386 METAMORPHOSIS 



these species. In a word, look at it as we will, we are compelled to believe that 

 the specific characters of Draba verna cannot have been evolved in the struggle 

 for existence, that they are not adaptive characters, and that they are in them- 

 selves useless. Many sub-species in addition to those of Draba verna are 

 similarly gregarious, others, however, such as those studied by Wettstein, 

 exclude their near relations from their habitats (Wettstein, 1898, Grundzuge 

 d. geographisch-morphologischen Methode der Pflanzensystematik. Jena). What 

 we have said as to sub-species applies to species also of other plants. In all 

 large genera we find species which live in the same habitats, but whose specific 

 characters are in no ways adaptive. Examine the differences, for instance, which 

 exist between Avena elatior and A. pubescens which grow beside each other in 

 the same field ; it is impossible to regard these as adaptive characters ; still less 

 can we regard as adaptive the characters which distinguish genera, families, 

 and higher groups. We are inclined to hold that in most cases (p. 395) where 

 two forms are distinguished only by dissimilar adaptive characters, either we are 

 not dealing with genuine species or that their real differences have not yet been 

 discovered. In the former case we should be dealing with modifications due to 

 habitat only, such as we so often meet with in water and land, light and shade, 

 mountain and lowland forms ; they differ from genuine species in this, that their 

 characters are not hereditary, but disappear again when, or soon after, the in- 

 ducing factors are removed. 



At the same time we do not mean to affirm that adaptive characters cannot 

 be inherited. We shall return to that point later on, but meanwhile it may be 

 noted that there are obviously two quite distinct kinds of characters possessed 

 by species, adaptive characters and organic characters, a distinction which has 

 been emphasized with special clearness by Nageli (1884). This distinction may 

 be made out at each stage in the classificatory system. There are aquatic and 

 terrestrial Algae just as there are aquatic and terrestrial Phanerogams, and 

 both Angiosperms and ferns include both xerophytic and hygrophytic types. 

 In smaller groups also adaptive and organic characters may be distinguished. 

 The problem as to the origin of species is primarily the problem as to the origin 

 of definite organic characters ; however, since all species exhibit adaptive 

 characters as well, we must study their origin also. 



From what has been already said it must be admitted that Darwin's 

 principle of natural selection cannot explain the origin of species. Let us 

 see whether it is sufficient to explain adaptive characters. Here again, how- 

 ever, we meet with great difficulties. According to Darw^in, the differences 

 between competing individuals are not great but they become gradually inten- 

 sified by summation in the course of an indefinite number of generations. Con- 

 sider, for example, some peculiarity of a plant obviously of service to it, such as 

 the prickles of a rose which aid it in climbing, or the spines of a thistle which 

 protect it from the attacks of animals ; according to Darwin's theory these 

 prickles and spines began as excrescences of minimum height on a previously 

 smooth plant and attained their present structure and dimensions gradually. 

 Only after the organ had become sufficiently prominent, however, could it have 

 been of any use to the plant ; in a word, it is easy to understand how Darwin's 

 theory may explain the improvement of an organ already in existence, but it 

 does not make clear how it first arose. 



Having seen that the mode of operation of natural selection as defined by 

 Darwin does not afford a satisfactory explanation of adaptive, let alone specific, 

 characters, let us next inquire as to Darwin's interpretation of variation and of 

 heredity. It will be necessary to study these two questions in conjunction, 

 since the important point for consideration is whether or not variations are 

 inherited. Darwin assumed that every new character, however it arose, was 

 capable of being inherited, but this assumption has yet to be proved, and 

 especially in reference to each type of variation, of which we may distinguish 



