Random Activity of Pollen Vectors in Isolated Plots of Upland Cotton 1

OR estimates of the amount -of natural crossing among F two or more genetic stocks of Upland cotton in an isolated plot to be unbiased, the population of pollen vectors operating therein must be assumed to be operating at random. The random activity postulated here does not refer to deviations which are attributable to stock preferences. Although certain physiological and morphological characteristics might attract more insects to one stock than another, the consequences of these characteristics will be reflected in the measure of crossability. It is important, however, that pollinations do not result from a localized pattern of visitations. The most active agencies effecting the transfer of pollen in cross-pollinated and partially cross-pollinated plants are the wind and various species of insects. In cotton the insects are the most prevalent, if not the exclusive, agents of natural crossing. Me11 (7) suggested the possible occurrence of -wind pollinations but offered no experimental evidence. Allard (1) believed that wind could distribute a considerable amount of pollen in cotton fields. Brown ( 3 ) believed that the pollen was blown too late in the day for it to be of any consequence in effecting pollination. He indicated that plants covered with screens or nets were not crossed and that there was no evidence of natural crossing except by means of insects. Balls (2) , using a method similar to Allard’s for the detection of winddisseminated pollen in Egypt, obtained negative results. Information regarding the specific role of insects in the pollination of cotton is for the most part rather fragmentary. Allard (1) indicated that the wild bee, Melissodes bimaculutu Le P., and the honey bee were probably the most abundant visitors to cotton flowers in northern Georgia. He also listed the insects observed to visit cotton flowers in that region, 26 of which were different species of Hymenoptera and seven were species of beetles. Shoemaker ( 9 ) at Palestine, Tex., observed two species of butterfly visiting cotton blooms continually, but he regarded them as being of slight importance in pollination. Common bumble bees, Bombus americanorzlm Fabr. were the most active of the visitors but were not regarded as important carriers except when they visited flowers for the purpose of obtaining pollen. Kearney ( 4 ) reported that various Hymenopteran species were the most efficient carriers of cotton pollen at Sacaton, Ariz. He found the honey bee and the wild bees, Melissodes spp., to be the most important cotton pollinators in this locality, the honey bee holding first rank especially among