Humpback, but it is more likely that the figure is incorrect, as the nasals are made to end against the inferior margins of the premaxillæ, which are inclined outward, so as to leave a much wider space between the superior margins. The distance between the superior margins is about 7 %, which is a very close approximation to the breadth of the nasals in other specimens. The inclined position of the premaxillæ in this figure causes the narial space to appear much shorter than in skulls I have examined. Another peculiarity of the figure is the very strong emargination of the orbital processes of the frontal anteriorly, and their emargination posteriorly also. This peculiarity may likewise be a characteristic of the Greenland Hump. back, but may, on the other hand, be merely an inaccuracy in the figure. In the absence of any other figure of the upper surface of the skull of a Greenland Humpback it is difficult to decide the points at issue.

VERTEBRÆ.

The proportions of the vertebræ in the Tay River whale, and of some American specimens, including the type of M. osphyia, are given in the following table:

MEGAPTERA NODOSA (BONNATERRE). EUROPEAN AND AMERICAN.

SKELETON.

With the proper allowance for difference in age, the specimens show a correspondence indicative of specific identity. The positions in the column at which the various processes become obsolete and the arterial foramina appear are as follows:

My notes on No. 269 from Greenland, in the Brussels Museum, show that 9 chevrons are in position. The figure of Megaptera in Van Beneden and Gervais's Ostéographie (pls. 10, 11, fig. 1) shows 12 chevrons. The young specimen from Cape Cod, in the U. S. National Museum, No. 16252, has 9 chevrons. The Tay River (Scotland) specimen had 10 chevrons.

SCAPULA.

The scapula of Megaptera is peculiar on account of its evenly convex superior border and the rudimentary condition of the acromion and coracoid processes. (See text figs. 73–78 and pl. 34, fig. 4; pl. 36, figs. 3-5.)

The percentages of the antero-posterior breadth and of the vertical height (from the margin of the glenoid cavity to the middle of the superior margin) to the length of the skull in various European and American specimens are as follows:

Arranging the measurements of breadth of scapula according to the length of

the skull, without reference to locality, we have the following:

'From Rudolphi's figure. Type of M. longimana.

2

Left. The right 33.6%.

=

MEGAPTERA NODOSA (BONNATERRE). EUROPEAN AND AMERICAN. SCAPULA.

We find here, beautifully brought out, a gradual increase in the relative breadth of the scapula, with the increase in the size of the skull. Unless the series represented one species, it is hardly likely that this gradation would be obtained.

The scapulae of the types of M. bellicosa and M. osphyia, like that of the Tay River (Scotland) whale, show a low, blunt spine, a very narrow prescapular fossa, and a slight elevation on the anterior border (pl. 34, fig. 4; pl. 36, fig. 3). The anterior border is nearly straight, though somewhat irregular in the upper three quarters, while the posterior border is evenly concave. A rudimentary coracoid is discernible in the United States specimens, as in the Tay River (Scotland) whale, and in the Greenland skeleton No. 269 in the Brussels Museum.

RADIUS AND ULNA.

Struthers has published a figure (87, fig. 6) of the forearm of the Tay River whale, which shows well the shortness and strong curvature of the ulna and the expansion of the radius at the distal end, but hardly gives the impression of massiveness which these bones have. Malm published a figure (after a photograph) of the radius of a specimen in the Stockholm Royal Museum, received from St. Bartholomew Id., West Indies, where it was collected by Dr. Goës (66, fig. 4a).

This last is comparable with the radius of the type of M. bellicosa, which was also from the West Indies, and probably from St. Bartholomew Id., and was collected by Dr. Goës (see p. 97). The two radii are exactly alike, except that the Stockholm specimen appears to be a little narrower at the proximal end.

The proportion of the breadth of the radius at the distal end to its length in various American and European specimens of Megaptera is as follows:

'The measurements of this radius given by Malm (66, 38) make the breadth at the distal end only 33% of the length, but it is obvious by examination of the figure that the measurements are incorrect. The above proportion is from the figure, which is after a photographic original.

From the figure.

MEGAPTERA NODOSA (BONNATERRE). AMERICAN AND EUROPEAN. SCAPULA.

FIG. 73.-GREENLAND. AD. (FROM VAN Beneden and GERVAIS.) (FROM STRUTHERS.) FIG. 75.-CAPE COD, MASS. (FROM A PHOTO.) (FROM A PHOTO.) FIG. 77.-TYPE OF M. OSPHYIA. (FROM A PHOTO. BELLICOSA. (FROM A PHOTO.)

FIG. 74.-TAY RIVER, SCOTLAND. ô. FIG. 76.—PROVINCETOWN, MASS. IM. 오. OBLIQUE VIEW.) FIG. 78.-TYPE OF M.

The proportion for the Tay River whale is from Struthers's figure (89, fig. 6). His measurements give only 32.7%. The cause of this discrepancy is not obvious. All the other specimens show a close agreement.

The correspondence in the bones of the forearm between the types of M. osphyia and M. bellicosa are seen on comparing plate 34, fig. 4, and plate 36, fig. 3.

The proportion of the length of the radius and ulna to the length of the skull in various American and European specimens is shown in the following table:

MEGAPTERA NODOSA (BONNATERRE). EUROPEAN AND AMERICAN. RADIUS AND ULNA.

The number of ossified phalanges (exclusive of metacarpals) in the European Humpback has been given by Struthers (87, 38), and of the Greenland Humpback by Eschricht (37), Van Beneden (8, 134), and others, as follows:

MEGAPTERA NODOSA (BONNATERRE). EUROPEAN AND AMERICAN. PHALANGES.

Type of M. osphyia.

Type of M. bellicosa.

In d'Alton's Die Skelete der Cetaceen, 1827, pl. 3, fig. e, the hand of a Humpback whale, which from the text appears to be the type of B. longimana, is represented with the following phalangeal formula: 2, 7, 6, 2.

6

'According to my own notes on this skeleton, there are 6 phalanges in the 4th digit.

In a foetus 45" long. Eschricht's figure of a foetus 35" long, from Greenland, appears to show the following ossified phalanges: 2, 8, 8, 3 (37, 79). Eschricht also gives the formula for the adult as 3, 9, 9, 3, but does not state from what specimen or specimens this was derived (37, 14 1 ). It appears to include the metacarpals.