108 BLASTOGENIC VARIATIONS. 



the same size as when they were fresh, even after they 

 had been kept 33 or 45 hours. The larvae from fresh 

 ova and stale sperm, on the other hand, were in each 

 case distinctly larger than the normal, they differing on 

 an average by + 5.8 per cent., whilst those from stale 

 ova and fresh sperm were distinctly smaller, they differ- 

 ing by - - 4.9 per cent. In one of these latter observa- 

 tions there was, for some unknown reason, a slight in- 

 crease in size, but there can be no doubt that on the 

 whole the tendency was towards diminution. On tak- 

 ing means of all the values obtained in this and in other 

 similar experiments, it was found that as an average of 

 eight observations, the stale 9 stale $ larvae were di- 

 minished by .7 per cent. in size; as an average of eleven 

 observations, the fresh $ stale $ larvae were increased 

 by 4.0 per cent., and as an average of ten observations, 

 the stale $ fresh $ larvae were diminished by 6.9 per 

 cent. There can be no doubt, therefore, that variations 

 in the degree of freshness of the sexual cells, that is to 

 say, in the comparative state of nutrition of the germ- 

 plasm as a whole, do have a very appreciable effect upon 

 the size of the subsequently developing larvae. It is to 

 be particularly noticed that the effect produced differs 

 entirely according to the sex-cells acted upon, and hence 

 affords distinct evidence of the possibility that different 

 portions of the same sex-cell may also react differently 

 to one and the same change of nutrition. 



Perhaps a more convincing proof of the influence of 



the nutritional condition of the sex-cells on the offspring 



they produce is afforded by certain other observations 



on these larvae. On two separate occasions * series of 



*Phil. Trans. 1895, B, p. 585, and 1898, B. p. 483. 



