Bent Life History of the Dark-eyed Junco

Published by the Smithsonian Institution between the 1920s and the 1950s, the Bent life history series of monographs provide an often colorful description of the birds of North America. Arthur Cleveland Bent was the lead author for the series. The Bent series is a great resource and often includes quotes from early American Ornithologists, including Audubon, Townsend, Wilson, Sutton and many others.

Bent Life History for the Dark-eyed Junco - the common name and sub-species reflect the nomenclature in use at the time the description was written.


Contributed by STEPHEN W. EATON


The northern slate-colored junco, or "common snowbird" as persons who know it only in winter often call it,is one of the most distinctive of our common sparrows. With its uniform pale gray upperparts sharply defined against its white belly, aptly described as "leaden skies above, snow below," it is not likely to be confused with anything but other closely related juncos, and then only in the western parts of its wintering range. A friendly little bird that breeds across the continent from Alaska to Labrador and Newfoundland and from the limit of trees southward into the northern United States, it is the summer companion of the canoeist in the Canadian forests and of the mountain hiker of Appalachia. In winter it retreats southward throughout most of the United States in small, congenial flocks of 15 to 25 individuals. These sometimes forage over the snow-covered fields with the tree sparrows searching for the seeds of weeds that escaped the cultivator, and they commonly frequent the yards of homes where food has been put out for them, which they much prefer to scratch from the ground than to pick from an elevated feeder.

Essentially an inhabitant of the more open northern woodlands and forest edges, it is generally common throughout its breeding range in the Hudsonian and Canadian life zones, except in the deeper woods, but tends to dwindle in numbers toward the north. Typical is E. A. Preble's (1908) comment: "This common species, sometimes called 'tomtit' in the North, is the sole representative of its genus throughout most of the wooded parts of the Athabaska-Mackenzie country. Over this vast region it is a common summer resident, being one of the earliest of the smaller migrants to arrive in spring and a rather late lingerer in autumn."

Francis Harper (1953) notes that "Apparently the numbers of this species diminish rather decidedly toward the tree limit in most parts of northwestern Canada although Porsild (1943: 43) reports it well beyond the tree limit at the Mackenzie Delta." Lawrence Walkinshaw writes Mr. Bent of finding the males singing from the tree tops 20 to 25 feet above the ground in the spruce bog areas along the Kuskukwim River in Alaska, and adds: "Where the tree line disappeared, so did the juncos."

Spring: The migrating juncos rush across most of the eastern and midwestern United States about mid-April passing, as they go, their southern relatives already singing on their territories. In Illinois NI. C. Shank (1959) reports they build up fat reserves before migrating, but D. W. Johnston (1962) finds the wintering populations leave Wake Forest, N.C., before they deposit any fat. The birds are restless and hyperphagic, and move northward rapidly in flocks of up to 100 individuals. In the East they are often accompanied in the earlier part of the migration by fox and tree sparrows; later along the Saskatchewan River they may be accompanied by tree and clay-colored sparrows (Houston and Street, 1959).

Territory: The males usually arrive on the breeding grounds well in advance of the start of nesting. During 10 years of observation near Olean in southwestern New York State (Eaton, 1965) I heard the average first territorial singing on March 12, but most males here do not start their territorial song in earnest until about March 21. Some 300 miles farther north Mrs. L. de K. Lawrence writes (in litt.) the juncos arrive at her home in Rutherglen, Ontario in late March or early April, with a mean arrival date of April 2 for 13 years.

The male proclaims his territory by singing from the top of the tallest trees within it, which may be 50 to 75 feet above the ground. Nero (1963) writes from the Lake Athabasca, Saskatchewan region: "On May 18 I found two males apparently engaged in a territorial dispute. The aggressor approached with its breast feathers raised and spread, forming a broad front, and with its tail widely spread and alternately depressed and elevated. Its pinkish bill was very conspicuous against the dark feathers of the head."

Individual territories appear to vary greatly in size, probably because of the scarcity of choice nest sites. The area a male defends vigorously has never been determined experimentally with models and recorded songs, but casual observations of the location of song perches near Olean suggest it is about 2 or 3 acres. Where ideal nest sites are more plentiful, the territories are probably smaller. Each usually seems to include some sort of opening in the forest canopy surrounding a rock outcrop or an exposed soil bank. The species' tendency to build in or near some sort of vertical wall probably helps to explain many unusually placed nests.

Courtship: The male may continue to sing for some days before a female enters his territory. Mrs. Lawrence (1956) thus describes the early courtship between one of her banded male juncos and a female who appeared 11 days after he arrived in 1953:

Her behavior indicated plainly that her sexual drive had not yet reached high intensity. She faced him as he pursued her, showing him her breast, or hopped aside or away to evade his approach, thus displaying her urge to escape to the point of aggression.

The male pursued her doggedly with wings drooped and tail lifted. Every time when the female withstood him, he stooped and with great intensity pecked at the ground and at his aluminum band on the right tarsus.

Obviously, this pecking at the ground and at the aluminum band, both irrelevant actions in the present situation, were displacement activities, a "substitute behavior" * * * as his sexual drive was denied by the female's condition of unreceptiveness.

Generally the first one or two days seem to be spent in establishing and strengthening the pair bond. The male follows his mate about as she feeds within the territory and the two birds remain close together, seldom more than 50 feet apart. Both birds, and particularly the male, display by hopping about the other on the ground with the wings drooping and the tail fanned laterally so that the white outer rectrices are conspicuous. The male now sings much less frequently, but he still leaves his mate occasionally to proclaim his occupancy of the property by song from one of his favorite perches.

Nesting: The junco's ground nest is built by the female, but the male often helps by bringing material for it. Cordelia J. Stanwood, who studied this species extensively at her home in Ellsworth, Maine, wrote Mr. Bent about the activities of a pair building their nest one wet May "under a mass of brush and leaves and sheltered by a small spruce. Both birds brought some of the damp materials and they appeared to care little how wet they were, but the female seemed to do the greater amount of the molding." She continues:

"The nest site varies according to its situation. I have seen the juncos brooding amongst the roots of a growing clump of gray birches, partially under stumps and rocks, below a tuft of leaves, in a brush heap shaded by small evergreens, beneath bracken, and many within the side of a bank or knoll. The wall of a knoll covered with birdwheat moss [Polytriehum] or the side of a steep bank just under the overhanging sod seems to be the most typical site for a j unco nest. A depression is made or enlarged in the side of the bank or knoll, and the moss or overhanging sod form a natural roof. On a pasture hillside the abode of the junco may be a little cup-shaped structure of straw in the midst of a blueberry patch; in a damp wood it will be a deeper structure with thick walls of moss, twigs, and hay with a substantial lining of fine hay or hair. The brooding female often draws her tail into the nest as the ovenbird does, so that it is well nigh impossible to distinguish the bird or the cradle when looking directly into the nesting cavity."

In wooded country the junco typically nests at the edges of openings in the forest canopy, such as those made by a stream, a logging road, or a clearing. Preble (1908) describes an Athabasca nest that "was built on the steep side of the river bank, and was quite bulky, the outer portion being constructed of fine twigs, strips of bark, and feathers. Thisfoundationinclosedacup-shapednestofdrygrass, thickly lined with gray dog's hair." E. W. Jameson, Jr. (in litt. to Mr. Bent) describes a nest he found on the Gasp6 Peninsula in 1940 "on an eastfacing slope of birches and alders. The ground was covered with grass, dead leaves, and bunchberries. The nest itself was in a cavity four inches in underneath a dead stump, the opening protected by a clump of club moss (Lycopodium). Both parents were feeding insects to the four half: grown young.~~ B. P. Bole, Jr. (1941: 1942) describes the nesting of a small colony of juncos on Little Mountain, just east of Cleveland, Ohio, and in a nearby hemlock-studded ravine known as Stebbens Gulch, which is typical for the species in western Pennsylvania and southwestern New York where similar Paleozoic rocks outcrop:

Every one of the junco nests found on Little Mountain was in exactly the same type of place. On this sandstone mesa the brows of the ledges and rocky outlying chunks of puddingstone have curling forelocks of Polypody fern, and it is under the overhanging fronds of these that the Juncos place their nests. As the ferns are on the very edges of the cliffs, it is frequently a matter of some danger to get into positions from which the nests can easily be seen or discovered.

The nests themselves are made of rootlets of various ferns, that of the Polypody being especially favored. There is a thin lining of dry sedges and grass. The whole stracture is very compact, and is placed well down in the roots and hanging dead fronds of Polypody. When danger threatens, the female bird tumbles out and downwards into the crevasse facing her; in this she flies for twenty feet or more before rising into the low yellow birches and hemlocks lining the ledges.

The junco often builds in rather unusual situations. Forbush (1929) cites a junco nest on a ledge beneath the gable of a house in Nova Scotia. Wendell P. Smith (1936) writes of a nest of dried grasses and fern stalks and other vegetation 8 feet above the ground in a trellis overgrown with woodbine (P.sed,era vitacea). Houston and Street (1959) describe a nest in Saskatchewan built in a half-pound tobacco can lying on its side and which contained three junco eggs and three cowbird eggs. Basil J. Wilkinson showed me a nest near Olean, New York, from which young were successfully fledged, in a wind-vane bird-feeder mounted on an 8-foot iron pipe. The base of the triangular feeder was open, and the two sides were glass. The nest was jammed into the apex angle, just as it might have been into a niche in a rock ledge.

Throughout the eastern parts of its range the species is apparently double-brooded. In southwestern New York, I (1965) found two laying peaks, the first at the end of April, the second the first of July. In Maine where Palmer (1949) notes first sets from the first week in May to the first week in June, be states: "A second brood is raised, the eggs being laid from late June to late July." Peters and Burleigh (1951a) found flying young near St. Johns June 9 and add: "Perhaps two broods are raised in Newfoundland for Arnold found a nest with three eggs in the Humber River valley on July 18, 1911."

Eggs: The northern slate-colored junco usually lays from three to five and rarely six slightly glossy eggs. They are generally ovate, although some may tend to be either elongated or short ovate. The ground is grayish or very pale bluish-white with speckles, spots, and occasional blotches of reddish-browns such as "Verona brown,~~ ~'russet," "chestnut," or "Brussels brown," with undermarkings of "pale mouse gray." in most cases the markings are concentrated toward the large end where they frequently form a wreath. There is considerable variation, some being only very faintly speckled, others quite heavily spotted and with a few blotches, but in all considerable groundcolor shows. Often the spottings are quite dull, and the gray speckles may sometimes predominate. One set of eggs in the MCZ is all white and unspotted. The measurements of 50 eggs of the nominate race average 19.4 by 14.4 millimeters; the eggs showing the four extremes measure 21.1 by 14.2, 20.9 by 16.2, 17.8 by 14.2, and 19.3 by 13.2 millimeters.

Young: Incubation is apparently by the female alone and usually lasts 12 to 13 days. V. A. Greulach (1934) reports a 12-day incubation period for a nest in Allegany State Park in southwestern New York. In two nests I recently (1965) timed in the same region the elapsed times from the lest egg laid to the last to hatch were 12 and 13 days, respectively.

Both parents feed the young and attend to nest sanitation. During the first few days they eat the nestlings' fecal sacs, but on the fourth or fifth day start to carry them away instead, usually flying to a perch not far distant and wiping the sac off on a limb. At one of my nests the male always flew to a nearby telephone wire to wipe the sac from his bill; the wire was soon speckled white for a considerable distance before a shower cleaned it up.

V. A. Greulach (1934) comments: "The male removed 27 fecal sacs to the female's 14 during the periods of observation. In all cases where the disposition of the sacs was noted they were wiped off on tree branches. The brooding was apparently all done by the female, and she was not observed brooding after the young were seven days old."

Mrs. Standwood wrote Mr. Bent as follows about a nest she watched from a blind at Ellsworth, Maine:

"In the early stages of nursery life the parent birds fed the nestlings 'regurgitated' or partly digested food, together with a few tender moths and caterpillars. Later I saw them feed yellow grubs, millers, many spruce bud-moths, caterpillars, and crane flies. During one period of many hours of watching, the parents fed the young nothing but great numbers of smooth, green caterpillars.

"The youngsters begin to open their eyes at the end of the second day and, as in other sparrows, their feathers begin to show about the seventh day. At this time the active youngsters begin to show fear by snuggling down in the nest when a person approaches it. I have seen young birds still in the nest on the 11th and 12th days, but know they could leave earlier if danger threatened them."

Greulach's (1934) young left the nest when 12 days old. In two nests near houses the young I (1965) followed left the nest in 9 days, and I know that a number of these were raised to independence. After leaving the nest the young remain at least partially dependent on their parents for about 3 weeks.

One brood I banded Aug. 3, 1959, just before they left the nest, I was able to follow for an extended period. I saw the father, a crippled bird readily identified, feeding them on August 24 and 27. On August 30, however, one of the young perched on the feeder next to its father and crouched in the begging posture with vibrating wings, but without giving the usual begging call. The old bird stretched upward into the aggressive posture a few times, and when the youngster continued to beg, the father flew at it and chased it a short distance without feeding it. The banded young and their father were still nsiting the feeder daily on September 19, about 46 days after leaving the nest. At this time the old bird had almost completed his postnuptial molt; the young still had a few juvenal feathers in the head and their undertail coverts had not quite completed their full growth.

Plumages: Mrs. Stanwood noted in a letter to Mr. Bent that "When they first peck their way from the shell, young juncos are a reddish, burnt-orange color, and well covered with burnt-umber down." Dwight (1900) on the other hand calls the natal down "slate-gray." He notes the juvenal plumage is acquired by a complete postnatal molt, and describes it as:

"Above, drab, plumbeous on crown; sides of head and nape streaked with dull black, the feathers especially of the back edged with bistre. Wings and tail slaty black edged with olive-gray, the tertiaries and and wing coverts with dull cinnamon, the greater coverts tipped with bull. Two outer rectrices pure white. Feet pinkish bull, dusky when older. Bill dusky pinkish buff, flesh-color when older and in dried specimens becoming dull ochre-yellow."

He describes the first winter plumage as "acquired by a partial postjuvenal moult in August and September, which involves the body plumage and the wing coverts, but not the rest of the wings nor the tail.

"Above, including wing coverts, sides of head, throat, breast and sides slaty gray, darkest on the crown and veiled with bistre edgings, especially on the back, more faintly with paler brown or ashy gray on the throat. Abdomen and crissum pure white, sometimes faintly washed with vinaceous cinnamon."

The first nuptial plumage is "acquired by wear through which the brown and ashy edgings are finally lost, birds becoming ragged but not much faded by the end of the breeding season. A few new feathers are acquired on the chin early in April, but no regular moult is indicated."

The adult winter plumage is "acquired by a complete postnuptial moult beginning the middle of August. Practically indistinguishable from first winter, but the tertiaries usually edged with gray instead of faded cinnamon, the xvings and tail blacker and showing everywhere fewer brown edgings." The adult nuptial plumage is acquired by wear as is the first nuptial, from which it is practically indistinguish.. able.

The sexes are indistinguishable in the natal down and juvenal plumages. In first winter and subsequent plumages the female is similar to the male, but the gray is much paler and the plumage everywhere more veiled with brown.

Wood (1951) throws new light on the amount of white in the junco's three outer tail feathers. The outer pair are always pure white, but the amount of white on the inner two, most notably on the third pair, increases greatly in the first adult postnuptial molt. Feathers lost or plucked during the first xv-inter are replaced by feathers having the design of those of the succeeding molt, with more white.

Food: Martin, Zim, and Nelson (1951) say "Juncos, like many other members of the sparrow family, are primarily ground-feeding seed eaters. They are partial to seeds of common weeds. In summer, insects constitute about half or more of their diet." For the northern slate-colored junco "Caterpillars, beetles, and ants seem to be the choice items of the animal diet, the balance being made up of wasps, bugs, grasshoppers, other insects, and spiders." Heading a long list of mostly weed plants whose seeds the junco is known to eat, they list those most frequently identified in their stomach contents as ragweed, bristlegrass, dropseedgrass, crabgrass, pigweed, and goosefoot.

In southwestern New York I have watched them feeding on the fall cankerworm, Alsophila pometaria, in late autumn. During the xnnter I once saw them eating the seeds of the wild black cherry, and they often eat the seeds of hemlock and yellow birch from the snow surface. Though they usually eat hemlock seeds from the ground, they can and do extract them from the cones on the trees. They feed avidly on the springtails (Collembola) that swarm abundantly about the bases of the trees in February, and they will go out of their way to capture, either on the snow surface or in the air, a small species of gnat that hatches out of the small streams about this time. They also join the early phoebes and bluebirds in preying on the late March or early April hatch of the stonefly, Pteniopteryx nivalis. Francis H. Allen wrote Mr. Bent of a large flock he watched at Cohasset Nov. 2, 1935, whose members "frequently flew into the air to catch flies. The flight was usually, if not always, from trees or bushes and not from the ground. They continued this off and on for nearly an hour."

Voice: Of the song with which the junco proclaims his territory, F. II. Allen wrote Mr. Bent: "The jingling trill of this junco is well known. It is usually a simple trill, but, as with some other birds whose normal song is a single trill, one will occasionally be heard singing two or even three trils on different pitches but joined together to form a single song." In southwestern New York this song is given mainly in February, March, and April before pair formation and egg laying. After incubation begins it is heard much less frequently, though there is a noticeable recrudescence during late June and early July, and an occasional autumnal upsurge of it in October. As Aretas A. Saunders (MIS.) describes it:

"The nonhal song of the northern slate-colored junco is a simple trill, all on one pitch, or a series of rapid notes, sometimes barely slow enough to count. It resembles that of the chipping sparrow, but is rather more musical in quality. When the notes of the song are slow enough to count they vary, in my records, from 7 to 23 notes, averaging about 12. The length of the songs varies from 1.4 to 2.8 seconds, averaging about 1.9. The pitch varies from B"' to G"".

"There is a considerable amount of variation in junco songs from the simple trill that is all on one pitch. Some songs vary a bit up or down in pitch, and some vary in time. I believe this bird shows as great a tendency to vary fron the normal type of singing as does the towhee. In the Adirondacks I heard a bird singing a song of three prolonged whistles. I chased it about for parts of three days and finally identified it as a junco. Possibly this bird got its song from a white-throated sparrow, but if so it did not sound enough like that bird for me to think it was such."

In notes she sent Mr. Bent, Mrs. Lawrence comments on "the lovely tinlding chorus by the j uncos in early spring, as if a myriad of woodland sprites were shaking little bells in an intensive competition," and she syllabizes three variations of the junco song as follows: tiiiiiii, tiiii-tiiii, and tuituituitilihi. She also describes a "conversational subsong" between members of a pair heard before and during the egg-laying period as "a rough zreet, zreet, zreet followed by a lengthy sotto-voce warbling." E. H. Eaton (1914) quotes Bicknell's description of this as "a whispering warble usually much broken but not without sweetness and sometimes continuing intermittently for many minutes," and which Florence Merriam calls "low, sweet, and as unpretentious and cheery as the friendly bird himself."

Mrs. Lawrence also sent Mr. Bent the following variations she detected in the junco's call notes in different situations:

Location: a simple tit-til-tit Alarm 1: an explosive teket, tchet Alarm 2: bzzz, bzzzz Scolding: a smacking tack, tack, tack Fighting: tuit, tuit, interspersed with a twanging note and a variety of smacking and buzzing notes Feeding: A throaty tulut, tulut seems to serve as a call to come together.

Behavior: Juncos usually progress on the ground by hopping in fall and winter, but occasionally run in short spurts when chasing a nval or to capture moving food. During the nesting season they may also hop, but more often one sees them walking with short, mincing steps, moving along not unlike a mouse.

F. H. Allen wrote Mr. Bent: "The juncos scratch for food, though not so often nor as vigorously as the fox sparrows do. They scratch by hopping forward and then back with both feet at once. When a thin layer of snow lies on the ground, a bird will scratch away a roughly circular hole 3 or 4 inches in diameter to get at the grain underneath.

"On the whole they are rather scrappy when feeding together and with other birds. Individuals vary in pugnacity, and sometimes females at a winter feeding station will drive off males. On Mar. 3, 1942 in West Roxbury, Mass., a male junco feeding on our lawn with a few other juncos and a number of English sparrows kept his white outer tail-feathers showing conspicuously for at least 5 minutes. He held the tail motionless without flicking. As the crowd thinned the white on one side was concealed for a time, and then when he was left alone that of the other side disappeared too. It looked as though the white rectrices where used as a t.hreat on this occasions a display I had never before seen except in momentary flashes."

Forbush (1929) describes how juncos he watched near the top of Mt. Washington in early August "drank from 'the Stream of a Thousand Falls,' which is formed by the melting of the snow, and then bathed in the frigid waters with much fluttering and splashing of spray, reminding me of other Juneos which I have watched in midwinter, similarly engaged in bathing, but in light dry snow, just as other sparrows take dust baths in hot weather."

W. S. Sabine (1957) comments on the flight behavior of a flock of juncos on their visits to a feeder in Ithaca, N.Y., on late winter afternoons. She found the birds, at what was probably their last feeding of the day, always departed in a regular pattern. As each bird finished feeding it perched quietly for a minute or so at the station, then joined others assembled in an arbor vitae clump about 40 feet away where they "made small movements" for about 5 minutes. The whole flock then left the arbor vitae together, closely following one another to an adjacent leafless deciduous tree, climbed high in it, and then flew from tree top to tree top along a ridge to the northeast, always in the same direction. She concludes "It seems a reasonable conjecture that the flock had a common goal, and this in turn suggests the hypothesis that a common roost may be a feature of the integration of junco flocks."

Hamilton (1940), also at Ithaca, found juncos roosting at night in winter "on the ground at the base of a Taxus thicket." While nightbanding robins near Olean on April 13 I flushed four juncos from roosts 3 to 8 feet from the ground in thick Norway spruces. On Dec. 28, 1960 I flushed a junco after dark from a nest 2 feet from the ground in a hemlock hedge near my house, and on Jan. 25, 1961 I again flushed a bird from the same nest at night. Thus, old nests occasionally function as winter roosting sites.

Field marks: A slate-colored bird slightly smaller and more slender than a house sparrow, with uniform gray head, back, breast, and sides contrasted sharply against the white belly, this junco is seldom confused with any other species except some of its western relatives, such as the Oregon junco, which has a much darker head contrasting with a browner back. The pale bill is conspicuous in the field, and the white outer tail feathers are especially prominent in flight.

Enemies: Essentially birds of open woodlands and forest edges, the june05 are subject to attack by accipitrine hawks and other predators. Red squirrels, chipmunks, weasels, and martens must take some eggs and young from the nests. Northern shrikes harry the wintering flocks fairly frequently. Cowbird parasitism is apparently not of great moment to the species' reproduction. Friedmann (1963) states:

The slate-colored junco is an infrequently reported host; probably it is molested very slightly by the brown-beaded cowbird. Eighteen instances have come to my attention. Three races have been recorded as victims: cismontanus in British Columbia; caroUneasis in Virginia and West Virginia; hyemalis in Alberta, Saskatchewan, Ontario, Quebec, Nova Scotia, New York, Pennsylvania, and Ohio. * ** Both cismontcnus and hyemelis have been known to rear young cowbirds.

In the Peace River District of British Columbia, Cowan (1939, p. 59) found that no fewer than four out of five junco nests which were observed were parasitized: evidence which suggests that in this region the bird is a commoner host than it has been found to be elsewhere.

The Communicable Disease Center of the Public Health Service at Atlanta, Ga. has reported finding antibodies of the St. Louis strain of encephalitis in the northern slate-colored junco. Allen McIntosh of the Animal Disease and Parasite Research Division at Beltsville, Md. writes (in litt.): "There are 61 references to parasites from this host; the following genera of parasites having been reported: Haemoproteus, Leucocytozoon, Plasmodium, Trypanosoma, Eurytrema, Zonorchis, Diplotraema, Taenia, Filaria, Strongyloides, Syngamus, Amblyomma, Analges, Analgopsis, Bruelia Degeriella, Docophorus, Haemaphysalis, Ixodes, Machaerillaemus, Nirmus, Ornithoica, Ornithomyja, Phiopterus, Physostomum, Ricinus, and Trombicula.

Fall and Winter: About the time the first wintry blasts begin to blow across the great coniferous forests of the North, the juncos start moving southward. E. A. iPreble (1908) last noted them along the Mackenzie River 50 miles below Fort Simpson on October 16. Houston and Street (1959) say the fall migration along the Saskatchewan River usually ends in late October, but some years the birds are common until mid-November. At Pimisi Bay, Ontario, Mrs. Lawrence reports in a letter to Mr. Bent that most of the juncos leave in October, a few late stragglers occasionally remaining into November. E. H. Eaton (1914) writes that in New York State:

"In the fall, migrants begin to appear from the 11th to the 28th of September, in the southernmost parts of the State sometimes not before the 4th to the 12th of October. Among the members of the sparrow family, this species rivals the Song sparrow, Vesper sparrow, Savannah sparrow and Chipping sparrow for the place of greatest abundance during the spring and fall migration, probably being as abundant as the Song sparrow in most localities * * *

In her studies of the wintering flocks of this junco at Ithaca, N.Y., Winifred S. Sabine (1956) found "that although the migrant individuals which are to become winter residents arrive irregularly over a period of several weeks, they somehow manage to form themselves into distinct, stable winter flocks with mutually exclusive foraging territories." She continues:

The junco flock is an association of birds which is firm in the identity of the individuals associated. * * * In a given small area a single group will be seen and no other. The formation of firm associations and the occupation of definite foraging areas take place at once among the earliest arrivals; it becomes obvious as soon as the first migrants are marked. The late comers are integrated into existing groups. The flock thus formed does not fly about as a unit, however. There appears to be no limit to the size of a foraging group. It may include the whole flock or it may consist of a single bird. The entire flocking procedure is marked by the continual forming and dissolving of groups of unpredictable size consisting of individuals that consort together and are daily visitors at the feeding sites.

Range: Western and northern Alaska, Mackenzie, northern Ontario and Labrador south to northern Mexico and the Gulf coast.

Breeding Range: The northern slate-colored junco breeds from western and northern Alaska (Brooks Range, Kobuk River, Yukon Delta), central Yukon (Ogilvie Range), northwestern and central Mackenzie (Mackenzie River Delta, Fort Anderson, Fort Reliance), northern Manitoba (Churchill), northern Ontario (Shagamu River), northern Quebec (Richmond Gulf), Labrador (Tikkoatokuk Bay), and Newfoundland south to south central Alaska (Lake Clark, Seldovia, Prince William Sound), southern Yukon (Lake Marsh), northeastern British Columbia (Muncho Pass), central Alberta (Edmonton district), central Saskatchewan (\icLean), southern Manitoba (Treesbank), central Minnesota (eastern ~vIarshaIl County, Minneapolis), southeastern Wisconson (Jefferson, Burlington), central Michigan (rarely south to Ingham County), southern Ontario (London), northeastern Ohio (Geauga and Trumbull counties), northern and western Pennsylvania (Pocono Mountains; intergrades with J. h. carolinensis in Appalachian Mountains), southeastern New York (Hardenburg, Bald Mountain near Dover), Connecticut (Union; rarely to Jiadlyme), and Massachusetts.

Winter Range: Winters chiefly south of the breeding range and east of the Rocky Mountains, but sparsely to the westward, from southeastern Alaska (Juneau), southern British Columbia (North Vancouver, Okanagan Landing), northwestern Montana (Fortine), southern Saskatchewan (Eastend, McLean), southern Manitoba (Brandon, Winnipeg), northern Minnesota (Bagley, Duluth), western Ontario (rarely north to Port Arthur), northern Michigan (rarely north to Munising and Sault Ste Marie), central Ontario (Algonquin Park, Ottawa), southern Quebec (Montreal; Anticosti Island, rarely), and Newfoundland (Avalon, Tompkins) south to northern Baja California (Cocopah Mountains), northern Sonora (Sonoyta), central Chihuahua (Chihuahua), southern Texas (Presidio, Rockport, Port Arthur), southern Louisiana (New Orleans), southern Mississippi (Cat Island), southern Alabama (Mobile), and northern Florida (Pensacola, New Smyrna).

Casual records: Casual in eastern Siberia, the islands of the Bering Sea (Little Diomede Island, St. Lawrence, Sledge, Nunivak, and Pribilof islands, and South Bight of the Aleutian Islands), the arctic coast of Alaska (Cairn, Wales, Point Barrow), Banks, Southampton and Baffin islands, southern Florida (Chokoloskee), Bermuda, and the Bahamas (New Providence).

Accidental in Ireland and Italy.

Migration: Data deal with the species as a whole. Early dates of spring arrival are: Maryland: Laurel, February 25. New Ilampshire: New Hampton, March 7 (median of 21 years, March 28); Concord, March 10. Maine: Lake Umbagog, March 21. Quebec: Montreal area, March 10. Illinois: Chicago, February 26 (average of 16 years, March 12). Wisconson: northern Wisconson, March 25. Minnesota: Cannon Falls, March 1 (average of 21 years for southern Minnesota, March 11). South Dakota: Sioux Falls, March 24 (average of 4 years, April 4). North Dakota: Cass county, March 15 (average, March 21). Manitoba: Treesbank, March 17 (average of 25 years, March 30). Montana: Libby, March 20. British Columbia: North Vancouver, April 8. Yukon: Sheldon Lake, April 20. Alaska: Forty Mile, April 22.

Late dates of spring departure are: Florida: northwestern Florida, April 19; Tallahassee, April 11. Alabama: Birmingham, April 21; Russellville, April 18. Georgia: Atlanta, April 24. South Carolina: Charleston, May 2 (median of 5 years, April ii). North Carolina: Raleigh, May 4 (average of 13 years, April 12). District of Columbia: May 17 (average of 36 years, April 30). Maryland: Baltimore County, May 30; Montgomery County, May 24; Laurel, May 14 (median of 6 years, May 2). Pennsylvania: State College, May 15; Beaver, May 10. New Jersey: Plainfield, May 13. New York: Westchester County, June 5. Connecticut: Fairfield, May 12. Massachusetts: Martha's Vineyard, June 7; Concord, May 22. New Hampshire: Concord, May 7. Louisiana: Baton Rouge, March 29. Arkansas: Fayetteville, April 25; Little Rock, April 23. Tennessee: Nashville, May 13. Missouri: St. Louis, May 14 (median of 14 years, April 20). Illinois: Chicago, May 24 (average of 16 years, May 10); Urbana, May 23 (median of 20 years, May 4). Indiana: Wayne County, May 10. Ohio: central Ohio, May 30; Oberlin, May 20 (median of 18 years, May 1). Michigan: Detroit area, May 30 (mean of 10 years, May 24). Iowa: Decorah, May16; Sioux City, May 15. Wisconsin: northern Wisconsin, May 5. Minnesota: Minneapolis, May 24 (average of 7 years for Minneapolis, May 14). Texas: Amarillo, April 13. Oklahoma: Cleveland County, April 25. Kansas: northeastern Kansas, May 2 (median of 14 years, April 20). Nebraska: IPlattsmouth, May 1. South Dakota: Sioux Fails, May 8 (average of 5 years, April 22). North Dakota: Cass County, May 27 (average, May 16). New Mexico: Los Alamos, April 17. Arizona: Tucson, May 1. Utah: Ashley Creek Marsh, May 6. Wyoming: Laramie, May 8 (average of 9 years, April 17). Idaho: Moscow, April 22 (median of 11 years, April 1). Montana: Fair'dew, April 20; Gallatin County, April 2. California: Berkeley, April 11. Oregon: Wasco County, April 12. Washington: Lake Crescent, April 14.

Early dates of fall arrival are: Washington: King County, September 30. Oregon: Lake County, September 16. California: Yermo, September 22. Idaho: Potlatch, October 2 (median of 11 years, October 21). Wyoming: Laramie Mountains, September 14 (average of 5 years, October 7). Utah: October 6. Arizona: Prescott, September 8; Wikieup, October 7. New Mexico: northern New Mexico, October 26. North Dakota: Cass County, September 5 (average, September 13). South Dakota: Sioux Falls, September 24. Nebraska: Santee, September 20. Kansas: northeastern Kansas, September 23 (median of 19 years, October 10). Olilahoma: Oklahoma City, September 1. Texas: Amarillo, October 9. Minnesota: Minneapolis, August 18 (average of 14 years, September 19). Wisconsin: northern Wisconsin, September 20. lowa: Decorab, September 16; Sioux City, September 21. Michigan: Detroit area, September 13 (mean of 10 years, September 18). Ohio: central Ohio, September 15 (average, September 26); Buckeye Lake, September 23 (median, October 2). Indiana: Wayne County, September 15 (median of 17 years, October 1). Illinois: Chicago, September 5 (average of 16 years, September 16). Missouri: St. Louis, September 15 (median of 14 years, October 2). Tennessee: Knox County, October 17. Arkansas: Fayetteville, October 12; Little Rock, October 15. Mississippi: Oxford, October 23. Louisiana: New Orleans, October 5; Baton Rouge, October 12. New Hampshire: Concord, September 23. Massachusetts: Belmont, September 4; Martha's Vineyard, September 19 (median of 5 years, September 30). Connecticut: West Hartford, August 18; Portland, September 15. New York: Long Island, August 15; Oneonta, September 5. New Jersey: Englewood, September 17. Pennsylvania: Hawk Mountain, September 3; Pittsburgh, September 11. Maryland: Talbot County and Laurel, September 5 (median of 7 years, September 29). District of Columbia-September 14 (average of 31 years, October 7). North Carolina: Durham, October 18; Raleigh, October 23 (average of 15 years, October 31). South Carolina: Clemson, October 3. Georgia: Atlanta, October 17. Alabama: Gadsden, October 7; Coffee County, October 20. Florida: Melbourne, October 6; Chokoloskee, October 11.

Late dates of fall departure are: Alaska: Chena River, October 17. Yukon: MacMillan Pass, September 3. British Columbia: Okanagan Landing, November 26. New Mexico: Los Alamos, November 9. Manitoba: Treesbank, November 12 (average of 24 years, November 4). North Dakota: Cass County, November 15 (average, November 11). South Dakota: Sioux Falls, November 15 (average of 3 years, October 28). Minnesota: Minneapolis, November 30 (average of 10 years, November 23). Wisconsin: northern Wisconsin, November 30. Ohio: Buckeye Lake, median, November 28. Illinois: Chicago, December 12 (average of 16 years, November 27). Quebec: Montreal area, November 18. Maine: Lake Umbagog, November 9. New Hampshire: New Hampton, December 18 (median of 21 years, November18); Concord, November 29. Maryland: Laurel, December 5.

Egg dates: Alaska: 47 records, May 4 to July 13; 30 records, May 30 to June 23.

Maine: 22 records, May 23 to July 23; 14 records, May 29 to June 10.

Maryland: 5 records, May 18 to July 9.

New Brunswick: 73 records, May 6 to August 9; 40 records, May 29 to June 12.

Nova Scotia: 49 records, May 4 to July 30; 23 records, May 30 to July 18.

Ontario: 26 records, April 28 to July 19; 13 records, June 11 to June 28.

Quebec: 15 records, June 3 to June 27.

Contributed by ALEXANDER SPRUNT, JR.


The early morning mist hung in a heavy gray curtain over the balsams, drifted in ragged wisps among the huckleberry bushes and stood shroud-like over the lichen-covered rocks. Suddenly atop one of these age-old formations a small gray and white bird took shape, wraith-like against its somber background. Tilting its head upward, the bird opened its beak and poured forth a simple but musical trilL which echoed sweetly amid the swirling banners of mountain fog. A Carolina slate-colored junco greeted the advent of another day.

If any one bird could be said to typify a region essentially, this race of the slate-colored jun00 deserves that accolade for the Blue Ridge Mountains of eastern United States. From the 3,000-foot level to the highest point in eastern North America at Mount Mitchell (6,684 feet), it is the characteristic avian form. When William Brewster (1886) described it as "Differing from J. kyemalis in being larger, with lighter, bluer, and more uniform coloration, and a horncolored, instead of pinkish-white or yellowish bill," he placed its type locality at Black Mountain in Buncombe County, N.C.

For some 20 summers I lived much of the time within 3 miles of Black Mountain and had intimate contact with the bird from the Shenandoah Valley of Virginia to the Great Smoky Mountains at the North Carolina: Tennessee line. During that time I learned something of the bird's way of life at first hand. Certainly, anyone interested in ornithology who has ever visited that region will agree that this jun00 cannot fail to attract attention. Preeminently a mountain dweller, it is as characteristic of the area as the very rocks, the rhododendron "hells," the "balds" and steep ravines, and the the vegetation itself.

Spring: The Carolina slate-colored junco is one of the few examples of eastern American birdlife that can be considered to indulge in vertical migration. Some birds do not leave the mountains at all throughout the winter; those that do usually return at the first marked indications of the change of season, which varies somewhat from year to year. As spring comes on, the birds that descended from the high country to winter at lower levels return to their mountain haunts. This is often only a short distance measured in miles, but it takes them into a completely different environment because of the elevation.

James T. Tanner (1958) comments "The 'migration' of Juncos from the valleys to the mountains in spring is gradual and indefinite. On March 19, 1952, for example, there were very few Juncos in their nesting areas at Mt. LeConte. By April 3 Juncos were common and many were singing there. It appears that March 25 is approximately the time at which the males, at least, move into their nesting territories. Yet some individuals may stay in the valleys around Gatlinburg, at the foot of the mountains, as late as April 20."

Courtship: As my earliest visits to the high country were always about the time nest building was in progress, I have never witnessed the courtship activities of this junco. Tanner (1958) describes the onset of breeding activities as follows:

Each male claims and defends a territory, but the resulting combats and chases are tame and brief, and the territories are not extensive. Early in the season I have frequently seen two pairs feeding side by side, and later birds of different pairs have met with no sign of fight. This behavior suggests that the defended territories are relatively small and leave unclaimed area, and "anyman's land", outside of and around them which may be used for feeding by any Junco.

The male Junco has a display, presumably a courtship display, which is not well known because it is inconspicuous and observed only by those who make a point of watching these birds. It is usually performed on a perch near the ground when the female is nearby. He spreads and droops his tail, droops his wings, and frequently sings a quiet, Goldfinch-like warble which carries only a short distance, very different from the regular song. He displays most frequently in the early part of the nesting season, less frequently later, and has almost entirely ceased before the second brood is started.

However, D. Ralph Hostetter (1961) has the following to say regarding it: "I believe the behavior that I recognize as courtship plays a more prominent part in the life of the birds after pairing than before pairing. * * * If it precedes pairing, it certainly is continued throughout nest building." He continues:

Courtship display on the part of the male begins shortly after the song is in its prime. * * * The male flew about repeatedly from limb to limb on a small tree. He uttered a series of short whistles, "chee-eps," and "tsips", sometimes followed by a short trill. This series of whistles, etc., was continued for about five minutes. The female on the ground seemed to pay no attention to him whatever. During this vocal demonstration the male was about six feet above the ground and the female almost directly beneath him. There was no spreading of wings and tail feathers. Finally he flew into the brush and was lost, followed by the female.

On other occasions there is a pretty feather display accompanying the song. * * * Usually this performance * * * is as follows. The male perches on a small limb above the female who may be feeding or even carrying nesting materials. He sings softly the simple junco phrase or enriches it with various whistles and trills, lowers his wings, and spreads his tail so that the white marginal feathers show beautifully. His head may be thrown back slightly with bill pointed upward, or ho may watch the female.

During all this he is nervously changing his position on the limb or changing limbs. The lowering of the wings and the spreading of the tail is one act of short duration, but is frequently repeated. * * * As the wings are dropped from their dorsolateral position, they are slightly turned out away from the body, and sometimes spread open, just a little, like a fan

Nesting: Tanner (1958) describes nest building as follows:

The building of the first nest is a desultory business. The female, which does all the building, will work at it a while, cease for a period of feeding, return again, and so off and on for several days. Her mate may accompany her as she works, but cooperates not at all. The nest is built of moss, rootlets, and stems, and its deep cup is lined frequently with the slender bristles that support the fruiting capsules of mosses. The typical location of a nest is on the ground or in the mosses and ferns growing on rocks. It is usually well hidden by ferns or other low plants, and placed on a hank or rock with a clear space before it,except for the immediate covering, so that an incubating bird can fly outward and downward from the nest when disturbed. Of 84 nests that I found, 74 were located on the ground or rocks as described. One was in an upturned tree root, four were located from one to three feet above the ground in shrubs, and five were placed on the branches of spruce trees, where spreading twigs formed a platform, from five to forty feet above the ground.

The nesting site varies rather considerably indeed for a generally low nesting bird. The conventional site is under a clod, on the sides of a bank, or amid roots of a windfall, but these by no means exhaust the bird's choice. I have found nests directly on, or rather in the ground, depressed into the soil of open meadows until the rim of the nest is even with the ground. Some are concealed from above by grass tufts, some are not. Though I have found several nests in bushes and young evergreens, such have not been over 5 feet from the ground. D. J. Nicholson wrote me, however, of finding one 17 feet from the ground in the tall sprouts of a locust tree at an elevation of 4,750 feet in the Mt. Pisgah area of North Carolina, and another 16 feet up and out on a limb from the trunk of a balsam on Clingman's Dome in the Great Smokies.

An unusual nest site that vividly illustrates the bird's disregard of people was a nest started in a swinging fern basket on our front porch July 7, 1930. The basket hung about a foot from the entrance door that the family used continuously. The bird used a good deal of the coarse threads from a floor mat just to one side of the basket, and often worked at detaching them while people were on the porch a few feet away. The next four consecutive summers juncos built and occupied nests on the rafters of our open-front garage.

The nest is a well made structure, deeply cupped and substantial. It averages 4.5 inches in width, 2.5 inches in inside diameter, and 1.1 inches in depth. The materials used in its construction are consistently uniform, largely grasses, bark strippings, rootlets, and moss. The lining is quite often horsehair and I have found both gray and white used. At times hair of other animals is used, and now and then evergreen needles.

Two and occasionally three broods, I believe, are raised in the North Carolina Blue Ridge. Eggs of the first clutch number four rarely five; later clutches are smaller, usually three in number. On the time of laying Tanner (1958) comments as follows: The laying of the first egg * * * depends on both the advance of the season and the altitude of the nest in the mountains, and there is also variation between different females or pairs in the same area. The earliest first egg date which I know definitely, April 20, was in a warm spring and at a low altitude. The avenge date of hying the first egg is about eleven days later for each thousand feet increase in altitude. The latest date for the laying of the first egg, of what I am sure was the season's first nest of the pair, was May 31.

The remaining eggs of the clutch are laid one each day, usually in the morning, until the set is complete. Four eggs is the usual number in the early nests. The female begins incubating with the laying of the next to last egg. While incubating she sits still with little or no turning, and will flush from the nest only when closely approached; I have had my face within a foot of an Incubating bird before she left. This, and the fact that the nests are usually well hidden, makes it difficult to find nests during this stage. The male does not incubate, but he stays in the vicinity, foraging for himself and occasionally singing. When the female leaves the nest for food, she frequently flies directly to the male and then moves off to search for food. She forages industriously with the male usually following her, behind and a little above. He may perform the courtship display. The female usually moves in a circle back to the nest, which she frequently approaches by hopping along the ground. During the incubation period she spends about three times as much time on the nest during the day as she does off for feeding, but the periods on and off are irregular. She passes the night on the nest.

The average incubation period is about twelve days, and since incubation begins with the next to the last egg, the eggs of a four egg set begin to hatch fourteen days after the first egg is laid. They do not hatch simultaneously, partly because the last egg is not laid until after incubation has started.

Eggs: Those of this subspecies do not differ in color from those of the nominate race. The measurements of 40 eggs average 20.1 by 15.2 millimeters; the eggs showing the four extremes measure 21.2 by 14.7, 21.1 by 16.0, and 18.8 by 14.6 millimeters.

Young: Again according to Tanner (1958):

"Both parents bring food to the young, working hard and steadily. Whenever one gets a billful of food, it flies directly to the nest and remains there very briefly, except when the female settles to brood for a short time, which she does more frequently when the young are small. The food brought consists of insects; moths and small caterpillars are common. The size of the insects increase with the size of the nestlings. Because of the feeding activity of the parents, and also because they are likely to scold persistently if a person comes near, nests are most easily found at this stage.

When hatched, the nestlings are naked except for a little down. Two days after hatching the sheaths of the body feathers appear as dark "pinfeathers". Five days after hatching the eyes open, and a day later the primary feathers of the wing begin to k eak out of the tips of their sheaths. Nine days after hatching the tall feathers begin to break out at the tip and the body is fairly well covered with feathers. The young may leave on the tenth or eleventh day if the nest is disturbed, but normally they will leave on the twelfth or thirteenth day. By this time the wings are well developed, the tail is almost an inch long, and they can fly clumsily but surprisingly well.

Young out of the nest are fed by the parents at least until the former are full grown. The length of care seems to depend on whether or not a second nest is started, in which case the young are earlier left by themselves. The majority of Juncos in the Smokies will nest a second time; the exceptions are some pairs at the very lowest elevations which appear to nest only once. The behavior during the second nesting is the same as during the first.

Plumages: Richard R. Graber (1955) comments that this race is very much like J. it. hyemalis in all plumages, but slightly grayer and less brown throughout. He describes the juvenal plumage of carolinensis as follows:

"Crown and nape dill gray, uniformly streaked with black. Back tinged slightly with brown, rather sparsely streaked with black. Rump gray brown, mottled with blackish. Upper tail coverts drab gray brown, obscurely streaked with blackish. Outer two pairs of rectrices largely white, third from outside with some white, others black. Remiges black; primaries and secondaries edged with whitish, tertials with buff or huffy gray. Coverts edged with buff. Lores, eye ring, and sides of head rather flat gray (uniform, except auriculars, lightly flecked with dusky). Chin and throat light gray or whitish, streaked and spotted (obscurely in some specimens) with blackish. Chest buff-tinted gray, or buff; sides and flanks buffy. Chest and sides streaked rather heavily with blackish. Belly and crissum unmarked white. Legs gray."

Food: Though not extensively studied so far as I am aware, the food of this race parallels that of J. h. hyemali~s. Seeds form much of it, those of both grasses and weeds, particularly in the colder months. During the warmer seasons it eats a higher proportion of insects, and it feeds the nestlings exclusively on animal food. It forages in the usual junco manner, scratching among leaves and ground debris beneath thickets and brush, and it often visits feeding stations immediately adjacent to human dwellings.

Voice: The song of this junco is very similar, practically identical in fact, to that of hyemali.s. It is a melodious trill, often uttered for considerable periods and frequently from an elevated perch. The call note can hardly be described as anything but a sharp "chip," easily recognized when once learned.

Behavior: The outstanding characteristic of this junco is its tameness. About such heavily populated resort towns as Blowing Rock, Linville, and Highlands, as well as along the Skyline Drive in Virginia, southward along the Blue Ridge Parkway, and in America's most visited national park, Great Smoky Mountain National Park, the bird is almost underfoot everywhere. Along roadsides, trails, and streets, in yards and gardens, the familiar "chip," trilling song, and flash of white enter tail feathers are conmon sights. The bird is omnipresent on manicured estates as well on remote summits and peaks. Its universal local name is "snowbird."

Fall and Winter: As fall comes and the first frosts occur, carolinensis begins to drift to lower levels. At an elevation of 2,800 feet near Black Mountain where the bird is not present during the summer I saw the first ones on Sept.. 30, 1931, after four days of frost with day time temperatures ranging from 370 to 420 F. As winter comes on, the birds drop farther into the valleys and along the Piedmont, where they may join the flocks of migrant hyemalis. Generally they stay close to the foothills and move only casually southward or out onto the coastal plain. That this junco can and does remain at high altitudes through the winter is attested to by T. D. Burleigh's (1941) record of "small flocks * * * feeding contentedly in the shelter of fir thickets" with a foot of snow on the ground, with the temperature zero near the summit of Mt. Mitchell on Dec. 23, 1930.

Range: Chiefly Appalachian Mountains from Maryland and West Virginia to Georgia.

Breeding Range: The Carolina slate-colored junco breeds in mountains from northeastern West Virginia (Terra Alta~ and western Maryland (Accident, Fiuzel) south through extreme eastern Kentucky (Black Mountain), western Virginia, and western North Carolina to eastern Tennessee (Unicoi Mountains), northern Georgia (Ellijay), and northwestern South Carolina (Sassafras Mountain).

Winter Range: Winters chiefly on breeding grounds, descending in part to lower elevations in the mountains and the adjacent valleys; casually to central Maryland (Howard County), central Virginia (Amelia), central North Carolina (Raleigh), coastal South Carolina (Mount Pleasant), and central Georgia (Augusta).

Egg dates: North Carolina: 42 records, April 20 to August 11, 15 records, May 15 to June 9; 13 records, July 8 to July 22. Virginia: 6 records, May 6 to June 6.

Contributed by OLIVER L. AUSTIN, JR.


This name denotes the junco population breeding in the basin country east of the western coastal ranges from south central Yukon southeastward to east central British Columbia and west central Alberta, which A. H. Miller (1941b) characterizes as "a group of hybrid origin, now stabilized, which occupies a geographical area in the breeding season to the exclusion of other forms." Males differ from nominate hyemalis in having the head darker than the back and the edge of the slate color on the chest convex instead of concave; females have the sides washed with brown or pinkish rather than gray. The form ostensibly intergrades with the several others whose nesting territories its breeding range adjoins, and individuals taken on the wintering grounds to the southward cannot always be identified with certainty. And, as Miller adds, "In other regions birds of identical appearance may be produced by hybridization."

Practically nothing has been published on this population's habits, which are not likely to differ essentially from those of the adjoining populations described in detail elsewhere in this volume.

The measurements of 34 eggs in the MCZ assigned to this race average 19.3 by 14.7 millimeters; the eggs showing the four extremes measure 21.1 by 14.2, 19.3 by 15.5, 17.8 by 14.8, and 19.8 by 13.2 millimeters.

Range: South central Yukon and British Columbia south to northern Baja California, southern Arizona, and central Texas.

Breeding Range: The Cassiar slate-colored junco breeds from south central Yukon (Carcross) south to central interior British Columbia (Hazelton district, Sinkut Mountain, Tupper Creek) and west central Alberta (140 miles west of Edmonton). (Some hybridization between this form and J. o. mont anus occurs at the western and southern borders of breeding range.)

Winter Range: Winters from southern British Columbia (Vancouver, Okanagan Landing, Cranbrook), Nebraska (Long Pine, Lincoln), Minnesota (Minneapolis), and Wisconsin (Beaver Dam) south to northern Baja California (Laguna Hanson), southern Nevada (Charleston Mountains), southern Arizona (Yuma, Chiricahua Mountains), New Mexico (Las Vegas), central Texas (Waring), and southern Michigan (Kalamazoo, Washtenaw, and Monroe counties); casually east to southern Ontario (Toronto), eastern New York (Hastings), Massachusetts (Wellesley), Kentucky (Graves and Jefferson counties), Virginia (Herndon, Arlington), Tennessee (Germantown), Georgia (Athens, Decatur, and Chatham County), Mississippi (Saucier), Arkansas (Delight), and Louisiana (Catahoula Lake).

The following subspecies are discussed in this section: Junco oreganus montanus Ridgway, J. o. shufeldti Coale, J. o. thurberi Anthony.

Contributed by JAMES H. PHELPS, JR.


To me, Oregon junco belong to the forest "edge," particularly the edge of the coniferous forest. An "edge" is that irregular meandering border between a mountain meadow and the timber; a fading woods road no longer used and now crowded with new second growth; an opening in the forest surrounding a pond or along a stream; or an old burn, full of snags, fallen logs, stumps, and a tangle of shrubs and other low growth. Juncos are everywhere in such situations. When disturbed, their white outer tail feathers flash attention as they scatter from the ground to nearby conifers; at certain seasons one may think juncos are the only birds in existence.

Aspen (Populus tremuioides) forests, which often grow much interrupted as subclimax within the true forest, are well liked by nesting juncos. Open park-like stands of yellow pine (Pinus ponderoaa) and lodgepole pine (P. contorta) of the interior west have much suitable and attractive edge. High mountains of Baja California, with similar but perhaps drier forest cover, have juncos. The dry forests of the California interior and southern mountains differ greatly from the forests of the humid coast, but Oregon juncos are found in all of them. Compact low fir growth at timberline attracts nesting juncos.

The northwest "rain coast" has forest and forest understory of such density that it must be seen to be appreciated. Here juncos find the necessary openings in the forest are scattered and infrequent; the sea shore becomes edge. In the humid country west of the Cascade Mountains, Oregon juncos are city and country dooryard birds, sometimes permanent residents. Juncos are common in city parks and on college campuses in many parts of the West; the natural and planted vegetation, including various conifers, has the edge effect they favor.

The many races of the Oregon junco tolerate variable conditions of moisture, forest vegetation, heat and humidity, from sea level to timberline. Adequate ground cover in the way of grasses or flowering broad-leafed herbs and the like that continues green or at least some-. what succulent is necessary during the nesting season. Philip Dumas (1950) says: "A scattered shrub layer may or may not be present. The herb layer, however, is prominent and made up of many grasses and herbs."

Apart from nesting, juncos are not closely dependent upon edge, yet the attraction continues. The forest growth provides shelter and protection; the open spaces and bare ground provide opportunities to forage.

Oregon juncos are as well known to westerners as slate-colored juncos (J. kyemalis) are to easterners. A. A. Saunders (1936b), having had field experience with the slate-colored, white-winged (J. aikeni), pink-sided (now J. o. mearnsi), and Oregon juncos, remarks: "[They] differ from each other in coloration only. Their calls, songs, and nesting habits, so far as I have observed them, are all alike."

As defined in the 1957 edition of the A. 0. U. Check-List, which we are following, Junco ore ganu.s is divided into eight races: montanus,, ore ganus, shufeldti, thurberi, pinosw~, pontili,s, and townsendi. J. K. Townsend (1837), who described Fringilia oregana, spelled "Oregon" with an "a," in common use at that time, and this spelling is retained. Life histories of the races montanus, shuJeldti, and thurberi are given in this account; the other races are given in subsidiary accounts following. The race mordanu.s is known to many as the Montana junco and many references in the literature are cited accordingly; similarly the race .shufeldti is known as Shufeldt's junco and the race thurberi is known as the Sierra junco or Thurber's junco.

In July 1911, J. H. Riley (1912) found slate-colored juncos and Oregon juncos nesting together without hybridizing. A. H. Miller (1941b), with more complete knowledge of distribution of the juncos at his disposal than Riley had, indicates montanus is the race involved; Riley says: "Junco h. hyemalis is the common junco east of the main divide, and Junco o. shuiJeldti west of it, but we found them both breeding together at Henry House, Alberta, Yellowhead Pass, B.C., and at the foot of Moose Pass, B. C.; all of these localities are just over the main divide, either east or west. This convinces me that both belong to distinct species as no intermediates were taken."

Nevertheless, Oregon juncos do hybridize with other junco species, and different races of the Oregon junco intergrade with one another where their ranges meet. In some parts of the West as many as 10 forms of juncos may occur during migration and the winter season. Hence it is not surprising that their nomenclature has suffered much change and confusion. For a full account of the relationships of the different juncos, see A. H. Miller (1941b).

Spring: Of the three races discussed in this account, montanus migrates more or less completely, the race .shuJeldti less so, and the race thurberi partially and, in some cases, not at all.

A short dispersal to nearby second growth or cutover timber lands suffices for some juncos wintering in the mild Pacific climate; others make a vertical migration from coastal or interior valleys to the higher mountains. Migratory Oregon juncos wintering on the low mountains of New Mexico or in the deserts of Arizona have hundreds of miles to go to reach chosen summer homes.

Where spring migration is noticeable, Oregon juncos appear first at lower elevations and move higher as the season advances. This suggests that spring migration follows natural pathways such as river valleys and mountain passes, pausing if inclement weather occurs. In our intermountain country, with its interesting mixture of isolated mountain ranges and intervening valleys, I have noticed this pattern. On May 10, 1959 several inches of wet, clinging snow covered the low country in our part of Idaho; snow bad melted from roadsides and trails only; higher elevations had more snow. Blame Lyon and I saw juncos everywhere as we hiked into the foothills. Juncos flew from beneath roadside cuts, from under shrubs covered and bent with snow, and from the path ahead of us. It seemed to us as though the birds were marking time until they could continue migration.

We could easily recognize more than one race. Included in the many small irregular sized flocks, besides our local breeding race mearnsi, were other Oregon juncos with darker heads, more brownish sides, or more brownish backs than mearnsi. Two or three slate-colored juncos were seen this day, too, confirming that migratory juncos were indeed present. Two weeks later juncos were gone from the low country and only our local breeding birds were seen in the mountains.

Migration dates in spring, as given by A. A. Saunders (1912a) for western Montana near Butte, are March 10 to April 5. L. R. Dice (1918), at Prescott in southeastern Washington, for the years 1905 to 1913, gives April 13 and May 1 as last spring observations, indicating migration had taken place from the low country. In the Lahontan Valley, Churchill County, Nev., an irrigated farming area, J. R. Alcorn (1946) says: "Juncos (of all races) were seen frequently each month from September to April, inclusive." The status of the different races for California as summarized in Grinnell and Miller (1944) indicates spring migration takes place in March and April. R. B. Rockwell (1908) and others say that the bulk of migration northward in Colorado is about April 1.

T. T. and Elinor McCabe (1927), at their banding station in interior British Columbia, report that during the 1927 spring migration "the great rush began on the 10th lof April]." They comment also on the excess of males over females, about four to one. At Coeur d'Alene, Idaho, Henry J. Rust (1915) writes: "Arriving as early as February 22, becoming common by first week in April." A. R. Phillips (1933) says wintering Oregon juncos were seen in the Baboquivari Mountains, Ariz., until Mar. 16 and Apr. 1,1932.

The experiments of Albert Wolfson (1942) with resident and migratory races of Oregon junco deserve mention. Wolfson says that internal and external factors were found to regulate the spring migration of the Oregon junco. The external factor is the length of day. "As the days increase in length the birds are awake for longer periods." Longer periods of being awake stimulate an increase in production or release of the production from various glands, which in turn causes physiological changes in the bird such as recrudescence of the gonads and the deposition of large amounts of fat. The internal stimulus triggers the actual migration by releasing the nervous mechanism that controls the migratory behavior. Resident birds show no deposition of fat, says Wolfson. On the other hand, the testes of the residents become active and increase in size at a faster rate than those of the migrants.

Courtship: References in the literature to courtship are sparse and fragmentary. Perhaps this is because courtship often begins in the flock, and unless an observer has long acquaintance with a particular flock he hesitates to record the actions of a lively group of small birds that look quite alike superficially.

A. H. Miller (1941b) describes a male junco courting a female with young, replacing a male that had been collected. While the particular reference pertains to the race mearnsi, I can find nothing to indicate the other Oregon junco races act differently. Miller writes: "A new male was on hand, singing, following her with tail fanned, and twittering in characteristic mating behavior." Miller also had some pairs of juncos in an aviary, and in an additional comment, he writes (pers. comm.): "What was observed there, of course, may have been more a matter of pair formation than actual courtship prior to copulation. However, the tail spreading and twit-. tering certainly comes in at that early stage of pair relations. In the cage I saw copulation at least once, and there was no conspicuous behavior preceding it: that is, no special notes or performance."

Just when the mating process began in the flock Winifred S. Sabine (1955) studied was not learned; but she did notice an increasing tendency for the two mated birds to arrive and depart from the feeding station together, and that as spring approached the dominant male showed an increasing intolerance towards juncos other than his mate. Mrs. Sabine (1952) suggests that though she observed no courtship gestures as such, probably her observations were incomplete for she "has seen them in a difFerent pair of the species and also has seen elaborate displays in a migratory flock of J. hyemali8."

Among spring flocks of foraging juncos, an individual bird may detach itself, fly a more or less brief circle, and alight not far from whence it started, More often than not another junco will give chase, as it were, flying the same course. A person observing the the flight is immediately aware of the flashing white outer tail feathers. Whether or not this behavior should be interpreted as courtship needs further study.

Nesting: I have never found nesting pairs at any great distance from water. The water may be a small seep or a minute trickle from a melting snow bank, but it is water.

Junco nests are usually on the ground in a cup-shaped depression. Typically they are made of grasses or fine stems and placed at the base of a shrub or plant. Animal hairs are often incorporated into the nest structure or used for lining. Chester Barlow (1901) writes of nests in the California Sierras built on the ground in the sides of shallow ditches and concealed by the vegetation. One nest and its surroundings were composed entirely of pine needles. H. T. Bohuman (1903) tells about two nests found in sides of "railroad cuts" near Portland, Oreg. He writes: "Both were constructed of an outer layer of coarse grasses, then a thick layer of finer grasses, and a lining of cow hair. The inner cavity measures two and one-fourth inches across and one and one-fourth inches deep, while the outer measurements are two, and two and one-half inches in depth respectively." Bohlman comments that one nest was lined with white cow hair, the other with black.

To T. D. Burleigh (1930), juncos are one of the characteristic birds of the scattered stretches of open fir woods of the Puget Sound country: "Here birds were frequently flushed from nests that almost invariably were sunken in the green moss that covered the ground, and protected and concealed by a dead fir limb or, rarely a clump of dead ferns. They were substantially built of weed stems and fine grasses, in one case with green moss intermixed, and lined, sparingly at times, with horse hair."

Burleigh (1930) found one nest "snugly built in an old rusty tin can lying at the edge of an open field, and twenty feet from the nearest underbrush." Milton S. Ray (1903), F. S. Hanford (1913), and D. I. Shephardson (1917) report similar tin can nests. Joseph Ewan (1936) reports a nest on a rafter in an old hay-filled barn; D. S. DeGroot (1934) and others tell of nests in crevices of rock ledges; Ray (1918) reports a nest built in the corner of an empty box in full view of hotel guests passing to and fro; and W. E. Griffee (1944) tells of nests built on stringers between the joists of wooden barracks with no enclosed foundation.

S. G. Jewett (1928) found two Oregon junco nests in trees. His unique observations were made on the same day; the nests were about a mile from each other, both in lodgepole pines each about 8 feet from the ground. He describes them as "loosely built." W. E. Griffee (1947) describes a junco nest he discovered in his own back yard in Portland, Oreg. He writes he "was amazed to find it fully 20 feet above ground and 8 or 9 feet out from the trunk on a thick brushy lower branch of a Douglas fir." H. B. Kaeding (1899) notes most nests are built on the ground, but nests in trees and deserted woodpecker holes have been recorded.

Concerning the actual nest construction May R. Thayer (1912) writes: "Our first intimation that they were building in our immediate vicinity came to us on June 8 [1908], when I noticed the female picking up hairs that Donald, the collie, had scattered on the walk. * * * For two or three days, they busied themselves with the completion of the nest. The actual construction of the nest seemed to fall to the share of the female, while the male watched over her, encouraging her by his presence and his music. They always came together for the dog hairs. Often he would perch on a tall stump beside the walk, and watch her while she worked, singing with the greatest energy.~~ A. H. Miller (pers. comm.) tells me the female "builds the nest although, as in so many cases where this is true, it is not unheard of that the male may toy a bit with material; but I have no evidence that he gathers up a substantial amount of material and goes and puts it in the nest."

Nesting begins early along the Pacific coast and at lower elevations. Birds farther north, or at higher elevations and in the interior nest at later dates. H. B. Kaeding (1899) writes: "During the summer june05 may be found up as high as 10,000 feet in the Sierras, but not as a rule lower than 3,000 feet, breeding. The breeding dates vary with the altitude and eggs may be found at 4,000 feet in May and as late as July 15 fresh eggs have been taken at 9,000 feet altitude."

In the vicinity of Portland, Oreg., Griflee and Rapraeger (1937) give April 22 as the earliest date, June 8 as the latest date, and the height of nesting season, when about 50 percent of the birds start incubating, as April 28 to May 3. The earliest date Burleigh (1930) recorded for the lowland Puget Sound area was May 9: a nest with four fresh eggs. At Coeur d'Alene, nort.hern Idaho, Henry J. Rust (1915) cites the following examples: "Pair noted gathering nest material March 27; five nests, each containing five eggs, examined May 8; young able to fly, May 19; nest with five newly hatched young, June 18; nest with five fresh eggs, June 27."

The variation in dates suggests that Oregon juncos raise two broods in a season. In this regard D. S. DeGroot (1934) writes:

Early in July (1933] a pair of Juncos * * * was located feeding four halfgrown young in a beautifully hidden nest in a crevice in a rock. These birds were watched daily and the parents quite definitely identified by certain peculiar markings and actions. On July 15 the young of this nest were out learning to fly and three days later they had disappeared in so far as I could ascertain. Two days later the female was seen carrying nesting material and on July 29 she was flushed from her new nest which at that time contained two fresh eggs. This nest was located not twenty-five feet from the first one and was similar in many respects although it was placed in a patch of skunk cabbage instead of a rock crevice, as in the first case. On August 3, the ne~t contained four eggs and the female was sitting. On the 16th the eggs had hatched and at this writing [August 26, 1933] the young are about ready to fly. Another nest of this species was located in the same area at about the same time and similar observations were made, thus confirming my suspicion that this species is one which nests rather commonly twice each season.

I have not been able to find other confirmation in the literature of A. W. Anthony's (1886) claim that in Washington County, Oreg., the species "breeds everywhere, raising three and often four broods."

Eggs: Usually four eggs are a set, sometimes three, occasionally five, and an authentic set of one is known. Milton S. Ray (1919) describes 75 sets totaling 300 eggs of thurberi he collected over a period of nearly 20 years. He says the eggs are: "Usually ovate or rounded ovate, sometimes short ovate, rarely elongate ovate" with a "slight, very slight, or scarcely perceptible gloss. * * * Ground color white, faintly tinged with lichen green * * * spotted and blotched or splashed with hazel, * * * chestnut, * * * and light vinaceous gray * *

Wilson C. Hanna (1924) gives the weight of a set of four Oregon junco eggs. The set averaged 2.08 grams, the largest egg weighed 2.15 grams and the smallest 2.00 grams.

W. G. F. Harris (MS.) thus describes the eggs of the species as a whole: The Oregon junco lays from three to five and most comminly four eggs. They are slightly glossy, and ovate, sometimes tending toward short ovate. The ground is white or very pale bluish-white, speckled, spotted, and sometimes blotched with reddish-browns such as "snuff brown," "russet," "cinnamon brown," "Brussels brown," and "chestnut," with undermarkings of "pale mouse gray." The spottings are generally concentrated toward the large end where they frequently form a wreath. There is considerable variation, some eggs are only sparsely marked while others may be quite heavily spotted or even blotched. The eggs of this species are indistinguishable from those of Junco hyemalia. The measurements of 142 eggs average 19.3 by 14.6 millimeters; the eggs showing the four extremes measure p21.6 by 15.3, 18.3 by 16.8, 17.1 by 14.9, and 19.3 by 1322 millimeters.

Young: D. S. DeGroot (1934) gives dates that indicate incubation lasts about 13 days. So far as I can determine incubation is entirely by the female. A. H. Miller (pers. comm.) tells me "in the aviary, as also in the field, I find that the female alone incubates."

A. D. DuBois, quoted by A. A. Saunders (1921), tells about a nest in which the eggs hatched May 30 and the young left the nest June 12. These dates suggest the young need 13 to 14 days to fledge. The young left the nest 11 days after hatching in a nest Winton Weydemeyer (1936) reports upon.

Irene G. Wheelock (1905) writes that on the day of hatching in a nest she was observing the young were fed 15 times during a 3-hour period in the morning. During the second day feedings were more frequent. She writes: "In two horns, from 9 to 11 a.m., the male came to the nest six and the female eight times. From 1 to 2 p.m. there were 11 feedings." The young were fed both vegetable matter consisting of seeds which she could not identify further and animal food consisting of insects. Large insects had the wings and legs carefully removed by the parents before being given to the young.

Plumages: A composite description of the three races discussed in this account is as follows: The head and breast including the chin and nape are collectively called a hood; the margin of the hood is convex posteriorly. The hood is somewhat darker on the top and sides of the head and is sharply defined from the back and side plumage. Depending upon the race, the hood may be deep to dark neutral gray, sooty gray, and black in some birds. Females average lighter in hood color than males, but this is a relative and not an absolute comparison. The back and scapulars are confluent and are a shade of brown, grayer or brighter depending on the race. Sides may be yellowish, buff, or light ruddy brown; the belly is white. The outer three pairs of tail feathers are white, varying slightly in amount in the three races. In all races the feet are red-brown, the bill flesh color, and the iris some shade of brown.

According to A. H. Miller (1941b) the race montamus averages larger than shufeldti and about 80 percent can be separated by measurement. Also, about 80 percent of the races shuJeldti and thurberi can be separated by their back color; in fresh plumage shufekfti is bister, contrasting with Verona in thurberi. At best, however, the races are difficult to separate in the field, and the literature and observations of the many workers over the years are thoroughly intermingled.

No descriptions of the juvenal plumages of the races montanus, shufeldti, or thurberi are available. An unpublished thesis by Richard R. Graber describing the juvenal plumages of oreganus and mearnsi is quoted in the respective accounts of those races. Oregon juncos do not keep the juvenal plumage for any length of time. I rather suppose that all races are like the race mear'nsi. Young of this race that I saw in the Centennial Mountains of the IdahoMontana border country (Clark County, Idaho) were already beginning to get their dark hoods in early August 1960, although their bodies were still streaked.

Gordon W. Gullion wrote to Mr. Bent: "Members of the Oregon jun00 species breed abundantly in the Yakima Park area in the northeastern part of Mount Rainier National Park in Washington state. While conducting banding operations in that area during the summer of 1947 I had the opportunity to trap a larger number of juvenal birds, often still retaining some of their natal down. Since these juncos remain in the area until late fall, it was possible to obtain three records on the time requlred for attaining adult plumage.

"One streaked immature bird banded (47: 64082) on July 24 had the full adult male plumage 37 days later on August 30. Another (46: 23936) banded in immature plumage on August 1 had the blackheaded adult plumage in 28 days on August 28. The third bird, banded as an immature (46: 23945) on August 14, had most of its adult plumage in just 22 days on September 5. From these data it seems probable that juncos attain their adult plumage in 2 or 3 months after hatching."

A. H. Miller (1941b) writes:

The annual molt of juncos is the only one of any importance. The prenuptial molt is nearly obsolete and effects no material change in appearance. The postjuvenal, or first fall molt, is incomplete; primaries, secondaries, and greater primary coverts are not replaced, and the tail usually is not molted at this time. Juvenal wing and tail feathers carried through the first year are indistinguishable from those of birds a year or more old. The body plumage of the fall immatures in a number of races is different in average coloration from that of adults. But no absolute differentiation has been revealed: a situation also true of sexual dimorphism.

As stated above, on the average females have the hood lighter in color. In the three races being discussed here, the difference is usually apparent. Brownish overcast of the head is common in all races (somewhat less so in thurberi than in the other two); it is to be looked for most often in young and females. Miller agrees with E. Mayr (1933) that the brownish overcast can be "properly viewed as a retarded phase of plumage." White wing spotting is a rare variant. Other odd color patterns occur, especially about the hood, and Miller says of this in mantanus: "They appear to be due to incomplete molts with retention of some feathers of a previous retarded plumage."

A number of albino Oregon juncos are on record. Louis B. Bishop (1905) writes:

A beautiful, albinistic, male junco was collected at Witch Creek [California) on Nov. 10, 1904, in company with a typical female Thurber's junco * * ~. The bill, tarsi, toes, and nalis axe pinkish white; forehead, lores, infra-orbital region, chin, lower breast, abdomen, wings and tail, white; the wings and tail slightly mottled with ashy; a slight pinkish suffusion on the sides and the greater wing-coverts; and the rest of the plumage, including the throat, and entire upper parts, blackish slate or slate-color, edged with grayish white. In coloring, therefore, this bird is nearer hyemalis than to thurberi.

Bishop also took a female at Witch Creek, Dec. 14, 1903, that had the chin and part of the throat grayish white.

S. G. Jewett and Ira N. Gabrielson (1929) write: "On December 8, 1924, a perfect albino junco was killed on Sauvies Island [near Portland, Oregon] and brought to the State Game Commission.~~ R. C. McGregor (1900b) collected an Oregon junco having a narrow collar of white about the neck at Saint Helena, Calif., January 1899 which he classified as Junco hyemalis oregonus, though to just what race it would now be referred cannot be determined. A. W. Anthony (1886) collected an almost pure male albino, mated to a normal female, near Beaverton, Oreg., some time in May 1885. Anthony had noticed the bird the year before but was unable to secure it; local residents told him the albino had been present at least two breeding seasons before Anthony first saw it.

Hybridization and intergradation.: The races montanuz and mearnsi, the latter at one time separated specifically as the pink-sided junco, intergrade. The races montanus, skufeldti, and oreganu.s intergrade; shuJeldti and thurberi intergrade; thurberi surrounds in range and intergrades with pinosu.s. The resident races of Baja California, pontiis and tow'nsendi, are not known to intergrade or hybridize with other juncos.

The race tliurberi hybridizes with the gray-headed junco (Junco caniceps) as does the race mearnsi. This is discussed in the species account of the gray-headed junco.

Hybridization between the slate-colored junco (Junco hyemalis) and the Oregon junco (Junco oreganus) is likely whenever a junco of one species in breeding condition finds itself surrounded only by june05 of the other species, also receptive. As individuals of each species may be found well within the range of the other, it follows that a wide scattering of hybrids does occur. When the two interbreed the color areas of the side, back, and head are modified, and every degree of mixture in plumage can be found. For a full discussion of the relationships between the different juncos, see A. H. Miller (1941b).

Food: G. W. Salt (1953) aptly classifies the junco as a "groundseed forager." Seeds of many plants are the main food of the Oregon junco. Weed seeds and waste grain are important during fall and winter. Insects are eaten when available and are important food during the nesting season. A. C. Martin, H. S. Zim, and A. L. Nelson (1951) say: "Beetles (especially weevils), ants, caterpillars, grasshoppers, leafhoppers, together with some spiders, wasps, and flies are the chief items in the animal diet of this junco."

F. E. L. Beal (1910) analyzed the contents of 269 Oregon junco stomachs, most of them taken in fall and winter. His first analysis '~gives 24 percent of animal matter to 76 of vegetable." His statement about animal food, mostly insects, agrees well with the statement of Martin, Zim, and Nelson. Beal adds: "Caterpillars are apparently the favorite insect food, forming 9.4 percent of the diet." Weed seeds are eaten in every month and on a yearly basis amount to 61.8 percent of the food, in September nearly 95 percent.

G. F. Knowlton (1937h) reports the contents of 25 stomachs from Oregon juacos as follows: "180 Ilomoptera in 18 stomachs, composed entirely of 13 adult and 167 nymphal beet leafhoppers in 12 stomachs; 3 Hemipteria in 3 stomachs; 9 adult and 11 larval Coleoptera in 4 stomachs; 2 adult and 36 larval Lepidoptera; 1 Diptera; 4 Hymer&optera, all being ants; 1 spider; 134 weed seeds, a number of which were from sunflowers; numerous plant fragments."

Martin, Zim, and Nelson (1951) report the analysis of 265 Oregon junco stomachs, most of them collected in winter in California. They say that in California cultivated (Aven~ sativa) and wild oats (A. Jatue and A. barbata) make up 10 to 25 percent of the fall and winter food, with common chickweed (Stellaria media) being equally important. Cultivated and wild barley (Hordeum mdgare) amount to 5 to 10 percent of fall and winter food, as do the small, shiny seeds of various pigweeds (Amaranthus sp.). The black shiny seeds of both redmaid (Calandrinia caulescens) and minerslettuce (Montia perJoliata) are prominent spring foods; also eaten in season are cryptantha (Cryptl2ntha sp.), goosefoot (Eleu&ine sp.), knotweed (Polygonum sp.), filaree (Erodium sp.), and melicgrass (Melica sp.) ;with lesser amounts of pine (Pinus sp.), silene (Silene sp.), wheat (Triticum sp.), eriogonum (Eriogonum sp.), annual bluegrass (Poa annua), star-thistle (for the most part Centaurea melitensis), mayweed (Anthemis cotula), scarlet pimpernel (Anagallia sp.), sheepsorrel (Rumex acetosella for the most part, also called sour grass or red sorrel), woodsorrel (the most abundant being Oxalis stricta), tarweeds (Madia and Ilemizona sp.), Pacific poisonoak (Toxicodendr-um diversiobum), and plant galls.

Of 39 Oregon juncos Lowell Adams (1947) trapped as a part of a study of damage caused to seeds of forest trees, the stomachs of six contained Douglas fir (Pseudotsuga taxijolia) seeds. Lincoln Ellison (1934) says that at the Priest River Forest Experiment Station in northern Idaho juncos became active about the seedbeds soon after they were uncovered in the spring, and they ate "In order of preference, the seeds [ofi: Western white pine, Pinus monticola; Douglas fir, Pseudot8uga ta.rifolia; western larch, Lzri~, occidentali8 * * *

"From the appearance of the nipped cotyledons of some young white pine and Douglas fir seedlings, it was surmised that the juncos had picked off the attached seeds which still contained some nutrient matter."

Mrs. H. J. Taylor (1920) watched a red-breasted sapsucker (SpA yrapicus varius subsp.) working and feeding in an old pepper tree (Schinus molle). Sap exuded from the holes the sapsucker made and attracted insects. If the sapsucker was absent, other birds including the Oregon junco fed from the holes, whether on the sap or on insects, Mrs. Taylor was unable to tell.

S. G. Jewett (1938) notes Oregon juncos among the small birds drinking and bathing in the highly mineralized water flowing from small springs near Paulina Lake, Desehutes County, Oreg. The water is unpleasant to human taste.

John M. Robertson (1931) says juncos are among the many birds in California that utilize the small black seeds produced in great abundance by the imported blue gum tree (Eucalyptus globulus). L. L. llargrave (1939) reports Oregon juncos wintering in Arizona eat the exposed seed pulp of fully ripened pomegranates that remain hanging upon the bush. E. D. Clabaugh (1930) suggests that to trap Oregon juncos for banding "Chick-feed, cracked corn, bread crumbs and bird-seed are all good baits."

Behavior: The Oregon junco is a social bird. Except during the breeding phase of the life cycle the species gathers in irregular-sized flocks. I. N. Gabrielson and S. G. Jewett (1940) say: "After the nesting season the birds roam the country in small family flocks that gradually merge into larger groups that sometimes number into the hundreds."

J. Eugene Law (1924) observes the following of the August flocks in the San Bernardino mountains of California: "The personnel of these flocks of Juncos must be purely fortuitous at this season. Family ties seemed to be entirely broken except for an occasional late-born youngster. The groups, if they can be called groups, were constantly milling about over any part of the meadow, and they did not leave a spot in the same order or numbers in which they arrived. From numbers feeding on open ground scattering individuals frequently left without disturbing many which remained."

Of the wintering flocks Mrs. Winifred- S. Sabine (1956) writes: "There appears to be no limit to the size of a foraging group. * * * The entire flocking procedure is marked by the continual forming and dissolving of groups of unpredictable size consisting of individuals that consort together and are daily visitors at the feeding sites."

The "purely fortuitous" composition of the flock remarked upon by Law is questioned by Mrs. Sabine (1959) who finds that Oregon juncos have a definite pecking order which she charts as she determined it among winter flocks. Mrs. Sabine's work suggests this intolerance serves as a spacing device which allows each bird of this social species a necessary area of privacy, main tamed by the dominant bird of the pecking order and by each subordinate bird, thereby avoiding conflict with all others dominant over it. The increased intolerance that begins with spring serves to disperse the species for the breeding phase of the life cycle.

James 0. Peterson (1942b) fed Oregon juncos cornmeal spread upon the snow at his feeding station at Cuyamaca Peak, elevation 5,000 feet, in San Diego County, Calif. He reports: "As an individual junco came to feed, the tail would be spread each time a morsel of food was picked up. Succeeding birds would not alight on the feeding area at random, but would perch on a shrub or some other elevated point in order to view the flock. Careful appraisal would soon reveal that at one or two points the feeding birds were flashing their tail marks very rapidly. Invariably it was to one of these points that the new and hungry bird would fly, and in alighting force the feeding bird to vacate the spot where the cornmeal was piled. The failure of the feeding bird to obtain food was a sign for the flock to break up."

Oregon juncos often forage in mixed flocks and otherwise associate with other small birds, such as chickadees (Parus sp.), common bush-tits (Psaltriparus minimus), varied thrushes (Izoreus nassius), ruby-crowned kinglets (Reg'ulus calendula.) , chipping sparrows (Spizella passer'trta), white-crowned sparrows (Zonotrichia leueophrys), and others. M. S. Ray (1911a) writes about large numbers of greentailed towhees (Chlorura chiorura) and Oregon juncos associating together among the dry meadowlands of late summer near the southern end of Lake Tahoe, Calif. J. A. Allen and William Brewster (1883) mention collecting an "Oregon Snow-bird" near Colorado Springs, Cob., Apr. 26, 1882, and collecting another the following day, both in the company of white-crowned sparrows.

In the lowland Puget Sound country of western Washington I have seen juncos and varied thrushes foraging in the same leaf litter and debris, but whether the larger thrushes ignored or merely tolerated the smaller juncos was not learned. The shyer thrushes always flushed before the juncos did.

Howard Twining (1940) says that in the high Sierra Nevada the juncos are the birds most commonly associated with the gray-crowned rosy finches (Leucostiete tephrocotis dawsoni). He writes:

On July 21, 1936, a male rosy finch was catching mayflies in the air above the north shore of Leuco Lake. A junco in juvenal plumage was performing almost identical actions. The two would leave the rocks at about the same time and bead for the same insect. As the rosy finch caught the insect, the junco would feint at it, then both would return to the rocks. This performance continued for about half an hour before the rosy finch left for the opposite side of the lake. It may be that the young juncos learn this method of catching insects from the rosy finches. On other occasions I saw flocks of young juncos mixed with rosy finches on the lake shore, catching insects in a similar way.

Lincoln Ellison (1934) saw juncos active about the seed beds of a forest tree nursery, and in cutover or partly cutover areas, but insofar as be could tell never bothered nearby seed beds in the dense virgin timber. He says: "The birds took seeds from all surfaces: duff (the layer of litter which covers the soil under the forest, in this case 1 to 2 inches thick and composed of dead leaves and twigs * * * ), mineral soil, and burnt mineral soil; but they scratched for seeds below the surface only on those beds covered with duff. They never scratched in a mineral soil bed, although there was invariably a wealth of seeds less than a quarter of an inch below the surface."

Seeds of western white pine were planted in the northwest corner of the different seed beds, and the certainty which the birds showed in flying from one northwest corner to another for their favorite seed indicated to Ellison "a surprisingly exact memory for places." Ellison concludes: Apparently [the juncos' intelligence] functions expertly within the limits of ordinary circumstances, as illustrated by their returning so accurately to the food source, and scratching among leaf-litter. Such stimuli and reactions are within the realm of the birds' experience. But when extraordinary conditions obtain: and the presence of forest seed buried in exposed soil is probably extraordinary, compared to its common presence on the soil surface or buried in duff: their intelligence does not turn those conditions to advantage.

Moreover, that they did not show any interest in the soil surfaces after the surface seed had been picked of! (although they returned for the attached seeds once the seedlings had broken ground), makes it seem probable that juncos do not detect such seed by smell, but by Right alone.

To study the winter society of the Oregon junco, Mrs. Winifred S. Sabine (1955) marked individual juncos by rubber cementing different combinations of trout-fly feathers to the top pair of the birds' tail feathers close to the body. Some members of the flock could not be caught, but enough were marked to simplify the problem and reduce the number of possible errors. She observed the juncos over a prolonged period as they frequented different feeding stations.

Her studies suggest the social activities of the winter society involve an orderly but rather complex type of flocking. She says (1956): "Although the migrant individuals which are to become permanent residents arrive irregularly over a period of several weeks they somehow manage to form themselves into distinct, stable winter flocks with mutually exclusive foraging territories."

Juncos in winter become birds with a definite routine. Each becomes a member of a flock, defined as an association of a group of juncos. The structure of the flock is formed with the earliest arrivals and with the use and occupancy of the foraging areas. Mrs. Sabine says this became obvious to the observer as soon as the first migrants were marked. New arrivals to the flock are integrated into the existing groups without difficulty. Individual birds of an established flock may visit or become temporarily attached to another established flock and are tolerated. Whether birds already established in the flock are dominant over new arrivals was not learned.

Voice: R. T. Peterson (1961) says, concisely: "Song, a loose musical trill on the same pitch. Note, a light smack; twittering note."

To my ear, the usual note between birds of a flock or between the two of a pair is a quiet "tsip." '1'o others, the sound is like a"sharp, metallic, kissing click," as T. G. Pearson (1939) describes the slatecolored junco call note. A. A. Saunders (193Gb) says to him the calls and songs of the slate-colored junco and the Oregon junco are alike.

The call note, once a person learns it, draws attention to a flock before the flock is seen. As an intruder, human or otherwise, gets closer to the flock and the members become alarmed, the call notes come more quickly, erratically, and are obvious expressions of the birds' alarm. Other birds or animals may be affected, according to Thane Riney (1951) who notes: "Sounds such as * * * erratic'thup' notes of the Oregon Junco elicit alarm reactions in deer."

H. H. Sheldon (1907) and Richard Hunt (1920) mention the Anna hummingbird (Ca lypte anna) having a sharp note like the junco; Hunt is more specific than Sheldon, calling it "a sharp smaciiing tip-note."

Peter Marler, Marcia Kreith, and Miwako Tamura (1962) made a series of experiments on song development in hand-raised Oregon juncos. They write: "The song of wild Oregon Juncos has several distinctive properties. It consists, with rare exceptions, of a trill of similar, repeated syllables. The length of the song, the number of constituent syllables, and the duration of those syllables vary relatively little. The fine structure of the syllables themselves shows great individual variability. Each individual has several song types, one of which may be given for long periods without interruption by another type.~~ None of the hand-raised jimeos had the opportunity to hear any songs or calls of wild juncos after being taken from the nest. They were exposed to songs of other species. Marler and his associates describe their experiments, which cannot be detailed in this account. In summary, they say:

Eight male Oregon Juncos were taken from the nest and raised by hand in varying degrees of acoustic isolation. Each developed several song types. In comparison with wild juncos, the songs of the experimental birds were somewhat longer, with fewer, longer syllables. They were more variable. However, there was appreciable overlap, so that each male had at least one "wild-type" song. Some abnormal songs developed from imitations of other species. In addition, the birds raised in a rich auditory environment had more song types and a more more elaborate syllable structure, derived not from imitations but from unspecific stimulation to improvise. Vocal inventiveness is established as a significant factor in the development of song in Oregon Juncos.

Field marks: Oregon juncos are sparrows with a dark-colored head, throat, and breast, called a "hood," and pinkish, yellowish, or light brown sides definitely separated from the "hood." The back is a shade of brown, never slate color. The outer tail feathers and underparts are white. This description includes the races montanus, .shufeldti, thurberi, ore gaines, and mearnsi.

The tone of brown on the back, the yellow, buff or pink on the sides, the tone of black on the head varying from gray to black, and their dimensions help the museum worker distinguish one race from another, but normally these characteristics are not to be relied upon for field identification. A. H. I\Iiller (1936) says: "I know of no one who dares claim ability to identify as to subspecies a living, fidgeting Oregon Junco that he may be banding. Identified specimens readily available for comparison can increase a person's accuracy in such a situation, but, even then, many errors will be made. * * * The most detailed study of specimens of some of these races enables one to ident.ify with accuracy only about eighty per cent of wintering birds. Obviously, birds of all these races should be designated merely as Oregon Juncos (J. ore ganus) in field and banding work."

The exception is the race mearnsi, formerly separated as the pinksided junco. Its sides are a rich pinkish cinnamon that often almost meets across the breast, and its hood is a clear gray.

Enemies: The survival rate of juncos in the wild tells us that its enemies are many. J. M. Linsdale (1949) and coworkers banded 233 Oregon juncos during an 11-year period beginning in the fall of 1937. After 1 year 20 were again trapped, 19 were trapped after 2 years, and retrapping success decreased until in the 8th year 1 bird living for that length of time was retrapped. Thus one junco lived at least 8 years after it was banded.

Loye Miller's (1952) imitations of pigmy owl (Glaucidium gnoma) calls at Saragossa Springs, San Bernardino Mountains, Calif., ,June 25, 1930, brought a quick response from a number of small birds including Oregon juncos. Charles Michael (1927) writes that after a pigmy owl robbed a downy woodpecker nest of a practically fullgrown nestling, a pair of juncos and other small birds gave distress cries and scolded the owl with no effect. All efforts by G. B. Castle (1937) failed to make a pigmy owl drop a junco from its talons.

Loye Miller (1952) observes that small passerine species do not react to the larger owls, either by sight or by sound. H. S. Fitch (1947) found the remains of a single Oregon junco among the 1,471 prey items he identified from 654 horned owl (Bubo virginianus) pellets.

A pair of Oregon juncos and several pairs of other small birds bad nests in the vicinity of a goshawk's (Accipiter gentiis) nest being studied by James B. and Ralph E. Dixon (1938). The Dixons write: "All of these birds had fear of the hawks * * * as shown by the fact that the only way we could tell the old hawk was approaching the nest was by the alarm notes these various birds sounded before the hawk appeared upon the nest."

Stomach analyses and personal observations of J. A. Munro (192gb) in British Columbia record the Oregon junco as being prey to the sharp-shinned hawk (Accipiter veloz) and the black pigeon hawk (Falco c. sucldeyi).

Cowbirds rarely parasitize the Oregon junco, for they seldom invade its usual nesting habitat. Ian McTaggart Cowan (1939) records one instance of the brown-headed cowbird (Molothrus ater obecurus) parasitizing the nest of an Oregon jun00 (J. o. montanus) at Tupper Lake, Peace River, British Columbia, May 20, 1938. The nest was later deserted. Friedmann (1963) reports another from the files of the British Columbia Nest Records Scheme, "a nest with 4 eggs of the junco and 1 of the cowbird, found 35 miles south of Vernon, Okanagan Lake, June 17, 1959."

Allen McIntosh (pers. comm. to Oscar M. Root) lists the following genera of parasites of the Oregon junco: Isospera, Dasypsyllus, Euschongcistia, Ornithomyja, Ricinus and Physostomum. R. 0. Malcomson (1960) specifically lists: Penenirmus mirinotatus, Ricinus hastatu.s, and R. pallidus. Oregon juncos were among the birds Don Bleitz ï (1958) found infected with foot pox, a viral infection called avian lymphomatosis.

John J. Williams (1900) tells of young june05 just out of the nest preyed upon by a "medium-sized chipmunk," not otherwise identified. From the efforts passerine birds make to drive chipmunks from the vicinity of nests, Williams assumes predation by chipmunks to be significant. Lyman Belding (1901), early day California ornithologist, did not agree, saying: "I have seen at least a hundred nests of the junco and can only remember one that was disturbed by bird or animal, the exception being a nest that contained four young which were killed by a gopher snake." Carl Sharsmith (1936) accuses the Belding ground squirrel (Citellus b. beldingi) of a carnivorous diet on occasion, and mentions it killing the junco.

J. M. Linsdale (1931) cites an example of the Oregon jun00 being poisoned by grain placed for rodents, in this instance hulled barley treated with thallium.

Certainly the environment must also be considered. Juncos nesting at high elevations in the mountains suffer from sudden or late season storms. A. M. Ingersoll (1913) tells of severe June storms in the Sierra Navada of California. lie writes: "All new snow that fell on June 23 [1912] melted away within forty-eight hours. Two nests that held eggs when discovered were later found to contain dead nestlings. Two nests held dented and cracked eggs after the snow. One nest and five young were destroyed by some mammal* * 'k." He adds that severe weather at a critical period in the life cycle increases chances of predation.

Fall: With family cares behind them, as fall approaches the juncos start gathering in flocks to forage together on the matured seeds that are plentiful and readily available in most of the coun tryside. I believe Oregon juncos that summer in the high mountains of our western states stay in the mountains until fall storms drive them down. With the first hard snow, which can come in early October in the high country, the juncos quickly move to lower elevations or leave altogether. The fall migration may involve long flights for many shufeldti and montanus. For many thurberi the fail migration is merely a withdrawal from the high Sierra to adjacent foothills and valleys.

Migrating juncos may be found in a wide variety of habitat: edges of chaparral, fence rows bordering farm and ranch lands, juniper (Juniperus utaJien~is or J. scopulorum) and pifion (Pinus edutis) woodland, sagebrush (Artemisia tridentata), city cemeteries, brush tangles, riparian plant growth along streams, suburban yards, orchards, and country roads with weed- or brush-covered rights-of-way. Here again, the "edge" effect appears to be overwhelmingly attractive to migrating juncos. They prefer habitats consisting of some tree or bush cover adjoining patches of open ground. Presence or absence of snow affects their movements. Juncos usually desert parts of the country subject to heavy or prolonged snowfall, except for a few individuals held by artificial feeding or those able to peck out a living around a barnyard.

Winter: Oregon juncos of the three races discussed in this account commonly occur in winter in the lower mountains of Colorado, New Mexico, and Arizona, on the great plains adjacent to the Rocky Mountains, the coastal and lowlands of Oregon, Washington, and California, the interior valleys and foothills of California, suitable places in the Nevada and Arizona deserts, similar locations in other western states, and at lower elevations within the breeding ranges of some of the forms as far north as British Columbia. Oregon juncos of different races become thoroughly intermingled in the winter flocks, and according to Winifred S. Sabine (1955) the same winter flock may contain both local and migrant individuals.

C. F. Batchelder (1885) writes about wintering juncos he saw near Las Vegas Hot Springs, San Miguel County, N. Mex., in December 1882:

Passing the various hot springs that come boiling to the surface at numerous points along the stream, a short walk up the cacon brings you to one of the openings where the retreating hills leave a level stretch of a few acres. Among the thick clumps of low scrub oaks that are scattered over it, or in the large patches of tall dead weeds, I was sure to find companies of Juncos (Juswo ore~arnua and J. canweps) busily searching the ground for fallen seeds. Of all the species that I met with, the Juncos were decidedly the most abundant. They were to be seen everywhere; it was hard to find a spot they did not like; but these were their favorite haunts. Among the pines on the hills, or in the thickets of willows down the river, they were in small parties, but here they were in large flocks. They moved about a good deal, straggling along one or two at a time, though occasionally a number would fly in a tolerably compact flock. They were shyer and more restless than J. h3emalis, and quicker in their motions.

Batchelder noticed no difference in the habits of the Oregon and gray-headed juncos. "They were always together in the same flocks, and seemed on the best of terms." Undoubtedly most of the wintering Oregon juncos he saw belonged to the race montonus, which Miller (1941b) considers the predominant "dark-headed" Oregon jun00 wintering in this region.

Grinnell and Wythe (1927) say of the race thurber-i: "Abundant winter visitant throughout the whole [San Franciscol Bay region. * * * Inhabit tree-covered areas, showing preference for conifers, but also affecting oaks and eucalyptus, where they forage in scattering companies either in the foliage or on the ground beneath or adjacently."

Jean M. Linsdale (1929) writes: "Some birds, for example juncos, regularly roost at night in crevices in road-cuts. The dark bare soil of roads and their banks is often freed of snow sooner in winter than vegetation-covered adjacent ground. Birds seek out such places." Many of the migratory Oregon juncos Linsdale (1949) and coworkers banded at the Hastings Reservation, Calif., were retrapped again in succeeding winters which suggests that individuals tend to winter in the same place year after year.

Montana Oregon Junco (J. o. montanus)

Range: British Columbia and Alberta south to northern Mexico and central Texas.

Breeding Range: The Montana Oregon jun00 breeds from central interior British Columbia (Hazelton district, near Takia Lake, McGregor River) and extreme western Alberta (Yellowhead Pass, Banif, Didsbury) south through interior British Columbia (east from crests of coast ranges) and eastern Washington (east of Cascade Range) to central and northeastern Oregon (Maury Mountains, Home), central western Idaho (Heath, Lardo), and northwestern Montana (near Florence, St. Marys Lake). (Breeds and hybridizes, sporadically, in parts of the breeding ranges of J. h. cismontanu.s and J. It. hyemali.~, north to Circle, Alaska, and Fort MoMurray, northern Alberta.)

Winter Range: Winters from southern British Columbia (Boundary Bay, Arrow Lake), western Montana (Fortine), Wyoming (Thermopolis, Guernsey), and South Dakota (Faulkton, Yankton) south to northern Baj a California (latitude 320 N.), northern Sonora (Caborca), Chihuahua (30 miles west of Mifiaca; Chihuahua), central Texas (Fort Clark, Austin, Gainesville), and eastern Kansas (Lawrence).

Casual records: Casual in winter or in migration to Banks Island (Sach's Harbour), Manitoba, (Aweme), Michigan (Marquette and Waslitenaw counties), Illinois (Waukegan, Ipswich), New Jersey (East Orange), Maryland (Laurel), North Carolina (Wake County), Arkansas (Winslow), Louisiana (Grand Isle), and southern Texas (San Antonio, Galveston Island).

Migration: D at a deal with the species as a whole. Early dates of spring arrival are: Montana: Libby, February 27 (median of 9 years, March 1). British Columbia: Departure Bay, March 4. Alaska: Sitka, March 6.

Late dates of spring departure are: New Mexico: Los Alamos, May 9 (median of 6 years, April 24). Arizona: Camp Verde, April 11; White River, April 8. Utah: Mexican Hat, May 1. Wyoming: Laramie, May 9 (average of 6 years, April 29). Nevada: Mercury, April 14. Washington: Tacoma, June 2.

Early dates of fall arrival are: Washington: Stevens Pass, August 23. Nevada: Mercury, October 7. California: Siskiyou and Yolo counties, September 27. Wyoming: Laramie, August 31 (average of 8 years, September 22). Utah: September 9. Arizona: near Prescott, September 8; south rim of Grand Canyon, September 22. New Mexico: Los Alamos, October 10 (median of 5 years, November 8).

Late dates of fall departure are: Alaska: Petersburg, October 8. British Columbia: Sumas Prairie, October 29. Montana: Libby, October21 (median of 9 years, September 18).

Egg dates: (Refer to the entire species) Alaska: 1 record, June 28. Alberta: 70 records, May16 toJuly 10;36 records,May 25 toJune 8.

Baja California: 9 records, May 14 to June 12.

British Columbia: 16 records, May 7 to July 8; 8 records, June 8 to June 23.

California: 164 records, March 20 to August 3; 82 records, May20 to June 12.

Montana: 6 records, May 5 to July 16. Oregon: 26 records, April 12 to July 15; 14 records, May 9 to June 1.

Shufeldt's Oregon Junco (J. o. shufeldti)
Range: Southwestern British Columbia to southern California, chiefly west of crest of Cascades.

Breeding Range: The Shufeldt's Oregon junco breeds from southwestern British Columbia (western slopes of coast ranges; intergrades with J. o. ore ganu.s on Vancouver Island) south, from the forests of the Cascade Range to the coast, through western Washington and western Oregon (to latitude 430 N.; intergrades with J. o. thurberi in Crater Lake area and the Rogue River Basin).

Winter Range: Winters at low elevations throughout the breeding range, south through California (south to Witch Creek, chiefly on coastal drainages), and southeast, sparsely, through eastern Washington, northern Idaho (Fort Sherman), Utah, and Colorado (Wray; northwest Baca County) to southern Arizona (Iluacliuca Mountains), southern New Mexico (Ancho, Las Cruces), central Chihuahua (Chihuahua), and western Texas (Brewster County).

Thurber's Oregon Junco (J. o. thurberi)
Range: Southern Oregon south to northern Baja California.

Breeding Range: The Thurber's Oregon junco breeds from southern Oregon (east to Hart Mountain) south through northern coastal California (Sonoma and Napa counties; intergrades with J. o. pinosus in Main County) and the interior mountains of California to Santa Barbara and San Diego counties (Laguna Mountains), and east to extreme west central Nevada (Galena Creek). (Hybridizes occasionally with J. c. caniceps in eastern Mono, Inyo, and San Bernardino counties, California, and along the southwestern border of Nevada.)

Winter Range: Winters at low elevations on or near the breeding grounds north to Rogue River, Oregon, and south throughout coastal, southern, and insular California to northern Baja California (latitude 300 N.); sparsely east to northern Sonora (Saric), eastern Arizona (Flagstaff, Fort Apache, Chiricahua Mountains), and southwestern New Mexico (Big Burro Mountains).

Contributed by JAMES H. PHELPS, JR.


Yellowstone National Park is a vast volcanic plateau, broken here and there by meadows, streams or lakes, and forested for the most part with lodgepole pine (Pinus contorta). The world famous geysers and related scenic phenomona sprinide the landscape and attract most of the attention, but a naturalist soon becomes aware of the abundant bird and animal life here, unafraid of man. Of course one soon notices the pink-sided Oregon juncos, so named for the rich pinkish cinnamon of their sides. I believe they are the most common birds in Yellowstone.

Yellowstone is indeed junco country. The mixture of meadow and pine affords much of the "edge" that all juncos favor. Most of the undergrowth is not dense, but is open and grass-covered beneath the pines.

I have seen pink-sided Oregon juncos equally at home in the isolated mountain ranges of southern Idaho, the high country near Mount Borah which is the highest peak in Idaho, the dense forests of aspen (Popuius tremutoides) and Douglas fir (Pseudoteuga taxijolia) near Grand Teton National Park, Wyo., and the high mountains of the Continental Divide along the Idaho: Montana border. Westward through the Sawtooth Mountains mearn8i intergrades with montanus and southward toward Utah and southern Wyoming this race hybridizes with J. c. caniceps, the gray-headed junco. In its habits I detect no significant differences from those of other related Oregon juncos.

A. H. Miller (1941b) says that mearnsi tolerates drier forests than skufeldti or montanus. Nevertheless an adequate ground cover of grasses and small flowering plants must be available during the nesting season. Wherever the ground cover is suitable, one is almost almost certain to find nesting j uncos, especially if there is water nearby. In the Bannock Range, Bannock County, Idaho, for instance, the slopes have much aspen interspersed with big-tooth sage (Artemesia triderd ala) and here and there a Douglas fir; each opening in the aspen or "edge" not too far from water supports a pair of nesting uncos.

On the other hand, L. B. MeQucen (pers. comm.) found june05 nesting abundantly in the Lost River Range, Custer County, Idaho, and not necessarily near water. The forest growth of this arid range is open and park-like, with many "edge" situations, and a thick ground cover of grasses under the trees. The forest is Douglas fir and at higher elevations Engelmann spruce (Picea engelmanni) and some limber pine (Pinusliexilis). Mountain mahogany (Cercocarpus ledijotius) is quite common on terraces and ridges not otherwise occupied by the conifers, and McQuecn observed juncos nesting in this cover. The use of mountain mahogany by mearnsi is not reported previously, although caniceps uses it often. McQueen found chipping sparrows abundant in the same habitat as the juncos. In 1960 in the Caribou Range, Bonneville County, Idaho, near the Wyoming border, I found chipping sparrows more common than juncos in the drier places in the forest, and seldom found juncos at any distance from water, nor far from the "edge" of the denser growth.

Spring: M. P. Skinner (1920) says of the june05 in Yellowstone:

When the juncos first arrive in the spring they appear at low elevations but soon move higher. Even so, they are often so early that they have to seek shelter about barns and other buildings. In March they are seen generally on the bare spots of ground under limber pines and Douglas firs. * * * The late storms of spring catch the Juncos, hut they are adept at seeking shelter about the barns, under sheds, and in potato cellars * * ~'ï At other times they take refuge in lodgepole pines under bunches of foliage covered by a canopy of snow, behind the snow caught on an overturned root, under firs, and even under sage bushes if nothing better offers. * * * The first arrivals appear suddenly in March, and they gradually increase in number until June 1.

The pink-sided juncos, as remarked previously, associate with other juncos in migration. In Idaho we first notice the junco flocks of mixed races in such places as cemeteries where plantings of conifers alternate with bare roadways and plot.s of grass, and along country roadsides, foothill trails, and the outskirts of towns. Our Idaho weather is notoriously unpredictable; winter may linger into May one year, and the next year spring comes in Feburary. Juncos accommodate themselves accordingly and forage or take shelter where opportunity offers.

When we find pink-sided juncos in late May among the aspens and Douglas firs of the mountains, all seem to be in separate pairs and no flocks are in evidence, On May 23, 1959 in the Bannock Mountains at 6,000 to 7,000 feet elevation Blame Lyon and I saw upwards of 15 pairs in the scrub aspen groves and along a small brook edging open mature Douglas fir growth. The birds were either foraging quietly in the litter or sitting quietly on aspen branches, and we saw no flocks, lone individuals, or conflicts between pairs. Each pair stayed in a certain vicinity that I took to be its territory and could not be flushed away. If one bird of a pair flew the other followed; neither flew far, perhaps no more than 60 feet, nor deserted the general area. One bird, presumed to be a female, was seen carrying several weed stems and a bit of dandelion in her bill; perhaps nest building had begun. M. P. Skinner (1925) says: "The Juncos pair off late in May, and by the middle of June have made their nests. * * ~ Nesting: J. C. Merrill (1881) writes: "A nest taken June 13 was near the top of a ridge connecting two peaks, at an elevation of 8,000 feet. The nest was under a shelving stone, one of many exposed by a land slide, and was in a little hollow dug out by the parents. The nest was rather large, but well and compactly built, composed externally of coarse dry grasses, with an inner lining of fine yellow straws and hairs of the mountain sheep." Edson Fichter (pers. comm.) tells me of a nest he found July 22, 1959, on the ground at the base of a lupine, made mostly of dried grasses. The nest was 2Y4 by 2% inches, outside diameter. M. P. Skinner (1920) describes several nests. One was made of pine needles and vegetable stems placed under a little bunch of blueberries; another similar nest was more in the open under a tall cluster of lupines. Skinner also tells about a nest "built of grasses and lined with fine material, placed seven feet from the ground on the back wall of a shallow formation cave at Mammoth (Hot Springs, Yellowstone National Park)."

I have found no proof that the pink-sided race raises more than one brood each year. M. P. Skinner (1920) says: "Young birds have been seen to fly as early as the end of June at the lower elevations; and as late as August 12 I have found them in the same stage at higher altitudes. I have not been able to determine whether the mountaineers are second broods or not."

Late nestings sometimes occur. Eliot Blackwelder (1916) found a junco nest in the mountains of western Wyoming Sept. 1, 1912 "on the ground among flowers and grasses in a straggling grove of spruce trees and at an elevation of 9700 feet above sea. It contained four newly hatched young birds." As 3 inches of snow fell that night and another snowstorm came 5 days later, Blackwelder doubted that the late nesting succeeded at this high elevation.

Eggs: Four eggs compose the normal see, J. C. Merrill (1881) collected a set of five. He gives the measurements 0f four in inches as 0.81 by 0.60, 0.80 by 0.59, 0.84 by 0.60, and 0.83 by 0.60; the fifth he broke in blowing. He describes them as follows: "The ground color of three of these eggs is a dull yellowish-white, marked with spots and blotches of light reddish-brown and with a few blotches of lavender. The spots are scattered over the entire surface of the eggs, but are largest and most numerous at the larger end. The ground color of the fourth egg, the largest one, is a rather greenishwhite."

Young: Edson Fichter (pers. comm.) watched both adults feed the four young in a nest he discovered July 22, 1959. The average time between visits was 4 minutes 12 seconds, and one parent or the other or both averaged 32 seconds at the nest during 10 visits. He adds that the adults removed the fecal sacs; they apparently ate some of the smaller ones, others they carried away and dropped at a distance.

The adults brought food each time by the same route; alighting in the same small conifer about 8 feet from the nest, they dropped to the ground 2 or 3 feet from the nest and approached it by essentially the same path. When Fichter returned to the nest the next day he found the herbaceous vegetation surrounding and above the nest had been disturbed, apparently by cattle, and the approach route was not as consistent as before. Fichter watched the birds on two occasions for more than an hour at a distance of 3% feet from the nest. The adults were unafraid and continued to feed their young in the nest.

A. A. Saunders (1910b) tells about hunting for junco nests in Jefferson County, Mont. He says "a pair of Pink-sided juncos * * * appeared, scolded me, flew about my bed and finally followed me out of the swamp where I had searcht in vain for nest or young. Later I found another spot where a pair of Juncos evidently had a nest or young and where I past several evenings in succession. I searcht this spot for three evenings before I finally found a single young bird. This bird was well feathered but unable to fly and I almost steppt on it before I found it. When I caught it and it called in distress the parents became fairly frantic and flew at my hed, and fluttered in front of me almost within reach."

M. P. Skinner (1920) reports a female fluttering away with a pretended broken wing when he discovered her nest with newly hatehed young.

Plumages: Richard R. Graber (1955) describes the juvenal plumage of J. o. mearnsi as follows: "Forehead, crown, and nape gray, profusely streaked with black. Back cinnamon brown, marked with heavy black streaks. Rump and upper tail coverts bufly gray, obscurely streaked with black. Outer pair of rectrices white, second from outer largely white. Others black, narrowly edged with gray. Remiges black, primaries and secondaries light gray-edged. Tertials edged broadly with dull pink. Lesser coverts gray, medians (blacktipped) buliy white. Greater coverts edged with buff, tipped with buify white. Two obscure buffy white wing bars. Lores black. Sides of head gray, rather obscurely spotted with black. Chin and throat grayish-white spotted and streaked with black. Chest, sides, and flanks tinged with buIT, streaked with black. Belly and crissum white, unmarked. Leg feathers gray and white."

Iris is brown; bill, flesh colored; feet, red-brown.

Hybridization and intergradation.: The pink-sided Oregon junco, now the race mearnsi, and the race montanus, formerly called Montana junco, meet and intergrade along a more or less north: south line from the Sawtooth Mountains of Idaho northward. Miller (1941b) says: "The meeting * * * occurs in a region where there are no barriers to distribution. At least in Idaho and in parts of Montana they meet in areas of nearly continuous forest habitat. There is more discontinuity within the range of mearnsi in northeastern Idaho than in the region of principal intergradation."

Hybridization between the pink-sided race of the Oregon junco and the white-winged junco (Junco aikeni) is rare although known. A. H. Miller (1941b) says "occasional mear'nsi wander eastward from their breeding range in the Big Horn Mountains of Montana and Wyoming to breed with aikerti." Breeding ranges of the two species are quite possibly separated by no more than 20 miles in this area and Miller considers it remarkable that more hybridization does not occur. He says: "Here are two forms, each representative of an extreme modification of a distinct rassenkreis, that remain peculiarly apart, considering their migratory habit, when in close proximity geographically in the breeding season."

Hybridization between J. o. rnearmsi and J. c. caniceps is discussed under the gray-headed junco, p. 1120.

Food: The food of this race is apparently similar to that of other juncos. One sees theni picking up weed seeds or plant seeds, or one presumes so from watching the birds feeding along roadsides or borders of trails, in neglected mountain cabin gardens, cemeteries, and the like. During breeding season I have seen pairs with young bring various small worms or insects, but have not identified the food further.

M. P. Skinner (1920) says that in the spring in Yellowstone the pink-sided Oregon juncos pick up grain and weed seed on bare slopes and go about barns with Cassin's finches for dropped oats. As the automobile has not eliminated the horse from Yellowstone and numbers of them are kept for packing into roadless areas or to rent as saddle horses, this observation of Skinner's should still be true.

Behavior: M. P. Skinner (1920) writes:

Our Juncos are very quick, sprightly and restless in their ways, hopping about on the ground and keeping up an almost continual cheepiug in loud tones, except when busy with parential duties or flitting through the pines with a flash of the white feathers in their tails. On the ground they move along with quick little jerks of the wings and tail at each hop. Sometimes they scuttle out of the road and under the nearby trees at one's approach. Usually, though, they are very tame and can be observed at close range * * *ï They are quite fearless of the Red-tail and Swainson hawks, even when those big birds are screaming in the same tree within a few feet of them.

Pink-sided Juncos are very sociable little birds, associating in spring and fall with Mountain Chickadees, Nuthatches, Tree Sparrows, and with their cousins the Intermediate [now called .1. o. menfenus] Juncos, in the evergreens. At other times they may be soon with Pine Siskins, White-crowned Sparrows, Chipping Sparrows, Kinglets, Audubon Warblers and Townsend Solitaires. They are often with the Robins and Bluebirds, with Vesper Sparrows in spring on the sage flats, and even with horned Larks and I.aucostictes on the bare spots. * * * The Pink-sided Junco usually progresses by a series of short flights from tree to tree, or from bush to bush. The flight has a peculiar, halting catch to it, due, no doubt, to the short and fast moving wings.

hawking for insects has been observed in this race. M. French Gilman (1935) saw a pink-sided Oregon junco, at Death Valley, Jnyo County, Calif., "jump several times into the air and catch insects." rrhis behavior has also been observed in oreganus and thurberi.

Voice: I do not know whether I should say a junco may sometimes sing for the pure joy of hearing his own voice. But a male I heard singing from the top of a chokecherry near Palisades Dam, Bonneville County, Idaho, one midsummer afternoon almost caused me to think so. The bird was wet, as wet as he could be, from a bath in a nearby backwater pool of a brook. He was perched, his head thrown back, vigorously shaking and fluffing his feathers dry, singing rapidly and repeatedly all the while. He paused only now and then to pick at a feather, and then sang again as I watched from an adjacent path for more than 10 minutes. His song was a whistling "ting-ing-inging-ing" repeated rapidly in series. The calls and notes of mearnsi are the same as those of others of the species insofar as I can determine.

Field marks: The pink-sided race may he distinguished in the field from the other Oregon junco races.

The description given by A. H. Miller (1936) is excellent: "The species has broad areas of rich pinkish cinnamon on the sides. The sides are never brownish or vinaceous as in many Shufeldt's Juncos. Pink-sided Juncos have gray, not sooty or blackish heads; sometimes the feathers are tipped with buff. The back is dull brown. No one of these charactors can be relied on solely, hut the combination of rich side color and clear gray head is highly dependable."

Enemies: Concerning survival in the wild, John N. and Eleanor Hough (pers. comm.) at Boulder, Cob., banded a total of 3,101 pinksided Oregon juncos from 1946 to 1958. These yielded 63 returns to their station after 1 year, 21 returns after 2 years, 14 returns after 3 years, 4 returns after 4 years, 2 returns after 5 years, and 1 return after 6 years.

Juncos in Yellowstone are attracted to the warmth of caves, which often become death traps from gases. Edgar A. Mearns (1903) notes: "In any hollow capable of holding the heavy gas (supposed t.o be carbon dioxide) f~t~l to animal life, dead birds were liable to be found, provided that the usual accompaniment of beat and moisture (from steam), and sulphurous odors (from emanating gases) were found. Most of the dead creatures were birds * * ~. The effect on bird bodies was to cause rapid decay, the flesh quickly disappearing, then the bones, and lastly, the feathers."

Mearns visited the "Stygian Cave" at Mammoth Hot Springs and other similar formations near by. Pink-sided Oregon juncos were always found, usually numbering more than other species. During the September 1902 autumn migration, Mearns says an unusually large number of birds perished and although he kept no records, "The largest number were pink-sided juncos."

Fall and Winter: About the first week in August the pink-sided juncos may be found in small flocks: usually birds of a single family, the parents with their offspring. As one hikes through the high country little groups of juncos fly ahead, alighting now and then in a Douglas fir, limber pine or aspen, then fly again as the hiking party disturbs them. But they do not fly far and do not really seem disturbed by human intrusion. I have a suspicion that after leaving the nest young juncos with their parents may make an upward migration among the mountains. Where earlier in the season I found nesting pairs the birds were gone, but in the higher country they seemed more plentiful than before.

Toward the latter part of August the flocks become larger as family joins family. In August 1960 near the Continental Divide in Clark County, Idaho, I encountered flocks of juncos in quantities beyond counting. Juncos were everywhere, in the open places between the scattered conifers of the hillsides, and among the upturned snags and brush of the shore of a small lake which shall remain nameless because of the huge trout my brother and I have taken there and want to take again.

Streaked young predominate in the late summer flocks, but the young juncos get their adult plumage within 2 or 3 months after hatching, and the October flocks are nearly all in adult plumage.

While many remain in the high country as long as possible, even as late as October and November, so far as is known birds of this race migrate completely, and none stays in its snow-covered mountain breeding home through the winter.

Fall migration is irregular as there are records of juncos migrating or at their winter homes while others are still in the vicinity of the breeding grounds. E. R. Warren (1916) says for the Elk Mountain region, Gunnison County, Cob., that the junco arrives "as early as September 24, and remains throughout October." H. S. Swarth (1908) says "first seen on October 18" in the Iluachuca Mountains of Arizona.

Pink-sided Oregon juncos winter commonly on the lower mountains and foothills of Colorado, New Mexico and Arizona. At Boulder, Boulder County, Cob., John N. and Eleanor Hough (pers. comm.) report October 11 as their earliest date in the fall and May 1 as their latest date for the years 1946 to 1958. In their opinion "it is certainly the most common winter junco" at Boulder.

Open or scattered woodland or forest, if that is tbe term, covers many of the mountains that are winter range for this junco. In southern Pima County, Ariz., A. R. Phillips (1933) says: "Found from the crests of the ridges to the canyon bottoms (of the Baboqulvari Mountains), but commonest in the live oak belt."

Pink-sided Oregon juncos are found associating in winter flocks with gray-headed juncos, white-winged juncos, and other Oregon junco races, principally montanus.

Range: Southeastern Alberta and southwestern Saskatchewan south to northern Sonora, central Chihuahua, and western Texas.

Breeding Range: The pink-sided Oregon junco breeds from southeastern Alberta (Eagle Butte) and southwestern Saskatchewan (Cypress Hills) south through central Montana (west to Belt River Canyon and Madison County, east to Big Horn Mountains), eastern Idaho (head of Pahsimeroi River, 20 miles northeast of Preston) and northwestern Wyoming (Teton and Wind River mountains). (Hybridization with J. c. canieeps occurs to southward in southern Idaho (Cassia County; Swan Lake), northern Utah (to Summit County and Uinta Mountains), and southern Wyoming (Rattlesnake and Casper mountains).)

Winter Range: Winters from northern Utah (Salt Lake Valley), northeastern Wyoming (Newcastle), and western and central Nebraska (Crawford, Johnstown, Red Cloud) south to northern Sonora (Sierra Carrizal), central Chihuahua (Bustillos), and western Texas (Hemphill and Brewster counties); casually west to southern California (Potholes, Ramona) and east to eastern Nebraska (Omaha).

Contributed by JAMES H. PHELPS, Jr.


This race breeds in the coastal islands and adjacent mainland of southeastern Alaska and British Columbia south to Princess Royal Island and Calvert Island. It is truly a coastal bird, and the pure form does not appear to go inland beyond the tidal inlets and lower river valleys. A number of other species have developed local races in this region of dense forest and heavy rainfall, with attendant pecularities in their life histories.

As the nominate race Townsend (1837) described originally, oreganu~ might be known as the Oregon junco, but Grinnell and Miller (1944) use the name northwestern Oregon junco. Theed Pearse of Vancouver Island (pers. comm.) who courteously allowed me to use his field notes, refers to it as the "red" junco. And indeed the backs of oreganu.s are markedly ruddier than those of the neighboring sltufeldti or montanu.s, and in some individuals are a bright, rich red-brown. Heads are black, darker on the average than in the adjacent races. The impression is that of an intensely colored junco, new looking, as though freshly washed from constant exposure to the incessant rains.

S. G. Jewett (1942), who visited different islands and the mainland of southeastern Alaska in the summer of 1941, comments on the irregular distribution of the race:

Common locally, but entirely or almost lacking in other apparently suitable areas. At Hobart Bay on June 23, during a half-day's hunt, only two adults were seen; and on June 25 several hours spent ashore revealed only one adult male, seen and heard singing. On June 28 at Olive Cove, Etolin Island, and June 28 and 29 at Wrangell the species was very common, many adults and striped young being seen. Almost an entire day spent ashore at Anan Creek revealed just one young junco. At hake, Kupreanof Island, both young and adults were common in the Indians' garden patches along the beach on July 4.

A. H. Miller (1941b) says:

Juncos in much of this area are sparsely distributed, probably because of the prevailing density of the forest and undergrowth. They occur chiefly along the borders of muskegs, meadows, streams, and beaches and in the occasional tracts of parklike timber, especially on the outer islands. Cedar (Thuja), Sitka Spruce (Picee sitcheasis), and western Hemlock (Tsu~a heterophylle) dominate the forest. There is an abundance of epiphytic growth. No other junco experiences more overcast weather and precipitation during the breeding season. The sun may be obscured from view for many days in succession. Nearly all the occupied region is below 2000 feet in elevation. Some of the higher mountains of the islands may afford suitable junco habitat at timber line.

From Comox, Vancouver Island, Theed Pearse writes me: "Here the Oregon jimco is practically a winter bird and nests at the higher levels."

Spring: The earliest arrival at Windfall Harbor, Admiralty Island, in 1907 according to Grinnell (l909a~, was April 19. A. M. Bailey (1927) says the birds are seen "the year round" in southeastern Alaska, but intimates they are less plentiful in winter and become more common with the arrival of spring: "A few were noted during January near Juneau * * * . My notes read 'Wrangell February 27, Kupreanof ?vlarch 4, Craig the l2th,: fcw noted.' A few were seen in Hooniah Sound throughout our stay May 7-24, but not daily. By May 31, however, the Juncos were common about Juneau and a female was seen building her nest * * *

Theed Pearse (MS.) records juncos of this race singing at Courtenay and Comox, Vancouv& Island, Feb. 9,1928, Feb. 11, 1922, Feb. 24, 1917, and notes on Mar. 30, 1919, that "white outer tail feathers play a prominent part in display."

Nesting: A. M. Bailey (1927) writes: "Saw a female Junco carrying nesting material and watched her for some time. She made several trips, always to one spot which seemed to possess material to her liking. The male did not appear for some time, and then came hopping nonchalantly along. On the appearance of the little female, he promptly pursued her, she protesting vigorously. I noticed, however, that she was very careful not to drop a wisp of the nest lining." The nest, Bailey says, was on the ground, in a thick clump of hemlocks, and tucked back under a carpet of moss, and had four eggs. He tells about another nest completed by a pair, then abandoned,and a second nest built in the moss a few feet away. The nest had five eggs as did another he observed. I have been unable to learn whether sets of five eggs are more common for this race than for others, where the usual number is four.

Bailey mentions one female building her nest with little if any assistance from the male, and a pair that built two nests together. The young in one of the nests he observed "were nearly ready to leave the nest July 4." No information indicates that incubation or fledgling periods differ from those of the other races, nor that ore ganu.s rears more than one brood annually.

Plumages: Richard R. Graber (1955) gives the juvenal plumage for this race as follows: "Forehead and crown streaked profusely, brown and dark brown without great contrast. Nape concolor with crown medially, grayer laterally. Back russet, streaked with blackish brown. Rump gray brown, upper tail coverts dark brown, both obscurely streaked with darker brown. Tail largely blackish brown, outer two pairs of rectrices white. Remiges blackish; outer primaries edged with white, secondaries and tertials with rust. Coverts edged with rusty brown, greater coverts tipped with whitish or buffy white (two narrow wing bars). Lores dusky. Auriciilars drab gray brown Post-auriculars like nape. Sub-auriculars, chin, and throat heavily streaked, blackish and huffy white. Chest and sides strongly tinged with bufTy, other underparts white or huffy white. Chest, sides, and flanks heavily streaked with dark (blackish) brown. Leg feathers brown. Crissum unstreaked."

Of the adult plumage Miller (1941b) notes:

Buff feather tips are less prevalent than in montanus and shufeidli, yet some extremes occur. * * * The buff tippings of the nape are usually more ruddy in this race. * * * There are no instances of white on the wings or of red on the pileum deeper than the feather tips.

The only aspect of wear that is different from that in other Oregon juncos relates to the hack; there the darker red brightens very materially from the Prout's brown of fresh feathers, hut never becomes yellowish or whitish along the edges of the feathers sa in shufeldti, thurberi, and montanus.

Hybridization and intergradation: Miller (194 ib) states that "in south-central Yukon Territory, south of Lake Marsh, ore ganus characters in the hyemalis population become frequent." He also notes that montanus and ore ganus intergrade in the sector of the coastal region from Queen Charlotte Sound to the Portland Canal, British Columbia. Apparently the oreganus juncos of the mainland tidewater and coast intergrade with montanus juncos of the mountains and valleys of the interior and higher elevations whenever their habitats adjoin.

Food: In his field notes for Oct. 13, 1940, Theed Pearse records: "Juncos are omnivorous; some cooked salmon put out for jays and not attracting them was eaten by the juncos."

Behavior: A. M. Bailey (1927) mentions juncos jumping in the air "like so many flycatchers" to take insects on the wing. A similar observation is recorded for the race thurberi in the main species account.

George Willett (1921a) notes that Oregon juncos on the forested coastal islands of southeastern Alaska apparently move after nesting during the latter part of August, up the sides of the mountains and away from sea level, together with the varied thrush (Izoreus naevius) and fox sparrow (Passerella iliaca): "the three species being frequently found in close proximity in the woods on the mountain sides from about 1000 feet altitude to timber line. As the weather becomes cooler they work back down the mountains to the shore."

Field marks: The general impression is that of a very intensely colored Oregon junco, with black head, red-brown or ruddy back, and white outer tail feathers. This race cannot be separated with certainty in the field from adjacent races of the Oregon junco that it may associate with in winter or when migrating.

Fall and Winter: The race is only partially migratory. Because little snow falls under ordinary circumstances at the lower elevations of its breeding range, juncos can forage successfully all winter. Nevertheless many migrate southward and winter in western Oregon, western Washington, and northern coastal California. While there are sufficient records inland to indicate that some migrate elsewhere, it is rare outside of these limits.

Grinnell and Miller (1944) say of this race in winter in California: "Habitat: Open forest understory, woodlands, edges of chaparral, and fence rows. As with all wintering juncos, this race may be found in a wide varity of habitats; but there must always be some tree or bush cover and some patches of open ground. J. o. oreganus seems to favor the humid forests, shaded ground and denser brushland more than do related subspecies of juncos; this preference is reflected in the concentration of the winter population in the northern coastal district of the State."

Range: Southeastern Alaska and British Columbia to central and coastal California.

Breeding Range: The northwestern Oregon junco breeds in coastal districts of southeastern Alaska (Yakutat Bay, Dall Island, Ketchikan) and British Columbia (Queen Charlotte Islands, Fort Simpson, south to Calvert Island).

Winter Range: Winters from southeastern Alaska (from Juneau) south through coastal British Columbia (east to Okanagan Lake), western Washington, and western Oregon to central California (Monterey County); sparsely through coastal California to northern Baja California (Santa Eulalia); casually southeast to southern Idaho (Naanpa), Nevada (Carson City, Ruby Lake), central Colorado (Denver), southern Arizona (Huachuca Mountains), and western New Mexico (Fort Bayard).

Contributed by JAMES H. PHELPS, JR


The unique central coast of California from San Francisco southward through Santa Cruz and Monterey, with its marked boreal climatic influences, has a local race of the Oregon junco. L. M. Loomis (1893), who described it, comments: "The fact that a Junco should be found breeding at the sea level so far south in California is very significant and in itself is enough to suggest the existence at least of a local race."

The region is characterized by moderate temperatures, frequent fogs, high humidity, and high winds along the ocean shore and exposed headlands. Grinnell and Linadale (1936) note: "The concentration of rain in the mid-winter months and its aimost complete absence in summer are major factors in the composition of the fauna and in the seasonal behavior of the animals." Concerning the fog they add: "At inland localities it contains so little moisture that objects moving through it often remain dry, but at Point Lobos even stationary objects, the trees, bushes, grass and even the ground, are often dripping wet."

Loomis named the race for Point Pinos where he first saw his birds and recognized that they were different. Conditions at Point Pinos are much like those at 1~oint Lobos. He (1S94) writes: "The mean temperature (according to local information) is about 6O~ F. during each of the summer months. Heavy fogs, that almost amount to rain, are also frequent during summer."

It is now known of course that Point Pinos juncos are not confined to the coastal forest, the "humid coast belt," but inhabit a great variety of situations, more so thaii other races of Oregon junco. They nest in a number of plant associations, not all coniferous. Grinnell and Miller (1944) say: Almost any forcst or woodland, including plantings of eucalyptus, sufflccs if it affords shade and ground cover that remains green throughout the summer. Usoally, perhaps always, water may he ohtained, either from surface streams or from fog-drenched foliage, in summer, this race is more tolerant of low zonal conditions thau is thurberi. Yet in its typical ground-foraging activity, presence of shade is just as much an essential factor. Usage of trees f or nesting is more prey: alent in this race than in others, hut the bird is nevertheless predominantly a ground nester.

Summarizing information from both A. II. Miller (1941b) and Grinnell and Miller (1944), the following plant associations meet the requirements of these juncos for nesting: moist redwood (Sequoia 8ernpervrens) forests in canyons; dense, though comparatively arid, California live oak (Quercus agrijolia) woodland; Monterey pine (Pinus radiata) , pricklecone pine (P. muricata) , and Monterey cypress (Cypressus macrocarpa) forests; yellow pine (P. ponderosa) and Douglas fir (Pseudotsuga taziJolia) forests; black oak (Q. kdlloggii), goldencup oak (Q. chrysolepis), and madroito (Arbutus mertziesii) woodlands; and arid digger pine (P. sabimiana) and Coulter pine (P. coulteri) "forests."

Point Pinos juncos do not migrate, although as H. B. Kaeding (1899) comments "as soon as the young are fledged the birds wander." There are no records at other seasons more than a few miles from the breeding range. However in recent years the Point Pinos juncos have extended their range and now breed in city parks and suitable residential areas of San Francisco Bay cities. Amelia S. Allen (1933) believes that extensive plantings of Monterey pines and other changes in the landscape, all done by man, are responsible. She (1943) also writes: It has been interesting to follow the spread of this junco in Berkeley. Before 1914 I had worked out the earliest dates of arrival for the Oregon Junco as Octoher 17 and the latest date of departure as April 10. Between 1914 and the date of the finding of the first nest, I had several records of single birds found singing inter than April 10 * * * On May 15, 1917, on the grounds of the Claremont Country CIch, 1 found the first nest reported in the San Francisco Bay region. By 1918 the local Berkeley birds had spread from their center on the campus to our hillside. Thia junco is now one of the common breeding birds of the area.

Another possible extension is reported by A. II. TvIiller (1945), who writes concerning the collections of the late 0. P. Silliman: "Two juncos * * * taken on July 7 and 8, 1937, at San Ardo, 450 feet, Monterey County, seem to indicate a breeding station for the race pinosus in the floor of the middle section of the Salinas Valley; one of the birds was molting. This is an unexpected local extension of the breeding range of this race which nests commonly at higher elevations in the mountains on either side of the valley."

Spring: In the mild climate of this part of California the breeding activities of the resident june05 are much advanced over those of the migrants present as winter visitants. The migratory races face spring flights of varying lengths; the resident juncos are near or actually on their breeding territories, and they behave accordingly.

Grinnell and Linsdale (1936) say: "We soon noticed on our winter trips that the resident birds seemed to be paired and the flocks seemed to be made up mainly of winter visitants and transients. The pairs were stationed on the type of ground where the birds later nested and they behaved often as if they were defending the sites against invasion. This situation was detected a.s early as December 119341 and it was plainly evident by the middle of February [19351."

In the experiments Albert Wolfson (1942) made on resident and migrant Oregon june08, mentioned in the main species account (page 1053), the resident juncos heused were pinosus. He found that although the different races flock together in winter in the same environment, the residents and migrants show marked differences in their gonadal cycles: "The testes of the resi~lents recrudesce earlier and at a faster rate than those of the migrants." Because of the energy migrants need for their spring flights: "In the migrants there is a heavy deposition of subcutaneous and intraperitoneal fat at the time of migration. The residents show no such deposition of fat."

Grinnell and Linsdale (1936) write: "Definite evidence of the beginning of the nesting season was seen on March 20 119351, a cold, cloudy day when a pair in copulation was seen in a patch of monkey flower on the floor of the pine woods." They continue: "On March 24 [1935] nearly all of the juneos seen were definitely in pairs and scattered mainly through the pine woods. The birds spent most of the time sitting quietly on perches in the trees. They fed largely in the trees, but about one-third of the time on the ground. Occasionally one or both birds of a pair would indicate excitement by spreading the tail and trilling, but this was followed by another quiet period. There was some singing: often 3 or 4 birds could be heard at once in different directions, but the locations of singing birds were hard to determine."

Nesting: Though they are primarily ground nesters, as are the other races of Oregon junco, the Point Pinos Oregon juncos occasionally build their nests elsewhere. Grinnell and Wythe (1927) say: "Nests may be situated not only on the ground but on branches of cypress trees, and occasionally in appropriate recesses about the eaves of buildings."

H. 0. Jenkins (1906) describes a typical nest with four eggs, incubation partial, he found June 26, 1905 at Big Creek, Monterey County; it was placed on the ground in the redwood forest partially hidden by trailing roots, grasses, and blackberry vines. Joseph Dixon (1924) discovered a junco nest on the ground well concealed in a dense mat of ivy on the University of California campus at Berkeley, Mar: 16, 1923.

Milton S. Ray (1911b) writes:

Our first nest jMar. 27, 1910, San Mateo Countyj * * * was a strange departure from all previously recorded nest situations being placed 8 feet up in a Monterey cypress where it was well hidden in a thick clomp of foliage. The nest, a well built structure consisting almost entirely of pine needles, contained four eggs in which incubation had begun. A second nest of the junco was found 16 feet up in the Monterey cypress in an open situation well out on the limb, and contained fresh eggs. This nest is even a better built structure than the first one found. It is a very compact affair of pine needles, roots, grasses and weed stems and well lined with various animal hair.

Ray (1919) quotes Chase Littlejohn's observations for the 1918 season at Redwood City, a suburban town south of San Francisco: "I know of three junco nests being found about Redwood City last summer. These were built about the eaves of occupied dwellings and in one case inside the attic and all of them were of pinosus."

Grinnell and Linsdale (1936) give data on 24 nests they discovered at Point Lobos. Six of the 24 were built in trees, either pine (Pinus radiata) or cypress (C'upressus macrocarpa). They write: "The first nest was one foot north of the base of one of four small live oak trees. It was at the north margin of an opening in the pines. Both birds of the pair were carrying material to the nest, on March 24 [1935], when it was nearly ready for the lining." This is one of the few reports of both birds of a pair working on a nest. In other races the female usually does the bulk of the nest building. Their first nest with eggs was found Apr. 8, 1935 as was the first nest with small young. The last nest of the season with eggs not yet deposited was found May 20 "7 feet up in pine" and "nearly complete." They remark "young about to leave on June 20."

The dates of nesting for San Francisco and vicinity are about the same. Grinnell and Wythe (1927) say: "Nesting begins by the last of March and continues through May."

The only suggestion that Point Pinos Oregon juncos have more than one brood is the remark by Grinnell and Linsdale (1936) that: "In one instance a nest was built in a tree, apparently by the same pair which had just brought off a brood from a nest on the ground."

Eggs: ll. B. Kaeding (1899) writes that the nest and eggs of the Point Pinos Oregon junco apparently do not differ from those of the Sierra or Thurber juncos. The normal clutch of eggs is four, but one of the 24 nests found by Grinnell and Linsdale (1936) at Point Lobos had a set of five. Another had three normal-sized eggs and one not more than half the normal size; at a later date three young were seen in the nest.

Young: Joseph Dixon (1924) writes that the nest he found was completed and the first egg laid on Mar. 23, 1923: "At 6 o'clock on the evening of March 26 the nest contained four eggs which the female had begun to incubate.

"At 9 A.M. on April 9 the nest contained two eggs and two young which had hatched since the previous evening. Only two out of the four eggs hatched. The fledglings left the nest on April 16 when only seven days old. They were not at that time able to fly, but scrambled about readily beneath the tangled ivy and eluded my grasp easily." Dixon's dates indicate about 13 days were needed for incubation. He attributed the early departure of the young to numerous ''Argentine ants'' in the nest.

Plumages: In his original description Loomis (1893) gives the following characteristics: "Most nearly like * * * thurberi, but throat, jugulum, and fore breast slate-gray, varying to dark slategray, and upper portions of head and neck slate-gray, varying to blackish slate; bill averaging broader and longer."

hybridization and intergradation.: Grinnell and Miller (1944) say this race: "Jntergrades with J. o. thurberi in Main and San Luis Obispo * * * counties." Miller (1941b) notes the Mann woodlands, across the Golden Gate north of San Francisco, are separated from the southern limits of thurberi breeding range along the Russian River in Sonoma County by about 20 miles of countryside not suitable for breeding juncos, and: "A similar gap separates the breeding areas north and south of San Francisco Bay. The isolation afforded does not appear great, but in resident populations it means much more than in migratory races."

The intergrades between pinosus and thurberi show mixing or blending of head, side and back color, and intermediacy in tail pattern and measurements.

Food: The remarks in the main species account quoting the work of F. B. L. Beal (1910) on food habits of juncos in California apply in great part to this race. Beal writes: "It would be better to treat the * * * races separately, but as many of the stomachs were collected before the races were recognized, their exact identity is unknown.~~ Behavior: Apart from the nesting season when the birds are in pairs, Point Pinos Oregon juncos associate together in small flocks. These are quite possibly family groups. Grinnell and Linsdale (1936) note the juncos on the Point Lobos reserve in small flocks or groups of four or five, six or more, 12 or more, and other such numbers. While they mention a "flock of about 25" and a "scattered flock of 50," most of their references are to small flocks. They cite examples of small flocks or groups feeding at the margin of a roadway or meadow, scattered on moist ground covered with fine litter in shade beneath cypresses, or feeding on ground close to and beneath a pile of brush at the edge of the pine woods. They add: "From these and other records it was apparent that factors prominently important in accounting for the occurrence of juncos [at Point Lobos] were the open nature of the ground, the large proportion of the area in which shade was afforded, and the trees and tall shrubs which provided refuge, roosting places, and singing perches." In the San Francisco Bay region, Grinnell and Wythe (1927) state this form: "Forages and nests in groves of planted evergreen trees and native redwoods; comes familiarly about dwellings." A. H. Miller (pers. comm.) writes me: "Pinosus does tend to stay on or very close to its nesting territory, and I have never seen it form very large flocks, but it will mix in with wintering flocks of other races of Oregon juncos here locally and certainly is not as solitary and non-flocking as is Junco phaeomot'us."

Concerning the habit of the Point Pinos Oregon juncos of nesting in trees Grinnell and Linsdale (1936) remark: "It was noted at nearly all the nests in trees that the adults showed less concern and less reluctance to go to the nest when a person stood near the site than did most of the ground-nesting pairs." Regarding interspecific behavior, they report an instance of a junco pursuing two pileoated warblers (Wilsonia pwsilla pileolata) and of a male junco driving a green-backed goldfinch (Spinus psaltria hesperopAilua) from the bough holding a nest with young.

Reports of juncos catching insects in the air or flycatching have been reported for other races. Grinnell and Linsdale (1936) write: "A junco was seen several times on May 1, catching insects in the air. Again on October 12, in the evening, several were flycatching for large termites."

W. Otto Emerson (1905) had to have a large eucalyptus treeï removed that had gi~own on his place for many years. At dusk he noticed many small birds, among them Point Pinos Oregon juncos, Audubon warblers (Dendroica auduboni), and white-crowned sparrows (Zonotrichia 1. nuttalli) flying about the barn, near the place the tree had grown before being cut, in great confusion. He writes: "They were coming in from all directions and would fly to where they had been used to roosting, but their lodging house was gone. They came by fours and more, hovering in mid-air, and fluttered about ia circles, then alighted on the barn * * *ï Many dodged down into the cypress hedge in front of the barn, keeping up short flights to the fallen tree as it lay in the road."

Voice: Calls and notes of pinosus are no different from those of others of the species insofar as I can determine.

Field marks: During the breeding season when no other Oregon juneos are present, field identification of the race yinosus is no problem. During the winter when other races of Oregon jun00 and Point Pinos juncos associate together, separating them is most difficult. Mrs. Amelia S. Allen (1933) says the Sierra junco, the race thurberi, is "not accurately distinguishable from the resident Point Pinos junco in the field." The numerous citations in the literature for both races in the San Francisco Bay region indicate some of the confusion. While the Point Pinos juncos have ruddier backs than the Sierra race and paler sides and heads, there is so much overlapping that only the most experienced field observers can state with any accuracy whether or not ~imosus is present in a mixed winter flock.

Enemies: R. F. Johnston (1960) records this race as host to the brown-headed cowbird. On Aug. 14, 1958, on the University of Calif ornia campus at Berkeley, he ~vatched for one-hall hour an adult female Oregon jun00 "repeatedly feed an almost fully grown, juvenalplumaged" cowbird.

Fall and winter: Miller (1941b) writes: "There are no records of pinosus in the fall or winter for localities more than a few miles from the habitats occupied in the breeding season. Definition of breeding areas thus indicates geographic distribution throughout the year."

He explains the absence of migration by the fact that few places in its range have a severe winter climate or any snowfall. He notes: "In winter, flocks may wander into open country where there is limited cover in the form of isolated clumps of trees. They occur often in orchards in the open valleys. Many are known to stay on the breeding territory in small flocks, possibly family groups."

Range: The Point Pinos Oregon jun00 is resident in the coastal hills and mountains of central California from Golden Gate and Carquinez Straits to southern San B.enito and Monterey counties (intergrades with J. o. thurberi in San Luis Obispo County).

Contributed by JAMES H. PHELPS, JR.


A man, identified only as an American named Hanson and subsequently murdered, once "tried his hand" at ranching high in the mountains of an obscure corner of Mexico. His name came to be associated, not always with accuracy, with the mountains, the ranch, a shallow lake, and a race of Oregon junco.

Harry C. Oberholser (1919a) named the race pontilis from juncos E. W. Nelson and E. A. Goldman collected in the Sierra Ju~rez (or Hanson Laguna Mountains) in 1905, noting: "It has the very pale pinkish sides of Junco ore ga~nus townsendi, but in the color of both head and back is almost exactly intermediate between" thurberi and tOiii'nsendi.

Contrary to Dwight's (1918) opinion that townsendi was a subspecies of Junco mearnsi, at that time considered a separate species, Oberholser proposed: "The study of these specimens and their relationships with the two eontig~ious forms shows clearly that Junco oreganus pontilis directly connects Junco townsendi with Junco ore ganus thur6eri, and that, therefore, the former must be a subspecies of the latter."

The race is resident and isolated in the nesting season from other juncos, in a single mountain range in the northern part of the Sierra Ju4rez-San Pedro MArtir mountain group of northern Lower California, Mexico. According to E. W. Nelson (1922) the Sierra JuArez is a single main ridge about 4,000 feet in elevation at the United States: Mexican border, rising to about 6,000 feet near Hanson Laguna (or Hanson's Lagoon), and decreasing to 4,000 or 5,000 feet at the southern end. The top of the northern part of the Sierra near the type locality of the race at El Rayo (formerly Hanson's Ranch) is no more than three or four miles across. The summit is a long, narrow, rolling plateau, broken by many knolls, ridges of granite, and piled masses of huge, smooth granite boulders, 50 to 300 feet high, between which are small mountain basins or parks and flats. The east face is quite abrupt; the west slope is rolling or undulating.

According to Nelson's (1922) estimate of the area of suitable conimercial timber, the available junco habitat along the crest of the Sierra Ju~rez is no more than 37 miles long by 3 miles wide. A. H. Miller (1941b) says: "Mr. Laurence lvi. Huey estimates that there is about 20 to 25 miles of scattered Transition Zone forest along this section of the Sierra at elevations between 5000 and 6000 feet.

Granitic outcrops are frequent and, in combination with exposure, are causes of the interdigitation of tracts of Upper Sonoran forest. The junco habitat in summer is the parklike yellow pine forest."

The pontilis population is separated in the nesting season from the nearest thurbeti to the north in San Diego County, Calif., by possibly 40 miles of unsuitable territory; the mountains are not high enough for pine forest from 15 or 20 miles north of the boundary to 15 or 20 miles south. Similarly the decrease in height of the mountains north and south of San Matias pass separates pontilis from townsendi in San Pedro M~rtir Mountain, the gap being 40 miles or perhaps less.

I have found nothing in the literature or by correspondence to indicate any differences between pontilis and other Oregon juncos in habits, behavior, nesting, eggs, or other phases of life history.

Joseph Grinnell (1928b) remarks that these juncos are "apparently not very common." Perhaps the explanation comes from E. W. Nelson (1922), who says: "Water is scarce all along the summit of the Sierra Ju~rez. The Hanson Lagoon (or Laguna Hanson), about half a mile across, is the largest of several small shallow lakelets lying on the highest part of the range northeast of Ensenada. A few very small streams rise at the heads of gulches near the top of the west slope and flow down to the lower border of the foothills * * Water is still scarcer on the higher parts of the east slope than on the west * * ~" Fall and Winter: Other races of Oregon junco, particularly thurberi and to a lesser extent shufeldti, winter within the range of pontilis. Miller (1941b) says: "The fall-taken specimens all come from these [Sierra Ju~rez] mountains. One bird was taken at an elevation of 4200 feet at Los Pozos about 30 miles north of Laguna Hanson, October 31. There is no evidence of migration, therefore. There may be some tendency, as in toumsendi, to descend to lower levels in winter."

Range: The Hanson Laguna Oregon junco is resident in the Sierra Ju~rez (Los Pozos, Laguna Hanson) in northern Baja California.

Contributed by JAMES H. PHELPS, JR.


The isolated, even to this day, Sierra San Pedro M4rtir of Baja California, Mexico, is the home of the southernmost member of the Oregon junco complex. The San Pedro M~rtir is the main and highest mountain mass in the northern section of the peninsula, lying between lat. 3QO3gI and 3lo10~ N. Alfred W. Anthony (1889), who described the race, says: "The region embraces a series of small ranges which rise from an elevated me&t, having a mean elevation of about 8,000 feet, and an extent of 60 by 20 miles. In these mountains are born the only streams that this part of the peninsula affords, and an abundance of pine timber is found throughout the region." Anthony adds: "J. toumsendi is probably the most abundant bird to be found in the timbered parts of the San Pedro Mountain, and is, I think, resident * * *,' Nesting: A1I the information we have about nesting of this race comes from A. W. Anthony (1890, 1893). All the nests he (1890) found were made of soft, dry grasses and lined with finer grass stems and the hair of the mule deer. One nest was "very artfully concealed behind a thick bunch of grass and under the overhanging edge of a large granite boulder." Another "was sunken to the level of the ground, apparently in a cow track, and well hidden in the tall grass on the edge of a running stream." In another year he (1893) states the juncos were building nests upon his arrival in the pines May 5, but he found no eggs until May 10. One "nest was in an old woodpecker's hole in a large pine that had been blown down, with its top resting on a big boulder. The hole which was about six feet from the ground was on the under side of the trunk and the nest about on a level with the opening; it was composed of dry grasses and lined with deer hair. A nest * * * was found on May 26 in a hole in a rotten stub about ten feet from the ground * * *~" He mentions a number of nests, under logs, boulders, and ot.her locations, that he left in the hope of collecting full sets, but which be found destroyed when he returned. Apparently the open nature of the forest without much ground vegetation partly explains the numbers of junco nests he found in locations not considered usual for the species elsewhere.

Eggs: Three eggs, without exception, constitute a set according to A. W. Anthony (1890). The three sets he reports upon in detail vary widely in their markings. A set be collected May 1, 1889, has a faint bluish-white ground color, profusely and variously spotted with lilac and raw umber; the eggs measure 20 by 15; 19 by 15; and 18 by 15, in millimeters. In another set taken May 5, 1889, the eggs 'I are uniformly bluish-white with a few of the faintest minute specks of burnt umber, on the large end, which are not at first noticeable.

* * * Measurements 20X15; 19X15; 20X15 mm." In the third set taken May 6, 1889, two eggs "are marked with small spots of pale fawn color, with a few small spots and lines of burnt umber collected about the large end." The third egg shows a faint greenish wash and is also the most heavily marked of the set, being heavily blotched on the larger end with pale lilac, the lilac extending in small flecks over the entire shell. Measurements in millimeters are 19 by 15, 19 by 15 and 19 by 14.

Plumages: Anthony (1889) describes the plumage of his type male as: "Head, neck and breast all around, clear, slatey, gray, much darker than J. annectene, but paler than J. hyemalis oregonus. Lores and crown, blackish. Dorsal and interscapular region, ashy brown, in some specimens scarcely differing from the head and neck. Rump and upper tail coverts ashy, first two lateral rectrices pure white, the rest blackish with ashy edges. Wings, blackish, primaries and secondaries with grey edges. Sides, pink. Belly and inder tail coverts, white. Mandibles and feet, flesh color. Iris, hazle."

Of a female cotype he writes: "Head, neck and chest all around, uniform dark gray, lighter than male. Lores, blackish. Dorsal patch, very faintly defined. Rump, clear ashy. Pink of sides, paler and less extensive than male. Lower parts white. Outer three rectrices white, third edged with dusky. Iris, hazle."

Field marks: Townsend's Oregon junco is a gray-headed junco with narrow "pink" sides resident in the San Pedro M~rtir of Lower California. It shows some resemblance to the pink-sided junco, the race mearnsi of the Rocky Mountains, but the color area in the sides of towsendi is narrower. There are no records of mearnsi taken within the range of townsendi, although other Oregon junco races, thurberi and to some extent shufeldti, winter that far south and mixed flocks may be expected.

Fall and Winter: Joseph Grinnell (1928b) says: "Common resident on the Sierra San Pedro M~rtir. * * * A slight scattering in the fall carries a few individuals to somewhat lower levels close by." Weather conditions must play a part in flights to lower levels when such occur. E. W. Nelson (1922) writes: "The summits of these mountains are never covered with snow during the entire winter but usually only for periods of from two to four weeks. In midwinter snow falls far below the lower border of the pine timber. Anthony writes that it falls down to about 2,500 feet above sea level on the west base of the mountains and it sometimes reaches as far as Valladares."

Range: Townsend's Oregon junco is resident in the Sierra San Pedro M4rtir (lat. 31010'N. to 30036'N.) in northern Baja California. In winter, some movement to lower altitudes, sporadically to San Agust~n, latitude 300 N.




In the northern Great Plains, in western South Dakota and northeastern Wyoming, lies a mountain mass about 100 miles long and 75 miles wide, ranging in elevation from 3,500 feet to a few peaks over 7,200 feet. These mountains, the Black Hills and Bear Lodge Mountains, are the home of the white-winged junco, one of the most numerous and conspicuous species of birds here and, during the breeding season, practically limited to this region.

The white-winged junco is the only bird species whose breeding range is essentially limited to the Black Hills. It has, however, close relatives to the west, north, and east. In Minnesota and across Canada the slate-colored junco breeds in the spruce forest, and in Wyoming various races of the Oregon junco breed in the Big Horn Mountains and the main ranges of the Rockies. The Black Hills are virtually an island of coniferous forest surrounded on all sides by prairies, and this fact may account for the distinctness of its junco population.

All the other characteristic breeding birds of the Black Hills pine forests have fairly wide breeding ranges. Among them are such Rocky Mountain species as Audubon's warbler and the western tanager, which here reach the eastern limits of their breeding ranges, and boreal forest species such as the gray jay and red-breasted nuthatch, which reach the southern edge of their breeding range in the Black Hills. Species of broad ecological tolerance such as robin and chipping sparrow are also numerous here.

Though the white-winged juncos are most numerous in the Black Hills pine forests, they are not limited to them. They may also be found in spruce forests and in aspen stands. In winter they often frequent brushy stream bottom lands, but they usually avoid open grasslands.

Spring: Most white-winged juncos winter in the lower elevations of the Black Hills or farther south in the foothills of the Rocky Mountains. With the first warm weather of early spring, they begin moving back to their breeding grounds. At banding stations in the Rapid City region, this movement is evident early in March. The following two observations are probably evidence of migration. On Mar. 18, 1956, a color-banded individual that bad been part of a flock wintering at 3,500 feet elevation on the west edge of Rapid City was observed at another feeding station 13 miles northwest of the point of banding and at an elevation of 4,500 feet. In January 1963, I banded 19 juncos, retrapping 9 of them later the same month but only 2 after the first of March. Meanwhile, an apparently diherent group moved through during March, and in the third week of March I banded 10 new individuals, some of which remained for a few days and were trapped again.

R. B. Rockwell and A. Wetmore (1914) collected two specimens near Golden, Cob., on April 11.

Territory: Each male white-winged junco apparently establishes and defends a breeding territory of a few acres, which it advertises by a well-developed spring song. Although some winter flocks tend to remain together well into April, two males were noted in full song on Mar. 4, 1956. One of these sang from several perches in pines, covering a territory at least 200 yards in diameter. I am uncertain whether the size of the territory is limited chiefly by the pressure of other males on neighboring territories, or by the maximum distance the male wants to move away from the nest, but suspect that it is the latter, as I have never observed a singing male driving another from his territory.

Courtship: Duiring many hours of watching white-winged juncos in the spring, I have never seen behavior I would consider of definite courtship pattern. Territorial singing is of course conspicuous at this time, and pairs of birds frequently can be seen feeding together. No specific activities, however, appear to precede pairing and copulation.

Nesting: W. H. Over and G. M. Clement (1930) found 29 nests during several seasons of field work in the central Black Hills. Of these six were under logs, four under exposed tree roots, three under rock ledges, and several had used artificial nesting sites. One location, used several seasons, was the roof plate of a busy blacksmith shop, 9 feet above ground, approached from under the eaves; one nest was in an old gallon syrup can and another in an old tomato can, both discarded in the pine forest. Several nests were found in and around a sawmill, where nesting was apparently successful despite the bustle of human activity.

In the summer of 1956 I found a nest built into a depression in the ground under a projecting slab of limestone. The nest construction was an outer framework of coarse grasses and an inner lining of fine grasses. Of the five nests I found in 1958, one was under projecting limestone, two were under the root network of grass or shrubs, and two were in small niches in limestone cliffs. Measurements of these nests showed the following variations: outside diameter: 10.0 to 15.0 centimeters; inside diameter: 6.0 to 9.0 centimeters; inside depth: 3.0 to 5.0 centimeters. The nest studied in 1959 contained 9.0 grams of nesting materials (dry weight).

Summarizing the above observations, I think that white-winged juncos usually select a cave-like situation, with a roof not far above the nest. Exceptions are known, but do not seem to negate the rule.

Eggs: The white-winged junco usually lays three or four slightly glossy ovate eggs. They are white or creamy white, speckled and spotted with "auburn," "Brussels brown," "hazel," "sayal brown," or c~cinnamon brown," with undermarkings of "pale mouse gray." In most cases the rather fine markings are concentrated toward the large end, although some may have them well scattered. There is always a considerable amount of ground showing, and the spottings are often somewhat dull and weak. The measurements of 30 eggs average 20.1 by 15.4 millimeters; the eggs showing the four extremes measure p21.7 by 15.2, 20.6 by 15.6, and 17.8 by 14.7 millimeters.

Incubation: Activities observed at a nest in Palmer Gulch in the central Black Hills, elevation 5,000 feet, in June 1959, were as follows:

June 2 : parents carrying nesting materials June 4 : nest lining completed June 6 : one egg in the nest June 7 : two eggs in the nest, both of which I marked with ink June 11: four eggs in the nest June 20: four eggs still present in the nest at 4:00 p.m.

June 21: one young hatched before noon June 22: four young in the nest at 9:30 a.m.

July 2 : two young still in the nest, one on the ground beside it From the above observations, one egg proved to have been in the the nest for 15 days between laying and hatching, and all four eggs were known to be under incubation for a minimum of 10 days. A reasonable assumption is that one egg was laid each day, and that incubation began on June 9, immediately after the fourth egg was laid. Another assumption which cannot be proved is that all four young hatched on June 21. If these two assumptions are correct, the incubation period is 12 days. Intensive observations of the critical periods of laying and hatching would be necessary to measure the incubation period with greater precision.

Young: In the nest I watched in 1956 the young hatched some time after 9:30 a.m., June 16, and before 5:00 a.m., June 19. When I visited the nest on June 19, the young were very small and blackskinned with some gray down. They held their heads straight up, begging for food, and one gaped as I leaned over the nest. On June 21 they gaped in response to my tapping the edge of the nest, but not to my blowing on them or shading them with my hand. At that time they showed down on top of their heads and the feather tracts on wings and bodies were appearing.

On June 23 I arrived at the nest at 6:00 a.m. and tapped the edge of the nest. The young closest to me responded by gaping but the others did not, and I presumed that they had just been fed. Thirty minutes later, during which time they were not fed, tapping the nest produced an immediate gaping response in all four young.

On June 24, 1956, I handed and weighed the young and collected the following data (weights in grams):

Band number

June 24

June 25

June 26

June 27 (at fledging)



















When I replaced them in the nest, they all gaped at me, pushing so hard that I had trouble finding room for the last one. At this age their wing and tail feathers are beginning to break through the sheaths.

After being weighed on June 25, all the young were able to scramble back into the nest. They then had a single distinctive call note unlike that of the adults. The wing and tail feathers were well developed by that time, but the white wing bars were not yet evident.

The young left the nest between 6:00 p.m., June 26, and 9:00 a.m., June 27. On the latter day two of the young, still flightless, were found in the vicinity of the nest. One was hiding in a clump of grass, and one under a yucca. On subsequent visits to the nest during the next few days I was unable to find any of the family.

Dennis Carter reports (in litt.) the following observations from Jewel Cave National Monument, Custer County South Dakota, in 1958. "White-winged junco is one of the most numerous species here and, on July 1, I found a pair with young birds. The adults were very agitated by my presence and uttered sharp alarm notes. When I finally discovered a juvenile bird, one of the adults swooped down over my head. I found three young birds, and although they could not fly, they moved along the ground at a surprisingly rapid rate and attempted to climb into brushpiles and shrubs."

Plumages: The scant natal down is slate-gray in color. The skin of the newly-hatched bird is black.

Richard Graber (1955), in a comparative study of immature plumages of sparrows, describes the juvenal plumage as follows: "Forehead and crown gray, streaked profusely with black. Nape similar, tinged with brown. Back brown-tinged gray, streaked with black. Rump and upper tail coverts grayish, obscurely streaked with black. Outer three rectrices largely white; others black, edged gray (fourth from outside with white mark). Remiges black, primaries and secondaries edged light gray. Tertials edged with pinkish-buff, tipped with buffy-white. Lesser coverts grayish, medians black, narrowly white-tipped. Greaters edged with buff, ripped buffy white. Two narrow wing bars. Lores gray, eye ring white. Auriculars gray. Post auriculars white, sparsely spotted with dusky (cheek patch partially outlining auriculars). Chin and throat grayish white (lateral black streaks), obscurely flecked with dusky. Sides of chest gray. Chest, sides, and flanks tinged with buff, streaked with black. Belly white, crissum buffy white, both unmarked. Leg feathers gray and white. Much like J. hyemalis, though clearly distinguishable."

Robert Ridgway (1901) describes Junco aikeni as follows:

Adult male: Head, neck, chest, sides, flanks, and upper parts plain slate-gray, darker (slate color) on the head; middle and greater wing-coverts usually tipped with white, forming two distinct bands; three outermost tail-feathers wholly white, the third sometimes with a little dusky, the fourth with more or less of white * * *

Adult female: Similar to adult male, but rather paler gray, the upper parts (especially hack) tinged more or less with light grayish brown, the wing-bands usually less distinct, frequently obsolete, and the third tail-feather more often with a little dusky * * *

Winter birds, especially young, are more or less tinged with light grayish brown, especially on the back. In some adult males the tertials are edged with white.

Food: No stomach analyses have yet been made. Insects appear to constitute all the food of the nestlings, and are probably the chief food of adults during the summer. In winter the white-wing's diet is primarily weed and grass seeds picked up from the ground.

Behavior: The white-winged junco is essentially a ground bird. Nesting takes place primarily on the ground, and most feeding does also. Trees are used for singing perches during the breeding season, as resting places throughout the year, and probably as roosting sites in winter. The most obvious behavior patterns can be observed in fall and winter flocks. In one such flock 20 birds observed Oct. 4, 1956, were feeding in a gravel drive at 4,500 feet altitude. They repeatedly flew down to the ground one at a time, but after feeding a few minutes they all flew up into the nearby pines together. While on the ground, they all moved slowly in the same direction, hopping, picking up two or three bites, and then hopping again.

White-winged juncos have often been noted around feeding stations well after sundown. Mrs. Grace McIntyre (pers. comm.) comments that they have fed around her banding station at Devil's Tower, Wyo., long after twilight. Individuals I have banded and released after full darkness have, in contrast, seemed unable to find their way to suitable roosts. I have not yet exactly determined the minimum intensity of light juncos require for feeding.

Alden H. Miller has pointed out (pers. comm.) that the whitewinged junco seems less active than smaller juncos, but that basically its habits are similar to theirs.

The general impression of winter flocks is that the birds tend to act as a compact unit. Close watching, however, shows that a definite pecking order exists, and that when two individuals are within pecking distance the dominant one lunges at the other without moving from his place. I have noted this often at banding traps, where one individual invariably defends the entry platform from all the others.

My observations concerning tolerance of other species are limited. Oregon and slate-colored juncos are the most intimate associates of white-wings in the winter flocks of the Black Hills. Oregon juncos, though smaller, may drive white-wings away from a feeding station. Other wintering passerines, such as black-capped chickadees and whitebreasted nuthatches, often feed with junco flocks, but seldom so closely that one tried to drive another away. Juncos usually scatter when flocks of piflon jays arrive at the feeding station.

When removed from a banding trap, most white-wings are passive, but some will attempt to peck the bander's fingers if they are held directly in front of the bird's bill. Their reactions are intermediate between those of the very passive pinion jays and the very pugnacions black-capped chickadees.

Voice: White-winged juncos in flocks seem to communicate with each other by a high musical squeak consisting of a single note. I heard this note used frequently by a flock of 10 juncos foraging around my banding trap during a heavy snowstorm.

Another frequent winter call is a single chip, similar to that of the slate-colored junco, but more musical. It seems to be more of an alarm note, while the squeaking note seems to be used in proclaiming intra-flock dominance.

True singing is apparently confined to spring and early summer. Peterson (1961) describes the song as "a loose musical trill, similar to the songs of other juncos." To me, it sounds very much like the song of the chipping sparrow, but somewhat more musical and variable.

Field marks: The traveler in the Black Hills and Bear Lodge Mountains during the summer can recognize white-winged juncos as the small gray birds that fly up from the roadside or the forest floor showing prominent white outer tail feathers. Closer observation shows that actually the plumage is predominantly a uniform pearl gray, except for the white wing bars and the white belly. In winter when the white-wing often associates with slate-colored, Oregon, and gray-headed june05, it can be distinguished from the Oregon and grayheaded juncos by the absence of contrasting reddish-brown on the back and sides, and from the slate-colored junco by the lighter gray plumage and the presence of more white in the tail. The white wing bars are good confirmatory points, but are not striking field marks in most individuals. White-winged june05 are also somewhat larger than the other june05 with which they may be associated. Whitewings average 24 to 27 grams, while slate-colored, Oregon, and grayheaded juncos weigh 18 to 20 grams.

Enemies: The only reported cowbird parasitism on the whitewinged junco was found by A. H. Miller (1948) in southeastern Montana. The cowbird (Molothrus ater) is not a common breeding bird within the range of the white-winged junco, and many other species of small birds nest in situations much more accessible to the cowbird.

K. C. Emerson (pers. comm.) has sent me the following list of species of Mallophaga (chewing lice) known from the white-winged unco: Bruelia milgata (Kellogg); Pliilopterus mirimotatus (Kellogg and Chapman); Ricinus pallidus (Kellogg); Machaerilaemus sp.; Menacanthus sp. R. 0. Malcomson (1960) mentions finding two species of lice, Rieinus kastatus and Ricinus palliJus.

Fall: Fall flocking begins on the breeding grounds in mid-July. Harry Behrens tells me that he saw several flocks in the central Black Hills about July 15, 1956. For the most part, however, juncos are seen singly or in small groups until early September. Later in the autumn flocks ranging in size from 10 to 30 individuals are usual. At this time, too, juncos begin to move down from the higher elevations of the Black Hills to the foothills. At elevations of 5,000 feet and over, much of the ground is snow covered from early November until mid-April, but below 4,000 feet much ground is bare. R. B. Rockwell and A. Wetmore (1914) note that white-winged juncos arrived in the region of Golden, Cob., on October 24 and reached maximum abundance on November 7.

Winter: T he flocking of white-winged juncos in winter seems similar to that of the better known species of the genus. Juncos frequent feeding tables during periods of heavy snow cover, but when the ground is clear they sometimes disappear entirely from artificial feeding grounds. I have yet to determine the exact foraging area a flock of juncos uses during the winter season. One individual Harry Bebrens banded in November in a brushy hollow near the center of Rapid City, S.D., was retrapped 11 days later in the pine woods 3 miles northwest of the point of banding. This is unusual; later I started a banding station ~ mile south of Bebrens' station. I banded 100 juncos there, while Bebrens had banded 30 at his home. While we each had several retraps of our own birds, we did not catch any of each other's.

For 3 years I made a population study of an area of ponderosa pine forest on the edge of Rapid City. My estimates of the winter population were 61 white-winged juncos per 100 acres the first year, 45 the second year, and 35 the third. The cause of this variation was not clear, although it could be correlated with the persistence of snow cover, juncos being more numerous when the ground was bare, and I could not evaluate its significance.

Range: Southeastern Montana and western South Dakota south to New Mexico and western Oklahoma.

Breeding Range: The white-winged junco breeds from southeastern Montana (Rosebud Mountains, Long Pine Hills) and western South Dakota (Short Pine Hills in Harding County; Custer) south to northeastern Wyoming (near Newcastle) and northwestern Nebraska (Hat Creek).

Winter Range: Winters in the vicinity of the breeding grounds (chiefly lower elevations in the Black Hills) and south to southwestern Colorado (Plateau Valley, La Plata County), north central New Mexico (Sante Fe), western Oklahoma (Cimarron and Texas counties), and western Kansas; sporadically to northern Arizona (Flagstaff, White Mountains), and central Oklahoma (Fort Reno, Norman).

Casual records: Casual in eastern Nebraska (Omaha) and Ontario (Scarborough).

Migration: Early dates of spring arrival are: Wyoming: Laramie, March 10. Montana: Missoula, April 15.

Late dates of spring departure are: New Mexico: Los Alamos, April 4. Wyoming: Laramie, April 16.

Early dates of fall arrival are: Wyoming: Laramie, October 13. Arizona: White Mountains, November 21. New Mexico: Los Alamos, November 16; Taos County, November 19. Oklahoma: Cimarron County, October 19.

Late date of fall departure is: Montana: Carter County, December 5.

Egg dates: South Dakota: 17 records, May23 to June 29; 8 records, May 27 to June 3. Wyoming: 2 records, June 8 and June 17.



This junco is a permanent resident on Guadalupe Island, Baja California, Mexico. The island lies about 135 miles due west of the Baja California coast and about 250 miles south southwest of San Diego, Calif., at lat. 29010'N. and long. 118018'W. It is a true oceanic island, volcanic in origin, that has never been connected with the mainland. Its dimensions are approximately as follows: length, 22 miles; width, 4 to 7 miles, broadest at north central portion; maximum elevation, at north central portion, 4,500 feet. The junco is one of a number of endemic land birds found presently or formerly on Guadalupe Island, and like most of the others its ancestors probably reached the island from the mainland to the northeast. A. H. Miller (194 ib) discusses the possible origin and relationships of insular-is in considerable detail in his famous monograph of the genus Junco. lie concludes that the affinities of insular-is are with the Junco oreganus group. Although some authors consider the Guadalupe junco to be a subspecies of oreganus, Miller retains specific status for insularis and suggests that this form was derived from vagrants of a migratory junco population inhabiting the California coast in Pleistocene or pre-Pleistocene times.

The Guadalupe junco is distinguished principally by its relatively long bill and short wing and tail, the reduced size of the sternum (F. A. Lucas, 1891), and a virtual absence of sexual dimorphism in color. Both sexes of insularis resemble somewhat the females of Junco oreganus pinosus. The streaked juvenal plumage of insularis also resembles that of oreganus juveniles.

The flora and fauna of Guadalupe Island have a sad history of destruction by introduced house mice, house cats, goats, and overzealous collectors. Successive stages of change in the avifauna may be traced through the papers of R. Ridgway (1876), W. E. Bryant (1887), J. E. Thayer and 0. Bangs (1908), and T. R. Howell and T. J. Cade (1954). The first biologist to visit Guadalupe Island was the botanist Edward Palmer, and he collected specimens of birds in the spring of 1875 that were subsequently described as new by Ridgway (1876). Palmer's notes on the junco, as quoted by Ridgway, are as follows: These are the most abundant birds of the island, and are so tame that they may be killed with a stick or captured in a butterfly-net. While I was looking for insects under stones and logs, these birds would sometimes join in the search, and hop almost into my hands. They gathered chiefly ants and their eggs. At times, they even enter the houses, picking up anything edible they can find. Numbers boarded the schooner as we neared the island, and made themselves perfectly at home, roaming over every part of the vessel in search of food. * * * A nest with eggs was found April ******* the highest point of the island * * It was placed in a small crevice in the face of a rock.

Before the introduction of terrestrial mammals the island had well developed flora including many endemic forms (S. Watson, 1875), and the junco undoubtedly became established and differentiated under conditions that were originally very favorable. Within a few years after Palmer's visit, however, the introduced goats had destroyed most of the native flora; by 1906, barren conditions similar to those of the present day had already been reached (Thayer and Bangs, 1908). Nevertheless, despite the extensive destruction of habitat and the depredations of house cats, the Guadalupe junco is still plentiful and very tame. It is found on the northern half of the island from the summit down to the shores. The birds inhabit the remaining groves of pine (Pinus radiata) , oak (Quercus tomentella) , and cypress (Gupressus guadalupensis) at the higher elevations, and also the now numerous stands of wild tobacco (Nicotiana glauca) in the canyons and along parts of the shore. This latter plant, which the goats do not eat, became established some time after 1932 and has spread rapidly. It now provides food and cover for the resident small birds and for the frequent stragglers that reach the island at various times of year. Virtually all other shrubs, undergrowth, and seedlings have long since disappeared or are eaten as soon as they are large enough to provide a mouthful for the innumerable goats.

The extensive ecological changes that have taken place on Guadalupe Island in the past 90 years have probably resulted in some changes in the habits of the junco, and it is perhaps less tame now than it was in the 19th century. Palmer's account has already been quoted, and H. A. Gaylord (1897) mentions an attempt by a junco to alight on the end of a gun that was pointed at it by a collector. Howell and Cade (1954) found that in June 1953 the juncos showed no alarm unless approached within about 6 feet, but the birds moved away at closer range and could not have been caught in a butterfly net or struck by a stick. Apart from this relative tameness, their habits appear to be much like those of other june05. The birds stay on or near the ground most of the time, but they may be found in the lower parts of the trees or, less abundantly, up in the tops.

W. E. Bryant (1887) is the only ornithologist who has stayed on Guadalupe Island during the early part of the breeding season of the resident passerine birds. He arrived on Dec. 16, 1885, and remained for 3% months. The biota of the island had already suffered from the activities of man and his introductions, but no birds were yet extinct and Bryant's notes provide the best account of the avifauna as it once existed.

Bryant found the juncos inhabiting primarily the pine and cypress groves, remaining for the most part on the ground among fallen trees, or in the lower branches. He reported that the june05 were paired soon after the beginning of the year and were setting by Jan. 26, 1886, despite almost continuous cold fogs and winds. No courtship behavior has been described.

Bryant collected a male Jan. 2, 1886, with "testes large"; males taken Jan. 26 and Feb. 4,1886 bad "testes very large." Of two adult males and one adult female that we collected buring the second week of June 1953, one male had enlarged testes (6 by 6 millimeters) and the others did not have enlarged gonads. These data and the records of nests and young cited below indicate that insularis has a long breeding season, beginning in January and ending in June.

Nesting: Palmer (in Ridgway, 1876) reports a nest in a rock crevice, and Bryant (1887) states that the ground beneath a fallen pine was most often used as a nesting place. He describes one nest as being in a depression, flush with the surface of the ground, and so carefully hidden beneath a covering of brush t.hat it was found with difficulty * * i"." Another nest Bryant described was 6 feet above ground in a narrow cleft between two conjoined pine trees. It was constructed of "a few pieces of bark-moss, light-colored dry grass blades, and a tail feather of a petrel, all surrounding a quantity of grass blades, lined within with goat hair. It measures externally about 120 mm. in diameter by 80 mm. in height, with a receptacle 60 mm. in diameter and only 28 mm. in depth."

Brown and Marsden (in Thayer and Bangs, 1908) describe a nest found in May or June of 1906 "placed on the lower branch of a pine * * * bulky and made mostly of dried grass stems."

Eggs: W. E. Bryant (1887) describes a clutch of three fresh eggs taken on Mar. 10, 1886, as follows: "In color the eggs are a pale greenish white, marked with fine dots of reddish brown clustered around the larger end. They measure 19.5 x 15; 20 x 15.5; 20 x 16 millimeters."

Bryant believed that these eggs were a second clutch as he noted young birds already present in other nests at that date. H. B. Kaeding (1905) found both fresh eggs and fledged young on Mar. 22, 1903. Palmer (in Ridgway, 1887a) reports a nest with eggs on Apr. 12, 1875. Bryant, and Brown and Marsden mention nests containing four young, and it seems likely that three or four eggs is the usual clutch.

According to W. G. F. Harris (MS.) the eggs of the Guadalupe junco are very pale greenish white, speckled and spotted with reddish browns such as "sayal brown" and 'chestnut." The two sets of eggs in the National Museum were collected on Guadalupe Island, Baja California April 5 and 12 respectively. The measurements of these six eggs average 20.4 by 14.9 millimeters.

Young: Bryant (1887) noted birds a few days old on Mar. 10, 1886, and he collected fully fledged juveniles on March 16. Kaeding (1905) mentions fully fledged young on Mar. 22, 1903. Brown and Marsden (in Thayer and Bangs, 1908) found full grown juveniles on May 1, 1906, but they also found young still in a nest in May or early June of 1906. I found many full grown juveniles in streaked plumage in the second week of June 1953. I estimate that 10 percent of the juncos in the Nicotiana thickets were juveniles but that in the cypress grove the proportion was as high as 40 to 50 percent of the total.

Plumages: The Guadalupe junco appears to have the suime sequence of plumages as do other juncos. The early nestling stages have not been described, but the juvenal plumage closely resembles that of J. oreganus: streakcd with dark brown on the entire head, breast, back, and flanks, with the wings, tail, and abdomen essentially like those of an adult. This juvenal plumage is presumably lost in the late summer or fall in the first prebasic (postjuvenal) molt, and the basic plumage thus acquired has the same aspect as that worn by adults. There is a prebasic (annual) molt by adults following the nesting season, but no change in the aspect of the plumage. Two out of three adult birds we collected in June 1953 were undergoing this molt, nnd many other molting adults were seen at that time. Considering the early beginning of nesting in the Guadalupe junco, it is likely that adult birds have completed their molt by midsummer. As mentioned previously, there is no evident difference in coloration between the sexes.

Food: The Guadalupe junco subsists on both insects and seeds. Palmer (in Ridgway, 1876) mentioned that the birds gathered ants and their eggs as he turned over logs. In June 1953 we found most insects rather scarce and ants particularly rare; no other observers have mentioned the latter in the junco's diet. We noted that juncos foraged in loose soil in the cypress grove and that they probed into crevices in logs for moths and possibly other insects. The juncos also fed on the calyces and ovaries of the Nicotiana flowers, which were abundant in June 1953. No doubt many other kinds of plant and insect food were utilized when the island was in its primordial condition. Bryant (1887) states that "their food was principally of seeds, a partiality being shown for the green seeds of the 'wild lettuce.' " Whatever plant this may have been, it probably no longer exists on the island.

Voice: The call notes of the Guadalupe junco are similar to those of Junco oreganus. Bryant (1887) states that he heard the song only twice, delivered from the top of tall cypress trees, and that it resembled the trill of the chipping sparrow. I did not bear any song approximating a trill, and describe the songs I heard as "wheep-whit-whitwhit-wheep." Bryant (1887) also mentions that the juncos gave a sharp chipping note when alarmed. No other accounts of the vocalizations of insularis have been published.

Field marks: As ineularis is strictly resident on Guadalupe Island, and as no other juncos are to be found there other than accidentally, there is no problem in field recognition. The Guadalupe junco has a gray hood, black lores, a dark brown iris, brownish (not rufous) back and flanks, no wing bars, and a white-edged tail. The long bill is quite noticeable in the field and would probably serve to distinguish insularis from any other forms in the same genus.

Range: The Guadalupe junco is resident on Guadalupe Island, Baj a California.

Egg dates: Baja California: 3 records, March 10 to April 12.

The following subspecies are discussed in this section: Junco caniceps caniceps (Woodhouse) and J. c. dorsalis Henry.
Contributed by DONALD M. THATCHER


One of the most frequently observed and relatively least known avian species of the coniferous forests of the southern Rocky Mountains and the Great Basin is the gray-headed junco. Two races are recognized: the northern Junco caniceps caniceps and the southern Junco caniceps dorsalic.

According to present taxonomy the genus Junco includes 10 species, 4 of which are subdivided, making a total of 21 recognized forms. The over-all range of these 21 forms resembles a chain, double and interlocking in the West, extending from Panama and southern Baja California north to Alaska and northern Canada, thence east to Labrador and Newfoundland, and from there south to Georgia, wherever suitable habitat occurs. From southern Arizona and northern Baja California northward and eastward, junco habitat is essentially continuous or its "islands" not too widely separated, so the chain is practically unbroken: in a sense a "dine." Each form occupies its own separate and distinct range, but each mixes freely and hybridizes with its neighbors within a narrow belt, much as though all the darkeyed, northern members of the genus Jurnco were one species, as some taxonomists consider them. The southern forms, whose chief difference from the northern is their yellow eye color, also form somewhat of a dine, but without mixing, except within Junco phaeonotus, as their "island" ranges are more widely separated and no migration encourages mixing. Presumably because all juncos are ecological equivalents and occupy the same niche in essentially similar habitat, there is no sympatry of two or more species such as those in Dendroiea, Sitta, Par-as, and several other avian genera. The gray-headed junco is especially interesting as its two races and their intergrades appear to form a three-stage transitional link near the center of the "chain" between the yellow-eyed juncos southward and the dark-eyed ones northward, combining some of the characteristics of each group.

The two subspecies of the gray-headed junco have had an interesting and confusing taxonomic history, probably equaled by few other birds. S. W. Woodhouse (1853) described the northern form as a full species, Struthus caniceps, the "gray-headed snow finch." Five years later T. C. Henry (1858) described the southern subspecies also as a full species, Junco dorsalis, the "red-backed snowbird."

The two subspecies have since been considered together as: a single race of the Mexican junco, J. cinereus "variety" caniceps; two separate races of J. cinereus (later J. phaeonotus); and as races of the original Junco species, hyemalis. Junco caniceps has now for many years generally been considered a separate species, but one proposal made it a race of the Oregon junco, as J. oreganits eaniceps, while dorsalis was left as J. phaeonotus dorsalis. Jonathan Dwight (1918) considered dorsalis a mere hybrid of two species, J. caniseps and J. plvieonot us. The fourth edition of the A.O.U. Check-List (1931) listed the two forms as Junco canieeps and Junco pkaeonotus dorsalis. The present status, as two races of J. caniceps, was proposed by A. H. Miller (1932, 1934) and accepted by the fifth edition of the A.O.U. Check-List (1957), although even Miller (194 ib) recognizes dorsalis as "a fully established form which gives evidence in its peculiar sublimation of characters of origin by the hybridization of J. c. caniseps and J. p. palliatus," practically echoing Dwight's opinion. This turbulent history has resulted mostly from the fact that such field characteristics as eye color, voice, color of eggs, and manner of locomotion were lost in the skins upon which classification was based. Its taxonomic history emphasizes the position of Junco caniceps as the connecting link between the southern and the northern juncos exhibiting the general plumage of the southern and the field characters, lost in study skins, of the northern groups.

The gray-headed junco is the only junco known to breed in its pure form in Colorado, Utah, and Nevada, with the minor exception of a race of the Oregon junco, J. o. hruroeri, in extreme western Nevada. Pure, unmixed Junco can jeeps in the northern and western parts of its range confines itself, even where junco habitat extends far beyond, within the limits of those three states. Practically all breeding records from near those state lines and adjoining parts of California, Idaho and Wyoming are of birds in populations mixed and interbreeding with J. oreganus, either thurberi west or mearnsi north. Junco habitat in southeastern Arizona and extreme southwestern New Mexico is occupied by the similarly plumaged, but yellow-eyed, northern subspecies of the Mexican junco (J. phceonotus palliat'us). Here there is no intermixing of the gray-headed and its neighbor, the breeding ranges of the two species being separated by hot desert, although one point of separation is no wider than 37 miles.

Habitat: Miller (1941b) says of the habitat of J. c. caniceps:

The spotted distribution of this junco must be emphasized. It inhabits for the most part a series of mountaintop islands above 7000 feet in the arid Rocky Mountain and Great Basin ranges. Associations in which it breeds include coniferous forest types dominated either by spruce (Picee), Pseudotsu go, Pinas contorta, Pinup ponderosa, Pinup ./lezilis, or fir (A bies). It also breeds in pure stands of aspen (Pc pulus tremuloidee) and of mountain mahogany (Cercocer pus ledifotius). Compared with Junco ore ganus, it shows high tolerance for arid forest and ground cover. It may occupy aspen groves where there is no surface water within two to five miles of the nest site. Unshaded forest floor with the ground poor in humus and nearly lacking in green plant cover is unsuitable for breeding.

While this subspecies has a high tolerance of arid conditions, its tolerance of moisture also is high, as is indicated by three nests R. B. Rocliwell (1910) found June 16, 1910 at Columbine Lake, a 40-acre Canadian Life Zone lake in northern Colorado at 8,630 feet elevation surrounded by "a dense growth of pine and spruce extending in places to the water's edge." Of J. c. carticeps breeding in the Uinta Basin in northeastern Utah, A. C. Twomey (1942) reports: "The birds nested in all the mountains of the Basin from altitudes of 7,500 feet to timberline at 10,000 feet. Their range included the Douglas fir, yellow pine, blue spruce, aspen, lodgepole pine and Engelmann spruce-alpine fir forests."

In southwestern Utah, according to W. H. Behle (1943), the gray-headed junco occupies approximately the upper half of the Transition Zone ponderosa pine belt, which occurs between 6,200 and 7,600 feet, and the lower two-thirds of the Canadian Zone blue spruce, white fir, and aspen at 8,000 to 9,500 feet. Behie says:

"The gray-headed junco is the summer resident of the juncos of the region. It occurs in the mountains from 7,000 to 9,000 feet. At lower elevations within the limits mentioned they are found most often in side draws arising from main canyons; at higher elevations they frequented the dense cover bordering meadows and occasionally ventured out along the edges of the meadows." Its failure to breed in southern Utah above 9,000 feet and into the spruce: fir of the Hudsonian Zone to timberline, as it does elsewhere, has not been explained. H. W. Henshaw (1886) found ca~iceps breeding somewhat lower in New Mexico "everywhere throughout the timber belt above an altitude of 6,000 feet."

Owen A. Knorr tells me of an exceptional nesting of caniceps on the University of Colorado campus at Boulder, Cob., in May and June 1955, at the relatively low elevation of 5,400 feet. The nest was on northward-sloping ground in low-growing myrtle (Vinca), near the shaded edge of a ~-acre pond. Tree cover was large American elm, western cottonwood, blue spruce, and a few small Juniperus. Here, on the plains a mile east of the Rocky Mountain foothills, several hundred feet lower than the usual Transition Zone habitat of the species, and only a few yards from a busy road, the birds found suitable conditions and fledged a brood.

Another low nesting, but in normal habitat, is reported by Louise Hering (1954) just within the lower limit of the ponderosa pine forest at approximately 5,900 feet % mile south of Boulder. Junius Henderson (1912) mentions a breeding, probably at the same place: "Bragg's summer record of the Gray-headed Junco at Boulder, altitude 5,700 feet, July 4, 1904, should be added to the list, as it indicates a probable breeding record much below the usual elevation."

For detailed breeding-bird population studies that show the habitat preferences of the gray-headed junco in the forest associations in the Colorado mountains see Hering (1948, 1954, 1956, 1958, 1961, 1962, 1963), Lawbead (1949), Snyder (1950), Cassel (1952), Thatcher (1954, 1955a, 1955b, 1956), and Beidleman (1960). These studies, covering all major forest types from 5,500 to 11,200 feet in altitude, show in essence that gray-headed junco populations may be found in varying numbers in forested areas of almost any tree species which are well but not densely stocked, have numerous openings to provide edges, and are not too arid.

Of the habitat of J. c. dorsalis, Miller (1941b) says:

There is a large area in the Mogollon Mountains of Arizona and New Mexico over which the distrihution is nearly continuous. In addition to this, numerous isolated mountain ranges are occupied, much as in the breeding range of J. C. cari~cep~. Plant associations in which dorsaiss breeds consist of coniferous forests wherein the following types predominate: Pine. ponderosa, Abiea, and, less commonly, Pseudotsuga. Birds also breed in pure groves of aspens and in coniferous forests with some oaks intermixed. Compared with the habitats of J. c. caniceps, those in which I have found dorsalis are, i~ anything, less arid, more luxuriant forests, although occasionally birds will hreed in the dry lower portions of the yellow pine belt. Ground cover is rarely as poor in grass, low hushes, and humus as in many parts of the range of J. c. caaiceps in Utah and Nevada. However, the presence of good stands of yellow pine in the Transition zone provide [sic] junco hahitat at lower zonal levels than in Utah and Colorado.

E. A. Mearns (1890a) found in east central Arizona that "Typical dorsatis breeds very plentifully through the northern Mogollon and San Francisco Mountains, but does not appear until one has ascended a considerable distance into the pine belt. It is the most characteristic bird of this higher region * *

The observations of these and several other writers indicate that the breeding range of dorsalis extends lower than that of eaniceps. The ponderosa pine forest of the two more southern states, New Mexico and Arizona, is less arid and more luxuriant than most of that of Colorado, Utah, and Nevada, suggesting that the lower edges of the more northern pine forest may be too arid for suitable junco habitat. At the other altitudinal extreme, dorsatis seems generally not to breed as high as cardeeps. W. W. Cooke (in Bailey, 1928) states its breeding range as "Transition and Canadian Zones of high mountains in Arizona and north central to southern New Mexico," but probably not "much if any above 9000 feet," while the breeding range of canicepa is given as "Hudsonian and Canadian Zones," up even to 12,400 feet at one locality.

Spring: As spring approaches and the northern gray-headed junco (caniceps) starts northward and upward toward its summer home, the proportion of this species in the mixed flocks of juncos in their winter haunts thins noticeably. At Paonia, in central western Colorado, in the Upper Sonoran pifion-juniper and scrub oak country, Homer Griffin wrote me: "Gray-headed, pink-sided [J. ore ganus mearnsi], and Oregon IJ. o. montanus] juncos were present all winter; grayheaded juncos made up the majority of the population during March and pink-sided juncos during April." On the plains in the vicinity of Denver, Cob., the gray-headed has almost entirely left for the mountains by May 1, while the more northern Oregon and slate-colored (J. hyemalis) juncos are still present in numbers.

The gray-headed junco becomes plentiful in the vicinity of its breeding territories long before conditions are suitable for nesting. F. C. Lincoln (1920) took a specimen in Colorado at a garbage pit at 10,400 feet, Mar. 31, 1915, when "snow was from 3 to 15 feet deep and this bird's presence is more or less * * * a mystery to me." Up to timberline in the Colorado mountains the males sing freely in the tops of the tallest pines or spruces while the ground is still covered by two to three feet of snow. But during severe late snowstorms, which may occur as late as the last week in May, large numbers often reappear in the lower foothills or at the edge of the plains, wherever brush patches and bare ground offer food. At such times they seldom sing. The latest ca'nieeps record on the Colorado plains is a female Niedrach and Rockwell (1939) report "taken in Denver, June 17, 1918." Joe T. Marshall (MS.) observed seven or eight caniceps in the Santa Catalina Mountains of southern Arizona Apr. 5,1951, in a "very tame" flock with six or seven Mexican juncos, a J. o. n-iearnsi, and a J. o. thurberi. A month later, May 4, he saw a caniceps with Mexican juncos at a picnic table in the same locality. This is approximately 250 miles south of the breeding range of carciceps. In southeastern Arizona Seymour Levy (in Phillips, Marshall, and Monson, 1964) saw caniceps "to May 24 (1957 in Santa Rita Mountains ** and there are exceptional dates of June 5 at Fort Huachuca and June 6 in Guadalupe Canyon, all some 250 to 300 miles south of the breeding range of caniceps.

The southern race, dorsalis, shows less seasonal movement than caniceps, and often winters with flocks of caniceps and Oregons in the pines and adjacent brushy places at or near its summer home. In March dorsalis individuals leave the mixed flocks and take up their summer territories while the higher country is still snow-covered. Edouard Jacot (MS.) writes that caniceps in the White Mountains of Arizona is "not as solitary as Red-backed [dorsalis], which is often seen alone in the timber." At Granville, Ariz., at the lower edge of the Transition Zone, he recorded the last dorsalis Apr. 19, 1935, in the "deciduous white oaks" (Q. gambelli), where wintering or migrating caniceps remained as late as May 7. On Feb. 13, 1937 Lyndon Hargrave (MS.) wrote of dorsalis at Flagstaff, Ariz., well within the pines of the Transition Zone: "Probably all my banded winter birds left last night. Abundant 12th." Allan R. Phillips recorded the "close of migration" there the previous year on April 23, after which date only one pair remained of the six or eight birds previously recorded daily in the vicinity.

Territory: While the caniceps probably all leave their breeding territories for the winter, apparently many dorsalis that breed in the ponderosa pine forest remain throughout the year in the vicinity of the breeding grounds. Hargrave (1936) reported banding an adult male on its breeding territory in Flagstaff Jan. 30, 1935, which he retrapped several times from May 8 to September 3, and again Noveniher 8 and the following February 14 (MS.). Apparently it remained in the general vicinity the entire year. Its mate, ho~vever, was observed or trapped only from March 24 to September 3, and probably wintered elsewhere.

Some, probably most, dorsalis breed near the territory in which they were raised. Hargrave (MS.) reports, also at Flagstaff, a female he banded as a juvenile, Aug. 13, 1936, and which he believed nested the next year within 50 yards of the point of banding. The bird nested there 2, 3, and 4 years after banding, and was last seen there in May of the 5th year, 1941.

Recorded observations of territorial conflict between gray-headed juncos are few. During population studies by the writer (Thatcher 1955a, 1955b, 1956) and others in three areas in the eastern foothills of the Colorado Rockies, approximately 84 supposed territorial pairs were observed during 5 breeding seasons, in or immediately outside the study areas. Only one brief fight between two males of this species was recorded, June 18, 1953. On May 12, 1952, two gray-heads, apparently a pair in their nesting territory in open ponderosa pine, attacked and drove off a migrating slate-colored junco. In the same locality on Mar. 16, 1960 I attracted two gray-headed juncos and a single J. o. mearnsi with "tic" notes from a pewter-and-wood "bird call." The gray-heads ignored the migrant mearnsi but presumably regarded the bird call as a rival.

Louise Hering (1948) found five pairs of caniceps believed to be nesting on a 75-acre study plot in the Black Forest of Colorado. She reports: "Three pairs of juncos were far separated on the tract, while two breeding pairs remained near each other throughout the season. The males of the latter two pairs sang rather often but both families fed on the forest floor without any apparent territorial conflict." A. C. Twomey (1942) in Utah took a female hybrid canicepsmearnsi "from a nest * * * within forty feet of a nesting pair of canweps."

Courtship: The mating behavior of the gray-headed junco probably differs little, if at all, from that of the hetter known Oregon and slatecolored juncos. Twomey (1942) wrote of our species in Utah during the nesting season: "The birds at that time could he heard singing from all corners of the forest. The male always chose the top of a tall pine as a singing post. Considerable activity, consisting chiefly of pursuit and nest-building, was observed here."

Miller (1941b), in discussing hybridization between J. c. carticeps and J. o. mearnsi in northeastern Utah. describes interesting mating behavior of some of the hybrids:

A pair of birds about my camp west of Garden City was feeding young on July 22 [1931]. The male was collected at 5 am.; it was pure meern2i on back and sides, but with head intermediate. The female, which could be seen to have normal mearnsi color on the sides and back, was left with the young. At 10 a.m. the small young were in the same group of busbes with the female, and a new male was on hand, singing, following her with tail fanned, and twittering with characteristic mating behavior; she did not drive him away. This bird was taken and found to have a mixed yellow back, mearnsi sides and intermediate head. Female X, as she now became known, had another male attached to her party shortly after 11 a.m. This male was mearnsi in all characteristics except for intermediate tone of head. A fourth male came to female X at noon and proved to have pure mearnsi back, pure caniceps sides, and intermediate head. I am doubtful that these males were all unattached previous to their interest in female X. Males of various sparrowt~ are known to be polygamous on occasion. There was no doubt of the attraction of the female for all of them, however. Not knowing the history of the case, an observer would have considered each to have been her normal mate. No intolerance was evidenced by the female. Some of the males gathered food for the young. This indicates disregard on the part of the junco for differences in colors of sides and backs.

Nesting: The gray-headed junco is with rare exceptions a ground nester and is ordinarily the only such passerine within its usual habitat, except the larger and much less plentiful Townsend's solitaire (Myadestes townsemdi), whose nest is not likely to he confused with the junco's. At timberline the junco's nest may be placed under a rock like that of the water pipit (Anthus spirtoletta) and the two may be confused readily. Ground nests of the Wilson's warbler (Wilsonia pusilla) and the Lincoln's sparrow (Melospiza lincoln.ii) in wet places, of the white-crowned sparrow (Zonotrichia leucophrys) at timberline and of the western flycatcher (Empidonax difficilis) in roadside or stream banks might be confused with the junco's.

Nesting activity commences as soon as most of the winter's snow disappears, and often before the last heavy spring snowfalls. My earliest observation of caniceps nesting activity in north central Colorado was Apr. 28, 1954, when I (Thatcher 1954) saw one of a pair carrying nest material near the upper limit of the ponderosa pine, at 7,800 feet during a "warm, dry spring when breeding activity was two to three weeks earlier than usual." The earliest nest in the same area had five eggs when I found it May 29, 1952, and young 3 days later. Assuming 5 days for laying and 12 for incubation, the nest was probably completed by May 15.

The latest definite nesting activity date is contributed by Louise Hering (MS.) of two young that seemed "to be just out of nest" at Grand Lake, Cob., at approximately 8,500 feet, Sept. 4, 1949. As parental feeding of the young probably continues 18 to 20 days more, the nesting period of caniceps in Colorado presumably extends through a period of at least 21 weeks: April 28 to September 22.

The May 29 caniceps nest mentioned above was at the edge of a large opening in the pines on a slight southwesterly slope, hidden in a shallow hollow beneath the downhill side of a clump of mountain mubley (Muhienbergia montana), the major grass of the south-slope pine forest. A heavy, 3-inch snowfall 6 days before the nest was discovered had no adverse effect on it. Three other nests found nearby during 5 years of population studies were similarly located in smooth, rounded hollows under clumps of mountain muhley, with no other immediate protection.

The usual Colorado nest of the gray-headed junco is well described by Aiken and Warren (1914): "The nest is built on the ground, sometimes in a cavity in a roadside bank or a stream bank. One found by Rockwell and Warren in Jefferson County was sunken in the ground so that the rim was flush with the surface. The nest proper was made of grass, coarse outside, lined with finer, with a few horsehairs intermingled. This was under a Douglas's fir tree, and nearly covered by a spreading branch which grew out almost at the foot of the tree and actually rested on the ground over the nest. This nest contained four fresh eggs * * t Taken May 30, 1912."

In northeastern Utah, Twomey (1942) found: The first nests * * * at Green Lake, Uinta Mountains, on June 10 [1937]. * * * From June 17 to 20, at Indian Canyon, eighteen nests with fresh eggs were found in the mixed blue spruce, Douglas fir, yellow pine and aspen forest. The nests, averaging four or five eggs, were always on the ground, generally under a protecting shrub or a log. These were the most common nesting birds of this region.

At Paradise Park, between July 7 and 10, numerous nests were located, all containing eggs advanced in incubation. Birds in juvenal plumage were seen in large numbers at Bald Mountain in the Engelmann spruce-alpine fir forest from July 16 to 20. These juncos were nesting in larger numbers here than at any other place visited in the [Uinta] Basin.

Florence Merriam Bailey (1904) found several caniceps nests at 11,000 feet in north central New Mexico, near the headwaters of the Upper Pecos River, "nests being found everywhere in the open. * * * All of the nests were on the ground, completely hidden by tufts of grass or bunches of weeds, being discovered only by flushing the brooding bird."

Denis Gale (MS.) describes nests he found in the Gold Hill: Ward district of the Eastern Slope of the Colorado Rockies, at 8,000 to 10,000 feet: "Nests on the ground. Fond of selecting sheltered places, side of hill or bank, concealed xvith care and cunning, set well into the ground, with sometimes only the smallest aperture for the entrance and exit of the bird; seldom selecting a bare place, but preferring seine shrub, plant, tree or stump or roots of such, under or close to which it excavates sufficiently for its purpose. Bird somewhat difficult to flush. Nest of coarse grasses outside, lined with fine grasses, hair and feathers sometimes. Eggs four and five * * One nest measured 4% by 3% inches outside: 2% by 1% inches inside." From Gale's data I estimate the earliest date of laying to be May 20, 1890 and the latest clutch was started July 8,1886.

I found four nests of caniceps in 1958 on a north-facing slope in penderosa pine at 7,800 feet 18 miles west of Denver, Cole. Nest 1, found May 25, contained four eggs and was in a slight hollow under a small, fallen, dead tree branch, overgrown with kinnikinnick (Aretostaphylos uva-ursi). When visited June 27, the nest had been torn apart and the young apparently taken by a predator; an adult was singing in a nearby tree.

The otber three nests were in a somewhat more open area, 200 to 350 yards west from nest 1, and were in concurrent use in July, presumably for second, or possibly third, broods. Nest 2 had four eggs when found July 12, and three young July 19; one egg did not hatch. Built under a small common juniper (Joniperu.s comm~cnis var. moutana) on level ground instead of in the more usual hollow, it was a large and bulky structure of grasses with a few pine needles and strips of inner bark. It measured: inside, 1% inches deep by 2% inches in diameter; outside, 3 inches deep by 4% to 7 inches in diameter. Nest 3 was found July 19, hidden by grass and concealed beneath a Canada buffaloberry (She pherdia canaxlensis), 50 yards down-slope from nest 2. It contained young both when found and 7 days later. This nest, too, was bulky, but had less than half the bulk of nest 2. Of the four, it was the only one containing hair, apparently of deer, and feathers, a few', gray and probably from the parent bird. This nest measured: inside, 1% inches deep by 2% inches in diameter; outside, 2 inches deep by 4 to 5 inches in diameter. Nest 4 was 150 yards west of 2 and 3, surrounded by grasses, alongside a Douglas fir seedling 12 inches tall, and 8 feet from the nearest large tree, a 10-inch pine. As it was in a hollow, this nest consisted of no more than a lining, % to % inches thick, of fine grasses and plant fibers and a few pine needles. It contained young 3 or 4 days old when found, July 27. The dissimilarity of nests 2, 3, and 4 can best be expressed by their dry weiglits, which were 33, 14 and 2 grams, respectively.

Nests 1 and 2 I found by searching when the parents scolded as I approached. The birds of nest 3 behaved much differently; the sitting bird flushed only when almost stepped upon, left without a sound, and never returned to scold on the three occasions I visited it. Nest 4 was found by watching one of the parents carrying food until it dropped from a pine overhead to the ground near the nest.

Neil Frederick Eadley recently spent two summeiu studying the species near Boulder, Cob. From about 7,000 feet altitude to timberline at 11,500 feet in the Colorado Front Range he found 34 grayheaded junco nests, most of them in the "upper montane" and "subalpine" zones between 9,300 and 11,000 feet. He has sent me the following notes from his unpublished study:

All the nests I found were on the ground and usually well concealed under some form of shelter. Favorite locations were beneath fallen logs, stumps, rocks, conifer seedlings, small shrubs, or tufts of grass, or in the banks of streams and gullies. Nests tend often to be somewhat "tunneled" into the ground, with only a very small aperture for the entrance and exit of the bird.

The outside of the nest is usually composed of coarse grasses and old decayed leaves; the inside is of finer grasses and always lined with some type of hair or feathers. Dr. Horace Quick helped me identify hairs from a number of different nests. We found not only the usually cited cattle and horse hairs, but also cervid hairs of either deer or elk, dog hairs, human hairs, hair from a snowshoe hare in winter pelage, and both fur and guard hairs of squirrels.

The nesting of dorsalis is similar to that of caniceps, but generally starts earlier in the spring, for this race winters either within or much nearer its summer range than does caniceps. llargrave (1936) saw a dorsalis carrying nest material at Flagstaff in northern Arizona, Mar. 24, 1935. The first brood left a nest in a vine at the same place, according to Katherine Bartlett (MS.), on Apr. 28, 1940. E. C. Jacot (MS.) saw a dorsalis in the White Mountains of Arizona, at 8,000 feet elevation, Apr. 7, 1935 "carrying nesting material: a pine needle." Probably the latest recorded nesting date for dorsalis appears in F. M. Bailey (1928): ~ * * On August 17, 1919, a nest with eggs was found 30 miles southwest of Chloride in southwestern New Mexico * * * at 7,200 feet (Ligon, 1916: 1918)." Assuming a requirement of at least 30 days after hatching for completion of breeding activity, this would extend the nesting season for dorsalis over a period of 25~ weeks, March 24 to September 17, 4 or 5 weeks longer than that of caniceps.

Of dorsalis nesting in northern Arizona, Mearns (1890b) says:

Sets of fresh eggs were found from May 22 to July 22, 1887, the nesting season varying considerably with the altitude, but the clutch seen on the last date probably belonged to a second brood. A typical nest was found on May 30, 1887, in pine woods near the bottom of a ravine on Mormon Mountain. At a short distance was a deep snow-bank. The male parent flew from the nest, beneath my horse's feet, where I found it concealed in a thick bunch of wire-grass. It was composed of fine roots, stems of plants, grasses, and an occasional feather, loosely put together in the manner of most ground nests. It contained four eggs * *

I found its nest close to the upper edge of timber on San Francisco Iviountain about the middle of June, and another nest on the very top of Baker's Butte 18077 feet], containing eggs, on the 22nd of July.

F. lvi. Bailey (1928) describes a dorsalis nest: "A typically well concealed nest of the Red-backed Junco found by Mr. Ligon in the Chloride region was hidden under a small pine that had spread over the ground and caught dead leaves, making a thick supporting mat as well as a dark base for the inconspicuous nest of bark and dead grass with its slight lining of hair. To further protect it from prowling enemies, a small bowlder [sic] stood beside it, blocking the entrance." Mrs. Bailey (1927) mentions that dorsalis also nests in clumps of oaks on hillsides," as does XV. I. Mitchell (1898):

"Abundant. Most common at 8,000 feet, breeding in clumps of scrub-oak on hillsides" in north central New Mexico. E. C. Jacot (MS.) describes an Arizona dorsalis nest found June 7, 1935, by Mrs. Jacot "in a thicket of 3- to 4-foot yellow pines; the nest was hidden near the foot of one of the small pines. It was composed almost entirely of very fine rootlets and some grass." One found in the same locality, Aug. 13, 1937 by Hustace H. Poor (MS.) near a stream was in a "dark, shady site under broad-leafed weeds."

Like the other juncos, the gray-headed sometimes nests off the ground. W. W. Cooke (1900) says of caniceps: "Breeds abundantly at Breckenridge [Colorado], and in 1898 one nested there under the eaves of Mr. [Edwin] Carter's house." One caniceps nest of 24 found by D. D. Stone (1884) in Colorado was "placed in a small pine, three feet from the ground, in a heavy bunch of timber." A note filed at the museum at Grand Canyon National Park, by K. Wing, June 5, 1950, mentions a nest "in the rafters of the cafeteria porch" at the North Rim of the Canyon. Lyndon L. Ilargrave (1936) tells of a dorsalis "nest in vines under the eaves of the house" at the Museum of Northern Arizona at Flagstaff in 1934. Unpublished notes by Hargrave and others at the museum mention a later nest at the same location "in vine, northwest corner of patio," about 8 feet above the ground. Not only was this nest unusual in its location, but it was used several times: in 1937 (probably); in 1938, number of broods not known; in 1939, two broods; and in 1940, two broods. After the second brood left in 1940, a second nest, in which the third brood of that season was raised, was built on a roof plate "one foot over and two up" (note by Katherine Bartlett) from the first nest.

Although there seems to be no definite proof of caniceps raising more than one brood in a season, the many late nestings indicate that it probably raises at least two. As N. F. fladley (MS.) comments: "The long period over which nests are found in Colorado suggests second broods are attempted whenever possible. The number of broods raised probably depends on how early the first nesting begins and the weather conditions when the second brood is to be started. If the first nesting is not successful, caniceps will attempt a second orood, as we observed several times. In each case the second nest was near the site of the first and built under a similar shelter, that is if the first nest was under a lodgepole seedling, so usually was the second."

Of dorsalis, however, we have Hargrave's (1936) detailed record, mentioned above, of the raising of three broods. Following is his summary, in part:

The summer's observations on breeding Red-backed Juncos at Coyote Range, Flagstaff, Arizona, have shown (1) that the male probably wintered within his prospective summer territory inasmuch as he returned to the locality on November 8, 1033, where he was banded on January 30 of the same year and where he was recaptured in early March; (2) that the female probably appeared on her breeding ground after the male had established himself; (3) that nesting activities were under way by March 24; (4) that one brood was hatched near the first of June, another in late July, and the last in late August; (3) that the male was father to three broods and the female was mother to the first and third and probably the second broods * * *

Eggs: The gray-headed junco lays from three to five slightly glossy eggs. 'rhey are usually ovate, although some may tend to short ovate. The ground is white, or very pale bluish-white, speckled, spotted, and occasionally blotched with "huffy brown," "sayal brown," "pecan brown," or "russet" with undermarkings of "pale mouse gray." These spottings may be scattered over the entire surface, but are usually concentrated toward the large end, frequently forming a wreath. The markings are often dull and somewhat clouded, and tend more toward yellow-browns whereas both Junco hyemalis and Jztnco oreganus have red-brown spottings. Sometimes a set or a single egg may be found that is plain bluish-white with only a few specks so small that they are hardly visible. The measurements of 47 eggs average 19.9 by 15.2 millimeters; the eggs showing the four extremes measure p2.0 by 15.8, 20.6 by 16.0, 18.7 by 15.0, and 19.3 by 14.1 millimeters.

Young: There appears to be no record of close observation of the complete nesting cycle of the gray-headed junco. Incubation time can be assumed to be 11 or 12 days, as given by Bergtold (1917b) for the slate-colored junco. Miller (1938) notes that two eggs laid in captivity by a "Point Pinos junco" (J. o. pinosus) mated with a male dorsalis hatched in 12 days. While determining the sex of an incubating bird is difficult, unless the bird is collected or the male is singing nearby, a bird flushed from the nest is generally assumed to be a female. Mearns (lS9Oa), however, as noted above, says of dorsalis: "The male parent flew from the nest," while it contained eggs.

From his observations on caniceps near Boulder, Cob., N. F. Hadley (MS.) writes:

The female alone incubates the eggs. The male will occasionally fly down to the nest and feed her while she incubates and then fly away, or both may fly away together. The male sometimes remains at the nest edge while the female flies off, but I never saw one enter the nest and assume an incubating position. The female on the nest continually changes her position and turns the eggs, using both her bill and feet for the purpose, commonly once every five to ten minutes.

The male's arrival with food can often be foretold by the incubating female's actions. Alerted by some sort of signal from the male, not always heard, she perks up her head and turns it from side to sIde in anticipation. When ants or other insects pass by within reach, she will reach out to grab them. She also leaves the nest at intervals to feed herself. I observed one pair feeding together on the floor of a lodgepole stand over 200 yards from the nest. The frequency and duration of her absences from the nest and of her periods of attentiveness to incubation depend on time of day and weather conditions. During morning hours when the sun's illumination is at its peak due to slope and nest exposure, she tends to keep the eggs or young covered to prevent overheating. She tends to incubate longer and more continuously during raw, wet weather.

The eggs in a clutch seldom hatch simultaneously. Most hatch in the morning, but often one or two hatch in the morning and the remainder in the afternoon, or even the next day. After the first eggs hatch the female seems to move the others so that they receive optimal warmth. The actual emergence of the young from the egg is very rapid, often helped by the female pulling the shell with her beak. As soon as the fledgling is free, the female eats the egg shell.

Both male and female feed the hatchlings a diet that appears to be entirely of insects. When the female is brooding, the male sometimes deposits food at the edge of the nest where the female can take it for herself, or give it to the young. The female is very attentive to nest sanitation. At each feeding visit she scours the nest and removes the fecal sacs, usually depositing them on the limb of a nearby spruce or lodgepole. The young are apparently not fed in sequence; the same bird may be fed several times in succession if it proves the most aggressive. The nestlings' bright red mouth linings combined with theft wide yellow bills give the feeding adult a large, easily visible target.

The young spend approximately 10 to 11 days in the nest after hatching, though one brood remalned 13 days. The nestlings' eyes become slit-like on the fourth day, and open fully by the sixth day. As they become older (7 to 10 days) they venture out of the nest to receive food from theft parents, sometimes as far as two feet, always returning immediately to the nest. They also leave the nest temporarily to try out their wings.

It was thought the nestlings would make theft final move out of the nest in response to the parents' enticing them with food, but they sometimes desert it of their own accord when the parents are not present. Not all leave the nest at once; one or two may remain several hours after their more venturesome siblings have gone. Once out of the nest they become very difficult to follow. They spend at least three or four days hopping around on the ground before they can fly with any degree of coordination. During this time their parents watch them solicitously from nearby trees and bushes, flying down to feed them occasionally.

The following history of the young in a nest of dorsalis at Flagstaff, Ariz., is from unpublished notes by A. R. Phillips at the Museum of Northern Arizona. The nest containing four eggs was discovered May 25, 1936 under a tussock of grass, opening toward the west, which was shaded in the morning. The next day at 2 p.m. it contained four young "mostly naked with some gray down on head and rear end." May 29, at 3 days, the four young weighed 27.5 grams. May 30: "still nearly naked, eyes not open~~; weight 35.5 grams. May 31: eyes still closed; weight 44.5 grams. June 1: eyes opening; weight 48 grams. June 2 (8:24 a.m.) : eyes open; feathers of belly tracts becoming prominent; weight 60.5 grams. June 3 (8:15 a.m.): "One young squealed (first sign of fear instinct) and both parents came to the nest"; weight 62 grams. June 4 (11:40 a.m.) : weight 64 grams. June 5 (11:12 a.m.): weight 70 grams; (1:25 p.m.): only three young in nest, one probably having been "lured away by the parents"; at this time the young were 10 days old.

Hargrave (1936) found that the dorsalis male parent did most of the feeding of the first broods after they left the nest. The male in his report fathered three broods in one season (determined by trapping and banding with the young as decoys). After the third brood left the nest "both adults were observed" feeding them.

Plumages: As previously mentioned, A. R. Phillips states that one-day-old young of dorsalis were naked with some gray down on head and rump. R. B. Rockwell (1910) described newly-hatched young he saw in Colorado as "pinkish little creatures irregularly covered with very fine grayish down." R. R. Graber (1955) says one of two Colorado caniceps nestlings had "smoky gray [natal] down on the rump," while the other had "down on the side of the crown." Graber describes the juvenal plumage: "Forehead and crown light gray, heavily streaked with black. Nape tinged with buff. Back mahogany red, streaked with black. Rump buffy, obscurely streaked with blackish. Upper tail coverts buffy gray, obscurely spotted with blackish. Rectrices blackish gray except outer two pairs white and third from outside about half white. Remiges black, narrowly whiteedged (tertials edged with mouse gray). Coverts edged with gray. Secondary coverts tipped with whitish (two obscure wing bars). Lores dark gray. Auriculars gray, postauriculars like nape. Subauriculars streaked, blackish and white. Chin and throat white, obscurely spotted with gray. Underparts largely white, the sides and flanks light huffy. Chest and sides streaked with blackish (triangular marks with apex anterior). Belly and crissum white. Legs gray."

Two New Mexico dorsab~s in juvenal plumage, apparently somewhat more advanced than the caniceps described above, were similar to the caniceps but in general lighter-colored. These Graber describes: "Forehead light gray. Crown and nape gray, streaked with black (nape more sparsely). Back light rusty (burnt sienna of Ridgway), sparsely streaked with black. Rump and upper tail coverts light, huffy-tinged gray, sparsely streaked with blackish. Rectrices largely dark gray, outer two pairs white (some white on third from outside). Remiges black, narrowly light edged (tertials broadly edged with gray). Coverts edged with gray and buff, secondary coverts tipped narrowly with white (narrow wing bars). Lores black. Auriculars and postauriculars gray, unmarked. Chin unmarked whitish. Other nnderparts \x-hite. Throat, chest, sides, and flanks finely spotted and streaked with blackish. Legs gray."

Miller (1941b) briefly describes the adult plumage for both races:

Iris dark brown; lower mandible flesh-colored, upper mandible black or flcshcolored. Back with sharply defined mahogany red area, confined normally to interseapular region; * * * tips of feathers grayish * * ~'ï Lores and ocular region black, contrasting with neutral grays of head. Sides gray without line of demarcation separating them from upper breast. Tail always with two, and usually three, outer feathers partly or completely white. * * *

Sexual differentiation in plumage slight. Females average slightly lighter in colors of head and side in race caniceps only. Fourth rectrix more often pure black, fifth and sixth rectrices less often pure white in females than in males. Degree of sexual differentiation in rectrices variable, depending on population involved. * * * Females often have traces of buff on tips of feathers of sides, and reddish- or buff-tipped wing coverts. A distinctive immature or retarded plumage not recognizable; occasionally young males have buff-tipped sides, and this feature predominates in young females.

The red of the back alters with the season, becoming brighter and yellower to approach burnt sienna and Sanford's brown. The gray tips are worn off early, hut this does not seem to affect the tone of color importantly. Brightening is not pronounced until late April, but in the succeeding six weeks the greatest alteration takes place. * * * The grays of. the head pale slightly, so that the mass effect in worn plumage is that of lighter hood, except in the extreme state of wear of midsummer, when exposure of the basal downy barbules gives a sooty appearance.

For J. c. caniceps, Miller (i941b) continues: "Features present in all individuals that distinguish them from J. c. dorsalis are: (1) darker neutral gray (light neutral gray and neutral gray) of hood, especially that of throat, and (2) flesh-colored upper mandible." Miller also states that measurements of caniceps tail, bill, tarsus and middle toe average slightly less than those of dorsalis, and the "amount of white in the tail averages less." He says further, of caniceps: "Reddish color of variable extent is not uncommon on the pileum. It appears in all populations. In the total of 772 J. c. caniceps examined, 48, or 6.21 percent, have some feathers distinctly red, not merely buff tipped." Of dorsatis, he adds: "Red on the pileum occurs in 1.7 percent, compared with 6.2 percent in carviceps and 1.7 percent in palliatus.

I have seen in Colorado in early spring a caniceps with the entire pileum red, much like that of the chipping sparrow (Spizella passerina) in extent. Rockwell and Wetmore (1914) mention an "immature female [caniceps, taken in November, with] a rufous line on each side of the croxvn," and another female with "faint rusty tips to the feathers on the occipital and nuchal regions."

A regular but infrequent variation within dorsali.s is the appearance of the mahogany red of the back "on the outer webs of the inner secondaries and greater secondary coverts," which appeared in 9, or 3.9 percent, of the 230 specimens Miller examined (1941b). This feature suggests affinity to phaeonotus, but places of its occurrence are not related to proximity to the range of that species. A. K. Phillips (MS.) at Flagstaff in March 1936, banded "a remarkably red bird ï [dorsalis], the color of the back invading much of the body: to be specific, the sides broadly tinged, the crown somewhat so, the second aries edged and their greeter coverts mostly the same color as the back!" Another variation within dorsali.s is the color of the lower mandible, which occasionally is dark, but probably never as dark as the black maxilla. Seven birds Phillips banded at Flagstaff, all but one in winter and hence not juveniles, had lower mandibles ranging from "pinkish lavender" through "bluish," "dark lavender" (a juvenile banded in June), "dark blue," "blue black," and "black."

F. M. Bailey's (1902) comment on the plumage of dor.salis applies equally well to cani~eps: "The coloration of most of the juncos is not particularly protective except as the color pattern disguises the bird's form, but the red-backed on the pine plateau of San Francisco Mountain, Arizona, spends a large part of its time about the fallen pine-tops, where the red of its back and the red of the dead pine needles and old bark make a protective combination that, added to the gray of the body, which offsets the gray of the branches, results in a most effective disguise."

Food: Juncos are mainly terrestrial and obtain practically all of their food on or very near the ground. Unless hard-pressed by hunger they seldom feed on a shelf or other raised feeding station, but prefer to pick up the seed spilled on the ground by other species. F. M. Bailey (1928) mentions dorsalis in July "feeding among the dead leaves, * * * scratching much like chickens" in the Capitan Mountains of New Mexico, and mentions Major Goldman's finding them in the Mogollon Mountains in late October, when "Small parties were everywhere hopping about, scratching among the leaves through the thin snow, and when startled rising and alighting in the lower branches of trees."

Unusual feeding behavior of a "red-backed" junco is reported by E. C. Jacot (MS.) in the White Mountains of Arizona on Mar. 13, 1935 "feeding from the bark of a pine tree 20 to 30 feet from the ground, as painted redstart~s often do. Perched on a limb and flew to the trunk keeping tail somewhat fanned, then back to the limb."

No detailed study has been made of the food of the gray-headed junco, which probably differs little from that of the other, betterknown northern junco species. Of eardceps in the Sangre de Cristo Mountains of New Mexico, F. M. Bailey (1928) says: "Among the insects fed to the young birds were a caddice [sici fly and a green caterpillar." A. R. Phillips notes that the dorsali8 young he studied at Flagstaff, Ariz., when 4 days old were fed green seeds and "apparently a green caterpillar." One of Miss Bartlett's Flagstaff notes says that at the "vine" nest previously mentioned Milton A. Wetherill observed that the female parent "fed young birds with soft parts of grasshoppers, carrying hard parts away." There, also, Hargrave (1936) reports: "On one occasion both adults were observed carrying millet [from a trapping and feeding stationi to young fifteen feet above the ground in a pine tree." William H. Belie (1943) says of caniceps in southwestern Utah: "Their diet seemed largely insectivorous while feeding young, for nearly every individual shot had the mouth and throat filled with bodies of moths, measuring worms, or larval insect forms." Louise Hering (1948) in her Black Forest, Cob., study noted young caniceps being fed "moths and worms" obtained from kinnikinnick. At a Colorado caniceps nest at timberline in July, I watched both parents bringing the young small grasshoppers.

In the White Mountains of Arizona, F. G. Watson (MS.) noted of dorsali~s seen daily about a cabin at 9,400 feet elevation in July 1936, that "they feed on crumbs and grain * * ¶" and A. R. Phillips (MS.) on Oct. 20, 1936, mentions one (of two or three) eating dandelion (Taraxacum) seeds. F. M. Bailey (1928) says of flocks of caniceps driven to lower elevations in the Gallinas Mountains of central New Mexico by an early October snowstorm: "They were everywhere but especially abundant in the weed patches on the edge of the scrub oak thickets into which they flew when flushed, and the stomachs of two taken were full of seeds, including a large per cent of pigweed." Edward R. Warren (19 lOa) quotes John W. Frey, from central Colorado, regarding caniceps: "Thousands of these birds wintered here [1908: 9] on the tumbleweed seed." I have observed wintering caniceps in Colorado eating the seed of an abundant species of cheat grass (Bromu.s sp.). According to D. I. Rasmussen (1941), who observed mixed winter flocks of Oregon and gray-headed juncos on the Kaibab Plateau, in northern Arizona: "They are active on the ground, and their food consists of all available plant seeds, grasses, herbs and shrubs."

E. D. McKee (1934) noted wintering juncos at Grand Canyon eating seeds of the pifions (Pinu.s edulis and P. monophylla): "When cracked nuts are put out it has been found, strangely enough, that natural seed-eaters such as juncos and chipping sparrows actually prefer them to various types of grain and that robins hold them in equal esteem with their much-loved raisins. Even bluebirds, Cedar Waxwings, and Cassin's Purple Finches will eat them with relish." At the South Rim of the Canyon, Oct. 20, 1948, L. Scheilbach (MS.) "observed 2 [dorsalis] distinctly pecking open pinyon pine nuts and extracting the nut meat * * *ï" On a Colorado mountain top at 11,300 feet F. V. Ilebard and A. W. Gardner (1954) observed a flock of caniceps in early April feeding with red and white-winged crossbills, pine siskins and pine grosbeaks, in Engelmann spruce and limber pine, where "spruce seeds were the main source of food."

Behavior: Nesting gray-headed juncos, except when incubating closely, generally take alarm readily and scold an intruder vigorously with their characteristic, sharp, rapidly-repeated "t.ic" note, which E. C. Jacot (MS.) describes as sounding "like snapping two nickels together." A bird on the nest usually flushes only when nearly stepped upon, and in most cases remains nearby, scolding loudly. But an occasional one flushes silently and disappears until the intruder leaves. J. K. Jensen (1923) notes a New \rexico caniceps S. R. Hammitt found that "flew off the nest almost under his feet, and although he waited patiently for more than an hour, it never returned." Jensen adds, "I have several times seen the birds building, in which case the nests have always been abandoned." F. M. Bailey's account. of H. W. Henshaw's flushing from her nest a female that "glided off through the grass, fluttering about and feigning lameness" is unusual.

In an account of pine grosbeak breeding behavior in northeastern Utah, Norman R. French (1954) writes: "A pair of Gray-headed Juncos nested near the base of the same tree in which the grosbeaks were nesting. The grosbeaks sometimes alighted low in the tree and worked up to their nest. The juncos invariably came into the tree and worked down to their nest. As a result the paths of the birds sometimes crossed. When a junco was on a perch and one of the grosbeaks came toward it the junco immediately flew, usually to a lower perch in the tree."

As mentioned previously, juncos will respond to "tic" notes made with a pewter-and-wood bird-call in their breeding territories. Once, in July, I used the call to attract mountain chickadees, unaware of two or three fledgling juncos near by. The parent juncos responded with vigorous scolding to the "tic" call. Wintering juncos generally ignore a call, but occasionally foraging mixed flocks, largely of Oregons, will respond to one as readily as do chickadees or nuthatches.

Although fairly tolerant of moderately arid conditions, the grayheaded junco makes frequent use of available water. L. Hering (1948) states that caniceps was one of six species of birds breeding in Colorado pine "seen bathing in the creek." C. XV. Stillinan (MS.) records "A group of at least six [dorsalis] close to a spring and small stream in bushes and grassland. Seen bathing in a stream" Aug. 5, 1937. A. R. Phillips ''saw one or two bathing in Horse Creek," Nov. 22, 1936.

Voice: The voice of the gray-headed junco is in most respects similar to that of the slate-colored junco as A. A. Saunders (1935) describes it:

The song of tl~o junco is a very simple one. Normally it is a simple trill * * * or a scrics of rapid notes all on the same pitch * * ~. The number of not~, wheii they can be counted, varies from eight to twenty or more. The quality is rather musical, decidedly more so than the quality of the Chipping Sparrow. There are occasional variations in time, with the notes at the beginning faster or slower, and pitch sometimes varies up or down a half-tone or a tone.

Occasional individuals have some peculiar abnormal form of the song, sometimes so different that one must see the bird to he sure of identification. Such individuals seem to sing in that way always, and may he known and followed from year to year by the peculiarity. An individual has little variation in its singing.

Early in the sprine, when singing first begins, one may sometimes find a Junco singing a faint, varied song with mixed pitches. It is a highly attractive song when one is near enough to hear it clearly, hut I am inclined to think it is primitive in character.

In winter, birds produce a note like "tsehehehc" which is rather musical and pleasing. Alarm notes are "tsHc" and "fUtCtCtUp" the latter used commonly when the nest contains young, but rarely heard from winter birds.

In the Colorado foothill ponderosa pine forest where both J. c. caniceps and the chipping sparrow are numerous, I found that frequently the quality of a less musical song of the junco matches the chipping sparrow's song so well that distinguishing unseen birds is extremely difficult, even when the two are singing simultaneously. They can usually be told apart by the timing and duration of the songs. The junco's song generally consists of 1%- to 2-second trills, given eight to ten per minute, while the chipping sparrow's are of approximately three seconds duration given four to six times per minute. Later in July I found this relative timing not completely reliable when a supposed junco near a known junco's nest at timberline and singing at the rate of eight trills of 1 'A to 2 seconds duration per minute turned out to be a chipping sparrow. It sang 191 times in 24 minutes, stopped for 1% minutes, then resumed singing. Not only was this sparrow singing the junco's song, but it was 3,000 feet above its usual foothill habitat.

I consider the song of the gray-headed junco more varied than that of the slate-colored. A gray-head in early spring, Mar. 29, 1960, in its breeding habitat sang three different songs, all of the more musical, "junco" quality, averaging ten per minute. One variation, lasting approximately two seconds, consisted of a two-part trill, swe-awe-sweswe-swe-te-te-te-te-te; the second, approximately one second long, was a simple trill of seven or eight notes, te-te-te-te-te-te-te, like the second part of the longer song; and the third, also of one second's duration, resembled the first part of the longer song, ewe-ewe-ewe-ewe-ewe-awe. Another song, heard the next day at the same place, alternated two simple trills, one quite musical, typically "junco," and the other less musical, much like the chipping sparrow's. Four or five of these alternating trifle comprised a series.

Joe T. Marshall, Jr. (pers. comm.) recorded an early July song of a dorealis singing at noon and later in the day near the top of a dead aspen in the Canadian Life Zone of the Arizona White Mountains: "Steadiness, quality and variety suggest phaconotus: Tswee tswee tsee tswee chit chit chit chit. There was another series of tsuwee tsuwee tsuwee tsuwee chit chit chit chit and one of just six chits."

While the junco characteristically sings from tall trees in spring, it occasionally uses lower perches. One of a pair of caniceps I watched at timberline in Colorado June 24, 1Y58, sang on the ground as it foraged and occasionally from a twig a foot or two above the ground.

At Flagstaff, Ariz., in the Transition pine forest at 6,900 feet, L. L. Hargrave (MS.) in 1936 recorded the earliest territorial songs of two dorsalis, one February 14 and the other March 3. The next year he heard juncos (perhaps Oregons instead of "red-backs") singing a "soft twittering song, very faint: not heard far" on February 4; the first territorial song by a dorsal is, he heard the same date as the preceding year: February 14. E. C. Jacot (MS.) at Alpine, southeast of and 1,100 feet higher than Flagstaff, recorded the gray-headed junco, probably both subspecies, singing "since the 11th" of March 1935, when there were 10 to 15 inches of snow on the ground, with open patches at bases of trees and along streams and roads. On March 18, Mr. Jacot recorded of caniceps: "At Alpine as usual. Seem to be in full song," and of dorsalis: "red-backed in full song. They and grayheaded seem to sing almost as soon as they are perched in tree or bush after being flushed." The latest dorscdis singing dates also are from the White Mountains. F. G. Watson (MS.) at the Phelps Ranger Station at 9,400 feet heard them "singing about the cabin each day" July 9 to 28, and also on Baldy Peak at timberline, 11,200 feet, July 12. Thus the period of singing by dorsalis in 1936 included nearly twenty-three weeks: February 14 to July 28.

My earliest seasonal record for singing by caniceps in Colorado is Mar. 15, 1951, in its breeding habitat at 7,500 feet. On May 2,1954, none of approximately 15 caniceps at the edge of the foothills at 6,000 feet, driven down from the mountains by a severe snowstorm, was heard to sing, although 7 weeks earlier on March 13 in 1955, at the same place several individuals of a mixed flock of approximately 35 J. o. montanus, J. o. mearnsi, and J. hyemalis (no J. caniceps) were in song. On May 21, in both 1955 and 1960, two of seven or eight birds seen each time in the Hudsonian Zone at 11,000 feet were. singing, although 2 to 3 feet of snow delayed nesting 4 or 5 weeks During a 5-year population study in ponderosa pine (Thatcher, 1956) the latest singing date was July 17, 1955, when the male of only one of six observed pairs sang. A few days earlier on July 12 in 1952, four of seven were in song. My extreme latest song dates are July 31, one in a cool aspen forest at 7,750 feet and one of eight adults at timberline, 11,700 feet. The song period of caniceps thus is approximately 20 weeks, March 15 to July 31, 3 weeks less than that of dorsalis.

While timing early morning singing in the ponderosa pine forest, June 6, 1954, I recorded the first song of caniceps at 4:23 a.m., 16 minutes after sunrise and 55 minutes after the first birds, western flycatcher and common nighthawk, were heard. I heard notes of 12 additional species before the junco's first song.

Field marks: Juncos usually are readily distinguished from other small birds occurring in their range by the combination of white outer tail feathers and the characteristic "tie" notes they almost invariably give in flight. The combination of mahogany-red saddle, light gray head, and gray sides distinguishes the gray-headed from all other juncos except those that breed south of its breeding range. Some races of the Oregon junco have somewhat reddish backs, but their heads are black or nearly so. The "pink-sided" Oregon junco (J. o. mearnsi) has a light gray head, but its back is brownish with almost no hint of red, and its sides are distinctly reddish, more so than those of the black-headed Oregons. Tf the lighter gray head fails to distinguish the gray-headed from the slate-colored and the white-winged juncos when the red back is obscured, the lack of a sharp line between gray of breast and grayish-white of abdomen should do so.

The light, flesh-colored upper mandible of caniceps and its slightly darker head, "light neutral gray and neutral gray" of ~vIil1er (1941b), distinguish that race from the more southern dorsali.s with its blackish upper mandible and "pale or pallid neutral gray" (Miller) head. The breeding gray-heads of the Kaibab Plateau in northern Arizona and of the Zuni Mountains in central western New Mexico are mostly intermediate. This condition is most obvious in the color of the upper mandible, which is flesh-colored or pinkish with tip and base black in highly variable proportions, as described by Miller (1941b), who later (1949) terms the Kaibab birds "a single hybrid swarm, not a geographic gradient either of blending or of alternating characters or both."

The pale upper mandible readily distinguishes those J. e. caniceps wintering in southern Arizona from the resident Mexican junco, but other characteristics must be observed for identification of the fe~v dorsalis which winter there. These are, for the Mexican junco: yellow eye, yellowish feet and lower mandible, and considerable red on wings; and for the gray-headed junco: dark brown eye, flesh-colored feet, grayish lower mandible, and little or no red on wings. In addition the voices differ considerably. The song of the Mexican is more varied than that of the gray-headed, being described by R. T. Peterson (1948h), in comparison with those of the Oregon and slate-colored juncos, as "a more complicated finch-like song, which involves two and sometimes three pitches * * *," and the call resembles more the "tsip" of the chipping sparrow. The gaits of the two differ noticeably also, for, as Peterson (1948b) says: "Furthermore, the Arizona [Mexican] Junco creeps along in strange mouse-like fashion. * * * whereas other juncos habitually hop." According to L. N. Nichols (1936) the Mexican junco "is said to have less the manners of a Junco than of a Water Thrush," and Allan Brooks (1914) agrees that "it walks daintily and deliberately over the floor of the forest like a titlark or water-thrush, instead of the shuffling hop of the junc.o and sparrows."

Owing to the free interbreeding of the gray-headed and Oregon juncos where the breeding range of the former meets those of J. o. thurberi and J. o. mearnsi, birds of mixed plumage are numerous in the overlapping ranges and occur commonly throughout the winter range of caniceps. Many of these have the red back of caniceps with the pinkish sides of mearnsi. in some individuals the red back is diluted or mixed with yellowish, or the pink of the sides is spotty or on only a few feathers. A few may have the dark head of thurberi or the brown or yellowish back of the Oregons, with the gray sides of caniceps. Variations are almost unlimited and most confusing to the close observer. The mated female J. o. thurberi and male J. c. caniceps x J. c. mearnsi Miller (1935) collected at the meeting point of the ranges of the three forms could certainly have produced puzzling offspring! Enemies: The enemies of the gray-headed junco are presumably those common to most small ground birds, but we have little definite information on this subject. The eggs or young of only two of seven nests I have observed until completion were taken by what I have assumed were small mammals. Upon investigating an unusually vigorous scolding by a pair of gray-heads at 9,500 feet in Colorado, in mid-August, I found a fledgling junco a day or two out of the nest in the jaws of a large garter snake and released it, apparently unharmed.

One of the young of the campus nest, at Boulder, Cob., previously mentioned, was said by 0. A. Knorr (pers. comm.) to have been taken by a screech owl nesting nearby. R. G. Beidleman (1957) mentions long-eared owls which "evidently preyed on the juncos" on a winter population study area in Colorado ponderosa pine, where four species of juncos, including caniceps, were present. Miss Oppie Reames (pers. comm.) reports a pygmy owl (Glaucidium guoma) eating an Oregon junco in southern Colorado in late October, undoubtedly from a mixed flock with a high percentage of gray-heads. Penis Gale (MS.) presumed predation by the northern shrike in northern Colorado. His notes state for Oct. 17, 1889: "Saw a Northern Shrike * * Evidently has crossed the range in the wake of Juncoes [sic] and was making fur the valley." Howard Rollin (pars. comni.) saw a northern shrike in January "chase and kill a pink-sided junco" at his ranch on the Colorado plains during severe weather when mice, the shrike's usual diet, remained in shelter. An occasional gray-head loses its life at a banding or feeding station; at the Grand Canyon a banded dorsalis was reported killed in November "by an Abert Squirrel in a trap" (unpublished banding notes), and at nearby Flagstaff, Hargrave (1936) reports an immature bird killed in a trap by a sharp-shinned hawk in February. Also at Flagstaff, Hargrave (MS.) says the first brood raised in 1940 in the "vine" nest previously mentioned was "killed hy dogs the day after they left the nest."

Although H. F. Friedmann (1949) lists the white-winged and the Cassiar (J. hyemalis ci.smontanws) juncos as victims of cowbird parasitism, there seem to be no records of victimization of the grayheaded junco. The breeding habitats of this junco and the br6wnheaded cowbird (Molotkrus ater) seldom, if ever, overlap.

Fall: After completing their breeding activities the northermi caniceps form small groups, presumably of adults and the young of the season, and start a southward movement. Such groups are frequently encountered weeks before late breeding is finished in the same area. Twomey (1942) observed small flocks as early as July 25 at 6,000 feet in northwestern Colorado. The late-summer migration appears to be southward and frequently upward as higher mountains are encountered. Occasionally small moving bands occur 2,000 feet or more above timberline, where I have seen them as late as August 29. The first birds usually reach the "lowlands" about October 1, but their advent may be hastened by an early mountain snowstorm. Grayheads soon join the migrating flocks of other junco species from the north, which eastward are mostly J. o. mearns~, and westward J. o. montan'us. In Arizona and New Mexico, the J. a. caniceps and the Oregon juncos associate commonly with the local, less migratory dorsalis.

The latitudinal limits of tbe winter range of caniceps are well south of those of the breeding range: generally 100 to 150 miles in the north and 500 miles or more in the south. Thus presumably the birds migrate not only vertically but a considerable distance southward. In his study of the winter birds in Utah, Hayward (1935) concluded that heavy snowfall covering "a large part of the available ground food * * * for a considerable length of time" determines the northern limit of the winter range of the gray-headed and other juncos.

To learn whether the wintering or migrating gray-headed juncos at Boulder, Cob., breed, as is generally presumed, in the mountains immediately to the west or northwest, Dr. and Mrs. John N. Hough color-banded 234 birds from October through April of 1957: 58 and 1958: 59. Not one of these birds was report.ed during the three summers, 1958: 1960, though several observers watched for the colored bands west of Boulder and in Rocky Mountain National Park 20 to 40 the birds miles northwest. This su~ests that migrating through or wintering at Boulder breed some distance farther north in Wyoming or extreme northern Colorado.

J. c. dorsalis has more of a vertical than a horizontal migration, and few birds move any great distance outside the pine forest of the breeding range. H. S. Swarth (1924b) writes from northern Arizona: "One specimen was collected in juniper woods September 28 * * ~. This was the only occasion on which dorsalis was seen below the yellow pine belt. When the migrating northern juncos arrived, the mixed flocks of caniceps, skuJeldti [probably morttanus (Miller, 194 ib)] and others were abundant at the lower edge of the pines and in the pifions and junipers, but no specimen of dorsalis was found in any of these aggregations. I found dorsulis at a higher altitude, in small flocks, and never accompanied by any other species." A. R. Phillips (Phillips, Marshall, and Monson, 1964) says dorsalis "'~ * * seems a fine example of an altitudinal migrant, at first glance. During severe winters, at least, it occurs commonly in the Verde and upper Gila Valleys * * * just below the Mogollon and Natanes Plateaux. It is notable, however, that these valleys lie to the south of the breeding range. * * * the available evidence indicates that these birds move downward in a southerly direction only."

In summarizing his account of the raising of three broods of dorsalis, Hargrave (1936) comments on fall behavior: "~ * * (6) that shortly after leaving the nest the young of all but the last brood moved out of the nesting territory; (7) that the third and last brood remained within the nesting territory until the postjuvenal molt was nearly completed; (8) that the parents apparently left the nesting territory together and ahead of their offspring of their third brood; (9) that, after the summer adults had left, the young of the third brood remained and mixed with others of their race; and (10) that the members of the third brood apparently left the nesting territory together after the fall migration was well under way." The young of the second brood left the nesting territory August 3, 1935. The first birds from outside were trapped September 10 and 11, "indicating that the fall movement was under way." The next season, Hargrave and A. R. Phillips (MS.) noted that dorsalis juveniles became numerous at Flagstaff August 15, nearly 4 weeks earlier than the preceding year, but no adults were seen at that time.

Winter: While in general the northern gray-headed juncos, caniceps, move downward and southward for t.he winter, a considerable number winter in the mountains, well into the Canadian Zone in suitable places such as near corrals and bird-feeding stations in towns. There the gray-heads are usually in company with smaller numbers of Oregon juncos and an occasional slate-colored or white-winged junco. Frequently one or two gray-heads and perhaps an Oregon or a slatecolored occur in company with the white-winged juncos that typically winter in flocks of 10 to 30 in the more open stands of ponderosa pine of the eastern Rocky Mountain foothills. In similar foothill situations I have seen small groups of gray-heads in mid-winter, usually unaccompanied by other juncos, in close company with mixed flocks of mountain chickadees and pygmy nuthatches, and perhaps one or two white-breasted or red-breasted nuthatches or brown creepers. Many gray-heads winter with the other juncos along the lower edges of the coniferous forest, chiefly in brushy ravines and patches of Crataegus, scrub oak, mountain mahogany (Cercocarpus montawus) and other "brush," which furnish both shelter and a plentiful food supply not covered by snow. Often what appear at first to be only three or four juncos will, upon taking flight two or three at a time to the next brush patch, turn out to be a mixed flock of fifty or more. Such flocks may also contain considerable numbers of tree sparrows.

Just as dorsalis migrates much less than the more northern ea'niceps, the latter is considerably less migratory than the more northern races of the Oregon junco, sh'ujeldti, mont anus, and mearusi. Miller (1941b) says that "Compared with J. c. caniceps, mearnsi leaves the breeding range more completely in winter and spreads out on the plains in greater number. The extreme limits of record are similar in the two forms; they are commonly associated in flocks in winter in Colorado and in the oak belt of New Mexico and Arizona."

At Boulder, Cob., near the northeastern extreme of the winter range of caniceps, the John Houghs (pers. comm.) banded 4617 juncos from October through April, 1946 through 1959. Their relative frequencies were: gray-headed 24 percent, white-winged 6 percent, slate-colored 6 percent, Oregon 64 percent (approximately 8 percent montanu.s and 56 percent mearnsi). The average junco numbers for the 13 Christmas counts between 1945 and 1961 at and near Boulder as reported in Audubon Field Notes show the following frequencies: gray-headed 14 percent, white-winged 4 percent, slate-colored 4 percent, J. o. monto,nus 11 percent, and J. o. mearnsi 67 percent. Summaries of the Christmas counts of the past 30 years within the winter range of the gray-headed junco show its relative numbers increase southward to the region of greatest concentration, a small area in north central New Mexico around and including the Sangre de Cristo Mountains. There approximately 45 percent of the juncos east of the mountains and 75 percent of those west were gray-heads. Immediately to the south the northern Oregons outnumber the gray-heads, mearn&i east and montanws west of the mountains.

In the western part of the winter range of the species on the Kaibab Plateau in northwestern Arizona, D. I. Rasmussen (1941) found that ~tThe most abundant birds found during the winter season in this [pifion-juniper-woodrat] association are the june05. The red-backed junco [dorsalis], which breeds in the upper portions of the mountain, is present in winter along with Shufeldt's junco [actually J. o. montanij.s (Miller, 1941b)], with perhaps individuals of the pink-sided junco and the gray-headed junco [J. c. caniceps]. Flocks of juncos were observed in the snow-covered foothills in flocks of twenty-five to one hundred individuals in the winter of 1930: 31. Both the 'black heads' and the 'gray heads' were present. The former exceeded the lattex in numbers of three to one."

At Grand Canyon National Park 2,476 june05 were banded from October 1932 through February 1939. In round numbers 60 percent of these were J. o. montanus (recorded as "Shufeldt's") and 40 percent J. caniceps, plus four "pink-sided" and one slate-colored (totalling 0.2 percent). Observations made at the same place 20 years later in 1956 and 1957 by Louise Hinchiife and W. E. Dilley (MS.) found mordanus the most numerous junco, often 6 to 1 of the others, followed by caniceps, including dorsalis and intergrades, and mearnsi in varying numbers, with a few slate-colored and "Cassiar" juncos. Probably at least % of the gray-headed juncos banded at Grand Canyon were caniceps-dorsali.s intergrades native to the vicinity. Records were kept of the color of the upper mandibles of 23 banded in November and December 1932. Miller (1941b) would probably have classified 10 of these as pure dorsalis, as the mandibles of six were entirely black and those of four "mostly black" or "nearly all black." Six presumably were Miller's pure caniceps, the upper mandible of one having been recorded as "pink" and those of five as "very little black," "practically no black," or "flesh-colored except black tip." The mandibles of the other seven were intermediate in color, the birds presumably being caniceps: dorsalis intergrades from the nearby Kaibab Plateau population.

In the San Francisco Mountains southeast of the Grand Canyon H. S. Swartli (1924b) found that "By the middle of October [1922] canweps was present in fair abundance, in the piflon-juniper belt to some extent, but in greater numbers in the yellow pine belt. Flocks of juncos were frequently encountered composed of as many as fifty or sixty individuals. Nine-tenths of such a flock would consist of canweps and shufeldti [montanus] in about equal numbers, with a few mearnsi and perhaps an occasional hyemalis". At Flagstaff a few miles south, in 1935, L. L. Hargrave (MS.) recorded that a flock of about 25 juncos, all caniceps and dorsalis, suddenly disappeared after a one-inch snowfall November 2, except two caniceps and one dorsali.s, "even though the feeding station was well baited; very surprising," and that most did not return for nearly a month.

While gray-headed juncos accompany the other juncos in the winter, both races often are partly segregated within the flocks from the Oregon and slate-colored juncos. In early November in the Arizona White Mountains, A. R. Phillips (MS.) noted that a group "which detached itself from the main group and fed off to one side of the field among small yellow pines was composed of several dorsalis and a few caniceps." Phillips continues: "These two races often segregate within larger flocks, as do the ore ganws group; thus a flock will contain several races, but often a part of the flock feeding in one spot will be entirely composed of northern or southern races." Later in November, he recorded that dorsalis outnumbered caniceps by two or three to one, and that the dorsalis "also seemed more active than any other race and were always chasing each other about." Ia February and March, on several occasions dorsalis was noted with other juncos, but almost always feeding aside from the main flock and staying closer to thick brush; on one occasion in late March they were, however, recorded as "not in brush this time, but feeding with others by the road." In the same area, E. C. Jacot (MS.) agrees that caniceps is "evidently rarer away from timber than ore ganus group."

Northern Gray-headed Junco (J. c. caniceps)

Range: Southern Idaho and southern Wyoming south to northern portions of Sonora, Sinaloa and Durango, and western Texas.

Breeding Range: The northern gray-headed junco breeds in mountains from southern Idaho, Nevada (Santa Rosa and Jarbidge mountains), Utah (Porcupine Ridge, Uinta Mountains), and southern Wyoming (Uinta Mountains, Medicine Bow Range) sout.h through central and east central Nevada (Toiyabe Mountains, Wilson Peak), the White \Iountains of California, Utah, and western and central Colorado to northern Arizona (intergrades with J. c. dorsalis on Kaibab Plateau) and northern New Mexico (Chuska Mountains; Pecos; intergrades with J. c. dorsatis in Zuni iMlountains).

Winter Range: Winters in lower mountains and plains of breeding area, north to northern Utah (Utah Lake Valley), northern Colorado (Rocky Mountain National Park); and from western Nebraska (Scotts Bluff County) and eastern Colorado (Fort Morgan, Yuma) south to northern Sonora (Rancho Carrizal), northern Sinaloa (Babizos), northern Durango (Ci6naga de las Vacas), and western Texas (Palo Duro Canyon, Chisos Mountains); rarely to southern California (San Diego, Los Angeles and San Bernardino counties).

Casual records: Casual north to eastern Montana (Glendive) and east to south central Nebraska (Red Cloud), central Oklahoma (Norman), Arkansas (Little Rock), and Louisiana (Shreveport).

Migration: T he data deal with the species as a whole. Early dates of spring arrival are: Utah: Moab, April 2. Wyoming: Laramie, March 23 (average of 8 years, April 12).

Late dates of spring departure are: Nebraska: Red Cloud, April 18. Texas: El Paso, April 25. Oklahoma: Cimarron County, April 20. New Mexico: southern New Mexico, April 18.

Early dates of fall arrival are: California: Deep Springs, September 20. Arizona: Hualpai Mountains, September 27. New Mexico: Manzano Mountains, October 8. Oklahoma: Cimarron County, October 20. Texas: El Paso, September 13.

Late dates of fall departure are: Wyoming: Laramie, November 7 (average of 6 years, October 13). Utab: Roundy Reservoir, October 25.

Egg dates: Colorado: 45 records, May 8 to July 18; 24 records, June 16 to July 4.

New Mexico: 3 records, July 15 to August 7. Utah: 5 records, June 18 to July 19.

Red-backed Gray-headed Junco (J. c. dorsalis)
Range: Arizona, New Mexico, and western Texas.

Breeding Range: The red-backed gray-headed unco breeds from north central Arizona (south rim of the Grand Canyon) and central New Mexico (Magdalena Mountains) south to east central Arizona (Sierra Ancha, Hannagan Meadow), southern New Mexico (head of Mimbres River, Guadalupe Mountains), and extreme western Texas (Guadalupe Mountains).

Winter Range: Winters on or near the breeding grounds (north to the Grand Canyon), south in small numbers to southern Arizona (Pajaritos, Huachuca, and Cbiricahua Mountains), extreme southwestern New Mexico (Big Hatchet Mountains), and southwestern Texas (Chisos Mountains).

Egg dates: New I'vfexico: 3 records, July 15 to August 7.

Contributed by RICHARD C. BANKS


This species was described by Robert Ridgway (1883b; see Deignan, 1961) on the basis of two specimens taken in the Laguna Mountains of Baja California by Lyman Belding. A short while after the original description, two additional specimens were obtained and described by Belding (1883b). Further work showed that Baird's junco lives throughout the mountains of the Cape region of Baja California. It has never been found elsewhere.

This is one of the yellow-eyed group of juncos, most closely related to the juncos of Mexico and Central America (Miller, 1941b). In some of its characteristics it is similar to J. alticola. of Chiapas and Guatemala, in some to J. juivescens of Chiapas, and in some to J. phaeonotus, which ranges broadly through Mexico to the east of the range of bairdi (Miller, 1941b; Davis, 1959). A. H. Miller (1941b) notes that "It possesses several peculiarities of color and proportion that sharply set it off from all others, yet, speaking broadly, it is a pale, dwarfed representative of the Central American juncos."

Miller's (1914b) study of the genus Junco led him to conclude that the progenitors of J. bairdi had colonized the isolated mountains of the Cape region by direct flight from the east or south, in times earlier than Pleistocene. John Davis (1959) considered Baird's junco to be one of a number of species that originated in the Madro-Tertiary flora of Mexico and followed this flora into the Cape region, where later climatic changes left it isolated. In isolation it retained a number of features which characterized the original pioneering population.

Relatively few ornithologists have seen Baird's junco in its natural habitat, and even fewer have written about it. Much of the information in this report is taken from unpublished field notes of Chester C. Lamb, who collected in the mountains of the Cape region of Baja California in the winter of 1928 and the summer of 1929.

Description: Back and scapulars cinnamon brown, becoming buffy brown on rump. Sides cinnamon, the color extending onto the breast and sometimes meeting in a band across the breast. The cinnamon fades with wear, becoming more yellowish. Lores blackish. Top and sides of head and hindneck uniform light gray, the nape often tinged with buff, at least in unworn plumage. Throat and upper breast pale mouse gray, lower breast, abdomen and under tail coverts white. Tail with outer two feathers partly, the outer often completely white, the third sometimes with a white spot. Iris yellow; lower mandible yellow, upper mandible brown. Wing and tail short, bill and feet moderately large, proportionately.

Sexual dimorphism is limited to size and tail pattern, the female being slightly smaller and averaging somewhat less white in the tail (Miller, 1941b; Ridgway, 1901.)

In juvenal plumage this is the least heavily streaked junco (R. R. Graber, 1935).

Habitat: Nelson (1921) characterized this bird as an inhabitant of the Upper Son oran Life Zone of the Cape region of Baja California. According to Miller (1941b), it is one of the few juncos that breed under zonal conditions lower than Transition. Frazer found it inhabiting the pine and oak woods (Brewster, 1902). Miller (1941b) amplified the description of the habitat: "The region inhabitated is forested predominantly with oak (Quercus devia), also with pifion (Pinws cembroides) and madroim (Arbutus penninsularis), and with cottonwoods (Populus monticola) in the canyons." He further states that "The nature of the region compares most closely * * * with the Upper Sonoran oak belt of Arizona * *

Belding (1883) found this species very common over 3,000 feet elevation, and Miller (1941b) reports that most specimens have been taken above 4,000 feet. Only a single bird has been taken as low as 1,800 feet, at El Triunfo (Brewster, 1902).

Habits: Belding (1883), when reporting the capture of the third and fourth specimens of this bird, said that "Nothing worthy of note in connection with its habits was noticed." Brewster reported (1902) that like most juncos they were tame and familiar, and that "they often came into a shed where Mr. Frazer prepared his specimens, and hopped about his feet, under the table, or pecked at the dried venison suspended from the roof."

These juncos apparently stay close to the ground, often taking advantage of the lower, bare limbs of pines or oaks for perches. Chester Lamb's observations were consistently of birds "feeding on the ground," "perched on a low bare twig in an oak seedling," or "on an old stump among some brush." They readily responded to "squeaking." After a day of intermittent showers in early August, Lamb wrote: "During these rains all the birds but the Baird juncos and large-billed towhees [Pipio erythrophthaimus magnirostrisl disappear, but these birds can be seen at the edge of the meadows busily feeding during most of the storms." In mid-December, after noting observations of birds apparently singly rather than in flocks, he summarized that they were "common, but very shy and retiring at this time of year.~~ On I~Iay 22, 1965, seven jimeos ~~-ere foraging together in a space about 30 by 30 feet in the meadow, near a small group of pines. There was one encounter which seemed to enforce individual distance. When disturbed by my approach, they flew to a small pine where, after some initial squabbles in finding places, they remained quietly together. Despite the tendency for gregariousness, there were numerous chases during May that seemed to enforce territorial boundaries. One bird was involved in two such chases within a few minutes, about 40 yards apart.

On one occasion, a bird sitting in an oak flicked its wings rapidly when alarmed.

Baird's juncos remain in winter in the highest part of their breeding range. "Apparently there are no winter flocks of any proportion, the birds staying close to the nesting grounds * * "'." (Miller, 1941b.) No seasonal movements are known.

Nesting: Nesting apparently occurs in late spring and early summer. Brewster (1902) says that Frazer found no nests in the Sierra Laguna during his stay from late April to early June, but "late in May a bird was seen collecting building material." At Laguna Valley, at 6,000 feet elevation, Chester Lamb found nests in many stages during July, 1929, and was led to believe that the birds nest twice. On July 7,12, and 15, he found nests with "heavily incubated" eggs, and on July 8 he "saw one carrying nesting material in its bill." Thayer (1909) mentions that W. W. Brown was too late for eggs of this species when he reached the mountains on August 2. Dependent young have been recorded from June 12 to July 17 (Miller et al., 1957), but the dependent period must extend into early August.

In contrast to this previously available information, which suggests a nesting season from late May through July, we found well-grown, but still dependent, young as early as May 20, which indicates that some birds, at least, start breeding early in May. Despite the fact that we saw young frequently, we were unable to find any nests.

Nests of Baird's juncos are either on the ground or in low trees. Of eight nests mentioned in Lamb's notes or collected by him, two he described as being "located in [the] leafy extremity of a branch 6 feet up in a small pine," and one was "6 feet Up, against the trunk in a small pinyon pine." Two nests were hidden in depressions in the ground; one was "under a small stick" and the other "in a hole in the ground 3 inches back from [the] entrance." Another nest was "placed on the ground among some low flowering weeds." Two nests were "made of fine weeds and grasses, lined inside with horse hair."

Eggs: Each of the nests Chester Lamb found in 1929 contained two eggs, and this apparently is the normal complete set. The eggs are whitish with small flecks of dark reddish-brown, most heavily spotted on the large end.

Food: In May 1965, most foraging was observed on the ground, among dry leaves and weeds, but one bird was seen feeding on the terminal ends of oak limbs about 20 feet above the ground. One bird tried twice, unsuccessfully, to catch an insect (fly?) that was slowly flying about 4 inches above the ground.

Voice: The song is described in my notes both as warbler-like and as reminiscent of that of the black-throated sparrow. It is a complex song, and is quite variable. It seemed that the song of every bird was different, and it is possible that individual birds sang more than one song, but that point was not determined.

Weights: Weights of two adult males were 17.8 and 16.7 grams; an adult female weighed 16.7 grams.

Enemies: Baird's juncos are undoubtedly exposed to all the hazards that beset other ground nesting birds. The only record of predation, however, is in Chester Lamb's notes for July 17, 1929: "Early this morning in the meadow [I] saw a large wild cat [Lynz rufus penim.sularisl. It had just despoiled a ground nest of a Baird junco."

Range: Baird's junco is resident in the Victoria Mountains (La Laguna,Mount Miraflores) of the Cape district of Baja California; casual downslope to Triunfo.