B ihliographical Notices. 71 



minate cervical vertebra is regarded as indicating a Siwalik stork 

 or allied form. The genus Mergits is quoted with doubt on the 

 e'\'idence of a cervical vertebra. Struthio asiaticus was an ostrich 

 closely allied to the existing species, and the author doubts whether 

 the slight differences in their cervical vertebrae can be of more than 

 individual or varietal value ; so that this species is regarded as pro- 

 visional. 



The Dromceus sivalensis has since been withdrawn by the author 

 on the ground that the bones must be referred to an Artiodactylate 

 mammal. 



Part 5 is devoted to Mastodon teeth from Perim Island. They 

 comprise the first and last upper true molars of Mastodon pandionis, 

 and the second right upper true molar, upper milk-molars, and 

 penultimate lower molar of Mastodon perimensis; and having 

 described these teeth in detail, the author regards it as evident that 

 the tetralophodont M. perimoisis is a more specialized form than 

 the trilophodont M. pandionis. In order to convert the teeth of 

 the latter into the former type it is necessary that the anterior 

 accessory columns should be less developed, so that the valleys 

 would be more open. A fourth ridge should be developed in the 

 intermediate molars, and a fifth ridge and double talon in the last 

 molar. Both species have cement in the valleys. It is considered 

 as likely that Mastodon sivalensis is a descendant from the stock of 

 Mastodon pandionis, and it is thought probable that Mastodon pan- 

 dionis and Mastodon pentelici are both branches from the older 

 stock of Mastodon angustidens. 



Iq this volume a considerable mass of material is made known 

 and illustrated by figures, which for the most part are excellent ; 

 and it is a great gain to palaeontology for naturalists to be in 

 possession of the author's descriptions. Admirable in many ways, 

 the work suffers from the disjointed manner in whicTi the growth 

 of material has caused the parts to be issued ; and some of the 

 descriptions rather convey the impression of unnecessary haste in 

 publishing what might perhaps have been perfected by fuller con- 

 sideration. There is an appearance of desiring to leave nothing 

 for those who may come hereafter, and yet at the same time to leave 

 open a way for retiring from positions which future research may 

 make untenable. Many of the species instituted by the author 

 seem to us to be founded on characters which would justify us in 

 extending to them the term provisional, which so often characterizes 

 species founded by others. Space might sometimes have been 

 gained which could with advantage have been taken for more ex- 

 tended description. Not but what the descriptions are excellent in 

 their way, only they could in many cases have been fuUer with 

 advantage. The author's strong interest is rather with what may 

 be termed genetic comparisons. He has done much to unravel the 

 affinities of species by comparing them with existing and fossil 

 allies ; and the speculations on descent of species are usually justi- 

 fied by the nature of the materials and the interest of the problems 

 involved. But suggestive as this pursuit of evolution undoubtedly 



