10 Investigations on the Life-ffisfory 



TABLE I. 

 WKIGHT PER FISH OF STANDARD LENGTH FEMALE SALMON. 



The figures given in the last column show : 



1. That the difference in the average weight, per fish of standard 

 length, in July and August on five different parts of the coast does not 

 exceed 4 per cent. 



2. That in October and November the variation in the average 

 weight, per fish of standard length, from the lower waters of the Tweed 

 and Spey respectively is less than ] | per cent. 



The large number of fish dealt with, the wide area over which they 

 were taken, and the very slight variation in the results, seem collectively 

 to afford good grounds for believing that these figures represent with a 

 considerable degree of accuracy the condition of female salmon coming in 

 from the sea in each of the periods specified. 



A comparison of these averages with those of the estuary fish examined 

 by Dr. Noel Baton (which are given in Table IV., p. 66), shows that 

 the average weight per fish of standard length of the latter is in 

 May and June 5-3 per cent, below the former, in July and August 7'3 

 per cent, below, and in October and November not quite 1 per cent. 

 above. The smallness of the differences in these percentages would seem 

 to indicate that the fish examined by Dr. Noel Paton may be token as 

 tair samples of fish coming in from the sea in the periods named, and, in 

 any case, that the excess of weight, per fish of standard length, of the 

 mouth fish as compared with the upper-water fish is not due to his 

 samples of the former being fish in exceptionally good condition. 



A similar calculation has been made with "regard to males, and is 

 given in the following table : 



n pro^rliontoTheir T^ "'Th ^^^ T* nleasured whe dive. Their weight, therefore, 

 "terbe^tkine^l S Fgrea ' rt a " that f fish which were weighwl a fewboun 



