98 



SCIENCE 



[N. S. Vol. XLI. No. 1046 



that variation has not occurred merely by large 

 losses subsequently fractionated so as to form 

 intermediates. Not merely intermediates arise 

 but also those which transcend any known 

 original sports. Original black races have 

 become blacker ; original yellow races have be- 

 come yellower ; white-spotted races have become 

 more spotted still at the will of the breeder. 

 Large races also become larger, and small 

 races smaller, under the hands of the fancier. 

 He does not limit himself to the production of 

 intermediates. 



To suggest further that all variation tran- 

 scending limits previously existing is due to 

 loss of inhibitors and so is really retrogressive 

 is scarcely satisfying. It is a formal evasion 

 of the difSculty but in no sense a solution of 

 it. It belongs with the box-within-box idea 

 of development. I agree with Bateson that 

 variety formation within the higher animals 

 and plants seems to be very frequently by a 

 process of loss but I can not believe that this 

 is the exclusive process concerned in the for- 

 mation of new species or even of varieties. It 

 needs but to carry the idea to its logical con- 

 clusion to show its absurdity. Is man merely 

 an amoeba simplified by loss of inhibitors? I 

 can not believe so. I can not believe that 

 the original proteid molecule has since its 

 original synthesis only grown simpler. New 

 radicles have undoubtedly become attached to 

 it as side-chains replacing or supplementing 

 old ones and changing its properties. The liv- 

 ing substance is not merely losing constitu- 

 ents; it is also gaining new ones. Similarly 

 organisms, morphologically and physiologically, 

 change not merely by losses but also by gains. 

 It is impossible to explain evolution satis- 

 factorily by either process alone. The two go 

 hand-in-hand and no doubt are constantly 

 occurring among organisms. Change is uni- 

 versal. Mere subdivision of a species into 

 two groups of individuals, which are pre- 

 vented from intermingling, seems to be suffi- 

 cient in time to make the two groups speci- 

 fically distinct. Each keeps on changing in 

 so many different ways that it would be little 

 short of a miracle if both changed similarly 

 and simultaneously in all respects. Direct en- 



vironmental effects are insufficient to account 

 for such organic changes, for among the best- 

 knovm illustrations of divergent evolution are 

 the animals of oceanic islands, close together 

 and subjected to the same climatic agencies, 

 undoubtedly descended from common ances- 

 tors at no remote period, yet having become 

 distinct, probably through numerous spontane- 

 ous changes which isolation prevented from 

 being ground dovpn to a common level by 

 inter-crossing. 



These are commonplaces of evolutionary 

 knowledge, familiar to everyone since Darwin 

 and Wallace first called attention to them, 

 yet we are in danger of overlooking them for 

 the moment in our enthusiasm over a new 

 method of attacking the obstinate problems of 

 evolution. It may not be superfluous there- 

 fore to call renewed attention to them in this 

 connection. Spontaneous variation is stiU 

 with us and is as widespread as it was in 

 Darwin's time. It is doubtful whether un- 

 varying " completely homozygous " organisms 

 occur anywhere outside the text-books. In 

 the case of organisms known to be varying 

 genetically there is abundant evidence that 

 small variations are heritable no less than 

 large ones, and we are by no means " done with 

 the idea " that small variations are capable of 

 summation. 



With Bateson we must deplore the necessity 

 of engaging merely in destructive criticism. 

 It is indeed " a low kind of work." It would 

 be so much easier, pleasanter, and more satis- 

 fying to adopt a single explanatory principle 

 for evolution and build on this. But it would 

 be foolish to go on building lofty superstruc- 

 tures of hypothesis on an insecure foundation, 

 and the more carefully we scrutinize the muta- 

 tion theory the more serious do our doubts 

 become whether it is a secure foundation for 

 evolutionists to build on. W. E. Castle 



BtJssEY Institution, 

 FoKEST Hills, Mass., 

 December 12, 1914 



MASTODON TUSK IN GLACIAL GRAVELS 



To THE Editor or Science : A tusk of a pro- 

 boscidean, probably Mastodon americanus, waa 



