Mammals to Aquatic Life. 173 



peut ^tre consid^r^e corame un caract^re tout-^-fait primitif " 

 — I can bj no means agree to. 



To sum up our results. In the aquatic Mammalia there 

 sets in a retardation in the ossification of the skeleton of the 

 hand, a consequence of which is tlie formation of double 

 epiphyses in each finger-joint. Since each finger-joint thus 

 acquires three divisions, which, owing to reduction, that is the 

 loss of separate functions, become continually more alike, the 

 number of original finger-joints will be doubled. The number 

 of the secondary finger-joints will therefore in the highest 

 case amount to twelve. 



The question now suggests itself whether this process can 

 still be followed embryologically. It must be admitted at the 

 outset that this is doubtful, since this transformation took 

 place so long ago, and the principle of making the separate 

 parts resemble one another in form was so speedy in its effect, 

 that it seems practically hopeless to expect that we can still 

 trace the various stages in the development of the embryo. 

 I might express myself in the words of Pfitzner* : — " In osteo- 

 logy in particular is ontogeny an auxiliary as inefficient as it 

 is untrustworthy ; we have to rely almost entirely on com- 

 pai-ative anatomy and variation." All that we can determine 

 is that from an embryonic tissue there develop pieces of carti- 

 lage which are separate from one another and which correspond 

 to the secondary phalanges. There is no question of supple- 

 mentary division of the cartilaginous tissue ; all changes which 

 subsequently took place are traceable to fusion. The whole 

 of the secondary phalanges of the whale's hand are therefore 

 to be found already in their places. Any supplementary 

 division of the cartilage is consequently excluded. 



This method of formation renders it impossible to trace tlie 

 origin of hyperphalangy embryologically ; I would only 

 mention that in a very small embryo of Glohiocephalus there 

 is a considerable difference in size between the secondary 

 metacarpalia and the remaining divisions of the phalanges, in 

 that the former, particularly in the first finger, are very much 

 more minute ; in the next stages of development, however, 

 this difference is already obliterated. This difference in size 

 points to the mode of origin of the secondary metacarpal. 

 The fact that it still finds expression here is intelligible when 

 we reflect that the process of hyperphalangy must liave com- 

 menced at the tip of the finger, and with the division of tlie 

 original metacarpal has reached its latest stage. 



!Now if our theory is correct the number of phalanges should 



* W. rfitzner, " Die kleine Zehe," Archiv f. Anat. u, Physiol. 18!K), 

 pp. V2 et seq. 



