Govoni and Olney Predation on fish eggs by Mnemiopsis leidyi in and around Chesapeake Bay 



183 



formalin solution. Ichthyoplankton was preserved in 

 either 5% formalin or 95% ethanol. All M. leidyi and 

 fish eggs were counted except in those samples of ex- 

 ceptionally high ctenophore volume, where ctenophore 

 number was estimated by volumetric subsampling and 

 multiplication. Counts of ctenophores and fish eggs 

 were averaged for replicate collections taken at a depth 

 stratum and diel interval. 



Estimation of co-occurrence 



Our intention was to assess the small-scale co-occur- 

 rence of ctenophores and eggs relative to the water 

 masses overlying these stations and to then evaluate 

 potential predation. Because the depth of each sample 

 occasionally varied from the nominal and the trawl con- 

 sequently fished through hydrographic discontinuities, 

 some collections were omitted from consideration. Col- 

 lections omitted were those in which salinity values, 

 recorded during each 30-60 second fishing interval, 

 varied outside a range of 1.5%o. This procedure elim- 

 inated seven of 35 collections at station El and none 

 at E4, the two stations where ctenophores and fish 

 eggs were consistently present and where we focused 

 our assessment of potential predation. 



Estimates of potential predation 



We estimated potential predation, for each depth and 

 diel interval, as the product of clearance rate (the 

 volume of water cleared of all prey per unit time per 

 ctenophore), times the end points of the range of den- 

 sity of ctenophores (the number of ctenophores per unit 

 volume, averaged for replicates), times the end points 

 of the range in density of fish eggs (again averaged for 

 replicates). A clearance rate of 168L/day was used 

 from Monteleone and Duguay (1988), who found that 

 the clearance rate of fish eggs was independent of egg 

 density (as well as the presence of alternate prey) and 

 was positively and linearly related to experimental 

 vessel size. This clearance rate was the highest rate 

 observed for ctenophores 4.5-5.0 cm in length feeding 

 in the largest vessels employed and falls roughly within 

 the range of values reported elsewhere (Larson 1987). 

 A sample of 10 preserved ctenophores from our col- 

 lections averaged 8.5mL in volume which converts to 

 an average length of 4 cm (Kremer and Nixon 1976). 

 We did not account for shrinkage. 



Results 



Distribution and co-occurrence 



Mnemiopsi leidyi and fish eggs were consistently pres- 

 ent only at stations El and E4 (Fig. 1). Pulses in den- 



sities of M. leidyi were evident, but did not conform 

 to specific diel intervals or tidal phases (Figs. 2, 3). Egg 

 density showed a diel pattern, with peak densities from 

 dusk to dawn. Eggs of Anchoa mitchilli accounted for 

 an average of 74.2% (range 23.0-98.5%) of the fish 

 eggs at station El and 32.3% (range 0-62.9%) at E4. 

 Mnemiopsis leidyi and fish eggs were, for the most 

 part, vertically segregated at station El, but co- 

 occurred, particularly in surface water, at E4. Vertical 

 segregation at El (Fig. 2) reflected the physical 

 stratification of the water column with a warm, low- 

 salinity, surface-layer characteristic of the Chesapeake 

 Bay plume overlying a cool, higher-salinity, bottom- 

 layer characteristic of coastal shelf water (Boicourt et 

 al. 1987). At El, in the southern reaches of the mouth 

 of the Bay, surface collections within the plume yield- 

 ed higher egg densities, while subsurface collections 

 yielded higher M. leidyi densities. Station E4, in the 

 northern reaches and outside the plume, was 

 unstratified with no thermo-, halo-, or pycnocline (Fig. 

 3). Water at this station apparently was a mixture of 

 Chesapeake Bay water and coastal shelf water, likely 

 the result of tidal, rather than wind, mixing. Winds, 

 often responsible for mixing at the mouth of the Bay 

 (Ruzecki 1981), were light to moderate during this 

 sampling period (l-8m/second). 



Potential predation 



Overall, potential predation was greater in the un- 

 stratified northern reaches of the mouth of the Bay out- 

 side the plume (E4) than in the southern reaches 

 stratified by the plume (El), because of greater tem- 

 poral and spatial co-occurrence of M. leidyi and fish 

 eggs there. Range estimates of potential population 

 predation were 0.1-14.7 eggs per m 3 /day at El, and 

 0-174.3 at E4 (Table 1). 



Discussion 



The assessment of ichthyoplankton predation in the 

 field has been based historically on the examination of 

 predator gut contents or on the strength of a negative 

 correlation of predator and prey densities, even though 

 biases may result from the lability of fish eggs and lar- 

 vae in the guts of predators, from the feeding of pred- 

 ators within the collecting device used to sample 

 predator and prey (Purcell 1985), and from the spurious 

 inference of cause and effect drawn from correlation 

 analysis (Frank and Legget 1982, 1985). Few have 

 resolved successfully the first two problems (Bailey and 

 Houde 1989, Purcell 1989, Purcell and Grover 1990). 

 In regard to the latter, the importance of small-scale 

 spatial and temporal distribution of predator and prey 



