It would not be surprising if this genus should prove to be related to Anoplonassa Cope, which has the long symphysis mandibuli of the Physeter, with the nearly edentulous character of the Choneziphiida. Char. specif.-The species is founded on a rostrum similar in its mode of preservation to Ziphioids in general. Nothing is preserved posterior to the nares, and the edge of the left maxillary, with the end of the muzzle, is broken off. All the parts are coössified. verse. The anteroposterior diameter of the basin exceeds by a little the transThe bottom is nearly regularly concave, with a few shallow fossæ at the right side. The part of the median vomerine ridge which forms the anterior half of the left border of the basin is thicker than that which bounds the posterior half of the same, and it presents an angular tuberosity horizontally to the left. The fractured edge of the left maxillary shows that it was thin at this point, and at no point had it when perfect any considerable horizontal extension to the left side. It is separated from the vomerine ridge by a groove, which extends from the left maxillary basin to a foramen, which is situated at a point in advance of the greatest width of the vomerine rostrum. A corresponding foramen exists on the right side, which is opposite the anterior border of the central basin. The vomerine rostrum contracts rapidly forwards, forming a prominent rounded ridge, and the premaxillaries rise to it on each side. The maxillaries are little expanded, and their superior surface is subhorizontal and is moderately rugose. It is much thicker on the right side than on the left, and is probably also wider on the right side. On the left side it is bounded below by a deep groove from opposite the anterior part of the left maxillary basin to a point in front of and below the supramaxillary foramen of the same side already described. On the right side there is a similar foramen which bears the same relation to the supramaxillary foramen of the same side. In front of the foramen on the left side the groove continues. On the right side the groove does not continue, but is succeeded forwards by a narrow vertical wall to the anterior extremity of the specimen. On the inferior side the middle line is deeply grooved to a point opposite the middle of the superior median basin. On each side of the groove the palatal aspect of the maxillary slopes slightly upwards, and on the left side rolls up to a thickened inferior border of the inferior or submaxillary groove. On the right side the palatal face of the maxillary turns up more obliquely to the border of the superior median basin. 1895.] Measurements. Width of rostrum at end of specimen...... Width of lateral edge of right maxillary at inferior maxil lary foramen................. Length of inferior keel of vomer... [Cope. MM. 35 25 120 This species exhibits the most unsymmetrical cetacean cranium known to me. Its size was probably about that of the Choneziphius indicus of the present ocean. Its exact locality is unknown, but it probably was exposed to the action of water for a considerable time, after being washed from its position of deposit. It has been bored by Pholades in several places. BALENIDE. I have remarked that the Mysticete with its single family the Balanida "would seem to bave derived their descent from some form allied to the Squalodontidæ, since their nasal bones are more elongated than those of the Odontoceti, and in Plesiocetus" (Cetotherium) "the superior cranial bones show some of the elongation of that family." This elongation of the superior cranial wall is not seen in the genus Squalodon, but is moderately developed in the genus Prosqualodon of Lydekker, founded on the P. australis Lydd. (l. c.) from Patagonia. It is exhibited in a still more marked degree by the genus Agorophius g. n. Cope, which is represented by the Zeuglodon pygmæus of Müller, which was referred to Squalodon by Leidy. The form of the skull in this genus approaches distinctly that of Cetotherium of the Balænidæ, and the permanent loss of the teeth would probably render it necessary to refer it to the Mystacocete. Stages of transition from some such genus as Agorophius to the typical whalebone whales are represented by several genera from the Yorktown beds. Theoretically the loss of teeth by failure to develop would be accompanied by the loss of the interalveolar walls, leaving the dental groove continuous and separate from the dental canal. A genus displaying these characters has not been discovered, but I have no doubt that it will be. The new genus Siphonocetus Cope exhibits the groove roofed over by ossification of the gum, and distinct from the dental canal. The genus Ulias indicates that a still farther degeneracy took place, in the fusion of the dental groove and dental canal, while the groove remained open. In Tretulias the same condition persists with the addition that the gingival passages and foramina are present, as in the genus Siphonocetus, and in "On the Cetacea," American Naturalist, 1890, p. 611; Meyer, Studien über Säugethiere, Jena, 1886, p. 191. + Extinct Mamm. Dakota and Nebraska, 1869, p. 420, Pl. xix, Fig. 8. It is doubtful whether this genus should be referred to the Zeuglodontidæ or the Squalodontidæ. The relations of the maxillary and premaxillary bones posteriorly resemble most those of the latter family. It differs from Prosqualodon in the contact of the temporal fossæ on the middle line above, and in the greater elongation of the frontal and probably nasal bones. the later genera. In Cetotherium and in later Balanidæ the groove and canal are fused, the gingival roof is complete, and it is perforated. It would appear, then, that Ulias may be descended from the undiscovered genus above mentioned, while Tretulias is descended from Siphonocetus. The exclusively Neocene genera may be tabulated as follows: I. Alveolar groove and dental canal distinct. Alveolar groove open.. Alveolar groove roofed over and perforate.... .Not discovered. .....Siphonocetus Cope. II. Alveolar groove and dental groove confluent in a gingivodental canal. . Ulias Cope. Tretulias Cope. Cetotherium Brandt. Gingivodental canal open; no gingival canals. SIPHONOCETUS EXPANSUS Cope. Acad. Philada., 1869, p. 11. Balana prisca Leidy, ProEschrichtius priscus Leidy, 11. Leidy, Extinct Mamm. Eschrichtius expansus Cope, Proceeds. The two mandibular rami ascribed to this species are the property of the Maryland Academy of Sciences. The collection of Johns Hopkins University contains a fragment of a ramus of an individual of rather smaller size than the types. Plate vi, Fig. 5. SIPHONOCETUS CLARKIANUs Cope, sp. nov. Plate vi, Fig. 4. In the collection of the Johns Hopkins University there are portions of mandibular rami of two species of Siphonocetus. The cranium of this genus is unknown, but it is probably similar in character to that of the Cetotherium of Brandt. This genus differs from Balaenoptera in having the elements between the supraoccipital and the nasals much elongated, so that there is a sagittal crest of greater or less length, and in the nonunion of the dia- and parapophyses into a vertebral canal, in which it agrees with Eschrichtius of Gray. Some of the rami described belong possibly to species of Balaenoptera, and it remains for future discoveries to ascertain which these are. One of the species above referred to is the Siphonocetus expansus Cope.t The other species differs from all of those known to me. In dimensions it exceeds those of any of the species described in this paper, and is only exceeded by the species which I have described (1. c.) under the names * See American Naturalist, 1890, p. 611, where these genera are characterized; but Van Beneden's name, Plesiocetus, is used for Cetotherium, and the latter name for Eschrichtius of Gray. † Proceeds. Academy Philada., 1869, p. 11. 1895.] [Cope. of Cetotherium leptocentrum and C. cephalus. It compares more closely in dimensions with the C. polyporum from the Chesapeake formation of North Carolina. From the last named, and from the C. cephalus, it differs in the robust form of the ramus, resembling in this respect rather such species as C. palæatlanticum, S. priscus (Leidy), and S. expansus. The fragment representing the S. clarkianus is from the part of the ramus anterior to the base of the coronoid process, and is about 350 mm. in length. Both faces are convex, but the external is more strongly so than the internal. The superior part of the latter is, however, not horizontal as in the S. priscus, nor is the internal face subhorizontal as in S. expansus. The two faces unite above at an obtuse angle, which if perfect, would be nearly right. The inferior edge is on the contrary a ridge which would be acute were it not rounded. The section of the ramus is therefore lenticular, with one side more convex than the other. Posteriorly the external convexity becomes greater, and the internal convexity rises towards the base of the coronoid, leaving a gentle concavity above the inferior border. The external foramina are large, distant, and only a little further below the superior ridge than those of the inferior interual row. The latter are in two series; those of the superior smaller and quite near the superior edge; the others larger and situated lower down, and separated by intervals of about 40 mm. No trace of Meckelian or alveolar grooves. The presence of two internal series of foramina distinguishes this species from any of those known to me. The rami are less compressed than those of the C. pusillum, while the external position of the external foramina distinguishes it from the S. priscus (Leidy). The presence of an acute-angled ridge below distinguishes it strongly from the C. palæatlanticum. The species was larger than the Cetotherium megalophysum above described, having probably attained a length of forty feet. I have dedicated it to Prof. William B. Clark, of the Department of Geology and Paleontology of the Johns Hopkins University, to whom I am under obligations for the opportunity of studying most of the material here described. The label attached to the specimen states that it was dredged up near Point-no-Point, Chesapeake Bay, and was presented to Johns Hopkins University by Conrad Miller. ULIAS MORATUS, gen. et sp. nov. Plate vi, Fig. 1. Char. gen.-Mandible with the gingivodental canal open throughout most of its length, closed only near its apex. Gingival foramina represented by a few orifices on the alveolar border near the distal extremity. Cope.] [April 5, This form is of much interest as representing in adult life a stage which is transitional in typical Balænidæ. The alveolar groove is continuous with the dental canal, and is permanently open. It is probable then that this genus possessed teeth during a longer period than the existing Balænidæ, and that they were retained in place by a gum so long that the canal could not close as is the case in the latter. The absence of the long series of mental foramina characteristic of the true whales is further evidence to this effect. Van Beneden, in his Descriptions des Ossemens Fossiles des Environs d'Anvers, describes a Balænid under the name of Balanula balanopsis which is represented by numerous individuals. I agree with Lydekker (Catalogue of Fossil Mammalia in the British Museum, Vol. v) that Van Beneden has not given sufficient reason for separating the species generically from Balæna. There is also considerable diversity between individuals referred to the species. In a small specimen, a narrow alveolar groove is present, but Van Beneden makes no reference to the character in his description. As the groove is closed in large specimens figured, it is possible that M. Van Beneden regarded the character as one of immaturity. This may be the case, as the groove figured by Van Beneden is a very narrow one, and is quite different from the widely gaping channel in the Ulias moratus, which is founded on an adult animal. Char. specif.-This species is founded on a nearly entire right mandibular ramus. The condyle and angle are wanting, as is also a piece from the proximal part of the distal third of the length. This piece was found with the rest of the specimen, but has been, for the present at least, mislaid. The ramus is moderately curved horizontally, but is not decurved except towards the angle. A slight convexity of the inferior margin exists at the anterior part of the proximal two-fifths of the length. The superior border is occupied with the widely open alveolar groove, which gradually contracts in transverse diameter distally, so as to be closed for the terminal fourth of the length. On this region two or three large foramina issue from it on the middle line above, and the terminal mental foramen issues at the superior extremity of the distal end, a little below the internal ridge on the external side of it. Of the borders of the alveolar groove the internal is much lower than the external on the proximal sixth of the length. The edges are then equal for a short distance, and are acute. The internal then becomes the more elevated, and continues so until its point of union with the external. The internal wall of the groove is at first narrow, and its superior edge from being acute becomes narrowly rounded, but becomes more obtuse distally as the wall becomes thicker. The internal side of the ramus is very little convex. The external side of the ramus is strongly convex in vertical section, hence it is that the external edge of the groove becomes wider as it becomes lower, until at the beginning of the distal third of the length it forms a plane distinct from the convex external face. This external convexity growing rapidly less, the |