, Nurminiementie 2, FI-93600 Kuusamo, Finland. H. Ylönen. University of Jyväskylä, Department of Biological and Environmental Science, Konnevesi Research Station, P.O. Box 35, FI- 40014 Jyväskylä, Finland. Corresponding author: Sirpa Kaunisto (e-mail: [email protected]).

Can. J. Zool. 90: 1044–1050 (2012) doi:10.1139/Z2012-072 Published by NRC Research Press Kaunisto et al. 1045

tive abilities of predators to exploit new prey resources and predators for the deer ked. Bank vole is regarded as being evolution processes leading to balanced predator–prey inter- granivorous–omnivorous (Hansson 1971, 1985; Eccard and action need time frames, which differ in length (Van Baalen Ylönen 2006) and thus may consume pupae of deer ked as et al. 2001; Carlsson et al. 2009). In contrast, predators may well, but herbivorous field voles (Hansson 1971) would pre- rapidly be able to target invaders, particularly if an invasive fer the alternative plant food provided. The present work is species lacks appropriate counter-adaptations against novel the basis for future studies on the interaction between enemies, thereby potentially impeding invasion. There are ground-dwelling predators and the deer ked. several examples in the literature that native predator species switch to consume novel species (see review in Carlsson et Materials and methods al. 2009). These are the two opposite endings of the continuum. Especially during summer time in northern temperate Natural history of the deer ked areas, there are many invertebrate prey alternatives available The deer ked is a Palearctic species, which has been intro- for predators. Thus, predation pressure may not be particu- duced also to the Nearctic region (e.g., Maa 1969; Dehio et larly high against one single prey species, which can enable al. 2004; Matsumoto et al. 2008). In Finland, the deer ked its establishment and even invasion (Holt 1977; Van Baalen has spread ca. 700 km north up to the southernmost part of et al. 2001; Koss and Snyder 2005). the reindeer husbandry region (Hackman et al. 1983; Väli- Small mammalian predation may be a major force in the mäki et al. 2010). The moose (Alces alces (L., 1758)) is the population regulation of many terrestrial invertebrate species, most common breeding host (i.e., support reproduction) in communities of small mammals having a notable impact on Fennoscandia (Välimäki et al. 2011). A single moose bull prey populations (e.g., Holling 1959; Hanski and Parviainen may host as much as 17 500 blood-feeding deer keds (Paak- 1985; Churchfield et al. 1991). Predation on pupal stage of konen et al. 2010). Also, the wild forest reindeer (Rangifer insects has been studied in certain species both with labora- tarandus fennicus Lönnberg, 1909) (Kaunisto et al. 2009) tory and field experiments (e.g., Frank 1967a, 1967b; Hanski and occasionally the semi-domestic reindeer (Rangifer taran- and Parviainen 1985; Hastings et al. 2002). Small mammals dus tarandus (L., 1758)) may serve as hosts for this ecto- are notable predators on pupal stage of different species such parasite (Kynkäänniemi et al. 2010). In central Europe, the as the autumnal moth (Epirrita autumnata (Borkhausen, main breeding hosts are red deer (Cervus elaphus L., 1758) 1794)) and the winter moth (Operophtera brumata (L., and roe deer (Capreolus capreolus (L., 1758)), and, to a 1758)) (e.g., Frank 1967b; Tanhuanpää et al. 1999). In north- lesser extent, fallow deer (Dama dama (L., 1758)) (Haarløv ern temperate areas, nonhibernating small mammals like 1964; Dehio et al. 2004). Evidence of white-tailed deer bank voles (Myodes glareolus (Schreber, 1780)) and soricine (Odocoileus virginianus (Zimmermann, 1780)) as a breeding shrews feed on small arthropods throughout the year, includ- host in Finland is uncertain, but in North America the deer ing the winter period in the subnivean space (see Hanski and ked is known to parasitize on white-tailed deer with relatively Parviainen 1985; Churchfield 2002; von Blanckenhagen et al. low prevalence (Matsumoto et al. 2008; Välimäki et al.

For personal use only. 2007). 2011). The deer ked has recently been of high interest, be- Our study species, the deer ked (Lipoptena cervi (L., cause it can cause negative physiological and behavioural 1758)), is a blood-feeding ectoparasite on cervids (Bequaert changes in cervids, as well as allergic reactions and incon- 1953). Invasion of this louse fly (Hippoboscidae) has been venience in humans (Rantanen et al. 1982; Laukkanen et al. relative rapid in northeastern Fennoscandia during previous 2005; Kortet et al. 2010; Kynkäänniemi et al. 2010). four decades (Välimäki et al. 2010; Kaunisto et al. 2011). Adult emergence from pupa takes place in late summer or The life cycle of the deer ked includes obligate host- early autumn within ground litter, and after finding a suitable dependent and off-host stages. The off-host pupal stage is reproductive host, adult deer ked drops off its wings and likely exposed to extrinsic mortality factors such as predators stays the rest of its life on the same cervid host (Haarløv on the ground layer for several months. In the present study, 1964; Hackman et al. 1983). Adults feed regularly on the we explored whether the deer ked has small mammalian blood of the host and viviparous (see Meier et al. 1999) fe- predators that could potentially affect the population of deer males produce repetitively only one, already prepupated larva ked in the northern Boreal region. To date, there is only one at a time through the winter and spring. Complete blackish study on avian predation on the deer ked (Kaunisto et al. and oval-shaped pupae (on average, diameter 3 mm, mass 6– 2012), even though predation was proposed as a possible reg- 13 mg; see also Kaunisto et al. 2011) drop off the host onto

Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of Eastern Finland on 08/28/12 ulative factor for the deer ked almost 50 years ago by the ground or snow in forest and field habitats, with off-host Haarløv (1964). At first, we investigated whether pupae of pupal stage lasting as long as 10 months (depending on sea- deer ked are consumed by three small-mammal species; the son when pupae are produced). bank vole, the field vole (Microtus agrestis (L., 1761)), and In the present study, we focused on potential small mam- soricine shrews, namely the common shrew (Sorex araneus malian predators on pupae of deer ked in forest and field L., 1758). These three species are among the most common habitats such as bank and field voles. We studied also insec- mammalian species in Finland, inhabiting forest and field tivorous soricines like the common shrew that inhabit mainly habitats (Sundell et al. 2012). Thus, these species could po- forest ground layer in Finland and are thus among potentially tentially exploit pupae of deer ked as a food resource. Sec- important predators. The main host species of the deer ked in ondly, we studied whether pupae are still consumed even if Finland, the moose (Välimäki et al. 2011), prefers forest there is alternative food present. If yes, this would indicate clear-cut and young forest regeneration areas as overwinter- that pupae are more or less the preferred food. We hypothe- ing sites (Heikkinen 2000). These habitats likely have very sized that insectivorous shrews would be the most likely high prevalences of pupae of deer ked, owing to frequent

Published by NRC Research Press 1046 Can. J. Zool. Vol. 90, 2012

visit rates of hosts. All the small-mammal species mentioned treatment, we put the food pellet and fresh grass onto the above co-occur in these habitats (Sundell et al. 2012), provid- floor of the cage, near the pupal dish, and randomized the ing a potential predator guild for pupal stage of deer ked. order of different food alternatives between the cages. The experiment lasted for 12 h in both treatments. Common shrew as potential predator To study whether shrews feed on pupae of deer ked, we Bank vole as potential predator captured shrews in the wild with Ugglan vole or shrew traps We used bank voles both from the laboratory colony and in July and August 2009. We captured 35 common shrews in from the wild in our experiments in August 2008. Voles their natural forest habitats in Konnevesi (62°41′N, 26°16′E), from the laboratory colony were captured from the wild in located in central Finland. Also moose and deer ked had early summer and were kept in the laboratory for 4–6 weeks. been observed in the same area during previous years and Wild bank voles were captured from the Konnevesi area on the year when the current study was performed. The feeding previous days and they had been in cages for 1–2 days before experiment took place outdoors immediately after capture be- our experiment. For the experiment, we placed the voles one cause the soricine shrews are forced to feed very often owing by one into similar cages as those used in the other treat- to their high metabolic rate, low energy storage capacity, and ments (see above) with a drinking bottle for water. We used thus low starvation tolerance (Vogel 1976; Hanski 1984; a total of 42 laboratory voles and 19 wild-caught voles that Sparti 1992; McDevitt and Andrews 1995). We placed the were divided equally into both treatments. shrews individually into plastic cages (42 cm × 25 cm × In the first treatment, only five pupae of deer ked were 15 cm) covered with metal lids during the experiment. available. In the second treatment, five pupae of deer ked There were two treatments. In the first treatment, we pro- and one Labfor® food pellet were available. The feeding trials vided only five pupae for the shrews. In the second treat- lasted for 19 h during which we observed the survival of pu- ment, we provided five pupae and an alternative food, which pae at regular intervals. We focused only on deer keds used was a fresh piece of red worm (Eisenia fetida (Savigny, as food and did not measure consumption of the alternative 1826)). We chose this alternative food because worms are food in the second treatment. very common nutriments used by the shrews in their natural habitats (Edwards and Bohlen 1996). The piece of red worm Statistical analyses was approximately similar in size to the five pupae. We We used Kaplan–Meier survival analysis to study possible served the food in small Petri dishes (diameter 5 cm, height effect of the treatment on pupal survival time in common 8 mm), all five pupae in one dish and the small piece of red shrews and field voles. We set treatment (pupae vs. pupae + worm in the other dish. We also served water in one dish in alternative food) as a factor. We used mean survival time of both treatments, because shrews cannot use drinking bottles all five pupae served to one small mammal owing to the de- like voles (see below). We randomized the position of the pendence of five pupae. We set the status values as 0 (pupae Petri dishes between different cages. All pupae were in the still left at the end of the experiment) and 1 (no pupae left For personal use only. same dish to avoid their rolling and to minimize errors dur- and all eaten at the end of the experiment) and censored the ing observation periods. We also added a bunch of grass pupae which survived in both treatments. into the cages to minimize stress and stress-related behaviour To study possible pupal consumption by bank voles, we of the shrews. pooled the data on wild bank voles and wild bank voles with We divided shrews randomly into the treatment groups (17 longer laboratory history using Cox regression analysis. In shrews in the first treatment and 18 shrews in the second the Cox regression analysis, we set the mean survival time treatment). Because shrews are very sensitive to stress, ex- of five pupae served to one bank vole as a dependent factor. periments were allowed to proceed for a maximum of 1 h. Treatment (pupae vs. pupae + alternative food) and back- We regularly conducted observations on possible consump- ground of bank voles (wild vs. wild with laboratory history) tion of pupae and alternative food. After the experiments, we were categorical covariates. We set the status values as 0 (pu- released the shrews back to nature at least 1 km from new pae still left at the end of the experiment) and 1 (no pupae capturing sites to minimize pseudoreplication (i.e., capturing left and all eaten at the end of the experiment) and censored the same animals). the pupae which survived in both treatments. To detect whether the shrews and the field voles prefer pu- Field vole as potential predator pae of deer ked or alternative food in the second treatment,

Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of Eastern Finland on 08/28/12 We conducted a similar feeding experiment with the field we used a c2 test (the cells had expected counts greater than voles as we did with the shrews in August 2009. We cap- 5). In bank voles (the data on bank voles with some labora- tured a total of 36 field voles, 18 individuals for both treat- tory history and wild bank voles were pooled), we used a c2 ments. We captured field voles in their natural field habitats test to test for possible differences in the amount of eaten pu- in Konnevesi by using Ugglan vole traps. For the experiment, pae between the two treatments. For all statistical tests, we we put the field voles individually into plastic cages used SPSS for Windows version 17 (SPSS, Inc., Chicago, Il- (42 cm × 25 cm × 15 cm) covered by metal lids with a linois, USA). drinking bottle for water. In the first treatment, we provided five pupae in a small Results Petri dish (diameter 5 cm, height 8 mm). In the second treatment, we provided five pupae and an alternative food, which Common shrew as potential predator was a commercial rodent food pellet (Labfor®, mean mass of Pupal survival time, i.e., pupal consumption, in the first the pellet 7.7 g) and a handful of fresh grass. In the second treatment (survival time 42.8 ± 5.5 min (mean ± SE)) did

Published by NRC Research Press Kaunisto et al. 1047

not differ from that in the second treatment with alternative Fig. 1. Survival times (min) of the pupae of deer ked (Lipoptena food (39.9 ± 5.5 min), as indicated by the results of the sur- cervi) in two treatments (only pupae vs. pupae + a piece of red vival analysis (log rank = 0.25, df = 1, p = 0.62) (Fig. 1). worm (Eisenia fetida)) in the experiment with the common shrews Hence, the pupal consumption was similar whether the (Sorex araneus). shrews had access to alternative food or not. In both treatments, at least half of the shrews did not eat pupae at all (Ta- ble 1). In the second treatment, lower proportion of shrews 2 fed on the pupae than on the red worm (c 1 = 4.5, p = 0.034) (Table 1). Only three shrews did not½ eat� the worm; they also left all pupae untouched.

Field vole as potential predator Results of the field voles showed a clear difference between the two treatments in pupal survival, i.e., consumption times (Fig. 2). Pupal survival time in the first treatment (survival time 6.9 ± 1.0 h (mean ± SE)) differed from that in the second treatment with alternative food (10.7 ± 0.8 h), as indicated by the results of the survival analysis (log rank = 13.3, df = 1, p < 0.001). In the first treatment, notably higher proportion of the field voles fed on the pupae than in the other treatment with alternative food (Table 1). The field voles preferred the alternative food containing vegetable diet. Very high proportion (83.3%) of the field voles did not eat pupae at all in the second treatment (Table 1). A higher pro- 2 portion of the field voles ate grass (c 1 = 13.49, p < 0.001) 2 ½ � and pellet (c 1 = 16.00, p < 0.001) than pupae of deer ked in the second½ treatment.� deer ked, as higher proportion of shrews fed on the red Bank vole as potential predator worm than pupae in the second treatment. The low feeding Cox regression analysis with the bank voles showed that rate of pupae among the shrews was opposite to our predic- pupal survival, i.e., consumption time, did not differ between tions. Low feeding rate of pupae could be partly explained by the two treatments (odds ratio = 0.66; Wald statistic = 1.65, stress that shrews may have experienced during the experi- df = 1, p = 0.199). Hence, alternative food had no impact on

For personal use only. ment, but because the feeding rate on the red worm was pupal consumption when the data on the wild bank voles and much higher, sensitivity to stress may have not been the de- laboratory colonies were pooled (Fig. 3). There was neither finitive factor. Nearly all soricine shrews have very high met- effect of vole s background (wild vs. wild with laboratory ’ abolic rates and they need to forage constantly and only few history) on the pupal consumption (odds ratio = 0.9; Wald hours without feeding (Barnard and Hurst 1987) can lead to statistic = 0.09, df = 1, p = 0.76) nor significant interaction death (Vogel 1976). Worms may contain more protein than between the two covariate factors, treatment and background pupae of deer ked, but the absence of alternative food did of voles (odds ratio = 0.8; Wald statistic = 0.2, df = 1, p = 0.65). However, the differences in the total numbers of eaten not raise pupal consumption, which may suggest that pupae pupae between the two treatments can be considered margin- of deer ked are not so palatable food for the shrews. The field voles preferred alternative food including grass ally significant (c2 = 3.81, p = 0.051). According to this, 1 and pellet instead of pupae of deer ked, as we predicted. bank voles more likely½ � consumed all five pupae if there was However, if there was no alternative food present, the pupal no access to alternative food (Table 1). consumption was much higher. This result is expected because in the wild, the main diet of field voles consists of Discussion green parts of vegetation; however, when the principal nutri-

Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of Eastern Finland on 08/28/12 The results indicated that all three small-mammal species ments are scarce, as in late winter, field voles might include can prey on pupae of deer ked. This was especially true for animal protein like insects to their diet (Hanski and Parviai- bank voles. Furthermore, it appeared that field voles under nen 1985). food limitation do consume pupae of deer ked, but not if an The bank voles, both the voles with the longer laboratory alternative food is available. Contrary to our predictions, in- history and the wild-caught individuals, consumed pupae re- sectivorous and most of the time food-constrained shrews gardless of availability of alternative food. Relative high pro- consumed less pupae compared with granivorous– portion of voles which ate pupae in both treatments reflects herbivorous voles in the experiment. the fact that the bank vole is a granivore–omnivore species, In our experiments, the pupal consumption of common its food consisting mainly of seeds, berries, and green vegeta- shrews was similar whether the shrews had access to alterna- tion, as well as insects, earthworms, and other invertebrates tive food or not. In both treatments, at least half of the (von Blanckenhagen et al. 2007). The highest peak in food shrews did not eat pupae at all. However, red worm was ob- consumption of bank voles in the wild (see, e.g., Eccard and viously a more palatable food for the shrews than pupae of Ylönen 2001; von Blanckenhagen et al. 2007) overlaps with

Published by NRC Research Press 1048 Can. J. Zool. Vol. 90, 2012

Table 1. Percentages of small mammals categorized according to food consumption of small mammals in two treatments (pupae of deer ked (Lipoptena cervi) only vs. pupae + alternative food).

All five pupae One to four Pupae not Species Treatment eaten (%) pupae eaten (%) eaten (%) Alternative food Common shrews, Sorex araneus Only pupae 35.3 5.9 58.8 Pupae + piece of worm 38.9 11.1 50 83% ate worm; 50% ate pupae + worm

Field voles, Microtus agrestis Only pupae 61 16.7 22.2 Pupae + pellet + grass 11.1 5.6 83.3 77.7% ate grass (>80% of grass); 83.3% ate pellet (<80% of pellet)

Bank voles, Myodes glareolus Only pupae 67.7 29 3.2 (wild and laboratory) Pupae + pellet 40 50 10 na

Fig. 2. Survival times (h) of pupae of deer ked (Lipoptena cervi) in Fig. 3. Survival times (h) of pupae of deer ked (Lipoptena cervi) in two treatments (only pupae vs. pupae + grass + pellet) in the ex- two treatments (only pupae vs. pupae + pellet) in the experiment periment with the field voles (Microtus agrestis). with the bank voles (Myodes glareolus). The data on wild bank voles and wild bank voles with longer laboratory history are pooled. For personal use only.

the highest occurrence of the pupae of deer ked during spring time. Hence, it could be assumed that both bank voles and high prey density seems to be higher than at low prey density field voles under food limitation, i.e., late winter and early (see Yasuda and Ishikawa 1999; Ioannou et al. 2009). Hence, spring when no new vegetative growth has occurred yet, bedding sites can be real “hot spots” for predators because could be notable predators of pupae of deer ked. Indeed, our denser areas of pupal groups are likely more conspicuous.

Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of Eastern Finland on 08/28/12 marginally significant result showed that bank voles con- The moose may move for tens, or even hundreds, of kilo- sumed all five pupae more likely if there was no access to metres during a year (Heikkinen 2000) and thus the pupae alternative food. Furthermore, as densities of both vole spe- may drop off the host far away from each other. Thus, pupal cies fluctuate multiannually, at least the peak of population prevalence and densities within the forest or meadow floor densities could affect the populations of deer ked locally. To litter is not evenly distributed but concentrated probably answer these questions, further studies on the ability of voles around bedding sites. Pupal stage may last as long as and shrews to find pupae in their natural ground-layer habi- 10 months in some individuals, depending on the birth time, tats are needed. and during that period also pupae outside the bedding sites Highest densities of pupae of deer ked can be observed on may encounter several possible vertebrate and invertebrate bedding sites of most infested cervid hosts (Kaunisto et al. predators. Especially in late winter, during the period when 2009; Välimäki et al. 2011). Although little is known how snow melts revealing bare ground without new vegetation, prey density affects prey detection of predators and the accu- small mammals may encounter pupae more easily than later racy of attacks, in some cases predator foraging efficiency at in the growing season. At this turning point, small mammals

Published by NRC Research Press Kaunisto et al. 1049

often face food limitations, which may further increase small Eccard, J.A., and Ylönen, H. 2001. Initiation of breeding after winter mammalian predation on pupae of deer ked during this time. in bank voles: effects of food and population density. Can. J. Zool. All three small-mammal species studied are common in 79(10): 1743–1753. doi:10.1139/z01-133. the same habitats as moose, the primary host of deer ked Eccard, J.A., and Ylönen, H. 2006. Adaptive food choice of bank (Välimäki et al. 2011; Sundell et al. 2012) and consequently voles in a novel environment: choices enhance reproductive status pupae of deer ked. In addition to the three species studied, in winter and spring. Ann. Zool. Fenn. 43(1): 2–8. mice can also feed on pupae of deer ked. For example, the Edwards, C.A., and Bohlen, P.J. 1996. Biology and ecology of wood mouse (Apodemus sylvaticus (L., 1758)) can frequently earthworms. Chapman and Hall, London. Frank, J.H. 1967a. The insect predators of the pupal stage of the prey on insects in forest habitats (Khammes and Aulagnier winter moth, Operophtera brumata (L.) (Lepidoptera: Hydriome- 2007). It is very likely that all of the above mentioned small nidae). J. Anim. Ecol. 36(2): 375–389. doi:10.2307/2920. mammals act as important predators on pupae of deer ked, Frank, J.H. 1967b. The effect of pupal predators on a population of especially when alternative food is scarce. There are regional winter moth, Operophtera brumata (L.) (Hydriomenidae). J. differences in the invasion success of deer ked inside its dis- Anim. Ecol. 36(3): 611–621. doi:10.2307/2816. tribution area, presumably explained by climatic or host- Haarløv, N. 1964. Life cycle and distribution pattern of Lipoptena related factors (Härkönen et al. 2010; Välimäki et al. 2011). cervi (L.) (Dipt., Hippobosc.) on Danish deer. Oikos, 15(1): 93– Based on the present results, predation may also have a role 129. doi:10.2307/3564750. as a regulative factor. Hackman, W., Rantanen, T., and Vuojolahti, P. 1983. Immigration of We demonstrated for the first time, although under labora- Lipoptena cervi (Diptera, Hippoboscidae) in Finland, with notes tory conditions, that deer keds have small mammalian preda- on its biology and medical significance. Not. Entomol. 63: 53–59. tors. However, more detailed field research will be needed in Hanski, I. 1984. Food consumption, assimilation and metabolic rate the future to evaluate the magnitude of predation on deer ked in six species of shrew (Sorex and Neomys). Ann. Zool. Fenn. in the wild. This information would be valuable also in the 21(2): 157–165. wider context, as deer ked provides a good model organism Hanski, I., and Parviainen, P. 1985. Cocoon predation by small to study interactions between predators and invasive parasitic mammals, and pine sawfly population dynamics. Oikos, 45(1): 125 136. doi:10.2307/3565230. insect. – Hansson, L. 1971. Small rodent food, feeding and population Acknowledgements dynamics: a comparison between granivorous and herbivorous species in Scandinavia. Oikos, 22(2): 183–198. doi:10.2307/ We thank Konnevesi Research Station and the staff for the 3543724. help during execution of this study. We also thank two anon- Hansson, L. 1985. Clethrionomys food: generic, specific and regional ymous reviewers for improving our contribution. This study characteristics. Ann. Zool. Fenn. 22(3): 315–318. was funded by the Ella and Georg Ehrnrooth Foundation Härkönen, L., Härkönen, S., Kaitala, A., Kaunisto, S., Kortet, R., (grant for S.K.), the Alfred Kordelin Foundation (S.K.), Laaksonen, S., and Ylönen, H. 2010. Predicting range expansion Societas pro Fauna et Flora Fennica (S.K.), University of of an ectoparasite—the effect of spring and summer temperatures For personal use only. Jyväskylä (S.K.), and the Biological Interactions Graduate on deer ked Lipoptena cervi (Diptera: Hippoboscidae) perfor- School (S.K.). This study complies with the current laws of mance along a latitudinal gradient. Ecography, 33(5): 906–912. Finland. doi:10.1111/j.1600-0587.2009.05890.x. Hastings, F.L., Hain, F.P., Smith, H.R., Cook, S.P., and Monahan, J.F. References 2002. Predation of gypsy moth (Lepidoptera: Lymantriidae) pupae in three ecosystems along the southern edge of infestation. Barnard, C.J., and Hurst, J.L. 1987. Time constraints and prey Environ. Entomol. 31(4): 668–675. doi:10.1603/0046-225X-31.4. selection in common shrews Sorex araneus L. Anim. Behav. 35(6): 668. 1827–1837. doi:10.1016/S0003-3472(87)80075-1. Heikkinen, S. 2000. Hirven vuosi. Suomen Riista, 46: 82–91. Bequaert, J.C. 1953. The Hippoboscidae or louse-flies (Diptera) of [In Finnish.] mammals and birds. Part I. Structure, physiology and natural Holling, C.S. 1959. The components of predation as revealed by a history. Entomol. Am. 32:1–209. study of small-mammal predation of the European pine sawfly. Carlsson, N.O.L., Sarnelle, O., and Strayer, D.L. 2009. Native Can. Entomol. 91(5): 293–320. doi:10.4039/Ent91293-5. predators and exotic prey—an acquired taste? Front. Ecol. Holt, R.D. 1977. Predation, apparent competition, and the structure of Environ, 7(10): 525–532. doi:10.1890/080093. prey communities. Theor. Popul. Biol. 12(2): 197–229. doi:10. Churchfield, S. 2002. Why are shrews so small? The costs and 1016/0040-5809(77)90042-9. PMID:929457. Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of Eastern Finland on 08/28/12 benefits of small size in northern temperate Sorex species in the Holway, D.A., Suarez, A.V., and Case, T.J. 1998. Loss of context of foraging habits and prey supply. Acta Theriol. intraspecific aggression in the success of a widespread invasive 47(Suppl. 1): 169–184. doi:10.1007/BF03192486. social insect. Science, 282(5390): 949–952. doi:10.1126/science. Churchfield, S., Hollier, J., and Brown, V.K. 1991. The effects of 282.5390.949. PMID:9794767. small mammal predators on grassland invertebrates investigated by Ioannou, C.C., Morrell, L.J., Ruxton, G.D., and Krause, J. 2009. The field exclosure experiment. Oikos, 60(3): 283–290. doi:10.2307/ effect of prey density on predators: conspicuousness and attack 3545069. success are sensitive to spatial scale. Am. Nat. 173(4): 499–506. Davis, M.A. 2009. Invasion biology. Oxford University Press Inc., doi:10.1086/597219. PMID:19231967. New York. Kaunisto, S., Kortet, R., Härkönen, L., Härkönen, S., Ylönen, H., and Dehio, C., Sauder, U., and Hiestand, R. 2004. Isolation of Bartonella Laaksonen, S. 2009. New bedding site examination-based method schoenbuchensis from Lipoptena cervi, a blood-sucking arthropod to analyse deer ked (Lipoptena cervi) infection in cervids. causing deer ked dermatitis. J. Clin. Microbiol. 42(11): 5320– Parasitol. Res. 104(4): 919–925. doi:10.1007/s00436-008-1273- 5323. doi:10.1128/JCM.42.11.5320-5323.2004. PMID:15528732. 0. PMID:19050924.

Published by NRC Research Press 1050 Can. J. Zool. Vol. 90, 2012

Kaunisto, S., Härkönen, L., Niemelä, P., Roininen, H., and Ylönen, introduced species. In Biological invasions: a global perspective. H. 2011. Northward invasion of the parasitic deer ked (Lipoptena Edited by J.A. Drake, H.A. Mooney, F. di Castri, R.H. Groves, F.J. cervi): is there geographical variation in pupal size and develop- Kruger, M. Rejmánek, and M. Williamson. John Wiley and Sons, ment duration? Parasitology, 138(3): 354–363. doi:10.1017/ New York. pp. 351–367. S0031182010001332. PMID:20880422. Pimm, S.L. 1991. The balance of nature? Ecological issues in the Kaunisto, S., Välimäki, P., Kortet, R., Koskimäki, J., Härkönen, S., conservation of species and communities. University of Chicago Kaitala, A., Laaksonen, S., Härkönen, L., and Ylönen, H. 2012. Press, Chicago. Avian predation on a parasitic fly of cervids during winter: can Price, P.W., Denno, R.F., Eubanks, M.D., Finke, D.L., Kaplan, I. host-related cues increase the predation risk? Biol. J. Linn. Soc. 2011. Insect ecology: behavior, populations and communities. 106(2): 275–286. doi:10.1111/j.1095-8312.2012.01869.x. Cambridge University Press, Cambridge. Khammes, N., and Aulagnier, S. 2007. Diet of the wood mouse, Rantanen, T., Reunala, T., Vuojolahti, P., and Hackman, W. 1982. Apodemus sylvaticus, in three biotopes of Kabylie of Djurdjura Persistent pruritic papules from deer ked bites. Acta Derm. (Algeria). Folia Zool. (Brno.), 56(3): 243–252. Venereol. 62(4): 307–311 PMID:6183862. Kortet, R., Härkönen, L., Hokkanen, P., Härkönen, S., Kaitala, A., Schoener, T.W., and Spiller, D.A. 1995. Effect of predators and area Kaunisto, S., Laaksonen, S., Kekäläinen, J., and Ylönen, H. 2010. on invasion: an experiment with island spiders. Science Experiments on the ectoparasitic deer ked that often attacks 267(5205): 1811–1813. doi:10.1126/science.267.5205.1811. humans; preferences for body parts, colour and temperature. PMID:17775810. Bull. Entomol. Res. 100(3): 279–285. doi:10.1017/ Sparti, A. 1992. Thermogenic capacity of shrews (Mammalia, S0007485309990277. PMID:19698197. Soricidae) and its relationship with basal rate of metabolism. Koss, A.M., and Snyder, W.E. 2005. Alternative prey disrupt Physiol. Zool. 65(1): 77–96. biocontrol by a guild of generalist predators. Biol. Control, Sundell, J., Church, C., and Ovaskainen, O. 2012. Spatio-temporal 32(2): 243–251. doi:10.1016/j.biocontrol.2004.10.002. patterns of habitat use in voles and shrews modified by density, Kynkäänniemi, S.M., Kortet, R., Härkönen, L., Kaitala, A., season and predators. J. Anim. Ecol. 81(4): 747–755. doi:10.1111/ Paakkonen, T., Mustonen, A.M., Nieminen, P., Härkönen, S., j.1365-2656.2012.01956.x. PMID:22325037. Ylönen, H., and Laaksonen, S. 2010. Threat of an invasive Tanhuanpää, M., Ruohomäki, K., Kaitaniemi, P., and Klemola, T. parasitic fly, the deer ked (Lipoptena cervi), to the reindeer 1999. Different impact of pupal predation on populations of (Rangifer tarandus tarandus): experimental infection and treat- Epirrita autumnata (Lepidoptera; Geometridae) within and out- ment. Ann. Zool. Fenn. 47: 28–36. side the outbreak range. J. Anim. Ecol. 68(3): 562–570. doi:10. Laukkanen, A., Ruoppi, P., and Mäkinen-Kiljunen, S. 2005. Deer 1046/j.1365-2656.1999.00305.x. ked-induced occupational allergic rhinoconjunctivitis. Ann. Välimäki, P., Madslien, K., Malmsten, J., Härkönen, L., Härkönen, Allergy Asthma Immunol. 94(5): 604–608. doi:10.1016/S1081- S., Kaitala, A., Kortet, R., Laaksonen, S., Mehl, R., Redford, L., 1206(10)61141-6. PMID:15945565. Ylönen, H., and Ytrehus, B. 2010. Fennoscandian distribution of Lockwood, J.L., Hoopes, M.F., and Marchetti, M.P. 2007. Invasion an important parasite of cervids, the deer ked (Lipoptena cervi), ecology. Blackwell Publishing, Oxford. revisited. Parasitol. Res. 107(1): 117–125. doi:10.1007/s00436- Lodge, D.M. 1993. Biological invasions: lessons for ecology. Trends 010-1845-7. PMID:20379833.

For personal use only. Ecol. Evol. 8(4): 133–137. doi:10.1016/0169-5347(93)90025-K. Välimäki, P., Kaitala, A., Madslien, K., Härkönen, L., Várkonyi, G., PMID:21236129. Heikkilä, J., Jaakola, M., Ylönen, H., Kortet, R., and Ytrehus, B. Maa, T.C. 1969. A revised checklist and concise host index of 2011. Geographical variation in host use of a blood-feeding Hippoboscidae (Diptera). Pac. Insects Monogr. 20: 261–299. ectoparasitic fly: implications for population invasiveness. Matsumoto, K., Berrada, Z.L., Klinger, E., Goethert, H.K., and Oecologia (Berl.), 166(4): 985–995. doi:10.1007/s00442-011- Telford, S.R.I.I.I., III. 2008. Molecular detection of Bartonella 1951-y. PMID:21390490. schoenbuchensis from ectoparasites of deer in Massachusetts. Van Baalen, M., Krivan, V., Van Rijn, P.C.J., and Sabelis, M.W. Vector Borne Zoonotic Dis. 8(4): 549–554. doi:10.1089/vbz.2007. 2001. Alternative food, switching predators, and the persistence of 0244. PMID:18454595. predator–prey systems. Am. Nat. 157(5): 512–524. doi:10.1086/ McDevitt, R., and Andrews, J.F. 1995. Seasonal variation in the 319933. PMID:18707259. metabolic rate of the pygmy shrew, Sorex minutus: can resting Vermeij, G.J. 1996. An agenda for invasion biology. Biol. Conserv. metabolic rate be measured in post-absorptive shrews? J. Therm. 78(1–2): 3–9. doi:10.1016/0006-3207(96)00013-4. Biol. 20(3): 255–261. doi:10.1016/0306-4565(94)00048-N. Vogel, P. 1976. Energy consumption of European and African Meier, R., Kotrba, M., and Ferrar, P. 1999. Ovoviviparity and shrews. Acta Theriol. 21: 195–206. viviparity in the Diptera. Biol. Rev. Camb. Philos. Soc. 74(3): von Blanckenhagen, F., Eccard, J.A., and Ylönen, H. 2007. Animal 199–258. doi:10.1017/S0006323199005320. protein as a reproductive constraint in spring reproduction of the Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of Eastern Finland on 08/28/12 Paakkonen, T., Mustonen, A.M., Roininen, H., Niemelä, P., Ruusila, bank vole. Écoscience, 14(3): 323–329. doi:10.2980/1195-6860 V., and Nieminen, P. 2010. Parasitism of the deer ked, Lipoptena (2007)14[323:APAARC]2.0.CO;2. cervi, on the moose, Alces alces, in eastern Finland. Med. Vet. Yasuda, H., and Ishikawa, H. 1999. Effects of prey density and spatial Entomol. 24(4): 411–417. doi:10.1111/j.1365-2915.2010.00910.x. distribution on prey consumption of the adult predatory ladybird PMID:20868432. beetle. J. Appl. Entomol. 123(10): 585–589. doi:10.1046/j.1439- Pimm, S.L. 1989. Theories of predicting success and impact of 0418.1999.00420.x.

Published by NRC Research Press dissertations

Sirpa Kaunisto An invasive ectoparasite of

cervids, the deer ked: predation and tolerance cold dispersion, ked: deer the of cervids, ectoparasite invasive | An Kaunisto | No 87 | Sirpa dispersion, cold tolerance and predation

Publications of the University of Eastern Finland Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences No 87 Dissertations in Forestry and Natural Sciences

isbn 978-952-61-0947-3 (printed) issnl 1798-5668 issn 1798-5668 isbn 978-952-61-0948-0 (pdf) issn 1798-5676 (pdf)