Prey Selection by Chaoborus in the Field and Laboratory
Larvae of Chaoborus species are voracious predators of zooplankton in aquatic ecosystems, capable of altering the species composition and size structure. To estimate the impact of Chaoborus larvae on zooplankton in Campus Lake, Jackson Co. IL, a 24-hour field survey and a controlled predation experiment in the laboratory were conducted. Specifically, the hypothesis tested was that Chaoborus larvae select Daphnia over copepods, two zooplankton taxa on which the larvae are known to feed and which occur in Campus Lake. In the lake, Chaoborus only positively selected rotifers, a small taxon of zooplankton, but remains of Bosmina, Daphnia and copepods were also found. In the laboratory experiment, overall predation rates were low and Chaoborus were not strongly selective when fed Daphnia and copepods. Thus, the original hypothesis was not supported. Data from the field sampling suggest that Chaoborus prey on the smallest prey taxon available, meaning they could strongly influence the plankton composition on a seasonal basis. Furthermore, the lack of prey parts in the guts of Chaoborus from the laboratory experiment suggests that predation estimates based on gut content analyses from field surveys may underestimate the predatory impact of Chaoborus on large zooplankton such as Daphnia and deserves further investigation.
Biotic communities can be strongly influenced by predation. Highly selective or abundant predators can severely reduce or even eliminate some of their prey populations. In aquatic ecosystems, especially lakes, such prey reductions can lead to cascading effects (Carpenter et al. 1985) and may result in undesirable lake conditions. Generally, in lakes with a high density of fish, the average body size of zooplankton is small because fish tend to select large-bodied prey (Brooks and Dodson 1965). In contrast, in lakes with abundant invertebrate predators, the average body size of zooplankton is large because invertebrate predators are gape-limited and select small-bodied prey (Vanni 1988, Wilhelm et al. 2000). When the abundance and size of herbivorous zooplankton is reduced, phytoplankton and bacteria are released from grazing pressure that often results in the formation of dense algal blooms (Dawidowicz 1990). The degradation of these blooms in turn results in oxygen depletion and anoxic bottom waters (Smith 2001), which are inhospitable to life forms with aerobic metabolism.
Chaoborus larvae (Figure 1 A) are one example of invertebrate predators with the ability to greatly affect zooplankton populations. Chaoborus larvae are the life stage of the terrestrial phantom midge, a non-biting fly similar to a mosquito, which are restricted to aquatic habitats where they develop through four instars (Saether 1972). The larvae are voracious predators and can cause shifts in the size structure of zooplankton communities and eliminate species (Neill 1981, Vanni 1988). For example, Bosmina (Figure 1 B) was completely absent from small bog lakes in northern Michigan because of predation by Chaoborus (Von Ende and Dempsey 1981). The larvae have well-developed mandibular teeth, are nearly transparent and ambush their prey (Saether 1972). The first two instars are entirely planktonic, while the third and fouth instars undergo diurnal vertical migrations (Stahl 1966). They are benthic during the day and planktonic at night (La Row and Marzolf 1970). Typically, the migration cycle involves an ascent of daytime benthic larvae up to the epilimnetic zone around sunset. The larvae descend again in the early morning hours near dawn (Juday 1921; Eggleton 1932; Berg 1937; Stahl 1966; Roth 1967). This vertical migration is thought to be a strategy to avoid visual fish predators because Chaoborus are often an important food source for fish (Lampert 1993).
To feed, Chaoborus larvae ambush their prey by hanging motionless in the water column until a prey comes close. Hydrodynamic disturbances caused by prey movement are detected by the mechanosensory setae of Chaoborus (Blais and Maly 1993). Chaoborus then attack the prey using convulsive movements. Because of their relative small size (6-23 mm), Chaoborus larvae are limited to feeding on small zooplankton such as cladocerans (Bosminaand Daphnia, Figure 1 B and C), copepods (Figure 1 D), and rotifers (Figure 1 E) (Vanni 1988; Moore et al. 1994). Moore et al. (1994) found that consumption of prey was positively related to larval development (instar) of Chaoborus punctipennis. Late instar (third and fourth) larvae consumed more copepods and Daphnia than the first and second instar larvae. However, all instars consumed rotifers. As mentioned above, Chaoborus can eliminate certain zooplankton species entirely from lakes and their impact is especially high when they are abundant.
In Campus Lake, a relatively shallow (max. depth 4.8 m) and small (16 ha) reservoir located on the Campus of Southern Illinois University in Jackson County IL, Chaoborus larvae are known to occur at high densities (up to 20 000 m-2 Wilhelm; unpublished data) which are among the highest recorded in the literature. Thus, they should have a pronounced predatory impact on the zooplankton community, which includes all of the common prey items of Chaoborus (Figure 1). To-date, no studies of the influence of Chaoborus at such high densities have been undertaken. To examine the impact of Chaoborus on the zooplankton community of Campus Lake, a field and laboratory study were undertaken. The field survey was used to examine predation rates and prey selection, while the laboratory study was used to test the hypothesis that Chaoborus larvae select Daphnia over copepods. There was also a need to estimate predation rates under controlled conditions because the impact of Chaoborus in lakes is often inferred from laboratory feeding trials (e.g., Pastorok 1980).
Materials and MethodsStudy Site:
Campus Lake is located in the campus of Southern Illinois University at latitude of 37° N in extreme southern Illinois. Student dormitories occupy the north shore of the lake while the southern portion of the watershed is predominantly forested. Rainwater and runoff provide the only inflow to the lake. From approximately mid March to mid September, the lake is thermally stratified. The fish community includes sunfish (esp. bluegill), channel catfish, grass carp, and bass, while the invertebrate zooplankton populations include predominantly rotifers, cladocerans, and copepods. As benthic invertebrates, Campus Lake has members of the families Chaoboridae, Chironomidae, and Ceratopogonidae. The lake is also influenced by artificial night light (ANL) from the dormitories, lights from a beach on the south shore, and the boat dock. General water chemistry and lake usage data are summarized by Muchmore et al. (2004).Field Sampling:
To determine the vertical migration pattern and prey selection of Chaoborus larvae in Campus Lake, a single 24-hour survey was performed on September 11, 2004. Samples were taken at the deepest site in Campus Lake at 6 hr intervals starting at 06:00. This date was chosen to obtain maximal rates of predation by the dense Chaoborus and to accommodate scheduling restrictions of the author. Because Chaoborus overwinter in the lake as fourth instar, larvae predation should have been maximal at this time. Once the lake cools in late fall after it destratifies, the metabolic rate and hence the predation rate of Chaoborus is slowed by the cold water temperature similar to other poikiliotherms. Ideally, the study would have been repeated a second time in the spring before all the Chaoborus emerged, however, the author's academic schedule limited the field research to the fall semester.
At each sampling time, profiles of dissolved oxygen (D.O.), water temperature (oC), and conductivity (μS) were taken at 0.5 m intervals from the surface to the bottom with a YSI model 85 multi-meter. At 06:00, 12:00, and 24:00 Chaoborus were sampled by three to five replicate vertical hauls of a regular plankton net (64 μm mesh, 0.2 m mouth diameter) hauled from 4.5 m. The contents of the net were rinsed into small containers using 64 μm filtered lake water. If Chaoborus were seen in the containers, soda water was added to relax them before a buffered formalin solution was added to preserve the sample (4% final concentration). This was done to prevent the expulsion of crop contents and prevent the loss of prey. At 18:00, the original net was accidentally dropped into the lake and a 64 μm mesh Wisconsin-style plankton net (0.12 m mouth diameter) was used to sample instead.
To sample Chaoborus from the sediment, a standard 0.15 m x 0.15 m Ekman grab sampler was used to collect three sediment samples at 06:00 and 09:00. Each sediment sample was individually rinsed through a 500-μm sieve with tap water to remove sediment after which all Chaoborus retained on the sieve were picked off with fine forceps and placed into labeled vials. These were also relaxed with soda water and then preserved with buffered formalin to make a final concentration of approximately 4%.
Chaoborus from the field samples were counted and densities calculated by dividing by the volume sampled with each plankton haul (171 and 60 L for the regular and Wisconsin-style nets, respectively) for water column samples, and by the area of the Ekman grab (0.0225 m2) for sediment samples. Mean densities, standard deviations and standard errors were calculated from the replicates. To test if densities differed with sampling time, analysis of variance (ANOVA) to compare densities in the water column was used, and a t-test was used to compare densities in the sediment (sediment was only sampled at two time intervals).
To determine the selection of zooplankton by Chaoborus in the lake, gut contents of Chaoborus from 06:00 time interval of the 24 h survey were examined. This sample was chosen to maximize the chance of finding Chaoborus with food in their guts because they were returning to the sediment after feeding in the water column during the night. A small amount of glycerin jelly was placed on a slide and liquefied when heated to 50º C on a hotplate. The head and crop of Chaoborus larvae were separated from the body using watchmakers forceps and insect pins, placed into the glycerin jelly and opened to release the contents. A coverslip was then applied to the slide and after cooling it was sealed with clear nail polish.
All slides were then completely examined under a compound microscope at 400X. Rotifers were easily identified from their intact lorica, while copepods and cladocerans were identified by diagnostic parts. For Daphnia, these included post abdominal claws, mandibles and second antennae. For Bosmina, typically, the carapace and characteristic first antennae could be identified, while for copepods caudal rami, mandibles and antennae were used. For all paired appendages, the number counted was divided by two to obtain the number of prey. As well, prey counts were only increased if the number of diagnostic parts exceeded those of one individual. For example, if one post abdominal claw and two second antennae were counted, one Daphnia prey item was recorded.
Selectivity was calculated by Manly's alpha for constant prey populations using equation 1 (Krebs 1989); where αi is Manly's alpha (preference index) for prey type i, ri and rj are the proportions of prey i or j in the diet. The values ni and nj represent the proportion of prey type i or j in the environment. An α of 0.5 indicates no selection for a prey item, while an α > 0.5 or an α < 0.5 indicates selection for or against a prey item, respectively. The closer α is to 1 or 0, the stronger the selection for or against a prey item, respectively (Krebs 1989). To calculate the proportion of prey in the environment, the average zooplankton density was determined from the three replicate vertical haul samples taken at 06:00.Laboratory Experiment:
To examine the predation rate and selectivity of Chaoborus when given a choice of Daphnia and cyclopoid copepods, a laboratory feeding experiment was undertaken. Zooplankton for the feeding experiment, were collected from Campus Lake on October 19th, 2004 with vertical hauls of a 64-μm Wisconsin-style plankton net. To obtain Chaoborus predators, lake sediment was collected with an Ekman grab and processed as for the 24 hr survey with the exception that larvae were transferred to 64 μm-filtered lake water. Additional lake water was brought into the laboratory and filtered to run the experiment. The water was first filtered through a 64-μm Wisconsin-style plankton net, then through a 10-μm phytoplankton net, and finally vacuum filtered through Whatman GFC filters (~2 μm pore size). Eight glass beakers were filled to 1.9 L with the filtered water. Using a dissecting microscope and large bore eyedropper, 26 Daphnia and 26 copepods of approximately the same size were transferred into each experimental beaker. This prey density and ratio were used to reflect the density and ratio in the lake in September. Five large Chaoborus were then taken out of the holding jar and one placed in each of the five experimental beakers. Three beakers did not receive Chaoborus and served as controls. The beakers were placed near a north-facing window to simulate natural diurnal light regimes. A cardboard wall was built on the laboratory side of the window to block light from the room. The Chaoborus were left to feed for 48 hr. After 24 hr, the experiment was checked for dead Chaoborus. To stop the experiment after 48 hr, the contents of the eight beakers were filtered through a 64-μm sieve and rinsed into petri dishes. Each dish was then observed under a dissecting microscope to identify and count dead and remaining zooplankton.
Selectivity was calculated using Manly's Alpha as above. I used this formula instead of the modified version to account for decreasing prey density (Krebs 1989) because few prey were consumed over the course of the experiment which did not appreciably decrease their density. The mean ± standard error (SE) selectivity of Chaoborus for Daphniaand copepods was calculated and graphed.
In the lake, Chaoborus showed pronounced diurnal vertical migration over 24 h (Figure 2 A). The density of Chaoborus ranged from < 0.06 L-1 at 12:00 to 0.32 individuals L-1 at midnight (Figure 2 A). These densities differed significantly, with the 12:00 and 18:00 samples having lower densities (ANOVA F1,12 = 10.91, P < 0.001) than the night samples (Figure 2 A). In the sediment, the density of Chaoborus was approximately 4,100 individuals m-2 and did not differ (t-test P = 0.898) between 06:00 and 09:00 (Figure 2 B).
Analysis of gut contents of Chaoborus collected during the 24 hr survey showed overwhelming (α = 0.943) selection for the rotifer Keratella (Figure 3). Seventeen Keratella were found in the gut contents of Chaoborus examined compared to only six in the water column samples. Bosmina,Daphnia, and copepods were not positively selected compared to densities in the environment (Figure 3), but their remains did occur in the gut contents.
In general, the predation rate was 1.7 prey Chaoborus-1day-1 (Table 1). For individual prey it ranged from a low of 0.03 prey Chaoborus-1day-1 for copepods to a high of 0.90 prey Chaoborus-1day-1 for rotifers (Table 1).
In the laboratory experiment few prey were consumed. Only 9 and 4 of 140 and 98 Daphnia and copepods, respectively, were consumed over the 48 hr resulting in rates of 0.900 and 0.400 prey Chaoborus-1day-1 (Table 1) which are higher than those recorded for the field. Chaoborus selected slightly for Daphnia(Figure 4). However, selection by individual larvae varied greatly and the individuals in replicates one and four fed on more copepods than Daphnia, while the individual in replicate five did not consume any copepods (Figure 4 B). The larva in replicate two was "kinked" in the middle and only spasmed in place at the end of the experiment. However, when checked at 24 hr, it appeared healthy and moved normally. Since Chaoborus are ambush predators and need to cover large distances quickly, this injury probably reduced the overall feeding ability of the Chaoborus in replicate two. Since the Chaoborus was healthy at 24 hr, the data are presented for completeness but were excluded from selectivity calculations (Figure 4 B and C).
Chaoborus larvae in Campus Lake undertook diurnal vertical migrations, similar to those observed in other lakes (La Row and Marzoff 1970). However, a number of large larvae remained in the water column during the day, which was unexpected. Although some larvae can be expected in the water column because first and second instars are planktonic, third and fourth instars are generally reported to be benthic by day. The presence of late instar larvae in the water column of Campus Lake may be due to night light pollution from campus buildings. The sampling site had so much light at night from lights at the beach and student dorms that headlamps were only needed to see Chaoborus in the collected samples. Thus, it is possible that this brightness' at night disrupted the normal migration pattern, forcing larvae into the water column whenever they were hungry, rather than in synchrony with nighttime darkness. Although the observed vertical migration pattern follows the traditional pattern, the presence of large numbers of Chaoborus in the sediment at 06:00 suggests that densities in the water column may have been higher still had all individuals left the sediment. This could be a further indication of disrupted migration patterns in Campus Lake.
The analysis of gut content of Chaoborus from Campus Lake showed almost exclusive selectivity for rotifers, specifically Keratella. Although Chaoborus larvae are known to feed on rotifers (Moore et al. 1994), such strong selectivity has not been reported. This strong selectivity resulted from the occurrence of Keratella in the guts compared to its near absence from the lake. This may suggest that Chaoborus become selective for prey, which is seasonally abundant and are then slow to switch to other prey types when the density of the preferred prey declines. Of the other prey taxa, Chaoborus exhibited a higher selectivity for Bosmina than for Daphnia and copepods. Bosmina were among the small prey items consumed by Chaoborus, which may reflect a handling/capture size-limitation. Although Copepods were the most abundant taxon in the lake, few were found in the guts of Chaoborus. Copepods have a quick and well-developed snapping escape response to predators. It is much faster than that of either Bosmina or Daphnia, which sink passively as an escape response (Kerfoot 1975). Because Chaoborus are quick to attack prey, a slowly sinking small Bosmina would be much easier to catch and handle than an escaping copepod. This snapping/catapulting escape mechanism could have prevented copepods from being captured and eaten. Thus it is likely that part of Chaoborus selectivity is related to prey size and escape ability. Chaoborus may have also selected against copepods because of their morphology. Moore and Gilbert (1987) found that as space in the crop of Chaoborus became limited during feeding, soft bodied prey could be compacted more easily. This also may influence selection because copepods have a harder carapace than cladocerans.
Chaoborus showed slight selection for Daphnia (Figure 4) in the laboratory experiment in which all of the Chaoborus consumed at least one Daphnia. The presence of dead and mangled prey items in the beakers at the end of the laboratory experiments clearly demonstrated that the Chaoborus fed. However, when slides of the gut contents were made, no identifiable parts of Daphnia or copepods were found in the crop contents. This suggests that Chaoborus selected soft body parts and were sloppy feeders. This also lends support to the above argument that soft tissue is more easily compressed into a limited space. This result could indicate that feeding rates based on gut content analyses underestimate the number of large prey items such as Daphnia and copepods killed by Chaoborus. As mentioned above, the predation rate of Chaoborus and impacts on populations of lake plankton are often based on gut content analyses from laboratory experiments (e.g., Pastorok 1980). This could mean that such calculations underestimate the number of prey killed by Chaoborus in the real world. A follow-up experiment to quantify the sloppy feeding of Chaoborus would be helpful to correct predation estimates based on gut content analyses from other field studies.
Although the predation rates in the laboratory experiment appeared low, they were higher than in the lake (Table 1). Considering the sloppy feeding of Chaoborus in the laboratory, this may indicate that the rates estimated from field collected animals underestimated the actual number of prey consumed. Because collections were only made on one day, rates were compared to those determined from another study on Campus Lake performed in August 2003 (Jacobs 2003). Jacobs (2003) also analyzed gut contents from field collected animals and reported predation rates of 0.27-1.36 rotifers Chaoborus-1day-1, 0.4-1.23 Bosmina Chaoborus-1day-1, and 0.15-0.36 copepods Chaoborus -1day-1. These rates encompass the rates determined except for my field-estimated rate for copepods, which was much lower. Thus, it is very likely that the laboratory rates reflect actual prey selection and rate by Chaoborus in the absence of rotifers or other small prey such as Bosmina.
The feeding rate of Chaoborus in the laboratory experiment could have been influenced by factors including presence of walkway lights, lack of sediment and prey offered. Light from the laboratory was blocked from the experiment, however, light from paths outside the window may have disrupted the night time feeding behavior of the larvae. Also, lack of sediment in the bottom of the beakers, into which the Chaoborus could burrow during the day, and which would have complicated the experimental setup, may have stressed the larvae and influenced feeding activity. Finally, the size of prey offered may also have influenced the feeding rates. Rotifers and Bosmina, which are much smaller than Daphnia and copepods were selected in the lake. Thus presenting only large prey may have elicited a higher attack response from the larvae.
Overall, the high predation rate and strong selectivity of Chaoborus for rotifers indicates this taxon is at greatest risk in Campus Lake. The smallest of the cladocerans, Bosmina, are also at risk, while Daphnia and copepods only face weak predation pressure. The hypothesis that Chaoborus select Daphnia over copepods was not supported. Because Chaoborus are gape-limited predators, an abundant Chaoborus population should promote populations of large-bodied zooplankton, one objective in the Campus Lake restoration (Muchmore et al. 2004).
This research was funded by a SIUC REACH (Research Enriched Academic Challenge) grant through ORDA (the Office of Research Development and Administration). I thank Dr. Frank Wilhelm for his help throughout all aspects of the project. Alicia Jacobs, Mike Venarsky and Nick Gaskill, graduate students in the Limnology Laboratory, also provided assistance.
Berg, K. (1937) Contribution to the biology of Corethra meigen (Chaoborus lichtenstein). Biol. Meddr. 13, 1-101.
Blais, J. M. and E. J. Maly (1993). Differential predation by Chaoborus americanus on males and females of two species of Diaptomus. Canadian Journal of Fisheries and Aquatic Sciences 50, 410-415.
Borkent, A. (1979) Systematics and bionomics of the species of the subgenus Schadonophasma Dyar and Shannon (Chaoborus, Chaoboridae, Diptera). Quaestiones Entomologicae 15, 122-255.
Brett, M. T. (1992) Chaoborus and fish-mediated influences on Daphnia longispina population structure, dynamics and life history strategies. Oecologia 89, 69-77.
Brooks, J. L. and S. I. Dodson. (1965) Predation, body size, and compositions of plankton. Science 150, 28-35.
Carpenter, S. R., J. F. Kitchell and J. R. Hodgson (1985) Cascading trophic interactions and lake productivity. Bioscience 35, 634-639.
Dawidowicz, P. (1990) Effectiveness of phytoplankton control by large-bodied and small-bodied zooplankton. Hydrobiologia 200/201, 43-47.
Eggleton, F. E. (1932) Limnetic distribution and migration of Corethra larvae in two Michigan lakes. Paper of the Michigan Academy of Science 15, 361-388.
Elser, M. M., C. N. Von Ende, P. Sorrano and S. R. Carpenter (1987) Chaoborus populations: response to food web manipulation and potential effects on zooplankton communities. Canadian journal of Zoology 65, 2846-2852.
Jacobs, A. M. (2003) The impact of Chaoborus larvae on zooplankton abundance and size structure in a shallow Midwest U.S. Undergraduate Research, Department of Zoology, Southern Illinois University, Carbondale, IL.
Juday, C. (1921) Observations on the larvae of Corethra punctipennis. Biological Bulletin of the Marine Biology Laboratory 40, 271-286.
Kerfoot, W. C. (1975) The divergence of adjacent populations. Ecology 56, 1298-1313.
Krebs, C. J. (1989) Ecological Methodology. Harper Collins Publishers Inc., New York, N.Y.
Lampert, W. (1993) Ultimate causes of diel vertical migration of zooplankton: New evidence for the predator-avoidance hypothesis. Archiv für Hydrobiologie Beiheft 39, 79-88.
La Row, E. J. and G. R. Marzolf (1970) Behavioral differences between 3rd and 4th instars of Chaoborus punctipennis Say. American Midland naturalist 84, 428-436.
Moore, M. V. (1988) Differential use of food resources by the instars of Chaoborus punctipennis. Freshwater Biology 19, 249-268.
Moore, M. V. and J. J. Gilbert (1987) Age-specific Chaoborus predation on rotifer prey. Freshwater Biology 17, 223-236.
Moore, M. V., N. D. Yan and T. Pawson (1994) Omnivory of the larval phantom midge (Chaoborus spp.) and its potential significance for freshwater planktonic food webs. Canadian Journal of Zoology 72, 2055-2065.
Muchmore, C., J. Stahl, E. Talley and F. M. Wilhelm (2004) Phase I Diagnostic / Feasibility Study of Campus Lake, Jackson County, Illinois. Southern Illinois University at Carbondale. 1-203.
Mumm, H. and A. F. Sell (1995) Estimating the impact of Chaoborus predation of zooplankton: A new design for in situ enclosures studies. Archiv für Hydrobiologie 134, 195-206.
Neill, W. E. (1981) Impact of Chaoborus predation upon the structure and dynamics of a crustacean zooplankton community. Oecologia 48, 164-177.
Pastorok, R. A. (1980) Selection of Prey by Chaoborus Larvae: A Review and New Evidence for Behavioral Flexibility. In Evolution and Ecology of Zooplankton Communities. Edited by W.C. Kerfoot. University Press of New England, Hanover, N.H. pp. 538-554.
Pastorok, R. A. (1980) The effects of predator hunger and food abundance on prey selection by Chaoborus larvae. Limnology and Oceanography 25, 910-921.
Roth, J. C. (1967) Notes on Chaoborus species from the Douglas Lake region, Michigan, with a key to their larvae (Diptera: Chaoboridae). Papers of the Michigan Academy of Science, Arts, and Letters 11, 63-68.
Roth, J. C. (1968) Benthic and limnetic distribution of three Chaoborus species in a southern Michigan lake (Diptera :Chaoboridae). Limnology and Oceanography 13, 242-249.
Saether, O. A. (1970) Family Chaoboridae. In Das Zooplankton der Binnengewässer. Binnengewässer. 15, 257-280.
Smith, D. G. (2001) Pennak's Freshwater Invertebrates of the United States 4th ed. Porifera to Crustacea. John Wiley and Sons, Inc., New York, N.Y.
Stahl, J. B. (1966) The ecology of Chaoborus in Myers Lake, Indiana. Limnology and Oceanography 11, 177-183.
Vanni, M. J. (1988) Freshwater zooplankton community structure: introduction of large invertebrate predators and large herbivore to a small-species community. Canadian Journal of Fisheries and Aquatic Sciences 45, 1758-1770.
Von Ende, C. N. and D. O. Dempsey (1981) Apparent exclusion of the Cladoceran Bosmina longirostris by invertebrate predator Chaoborus americanus. Am. Mid. Nat. 105, 240-248.
Von Ende, C. N. (1982) Phenology of four Chaoborus species. Environmental Entomology 11, 9-16.
Wilhelm, F. M., D. W. Schindler and A. S. McNaught (2000) The influence of experimental scale on estimating the predation rate of Gammarus lacustris (Crustacea: Amphipoda) on Daphnia in an alpine lake. Journal of Plankton Research 22, 1719-1734.