1002 U.S. NATIONAL MUSEUM BULLETIN 23 7 part 2 



J. Grinnell (1928b) gives the range of this subspecies as the southern 

 portion of Baja CaUfornia, north to about latitude 26°, with inter- 

 gradation wdth A. b. deserticola taking place to about 27° N. This 

 includes most of the southern islands in the Gulf of California and 

 Magdalena and Santa Margarita islands on the Pacific side of the 

 peninsula. The population on Cerralvo Island has recently been 

 described as A. b. belvederei (Banks, 1963a). 



Walter E. Bryant (1889) says that "On Santa Margarita and Mag- 

 dalena Islands they were the most common and generally distributed 

 species. Breeding far from any water, nests were found in bushes 

 from one to five feet above the ground." In the Cape region, I 

 have found this species in fauiy open desert, but not where the under- 

 brush is thick nor in the thorn forest. 



A nest containing three eggs was found on Cerralvo Island (Banks, 

 1963b). Several other persons have mentioned finding nests, but 

 have not recorded the number of eggs. 



The data relating to the time of nesting of this subspecies are 

 confusing and contradictory. A. J. van Rossem (1945b) indicates a 

 February breeding season on Magdalena and Santa Margarita islands, 

 but presents evidence to show that elsewhere in the Cape region 

 on the mainland the birds nest in October. On the islands in the 

 GuK of CaUfornia he mentions March as the breeding month. More 

 recent data from Magdalena Island (Banks, 1964a) contradict the 

 February date without clearly indicating an alternative. While 

 there is some evidence of nesting near La Paz in the fall, other evi- 

 dence also suggests a spring breeding season. 



It seems that the population of each island in the GuK of California 

 has adjusted its breeding cycle to its own pecuHar circumstances. 

 Thus, Richard C. Banks (1963c) found evidence that breeding was 

 in progress or about to begin on San Marcos, Coronados, and Santa 

 Catahna islands in late March and early April of 1962, but not on 

 Monserrate Island. The bkds on Espiritu Santo, Monserrate, and 

 Danzante islands were in breeding condition in early May 1963 

 (Banks, 1964a), and nesting began on Cerralvo Island in mid-May 

 1962, perhaps continuing throughout the summer (Banks, 1963b). 



These bkds have been noted feeding at cacti of the genera Mam- 

 millaria and Pachycereus. Several seen foraging on Ocotillo (Fou- 

 guieria) on Coronados Island were apparently taking aphids from the 

 leaves. On Cerralvo Island most foraging took place among annual 

 plants in washes, but a bird was observed feeding at the flower of a 

 cactus (Banks, 1963b, 1963c). 



