DISTRIBUTION AND GENERAL NOTES ON THE SPECIES 295 



spawning and {b) upon the extent to which spawning is Umited in time. On the assumption that 

 M. hubbsi conforms to this rule (as European hake are known to do), a purely hypothetical completion 

 of the harmonic curves has been shown by the pecked lines in Fig. 24, for that part of the seasonal 

 cycle for which we have no data. 



Obviously such curves must regain a similar level for the beginning of the next seasonal cycle, 

 and there are good grounds for belief that recovery after spawning is rapid, so that they will rise more 

 steeply in autumn than in winter. The hypothetical portions of the curves are therefore almost cer- 

 tainly more accurate than might seem probable at first sight. 



Before we proceed to a more detailed examination of the evidence provided by our ponderal indices, 

 and their shortcomings, it may be pointed out that the generalized curves shown in Fig. 24 provide 

 valid evidence of one important analogy with Hickling's findings on the European species— vahd 

 because it depends upon the portions of the curves derived from actual data and not from hypothesis. 

 It can be seen that while the fall in condition of males and of females in early summer is nearly 

 parallel, the females begin to recover first. The inference is that the spawning season of a majority of 

 males is longer than that of a majority of the female fish. The agreement with Hickling's (1930*. 

 pp. 33-4) observations on M. merliiccius seems complete. 



The generalized curves show a greater annual variation among males than among females. Since 

 eggs are larger than sperms it might appear that this is anomalous, but it is probable that it is not so, 

 for the following reasons: the smallest length classes of males contain a far higher proportion of 

 mature fish than corresponding length classes of females, and as the male growth rate falls off the 

 metabolic strain of spawning increases far more rapidly than in females (cf. Hicklmg, if)Zob, p. 36; 

 1933. PP- 43-4)- Consequently these curves show a greater annual variation among males just because 

 they are generalized: the data for males include a greater proportion of spawning fish. 



When we come to consider the seasonal variation in condition within individual length groups 

 certain undoubted anomalies appear. In the following table the data used to compute the relation 

 between K and length are also subdivided to show the means at mean dates within length classes. 

 Several untoward features are at once apparent. Males of the 21-30 cm. length class showed greater 

 seasonal variation than males of 31-4° cm., and females of the smaller length classes a greater seasonal 

 range than larger females. 



Table 14. Variation in ' average' K of Merluccius hubbsi in relation to length, and seasonal 



variation of K zvithin length classes 

 (The left-hand columns are the data of Fig. 23.) 



Length 

 class 



21-30 

 31-40 

 41-50 



'True' 

 mean 

 length 



27-4 

 35-7 

 43-1 



'Average' K 



in relation 



to I 



0-705 

 0-688 

 0-654 



0-678 

 0-695 

 0-685 

 0-649 

 0-628 

 o-6i8 



Males 



Females 



Mean K at mean dates within I classes 



31 Oct. 



0-791 

 0-731 

 0-751 



0-736 

 0-767 

 0-747 

 0-686 

 0-662 

 0-633 



16 Dec. 



0-705 

 0-685 

 0-657 



0-677 

 0-647 

 0-667 

 0-636 

 0-614 

 0-621 



24 Mar. 



0-630 

 0-647 

 0-555 



0-621 

 0-671 

 0-641 

 0-624 

 0-608 

 0-600 



