published early in 1908 (Bull. Amer. Mus. Nat. Hist.) was known only from the wings. A photograph of a wing was sent to Dr. Needham, who wrote: "It is indeed a most interesting fossil, another synthetic type. De Selys' Podagrion group. of Agrioninæ includes the most primitive members of that subfamily, and this fossil is more primitive in several characters than any living forms." Very fortunately, a splendid specimen of Phenacolestes parallelus was uncovered this year by Mr. Geo. N. Rohwer. As the illustration shows, it is nearly complete, lacking, however, the apex of the abdomen. The wings are not so heavily clouded as in P. mirandus, the type of the genus, and there are differences in the venation. P. parallelus was originally described from the apical half of a wing.

SOME FOSSIL BEES

In 1906 (Bull. Mus. Comp. Zool.) I described a bee's wing found at Florissant by Scudder, regarding it as the type of a new Anthophorid genus, Calyptapis. A very

fine example, showing the body, was found this year, and from a close examination I am able to ascertain its true position. It is not an Anthophorid at all, but is a genus of Bombidæ, in other words a bumble-bee. The genus is valid, and gives the first indication of the former history

of this group in America. The insect was especially interesting to me, because I had just been studying the bees in Baltic Amber, which include various genera and species of still earlier bees related to Bombus.

Another bee of great interest was a species of Anthophora, with the mouth-parts exserted and plainly visible. Some of the amber bees show the mouth-parts very well, but it is extremely rare for those in shale to show anything of the kind. The genus Anthophora is common in Colorado to-day, but it was not previously known from the American Tertiaries.

A PROBLEMATICAL FLOWER

Last year we found, among other flowers, one which was so interesting, and so well preserved, that Dr. Arthur Hollick made it the subject of a special article in Torreya, September, 1907. Dr. Hollick named it Phenanthera petalifera, new genus and species, but was unable to place

FIG. 7. Fossil flower, Phenanthera petalifera Hollick.

it definitely in any known family. A new specimen, figured herewith, is clearly of the same species, and on the whole confirms Dr. Hollick's description. The stamens, with long filaments and large anthers, are certainly eight in number. The supposed appendages of the calyx seem to me to be emarginate, and to resemble rather closely the small petals of certain Ribes. Following this clue, the large, thin "petals" may be interpreted as petaloid calyx-lobes, also as in Ribes. The short pedicels, about the length of the hypanthium, suggest that

the flowers were borne in clusters, and so in all respects they seem to agree sufficiently with Ribes, except for the insuperable difficulty of the eight stamens. The eight stamens would agree with Weinmannia, but the flower otherwise seems discordant, judging from the descriptions-I have never seen a Weinmannia flower. Both Weinmannia and Ribes are represented by leaves in the shale.

THE PROBLEM OF THE PROTEACEÆ

The Proteacea constitute a rather large and very characteristic family, with over 950 living species, almost confined to the Southern Hemisphere. Nearly 600 are Australian; New Caledonia has 27, New Zealand 2, Chile 7, tropical South America 36, South Africa over 250, Madagascar 2, and the mountains of tropical Africa about 5. These particulars are taken from Engler (1894), probably the numbers should now be somewhat increased. The genus Helicia, with some 25 species, is IndoMalayan, and extends north of the equator as far as the Himalayas.

One of the most remarkable discoveries-if such it be -of paleobotany is that of the occurrence of Proteacea in abundance in the Tertiaries of the Northern Hemisphere. In Ettinghausen's work on the fossil flora of Häring (1853) numerous remains of leaves are figured, together with drawings of recent species of Proteaceæ. The resemblances are not merely close; it is not too much to say that the oligocene leaves look practically identical with their modern representatives. Furthermore the resemblances are not shown in one or two types only, but extend throughout a considerable series; nor are they confined to the leaves-the determinations in some instances are fortified by characteristic-looking seeds. Even the peculiar fruits of Persoonia are shown. Such evidence looked convincing enough to Ettinghausen, and a priori, there seemed to be no obstacle. The distribution of the Proteacea to-day seemed to be that of a group once world-wide, but now driven to the ends of the earth by the stress of competition. This would agree well with

the case of the marsupial mammalia, and others such as the recently elucidated one of the Chrysochloridæ, or golden moles.

On the other hand, it was pointed out that there were other leaves resembling those of the Proteacea. In 1870 Bentham went so far as to say, in regard to detached leaves, "I do not know of a single one which, in outline or venation, is exclusively characteristic of the order, or of any one of the genera." Quite recently Dr. Schönland (Trans. S. African Phil. Soc., 1907, p. 321) has written: "The supposed identifications of southern types of plants in the Tertiary deposits of the Northern Hemisphere are considered by most eminent botanists, such as Sir Jos. Hooker, the late Mr. G. Bentham, A. Schenk, etc., as worthless. Laurent has recently tried again to prove that the Proteacea originated in the North, but the evidence on which he relies seems to be altogether untrustworthy." Without having seen the European fossils, it may be hazardous to attempt any contribution to this controversy; but it must be pointed out that those who regard the paleontological evidence with contempt seem to have forgotten one or two things. They have not sufficiently remembered the great antiquity of the genera of flowering plants, as shown by indisputable evidence; they have failed to consider the great lapse of time, which would permit migrations from one end of the world to the other (continuous land provided), even at the slowest rate; and more especially, they seem to have forgotten the unquestioned cases of Sequoia, Comptonia, Liquidambar, etc., in which wide-spread types have been reduced to comparatively small areas within quite recent geological times. It may also be added, that they have overlooked the analogous cases among animals, which can by no means be explained away. With all this, it must be confessed that the dicta of paleobotany are not so reliable as we could wish, and that an attitude of scepticism. is often more than justified.

Lesquereux believed that he could recognize a considerable series (8 species) of Proteace in the Florissant

shales. They are by no means so convincing as the European fossils; but they appear to represent an element now wanting in the North American flora, and no one has been able to show that they are not Proteaceous. I give figures of two of the most characteristic-Lomatia acutiloba and Lomatia tripartita. Our new material of L. tripartita is especially interesting as showing-what Lesquereux did not know-that it has compound leaves.

These leaves are exceedingly variable, and have very much the cut of certain species of Phacelia.

This question of the Proteaceæ is one of wide importance, for it is not only a test of the accuracy of paleobotanical conclusions, but, according as it decided one way or the other, it provides or removes an argument for the former existence of great southern lands between the present continents.