308 NOTES ON BUMBLEBEES 



(Curtis), and one new variety of Bremus sylvicola (Kirby), both 

 species being new records for Arizona. 



Bremus kirbyellus var. alexanderi new variety 



Queen and worker. — Taxonomically almost identical with the typical 

 hirhyellus (Curtis), which occurs in regions farther north. Differs slightly 

 in structure from the typical queens of kirbijellus that I have studied, by 

 having a comparatively shortermalar space, a condition particularly noticeable 

 in the worker caste. Yellow pile of mesopleura not extending to the base of 

 the legs, and with pile on posterior margin of this area entirely black. Meta- 

 pleura and medium segment with black pile. Posterior margin of the fourth 

 together with the entire fifth dorsal abdominal segment ferruginous, and also 

 the sixth dorsal abdominal segment in the workers. Legs and ventral portions 

 of the body black, except that the tips of some of the setae bordering the 

 corbicula have a slight ferruginous tinge, particularly pronounced in the case 

 of some workers. Coloration otherwise as in typical specimens of kirbyellus. 



Holotype. — queen, July 11, 1917, Patagonia Mountains, 

 Arizona. Paratopotype queen, July 13, 1917. Morphofype 

 worker, July 14, 1917, Patagonia Mountains, Arizona. Parato- 

 potype workers, one collected July 11, 1917, three on July 13, 

 1917, and one on July 14, 1917. One paratijpe worker, June 5, 

 1917, Oracle, Arizona. All the specimens collected by Mr. E. J. 

 Oslar. 



One paratopotype worker deposited in the collections of each 

 of the following institutions: American Entomological Society, 

 American Museum of Natural History and the United States 

 National Museum. Holotype and remaining paratype speci- 

 mens in the collection of the author. 



Named in honor of my friend Dr. C. P. Alexander, to whom I 

 am indebted for many favors. 



The male of this variety will undoubtedly be obtained in 

 future collections from this region, a caste which may necessitate 

 the elevation of this form from varietal to specific rank. In 

 many respects B. kirbyellus alexanderi resembles also B. polaris 

 (Curtis), another species of the same group, and if the specimens 

 were from Arctic America might readily pass as that species. 

 I have deemed it advisable to describe my specimens as a variety 

 of kirbyellus because this species is found in the high mountain 

 ranges of the western United States, whereas polaris is only 

 known from Greenland, Labrador, Alaska and other far northern 



