sign in sign up
<<
Home , Abacion, Abacionidae, Alloporus, Blaniulidae, Brachyiulus, Brachyiulus pusillus

Some Aspects of the of (Diplopoda)

Thesis

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

By

Monica A. Farfan, B.S.

Graduate Program in Evolution, Ecology, and Organismal Biology

The Ohio State University

2010

Thesis Committee:

Hans Klompen, Advisor

John W. Wenzel

Andrew Michel Copyright by

Monica A. Farfan

2010 Abstract

The focus of this thesis is the ecology of invasive millipedes (Diplopoda) in the . This particular group of millipedes are thought to be introduced into

North America from and are now widely found in many urban, anthropogenic in the U.S. Why are these such effective colonizers and why do they seem to be mostly present in anthropogenic habitats? In a review of the literature addressing the role of millipedes in nutrient cycling, the interactions of millipedes and communities of fungi and are discussed. The presence of millipedes stimulates fungal growth while fungal hyphae and bacteria positively effect feeding intensity and nutrient assimilation efficiency in millipedes. Millipedes may also utilize enzymes from these organisms.

In a continuation of the study of the ecology of the family Julidae, a comparative study was completed on associated with millipedes in the family Julidae in eastern

North America and the United Kingdom. The goals of this study were:

1. To establish what mites are present on these millipedes in North America

2. To see if this fauna is the same as in Europe

3. To examine host association patterns looking specifically for host or

specificity.

In 2008-2009 millipedes and mites were collected in the eastern U.S. and in the

ii U.K. Millipedes were identified to and mites were identified to morphospecies.

Prevalence of mites for most species is low to medium (< 1/3) and average intensity is low, < 5 mites. Little evidence was found for host specificity in family Julidae, although this may be a possibility in some other taxa. There was some evidence for locality specificity on the part of the mites, and specificity for area or habitat cannot be rejected.

Overall this study shows that millipedes seem to have the same mites as other ground dwelling .

iii Dedication

Dedicated to my husband, Lee T. Ayres, and my grandmother, Edith Catherine Myers Saulsbury.

iv Acknowledgements

Dr. Hans Klompen for advice, moral and financial support and review of a lot of drafts of this thesis.

Dr. Andrew Michel for assistance in data collection and analysis and review of this thesis.

Dr. John W. Wenzel for sound political advice and review of this thesis.

Dr. Petra Sierwald and the Department of at the Field Museum of Natural History, Chicago, IL for partial funding of this research.

Dr. Henrik Enghoff for collection of samples.

Dr. Helen J. Read for collection of samples.

Lee T. Ayres for field support, data analysis, review of later drafts of this thesis and personal support to allow for the completion of this research.

v Vita

June 1991 …...... Holy Name Catholic School

1999 …...... B.F.A with Distinction in Art,

The Ohio State University

2002 …...... M.F.A. Studio Art, The School of

the Art Institute of Chicago

2004-2005 …...... Faculty, The School of the Art

Institute of Chicago, Chicago, IL

2002-2004 …...... Faculty, Robert Morris College,

Chicago, IL

2008 …...... B.S. with Research Distinction

in Entomology, The Ohio State

University

2009-2010 …...... Graduate Teaching Assistant, The

Ohio State University

Publications

Schmidt-Rhaesa, A., M. A. Farfan, and E. C. Bernard. First Record of Millipedes as Hosts for Horsehair (Nematomorpha) in North America. Northeastern Naturalist. 2009: 16(1):125-130.

Shear, William A., Julian J. Lewis, and Monica Farfan. Diplopoda, ,

vi , Pseudotremia salisae Lewis: Distribution extension north of the Ohio River in Ohio and Illinois, U. S. A. Check List 2007: 3(1).

Fields of Study

Major Field: Evolution, Ecology, and Organismal Biology

vii Table of Contents

Abstract …...... ii

Dedication …...... iv

Acknowledgements …...... v

Vita …...... vi

List of Tables …...... xi

List of Figures …...... xii

Chapter 1: Millipedes As Ecological Engineers in the Process of Nutrient Cycling …..... 1

1.1. INTRODUCTION …...... 1

1.2. THE PRIMARY STEPS IN LITTER BREAKDOWN …...... 3

1.2.1. Mineralization …...... 4

1.3. THE ROLE OF MILLIPEDES IN LITTER BREAKDOWN …...... 6

1.3.1. Mechanical processing …...... 6

1.3.2 Factors affecting ingestion and fecal production rates …...... 7

1.3.3 Chemical and assimilation by millipedes …...... 13

1.3.4. Millipedes as nutrient stores …...... 17

1.4. THE EFFECT OF ACTIVITY ON FUNGI AND BACTERIA

…...... 20

1.4.1. Chemical and Structural Characteristics of Fecal Pellets …...... 20

viii 1.4.2. Effects of millipede activity on microbial and fungal populations ..… 25

1.5. EFFECTS OF FUNGI AND BACTERIA ON MILLIPEDES …...... … 28

1.5.1 Consumption of microbes and fungi …...... 29

1.5.2 Coprophagy …...... 32

1.6. APPLICATIONS OF MILLIPEDE ECOLOGY RESEARCH …...... 33

1.7. DISPERSAL AND COLONIZATION BY MILLIPEDES …...... 36

Chapter 2: Atlantic Millipedes and their Associates…...... 39

2.1 INTRODUCTION …...... 39

2.1.1 Goals of Study …...... 49

2.1.2 Hypotheses …...... 49

2.2. METHODS AND MATERIALS …...... 50

2.2.1 Collection Localities …...... 50

2.2.2 Collection Methods …...... 52

2.2.3 Determinations …...... 52

2.2.4 Prevalence and Average Intensity …...... 53

2.3. RESULTS …...... 55

2.3.1 Astigmata …...... 55

2.3.2 Rhizoglyphus …...... 57

2.3.3 Sancassania …...... 57

2.3.4 Schwiebea …...... 57

2.3.5 …...... 67

2.3.6 Mite-Millipede Associations …...... 67

ix 2.3.7 Julid Millipede Collections …...... 67

2.3.8 Non-Julid Millipede Collections …...... 72

2.3.9 Mite Taxa and Relation to Locality …...... 73

2.4. CONCLUSIONS …...... 74

2.4.1 Caveats …...... 79

2.4.2 Additional findings: Non-julid millipede hosts …...... 80

2.5 SUMMARY …...... … 82

References …...... 84

Appendix A: Accession Codes for Millipede Vouchers …...... 96

Appendix B: Accession Codes for Mite Vouchers …...... 99

x List of Tables

Table 1.1. Factors affecting ingestion and rate of fecal production of millipedes.…...... 8

Table 1.2. Factors influencing assimilation efficiency in millipedes.…...... 15

Table 1.3. Studies of nutrient reserves in millipedes.…...... 18

Table 1.4. Differences in chemical and structural components from litter to …..... 21

Table 2.1. Known records of associations between Mesostigmatid mites and millipedes in literature and at the Ohio State Acarology Laboratory…...... 40

Table 2.2. Known records of associations between Astigmatid mites and millipedes in literature and at the Ohio State Acarology Laboratory. …...... 46

Table 2.3. Latitude and Longitude of localities where millipedes were collected in 2008-

2009...... 51

Table 2.4. Characteristics used to distinguish morphospecies of mites in the

Schwiebea collected in 2008-2009.…...... 61

Table 2.5. Prevalence and average intensity of mites collected from millipedes in 2008-

2009...... 68

Table 2.6. Prevalence of identified mite associates on millipedes collected between

2008-2009 by locality...... 75

Table 2.7. Mites present on millipede species collected in both the U.S. and the

U.K...... 77

xi List of Figures

Figure 1. Venter of Rhizoglyphus echinopus. From OConnor, 2009, in A Manual of

Acarology, Krantz & Walter eds., 2009.…...... 54

Figure 2. Venter of Rhizoglyphus A …...... 56

Figure 3. Venter of Rhizoglyphus B …...... 58

Figure 4. Venter of Sancassania B...... 59

Figure 5. Venter of Schwiebea A …...... 63

Figure 6. Venter of Schwiebea C …...... 64

Figure 7. Venter of Schwiebea D …...... 64

Figure 8. Venter of Schwiebea E …...... 65

Figure 9. Venter of Schwiebea F…...... 65

Figure 10. Venter of Schwiebea G …...... 66

Figure 11. Venter of Schwiebea H…...... 66

xii Chapter 1: Millipedes As Ecological Engineers in the Process of Nutrient Cycling

1.1. INTRODUCTION

Millipedes ( Diplopoda) are part of a large guild of arthropods which are the first actors in the mechanical processing of detritus, a low nutrient food source

(Hopkin and Read, 1992). They exist in different types of habitats throughout the world and in a large diversity of forms. The wide variety of sizes of these animals allows them to utilize different habitats, specializing as mostly soil-bound as juveniles while less soil-bound (even arboreal) as adults. Generally, millipedes are desiccation intolerant due to the lack of water-proofing in the and need to, at least periodically, recoup any moisture lost to maintain proper ion balance. Each species must have a set of behavioral and physiological tools for maintaining moisture which are specific to their climate and microhabitat. For these reasons, millipede ecology is an area where a number of questions have yet to be investigated.

Millipedes are associated with many other organisms that also inhabit soil surface and subterranean environments. These include, but are not limited to, fungi, , nematomorphs, bacteria, , insects and mites. In the soil system there is close contact between many of the inhabitants, both directly and indirectly through their activities. It is possible that many of these relationships provide support for millipedes to maintain life in disturbed, arid, artificial, or otherwise foreign and/or inhospitable

1 habitats.

I am interested in the ecology of millipedes in the family Julidae. Many species are thought to be introduced from Europe into a wide variety of habitats and are now widespread in many parts of the world. They are common in the U.S. in residential gardens, parks, and other disturbed habitats where small quantities of detritus are deposited. These millipedes are quite common, but we do not completely understand the mechanisms they utilize that assist them in colonizing new habitats. In to better understand why these animals have proliferated in these anthropogenic habitats, it seems appropriate to asses their role in the environments they inhabit and their interactions with other organisms, especially fungi and microbes. This review summarizes what is known regarding the ecological role of millipedes and clarifies the need for further study in certain areas. This review is organized in the following topics:

a) The primary steps in litter breakdown (section 1.2).

b) The role of millipedes in litter breakdown (section 1.3)

c) The effect of millipede activity on fungi and bacteria (section 1.4)

d) Effect of fungi and bacteria on millipedes (section 1.5)

e) Application of millipede ecology research (section 1.6)

f) Dispersal and colonization by millipedes (section 1.7)

Millipedes (and similar organisms) have been given the title “ecological engineers” (Coleman et al, 2004) as a description of their function in ecosystems irrespective of the ecosystem climate, vegetation, or soil type. Using the engineering concept as an organizing principle, the materials and other workers on the ecological

2 “site” shall be discussed in the context of the concept of nutrient cycling.

1.2. THE PRIMARY STEPS IN LITTER BREAKDOWN

In natural ecosystems, organic material deposited by plants is often unavailable because many of the nutrients are tied up in the physical structure of the plant. Physical breakdown of the organic material (leaf litter, dead wood) must be accomplished for compounds to be released, transformed through mineralization by microorganisms, and recycled to the vegetation. Mechanical breakdown can happen by a few different routes.

The slowest is through seasonal weathering of the leaves over time. This degrades the waxy outer surfaces of leaves and bark eventually breaks down the structural components of the leaves or wood (Chapin et al, 2002). This takes much time and, even after many months, may not happen in any appreciable amount. Over 150 weeks, Bonkowski et al

(1998) saw a decrease in litter weight of only 19.4% when detritivores are excluded. A more efficient approach is mechanical breakdown by detritivorous soil , such as , insects, mites, and millipedes. Indeed, the main role of larger organisms like millipedes is fragmentation of detritus to allow more surface exposure for fungi and bacteria to act upon and to stimulate growth of these mineralizers (Hanlon, 1981b;

Nicholson et al, 1966; Webb, 1977). Their activities penetrate the protective layers of plant material and allow fungi and bacteria to propagate. Millipedes encompass one of the largest groups of , both physically and in diversity. In many habitats they are the main fragmenters of detritus on the habitat floor.

3 1.2.1. Mineralization

The actual decomposition of complex molecules from fragmented leaf litter and wood is accomplished almost exclusively by microorganisms that reside in the soil system, specifically fungi and bacteria. These organisms are the only ones possessing the enzymes capable of breaking down the complex compounds produced by plants.

Fungal hyphae make use of ligninases that can degrade cell walls to access intracellular compounds (Chapin et al, 2002). The hyphal network created by a is able to extract carbon and calcium from plant structures with cellulases and other enzymes from different locations and move them throughout the fungal body. Fungi are a major sink for calcium in forest ecosystems and are an important source of calcium and sodium for soil-bound invertebrates (Cromack and Todd et al, 1977). Mycorrhizal fungi function differently in that they exchange nitrogen, amino acids and water with the plants in return for carbohydrates from the plants. Within the fungal taxa, there are enzymes for the degradation of almost all plant compounds.

Bacteria also produce many enzymes for the decomposition of plant compounds such as cellulases, hemicellulases, and other cellulolytic enzymes (Szabo et al, 1992).

The types of bacteria present in are diverse and it is thought a majority of the species present have not been discovered or described. This diversity of bacteria allows for a large range of enzymes to be produced. In soil systems it is not uncommon for bacteria of different species to benefit from the formation of bacterial consortia in order to more efficiently decompose and utilize plant compounds (Chapin et al, 2002). Chapin et al (2002) describes these consortia as “assembly lines” in which each species must

4 produce its set of enzymes for plant compounds to be decomposed to a point where any of the bacteria present can make use of the products. In times of low resources, fungal hyphae have an advantage over the bacteria because hyphae have the ability to spread widely and utilize nitrogen and carbon compounds from locations that are far from each other. Bacteria are restricted to the resources that directly surround them.

Due to the diversity of, and metabolic restrictions on, different types of fungi and bacteria, the cooperation of the various microbes in a forest ecosystem is key to the mineralization of the compounds created by plants. Through the actions of enzymes of macrofauna (soil invertebrates > 1 cm), mesofauna (invertebrates < 1 cm), bacteria, and fungi these elements are returned to the soil or are otherwise exchanged with the vegetation. Additionally, microorganisms are producers of many water-soluble growth factors which can be transferred between each other and to higher trophic levels

(Webster, 1970).

Nutrient cycling thus involves the combined activity of the entire community of soil organisms. Of course none of these organisms is working strictly for the good of the other organisms in its actions. The activity of an organism is for the survival of the organism and not intended to provide a service. The co-evolution of forest organisms developed over time for guilds to take advantage of available resources. As detritivores, millipedes can be significant players in one element of the cycle, communition of plant material.

5 1.3. THE ROLE OF MILLIPEDES IN LITTER BREAKDOWN

1.3.1. Mechanical processing

Fragmentation of litter and inoculation of the litter with microorganisms, rather than actual mineralization, are considered the most important roles of millipedes in the forest ecosystem (Ghilarov, 1962, Gist and Crossley, 1974). Consumption decreases the overall volume of detritus present due to compaction by the millipede. In experiments involving the presence of different taxa of invertebrates, Cárcamo et al (2001) found that litter mass is reduced fastest in Canadian coastal forests when millipedes are present.

When soil arthropods are eliminated from the soil/litter horizon litter mass loss is often impacted negatively (Whitkamp and Crossley, 1966). Efficient deconstruction of the organic matter on the forest floor is clearly promoted by the larger invertebrates present.

Depending on the size of the millipede, the change in surface area to volume ratio from litter to feces is raised or lowered. Generally speaking, smaller millipedes have a higher surface-area-to-volume ratio. Litter is broken-down into small pieces, the size of which are determined by the teeth on the pectinate lamella (Köhler and Alberti, 1990).

Generally, the higher the density of teeth, the smaller the food particles. Feces produced by millipedes from the ingestion of litter, decaying wood, fungi, fruit, carcasses, and/or other organic materials are usually deposited as pellets of various size depending on the size of the millipede. The size of feces of most adult millipedes generally falls into the category of “macrofeces” (McBrayer, 1973). This is a term used by McBrayer to refer to feces from an greater than 1 cm in length. With the fragmentation and compaction of the comminuted litter by passage through the gut, the suitability of the litter for other

6 trophic groups is significantly altered. Earthworms are known to rely on millipede feces as a food source in some habitats. In one such habitat, removal of the millipedes resulted in the loss of 37% of the earthworms' initial biomass (Bonkowski et al, 1998).

The feeding activity of the millipedes is also important in the vertical redistribution of organic matter in an ecosystem, hence the comparison to an engineer. As the animal feeds and travels vertically through the soil system, the aeration of the soil influences the flow of nutrients and propagation of microbes (Anderson, 1988).

Additionally, the fine particulate of the now comminuted litter has greater ability to move into the lower soil horizon by gravity. The combination of these modifications to the soil surface and litter allows for increased transfer of nutrients into the lower layers of the soil horizon through the flow of leachates.

1.3.2 Factors affecting ingestion and fecal production rates

The reengineering of soil though ingestion and fecal production of millipedes varies based on temperature, moisture, and chemical and physical quality of food available. Additionally, the size of the millipedes may have an influence on the rate of ingestion. An overview of previous research on the influences of ingestion and fecal production is provided in Table 1.1.

As with all organisms, optimal temperature for respiration and reproduction is related to the local climate. Temperate species have a lower optimum temperature for ingestion than species (Wooten and Crawford, 1975) (Table 1.1). Unsurprisingly, activity levels of temperate species of millipedes increase as temperature increases in

7 Table 1.1. Factors affecting ingestion and rate of fecal production of millipedes.

Effect Finding Source Temperature temperate -optimal at 21°C Striganova, 1972b -optimal between 19.5-21.5 °C Gere 1956

desert -optimal at 30 °C Wooten and Crawford, 1975

Moisture -optimal at 60% moisture by weight Elliot, 1970 -positive relationship to ingestion Madge, 1969

Food “quality” decomposition -positive relationship to ingestion Gere 1956; Kheirallah, 1979 -older litter favored over fresh litter Striganova, 1970b -25/75 litter/humus preferred to mull Bocock, 1964

litter species -Fraxinus sp. and Armeniaca sp. Striganova and Prishutova (1990) preferred -hornbeam preferred Striganova, 1969a

toxin content > toxin content = > ingestion Neuhauser and Hartenstein,1978

nutrient content -ingestion positively related to Ca+ Lyford, 1943

mineral proportion -25/75 organic/mineral mix is optimal Dangerfield, 1993

coprophagy -reduces quantity of ingestion McBrayer, 1973; Dangerfield, 1994

body mass -positive relationship Reichle, 1968; Lawrence and Samways, 2003; Dangerfield and Telford, 1989 -negative relationship, by % body Elliot, 1970 mass -negative relationship Gere, 1956 -no relationship Dangerfield, 1993

Temporal factors feeding time -positive relationship to pellet Kheirallah, 1979 production

8 spring (Baker, 1974; Bocock and Heath, 1967). Gist and Crossley (1975) found that the spring rise in temperature seemed to be a precursor to the times of highest activity for detritivorous arthropods. Mild winters allow millipedes to take advantage of a longer growing season and this also influences the overall annual ingestion rate by millipedes. In tropical areas, e.g. Zimbabwe, increased millipede activity is also correlated with temperatures in spring, with further emergence and high feeding activity brought about by the rainy season (October-March) (Dangerfield and Telford, 1991). Long term studies have also shown that seasonal precipitation influences litter breakdown by millipedes. In a long-term study, with low precipitation were directly related to lower litter disappearance in relation to other years (Van der Drift, 1962). Even ephemeral moisture changes associated with diurnal and nocturnal fluxes have been seen to affect ingestion and litter disappearance. Litter dampened by condensation is often fed on by millipedes during the dry season in arid regions (Madge, 1969) and it is thought that this is primarily responsible for driving litter consumption in savannahs.

There is evidence that moisture also encourages consumption in the laboratory and in the field. Increased activity of soil fauna (including millipedes) is often observed in the wet season compared to the dry season (Madge, 1969). Moisture is so important that even desert millipedes would not ingest enough litter for the researchers to collect data on ingestion or assimilation without adding some water to the litter mix (Wooten and

Crawford, 1975). When millipedes were placed in experimental plots with plants, the absorption of water by the plants caused a decrease in ingestion by the millipedes leading to a smaller increase in biomass (Bonkowski et al, 1998). In their natural environment,

9 similar patterns have been observed. In the dry steppe subzone of the former USSR, millipedes supplement their litter diet with moist, living grasses as the season progresses and litter and soil surfaces dry. This usage allows millipedes to make better use of the leaf litter which, Striganova and Prishutova (1990) say, is obligatory for them to ingest. It is known that millipedes are prone to desiccation, as is common among soil fauna, so it makes sense that their highest ingestion and fecal deposition happen when moisture is high.

Quality of the food often seems to dictate how much food will be consumed.

Unfortunately “quality” can refer to any number of characteristics of the food. One factor is the level of degradation of available litter. Often there is a positive relationship of decomposition level to ingestion (Gere, 1956; Striganova, 1970b; Bocock, 1964). This effect trumps species effects as experiments have shown that more decomposed litter is almost always preferred no matter the species (Kheirallah, 1979). However, there is some evidence of preference based on the amount of nutrition within a particular species of litter. Striganova and Prishutova (1990) and Striganova (1969a) found millipedes did exhibit preference for litter from some tree species. Certain nutrient requirements of millipedes could be the reason for observed preferences. For example, calcium content of litter could influence quantity ingested since millipedes require a lot of this element for maintenance of their exoskeleton (Lyford, 1943). Notably, it is debated in the literature whether ingestion rate is a good indicator of food “quality”. Assimilation rate of the food may be more important in determining what is “good” food (Striganova, 1969b). Second, these investigations did not control for the presence or effect of microbes. It is

10 conceivable that simple sugars produced by fungal or bacterial activity may act as an attractant to the millipedes.

In many cases, millipedes consume foods which are poor in nutrients. Their consumption of these poor quality foods must be coupled with the consumption of higher quantities. Millipedes utilize sources of roughage to facilitate the quick throughput of foods with low nutritional content. For example, the ingestion of mineral soil assists in quickening digestion (Sibly, 1981). Toxins present in a food source may also affect the amount of food ingested. Millipedes have been known to consume wood types with high amounts of lignin and phenolics and few carbohydrates, perhaps indicating these are low quality resources which require large meals (Neuhauser and Hartenstein, 1978). An alternative hypothesis is that the toxins themselves accelerate ingestion in millipedes just as tannins and other plant-defense chemicals promote quick digestion in some herbivores.

Again, ingestion may not be the best determinate of high food quality

Coprophagy may also affect ingestion rates. There is evidence that millipedes who have access to their own feces consume less than half the amount of litter they would if restricted in their fecal consumption. This may be due to the higher nutrient content of the feces resulting from microbial processing of the feces over time

(McBrayer, 1973). As a side note, the high frequency of coprophagy in laboratory experiments, possibly created by an unnatural situation in which feces are plentiful, and unavoidable, may alter ingestion data in a way that makes these results unreliable as support for what is happening in the field. (Dangerfield, 1994). It seems that the most meaningful ingestion results may come from carefully planned field experiments instead

11 of laboratory feeding experiments due to the uncertainty of the coprophagy as a dietary requirement.

The relationship of body mass to ingestion has been studied with mixed results.

Juvenile millipedes tend to eat more litter per unit body weight, but adults consume the most litter overall (Lawrence and Samways, 2003; Dangerfield and Telford, 1989; Elliot,

1970). Metabolism is usually inversely related to body weight. Species (Dangerfield and

Milner, 1993) and age of the millipedes (Elliot, 1970) may also significantly impact rates of ingestion. It should be noted, however, that there have not been any studies in which species of millipede has affected ingestion and nitrogen mobilization rates in any other way than through differences in size or activity level (Anderson and Huish et al, 1985).

This is not to say it is not a variable, but it has not been shown to be a significant variable yet. Therefore, the results of research involving millipedes nutrient cycling and ingestion can, generally, be used to comment on the guild in general, with size being a significant influence on ingestion.

Temporal factors, including past feeding history, may influence overall ingestion

(Dangerfield, 1995). For millipedes that were starved and then given food after two weeks, there was a higher average number of fecal pellets produced versus those who were given a consistent supply of food while in culture. It has been hypothesized

(Dangerfield and Telford, 1991) that in areas where starvation regularly occurs, such as arid environments, millipedes tend to rely on ephemeral food resources which are of high quality (i.e. fruits, seeds, fungi). Increased digestion during times of available and high quality of food may be able to compensate for the times of starvation, but it means the

12 millipedes must be large and capable of becoming quickly mobile in order to find food.

This is a strong argument for the role of carnivory in millipedes (Hoffman and Payne,

1989). Carnivory is a mostly ephemeral type of feeding in millipede. It has been observed mostly in taxonomic orders that include large and relatively fast millipedes: ,

Chordeumatida, , and . These groups represent the only millipedes present in habitats which are warmer and more arid than most temperate climates.

In summary, millipede ingestion and fecal production has been shown to increase with the start of the local “growing season”. In most cases, a mixture of mineral soil and organic detritus causes the most digestive activity and fecal production. Moisture is important to millipedes because they are desiccation prone, and there is evidence that moist detritus is more attractive as a food source than dry material. Sometimes this moisture is obtained from live vegetation in the form of roots and tender seedings. Other times moisture is a temporal resource and millipedes will use this to their advantage and feed in certain parts of the day. Overall, high nutrient content of food seems to have the effect of causing lower ingestion. The practice of coprophagy also seems to cause a similar phenomenon. What this reveals is the general need of a soil-bound for moisture and a basic quantity of nutrition. For the average millipede, this comes by ingestion of a high quantity of low quality food.

1.3.3. Chemical decomposition and assimilation by millipedes

Along with the comminution of litter, biochemical decomposition may be caused

13 indirectly and, possibly, directly by millipedes. There is some debate over whether the digestion of cellulose is accomplished by the enzymatic activity of millipedes themselves or by microbes in the millipede digestive system. Cellulose is digested in millipedes and cellulases are present in the hindgut of the millipede foetidissimus (Mur.)

(Striganova, 1970a, 1970b, 1972a). Striganova suggests millipedes actively assimilate

30-40% of the litter consumed although the possibility of microbial interaction was not studied. Cellulolytic enzymes of many types were also found in sp.,

Glomeris sp., and sp. millipedes (Nielsen, 1962). However, cellulase, pectinase, and xylanase were not found in the former two genera . Since these are more complex compounds, this suggests millipedes are not involved in preliminary degradation of plant compounds by themselves. The question of whether or not the enzymes are of millipede or microbe origin was not investigated although the possibility is mentioned

(Nielsen, 1962). Nielsen suggests that the origin of the enzymes is not of consequence from an ecological standpoint as long as the relationships are constant and consistent.

Millipedes that live in arid habitats may have cellulases or other enzymes for polysaccharides of their own. ornatus (Girard) has a high assimilation rate possibly due to using their own enzymes (Wooten and Crawford, 1975). Later research with O. ornatus found the enzymes most common in the gut (hemicellulase, ß- glucosidase, and pectinase) are of bacterial origin and play a significant role for millipedes in the assimilation of nutrients from detritus (Taylor, 1981b). Existing research thus gives little support for enzymes of millipede origin as the main cause of degradation.

Assimilation by millipedes, like ingestion, is also influenced by temperature,

14 Table 1.2. Factors influencing assimilation efficiency in millipedes.

Effect Optimal condition Assimilation efficiency Source

0-4.99% Gere, 1956 6-10.5% Bocock, 1963 30-36.6% Brüggl, 1992 30-40% Striganova, 1970b

Nutrient content ↑ carbs, ↑amino acids Rawlins, 2006 ↑ protein Kheirallah, 1978

Temperature 21 °C 78-98% Striganova, 1972b 25 °C 31.40% Wooten and Crawford, 1975 14.2 °C highest assimilation Saito, 1967 20.2 °C highest assimilation Saito, 1967

Litter species 37.60% Striganova, 1969a hornbeam 39.40% Striganova, 1969a Populus sp. highest assimilation Striganova and Prishutova, 1990

Body mass positive relationship Dangerfield and Milner, 1993

Decomposition high decomposition 14.96-20.2% Gere, 1956 fresh litter 3.77-12.37% Gere, 1956

15 moisture and chemical and physical quality of food. Studies regarding factors influencing millipede assimilation rate are listed in Table 1.2. Generally, juvenile millipedes consistently show greater variation in assimilation rate than adults. This is probably dueto periodic differences in rate of growth. As with ingestion, assimilation rate is optimal at higher temperatures for desert millipedes than millipedes from temperate climates

(Wooten and Crawford, 1975). At 24°C, the rate of assimilation was highest (Wooten and

Crawford, 1975). Optimal temperature for assimilation in temperate species is around

21°C. It also seems that assimilation can fluctuate significantly over time. In a study run two times and a few years apart, millipedes had highest calculated assimilation rates when temperatures averaged 14.2°C and 20.2°C (Saito, 1967), respectively. Clearly, temperature cannot be the only factor affecting assimilation rates.

Assimilation can also be a function of the nutritional or chemical quality of the litter. The reduction of certain chemicals in feces can reveal what compounds millipedes are utilizing in litter. For instance, in millipedes feces may show large decreases in sterols, short chain fatty acids, tricylglycerols, amino acids, and carbohydrates (Rawlins et al, 2006). These can be used directly as energy sources or for maintenance of cellular structures. The presence of some simple sugars can even be a chemical attractant.

Contrary to Striganova and Chernobrovkina (1992), Rawlins et al (2006) also found a decrease in amino acids from the litter to the feces along with a decrease in the diversity of amino acids. Assimilation in the millipede gut is likely responsible for this change in distribution. Nutrient content may also be responsible for the difference in assimilation rates between different litter types (Striganova, 1969a). It is logical to conclude that

16 assimilation may be higher when litter high in the most soluble carbohydrates, lipids, and amino acids is consumed, as opposed to litter with high amounts of lignin and other recalcitrant compounds.

Anything that causes litter to be a more nutritious meal is a secondary influence on assimilation. Degradation of litter by microbes and fungi can influence how much is assimilated. Again, the type of litter species may be influential in assimilation.

Kheirallah's research (1978) investigated weather the rapid growth of millipedes is caused by the quantity or quality (feeding on the preferred litter species) of food consumed. The result showed that millipedes fed on the higher protein litter (the millipedes' preferred litter) developed more rapidly (time per stadia was 10 days shorter in stages of normally high growth) (Kheirallah, 1978). Striganova and Prishutova (1990) found some evidence for species preference as well. Other factors could be influencing assimilation, but these have not been thoroughly studied. As with ingestion, body mass may have an effect on assimilation rate. Dangerfield and Milner (1993) found that with increased body mass, assimilation also increased but that this was only significant in one genus of millipedes out of the six studied. Rates of assimilation may vary widely, based on current research. Unlike ingestion, assimilation can be selective for certain compounds. Determining which compounds are favored requires much testing and may differ from one sampling time to another even with the same millipede species.

1.3.4. Millipedes as nutrient stores

The results of the assimilation of degraded compounds by millipedes is that

17 Table 1.3. Studies of nutrient reserves in millipedes.

Nutrient Finding Source

Ca+ 9-11% of total in soil system Shaw, 1968 15-25% of return to the system Coleman et al, 2004 ↑ Ca+, ↑ sclerotization Gist and Crossley, 1975; Reichle et al, 1969 ↑ Ca+ of all soil fauna Carter and Cragg, 1976 45% Ca+ content McBrayer, 1973

K+ ↓ K+, ↑ sclerotization Gist and Crossley, 1975 ↑ K+, ↑ sclerotization Reichle et al, 1969 ↑ when coprophagy practiced McBrayer, 1973

P 19-22% of total in soil system Shaw, 1968 ↑ when coprophagy practiced McBrayer, 1973

18 millipedes act as stores (or sinks) for certain mineral nutrients (Reichle et al 1969).

Studies on nutrient reserves of millipedes are summarized in Table 1.3. Millipedes require large amounts of calcium to be able to maintain their exoskeleton which is highly sclerotized. Many researchers have found a positive relationship of sclerotization to Ca+ reserves in millipedes. Gist and Crossley (1975) looked into the mineral contents of a standing crop of millipedes in the Appalachian Mountains and found they have the second highest amount of calcium (only terrestrial gastropods have more). The high calcium content of millipedes is evident in the decomposition of millipedes carcasses.

This proceeds very slowly (~ 5 years to decay completely) which is thought to be due to high calcium and magnesium contents of the millipede exoskeleton (Seastedt and Tate,

1981). In total, the macro- community was seen to release 12% of the annual calcium input into the litter environment. Notably, this is significantly less than the release through micro-detritivores. Gist and Crossley (1974) found that microarthropods

(mites, collembola) seem to affect mineral movement more because of their more consistent activity throughout the and not just in the summer. The main source for calcium for millipedes may be digestion of fungi (Cromack and Todd et al, 1977), which are known to be a large source of calcium (Taylor, 1982a).

Other ions are also stored by millipedes, but the concentrations tend to vary and the reasons for this are unclear. Like calcium, potassium is a major component of fungal bodies and consumption of fungi influences potassium content of the millipede due to the presence of fungi in the gut (Gist and Crossley, 1975). There is evidence potassium intake is affected by coprophagy (McBrayer, 1973), as a significantly higher amount of

19 potassium was assimilated by millipedes who participated in coprophagy. Other studies found no evidence for the differential retention of other ions compared to other soil arthropods (Carter and Cragg, 1976).

Sometimes ion concentrations can be shown to differ between different trophic levels. This has been seen with sodium ions (Reichle et al, 1969). Generally concentrations decrease up the food chain. Some of this is due to the consumption of fungi by detritivores/fungivores (Cromack and Sollins et al,1977) as these tend to be in lower trophic levels. As with potassium, sodium assimilation was greatly affected by coprophagy (McBrayer, 1973). The relationships of ion concentration in the millipede to the food source is very likely to be influenced by food sources, species of millipede and, perhaps, the stage of the millipedes at the time of collection.

1.4. THE EFFECT OF MILLIPEDE ACTIVITY ON FUNGI AND BACTERIA

1.4.1. Chemical and Structural Characteristics of Fecal Pellets

The excrement produced by millipedes has chemical and physical characteristics that the parent litter does not, which influence nutrient cycling and the propagation of microbes. The following factors have been found to be different in millipede feces versus the original litter (Lodha, 1974): a) more available carbohydrates and amino acids b) higher nitrogen levels c) more moisture due to compaction of the pellets or otherwise, d) a pH of ~6.5 (which is generally an increase from the litter) and e) the change in physical structure of the pellet compared to the parent litter, namely, increase in surface area.

Research of the change from litter to feces is given in Table 1.4.

20 Table 1.4. Differences in chemical and structural components from litter to feces.

Characteristic Effect Source carbohydrates increase Nicholson et al, 1966; Cárcamo et al, 2000; Rawlins et al, 2006; amino acids increase Striganova and Chernobrovkina, 1992 decrease Rawlins et al, 2006 nitrogen increase Anderson and Ineson, 1983; Anderson et al,1983; Anderson and Ineson, 1984; Anderson and Leonard et al,1983; Bano, 1992; Bocock, 196; Cárcamo et al, 2000; Cárcamo et al, 2001; Dangerfield and Telford, 1989; Kaneko, 1999; Mancuzzi, 1970; Nicholson et al, 1966; Smit and van Aarde, 2001; Webb, 1977; moisture increase Nicholson et al, 1966 pH increase (to ~ 6.5) Cárcamo et al, 2000 increase Webster, 1970 physical structure coherence increase Dangerfield and Telford, 1989; Hanlon, 1981a; Romell, 1935 surface area increase Webb, 1977

21 With the decomposition of litter, holocellulose is degraded to soluble carbohydrates used by organisms very soon after production as seen by the high respiration quotient in the feces the first weeks (Nicholson et al, 1966). This increase in carbon utilization was confirmed in other studies (Cárcamo et al, 2000; Rawlins, et al

2006). It is is due to the increased surface area available for microbial activity. Many nutrients increase in quantity and diversity on feces after they are deposited which is also evidence of the increase in microbial activity in feces (Striganova and Chernobrovkina,

1992). As with carbohydrates, the most soluble of these are used quickly by bacteria and fungi so they are usually not present in quantitative assays.

The processing of litter by millipedes decreases the carbon-nitrogen ratio in the fecal pellets. This is mostly caused by the increase in nitrogen compounds with relatively little change taking place in the amount of carbon compounds. Increases in nitrogen have been found by many researchers (Table 1.4). Webb (1977) saw an increased level of nitrogen fixation through the production of NH3 in the pellets. It is important to note that when doing research on carbon-nitrogen ratios, that the age and decomposition of the

“parent” leaf litter could be an important factor in understanding the results obtained.

Older leaf litter will have a higher carbon nitrogen ratio than fresh litter because of reduced microbial communities on the fresh litter (Mancuzzi,1970). The presence of millipedes positively changes concentrations of many nutrients, not just ammonia.

Phosphorus, calcium, magnesium, and carbon concentrations also increase (Smit and van

Aarde, 2001; Kaneko, 1999).

There is a great deal of support from past research for the increase in nutrient

22 mobilization caused by the leaching of compounds (especially ammonia) from millipede feces. It seems likely that the increased levels of ammonia in the feces also causes the increase in pH of fecal matter. This rise in pH and the availability of ammonia make millipedes feces very hospitable to microflora and other organisms (Cárcamo et al, 2000;

Ghilarov, 1962). Cromack and Todd et al (1977) mention another possibility for the increase in pH in feces is the decomposition of organic salts, such as calcium oxalate, which is present in fungi and often digested by millipedes themselves or by bacteria in the millipede gut. This processing makes for a much weaker acid and could account for at least some of the observed increase in pH. The increase in pH is common to a wide range of forest and litter types, from Douglas fir to birch (Cárcamo et al, 2000).

The changes in physical structure of the litter when digested are thought to be the most important contributions by millipedes. These effects are the consequence of fragmentation of litter, but also the compaction of the litter into feces. Persistence of the feces is influenced by its cohesion and this can affect the release of nutrients from the pellets or surface availability for microbes and fungi. It is generally thought that fecal pellets are capable of disintegration due to the smaller particulate, but this is often not the case. In some locations with high density populations of millipedes, persistence could cause a nutrient sink. Sometimes millipedes feces are not deposited as a solid form but as a “semi-fluid drop” (Romell, 1935). As these dry they form large, hard clumps of very high persistence. Living bacteria were found more often in the soil around the excrement clumps than in the clumps themselves. Hanlon (1981a) found that bacterial and fungal respiration was highest on feces when particulate was very small (0.053-0.1mm) and

23 compaction moderate. Overall, fungi preferred less compacted fecal pellets.

Webb (1977) found that the reduction in the surface to volume ratio (surface area being less than the litter lamellar thickness) in the production of macrofeces is detrimental to the populations of decomposing microbes. More microbes were found on litter, in this case. Conversely, if fecal pellets have more surface area than the litter lamellar thickness, more microbes will be present. The physical conglomerate-feces differentiation theory (Webb, 1977) explains how small arthropod feces encourages microbes and decomposition of litter more effectively than without processing by millipedes. Without the processing by millipedes, litter eventually decomposes into smaller particles. Conglomerates of these smaller particles form resulting in a negative relationship of particle size to conglomerate size. This reduces the surface area to volume ratio, reduces fungal and microbial populations, and decreases decomposition. However, when small arthropods digest litter, particle size becomes very small but this is due to mechanical breakdown of the litter by the arthropod (in our case, the small millipede).

Small millipedes compact these particles as they pass through the digestive system. A direct relationship of pellet size to the size of the millipede is observed and the particles are formed into pellets which are smaller than the conglomerates that would form naturally, thus increasing surface area to volume ratio and increasing decomposition. This is how very small soil animals can break the limitation of reduction of free particle size

(Webb, 1977).

Cohesiveness of the pellets is related to how compact they are. High compaction pellets hold more moisture than the litter. The moisture content of millipede feces can be

24 very different from the parent litter and this can change conditions for the growth of microbes. The content of millipede feces was found by Nicholson et al (1966) to be a majority water (70-90% of fecal content) and this level of moisture has been seen to be an influence on the growth of various microbes (Webster, 1970). This can also contribute to the propagation of anaerobic bacteria inside or fungi and aerobic bacteria on the outside of the pellet, but may deter aerobic bacteria growth on the inside.

The mechanical processing of litter is the most efficient way to increase populations of fungus and bacteria which facilitate decomposition. Millipedes are not the only soil invertebrates which do this, but unlike some soil animals (such as orbatid mites and collembola), millipedes exist in a wide diversity of sizes in the same habitat. This is an advantage because of the relatively large variety of size and quantity of feces. This can provide a wider variety of habitats for bacteria and fungi. These organisms specialize on the degradation of a certain fraction of the compounds produced by plants. It seems reasonable to think that the mix of feces from different stages of different millipede species also encourages a wider variety of .

1.4.2. Effects of millipede activity on microbial and fungal populations

Bacteria and fungi are microbes important in the process of nutrient cycling due to their utilization of enzymes to decompose many complex polysaccharides other organisms cannot. For millipedes, the presence of these microbes makes simple sugars and other nutrients, such as calcium and vitamins, readily available on litter and feces surfaces (Webster, 1970). I have mentioned that detritivores change litter in ways which

25 affect bacteria and fungi and their ability to carry out the mineralization of plant compounds in the litter. I will now summarize some of the effects of millipede activity on microbes and fungi.

The environment of the gut of the millipede is stimulating and hospitable for reasons which could include preferable pH, peristalsis, or the availability of secreted, digestible products by the millipede. As mentioned previously, one of the most important effects of millipede respiration and fecal production is the significant increase in the nitrogen content of the substrate, specifically in the amount of ammonium in millipede feces. There is higher diversity in communities of microorganisms in millipedes feces compared to the parent litter due to the increase in nitrogen compounds and the increase in pH in the feces (Cárcamo et al, 2000; Hanlon, 1981b). Bacterial counts show populations are significantly higher in millipedes feces as opposed to the original litter

(Anderson and Bignell, 1980).

Usually bacterial communities are not exclusive to the gut as they are often found in the soil as well. There is strong evidence, however, of well-developed colonies of

Pseudomonas stutzeri in the gut of hexasticha (Brandt) and possibly the guts of some other arthropods (Szabo et al, 1992). The conditions of the millipedes gut (high humidity and low oxygen) are such that it creates a suitable habitat for anaerobic bacteria

(Ghilarov, 1962). However, a limitation in growth caused by the lack of availability of monosaccharides has been observed by some (Anderson and Ineson, 1983). It is known that some bacteria have a difficult time decomposing carbon compounds to this level. So, perhaps, the survival of bacterial communities inside the millipede gut is limited because

26 other organisms which are capable of this are not present. One study contradicts the beneficial effects of millipedes on bacteria. When the soil arthropod community was eliminated, bacterial populations were seen to increase initially (weeks 4-10) (Whitkamp and Crossley,1966). Generally, though, it can be said that bacterial communities benefit from the comminution of litter by millipedes.

Feces suitable for fungal growth has a pH > 6.5. (Webster, 1970) so the stimulation of fungal communities in feces should not come as a surprise. However, in experiments on soil fauna feces, Hanlon (1981b) initially found no significant difference between fungal respiration on litter particulate versus feces. When the feces was inoculated with different organisms after passage through the gut, bacterial respiration was shown to be low and fungal respiration high. However, when inoculation is caused by passage through the gut of the millipede, bacterial standing crops increased by seven times. Fungi are sensitive to the disruption caused by millipedes and the populations are initially reduced in feces (Hanlon and Anderson, 1980). Millipedes can disrupt the movement of nitrogen from litter to mycorrhizae through their feeding, which increased mobilization of nitrogen in the substrate (Anderson and Leonard et al, 1985).

The vertical movement of invertebrates (including millipedes) through the soil system causes the spread of microorganisms and, indirectly, this increases decomposition of the litter. Fungi are often found on and in millipedes. Seven different genera of fungi were cultured from millipedes from a forest in Michigan, including many species of

Penicillium (Pherson and Beattie, 1979). Some nocardioform actinomycete species (such as Oerskovia turbata) inhabit the digestive system of millipedes in various parts of the

27 world which could lead to dispersal through feces deposition. It has been thought that the fungi most likely to be dispersed by millipedes are those that produce many spores and that have spores which readily germinate. However, most of the support for this has been from samples cultured on plates. Plates used for laboratory culturing favor the growth of organisms that are good colonizers (Visser, 1985). However, arthropod-assisted dispersal may be most important for poor colonizers due to their limited ability for self-dispersal. It is unknown how successful millipede dispersal of poor colonizing fungi may be, but it is not difficult to see how millipedes could harbor much smaller fungal spores or bits of fruiting body and disperse them.

Finally, millipedes contribute to the propagation of microorganisms through their death and decay. In laboratory cultures in which millipedes die and their are left on the substrate, fungal and mold growths have been seen upon the exoskeletons

(personal observation of the author). Decaying millipede remains have a larger amount of calcium than living millipedes (Seastedt and Tate, 1981). The cause is thought to be that colonization of the decomposing remains by fungi has raised the calcium levels. This means decaying millipedes are valuable resources for fungal (and possibly bacterial) populations. Some microbes residing on the forest floor do utilize chitinases, so their presence on millipede exuviae would be expected.

1.5. EFFECTS OF FUNGI AND BACTERIA ON MILLIPEDES

In a forest ecosystem, bacteria and fungal hyphae are associated with millipedes in a few ways. As mentioned, these microorganisms are the main mineralizers of plant

28 compounds on the forest floor. Bacteria and fungi are prevalent on the forest floor in soil and on litter and are often ingested by millipedes. There have been some observations of millipedes which have a symbiotic fungus on the outside of the body. These fungi are in the group ; Triainomycetes hollowayanus Rossi & Weir, Diplopodomyces callipodos Rossi & Balazuc, Troglomyces manfrediae Colla, Rickia dendroiuli Rossi, R. pachyiuli Bechet & Bechet, R. suddhartha Balazuc, and R. uncigeri Scheloske (Rossi and

Weir, 1998). These associations seem to be somewhat specific. While Rickia species are known from many different arthropods, the particular species from millipedes have only been found on specific millipede species. Specific requirements for some of the other fungal species mentioned, seem to be for location on the millipede body (usually, but not exclusively, the first few pairs of legs; sometimes on females only) (Rossi and Weir,

1998).

1.5.1 Consumption of microbes and fungi

One hypothesis for the interaction of millipedes and bacteria or fungi is that millipedes are attracted to the products of decomposition. These products are the results of the presence of microbes which can change the chemical and physical characteristics of the soil/litter making it more “appetizing” for millipedes. By utilizing cellulolytic enzymes bacterial and fungal colonies decompose the litter surface and the resulting sugars are what may attract millipedes (Nielsen, 1962). Sugar preference of five common species of millipedes present in England suggest that simple sugars are preferred (Sakwa,

1974). These sugars are also the most nutritious for the animals. However, millipedes

29 cannot degrade plant material to the stage where simple sugars are generated.

Consumption of fungi and bacteria may therefore be an artifact of consumption of litter for specific nutrient.

However, litter itself may not be the food source millipedes are seeking. An alternative hypothesis holds that microorganisms may be the real target food source for arthropods. In this view, arthropods use litter primarily as a vehicle for the consumption of microbes (Engelmann, 1961). Millipedes are known to selectively consume litter with bacteria and fungal colonies already present (Anderson and Bignell, 1980). Small millipedes will be most likely to eat selectively from available fungi due to limitations placed on fungivores by their size (Visser, 1985). Fungal hyphae can marshall large amounts of widespread nutrients. As noted above, calcium is one of these nutrients. Fungi store excess calcium in the form of calcium oxalates which can be degraded by bacteria and some other soil organisms to calcium ions which can be used for respiration and maintenance of exoskeletons, etc. (Cromack and Sollins et al, 1977). In a study by

McBrayer and Reichle (1971), millipedes were initially included in the category of saprophagous arthropods until results of experiments using radio-nucleotide uptake information were completed. Fungus had to have been ingested for there to exist such high concentrations of calcium in millipedes. For this reason, some researchers consider millipedes fungivores as well as detritivores.

Many other possible reasons exist for fungal consumption. One hypothesis suggests that millipedes utilize fungal enzymes to degrade ingested food and will seek certain fungi for this reason. Fungal isolates collected from the gut of the millipede O.

30 ornatus and its soil environment were taste-tested and millipedes were observed to ingest a number of different fungi (Taylor, 1982a). Forty-seven percent of these fungi were isolated from the millipede gut and not from the soil environment. This means that these millipedes are obtaining some fungal species from alternative food resources, including low-growing bark (Wooten and Crawford, 1975). Taylor (1982a) suggests that there could be a preference for fungi on this type of vegetation as they are able to produce enzymes which are not present in millipede glands or tissues (these were investigated by

Nuñez and Crawford, 1976). Such enzymes would allow millipedes to digest bark and lichens. Additionally, these fungi may provide other nutrients (e.g. vitamins) to the millipedes which may not be obtained in any other way due to the limited resources provided by a desert habitat.

Many enzymes are also produced by bacteria found in soil and millipede guts. In a study of bacteria from isopod and millipede guts, over 100 species of bacteria were isolated from millipede guts (Szabo et al,1992). Among these bacteria were degraders of many complex compounds. Degraders of dulcitol, gelatin, and malatose were only found in millipedes and not isopods. The bacteria themselves may also be providing nutrition, as suggested by Englemann (1961). Bacterial and fungal tissues have been found to be preferentially assimilated over plant tissue at a rate of approximately three times more

(Bignell, 1989). The amount of CO2 waste was highest when fungi were consumed, but assimilation efficiency is higher when bacteria is ingested. Visual investigation of the leaf litter extracted from the mesenteron (the midgut) of the millipede showed that bacteria were digested off the litter fragments which agrees with the assimilation efficiency data.

31 Interestingly, the passage through the hindgut tends to repopulate the feces with bacteria.

The resulting community structure is very different from that in the foregut due to the release of nutrients in the midgut cause by digestion (Bignell, 1989). The results from this research provide the most clear evidence yet that ingestion by millipedes results in significant alteration of bacterial and fungal communities.

1.5.2 Coprophagy

The roles of the microorganisms associated with millipedes and the roles of the millipedes themselves is complicated by the practice of coprophagy, the ingestion of feces of ones own or of another. Communities of microbes present on feces can be a source of nutrition for millipedes and/or a source of various enzymes which the millipedes cannot produce themselves. A great deal of enzymes are produced by coprophilous fungi and the fungal succession on feces changes the enzyme availability and nutritional content of the feces through time (Webster, 1970). Fungi are a large source of calcium, nitrogen, and soluble vitamins and provide the rise in pH required for the growth of other microorganisms. The results of Bignell's experiments (1989) in which bacteria were stripped from litter in digestion through the midgut of the millipedes led him to conclude that coprophagy is potentially important to millipedes although coprophagy was not observed during this study. In a famous laboratory study, McBrayer

(1973) found not only was there a caloric advantage for millipedes that had access to their own feces, but that these millipedes were unable to survive with out this access. On day 28 of his experiments the death of two subjects and the 10% weight loss of three

32 other subjects led to the termination of the experiment. Consuming the feces of another organism has also been observed. Vertebrate feces was preferred over mineral soil in millipede communities in Bangalore, India (Bano, 1992). Consumption of millipede feces by other organisms may also be required in some cases. communities deprived of millipede presence and millipede feces lost 37% of their initial weight

(Bonkowski et al, 1998). As mentioned earlier, coprophagy also changes ingestion rates of litter which can lead to an underestimation of litter consumption in the field. Many researchers of soil arthropods agree with the idea of the forest floor as an “external rumen” for soil arthropods and I think it would be very difficult for a millipede to intentionally avoid its own feces given their nature as slow-grazers.

1.6. APPLICATIONS OF MILLIPEDE ECOLOGY RESEARCH

The activities of millipedes and the other organisms in soil/litter systems influence each other in a cyclic fashion. Like a system of cogs, the activity of a guild of organisms perpetuates the activities of others in the forest or desert system. Taylor

(1982b) suggests that the importance of the association cannot be understated in harsh environmental situations where proper conditions for microbes (such as consistent moisture) are often not available (leading to sporadic microbial activity) and amount of detritus deposition is sometimes small. The noted preference of some millipedes for certain species of fungi found in their environments gives credence to the idea that millipedes and fungi have co-evolved to benefit each other in nutritional and distributional regards, respectively (Taylor, 1982a). Learning about these relationships is

33 then key to understanding the nutrient cycling in a given ecosystem. As a factory assembly line, certain processing must occur for the end result (the recycling of minerals and other nutrients) to occur.

Taylor and Crawford (1982) see the importance of nutrient cycling in desert environments as an investment in the understanding of the future of changing lands since current are expanding. Our understanding informs our choices when it comes to management for the sake of maintaining fertility in these regions. The influence of millipedes and microbes on nutritional resources is also important in areas of low resources but high economic value such as with crops in arid areas. In an area in

Zimbabwe discussed by Dangerfield and Telford (1989), the millipede community is the primary group of detritivores. Changes in the communities of millipedes and/or their microbial associates can have a great affect on fertility and crop yield. Studies have led them to recommend the addition of detritus to agricultural fields in order to provide crops with a higher degree of nutrition through the encouragement of millipede activity. This then stimulates the growth of microbial communities. Crossley (1977) mentions the value of looking at fauna and their influence on the nutrients of forest ecosystems with regard to increasing the values of commercial tree stands. Other recommendations have been made regarding row-crop agricultural systems and the role soil fauna could possibly play in yield (Crossley et al, 1989). As with integrated management strategies, another approach towards maintaining a sustainable farming industry, effective management of soil fauna through the use of research from previous studies could be another method for the acceleration of growth of certain crop products.

34 Studies in this field of research are complicated, tedious to plan and often hard to interpret. Crossley (1970) mentions the difficulty in studying soil processes is multi- leveled. of fauna involved in soil systems is still rudimentary in many cases.

The trophic levels in soil systems can be difficult to discern due to the intimate interaction between fauna, flora, and microbes. It is difficult to tease apart the specific functions of millipedes as detritivores in all ecosystems due to the differences in the biology of the millipedes themselves, but also due to the wide variation of the habitats and organisms millipedes interact with. The level of close physical contact between microorganisms and millipedes makes our investigation of behavior in the field physically difficult. Significant differences in millipede sizes and species' reactions to temperature and moisture are indicative of the complexity of soil ecosystems (Crawford,

1992). Applying the conclusions from nutrient cycling studies with millipedes to other soil systems involving detritivores should be done with caution because climate seems to affect millipedes more than other elements of the soil fauna due to their sensitivity to desiccation. The varying levels of sclerotization and mobility (due to size) of millipedes endow them with wide-ranging ability to affect their habitat for other organisms.

Ghilarov (1962) mentions another unfortunate issue with studies of microorganisms and soil invertebrates: most studies are performed by either microbiologists or zoologists and, therefore, an incomplete resolution to the question posed is usually the result. In order to effectively study direct and indirect effects of microorganisms, millipedes and nutrient cycling in forest systems, the use of experimental microcosms in the field (not the laboratory) consisting of animals, substrate

35 and living plant roots has been suggested (Anderson and Huish et al, 1985). The addition of lysimeters would also help in the discernment of the role precipitation plays in the system. The effect of roots on moisture and microorganism communities quite likely influences mineralization. It does seem clear that while many decades of research has been done in this area have led to considerable progress, benefits can still be gained by more experimentation in the field. These holistic approaches are thought to be more realistic and could reveal more regarding the functions of millipedes in nutrient cycling.

Nielsen (1962) said that the exact bacterial and fungal fauna producing cellulolytic enzymes is unimportant from an ecological standpoint, as long as the relationships are constant and consistent. This could be expanded to apply to the investigation of fauna and nutrient cycling as a whole. Perhaps the functions of the individual arthropod taxa are not as important as the constancy and consistency of the function within the guild. For microbiologists and biochemists, generalization of taxa into functional groups may not be a reasonable approach as specific chemical production is integral to proper identification to species (specifically with bacteria). However, for the landscape ecologist interested in conservation or improving nutrient cycling conditions in a nutrient depauperate or economically important habitat, concentration on guild-level effects is not only appropriate but fundamentally important to understanding the needs of the ecosystem.

1.7. DISPERSAL AND COLONIZATION BY MILLIPEDES

Millipedes (especially small millipedes) are often introduced into new habitats by

36 man. I am interested in how this seems to occur without much challenge given their intolerance to desiccation and what seems to be, from the evidence in this review, a reliance on other small organisms for nutrition. From the present body of literature, certain generalizations can be made which I think should be kept in mind as millipede colonization is studied further.

1. Support for fungal or bacterial species food preference or specialization by millipedes needs to be reviewed critically. Bacteria and fungi are so ubiquitous and diverse that even in new habitats, a replacement taxon with similar enzymes and nutrient content will most likely be present in a new habitat. Additionally, if the millipedes are introduced with substrate (which is usually the case), fungal or bacterial inoculation of the new substrate will occur.

2. Small millipedes (< 3 cm), as with small animals in general, are able to make use of microhabitats which researchers tend not to study. These are especially common in urban habitats (i.e. beneath flower pots, under ornamental statuary, etc.).

Their need for little space and their lack of dispersal ability mean they do not require the resources of a larger animal. They can sustain themselves on fewer resources for a short period of time, at least.

3. The practice of coprophagy, vital or not, has been shown to be beneficial for millipedes to participate. In transportation situations in which millipedes are, possibly, contained with little substrate, coprophagy may help sustain the millipede for an extended period of time.

Transportation of small millipedes, like the invasive ones in the family Julidae,

37 happens often and their dispersal can be considered a direct consequence of human activities. Their ability to thrive in anthropogenic, disturbed habitats is affected both positively and negatively by the management by humans. This disrupts bacterial and, especially, fungal communities on which millipedes feed. However, the continual addition of resources to these habitats in the form of mulch and fertilizers may provide the supplemental nutrition needed by soil organisms to counteract the effects of disturbance. The movement of resources into and out of these types of habitats is atypical and may be very important in survivorship of these millipedes.

38 Chapter 2: Atlantic Millipedes and their Mite Associates

2.1. INTRODUCTION

Mites evolved relationships with other arthropods approximately 100 million years ago with the diversification of both plants and arthropods in the late Mesozoic era

(Southwood, 1973). As more niches became available mites developed a wide variety of well-known symbiotic relationships with many arthropods (Lindquist, 1975) including many species in the orders Coleoptera, Diptera, Hymenoptera, and Lepidoptera.

There are also many reported relationships between mites and myriapods (Maes, 1983;

Southcott, 1987; Bloszyk et al, 2006), especially class Diplopoda, the millipedes and this is the focus of this study.

Millipedes are detritivores that generally live on the habitat floor and tend to interact with organisms that also live and feed in this space. They have been known to have close relationships with bacteria (Szabo et al, 1992), fungi (Rossi and Weir, 1998;

Pherson and Beatti, 1979), nematomorphs (Schmidt-Rhaesa et al, 2009), and mites, to name a few. What is known about the relationships between millipedes and mites is mostly from studies of relationships involving relatively large, tropical species of millipedes (generally > 3 cm). These generally involve mites in the suborder

Mesostigmata. The known records of millipede-mesostigmatid mite associations are detailed in Table 2.1. With some exceptions (Evans and Sheals, 1959; Fain, 1988; Fain,

39 Table 2.1. Known records of associations between Mesostigmatid mites and millipedes in literature and at the Ohio State Acarology Laboratory. Continued.

Mite Millipede host Country Source Gamasina (=Aceosejidae) aphidioides sp. Japan Ishikawa, 1986 angustus “Phyodesmus” sublimbatus Indonesia Evans and Sheals, 1959 Lasioseius frontalis Platyrrhachus mirandus Indonesia Evans and Sheals, 1959 Lasioseius polydesmophilus Platyrrhachus mirandus Indonesia Evans and Sheals, 1959 Lasioseius sugawari gracilis Japan Ishikawa, 1986 Trichaspididae Trichaspis China Gu et al, 1991 Iphiopsididae Iphiopsis sp. Orthoporoides sp. South OSAL coll. 40 Iphiopsis mirabilis millipedes Italy Berlese 1882 Iphiolaelaps millipedes Queensland, Womersley, 1956 Jacobsonia africanus Spirostrepta sp. Cameroon Fain, 1994 Jacobsonia andrei Spirostrepta sp. Batouri Fain, 1994 Jacobsonia audyi sp. Malaya Evans, 1955 Jacobsonia berlesei Indo-Malayan millipede Java-Malaysia Casanueva & Johnston, 1992 Jacobsonia puylaerti Pachybolus macrosternus Zaïre Fain,1994 Julolaelaps sp. Pachybolus insignus Ivory Coast OSAL coll. Julolaelaps buensis myriapoda Cameroon Maes, 1983 Julolaelaps cameroonensis myriapoda Cameroon Maes, 1983 Julolaelaps celestiae gigas East Africa Uppstrom and Klompen, 2005 Julolaelaps dispar juliform millipede Somalia Berlese, 1916 Julolaelaps excavatus large “julid” Madiakoko (Zaïre) Fain,1987b Julolaelaps idjwiensis large “julid” Madiakoko (Zaïre) Fain,1987b Julolaelaps kilifiensis spirostreptid millipede Kenya Kontschan, 2005 Julolaelaps lucator juliform millipede Somalia Berlese, 1916 syn. Julolaelaps indicus julid millipede India Vitzthum, 1921 Julolaelaps madiakokoensis large “julid” Madiakoko (Zaïre) Fain,1987b Julolaelaps moseri spirostreptid millipede Trinidad Hunter and Rosario, 1986 Continued. Table 2.1. Continued.

Mite Millipede host Country Source Gamasina Iphiopsididae Julolaelaps myriapodalis spirostreptid millipede West Africa Ryke, 1959 Julolaelaps nishikawai Nedyopus patrioticus Japan Ishikawa, 1986 Julolaelaps pararotundatus spirostreptid millipede West Africa Ryke, 1959 “ ” spirostreptid mllipede Kenya Kontschan, 2005 “ ” Ohio, U.S. OSAL coll. Julolaelaps parvitergalis Parafontaria sp. Japan Ishikawa, 1986 Julolaelaps parvunglatus Parafontaria sp. Japan Ishikawa, 1986 Julolaelaps paucipilis large juliform millipede Zaïre Fain, 1987b Julolaelaps peritremalis spirostreptid millipede West Africa Ryke, 1959 Julolaelaps rotundatus juliform millipede Somalia Berlese, 1916 Julolaelaps serratus myriapoda Cameroon Maes, 1983 4 1 Julolaelaps spirostrepti Archispirostreptus gigas Kenya OSAL coll. Julolaelaps vandaelensis myriapoda Cameroon Maes, 1983 Narceolaelaps americanus americanus North Carolina, U.S. Kethley, 1978 Narceolaelaps annularis Narceus annularis eastern U.S. Kethley, 1978 Narceolaelaps burdicki Tylobolus sp California, U.S. Kethley, 1978 Narceolaelaps gordanus Narceus gordanus Florida, U.S. Kethley, 1978 Scissuralaelaps bipartitus millipede on orchid Philippines Ishikawa, 1988 Scissuralaelaps breviseta sp. Philippines Ishikawa, 1988 Scissuralaelaps grootaeri unidentified “Iule” New Guinea Fain,1992 Scissuralaelaps hirschmanni Polyconoceras sp. New Guinea Fain,1992 Scissuralaelaps irianensis unidentified myriapod New Guinea Fain,1992 Scissuralaelaps joliveti Polyconoceras sp. New Guinea Fain,1992 Cosmolaelaps claviger Lacinogonus sp. Ivory Coast OSAL coll. Cosmolaelaps hortensis Oxidus gracilis Japan Ishikawa, 1986 polydesmoides polydesmid millipede Malaya Evans, 1955 Iphidolaelaps myriapoda millipedes Australia Womersley, 1956 Macrochelidae Macrocheles muscaedomesticae Parafontaria sp. Japan Ishikawa, 1986 Continued. Table 2.1. Continued.

Mite Millipede host Country Source Sejina Afroheterozercon sp. Lacinogonus tilde Ivory Coast OSAL coll. “ Pelmatojulus tigrinus Ivory Coast OSAL coll. “ Peridontopyge togoensis Ivory Coast OSAL coll. Afroheterozercon pachybolus Pachybolus macrosternus Zaïre Fain, 1988 Afroheterozercon Spirostreptus cornutus Zaïre Fain, 1988 Asioheterozercon sp. Phyllogonostreptus nigrolabiatus India OSAL coll. Asioheterozercon audax Spirostreptus Java, Indonesia Berlese 1910 millipede Malaysia Fain 1989 Discomegistus sp. Rhombocephalus giganteus Trinidad Trägårdh, 1911 "Heterozercon" elapsus Thyropygus sp. Sumatra, Indonesia Vitzthum, 1925; 1926 Heterozercon microsuctus spirostreptid Brasil Fain, 1989 4 2 Maracazercon joliveti spirostreptid Brasil Fain, 1989 Narceoheterozercon ohioensis Narceus annularis Ohio, U.S. Gerdeman and Klompen, 2003 Gerdeman et al, 2000

Trigynaspida Costacaridae Costacarus reyesi millipede Mexico Hunter, 1993 Euzerconidae Neoeuzercon diplopodophilus millipede Panama Funk, 1980 Nyssodesmus python Costa Rica OSAL coll. Continued.

Table 2.1. Continued.

Mite Millipede host Country Source Trigynaspida Cryptometasternnum queenslandense pill millipedes Australia Womersley, 1958 Diplogynium acuminatum millipede Brasil Canestrini, 1888 Neodiplogynium schubarti Sooretama aguirrei Brasil Trägårdh, 1950 Neotenogyniidae Neotenogynium malkini Orthoporus sp. Ecuador Kethley, 1973 Paramegistidae Meristomegistus vazquezi Aceratophallus sp. Mexico Kim & Klompen, 2002 Neomegistus julidicola juliform millipede South Africa Trägårdh, 1906; 1907 Neomegistus remus Proporobolus sp. Australia Baker & Seeman, 2008 Paramegistus confrater juliform millipede South Africa Trägårdh, 1906; 1907 4 3

Uropodina Chiropturopoda sp. Narceus gordanus Ohio, Florida, U.S. OSAL coll. Phaulodinychus sp. Euryurus leachii Ohio, S. Carolina, U.S. OSAL coll. “ Cylindroiulus latestriatus Ohio, U.S. OSAL coll. 1994; Gerdeman and Klompen, 2003; Gu et al, 1991; Ishikawa, 1986; Kethley, 1978;

Lawrence, 1939a; Uppstrom and Klompen, 2005), most studies do not report a precise

(genus or even family level) identification of the millipede host. This, of course, can lead to confusion when attempting collections of these mites in the field. Additionally, it becomes more difficult to study the ecological relationships when the species involved and, possibly, phylogeny are unknown.

It appears the known millipede-mite relationships can be categorized into two of the three types of symbioses, and . There are no known records of associations where the presence of the mites benefits the millipede (). Until recently, there has not been much support for parasitism either. Gerdeman et al (2000) observed the tendency of the heterozerconid Narceoheterozercon ohioensis (Gerdeman and Klompen, 2003) to collect around wounds in the exoskeleton of Narceus sp. This could allow for the in-take of fluids from the millipede, but feeding from these sites by N. ohioensis was not observed. Mites in the genus Narceolaelaps were also not observed feeding on their millipede associates, Narceus spp. (Kethley, 1978), despite sharing many morphological characteristics with other members of suborder Gamasina that have known adaptations for parasitism. Additionally, Kethley (1978) notes that both host specificity and site specificity are high further suggesting a parasitic relationship (p. 209). This particular genus creates its own acarinarium, or pouch, on the host that seems to function only as housing for a mite. It does this by placing itself under the millipede's tergite just after the host moults. The tergite tans and then is moulded around the mite. This evidence could be considered indicative of directly parasitic behavior.

44 Mite associations with small millipedes usually do not involve the suborder

Mesostigmata, but mostly refer to mites in the cohort Astigmata. Reported associations of astigmatid mite and millipedes are shown in Table 2.2. The relationships between astigmatid mites and small millipedes are suspected to be commensalistic (OConnor,

2009) although little experimentation has been done to discern the nature of the relationship. It appears the mite benefits and the millipede neither benefits nor incurs damage from the relationship. Parasitism is deemed unlikely because of the morphology of these mites.

The suborder Astigmata is best-known for having a deutonymph, the second of the 3 ancestral nymphal instars, highly modified for dispersal, the “hypopus” stage

(OConnor, 2009). This modified deutonymph is a non-feeding stage. It is characterized by reduced mouthparts, a very flat venter and convex dorsum, strong sclerotization of both dorsum and venter, and a special attachment organ on the posterio-ventral side of the mite. This organ typically features a number of sucker-like structures adapted for adhering to the smooth exoskeleton of an arthropod. In the family , this stage does not imbibe nutrients and fluid orally. The mites tend to be inactive which conserves resources. Generally, the modified astigmatid deutonymph is considered to be one of the best examples of a phoretic adaptation. Houck and OConnor (1991) define phoresy as “a phenomenon in which one organism (the phoretic) receives an ecological or evolutionary advantage by migrating from the natal habitat while superficially attached to a selected interspecific host for some portion of the individual phoretic's lifetime”. So in broad terms, this is a way for the phoretic to disperse and/or remove itself from a suboptimal

45 Table 2.2. Known records of associations between Astigmatid mites and millipedes in literature and at the Ohio State Acarology Laboratory.

Millipede host Country Source Astigmata feroniarum moreleti Australia Baker, 1985 Acaridae Caloglyphus julidicolus flavifilis South Africa Lawrence, 1939a Schwiebea sp. xystodesmid millipedes Appalachian Mtns., U.S. Swafford and Bond, 2009 Schwiebea nova Cylindroiulus sp. Misina-tető Mahunka, 1962

Canestriniidae Diplopodocoptes transkeiensis glomerid millipede South Africa Fain, 1987a odontopygidae millipede Kenya Fain, 1987a 4 6

Chetochelacaridae Chetochelacarus mamillatus julid millipede Zaïre Fain, 1987a

Lophonotacaridae Lophonotacarus minutus glomerid millipede South Africa Fain, 1987a odontopygid millipede Kenya Fain, 1987a

Unidentified mites mites Polydesmus inconstans Michigan, U.S. Snider, 1984 environmental situation which it may not be able to move from on its own. By traveling on an arthropod host, the mite has a better opportunity at finding a limited, possibly ephemeral, resource (Houck and OConnor, 1991). In many cases, the requirement is for higher quality food or moisture. When food quality begins to decline, these mites may induce the deutonymph stage and start seeking a host instead of developing straight into tritonymphs from the protonymph stage (Corente and Knülle, 2003). The option of developing into this stage is what has led these mites to be defined as facultative heteromorphs (Houck and OConnor, 1991). Another purpose for this instar has been suspected. The deutonymph has been shown to be a more hardy instar than the other development stages (OConnor, 1994) and has been shown to tolerate harsh temperature and moisture conditions. It can thus function as a protective stage for the mite.

The presence of this group of mites on small millipedes has been reported, but details are vague. What is present in the literature is generally not specific on the identification of the mite taxa involved and does not give information regarding host specificity. In some cases assumptions regarding the nature of the relationships are not well-substantiated. For example, in an exploratory study for biocontrol purposes by

Baker (1985), many deutonymphal Histiostoma feroniarum (Dufour) were collected from the millipede Ommatiulus moreletii (Lucas). The mite was fully identified (although there are reasons to doubt the identification), and Baker considered the mites “ectoparasitic”.

The latter may have been an incorrect assumption since there are no records of parasitic behavior by Histiostoma. A better case was made by Snider (1984) who explained the presence of many acarid deutonymphs inside the moulting chambers of Polydesmus

47 inconstans (Latzel) by the vulnerability of the millipede during the moulting period: deutonymphs may position themselves to be enclosed in the chamber, moult to a tritonymph and feed on the softened, newly moulted millipede. Astigmatid tritonymphs are feeding instars. That being said, the hypothesis of a parasitic relationship with small millipedes remains unlikely but the lack of research prevents its rejection.

The group of interest in this study are millipedes in the family Julidae. These millipedes are small (~3 cm), tend to be very common, and are found in a variety of anthropogenic habitats all over the world. They are assumed to be introduced to the New

World from Europe by the deposition of ballast as part of colonial-era shipping (Lindroth,

1957; Shear, 1999; Blower, 1985). Soil ballast was used at the time (1600s-1800s) and many soil invertebrates made their appearance in the New World and other former colonies this way, including some species of European earthworms. Currently, julid millipedes are found in a wide variety of man-made habitats such as residential gardens and parks. Dispersal of these millipedes within North America was, and probably continues to be, mainly through landscape management and transport of plants by humans. This history makes them an ideal test group for examining effects of recent long range dispersal and possible founder effects.

Between March 2008 and October 2009, small millipedes in the family Julidae in the Atlantic region (eastern U.S. and western Europe) were collected for a population genetics study. The number of these millipedes present in the eastern United States and their seemingly synanthropic nature have led me to inquire about their mite associates.

What are the types of mites associated with these millipedes? Are the associates the same

48 in the U.S. as in Europe? It has been speculated that the mites found on these millipedes will not exhibit host specificity (OConnor, personal communication) as the same genera can been seen on different ground dwelling arthropods. Olynyk and Freitag (1979) demonstrated such a pattern in a study of mite associates of carabid beetles. Schwiebea sp., Sancassania sp., sp., Anoetus sp., and Kuzinia sp. were present and collected from no fewer than four different carabid species each. They concluded there was no evidence for host specificity.

2.1.1 Goals of Study

I am interested in the study of millipedes in the family Julidae and their mite associates as they are common (synanthropic?) species in the U.S. but we know little about how they interact with other organisms, such as mites. To start to determine the nature of the relationship, I sampled millipedes from across the Eastern USA, and checked them for mites. I am interested in the following questions:

1. What small species of millipedes are present across the eastern U.S.?

2. What mites are associated with these species of millipedes?

3. Is there evidence of host specificity?

2.1.2 Hypotheses

The null hypothesis (H0) is that there is host specificity in relationships between the millipedes in the family Julidae and the phoretic mites found to be associated with them. If this is the case, I expect to see specific mite species associated with specific

49 millipede host species on a consistent basis. Second, if mites are host specific and there was a founder effect during colonization of the New World, diversity of specific mites in the U.S. is likely to be less than in Europe. It would be unlikely that a small samples of introduced millipedes would harbor the entire diversity of specific associates.

The alternate hypotheses (HA) is that the presence of mites on a millipede is based on locality, not host specificity. This hypothesis assumes that mites are specific to a certain area or off-host habitat. In that case a wide range of hosts occurring in the preferred habitat would be suitable as phoretic host. I would expect to see (1) specific mite species present on a number of different millipede host species but restricted to certain collection localities; (2) "American" mites on U.S. representatives of European millipede species, and (3) similar mite diversity on U.S. and European populations of the same millipede species.

2.2. METHODS AND MATERIALS

2.2.1 Collection Localities

Millipede specimens were collected from localities in the eastern United States from March through October in 2008 and 2009 and the United Kingdom in April 2009.

Sites in the U.S. were chosen due to their proximity to ports. In many cases, these millipedes are found in locations where there were ports during the colonial era, although these are not the only places where they are now found. This study started with local millipede species in Columbus, OH, which was not a port city. Table 2.3 details locality information for all geographic locations from which millipedes and mites were collected.

50 Table 2.3. Latitude and longitude of localities where millipedes were collected in 2008-2009.

United States Site name Lat./Long. North Kingstown, RI Richard Smith Grove 41.5814°, -71.4600°

Wilson Park 41.5740°, -71.4551°

Chicago, IL Grant Park 41.8725°, -87.6233°

Lincoln Park 41.9176°, -87.6346°

Cleveland, OH Lakewood 41.4971°, -81.7952°

Rockefeller Greenhouse 41.5355°, -81.6284°

Rockefeller Park 41.5208°, -81.6178°

Delaware, OH Highbanks Metropark 40.1531°, -83.0328°

Columbus, OH Franklin Park 39.9633°, -82.9505°

Smith Laboratory 40.0023°, -83.013°

Townshend Hall 39.9996°, -83.0157°

Whetstone Park 40.0434°, -83.0261°

Pike Co., OH Bryant Residence 39.0407°, -83.1083°

Logan campsite no data available

Baltimore, MD Cylburn Arboretum 39.3530°, -76.6516°

Druid Hill Park 39.3204°, -76.6394°

Charlotte, NC McAlpine Creek Park 35.1511°, -80.7424°

Reedy Creek Forest Preserve 35.2620°, -80.7199°

Sheffield Park 35.2068°, -80.7693°

Ft. Mill, SC Sun City 34.9291°, -80.8533°

United Kingdom Buckinghamshire Slough 51.5675°, -0.6231°

Cornwall Eden Project 50.3625°, -4.7449°

Glendurgan Garden 50.1068°, -5.1167°

Looe River Bank 50.3584°, -4.4660°

Luxulyan Valley no data available

Trelissick Garden 50.2166°, -5.0306°

51 2.2.2 Collection Methods

Millipedes were collected by hand except in one instance where litter from

Whetstone Park (Cols., OH) was returned to the OSU Acarology Laboratory for extraction by Berlese funnel. Hand collection was done from wood mulch, leaf litter, and soil. Much collecting was done at the soil/litter interface. Millipedes (with mites associates attached) were collected individually into 1.5 mL vials containing 95% ethanol. These specimens are stored at -20°C in the OSU Acarology Laboratory.

Millipede specimens used in this study are deposited in the Ohio State University

Acarology Collection, Columbus, Ohio, USA under the accession numbers identified in

Appendix A.

Upon returning to the lab the ethanol in the vials was inspected for mite associates which release their hold on the millipede after it is placed in the ethanol. Mite specimens were placed into a 0.92 mL (0.25 dram) glass shell vial containing 95% ethanol, stuffed with cotton, and then placed within the larger vial containing the millipede. These are stored at -20°C in the OSU Acarology Laboratory. Mite specimens used in this study are deposited in the Ohio State University Acarology Collection, Columbus, Ohio, USA under the accession numbers identified in Appendix B.

2.2.3 Determinations

Millipedes were dissected and, in some cases, genitalia were mounted on glass slides using the Strandtmann modification of Hoyer's mounting medium. Specimens were

52 identified using the following keys: Blower (1985), Shear (unpublished key), and Shelley

(1978; 1988; 2002). Hoffman's checklist (1999) was used as a reference for regional distributions.

Mites were chosen to be mounted on slides because they appeared to differ morphologically from other mite morphospecies. This was assessed by observation under a dissecting . Approximately 11% of mites collected were mounted for identification. The selected mites were removed from the ethanol and cleared for identification purposes using either lactophenol or Nesbitt's solution. Mites were mounted onto glass slides using the Strandtmann modification of Hoyer's mounting medium.

Specimens were identified to genus using keys to astigmatid mites by OConnor

(unpublished). In many cases, the mite species collected have not been described.

Species were given morphospecies determinations based on shape and amount of fusion of the coxal fields and coxal apodemes on the venter (Figure 1). These characteristics were chosen because combinations of these features are different among morphospecies in this study and are unlikely to be artifacts of the mounting process.

Therefore, they should be consistent within each morphospecies.

2.2.4 Prevalence and Average Intensity

Prevalences of the identified mites were calculated for each host species and each locality. Prevalence is defined as: (number of hosts with a particular parasite species) /

(number of hosts examined) (Margolis et al, 1982). Additionally, the average intensity

(average mites per infested host) for each host species was calculated (Margolis et al,

1982). Because not all mites were mounted for identification, prevalence calculations for

53 54 millipedes species where N > 20 are intended to give a minimum estimate for prevalence of a mite morphospecies per host. This is probably the opposite case for millipedes species collected < 20 times.

2.3. RESULTS

Representatives of a total of six genera of mites were associated with millipedes collected in this study. This included two families in the cohort Astigmata; Acaridae and

Histiostomatidae, and two families in the suborder Mesostigmata; Laelapidae and

Uropodidae.

2.3.1 Astigmata

The astigmatid mites represented 13 morphospecies: Rhizoglyphus A and B,

Sancassania A and B, Schwiebea A, C, D, E, F, G, and H, Thyreophagus sp., and

Histiostoma sp. Specimens of Rhizoglyphus, Sancassania, and Schwiebea were separated into morphospecies by the shape and amount of fusion of the apodemes and the shape of the coxal fields on the venter. In some cases, the differences between the species are subtle. The following are the distinguishing characteristics of the morphospecies within the genera Rhizoglyphus, Sancassania, and Schwiebea. Figures show conoidal setae as filled-in circles and simple setae as open circles. The size of the conoids in the diagram is indicative of their relative size.

55 56 2.3.2 Rhizoglyphus

Rhizoglyphus A- Figure 2. The posterior edge of apodeme II is not fused with anterior edge of apodeme III. These are both poorly developed. Median ventral apodeme

III is well-developed. The median ventral apodeme of coxal field III is straight and almost perpendicular to anterior apodeme III.

Rhizoglyphus B- Figure 3. The posterior edge of apodeme II is fused with anterior edge of apodeme III. Both of these are well-developed where they fuse. Median ventral apodeme III is poorly developed. The median ventral apodeme of coxal field III is acutely rounded.

2.3.3 Sancassania

Sancassania A- The coxal apodemes of this morphospecies are very poorly developed. After being slide-mounted, the venter of each specimen fragmented as if soft and highly unsclerotized. Characters could not be described.

Sancassania B- Figure 4. The posterior edge of apodeme II is not fused with the anterior edge of apodeme III. The posterior edge of apodeme II is poorly developed and rounded. Anterior apodeme III is well-developed and straight. The apodemes of coxal field III are all well-developed.

2.3.4 Schwiebea

Morphospecies of Schwiebea were determined using eight morphological characteristics of the venter; shape of coxal fields II, shape of posterior edge of coxal fields II, anterior edge of coxal fields III, shape of coxal fields III and IV, quality of development of apodemes III and IV, the proximity of the median ventral apodemes to

57

58 59 each other, and the size of the conoidal setae in coxal field I. The most important characteristics in determining these species are described here. Full character descriptions used to determine morphospecies are available in Table 2.4.

Schwiebea A- Figure 5. Coxal fields II are oval, not widened distally (opposed to

Schwiebea D). The posterior edge of coxal field II and anterior edge of coxal field III areconvex and concave in their curvature from the middle of the mite, respectively.

Schwiebea C- Figure 6. Coxal fields II are roughly quadrilateral with straight apodemes. Posterior edge of coxal fields II and anterior edge of coxal fields III are both concave and poorly developed.

Schwiebea D- Figure 7. Similar to Schwiebea A with three exceptions. Coxal fields II are larger and much more widened distally. The median ventral apodemes are in very close proximity to each other (almost touching).

Schwiebea E- Figure 8. All apodemes are very well-developed. Coxal fields II are roughly quadrilateral. Posterior edge of coxal fields II and anterior edge of coxal fields III are straight and parallel to each other crossing the venter. Coxal fields IV have a small lobe on the posterior edge.

Schwiebea F- Figure 9. Coxal fields II are roughly quadrilateral and there is a distinct curve medially at the end of apodeme II. Coxal fields III and IV are poorly- developed.

Schwiebea G- Figure 10. Similar to Schwiebea E in that apodemes are very well- developed and coxal fields II are roughly quadrilateral. Posterior edge of coxal fields II and anterior edge of coxal fields III are straight and parallel to each other crossing the

60 Table 2.4. Characteristics used to distinguish morphospecies of mites in the genus Schwiebea collected in 2008-2009.

Schwiebea A (Fig.5) Schwiebea C (Fig. 6) Schwiebea D (Fig. 7)

Coxal fields II roughly oval, well- roughly quadrilateral; roughly oval, widened developed apodemes straight, well- distally, mostly well- developed developed

Posterior edge of coxal convex, poorly concave, poorly developed convex, poorly developed fields II developed medially medially

Anterior edge of coxal roughly concave, concave, poorly developed slightly concave, well- fields III well-developed developed

Coxal fields III & IV III: rounded medially III: rounded medially III: rounded medially, almost IV: roughly IV: roughly triangular touching IV: roughly triangular triangular

Apodemes III & IV well-developed poorly developed anterior III well-developed; where posterior III and anterior IV meet, poorly developed

Median ventral III: not touching III: not touching IV: not III: ALMOST touching apodemes IV: touching, extend touching, extend to genital IV: touching, terminates to genital opening opening half-way to genital opening

61 Table 2.4. Continued.

Schwiebea F (Fig. 9) Schwiebea G (Fig. 10) Schwiebea H (Fig. 11)

Coxal fields II roughly quadrilateral; roughly quadrilateral; roughly oval, well- anterior apodemes II end fields I & II contiguous, developed in distinct curve medially well-developed

Posterior edge of coxal slightly concave, straight, horizontal, median section of fields II quadrilateral edge poorly quadrilateral edge, well- posterior curve poorly developed developed developed

Anterior edge of coxal slightly concave, poorly straight, horizontal, slightly concave at fields III developed quadrilateral edge, well- lateral edges, well- developed developed

Coxal fields III & IV III: roughly triangular, not III: fields are contiguous, III: roughly triangular, touching IV: roughly quadrilateral IV: roughly not touching IV: triangular, not touching, deltoid, touching roughly arrow-shaped divided into two fields medially, acute tapering medially, ending in simple setae; divided into two fields

Apodemes III & IV poorly developed well-developed mostly well-developed

Median ventral apodemes III: not touching IV: not III: N/A IV: in full III: not touching, poorly touching, extend to genital contact, extend to genital developed IV: poorly opening opening developed in anterior, but well-developed in the posterior extension, touching

Coxal field I conoidal large very small very small setae

62 63 64 65 66 venter. Coxal fields IV are the same in that they have a small lobe on the posterior edge.

They are different because they have a deltoid shape and the anterio-medial end tapers acutely.

Schwiebea H- Figure 11. Coxal fields II are really rounded on the posterio-medial edge. These are also poorly developed toward the medial edge. Coxal field IV has a unique “arrow” shape that none of the other morphospecies have.

2.3.5 Mesostigmata

Mites in the families Laelapidae and Uropodidae were determined to genus. Two genera of laelapids were identified as Cosmolaelaps and Holostapis. The uropodid genus was Phaulodinychus.

2.3.6 Mite-Millipede Associations

In 2008 and 2009, 23 species of millipedes were collected in the United States and the United Kingdom. These specimens represent 9 different families with a majority of the specimens in the family Julidae, the focus of this study. Mite prevalence and average mites per host were calculated for all millipede species collected (Table 2.5).

2.3.7 Julid Millipede Collections

Thirteen species of millipedes in the family Julidae were collected during 2008-

2009 (Table 2.5). Six species were collected in quantities > 100; pusillus

(Leach), Cylindroiulus caeruleocinctus (Wood), Cylindroiulus latestriatus (Curtis),

67 Table 2.5. Prevalence and average intensity of mites collected from millipedes in 2008-2009.

y

t

i

B A s

B A

)

n s s s a a C D E A e u u i i

t e h h n n a a a a n d i p p a a e e e e

e s s y y b b b b e l l p i s s e e e e l g g g i i i i l a a i a o o c c w w w r w z z m i i n n h e h h h ( h h a a c c c c Species v N A R R S S S S S S

Abacion lactarium 8 3 12.50% 12.50%

Blaniulid sp. 104 2 1.89% 1.89%

Blaniulus guttulatus 93 1.2 3.77% 6 8 Choneiulus palmatus 8 2.4

Nopoiulus kochii 76 1.6

Cleidogona caroliniana 7 2

Euryurus leachii 14 11.27 7.14%

Glomeris marginata 4 2.67 50.00%

Brachyiulus lusitanus 16 1

Brachyiulus pusillus 208 2.3 0.52% 1.57%

Cylindroiulus britannicus 35 2 2.86% 2.86%

Cylindroiulus caeruleocinctus 365 2.77 2.19% 0.82% 0.27% 0.27% Continued. Table 2.5 Continued.

y

t

i

B A s

B A

)

n s s s a a C D E A e u u i i

t e h h n n a a a a n d i p p a a e e e e

e s s y y b b b b e l l p i s s e e e e l g g g i i i i l a a i a o o c c w w w r w z z m i i n n h e h h h ( h h a a c c c c Species v N A R R S S S S S S

Cylindroiulus latestriatus 337 1.66 0.30% 0.30%

Cylindroiulus londinensis 2 5 50.00%

Cylindroiulus punctatus 109 3.32 0.92% 3.67%

6 9 Cylindroiulus truncorum 171 1.89 1.17% 1.75% 0.58%

Cylindroiulus sp. 186 2.74 0.54% 2.15%

Julus scandinavicus 19 3.17 5.26% 5.26%

Ommatoiulus sabulosus 9 2

Ophyiulus pilosus 373 2.64 0.27% 0.54%

Tachypodoiulus niger 2 5

Uroblaniulus carolinensis 8 2.67 12.50%

Polydesmus angustus 6 6 16.67%

Polydesmus sp. 8 1 12.50%

Paraspirobolus lucifugus 5 2.86 20.00% Continued. Table 2.5 Continued.

.

p . . s p p

s s s .

. u p

s s s h p u p F

G H c s

g a

a l y a a a a s e i n m e e e i h p a o b b b l p d t a e e e o t o o s i i i l s e o m u w w w i r o t s l a h h h y s o i o h Species c c c h S S S T H C H P

Abacion lactarium

Blaniulid sp. 1.89% 1.89%

Blaniulus guttulatus 0.94% 7

0 Choneiulus palmatus

Nopoiulus kochii

Cleidogona caroliniana

Euryurus leachii 7.14% 7.14% 14.29% 14.29%

Glomeris marginata

Brachyiulus lusitanus

Brachyiulus pusillus

Cylindroiulus britannicus 2.86% 2.86%

Cylindroiulus caeruleocinctus 1.37% Continued. Table 2.5. Continued.

.

p . . s p p

s s . s

p . u

s s s h p u p F

G H c s

g a

a l y a a a a s e i n m e e e i h p a o b b b l p d t a e e e o t o o s i i i l s e o m u w w w i r o t s l a h h h y s o i o h Species c c c h S S S T H C H P

Cylindroiulus latestriatus 0.30% 0.30%

Cylindroiulus londinensis 50.00%

Cylindroiulus punctatus 0.92% 0.92%

7 Cylindroiulus truncorum 0.58% 1

Cylindroiulus sp. 1.08%

Julus scandinavicus

Ommatoiulus sabulosus 11.11%

Ophyiulus pilosus

Tachypodoiulus niger 50.00%

Uroblaniulus carolinensis 12.50%

Polydesmus angustus 16.67% 16.67% 16.67%

Polydesmus sp.

Paraspirobolus lucifugus 40.00% Cylindroiulus punctatus (Leach), Cylindroiulus truncorum (Silvestri), and Ophyiulus pilosus (Newport). The average intensities for these six species ranged from 1.66-3.32 mites per host with an average of 2.43 mites per host. This is probably very close to the average intensity in the family Julidae.

Nearly all mite specimens collected from julid millipedes were astigmatid deutonymphs. Morphospecies of Schwiebea were the most commonly collected from these millipedes (Table 2.5). Rhizoglyphus A was collected from the most julid species in this study (n = 9). This was followed by Histiostoma sp. (n = 6). Schwiebea D, E, and G and Phaulodinychus sp. were collected from one julid host species only. C. caeruleocinctus had five species of mite associates which was the highest diversity among julids. Tachypodoiulus niger (Leach) and Ommatoiulus sabulosus (Linné) had the fewest associate species (one each). Mite species of Thyreophagus specimens were only found on O. sabulosus and Cylindroiulus britannicus (Verhoeff). Phaulodinychus sp. was collected only once on a julid millipede, C. latestriatus. There were zero laelapid mites collected from julid millipedes.

Prevalences ranged from 0.27%-50.0%. Prevalence of Schwiebea H on T. niger was 50.0% (n = 2). This was also the case with Cylindroiulus londinensis (Leach) (n = 2) in regards to the two mite morphospecies collected from it (Rhizoglyphus A and

Histiostoma sp.). The smallest prevalence was calculated for Rhizoglyphus A on O. pilosus (0.27%).

2.3.8 Non-Julid Millipede Collections

72 Eight families other than family Julidae were collected in 2008-2009 (Table 2.5).

Schwiebea C was collected from the most non-julids (n = 5). Polydesmus angustus Latzel had the highest number of mite associates with four morphospecies of Schwiebea. All four of these were collected from one individual. Histiostoma sp. and three species of

Schwiebea have been collected from the combined specimens in the family .

The two taxa collected from Uroblaniulus carolinenesis Causey and Paraspirobolus lucifugus (Gervais) were Histiostoma sp. and Schwiebea C. Abacion lactarium (Say) was the only species, julid or non-julid, to have Rhizoglyphus B as an associate. The euryurid

Euryurus leachii (Gray) was the only millipede host to have mite associates in the family

Laelapidae. Representatives of two genera in this family were collected, Cosmolaelaps sp. and Holostapis sp. These were also the only mites collected as adults, not deutonymphs. Additionally, a considerable number of deutonymphs of the uropodine

Phaulodinychus sp. were collected from this millipede. E. leachii was the millipede with the highest intensity (11.27) and the largest diversity of mites within the non-julids with five mite associates.

The prevalence of mites found on non-julid millipedes was calculated. For many of these groups < 20 individuals were collected. 50.0% of the Glomeris marginata

(Villers) (n = 4) collected had only Schwiebea E associated with it. Blaniulids were collected the most and had prevalences between 0.94%-3.77. Schwiebea D, E, and F, and

Histiostoma sp. were collected.

2.3.9 Mite Taxa and Relation to Locality

73 Of the sixteen mite taxa collected from millipede hosts in 2008-2009, only four were collected from localities in both the U.S. and in the U.K., Rhizoglyphus A,

Schwiebea C, Schwiebea F, and Histiostoma sp. (Table 2.6). Mite taxa that were found exclusively in the U.S are Rhizoglyphus B, Sancassania A, Sancassania B, Schwiebea D,

Cosmoglyphus sp., Holostapis sp. and Phaulodinychus sp. Taxa found only in the U.K.

Schwiebea A, Schwiebea E, Schwiebea G, Schwiebea H, and the genus Thyreophagus.

The highest diversity of mite taxa was collected in Cornwall, U.K. Only one mite taxon each was collected in Chicago, IL, Columbus, OH, and Lakewood, OH.

Of the seven millipede species collected in both the U.S. and the U.K. only

Cylindroiulus caeruleocinctus had the same mite associates in both countries,

Rhizoglyphus A and Histiostoma sp. (Table 2.7). Mite associates of the other six species of millipedes did not appear in both countries on the same julid host species.

2.4. CONCLUSIONS

In evaluating the diversity of mite species data collected from the millipedes in family Julidae, I have concluded from this preliminary study that the mite associates are, for the most part, not host-specific. With the exception of four of the mite taxa

(Schwiebea D, Schwiebea E, Schwiebea G, and Phaulodinychus sp.), all other taxa were collected from more than one julid host (Table 2.5). These four mite taxa were also associated with non-julid hosts.

The results of this preliminary study indicate that the acarid associates of small anthropophilic julid millipedes are most likely not host specific. What may appear to be

74 Table 2.6. Prevalence of identified mite associates on millipedes collected between 2008-2009 by locality.

B A

B A

)

s s s a C D E a A u i

u i

e h n h n a a a a d p a p a e e e e e y s y s b b b b l l p i s s e l e e e g g i i i i l a a i o o c c w w w w z z m i i n n h h h h ( h h a c c c c Locality a N R R S S S S S S

North Kingstown, RI 175 0.57% 2.86% 0.57%

Chicago, IL 452

Lakewood, OH 61 1.64%

Cleveland, OH 119 1.68%

Delaware, OH 23 8.70% 4.35% 4.35%

Columbus, OH 1054 0.66%

Southern Ohio 7

Baltimore, MD 102 7.84% 0.98% 2.94%

Charlotte, NC 161 0.62% 1.24% 1.24%

Fort Mill, SC 15 6.67%

Cornwall, UK 59 1.69% 1.69% 1.69% 1.69%

Eden Project, UK 56 1.79% 3.57%

Slough, UK 27 7.41% 18.52% 7.41% Continued.

75

Table 2.6. Continued.

.

p . . s p p

s s . s

u p . s

s )

s h p u s p

F G H c s g a e a l y a a a a s d e i n m e e e i e h p a o b b b l p d p t i a l e e e o o t o s i i i l l i s e o m u w w w i r o t m l s a y h h h ( s o o i h h Locality c c c N S S S T H C H P

North Kingstown, RI 175 0.57% 2.29%

Chicago, IL 452 0.66%

Lakewood, OH 61

Cleveland, OH 119 0.84%

Delaware, OH 23

Columbus, OH 1054

Southern Ohio 7 14.29% 28.57% 33.33%

Baltimore, MD 102 1.96%

Charlotte, NC 161 0.62%

Fort Mill, SC 15 6.67% 6.67%

Cornwall, UK 59 1.69% 1.69% 1.69% 1.69%

Eden Project, UK 56 1.79% 1.79% 5.36%

Slough, UK 27 3.70% 3.70% 3.70% 3.70%

76 Table 2.7. Mites present on millipede species collected in both the U.S. and the U.K. . p s .

A

p A B s

s s u

a C E H A D G a F i u

i

g a n h n a a a a a a a a a m p a e e e e e e e h s y s o b b b b b b b l p t s s e e e e e e e g o s i i i i i i i a a o e o c c w w w w w w w i r z t i n n h h h y h h h h s h i a c c c c c c c a Country N h R S S S S S S S S S T H

Blaniulidae, Blaniulus guttulatus U.S. 103 X U.K. 3 X Blaniulidae, fascus U.S. 29 U.K. 1 Julidae, Cylindroiulus britannicus U.S. 11

7 U.K. 24 X X X X 7 Julidae, Cylindroiulus caeruleocinctus U.S. 361 X X X X U.K. 1 X X X Julidae, Cylindroiulus punctatus U.S. 75 X X U.K. 34 X X Julidae, Ophyiulus pilosus U.S. 361 X X U.K. 11 , Polydesmus angustus U.S. 1 U.K. 5 X X X X host preference in some cases may be opportunism on the part of the mite. It has been noted by OConnor (1998) that these acarids are often associates of a wide variety of hosts and are cosmopolitan. There is precedent for this in terms of millipede-associated

Mesostigmata. While most Heterozerconidae are associated with large tropical millipedes, one genus, Amheterozercon Fain, is associated with and amphisbaenids (Flechtmann and Johnston, 1990). In contrast, most Paramegistidae are associates of carabid beetles or millipedes, except for the genus Ophiomegistus Banks which is associated with and snakes (Klompen and Austin, 2005). These records suggest that these mites favor a host with a specific type of locomotion and general distance from the ground. In forest habitats, animals of this description abound.

Julid millipedes collected both in the U.S. and the U.K. did not seem to have mite morphospecies that were consistently found on one particular species (Table 2.7) with one exception (C. caeruleocinctus). This supports the above conclusion of little evidence for host specificity. In a comparison of mite associates found on millipedes and the locality in which they were found (Table 2.6), 12 of the 16 taxa were collected in either the U.S. or the U.K. The remaining four were collected from both countries. This is an indication that the mites are most likely not seeking specific hosts but, instead, are favoring hosts with associations with specific habitats.

An unusual finding regarding mite associates was the collection of a species in the genus Thyreophagus collected in the United Kingdom. Thyreophagus is a mite thought to be associated only with subcortical insects (OConnor, 1982) or are very specific in host choice in some other way (OConnor, 1984). This is very different from the lifestyles of

78 the other acarids collected in this study that seem to be very general in their host choice.

O. sabulosus and C. britannicus were the hosts of this mite. They were collected from an outdoor garden a greenhouse/conservatory in Cornwall, U.K. which were a good distance from each other. Past reports of the hosts of Thyreophagus sp. are mostly from the U.S., however. A further survey of ground arthropods in Cornwall could help clarify the degree of specificity of this genus in the area.

2.4.1 Caveats

In general, more mites were present on millipedes than were mounted and identified for the study. Total numbers of mites present are known as are the average intensities (for all mite species combined), but the numbers divided by mite species are underestimates, because not all individual mites were identified. For example, with the specimen of P. angustus (Table 2.2) collected from Buckinghamshire, U.K. not all mites were prepared and mounted for identification. I found this millipede was host to four morphospecies of Schwiebea. This may be typical for many millipedes of this size (julid or non-julid) but these instances can complicate exact evaluations of species present on a millipede without completing the time consuming practice of mounting and identifying every mite from every millipede.

Additionally, understanding the likelihood of finding a particular mite species on a millipede in a particular habitat would most likely require a survey of the mite taxa in the habitat. There would be some difficulties with this however. Since many of these mites (a majority are acarids) are fungivores, it would be necessary to survey soil,

79 litter/mulch, and the fungal resources available in the area. It is unknown how far millipede taxa in this survey may travel for fungus (if they travel to large resources), but this is something that should be taken into consideration when designing a study to evaluate possible millipede associates further. However, the results provide a good start towards better understanding of the overall taxa found on these small millipedes and also what their role is in the relationship.

2.4.2 Additional findings: Non-julid millipede hosts

In evaluating the non-julid millipede hosts, there is some evidence that some millipede hosts may have specific mites associated with them. Three mite taxa each were associated with one host species (Table 2.5). Rhizoglyphus B was associated with A. lactarium and both Cosmolealaps sp. and Holostapis sp. were associated only with E. leachii. The last two genera were collected as adults and could be parasites of E. leachii.

E. leachii is a polydesmid, a “flat-backed” millipede, and these tend to have more tissue exposure between segments than the “cylindrical” julids, spirostreptids, spirobolids, and callipodids. If the laelapid taxa are parasitic, E. leachii would most likely, be a better resource to utilize.

E. leachii was also observed to be the most common host of the uropodid

Phaulodinychus. This mite was usually attached to the legs, an observation valid for specimens collected in this study as well as specimens which were observed but not collected. The presence of uropodid deutonymphs on millipedes is thought to be a rare occurrence, and has not been reported before. They have been collected from

80 (Bloszyk et al, 2006). A few other collections from millipedes has been reported to the

OSU Acarology Laboratory. Phaulodinychus are fungivores or predators and are known for their associations with social insects (Eickwort, 1990). This may also be a situation in which there may be some degree of host specificity.

The second interesting finding is that Abacion lactarium () did not have any mite genera present as an associate besides species of Rhizoglyphus. This is a millipede which is readily collected from ravines and more forested habitats in some surprisingly urban areas (Shelley, 1984; pers. obs.). It was found in city parks near the main campus of the Ohio State University (Iuka Park) and with great frequency in

Whetstone Park, another city park in the residential part of Columbus, OH. It has some lifestyle traits that are different from a number of other millipedes. It is a detritivore but is known as a scavenger, feeding from small vertebrate carrion and arthropods (Marek and

Shelley, 2005; Hoffman and Payne, 1969). When raised in culture, they are fed “trout chow”, a mix of pellets of dried fish, as well as leaf litter. It is difficult to know if necrophagy is a lifestyle which would affect host preference of an acarid. Another possibility is that this has something to do with the composition of the defensive chemicals secreted by A. lactarium. The sister species of this millipede, A. magnum, had been studied and was found to secrete a p-cresol (a phenol) which has not been discovered to be secreted by any other millipedes (Eisner et al, 1978). It is very common for julids to secrete benzoquinones and for polydesmids to secrete cyanic compounds, so this may provide enough difference for the average acarid that Abacion sp. is undesirable.

Additonally, A. lactarium is quite fast compared to other millipede species found in

81 similar habitats (pers. obs.). Because acarid deutonymphs are relatively slow, it is possible that they do not have as much opportunity to board an Abacion sp. It is unclear why Rhizoglyphus B was found on only this host species.

2.5 SUMMARY

Sixteen taxa of mites were collected from millipedes in the U.S. and in the U.K. in 2008-

2009. Twelve taxa of mites were collected from 13 species of julid millipedes. There was no evidence of host specificity by the mites. Twelve of the 16 mite species collected from all millipedes, julid and non-julid, were collected from either the U.S. or the U.K. Is seems that the mites collected from these millipedes are similar to those collected from other ground-dwelling arthropods.

82 References

Anderson, J.M. 1988. -mediated transport in soils. , Ecosystems, and Environment 24: 5-19.

Anderson, J.M. and D.E. Bignell. 1980. Bacteria in the food, gut and feces of the litter feeding millipede Glomeris marginata (Villers). and Biochemistry 12: 251-254.

Anderson, J.M., S.A. Huish, P. Ineson, M.A. Leonard, and P.R. Splatt. 1985. Interactions of Invertebrates, Micro-Organisms and Tree Roots in Nitrogen and Mineral Element Fluxes in Deciduous Woodland Soils. In Ecological Interactions in Soil: Plants, Microbes and Animals, edited by A. H. Fitter, D. Atkinson, D. J. Read and M. B. Usher. Vol. 4, 377-392. Oxford; Boston: Blackwell Scientific.

Anderson, J.M. and P. Ineson. 1983. Interactions between Soil Arthropods and Microbial Populations in Carbon, Nitrogen, and Mineral Nutrient Fluxes from Decomposing Leaf Litter. In Nitrogen as an Ecological Factor, edited by J. A. Lee, S. McNeill and I. H. Rorison, 413-432. Oxford: Blackwell Scientific Publications.

———. 1984. Interactions between Microorganisms and Soil Invertebrates in Nutrient Flux Pathways of Forest Ecosystems. Exeter, Cambridge University Press, .

Anderson, J.M., P. Ineson, and S.A. Huish. 1983. Nitrogen and cation release by macrofauna feeding on leaf litter and soil organic matter from deciduous woodlands. Soil Biology and Biochemistry 15: 463-467.

Anderson, J.M., M.A. Leonard, P. Ineson, and S.A. Huish. 1985. Faunal biomass: A key component of a general model of nitrogen mineralization. Soil Biology and Biochemistry 17, no. 5: 735-737.

Baker, A.N. 1974. Some aspects of the economic importance of millipedes. Symposia of the Zoological Society of London 32: 621-628.

Baker, G.H. 1985. Parasites of the millipede (Lucas) (Diplopoda: Iulidae) in portugal, and their potential as biological control agents in Australia. Australian Journal of Zoology 33: 23-32.

Baker, M.R. and O.D. Seeman. 2008. Mites and millipedes: A new Neomegistus (: Mesostigmata: Paramegistidae) from Australia. Systematic and Applied Acarology 13, no. 3-4: 204-213.

83 Bano, K. 1992. The role of the Millipede Jonespeltis splendidus Verhoeff in an ecosystem (Diplopoda, , ). Berichte Des Naturwissenschaftlich-Medizinischen Vereins in Innsbruck Suppl. 10: 327-332.

Berlese, A. 1916. Centuria prima di Acari nuovi. Redia 12: 31-32.

Bignell, D.E. 1989. Relative assimilations of 14C-labeled microbial tissues and 14C-plant fibre ingested with leaf litter by the millipedes Glomeris marginata under experimental conditions. Soil Biology and Biochemistry 21, no. 6: 819-827.

Bloszyk, J., J. Klimczak, and M. Lesniewska. 2006. Phoretic relationships between Uropodina (Acari : Mesostigmata) and centipedes (Chilopoda) as an example of evolutionary adaptation of mites to temporary microhabitats. European Journal of Entomology 103, no. 3: 699-707.

Blower, J.G. 1985. Millipedes: Keys and Notes for the Identification of the Species . Synopses of the British Fauna (New Series)., edited by Doris M. Kermack, R. S. K. Barnes. Vol. No. 35. London: E. J. Brill/Dr. W. Backhuys,.

Bocock, K.L. 1963. The Digestion and Assimilation of Food by Glomeris. Oosterbeek, Netherlands,

———. 1964. Changes in the amounts of dry matter, nitrogen, carbon and energy in decomposing woodland leaf litter in relation to the activities of the soil fauna. Journal of Ecology 52: 273-284.

Bocock, K.L. and J. Heath. 1967. Feeding activity of the millipede Glomeris marginata (Villers) in relation to its vertical distribution in the soil. In Progress in Soil Biology, 233-240. Amsterdam: North Holland Publishing Company.

Bonkowski, M., S. Scheu, and M. Schaefer. 1998. Interactions of earthworms (Octolasion lacteum), millipedes (Glomeris marginata) and plants (Hordelymus europaeus) in a beechwood on a basalt hill: Implications for litter decomposition and soil formation. Applied Soil Ecology 9: 161-166.

Brüggl, C. 1992. Gut passage, respiratory rate and assimilation efficiency of three millipedes from a deciduous woods in the Alps (Julidae: Diplopoda). Berichte Des Naturwissenschaftlich-Medizinischen Vereins in Innsbruck Suppl. 10: 319-326.

Canestrini, G. 1888. Intorno ad alcuni Acari ed Opilionidi dell America. Atti Della Società Veneto-Trentina Di Scienze Naturali 11, no. 1: 100-111.

Cárcamo, H.A., T.A. Abe, C.E. Prescott, F.B. Holl, and C.P. Chanaway. 2000. Influence of millipedes on litter decomposition, N mineralization, and microbial communities on a coastal forest in British Columbia, Canada. Canadian Journal of Forest Research 30, no. 5: 817-826.

84 Cárcamo, H.A., C.E. Prescott, C.P. Chanaway, and T.A. Abe. 2001. Do soil fauna increase rates of litter breakdown and nitrogen release in forests of British Columbia, Canada? Canadian Journal of Forest Research 31: 1195-1204.

Carter, A. and J.B. Cragg. 1976.Concentrations and standing crops of calcium, magnesium, potassium, and sodium in soil and litter arthropods and their food in an aspen woodland ecosystem in the Rocky Mountains (Canada). Pedobiologia 16: 379-388.

Casanueva, M.E. 1993. Phylogenetic studies of the free-living and arthropod associated Laelapidae (Acari: Mesostigmata). Gayana Zoologia 57, no. 1: 21-46.

Casanueva, M.E. and D.E. Johnston. 1992. Systematic studies on Jacobsonia (Acari, Mesostigmata), a mite associated with indo-malaysian millipedes. Boletin De La Sociedad De Biologia De Concepcion 63: 55-63.

Causey, N.B. 1953. On five new North American millipeds and records of some established species. American Midland Naturalist 50, no. 1: 152-158.

Chapin, F.S., III, P.A. Matson, and H.A. Mooney. 2002. Principles of Terrestrial Ecosystem Ecology. New York: Springer-Verlag.

Corente, C. and W. Knülle. 2003. Trophic determinants of hypopus unduction in the stores-product mite Lepidoglyphus destructor (Acari: Astigmata). Experimental and Applied Acarology 29: 89-107.

Crawford, C.S. 1992. Millipedes as model detritivores. Berichte Des Naturwissenschaftlich-Medizinischen Vereins in Innsbruck Suppl. 10: 277-288.

Cromack, K., P. Sollins, R.L. Todd, D.A. Crossley Jr., W.M. Fender, R. Fogel, and A.W. Todd. 1977. Soil microorganisms-arthropod interactions: Fungi as major calcium and sodium sources. In Role of Arthropods in Forest Ecosystems, edited by W. J. Mattson, 78-84. New York: Springer-Verlag.

Cromack, K., R.L. Todd, R. Fogel, A.W. Todd, W.M. Fender, M.E. Crossley, and D.A. Crossley Jr. 1977. The role of oxalic acid and bicarbonate in calcium cycling by fungi and bacteria: Some Possible Implications for Soil Animals. Uppsala, Sweden, Swedish Natural Science Research Council.

Crossley, D.A., Jr. 1970. Roles of Microflora and Fauna in Soil Systems. East Lansing, MI.

———. 1977. The Roles of Terrestrial Saprophagous Arthropods in Forest Soils: Current Status of Concepts. In Role of Arthropods in Forest Ecosystems, edited by W. J. Mattson, 49-56. New York: Springer-Verlag.

85 Crossley, D.A., Jr., D.C. Coleman, and P.F. Hendrix. 1989. The importance of the fauna in agricultural soils: Research approaches and perspectives. Agriculture, Ecosystems, and Environment 27: 47-55.

Dangerfield, J.M. 1993. Ingestion of mineral soil/litter mixtures and faecal pellet production in the southern millipede uncinatus (Attems). Pedobiologia 37: 159-166.

———. 1994. Ingestion of leaf litter by millipedes: The accuracy of laboratory estimates for predicting litter turnover in the field. Pedobiologia 38, no. 262: 265.

———. 1995. Effects of starvation and mechanical manipulation of leaf litter on faecal pellet production and assimilation in some millipedes from southern africa: Implications for feeding strategies. European Journal of Soil Biology 31: 199-206.

Dangerfield, J.M. and A.E. Milner. 1993. Ingestion and assimilation of leaf litter in some tropical millipedes. Journal of the Zoological Society of London 229: 683-693.

Dangerfield, J.M. and S.R. Telford. 1989. Are millipedes important for soil fertility? Zimbabwe Science News 23: 66-68.

———. 1991. Seasonal activity patterns of julid millipedes in Zimbabwe. Journal of Tropical Ecology 7: 281-285.

Eickwort, G.C. 1990. Associations of mites with social insects. Annual Review of Entomology no. 35: 469-488.

Eisner, T., D. Alsop, K. Hicks, and J. Meinwald. 1978. Defensive secretions of millipeds. In Handbook of Experimental Pharmacology: Arthropod , edited by Sergio Bettini. Vol. 48, 41-72. New York: Springer-Verlag.

Elliot, H.J. 1970. The role of millipedes in the decomposition of Pinus radiata litter in the Australian capital territory. Australian Forest Research 4: 3-10.

Englemann, M.D. 1961. The role of soil arthropods in the energetics of an old field community. Ecological Monographs 31, no. 3: 221-238.

Evans, G.O. 1955. A review of the laelaptid paraphages of the Myriapoda with descriptions of three new species (Acarina: Laelaptidae). Parasitology 45, no. 3-4: 352-368.

Evans, G.O. and J.G. Sheals. 1959. Three new mesostigmatic mites associated with millipedes in Indonesia. Entomologische Berichten, Amsterdam 19: 107-111.

Fain, A. 1987a. Notes on mites associated with Myriapoda. I. Three new astigmatic mites from Afrotropical Myriapoda (Acari, Astigmata). Bulletin De l'Institut Royal Des

86 Sciences Naturelles De Belgique, Entomologie 57: 161-172.

———. 1987b. Notes on mites associated with Myriapoda: II. Four new species of the genus Julolaelaps Berlese, 1916 (Acari, Laelapidae). Bulletin De l'Institut Royal Des Sciences Naturelles De Belgique, Entomologie 57: 203-208.

———. 1988. Notes on mites associated with Myriapoda. III. Two new species of the genus Heterozercon Berlese, 1888 (Acari, Mesostigmata) from Afrotropical myriapods. Bulletin Et Annales De La Société Royale Belge d'Entomologie 124: 237-242.

———. 1989. Notes on mites associated with Myriapoda. IV. New taxa in the Heterozerconidae (Acari, Mesostigmata). Bulletin De l'Institut Royal Des Sciences Naturelles De Belgique, Entomologie 59: 145-156.

———. 1992. Notes on mites associated with Myriapoda: V. The genus Scissuralaelaps Womersley, 1945 (Acari, Mesostigmata): Description of four new species from New Guinea. Bulletin De l'Institut Royal Des Sciences Naturelles De Belgique, Entomologie 62: 109-116.

———. 1994. Notes on mites associated with Myriapoda: VI. The genus Jacobsonia Berlese, 1910 (Acari, Laelapidae): Description of four new species from New Guinea. Journal of African Zoology 108: 585-595.

Flechtmann, C.H.W. and D.E. Johnston. 1990. Zeterohercon, a new genus of Heterozerconidae (Acari: Mesostigmata) and the description of Zeterohercon amphisbaene n sp. from Brasil. International Journal of Acarology 16, no. 3: 143- 148.

Funk, R.C. 1980 (1981). Generic revision of the family Euzerconidae (Mesostigmata: Celaenopsoidea) with description of seven new genera. International Journal of Acarology 6, no. 4: 313-349.

Gerdeman, B.S. The Biology of the Heterozerconidae. Ph.D., Ohio State University, 2002.

Gerdeman, B.S. and H. Klompen. 2003. A new North American heterozerconid, Narceoheterozercon ohioensis, n. gen., n. sp., with first description of immatures of Heterozerconidae (Acari: Mesostigmata). International Journal of Acarology 29, no. 4 : 351-370.

Gerdeman, B.S., H. Klompen, and L. Tanigoshi. 2000. Insights into the biology of a mite - millipede association. Fragmenta Faunistica, Warszawa 43 (Suppl.): 223-227.

Gere, G. 1956. The examination of the feeding biology and humificative function of Diplopoda and Isopoda. Acta Biologica Academiae Scientiarum Hungaricae 6: 257-

87 271.

Ghilarov, M.S. 1962. On the Interrelations between soil dwelling invertebrates and soil microorganisms. Oosterbeek, Netherlands, .

Gist, C.S. and D.A. Crossley Jr. 1974. A Model of Mineral Cycling for an Arthropod Foodweb in a Southeastern Ecosystem. Augusta, Georgia.

———. 1975. The litter arthropod community in a southern Appalachian hardwood forest: Numbers, biomass and mineral element content. American Midland Naturalist 93, no. 1 Jan.: 107-122.

Gu, Y., C. Wang, and J. Li. 1991. A new genus and species of gamasides off Julus terrestris and a new family proposed (Acari: ). Acta Zootaxonomica Sinica 16: 428-431.

Hanlon, R.D.G. 1981. Some factors influencing microbial growth on soil animal faeces I. bacterial and fungal growth on particulate oak leaf litter. Pedobiologia 21: 257-263.

———. 1981. Some factors influencing microbial growth on soil animal faeces II. bacterial and fungal growth on soil animal faeces. Pedobiologia 21: 264-270.

Hanlon, R.D.G. and J.M. Anderson. 1980. Influence of microarthropod feeding activities on microflora in decomposing oak leaves. Soil Biology and Biochemistry 12: 255- 261.

Hoffman, R.L. 1999. Checklist of the Millipeds of North and Middle America. Vol. 8. Martinsville, VA: Virginia Museum of Natural History,.

Hoffman, R.L. and J.A. Payne. 1969. Diplopods as . Molecular Ecology Resources 50, no. 6: 1096-1098.

Hopkin, S.P. and H.J. Read. 1992. The Biology of Millipedes. 1st ed. Oxford, England: Oxford University Press.

Houck, M.A. and B.M. OConnor. 1991. Ecological and evolutionary significance of phoresy in the Astigmata. Annual Review of Entomology 36: 611-636.

Hunter, P.E. 1993. A new family of mites, Costacaridae (Mesostigmata: Trigynaspida: Celaenopsoidea), associated with millipedes in Mexico. Israel Journal of Zoology 39, no. 2: 185-191.

Hunter, P.E. and R.M.T. Rosario. 1986. A new species of Julolaelaps Berlese (Acari: Laelapidae). International Journal of Acarology 12, no. 2: 63-67.

Ishikawa, K. 1986. Gamasid mites (Acarina) associated with Japanese millipeds. Reports

88 of Research Matsuyama Shinonome Jr. College 17: 165-177.

———. 1988. Two new species of sSissurolaelaps (Acarina, Laelapidae) associated with Philippine millipeds. Bulletin of the National Science Museum, Tokyo, Serie A 14, no. 4: 173-178.

Kaneko, N. 1999. Effect of millipede Parafontaria tonominea Attems (Diplopoda: ) adults on soil biological activities: A microcosm experiment. Ecological Research 14: 271-279.

Kethley, J.B. 1973. Developmental chaetotaxy of a paedomorphic celaenopsoid, Neotenogynium malkini n. g., n. sp. (Acari: : Neotenogyniidae, n. fam.) associated with millipedes. Annals of the Entomological Society of America 67, no. 4: 571-579.

———. 1978. Narceolaelaps n. g., (Acari: Laelapidae) with four new species parasitizing spirobolid millipedes. International Journal of Acarology 4, no. 3: 195-210.

———. 1983. The deutonymph of Epiphis rarior Berlese, 1916 (Epiphidinae n. subfam., Rhodacaridae, Rhodacaroidea). Canadian Journal of Zoology 61, no. 11: 2598-2611.

Kheirallah, A.M. 1978. The consumption and utilization of two different species of leaf litter by a laboratory population of gracilis (Diplopoda: Polydesmoidea). Entomologia Experimentalis Et Applicata 23: 14-19.

———. 1979. Behavioral preference of Julus scandinavicus (Myriapoda) to different species of litter. Oikos 33: 466-471.

Kim, C. and H. Klompen. 2002. A new genus and species of Paramegistidae (Mesostigmata: Trigynaspida) associated with millipedes from Mexico. Acarologia, Paris 42, no. 1: 39-52.

Klompen, H. and C.C. Austin. 2007. A new species of Ophiomegistus Banks (Acari: Paramegistidae) from papua New Guinea. Zootaxa 1387: 47-57.

Köhler, H.R. and G. Alberti. 1990. Morphology of the mandibles of the millipedes (Diplopoda, Arthropoda). Zoologica Scripta 19: 195-202.

Kontschan, J. 2005. Two species of Julolaelaps Berlese, 1916 (Acari: Mesostigmata: Laelapidae) associated with millipedes from Kenya. Annales Historico-Naturales Musei Nationalis Hungarici 97: 257-260.

Lawrence, J., M. and M.J. Samways. 2003. Litter breakdown by the Seychelles giant millipede and the conservation of soil processes on Cousine Island, Seychelles. Biological Conservation 113: 125-132.

89 Lawerence, R.F. 1939a. A new mite attached to the sex organs of South African millipedes. Transactions of the Royal Society of South Africa 27: 226-231.

———. 1939b. Notes on the habits of the two mites living on South African millipeds. Transactions of the Royal Society of South Africa 27: 233-239.

Lindquist, E.E. 1975. Associations between mites and other arthropods in forest floor habitats. The Canadian Entomologist 107: 425-437.

Lindroth, C.H. 1957. The Faunal Connections between Europe and North America. Stockholm: Almqvist & Wiksell.

Lodha, B.C. 1974. Decomposition of Digested Litter. In Biology of Decomposition, edited by C. H. Dickinson and G. J. F. Pugh. Vol. 1, 213-241. New York: Academic Press.

Lyford, W.H. 1943. Palatability of freshly fallen leaves of forest trees to millipedes. Ecology 24: 252-261.

Madge, D.S. 1969. Litter disappearance in forest and savannah. Pedobiologia 9: 288-299.

Maes, K. 1983. Scientific reports of the Belgian Mount Cameroon expedition 1981. VIII. description of four new species of the genus Julolaelaps (Acarina: Laelapidae). Revue De Zoologie Africaines 97, no. 1: 211-220.

Mahunka, S. 1962. Neue beiträge zur kenntnis der einheimischen acariden-fauna (Acari). Opuscula Zoologica (Budapest) 4, no. 2-4: 97-103.

———. 1992. New and interesting mites from the geneva museum. XXXVII. new mites from the Pacific islands (Acari: Anoetidae, Scutacaridae). Revue Suisse De Zoologie 99, no. 2 : 395-403.

Marcuzzi, G. 1970. Experimental observations in the role of Glomeris spp. (Myriapoda Diplopoda) in the process of humification of litter. Pedobiologia 10: 401-406.

Marek, P. and Shelley, R.M. Expert Burrowers.Myriapoda.org. http://www.myriapoda.org/diplopoda/millipede_general_info.html (accessed March 18, 2010).

Margolis, L., G.W. Esch, J.C. Holmes, A.M. Kuris, and G.A. Schad. 1982. The use of ecological terms in parasitology (report of an ad hoc committee of The American Society of Parasitologists). Journal of Parasitology 68, no. 1: 131-133.

McBrayer, J.F. 1973. Exploitation of deciduous leaf litter by Apheloria montana (Diplopoda: Eurydesmidae). Pedobiologia 13: 90-98.

90 McBrayer, J.F. and D.E. Reichle. 1971. Structure and feeding rates of forest soil invertebrate populations. Oikos 22, no. 3: 381-388.

Neuhauser, E.F. and R. Hartenstein. 1978. Pheolic content and palatability of leaves and wood to soil isopods and diplopods. Pedobiologia 18: 99-109.

Nicholson, P.B., K.L. Bocock, and O.W. Heal. 1966. Studies on the decomposition of the faecal pellets of a millipede (Glomeris marginata (Villers)). Journal of Ecology 54, no. 3 Nov.: 755-766.

Nielsen, C.O. 1962. Carbohydrases in soil and litter invertebrates. Oikos 13, no. II: 200- 215.

Nuñez, F.S. and C.S. Crawford. 1976. Digestive enzymes of the desert millipedes (Girard) (Diplopoda: ). Comparative Biochemistry and Physiology 55A: 141-145.

OConnor, B.M. 1982. Evolutionary ecology of astigmatid mites. Annual Review of Entomology 27: 385-409.

———. 1994. Life History Modifications in Astigmatid Mites. In Mites: Ecological and Evolutionary Analyses of Life-History Patterns, edited by Marilyn A. Houck, 136- 159. New York: Chapman & Hall.

———. 2009. Cohort Astigmata. In A Manual of Acarology, edited by Gerald W. Krantz and David E. Walter. 3rd ed., 565-658. Lubbock, Texas: Texas Tech University Press.

———. Keys to the Selected Genera of Astigmatid Mites. Acarology Summer Program 2009. Unpublished.

Olynyk, J.E. and R. Freitag. 1979. Some phoretic associations of ground beetles (Coleoptera: Carabidae) and mites (Acarina). The Canadian Entomologist 111: 333- 335.

Oudemans, A.C. 1915. Notizen über Acari. XXII. Archiv Für Naturgeschicthe 81: 131- 134.

Pherson, D.A. and A.J. Beattie. 1979. Fungal loads of invertebrates in leaf litter. Revue d Ecologie Et De Biologie Du Sol 16: 325-335.

Rawlins, A., J, I.D. Bull, N. Poirier, P. Ineson, and R.P. Evershed. 2006. The biochemical transformation of oak (Quercus robur) leaf litter consumed by the (Glomeris marginata). Soil Biology and Biochemistry 38: 1063-1076.

Reichle, D.E. 1968. Relation of body size to food intake, oxygen consumption, and trace

91 element metabolism in forest floor arthropods. Ecology 49, no. 3: 538-542.

———. 1977. The Role of Invertebrates in Nutrient Cycling. Uppsala, Sweden, Swedish Natural Science Research Council, .

Reichle, D.E., M.H. Shanks, and D.A. Crossley Jr. 1969. Calcium, potassium, and sodium content of forest floor arthropods. Annals of the Entomological Society of America 62, no. 1: 57-62.

Romell, L. 1935. An example of myriapods as mull reformers. Ecology 16, no. 1: 67-71.

Rossi, W. and A. Weir. 1998. Triainomyces, a new genus of Laboulbeniales on the pill- millipede tuberculatum from New Zealand. Mycologia 90, no. 2: 282- 289.

Ryke, P.A.J. 1959. A revision of the hypoaspid mites associated with Myriapoda with descriptions of three new species of the subgenus Julolaelaps Berl. (Acarina: Laelaptidae). Parasitology 49, no. 1-2: 6-22.

Saito, S. 1968. Productivity of high and low density populations of Japonaria laminata armigers (Diplopoda) in a warm-temperate forest ecosystem. Researches on Population Ecology 9: 153-164.

Sakwa, W.N. 1974. A consideration of the chemical basis of food preference in millipedes. Symposia of the Zoological Society of London 32: 329-346.

Schmidt-Rhaesa, A., M.A. Farfan, and E.C. Bernard. 2009. First record of millipedes as hosts for horsehair worms (Nematomorpha) in North America. Northeastern Naturalist 16, no. 1: 125-130.

Seastedt, T.R. and C.M. Tate. 1981. Decomposition rates and nutrient contents of arthropod remains in forest litter. Ecology 62: 13-19.

Shaw, G.G. 1968. Population size, ecology and mineral reservoir of the millipede, Narceus annularis (RAF.). Ecology 49, no. 6: 1163-1166.

Shear, W.A. Key to Milliped Families of the Eastern United States. unpublished key, Hampden-Sydney, VA.

Shelley, R.M. 1978. Millipeds of the eastern Piedmont region of North Carolina, U. S. A. (Diplopoda). Journal of Natural History 12: 37-79.

———. 1984. A synopsis of the millipede genus Abacion Rafinesque (Callipodida: Caspiopetalidae). Canadian Journal of Zoology 62: 980-988.

———. 1988. The millipeds of eastern Canada (Arthropoda: Diplopoda). Canadian

92 Journal of Zoology 66: 1638-1663.

———. 2002. The millipeds of central Canada (Arthropoda: Diplopoda), with reviews of the Canadian fauna and diplopod faunistic studies. Canadian Journal of Zoology 80: 1863-1875.

Sibly, R.M. 1981. Strategies of Digestion and Defecation. In Physiological Ecology. an Evolutionary Approach to Resource use, edited by C. R. Townsend and P. Calow, 109-139. Oxford: Blackwell Scientific Publications.

Smit, A.-. and R.J. Van Aarde. 2001. The influence of millipedes on selected soil elements: A microcosm study on three species occurring on coastal sand dunes. Functional Ecology 15: 51-59.

Snider, R.M. 1984. The ecology of Polydesmus inconstans (Diplopoda: Polydesmidae) in Michigan wood lots. Pedobiologia 26: 185-196.

Southcott, R.V. 1987. A new larval mite (Acarina: Trombidioidea) ectoparasitic on an Australian , and the reclassified. Transactions of the Royal Society of 3, no. 1: 43-52.

Southwood, T.R.E. .1973. The insect/plant Relationship - an Evolutionary Perspective. In Insect/Plant Relationships, edited by H. F. Van Emden. Vol. 6, 3-30. London: Halsted Press.

Striganova, B.R. 1969a. Assessment of ability of different species of millipedes (Diplopoda) to assimilate leaf litter. Zoologicheskii Zhurnal 48, no. 6: 821-826.

———. 1969b. Distribution of diplopods in mixed forests of north Caucasus and their role in the decomposition of forest litter. Zoologicheskii Zhurnal 48, no. 11: 1623- 1628.

———. 1970a. Decomposition of cellulose in gut of Pachyiulus foetidissimus (Mur.) (Juloidea, Diplopoda). Doklady Akademii Nauk SSSR 190, no. 3: 703-705.

———. 1970b. Significance of Diplopod Activity in Leaf Litter Decomposition. Paris.

———. 1972a. A Comparative Account of the Activity of Different Groups of Soil Invertebrates in the Decomposition of Forest Litter. Kazan.

———. 1972b. Effect of temperature on the feeding activity of Sarmatiulus kessleri (Diplopoda). Oikos 23: 197-199.

Striganova, B.R. and N.P. Chernobrovkina. 1992. The effect of Diplopoda on the dynamics of amino acids in the soil. Berichte Des Naturwissenschaftlich- Medizinischen Vereins in Innsbruck Suppl. 10: 297-303.

93 Striganova, B.R. and Z.G. Prishutova. 1990. Food requirements of diplopods in the dry steppe subzone of the USSR. Pedobiologia 34: 37-41.

Swafford, L. and J.E. Bond. 2009. The symbiotic mites of some Appalachian Xystodesmidae (Diplopoda: Polydesmida) and the complete mitochondrial genome sequence of the mite Stylochyrus rarior (Berlese) (Acari: Mesostigmata: ). Invertebrate Systematics 23, no. 5: 445-451.

Szabo, I.M., E.G.A. Nasser, B.R. Striganova, Y.R. Rakhmo, K. Jäger, and M. Heydrich. 1992. Interactions among millipedes (Diplopoda) and their intestinal bacteria. Berichte Des Naturwissenschaftlich-Medizinischen Vereins in Innsbruck Suppl. 10: 289-296.

Szabo, I.M., H. Prauser, G. Bodnar, C.T. Loc, K. Ravasz, E.A. Hossein, and K. Maria- Ligeti. 1990. The indigenous intestinal bacteria of soil arthropods and worms. In Microbiology in Poecilotherms, edited by R. Lesel, 109-118. Amsterdam: Elesevier.

Taylor, E.C. 1982a. Fungal preference by a desert millipede Orthoporus ornatus (Spirostreptidae). Pedobiologia 23: 331-336.

———. 1982b. Role of aerobic microbial populations in cellulose digestion by desert millipedes. Applied and Environmental Microbiology 44, no. 2: 281-291.

Taylor, E.C. and C.S. Crawford. 1982. Microbial gut symbionts and desert detritivores. Scientific Reviews on Arid Zone Research 1: 37-52.

Trägårdh, I. 1906. Neue acariden aus natal und Zululand. Zoologischer Anzeiger 30, no. 26: 870-877.

———. 1907. Description of two myriopodophilous genera of Antennophorinae, with notes on their development and biology. Arkiv För Zoologi 3, no. 28: 1-35.

———. 1911. Discomegistus, a new genus of myriopodophilous from trinidad, with notes on the Heterozerconinae. Arkiv För Zoologi 7, no. 12: 1-21.

———. 1950. Studies on the , Diplogyniidae and Schziogyniidae. Arkiv För Zoologi 1, no. 25 : 361-449.

Uppstrom, K. and H. Klompen. 2005. A new species of Julolaelaps (Acari: Iphiopsididae) from African millipedes. International Journal of Acarology 31, no. 2: 143-147.

Van der Drift, J. 1962. The disappearance of litter in mull and mor in connection with weather conditions and the activity of the macrofauna. Oosterbeek, Netherlands.

Visser, S. 1985. Role of the soil invertebrates in determining the composition of soil

94 microbial communities. In Ecological Interactions in Soil: Plants, Microbes and Animals, edited by A. H. Fitter, D. Atkinson, D. J. Read and M. B. Usher. Vol. 4, 297-317. Oxford; Boston: Blackwell Scientific.

Vitzthum, H.G. 1921. Acarologische beobachtungen. Archiv Für Naturgeschicthe 86: 14- 22.

———. 1925. Fauna sumatrensis (Beitrag no. 5) Acarinae. Supplementa Entomologica 11: 37-48.

Walter, D.E. and G.W. Krantz. 2009. Oviposition and Life Stages. In A Manual of Acarology, edited by Gerald W. Krantz and David E. Walter. 3rd ed., 57-63. Lubbock, Texas: Texas Tech University Press.

Webb, D.P. 1977. Regulation of deciduous Forest Litter Decomposition by Soil Arthropod Feces. In The Role of Arthropods in Forest Ecosystems, edited by W. J. Mattson, 57-69. United States of America: Springer-Verlag.

Webster, J. 1970. Coprophilous fungi. Transactions of the British Mycological Society 54: 161-180.

Whitkamp, M. and D.A. Crossley Jr. 1966. The role of arthropods and microflora in the breakdown of white oak litter. Pedobiologia 6: 293-303.

Womersley, H. 1956. On some new Acarina-Mesostigmata from Australia, New Zealand and New Guinea. Linnean Society's Journal-Zoology 42, no. No. 288: 505-599.

———. 1958. Some Acarina from Australia and New Guinea paraphagic upon millipedes and cockroaches, and on beetles of the family Passalidae. Pt 2. the family Diplogyniidae (Mesostigmata-Trigynaspida). Transactions of the Royal Society of South Australia 81: 13-29.

Wooten, R.C. and C.S. Crawford. 1975. Food, ingestion rates and assimilation in the desert millipede Orthoporus ornatus (Girard) (Diplopoda). Oecologia 20: 231-236.

95 Appendix A: Accession Codes for Millipede Vouchers

96 Millipede species Accession number

Abacion lactarium N/A

Blaniulus guttulatus OSAL0100704 ♂

Brachyiulus lusitanus OSAL0100959 ♂ *OSAL006951 ♂

Brachyiulus pusillus OSAL0100593 ♂

Choneiulus palmatus OSAL0100811 ♂

Cleidogona caroliniana OSAL0100283 ♂

Cylindroiulus britannicus OSAL0100455 ♀ OSAL0100394 ♂ *OSAL006949 ♀ *OSAL006950 ♂

Cylindroiulus caeruleocinctus OSAL0100759 ♂ *OSAL006948 ♂

Cylindroiulus latestriatus OSAL0100650 ♂ *OSAL006947 ♂

Cylindroiulus londinensis OSAL0100375 ♂ *OSAL006955 ♂

Cylindroiulus punctatus OSAL0100417 ♂

Cylindroiulus truncorum OSAL0100935 ♀ OSAL0100536 ♂ *OSAL006952 ♂

Euryurus leachii OSAL0100840 ♂

Glomeris marginata OSAL0100376 ♂

Julus scandinavicus OSAL0100358 ♂

Nopoiulus kochii OSAL0100282 ♂

Ommatoiulus sabulosus OSAL0100403 ♂ * denotes slide of genitalia

97 Millipede species Accession number Ophyiulus pilosus OSAL0100423 ♀ OSAL0100613 ♂ *OSAL006954 ♀ *OSAL006953 ♂

Paraspirobolus lucifugus OSAL0100386 ♂

Polydesmus angustus OSAL0100439 ♂

Tachypodoiulus niger OSAL0100384 ♂

Uroblaniulus carolinensis OSAL0100882 ♂ * denotes slide of genitalia

98 Appendix B: Accession Codes for Mite Vouchers

99 Mite species Accession number

Cosmolaelaps sp. OSAL006960

Histiostoma sp. OSAL006789

Holostapis sp. OSAL006958

Phaulodinychus sp. OSAL006790

Rhizoglyphus A OSAL0083451

Rhizoglyphus B OSAL006896

Sancassania A OSAL006911

Sancassania B OSAL0083471

Schwiebea A OSAL006979

Schwiebea C OSAL006942

Schwiebea D OSAL0083456

Schwiebea E OSAL0083474

Schwiebea F OSAL006853

Schwiebea G OSAL0083448

Schwiebea H OSAL006939

Thyreophagus sp. OSAL006843

100