372 INTRODUCTION TO CYTOLOGY 



reason that many of them have been carried out with angiosperms, in 

 which hermaphroditism is so common and which are regarded by some 

 botanists as all potentially bisexual. As a case illustrating the latter 

 point may be cited the hemp plant, Canndbis saliva. This species is 

 normally unisexual, but if the flowers are removed it will produce flowers 

 of the other sex (Pritchard 1916). 



Intersexes of many grades between the normal plants in dioecious 

 angiosperms are frequently found, as in Myrica Gale (Davey and Gibson 

 1917), Cannabis saliva, Salix amygdaloides, and Morus alba (Schaffner 

 1919a6), whereas the relative proportions of maleness and femaleness in 

 hermaphroditic forms are apparently very easily influenced by the 

 environment. In Planlago lanceolala (Bartlett 1911; Correns 1908; 

 Stout 1919) the stamens and pollen are developed in various degrees in 

 different flowers, or even in the same flower; in some cases they are only 

 rudimentary, leaving the flowers functionally female. Stout inclines 

 to the view that dioecism results from the suppression of femaleness or 

 maleness in organisms originally monoecious, and points out that the 

 many "degrees of maleness" in Planlago fill the gap completely in this 

 case. He accordingly concludes, in agreement with Riddle, Banta, 

 and Goldschmidt, that sex is a labile, reversible character; that maleness 

 and femaleness, both present in the somatic cells of all sporophytic 

 individuals, are relative and not absolute conditions; and that "sex- 

 determination, at least in hermaphrodites, is fundamentally a phe- 

 nomenon of somatic differentiation that is ultimately associated with 

 processes of growth, development, and interaction of tissues, and subject 

 to modification or even complete determination by them." The sex- 

 chromosome theory is held to be inadequate in the case of hermaphro- 

 dites: sex is an "epigenetic," not a "preformed" character". 



Yampolsky (1919, 1920) thinks it probable that male and female 

 gametes with graded potencies are produced in Mercurialis annua, and 

 that the sexes cannot be due to a segregation of sex-factors at reduction. 



The sex of the haploid phase (gametophyte) of the plant life cycle, 

 when this phase is unisexual (dioecious), behaves as a much more nearly 

 irreversible character, as shown by the experiments on mosses, Marchanlia, 

 and Phycomyces cited early in the present chapter. How widely this holds 

 true in thallophytes and bryophytes is not known. Allen regards 

 the quantitative theory of sex as quite inapplicable to the case of 

 Sphcerocarpos. In the homosporous pteridophytes the gametophytes are 

 commonly monoecious, and the fact that here, as in many liverworts, 

 the male organs appear early and the female organs later suggests a 

 physiological basis of sex-determination. Some fern gametophytes, on 

 the contrary, are dioecious under all ordinary conditions. In Onoclea 

 (Wuist 1913), such a normally dioecious form, the female gametophytes 

 under certain experimental conditions produced antheridia in addition 



