July i6, 1891] 



NATURE 



245 



the Pterodactyles and Dinosaurs calls for no separate 

 mention, although we are led to wonder why the Croco- 

 diles are so widely separated from these groups. 



The whole chapter on birds is decidedly feeble ; and 

 we must certainly take exception to the statement that 

 Hesperornis and Ichthyornis respectively connect Archceo- 

 picryx with the Ratitae and Carinatae. 



Turning to the last chapters, on the Mammals, we find 

 the author adopting the view that the Monotremes have 

 had an origin totally separate from the other two sub- 

 classes. We then have a notice of the Secondary Mam- 

 mals, in which we observe a lamentable lack of attention 

 to recent work on their affinities, and also to the 

 synonymy of the various genera. We also notice that the 

 Jurassic Plagiaulax and its allies are still referred to the 

 Diprotodont Marsupials (p. 327) ; so that on these points 

 the author's evolutionary views are totally out of date. 

 Following the Marsupials, we have a very fair, although 

 brief, account of the most recent conclusions on Mam- 

 malian phylogeny, which needs but few remarks. We 

 notice, however, that the author adopts M. Boule's views 

 as to the dual origin of the Canidcc, according to which 

 the Foxes (Alopecoids) are considered to have originated 

 from Cynodiciis, while the Wolves, Jackals, and Dogs 

 (Thooids) trace their descent to Amphicyon. To our- 

 selves, indeed, it has always appeared difficult to under- 

 stand how these two groups of Canidcz have become so 

 much alike if they had this dual origin ; and this diffi- 

 culty is increased by the author's statement that those 

 Thooids known as Cyan differ from the other members of 

 that group in having originated from Cynodictis. 



On p. 343 the author makes a slip in stating that the 

 Hydracoidea are now represented only by a single genus ; 

 while later on he appears to be uncertain whether the 

 Siwahk beds should be regarded as Upper Miocene (pp. 

 349> 350) or Upper Pliocene (p. 366). Again, we notice 

 on p. 353 some want of acquaintance with the recent 

 literature relating to the ancestry of the horse, Orohippus 

 being identified with PHolophus, whereas the latter is 

 really the same as Hyracotherium, while the former is 

 identical with Pachynolophus. We are in full accord 

 with the author when he states, on p. 361, that Chalico- 

 therium (with which the supposed Edentate Macro- 

 theriiim is now known to be identical) is an aberrant 

 Ungulate, although we must be permitted to differ from 

 him when he adds that it shows signs of affinity with the 

 Edentates. 



We must likewise take exception to the statement, on 

 p. 370, that the Indian Nilgai is in any sense the pro- 

 genitor of the Oxen ; while the view expressed on the 

 same page, that the Buffaloes, Bisons, and true Oxen have 

 severally originated from three distinct groups of Ante- 

 lopes, can be only regarded as another instance of the 

 author's partiality for multiple phylogenies. Although 

 M. Priem is careful not to commit himself to the view 

 that the Cetaceans have been derived from the extinct 

 EnaHosaurian reptiles, yet the prominence which he 

 gives to the statement of that view may be taken as a 

 sign that he has not thoroughly purged himself from that 

 heresy. 



Finally, although we have felt bound to call attention 

 to a certain amount of imperfection in the later chapters, 

 yet, as a whole, we can conscientiously recommend the 

 NO. I 133, VOL. 44] 



work before us to those readers who are desirous ot 

 obtaining in a compact form a summary of the evidence 

 afforded by palaeontology of the progressive evolution of 

 animal forms. R. Lydekker. 



METALLURGY. 

 Leqons stir les Mitaux. Par Prof. Alfred Ditte. (Paris : 



Dunod, 1891.) 

 Traite pratique de Chimic Metallurgique. Par le Baron 



Hans Jiiptner von Jonstorff. Translated from the 



German by M. Ernest Vlasto. (Paris : Gauthier- 



Villars, 1891.) 



THESE two volumes, recently published, are both of 

 unusual interest. The first, by Prof. Ditte, who 

 is well known to English readers by his " Expos^ de 

 quelques Propri^t^s g^ndrales des Corps," may be said 

 to mark a new departure in teaching the chemistry of 

 metals. He points out that the principles of thermo- 

 chemistry do not merely enable reactions to be explained, 

 but to be predicted, and, on the other hand, when two 

 sets of reactions are simultaneously possible, the laws of 

 dissociation render it possible to rigorously define the 

 conditions of equilibrium which are established in the 

 chemical "systems" under consideration. It is often 

 possible, with the guidance afforded by these laws, 

 to say, in the absence of direct experiment, why one 

 reaction is impossible and another certain to occur ; or 

 why a certain reaction begins without difficulty, and is 

 arrested at a definite stage ; or why a reaction which 

 takes place readily under certain conditions cannot be 

 effected under others that do not appear to differ greatly 

 from those which were favourable to it. As a pupil of 

 Deville, the author might have been expected to develop, 

 in a treatise such as this, the teaching of his great master, 

 and he has admirably performed his task. The classi- 

 fication of the work is excellent, the metals being first 

 considered collectively, and then in detail with numerous 

 tables of the data and constants which are so frequently 

 required by metallurgists. 



The work begins with a very clear account of Berthelot's 

 labours in mdcanique chimique, special care being de- 

 voted to the description of the calorimetric investiga- 

 tions, and to the appliances adopted in these important 

 researches. 



It appears to be a great advance for us in this country 

 to read a chemical treatise in which the thermal values 

 of the equations are stated in calories, side by side 

 with the formulae. As the book is too long to review 

 in detail, it may be well to indicate the nature of one 

 section only, as showing the author's care and thorough- 

 ness in the selection and arrangement of the mate- 

 rials. Take, for instance, the few pages devoted to 

 carbides. The author points out that carbon in uniting 

 with metals sometimes gives rise to the formation of true 

 compounds, and at others to solutions of carbon in the 

 metal. He then describes the orange-yellow product 

 obtained by the action of carbon on metallic copper, and 

 passes to the association of carbon with nickel, which 

 does not confer upon nickel the property of being 

 hardened by rapid cooling. The definite carbides of 

 manganese, as well as the indefinite associations of 

 carbon with iron and manganese, receive due attention, 



