430 BULLETIN 162, UNITED STATES NATIONAL MUSEUM 



three and one-half miles south of Arlington extended over an area a quarter of a 

 mile square, while another three miles beyond occupied a grove nearly half a 

 mile wide and an equal distance in length. The birds maintained regular flights 

 across country and gathered in flocks to feed, so that they were conspicuous fig- 

 ures in the bird life of the region. It was difficult to estimate the number pres- 

 ent, as they were scattered about in dense groves of mesquites, but it is believed 

 that there were at least two thousand pairs in the largest colony examined. 

 The total number present in the area was large. It appeared that the period 

 for breeding among these birds was somewhat irregular. A part of them 

 evidently began to nest soon after their arrival, as a number that were feeding 

 young were observed on June 6. Others were nest-building on June 17, 

 so that the entire period of reproduction was somewhat prolonged. In the 

 colonies nests were scattered about irregularly through the mesquites. Some- 

 times two or three nests were placed in the same tree, or again one pair 

 occupied a tree alone. There was no crowding and apparently the birds, while 

 gregarious, were too truculent to permit close proximity of nests. Often two 

 or three trees, suitable in every way for the primitive needs of these doves, 

 intervened between occupied sites. 



In most cases the nest, slight in structure, though usually somewhat larger 

 and bulkier than that of the Mourning Dove, was placed in a mesquite, though 

 a few were observed on the desert in palo verdes. Nests were built on in- 

 clined living limbs where forking of small branches gave a firm, broad support. 

 The site varied from six to twenty feet from the ground, with about eight feet 

 as an average height. In most of those that were examined the structure was 

 composed of dead twigs of the mesquite, small in diameter, and from six to ten 

 inches long. For the inner layers small twigs were chosen that had been dead 

 for some time, so that the spines, abundant on mesquite limbs, crumbled at a 

 touch and caused no discomfort to the brooding bird or to the young. The 

 nest was flat and had merely enough depression to receive the eggs that often 

 were visible through the loosely interlaced twigs at the sides. 



Gilman (1911), who has had extensive experience with this dove, 

 says: 



Nests are always, as far as my observation goes, placed in trees or shrubs 

 at varying distances from the ground. The average height was ten feet and 

 extremes ranged from four to twenty-five foot. The only nest as low as four 

 feet was built in a mesquite tree and placed on top of an old Thrasher's nest. 

 This may have been a shiftless bird ; but I found several others using old Cactus 

 Wren's nests as foundation, and one had made use of a deserted Verdin's home. 



In choice of nesting sites the bird shows a decided preference for mesquite, 

 as about 70 per cent of nests noted were in that plant. About 20 per cent 

 were in willows, and 3 per cent in Cottonwood, Opuntia fulgida or tree cholla, 

 and Prosopis odorata or screw-bean. Baccharis gluten brought up the rear with 

 1 per cent. The dove is usually very wild on the nest, flying off whenever ap- 

 proached as close as twenty-five feet. Rarely is the broken-wing play made, 

 though I have seen a few mild attempts at it, and occasionally one will allow an 

 approach as close as fifteen feet to the nest before taking flight. 



In Arizona we found the white-winged doves nesting in mesquite 

 and hackberry trees in the mesquite forest; the nests were on hori- 

 zontal branches, 10 or 12 feet above the ground, and were made 

 entirely of grass, weed stems, and straws; a nest found in the San 

 Pedro Valley was 12 feet from the ground in a large willow. 



