the actual progenitors of the Mammalia, and that they should rather be regarded as a divergent branch of the stem which had given origin to the Mammalia. Since that date observations made on the structure of the South African Anomodonts have shown such an intimate connection between that group and the Labyrinthodont Amphibians, that there can be no hesitation in regarding the one as the direct descendant of the other; and we may probably regard the Mammalia as having originated from the same ancestral stock at the time the Amphibian type was passing into the Reptilian. From this point of view, some of the mammalian features found in the more specialised Anomodonts may probably be regarded as having been acquired during a parallel line of development. Both the Anomodontia and the Mammalia differ from the Amphibians in the loss of the splint-like parasphenoid which underlies the basisphenoid axis of the skull, and by the ossification of that axis; but while the former have become monocondylic by the participation of the basioccipital in the support of the cranium, the latter retain the Amphibian dicondylic plan. The skull of the Anomodonts presents mammalian resemblances not found in any other Reptiles, this being especially noticeable in the region of the squamosal; and it is only in this group and mammals that the temporal or zygomatic arch is a squamoso-maxillary one (see p. 37). The resemblance between the pectoral and pelvic girdles of the Anomodonts and those of the Monotreme Mammals is noticed under the head of the latter, where reference is also made to the similarity in the structure of the humerus in the two groups. The pes of the Amphibia and Anomodontia agree in having a distinct intermedium, tibiale, fibulare, and centrale, whereas in other Reptiles these bones are not generally distinct; in Mammals the intermedium, fibulare, and centrale are distinct, and according to Cope’s interpretation there may be a distinct tibiale. _Classification._—In the present condition of the world, mammals have become so broken up into distinct groups by the extinction of intermediate forms, that a systematic classification is perfectly practicable. Most of the associations of species, which we call “orders,” and even the “suborders” and “families,” are natural groups. In isolating, defining, and naming them, we are really dealing with facts of nature of a totally different order from the artificial and fanciful divisions formed in the infancy of zoological science.