66 S. J. HOLT 



factors. The evolutionary consequences of such mechanisms might be worth 

 further examination. 



CONCLUSION 



I hope I have said enough to illustrate my reasons for beHeving that compara- 

 tive studies of fish populations would be expected, even at this rather early 

 stage in our knowledge of the dynamics of exploited animal populations, to 

 give some practically useful results. Several kinds of comparison are possible 

 and would be interesting. They include comparisons between species, 

 considering systematic position; comparison between different stocks of the 

 same species; comparison between sexes; comparison between different year 

 classes or spawning classes of the same stock; comparison between individuals 

 in a particular stock having recognizably differing rates of growth and 

 maturity (as observed by back-calculating life-history from scales, etc.). 



Gross comparisons, as of total stock size in relation to systematic position 

 (e.g. Clupeoids seem to have larger populations than do Perciforms) might 

 be useful, but 'dissection' of the population is likely to be more fruitful and 

 less misleading, in the same way as internal anatomy is more useful than 

 general external form in comparing individual organisms. However, to 

 relate the mortality, maturity and growth parameters eventually to popula- 

 tion size it will be necessary to make comparisons, not attempted here, 

 between these parameters and the specific fecundity and recruitment rates. 

 Thus, in a recent review of hterature, Woodhead (i960) has drawn attention 

 not only to complex relations between feeding level, growth rate, and age 

 and size at maturity, but also to the fact that the number of eggs laid per 

 spawning (and also sometimes, the annual fecundity, through changes in the 

 frequency of spawning) are not simply proportional to the body size, but 

 depend on the previous rate of growth. 



Difficulties in making population comparisons arise from: absence of 

 standardization in presenting results of, for example, maturity determination; 

 and the problem of drawing up a set of parameters all of which have been, 

 or can be, estimated for many populations, taking account of environmental 

 differences. Other difficulties, very important in practice, arise out of the 

 need to use pubHshed data. These may be difficult to find in a scattered 

 literature and their use for purposes not intended by the authors may be 

 dangerous, especially when their true meaning in their original context is 

 misunderstood because that context was not fully defined. These difficulties 

 should not, however, deter fisheries biologists from attempting to formulate 

 hypotheses which unite into a single whole the considerable volume of 

 accumulated data pertaining to fish populations. 



