CERTAIN LAWS OF VARIATION. 199 



which proved itself so favourable to the growth of 12 

 hours' blastulse, would correspond to a temperature of 

 23 acting on 4 hours' blastulse, and one of about 19 

 acting on ova shortly after impregnation. 



To return to our conclusion as to the diminishing 

 effect of temperature with development, it seems rea- 

 sonable to imagine that what is true for one environ- 

 mental condition is probably true for others also. Ob- 

 servations were in fact made to test the effects of a 

 change of salinity, and they also proved the ova to be 

 very much more sensitive in the earlier stages of their 

 growth than in the later.* Here again, however, a 

 double effect was produced, adverse for the first 

 few hours of exposure, and favourable for the later 

 ones. 



Due reflection will, I believe, incline one to infer that 

 what is true for Echinoid larvae is true for most multi- 

 cellular organisms. In fact, it would seem to be a law 

 of general application that the permanent effect of en- 

 vironment on the growth of a developing organism 

 diminishes rapidly and regularly 'from the time of im- 

 pregnation onwards. 



It is curious that this principle, enunciated by the 

 author in 1900, should have been laid down by De 

 Vries,t only a few months later, as the result of his ob- 

 servations on plants. Thus, judging from the effects 

 of nutrition (manuring, planting out, good light, and 

 watering), he concluded that: (1) The younger a plant 

 is, so much the greater is the influence of external con- 

 ditions on its variability. (2) The nutrition of the 



* Vide Proc. Roy. Soc., vol. Ixvii. p. 97, 1900. 

 f" Die Mutationstheorie," p. 373. 



