EVOLUTION 963 



to be frequent in snapdragons. In familiar metaphor, the mutants 

 are simply trying to go one better than their parents. 



After a profound study of the garden races and wild species of 

 snapdragons, Baur has come to the conclusion that they have been 

 and are in many cases the result of the summation of small muta- 

 tions, comparable to those that are of everyday occurrence in the 

 garden. In natural conditions the summation may be put to the 

 credit of natural selection, which has sifted the mutations in refer- 

 ence to the diverse and changeful conditions of locality and climate. 

 This is an eminently Darwinian conclusion; for Darwin thought 

 much more of the creeping than of the leaping of the eternal Proteus. 

 But whereas Darwin was vague in regard to his raw material, of 

 "small variations", Baur is very precise in regard to his "small 

 mutations", except as regards their cause. 



This investigator, who confirms Darwin's belief in the cumulative 

 importance of little changes which we might call evolution- jerks, 

 is far from saying that these furnish the whole of the raw material 

 of progress. On the contrary, while he lays chief emphasis on minute 

 changes in the hereditary "factors" or "genes", changes which occur 

 abundantly even in "pure lines" (all descended from one parent), 

 he admits that new departures may arise by combinations of different 

 sets of factors in crossing different strains. But the existence of these 

 diiferent strains depends in snapdragons on the previous summing 

 up of small mutations. 



There are species of Rose with seven chromosomes or nuclear 

 rods in each cell — surely a suitable number for a perfect flower! 

 But there are five known species with 14 chromosomes, and others 

 with 28, 35, 42, 56, and so on. In the Rose, therefore, and in some 

 other genera, evolution has implied multiple combinations of differ- 

 ent sets of chromosomes. Thus there are several distinct 56-chromo- 

 some — "tetraploid" — species of Rose, which have arisen through 

 the combination of different "diploid" or 14-chromosome species. 

 This is a very interesting mode of evolution, of which we shall hear 

 much in years to come; but it cannot be the method observable in 

 snapdragons, since all the species of Antirrhinum that have been 

 studied have the same number of chromosomes, namely eight. 



But what, it may be asked, of the large mutations, the "transilient 

 variations" or freaks; do they not occur among snapdragons? 

 Assuredly they do, Baur answers; and it is to them that the experi- 

 mental gardener mainly devotes himself, making them the subject 

 of his artificial selection. His sieve is not fine enough in its mesh, 

 speaking metaphorically of course, for the small mutations on 

 which the sieve of Natural Selection operates. Conversely, the sports 

 or large mutations, which catch man's fancy and come under his 

 segis, tend to be eliminated by Natural Selection as too extreme. 

 The gist of the matter is that in snapdragons the majority of the 



