Part 18

Ridgway (1902) describes the northern bird as "similar to _C. a. americana_, but slightly larger, with smaller bill and darker, richer coloration; adult male with blue of upper parts deeper, and black of lores more intense; lower throat or upper chest (sometimes both) blackish or dusky (the feathers sometimes tipped with chestnut), forming a more or less distinct, often very conspicuous band; lower chest orange-tawny, tawny, or chestnut (the feathers usually margined with yellow) forming usually a distinct and often abruptly defined patch; sides usually more or less tinged or spotted with chestnut."

In the same work, he describes a third form, _C. a. ramalinae_, as "similar in coloration to" the northern bird, "but smaller even than _C. a. americana_." He gives as its range the Mississippi Valley, from Mississippi, Louisiana, and Texas to Minnesota and Michigan. This western race is not recognized in the 1931 A. O. U. Check-List.

Our experience with the northern parula warbler in Bristol County, Mass., well illustrates the successive changes that nature and man have wrought in the distribution of so many of our birds. Many years ago, perhaps early in the last century or before, some hardy pioneers hewed out a clearing in the forest that clothed the slopes of Rocky Hill in Rehoboth, Mass., planted an apple orchard, and surrounded it with stone walls. All traces of the old farm, if ever there had been one, disappeared before I first visited the locality in 1888, and the forest had begun to encroach on the old clearing. The apple trees even then showed signs of old age and were profusely covered with long festoons of that picturesque tree lichen, often called beard-moss or old-man's-beard (_Usnea barbata_, _U. longissima_, or _U. trichodea_). This old orchard was a mecca for all local oologists, and many a set of eggs of the blue yellow-backed warbler was taken from it during succeeding years. As time passed, the old trees gradually died, the _Usnea_ disappeared, the warblers ceased breeding there, and the forest eventually reclaimed the land until today only the ancient stone walls remain to mark the locally famous haunt of the blue yellow-backs.

I can remember several other old, neglected orchards that were similarly decorated with the long, gray-green lichen and that were inhabited by parula warblers as nesting sites, but they all suffered the same fate; the orchard trees decayed and were replaced by woods and thickets. During the early part of the present century this warbler continued to breed commonly in Bristol County wherever it could find trees infested with _Usnea_--around the edges of swamps and along the shores of ponds, lakes, and sluggish streams; but now this lichen seems for some reason to have entirely disappeared from the County, and the parula warbler has likewise disappeared, although it may still breed in a few similar localities on Cape Cod, Mass., where I have found it a few times in more recent years.

Localities such as those described above seem to be typical of the breeding haunts of the northern parula warbler, at least in New England, southern New York, and New Jersey. Whether the presence of _Usnea_ is a sine qua non for the breeding haunts of this wood warbler is an open question; but it may safely be said that where this lichen grows in abundance one is almost sure to find it breeding; and conversely, where this lichen is scarce or lacking, the warbler breeds sparingly or not at all.

Farther westward, northward, and southward, where _Usnea_ is scarce or entirely absent, these warblers seem to find congenial haunts in hemlock ravines and in other coniferous woods and swamps; but even there they are more likely to be found where there is at least some of one species or another of this lichen, or where the somewhat similar Spanish moss (_Tillandsia usneoides_) grows.

_Spring._--Parula warblers that have wintered in the West Indies reach southern Florida during the first week in March. Dr. Wetmore (1916) says that it "was the most common of the migrant warblers in Porto Rico. * * * Migratory movement was apparent among them by February 14, and after this the birds were very restless, especially during early morning, and there was tendency to work from the east to west. In March and April there were distinct waves of migration." But it is well on toward the middle of May before the first migrants reach the northern limits of their breeding range.

Professor Cooke's (1904) records show that the migrants from Mexico and Central America reach the Louisiana coast by the very last of February or early March, while the first arrivals on the lower Rio Grande, in Texas, come two or three weeks later. He observes:

A comparison of the dates shows, first, that the parula warbler arrives in Texas much later than in either of the other States, and hence does not reach the Mississippi Valley by way of Texas; second, that it arrives in northern Florida at least ten days later than it attains the same latitude in Louisiana. From these two facts it would appear that Louisiana is reached by direct flight across the Gulf of Mexico. The average date of arrival at New Orleans coincides closely with the date when the first migrants arrive at the southern end of Florida. It would seem that the birds of Mexico and Cuba are prompted to move northward at the same time, but the flight over the Gulf of Mexico being so much longer than that from Cuba to Florida, the Mexican birds reach a higher latitude by their initial flight.

There are other interesting details in Cooke's account to which the reader is referred.

During migration the parula warbler does not frequent haunts typical of its breeding ground; in fact such are not to be found in much of the country over which it travels; nor does it especially frequent the coniferous woods to which it is partial in summer. It is to be found almost anywhere, in many kinds of trees, though it seems to show a decided preference for deciduous woods. There, it may often be seen drifting through the highest tree-tops in mixed groups of migrating wood warblers, gleaning insects amidst the freshly opening foliage. Referring to the Buckeye Lake region in Ohio, Milton B. Trautman (1940) writes: "The transient Parula Warblers usually displayed a preference for large pin oak and shingle oak trees and a marked preference for one shingle oak in particular. This oak was in the Lakeside Woods, and more Parula Warblers were observed in it than in all of the remaining trees of the woodland. A transient often displays a marked preference for certain types of trees, but it appears unusual for a single tree among many of the same kind to retain yearly so marked an attraction for a particular bird species."

The migrating parula warbler is often seen in roadside trees and in shade trees in parks and gardens. It even visits our orchards, where one of the most charming sights of springtime is to see this gay-colored, tiny warbler flitting about in search of insects among the apple blossoms, a delightful bit of color contrast in a beautiful setting.

_Nesting._--The nests of the northern parula warbler that we used to find in southern Massachusetts were all located in haunts similar to those described, and mostly in old orchards heavily festooned with beard moss (_Usnea_). We could usually find three to five nests in a well-populated orchard, but they were so well hidden in the hanging moss that we may have overlooked some. A casual observer would never notice one, but with practice we learned to recognize a rounded, cuplike, thick place in a bunch of _Usnea_ as indicating a nest. The nests were usually made in bunches of moss that hung from horizontal or sloping branches and were from 5 to 15 feet above the ground, more being below than above 12 feet. Some nests were in red cedars, or savins (_Juniperus virginiana_), scattered among other trees or growing in open stands by themselves; they were located in bunches of _Usnea_ close to the center of the tree and often within reach from the ground. Occasionally, isolated trees on the edges of swamps or on the shores of ponds were sufficiently covered with the lichen to contain nests, and these were sometimes as much as 20 feet above the ground. The nearest approach to a colony that I ever found was in a small cedar swamp, not over an acre in extent, that jutted out from the shore of a lake into rather deep water.

The white cedars (_Chamaecyparis thuyoides_) were growing in water that was waist deep or more in places, and the whole place was so obstructed with fallen trees and sunken snags that it was very difficult to explore thoroughly; I managed to find some half a dozen nests, and there may have been others, for many of the trees were well "bearded."

The nests that we have found have all been very simple affairs, apparently merely pockets hollowed out in bunches of hanging _Usnea_, with side entrances slightly above the cups. Some nests were small and suspended only 2 or 3 inches below the supporting branch, practically open baskets accessible from directly above; others were found in long, thick bunches, a foot or more in length, with long streamers hanging below the nest. External measurements were therefore quite variable. Many of the nests were unlined, save with a soft bed of fine shreds of _Usnea_, some were scantily lined with a few pieces of fine grass, two or three pine needles, one or two horsehairs, or a few bits of buff-colored down from the stems of ferns; rarely, a nest was more elaborately lined with the latter material but never as profusely as are the nests of other wood warblers.

Apparently the nests were also difficult to find in Connecticut; "J. M. W." (C. L. Rawson, 1888), who has probably taken more eggs of the blue yellow-backed warbler than any other man, says that the older ornithologists did not realize "that the three Southern New England States were about the centre of its breeding range," until he began sending eggs to Dr. T. M. Brewer. Thomas Nuttall (1833) remarked: "The nest and eggs are yet unknown."

Rawson found the parula warbler nesting in colonies near Norwich, Conn., and says:

I know a swamp where may be found seventy-five pairs of these summer residents. The first time I visited the Preston colony on the 31st of May, I took eight sets of four. The first time I visited another large community in this county on June 5, on a point of land trending into salt water, I took eleven sets of four. * * *

The nests are built on dead or green trees, and on savins or deciduous trees, at varying heights. I took one from the single filament of moss caught on the green twig of a birch, within five inches of the ground, and others close to the trunks of great oaks fifty feet in the air. On the lower swamp, huckleberry brush in the littoral colony is a favorite site.

William Brewster (1906) mentions only one nest taken in the neighborhood of Cambridge, Mass., a region where _Usnea_ is scarce:

In shape and general plan of construction the nest closely resembles that of a Baltimore Oriole. It has no hole in the side but instead a wide-mouthed opening at the top through which the bird entered it as the Oriole enters her nest. The upper edges and sides were securely fastened to the fine terminal twigs of a drooping branch where the nest hung suspended among the evergreen foliage of the hemlock, precisely as the Oriole's hammock swings in the dropping spray of an elm. The Warbler's nest has a scanty lining of pine needles and fine grasses but it is otherwise composed entirely of _Usnea_, loosely woven or perhaps merely felted together, evidently by the parent birds. They must have been at some pains to collect this material, for the closest scrutiny on the part of a friend and myself failed to reveal more than a few small and scattered tufts of _Usnea_ in the surrounding woods.

Henry Mousley (1924, 1926, and 1928), of Hatley, Quebec, made three attempts to make complete studies of the home life of the northern parula warbler, none of which covered the whole cycle for reasons beyond his control.

The nests were suspended from the branches of coniferous trees, at heights ranging from 26 feet in a spruce to 40 feet in a balsam fir. One of these nests was watched for a total of 24 hours, from May 22 to 31, during the process of construction; during this time the male sang 549 times from a little birch and went with the female to the nest, but brought no material; the female, however, made 206 trips with material, an average of one load every 5.4 minutes. The nest was made entirely of _Usnea_, all brought in, and lined with "some black hair-like rootlets, with two bits of plant down"; it was strengthened with a few fine grass stems. It weighed only 100 grains, or .23 ounce! "Outside diameter 3.25, inside 1.75 inches; outside depth 2.50, inside 1.75 inches. The female after selecting some of the longest threads of the hanging bunch of _Usnea_, attached them to a little twig a few inches off, following this up with that curious process--inherent--of moulding the nest, which in this case, was really an acrobatic performance, there being of course no apparent nest to mould, just a few strands, through which the bird's tail and wings protruded."

Outside New England, where _Usnea_ is scarce, the nests are often built in hanging clusters of twigs of hemlocks or spruces, with the use of more or less of this lichen when available. In the lower Mississippi Valley, Spanish moss (_Tillandsia usneoides_) offers a popular substitute and is generally found growing in profusion. But some nests are built of various other materials. George H. Stuart, 3d, writes to me of a nest he found at Pocono Lake, Pa., on June 22, 1916: "This remarkable nest was placed in a horizontal limb of a spruce, 20 feet up and 12 feet from the trunk, near the tip and overhanging a road near the lake." It was "composed mainly of fine dry grasses and the thinnest of bark shreds, with a few bits of down, fashioned together oriole-like, though loosely, with a few coarse grasses projecting suggesting the handiwork of the magnolia warbler. The tiny basket was suspended from the under side of the branch, partially supported by inclining twigs. In form it is an inverted cone or pear, measuring 3 inches deep by 3 inches wide at the rim, the thin walls tapering down to a narrow, pointed bottom. The thinness of the walls in places revealed the eggs from a side view."

Mrs. Nice (1931) reports a curious nest, found by Mr. Kirn near Copan, Okla.; it was fastened to ivy leaves and to a stick which was hanging down, held by the vine. "In this hanging, swaying cluster about two feet long, the nest was built almost entirely of box elder blossoms held together by spider webs on the outside, and sycamore seed down on the inside with a light lining of fine strips of weed stems."

Several nests have been reported as built of various materials in bunches of leaves and other rubbish deposited by freshlets on branches over streams.

Because of the bird's habit of using various materials and sites in its nest building, it may be well to mention some nesting records from the southern Gulf States. Andrew Allison wrote to Dr. Chapman (1907):

The invariable nesting site is a clump of Spanish moss--where this it to be had; I have not observed nests from beyond the range of this plant. The nest is generally placed near the branch from which the long filaments of the 'moss' depend, so that it is well concealed. The height from the ground varies from about eight feet upwards. * * * The nest is nearly hemispherical in shape, opening directly upward. The usual material, in lower Louisiana, is thistle down, which is abundant during the nesting season. Animal hairs are not used, I think. A nest from Bay St. Louis was composed of the very black horse-hair-like inner fiber resulting from the decay of _Tillandsia_.

M. G. Vaiden writes to me that he found a nest near Belzoni, Miss., in a heavy oak swamp where there were clusters of _Usnea_ on practically all of the trees. The nest was 16 feet above the ground and 12 feet out on a limb of an oak; it was made like our northern nests and lined with the "moss" and fine rootlets. Another described in his notes was entirely different. It was in a section of Mississippi where there was no _Usnea_ growing within 60 miles. The nest was 6 feet from the ground and 4 feet out in the crotch of a limb of a hackberry tree. A pretty nest, it was nicely constructed of leaves and bark from cypress trees, and was lined with small rootlets and very fine twigs.

_Eggs._--The usual set for the northern parula warbler consists of 4 or 5 eggs; 3 sometimes constitute a full set, and as many as 6 or 7 have been found in a nest; there are 3 sets of 7 in the J. P. Norris collection. The eggs are ovate or short ovate, have only a slight gloss, and are white or creamy white, speckled and spotted with shades of "russet," "chestnut," "bay," and "auburn," with a few underlying spots of "brownish drab." There is much variation; on some eggs the "brownish drab" color is entirely lacking, while on others spots of this color are the most prominent markings; again, the eggs may be almost immaculate, or may have just a few indistinct freckles of "pale wood brown" at the large end. The measurements of 50 eggs average 16.5 by 12.1 millimeters; the eggs showing the four extremes measure =18.3= by 12.7, 16.9 by =12.9=, =14.8= by 11.9, and 16.3 by =11.2= millimeters (Harris).

_Young._--The period of incubation does not seem to have been determined, nor do we know how long the young remain in the nest. Incubation of the eggs and brooding of the young is performed mainly by the female, but the male assists in both to some extent. I have seen a male leave a nest in which there were eggs; and Mr. Mousley (1924) saw a male brood the young for a period of 4 minutes in the absence of the female, but he left as soon as she returned. Both parents feed the young. Mr. Mousley's table shows that during a watching period of 15 hours the male fed the young 45 times and the female fed them only 21 times; the average rate of feeding was once in 13.6 minutes; during this time the male brooded once and the female 34 times, a total of 11 hours and 27 minutes. He "noticed that the food the male brought consisted almost invariably of soft green larvae, whereas, that of the female more often than not consisted of insects, and the portions she brought were usually smaller in proportion than those of her partner."

_Plumages._--Dr. Dwight (1900) says that the natal down is "smoke-gray." The sexes are alike in the juvenal plumage, which Ridgway (1902) describes as "above plain slate-gray, slightly tinged with olive-green; middle and greater wing-coverts narrowly tipped with white; chin and upper throat pale yellowish; lower throat, chest, sides, and flanks plain light gray (intermediate between mouse gray and gray no. 6); abdomen, anal region, and under tail-coverts white; remiges and rectrices as in adults."

A postjuvenal molt, involving all the contour plumage and the wing coverts but not the rest of the wings or the tail, begins about the middle of July. This produces a first winter plumage in which old and young birds are very much alike and the sexes are recognizable. The young male differs little from the adult male, but the bluish gray of the upper parts is more heavily tinged with olive-green, the yellow of the under parts is duller, and the dark throat band is more or less obscured by yellowish tips on the feathers. The young female differs from the adult female in a similar way and is without any brown throat band.

Dr. Dwight (1900) says that the first nuptial plumage is "acquired by a partial prenuptial moult which involves chiefly the head, chin and throat, but not the rest of the body plumage, the wings nor the tail. The ashy blue crown feathers faintly dusky centrally, the blackish ones of the sides of the head with a white spot above and below the eye and the yellow or chestnut-tinged chin feathers as far as the pectoral band or farther are assumed by moult. Wear brings the back into contrast with the nape and whitens the lower parts. The wings and tail are browner and more worn than in the adult, especially the primary coverts."

A first postnuptial molt in July and early August, which is complete, produces the fully adult plumage. Fall males are similar to spring males, but the blue areas are more or less tipped with greenish and the throat bands with yellowish. Fall females differ in the same way from the spring birds, and there is little, if any, chestnut and no blackish in the throat band.

Subsequent molts and plumages are the same as described above for the young birds.

Charles C. Ayres, Jr., writes to me of a bird he observed near Ottumwa, Iowa: "It was a typical parula warbler with the exception that the blue-gray color extended over the throat and terminated abruptly on the upper breast. Immediately below the termination of the blue-gray color was the well-defined orange-brown breast band, below which the rest of the breast was yellow."

_Food._--The parula warbler is almost wholly insectivorous. Its food is mainly obtained in the deciduous trees, where it is often seen among the branches and twigs or hanging downward under a cluster of leaves or blossoms like a chickadee searching for small insects, beetles, flies, moths, larvae, and egg clusters. Some flying insects are taken on the wing; and occasionally the bird may be seen feeding on the ground.

Dr. Wetmore (1916) reports on the contents of 61 stomachs from Porto Rico, which contained 97.7 percent animal matter and only 2.3 percent vegetable matter. The latter "consisted of seeds of small berries of the camacey (_Miconia prasina_) and others." In the animal food, beneficial insects and a large number of spiders amounted to about 35 percent, and the remainder were all harmful pests. "Lantern flies (Fulgoridae) (19.09 percent) were identified in 29 stomachs. * * * Other bugs (3.69 percent) comprise small numbers of leaf bugs, species of the chinch bug family, stink-bugs, and a few predaceous assassin bugs. The birds are fond of beetles, and this order supplies 22.53 percent of the food, nearly all being injurious species. Ladybird beetles (1.36 percent) were present in 14 stomachs. Longicorn beetles (1.68 percent) were taken 11 times, and leaf beetles of several species (7.95 percent) were eaten by 30 of these birds." Other beetles taken included darkling beetles, skin beetles, scarred-snout weevils, coffee leaf-weevils, stalk borers, and curculios. Among other items were a few ants and other small Hymenoptera (3.57 percent) and flies (1.19 percent). Caterpillars were found in 18 stomachs and moths in 4. Spiders (29.53 percent) were identified in 29 stomachs. Stuart T. Danforth (1925), from Puerto Rico, adds berries of _Varronia angustifolia_ and fleabeetles, and says that large moth eggs were eaten by two birds, forming 25 percent of their food. Forbush (1929) says that "it feeds much on small hairless inch-worms, such as the fall canker-worm and the spring canker-worm, and on the younger and smaller hairy caterpillars, such as the gipsy and the tent caterpillar."

_Behavior._--The parula warbler is less active in its movements, more sedate and deliberate, than most of the other tree-top wood warblers.