196 CERTAIN LAWS OF VARIATION. 



opment than during the later. In the foregoing 

 table are collected all the observations made by both 

 methods, the mean effect above mentioned as being 

 produced by an hour's exposure to a temperature of 8 

 at the time of impregnation being also included. We 

 accordingly see that the unfavourable influence of the 

 cold steadily diminishes from the time of impregnation 

 up to the 15th hour; whilst in the other series of obser- 

 vations the favourable influence of the warmth also 

 diminishes rapidly, though not so regularly. In one or 

 two cases, for some unknown reason, apparently no 

 effect was produced; but allowing for these by taking 

 means of the values obtained at more or less similar 

 hours, we still find a rapid and fairly regular decrease. 

 Thus, at about the 120th hour the average effect pro- 

 duced was only about a fiftieth of that at the fifth hour. 

 In still a third series of observations, some of the ova 

 were kept for various periods at about 26 C. Dur- 

 ing the first few hours of development this tempera- 

 ture had an extremely unfavourable action, produc- 

 ing a diminution in size varying from 20.76 to 7.36 per 

 cent. Still further exposure to it, however, produced 

 a favourable effect on growth, though this was never 

 sufficient entirely to counteract the previous adverse in- 

 fluence. The results obtained in two of the experi- 

 ments made are given in the table below. Here we 

 see that from the end of the first to the end 

 of the fourth hour average diminutions of 3.89 and 

 6.45 per cent, per hour were produced. During the 

 next four hours this adverse effect dwindled down to 

 nothing, or almost nothing, and in the next four hours 

 a distinctly positive effect set in. This probably con- 



