Ectoparasitic effects on host survival and reproduction: the Drosophila-Macrocheles association.
Parasites are ubiquitous in natural communities (Price 1980). By impairing host survival and reproduction, many parasitic species can affect disparate levels of biological organization. These levels range from the genetic and demographic structure of local host populations (Anderson and Crombie 1985, Crump and Pounds 1985, Hamilton et al. 1990) to the composition of entire ecological communities (Minchella and Scott 1991).
During the course of infection, parasites assimilate host nutrients that would otherwise remain available to the host. But parasites may also impair host feeding and assimilation efficiency (Holmes and Zohar 1990). As a consequence, many cases of parasitism lead to disturbed host physiological functions (Thompson 1983), physical emaciation, and elevated mortality (Holmes and Zohar 1990). Parasites may also utilize nutrients otherwise destined for host egg production (Hurd 1990, 1993), or indirectly cause diversion of nutrients from oogenesis by altering fat body metabolism and/or perturbing neuroendocrine control mechanisms (Holmes and Zohar 1990). Parasite-mediated fecundity depletion has been demonstrated in Argentine stem weevils, Listronotus bonariensis (Malone 1987), pyralid moths, Sceliodes cordalis (Mercer and Wigley 1987), and western fence lizards, Sceloporus occidentalis (Schall 1983) infected with protozoans, and in Drosophila putrida (Jaenike 1992) and Tribolium confusum (Keymer 1980) infected with nematodes and cestodes, respectively.
A growing body of the parasitological literature focuses on mechanisms by which microparasites (bacteria, viruses, protozoa, and fungi) reduce host fitness, primarily because of their role in regulating natural populations of many animals, including humans (Park 1948, Anderson and May 1981, May 1983, Ewald 1994). Baculovirus infections, for example, have been implicated in major oscillations in abundance of some temperate forest Lepidoptera (May 1983). Macroparasites, in contrast, which include parasitic helminths and arthropods, are not typically implicated in dramatic oscillations in host abundance. Nevertheless, their effects on host fitness can be pronounced, and some species can reduce host numbers and rates of exponential increase (Lanciani 1975, May and Anderson 1979, Keymer 1981, Kaya and Gaugler 1993). However, relative to parasitic helminths, less is known about how and to what extent parasitic arthropods, such as mites, can influence host fitness and population dynamics.
The present paper examines effects of infestation by mites Macrocheles subbadius (Berlese) (Macrochelidae: Mesostigmata) on multiple fitness components in Drosophila nigrospiracula Patterson and Wheeler (Drosophilidae: Diptera). Macrochelid mites have previously been believed to associate only phoretically with flies, i.e., to assume an entirely passive role while attached to their "host" (see Study organisms). Attachment is viewed merely as a means of facilitating mite dispersal to fresher substrates better suited for feeding and reproduction. In the present study, however, I demonstrate that M. subbadius is actually ectoparasitic, and examine dose-dependent effects of mites on the period prior to onset of oviposition, lifetime productivity, and survivorship of adult flies. I also address the potential of mites to affect fly populations in nature by drawing upon the results of a previous study by Polak and Markow (1995). This study demonstrated that the distribution of mites can be strongly aggregated in natural host populations, and that the degree of aggregation is positively correlated with a highly variable ecological parameter, age of the cactus necrosis.
Macrocheles mites occur worldwide. They feed and oviposit in a wide spectrum of substrates ranging from rotting plant tissue and moist soil to animal dung. In these materials, they primarily consume bacteriophagic nematodes (Rodriguez et al. 1962) and small arthropods, including the first and second instar larvae of flies (Pereira and de Castro 1945). Axtel (1963a, b) has suggested that by consuming the immature stages of flies, M. muscaedomesticae exerts a controlling effect on some house fly (Musca domestica) populations in nature. Nothing is known, however, about the capacity of these mites to reduce fly numbers in nature by feeding directly on adult hosts.
Macrochelid mites have been recovered from the body surface of adult flies, as well as from coprophagous scarab beetles and even from some rodents (reviewed by Krantz 1983). M. subbadius has been recovered from the house fly and the stable fly (Stomoxys calcitrans) (Axtell 1964), D. mettleri (M. Polak, unpublished data), D. mojavensis (T. A. Markow, personal communication), D. pachea (S. Pitnick, personal communication), and the cactus fly Odontoloxozus longicornis (Neriidae) (M. Polak, unpublished data). Female M. subbadius, in both adult and deutonymph stages of development, attach to adult Drosophila nigrospiracula (M. Polak, unpublished data). This fruit fly is endemic to the Sonoran desert of North America (Heed 1978). Courtship and mating occur on the outer skin of necrotic saguaro cacti, Carnegiea gigantea, and larvae develop within the decaying tissue (Markow 1988). Approximately 95% of attached mites occur on the ventral surface of the fly abdomen near the junction with the thorax. Mites occasionally cling to the dorsal surface of the abdomen, as well as to the neck and face of flies (Polak 1993). Under laboratory conditions, mites can remain attached for the entire life of the host. In nature, mean duration of infestation is unknown. Wade and Rodriguez (1961) speculated that M. muscaedomesticae can ". . . suck the body fluid . . ." of adult house flies, but they failed to substantiate their claim. In fact, Kinn (1966) found evidence to the contrary: M. muscaedomesticae failed to take up Nile Blue stain that had been injected into adult hosts of these mites.
Laboratory cultures of flies and mites
Drosophila nigrospiracula stocks were initiated from field-caught flies netted at saguaro cacti located [approximately equal to]72 km east of Phoenix, Arizona, USA. Stock flies were maintained in mass culture for no longer than six generations to avoid inbreeding. Stocks were cultured in banana-agar medium with live yeast and autoclaved necrotic saguaro cactus. The agar medium contained fresh bananas, white corn syrup, and Premier Malt extract. Stocks were maintained at room temperature at [approximately equal to]10L: 14D photoperiod. Experimental flies were collected on the day they eclosed and stored in uncrowded vials lined with banana-agar medium in an incubator with a 12L: 12D photoperiod and a 26 [degrees] C day and 22 [degrees] night temperature cycle. Unless otherwise stated, flies were provided with Fleischmann's dry live yeast.
A laboratory culture of M. subbadius was initiated from several hundred mites removed with forceps from field-caught adult D. nigrospiracula. Mites were maintained in 0.5-L glass jars containing a rich organic medium: wheat bran, alfalfa flakes, yeast, double-distilled water, and a bacteriophagic nematode, Rhitis inermiformis (Royce and Krantz 1991).
Adult flies were experimentally infested with M. sub-badius as follows. Approximately 10 g of the wheat bran-alfalfa from culture jars containing mites was transferred to plaster-of-Paris/charcoal-lined 200-mL glass bottles containing cardboard squares embedded into the plaster. Thirty flies were introduced into each bottle, left overnight, and then recovered for use in experiments.
Are mites ectoparasites of adult flies?
I performed a laboratory experiment to track radio-labelled (14C) amino acids from labeled adult Drosophila to mites. Larvae were cultured in medium containing a 14C L-amino acid mixture (ICN Lot Number 10147) and labeled females were harvested. Mites were experimentally attached to labeled flies as described above. After 24 h at 25 [degrees] C, mites were removed with fine forceps, the gnathosoma removed, and washed in phosphate-buffered saline. Control mites were extracted from culture jars. Mites from both groups were processed for scintillation counting (Pitnick et al. 1991).
I tested the efficacy of the washing procedure to remove radiolabel contamination from the surface of mites. Six mites were immersed into a puddle of haemolymph released from a radiolabeled fly onto a glass microscope slide. When the haemolymph was fully dry, three mites were subjected to the above washing procedure. The other three were processed for scintillation without washing.
Ectoparasitism and wet mass of flies collected in the field
I tested whether the amount of material ingested by mites represents a significant proportion of fly body mass. Flies were captured with an insect net at necrotic cacti on nine occasions between 4 June 1993 and 18 June 1994. A sample of infested and uninfested flies were individually aspirated into vials lined with banana-agar medium. In the laboratory on the same day of collection ([approximately equal to]1.5 h post-collection), sex, thorax length (estimate of body size), and mite load for each fly were recorded. Mites were removed from infested flies, and wet masses of previously infested and uninfested flies were obtained using a Cahn C-31 microbalance (Cerritos, California, USA) to the nearest 1 [[micro]gram]. To determine the relationship between mite load and wet mass of flies, I constructed a multiple regression model for each sex separately in which the dependent variable was wet mass, and thorax length and mite load were entered as covariates.
Ectoparasitism and fly longevity and oviposition latency
I held newly eclosed females for 48 h in banana-agar vials. Experimental females were then placed into infestation chambers with mites for another 48 h, and then placed separately into vials with a single male. Dead males throughout were replaced with live individuals. Control females consisted of two types, which, as revealed by later analysis, did not differ with respect to either longevity or oviposition latency (see Results). "Resistant" individuals were exposed to mites, but not infested by mites, within infestation chambers. "Unexposed" females were not exposed to mites and consisted of a random sample of flies collected from stock bottles and held in vials while others were in the infestation chambers. Experimental and control females were held individually in vials within an incubator starting at day 5 post-eclosion. Vials were checked every 12 h for dead females as well as for any eggs that had been laid. Dead females were removed from vials and their thoraces were measured using an ocular micrometer. Longevity was taken as the time between the day of eclosion and time of death. Oviposition latency was calculated as the time between eclosion and when eggs were first noticed in the vial. Longevity data were [log.sub.10](y + 1) transformed to meet assumptions of parametric analysis. Analysis of covariance (ANCOVA) was performed on longevity, with thorax length entered as covariate. Preliminary analysis indicated that the "TRT x throrax length" interaction was not significant (P = 0.95), thereby confirming the equal slopes assumption of ANCOVA (Neter et al. 1990). ANCOVAs, ANOVAs, and regression analyses throughout were performed with the general linear model (GLM) and regression (REG) procedures of SAS (1989).
Ectoparasitism and female egg load in the field
To test whether mites influence female fecundity, I randomly collected females with an insect net at a rotting saguaro on 16 April 1993. Females were individually aspirated from the net, placed into a banana-agar vial, and immediately killed with ether. Upon return to the laboratory, vials containing females were frozen. For each thawed female I later counted the number of attached mites that she carried and measured her thorax length using an ocular micrometer of a dissecting microscope. I then dissected her abdomen and counted all mature, or nearly mature eggs (stages 11-14, Mahowald and Kambysellis 1980) present within ovaries. To assess the effect of female size and mite load on fecundity, I performed multiple regression analyses in which egg load was the dependent variable.
I replicated this work [approximately equal to]1 and 12 mo later at different sites, but employed a modified procedure. Each female, upon dissection, was checked for sperm within her uterus and seminal receptacle. Only inseminated females were included in subsequent regression analyses.
Ectoparasitism and female fecundity in the laboratory
Female flies were collected from stock bottles on the day of eclosion and held with males for 3 d and then infested following procedures described above. On day 4, females were placed separately with two males into 30-mL vials lined with banana-agar medium and containing either necrotic saguaro only, or necrotic saguaro supplemented with ad libitum live yeast. Fifteen infested and 15 uninfested females were assigned to each of the two diets (N = 60 females). Uninfested females, which served as the control group, included flies that had been exposed to mites in chambers but had evaded infestation. Starting on day 5, vials were checked daily for eggs and for dead females. Live females were transferred to new vials. Number of eggs present within each vial from which each female was removed was recorded. Thorax length and day of death were recorded for each female. Subsequent ANCOVAs were performed on female fecundity with thorax length and longevity as covariates.
Ectoparasitic effects on post-infestation productivity
It is possible that mites physically interfere with oviposition behavior, thereby causing reductions in female fecundity. To test this hypothesis, I contrasted productivity between uninfested females and females immediately after their mites had been removed (i.e., post-infestation). I used two age categories of flies to test for an interaction between degree of infestation and maternal age. Flies in the "Old" category were collected on the day they eclosed and stored in banana vials with yeast in groups of 15 (with 15 males). Flies were changed to fresh vials every 5 d. Thirty days after "Old" females were collected, the "Young" group was collected and held for 2 d. Females were then experimentally infested with either 0, 1 or [greater than] 1 mite and held with males for 4 d in vials with no yeast. Mites were removed with fine tweezers while flies were anesthetized using C[O.sub.2]. Control flies (i.e., with 0 mites) were also anesthetized. Groups of two females of the same age and mite load category were placed with three males in 30-mL vials containing 2.0 g powdered mashed potatoes (Betty Crocker Potato Spuds), 9.5-mL double-distilled [H.sub.2]O, and 4-5 mg active dry yeast sprinkled on the surface. After females were allowed to oviposit for 48 h in the incubator, parentals were removed from vials and total number of progeny that emerged from each vial was recorded. Ample pupation sites were available.
Consumption of fly haemolymph by mites
Most mites used their mouthparts to attach to the ventral surface of their host's abdomen [ILLUSTRATION FOR FIGURE 1 OMITTED]. Mite chelicerae are large and possess distinct serrated edges (see Appendix), which they utilize to open fly eggs and larvae. These structures are also effective in piercing adult abdominal cuticle, as evidenced by oozing and encrusted haemolymph at sites of mite attachment on adult Drosophila and cactus flies (Odontoloxozus longicornis: Neriidae) (N = 3) in the field.
I first examined whether the washing procedure could thoroughly eliminate fly-derived contamination from the surface of mites that had attached to radio-labeled flies. Mites bathed in radiolabeled haemolymph, but subsequently washed, emitted background levels of radioactivity, indicating that washing was entirely successful [ILLUSTRATION FOR FIGURE 2a OMITTED].
Mites that had infested radiolabeled flies had radioactivity levels more than six times as high as control mites [ILLUSTRATION FOR FIGURE 2b OMITTED]. There was also a significant positive correlation between the disintegrations emitted collectively by groups of mites and the number within each group attached to a labeled fly (r = 0.89, N = 14, P [less than] 0.0005). This relationship reaches an asymptote [ILLUSTRATION FOR FIGURE 3 OMITTED], suggesting either that competition for limited haemolymph occurs between mites attached to a fly, or that feeding efficiency by mites decreases as a function of mechanical interference. Thus, M. subbadius is a haematophagous ectoparasite of adult D. nigrospiracula.
Ectoparasitism and wet mass of flies collected in the field
Mean mite loads for males collected in 1993 and 1994 in the following analysis (mean [+ or -] 1 SE) were 1.06 [+ or -] 0.22 mites (range = 0-5 mites, N = 45 males) and 0.94 [+ or -] 0.24 mites (range = 0-4 mites, N = 51 males), respectively. For females in turn, these values were 1.05 [+ or -] 0.16 mites (range = 0-5 mites, N = 46 females) and 0.87 [+ or -] 0.19 mites (range = 0-4 mites, N = 60 females).
Abdomens of both male and female flies that carried heavy mite loads were visibly flaccid, suggesting that mites ingest considerable quantities of haemolymph while attached to their host. Multiple regression analysis (with thorax length and mite load entered as covariates) revealed a significant negative effect of mite load on the wet mass of adult flies of both sexes (Table 1). After controlling for the strong positive effect of body size on wet mass of adult flies, mite load explained from 19 to 30% of the variation in adult wet mass (see coefficient of partial determination [COPD] values in Table 1). COPD values measure the marginal contribution of one independent variable, given that the other(s) are already entered into the regression model (Neter et al. 1990).
[TABULAR DATA FOR TABLE 2 OMITTED]
Ectoparasitism and fly longevity in the laboratory
ANCOVA, with thorax length entered as a covariate, revealed significant differences in longevity across "resistant," unexposed and infested females ([F.sub.2,64] = 19.63; P [less than] 0.0001). Multiple contrast analysis (using Tukey's HSD method) revealed that mean longevity did not differ between "resistant" (mean [+ or -] 1 SE, 24.4 [+ or -] 1.62 d, N = 20) and unexposed females (29.3 [+ or -] 1.92 d, N = 20 females; P = 0.25). This nonsignificant result indicates that both groups of flies can serve as controls. In contrast, longevity of infested females (14.6 [+ or -] 1.62 d, N = 28) was significantly less than that of both control groups (P [less than] 0.001). Infested females harbored from 1-9 [Mathematical Expression Omitted] mites in this experiment.
The overall multiple regression model, with mite load and thorax length as covariates, explained a significant amount of the variation in female longevity ([r.sup.2] = 0.46, [F.sub.2,65] = 28.03, P [less than] 0.0001) and revealed a significant effect of both mite load and thorax length on this dependent variable (Table 2). The effect of mite load, however, was more pronounced: mite load and thorax length, respectively, explained 44 and 9% of the variation in the dependent variable. In a separate regression analysis, the Mite x Thorax interaction was found to be nonsignificant (t = 0.004, P = 0.28).
Oviposition latency in laboratory flies
As for longevity, oviposition latency (time between eclosion and onset of oviposition, mean [+ or -] 1 SE) did not differ between "resistant" (142.3 [+ or -] 2.74 h, N = 28 females) and unexposed females (145.2 [+ or -] 3.83 h, N = 30 females; two-sample t = 0.61, df = 56, two-tailed P = 0.54). Thus, since both groups can serve as controls, data for these groups were pooled. Mean oviposition [TABULAR DATA FOR TABLE 1 OMITTED] [TABULAR DATA FOR TABLE 3 OMITTED] latency of the pooled sample of females (143.8 [+ or -] 2.37 h, N = 58 females) was significantly less than that for infested females (153.6 [+ or -] 3.21 h, N = 20 females; two-sample t = 2.20; df = 76, P = 0.031). Among infested females, the correlation calculated between mite load and oviposition latency was not significant (r = 0.34, P [greater than] 0.05, N = 20 females), nor was there a significant correlation between female thorax length and oviposition latency (r = 0.12, P [greater than] 0.6, N = 19 females).
Ectoparasitism and female egg load in the field
In each of the three samples of flies collected at different saguaro cacti, differences in mite load explained a significant amount of the variation in ovarian eggs carried by females (Table 3). Analysis of the April 1993 sample, however, did not distinguish between fertilized and unfertilized females (see Methods). Consequently, the significant negative relation between mite and egg load for this sample could have been the result of elevated susceptibility of immature females (e.g., teneral individuals). In an effort to eliminate this possibility, the May 1993 and 1994 samples comprised exclusively inseminated females. Female D. nigrospiracula become sexually receptive 3 d post-eclosion (Markow 1988). Results of these analyses did not differ from those of the April 1993 sample (Table 3).
The relation between female thorax length and fecundity was positive and significant only for the sample collected during April-May 1994 (Table 3); in this period, mite load and thorax length explained 38 and 33% of the variation in egg load, respectively. For each sample, the "Mite x Thorax" interaction was not significant (all P values [greater than] 0.1).
Ectoparasitism and female fecundity in the laboratory
To examine experimentally the effect of infestation on net egg output of females maintained in the laboratory, I contrasted number of eggs laid by infested and uninfested females over their lifetimes under two diet regimes. Mean mite load did not differ significantly between females maintained on the cactus ([Mathematical Expression Omitted] mites, range = 1-3 mites, N = 15 females) vs. the supplemented diet ([Mathematical Expression Omitted] mites, range = 1-3 mites, N = 15 females; two-sample t = 0.72, P = 0.47).
The magnitude of the effect of mites on net egg output differed between the two diet treatments. On the cactus diet, females that were experimentally infested laid significantly fewer eggs (mean [+ or -] 1 SE, 49.9 [+ or -] 14.1 eggs, range = 0-138 eggs, N = 15 females) than un-infested females (153.5 [+ or -] 31.9 eggs, range = 0-438 eggs, N = 15 females; two-sample t = 2.97, P = 0.0061). But this effect was not significant among females that had ad libitum access to yeast (infested: 633.3 [+ or -] 97.7 eggs, range = 0-1252 eggs, N = 15 females; uninfested: 590.9 [+ or -] 86.4 eggs, range = 79-1112 eggs, N = 15 females; t = 0.32, P = 0.75)
The significant effect of mites on egg output revealed in the above analysis appears to be mediated through mite effects on female longevity. Recall that I have already demonstrated a significant effect of mite load on longevity and that control females exposed to mites (but uninfested) do not to differ in longevity relative to unexposed individuals. From ANCOVA on egg output of females on the cactus diet, with body size as the only covariate, there was a highly significant effect of mite treatment (TRT, i.e., presence or absence of mites) ([F.sub.1,22] = 27.63, P = 0.0001, model [r.sup.2] = 0.64). Adding longevity as another covariate improved the overall model ([r.sup.2] = 0.81), but the significant effect of mite treatment was lost. Longevity remained the sole significant predictor of lifetime egg output (Table 4).
In contrast, among females maintained on the yeast-supplemented diet, mite load had no significant effect on egg output with body size entered as the covariate ([F.sub.1,27] = 1.79, P = 0.19, model [r.sup.2] = 0.35). Furthermore, in the expanded model with both body size and longevity entered, body size was the sole significant predictor of egg output (Table 4). These results indicate that flies maintained on the supplemented diet overcame the debilitating effect of infestation on longevity, which caused reductions in fecundity on the restricted, cactus diet.
Ectoparasitic effects on post-infestation productivity
Productivity in this experiment was measured as the total number of offspring produced by pairs of females. In the two-way analysis of variance, both age and mite burden had a significant effect on productivity, and the effect of age was more pronounced (Table 5). However, the Age x Mite interaction was not significant. Nevertheless, multiple contrasts revealed that mite effects were significant only among "Old" females [ILLUSTRATION FOR FIGURE 4 OMITTED]. These results indicate that debilitating effects of mites on this component of fly fitness do occur post-infestation. Mechanical constraints on oviposition imposed by the mere physical presence of mites are therefore unlikely to be the major cause of reduced fecundity of infested females demonstrated in Ectoparasitism and female fecundity in the laboratory.
Table 4. Results of ANCOVA showing the effect of infestation treatment (TRT, i.e., infested or not), body size, and longevity on lifetime productivity of females under two diet regimes in the laboratory.
Diet Source df MS(*) F P
Cactus only TRT 1 4.93 1.90 0.18 Body size 1 7.99 3.09 0.093 Longevity 1 48.25 18.64 0.0001 Error 21 2.59 Model [r.sup.2] 0.81
Yeast supplement TRT 1 148.45 1.67 0.21 Body size 1 916.66 10.31 0.0035 Longevity 1 4.84 0.05 0.82 Error 26 88.95 Model [r.sup.2] 0.35
* Type II mean squares. Mean square values have been multiplied by [10.sup.-3].
Results of the present study dispel the belief that macrochelid mites form only phoretic associations with adult flies. Experiments with radiolabeled hosts showed that mites ingest haemolymph from the abdominal cavity of adult flies of both sexes. In addition, analyses of field-caught flies burdened with 0-6 mites revealed a highly significant negative correlation between mite load and wet mass of flies. Males (N = 5) carrying two mites sampled in 1994 from a single necrosis, for example, were 13% lighter than uninfested males of equal body size (N = 7). These findings indicate that haemolymph extraction by the present species of mite results in a measurable reduction in host haemolymph volume.
The composition of insect haemolymph represents a steady state between flux of water, oxygen, salts, and organic nutrients such as amino acids, protein and carbohydrates (Roeder 1953, Demerec 1965). Consequently, disturbances to haemolymph volume by mites could alter fly water balance and pressure gradients essential for blood circulation, as well as reduce net nutrient availability to metabolic processes (Roeder 1953). Other parasites of insects, especially helminths, are known to cause significant changes in blood constituents in several host species (Gordon et al. 1978, Thompson 1983, Hurd 1990, 1993). But helminths appear to be much more specialized in their feeding habit than M. subbadius mites, and almost certainly possess a higher degree of engagement with the metabolic and endocrinological pathways of their host (see, e.g., Gordon and Webster 1971, Rutherford and Webster 1978, Gordon 1981, Poinar 1983, Thompson 1983). Nutrient depletion by mites probably reduces physiological competence of flies and impairs their performance in a variety of energetically demanding tasks, especially those such as courtship, mating, and flight (Roeder 1953). In fact, Polak and Markow (1995) have demonstrated a pronounced effect of infestation on mating success of flies of both sexes. Not surprisingly, this effect was more pronounced in males, probably because males must work closer to their physiological limit relative to females during courtship (Polak and Markow 1995).
Table 5. Results of two-way ANOVA showing the effect of age and degree of infestation on number of progeny produced by post-infestation females.
Source df MS(*) F P
Age 1 26.08 18.11 0.0001 Mites 2 7.46 5.11 0.0071 Age x Mites 2 2.29 1.59 0.21 Error 105 1.44 [r.sup.2] 0.23
* Type I mean squares. Mean square values have been multiplied by [10.sup.-3].
Loss of nutrients stored in depots such as the fat body and flight muscles are frequently also reported consequences of parasitism (Porter 1970, Reader 1971, Thompson 1983, Hurd 1990). If reductions in stored energy reserves occur in infested D. nigrospiracula, resultant impairment of nutrient mobilization could restrict performance in high-endurance tasks such as long-distance dispersal, which is dependent on the amount of glycogen held in reserve (Roeder 1953). Effective dispersal by D. nigrospiracula and other desert flies is required for tracking habitable cacti throughout their arid environment. Available sites are especially scarce during summer months (Breitmeyer 1994), when flies are routinely challenged with flights of [greater than] 1 km/d (Johnston and Heed 1976). Smith (1988) also suggested that insect flight activity could be impeded by actual damage to flight musculature from feeding tube formation by Arrenurus water mites.
Parasitic effects on individual reproductive output
My experiments demonstrated that as mite load increased, females suffered longer delays prior to oviposition, laid and carried fewer eggs, produced fewer offspring, and lived shorter lives. A plausible explanation for these fitness consequences is nutrient deprivation resulting from host haemolymph extraction by mites. Evidence supporting this contention is twofold. First, although infestation was asymptomatic for egg production among well-fed females, the same mite loads caused significant effects on a restricted diet. Second, previously infested females whose mites were experimentally removed produced fewer offspring than uninfested females. Thus, the effect of ectoparasitism is similar to that of starvation (see also Becker 1980).
In comparing fecundity of infested and uninfested (but exposed) laboratory females, I have assumed that females were equal in their condition (e.g., energy budget) prior to being experimentally infested. To help meet this assumption, I maintained all females prior to infestation in uncrowded vials under similar food, temperature, and light conditions. Otherwise, reduced fecundity of infested females could have resulted if generally less fit females were also more susceptible. Admittedly, such a possibility cannot be excluded as a cause of my field results. On the other hand, if "resistance" and fitness were negatively correlated (consult Simms 1992 for examples), my results are conservative.
In the field study examining host fecundity, I was able to exclude the possibility that the negative relationship between mite and natural egg loads was due simply to higher susceptibility of immature females (recently eclosed, teneral individuals). I did this by including into separate analyses only reproductively mature females (identified by the presence of sperm within their uterus and/or seminal receptacle). Therefore, oocyte resorption by infested females and/or reduced sequestration of protein by the ovaries probably contributed to this negative relationship. Diversion of nutrients away from oogenesis could compensate for a reduction in overall levels of haemolymph nutrients (Gordon et al. 1973, Hurd and Arme 1986), and serve to increase the probability of surviving an episode of infestation and of dispersal (Polak and Markow 1995).
In contrast to my field data, results of laboratory work on fecundity do not appear to reflect processes relating to ovarian provisioning. Instead, my results indicate that differences in reproductive output resulted from mite-induced effects on survivorship. The relatively favorable conditions experienced in the laboratory may have had a partial compensatory effect on female nutritional stress, particularly since movements of flies were enormously restricted in their 30-mL holding vials. In nature, however, the combination of a limited diet, greater activity levels, and fluctuating environmental conditions could magnify the effect of infestation on female reproductive physiology. Therefore, reduced survivorship and oocyte provisioning may simultaneously act to reduce female reproductive output in nature.
Reduced host survivorship has been reported in other dipterans, including mosquitoes, Anopheles crucians (Lanciani and Boyt 1977), and ceratopogonid midges, Dasyhelea mutabilis (Lanciani 1986), parasitized by larval water mites. However, Forbes and Baker (1990) failed to find a significant relationship between degree of Arrenurus mite infestation and longevity of adult damselflies, Enallagma ebrium, under laboratory conditions. In my laboratory study, uninfested and large flies lived longer than infested and smaller individuals, respectively. However, these flies were maintained in small vials under controlled environmental conditions and they enjoyed continuous access to a sugar food source (banana-agar medium, see Methods). In nature, the weakening effects of mites could be greater due to adverse environmental conditions. David Lack (1954) emphasized an interaction between bad weather conditions, food shortage, and parasitism in generating significant density-dependent mortality in natural populations.
Parasitic effects on host populations
Intensities of infestation used in my laboratory experiments were comparable to those encountered in nature. Thus the observed fitness consequences of infestation imply that action of mites can reduce fly numbers in nature (Anderson 1979). Polak and Markow (1995) have demonstrated that both the intensity and prevalence (fraction of flies infested) of parasitism by mites increases as a function of rot age. For example, at a fallen cactus monitored over a period of 33 d, intensity rose from 0 to 1.25 mites per fly. At another old rot, however, intensity rose to 7.8 mites per fly and prevalence neared 100%. At the monitored rot, degree of aggregation of mite numbers equalled 0 (random distribution, where variance = mean) when the rot was young, but aggregation (degree of clumping) increased rapidly and linearly as a function of rot age (Polak and Markow 1995). Aggregated distributions of parasites are expected to attenuate any regulatory effects since most parasites will be harbored by a relatively small fraction of the host population (Anderson 1980, Jaenike and Anderson 1992). However, the sharp rise in prevalence as well as intensity of mite infestation with rot age indicates that the regulatory effects of mites on flies could be pronounced.
The aggregated distribution of mites together with dose-dependent effects of host fitness, also suggests that mites are capable of density-dependent regulation of their own numbers. This regulation would probably be mediated primarily through effects on fly mortality and dispersal capacity; flies with the heaviest mite burdens would be least successful at colonizing fresh substrates. Therefore, mites clinging to unsuccessful flies would be lost from the overall parasite population. This dampening effect on mite numbers has important implications for the long-term coexistence of this association (Anderson and May 1978), as it probably contributes to the stability of the Drosophila-Macrocheles dynamic.
I thank John Alcock, Elizabeth Davidson, Therese Markow, and Ronald Rutowski for discussion and comments on the manuscript. I also thank K. Bolte, Agriculture Canada, for the electron micrograph, G. W. Krantz for identifying mite mouthparts, and B. Terkanian for the illustration of the infested fly. Financial support for research was provided by the DuPont Company (fellowship to the author), National Science Foundation grant BSR 89-19362 to T. A. Markow, and the Department of Zoology at Arizona State University.
Anderson, R. M. 1979. Parasite pathogenicity and the depression of host population equilibria. Nature 279:150-152.
-----. 1980. Depression of host population abundance by direct life cycle macroparasites. Journal of Theoretical Biology 82:283-311.
Anderson, R. M., and J. A. Crombie. 1985. Experimental studies of age-intensity and age-prevalence profiles of infection: Schistosoma mansoni in snails and mice. Pages 111-145 in D. Rollinson and R. M. Anderson, editors. Ecology and genetics of host-parasite interactions. Academic Press, New York, New York, USA.
Anderson, R. M., and R. M. May. 1978. Regulation and stability of host-parasite population interactions. I. Regulatory processes. Journal of Animal Ecology 47:219-247.
Anderson, R. M., and R. M. May. 1981. The population dynamics of microparasites and their invertebrate hosts. Philosophical Transactions of the Royal Society of London B 291:451-524.
Axtell, R. C. 1963a. Manure-inhabiting Macrochelidae (Acarina: Mesostigmata) predaceous on the house fly. Advances in Acarology 1:55-59.
-----. 1963b. Effect of Macrochelidae (Acarina: Mesostigmata) on house fly production from dairy cattle manure. Journal of Economic Entomology 56:317-321.
-----. 1964. Phoretic relationship of some common manure-inhabiting Macrochelidae (Acarina: Mesostigmata) to the house fly. Annals of the Entomological Society of America 57:584-587.
Becker, W. 1980. Metabolic interrelationship between parasitic nematodes and molluscs, especially Schistosoma mansoni in Biomphalaria glabrata. Zeitschrift for Parasitenkunde 63:101-111.
Breitmeyer, C. M. 1994. Factors influencing population size in Sonoran desert Drosophila. Thesis. Arizona State University, Tempe, Arizona, USA.
Crump, M. L., and J. A. Pounds. 1985. Lethal parasitism of an aposematic anuran (Atelopus virius) by Notochaeta bufonivora (Diptera: Sarcophagidae). Journal of Parasitology 71:588-591.
Demerec, M. 1965. Biology of Drosophila. Hafner, New York, New York, USA.
Ewald, P. W. 1994. Evolution of infectious disease. Oxford University Press, Oxford, UK.
Forbes, M. R. L., and R. L. Baker. 1990. Susceptibility to parasitism: experiments with the damselfly Enallagma ebrium (Odonata: Coenagrionidae) and larval water mites, Arrenurus spp. (Acari: Arrenuridae). Oikos 58:61-66.
Gordon, R. 1981. Mermithid nematodes: physiological relationships with their insect hosts. Journal of Nematology 13:266-274
Gordon, R., W. J. Condon, W. J. Edgar, and S. J. Babie. 1978. Effects of mermithid parasitism on the haemolymph composition of the larval blackflies Prosimulium mixtum/fuscum and Simulium venustum. Parasitology 77:367-374.
Gordon, R., and J. M. Webster. 1971. Mermis nigrescens: physiological relationship with its host, the adult desert locust Schistocerca gregaria. Experimental Parasitology 29:66-79.
Gordon, R., J. M. Webster, and T. G. Hislop. 1973. Mermithid parasitism, protein turnover and vitellogenesis in the desert locust, Schistocerca gregaria. Comparative Biochemistry and Physiology 46B:575-593.
Hamilton, W. D., R. Axelrod, and R. Tanese. 1990. Sexual reproduction as an adaptation to resist parasites (A review). Proceedings of the National Academy of Sciences (USA) 87:3566-3573.
Heed, W. B. 1978. Ecology and genetics of Sonoran desert endemic Drosophila. Pages 109-126 in P. Brussard, editor. Ecological genetics: the interface. Springer-Verlag, New York, New York, USA.
Holmes, J. C., and S. Zohar. 1990. Pathology and host behaviour. Pages 34-63 in C. J. Barnard and J. M. Behnke, editors. Parasitism and host behaviour. Taylor and Francis, New York, New York, USA.
Hurd, H. 1990. Physiological and behavioral interactions between parasites and invertebrate hosts. Advances in Parasitology 29:271-318.
-----. 1993. Reproductive disturbances induced by parasites and pathogens of insects. Pages 87-105 in N. E. Beckage, S. N. Thompson, and B. A. Federici, editors. Parasites and pathogens of insects. Volume 1: Parasites. Academic Press, San Diego, California, USA.
Hurd, H., and C. Arme. 1986. Hymenolepis diminuta: influence of metacestodes upon synthesis and secretion of fat body protein and its ovarian sequestration in the intermediate host, Tenebrio molitor. Parasitology 93:111-120.
Jaenike, J. 1992. Mycophagous Drosophila and their nematode parasites. American Naturalist 139:893-906.
Jaenike, J., and T. J. C. Anderson. 1992. Dynamics of host-parasite interactions: the Drosophila-Howardula system. Oikos 64:533-540.
Johnston, J. S., and W. B. Heed. 1976. Dispersal of desert-adapted Drosophila: the saguaro-breeding D. nigrospiracula. American Naturalist 110:629-651.
Kaya, H. K., and R. Gaugler. 1993. Entomophagic nematodes. Annual Review of Entomology 38:181-206.
Keymer, A. E. 1980. The influence of Hymenolepis diminuta on the survival and fecundity of the intermediate host, Tribolium confusum. Parasitology 81:405-421.
-----. 1981. Population dynamics of Hymenolepis diminuta in the intermediate host. Journal of Animal Ecology 50: 941-950.
Kinn, D. N. 1966. Predation by the mite, Macrocheles muscaedomesticae (Acarina: Macrochelidae), on three species of flies. Journal of Medical Entomology 3:155-158.
Krantz, G. W. 1983. Mites as biological control agents of dung-breeding flies, with special reference to the Macrochelidae. Pages 91-98 in M. A. Hoy, G. L. Cunningham, and L. Knutson, editors. Biological control of pests by mites. University of California Special Publication 3304.
Lack, D. 1954. The natural regulation of animal numbers. Clarendon, Oxford, UK.
Lanciani, C. A. 1975. Parasite-induced alterations in host reproduction and survival. Ecology 56:689-695.
-----. 1986. Reduced survivorship in Dasyhelea mutabilis (Diptera: Ceratopogonidae) parasitized by the water mite Tyrrellia circularis (Acariformes: Limnesiidae). Journal of Parasitology 72:613-614.
Lanciani, C. A., and A.D. Boyt. 1977. The effect of a parasitic water mite, Arrenurus pseudotenuicollis, on the survival and reproduction of the mosquito Anopheles crucians (Diptera: Culicidae). Journal of Medical Entomology 14:10-15.
Mahowald, A. O., and M.P. Kambysellis. 1980. Oogenesis. Pages 141-224 in M. Ashburner and T. R. F. Wright, editors. The genetics and biology of Drosophila. Volume 2. Academic Press, London, UK.
Malone, L. A. 1987. Longevity and fecundity of Argentine stem weevils, Listronotus bonariensis (Coleoptera: Cuculionidae), infected with Microsporidium itiiti (Protozoa: Microspora). Journal of Invertebrate Pathology 50:113-117.
Markow, T. A. 1988. Reproductive behavior of Drosophila melanogaster and D. nigrospiracula in the field and the laboratory. Journal of Comparative Psychology 102:169-173.
May, R. M. 1983. Parasitic infections as regulators of animal populations. American Scientist 71:36-45.
May, R. M., and R. M. Anderson. 1979. Population biology of infectious diseases: Part II. Nature 280:455-461.
Mercer, C. F., and P. J. Wigley. 1987. A microsporidian pathogen of the poroporo stem borer, Sceliodes cordalis (Dbld) (Lepidoptera: Pyralidae). III. Effects on adult reproductive success. Journal of Invertebrate Pathology 49:108-115.
Minchella D. J., and M. E. Scott. 1991. Parasitism: a cryptic determinant of animal community structure. Trends in Ecology and Evolution 6:250-254.
Neter, J., W. Wasserman, and M. H. Kutner. 1990. Applied linear statistical models. Richard D. Irwin, Homewood, Illinois, USA.
Park, T. 1948. Experimental studies of interspecies competition. I. Competition between populations of the flour beetles, Tribolium confusum Duval and Tribolium castaneum Herbst. Ecological Monographs 18:265-307.
Pereira, C., and M.P. de Castro. 1945. Contribuicao para o conhecimento da especie tipo de Macrocheles Latr. (Acarina): M. muscaedomesticae (Scopoli, 1772). Arquivos do Instituto Biologico, Sao Paulo 16:153-156.
Pitnick, S., T. A. Markow, and M. F. Ridley. 1991. Transfer of ejaculate and incorporation of male-derived substances by females in the nannoptera species group (Diptera: Drosophilidae). Evolution 45:774-780.
Poinar, G. O., Jr. 1983. The natural history of nematodes. Prentice-Hall, Englewood Cliffs, New Jersey, USA.
Polak, M. 1993. Parasites increase fluctuating asymmetry of male Drosophila nigrospiracula: implications for sexual selection. Genetica 89:255-265.
Polak, M., and T A. Markow. 1995. Effect of ectoparasitic mites on sexual selection in a Sonoran desert fruit fly. Evolution 49:660-669.
Porter, C. A. 1970. The effects of parasitism by the trematode Plagioporus virens on the digestive gland of its snail host, Flumenicola virens. Proceedings of the Helminthological Society of Washington 37:39-44.
Price, P. W. 1980. Evolutionary biology of parasites. Princeton University Press, Princeton, New Jersey, USA.
Reader, T. A. J. 1971. Histochemical observations on carbohydrates, lipids and enzymes in digenean parasites and host tissues of Bithynia tentaculata. Parasitology 63:125-136.
Rodriguez, J. G., C. E Wade, and C. N. Wells. 1962. Nematodes as a natural food for Macrocheles muscaedomesticae (Acarina: Macrochelidae), a predator of the house fly egg. Annals of Entomological Society of America 55:507-511.
Roeder, K. D., editor. 1953. Insect physiology. John Wiley, New York, New York, USA.
Royce, L. A., and G. W. Krantz. 1991. A new rearing method for nematophagous mites. Pages 619-622 in E Dusbabek and V. Bukva, editors. Modern acarology. Academia, Prague, Czech Republic.
Rutherford, T A., and J. M. Webster. 1978. Some effects of Mermis nigrescens on the haemolymph of Schistocerca gregaria. Canadian Journal of Zoology 56:339-347.
SAS. 1989. SAS user's guide: statistics. Version 5. SAS Institute, Cary, North Carolina, USA.
Schall, J. J. 1983. Lizard malaria: cost to vertebrate host's reproductive success. Parasitology 87:1-6.
Simms, E. L. 1992. Costs of plant resistance to herbivory. Pages 392-425 in R. S. Fritz and E. L. Simms, editors. Plank resistance to herbivores and pathogens. University of Chicago Press, Chicago, Illinois, USA.
Smith, B. P. 1988. Host-parasite interaction and impact of larval water mites on insects. Annual Review of Entomology 33:487-507.
Thompson, S. N. 1983. Biochemical and physiological effects of metazoan endoparasites on their host species. Comparative Biochemistry and Physiology 74B:183-211.
Wade, C. E, and J. G. Rodriguez. 1961. Life history of Macrocheles muscaedomesticae (Acarina: Macrochelidae), a predator of the house fly. Annals of the Entomological Society of America 54:776-781.
|Printer friendly Cite/link Email Feedback|
|Date:||Jul 1, 1996|
|Previous Article:||Paleobiology, community ecology, and scales of ecological pattern.|
|Next Article:||Altered behavior of parasitized killifish increases susceptibility to predation by bird final hosts.|