36 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 57. 



Now, no one would ever think of speaking of these hyo-mandibiilar 

 elements as membrane bones, yet it would appear that their ossifi- 

 cation takes place largely if not exohisivel}'' through the iniluence of 

 the perichondrial membrane surroimding them. In the case of 

 Gymnura and the kangaroo the ossification of the premallear ele- 

 ment produces a solid, more or less thickened, three-cornered bony 

 nodule, and this is to a certain extent true of this region of Erinaceus. 

 In Choloepus again tliese elements are seen developing in and around 

 the thick cartilaginous mold by which they are preceded, in a manner 

 not at all dissimilar to that of tlie incus and malleus in the same 

 stage of development, and what is here said of the Unau applies 

 equally well to all other forms in which it is found. That the ossifi- 

 cation of this element or these elements was originally the same and 

 in no manner different from that of the incus and malleus, I do not 

 think that there can be any reason to doubt, and if there are any dif- 

 ferences at the present time they are due wholly to the fact that the 

 latter are now functional elements still retaining their vitality and 

 vigor, widle the former are inconsiderable degenerate vestiges prac- 

 .tically on the point of disappearance. If it can ever be shown that 

 the cartilage itself is in the least involved in the process of ossifica- 

 tion of tliese elements, in any of the species in which it exists, we can 

 then conclude beyond any possible fear of error that it is a true car- 

 tilage bone and undoubtedly represents some missing element which 

 originally pertained to the reptilian or batrachian suspensorium. 



In all those cases wherein these elements appear in the premallear 

 tract of the developing embryo, whether they be one or more than 

 one, and continue into the half-grown or adult stages, it will be 

 observed that the position is always the same, namely, to the inner 

 side of the glenoid cavity and slightly posterior to it, passing around 

 the inner edge of the tympanic ring, with which it often becomes 

 intimately associated, and frequently having a free extremit}^ pro- 

 jecting forwards toward the tip of the pterygoid, with which it is 

 connected by ligament. This latter fact is highly suggestive, since it 

 betokens a former bony connection betv/een these elements, just as 

 the stylo-hyoid ligament in the human subject connects the styloid 

 process and the hyoid bone, and represents the missing bony element 

 in tlie hyoid arch (the epi-hyal). Again, when we study this bone 

 in such types as Gymnura and the kangaroo, we begin to realize that 

 it has not only the exact position that would be assumed by a ves- 

 tigial quadrate, but actually resembles certain parts of this bone in 

 the reptilia. Thus we can imagine that the anterior pointed ex- 

 tremity represents the pterygoid process of the quadrate, the pos- 

 terior extremity represents the divided posterior bar which joins the 

 squamosal on the inner side and behind, and the blunt angidar pro- 

 jection would be the remains of the articular head of the quadrate. 



