The Recapitulation Theory in Biology 31 



horses, tapirs, and rhinoceroses, permits of a comparison of 

 phylogenesis with the ontogenesis of the horns in bovine mam- 

 mals the conclusion is that ontogeny closely recapitulates 

 phylogeny, that the genesis is gradual or continuous, that the 

 horns arise definitely and determinately." 49 



As is well known the ancestry of the horse is one of the few 

 vertebrate ancestries that have been reconstructed on palaeon- 

 tological evidence, and it is therefore of peculiar interest to 

 note the results of a comparison of the embryogeny of the horse 

 with its racial history. This comparison has been made in part 

 by Ewart, who, in the course of his report, raises and answers 

 this question, 



"What evidence is afforded by the horse in favour of the re- 

 capitulation theory? 



Evidently if the reservations made on the plea of abbreviation 

 in development, etc., are almost unlimited, a very remote re- 

 semblance in an embryo to a supposed ancestor might be con- 

 sidered sufficient evidence of recapitulation. . . .There must, 

 however, be some limit to the use of the word recapitulation .... 

 Recapitulation, if it means anything at all, must be held to mean 

 that, say in the case of the horse, the embryo should after a time 

 not only resemble the unknown embryos of its extinct ancestors, 

 but that it should resemble the fully-developed ancestors. This 

 implies that during its ontogeny the horse should not only ad- 

 vance in a zigzag fashion along the trunk of the animal tree, but 

 also when it reached its own particular branch that it should 

 during its onward course make deflections in the direction of 

 Hyracotherium, Mesohippus, etc., until having, at a respectful 

 distance, done hurried homage to its ancestors, it at last reaches 

 its final goal, and presents its true characters .... It is impos- 

 sible with any degree of accuracy, either to speak of a Hyraco- 

 therium or a Mesohippus stage, or even of a Hipparion stage. 

 If, on the other hand, all that is meant by recapitulation is that 

 the developmental record of any given form is represented by 

 a series of zigzags or curves instead of a straight line, the 

 horse may be claimed as supporting it. As is to be expected 

 in forms that have evidently descended from the same ancestors, 

 there is a certain amount of agreement between horse embryos 

 and the so-called fossil horses; and were it possible to know the 

 development of these extinct forms, the points of agreement 

 would be doubtless increased. Further, as might also have been 

 expected, the points of agreement become more numerous and 

 more evident as Hipparion a not very ancient form is reached. 



Science, Vol. XXXV. 1912, pp. 595, 596. 



