Measurements. Length anterior to parietals above.... Length of occipital from base of foramen magnum to apex (on curve) ..... Width of occipital condyles and foramen.. M. .225 .290 .140 The mandible of this species is unknown. The size is not far from that of the Cetotherium pusillum and Siphonocetus expansus of Cope. Should either of these turn out, on the discovery of the skull, to be Cetotheriform, it will become necessary to compare them with the present species. The total length of the animal was about twenty or twenty five feet. CETOTHERIUM CRASSANGULUM, sp. nov. This species is represented by an imperfect skull, to which adhere three cervical vertebræ, the posterior parts of both mandibular rami, parts of the hyoid arch and a humerus. The sagittal part of the skull is crushed and the frontal bones somewhat displaced outwards. Large portions of the nasal bones remain, but the premaxillaries and maxillaries are mostly wanting. By excavating at the proper point the right otic bulla was brought to light. The presence of this structure, together with a considerable part of the mandibular rami, enables me to compare the individual with known species and to determine its specific reference with certainty. The coössification of all the epiphyses shows that the animal is adult. The species belongs to that group in the genus Cetotherium which is characterized by the presence of a developed angle of the mandible, but where it is short and broadly truncate. The angle is, however, scarcely separated from the condyle, and partakes of the articular surface, apparently much as in the Balanoptera emarginata of Owen. It is further distinguished from such species as C. priscum, C. meyerii and C. klinderii by the oval and little-compressed form of the otic bulla, resembling in this part rather the C. rathkei. From the C. megalophysum the bulla differs by the much smaller dimensions as well as the different form. Thus while the exoccipital width of the skull of the C. crassangulum is half as great again as that of the C. megalophysum, the length of the bulla is only about three-quarters that of the latter. The bulla of the C. cephalus Cope is very different in form from that of either species. It is of the compressed type and is a little smaller than that of the C. crassangulum. It is truncate both anteriorly and posteriorly, which is not the case with that of the latter species. Although the sagittal crest is crushed away, it is evident that it is much less elevated relatively to the width of the skull than in any of the species so far known. This elevation, allowing for the injury, was about equal to that in the C. megalophysum, measured from the floor of the foramen magnum. The exoccipital width is one-half greater than that of the latter species. The nareal orifice was about as far in advance of the supraoccipital angle as in the C. megalophysum and much further than in any existing whale. The supraorbital portions of the frontal are wanting and the nasals are spread apart laterally in the matrix. Their lateral portions are produced forwards on each side of the nares for a considerable distance, as vertical plates, in a manner which I have not observed in any other Balænid. The proximal extremity of the bone has the vertical laminate suture usual in the family. The bulla is oval and, viewed from below, the extremities are regularly rounded, the posterior but little wider than the anterior. The inferior side is regularly convex in all directions and the interior edge is flattened as in various other species of Cetotherium. The two lateral internal longitudinal angles come together well externally on the anterior end, thus leaving a very short anterior keeled edge. The fate of the angles is not visible posteriorly, as the bulla is in place, but they do not seem to come together. There is preserved of the right mandibular ramus .595 m. from the angle forwards, and a corresponding part of the left ramus, measuring .740 m. The condyles and angles of both sides are preserved. The condyle is compressed, and the articular face presents both upwards and backwards. The angle is broadly truncate, its outline a broad parallelogram, which, when placed vertically, presents its lateral upper angle to the condyle, which obliquely truncates the same. The two surfaces are separated by a shallow groove for only a part of this contact; elsewhere they are continuous. The presence of a coronoid process cannot be positively demonstrated, owing to the position of the rami in the matrix. The ramus of the left side displays its characters at the anterior extremity of the fragment. The convexity of the external face, as usual, exceeds that of the internal, but both are rather flat and meet above at an angle which is a little less than a right angle. Foramina are very few in the portion of the ramus preserved. There is a very shallow groove on the internal side of the superior angle, which is pierced by a single small foramen. On the external side a single foramen of still smaller size pierces the external wall anterior to the position of the internal foramen mentioned, and three times as far below the superior angle. The inferior edge of the left ramus is preserved at a position not far posterior to that just described of the right ramus. It presents an obtuse angle, indicating that a more acute angle exists anterior to its position. Posteriorly the right ramus is rounded more broadly below. The characters of the ramus differ from those of any of the North American species so far known. In S. clarkianus and C. polyporum the foramina are much more numerous at the corresponding locality. The form of the part is different in the S. expansus, C. pusillum and C. cephalus. In C. polyporum and C. cephalus the form is more compressed and the superior edge more acute; in the other species named it is less so. There is no meckelian groove as in the C. palæatlanticum. The three cervical vertebræ diminish in anteroposterior and transverse diameters from the first to the third. The diapophyses of the atlas arise opposite to the base of the neural canal and are short. The distal end of each is depressed. No tuberculum atlantis. The axis is slightly concave transversely below between the bases of the parapophyses. The latter are directed obliquely backwards at an angle of 450 from the articular surface and are vertically expanded at the distal extremity. The parapophyses of the third cervical are as long as those of the second, but are more slender. The humerus is of the size of that of the Cetotherium brialmontii, as represented by Van Beneden, and has the moderately elongate form characteristic of the species of Cetotherium so far as they are known. The tuberosity and crest are broken off, but a portion of the smooth surface which connects the former with the head remains. The distal end is somewhat crushed and the olecranar facet is not as well distinguished from the remainder of the ulnar facet as in most specimens. The humerus is distinguished by two peculiarities. The crest ceases near the middle of the length, and there is a wide and medially deep fossa on the inner side of the shaft immediately beyond the line of the distal end of the crest. Van Beneden figures a somewhat similar fossa in the Cetotherium hupschii Van B., but in that species its position is more proximal and it is bounded anteriorly by the distal portion of the crest. The other species of Cetotherium, and those of Balaenoptera figured by Van Beneden, do not present this fossa. Measurements. Total length of fragment from line of paroccipital M. ......870 ....660 Length on occipital bone from foramen magnum to apex of Length of parapophysis of second cervical from centrum.....135 Anteroposterior diameter of head.. Anteroposterior diameter of shaft at middle.. { .... .130 ..115 Diameters of distal end transverse (somewhat crushed)... .075 anteroposterior. .170 For the opportunity of describing this specimen I am indebted to the Rev. John T. Goucher, President of the Woman's College of Baltimore, who kindly placed the specimen at my disposal. I am also indebted to Prof. Arthur Bibbins, of the same institution, who first drew my attention to it. The specimen was presented to the Woman's College by Dr. Richard Eppes, of City Point, Va., who obtained it from the Yorktown bed at Tarbay, not far from that place. Dr. Eppes discovered in the year 1854 the specimen which became the type of the S. priscus of Leidy, and it is through his hospitality that I have been enabled to visit recently the locality, seven miles below City Point on the James river, where the skull of the C. crassangulum was found. BALENOPTERA SURSIPLANA, sp. nov. Five species of Balænidæ from the Yorktown bed are known from otic bullæ. These are Balana mysticetoides Emmons, Mesoteras kerrianus Cope, Cetotherium cephalus Cope, C. megalophysum Cope, and C. crassangulum Cope. The present species will be the sixth. It is established on a bulla from the Yorktown formation of Maryland, and is in excellent preservation, the middle portion of the inferior thin wall being absent. It is not necessary to compare this species with any of those of the genus Cetotherium. On comparison with the Balanopteræ described by Van Beneden, it is to be observed that they all differ from the present form in the convexity of superior face, where the dense layer or lip has a different chord or face from that of the space which separates it from the internal longitudinal marginal angle. In the B. sursiplana there is but one superior plane from the eustachian orifice to the internal edge, which is absolutely flat. In all these species also the dense layer of the lip is reflected on the superior edge of the external thin wall at its anterior end. In the present species this layer is reflected in a very narrow strip underneath the free border, which overhangs it. In all these species also the anterior extremity, as viewed from above or below, is angulate, the angle marking the end of the inner border of the dense layer or lip. In B. sursiplana the anterior extremity, viewed in the same way, is truncate. The species which appears to approach nearest is the B. definita Owen, which is figured by Lydekker.* This otolite appears to be flatter above than the species described by Van Beneden, although the figure is not clear in this respect. It has the oblique upwards and backwards looking face at the posterior extremity, which is a conspicuous feature of the B. * Quar. Journ. Geolog. Society, 1887, Pl. ii, Fig. 3, p. 11. Cope.] [April 5, sursiplana, although it is not so sharply defined by a strong transverse convexity of the superior surface, as in the latter. Nor is there as strong a bevel of the anterior extremity of the superior face when viewed from within, as in B. definita. An equally conspicuous difference is to be seen in the form of the inferior wall. According to Lydekker this surface, when the bulla is viewed from within, consists of three planes separated by rounded angles, of which the median is longer than those at the ends. In the B. sursiplana this surface is regularly convex from end to end. In size this species is like that of the large Balaenopteræ, including the B. definita. BALENA AFFINIS Owen, Brit. Foss. Mamm. and Birds, 1846; plate opp. p. xlvi. Lydekker, Catal. Foss. Mamm. Brit. Mus., v, p. 17, Fig. 7. Two otic bullæ agree with the figures and descriptions of this species, except in their smaller dimensions. The smallest given by Lydekker is 120 mm. The larger specimen in the Johns Hopkins collection is 105 mm., and the smaller is 95 mm. All of the material described in this paper, excepting the type specimen of the Cetotherium crassangulum, belongs to the Museum of the Johns Hopkins University of Baltimore; and I wish to express my obligations to the authorities of that institution, and especially to Prof. William B. Clark, in charge of the Department of Geology, for the opportunity of studying it. EXPLANATION OF PLATE. Diagrammatic sections of the left mandibular rami of species of Balænidæ, one-half natural size. Lettering E. G., External gingival canal; I. G., Internal gingival canal; G. G., Gingival groove; D. C., Dental canal. |