of the Vertebrate Skeleton. 41 



upon growth is only seen in the great length of the ribs. 

 Now, if the body were stiifer in the middle, and flexible chiefly 

 in the neck and tail, then, instead of intermittent pressm-e 

 being distributed uniformly, it would be manifested chiefly at 

 the two extremities of the stifter part, which, touching the 

 ground, would be lifted by the movements of the head and 

 tail. If, then, a large part of the pressure and tension which, 

 distributed over the body, elongate the ribs of Ophidians, were 

 accumulated in this or some such way (by movement of the 

 body) at these points, whatever osseous structures pre-existed 

 there would grow ; and potential growth would tend to make 

 the parts at the anterior end of the body correspond with those 

 at the posterior end. What parts, then, would there be exist- 

 ing in such places ? Clearly some element of the abdominal 

 rib — elements, it may be presumed, which become the coracoid 

 bones and the iscliia. As the ribs become segmented into a 

 number of parts in different animals, it is not easy to guess 

 how many were developed ; but as the facts of the case only 

 require two (coracoid and scapula, and ilium and ischium), 

 these may be presumed to be the second and third segments 

 of the rib. Now the consequence of setting up a special ten- 

 dency to grow in these elements can in no Avay interfere with the 

 groAvth of the original rib, which, being joined to these hsemal 

 elements by overlap and by muscles, would, I suppose, slide 

 over the outside of these new growths, which would extend 

 inside of it. And I should regard the epi]ileuron as eventually 

 forming the clavicle and the pubis, while the suprascapular 

 is an effort of potential growth to reproduce the original rib 

 from which the arch-elements 'have become detached. 



But how account for the limbs ? Did they spring into ex- 

 istence ready formed, or grow gradually ? and, in cither case, 

 hoAV ? I cannot but be impressed with tlie forked character of 

 the limb, dividing in its second segment, as reproducing the 

 forked character of the visceral arches of the cranium and of 

 the vertebrse ; and therefore I believe that, in the absence 

 of any other evidence of a distal osseous fork, we can 

 only look for the proximal element of a limb in the proximal 

 element of a rib. And so I conceive that the increased mus- 

 cular ])ower of the pectoral or pelvic girdle might detach the 

 proximal part of the rib from its attachment with the vertebra 

 and draw it on to the already expanded hasmal elements — and 

 that potential growth, such as reproduces the lizard's tail and 

 the salamander's legs, would cause its distal segments to be 

 developed anew at the distal end, although the proper distal 

 segments now gave attachment to the proximal end. With 

 the bone would necessarily follow the muscles ; and potentially 



