THE WHAXEBOJTE WHALES OF THE WESTERN NORTH ATLANTIC. 93 



(posteriorly) of the centrum of the 33d vertebra is 12 inches, and the neural arch 

 and spine 13^ inches, making a total of 25^ inches. The height of the arch 

 and spine is, therefore, about equal to the depth of the centrum, and not "nearly- 

 double," as stated in the original description. In skeleton No. 21492 the diameter 

 of the centrum of the 34th vertebi'a ' is 9.75 inches, and with the neural arch and 

 spine the total height of the vertebra is 18 inches. This character of the lumbar and 

 caudal spines is clearly fictitious, and as it was really the principal one on which the 

 species osphyia was based, we are justified in the assumption that the type repre- 

 sents the ordinary Humpback of our Atlantic coast, and is to be so regarded unless 

 other charactei-s than those enumerated by Cope can be detected. 



In 1884, Cope in reviewing my Catalogue of Aquatic Mammals, above 

 mentioned (30, 1123-1124), took exception to this view, and accused me of 

 inaccuracy in stating that the high neural spines had been put forward as the 

 principal character of the species. He quoted from his original description, as fol- 

 lows: "The shorter bead and fins, the peculiarly high neural spines'" and peculiari- 

 ties of some of the cervical vertebrae, would seem to distinguish this [species] from the 

 longimanaP As, however, a Megciftera with the skull "one-fifth, or less" the total 

 length, and the flipper " one-fifth " the total length, as first reported by Cope, would 

 be a decided anomaly, I regarded these dimensions with suspicion, and an examina- 

 tion of the skeleton showed that they were due to the imperfection of the specimen. 

 The characters of the cervicals mentioned by Cope, in so far as they differ from 

 those of any specimen of Megaptera, seemed to be of little importance, as above 

 noted. The supposed great elevation of the neural spines of the dorsal and lumbar 

 vertebrae^ seemed possible, and hence the really important cluiracter ; and so, indeed, 

 it would be, if established. 



In 1868 Cope (37, 194) made further reference to the type, stating that the 

 skeleton (as it then was) measured 34 feet, but that as it lacked some of the caudal 

 vertebrae and the intervertebral cartilages had shrunk, the proper length was perhaps 

 42 feet. He describes several additional features of the skull and skeleton, all of 

 which are to be found in the specimen in the National Museum, No. 21492, except 

 that which relates to the union of the neural arches of the 3d and 4th cervical 

 vertebrae. This is, however, an individual rathei- than a specific character. 



In 1871, in describing another species (29, 107), Cope makes a few additional 

 comments on M. osphyia. He remarks that in this species " the head and fin are 

 even shorter than in M. langimana, and the coronoid process equally rudimentary. 

 . . . The width of the orbital plates [orbital process of the frontal] distally is 

 .5 their length in the type of M. osphyia^ As regards the orbital process of the 

 frontal it is to be remarked that the proportions given by Cope for M. osphyia are 

 the same in the two skulls in the National Museum (Nos. 21492 and 16252) from 



' The 33d cannot be measured. 



" This refers to the spines of the dorsal and lumbar vertebr.ie, and not to those of the cervicals. 



F. W. T. 



" And also of the caudals, as Cope mentions particularly the 33d vertebra among them. It is 



really the 2d caudal. 



