Hayti. Where the species belong to continental genera not represented in North America, they are more generally of Mexican and Central American forms and rarely of strictly South American.

The following table of resident land birds of Cuba and Jamaica, exclusive of diurnal Raptores and Columbidæ, although approximately complete only, may serve to illustrate more fully the preceding remarks.

The species of truly West Indian birds are remarkable for their local distribution, comparatively few being found on more than one of the larger islands, and, what is still more remarkable when the contrary is the case, an intervening island may be destitute of the species. Thus Cuba lacks several species common to the Bahamas and to Jamaica.

Professor Agassiz (Types of Mankind, 1854), has urged very strongly the recognition of an Arctic and an Antarctic region or "realm," a point in favor of which there is much to be said, but which cannot be discussed in the present article. He also anticipates Dr. Sclater in regard to some of his views, but the facts at command at the time did not allow him to define the boundary lines of the regions with the same precision. Still more recently Dr. Pelzeln (Reise der Novara, 1865) insists likewise upon an Antarctic region.

Proceeding now to the, especial subject of the present article, the mapping out of North America with reference to the geographical distribution and migrations of North American birds, it may be premised that in the article above referred to by Professor Agassiz, in Nott & Gliddon's Types of Mankind, we find the first attempt to mark off the zoological provinces of the New World-and very successful considering the insufficient data accessible at the time. In 1859 Dr. Leconte sketched out their

Phonipara.

Coleoptera of Kansas and New Mexico, Dec. 1859, Smithsonian Contributions, vol. xi.

I may also refer to incidental mention of the same law in a paper by myself on the birds of Cape St. Lucas, in the Proceedings of the Philad. Academy for Nov. 8th, 1859, p. 299.

AM, JOUR. SCI.-Second SERIES, VOL. XLI, No. 121.-JAN., 1866.

boundaries, in North America, with more precision, having particular reference to the distribution of Coleopterous insects.

The subdivisions by Dr. Leconte of these provinces, as based on the study of their Coleoptera, do not agree strictly with those of the ornithological fauna, especially in the considerable number of local areas which he has adopted. This difference is, however, easily intelligible when we bear in mind the much superior power of flight and innate tendency to migration of the bird as compared with the insect; the distribution of reptiles agreeing much better with his outline than that of birds.

To present the general principles of distribution to which I have been led by an examination of the large collection of specimens in the museum of the Smithsonian Institution, I may say that as far as its ornithology, and to a considerable degree its vertebrate zoology in general is concerned, North America appears to be divided into two great regions, an eastern and a western, which in the United States are of approximately equal extent, but very unequal farther north. The eastern division extends from the Atlantic seaboard, westward across the Alleghanies (which affect the distribution of species but little) and over the valley of the Mississippi and its fertile prairies to about the 100th degree of longitude, or to the beginning of the sterile plains. Its western border is not sharply defined, nor strictly in a meridian line, but somewhat oblique, and interdigitates with the western division by extending westward along the river bottoms, some species, as Galeoscoptes Carolinensis, Vireo olivaceus, &c. occurring as far west as Fort Benton, or even Fort Colville.

The western division begins at the western border of the eastern, or along the sterile plains of the trans-Mississippi country and extends across to the Pacific ocean. The character of the ornithological fauna of this division is much the same through and beyond the Rocky Mountains to the eastern slope of the Sierra Nevada and Cascade Mountains of California and Oregon, but changes somewhat on the western slope and thence to the Pacific, and although to a considerable extent uniform, yet exhibits some modifications which may warrant a separation into a western and middle division, making three in all, which we may call provinces, of very unequal extent, and exhibiting further modifications or subdivisions with latitude, as I shall proceed to explain, taking into consideration the whole continent north of Mexico.

As previously remarked, the eastern province or division extends from the Atlantic ocean to about the meridian of 100° west from Greenwich, or 23° west from Washington. The line of division on the Gulf of Mexico, starts near the eastern border of Texas, perhaps between the Brazos and the Sabine, and fol

lowing up the direction of the former river to the approaches of the Great Desert nearly on the meridian mentioned, proceeds northward, forced sometimes more or less westward, especially along the Platte, sometimes eastward. It crosses the Platte between Forts Kearney and Laramie and intersects the Missouri between Fort Randall and Fort Pierre, perhaps near Fort Lookout, as it is between the first mentioned two points that in ascending the river we find the change to take place in the ornithology of the country. Soon after crossing the northern boundary of the United States and to the western side of Lake Winnipeg, the line rapidly inclines westward, especially beyond the Saskatchewan, and extends to the Rocky Mountains, including the valleys of Athabasca and Great Slave Lakes, and both sides of the Mackenzie River, north to the Arctic ocean, even crossing the Rocky Mountains to the Porcupine river and into Russian America at least to 145°, or beyond the forks of the Yukon, where Mr. Kennicott found many of the most characteristic summer land birds to be almost identical with those of Slave Lake, Lake Winnipeg, and Northern Canada.

The western province occupies the western slope of the Cascade and Sierra Nevada ranges of mountains in the United States, although its extent southward along the peninsula of Lower California is not well determined. To the northwest it extends at least to the 140th meridian, beyond that probably replaced by a more Arctic fauna. We are not sufficiently familiar with the birds occurring between the northern Rocky Mountains and the coast, to tell how far inland in Stickin Territory or even in northern British Columbia, the coast fauna extends, perhaps not farther than in California or Oregon, although it is possible that, owing to the absence of a continuous longitudinal range of great height, the western and middle regions may there be more thoroughly blended into one.

The middle province, or that of the great plateau, occupies the space between the two just mentioned, probably not passing in its integrity, or as a peculiar province north of the valley of the Saskatchewan and is thus wedged in between the two. As already stated, it extends along the eastern slope of the Cascade and Sierra Nevada mountains, and apparently along the east side of Lower California to Cape St. Lucas, at least the birds of the Cape, as will hereafter be explained, belong much more emphatically to it than to the Western province. A break in the mountains opposite San Diego explains the appearance at that point on the coast of a few species like Tyrannus vociferans, Sialia arctica, Polioptila melanura, &c., so characteristic of the middle province. The southern boundary of this province during the summer may be considered as occupying the valleys of the Rio Grande and Gila but along this line it is greatly mixed up with

the peculiar fauna of Northern Mexico, which as far as its summer birds indicate, is almost entitled to be considered as a fourth main province.

The eastern province to the north merges into the Arctic, and southward exhibits a very important subdivision in the hot region of the south Atlantic and Gulf States, which is bounded to the north by the isothermal of 80°, extending however up the coast to the Dismal Swamp of Virginia, or even to the James river. To the west it ranges along the isothermal of 83° or 85° following the line to the N.N.W. along the valleys of the Brazos, Red river, the Washita and the Canadian. Most of the species belonging to this subdivison reach along the valley of the Mississippi to a point far north of their limit on the Atlantic slope; the Swallow-tailed-Hawk, Parakeet, and other characteristic species, being well known visitors to Cairo, St. Louis, and even as far north as Wisconsin. This subdivision of the eastern province experiences a still further modification in the southern part of Florida in consequence of the proximity of the Bahamas and Cuba, which causes stragglers of the West India fauna to enter its limits, especially along the south eastern keys. Some of these are Certhiola Bahamensis, Progne cryptoleuca? Vireo barbatula, Quiscalus aglaeus (Q. baritus, Baird, B. N. A., 556), etc. The only really peculiar indigenous land bird in Florida is the Florida Jay (Cyanocitta Floridana), seldom, if ever, found out of that State. As far as is known, there is no corresponding southern subdivision on the west coast in the western province, although California and Washington Territory have each some peculiar species.

As in the eastern province, so in the middle, there is a subdivision along the southern border inhabited by species belonging more particularly to northern Mexico, and occupying the valleys of the Rio Grande and Gila, extending northward along the Rio Grande and Colorado far into the United States. It is the species of this subdivision, that, with those peculiar to Cape St. Lucas, characterize the summer fauna of the latter region. In winter, both there and along the Mexican boundary line, these species are mingled with others coming from the more northern portions of the middle province.

In addition, however, to possessing certain species of the boundary line fauna, Cape St. Lucas has other peculiarities which entitle it to especial consideration.'

It forms a distinct subdivision of the boundary sub-province even more peculiar in its relations than Florida, where the characteristic species (excepting the Florida Jay) are stragglers of the West Indian type from the Bahamas, while as shown by the

See Baird, Pr. Acad. Nat. Sci., Nov. 8th, 1859.

indefatigable researches of Mr. Xantus, there are at least twenty species found at Cape St. Lucas not known elsewhere.

Very few of the birds of the coast of California, or of the western province, winter at Cape St. Lucas, the species being almost entirely those of the middle province. The new and peculiar species in all cases belong to genera of the middle province, especially of its boundary subdivision, and no genera are peculiar to it. Furthermore, in no instance do we find species of the Tierra Caliente of Mexico not belonging to the United States, nor of any Mexican genera that do not possess representatives in the United States. The difference between the species of birds of Cape St. Lucas and of Mazatlan is very great, although separated only by the breadth of the Gulf of California.

From all these considerations we are legitimately entitled to claim Lower California, or at least its southern extremity, as belonging to temperate North America, even more positively than Florida itself.

Peculiarities in regard to the size of Cape St. Lucas birds will hereafter be referred to.

There is of course an Arctic province which melts gradually into those great provinces mentioned, and along the mountainous ranges extending far southward, in fact almost into Mexico, as shown by the occurrence in summer at Cantonment Burgwyn, near Lat. 37°, of Lagopus leucurus, Pinicola Canadensis, Curvirostra Americana, Hesperiphona vespertina, etc., while the two last mentioned species, with Carpodacas cassini, are even found in summer on the highlands about Orizaba, as shown by specimens. transmitted to the Smithsonian Institution by Dr. Sartorius. Similar intrusions of species belonging to the North Mexican fauna take place up the valleys of the Colorado and the Rio Grande, and of those of the eastern province westward along the Missouri and along the Canadian, etc., but they do not affect the general plan. Although characteristic of the eastern province, as already stated, the Cat-bird, (Galeoscoptes Carolinensis), Red-eyed Vireo (V. olivaceus), and Wild-pigeon (Ectopistes migratoria), are found along the northern boundary of the United States to the Cascade Mountains, while specimens of Dendroica coronata have even been taken at Fort Steilacoom on Puget Sound. On the other hand, Turdus nævius has been shot on Long Island and in New Jersey, Helminthophaga celata in the Atlantic states, and Zonotrichia Gambelli, and Spizella pallida,

* See Xantus, Pr. A. N. S., Nov. 1859.

I am informed by Dr. Cabot that a third specimen has recently (Dec. 1864) been shot near Boston and presented to the Natural History Society. As it has been met with as far east as Fort Franklin, it may not improbably reach our eastern coast in company with some of our eastern species bred in the Mackenzie River valley and returning southward to the Atlantic.