DEVELOPMENT OF THE CARAPACE OF THE CHELONIA. 341 



are bridged between by many intermediate processes. Thus the terminal phalanx of 

 the finger is an intra-cartilaginous bone, and developes in all respects like the middle 

 phalanx, except that its distal end shoots out towards the tip of the finger just like the 

 end. of the costal plate of the tortoise shoots out towards the marginal plate. We cannot, 

 however, consider the costal plates simply as ribs, for most would agree that the term 

 rib indicates a long cylindrical bone enclosed in a periosteum. We may, however, with 

 propriety consider the costal plate as a rib expansion. 



If, finally, we contrast the carapace of the Chelonia with the body- wall of a crocodile, 

 the following differences are apparent. In the crocodile we have segmentation into 

 alternate muscle plates and rib cartilages enclosed in periosteum, the outer layers of 

 which have differentiated already into fibrous tissue. These structures grow, the muscle 

 plates forming intercostal muscles, the periosteum being only osteogenetic on its inner 

 surface secreting a cylindrical rib. In the tortoise the carapace is segmented, but only 

 to the extent of having cartilaginous ribs within an otherwise undifferentiated body-wall. 

 The ribs are not ensheathed by any specially differentiated periosteal membrane, and so 

 the bone where it developes around the cartilages gradually comes to involve the whole 

 of the intercostal spaces. In the turtles, parts of the ribs, viz., those which join the 

 marginal plates, preserve their rib-like character. They have not expanded to form 

 true costal plates, because they are invested by a restraining periosteum, in which true 

 adult fibrous tissue has already formed. 



The study of Chelonia, in which the ossification of the carapace is incomplete, is very 

 instructive in this relationship. Owing to the lack of material I have not been able to 

 devote as much attention to this as I should have desired, but such observations as I 

 have been able to make are worth recording. 



If a young Horopas areolatus be dissected, one can remove the scutes and the 

 membranous carapace, leaving behind true bony vertebrse and ribs. I have made care- 

 ful sections of a Horopas, and find that the bony vertebrse and fully-formed bony ribs 

 show the merest indications of osseous extension into the surrounding carapace. I am 

 told that my specimen corresponded in point of general development with one of the 

 larger fresh-water tortoises I have figured, yet the condition of the carapace was very 

 different. The ribs were cylindrical bony tubes, with here and there indications of a 

 lateral expansion. They were embedded in a tissue which was no longer embryonic, it 

 was fairly differentiated fibrous tissue. The ribs were not enclosed in a tube of perios- 

 teum, hence the attempts at lateral expansion. They were surrounded, however, by 

 tissue which had already differentiated, and had therefore not the same tendency to be 

 involved in any osteogenetic changes. 



One may conclude with the following general statement. The chest-wall of a typical 

 vertebrate segments into rib cartilages, surrounded by perichondrial tubes, osteogenetic 

 on their inner surfaces alone, and muscle plates embedded in connective tissue. The 

 cylindrical ribs form on the inner surface of the growing perichondrium (periosteum), and 

 the muscle plates form the intercostal muscles. 



