116 THE MAMMALIA OF THE DEEP EIVEK BEDS. 



more nearly allied any two organisms are, the more likely will they he to independ- 

 ently acquire similar modifications. For development is the resultant or outcome of 

 the interaction of two great groups of factors, viz., the nature of the organism and 

 the character of the environment, and obviously the more nearly alike these two 

 classes of factors are, the greater the similarity to be expected in the result. To use 

 Darwin's words: "The members of the same class, although only distantly allied, 

 have inherited so much in common in their constitution that they are apt to vary 

 under similar exciting causes in a similar manner; and this would obviously aid in 

 the acquirement through natural selection of parts or organs strikingly like each 

 other, independently of their direct inheritance from a common progenitor." * Thus 

 there are certain characters which have repeatedly arisen in the artiodactyls, but are 

 not known outside of that group. An example of this is given by the teeth ; I have 

 elsewhere shown that the selenodont molar has been, in all probability, independently 

 acquired by at least three very distinct phyla, not including others in which the 

 molar pattern is slightly aberrant from the typical four crescent plan, but outside of 

 the suborder no tooth is known which presents more than a superficial resemblance to 

 this pattern. That this is not, however, the limit of the process, is shown by the spout- 

 shaped odontoid process of the axis in many ungulates, the bicipital tubercle and double 

 bicipital groove of the humerus in the horse and camel, and the numerous resem- 

 blances between artiodactyls and perissodactyls which have not been inherited from 

 their common ancestors, the Condylarthra. Indeed we are as yet very far from being 

 able to set a limit to the possibilities of this mode of development, not to mention at 

 all the phenomena of convergence. What is here contended for is the principle (at 

 tirst sight the most obvious truism) that numerous and close resemblances of struc- 

 ture are prima facie evidence of relationship, even though many of these resem- 

 blances be due to parallelism, and further, that such parallelisms, when properly 

 understood, may be of great value in morphological and phylogenetic speculation. 

 Even when we exclude Ancliitherium (and very probabl} 7 the same reasoning will 

 apply to Hipparion) from the direct line of equine descent, we find that most of the 

 conclusions as to the steps of modification in this line which were deduced from the 

 older hypothesis (except, of course, so much of it as referred to the relationships of 

 JPalceotJierium) are still valid and need to be revised only in comparatively few de- 

 tails. Should it prove to be the case that A. aurelianense and A. equinum have no 

 nearer connection than through some species of Mioldppus, which was ancestral to 



* My attention was called to this passage by its quotation in Prof. H. T. Fernald's paper on " The Rela- 

 tionships of Arthropods" (Baltimore, 1890), which the author has kindly sent me. I have been much interested 

 to see how well Fernald's results in Arthropods agree with rny own in mammals. 



