Diptera: Calliphoridae) at Constant and Fluctuating Temperatures By

The Development of Protophormia terraenovae (Robineau-Desvoidy) (Diptera: Calliphoridae) at Constant and Fluctuating Temperatures by

Jodie-Ann Warren B.Sc., Simon Fraser University 1999

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF ARTS

In the School of Criminology

O Jodie-Ann Warren 2006

SIMON FRASER UNIVERSITY

All rights reserved. This work may not be reproduced in whole or in part, by photocopy or other means, without permission of the author $?.? SIMON FRASER aQ3 UN~~ER~A~brary

DECLARATION OF PARTIAL COPYRIGHT LICENCE

The author, whose copyright is declared on the title page of this work, has granted to Simon Fraser University the right to lend this thesis, project or extended essay to users of the Simon Fraser University Library, and to make partial or single copies only for such users or in response to a request from the library of any other university, or other educational institution, on its own behalf or for one of its users.

The author has further granted permission to Simon Fraser University to keep or make a digital copy for use in its circulating collection, and, without changing the content, to translate the thesislproject or extended essays, if technically possible, to any medium or format for the purpose of preservation of the digital work.

The author has further agreed that permission for multiple copying of this work for scholarly purposes may be granted by either the author or the Dean of Graduate Studies.

It is understood that copying or publication of this work for financial gain shall not be allowed without the author's written permission.

Permission for public performance, or limited permission for private scholarly use, of any multimedia materials forming part of this work, may have been granted by the author. This information may be found on the separately catalogued multimedia material and in the signed Partial Copyright Licence.

The original Partial Copyright Licence attesting to these terms, and signed by this author, may be found in the original bound copy of this work, retained in the Simon Fraser University Archive.

Simon Fraser University Library Burnaby, BC, Canada

Summer 2008 ABSTRACT

Protophormia terraenovae (R-D) was observed over a range of constant temperatures, 9.8 to 32OC as well as at fluctuating temperatures of 4 to 28 and 9 to 23OC and the minimum developmental times and mode of development to reach each stage were recorded. A comparison of the actual minimum temperature threshold was made to findings from the linear method. The minimum temperature thresholds for the lSt,and presumably the 2ndand 3rd instars fall within 9.8 and 1I0C and those of the post feeding, pupal and adult stages fall within the range of 11 to 13OC. However, in all cases the actual minimum temperature thresholds were radically underestimated by the linear method.

In addition, a comparison of the development of P. terraenovae at fluctuating temperatures of 4 to 28 and 9 to 23OC, to their common mean temperature (16OC) indicated that development was faster at fluctuating temperatures.

Keywords

Development; Forensic Entomology; Minimum Temperature Threshold;

Protophormia terraenovae; Post-Mortem Interval For Edna Charlton

A True Inspiration! I would like to thank my committee members for their advice. Dr Lisa

Poirier was an exceptional help with piecing my thesis together and went to great lengths literally to help with the production of this thesis. Dr Paul Brantingham made a significant effort by stepping out of his niche into a habitat not frequented by many, other than entomologists. I would like to express my deepest gratitude to Dr Gail Anderson because without her constant support and understanding, I would not have had the courage to strive to this level of academics.

I would also like to thank Linnea Duke for coming to my aid at a moments notice when I needed assistance. I cannot thank Diane Strub and Melissa Austin enough for their contributions to the lab and to my experiments. I would like to thank Niki Huitson for her input and assistance with developing strategies to overcome obstacles in the lab.

Without the financial contributions of the Canadian Police Research

Centre, the Vancouver Foundation and the Terry Fox Gold Medal Award, this research would not have been possible.

Finally, this endeavour would not have occurred without the loving support of my family and friends. TABLE OF CONTENTS

Approval ...... ii ... Abstract ...... 111 Dedication ...... iv Acknowledgements ...... v Table of Contents...... vi ... List of Figures ...... VIII List of Tables ...... x Glossary ...... xii 1 Forensic Entomology ...... I 1. 1 Introduction...... I 2 Literature Review ...... 8 2.1 Theory ...... 8 2.2 Rationale ...... 11 3 Protophormia terraenovae Development at Constant Temperatures ...... 17 3.1 Introduction...... 17 3.1 . 1 Core Constant Temperature Experiments ...... 18 3.1.2 Objectives ...... 18 3.2 Methods and Materials ...... 19 3.2.1 Collection of Experimental Insects ...... 19 3.2.2 Rearing of the Insects ...... 20 3.2.3 Experimental Methods for Constant Temperature Experiments ...... 22 3.2.4 Environmental Chambers ...... 22 3.2.5 Insect Sampling ...... 24 3.3 Results (Constant Temperature Experiments) ...... 26 3.3.1 A Comparison with Previously Published Research ...... 46 3.4 Discussion (Constant Temperature Experiments) ...... 49 4 Protophormia terraenovae Development at Fluctuating Temperatures ...... 58 4.1 Introduction...... 58 4.1 .1 Core Fluctuating Temperature Experiment ...... 59 4.1.2 Objectives ...... 60 4.2 Methods and Materials ...... 60 4.2.1 Methods for Fluctuating Temperature Experiments ...... 60 4.3 Results (Fluctuating Temperature Experiments) ...... 63 4.4 Discussion (Fluctuating Temperature Experiments) ...... 72 5 Overall Discussion ...... 76 5.1 Conclusion...... 76 5.2 Forensic Entomology in Court ...... 80 Appendices ...... 85 Appendix A Datalogger Recordings for the 4 to 28OC Fluctuating Temperature Experiments...... 86 Appendix B Datalogger Recordings for the 9 to 23OC Fluctuating Temperature Experiments...... 87 Appendix C Datalogger Recordings for the Mean Constant 16•‹C Experiment...... -88 Reference List ...... 89

vii LIST OF FIGURES

Figure 2-1 The relationship between development rate and temperature for blow flies where Tminis the minimum temperature threshold, Tmin(estIis the estimated minimum temperature threshold, Tmaxis the maximum temperature threshold and Optde, is the optimum temperature for fastest rate of development ...... I0 Figure 3-1 The percent minimum developmental time (a) and mode of development (b) of P. terraenovae in each stage at constant temperatures of II, 13, 15, 1G(chapter 4), 20, 25, 28, 30, and 32OC ...... 40 Figure 3-2 Rates of the minimum development and mode of development of P. terraenovae to 1'' instar at the researched temperatures including a linear regression and extrapolation of the minimum rates of development ...... 42 Figure 3-3 Rates of the minimum development and mode of development of P. terraenovae to 2ndinstar at the researched temperatures including a linear regression and extrapolation of the minimum rates of development ...... 43 Figure 3-4 Rates of the minimum development and mode of development of P. terraenovae to 3rdinstar at the researched temperatures including a linear regression and extrapolation of the minimum rates of development ...... 43 Figure 3-5 Rates of the minimum development and mode of development of P. terraenovae to the post feeding stage at the researched temperatures including a linear regression and extrapolation of the minimum rates of development ...... 44 Figure 3-6 Rates of the minimum development and mode of development of P. terraenovae to the pupal stage at the researched temperatures including a linear regression and extrapolation of the minimum rates of development ...... 45 Figure 3-7 Rates of the minimum development and mode of development of P. terraenovae to the adult stage at the researched temperatures including a linear regression and extrapolation of the minimum rates of development ...... 45

viii Figure 3-8 A comparison of the minimum developmental times at four temperatures (15, 20, 25 and 30•‹C) to that of previously published data ...... Figure 3-9 A comparison of the mean minimum developmental times for P. terraenovae at 25 and 28OC to Kamal's (1958) development data at 26.7OC...... Figure 3-10 The actual findings of the development graphs where Tminis the minimum temperature threshold, Tmin(est,is the minimum temperature threshold determined by extrapolating the linear regression. Optd,, is the temperature at which optimum development occurs and Tmaxis the maximum temperature threshold. By extending the linear portion of the graph, it always underestimated the Tmin. This is unlike that which is reported in the literature where the line appears to overestimate that of the Tmin. (Figure 2-1 ) ...... 55 Figure 4-1 A comparison of the means of the minimum developmental time for P. terraenovae to reach each stage of development at fluctuating temperatures of 4 to 28OC and 9 to 23OC to each other and their common mean temperature of 16OC ...... 69 Figure 4-2 A comparison of the means of the mode of development for P. terraenovae to reach each stage at fluctuating temperatures of 4 to 28OC and 9 to 23OC to each other and their common mean temperature of 16OC ...... 69 Figure 4-3 Figure 4-3 The minimum developmental time (a) and mode of development (b) for P. terraenovae in each stage of development expressed as a percentage of complete immature development for fluctuating temperatures of 4 to 28 and 9 to 23OC and their mean constant temperature of 16OC ...... 71 LIST OF TABLES

Table 3-1 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 11•‹C ...... 27 Table 3-2 The mode of development for P. terraenovae to reach each stage at a constant temperature of 11•‹C ...... 28 Table 3-3 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 13•‹C ...... 29 Table 3-4 The mode of development for P. terraenovae to reach each stage at a constant temperature of 13•‹C...... 29 Table 3-5 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 15•‹C ...... 30 Table 3-6 The mode of development for P. terraenovae to reach each stage at a constant temperature of 15•‹C ...... 30 Table 3-7 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 20•‹C ...... 31 Table 3-8 The mode of development for P. terraenovae to reach each stage at a constant temperature of 20•‹C ...... 31 Table 3-9 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 25•‹C ...... 32 Table 3-10 The mode of development for P. terraenovae to reach each stage at a constant temperature of 25•‹C ...... 33 Table 3-1 1 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 28•‹C ...... 34 Table 3-12 The mode of development for P. terraenovae to reach each stage at a constant temperature of 28•‹C ...... 34 Table 3-13 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 30•‹C ...... 35 Table 3-14 The mode of development for P. terraenovae to reach each stage at a constant temperature of 30•‹C ...... 36 Table 3-1 5 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 32•‹C ...... 37 Table 3-16 The mode of development for P. terraenovae to reach each stage at constant temperature 32•‹C ...... 38 Table 3-1 7 The equations of the regression lines, the R-square values and the x-intercepts for the linear regressions of the mean minimum developmental times and mode development to each of the stages ...... 41 Table 4-1 The environmental chamber settings for the experiments set at fluctuating temperatures...... 63 Table 4-2 The minimum developmental times of P. terraenovae to reach each stage at fluctuating temperatures of 4 to 28OC ...... 64 Table 4-3 The mode of development for P. terraenovae to reach each stage at fluctuating temperatures of 4 to 28OC ...... 64 Table 4-4 The minimum developmental times of P. terraenovae to reach each stage at fluctuating temperatures of 9 to 23OC ...... 65 Table 4-5 The mode of development for P. terraenovae to reach each stage at fluctuating temperatures of 9 to 23OC ...... 66 Table 4-6 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 16OC ...... 67 Table 4-7 The mode of development for P. terraenovae to reach each stage at a constant temperature of 16OC ...... 68 GLOSSARY

Eclosion Hatch of the insect eggs

Founder Effect Those insects that may show some variation from the population that it was originally derived and consequently may not be an accurate representation of the species

Forensic The study of the science of insects and how it Entomology relates to law

Holarctic A region including both the Nearctic (North America) and the Palearctic (Eurasia) land regions

lsomegalen A diagram that incorporates data from growth Diagram curves and larval length. It is plotted as temperature versus time

lsomorphen A diagram that incorporates development data Diagram for all stages and is plotted as temperature versus time. Each line represents morphological changes

Maximum The upper temperature at which insect Temperature development ceases Threshold (Tma,)

Minimum The lower temperature at which insect Temperature development ceases Threshold (Tmin) The most frequently occurring number in a Mode distribution Necrophagous Feed on dead tissue

Optimal The ideal temperature for development at which Development greatest hatch occurs, least mortality and Temperature highest rate of development (O~tdev)

xii Photoperiod Is the period of daylight and darkness that occurs in each 24 hour cycle

Poikilothermic Refers to a body temperature that fluctuates with the environment

Protophormia A species of blow fly found in colder regions terraenovae (R-D) commonly referred to as both the Holarctic blow fly and the Blackbottle fly

Rate Summation Is a phenomenon that is observed with Effect sinusoidal development and fluctuating temperatures. Fluctuations at low temperatures to above and below a given mean temperature cause an increased development rate as compared to the mean temperature because those fluctuations above the mean temperature increase the development rate relatively more than those fluctuations below the mean can lower the rate. (Higley and Haskell 2001)

Sarcosapro- Can be divided into two key terms, sarc phagous meaning tissue and saprophagous meaning to feed on dead matter

Thermoperiods Daily temperature cycles

xiii I FORENSIC ENTOMOLOGY

1.I Introduction

Medicolegal entomology, commonly referred to as forensic entomology,' is the study and application of insects as they relate to legal events. Forensic entomology can be applied to investigations to determine if a discovered body has been relocated or disturbed, the presence and position of wound sites and whether there are any drugs in the body. Perhaps the most important and well known application is that of determining an estimated time since death or post- mortem interval (PMI) (Anderson 2001). It is vital to a death investigation to be able to determine the PMI upon discovery of human remains. The single most important reason in a non-suspicious death is to be able to provide this information to the grieving family members. However, in the case of a suspicious death, a PMI becomes essential to the investigation in providing a timeline of when the event occurred that can, therefore, direct law enforcement in their investigation (Anderson 2001 ). Within the first 72 hours, several methods can be used by a forensic pathologist to determine the PMI, but after 72 hours forensic entomology is the most accurate and frequently the only approach (Kashyap and

Pillai 1989; Anderson and VanLaerhoven 1996; Anderson 2001 ; Bourel et al.

2003).

1 The bolded terms are defined in the glossary As a body, whether human or animal, decomposes, it goes through rapid biological, chemical and physical changes. Each of these changes is attractive to a different group of insects. Megnin (1894), in France, first recognized that this insect sequence was predictable and since this time, many other researchers worldwide have confirmed this.

Four categories of insects inhabit decomposing remains. The first category is the necrophagous species, those insects that feed on the decomposing remains itself. The second category includes the predators that prey on the necrophagous species. Thirdly, there are omnivorous species that scavenge both the remains and the organisms that are attracted to the remains.

Finally, the adventitious species are present simply by chance or because the remains provide shelter (Lane 1975; Smith 1986; Goff 1993).

The species and even families of insects colonizing the remains in this sequence vary greatly depending on many parameters, such as geographic region, season and habitat (Anderson 2001 b). However, within these parameters, the sequence is predictable, allowing its use in estimating the PMI.

Some researchers describe the sequential colonization of a body as occurring in waves and anything from five to eight waves are often discussed (Megnin 1894;

Easton and Smith 1970; ErzinQioglu 1983; Smith 1986; Payne 1965; Goff 1993).

However, although each of the stages of decomposition is rather well defined, the insect attraction to them is not, as there is considerable overlap of insects throughout the stages (Schoenly and Reid 1987). The insects colonizing the body immediately after death are usually referred to as first wave insects and are primarily blow flies (Diptera:

Calliphoridae). These colonize remains within minutes or seconds of death, being attracted primarily to wound sites and natural orifices (Smith 1986, Anderson and

VanLaerhoven 1996) to lay eggs. These eggs hatch, in a predictable period of time, into first instar or first stage larvae, which feed on liquid protein at the orifices or wounds. They then proceed through the immature stages, feeding on the remains. These first colonizers are most valuable in estimating the elapsed time since death. The actual species that first colonize vary with region, season and habitat, but may often include Protophormia terraenovae (R-D) (Hedouin et al. 2001), the subject of this research.

As the body begins to bloat, second wave insects colonize, including species of Calliphoridae as well as flesh flies or Sarcophagidae. When the adipose or fatty tissue becomes rancid, undergoing butyric fermentation, the remains become less attractive to the early colonizers and more attractive to third wave insects, which feed on decomposing adipose tissue. These include a variety of Diptera and Coleoptera (beetles) as well as some Lepidoptera

(butterflies and moths). When protein or caseic fermentation begins, fourth wave insects are attracted, including dipteran families such as Piophilidae, Fanniidae,

Drosophilidae, Sphaeroceridae and Sepsidae and a variety of coleopteran species (Easton and Smith 1970; Erzin@oglu 1983; Smith 1986).

Ammoniacal fermentation of the body attracts other Diptera such as

Ophyra spp. (Muscidae) and new species of Coleoptera. When the body becomes desiccated, it is unattractive to some of the previous species, but more attractive to others. Eventually the body becomes entirely desiccated and attractive to only a few Coleoptera and Lepidoptera species. Finally, very little remains, with only a few species still able to obtain some nutrition associated with the body (Payne 1965; Easton and Smith 1970; Erzinqlioglu 1983; Smith

1986; Goff 1993)

There are two methods of determining a PMI. The first method looks at the successional waves of insects on the body. Because insects appear on the body in a predictable sequence, as they are attracted to the body at different stages of decomposition, this sequence can be used to estimate the PMI. If all attributes that may have an affect on this sequence are accounted for then a relatively accurate time since death can be established. The parameters that influence the sequence of colonization include the biogeoclimatic zone, the season, the particular habitat, whether the corpse is in direct sunlight or shade and scavengers (Anderson 2001 b).

The second method of determining time of death uses the life stages of the insect. It is used when death is more recent. The necrophagous blow flies are the most commonly used insects in this realm of study. Along with other arthropods, blow flies depend on animal carcasses, including human, for their development (Anderson 2001). In their immature stages, these sarcosaprophagous insects feed on the remains whilst maturing. The gravid adult female blow fly oviposits on the remains at the natural orifices of the body, unless there are wounds; the blow fly has a great affinity for blood and other bodily fluids. These eggs will hatch into first instar or first stage larvae. The larvae begin feeding and increase in size as they feed. They are limited in the size they attain as their body surface is chitinous in structure and, while flexible, does not expand (Goff 2000). For growth to continue, the blow fly must shed its skin and moult into the next stage, or instar (Goff 2000). Feeding becomes much more vigorous with each moult (Haskell et al. 1997) as their mouthparts become larger and further developed, so are able to break tougher tissues.

The third instar can be divided into two distinct behavioural stages. In the first of these, the insect feeds until it reaches its maximum size as it does in the other instars. The second stage, referred to as post feeding and often incorrectly referred to as prepupal, is the wandering stage (Greenberg and Kunich 2002).

The larva is no longer feeding and wanders to find a suitable place to pupate

(Byrd 2001), usually in a drier place such as soil, clothing, or carpet. The larva must then rely on the food it has stored in its crop. These stores become depleted as the insect prepares for its next stage. Once the crop is empty and a suitable pupation site has been located, the larva stops moving and shrinks in length. It begins to form fat bodies, which obscure the organs, giving the insect an opaque appearance (Anderson 2000; Greenberg and Kunich 2002).

Pupariation occurs and the outer cuticle hardens and tans or darkens into stages of reddish brown eventually becoming a dark brown colour (Greenberg and Kunich 2002). Pupation then begins, and the pupa metamorphoses into an adult within this dark brown casing, using a process called histolysis (Goff 2000). Many tissues completely dissolve and new structures develop, eventually forming an adult fly (Goff 2000).

After a period of time, the adult blow fly emerges from the pupal skin. The adult cracks open the anterior end of the casing, referred to as an ecdysial cap, by everting a membranous sac, its ptilinum repeatedly (Greenberg and Kunich

2002). The adult blow fly at this stage, however, has wrinkled wings and is not entirely functional. The cuticle must harden and the wings must expand (Goff

2000). Even then, the adult blow fly is not fully developed; the insect is anautogenous, requiring a protein source to develop its gonads (Stoffolano et al.

1995).

Blow flies progress through this life cycle at a predictable rate. As long as all the required variables are known, a forensic entomologist can determine the time it takes to reach each of these developmental stages. Nonetheless, several factors must be considered. Inactivity at night and rainfall can prevent adult blow flies from reaching a body quickly (Erzinqlioglu 1983). The nearby remains of another organism at a different stage of decomposition could attract insects that are older than expected. Drugs in the body can drastically affect insect development rates and must be considered when determining a PMI. Other factors that may delay or prevent the insects from inhabiting the body such as the remains being wrapped, confined or buried (Anderson 2001 b) may also influence the estimated PMI.

All circumstances affecting the decomposition of the body or preventing insects from reaching the body must be considered (Erzinqlioglu 1983). Providing a minimum estimated PMI is, therefore, most conservative from the forensic entomologist's perspective. The deceased may have expired before the insects

inhabited the body but it is extremely unlikely that the insects were there prior to death, unless the person suffered from cutaneous myiasis which only occurs

under certain rare circumstances (Sherman and Pechter 1988). The most

important consideration in determining a minimum PMI is the developmental rate and sequence for the species associated with remains found in a particular

location.

The blow fly has a predictable life cycle based primarily on the species

and temperature. Therefore, the development data for the species at those temperatures can be used to predict how long it takes that species to reach each

stage of development. This conclusion as to the length of time that the insects

have been on the body then becomes a reasonable estimate as to the minimum

amount of time that the person or animal has been deceased (PMI).

The focus of this thesis was to study the developmental times and rates of

one species of blow fly (P. terraenovae) under a variety of temperature regimes,

for the purpose of applying their development data to post-mortem intervals. 2 LITERATURE REVIEW

2.1 Theory

The blow fly follows a very predictable life cycle (Chapman 1980). Most insects are poikilothermic ectotherms and, therefore, their development

(Chapman 1980; Beck 1983; Ames and Turner 2003) and changes in their metabolic rate are temperature dependent (Chapman 1980). It is, therefore, possible to determine how long it takes for a particular species of blow fly to develop to each stage of its life cycle at a given temperature. It is understood that the rate of development increases with increasing temperature. The opposite is also true; the rate of development decreases with decreasing temperature. There is, however, a limit to each of these statements.

Temperature thresholds occur at both ends of the gradient. That is, there is a minimum temperature threshold (Tmi,, lower temperature threshold) and a maximum temperature threshold (T ,,,, upper temperature threshold), beyond which insect development is negligible.

Development beyond each of these thresholds is so slight that it is considered insignificant, as is the development that occurs with the onset of diapause or quiescence, two forms of insect dormancy (Myskowiak and Doums

2002). Insect dormancy is a strategic condition used to foster survival under unsatisfactory conditions, usually at lower temperatures. Death generally results when temperatures exceed the higher lethal limit (Greenberg and Kunich 2002).

Diapause is a result of a diminished release of ecdysteroids into the haemolymph (Richard and Saunders 1987; Vaznunes and Saunders 1989;

Saunders 1997), which delays development and promotes survival through unfavourable conditions (Myskowiak and Doums 2002; KoStal2006). Diapause is brought on by a combination of factors, including genetic and environmental cues (KoStal2006). For the most part, diapause is induced by a short day photoperiod (LD 12:12h or less) (Numata and Shiga 1995; Saunders 1997) that is experienced by the maternal parent (Vinogradova and Zinovjeva 1972;

Saunders et al. 1986; Vaznunes and Saunders 1989). The onset of diapause in

Calliphoridae is typically seen in the post feeding larval stage and adult stage

(Greenberg and Kunich 2002). P. terraenovae is said to over winter in the adult stage (Danilevskii 1961 as cited in Vinogradova 1987; Wood and Nordin 1976;

Numata and Shiga 1995), although Calliphora vicina (R-D), a species also found in subarctic regions, can undergo diapause in both larval and adult stages

(Vinogradova and Bogdanova 1985; Numata and Shiga 1995).

Quiescence is defined as a delay in insect metabolism (Myskowiak and

Doums 2002) that appears as a stall in insect development but occurs only for a short period of time (Myskowiak and Doums 2002). Quiescence occurs as an abrupt reaction to limitations in physiological resources but just as quickly as it is entered, it is exited once all requirements are met (KoStal 2006). Although development essentially comes to a standstill at the upper and lower thresholds, rate of development is expected to increase linearly from the lower threshold. In actuality, a linear relationship only exists in the central temperature range of development (Laudien 1973) (Figure 2-1). The development rate slows at each end of this central temperature range, producing a sinusoidal effect. The Tminis usually assumed to fit this linear relationship when in fact it is beyond the linear portion. Therefore, error is created when a linear regression is used alone to estimate Tmin.

Figure 2-1 The relationship between development rate and temperature for blow flies where Tmi, is the minimum temperature threshold, Tmi,,(,,, is the estimated minimum temperature threshold, T,,, is the maximum temperature threshold and Optd, is the optimum temperature for fastest rate of development Insect development slows to near zero as temperatures approach the Tmin creating difficulty in determining a precise Tmin(Ames and Turner 2003). In addition, development drops drastically beyond the optimal development temperature (that temperature at which development is fastest, and greatest hatch and least mortality occur (Laudien 1973)) until the Tmaxis met. Determining a precise Tmaxis just as difficult as determining a Tmin due to the drastic drop in survival of the insects beyond the optimal temperature (Wagner et al. 1984).

2.2 Rationale

Protophormia terraenovae was chosen as the species of study for three reasons: firstly, due to its prevalence in the Lower Mainland of British Columbia and throughout the world. P. terraenovae is well established in the Coastal

Western Hemlock biogeoclimatic zone in which the Lower Mainland falls

(VanLaerhoven and Anderson 1999). In fact, P. terraenovae is an abundant species in colder regions and is considered to have a Holarctic distribution

(Smith 1986; Grassberger and Reiter 2002) and has often been used in local forensic entomology cases.

Secondly, P. terraenovae was chosen because of a lack of research on this forensically important, widespread species. P. terraenovae has been studied to determine the effects of constant temperatures on its rate of development

(Davies and Ratcliffe 1994; Grassberger and Reiter 2002). Davies and Ratcliffe

(1994) also researched the effects of low temperature on P. terraenovae, as did

Myskowiak and Doums (2002). However, Myskowiak and Doums (2002) were looking primarily at the effects of refrigeration. Numata and Shiga (1995) studied the induction of adult diapause by temperature and photoperiod in P. terraenovae. Clarkson et al. (2004) raised P. terraenovae at outdoor fluctuating temperatures and compared the development to the development at the mean constant temperature of 20•‹C. Nonetheless, studies on P. terraenovae that look at development at both constant and fluctuating temperatures are rare.

Lastly, but somewhat related to the two previous points, it has been noted that some holarctic blow fly species do not display the same growth rates as seen in other geographic zones and, therefore, research should be done on the same species in different regions for such cases where the findings may vary

(Grassberger and Reiter 2001 ; Reiter and Grassberger 2002). Hence, results for

P. terraenovae in the Lower Mainland of British Columbia may differ from results for P. terraenovae found elsewhere.

An option that may be considered to determine minimum development times is the use of isomorphen- and isomegalen-diagrams (Grassberger and

Reiter 2001 ; 2002). The diagrams incorporate development over a range of temperatures to each stage (isomorphen) as well as include larval length

(isomegalen). Both can be used readily to determine the development stage of

any located species when development occurs at constant temperatures

(Grassberger and Reiter 2001 ; 2002) but there are no available

isomorphen/megalen diagrams for P. terraenovae of British Columbia nor do

ambient temperatures remain constant. When temperatures fluctuate, Reiter and

Grassberger (2002) suggest that a degree-hour or degree-day model is best. In

this case, PMI results should be expressed as thermal or heat units. A degree day model can conveniently be used when looking at constant and fluctuating temperatures, unless development is faster at fluctuating temperatures than it is

at the mean temperature.

Insect developmental data can be converted to thermal units, as either

accumulated degree days or hours, by multiplying the minimum time spent to

reach each stage, by the temperature at which the data were generated. The

number of degree days or hours is determined by a simple reverse summation

process (Greenberg and Kunich 2002).

Upper and lower temperature thresholds must be considered when

determining thermal units. For the most part, development of the insect does not

occur when these thresholds are reached. Typically, death of the insect occurs

when the upper threshold is crossed, while development stops for the time that

temperature is below the lower threshold (Greenberg and Kunich 2002). To

account for the lower threshold, the time spent below the threshold is subtracted

as it is assumed that development does not occur during this period (Greenberg

and Kunich 2002). However, in the case of ambient temperatures that fluctuate

to below the lower threshold, development ceases as the temperature drops

below the Tmin,unless a maggot mass has formed that raises the temperature.

Perhaps some may consider this cease in development as part of the

development duration. To account for this, ADDS (accumulated degree

days)should be determined consistently with either the lower threshold

subtracted or with a simple base of zero degrees Celsius assumed. The

threshold used and the method of threshold calculation should be stated. To integrate the threshold when determining the number of degree hours

(days), Grassberger and Reiter (2002) employ a calculation that determines a thermal constant (K). K=y(t-tL),where y is the development time, t is the temperature and tL is the lower temperature threshold. This equation works well for determining degree hours (days) at constant temperatures but must be adjusted for fluctuating temperatures and is therefore replaced with:

K~uctuatin~'ya(ta-t~) + Yb(tb-t~) + -.- + Yn(tn-t~)

Hence, an addition of all the ADDS at each temperature must occur rather than carrying out a calculation using the average temperature and total time as suggested by Greenberg and Kunich (2002) for varying ambient temperatures.

It is of utmost importance that an accurate Tminbe known otherwise an error will occur in determining the post-mortem interval. A Tminis necessary for determining the ADDIHs (accumulated degree dayslhours) that are required for the insect to reach a particular stage of development. Although the ADD and

ADH models have been effective thus far, there are many limitations. Most models are based on a single constant temperature and it is rarely the case that development occurs at a constant temperature. Ambient temperature varies with photoperiod (Beck 1983), hence decreasing at night. In addition, more than one temperature is required in order to conduct a regression analysis. Greenberg and Kunich (2002) propose that a mean temperature be used to determine the

ADH's for varying temperatures in the field. However, both Clarkson et al. (2004) and Davies and Ratcliffe (1994) found that fluctuating temperatures produced different development times from their mean temperature. Furthermore, maggot masses may form, increasing the temperature by many degrees (Greenberg and

Kunich 2002).

Frequently, the applied temperature threshold that is used is incorrect

(Greenberg and Kunich 2002) as it is usually an estimate based on the extrapolation of a linear regression (Liu et al. 1995; Grassberger and Reiter

2002; Ames and Turner 2003). Additionally, it can be difficult to determine the lower temperature threshold because insect development can be drawn out at times with very slow to almost no development at very low temperatures (Davies and Ratcliffe 1994; Ames and Turner 2003). Survival may also be reduced near the lower temperature threshold.

Several methods of determining the ADDS are employed (Pruess 1983) and as there is not a single strict method that is adhered to, variation in values is inevitable. In addition, most methods are based on approximations of the actual values, for example the use of estimated temperature thresholds and the application of minimum and maximum daily temperatures (Reiter and

Grassberger 2002). The use of all of these approximations increases the variability of the estimated PMI. Therefore, it can be said that as the PMI increases, the range of ADDIHs increases and creates more variability in the outcome (Reiter and Grassberger 2002).

Furthermore, applying ADDIH data that span the entire life cycle to particular stages of the life cycle will not necessarily work. An assumption that development occurs at the same rate for each stage is being made (Greenberg and Kunich 2002) when it is known that growth rates, temperature thresholds and development change with each stage of the lifecycle (Williams 1984).

Ames and Turner (2003) suggest that it is possible that a non-specific

ADH value is being used amongst entomologists as variability exists in the published data. They then propose that these differences may be a result of the geographic location of the species studied or that they occur simply due to the differences in egg batches as a result of genetic mutation.

Curvilinear regression methods have been applied to overcome the errors that occur with linear methods; however, they come with their own difficulties.

Curvilinear methods involve the application of an equation that best fits the development rate temperature curve rather than just a portion of the curve.

There is no one method that can be agreed upon among the scientific community and the application of this method is much more difficult than a linear method

(Howe 1967). The linear method seems to work just as well and sometimes better, so it predominates (Pruess 1983; Higley and Haskell 2001). Several other parameters, usually four to five, are required to determine the equation of the curve in the curvilinear methods (Howe 1967; Briere et al. 1999), which increases the opportunity for the accumulation of errors (Higley and Haskell 2001).

Because more parameters are required, the method is not a realistic option when determining a PMI in a death investigation. 3 PROTOPHORMIA TERRAENOVAE DEVELOPMENT AT CONSTANT TEMPERATURES

3.1 Introduction

Experiments have been performed concerning the effects of low temperature on the development of blow flies (Dallwitz 1984; Buei 1986; Block et al. 1990; Davies and Ratcliffe 1994; Johl and Anderson 1996; and Myskowiak and Doums 2002) but few experiments have examined P. terraenovae development at the range of temperatures most frequently encountered in the

Lower Mainland of British Columbia.

Davidson (1944) looked at the speed of development of insect eggs at constant temperatures. In 1975, Ash and Greenberg studied Phaenicia sericata

(Meigen) and Phaenicia pallescens (Shannon) but only considered three constant temperatures, 19, 27 and 35OC. Grassberger and Reiter (2001 ; 2002) studied the effects of a select range of constant temperatures on its development of some species of Calliphoridae including P. terraenovae. Numata and Shiga

(1995) researched the induction of adult P. terraenovae diapause by photoperiod at constant temperatures. The effects of refrigeration on blow fly development has been examined for C. vicina (Johl and Anderson 1996) and P. terraenovae

(Myskowiak and Doums 2002). Davies and Ratcliffe (1994) studied developmental rates of some Calliphoridae including P. terraenovae. Minimum temperature thresholds (Tmin6s)have been determined for P. terraenovae

(Marchenko 2001; Grassberger and Reiter 2002); however, as stated previously, geographic differences may have an effect on development rates.

3.1 .ICore Constant Temperature Experiments

A crucial element of this research was to determine an accurate Tminthat

may be applied to and not create variability in the accumulated degree days

(ADDS). In order to achieve this, development was studied over a range of temperatures. P. terraenovae were raised at a range of constant temperatures to determine the minimum thresholds and development rates. The rate of development for each stage was graphed and linear regressions were used to

determine the estimated thresholds.

3.1.2 Objectives

The specific objectives of this research were to:

1. Determine the development times and rates of each stage of P.

terraenovae collected from the Lower Mainland of British Columbia at a

variety of constant temperatures

2. Determine the minimum temperature thresholds for each stage of

development by both experiment and calculation

3. Compare the empirical minimum temperature thresholds to the thresholds

determined in a linear regression for each developmental stage

4. Determine the optimal temperature for development of each stage 3.2 Methods and Materials

3.2.1 Collection of Experimental Insects

Two methods of collection were used. Firstly, inverted cone traps were built from hollowed coffee canisters, wire mesh, and plastic bags (Anderson

2000; Byrd and Butler 1996; 1997; 1998). The P. terraenovae colonies originated in July 2001; traps were baited with pieces of beef liver and were put out in different locations of the Lower Mainland of British Columbia in order to catch blow flies of differing lineages. The second method of collection was to gather immature insects from animal carrion used in research for educational purposes. These insects were then reared to the adult stage, and the adult blow fly was anaesthetized or chilled to immobilize it so that it could be identified under a dissecting microscope.

As the P. terraenovae were identified, they were placed into four different

75cm3 cages. These flies were allowed to reproduce and the subsequent generations were utilized in these experiments. In order to ensure that the lab

colonies did not suffer from founder effect or adaptation to lab conditions,

continuous trapping and addition of insects occurred throughout the experiments.

The adult or stock colonies were provided with a food source of 50:50 milk

powder and cubed sugar as well as water (Byrd and Butler 1996; 1997; 1998) ad

libitum. The water container was closed to prevent drowning and dental wicks

provided a continuous source of water (Byrd 2001). 3.2.2 Rearing of the Insects

A protein source, beef liver, was placed into each of the stock cages so that the blow flies could feed and develop their reproductive organs (Harlow

1956; Stoffolano et al. 1995). A second, fresh protein source was used as an oviposition medium. Beef liver was selected for this purpose due to its inexpensive nature and its convenience of handling. It was also selected for its forensic investigative purposes as recommended by Byrd (2001). It has been suggested that cat food can be used as a rearing medium but a meat source, like beef liver, is preferable because it is closer forensically to humans and thus more relevant to this research (Byrd 2001). In addition, cat food can vary between brands, flavours and batches and so beef liver has a more consistent composition. However, beef liver was not a perfect rearing medium because it desiccated quickly (Byrd 2001). Therefore, the continuous addition of fresh beef liver to each of the cages was necessary.

For oviposition, fresh thawed beef liver was placed into black 35 mm film canisters positioned on their sides in the cages, providing an ideal environment for oviposition (Grassberger and Reiter 2001 ; 2002). After the eggs were laid, they were divided equally between control and experimental samples. Because the eggs are so tiny, it was necessary to do this under a dissecting microscope

(Meiji EMZ-5TR). It was important to do this step as rapidly as possible to prevent the eggs from desiccating due to the warm lights of the microscope.

Samples were not held underneath the microscope for more than ten minutes. A moistened children's paintbrush (~i~uitex@,size three, natural round) was used to transfer the eggs to each of the rearing containers.

The eggs collected from the stock colonies were divided into equal batches of roughly 100 and placed into separate 4.5 L wide-mouthed glass rearing containers. Some species die rather than pupate if they are confined in too small an area (Byrd 2001), however, the jars were large enough that they provided ample space for the insects to move and develop without restraint.

Each container held 125 mL of pine sawdust dampened with 15 mL of water.

The sawdust allowed for migration of the post feeding third instar larvae and the dampness prevented desiccation.

A paper towel was placed on top of the sawdust and under approximately

2009 beef liver to act as an interface between the two. The paper towel prevented the eggs and larvae from drowning by soaking up any extra fluids that may have been released by the larval rearing medium. The rearing medium, beef liver, was placed on the paper towel along with approximately 100 eggs. A tight fitting lid was not used on the containers in order to allow the release of ammonia (gaseous waste product), a toxin that may kill or interfere with the natural activities of the larvae (Byrd 2001 ). Instead, the containers were kept secure using paper towel lids and elastic bands. Industrial paper towel is well ventilated and prevented any unnecessary spillage if the container was inadvertently overturned.

Beef liver was added as the experiments advanced so that a food source was always available (as often as every two days when the insects reached 3rd instar) and nutrition did not become a limiting factor. With an adequate food source in each jar, 100 larvae did not result in increased metabolic heat; however, if the food source became depleted, larvae would have congregated on the remaining resource, possibly resulting in an increased temperature (Byrd

2001) that would have biased the experiments.

3.2.3 Experiments at Different Constant Temperatures

Recently laid P. terraenovae eggs were collected from three or four different cages and divided up into jars such that eggs from each cage were added to each temperature chamber. The experiments at 15,20, and 25OC were conducted first so only three jars were considered necessary at each temperature. At lower temperatures when it became apparent that mortality would be high, a fourth jar was added. By dividing the eggs collected from cage

A into three separate jars, each jar became the first experiment in each chamber.

Hence, Experiment A in each chamber originally came from cage A and

Experiment B in each chamber came from Cage B, and so forth. It was not necessary to have a control chamber in this case because the egg batches were divided up so that the other experimental chambers acted as controls for each of the other experiments. Development was observed and noted from hatch to emergence as an adult fly.

3.2.4 Environmental Chambers

The use of environmental chambers (~onviron@E7/2 plant chambers) allowed for temperature, relative humidity, and lighting to be controlled and set over a 24-hour clock. In all cases, relative humidity remained at a constant 75

?5% and lighting followed a photoperiod of 14:lO (L:D). The environmental chambers were computerized and therefore implemented, monitored and recorded the set information. A smartbutton@datalogger (ACR Systems Inc.) was used to record the temperature of the chamber at intervals of 30 minutes, for the purpose of monitoring and recording all changes. In addition, a s is herb rand^ mercury thermometer, was used to confirm the temperature settings of the chamber itself. Chambers were set at temperatures of 9.8, 11, 13, 15, 20, 25,

28,30, and 32OC.

The temperature 9.8OC was selected as the lowest of the experimental temperatures for two reasons. Firstly, preliminary experiments indicated that eclosion would not occur below or at g0C (Warren, unpublished). Secondly, the literature indicates that 9.8OC, by linear regression, is the Tminfor P. terraenovae from oviposition to pupariation and that it is 8.95OC for the entire immature development (Grassberger and Reiter 2002). Alternatively, Marchenko (2001) indicates that the lower threshold is even lower at 7.8OC from egg to pupa.

Eggs were collected from the stock cage within one and a half hours of the start of oviposition to ensure that development occurred in the desired environments (Howe 1967). All trials (jars containing eggs) were added to the chambers at the same time, within two hours of the beginning of oviposition, so that exposure was the same for all insects because blow flies develop at different rates at different stages of their life cycle (Levot et al. 1979; Williams 1984). 3.2.5 Insect Sampling

Every 12 hours, close to the photoperiod change from light to dark or vice versa and sometimes more frequently, the insects were examined.

Generalizations about the insects' appearances were made and the stage of development was established (for example, the number of spiracular slits was

used to determine the stage of the larvae). Observations about the populations were made. As the insects reached each stage of development, times were

recorded and general notes were made about the stage.

The number of posterior spiracular slits alone was used to determine

stage (although crop size and behaviour were also taken into consideration at the

post feeding stage). Measurements of length were not taken at any of the stages

as length has been shown to overlap in the different stages of development

(Greenberg 1991; Greenberg and Tantawi 1993) and can readily vary if

starvation is a factor (Anderson 2000). Furthermore, I wished to minimize insect

exposure to the warm lights of the microscope and to the temperature outside of

the environmental chambers as these may have influenced development time.

To determine if the insects had reached first instar, the entire larval

medium was placed under a dissecting microscope and examined for signs of

eclosion. Ascertaining whether the insects had reached any of the subsequent

larval stages was done by taking a sample of 20 insects and examining the

posterior spiracular slits under the microscope for insect stage. With each moult,

the insect develops another slit in each of its posterior spiracles or openings to

the trachea or breathing tubes that run the length of the insect's body. At 1'' instar, the insect has a single slit in each spiracle, at 2""nstar, the insect has two slits in each posterior spiracle and so forth.

The post feeding third instar stage was determined based on the behaviour of the insects, that is whether they had left the food source to wander in the sawdust, and their appearance, that is whether the crop (food storage organ) was very apparent or not. The pupal stage was distinguished from this stage by any evidence of motionlessness and tanning. Finally, when the blow fly had emerged, it was declared an adult. Emergence was considered most recent when the wings were still wet and wrinkled.

Minimum developmental times and rates were established when the first insect was observed to have changed to the following stage. The mode of development was defined as the time at which the largest number of insects entered the stage.

Although it has been suggested that temperature probes be inserted into each sample to determine if there is a difference between the environmental temperatures and the maggot temperature (Byrd and Butler 1996; 1997; 1998;

Byrd 2001), this was not done because the number of insects (100) appeared small enough compared to the large jar size (4.5 L) not to cause a sufficient difference in temperature. As well, minimal handling and interference to insect activity was favoured. I preferred to avoid disrupting the insects any more than seemed necessary. Disturbing the insects during each shift in photoperiod was considered sufficient. 3.3 Results

At 9.8OC , eggs did not hatch and, therefore, development was not observed at further stages and no data were recorded. However, egg eclosion did occur at 1I0C (Table 3-1). At this low temperature, hatch was minimal at only approximately 10%.

Three separate trials were required in order to collect data at both this temperature and 13OC, as the development of mould on the liver became an issue. The mould obscured the insects and appeared to increase egg mortality.

However, as long as the insects reached IS'instar before the mould formed, there appeared to be no effects on mortality. This was achieved in the third trial.

Even so, the minimum development time to hatch was missed in Jar 3 as it was obscured by a layer of mould. At 1I0C, P. terraenovae did not complete immature development but did reach 3rdinstar. The majority of the insects died during the 2ndinstar and in only one jar did P. terraenovae reach 3rdinstar. Approximately 4% of Jar 1 P. terraenovae reached 3rdinstar but died in this stage.

Table 3-1 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of ll•‹C

(days) Stage I Jar I Jar 2 Jar 3 Jar 4 I Mean S.E.

1 instar 3.5 4.0 obscured 6.0 Znd instar 12.0 10.5 15.0 11.5 3rd instar 21 .O died died died Post feeding died nla nla nla Pupal stage nla nla nla nla Adult stage nla nla nla nla

The mode of development for P. terraenovae to reach 3rdinstar at a constant temperature of 11•‹C is based on a single jar of insects (Table 3-2). Most of the insects died as 2ndinstar larvae. Table 3-2 The mode of development for P. terraenovae to reach each stage at a constant temperature of 1I0C

Stage Jar 1 Jar 2 Jar 3 Jar 4 I Mean S.E. 1 lStinstar 6.5 7.0 8.5 8.75 7.69 k0.55 2ndinstar 14.0 14.0 15.0 11.5 13.62 f 0.75 3rd instar 26.0 died died died 26.0 Post feeding died n/a n/a n/a n/a Pupal stage n/a n/a nla n/a nla Adult stage n/a n/a n/a n/a n/a

P. terraenovae completed immature development at 13OC as shown in

Tables 3-3 and 3-4. On the third trial, as mentioned above, mould did not prove to be an obstacle and the experiment was completed. However, complete development only occurred in two of the four jars. The majority of insects in Jar 2 died before reaching 3rdinstar and the last insect died in the post feeding stage and did not pupate whereas the insects in Jar 3 did not hatch. Consequently, the mode of development to the post feeding stage in Jar 2 is based on a single insect. Of the approximately 100 eggs that were placed into each jar, only nine flies emerged in Jar 1 and four flies in Jar 4. The insects in Jar 3 were not viable and it is assumed that the mould prevented hatch as the insects from the same egg batch hatched in the other chambers. Table 3-3 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 13•‹C

(days) Stage Jar 1 Jar 2 Jar 3 Jar 4 I Mean S.E. I I

1St instar 3.0 3.5 died 4.0 Znd instar 8.0 10.0 n/a 10.0 3rdinstar 17.0 16.0 n/a 17.0 Post feeding 29.0 26.0 n/a 27.0 Pupal stage 31 .O died n/a 29.0 Adult stage 52.0 n/a n/a 55.0

Table 3-4 The mode of development for P. terraenovae to reach each stage at a constant temperature of 13OC

(days) Stage Jar 1 Jar 2 Jar 3 Jar 4 I Mean S.E.

1 instar 6.0 6.5 died 8.75 2ndinstar 9.5 11.0 n/a 11.O 3d instar 20.0 22.0 n/a 21 .O Post feeding 30.0 26.0 n/a 28.0 Pupal stage 31 .O died n/a 31 .O Adult stage 53.0 n/a n/a 55.0

The experiments at 15, 20, and 25OC were conducted first so only three jars (insects from three colonies) were used at each temperature. The minimum

time to reach 2ndinstar at 15"C, 6 days, was the same for all jars. Jar 1 varied

from the other jars at the other stages. Jar 1 also had a much greater adult

emergence versus egg ratio than the other jars raised at 15OC. With

approximately 100 eggs per experimental jar, 139 flies emerged in Jar 1 as

compared to fewer than 10 in the other jars. Jar 3 had one pupal case with a

blow fly that failed to emerge as well as one dead larva. Table 3-5 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 15•‹C

(days) Staae I Jar 1 Jar 2 Jar 3 Mean S.E.

1 instar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

The mode of development to each of the stages at 15OC is fairly similar for all the jars, but varies in Jar 3 the most in the transition to the pupal and adult stages

(Table 3-6).

Table 3-6 The mode of development for P. terraenovae to reach each stage at a constant temperature of 15•‹C

(days) Stage Jar 1 Jar 2 Jar 3 Mean S.E.

1St instar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

The minimum time to develop to each of the stages at 20•‹C is

comparable for each of the three jars and only varies by a very small amount

(Table 3-7). Table 3-7 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 20•‹C

(days) Stage Jar 1 Jar 2 Jar 3 Mean S.E.

1St instar zndinstar 3rdinstar Post feeding Pupal stage Adult stage

The development in each of the jars at 20•‹C was very similar and showed the most variation in Jar 2 at the post feeding stage (Table 3-8). Jar 2 was the only jar at this temperature that had a minor mould growth problem but a total of

88 flies emerged from this jar. In Jar 1, 200 flies emerged and a mere 45 flies emerged in Jar 3. Table 3-8 The mode of development for P. terraenovae to reach each stage at a constant temperature of 20•‹C

Staae Jar 1 Jar 2 Jar 3 1 Mean S.E

1St instar 2ndinstar 3'(' instar Post feeding Pupal stage Adult stage

The minimum developmental time for P. terraenovae to reach the first instar at a constant temperature of 25•‹C was one day in all three jars (Table 3-9).

The greatest variation in development, two days at the post-feeding stage, was seen in Jar 2 as compared to the other jars.

Table 3-9 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 25OC

(days) Stage Jar 1 Jar 2 Jar 3 Mean S.E.

lStinstar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

Unlike the minimum development times of P. terraenovae at 25OC, the mode of development in Jar 2 is rather consistent with only slight variation from the other two jars (Table 3-10). At this temperature mould was not a factor. In fact, mortality was minimal at this high temperature and 219 flies emerged in Jar

1, 105 flies emerged in Jar 3 but a mere seven flies emerged in Jar 2. There were no dead larvae or pupae in this jar.

Table 3-10 The mode of development for P. terraenovae to reach each stage at a constant temperature of 25OC

1 instar 2"d instar 3rd instar Post feeding Pupal stage Adult stage

The minimum development time to hatch to 1'' instar larvae at 2B•‹C was missed as it occurred between half a day and one day. The minimum development time to moult to 2ndinstar was two days in all four jars and was the same as the mode of development in Jars 2,3 and 4 as shown in Table 3-12. Table 3-11 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 2a0C

(days) Stage Jar 1 Jar 2 Jar 3 Jar 4 I Mean S.E.

lStinstar 2nd instar 3rdinstar Post feeding Pupal stage Adult stage

The greatest number of flies emerged in Jar 4 with 304 flies. In Jar 3, the final count included 80 emerged flies, five dead larvae and six dead pupae. Jar 2 only had 33 flies emerge and three dead flies that did not complete emergence from their casings. Finally, Jar 1 had a mere five blow flies emerge.

Table 3-12 The mode of development for P. terraenovae to reach each stage at a constant temperature of 28OC

(days) Stage Jar I Jar 2 Jar 3 Jar 4 Mean S.E

1 st instar Pdinstar 3rdinstar Post feeding Pupal stage Adult stage

Table 3-1 3 shows the minimum developmental times to reach each stage

at a constant temperature of 30•‹C. At this temperature, the minimum development time to hatch was also missed. In all jars, the minimum time to hatch occurred between half a day and one day. However, the mode of development to reach hatch occurred at one and a half days (Table 3-14). Both the minimum developmental time and mode of development to reach 2"dinstar at

30•‹C were the same in all jars at 2 days. Development was consistent among the jars and only appeared to vary notably in Jar 3 at the pupal stage

In Jar 4, 75 flies emerged whereas four flies did not emerge from the

pupal casing and a single dead larva was found. Only 40 flies emerged in Jar 3

and there were four dead larvae and one fly that did not emerge. In Jar 2, 41 flies

emerged as adults and a single fly did not emerge from the pupal casing. Only

14 adult flies emerged in Jar 1.

Table 3-13 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 30•‹C

Stage Jar I Jar 2 Jar 3 Jar 4 I Mean S.E.

lStinstar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage Table 3-14 The mode of development for P. terraenovae to reach each stage at a constant temperature of 30•‹C

(days) Stage Jar 1 Jar 2 Jar 3 Jar 4 I Mean S.E.

1 instar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

The minimum developmental times for P. terraenovae to reach each stage of development at a constant temperature of 32OC are shown in Table 3-15. Also at this temperature the minimum developmental times to reach hatch were missed as they too fell between half a day and one day. In all jars, the minimum development time to reach 2"dinstar was two days. The development in Jar 2 was not substantially different but was different from the other three jars. The insects in the jar took longer to reach the 3rdinstar and post feeding stages. The insects in Jar 2 did not complete development and died before reaching the

pupal stage. Unfortunately, this could not be explained as the insects from the

same colony developed normally in the other temperature chambers. Table 3-15 The minimum developmental times of P. terraenovae to reach each stage at a constant temperature of 32OC

(days) Stage Jar I Jar 2 Jar 3 Jar 4 I Mean S.E.

IS'instar 0.5

Similarly to 30•‹C, the modes of development to 1 and 2"d instar at 32OC were one and a half and two days, respectively (Table 3-16). However, unlike the minimum developmental time to reach each of the stages in Jar 2, the mode development to reach each stage in Jar 2 was relatively consistent with the other jars.

In Jar 4, 70 flies emerged and four flies did not exit from the pupal casings. A total of 40 flies emerged in Jar 3 with three dead pupae. The experimental insects died in Jar 2 before they could pupate and in Jar 1 only three flies emerged as adults and one fly did not emerge from the casing and finally one dead larva was found amongst the sawdust.

Jar 1 from each of the 28,30 and 32OC experiments all had low numbers emerge at the end and they were all from the same egg batch. It is possible that the egg batch was smaller to begin with and so each sample may have been less than 100 eggs. Table 3-16 The mode of development for P. terraenovae to reach each stage at constant temperature 32OC

Stage Jar 1 Jar 2 Jar 3 Jar 4 I Mean S.E.

1 instar 1.5 1.5 1.5 1.5 2ndinstar 2 2 2 2 3rd instar 3.5 4.25 4.25 3.75 Post feeding 6.5 6.5 6.5 6.75 Pupal stage 7.5 nil 8 7.75 Adult stage 12 nil 12.25 12.25

The standard error for each of the temperature recordings from the temperature chambers could not be included as the datalogger recordings were lost due to a computer malfunction. Only the recordings for 3Z•‹C were recovered and the mean constant temperature and standard error were determined to be

32.0k0.003•‹C

As illustrated in Figure 3-1, the longest stage of immature development was the pupal stage lasting an average of approximately 41 O/O of the total minimum developmental time and approximately 39% of the total mode of development. At 11 OC, 100% immature development was divided amongst only the first three stages, egg, 1'' and zndinstar, as development did not occur beyond these stages.

On average, excluding the 11 OC findings, P. terraenovae spends a minimum time of 7.3% developing in the egg stage and 9.3% as 1'' instar larvae.

In addition, the minimum percent of time as zndand 3rdinstar larvae is 12.5 and

16.2%, respectively. The percent minimum development time as post feeding larvae at each of the temperatures varied rather significantly. The percent of mode of development in the egg stage is 12.4 and only 5.8 as 1" instar. The minimum developmental times and modes of development for P. terraenovae were plotted for each stage of development as a function of development rate (days-') versus temperature (OC) in the following figures. The linear portions of each plot of minimum developmental rates and mode of development were extrapolated and used to determine the x-intercept or that point at which development is negligible according to linear models (Ames and

Turner 2003; Bourel et al. 2003). The equations, R* values and x-intercepts are included in Table 3-1 7 Egg 1st instar 2nd instar 3rd instar Post feeding Pupal

Percent Minimum lmmature Development in Each Stage

Egg 1st instar 2nd instar 3rd instar Post feeding Pupal

Percent Mode of lmmature Development in Each Stage

Figure 3-1 The percent minimum developmental time (a) and mode of development (b) of P. terraenovae in each stage at constant temperatures of 11,13,15,16 (chapter 4), 20,25,28,30, and 32OC Table 3-17 The equations of the regression lines, the R-square values and the x- intercepts for the linear regressions of the mean minimum developmental times and mode development to each of the stages

Stage Mode

1 instar

2ndinstar

3rdinstar

Post feeding

Pupal

Adult

Eclosion did not occur at 9.8OC, but did occur at 1I0C and thus the actual

Tminto reach IS'instar falls between these temperatures. Figure 3-2 illustrates the actual rate of development for P. terraenovae over the range of temperatures for both the minimum developmental times and mode of development. Linear regressions of both the minimum developmental and mode of development rates were completed and the x-intercepts were extrapolated. In all cases the estimated Tminunderestimates the actual Tmin.

Development at 32OC was included in the graphs but was excluded when the slope and placement of each line was determined. Because the rate of development appears to decline from 30 to 32OC in the following figures, it was treated as an outlier and as a data point that falls beyond the temperature of optimum development.

1 --t Minimum 1

~ Temperature (OC) Figure 3-2 Rates of the minimum development and mode of development of P. terraenovae to 1'' instar at the researched temperatures including a linear regression and extrapolation of the minimum rates of development

Not only does eclosion occur at 11•‹C but so does moult to 2" and 3rd instar and therefore the Tminto each of these stages is less than 1I0C. At this time it can only be assumed that the Tminfor all three stages is greater than 9.8OC as development is temperature dependent. If this is the case, then the extrapolations of the linear portions of the 2ndand 3" instar minimum developmental and mode of development rates are also underestimating the Tmin for both the zndand 3rd instars (Figures 3-3 and 3-4). 0 10 20 30 40 Temperature ("C)

Figure 3-3 Rates of the minimum development and mode of development of P. terraenovae to 2ndinstar at the researched temperatures including a linear regression and extrapolation of the minimum rates of development

-Minimum '1- Mode

Temperature ("C) I Figure 3-4 Rates of the minimum development and mode of development of P. terraenovae to 3rdinstar at the researched temperatures including a linear regression and extrapolation of the minimum rates of development Development at the post feeding, pupal and adult stages did not occur at

11•‹C but did at 13•‹C. Therefore, the actual TminKsfor each of the stages fall between 11 and 13•‹C. However, the extrapolations for each of the development rates for the post feeding, pupal and adult stages are 9.1 9 and 9.4, 8.56 and 8.02 and 9.42 and 9.1 7, respectively. The extrapolated Tminunderestimates the temperatures that fall in these ranges in all three cases (Figures 3-5, 3-6 and 3-

7).

I Temperature ("C) Figure- 3-5 Rates of the minimum development and mode of development of P. terraenovae to the post feeding stage at the researched temperatures including a linear regression and extrapolation of the minimum rates of development 20 Temperature (OC)

Figure 3-6 Rates of the minimum development and mode of development of P. terraenovae to the pupal stage at the researched temperatures including a linear regression and extrapolation of the minimum rates of development

1 +Minimum

Figure 3-7 Rates of the minimum development and mode of development of P. terraenovae to the adult stage at the researched temperatures including a linear regression and extrapolation of the minimum rates of development 3.3.1 A Comparison with Previously Published Research

Figures 3-8 and 3-9 illustrate that differences occur in published developmental data for P. terraenovae from around the world. At 15OC, both

Marchenko (2001) and Grassberger and Reiter (2002) found that P. terraenovae

reached both the pupal and adult stages faster than was found here. When the

standard error is considered in all cases at 15OC, the data do not show any

significant differences. At 20•‹C, Clarkson et al. (2004), using some of the same

insect colonies found much shorter development times than the other

researchers. To reach the pupal stage, both Marchenko (2001) and Grassberger

and Reiter (2002) found that it took 13.1 days (&I.I [Grassberger and Reiter

20021). 1 find that it takes 11.5f0.29 days to reach the pupal stage which is

within the range of Grassberger and Reiter's findings. At both 25 and 30•‹C, the

findings are extremely similar.

Caution must be used when applying developmental data for P.

terraenovae to different geographic areas as P. terraenovae may develop at a

different rate than that of the area of research. As slight as the differences may

be, an accumulation of errors may occur and therefore a much greater error is

compiled. LOOLOOLOOLOOLOO d-mmCVCV.r.r A separate comparison of minimum developmental times to reach each stage was made to Kamal's (1958) minimum developmental times at 26.7OC

(Figure 3-9). Kamal published some of the first developmental data for

Calliphoridae and his data are often used by forensic entomologists. A comparison of Kamal's results for P. terraenovae obtained at two of the closest temperatures examined here, 25 and 28OC, found that Kamal's development times substantially underestimate these times. Kamal's data have shown much more rapid development rates than found by any other researchers. Kamal's methods of determining development times have been questioned as the experiments were not set up in incubators that maintain a constant temperature but instead were established at room temperatures, with little temperature control

(Anderson 2000). Therefore, there is some question as to whether the temperature of 26.7OC (80•‹F)was maintained. Also, it seems questionable that any researcher was able to examine the colonies hourly for the duration of the experiments, which were reported to take years. This may partially explain why the times to reach each stage of development do not add up to the time it takes for complete development (Anderson 2000). w Warren (Minimum 25OC)

w Kamal (1958) (26.7OC)

Warren (Minimum 28•‹C)

Stage of developmen

Figure 3-9 A comparison of the mean minimum developmental times for P. terraenovae at 25 and 28•‹C to Kamal's (1958) development data at 26.7"C.

3.4 Discussion

At 9.8OC eggs did not hatch. It is assumed that temperatures were not

high enough for the insect to break the chorion in order to eclose (Davies and

Ratcliffe 1994). Because hatching did not occur at this temperature, the following stages were not observed. Placing recently hatched eggs into a chamber maintained at 9.8OC was considered, however this would have further delayed subsequent temperature experiments as only two chambers would have been available. It is quite possible that development may occur at the following stages; however, if this is the case, development in all likelihood would only occur to 2ndand 3rdinstars as it did not occur beyond these stages at a more optimum temperature of 1 I0C. Regardless, development to 2" and drdinstar most likely does not occur at 9.8OC as the insects seem to tolerate lower temperatures increasingly better in the earlier stages. Just as the post feeding, pupal and adult stages do not occur at lower temperatures than the earlier stages, most certainly moult to 2ndand 3rdinstars do not occur at temperatures lower than when egg hatch will occur. The difficulties reaching 2ndand 3rdinstar at 1I0C also support the improbability of P. terraenovae developing at a temperature as low as 9.8OC.

At 1I0C, very high mortality was observed and the majority of insects died during zndinstar (Table 3-1). Others died during transition to 3rdinstar or during the 3rdinstar. As the sample size was diminishing (only a single 3rdinstar larva was left), no definitive observation could be made. However, it appeared that the larva was shrinking in size and still alive but no longer feeding and it is not believed to have entered the post feeding stage. It was thought to have either entered diapause or to be dying. Unfortunately, the insect died so there was no opportunity to determine if diapause was entered.

In all replicates, development from oviposition to adult emergence occurred at 13OC (Table 3-2). This is the lowest temperature examined at which complete development occurred. Presumably, the Tminfor complete immature development and development from oviposition to each of the post feeding and pupal stages falls between 11 and 13OC which is somewhat comparable to the findings of Greenberg and Tantawi (1993). They found that development slowed rapidly at a constant temperature of 12.5OC and that total immature development lasted 2,176.8 hours (-91 d). However, the Tmin for development from oviposition to 1 instar falls between 9.8 and 1I0C and because development occurred at 1I0C for 2ndand 3rdinstars their Tmin falls below 1I0C and in all likelihood is greater than 9.8OC. Neither of these ranges for Tminsupports the previous findings for Tminand very well may be a result of geographic differences

(Grassberger and Reiter 2001 ; Reiter and Grassberger 2002) or may simply be a result of the selected method of experimentation.

Substantial differences were observed between development in each of the jars at the lower temperatures in comparison with that seen at higher temperatures. This may have been due to the reduced sample sizes from mould growth, very high natural mortality and reduced egg hatch at the low temperatures.

At those temperatures including and below 13OC, mould growth was found to be a nuisance and may have contributed to the higher mortality found at the lower temperatures. Survival of experimental insects was poor at the lower temperatures so sample sizes were reduced to very low levels. I considered adding a fungicide to an artificial diet rather than using beef liver but Voss (2000) researched the effects of methylparaben (a common fungicide) on the development of P. terraenovae and determined that it lengthens the development time of the insect. Furthermore, Tachibana and Numata (2001) found that an artificial diet alone increases the duration of the larval stages for Lucilia sericata

(Meigen). However, through observation, it was found that as long as the insects reached 1'' instar, mould did not continue to impact survival; therefore, beef liver was maintained as the rearing medium. At 15OC, P. terraenovae had an extremely high adult emergence to egg ratio from one of the experiment insect colonies as compared to the other experimental colonies which had fewer than ten adults emerge. This difference cannot be explained except that it was probable that the egg batches were unequal.

Development was much more rapid approaching the higher temperatures.

At 20•‹C immature development was almost complete and adult emergence occurred at a minimum of 20.5 k0.29 days (Table 3-7), whereas at 25OC adults began emerging at 14.42 k0.42 days (Table 3-9).

Observation of the minimum time required for hatch was missed at 28, 30 and 32OC as it occurred between observations, that is, between half a day and one day (Tables 3-1 1, 3-13, and 3-1 5).

Overall, at the temperatures examined, development was fastest at 30•‹C.

At those temperatures approaching the optimal temperature, it is recommended that the experiments be checked more frequently than at 12 hr intervals as development is much more rapid and easily missed as unfortunately constant observation is not an option (Howe 1967).

The mean minimal percent of time spent in each stage differs primarily at the post feeding stage with a difference of 16%. At 13OC, the insects spent only

5% of their immature development time in that stage as compared to 21 % at

30•‹C. The insects spent the greatest percentage of time in the pupal stage with averages of 40.7 (minimum developmental time) and 39.4% (mode of development). The graphs at each stage of development, as expected, display a sinusoidal shape along with a linear portion (Figures 3-2 through 3-7). From these data, it appears that 30•‹C is the optimum development temperature because immature development was fastest, least mortality occurred at this temperature and hatch was greatest. However, further experiments must be done and development observed at higher temperatures to determine if a drastic drop in development occurs following this temperature, as it appears to do at

32OC. Also, if it is the case that development drops drastically beyond these temperatures then experiments using more precise temperatures within the range of 28 to 32OC must be completed to determine at exactly which temperature optimum development occurs. Ideally, to determine the temperature curve, as Howe (1967) suggests, at least ten temperatures are required. They should be separated by no more than 2.5OC (which was done here except between 16 & 20•‹C and 20 & 25OC) and temperatures no more than 0.5 to 1.O•‹C should be added to either side of the thresholds and optimum temperature for up to 3.0•‹C (Howe 1967).

Unfortunately, temperature experiments were not conducted at temperatures that were high enough to determine the Tmaxand, therefore, further experiments must be done to determine this temperature. In the Lower Mainland of British Columbia, it is extremely rare for temperatures to reach this high and exposure to these temperatures would occur only under limited circumstances.

In addition, as Reiter and Grassberger (2002) note when determining the thermal units in a forensic case, for the most part, the Tmaxcan be ignored. Whilst fitting the lines to each of the graphs, the 32•‹C data was treated as an outlier with the assumption that it is beyond the optimum development temperature and therefore, not part of the linear portion. For each graph, development is either equivalent or there is a drop in development rate between

30 and 32•‹C; if the 32•‹C data point were included, it would reduce the x-intercept even further.

The practical method used here to determine the lower temperature threshold situates the Tminfor oviposition to lSt,and presumably 2ndand 3rd instar between 9.8 and 1I0C, whereas the minimum threshold for oviposition to post feeding, pupal and adult stages falls between 11 and 13OC.

Overall, the minimum time for development provides a better estimated

Tminthan that of the mode of development for the earlier stages of development,

1'' and 2ndinstar and post feeding stage, whereas the mode of development provides a better estimate for the later stages, 3rd instar, pupal and adult stages.

The x-intercepts are closer to the actual Tminand predominantly have a higher R~ value. In comparing the minimum thresholds determined by the linear method of extrapolation to that of the measurable method, the linear method underestimates the Tmineach time.

Frequently in the literature, the model drawn and used in estimating the

Tminappears as seen in Figure 2-1 and indicates that development occurs at temperatures lower than the linearly determined threshold (Janisch 1932; Logan et a1.1976; Wagner et a1.1984; Worner 1992; Liu et al. 1995; Reiter and

Grassberger 2002). However, the present data have shown that in each circumstance the linear model underestimates that of the actual or practically determined minimum threshold and therefore should more correctly appear as in

Figure 3-10. Consequently, a linear model should not be used alone to estimate the minimum threshold and may be used in conjunction with practical methods or at the very least it should be assumed that this value is underestimating the Tmin of that of P. terraenovae.

Figure 3-10 The actual findings of the development graphs where Tmi, is the minimum temperature threshold, Trnin(,,,, is the minimum temperature threshold determined by extrapolating the linear regression. Optd, is the temperature at which optimum development occurs and T,,, is the maximum temperature threshold. By extending the linear portion of the graph, it always underestimated the Tmin. This is unlike that which is reported in the literature where the line appears to overestimate that of the Tmin. (Figure 2-1)

A comparison to previous research has shown that a forensic entomologist should err on the side of caution when applying developmental data for one region to that of another as geographic differences may occur. In a comparison of these data to Kamal's 1958 data, Kamal's data was found once again to greatly underestimate actual developmental times. This has been reported many times by other workers (Anderson 2000). Therefore, extreme care should be taken with Kamal's data and more recent databases, in which all conditions are well known, should be used where possible.

For forensic purposes, the minimum developmental time to reach each stage is most important whereas the mode of development is important for entomological purposes and is sometimes of interest forensically. The earliest emerged insects provide the most conservative post-mortem interval in the sense that death could have occurred earlier but not after the insects arrived on the body. Therefore, it is assumed that the first insects to develop on the body will provide an absolute minimum post-mortem interval, as they were, in all likelihood, the first insects to find it. Therefore, working backwards to determine just how long they have been there should give the entomologist the closest estimate to time of death or PMI.

The mode of development to reach each stage is more useful in other forms of entomology such as pest management which requires determinations of when the most frequent number of insects of a species will enter a stage of development so that spraying, trapping or other means of management can be accomplished at the most appropriate and useful time.

The experimental temperatures ranged from 9.8 to 32OC. Slight fluctuations from these set temperatures occurred during experimentation. However, it has been suggested by Howe (1967) that these changes would have little effect on the temperature of the food and the insects themselves.

Disappointingly though, the datalogger recordings were lost in a computer malfunction and, therefore, it was impossible to determine the mean constant temperatures and standard errors for each chamber.

Some limitations arose with this research. Firstly, because the 100 eggs were approximated, mortality could not be determined conclusively. The number of eggs placed on each piece of liver was poorly estimated. In the future it is recommended that the eggs be counted out and divided equally under a dissecting microscope using a method that floats and separates them on graph paper (Wells and Greenberg 1992). This is probably the reason for differences in mortality between experimental jars at similar temperatures. Also, the data only allowed ranges to be determined for the Tminand optimum temperature for development and, therefore, future experiments at temperatures within the ranges must be completed to determine the genuine Tminand optimum temperature. Further experiments should be conducted to determine whether later immature stages will develop at 9.8OC even though eggs will not eclose.

This would also help to elucidate whether the eggs were inhibited from eclosing due to temperature, or due to mould growth. Finally, the maximum temperature threshold still needs to be determined and temperatures above 32OC need to be examined. 4 PROTOPHORMIA TERRAENOVAE DEVELOPMENT AT FLUCTUATING TEMPERATURES

4.1 Introduction

The natural environment does not fluctuate in a predictable manner, which makes it difficult to simulate for experimental purposes. To study the influence of fluctuating temperatures on insect development, one of five methods must be relied upon (Howe 1967). A first method involves transferring the insects from one temperature to another after a period of time has lapsed. The second method, similar to the first, is to transfer the insects to different temperatures according to a set pattern and the third, used by Clarkson et al. (2004), is to record the temperatures of the natural environment using a datalogger or thermograph. The fourth method, which by no means emulates the natural environment, is to set up conditions where the insects develop at temperatures based on a gradient as a result of insect mobility. The fifth and final method, used here, is to set an incubator to the desired conditions for development and cycle conditions every 24 hours (Howe 1967).

Clarkson et al. (2004) compared the development rates of P. terraenovae at fluctuating ambient temperatures to a laboratory set constant mean temperature of 20•‹C. Similarly, Davies and Ratcliffe (1994) compared the development of P. terraenovae, among other species, at alternating temperatures of 10 to 20•‹C and 15 to 26OC to that at the respective mean temperatures. They found that development was significantly faster at alternating temperatures. Ames and Turner (2003) studied the effects of a cold episode on development of Cailiphora vomitoria (L) and Calliphora vicina. They looked at the influence this would have on determining a PMI and found that the accumulated degree hours were not the same for cold treated insects as they were for the control insects at 20•‹C and that errors are multiplied when applying constant temperature accumulated degree hours to insects that are at fluctuating temperatures. There have been experiments that look at the effects of fluctuating temperatures on some species of blow fly (Dallwitz 1984; Byrd and

Butler 1996; 1997; 1998) but until now, no study has taken into consideration daily fluctuations to below the Tminfor P. terraenovae.

4.1 .ICore Fluctuating Temperature Experiment

This study was completed to determine if daily fluctuations

(thermoperiods) affect the speed of development of P. terraenovae, but more

specifically to determine if a fundamental drop in overnight temperature below

10•‹C halts development. This research takes into consideration the effects of

evening temperatures that are lower than 1O•‹C and how these influence P.

terraenovae development. For comparison, I reared flies from the same source

at a constant mean temperature. 4.1.2 Objectives

The specific objectives of this research were:

1. To compare development time and rates of P. terraenovae at fluctuating

temperatures from 4 to 28OC and 9 to 23OC with the common mean

temperature of 16OC

2. To compare development rates with 24 and 14OC fluctuations

4.2 Methods and Materials

4.2.1 Experiments at Fluctuating T atures

Insects were collected, reared, maintained and sampled similarly to the constant temperature experiments (Chapter 3). However, unlike the constant temperature experiments, one chamber was kept as a control chamber at the mean temperature of the experimental chambers. The control chamber was set at 16OC and the relative humidity was kept at a constant 75*5% in all chambers used in the experiment. The temperature 16OC was chosen as the control because it falls within the linear section of the development rate vs. temperature curve (Fig.3.7) for P. terraenovae and it is a temperature that is regularly seen in cases during insect season in the Lower Mainland. High humidity is common in early and late insect season in the Lower Mainland of British Columbia, so a relatively high humidity of 75% was used. Lighting was kept the same in all chambers but unlike the constant temperature experiments, a photoperiod of

12:12 (L:D) was followed, hence, the only variables that differed between the constant and fluctuating temperature experiments were temperature and photoperiod.

For the fluctuating temperature experiments, temperature and lighting changes were set to follow a sinusoidal curve and consequently, were similar to that of the natural environment. Incidentally, Davidson (1944) points out that when temperature is plotted with a poikilothermic animal's development rate, it follows this same sinusoidal curve. The photoperiod was set to coincide with that of the thermoperiod, given that they naturally cycle together (Beck 1983) A sinusoidal setting for both the thermoperiod and photoperiod were followed so that the lights shut off as the colder temperatures were approached and turned on as the warmer temperatures were approached.

The experimental temperatures were set at a range of 4 to 28OC and 9 to

23OC to imitate a rare drastic temperature change and one somewhat more likely, respectively. The experimental chamber temperatures dropped drastically during the "evening" section of the cycle (Table 4-1). The temperature settings for each incubator followed the same patterns during the entire experiment (Table 4-1).

At those settings, the 4 to 28OC chamber followed an actual 24 hour pattern of eight hours at 4OC, an hour each at 7, 13, 19 and 25OC, eight hours at 28OC and finally an hour each at 25, 19, 13, and 7OC. The 9 to 23OC chamber followed a continuous pattern of eight hours at g•‹C, an hour each at 11, 14, 18, and 2I0C, eight hours at 23OC and finally an hour each at 21, 18, 14, and 1I0C. Thus, between drastic changes in temperature, an average temperature of 16OC was always maintained. Appendices A through C illustrate the actual temperature recordings made by the dataloggers. The control chamber was set and maintained a constant temperature of 16OC but followed the same relative humidity and photoperiod, 12:12 (L:D),as the experimental chambers.

As suggested by the literature, sampling occurred every 12 hours during each period shift of the photoperiod (Byrd and Butler 1996; 1997; 1998) with the exception of times when a moult was imminent. Occasionally, when a moult was expected, the experiments were observed more frequently in an attempt to capture the change.

Recently laid eggs collected from each cage were divided into three separate jars so that approximately 100 eggs was placed into each jar. One jar was placed into each of the experimental chambers and one into the control chamber. Subsequent generations of four separate P. terraenovae stock colonies were used such that there were four repetitions of the same experiment happening at the same time. The eggs that were oviposited in Cage A became

Experiment A in each of the chambers, the eggs that were collected from Cage B became Experiment B in each chamber, and so forth. Table 4-1 The environmental chamber settings for the experiments set at fluctuating temperatures.

Incubator settings for fluctuating temperature experiments 4 to 28OC 9 to 23OC 12:12(L:D), 75% RH 12:12(L:D), 75% RH Time Temperature I Lighting Temperature I Lighting 0:oo 4OC Off 9OC Off 12OC Off 12.5OC Off 20•‹C On 19.5OC On 28OC On 23OC On 20•‹C On 19.5OC On 12OC Off 12.5OC Off 4OC Off 9OC Off

Development of the P. terraenovae in the jars was observed. Both the

minimum developmental times and mode of development were recorded, in

days, for each stage. Comparisons of this development were made between the

two fluctuating temperature regimes and the common mean temperature of 16OC.

4.3 Results

In all four jars maintained at temperatures fluctuating from 4 to 28OC, the

minimum number of days to hatch was 1.75 and the minimum number of days to

moult to 2"d instar was 3.75 days (Table 4-2). In fact, development was

consistent amongst the jars up to and almost including 3rdinstar. A small

difference was seen in Jar 1 as the post feeding stage was reached and this

difference was maintained at the following stages. Table 4-2 The minimum developmental times to reach each stage at fluctuating temperatures of 4 to 28OC

(days) Staae Jar I Jar 2 Jar 3 Jar 4 1 Mean S.E.

1St instar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

The minimum developmental times were similar to the mode of development times at 4 to 28OC (Table 4-3). However, the only real consistency among the jars for mode of development is to 2"d instar at 3.75 days.

Table 4-3 The mode of development to reach each stage at fluctuating temperatures of 4 to 28•‹C

(days) Stage I Jar 1 Jar 2 Jar 3 Jar 4 Mean S.E.

1St instar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

In every jar raised at 9 to 23OC, the minimum developmental time to reach

1" instar was identical to those raised at 4 to 28OC, that is, 1.75 days (Table 4-4).

Jars 3 and 4 demonstrated similar development through the experiment and Jars

1 and 2 were not that dissimilar. In Jars 1, 3 and 4, only 15, 17 and 15 adult flies emerged, respectively.

Approximately five flies in each jar did not emerge from their puparia. However, in Jar 2, 72 adult flies from approximately 100 eggs emerged and 10 reached only as far as the pupal stage.

Table 4-4 The minimum developmental times to reach each stage at fluctuating temperatures of 9 to 23OC

(days) Stage Jar 1 Jar 2 Jar 3 Jar 4 1 Mean S.E. I I IS'instar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

The mode of development at temperatures fluctuating from 9 to 23OC in

Jars 3 and 4 showed similar development with not a great deal of variation to reach each of the stages (Table 4-5). Jar 1 however, in general, developed slower than the other jars and Jar 2 developed faster at the later stages.

Furthermore, in Jar 2, P. terraenovae had two modes of development at the adult stage and both were earlier than the other jars (30.75 days and 31.25 days).

In Jars 3 and 4 only 46 adult flies emerged along with five adult flies with incomplete emergence and 51 adult flies emerged along with 18 adult flies with incomplete emergence, respectively. In Jar 1, 147 flies emerged and six remained in their pupal case. However in Jar 2, more than 200 flies emerged. Table 4-5 The mode of development to reach each stage at fluctuating temperatures of 9 to 23OC

(days) Stage I Jar 1 Jar 2 Jar 3 Jar 4 Mean S.E.

1'' instar 2ndinstar 3rdinstar Post feeding Pupal stage Adult stage

The temperature in the 16OC chamber was rather unstable at the beginning of the experiments and so the experiment in that one chamber was stopped and restarted 4 days later when the temperature was stabilized.

However, the adult flies in the cages laid their eggs only one hour later in the thermoperiod from the time of laying four days previous. The hour difference was thought to be negligible as the temperature did not change in the 16OC chamber and the insects were still added during the dark fluctuation of the photoperiod as they had been when they were added to the fluctuating temperature chambers.

The minimum developmental time at 16OC in Jar 3 differed radically from the development in the other jars and the insects took much longer to reach each stage (Table 4-6). Unfortunately, the minimum developmental times to reach each of the 3rdinstar, post feeding and pupal stages were missed in this jar and, therefore, the mean development is based only on three jars. The 3rdinstar was missed because the insects were hidden amongst the folds of the liver and paper towel. The post feeding and pupal stage were both missed because finding the insects in amongst the sawdust on the bottom of the jar proved to be difficult. At the time, it was decided that the results from three jars would be sufficient if the

change in insect stage was missed and that it was better to keep the insects in

the temperature chambers rather than search the jar for a lengthy period of time

at room temperature and affect the development time.

Table 4-6 The minimum developmental times to reach each stage at a constant temperature of 16OC

(days) Staae Jar I Jar 2 Jar 3 Jar 4 I Mean S.E.

instar 2.34 2.34 3.25 2.34 instar 5.75 5.75 6.75 5.25 instar 9.75 8.0 missed 9.75 Post feeding 12.75 17.25 missed 15.25 Pupal stage 20.25 20.25 missed 20.25 Adult stage 37.75 37.75 41.75 36.0

Although the mode of development to reach each of 3rdinstar and the

pupal stage were missed, the mode of development at 16OC was observed at the

pupal stage (unlike the minimum development time) and was determined to be

22 k0.31 days(Table 4-7). In Jar 3, only five flies emerged at the end of the

experiment and one fly only partially emerged from its pupal casing. In Jar 1, 75

flies emerged and four flies did not develop beyond pupae. In jar 4, 142 flies

emerged and 15 flies did not develop beyond pupae. Finally, in Jar 2, 167 flies

emerged and 10 flies did not develop beyond the pupal stage. Table 4-7 The mode of development to reach each stage at a constant temperature of 16OC

(days) Stage Jar 1 Jar 2 Jar 3 Jar 4 Mean S.E. I I

1St instar 3.0 3.0 3.75 3.0 2ndinstar 6.75 5.75 6.75 5.75 3rdinstar 11.75 11.75 missed 9.75 Post feeding 18.25 19.25 missed 16.75 Pupal stage 22.75 22.0 22.0 21.25 Adult stage 39.75 39.75 41.75 38.75

Figures 4-1 and 4-2 show the developmental differences observed at the fluctuating temperatures compared to each other and to their common mean temperature and they illustrate that development was fastest at the fluctuating temperatures of 4 to 28OC each time except for the mode of development to the

1'' instar, which was earliest at fluctuations of 9 to 23OC. Stage of Development Figure 4-1 A comparison of the means of the minimum developmental time to reach each stage of development at fluctuating temperatures of 4 to 28OC and 9 to 23OC to each other and their common mean temperature of 16•‹C

Stage of Development Figure 4-2 A comparison of the means of the mode of development to reach each stage at fluctuating temperatures of 4 to 28•‹C and 9 to 23OC to each other and their common mean temperature of 16OC Figure 4-3 illustrates the minimum developmental time (a) and mode of development (b) for P. terraenovae in each stage of development expressed as a percentage of complete immature development for fluctuating temperatures of 4 to 28OC and 9 to 23OC along with their common mean temperature of 16OC.

Interestingly, the percent of time spent in each stage at 16OC is quite similar to the average time spent in each stage at each of the three temperature regimes.

That is, unlike that derived at fluctuating temperatures, the percent of time spent in each stage at 16OC does not deviate greatly from the mean.

Clearly, the longest stage of development is the pupal stage. During complete immature development, the insects spend approximately half their time metamorphosing into an adult blow fly. The shortest stage at all three temperature regimes was the egg stage for both the minimum development time and for the mode of development. For the most part, the percentage of development in each stage is rather similar amongst the temperatures and only shows some variation at the mode of development. In the egg stage, the percent of immature development is much less at temperatures fluctuating from 9 to

23OC. It is only 5.9% as compared to 8.8 and 8% at temperatures of 4 to 28 and

16OC, respectively. At 1'' instar, P. terraenovae spends only approximately 6% of its development at 4 to 28OC as compared to 9.5 (9 to 23OC) and 7.8% (16OC). At the post feeding stage, P. terraenovae spends a greater percentage of time at temperatures fluctuating from 4 to 28OC than it does at the other two temperature regimes. Egg rn 1st rn 2nd rn 3rd rn Post feeding rn Pupal

Minimum Percent of lmmature Development to Each Stage

Egg rn 1st rn 2nd rn 3rd rn Post feeding rn Pupal

Percent Mode of lmmature Development to Each Stage

Figure 4-3 The minimum developmental time (a) and mode of development (b) for P. terraenovae in each stage of development expressed as a percentage of complete immature development for fluctuating temperatures of 4 to 28 and 9 to 23OC and their mean constant temperature of 16•‹C 4.4 Discussion

Early fluctuating temperature experiments began with simple devices like those created by Hagstrum and Hagstrum (1970) or relied on the uncontrolled natural environment itself (Howe 1967) to provide temperature fluctuations but now depend on devices such as the environment chambers used here. The chambers were set to follow the 24 hour fluctuation patterns as set out in Table

4-1, and the actual recordings made by the dataloggers in each chamber are presented in the appendices. The chamber set for fluctuations of 4 to 28•‹C was most accurate with its settings according to the recordings of the datalogger and frequent testing using a mercury thermometer. However, in the chambers set for

16•‹C and 9 to 23•‹C the temperatures varied from their settings and so it became necessary to start the 16•‹Cexperiment over when the temperature readings were on target. Minor fluctuations were occurring at this temperature but this was deemed acceptable as ambient temperatures vary naturally (Liu et al. 1995).

In the chamber set for fluctuations of 9 to 23OC, temperatures were recorded by the datalogger as low as 6•‹C and fell once to as low as 5.5"C. However due to frequent checks with the thermometer this was not a regular occurrence and the experiment was continued.

The insects were added to each chamber at the same time and therefore at the same time in the scheduled fluctuating temperatures. It just so happened that when the eggs were laid it was during the coldest temperature of the thermoperiod, hence they were added at 4, 9, and 16•‹C. Addition to the chambers at the coldest temperature may affect egg development and further research should be done to look at the differences that occur to development when the insects are added to the environmental chambers at a different time in the thermoperiod.

Figures 4-3a and b indicate that the percent of development in each stage is similar at all three temperature settings. The greatest percent of immature development occurs at the pupal stage, at approximately 50•‹/0. Also, the percent of development in each stage at the mean temperature, 16OC, is almost exactly the same as the mean percent of development for all three temperature regimes.

While some deviation from the mean percent occurs at 4 to 28 and 9 to 23OC, it does not occur at 16OC.

In some cases, large differences were seen between jars, in both developmental times and emergence rates. The latter may be explained by the fact that egg numbers were estimated rather than counted. In future studies, eggs should be lightly floated and separated on graph paper so that they may be observed under a dissecting microscope and divided into more equal samples

(Wells and Greenberg 1992). Despite this, differences were clearly seen between experiments. This may have many causes, but may suggest that a greater variation in developmental times occurs between insects raised at greatly fluctuating temperatures compared with those raised at a constant temperature.

It is probable that much larger replicates will be required to fully elucidate this.

Both the results for P. terraenovae found here and those results found by

Davies and Ratcliffe (1994) indicate that P. terraenovae develops faster at fluctuating temperatures than it does at the mean constant temperature. This is unlike the results of Clarkson et al. (2004) who found that development was faster to lSt,2"d and 3rd instar stages at constant temperature and that there was no difference to pupal and adult stages. In fact at all stages, as indicated by

Figures 4-1 and 4-2, development was fastest at extreme fluctuations of 4 to

28OC and slowest at 16OC. One exception was the minimum development time to lStinstar at 4 to 2g•‹C, which was the same under both fluctuating temperature regimes. In addition, the mode of development for P. terraenovae to reach 1'' instar at a fluctuation of 9 to 23OC (2.0 k0.25) was less than the mode of development required at 4 to 28OC (2.25 k0.20). Because P. terraenovae experiences a greater overall amplitude at fluctuations of 4 to 28OC than it does at 9 to 23OC, the insects will develop at a faster rate when the temperature is climbing from a lower temperature to a greater temperature, according to the rate summation effect. The rate summation effect is a phenomenon that is observed with sinusoidal development and fluctuating temperatures. Fluctuations at low temperatures to above and below a given mean temperature cause an increased development rate as compared to the mean temperature because those fluctuations above the mean temperature increase the development rate relatively more than those fluctuations below the mean can lower the rate (Higley and Haskell2001).

In conclusion, development was faster overall at the fluctuating temperatures and, therefore, further use of mean temperatures representing development at fluctuating temperatures must be done cautiously. Development occurred the fastest at the greatest temperature fluctuation (4 to 28OC) and was more comparable to the mean temperature development at a less drastic fluctuation. Perhaps the larger climb in temperature over the same length of time acted to boost development during each 24 hour cycle. As a result of the greater amplitude of temperature exposure that the insects were subjected to in the 4 to

28OC chamber and due to the rate summation effect, development may have come in spurts as temperatures rose from those lower temperatures below the mean temperature. The same would be true for the insects exposed to the 9 to

23OC fluctuations but those insects were only exposed to temperatures over an amplitude of 14•‹Cas compared to a change in temperature of 24•‹C. The insects in the 16OC chamber were strictly exposed to the mean temperature with only extremely minor fluctuations and therefore were not exposed to any effects of sinusoidal fluctuating temperatures. Nevertheless, similar comparisons must be done at other temperatures to see if this explanation holds true. 5 OVERALL DISCUSSION

5.1 Conclusion

Application of the results determined for P. terraenovae caught in the

Lower Mainland of British Columbia to P. terraenovae of other locations must be done with some trepidation as development results have been shown to differ in different areas. Furthermore, although wild caught P. terraenovae were added each insect season to replenish each of the insect colonies, applying the results of these experiments to wild populations should be done with some caution. It is possible that isolated insects will express genetic differences that would not normally be seen in average wild populations (Howe 1967).

Logan et al. (1976) suggest that not only does temperature affect development but so do factors such as humidity, diet quality, disease and food availability. All of these factors were controlled for and maintained the same in all experiments. Relative humidity was maintained at 75 k 5% in all cases and the beef liver was replenished frequently so that there was always a food source.

To prevent mutations and disease, the insect colonies were kept separate and wild caught flies were added each insect season. Hence the effects from these other factors on insect development should have been minimized.

Although some differences were seen, the insect development times determined here were relatively comparable to some recently published data (Marchenko 2001 ; Grassberger and Reiter 2002). However, in comparison to some of the very old published data (Kamal 1958), development rates were much slower. This is not surprising as Marchenko and Grassberger and Reiter used incubators to control for temperature, whereas Kamal did not have any proper temperature control, using only room temperature. When temperatures were too high, an air conditioner was installed part way through the experiments

(Kamal 1958). This suggests that his reported development rates may have come from a variety of different, unreported temperatures, rather than the very specific 26.7OC (80 OF) that is published (Kamal 1958).

The minimum temperature thresholds (Tmin ) using empirical methods of determination fall between 9.8 and 1 I0Cfor development from oviposition to each of 1", and presumably to 2ndand 3rd instar and are between 11 and 13OC for oviposition to the post feeding, pupal and adult stages. Therefore, in the constant temperature experiments, it was determined that by extrapolating the linear portion of the graphs (comparison of rate of development to temperature), an underestimation of the Tmin occurred at each stage.

The optimum temperature threshold is believed to fall near 30•‹C but must be confirmed, as it appears to be within the range of 28 to 32OC. There was a drop off in development from 30 to 32OC and so it can only be assumed at this point that the optimum temperature is close. Further experiments must be completed to reduce each of these Tmin ranges determined using empirical methods and to determine how close the optimum temperature threshold falls to

30•‹C. A maximum temperature threshold (TmaX)was not yet determined as development occurred at 32OC, the highest temperature investigated here, and development at temperatures beyond 32OC must still be completed to establish the Tmax.

Differences in development times between jars at similar temperatures for both the constant and fluctuating temperature experiments probably resulted because of poor representation of insects. An estimated number of eggs was placed into each jar and several jars probably received fewer than the expected

100 eggs. Hence, there would have been fewer than 100 insects to begin the experiment and this would have been compounded by natural mortality.

Furthermore, survival of experimental insects was poor (minimal) at the lower temperatures so sample sizes were reduced to negligible numbers and results were based on these reduced numbers, which may not be an accurate representation of the species. For example, the minimum developmental time at

1I0C to reach 3rdinstar is based on a mere four insects. It is recommended that further repetitions at the lower temperatures be done to verify the findings at these temperatures.

Mould formation on the beef liver proved to be one of the greatest obstacles for development at the lower temperatures and evidently had an affect on P. terraenovae hatch. Not only did it affect the hatch of the insects but it also obscured the insects in the experiments. However, once the insects reached 1'' instar, the mould did not impact the insects any further as they moved to the underside of the liver where the mould was minimal. It was also determined that previously published figures of development rate compared to temperature incorrectly represent the relationship between the linear extrapolation of the regression line and the actual lower development.

Figure 3-10, according to findings, is a more accurate depiction of the relationship as it more correctly places the linear extrapolation at a lower temperature than the actual Tmin.

Since determining that the Tminlsfor all stages fall between 9.8 and 13OC, it was quite simple to decide on a larger fluctuation of 4 to 28OC and a less drastic fluctuation of 9 to 23OC as both fluctuate to temperatures below the Tmin. In addition, both of these fluctuations have a mean temperature of 16OC which is a common temperature in investigations handled by the Forensic Entomology

Laboratory in the School of Criminology at Simon Fraser University.

It was determined that, at fluctuating temperatures, P. terraenovae reached each stage faster than at that of the common mean temperature, 16OC.

These findings agree with those of Davies and Ratcliffe (1994) but are different from those of Clarkson et al. (2004), using the same colonies. Perhaps the method selected for controlling the fluctuation is the reason for these different findings as Clarkson et al. chose to raise P. terraenovae under uncontrolled naturally occurring fluctuations.

Not only did P. terraenovae reach each stage faster at the fluctuating temperatures but it was fastest at the fluctuation with the greater amplitude, 4 to

28OC. It seems likely that with each boost in temperature, a boost to development occurred but with each drop in temperature, development did not drop at the same rate hence the reason for fastest development at fluctuations of

4 to 28OC. From these data, a rate summation effect is observed and therefore caution must be used when using mean temperatures to represent fluctuating temperatures or daily fluctuations. However, these data were collected at only a single temperature regime and further studies are required to prove this point.

Nevertheless, sample sizes were adequate in these experiments and vastly reduced sample sizes were never observed even with the fluctuations to temperatures well below the minimum threshold for this species.

A noteworthy argument needs to be made about the addition of the insects to the environmental chambers for the fluctuating temperature chambers.

Although the insects were added consistently at the same time, even for the constant 16OC experiment which was started on another day when the temperature was correct, at what point in the fluctuation does one add the insects to the chamber? It turned out that at the time of oviposition, the chambers were at the coldest temperatures of each setting and the insects were added at this point. Most likely, a difference would have been observed if the insects had been added at a different point in the thermoperiod and it is recommended that this be examined further.

5.2 Forensic Entomology in Court

It is important that all of these arguments be examined further as it is often the case where a forensic entomologist is called to court to give hislher opinion as an expert witness. As a witness, the forensic entomologist wants to state beyond all certainty the facts related to the evidence. The more the forensic entomologist can offer to the trier of fact, the greater their contribution to the case. Unlike a lay witness, who only provides direct evidence, the forensic entomologist is an expert scientific witness and is being called to draw inferences about the facts related to the insect evidence (Chayko and Gulliver 1999). There are, however, several rulings that must be followed before a forensic entomologist can be called to testify. In every trial, the forensic entomologist must be declared an expert witness to provide the expert entomology opinion evidence. To be declared an expert, the witness must have obtained knowledge of a particular field of which they intend to testify, either through study, experience, skill or training (Greenberg and Kunich 2002). The court preceding that declares a witness a qualified expert is referred to as a voir dire (Chayko and

Gulliver 1999).

To be an expert witness in the field of forensic entomology, board certification with the American Board of Forensic Entomology is becoming a recent and necessary development (Greenberg and Kunich 2002) along with the concomitant educational and research requirements. Publications in the field and attendance at related conferences as well as having been qualified as an expert on previous occasions are all important to becoming qualified as an expert in court (Chayko and Gulliver 1999).

Not only does the expert's testimony have to be declared admissible in court but the evidence itself must also be admitted. In order for entomological evidence to be admissible in court, it must fulfil the requirements expected of all scientific evidence (R v. Mohan 1994; Chayko and Gulliver 1999). The entomological evidence must be relevant and necessary to assist the trier of fact.

There must not be any exclusionary rule in place. In addition, a qualified expert must be approved to interpret and form an opinion of the evidence (R v. Mohan

1994; Chayko and Gulliver 1999; Greenberg and Kunich 2002).

The relevance of the evidence to the case is decided by the judge and by definition it is the determination of whether its existence has probative value or not (Greenberg and Kunich 2002). However, if decided to be logically relevant it still may be excluded as evidence because its prejudicial value overshadows its probative value or it is just simply too confusing to the trier of fact (R. v. Mohan

1994; Chayko and Gulliver 1999; Greenberg and Kunich 2002).

In order for expert evidence to be deemed necessary, it must be considered as knowledge beyond the scope of the trier of fact (whether it be a trial by judge or by judge and jury). The expert may then state hislher opinion regarding the evidence only when the trier of fact does not have the ability to formulate an opinion about the scientific evidence (R v. Mohan 1994; Chayko and Gulliver 1999; Greenberg and Kunich 2002). If the trier of fact has the ability to come to the same conclusions as the expert then the expert witness opinion testimony is not admissible (Chayko and Gulliver 1999).

Rules have been put into place to preside over the admissibility of evidence in courtrooms and are, in effect, a way to maintain structure and compliance in a courtroom. Thus, if there are any exclusionary rules regarding the admittance of expert opinion evidence then it must not be included (Chayko and Gulliver 1999). The importance of the forensic entomologist and the entomological evidence does not begin in court itself, rather both are important in the preparation for trial. Interpretation and evaluation of the findings must be done to assist counsel in preparation for trial (Chayko and Gulliver 1999). Continuity

(chain of custody) of possession of all evidence including entomological evidence must be upheld in order to introduce the evidence in court (Saferstein 2004). It must be shown that the evidence has not been tampered with or contaminated and that proper collecting and testing techniques were applied as well as accepted scientific theory (Greenberg and Kunich 2002).

The opinion reached by the forensic entomologist must logically relate the theory and facts of forensic entomology (Sekula and Eckel Hinton 2006) and therefore the importance of continuous research is multifaceted. Unfortunately, in 86 cases of wrongful conviction, as found in the United States of America by exoneration through DNA evidence, it was found that 27% were associated with false and misleading testimony by the forensic scientist (Saks and Koehler 2005).

Forensic entomology is a growing field with new data being developed daily. At present, most forensic entomologists use constant temperature data sources as these are most readily available. However, the data presented here suggest that fluctuating temperatures, as seen in nature, may speed up development, especially when unusually large fluctuations occur. Therefore, entomologists should exercise caution when extrapolating from constant data sources. This research suggests that, if mean temperature data are incorrectly applied to a fluctuating temperature regime, it is possible that it could indicate that death occurred sooner. Further research in this area is required, as dramatic differences were seen between fluctuating and constant temperatures (Figures 4-

1 and 4-2), however only two regimes were studied. Also, using traditional linear methods may underestimate the Tminand suggest development may have occurred at a lower temperature than possible. APPENDICES

REFERENCE LIST

Ames, C. and B. Turner (2003) Low temperature episodes in development of blowflies: implications for post-mortem interval estimation Medical and Veterinary Entomology, 17(2):178-186 Anderson, G.S. (2000) Minimum and maximum development rates of some forensically important Calliphoridae (Diptera) Journal of Forensic Science, 45, (4):824-832. Anderson, G.S. (2001) Forensic entomology in British Columbia: A brief history. Journal of the Entomological Society of British Columbia, 98: 127-135 Anderson, G.S. (2001 b) lnsect succession on carrion and its relationship to determining time of death. In Byrd, J.H., and J.L. Castner (Eds.) Forensic Entomology: The Utility of Arthropods in Legal Investigations Ch 5 CRC Press Ltd. Boca Raton, FI. Pp 143-176 Anderson, G.S. and S.L. VanLaerhoven (1996) Initial Studies on lnsect Succession on Carrion in Southwestern British Columbia, Journal of Forensic Sciences, 41 (4):613-62 1 Ash, N. and B. Greenberg, (1975) Developmental temperature responses of the sibling species Phaenicia sericata and Phaenicia pallescens. Annals of the Entomological Society of America 68(2): 197-200 Beck, S.D. (1983) lnsect thermoperiodism. Annual Review of Entomology, 28:9l - lo8 Block, W., Erzinqlioglu, Y. Z. and M. R. Worland (1990) Cold resistance in all life stages of two blowfly species (Diptera, Calliphoridae). Medical and Veterinary Entomology 4:2 13-2 19 Bourel, B., Callet, B. Hedouin and D. Gosset (2003) Flies eggs: a new method for the estimation of short-term post-mortem interval? Forensic Science International 135:27-34 Briere, J.-F., Pracos, P., Le Roux, A.-Y. and J.-S. Pierre (1999) A novel rate model of temperature- dependent development for arthropods. Environmental Entomology 28(1):22-29 Buei, K. (1986) The lethal effects of low temperatures on the immature stages of three species of flies, Japanese Journal of Sanitation Zoology 2: 133-1 40 Byrd, J.H. (2001) Laboratory Rearing of Forensic Insects. In Byrd, J.H. and J.L. Castner (Eds.) Forensic Entomology: The Utility of Arthropods in Legal Investigations Ch4 CRC Press Ltd Boca Raton Pp 121-1 42 Byrd, J.H. and J.F. Butler (1996) Effects of Temperature on Cochliomyia macellaria (Diptera: Calliphoridae) Development. Journal of Medical Entomology 33(6):901-905 Byrd, J.L. and J.F. Butler (1997) Effects of Temperature on Chrysomya rufifacies (Diptera: Calliphoridae) Development. Journal of Medical Entomology 34(3):353-358 Byrd, J.H. and J.F. Butler (1998) Effects of Temperature on Sarcophaga haemorrhoidalis (Diptera: Sarcophagidae) Development. Journal of Medical Entomology, 35(5):694-698 Chapman, R.F. (1980) The Insects, Structure and Function, London, Hodder and Stoughton, Pp 1-819 Chayko, G.M. and E.D. Gulliver (1999) In Chayko, G.M. and E.D. Gulliver (Eds) Forensic Evidence in Canada. 2nded. Chapters 1,2 and 3 Canada Law Book Inc. Aurora, Ontario Pp: 1-34 Clarkson, C.A., Hobischak, N.R. and G.S. Anderson (2004) A comparison of the development rate of Protophormia terraenovae (Robineau-Desvoidy) raised under constant and fluctuating temperature regimes. Canadian Society of Forensic Science 37(2):95-101 Dallwitz, R. (1984) The influence of constant and fluctuating temperatures on development rate and survival of pupae of the Australian sheep blowfly Lucilia cuprina Entomology exp. appl. 36:89-95 Davidson, J. (1944) On the relationship between temperature and rate of development of insects at constant temperatures. Journal of Animal Ecology 13:26-38 Davies, L. and G. G. Ratcliffe (1994) Development rates of some pre-adult stages in blowflies with reference to low temperatures. Medical and Veterinary Entomology, 8:245-254 Easton, A.M. and K.G.V. Smith (1970) The entomology of the cadaver. Medicine. Science and Law, 10:208-2 15 Erzinqlioglu, Y.Z. (1983) The application of entomology to forensic medicine. Medicine, Science and Law, 23(1):57-63 Goff, M.L. (1993) Estimation of postmortem interval using arthropod development and successional patterns. Forensic Science Review 5(2):82-93 Goff, M.L. (2000) A fly for the prosecution, How insect evidence helps solve crimes. Harvard University Press, Cambridge, Massachusetts Pp:1-225 Grassberger, M. and C. Reiter (2001) Effects of temperature on Lucilia sericata (Diptera: Calliphoridae) development with special reference to the Isomegalen- and Isomorphen-Diagram. Forensic Science International, Elsevier Science Ltd. 120:32-36 Grassberger M. and C. Reiter (2002) Effect of temperature on development of the forensically important holarctic blow fly Protophormia terraenovae (Robineau-Desvoidy) (Diptera: Calliphoridae). Forensic Science International, Elsevier Ltd 128: 177-1 82 Greenberg, B. (1991) Flies as forensic indicators. Journal of Medical Entomology 281565-577. Greenberg, B. and J.C. Kunich (2002) Entomology and the Law Flies as Forensic Indicators Cambridge University Press, United Kingdom Pp:3-306 Greenberg, B. and T.I. Tantawi (1993) Different developmental strategies in two boreal blow flies (Diptera: Calliphoridae). Journal of Medical Entomology 30:481-484 Hagstrum, D.W. and W.R. Hagstrum (1970) A simple device for producing fluctuating temperatures, with an evaluation of the ecological significance of fluctuating temperatures. Annals of the Entomological Society of America, 63(5): 1385- 1389 Harlow, PM (1956) A study of ovarial development and its relation to adult nutrition in the blowfly Protophormia terraenovae (R. D.) Journal of Experimental Biology 33:777-797 Haskell, N.H., Hall, R.D. Cervenka, V.J. and M.A. Clark (1997) On the body: insects' life stage presence and their post-mortem artifacts. In Haglund, W. D. and M.H. Sorg (Eds.) Forensic taphonomy: The post-mortem fate of human remains Ch 27 CRC Press Ltd. Boca Raton, FI. Pp 415-448 Hedouin, V., Bourel, B., Becart, A., Tournel, G., Deveaux, M. Goff, M.L., and D. Gosset. (2001 ) Determination of drug levels in larvae of Protophormia terraenovae and Calliphora vicina (Diptera: Calliphoridae) reared on rabbit carcasses containing morphine. Journal of Forensic Science, 46 (1 ):I2-1 4 Higley, L.G. and N.H. Haskell (2001) Insect development and forensic entomology. In Byrd, J.H., and J.L. Castner (Eds.) Forensic Entomology: The Utility of Arthropods in Legal Investigations Ch 9 CRC Press Ltd. Boca Raton, FI. Pp 287-302 Howe, R.W. (1967) Temperature effects on embryonic development in insects. Annual Review of Entomology 12.15-42. Janisch, E. (1932) The influence of temperature on the life history of insects. Transactions of the Entomological Society of London 80: 137-168. Johl, H.K. and G.S. Anderson (1996) Effects of refrigeration on development of the blow fly, Calliphora vicina (Diptera: Calliphoridae) and their relationship to time of death. Journal of the Entomological Society of British Columbia 93:93-98 Kamal, A.S. (1958) Comparative study of thirteen species of sarcosaprophagous Calliphoridae and Sarcophagidae (Diptera) I. Bionomics. Annals of the Entomological Society of America 5 1:26 1-270 Kashyap, V.K., and V.V. Pillai (1989) Efficacy of entomological method in estimation of post-mortem interval: a comparative analysis. Forensic Science lnternational 40:245-250 KoStal, V. (2006) Eco-physiological phases of insect diapause. Journal of Insect Physiology 52: 113-1 27 Lane, R.P. (1975) An investigation into blowfly (Diptera: Calliphoridae) succession on corpses. Journal of Natural History 9:581-588 Laudien, H. (1973), Changing reaction systems. In Precht, H., Christophersen, J., Hensel, H. and W. Larcher [eds.] Temperature and Life. Springer, New York Pp:355-399 Levot, G.W., K.R. Brown, and E. Shipp (1979) Larval growth of some calliphorid and sarcophagid diptera. Bulletin Entomological Research 69:469-475 Liu, S.-S., Zhang, G.-M. and J. Zhu (1995) Influence of temperature variations on rate of development in insects: Analysis of case studies from entomological literature. Annals of the Entomological Society of America 88(2): 107-11 9 Logan, J.A., Wollkind, D.J., Hoyt, S.C. and L.K. Tanigoshi (1976) An analytic model for description of temperature dependent rate phenomena in arthropods. Environmental Entomology 5(6): 11 33-1 140 Marchenko, M.I. (2001) Medicolegal relevance of cadaver entomofauna for the determination of the time of death. Forensic Science lnternational 120:89- lo9 Megnin, P. (1894) La Faune des Cadavres. Encyclopedie Scientifique des Aide- Memoire Masson, Gauthier-Villars et Fils, Paris p28 Myskowiak, J.-B. and C. Doums (2002) Effects of refrigeration on the biometry and development of Protophormia terraenovae (Robineau-Desvoidy) (Diptera: Calliphoridae) and its consequences in estimating post-mortem interval in forensic investigations. Forensic Science International, Elsevier Science Ireland Ltd., 125(2-3):254-261 Numata, H. and S. Shiga (1995) Induction of adult diapause by photoperiod and temperature in Protophormia terraenovae (Diptera: Calliphoridae) in Japan. Environmental Entomology 24(6): 1633-1 636 Payne, J.A. (1965) A summer carrion study of the baby pig Sus scrofa Linnaeus. Ecology 46(5):592-602 Pruess, K.P. (1983) Day-Degree methods for pest management. Environmental Entomology 12(3):613-619 Reiter, C. and M. Grassberger (2002) Post-mortem interval estimation using insect development data. Proceedings of the First European Forensic Entomology Seminar, May, 28-30, 2002, Rosny sous Bois, France, Pp 81 - 85 Richard D.S. and D.S. Saunders (1987) Prothoracic gland function in diapause and nondiapause Sarcophaga argyrostoma and Calliphora vicina. Journal of lnsect Physiology 33:385-392 R. v. Mohan (1994) Regina v. Mohan 2 S. C. R. 9, 1994 CanLll 80 (S.C.C.) www.canlii.org/ca/cas/scc/l994/1994scc34. html Saferstein, R. (2004) The crime scene. Criminalistics An Introduction to Forensic Science Ch2 Pearson Prentice Hall, New Jersey Pp:34-61 Saks, M.J. and J.J. Koehler (2005) The coming paradigm shift in forensic identification science. Science 309:892-895 Saunders, D.S. (1982) lnsect Clocks, 2"d ed. Pergamon Press, Oxford Pp: 273 Saunders, D.S. (1997) Undersized larvae from short-day adults of the blow fly Calliphora vicina, side-step the diapause programme. Physiological Entomology 22:249-255 Saunders, D.S., Macpherson, J.N., and K.D. Cairncross (1986) Maternal and larval effects of photoperiod on the induction of larval diapause in two species of fly, Calliphora vicina and Lucilia sericata. Experimental Biology 46151 -58 Schoenly, K. and W. Reid (1987) Dynamics of heterotrophic succession in carrion arthropod assemblages: discrete seres or a continuum of change? Oecologia 73: 192-202 Sekula, R.F. and S. Eckel Hinton (2006) Experts and the admissibility of evidence concerning scientific, technical and other specialized areas of knowledge. In Wecht C.H. and J.T. Rago (Eds.) Forensic Science and Law: Investigative Applications in Criminal, Civil and Family Justice Ch 13 Boca Raton CRC Press Pp: 285-294 Sherman, R. A. and E. A. Pechter (1988) Maggot therapy: a review of the therapeutic applications of fly larvae in human medicine, especially for treating osteomyelitis. Med Vet Entomol. Oxford :Blackwell Scientific Publications : 2(3): 225-230. Smith, K.G.V. (1986) A Manual of Forensic Entomology, British Museum (Natural History), London Pp:203 Stoffolano Jr., J. G., Duan, H. and C-M. Yin (1995) Crop and midgut filling and emptying in female Phormia regina (Diptera: Calliphoridae) fed a liver diet. Journal of Medical Entomology 32(2): 190-1 94 Tachibana, S.-I., and H. Numata (2001) An artificial diet for blow fly larvae, Lucilia sericata (Meigen) (Diptera:Calliphoridae). Applied Entomological ZOO~O~Y.36(4): 521-523. VanLaerhoven S. L. and G. S. Anderson (1999) lnsect succession on buried carrion in two biogeoclimatic zones of British Columbia. Journal of Forensic Science, 44(1):32-43 Vaznunes, M. and D.S. Saunders (1989) The effect of larval temperature and photoperiod on the incidence of larval diapause in the blowfly, Calliphora vicina. Physiological Entomology 14:471-474 Vinogradova E.B. (1987) Characteristics and regulation of imaginal diapause in the blackbottle fly Protophormia terraenovae R.-D. (Diptera, Calliphoridae). New York, Ny : Consultants Bureau Soviet Journal of ECOIO~Y17(3): 163-167 Vinogradova E.B. and T.P. Bogdanova (1985) Relationship of larval and imaginal diapause during ontogenesis of bluebottle fly Calliphora. Dokl. Academy of Science USSR 278555-557 Vinogradova, E.B. and K.B. Zinovjeva (1972) Maternal induction of larval diapause in the blowfly, Calliphora vicina. Journal of lnsect Physiology 1812401 -2409 Voss, M. (2000) Effects of methylparaben on the development and chemosensitivity of the blow fly Protophormia terraenovae Rob.-Desv. (Diptera: Calliphoridae). Canadian Journal of Zoology. 78:680-683 Wagner, T.L., Wu, H.-I., Sharpe, P.J.H., Schoolfield, R.M. and R.N. Coulson (1984) Modelling insect development rates: A literature review and application of a biophysical model. Annals of the Entomological Society of America 77:208-225 Wells, J.D. and 6. Greenberg (1992) Laboratory Interaction between introduced Chrysomya rufifacies and native Cochliomyia macellaria (Diptera: Callip horidae). Environmental Entomology 2 1(3): 640-645 Williams, H. (1984) Model for the aging of fly larvae in forensic entomology. Forensic Science International 25: 19 1-1 99 Wood, F.E. and J. H. Nordin (1976) Studies on the low temperature induced biogenesis of glycerol by adult Protophormia terraenovae. Journal of lnsect Physiology, Pergamon Press 22(12):1665-1673 Worner, S.P. (1992) Performance of phenological models under variable temperature regimes: Consequences of the Kaufmann or rate summation effect. Environmental Entomology 21 (4):689-699