has so little foundation as to weaken rather than strengthen his argument. It is only referred to as being on the level with the rest of the paper and to show what a large proportion of it is assumption, and how slight is its foundation of fact.

The method of determining the kind and use of implements, their mode of manufacture and the expected benefit which induced the prehistoric man to expend himself upon them, by comparing him with ourselves, with putting him in our places, or putting ourselves with our knowledge and skill, culture and spirit of invention, in his place, and then deciding everything he did from our standpoint, is one of the errors of modern archeologists and one which leads them far from the right path. This discussion leads Mr. Holmes into the processes of the manufacture of flaked stone tools, and he explains direct or free-hand percussion, declares its limitation, how it was the only method known in early times, and throughout pages 15 and 16, explains the details, giving philosophic dissertations upon the art of stone flaking or chipping and, of the development of the spirit and technology of art required in this work of making leaf-shaped blades. No better answer could be given to this theory than the exhibition of the finely flaked flint implements all prehistoric, coming, some of them, from Scandinavia, Mexico, and a large number belonging to the paleolithic period, found throughout the interior of France in what M. de Mortillet calls the "Solutrèen epoch," M. Reinach and others, the "Cavern epoch." Yet Mr. Holmes has never been able to reproduce one of them or to overcome the difficulties of their fabrication.

(To be Continued.)

THE GEOGRAPHICAL DISTRIBUTION OF BATRACHIA AND REPTILIA IN NORTH AMERICA.

By E. D. COPE.

As is well known, the aggregates of organic beings called faunæ and floræ correspond in part with the natural land divisions of the earth's surface, but not exactly. The first classification of the primary fauna was proposed by Dr. P. L. Sclater in 1858, as follows:

1. Paleartic.-Europe, Northern Africa, Northern and Central Asia.

2. Ethiopian.-Africa south of the Great Desert, and Mada

gascar.

3. Indian.-Southeastern Asia and the Malay Archipelago. 4. Australian.-Australia with New Guinea and the adjacent islands, New Zealand and Polynesia.

5. Nearctic.-North America as far south as Mexico.

6. Neotropical.-Central and South America and the West Indies.

Subsequently Dr. A. R. Wallace proposed that the name Oriental be used in place of Indian.

In 1868 Prof. T. H. Huxley proposed that the world's areas be arranged in two divisions, Arctogæa and Notogæa; the former including the Palearctic, Indian, Ethiopian and Nearctic of Sclater, and the latter including the Australian and Neotropical regions. To the last two he added the Novo Zealanian for New Zealand, and he proposes to change the name of the Neotropical to Austrocolumbian.

In 1871 Dr. J. A. Allen proposed the following faunal divisions: I. Arctic Realm; II. North Temperate Realm; III. American Tropical Realm; IV. Indo-African Tropical Realm; V. South American Temperate Realm; VI. African Temperate Realm; VII. Antarctic Realm; VIII. Australian Realm. In 1874 Sclater modified his system as follows: He retained the term Arctogea in the Huxleyan sense. To the Neotropical region he gave the name of Dendtrogea, and to the Australian

he gave the name Antarctogea, omitting New Zealand and Polynesia, which he constituted a fourth division, Ornithogea.

In 1878 Heilprin proposed the name Holarctic, to include Sclater's Palearctic and Nearctic regions. He also proposed two transitional regions; that of the Old World he called Mediterranean and that of the New World the Sonoran, the latter a term already introduced by Cope for a division of the Nearctic of Sclater.

In 1884 Gill proposed the following primary divisions or realms: 1. Anglogæan (N. American); 2. Eurygæan, or Eurasian; 3. Indogæan; 4. Afrogæan; 5. Dendrogæan, or Tropical American; 6. Amphigæan, or Temperate South American; 7. Austrogæan, or Australian; 8. Ornithogæan, or New Zealand; 10. Nesogæan, or Polynesian. Prof. Gill justly insisted on the importance of fresh water fishes as furnishing definitions of natural faunal realms and regions.

In 1890 Blanford published a system of geographic zoology in which he adopted the primary divisions of Huxley, and divided the Arctogaean region into the following: Malagasy, Ethiopian, Oriental, Aqulonian (= Palearctic and northern part of Nearctic), and Medio-Columbian (S. part of Nearctic).

In 1896 Lydekker proposed the following divisions: I. Notogæic Realm; regions: 1. Australian; 2. Polynesian; 3. Hawaiian; 4. Austromalayan. II. Neogæic Realm; regions: Neotropical. III. Arctogæic Realm; regions: 1. Malagasy; 2. Ethiopian; 3. Oriental; 4. Holarctic; 5. Sonoran. Lydekker makes use of paleontologic evidence in this connection. While this treatment of the subject is important from the point of view of origin, it is often irrelevant, since the distribution of vertebrate life in each geologic age was different from that in each other geologic age.

In an essay on the geographical distribution of North American Reptilia published in 1875, the present writer adopted the first system of Sclater. After a lapse of twenty years, the light thrown on the subject by various investigators suggests the following modifications. In the first place the recognition of the close similarity of the life of the northern regions of the earth, requires more definite formulation than was accorded it in

Sclater's first system, by the union of his three divisions of Nearctic, Palearctic and Indian into one, for which the name Arctogæa is appropriate. The enclosure of his Ethiopian division in it as proposed by Huxley, does not seem to me to be proper, in view of the important types of fishes and reptiles which characterize it; e. g., the Crossopterygian, Dipnoan and Scyphophorous fishes, and the Pleurodire tortoises. In the fishes, indeed, the Ethiopian region has as much affinity with the Neotropical fauna as with any other, in its Characin and and Cichlid families, and in the Dipnoan subclass. The presence of the Dipnoi and the Pleurodire tortoises ally it to the Australian fauna as well. It is for these reasons that Prof. Gill proposes to combine the southern hemisphere realms into a single "Eogaean" division. The northern affinities of the Ethiopian realm are, however, too many to permit us to regard this arrangement as a just expression of the facts. Thus, it has Insectivorous Mammalia, Firmisternial Batrachia Anura, and Cyprinid fishes, none of which are Australian or Neotropical types. The course that remains under the circumstances is to regard the Ethiopian Realm as fully distinct from the other three. The definitions of the four primary divisions are then as follows:

The Australian realm is peculiar in the absence of nearly all types of Mammalia, except the Ornithodelphia and the Marsupials; in the presence of various Ratite birds, in great development of the Proteroglyph serpents, and absence of the higher division of both snakes and frogs; i. e., Solenoglypha and Firmisternia; in the existence of Dipnoi (Ceratodus) and certain Isospondylous families of fishes. On the other hand many of the lizards and birds are of the higher types that prevail in India and Africa, viz.: the Agamida and the Oscines.

The Neotropical realm only possesses exclusively the Platyrhine monkeys and the great majority of the humming birds. It shares with the other Southern regions the Edentate and Tapiroid mammals; Ratite, Pullastrine, and Clamatorial birds; Proteroglyph snakes; Iguanid Lacertilia, the Agamids being entirely absent; Arciferous frogs; and Characin, Chromid, Osteoglossid, and Dipnoan fishes. It has but few types of the

Northern regions; these are a few bears, deer, and oscine birds. Insectivorous mammalia, Viperid serpents, and Ginglymodous, Halecomorphous and Cyprinid fishes are wanting, except on the northern border.

The Ethiopean realm is that one which combines the prevalent features of the Arctogean realm with the southern hemisphere types already mentioned, together with some found elsewhere only in the Indian region, and a very few peculiar. The two latter classes not being mentioned elsewhere, they may be here enumerated. The region shares, with the Indian alone, the Catarrhine monkeys, the Elephantida, Rhinocerotidæ, Nomarthrous Edentata and Chameleons. Its peculiar types are the Lemurida, Hippopotamidæ and Protelidæ, Cryptoproctidæ and Hyracoidea among mammals, and Polypteridæ and Mormyrida among fishes. It possesses in common with the Neotropical realm characinid, cychlidi, and dipnoan fishes, Pleurodire tortoises and Ratite and Trogonoid birds; and differs from it in the absence of arciferous Batrachia and crotalid snakes, and presence of dendraspid, causid, atractaspid and viperid snakes.

The Arctogean Realm is characterized by the absence of types conspicuous elsewhere, and by the presence of a few peculiar forms. Among fishes it lacks Dipnoi and Crossopterygia, Osteoglossidæ, Characinidæ and Cichlidæ. It lacks Pleurodire tortoises and Ratite birds. Gingly modous fishes and Urodele Batrachia are nearly confined to it, merely extending a little over the border of the Neotropical. Its Cryptodire tortoises extend both into the Neotropical and Ethiopian. Anguid lizards are confined to it. It shares most of its Mammalia with other regions. The Insectivora it shares with the Ethiopian, and its deer and camels with the Neotropical. The genus Ursus is very characteristic, one aberrant species only extending into the Neotropical.

From what has preceded it is seen that the primary differences between the fauna of the realms are to be found to a large degree in the lower vertebrata, the fishes, Batrachia and Reptilia. These forms furnish stronger distinctions than the birds and mammals, owing to their greater inability to traverse