12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 76 



tion of the zygoma accompanying the loss of the posterior teeth and 

 the consequent disuse of the masseter muscle. The outward pushing 

 of the lacrimal beyond the base of the jugal, characteristic of the 

 extreme type of toothed cetacean skull, is already well indicated in 

 the zeuglodont (see especially Andrews's pi. 21, fig. i, Tert. Vert. 

 Fayum, 1906). On the other hand, it will at once appear from com- 

 parison of the maxillary of a zeuglodont with that of a baleen whale 

 (pi. 4, figs. I and 3) that here a transition from the earlier to the 

 later type would involve the inexplicable complication of introducing 

 a large new structure, the freely projecting orbital plate (pi. 4, fig. 3, 

 o. pi.), in a position where it meets no recognizable mechanical need, 

 and in a condition which suggests a well advanced stage of 

 degeneration. 



THE THREE CETACEAN PHYLA OR SERIES 



The facts which have just been reviewed appear to be most simply 

 and fully explained by assuming that the known cetacea represent 

 three distinct lines of descent and that the directions in which these 

 lines w'ere to develop were determined by peculiarities of structure 

 established at or before the beginning of pelagic life. The idea that 

 the phylogeny of the cetacea has been multiserial ^ is not new. In dif- 

 ferent forms and for various reasons it has been put forward by 

 several authors who have dealt with the question of the group's 

 history. The whole subject has recently been reviewed by Kiikenthal 

 ( Sitzungsber. Preuss. Akad. Wissensch., Phys.-math. Kl., 1922, pp. 

 72-87, Meeting of March 16, 1922). As I understand it this assump- 

 tion does not imply that the separate lines represent convergence 

 from widely different ancestral stocks. Such heterogeneous origin 

 is made to appear improbable by the resemblances between zeuglodonts, 

 baleen whales, and toothed whales in features which are not readily 

 explained as the mere retention of primitive characters or as the 

 separate remolding into similar structure of originally unlike parts 

 applied to a single new and peculiar mechanical use. The auditoiT 



^ The term polyphyletic, often used in this sense, is open to the objection 

 that it has two meanings : nitiHiserial and polygenetic. Apart from the con- 

 text, therefore, we can never know whether a " polyphyletic group " is a 

 group consisting of several genetic Hnes coming up out of the past in a 

 direction parallel with each other and never uniting, or whether it is a group 

 formed by the uniting of several lines coming up from different directions. 

 Ambiguity would be avoided by the consistent use of multiserial to express the 

 first idea and polygenetic to express the second. For monophyletic it might be 

 well to substitute imiscrial. 



