To convey an adequate idea of the bird life of any given area, enumerations of species are not alone sufficient; the numbers of individuals of each species must also in some way be indicated. The usual terms "abundant," "common," "rare," and the like, are unsatisfactory in that their meaning varies both with the person employing them and with the kinds of birds considered. In the latter regard, the Western Chipping Sparrow and the Western Red-tailed Hawk might both be put down as "common," whereas the sparrow may have been observed in actual numbers ten times those of the hawk.

Counts of individual birds are fairly practicable when made in the breeding season on the basis of some unit of area such as an acre. At that season each adult pair is settled within a particular circumscribed locality, and the male is in song. But as soon as the young are out, and from then on throughout the year until the beginning of the next nesting season, most species of birds are moving about incessantly. Counts of individuals are then very difficult to make and furthermore are likely to be misleading because of their great variation in any small area from hour to hour and from day to day. And so, in our field work in the Yosemite region, we put into effect the following different method.

Instead of using a unit of area, we used a unit of time. Birds were listed, as to species and individuals, per hour of observation. In a general way this record involved area, too. Our censuses were practically all made on foot, and the distance to the right or left at which the observer could see or hear birds did not differ, materially, in different regions. The rate of the observer's travel did, of course, vary some; for example, when climbing a steep trail, or going through chaparral, progress was slower than when hiking straightaway along open ridges. Also, in some places, the greater density of the vegetational cover acted to limit the range of sight. But for each of these adverse features of the method there were certain compensations.

For recording a census, a piece of cardboard and a pencil were carried, the names of the various species of birds jotted down, and their numbers checked, as they came to notice. The presence of no species was assumed; but probabilities were given consideration in making identifications. In cases where birds were seen or heard, but their identity was not established with certainty, provisional names were entered, each followed by a question mark. Occasionally the bird could be identified only as to its general grouping, as "hawk." Species of very close resemblance were sometimes grouped together in a joint entry; for example, the crowned sparrows (Zonotrichia). The plus sign indicated that more were present than the actual number entered: the birds could not be counted with certainty; flocking birds, for instance, frequently could not be counted accurately.

At the close of the day or of the period of observation, we were accustomed to transfer our censuses from the field sheet (more or less scribbled, in lead pencil) to our permanent notebooks. If but few species of birds had been seen, these were entered seriatim with numbers of each observed, and comments; if a goodly census had been secured we entered the results in more formal, tabular style, on special sheets printed for this purpose (fig. 3). In either case, record was kept of exact time involved, approximate distance covered, nature of territory traversed, and weather conditions.

Fig. 3. A sample census sheet.

Totals were computed, both of species and individuals. Comparisons of these totals for different parts of the Yosemite region and for different seasons have brought forth some interesting conclusions. Outstanding among these generalizations are the following: The greatest bird population, both summer and winter, is found in the Upper Sonoran Zone. Next come the Lower Sonoran and Canadian zones. The Transition Zone has a fairly large population in summer, but its population drops far down in winter. The Hudsonian has the sparsest summer population, except, of course, for the Alpine-Arctic. The winter population below the snow line consists more largely of seed and berry eaters than of insect feeders; the summer population everywhere contains a predominating proportion of insect-eating birds.

We present below a series of censuses, selected from the more than 250 in our notebooks. The censuses given are chosen to illustrate, first, the nature of the avifauna in various representative parts of the Yosemite section, and, second, the marked changes in bird life taking place in Yosemite Valley from season to season through the year.

The series of censuses given for Yosemite Valley is more complete than for any other station in the section. It begins at the height of the nesting season with two censuses on separate days in two different parts of the Valley, embracing widely different sorts of habitats (associations) and consequently unlike assemblages of birds. The decline of song and general activity at the end of the nesting season is indicated in the census of July 30. That for October 25 shows replacement of the summer visitants by winter invaders. The censuses of December 10 and February 29 show how completely the Valley is deserted by birds with the advent of the midwinter snows; there are scarcely one-fourth as many birds present there in midwinter as in early summer. Return of summer species is already much in evidence in the list made on April 29.

In the census at Mono Lake Post Office on May 31 a 'wave' of migration is indicated in the numbers of species and individuals of warblers seen, which are in excess of what would be present there a month later, in the height of the nesting season. The census on the Big Oak Flat Road in December exhibits the congregation, in a favorable situation, of berry-eating species such as the Townsend Solitaire and Western Bluebird. Had it not been for the berry-laden mistletoe in the golden oaks on the talus slope (pl. 16a) this census in all likelihood would have been no larger than the one taken at the same season on the floor of the Valley.

The census-taker is struck by the variation in his records from hour to hour during the day, irrespective of kind of territory covered and of his own degree of alertness. This fluctuation is due in large part to the fact that there are two daily periods of marked activity on the part of birds, namely, in the early morning, within an hour or so after sunrise, and in the late afternoon, about two hours before sunset. Of these two periods, that in the morning is the most impressive; in other words the observer, by selecting the earlier hours for his census-walk, will make the highest score and also the most representative one. It is quickly apparent that in comparing the enumerations for different days and for different localities allowance should be made for this daily double fluctuation in the visibility and audibility of birds.

It is the earnest recommendation of the authors that observers in a position to do so will get into the habit of taking bird censuses. The method here advocated is a practicable one; we believe it can be adopted to advantage by anyone possessed of a fair acquaintance with bird species. A 'collection' of census records will afford basis for much future satisfaction. On the one hand, is the pleasure of recalling to mind pleasant days afield spent among the most attractive things in nature; and on the other hand is the intellectual enjoyment derived from comparing bird populations in kind and size from place to place and season to season, and from endeavoring to account for the fluctuations which are shown, on the basis of all the factors known to control the birds' existence.

SNELLING, altitude 250 feet, Lower Sonoran Zone, riparian association; January 6, 1915, 9:30 A.M.-12:00 M.; rain the night previous, the morning somewhat cloudy; distance covered about 3 miles, within a mile east of the town.

SNELLING, 250 feet, Lower Sonoran Zone, riparian association; May 26, 1915, 6:00-9:00 A.M.; warm, sunny; distance covered about 3 miles, close to Merced River, within a mile east of the town.

PLEASANT VALLEY westward toward Forty-nine Gap and return, 600 to 1100 feet, Upper Sonoran Zone, blue-oak, chaparral and grassland associations; February 27, 1916, 7:30-9:30 A.M., and 10:30 A.M.-12:00 M. (actual census time 3 hours 30 minutes); cloudy, with rain 9:30-11:30; distance covered about 6 miles, all on foot, chiefly along roadways.

PLEASANT VALLEY westward to high hill near Forty-nine Gap and return, 600 to 1700 feet, Upper Sonoran Zone, blue-oak, chaparral and grassland associations; May 24, 1915, 7:30 A.M.-12:30 P.M.; the day cloudy, considerable rain the night previous; distance covered about 8 miles.

SMITH CREEK (Dudley), 3000 feet, Transition Zone, riparian, grassland and forest associations; July 21, 1920, 8:10-11:10 A.M.; the day clear, hot; distance travelled about 6 miles on foot, from Dudley northeast over ridge, thence down a cañon to Smith Creek and back up to Dudley.

EL PORTAL and vicinity, 2000 feet, Upper Sonoran Zone with few Transition Zone elements, riparian, chaparral, and blue-oak and golden-oak associations; April 27, 1916, 7:00-11:40 A.M. (actual census time 4 hours); clear, hot day, little or no wind; distance covered about 5 miles, all on foot, within 2 miles of the settlement.

YOSEMITE VALLEY, altitude 4000 feet, Transition Zone, chaparral, meadow, forest and riparian associations, May 31, 1915, 2:00-6:00 P.M.; from Sentinel Hotel to LeConte Lodge, then across Stoneman Bridge and along Sequoia Lane, with many zig-zags to likely looking brush clumps or trees, or to run down doubtful songs.

YOSEMITE VALLEY, 4000-4500 feet, Transition Zone, forest, golden-oak and boulder-talus associations; June 3, 1915, 7:00-11:00 A.M.; clear day, windy; distance traveled about 8 miles, all on foot, from old Presidio down nearly to base of El Capitan and return, on Valley floor and up talus to base of cliff.

YOSEMITE VALLEY, 4000 feet, Transition Zone, forest, golden-oak, talus and chaparral associations; July 30, 1915, 7:30-10:00 A.M.; along north side of Valley from old Presidio to vicinity of Rocky Point and return; distance traveled about 4 miles.

YOSEMITE VALLEY, 4000 feet, Transition Zone, forest, chaparral and riparian associa tions; October 25, 1915, 2:35-5:35 P.M.; afternoon to late duck of evening; weather clear; distance traveled about 6 miles, from old Presidio along north road to Kenneyville, to Camp Curry, to village, to Presidio, then to Ahwahnee footbridge and return.

YOSEMITE VALLEY, 4000 feet, Transition Zone, chiefly in forest association; December 10, 1914, 7:50 A.M.-12:10 P.M.; eight inches of fresh snow on ground, and snow falling off and on during the morning; trees heavily laden with snow; from village via Camp Curry and Clark Bridge to Mirror Lake and up lower zigzags on Tenaya Lake trail, returning by same route.

YOSEMITE VALLEY, 4000 feet, Transition Zone, forest and riparian associations; February 29, 1916, 1:30-3:00 P.M.; about 3 feet of snow on ground; distance covered about 2 miles, west of village in vicinity of Camp Ahwahnee.

YOSEMITE VALLEY, 4000 feet, Transition Zone, forest, meadow, chaparral and riparian associations; April 29, 1916, 7:10-10:55 A.M. (actual census time 3 hours 30 minutes); a bright day with few thin clouds and slight wind; distance covered 8 miles, from village to Yosemite Falls Camp, to El Capitan bridge, and return to village.

CRANE FLAT TO MERCED GROVE BIG TREES, 6000-5500 feet, Canadian and Transition zones, coniferous and oak forest, meadow and riparian associations; June 15, 1915, 7:30-11:00 A.M.; the day clear and hot; distance traveled about 4 miles, in more or less direct course between the two stations but with many short side trips to examine 'unknowns' or to study rare species.

BIG OAK FLAT ROAD below Gentrys, 4500-5700 feet, Transition Zone, golden-oak, talus, yellow-pine, and chaparral associations; December 28, 1914, 9:40 A.M.-1:30 P.M; the day clear, crackling cold, trees covered with frost, snow on ground; distance traveled about 4 miles.

CHINQUAPIN and below, along Indian Creek, 6200 to 4500 feet, Canadian and Transition zones, forest association chiefly; June 11, 1915, 7:30-11:30 A.M.; distance covered about 7 miles, going and returning along different routes.

CHINQUAPIN to MONO MEADOW along "Glacier Point road," 6200-7700 feet, Canadian Zone, forest, riparian and chaparral associations; June 18, 1915, 7:20-10:30 A.M.; distance covered about 8 miles.

Above YOSEMITE FALLS, 6600-7300 feet, Canadian Zone, chaparral and forest associations; October 30, 1915, 10:20 A.M.-4:20 P.M.; distance traveled about 5 miles, from top of zigzags, along old Snow Flat trail to west branch of Indian Canon, and return.

TIOGA ROAD between Porcupine Flat and Snow Flat, and return, 8100 to 8700 feet, across Canadian-Hudsonian Zone boundary, forest and meadow associations; June 28, 1915, 6:55 A.M. to 1:45 P.M. (actual census time 5 hours 45 minutes); clear, moderately warm, much snow about, covering road in places; distance covered about 6 miles, by two observers jointly.

MERCED LAKE, 7500 feet, Canadian Zone, forest and riparian associations; August 20, 1915, 8:00-10:30 A.M.; distance traveled about 2 miles, all within one mile of upper end of lake.

VOGELSANG LAKE to Evelyn Lake and return, via Fletcher Creek, 10,350 feet, Hudsonian Zone, rock-slide, riparian and white-bark pine associations; September 4, 1915, 7:30 A.M.-2:30 P.M. (but actual census time 4 hours); heavy frost in morning, ice on quiet pools; distance traveled about 5 miles.

GLEN AULIN, 7700 feet, Hudsonian-Canadian Zone boundary, riparian and forest associations; September 30, 1915, 9:15 A.M.-12:15 P.M.; clear, moderately warm in sunlight, west wind; distance covered about 4 miles, in territory northeast of the Glen.

TUOLUMNE MEADOWS, 8600 feet, Hudsonian Zone, forest, riparian and meadow associations; July 7, 1915, 8:00-10:00, 11:00-11:45 A.M.; clear, sunny, warm; distance traveled about 4 miles, chiefly along road on south side of meadows.

YOUNG LAKE (near Conness Mountain) to Tuolumne Meadows, 10,000-8600 feet, Hudsonian Zone, forest, meadow and rock-slide associations; 8:30 A.M.-12:30 P.M., July 9, 1915; clear, hot, slight westerly wind; distance covered about 7 miles.

WARREN FORK OF LEEVINING CREEK, 9200-11,000 feet, upper part of Hudsonian Zone, open forest and cliff associations, chiefly; September 26, 1915, 7:10 A.M.-2:30 P.M.; after slight snowstorm on September 25, partially clear day; distance covered about 10 miles, from camp in canon up onto Tioga Crest, thence around head of cañon to southwest slope of Warren Mountain, and return.

FARRINGTON RANCH to Walker Lake and return, 6600-8000 feet, Transition and Canadian zones, meadow, riparian and sagebrush associations; May 9, 1916, 2:00-7:00 P.M. strong west wind; distance traveled about 12 miles, returning through sagebrush on ridge adjacent to Williams Butte.

FARRINGTON RANCH (near Mono Lake) to MONO CRATERS and June Lake, and return; Transition and Canadian zones, sagebrush, Jeffrey-pine and meadow associations; September 17, 1915, 7:50 A.M.-6:50 P.M.; distance covered about 25 miles, on horseback.

MONO LAKE POST OFFICE and vicinity, Transition Zone, willow-cottonwood, meadow and sagebrush associations; May 31, 1916, 7:00 A.M.-12:00 P.M.; weather bright and warm; distance traveled about 2-1/2 miles, chiefly along lake shore.