therium, which was considered by Paleontologists of twenty-five years ago, as a primitive form, especially in its foot structure, Anoplotherium certainly possesses a number of primitive characters in its manus and pes, such as the separation of the metatarsals, and the non-fusion of the podial elements, but the inadaptive reduction of its digits, as pointed out by Kowalevsky and the peculiar position of the pollux and hallux, excludes the possibility of placing Anoplotherium in the direct line leading to any of the living Artiodactyla.

I propose in this short paper to attempt to prove, that Anoplotherium has been probably derived from Dacrytherium, a closely allied genus, but whose foot structure is normal and which resembles that of many of the early Eocene Artiodactyla such as Cainotherium. Prof Cope' suggested that Cebochorus may have been the ancestor of Anoplotherium, but the structure of the skull in Cebochorus, is already quite modernized, nearly as much so as in the true pigs, consequently I am inclined to think that we shall have to look for some other form as ancestral to Anoplotherium.

The general form of the skull in Dacrytherium is like that of Anoplotherium, however, in Dacrytherium there is a strongly pronounced preorbital fossa, which is absent in Anoplotherium. The crowns of the upper teeth in Dacrytherium are low and primitive in structure. They exhibit rounded external crescents, which are not at all angular. In Anoplotherium, especially the large species, the crowns of the superior true molars are more lengthened than in Dacrytherium and the external crescents are angular and broad. We see this change in many mammalian phyla from extremely low crowned molars, to those which are tending to the hypselodont condition. As regards the intermediate stage, between Dacrytherium and Anoplotherium, as to the height of the molars, this is found in the genus Diplobune.

The lower true molars of Dacrytherium exhibit two internal cones, which is the normal number in the Artiodactyla. It is interesting to record, that I have noticed in a number of young jaws of Dacrytherium in which the true molars were just coming through, that the anterointernal cusp, which is single in the adult, shows a slight reduplication, which is the normal condition in Diplobune. The division of the metaconid is carried still further in the largest species of Anoplotherium, although I have examined many jaws from the Phosphorites of the Anoplotherium, and I can confidently state, that all gradations exist between the complete isolation of the two antero-internal cusps of the typical forms of Anoplotherium, and the single condition of these cusps, Artiodactyla, AMERICAN NATURALIST, Dec., 1888, p. 1083.

which is found in the supposed ancestral genus, Dacrytherium. Accordingly I am not acquainted with any good generic character at present, which will distinguish the so-called genus Diplobune from Anoplot herium, as in many cases in jaws from the Phosphorites, it is impossible to say whether they belong to Anoplotherium or Diplobune. Dr. Henri Filhol informed me that he was of the same opinion, in regard to the validity of the genus Diplobune.

In Dacrytherium the hind foot has at least four well developed toes and the internal digit is not placed at an angle with the others as in Anoplotherium. This structure of the pes is just what one would expect to find in a genus standing in ancestral relationship to the more specialized members of the Anoplotheriida. Granting that Dacrytherium fulfills in most of its characters, what we require of a form, supposed to be ancestral to Anoplotherium, there is still the presence in Daerytherium of a preorbital fossa, which is absent in the skull of Anoplotherium, and also another objection, is, that Dacrytherium has claw-like ungual phalanges, much as in Agriochorus. I believe, however, the extremely compressed ungual phalanges of Dacrytherium is of little weight against this genus being ancestral to Anoplotherium, for in the latter these phalanges are rather compressed, more so than in the normal Artiodactyles, and they could be easily derived from those of Dacrytherium. The structure of the skull is not known in all the species of Anoplotherium, and one of them may have had a skull with a preorbital fossa, which is so characteristic of Dacrytherium.

As is well known, the original specimens of the manus and pes of Anoplotherium commune, which are in the Muséum d'Histoire Naturelle, Paris, show only two well developed digits as restored by Cuvier. This restoration of the feet of Anoplotherium is shown by Schlosser and Zittell to have been an error on the part of Cuvier, and I quite agree with these authors on this point. Prof. Zittell in his "Traité de Palæontologie" in speaking of the structure of the feet in Anoplotherium remarks "La plupart des représentation de la patte d' Anoplotherium faites jusqus à present omettent par erreur à la patte antérieur l'index et le rudiment de pouce, à la patte postérieur le second doigt." I have examined a fine cast of the hind foot of Anoplotherium commune and I find that the restoration of the internal portion as completed by Cuvier is quite erroneous. The two small bones placed by him on the tibular side of the pes do not at all fit the facets on which they are placed. The broad and obliquely placed facet on Mt. 111 in A. commune is for the large and wide spreading second digit, this same structure of the metatarsal occurs in A. (Eurytherium) latipes of the upper Eocene of Débruge.

Summing up the principal changes which have occured in the evolution of Anoplotherium from Dacrytherium, I emphasize the following: 1. Increase in height of the crowns of the upper molars, and the redu plication of the metaconid of the lower molars, this division of the metaconid is found in an incipient condition in young jaws of Dacrytherium. Complete separation of the metaconid into two distinct cusps only occurs in some forms of Anoplotherium. 2. The hind foot of Dacrytherium is normal in structure, and has at least four toes, this is the primitive type of pes, from which the specialized foot of Anoplotherium has been derived.

Note. In my "Notes on the Fossil Mammalia of Europe," part III, AMERICAN NATURALIST, April, 1896, I find two mistakes, which should be corrected. On page 309, third and fifth lines from top, read Adriotherium, instead of Adiotherium as printed, and also page 310, eighth line from the bottom, read Anoplotheriide, in place of Suillines. -CHARLES EARLE.

BOTANY.'

De Toni's Sylloge Algarum.-Dr. De Toni' has recently issued the third volume of his Sylloge Algarum. It deals entirely with the Brown Algae or Phaeophyceae-the FUCOIDEAE as he calls them. A thousand species are described under one hundred and eighty genera, which are grouped into twenty-nine families. He divides the group into three orders, Cyclosporina, (Fucaceae) Tetrasporina (Dictyotes) Phaozoosporina (Phæozoosporea).

Splanchnidium rugosum the interesting plant which after careful study was placed by M. O. Mitchell and F. G. Whiting in the Photosporina, is retained in the Durvilleacea, the fruit being described as a polysporous oogone. The general appearance of the plant and the structure of the conceptacles suggest a close relationship with the fucoids, but if the above investigations are to be accepted the plant 1 Edited by Prof. C. E. Bessey, University of Nebraska, Lincoln, Nebraska.

2 Sylloge Algarum Omnium Hucusque cognitarum by J. Bapt. De Toni, Vol. III, Fucoidea.

3 On Splanchnidium rugosum Grev. the type of a new order of Algae, Phycological Memoirs, Pt. I. I., 1892.

bears zoospores in the conceptacles and not oogones, hence it must be placed in the Zöosporina.

The treatment of the Zoosporine is practically that of Kjellman in Engler and Prantl's, Pflanzenfamilien, except that the genera Lithoderma and Arthrocladia are placed in families by themselves, instead of in the Ralfsiaceae and Desmarestiacea respectively, and that De Toni has included five small, mostly, monogeneric families, the Phaothamniaceœ, Phaocapsacea, Hydruracea, Chromonodacea and Chromophytonacea not mentioned by Kjellman. In all the Zöosporinæ except the above families the zoospores as far as known are laterally biciliated and áre borne in some form of zöosporangia. In these families there are no zoosporangia and in at least a part of them the zoospores are not laterally biciliated and in general their relationship seems to be with the Chlorophyceae. It seems more natural to place them, as Wille has with some of them, in the Chlorophyceae next to their closely related genera.

The book is well arranged; priority in class, ordinal and family. nomenclature is strictly observed. It will be indispensible to the specialist in this line and a great help to the general student.-DE ALTON SAUNDERS.

The Flora of the Black Hills of South Dakota.-In a recent number of the Contributions from the U. S. National Herbarium (Vol. III, No. 8; issued June 13, 1896), P. A. Rydberg gives the results of his explorations (in 1892) of the Black Hills of South Dakota. The report, which includes about eighty pages, includes the following, viz. : Itinerary, Geography, Geology, Altitudes, Precipitation and Temperature, Floral Districts, General Remarks, and the Catalogue of Species. The plates are a Map of the Black Hills, Aquilegia brevistyla, Aquilegia saximontana and Poa pseudopratensis. The floral districts recog nized by the author are five, viz.: (1), the foothills and surrounding plains, (2), the Minnekata Plains, (3), the Harney Mountain Range, (4), the Limestone District, (5), the Northern Hills.

In summing up his discussion of the vegetation of these districts the author says, "From the foregoing can be seen what a varied flora the Black Hills have. There are found plants from the East, from the Saskatchewan region, from the prairies and table-lands west of the Missouri River, from the Rocky Mountains, and even from the region west thereof. In the foothills and the lower parts of the Hills proper the flora is essentially the same as that of the surrounding plains, with an addition of eastern plants that have ascended the streams. In the higher parts the flora is more of a Northern origin. Most of the plants

composing it are of a more or less transcontinental distribution, but often characteristic of a higher latitude. Some can be said to belong

to the Rocky Mountain Region. The only trees of western origin are Pinus ponderosa scopulorum, and Betula occidentalis; the others are eastern, or transcontinental. The flora resembles, therefore, more that of the region around the Great Lakes than that of the Rockies."

It merely remains to say that the nomenclature and capitalization (all specific names decapitalized) of this interesting and valuable report are of the most advanced type.-CHARLES E. BESSEY.

Trelease's Hickories and Walnuts of the United States.— Dr. Trelease has rendered a good service to the botanists of the country by publishing (in the Seventh Annual Report of the Missouri Botanical Garden) the results of his studies of the Juglandaceae of the United States, especially with reference to their winter characters. The species recognized are:

Hicoria pecan (Marshall) Britton.-Iowa to Southern Indiana, Kentucky, Louisiana and Texas, extending into Mexico.

H.myristica formis (Michx. f.) Britton.-Arkansas to Alabama, Texas and Mexico, and in South Carolina.

H. aquatica (Michx. f.) Britton.-Virginia to Florida, around the Gulf to Texas, thence north to Arkansas and southern Illinois.

H. minima (Marshall) Britton.-Canada and Maine to Minnesota and Nebraska, south to Texas and Florida.

H. glabra (Miller) Britton.-Atlantic region from Massachusetts and Pennsylvania to Florida.-var. odorata (Marshall) Sargent.— Mississippi valley eastward, and from Canada to the Gulf.-var. villosa Sargent.-Missouri, on flinty hills.-var. microcarpa (Nuttall) Sargent. Same range as var. odorata.

H. alba (L.) Britton.-Canada to the Great Lakes and Kansas, south to Texas and Florida.

H. mexicana (Engelm.) Britton.--Mexico, in mountains of Alvarez. H. laciniosa (Michx.) Sargent.-New York and Pennsylvania to Iowa, Kansas and the Indian Territory.

H. ovata (Miller) Britton.-Canada to Minnesota, south to Florida, Kansas and Texas.

Juglans cinerea L.-New Brunswick to Dakota, Kansas, and the Mountains of Georgia and Alabama.

J. rupestris Engelmann.-Texas, New Mexico and Arizona, extending into Mexico.

J. californica Watson.-Coast range of southern California.