138 Bulletin American Museum of Natural History. [Vol. XXVII, 



this would entirely meet Gadow's objection (cited above) to Gaupp's 

 "double articulation" theory. 



The slender character of the angular and articular bones, and the in- 

 stability of their connection with the dentary, combined with the backward 

 and upward pull of the powerful temporal muscles upon the large dentary, 

 may have conditioned the backward growth of that element until its posterior 

 (condylar) extension found secure lodgment against the squamosal. The 

 new joint would then be more in line with the pull of the temporal muscle 

 and would be mechanically better and more stable than the old joint. 



These considerations are in harmony with the arguments against the 

 view (p. 135), that the quadrate was transformed into the meniscus. That 

 the new condylar process of the dentary did not impinge altogether against 

 the quadrate, but partly above it, is indicated by Broom's detailed side view 

 of this region in Cijnognafhiis (op. cit., pi. xxxv, figs. 5, 6). In the front 

 view of this region in Gomphognathus {op. cit., pi. xxxv, figs. 2, 3) the condy- 

 lar process is represented as beloiv the level of the quadrate, but this is because 

 the jaw is depressed at an angle of 90° to the palate. 



When the condylar process of the dentary grew backward we may imagine 

 that it did not abut directly against the squamosal bone itself but merely 

 pressed against the muscles which occupy this region in reptiles. This 

 would agree with the conclusion reached by Kjejlberg (1904) who sees in the 

 meniscus merely the tendinous origin of some of the fibres of the external 

 pterygoid muscles, which fibres are attached to the meniscus in man 

 (Cunningham, 1902, p. 260) Echidna (Lubosch, 1906) and other mammals. 

 Lubosch (1906, pp. 591-594) regards the meniscus as representing a sepa- 

 rated portion of the periosteum of the lower jaw. 



In order to substantiate the conclusion that the mandibulo-squamosal 

 joint in mammals is a wholly new structure, into which the quadrate and 

 articular did not enter, we recall the facts: (1) that embryological research 

 gives no warrant for the belief that the mammalian jaw is composed of more 

 than one element (except for the occasional vestiges of a splenial) ; (2) that 

 the oldest known mammalian jaws, from the Triassic, Jurassic and Basal 

 Eocene never show any trace of sutures; (3) that in the Cynodonts the broad 

 ascending ramus, or corono-condylar region appears from Broom's researches 

 (1904.2) to be a part of the dentary. 



In proportion as the new squamoso-mandibular joint became established 

 the old quadrato-articular joint would probably dwindle in size and become 

 available for the change of function which is the chief condition for the 

 transformation of the quadrate and articular into the incus and malleus. 

 But there are two im])ortant difficulties which must be met before the view 

 that the cjuadrate became the incus, the articular the malleus may be con- 



