PREDATOR-PREY RELATIONSHIP BETWEEN PACIFIC HERRING, 



CLUPEA HARENGUS PALLASI, LARVAE AND 



A PREDATORY HYPERIID AMPHIPOD, HYPEROCHE MEDUSARUM^ 



Hein von Westernhagen and Harald Rosenthal- 



ABSTRACT 



Predatory efficiency of Hyperoche medusarum (Hyperiida, Amphipoda) on yolk-sac larvae of Pacific 

 herring, Clupea harengus pallasi, was studied in the laboratory under continuous light conditions: 1, 5, 

 10, and 50 herring larvae were exposed to 1, 2, 4, 8, and 16 hyperiids in 500-ml beakers. It was found that 

 the number of attacked larvae per unit time increased with rising predatory and/or prey density. 

 Individual mean predation rate was found to decline with increasing predator as well as prey densities, 

 prolonged exposure times, and the presence of alternative prey. 



Aside from starvation (Sette 1943; Schnack 1972), 

 one major cause of mortality in marine fish larvae 

 is assumed to be predation (Stevenson 1962), the 

 predators frequently being crustaceans, as de- 

 scribed by Garstang (1900), Lebour (1925), Davis 

 (1959), Lillelund (1967), Rosenthal (1967), Kabata 

 (1970), Lillelund and Lasker (1971), Theilacker and 

 Lasker (1974), and others. The pelagic hyperiid 

 amphipod Hyperoche medusarum occurs com- 

 monly off the Oregon coast (Lorz and Pearcy 1975), 

 in Californian waters (Hurley 1956), in the North 

 Atlantic (Shoemaker 1930; Bowman et al. 1963; 

 Dunbar 1963), in the North Sea (Sars 1895; Evans 

 and Sheader 1972), and in New Zealand waters 

 (Hurley 1955). In British Columbia waters it occurs 

 commonly in the upper layers (<30 m) of the water 

 column (Bowman 1953), and in Departure Bay 

 (Vancouver Island) its juveniles are frequently 

 found clinging to the exumbrellae of hydromedu- 

 sae (Westernhagen 1976). 



The cooccurrence of large numbers of juvenile 

 H. medusarum with newly hatched larvae of the 

 Pacific herring, Clupea harengus pallasi, was 

 incidentally discovered in 1974 at the pier of the 

 Pacific Biological Station, Departure Bay. Field 

 observations indicated that Hyperoche juveniles 

 preyed on herring larvae and occasionally on other 

 fish larvae. Since this was the first record on a 

 possible predator-prey relationship between H. 



^This study was sponsored by the international bureau of the 

 Gesellschaft ftir Kernenergieverwertung in Schiffbau and 

 Schiffahrt in connection with the German-Canadian agreement 

 on scientific and technical cooperation. 



-Biologische Anstalt Helgoland (Zentrale, 2 Hamburg 50, 

 Palmaille 9, Germany (Federal Republic of Germany)). 



medusarum and marine fish larvae, this study was 

 initiated to shed some light on the predatory 

 efficiency of this amphipod. 



MATERIAL AND METHODS 



For prey, yolk-sac larvae (8.0-9.5 mm TL (total 

 length)) of the Pacific herring incubated in the 

 laboratory were used. Immature H. medusarum 

 (1.48-1.80 mm TL) which had aggregated beneath 

 a light at night were caught with a pail and 

 separated from other plankton organisms with a 

 large bore pipette. 



Experiments were performed in filtered 

 seawater in 500-ml beakers (salinity 28''/oo; tem- 

 perature 9°C; constant light). The water surface of 

 the beakers was covered with 300-/xm mesh size 

 nylon gauze in order to keep the amphipods from 

 breaking through the surface. Because Hyperoche 

 specimens in their natural habitat were occasion- 

 ally found resting on the exumbrellae of medusae, 

 a strip of nylon gauze (50x20 mm) hanging from 

 the surface cover provided attachment for the 

 amphipods when needed. 



Different numbers of herring larvae 1, 5, 10, and 

 50 were exposed to 1, 2, 4, 8, and 16 hyperiids for 

 three exposure periods (2, 4, and 8 h). The number 

 of replicates for all predator/prey ratios were 4, 6, 

 and 5 for the 2-, 4-, and 8-h exposure periods. Some 

 additional experiments with 6- and 10-h exposure 

 periods were used for the computation of a mean 

 attack rate on the basis of HI h of observation. 

 Eleven trials using 25 herring and 25 flatfish 

 larvae with 16 amphipods were also conducted. 

 One additional control vessel (50 herring larvae, no 



Manuscript accepted February 1976. 

 FISHERY BULLETIN: VOL 74, NO. 3, 1976. 



669 



