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



STRUCTURE OF THE FLOWER IN RELATION TO POLLINATION. 



The large and showy cotton flower with its reproductive organs 

 so plac?d as to be readily accessible to all kinds of insects (Pis. I and 

 II) would seem to be admirably adapted to cross pollination, espe- 

 cially as the abundant secretion of nectar attracts large numbers of 

 pollen-carrying insects. The transfer of pollen is favored by the 

 fact that even during the height of the blossoming period the num- 

 ber of flowers opening daily on the individual plant rarely exceeds 

 three and is usually only one. 8 Yet the evidence presented in the 

 preceding pages indicates a strong preponderance of self-fertiliza- 

 tion. In seeking an explanation of this apparent anomaly the 

 structure and the later ontogeny of the flower will be considered. 



The description which follows is based upon the Pima variety, but 

 applies in all essential particulars to other varieties of the Egyptian 

 type. The points of structure relative to pollination in which the 

 flower of upland cotton {Gossypium hirsutum) differs from that 

 of Egyptian cotton will be mentioned for comparison. 



POSITIONAL RELATIONS OF THE REPRODUCTIVE ORGANS. 



Egyptian cotton, like other members of the genus Gossypium, 

 has the ovary and style inclosed in a sheath or tube formed by the 

 coalescent bases of the filaments of the stamens, and the pistil pro- 

 jects above the summit of this so-called staminal column (PI. I, 

 Fig. 1; PL II, Fig. 1). There is no sharp differentiation between 

 stigmas and style, the latter beginning to increase in diameter and 

 to become pubescent below the summit of the staminal column, but 

 under normal conditions pollen is deposited in quantity only on the 

 exserted portion of the pistil, and for convenience the term " stig- 

 mas ? will be used in referring to this portion only. The erect and 

 usually somewhat spirally twisted stigmas (PL I, Fig. 1) are coherent 

 except very near the apex and are slightly enlarged upward. The 

 stigmatic surface is not viscid but is densely pubescent, and this to- 

 gether with the spiny surface of the pollen grains secures their ad- 

 hesion to the stigmas. Unlike the condition in many of the Mal- 

 vaceae, the stigmas do not become spreading or reflexed after the 

 flower opens but remain erect. There is no evidence that the flower 

 is protandrous, as is the case in most of the Malvaceae. 9 The stigmas 

 from the apex to a little below the point where they emerge from the 

 staminal column are homogeneous in texture and pubescence, and 

 pollen grains adhere to and doubtless germinate upon all parts of 

 their surface. 



Reference to Plate I, Figure 1, and to Plate II, Figure 1, shows 

 that in Pima cotton the stigmas project far beyond the summit of 

 the staminal sheath, averaging in length, at 8 a. m., or about 1J 

 hours after the corolla has begun to open, 10 millimeters, or one- 



8 Darwin (13, p. 389), evidently having in mind plants on which numerous flowers are 

 in anthesis at the same time, states " Insects usually search a large number of flowers 

 on the same plant before they fly to another, and this is opposed to cross-fertilization." 



8 Knuth (S3, p. 206). According to K. Schumann (J,2, p. 32) all Malvaceae are pro- 

 tandrous. 



