THE SUDDEN ORIGIN OF NEW TYPES 405 



Even at the present day the small, heterogeneous remnant 

 of the Urodele Amphibians shows considerable diversity and 

 variety with regard to the glands of the lateral line. In 

 some cases these glands secrete a milky juice, although this 

 appears to be of a poisonous nature. It is therefore conceivable 

 that in some of their remote ancestors corresponding glands 

 may have taken to secreting a milky fluid capable of nourishing 

 their young. As soon as this change of function had become 

 established, it would obviously be more convenient and 

 advantageous for the nutritive organs firstly to travel from 

 a dorso-lateral to a ventro-lateral position, and secondly to 

 become reduced and concentrated in a few centres along these 

 lines. 



It will now be evident that the same principle which has 

 been discussed in the case of the ancestry of flowering plants 

 can be held to apply to the origin of mammals. Thus we 

 find in the first place a lateral line (or more than one) composed 

 of a repetition-series of similar glands — a condition which is 

 in itself conducive to a high degree of variability of the 

 members of the series. Under special conditions one or more 

 of the variations might have progressed in the direction of 

 the glands producing a nutritive secretion ; when once this 

 initial step had been taken all the subsidiary characters of 

 the Mammalia could easily have followed as a natural con- 

 sequence. Among the chief of these was the increase of intellect, 

 which is well known to vary directly as the maternal sacrifice. 

 As a necessary corollar}^ to the absence of the lateral line 

 in all reptiles, it is evident that — contrary to the received 

 and prevalent opinion — the mammals must have taken their 

 origin directly from Amphibia, not from Anomodont reptiles, 

 which were already highly specialised creatures and should 

 be more naturally regarded as a closely parallel side-branch 

 from a common ancestral amphibian stock. The earliest 

 mammals must have been smaller and more generalised in 

 structure than the Anomodonts. Indeed most attempts at 

 representing the true phylogenetic relationships of animals 

 suffer from the stem-form not being placed far enough back 

 in the line of ancestry. On the other hand, the lateral line 

 was well developed in the adult Stegocephalan amphibians, 

 with the exception of the Aistopoda. 



As an instance of the manner in which reduction in one 



