The osteological material of this species in the National Museum available for the study consists of (1) a skull from Puget Sound (Cat. No. 12177, U. S. N. M.), presented by Capt. C. M. Scammon about 1872, and very probably belonging to the individual from which the original description was drawn, and hence the type of the species; (2) a skull from St. Paul Id., Pribilof Group, Alaska (Cat. No. 61715, U. S. N. M.), collected by Mr. C. H. Townsend of the U. S. Fish Commission.

Measurements of these skulls, reduced to percentages of the total length, are given on page 197, with those of specimens of B. acuto-rostrata. It will be observed by reference to the measurements that the Pacific specimens agree in all but one or two of the proportions given with those from the Atlantic in the closest possible In each proportion in which one of the Pacific skulls differs from the Atlantic ones, the other harmonizes with the latter, so that there cannot be said to be a constant difference in any of the proportions between the specimens from the For convenience, the actual measurements of the two Pacific skulls,

manner.

two oceans.

and of the Norway skull, No. 13877, are given below:

BALENOPTERA ACUTO-ROSTRATA LAC. AND B. DAVIDSONI SCAMMON. SKULL.

2 Add 2 in. for breakage.

Greatest breadth of maxilla behind base of beak..

66

between outer borders of both premaxillæ..
inner

66

66

From vomer at ant. end of palate through to vertex.
Inner margin nasal processes of max. to end of orbital process of max..
Outer edge of premaxilla to end of orbital process of maxilla...

'All straight, unless otherwise stated.

In spite of the correspondence in general proportions between two Pacific skulls and the Norway and Massachusetts skulls, my associates, Dr. L. Stejneger and Mr. G. S. Miller, Jr., who examined them with me, while side by side in one of the halls of the Museum, pointed out certain characters in which the two Atlantic skulls appeared to them to differ from the two Pacific skulls. The principal of these were (1) that the nasal processes of the maxilla were bent toward the median line much more strongly in the Pacific than in the Atlantic skulls, and (2) that the orbital process of the maxilla was shorter and thicker in the former than in the latter. The characters will be seen by comparing the figures on plates 22 and 23. I also noted that in the Pacific skulls the vomer appeared to descend more opposite the anterior end of the palatines, giving a stronger curve to the inferior profile of the cranium, and that the palatines were broader posteriorly. I have endeavored to bring out some of these differences in the last three measurements of the foregoing table. These measurements reduced to percentages of the total length of the skull are repeated below:

Distance from inferior surface of vomer at ant. end of palatines to vertex, straight......

Inner edge of proximal end of nasal process of maxilla to distal end
of orbital process of maxilla, straight......
Outer edge of premaxilla to distal end of orbital process of maxilla,
straight....

It would appear from the foregoing that the vomer is deeper in the Pacific skulls, but the proportional length of the orbital process of the maxilla does not differ materially in the Norway and Pacific skulls. The breadth of this process, as shown by plates 22 and 23, is greater in the Pacific skulls than in the one from Norway. This greater breadth, however, is approximated in Eschricht's figure of an adult skull from Norway (37, pl. 9, fig. 1).

If any of these differences prove constant on examination of a larger number of specimens, it will probably be the greater depth of the vomer and the bending inward of the nasal process of the maxilla. As regards the latter, Eschricht's and Capellini's figures of European skulls present a substantial agreement with our skulls from Norway and Massachusetts.

'Type of B. davidsoni.

22 in. added for breakage.

A series of vertebræ belonging to a small Finback whale were found by me on St. Paul Id., Pribilof Group, Bering Sea, July 30, 1895. They were 27 in number, and included the 7th cervical, 11 dorsals, and 15 lumbars and caudals. It will be observed that the number of dorsals is the same as in B. acuto-rostrata.

Of this species Van Beneden remarked in 1889 (7, 165): "In our opinion it is a synonym of Balaenoptera rostrata" (= B. acuto-rostrata).

CHAPTER X.

CONCLUSIONS.

The conclusions reached in the foregoing pages are:

(1) That the species of whalebone whales occurring in the western North Atlantic Ocean are identical with those occurring in the eastern North Atlantic.

(2) That these species are the Bowhead, or Greenland Right whale, Balana mysticetus, the Black whale, Balana glacialis, the Humpback, Megaptera nodosa, the Sulphurbottom, Balaenoptera musculus, the common Finback, Balaenoptera physalus, and the Little Piked whale, Balaenoptera acuto-rostrata, and probably the Pollack whale, Balaenoptera borealis.

(3) That the range of one of these whales-the Humpback-extends southward at least as far as 18° North Lat.

(4) That the probability of the identity of the North Pacific species with those of the North Atlantic is strengthened by the evidence herein collected.

As modifications of the preceding statements, several particulars require to be brought forward. Both the Little Piked whale and the Humpback of Greenland may possibly possess characters entitling them to be regarded as separate subspecies. These differences, however, are quite as likely to be due to inaccuracy of observation. As the species are migratory, it is probable that the Greenland individuals mingle with individuals from farther south and are identical with them both specifically and subspecifically, but additional evidence is needed to prove this hypothesis.

As no specimens of the Pollack whale, Balaenoptera borealis, from American waters have been examined, it is not certain that the species is really the same on both sides of the Atlantic. As the other species are the same, the presumption is, of course, that the Pollack whale also undergoes no modification. This, however, requires to be demonstrated.

As evidence is strengthened regarding the specific identity of the whales of the North Atlantic and North Pacific, the belief that the same species of large whales range all over the globe is, of course, also strengthened. It is well-known that whales closely resembling Megaptera nodosa, B. acuto-rostrata, B. musculus, and B. physalus-to mention no others-occur in the South Atlantic and the Antarctic seas, and also the second and last, at least-about New Zealand.

Some competent zoologists have expressed the opinion that the species are cosmopolitan, but as already said in the case of the North American species, such opinions have not been based to any large extent on the critical examination of

considerable numbers of specimens from the regions mentioned. Such opinions have, of course, a certain interest and value, but knowledge will not be greatly increased without the study of new material.

Even should it be demonstrated that the species of large whalebone whales are cosmopolitan, it does not follow that the individuals constituting these several species range throughout the globe. The probabilities are much against such world-wide movements, and in the case of the Right whales it appears to have been demonstrated by Maury that individuals do not cross the equator. In this latter case, and perhaps in others, it would appear that the study of the migrations of separate groups of individuals, or schools, can be carried on profitably without regard to the general facts pertaining to the distribution of the species as a whole. The following diagnoses of North Atlantic species are intended to summarize the observations of earlier writers both American and European, as well as those detailed in the preceding pages. The diagnosis of Balanoptera borealis is based on Collett's admirable account of that species (21).

BALENA GLACIALIS Bonnaterre.

Black whale, Nordcaper, or Biscay whale. Plate 50, fig. 2.

Form massive. Head very large. Rostrum narrow and curved, with a protuberance near the anterior end ("bonnet "). Blowholes elevated and followed by a distinct depression. Lower lip very large, oblong, the free margin more or less sinuous.

Pectorals very broad, short, with a convex posterior margin and pointed tip. Color black throughout, or with more or less white on the throat and breast in some individuals.

Rostrum of skull very long and narrow; the anterior half strongly curved. Intermaxillæ broad, occupying nearly the whole upper surface of the rostrum. Nasals very large, broad, oblong. The free anterior border w-shaped. Orbital process of frontal very narrow, somewhat tubular, and only moderately bent backward, the orbital border very narrow, oblique. Occiput broad, with convex sides. Sternum broadly and irregularly triangular. Scapula broader than high; broad near the base. Vertebral formula: C. 7, D. 14, L. 11 (10-12), Ca. 23 (−26). Total 55 (-57).

MEGAPTERA NODOSA (Bonnaterre).
Humpback. Plate 50, fig. 1.

Form massive and peculiarly ungraceful, size moderate. Head flat and obtuse. Abdominal ridges few and broad, 14 to 30. Average total length, 48 feet; maximum, 55 feet.

Pectorals, from head of humerus, 32 per cent. of total length; lanceolate, with extremity recurved; anterior margin with ten or eleven very prominent sinuosities corresponding to the joints of the manus; posterior margin convex proximally, concave distally, with several small sinuosities at the extremity.