De \'ries : Age and Area 



169 



ception. We are not obliged to regard 

 a new species as coming into existence 

 at the expense of its ancestors, as was 

 assumed under the theory of natural 

 selection. The areas now occupied are 

 large enough to ofifer sufficient room 

 to newcomers, provided these are as 

 well suited to the obtaining local con- 

 ditions as the previous inhabitants. 



If the appearance of a new form 

 does not imply the disappearance of its 

 ancestors, the whole, or at least a great 

 part of the family tree of existing spe- 

 cies should still be surviving. It is 

 commonly assumed that such pedigrees 

 are often only incompletely represented 

 by the existing forms and that many 

 links have disappeared during the geo- 

 logical periods. Discontinuity in a 

 pedigree could always be easily ex- 

 plained in this way. The conclusions of 

 Willis, however, no longer allow the 

 constant use of this method. If, as a 

 rule, species survive beside the new ones 

 evolved from them, there is no good 

 reason to assume gaps of any impor- 

 tance between the members of any 

 given family. The pedigree of the 

 whole should still be represented in a 

 practically continuous way by the total- 

 ity of the living species. This is a 

 simple and clear principle, which may 

 free the systematists from superfluous 

 hypotheses and at the same time make 

 their conclusions concerning the affini- 

 ties of the genera and species within a 

 familv more convincing and more re- 

 liable. 



The Evolution of Species 



Species are thus assumed to have 

 evolved from one another by means of 

 small steps and not by the slow and 

 gradual accumulation of infinitesimal 

 variations. It is now generally con- 

 ceded that this latter theory can by no 

 means explain the first beginning of the 

 process. This view supposed the se- 

 lection to take place as a result of some 

 slight advantageous response to sur- 

 rounding conditions. But the first 

 steps cannot have had such a decisive 

 degree of utility, even if the whole 



specific difference could be shown to 

 possess it. In the differentiation of 

 new species the steps must therefore 

 have been of some definite importance. 

 But how large have they been? This 

 question has, at the present time, quite 

 another meaning than it had half a 

 century ago. We are now accustomed 

 to consider the qualities of organisms 

 as built up of units, which are usually 

 designed as hereditary factors or sim- 

 ply as factors. In experimental work 

 we rarely know whether a factor is 

 really simple or still a compound en- 

 tity, since in so many cases factors 

 originally assvmied as simple have been 

 split up into minor units in the course 

 of further researches. Leaving this 

 difficulty aside, the question remains 

 whether the steps which have produced 

 the existing species are to be considered 

 as involving each a single factor or a 

 small group of such. 



Willis replies that such changes may 

 at times occur of the necessary size to 

 give rise at once to Linnean species. 

 The elementary species of Draba verna 

 and of numerous other plants, described 

 by Jordan, are often looked upon as 

 constituting the steps by which species 

 have been evolved in nature. In order to 

 get real Linnean species we could sup- 

 pose large numbers of the small species 

 of Draba verna to die out, since the 

 differences between the remaining ones 

 would then evidently increase, and 

 might easily reach the degree necessary 

 for them to be admitted as good species. 

 But there is no good reason for the 

 theory of dying out, and the discoveries 

 of Willis show such a process to be 

 rather an exception in nature than a 

 general rule. The Jordanian varieties 

 show the same phenomena of dispersal 

 as do ordinary Linnean species. Often 

 they occupy as large areas, while they 

 still remain true-breeding, and show no 

 sign of variation. Jordan himself has 

 always described them as members of 

 well known good species, not as inter- 

 mediates between such. Evidently they 

 do not constitute the steps which pre- 

 viously might have linked existing spe- 



