1910.] The Need of a Nctv Osteology. 421 



raw material for ])hylogenetic conclusions; but in the leading existing text 

 books on vertebrate osteology tl|.is raw material seems to have been assimi- 

 lated and wrought into permanent philosophical results only to a very limited 

 extent. 



To the student of phylogeny the ])roblem of the relative antiquity of 

 different characters may be said to involve all the other general problems of 

 palaeontology and mammalogy, and especially such questions as the follow- 

 ing: What is the taxonomic rank or extension of the character in question ? 

 Is it a class character, a subclass, ordinal, superfamily, family, subfamily, 

 generic, or only a specific character ? Does it depend upon age or sex ? 

 How far is it "cfienotelic," i. e., of recent origin and related to particular 

 habits and habitat ? Does it depend upon mere size, or weight, or upon 

 strength of muscular pull? Or is it a fundamental and utiiversal character 

 like the foramen magnum, a character Avhich conditions survival in any 

 group whatever ? At what stage of its phyletic development does the char- 

 acter stand ? How far is it correlated with other characters and how far are 

 other characters correlated with it ? 



The solution of such questions, so far as it has progressed, naturally 

 has been largely by means of the "guess-try-guess-again" method, but the 

 point to be noted here is that basal assumptions and postulates should be 

 re-tested in order to make further advance sure. In the present state of 

 })hylogenetic research the need for new material is no greater than the need 

 for the reexamination of old material. A new and comprehensive treatise 

 on the osteology of living and fossil mammals is, in fact, urgently needed. 

 For although Flower's 'Osteology of the Mammalia' and Reynold's well 

 arranged text book on the 'Vertebrate Skeleton' have proved continuously 

 useful, yet these works contain no real clue to the interpretation of the facts 

 they record, so that the student in attempting to compare one form with 

 another is lost in the bewildering maze of analogical and homological re- 

 semblances. 



Like the above mentioned works the new osteology also should be i)ri- 

 marily an "osteography," abounding in good figures. It would of course 

 include living and fossil mammals; perhaps two or more representatives of 

 all the well known families. It would devote much more space to the more 

 ancient and generalized forms than to the more specialized and usually 

 better known forms. The forms should be arranged as far as possible in 

 accordance with the most nearly phylogenetic classification, and while this 

 feature would cause the work to become "out of date" to the degree that the 

 phylogenetic classification of the mammals became modified by the prog- 

 ress of the science, yet such an arrangement would be the best means of 

 testing certain theories of relationship and of showing the relative phylogen- 



