6o S. J. HOLT 



at recapture, which results in a straight line cutting the x axis at x = Lq^ and 

 having slope 



— Ky tanh & ~ — iC for small h 

 h 



where h = Kaji and a = time interval. 



It should be noted that it is equally applicable to the determination of 

 crustacean growth parameters from tagging experiments even when only the 

 increment at moulting and not the moulting frequency, is known. 



Now several experiments have shown that the presence of tags may affect 

 the growth offish, usually adversely. It seems however that a common effect 

 is to restrict feeding, and in this case it may be that L^o will be altered rather 

 than K. A review of existing data would probably clarify this point, but 

 unfortunately data for tagging experiments are rarely published in sufficient 

 detail for a full study of this kind. Further light could also be thrown on 

 the problem by a comparative analysis of data for growth of fish in aquaria 

 and ponds. It seems from preliminary analyses that although L^ may for a 

 particular species vary over a very wide range, estimates of K are obtained 

 which do not differ very greatly from those given by the natural curves of 

 growth in wild populations. 



MAXIMUM SIZES OF FISH IN CATCHES 



A possible approach to the estimation of asymptotic size might be the 

 examination of mean sizes of the largest fish in a series of samples. Certainly 

 one can obtain some idea of the order of magnitude ofL^^, but further than 

 that it is difficult to go because the sizes of the largest fish depend not only 

 on the sample size, but also of course on the mortahty rate (see Holt, 1957c). 

 It may be, however, that for certain purposes it would suffice to have 

 minimum estimates of Loo- A comparative study of samples from stocks for 

 which the growth parameters are known would be useful to give some idea 

 of the value of such crude estimates ofL^o. 



SIZE AT MATURITY 



It is commonly stated in the literature that many fish become mature at a size 

 which is some rather constant proportion of their final length (see discussion 

 by Svardson (1943)). The proportion usually quoted is about two-thirds, 

 and, if the weight of a fish is proportional to the cube of its length, this 

 corresponds with the inflexion point of the curve of growth in weight. This 

 observation suggests a study of sizes at maturity, which may be measured 

 by several methods in fish for which age determination is difficult — with 

 a view to their use as estimators of L^. 



