IV SIGNIFICANCE OF EXPERIMENTS 81 



seeds in the Alps again — this on the additional ground that 

 the advance would be useful also in the Alps. But it is 

 otherwise when, as was done by Nageli, plants growing on 

 the Alps (Hieracium) are transplanted to the Botanical Garden 

 of Munich, and their offspring are reared in gravelly soil (as 

 a substitute for the conditions of their Alpine home). That 

 these offspring, deprived of the rich garden soil, revert to the 

 original form seems to me less surprising than the reversion 

 of all the garden plants everywhere artificially cultivated for 

 a much longer time as soon as they grow in a wild condition ; 

 and thus Nageli's experiments, on which he lays so nuicli 

 stress, seem to me to prove nothing new against the modifica- 

 tion of forms by external influences. 



The effect of the gradual withdrawal of water upon the 

 Axolotl is a case in which art hits upon and hastens the 

 movement of evolution already in progress. Likewise, un- 

 doubtedly, the action of water of greater or less saltness on 

 the crustacean Artemia salina, shortly to be considered more 

 closely. It is of course also possible that one part of this 

 latter case is an instance of reversion, for examples show that 

 when animals are transferred to the life-conditions of their 

 nearest ancestral form, or retained in those conditions from 

 the young state onwards, they retain the characters of the 

 ancestral form which they possess as larva3. Thus we are 

 able to prevent Salamanders from losing their gills and giving 

 up branchial respiration by compelling them to remain in water 

 all their lives, and so cause them to remain similar to their 

 ancestors the Perennibranchiata, because their other characters 

 also, through correlation, remain permanently at the lower stage. 



Very pretty experiments in relation to this point have been 

 made by Fraulein von Chauvin ^ upon the Alpine Salamander, 

 Salamandra atra. The gill-bearing larva? of this animal 

 were taken out of the oviduct of the mother and put into 



1 Zeitschr. filr loiss. Zoologie, Bd. xxix. 

 G 



