to ask whether the glands with which the filaments and base of the corolla are beset may not furnish the nectar. In the American L. ciliata, L. quadrifolia and L. stricta, and on the filaments at least of the European L. vulgaris the glands are very numerous. But upon the flowers of stricta and quadrifolia the Macropis has not yet been found, although the flowers have been often watched; it seems, therefore, that the glands afford no attraction. We must conclude that it is with nectar that the pollen is moistened; and as it has been my good fortune to distinctly observe a female Macropis sucking nectar from the flowers of Rhus glabra, it is, evidently, from these and other flowers that the Macropis obtains the honey for the food both of itself and its young. But why does the Macropis moisten the pollen as it is collected? This is an unusual habit. The social bees moisten it in order that it may be retained on the pollen plates. The Scopulipede and Gastrilege bees retain the dry pollen with the hairs forming the pollen brushes. The Lysimachia pollen is not of so dry a nature that hairs would not hold it. An altogether new interest was given to the genus Macropis by Hermann Müller's observation that it alone of all the solitary bees of Germany moistened the pollen as collected, thus economizing in the expanse of hairs upon the legs.* The retaining hairs upon the posterior legs of Macropis are unusually short. By moistening the pollen they are enabled to retain much larger masses than they otherwise could. Such, also, is the habit, as I have observed, with the allied American genera Scrapter, Calliopsis, and Perdita (P. 8-maculata Say); and Fritz Müller has recorded the same habit for Centris, Tetrapedia and Epicharis in Brazil,† although in these latter genera the scopa is long. On account of the close resemblance which Macropis bears to the higher bees, Shuckard (British Bees) was led to believe that it would be found to agree with them in their noisy flight also. But repeated observations in the field, under the most favorable circumstances, have satisfied me that their flight is perfectly silent. Yet Shuckard is not correct when he says the other Andrenida are mute, for I have observed that certain species of Colleles, C. armata mihi and C. compacta Cress., and possibly, some of the larger species of Andrena, make, during flight, a distinct hum much like that of the honey-bee. * L. c., p. 47, and Anw. d. Darw. Lehre auf Bienen, p. 22 (1872). Nature, vol. x, p. 103. These observations by Fritz Müller are open to doubt. In Centris, as in our native genera Diadasia (n. g.) and Melissodes, the hairs of the scopa are conspicuously plumose, and the pollen would have a matted appearance even when dry. It can be stated with confidence that, even if the pollen is slightly moistened by these bees, it is not formed into a paste, as it is by the social bees. Up to the present time no French* or English author has questioned the validity and naturalness of the two groups, Abeille and Pro-abeille, into which Réaumur divided all the bees. Kirby adopted this classification, employing the names. Apis and Melitta; Latreille adopted it under the names Apiarie and Andreneta; and all subsequent authors have employed the same classification, either under these names or under Leach's family names Apidae and Andrenida. Yet the only characters given for separating the Apidae and Andrenide which are not entirely erroneous are: Apidae; labium longer than mentum, basal joints of labial palpi elongate, labium slender and not flattened. Andrenidae; labium shorter than mentum, basal joints of labial palpi not unlike the following joints, labium flattened. But in the genus Scrapter (placed among the Andrenida) the palpi are precisely as in Calliopsis (placed among the Apida), and, as I have observed, the labium in repose is of precisely the same length-in both extending to the tip of the basal joint of the palpi. The greater breadth of the labium in Scrapter can alone determine to which family it belongs, and this difference in breadth is imaginary rather than real. Moreover, in the genera Megalopta and Oxystoglossa, and some groups of the genus Nomia (genera placed among the Andrenida), the labium is as slender as in the Apidae; and in the genus Hyleoides (placed among the Andrenida) the joints of the labial palpi are proportioned just as in certain of the Apidae. Rejecting, therefore, the families Andrenido and Apide, and without proposing, at present, a more natural classification for the ANTHOPHILA, Macropis may be removed from connection with the short-tongued bees and placed between the Andrenoides and Scopulipedes. In the greater number of its characters it is allied to the Andrenoides, but in single characters of great value it bears relationship to other very diverse groups. With the Andrenoides it agrees in the venation of the anterior wings, which differs from that of Scrapter and Calliopsis in the pointed marginal cell only, in the cleft claws of the female, and in the habit of moistening the pollen as collected. With Andrena it agrees in the form of the tongue and palpi. With the Scopulipedes it agrees in the short anal lobe of the posterior wings and in general appearance. In the form of the basal joint of the posterior tarsi of the female it agrees with none but the social bees, which also have the habit of moistening the pollen as collected. * As Lepeletier failed to recognize the Bees as a natural group, he cannot be said to have presented any classification of them. MACROPIS Panz. (1809). Ocelli in a slight curve; face slightly narrowed beneath; clypens not elevated, yellow in the male; labium transverse, entire; mandibles stout, obtusely bidentate; maxillary palpi 6-jointed, the sixth and one-half of the fifth joints extending beyond the apical lobe of the maxilla; labium lanceolate, onethird the length of the mentum, the latter narrowing toward the base, the paraglossæ small; joints of the labial palpi decreasing in length successively, the basal joint equal in length to the second and third taken together. The flagellum in the female sub-clavate, the first joint ovate, the second narrowed toward the base and one-third longer than the first joint, the third and fourth joints equal and when taken together shorter than the second joint, the apical joint obliquely truncate; in the male the first joint of the flagellum is globose, the second scarcely longer than the first, the third scarcely one-half as long as the second, the fourth about equal in length to each of the following joints, the flagellum not clavate but longer than in the female. The anterior wings have two submarginal cells, the second receiving both recurrent nervures, the origin of the first recurrent nervure far beyond the origin of the cubital nervure; the stigma of good size; submarginal bullæ six, two on the first transverse nervure, one on the second, one on the first recurrent nervure, two on the second; basal lobe of the posterior wings extending beyond the middle of the submedial cell. Both sexes have the tarsal claws cleft and a distinct enclosure at the base of the posterior tibiæ. Posterior femora of the male swollen; posterior tibiæ in both sexes robust; basal joint of the posterior tarsi of the female quadrate, flattened, the upper angle not produced, the second joint attached at the lower angle; the posterior tibia and the basal joint of the posterior tarsi of the female clothed with a short, dense probescence upon which the pollen is collected in moist masses; basal joint of the posterior tarsi of the male armed with a regular comb of long teeth projecting from the inner margin of the lower face. Sixth segment of the abdomen of the female with a smooth enclosure on the disk. The seventh segment in the male with a triangular pyramidal projection on the disk, the apex of the projection obtuse, the anterior and longest side polished. ART. XXXIV.—Additional Remains of Jurassic Mammals; by O. C. MARSH. BESIDE the two mammals from the Jurassic beds of the Rocky Mountains already described by the writer,* two other specimens have recently been brought to light, from the same locality and horizon. Both are lower jaws, and apparently both pertain to the genus Dryolestes, and furnish important characters to distinguish it. In one of these specimens, the angle of the lower jaw is strongly inflected, thus indicating its marsupial nature. The other proves that the genus is quite distinct from Didelphys, as there were at least four premolars. The last lower premolar is compressed and trenchant, and not like the molars. This specimen differs from the jaws of Dryolestes priscus, in being more slender, less curved, and less compressed. The symphysial surface is long, and only moderately roughened. The fourth lower premolar is in perfect preservation. It has two fangs, and the crown is very sharp, and much compressed. There is a slight tubercle on the front margin, and a low distinct heel on the posterior border. The following measurements are from this specimen: Space occupied by four lower premolars, Depth of jaw below first premolar, Depth of jaw below fourth premolar, 6. mm 2.5 3. 2. 2. The species represented by this specimen may be called Dryolestes vorax. The animal appears to have been rather smaller than D. priscus. The only known remains are in the Yale Museum. Yale College, New Haven, August 8, 1879. POSTSCRIPT. Since the above was in type, another lower jaw has been obtained from the same locality and horizon as those already noticed. This specimen is quite distinct from those described from this country, and in some respects resembles the genus Triconodon of Owen, from the Jurassic of England. The molar teeth have each three pointed cones, as in that genus. In the present specimen, however, there are four lower molar teeth, instead of three. The middle cone on each tooth is the largest, while in Triconodon they are nearly of the same size. The last lower molar of the present specimen is only about half as large as those before it. This Journal, vol. xv, p. 459, 1878, and vol. xviii, p. 60, 1879. From Phascolotherium, with which it agrees more closely, the present genus may be distinguished by the greater number of teeth. A striking feature in this jaw is the coronoid process, the anterior margin of which forms a right angle with the ramus, immediately behind the last molar. The angle of this jaw is much extended backward, but not perceptibly inflected. The condyle is low, and but slightly above the dental series. The figure below gives the outline and general features of this specimen. Right lower jaw of Tinodon bellus, Marsh. Twice natural size. The principal dimensions of this specimen are as follows: Space occupied by eight posterior teeth,- Distance from last molar to posterior end of jaw, -- 10. min 9 7. 2.5 2. This specimen indicates a new genus, which may be called Tinodon, and the species Tinodon bellus. The animal thus represented was apparently an insectivorous marsupial,* and in size somewhat smaller than those above noticed. Yale College, August 16, 1879. SCIENTIFIC INTELLIGENCE. I. CHEMISTRY AND PHYSICS. 1. On the Spectrum of Ytterbium.-LECOQ DE Boisbaudran having received from Marignac a portion of his new earth ytterbia, has submitted it to spectroscopic examination. By using the chloride in solution in water, and the induction spark, he obtained a beautiful and distinctive spectrum formed for the most part of bands grouped between the solar lines D and F. Almost all of these bands are shaded from the left toward the right, the spectrum being so placed that the red end is at the left of the observer. The following are the bands observed, their positions being given in scale numbers: (1) Well marked band, slightly shaded from left to right, of intensity ß; the left border nebulous at 103, the apparent center at 1044. (2) Feeble band a little shaded from left to right, its apparent center at 107. (3) Well marked and strongly shaded band, of intensity a; left border sharp at 109, *The elevated coronoid process, and the absence of inflection at the angle, suggest the possibility that this jaw may have belonged to a placental mammal. The latter character, with others of importance, indicate a distinct family, which may be called Tinodontidae. |