There was a general decline at stations II 

 and III in midlake throughout the season, 

 whereas at station I in shallow water, two 

 maxima occurred, one in early July and the 

 other in September. Zooplankton density was 

 highest at station II. Part of the differences 

 between stations may be explained by the fact 

 that station III samples were weighted by the 

 inclusion of deeper plankton- scarce samples. 

 If only the surface and 5-m. samples for each 

 station are plotted, however, plankton density 

 at station II remains highest (fig. 29). A pos- 

 sible explanation of this phenomenon is that 

 station II may be in a region of upwelling 

 cool water resulting from deflection of water 

 masses upward along the dropoff near station II. 



The significant points of figures 28 and 29 

 are: the general decline in zooplankton abund- 

 ance through the season and the consistently 

 high plankton density at station II. The second 

 point has no obvious significance in the ecology 

 of sockeye salmon, but the first has the follow- 

 ing significant implications. 



The seasonal decline in zooplankton must 

 have been caused either by progressively less 

 favorable conditions for plankton production 

 or by cropping by fish. If the former is true, 

 insects are the preferred food and plankton is 

 second choice, as evidenced by the fact that 

 in the spring when zooplankton is most abund- 

 ant juvenile sockeye salmon feed much more 

 heavily on insects (figs. 24 and 28). 



If pelagic sockeye salmon (and other com- 

 peting species) are cropping the zooplankton 



Oi- 



27 4 14 24 3 13 23 3 



JUNE JULY AUG. SEPT. 



Figure 29.— Numbers of zooplankters per liter of lake 

 water, averages of surface and 5-m. depths only, 

 June 27 to Sept. 3. 1957, Brooks Lake. 



population as the season progresses, plankton 

 could be a limiting factor under conditions 

 of high densities of sockeye salmon. The 

 plankton-feeding sockeye salmon contained 

 from 3,000 to 5,000 plankters, and one con- 

 tained more than 10,000. At an average density 

 of 40 zooplankters per liter in lake water, a 

 fish containing 10,000 plankters would have to 

 ingest the entire population from 250 liters 

 of water. 



Zooplankton depth distribution is charac- 

 terized by large fluctuations resulting from 

 diurnal vertical migrations, with least num- 

 bers near the lake's surface during midday 

 (Welch, 1935). All plankton samples at Brooks 

 Lake were collected around midday, and as 

 might be expected, abundance of copepods and 

 cladocerans was usually least at the surface 

 (fig. 30). On the other hand, all gill net 

 catches of juvenile sockeye salmon were at 



a. 



UJ 



Figure 30.— Daytime seasonal and vertical distribution 

 of zooplankton groups at three stations in Brooks 

 Lake, June 27 to September 3, 1957. Width of each 

 spherical curve corresponds to the cube root of 

 individuals per liter at the indicated depths. 



45 



