This is a repository copy of Prevalence, transmission and intensity of infection by a microsporidian sex ratio distorter in natural Gammarus duebeni populations.
White Rose Research Online URL for this paper: http://eprints.whiterose.ac.uk/1007/
Article: Dunn, A.M. and Hatcher, M.J. (1997) Prevalence, transmission and intensity of infection by a microsporidian sex ratio distorter in natural Gammarus duebeni populations. Parasitology, 114 (3). pp. 231-236. ISSN 1469-8161 https://doi.org/10.1017/S0031182096008475
Reuse See Attached
Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [email protected] including the URL of the record and the reason for the withdrawal request.
[email protected] https://eprints.whiterose.ac.uk/ 231 Prevalence, transmission and intensity of infection by a microsporidian sex ratio distorter in natural Gammarus duebeni populations
A. M. DUNN* and M.J.HATCHER Department of Biology, University of Leeds, Leeds LS2 9JT, UK
(Received 13 May 1996; revised 16 August 1996; accepted 16 August 1996) This is a report of the prevalence, transmission and intensity of infection of a microsporidian sex ratio distorter in natural populations of its crustacean host Gammarus duebeni. Prevalence in the adult host population reflects differences in the intensity of infection in transovarially infected embryos and in adult gonadal tissue. The efficiency of transovarial parasite transmission to young also differs between populations, but this alone is insufficient to explain observed patterns of prevalence. Infection intensity may be important in determining future infection of target tissue in the adult and subsequent transmission to future host generations. We consider patterns of parasite infection in terms of selection on transmission and virulence.
Key words: transovarial transmission, parasite burden, microsporidian, sex ratio distorter.
distorter described as Octosporea effeminans (Bulnheim & Vavra, 1968). The microsporidian is Parasitic sex ratio distorters have important impli- transovarially transmitted in the cytoplasm of the cations for the ecology and evolution of their host ova and is not passed on through males. There is no populations. They have the potential to drive the evidence of horizontal transmission (Bulnheim, evolution of the host population sex ratio (Werren, 1978; Dunn et al. 1993). The parasite feminizes its 1987; Taylor, 1990; Hatcher & Dunn, 1995), impose host by converting putative males into functional selective pressures on host sex determination (Bull, females capable of transmitting the parasite to future 1983; Rigaud, Mocquard & Juchault, 1992; generations (Smith & Dunn, 1991; Dunn et al. Juchault, Rigaud & Mocquard, 1992; Hatcher & 1993). Hence, parasite-induced feminization boosts Dunn, 1995; Dunn et al. 1995) and may affect host the effective transmission base of the parasite, population size, stability and extinction (Werren, enabling parasite invasion and maintenance at higher 1987; Werren & Beukeboom, 1993). In this paper we prevalences in host populations (Hatcher & Dunn, examine prevalence, transmission and intensity of 1995). infection by a microsporidian sex ratio distorter Theoretical analyses (Werren, 1987; Taylor, 1990; infecting the crustacean Gammarus duebeni. Hatcher & Dunn, 1995) have shown that invasion Vertically transmitted parasites are transmitted and equilibrium prevalence of parasitic sex ratio from parent to offspring via the host’s gametes. If distorters in host populations depend upon the these parasites are uniparentally inherited, selection efficiency of parasite transmission, the efficiency of favours parasites which bias the host sex ratio parasite-induced feminization of the host, the rela- towards the transmitting sex (Lewis, 1941; Howard, tive fitness of infected hosts and the underlying host 1942; Hamilton, 1967). Vertically transmitted micro- population sex ratio. Although parasite prevalence sporidia have been found to distort host sex ratio has been shown to be sensitive to parasite trans- through killing male hosts which releases spores for mission efficiency (Werren, 1987; Hatcher & Dunn, horizontal transmission (Kellen & Wills, 1962; 1995), parasite burden and its relationship with Kellen et al. 1965; Hazard & Weiser, 1968) and transmission and prevalence has not been explicitly through feminization in which genetic males are considered. Here we examine parasite prevalence, converted into phenotypic females, so directly in- parasite transmission efficiency and parasite burden creasing the relative frequency of the transmitting in 3 G. duebeni field populations. sex (Smith & Dunn, 1991; Dunn, Adams & Smith, 1993). G. duebeni is host to a microsporidian sex ratio During the breeding season of 1995 we collected * Corresponding author. Tel: j0113 2332856. Fax: random samples of adult animals from 3 Gammarus j0113 2441175. E-mail: a.dunn!leeds.ac.uk. duebeni field populations: Budle Bay, Northumber-
Parasitology (1997), 114, 231–236 Copyright # 1997 Cambridge University Press A. M. Dunn and M. J. Hatcher 232 land, UK, Totton Marsh, Hampshire UK, and assessed by looking at the reduction in deviance Douarnenez, Finestere, France. Parasite prevalence caused by deletion of a term from the maximal and the intensity of infection in adult females were model. Data for parasite load and for clutch size estimated by light microscopy. Individuals were are count data, therefore we specified a Poisson fixed in 10% formalin, dehydrated and embedded in error distribution. We corrected for overdispersion # paraffin wax. Previous studies indicated that the with a heterogeneity factor (Hf l Pearson’s χ \..; microsporidian is restricted to the gonadal tissue of Crawley, 1993). Changes in deviance caused by infected females where spore (Dunn et al. 1993) and removing a factor from the model were compared vegetative stages (Dunn, unpublished observations) with χ# tables. Data for parasite transmission have been observed. Therefore, serial transverse efficiency were proportion data (number of infected sections through the thorax (the site of the gonad) eggs\total number of eggs) and were analysed were stained with Giemsa’s stain and examined for specifying a binomial error structure, taking the total the microsporidian and measurements taken of the number of eggs as the binomial denominator. A number of parasites infecting the mature oocytes of heterogeneity factor was used to correct for over- infected females. dispersion and changes in deviance caused by The efficiency of transovarial parasite transmission removing a factor from the model were assessed to young was estimated by examining early stage using an F-test. embryos produced by infected mothers for the presence or absence of infection. Parasite burden was measured by counting total parasite load in these embryos. To determine the infection status of Parasite prevalence females from the field it is necessary to kill them. Parasite prevalence differed significantly between Therefore, we collected embryos from a random the field sites (χ#,2..l28n74, P ! 0n001). The sample of infected and uninfected females from each highest level of infection was at Budle Bay site and screened the mother and her embryos for where 30% of females were infected (N l 70), at infection after breeding. G. duebeni form precopula Douarnenez 15n6% (Nl64) were infected and the pairs a few days before they mate. Pairs of animals lowest level of infection, 6n3% (Nl127), was found were set up in individual containers in brackish at Totton Marsh (Table 1). water (specific gravity 1005m, corresponding to field salinity) at 12 mC and examined daily. Females mate and lay their eggs when they moult, approximately Parasite burden in adults every 3 weeks. Eggs are laid into a marsupium Parasite burden in the oocytes of adult females also formed from a series of overlapping plates differed between the 3 field sites and the pattern of (oostegites) on the ventral surface of the thorax and parasite load was in accord with differences in are brooded for 3–4 weeks. Within 24 h of prevalence: mean parasite load per oocyte was fertilization, early-stage embryos (1–128 cells) were highest at Budle Bay, and lowest at Totton Marsh flushed from the marsupium using a syringe filled (Table 1). From Fig. 1, in infected Totton Marsh with brackish water. Clutch size and blotted wet females most (80%) oocytes had a very low parasite weight of the female were recorded. Embryos were burden (0–4 parasites) and the maximum burden freeze fractured, fixed and stained with DAPI (4,6- recorded was 52 parasites. In contrast, 64% of diamidino-2-phenyl-indole), a fluorescent dye for oocytes in infected Budle Bay females had a burden DNA, and examined using a Zeiss Axiovert 10 of more than 10 parasites with 12% having a burden fluorescent microscope. This enabled us to see host in excess of 90. nuclei and the nuclei of parasites lying in the Parasite load differed significantly between the cytoplasm of the embryo cells (Dunn et al. 1995). sites (χ#,2..l45n2, P ! 0n001). In addition, Under fluorescence microscopy, a parasite nucleus within a population, parasite burden differed can sometime be resolved into 2 nuclei, although it is between families. Taking the brood of an individual not possible to determine whether this is a diplo- mother to represent a family we found that incor- karyon or a dividing nucleus. However, the increase poration into the model of a (site.mother) inter- in parasite number during progressive stages of host action term caused a significant reduction in the embryogenesis (Dunn et al. 1995) suggests that these deviance of the data from the model (χ#,30..l are vegetative stages of the microspordian. Each 57, P ! 0n05). embryo was scored for the presence or absence of the parasite. Parasite burden was measured by counting Parasite transmission to eggs total parasite load in embryos at the 64 and 128 cell stage of development. The efficiency of transovarial parasite transmission Data were analysed using the generalized linear to eggs also differed between sites (F(#,%!) l 3n95, P ! modelling package GLIM (GLIM 3.77, Numerical 0n05). However, there did not appear to be a simple Algorithms Group, Oxford 1985). Significance was relationship between the efficiency of parasite trans- Infection patterns of a microsporidian 233
Table 1. Prevalence, intensity of infection and the efficiency of transovarial transmission of the microsporidian sex ratio distorter in Gammarus duebeni from 3 field sites: Budle Bay (BB), Totton Marsh (TM) and Douarnenez (Dou) (Parasite prevalence: percentage of adult females infected in the population. Parasite burden: mean number of parasites in mature oocyte pstandard error; mean parasite number in 64 cell embryo p..; mean parasite number in 128 cell embryo p.. Transmission efficiency: mean proportion of infected eggs in the brood of infected mothers p..)
Burden in Burden in Prevalence Burden in 64 cell 128 cell Transmission Site (%) oocytes embryos embryos efficiency
BB 30n024n4p6n268n3p5n655n9p12n80n95p0n033 (N l 70) (N l 29) (N l 88) (N l 6) (N l 161) Dou 15n68n0p1n735n8p4n90n77p0n072 (N l 64) (N l 49) (N l 14) (N l 180) TM 6n34n5p1n011n8p3n20n87p0n097 (N l 127) (N l 74) (N l 8) (N l 92)
infection was highest at Budle Bay (where prevalence is highest), an intermediate proportion was recorded at Totton Marsh, with Douarnenez females showing the lowest transmission efficiency (Table 1).
Parasite burden in embryos The pattern of parasite burden in 64 cell-stage embryos from Budle Bay and Douarnenez is illus- trated in Fig. 2 and Table 1. Low parasite prevalence precluded collection of sufficient data from Totton Marsh embryos to include in the analysis. From Fig. 2, 71% of Douarnenez embryos had a parasite Fig. 1. Frequency distribution for parasite burden in burden of less than 40 and parasite burdens ranged mature oocytes from female Gammarus duebeni infected from 14 to 72. In contrast, the variance was greater with the microsporidian sex ratio distorter. BB, Budle in Budle Bay embryos (Budle Bay mean burden l Bay, Northumberland, UK; Dou, Douarnenez, n l l Finestere, France; TM, Totton Marsh, Hampshire, 68 3, variance 2732, Douarnenez mean burden UK. 35n8, variance l 335) and parasite burden ranged from 1 to 240. Parasite burden was significantly higher in em- bryos from Budle Bay than in Douarnenez embryos (χ#,1.l 7n6, P ! 0n01). This is in accord with the pattern of prevalence seen in the field. There was also a significant difference between families within the two sites: incorporation in the model of a (site.mother) interaction term caused a significant reduction in the deviance of the data from the model (χ#,12..l31n9, P ! 0n01); it appears that parasite load is higher in some host families than others. We also collected data from a small number of 128 cell embryos from Budle Bay and Totton Marsh Fig. 2. Frequency distribution for parasite burden in (Table 1). Parasite burden differed significantly # embryos infected with the microsporidian sex ratio between-sites (χ ,1..l34, P ! 0n001). However, distorter. Data for 64 cell stage embryos from 2 field we treat these data with caution due to the problem sites: Budle Bay (BB) and Douarnenez (Dou). of small sample size. In addition, all 8 Totton Marsh embryos were from the same brood so it is not mission from mother to eggs, and parasite prevalence possible to dissociate inter-family differences from in the population. Although the proportion of the between-site differences in the host–parasite re- young of infected mothers which inherited the lationship. A. M. Dunn and M. J. Hatcher 234
The data also suggest that the parasite is not Undeen, 1990), mathematical models predict that, as replicating during this period of host development a result of its feminizing effect, the microsporidian (assuming no parasite death). From Table 1, it may be maintained in host populations through appears that mean parasite numbers in Budle Bay vertical transmission alone (Hatcher & Dunn, 1995). embryos decrease from the 64 to 128 cell stage. This study indicates that the parasite burden and However, variances are high and the sample size for the efficiency of transovarial parasite transmission 128 cell embryos is small. There is, in fact, no are important factors underlying the level of in- significant difference in the parasite burden in 64 and fection in natural host populations. Parasite burden 128 cell embryos (χ#,1.l0n41, P " 0n05). A also varies between host families within a site. The previous study of parasitism in Budle Bay embryos pattern of prevalence in the field reflects the intensity also found that parasite replication appeared to be of infection in oocytes of adults and in developing limited during early host embryogenesis (Dunn et al. embryos: parasite burden is highest at Budle Bay 1995). where prevalence is also high. Previous studies of vertically transmitted microsporidia record similar parasite loads per ovum (Kellen & Wills, 1962) and Fitness effects of parasitism also demonstrate a relationship between intensity of Female weight differed significantly between the infection in the adult female and the efficiency of 3 field sites (mean weight mgp..: Budle Bay vertical transmission (Milner & Lutton, 1980; 22n98p0n83; Douarnenez 19n96p0n58; Totton Canning et al. 1985). Parasite transmission efficiency is also highest at Marsh 26n01p0n80; F(#,"#*) l 38n2, P ! 0n01). This may reflect differences in temperature or food Budle Bay, although there is no simple relationship availability at the different field sites, both of which between parasite prevalence and transovarial trans- affect growth. However, parasite burden did not mission efficiency. Studies of other transovarially transmitted microsporidia reveal a range of trans- significantly affect female weight (F(","#*) l 0n23, P " 0n05). In addition, there was no significant effect mission efficiencies: e.g. transmission of Amblyo- on weight caused by an interaction between site and spora spp. to 90% of the adult mosquito hosts (Andreadis & Hall, 1979); transmission of Perezia parasite burden (F($,"#') l 0n52, P " 0n05). Hence, there is no evidence that parasite burden affects pyraustae to 50% of the eggs of the European weight at any site. cornborer host (Kramer, 1959); transmission of Clutch size was significantly related to the weight Nosema plodiae to 12% of eggs of the Indian meal of the mother (χ#,1..l67n4, P ! 0n001; larger moth host (Kellen & Lindegren, 1973). Kellen & females produced more eggs and this is in accord Lindegren (1973) also noted considerable interfamily with previous studies (Kolding & Fenchel, 1981; variation in the rate of Nosema transmission to host McCabe & Dunn, 1994; Dunn & McCabe, 1995). eggs. However, these microsporidia have both Taking female weight as a covariate, we found no vertical and horizontal transmission routes (Kramer, significant effect of parasite burden on clutch size 1959; Kellen & Lindegren, 1973; Sweeney et al. (χ#,1.l0n23, P " 0n05). Clutch size did not differ 1985) and Amblyospora has an indirect life-cycle between the sites (χ#,2..l4n77, P " 0n05). Simi- which involves 2 host species (Sweeney et al. 1985). larly, there is no significant effect on clutch size of Parasite spread and maintenance in the host popu- the interaction between site and parasite burden (χ#, lation is, therefore, less dependent on transovarial 3 .. l 4n4, P " 0n05). These data are in accord with transmission for these microsporidia than it is for a previous study which found no parasite-induced O. effeminans. reduction in host fecundity (Dunn et al. 1993). The pattern of parasite burden is recognized as an important factor for the dynamics of horizontally transmitted parasites (Anderson & May, 1978; May & Anderson, 1978; Anderson, 1991). However, parasite burden has not been explicitly considered in Although the transmission route for most micro- many studies of vertically transmitted parasites sporidia is horizontal there are a number of micro- (Fine, 1975; Taylor, 1990; Werren, 1987; Hatcher & sporidia for which vertical transmission is also an Dunn, 1995). A mismatch in previous studies of the important transmission mechanism (Chapman et al. G. duebeni system between estimates of transmission 1966; Canning, 1982) and this may be the sole from adult to adult host (69% in Dunn et al. 1993) transmission route for the sex ratio distorter infecting and estimates of transmission from adult to egg G. duebeni (Smith & Dunn, 1991; Dunn et al. 1993). (96% in Dunn et al. 1995) may be accounted for by Although we cannot discount the possibility of a relationship between parasite burden, growth rate horizontal transmission of the microsporidian in G. and infection in the developing host. Low parasite duebeni, either directly or through some intermediate numbers in oocytes may be sufficient to ensure host (e.g. Chapman et al. 1966; Andreadis & Hall, transmission to the embryos of the next host 1979; Sweeney, Hazard & Graham, 1985; Avery & generation but continued transovarial transmission Infection patterns of a microsporidian 235 though generations of hosts is dependent on the host susceptibility or resistance to parasite trans- presence of the parasite in the gonadal tissue of the mission and growth. The differences may reflect breeding female (parasites in other tissues are not clonal differences between parasite strains in the transmitted; Dunn et al. (1995); Hatcher, Tofts & different populations or may result from the impact Dunn (1996)). Initial parasite load and parasite of abiotic factors on parasite development (Bulnheim growth may dictate transmission to the target 1978). In further studies we will investigate the gonadal tissue and feminization of the host, factors importance of these factors for parasite–host dy- which have an impact upon parasite prevalence at the namics. population level. The precise mechanisms of parasite development and transmission within individual We thank English Nature and the Hampshire and Isle of hosts are currently under investigation. White Naturalists’ Trust for granting permission to sample Parasite numbers will represent a trade-off be- animals; Sheree Olbison and David Blakeley for technical assistance; Moira Dunn and Mark Scully for assistance in tween opposing selective forces. On the one hand, the field; an anonymous referee for helpful comments on selection will favour a high burden, which will the manuscript. This study was supported by NERC maximize the chances of transmission from adult to (GT5\F\92\ALS\1; GST\02\688) and the Leverhulme offspring and of infection of gonadal tissue in the Trust (F\122\AK). new host (Dunn et al. 1995; Hatcher et al. 1996). On the other hand, vertically transmitted parasites are dependent on host reproduction for their own transmission and so selection may favour a parasite , . . (1991). Populations and infectious diseases: ecology or epidemiology? Journal of Animal which minimizes the metabolic burden imposed on Ecology 60, 1–50. the host (Ewald, 1987; Bull & Rice, 1991; Bull, , . , . . (1978). Regulation and Molineux & Rice, 1991; Smith & Dunn, 1991). Any stability of host-parasite population interactions I. negative effects on host fitness (reduced reproductive Regulatory processes. Journal of Animal Ecology 47, success of infected hosts) must be traded-off against 219–247. the consequent loss of potential targets for infection. , . . , . . (1979). Significance of Furthermore, parasite transmission strategy is transovarial infections of Amblyospora sp. likely to reflect a trade-off between initial load in the (Microspora: Thelohaniidae) in relation to parasite egg and replication rate within the embryo, if low maintenance in the mosquito Culex salinarius. Journal initial parasite load is associated with increased host of Invertebrate Pathology 34, 152–157. survival (Dunn et al. 1995). Although a reduction in , . . , . . (1990). Horizontal transmission of Parathelohania anophelis to the egg production, egg hatch (Andreadis & Hall, 1979; copepod, Microcyclops varicans, and the mosquito, Hembree & Ryan, 1982) and subsequent survival Anopheles quadrimaculatus. Journal of Invertebrate (Kramer, 1959; Raina et al. 1995) is commonly Pathology 56, 98–105. associated with transovarially transmitted micro- , . . (1983). The Evolution of Sex Determining sporidia, we found no evidence of a relationship Mechanisms. Benjamin Cummings, California. between parasite burden and fitness in G. duebeni. , . . (1994). Virulence. Evolution 48, 1423–1437. Parasite burden had no effect on clutch size or on , . . , . . (1991). Distinguishing adult female weight. However, other fitness para- mechanisms for the evolution of co-operation. Journal meters which may be affected by parasite burden of Theoretical Biology 149, 63–74. (e.g. growth rate, onset of sexual maturity, lifetime , . ., , . . , . . (1991). Selection reproductive output) were not measured. A previous of benevolence in a host-parasite system. Evolution 45, 875–882. study revealed no reduction in survival or growth , . (1978). Interaction between genetic, rate of infected young (Dunn et al. 1993) but external and parasitic factors in sex determination of infection status rather than precise burden was the crustacean amphipood Gammarus duebeni. considered. An increased understanding of trans- Helgolander wissenschaftliches Meeresunters 31, 1–33. missibility and virulence in terms of within host , . . , . (1968). Infection by the processes of infection and parasite growth has been microsporidia Octosporea effeminans sp. n., and its sex identified as important for determining population- determining influence in the amphipod Gammarus level outcomes for the host (Bull, 1994). Future duebeni. Journal of Parasitology 54, 241–248. studies will investigate in more detail the relationship , . . (1982). An evaluation of protozoal between parasite burden and host fitness in the G. characteristics in relation to biological control of pests. duebeni system, and the implications for parasite– Parasitology 84, 119–149. , . ., , . ., , . . , . host dynamics. . (1985). Transmission of microsporidia, especially We found between population and between family Orthosoma operophterae (Canning 1960) between differences in the parasite–host relationship. generations of winter moth Operophtera brumata (L) Differences in the pattern of infection by the (Lepidoptera: Geometridae). Parasitology 90, 11–19. microsporidian sex ratio distorter in natural G. , . ., , . ., , . . , duebeni populations may result from differences in . . (1966). Host–parasite relationships of Thelohania A. M. Dunn and M. J. Hatcher 236
associated with mosquitoes in Louisiana , . . , . . (1973). Transovarial (Nosematidae; Microsporidia). Journal of Invertebrate transmission of Nosema plodiae in the Indian-meal Pathology 8, 452–456. moth, Plodia interpunctella. Journal of Invertebrate , . . (1993). GLIM for Ecologists. Blackwell Pathology 21, 248–254. Scientific Publications, Oxford. , . . , . (1962). The transovarial , . ., , . , . . (1993). Transovarial transmission of Thelohania californica Kellen and Lipa transmission and sex ratio distortion by a in Culex tarsalis Coquillett. Journal of Insect Pathology microsporidian parasite in a shrimp. Journal of 4, 321–326. Invertebrate Pathology 61, 248–252. , . , . . (1981). Patterns of , . ., , . ., , . . , . reproduction in different populations of five species of (1995). Evolutionary ecology of vertically transmitted the amphipod genus Gammarus. Oikos 37, 167–172. parasites: strategies of transovarial transmission of a , . . (1959). Some relationships between Perezia microsporidian sex ratio distorter in Gammarus pyraustae Paillot (Sporozoa, Nosematidae) and duebeni. Parasitology 111, S91-S110. Pyrausta nubilalis (Hubner) (Lepidoptera, Pyralidae). , . . (1995). Resource allocation to Journal of Insect Pathology 1, 25–33. young: seasonal patterns within and between , . (1941). Male sterility in natural populations of Gammarus duebeni populations. Oikos 73, 199–202. hermaphrodite plants. New Phytologist 40, 50–63. , . . (1987). Transmission modes and the , . . , . . (1978). Regulation and evolution of the parasite-mutualism continuum. stability of host-parasite population interactions II. Annals of the New York Academy of Science 503, Destabilising processes. Journal of Animal Ecology 47, 295–306. 249–287. , . . . (1975). Vectors and vertical transmission; , . , . . (1994). Is resource partitioning an epidemiological perspective. Annals of the New a response to brood sex ratio in an amphipod with York Academy of Science 266, 173–194. environmental sex determination? Oikos 69, 203–206. , . . (1967). Extraordinary sex ratios. , . . , . . (1980). Interactions Science 156, 477–488. between Oncopera alboguttata (Lepidoptera: , . . , . . (1995). Evolutionary Hepialidae) and its microsporidian pathogen, consequences of sex ratio distortion by cytoplasmically Pleistophora oncoperae (Protozoa: Microsporida). inherited feminizing factors. Philosophical Transactions Journal of Invertebrate Pathology 36, 198–202. of the Royal Society, London B 348, 445–456. , . ., , ., , . ., , . . (1995). , . ., , . , . . (1996). The Transovarial transmission of Nosema locustae effect of the embryonic bottleneck on vertically (Microsporida: Nosematidae) in the migratory locust transmitted parasites. Selected Proceedings of the 1st Locusta migratoria migratorioides. Parasitology Conference on Information Processing in Cells (in the 81, 38–44. Press). Research , . . , . (1968). Spores of Thelohania in , ., , . . , . (1992). The adult female Anopheles: development and transovarial spread of parasitic sex factors in populations of transmission, and redescriptions of T. legeri Hesse and Armadillidium vulgare Latr. (Crustacea, Oniscidea): T. obea Kudo. Journal of Protozoology 15, 817–823. effects on sex ratio. Genetics, Selection, Evolution 24, , . . , . . (1982). Observations on the 3–18. vertical tranmsission of a new microsporidian , . . , . . (1991). Transovarial pathogen of Aedes aegypti from Thailand. Mosquito transmission. Parasitology Today 7, 146–148. News 42, 49–54. , . ., , . . , . . (1985). , . . (1942). The genetics of Armadillidium Intermediate host for an Amblyospora sp. vulgare Latreille. II. Studies on the inheritance of (Microspora) infecting the mosquito, Culex monogeny and amphogeny. Journal of Genetics 44, annulirostris. Journal of Invertebrate Pathology 46, 143–159. 98–102. , ., , . , . . (1992). , . . (1990). Evolutionary consequences of Evolution of sex-determining mechanisms in a wild cytoplasmic sex ratio distorters. Evolutionary Ecology population of Armadillidium vulgare Ltr. (Crustacea, 4, 235–248. Isopoda): competition between two feminizing , . . (1987). The coevolution of autosomal and parasitic sex factors. Heredity 69, 382–390. cytoplasmic sex ratio factors. Journal of Theoretical , . ., , . ., , . . Biology 124, 313–334. , . . (1965). Host–parasite relationship of , . . , . . (1993). Population some Thelohania from mosquitoes (Nosematidae: genetics of a parasitic chromosome: theoretical Microsporidia). Journal of Invertebrate Pathology 7, analysis of PSR in subdivided populations. American 161–166. Naturalist 142, 224–241.