568 



NATURE 



[Oct. lo, 1889 



removed the primary sifting of young forms. This sifting 

 must occur under natural conditions, so as to allow only a 

 limited range of variations to reach the collector in his 

 museum. Clearly enough, this primary sifting, and all 

 later operations of the same kind due to natural conditions, 

 may under new circumstances be vastly modified in their 

 nature, and variations may be allowed to pass the sieve 

 which at another time are excluded. The range of varia- 

 tion, therefore, in even a very large museum series of a 

 wild species, can afford but an inadequate notion of the 

 variability of animals. We may, however, justly conclude 

 that, if the former is so large as Mr. Wallace shows it to 

 be, the congenital variations which occur, but never in 

 given conditions reach maturity, must comprise instances 

 which are very much more marked, and would furnish 

 abundant material for natural selection were the natural 

 conditions of the species to change. An attempt to de- 

 termine by experimental rearing, the range of congenital 

 variation (that is, oi possible adult variation) in such 

 animals and plants as are fitted for the inquiry, seems to 

 be well worth making. 



Mr. Wallace, who must have watched the early cri- 

 ticism of Darwin's theory with special keenness, makes 

 a good point when he insists that the objection that it is 

 difficult " to imagine a reason why variations tending in 

 an infinitesimal degree in any special direction should be 

 preserved" is a quibble. Darwin never used the word 

 " infinitesimal," but spoke of variations being " shght" or 

 of " small amount," and we agree with Mr. Wallace that 

 even those terms are open to the objection that they may 

 seem to imply that congenital variation is of less range 

 and frequency than it really is. 



Naturally enough, Mr. Wallace is not equally thorough 

 in his treatment of each of the various " difficulties and 

 objections" which he discusses, but the chapter thus 

 headed gives an interesting summary of the present state 

 of opinion. Among the matters discussed are the sup- 

 posed smallness of variations, the doubt as to the right 

 variations occurring when required, the beginnings of 

 important organs, useless or non-adaptive characters, 

 the instability of non-adaptive characters, the swamping 

 effects of intercrossing, and the effects of isolation. In 

 some of these instances Mr. Wallace's reasoning is very 

 clear and forcible ; in other cases it is much less so. 

 Mr. Cunningham has already pointed out, in a letter to 

 Nature (July 25, p. 297) a curious slip on Mr. Wal- 

 lace's part in his explanation of the gradual development 

 of the twisted condition of the head and eyes of flat-fish. 

 Mr. Wallace declines to admit the transmission of ac- 

 quired characters as a cause of variation and progressive 

 development ; yet, apparently without being conscious of 

 it, he attributes the movement of the eye of flat-fish from 

 one side of the head to the other, to the transmission of 

 a series of slight shiftings of the eye acquired in succes- 

 sive generations by the muscular effort of the ancestors 

 of our present flat-fish, which is (to use an expression 

 already known to the readers of Nature) " flat Lamarck- 

 ism." In relation to this, I may mention that the asym- 

 metry of the Gastropod Mollusca, the forward position 

 of the anus, and the twisted condition of the nerve-loop 

 in the Streptoneurous division of that class, had been 

 similarly attributed by myself to the cumulative effect of 

 a mechanical cause— the one-sided lopping of the shell- 



operating in successive generations. Like Mr. Wallace, 

 I had failed to notice that the explanation adopted was 

 an admission of Lamarckism. It seems to me possible to 

 explain the position of the flat-fish's eye by the selection 

 of congenital variations, since there is no doubt of the 

 advantage to the animal of having its two eyes on the 

 one side of the body. But I confess that the Gastropods 

 at present have not been satisfactorily explained. I have 

 not been able at present (and I say at present advisedly) 

 to find any evidence of advantage to the Gastropod in 

 the torsion of its visceral hump, such as would justify 

 the supposition that a monstrosity presenting this con- 

 dition in full development was favoured by natural selec- 

 tion ; still less does it appear how the steps of a gradual 

 torsion— that is, a series of approximations to complete 

 torsion— could be advantageous. It does not follow that 

 we must [admit Lamarckism ; but merely that we must 

 further examine Gastropod habits, structure, and deve- 

 lopment with this problem in mind. 



Mr. Wallace does not, in my judgment, give sufficien 



grounds for rejecting the proposition which he indicate 



as the main point of Mr. Gulick's valuable essay oi 



" Divergent Evolution through Cumulative Segregation.' 



By the bye, Mr. Gulick is one of the heretics who attribut 



some part in the production of species to other cause 



than natural selection, yet he is not a laboratory naturalist 



but one who, substituting land-shells for butterflies, ha 



precisely the same foundations and training as Ml 



Wallace himself. Mr. Gulick's idea is that there is an in 



herent tendency to variation in certain divergent lines, an 



that when one portion of a species is isolated, even thoug 



under identical conditions, that tendency sets up a dive* 



gence, which carries that portion further and further awa 



from 'the original species ; or, in other words, no tw 



portions of a species possess exactly the same averaf 



character, and the initial differences will, if the individua 



of the two groups are kept from intercrossing, asse 



themselves continuously by heredity in such a way as 1 



insure an increasing divergence of the forms belongin 



to the two groups, amounting to what is recognized 



specific distinction. Mr. Gulick's idea is simply the reco} 



nition of a permanence or persistency in heredity, whid 



ceteris paribus, gives a twist or direction to the variatiot 



of the descendants of one individual as compared wi' 



the descendants of another. Ireland is cited by ^ 



Wallace as an evidence that isolation has not been effe 



tive in modifying specific character of plants and anima 



If, however, unlike Mr. Wallace, we may look upon ma 



kind as subject to the same developmental causes, a 



only to the same causes, as animals, then Ireland woi 



seem to be a very interesting case of the production 



divergent character by isolation. All parties are agre 



that, whatever value is to be assigned to the fact, 1 



human inhabitants of Ireland, whether of Celtic or TeutJ 



ancestry, exhibit characters which are "divergent" 



those of the inhabitants of Great Britain, and, witl 



going into details, we may say that the isolation 



persistence of an original tendency seem to be the 



explanation of the divergence. , 



The subject of " correlated variations " is but h^ 



touched on by Mr. Wallace, and its immense import* 



in relation to the whole question of "useless organs "f; 



useless characters of growth and structure is not suffici© 



I 



