
Limnol. Oceanogr., 36(3), 199 1, 483-495 0 199 1, by the American Society of Limnology and Oceanography, Inc Invertebrate predation in Lake Michigan: Regulation of Bosmina longirostris by Leptodora kindtii Donn K. Branstrator and John T. Lehman Department of Biology, Natural Science Building, University of Michigan, Ann Arbor 48 109 Abstract The density of Bosmina longirostris (0. F. Mueller) declined loo-fold between 25 June and 20 August 1985 in Lake Michigan; a similar pattern of changing abundances was observed in 1986 but not in 1987 or 1988. Bosmina collected from periods before and during the decline in 1985 and 1986 were examined for size at maturity and clutch size. Population size-frequency distributions suggest that some animals began to mature at smaller sizes during the decline in 1985. In some cases, the average clutch size for Bosmina of similar body lengths was significantly reduced during the decline in 1985 and 1986. These data suggest that food limitation was a factor in the species decline in both years. Gut content analyses indicate that of seven potential predators, Leptodora kindtii (Focke) was probably most responsible for Bosmina mortality. There is ample evidence that Leptodora does not consume strictly fluid from its prey: juvenile Leptodoru ate mainly smaller prey that included Bosmina and the rotifer Conochilus; adults ate a broader size range of prey that included mainly Bosmina, Daphnia, and copepods. In 1987 and 1988, Bythotrephes cederstroemii Schoedler was abundant in Lake Michigan and was contemporaneous with collapse ofthe Leptodora population and concurrent increase of the Bosmina population in August and September. Predation by the large copepods Epischuru lucustris S. A. Forbes and Limnoculanus macrurus Sars did not appear to have any significant effect on the population dynamics of Bosmina. Lake Michigan has long been regarded as food resources or from reduced planktivory an ecosystem in which visual planktivory by invertebrates. That planktivory played a by fish exerts a dominant structuring influ- role in Bosmina’s dynamics seems plausible ence. The exposition by Wells (1970) doc- but because several putative predators de- umenting changes in zooplankton popula- clined simultaneously the relative impor- tions in response to predation by alewife, tance of each is indecipherable from his data. Alosa pseudoharengus (Wilson), helped to Recent changes of Lake Michigan’s zoo- establish size-selective predation as one of plankton community, wrought by a new in- the paradigms of plankton community ecol- vading species, provide an opportunity to ogy. Wells observed that the successful in- further understand factors controlling Bos- vasion of alewife into Lake Michigan was mina dynamics there. In summers 1985 and associated with depressed densities of 1986, Bosmina populations in offshore Lake Daphnia, Leptodora, Limnocalanus, Epi- Michigan experienced similar abundance schura, and other large-bodied crustacean cycles. Populations quickly reached peak taxa and with increased densities of smaller densities in late June of 3 1,000 m-2 (1985) taxa, particularly Bosmina longirostris. and early July of 70,000 me2 (1986), and Although Wells (1970) argued convinc- then decreased steadily to densities of 600 ingly that the large plankton were removed m-2 (1985) and 9,000 rnw2 (1986) by mid- by selective planktivory, no rationale was August. In 1987, however, major changes advanced for the numerical increase in Bos- occurred in the composition of the offshore mina. In retrospect, the taxon’s success may zooplankton community; they were attrib- have stemmed from relaxed competition for uted in part to predation by the invading cladoceran, Bythotrephes cederstroemii Acknowledgments (Lehman 1988). That summer, and in 1988, We thank James Bowers, Douglass Burdette, Robert Bosmina did not decline in August and Sep- Gensemer, Robert Dorazio, Glenn Warren, and the tember as it had in 1985 and 1986 during captain and crew of the RV Laurentian for assistance those months. In this report we use the pop- with field sampling. Robert Black and Patricia Chow- Fraser provided suggestions for the manuscript. ulation changes associated with this event This research was supported by NSF grants OCE 84- to investigate the controls on Bosmina in 15970, OCE 86-13880, and OCE 87-16187. Lake Michigan. 483 I J 484 Branstrator and Lehman General approach at 0.5-l m s-l. Large- and small-bodied Factors responsible for the seasonal de- zooplankton were collected by submersible clines of Bosmina in Lake Michigan in 198 5 pump discharged through either 63- or 130- and 1986 were hypothesized to be resource pm mesh (Dorazio et al. 1987; Lehman et limitation, predation, or a combination of al. 1990). Pump collections were integrated the two. Because populations in 1985 and over 10-m intervals. Temperatures were 1986 declined at summer epilimnetic water measured by mechanical bathythermo- temperatures that permitted continued graph (1985 and 1986) and by electronic growth in 1987 and 1988, ambient temper- bathythermograph(l987 and 1988). In 1987 ature was not considered to ‘be a cause. Bos- and 1988, zooplankton were collected ex- mina size-frequency distributions and size- clusively by closing nets fitted with 130-pm specific egg ratios (average clutch size per mesh. Zooplankton were immediately nar- body-length class) were examined for cotized with carbonated water and pre- changes that would suggest food limitation served in a sugar-Formalin solution (Haney as a factor. Gut analyses were performed on and Hall 1973). Samples collected by the the potential predators to determine if any trawl and closing nets were split by volume were eating Bosmina. On the basis of phys- with a Folsom plankton splitter. Represen- ical presence and potential for carnivory, tative fractioned samples were counted to these predators were young-of-the-year determine population densities of species (YOY) Coregonus hoyi (Gill), and the crus- by date and depth. Pump samples were sub- tacean zooplankton Mysis relicta (Loven), sampled by pipette and counted. Leptodora kindtii, Epischura lacustris, Lim- Bosmina-The potential effects of re- nocalanus macrurus, Senecella calanoides source limitation on Bosmina were deter- Juday, and cyclopoids [mainly Diacyclops mined by examining the animals for changes bicz.qidatus thomasi (S. A. Forbes)]. Plank- in size at maturity and clutch size. Goulden tivory on Bosmina by YOY C. hoyi was et al. (1982) reported that Bosmina grown immediately rejected based on results of gut under low-food conditions matures at a analyses from 1985 (Warren and Lehman smaller size than better fed individuals. The 1988), which showed that these vertebrates smallest sizes at which Lake Michigan Bos- consumed cyclopoid copepodids and Daph- mina reached second and third (primapa- nia pulicaria Forbes almost exclusively dur- rous) instars were determined from popu- ing the time when Bosmina declined. Evi- lation size-frequency distributions. Instars dence for the potential carnivory of M. were identified from modes of the distri- relicta (Bowers and Vanderploeg 1982), L. butions and numbers of barren vs. fecund kindtii (Mordukhai-Boltovskaia 1958), E. females in individual frequency cells. We lacustris (Chow-Fraser and Wong 1986), L. hypothesized that declining resources would macrurus, and S. calanoides (Nero and have caused Bosmina to mature at a smaller Sprules 1986) and cyclopoids (Fryer 19 5 7) size. made these species candidates for study. Somatic growth in Bosmina continues af- ter primaparity; thus later adult instars have Methods larger brood chambers and can carry more eggs. Kerfoot (1974) found that Bosmina Zooplankton collections -Zooplankton grown under low-food conditions carries were collected during summer 198 5-l 988 fewer eggs per clutch than well-fed individ- in Lake Michigan at a reference station uals; this relationship held over a range of (43ON, 86”4O’W), 36 km offshore from carapace lengths. We hypothesized that de- Grand Haven, Michigan (Lehman et al. clining resources would have forced Bos- 1990). In 1985 and 1986, large crustacean mina to decrease clutch size. zooplankton were collected by Tucker trawl To test both hypotheses, we measured with a mouth opening of 1 m* and a .net Bosmina for body length and clutch size. aperture of 300 pm and by l-m-diameter Plankton samples were filtered on 50-pm Puget Sound vertical closing nets with 130- mesh and rinsed with filtered tapwater (50 or 300-pm mesh. Trawl and nets were towed pm) to remove excess Formalin. After re- Bosmina in Lake Michigan 485 suspension, random subsamples of 5 ml were density in that stratum for each period. x2 extracted with a widebore Finn injection pi- homogeneity tests showed that E was not pette and discharged into a zooplankton significantly different by stratum on any date counting chamber. Animals were viewed on (P > 0.05). Values for the dependency of a Wild M5A stereo dissecting microscope egg development rate on temperature were at 25 x magnification. As encountered, in- taken from Vijverberg (1980). The popu- dividuals were removed to and mounted on lation egg development time was then com- glass microscope slides in 50% glycerin. puted as an average, weighted to account Coverslips were sealed with Permount his- for the time that different percentages of tological mounting medium (Fisher SO-P- eggs experienced different temperatures over 15). Individuals were measured for total 24 h. The instantaneous population growth body length (Kerfoot 1980) with the aid of rate (r) was taken as an ocular micrometer under 190 x magni- r = (In N, - In NJ/t fication on an Olympus BHA compound (3) stereo microscope. Clutch size was record- where A$ and No are the population densi- ed. ties at time t and 0, and t is the time between Bosmina egg volume was also estimated sample dates. The instantaneous death rate to determine whether egg size could explain (d) was taken as differences in clutch size across dates. For d=b-r. individuals carrying eggs in an early stage (4) of embryonic growth, dimensions of one egg Daily death rate (DR) was then taken as per clutch were measured.
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages13 Page
-
File Size-