36 BULLETIN 1134, U. S. DEPARTMENT OF AGRICULTURE. 



An answer to the question why Egyptian pollen is conveyed to 

 the stigmas of upland cottons in greater quantity than upland pollen 

 to the Egyptian stigmas is to be sought in a consideration of the 

 habits of the pollen-carrying insects in relation to the ontogeny of 

 the flower. At Sacaton, Ariz., the honeybee and at times wild "bees 

 of the genus Melissodes are the insects which enter most frequently 

 the corollas of Pima cotton. It has been observed that many of 

 the flowers are first entered when the petals are just beginning to 

 unfold, occasionally even before there is an actual aperture at the 

 apex of the corolla, under which conditions the bees almost invari- 

 ably come into contact with the reproductive organs, taking up and 

 depositing pollen. When the Pima corollas are in this early stage of 

 expansion the anthers of upland cotton frequently are still closed 

 or are just beginning to split, so that little, if any, pollen of this 

 type is available for transfer to the Pima stigmas. On the other 

 hand, many of the upland corollas at this time are quite as accessible 

 as those of. Pima to entry by the insects, and this accounts for the 

 frequent deposition of Pima pollen upon the upland stigmas before 

 any self pollen has reached the latter. 



By the time the upland anthers have begun to discharge pollen 

 freely most of the Pima corollas have opened to a degree which 

 allows honeybees to reach the nectaries by crawling down the inside 

 of the petals without touching the reproductive organs and to make 

 their exit in the same manner. It is a relatively infrequent occur- 

 rence for the honeybees to touch stigmas or stamens in entering or 

 leaving a well-opened Pima flower. On the other hand, the wild 

 bees (Melissodes spp.) apparently do so regularly. It would be 

 interesting to ascertain whether foreign pollen is conveyed in greater 

 quantity to the Pima flowers during periods when Melissodes are 

 visiting them in large numbers than when honeybees are the pre- 

 dominant visitors. 



POLLEN-CARRYINQ INSECTS AT SACATON. 



There is little doubt that natural cross-pollination in cotton is 

 effected almost solely by the agency of insects. The nature of the 

 pollen grains of Gossypium is unfavorable to their transportation 

 by currents of air. Allard (£, p. 256), however, found that glass 

 plates smeared with vaseline and exposed in cotton fields in northern 

 Georgia collected considerable quantities of cotton pollen. On the 

 other hand, Balls (£, p. 117) , using the same method for the detec- 

 tion of wind-disseminated pollen in Egypt, obtained negative re- 

 sults. 



No systematic study of the insects which visit cotton flowers has 

 been attempted in connection with these investigations, but nu- 

 merous specimens have been collected at Sacaton, Ariz., and notes 

 have been made upon the efficiency as pollinators of those which 

 most frequently enter the flowers. The writer is indebted for the 

 identification of the specimens to Dr. L. O. Howard, Chief of the 

 Bureau of Entomology, United States Department of Agriculture. 19 



19 The ' several groups were identified by the following specialists: Hymenoptera. by 

 S. A. Rohwer ; Coleoptera, by E. A. Schwarz ; Hemiptera, by Miss E. A. Wells and W. L. 

 McAtee ; Orthoptera, by A. N. Caudell ; Lepidoptera, by H. C. Dyar ; and Diptera, by 

 J. M. Aldrich. 



