Oct. 25, 1888] 



NA TURE 



613 



co lard us, and Phascolomys, could surely also find room 

 for the Multituberculate type. The evidence of the 

 homology of the incisors is, however, a weighty one in 

 the author's favour. 



Prof. Osborn places the Triassic Microlestes with the 

 Plagiaulacidce rather than the Bolodontida, but we think 

 the existence of a cutting fourth lower premolar ought to 

 be proved before this view can be definitely admitted. 

 There may also be considerable hesitation in accepting 

 the view expressed on p. 217, that there are five premolars 

 in the upper jaw assigned by Prof. Marsh to Ctenacodon; 

 but beyond these and other small points the author's 

 classification of this group appears to commend itself. 



We cannot say the same in regard to the classi- 

 fication of the second group, which, as we have 

 seen, it is proposed to split up into one distinct order, 

 into one sub-order provisionally referred to the Mar- 

 supialia, and a second assigned with more hesitation 

 to the Insectivora. In this group the author has, we 

 venture to think, found differences which, if they exist at 

 all, are by no means of the importance he attributes to 

 them ; while at least one case occurs to us, where, to say 

 the least, there is a considerable presumption that speci- 

 mens assigned to the two sub- orders may really be refer- 

 able to a single genus. Sufficient account does not, 

 indeed, appear to have been taken of the variation in the 

 dentition of different recent genera of Marsupials which 

 are usually included in a single family ; as, for example, 

 Thylacinus, Dasyurus, and Myrmecobius among the Poly- 

 protodonts, and Phalanger, Pseudochirus, and Phascol- 

 arctusm. the Diprotodonts. In the case of obscure fossil 

 forms like the present, it appears to us that there ought to 

 be the greatest hesitation in making groups of higher 

 value than family rank ; and that even in the case of 

 families their limits ought to be much more loosely 

 drawn than among existing forms, where we have full 

 evidence before us. It is, indeed, far more advantageous 

 to keep all such obscure forms more or less closely 

 associated until absolutely decisive evidence is forth- 

 coming as to their right to wide separation. In the 

 present instance, however, the author has, to put it in the 

 mildest form, by no means adduced any such decisive 

 evidence ; while, as already mentioned, there is a strong 

 presumption that in certain particular cases he has widely 

 separated closely allied, if not absolutely identical, forms. 

 The first so-called order — the Protodonta — is formed 

 for the reception of the American Triassic Dromatherium ' 

 and Microconodon ; if, indeed, the latter be really entitled 

 to generic distinction. The grounds for the ordinal 

 distinction of these forms are that the roots of the cheek- 

 teeth are not fully divided ; but stronger evidence than 

 this is required before these obscure forms can be definitely 

 regarded as entitled to constitute more than a family. 

 And even if they belong to an order distinct from the 

 Marsupials, there is no evidence to show that they are 

 not Monotremes, or perhaps rather Prototheria. 



The sub-order Prodidelphia is defined as including 

 primitive Marsupials, generally characterized by the pre- 

 sence of four premolars and numerous molars, the latter 

 having distinctly divided roots. It is, however, added (on 

 p. 259) that "no definite sub-ordinal character can be 



1 Prof. Osborn proposes. to alter the spelling of this name to Dromo- 

 tlicrium. 



assigned ; but in view of the retention of several features, 

 and of their ancestral position, these Mammals may be 

 distinguished from the recent Marsupials as the sub-order 

 Prodidelphia." In our own judgment, the formation of a 

 large group which confessedly cannot be distinguished 

 from one already established is unjustifiable, and not 

 conducive to any advantage. The first family of this 

 group is the Triconodontidce, in which, as shown above, 

 our author includes a large number of genera. The 

 genus Triconodon, together with the allied or identical 

 American Priacodon, has, however, such a totally different 

 fades from all the other forms, that we are inclined to 

 follow Prof. Marsh in regarding it as alone constituting 

 the family. We are, morever, rather at a loss to find the 

 value of the characters which Prof. Osborn regards as 

 distinctive of the enlarged family ; for, whereas he states 

 in the definition of the family (on p. 227) that the " condyle 

 is low," on the opposite page the genus A mphitylus is 

 described as having the " condyle lofty." Some very 

 interesting observations are recorded (p. 198) as to the 

 changing and development of the teeth in Triconodon, in 

 which it is concluded, as had been previously indicated by 

 Mr. O. Thomas, that the replacement was limited, as in 

 modern Marsupials, to a single premolar ; while it is further 

 shown that in many instances it appears probable that the 

 last true molar was never developed. In classing Phasco- 

 lotherium, of the Stonesfield Slate, in the Triconodontidce, 

 the author appears to have been greatly influenced by 

 regarding Trico7iodo7i as having the condyle placed low 

 down on the mandible. We have, however, considerable 

 doubts whether this is a character of much importance, 

 as it varies so much in the allied Phascolotherium and 

 Amphitylus. In considering that the whole of the seven 

 cheek-teeth of Phascolotherium are true molars, the author 

 departs very widely from the view taken by Sir R. Owen, 

 and a great deal more evidence is required before it can 

 be considered proved that at least the first two of these 

 teeth are not premolars. 



In making such mention as space permits of some of the 

 other genera, we must take those included under the Prodi- 

 delphia and the so-called Insectivora Primitiva together. 

 In this connection it appears that a great deal depends on 

 the interpretation of the dental characters of the original 

 genus Amphithcrium, to which Prof. Osborn refers the 

 fragment of a mandible figured on p. 192. It is stated, 

 with great fairness, that when the author examined this 

 specimen he regarded it as totally distinct from Amphi- 

 therium, but that comparisons of his drawings with 

 figures led him to change his opinion. On p. 192 it is 

 observed that " when these mutilated crowns [of the 

 type] are compared with the perfect crowns of the newly- 

 acquired jaw, there can be no doubt that they belong to 

 the same pattern. If this be the case, the latter specimen 

 is of great interest, as it enables us for the first time to 

 fully characterize the molar dentition of Amphithcrium." 

 We have purposely italicized portions of the above sen- 

 tences, since they show a somewhat curious instance of 

 the author's method. Thus, in the first sentence the 

 teeth of the new jaw are definitely stated to be of the 

 type of those of Amphithcrium, while in the second a 

 provisional element is introduced ; and yet subsequently 

 this jaw is again definitely taken as affording the true 

 structure of the Amphitherium molar. Far be it from us 



