Decembee 6, 1907] 



SCIENCE 



797 



of the mammal-like reptiles have abdominal 

 ribs hitherto been found and we may feel 

 quite certain in concluding that in the An- 

 omodonts at least they did not occur, and had 

 they occurred in the Therocephalians it is 

 likely they would have been discovered. The 

 combination of characters seems to point to 

 Galechirus being a primitive Therapsidian 

 reptile most closely related to the Theroce- 

 phalians, but with distinct Ehynchocephaloid 

 affinities. As the American Pelycosaurs are 

 undoubtedly Ehynchocephaloid, it seems not 

 improbable that the common ancestor of the 

 Pelycosaurs and the African mammal-like 

 forms may have been an early Ehynchoceph- 

 aloid reptile rather than a Cotylosaurian, as 

 some of us had thought. 



Another type which I have discovered in the 

 last few months seems in some respects almost 

 as important as Galechirus in that it is an- 

 other " missing link " discovered. A few 

 years ago I pointed out that the old order 

 " Theriodontia " of Owen was an unnatural 

 group including two well-marked divisions — • 

 the Therocephalia with a Ehyncephalian type 

 of palate, single occipital condyle, simple 

 molars and large ang-ular and surangular 

 bones in the lower jaw, and the Cynodontia 

 with a mammalian type of palate, two con- 

 dyles, complex molars and the lower jaws al- 

 most wholly formed by the dentary. The 

 Therocephalians are almost entirely confined 

 to the Middle and Upper Permian beds, the 

 Cynodonts to the Middle and Upper Triassic 

 beds. The new type, which I propose to call 

 Bauria, was found in Upper Triassic beds, and 

 while it must be placed among the Cynodonts 

 it shows affinities with the Therocephalians 

 not present in the other known genera. In 

 general shape the skull is not unlike that of 

 Trirachodon, but about one half larger than 

 T. hannemeyeri. The dental formula is 

 14, el, m 10 above and apparently the same 

 below. The molars are remarkable by being 

 simple uneusped teeth with flattened tops. 

 There is a secondary palate as in typical Cyno- 

 donts, but unlike all previously discovered 

 forms the postorbital arch is incomplete, the 

 postorbital bone not meeting the jugal. The 



appearance thus produced is very mammal- 

 like. There is apparently no parietal fora- 

 men. The squamosal is more like that of the 

 Therocephalians than that of the Cynodonts, 

 and the quadrate is very minute. The occip- 

 ital condyle is intermediate between that of 

 the Therocephalians and the Cynodonts in that 

 while it is really double the two parts are so 

 close to each other that it is practically single. 

 The lower jaw is almost typically Theroce- 

 phalian, the angular and surangular being 

 large and the dentary only forming the an- 

 terior two thirds of the jaw. While Bauria 

 is thus typically Cynodont in the structure 

 of its palate and must therefore be placed in 

 the Cynodontia, in the simplicity of its molars, 

 the condition of the occipital condyle, and in 

 the structure of the lower jaw it shows dis- 

 tinct affinities with the more primitive Thero- 

 cephalians. 



In the Lower Triassic beds an imperfect 

 skeleton of a small Mesosaurus-like reptile has 

 been discovered. It is less typically an 

 aquatic form and has slender ribs. Abdom- 

 inal ribs are well developed. The skull, which 

 is imperfectly preserved, is long and pointed 

 and, so far as can^be made out, is Ehyncho- 

 cephaloid in its characters. Until other speci- 

 mens throw further light on the form it will 

 provisionally be placed in the Mesosauria. It 

 has been named Heleosauriis. 



E. Broom 



Victoria College, 

 Stellen Bosch, South Africa, 

 October, 1907 



NOTE ON THE FERMENTATIVE REACTIONS OF THE 

 B. COLI GROUP 



In view of the fact that Prescott' and 

 others have recorded the presence of organisms 

 resembling Bacillus coli on grains, it seemed 

 to us of interest to make a somewhat careful 

 comparison of these forms with intestinal B. 

 coli in regard to their power of fermenting 

 carbohydrates. The success of Gordon and 

 Houston^ in clearing up the relations of the 



^ Science, N. S., XV., 363 ; Medicine, XI., 20 ; 

 " Biological Studies " by the pupils of William 

 Thompson Sedgwick, Boston, 1906. 



^Report of the Medical Officer to the Local 



