WESTERNHAGEN and ROSENTHAL: PREDATOR-PREY RELATIONSHIP 



5.0 

 1 larva /500ml 



</ZZl7i 



12 4 8 



Number of Hyperoche/500ml 



12 U 



Num ber of Hyperoche/ 500ml 



Figure 2.-Mean number of yolk-sac larvae of Clupea harengus pallasi attacked by Hyperoche medusarum after different exposure 



times. Water temperature 9°C. 



does not necessarily lead to a corresponding in- 

 crease in predation. At herring larvae densities of 

 5/500 ml and 10/500 ml, one individual hyperiid 

 attacked 0.1 larvae/h and 0.16 larvae/h, respec- 

 tively. At 50 larvae/500 ml the attack rate was 0.45 

 larvae/h. Assuming a linear increase in attack 

 rate, we would have expected rates of 1.0 and 0.8 

 larvae/h. 



Alterations in predation rates of Hyperoche 

 were obtained when heterogenous prey was 

 offered (25 herring larvae -i- 25 flatfish larvae), and 

 Figure 3 shows that predation on larvae was 

 remarkably reduced. Of the 0.07 larvae attacked 

 per hour by one hyperiid, 0.055 (78%) were herring 

 larvae and 0.015 (22%) flatfish larvae, thereby 

 showing a pronounced preference for herring. 



DISCUSSION 



Figure 1 shows a clear, direct relationship 

 between number of attacked larvae and both 



larval and hyperiid density. Increase in larval as 

 well as predator density lead to increasing attack 

 rates per hour. Because searching and contacting 

 are random, this response was expected and has 

 been described by Murdoch (1971) for predator- 

 prey interaction. That relatively more larvae are 

 attacked per hour during short exposure periods 

 than during long ones (Figure 2) can be partially 

 explained by a rapid thinning out eff'ect on prey in 

 confined containers, a problem discussed by Mur- 

 doch (1969) for the predation of Thais and 

 Acanthina on Mytilus and Balanus. These data 

 suggest that short observation periods are prefer- 

 able in experiments of this type, a point 

 frequently neglected in experiments with expo- 

 sure times of 20 and more hours (Lillelund 1967; 

 Lillelund and Lasker 1971; Theilacker and Lasker 

 1974; Ambler and Frost 1974), leading to an under- 

 estimate of the actual possible predation rate. 

 An additional factor may be the degree of satia- 

 tion, which could be shown for invertebrates to 



671 



