rounded, and with numerous long and soft bristly hairs. This is, of course, very different from the long, extensile, acutely pointed tongue of other Woodpeckers, with its tip armed with a few strong, sharp, short, recurved barbs.

Dr. Hoy and Dr. Coues maintain that the food of these Woodpeckers consists mainly of the cambium or soft inner bark of trees, which is cut out in patches sometimes of several inches in extent, and usually producing square holes in the bark, not rounded ones. As may be supposed, such proceedings are very injurious to the trees, and justly call down the vengeance of their proprietors. This diet is varied with insects and fruits, when they can be had, but it is believed that cambium is their principal sustenance.

This strongly marked genus appears to be composed of two sections and three well-defined species; the first being characterized by having the back variegated with whitish, and the jugulum with a sharply defined crescentic patch of black, though the latter is sometimes concealed by red, when the whole head and neck are of the latter color, and the sharply defined striped pattern of the cephalic regions, seen in the normal plumage, obliterated. Comparing the extreme conditions of plumage to be seen in this type, as in the females of varius and of ruber, the differences appear wide indeed, and few would entertain for a moment a suspicion of their specific identity; yet upon carefully examining a sufficiently large series of specimens, we find these extremes to be connected by an unbroken transition, and are thus led to view these different conditions as manifestations of a peculiar law principally affecting a certain color, which leads us irresistibly to the conclusion that the group which at first seemed to compose a section of the genus is in reality only an association of forms of specific identity. Beginning with the birds of the Atlantic region (S. varius), we find in this series the minimum amount of red; indeed, many adult females occur which lack this color entirely, having not only the whole throat white, but the entire pileum glossy-black; usually, however, the latter is crimson. In adult males from this region the front and crown are always crimson, sharply defined, and bordered laterally and posteriorly with glossy-black; and below the black occipital band is another of dirty white; the crimson of the throat is wholly confined between the continuous broad, black malar stripes, and there is no tinge of red on the auriculars; there is a broad, sharply defined stripe of white beginning with the nasal tufts, passing beneath the black loral and auricular stripe, and continuing downward into the yellowish of the abdomen, giving the large, glossy-black pectoral area a sharply defined outline; the dirty whitish nuchal band is continued forward beneath the black occipital crescent to above the middle of the eye. The pattern just described will be found in ninety-nine out of a hundred specimens from the Eastern Province of North America (also the West Indies and whole of Mexico); but a single adult male, from Carlisle, Penn. (No. 12,071, W. M. Baird), has the whitish nuchal band distinctly tinged with red, though differing in

this respect only, while an adult female, from Washington, D. C. (No. 12,260, C. Drexler), has the lower part of the throat much mixed with red.

Taking next the specimens from the Rocky Mountains and Middle Province of the United States (S. nuchalis), we find that all the specimens possess both these additional amounts of the red, there being always a red, instead of dirty-white, nuchal crescent, while in the female the lower part of the throat is always more or less red; in addition, the male has the red of the throat reaching laterally to the white stripe, thus interrupting the black malar one, which is always unbroken in the eastern form; and in addition, the auriculars are frequently mixed with red. Proceeding towards the Columbia River, we find the red increasing, or escaping the limits to which it is confined in the normal pattern, staining the white and black areas in different places, and tingeing the whitish which borders the black pectoral area.

Lastly, in the series from the Pacific coast (S. ruber), we find the whole normal pattern rendered scarcely definable—sometimes entirely obliterated—by the extension of the red, which covers continuously the whole head, neck, and breast; but nearly always the normal pattern may be traced, the feathers of the normally black areas being dusky beneath the surface, and those of the usual white stripes very white for the concealed portion. Usually, in this form, the red of the breast covers only the black pectoral area; but in extreme specimens it reaches back to the middle of the body beneath, and stains the white spots of the back.

With the increase of the red as we proceed westward, there is also a decrease in the amount of white above; thus, in varius the whole back is irregularly spotted with dirty white and black,—the former predominating, the latter most conspicuous as a medial, broken broad stripe,—and the lateral tail-feathers are much variegated by white spots. In nuchalis the back is mostly unbroken glossy-black, with two parallel narrow stripes of white converging at their lower ends; and the lateral tail-feather is almost wholly black, having merely a narrow white border toward the end. S. ruber is most like nuchalis, but has the white still more restricted.

In varius the bill is dark brown, in nuchalis it is deep black, and in ruber wax-brown. In varius the yellow of the lower parts is deepest, in nuchalis just appreciable.

Species and Varieties.

A. Wing with a white patch on the middle and greater coverts. Markings along the sides with a longitudinal tendency.