EXPERIMENTAL WORK.. 



205 



The temperatures of the air were read from a thermometer in 

 the centre of the hatching-house, generally about noon ; the tem- 

 peratures of the water were read from a thermometer, whose bulb 

 was immersed in the distributing-spout, and the temperature was 

 consequently that of the water before it reached the ova. The per- 

 centage of incubation was calculated from the temperature of the 

 water, read in fifths of a degree (the nearest whole degree only being 

 entered in the book afterwards), and a further correction was made 

 generally from two other observations, so that the percentage of 

 incubation is really calculated from a very approximate mean of 

 the temperature of the preceding twenty-four hours. The next 

 column is the addition of the percentage of all the preceding per- 

 centages, so that, by subtracting the total opposite the day on 

 which any particular batch of ova was laid down, its accomplished 

 percentage of the period of incubation is at once apparent. I 

 think it is best to generalise solely from the ova of the S. levenensis, 

 for this reason, all the spawners were netted in the lake at the 

 mouth of the South Queich, and were all mature fish of 1 Ib. and 

 over, while the burn trout were small and very much mixed. 



