1916 



Smith; on Salmon Development 



99 



About the first of April the fish were transferred from troughs A and 

 B to D and C respectively {Plate 17). They now occupied, in each trough, 

 sections 3, 4 and 5, the partitions separating these sections having been 

 removed. Those in C were in a white trough and kept constantly in the 

 light as they had been in B, while those in D were in a black trough and 

 covered from the light. 



With a view to obtain, if possible, a more accurate average weight 

 it was determined to weigh a larger number of fish from each group ; ac- 

 cordingly on April 6, 150 fish were taken at random from each trough 

 and after drying uniformly were carefully weighed. The comparative 

 weights are given in table 8. According to these figures the percentage 



Table 8. Comparative] :weights of fish taken Apr. 6 



Group 



No. of fish 



Fish from dark 

 Fish from light 



150 

 150 



Total weight 

 in grams 



127.0600 

 110.0500 



Average 



weight 



in grams 



.8471 

 .7336 



Difference in average weights 



.1135 



of difference is less than in the previous estimate being only 15.47 in favor 

 of the fish from the dark, over those from the light. 



Twenty-five fish were picked at random from each of these last groups 

 and measurements made. In table 9 are recorded the maximum and min- 

 imum dimensions between which the fish varied, together with the average 



Table 9. A comparison of the dimensions of 25 fish from each group 



group in table 8 



of each dimension. The figures show not only a very decided difference in 

 size between the maximum and minimum dimensions within each group, 

 but also indicate an increasing irregularity in growth, the largest from the 

 dark trough being 18.42 per cent longer than the smallest, and the largest 

 from the light 17.14 per cent longer than the smallest. 



Another estimate of relative weights and growth was made April 9, 

 again using large numbers of fish. The fish were placed in a four per 



