The Sex Expression of Flowers • 133 



sterility and otherwise suppress anther development, but often only 

 in conjunction with other strong morphogenetic effects. A feminiz- 

 ing effect of carbon monoxide has been observed by J. and Y. 

 Heslop-Harrison (1957) in a monoecious race of Mercurialis. This 

 was accompanied by formative effects resembling those caused by 

 auxins. 



Three other chemical effects should be mentioned. High 

 nitrogen levels generally promote pistillate as opposed to staminate 

 expression; this has been observed on monoecious species and on at 

 least one hermaphrodite, the tomato. The question of whether mam- 

 malian sex hormones may affect sex expression in higher plants has 

 attracted surprisingly little attention. A single major investigation 

 (Love and Love, 1945) with Melandrium showed highly significant 

 effects in spite of high toxicity. Although similar work on a few 

 other plants has found nothing of interest, the problem may still 

 be worth pursuing. 



The gibberellins have so far been little studied with regard to 

 these phenomena, but may prove to be of great importance. Galun 

 (1959) has found that gibberellic acid, unlike auxin, causes a trend 

 toward "maleness"— prolonged staminate and delayed pistillate 

 expression— in the cucumber; this effect is partially counteracted 

 by naphthaleneacetic acid. Moreover, certain cucumber strains that 

 normally produce only pistillate flowers will produce staminate 

 flowers as well following gibberellic acid treatment. Besides its 

 theoretical interest, this result also holds promise for practical 

 breeding work (Peterson and Anhder, 1960). 



So far, the only important hypothesis on the control of sex 

 expression is that derived primarily from work with applied auxin; 

 it envisages auxin level in the plant as the major controlling factor. 

 Daylength, temperature, and other factors are considered to act 

 through their effects on auxin level. Probably the most detailed 

 statement is given by Heslop-Harrison (1957). In essence, optimum 

 auxin levels for flowering are considered to be lower than those for 

 vegetative growth; within the flowering range, the optimum for 

 staminate expression is lower than that for pistillate expression. 

 In a sense this hypothesis contradicts the suggestion, noted earlier, 

 that the pistillate expression represents a more intense flowering 

 condition than the staminate. As a working hypothesis, however, it 

 has proved fruitful. Experiments on the relationships between 



