70 S. J. HOLT 



S. J. Holt: Some data for starved fish which were actually shrinking 

 in an aquarium gave a K value more or less like that in wild populations. 

 K seems not to vary much with feeding or habitat (except temperature) 

 — Loo however, varies greatly. 



L. B. Slobodkin: The constancy o£ M/K ratios seem to suggest that 

 predation is unimportant, and that the non-fishing mortality is basically a 

 physiological product. 



S. J. Holt: I think it is better to look at the question from an evolu- 

 tionary viewpoint. For survival of the population sufficient fish must reach 

 maturity. Even if M had been determined by predation, K might be adjusted 

 by natural selection so as to allow enough fish to reach the breeding state. 



L. B. Slobodkin: It would seem to imply that the rough level of 

 predation is the same around the world wherever the species is found, if K 

 is a species constant. 



S.J. Holt: But K does vary within a species — it is temperature depend- 

 ent. There is some evidence of the physiological determination of mortality 

 from comparative studies of the two sexes, in a few species. Where both 

 sexes have the same growth rates, there is normally no difference in mortality. 

 Normally, however, the K of males exceeds that of females, and in such 

 cases M of males also seems to be higher. This allows both sexes to be 

 considered together, provided that the sex ratio is not altered by selective 

 fishing. 



F. T. K. Pentelow: Does the diagram you have shown imply that 

 unmarked fish live longer ? 



S. J. Holt: No. Marked fish grow more slowly. There is sometimes 

 evidence of differential mortahty, but I have omitted it from this discussion. 



E. B. Worthington: The aim of this approach would seem to be to 

 give practical advice by extrapolation to parts of the world where data are 

 lacking. Has this in fact been done? 



S. J. Holt: No. The present position is one of intensive study of the 

 available data, to derive ideas about the steps which must next be taken. 

 Full details about well-studied species are essential. Only a little information 

 is available for tropical species. One can already see the utility of the method, 

 however, in helping to resolve uncertainties in the available data. For 

 example, the Tilapia species of the African lakes show scale rings which 

 might result from either annual or biennial spawning. If they are treated as 

 annual, a value of K is obtainable which may be a multiple or factor of the 

 true value: the graphs in fact suggest that the rings are six-monthly. 



