William H. Behle

A New Race of the Purple Martin from Utah

The Purple Martin (Progne subis) has long been known to show geographic variation in certain of its characters, and much has been said about the nature and degree of the variation (Brewster 1889:92; Mearns 1902:919; Ridgway 1904:35; Dwight 1905:37; Miller 1906:177; Grinnell 1928:123; van Rossem 1931: 269 and 1945:164; Hellmayr 1935:13; Brandt 1951:669; A.O.U. Check-list 1957:365; Miller, Friedmann, Griscom, and Moore 1957:107; Mayr and Greenway 1960:86; Phillips, Marshall, and Monson 1964:102; Johnston 1966). It seems certain that the species is weakly polytypic. Only two infraspecific differentiates are currently recognized. The race P. s. subis ranges through the eastern and midwestern United States, while P. s. hesperia occupies Lower California, Sonora, and extreme southern Arizona. Size differences are apparent between these two races in both sexes, but color differences are confined to females. Johnston (1966) suggests an adaptive basis for the pattern of geographic variation in the color characteristics, concluding that the color of the feather coats of females

The Pelican Colony at Gunnison Island, Great Salt Lake, in 1943

The colony of White Pelicans (Pelecanus erythrorhynchos) nesting on Gunnison Island in Great Salt Lake is one of the largest now existing. Consequently it seems desirable to place on record any data which may accumulate from time to time pertaining to the numbers of birds there or their natural history. On May 29, 1943, the writer had occasion to make a hurried visit to the island. Forty young pelicans were banded and the following observations were made. A landing was effected on the east side of the island about noon. At that time flocks of adults were streaming in from the east, evidently returning from foraging. The flocks were either in irregular V-shaped arrangement or strung out in a straight line. While our landing was being made the flocks would go directly to their colonial nesting sites, but almost invariably one or two individuals would break away from the group and circle over us as though they were scouts investigating the intruders. Some of these returning groups had as few as four pelicans. One had 80 individuals but usually they numbered between 30 and 40 birds per flock. Once three groups of about 40 birds each merged as they approached the island, thus making one large flock of about 120 birds. A few flocks were eastward bound but they were not nearly as numerous as returning flocks. From the direction of approach of most of the birds, it seemed that they were coming from the Bear River Migratory Bird Refuge which is northeast of the island about 35 miles. This area probably serves as the principal foraging ground for the pelicans of Gunnison Island. One large flock was seen coming in from the south, possibly from Farmington Bay or even so far as Utah Lake. In going to and from the Bear River marshes, the pelicans must fly over the Promontory Range of mountains which runs north and south. The crest of the range is from 2000 to 3000 feet above the lake level. The following day we were on the east side of the Promontory Range about noon and noticed that flocks of pelicans coming from the feeding grounds approached the mountains, then started to circle, all the while gaining altitude until finally they were high enough to fly over the mountain mass. It is not known whether they circled down on the other side or took off directly toward the island. When the flocks approached the island, all seemed to be about the same height above the water. After 3:00 p.m. the number of returning flocks seemed to lessen, although a few continued to come in all afternoon. Since members of each flock went to the same nest-

The Song Sparrows of the Virgin River Valley, Utah

Study of a series of Song Sparrows (Melospiza melodia), recently collected in Washington County, Utah, has brought to light one more case of racial differentiation in this exceedingly variable species. Correlated with this differentiation is a significant ecological separation of populations of Song Sparrows in southwestern Utah. The specimens of the series are easily separable into two distinct groups. Those taken from stream-side thickets from 4200 to 6800 feet in the Pine Valley Mountains are dark, heavily streaked and apparently typical of M. m. fallax, whereas others collected at swamps 25 miles away at 2800 feet along the Virgin River near Saint George are of a conspicuously lighter brown color and also lack the black streaks on the feathers of the dorsum which are so characteristic of fallax. These features suggest an approach to the race saltonis, yet the Virgin River Valley birds are darker than those of saltonis. While the Song Sparrows of the Virgin River Valley are thus intermediate in certain respects between fallax and saltonis, several circumstances argue against the view that they constitute an intergrading population. Although the Virgin River series exhibits the wide range of individual variation characteristic of Song Sparrows as a whole, there is, nevertheless, closer agreement among individuals of the group in their color characters than there is in comparable series of many other races. In addition, they exhibit a small range of variation in wing length. These features suggest a racial stock of relatively uniform genetic constitution, which view is strengthened when one considers the wide range of variation in saltonis-fallax intergrades, where the ranges of these two races adjoin. For instance, five breeding and two fall-taken adults from Cochise and Santa Cruz counties, Arizona, which we have examined (also reported by Swarth, Proc. Calif. Acad. Sci., ser. 4, 18, 1929:329) vary in wing length from below average for saltonis to a value extremely high for fallax. One breeding female in contrast to the others has conspicuous black dorsal streaks. The dorsal ground color of one specimen is grayish. Another has a back typical of saltonis. The rest have an intensification of the bright reddish-brown of saltonis, rather than the subdued chestnut color of the Virgin River birds. The lack of resemblance of the Virgin River series to these highly variable intergades might be taken as evidence that the Virgin River Valley birds are not themselves intergrades. The uniformity of the Virgin River population may be explained in part by its being more or less isolated from surrounding races. Between it and saltonis to the south lies a great desert country with infrequent oases. Even though the Virgin River is a tributary of the Colorado River, there is, for the most part, an absence of suitable Song Sparrow habitat, especially in regions below Boulder Dam where the Colorado River cuts through narrow gorges and has little or no stream-side vegetation along its course. The Virgin River population is in closer geographic proximity to fallax but is separated from that form to some extent by prevailing desert conditions and by lack of suitable habitat. There is seemingly an ecological separation, in addition, between the two races.

Status of the Cormorants of Great Salt Lake

often be impossible to go further, but in many winter or summer falcon ranges, accurate identification of pellet material, at least to family or genus may be made, and not rarely to species or even subspecies. The bulk of the feathers in a pellet are usually from head, neck and breast. These feathers may be slightly altered in color, but apparently not in pattern, if any exists. Frequently the beak or claws or both are present and are useful in identification. A knowledge of local distribution and habitat will help narrow the field, as in mammals, and it should be remembered that coverhaunting birds are not ordinarily available to any falcon. Finally, at a given site there may be a dozen or so pellets of exactly the same color, size, and consistency. Two or three of these may be positively identifiable through beak, claws, primaries, etc. Since falcons so habitually pursue one species of prey at a time, there is an extremely strong likelihood that the beakless and clawless pellets of the lot represent the same species as the others, especially if from state of dryness and weathering they seem to form a compact age series. Similar deductions may often be made from the scales of reptiles, especially since the skin of one or more feet of a lizard may often be floated out of a pellet almost entire. Insects may be counted, and at least partly identified, by the very resistant mandibles, even if the rest of the head is broken beyond use. Other parts of insects in recognizable condition are of irregular occurrence in falcon pellets.

Mohave Desert Avifauna in the Virgin River Valley of Utah, Nevada, and Arizona

The Mohave Desert biome, which occurs mostly in California, extends northward along the Virgin River Valley into extreme northwestern Arizona, southeastern Nevada and southwestern Utah. The approximate northern limits in Nevada have been mapped by Gullion et al. (1959:279). Included are the Meadow Valley Wash, Muddy River and Pahranagat Valley. In northwestern Arizona its area of occurrence is the localized lower Beaver Dam Wash area near Littlefield and the adjacent Virgin River Valley. In Utah this warm southern desert occurs as a relatively narrow tongue along the floor and immediate slopes of the Virgin River Valley up to the mouth of Zion Canyon near Springdale and Coal Pits Wash, as well as the lower stretches of tributary streams such as La Verkin, Ash and Santa Clara creeks and the Beaver Dam Wash. The Virgin River Valley is characterized by long, hot summers, a feature shared with the more southern deserts of Arizona and California, but correlated with the higher elevation (2,880 feet at St. George), the winters are colder. The most conspicuous indicator plants found in this northern portion of the Mohave Desert are cholla cactus, Joshua tree, mesquite, desert willow and creosote bush. The desert is discontinuous