Mabch 8, 1907] 



SCIENCE 



381 



differentiation. I do not refer to such pre- 

 potency as is characterized by plant and 

 animal breeders as the power of certain 

 individuals in a given cross to transmit 

 their qualities to the offspring with the 

 suppression of those of the other parent, 

 but to the prepotency established by Dar- 

 win's classic experiments in cross pollina- 

 tion and whose correlative is the partial or 

 complete self sterility observed in so many 

 hermaphroditic flowers. This prepotency 

 is a very widely spread phenomenon among 

 plants and appears almost with the first 

 development of gametic fusions. It is a 

 common fact among algse that even before 

 there is any sexual differentiation between 

 the gametes, the tendency to conjugation 

 is very much stronger between swarm 

 spores from mother cells of separated 

 origin. 



Sex differentiation itself seems to arise 

 in entire independence of this prepotency 

 as an expression of the universal tendency 

 to specialization of function with differ- 

 entiation of structure, on the one hand, 

 toward increased motility with rela- 

 tively reduced size, on the other, toward 

 increased size and storage of metaplasmie 

 materials as is so perfectly shoviTi in 

 the Volvox series and many other groups 

 of algse. 



As to the stage in the life cycle at which 

 sex differentiation occurs, plants show the 

 widest possible variety. Marehal has re- 

 cently shown that the spores from a single 

 capsule in some of the dioecious mosses are 

 of both sexes, but that the sex of each spore 

 and of the protonema and moss plants 

 which come from it is fixed. It is quite 

 possible that here sex differentiation is ac- 

 complished in connection with the reduc- 

 tion divisions and spore formation — the 

 same stage as that at which the differential 

 distribution of chromosomes is described 

 for certain insects, but far removed from 

 egg formation and fertilization. The stage 



of sex differentiation in Equisetum may be 

 the same as that in these mosses with dioe- 

 cious gametophytes. 



In the ferns and lycopods, however, sex 

 differentiation is accomplished in the 

 gametophyte. In the ferns the older and 

 first formed regions of the prothallus pro- 

 duce antheridia and later on the newer 

 growth produces archegonia. In the het- 

 erosporous fern allies Marsilia, Isoetes, etc., 

 and in hermaphroditic seed plants, sex dif- 

 ferentiation appears in the spore sacs and 

 the asexual spores themselves. The gam- 

 etophytes are thus unisexual and their sex 

 is determined by the sex of the spores 

 from which they come. 



In dioecious seed plants sex differentia- 

 tion apparently may occur in the egg so 

 that the entire sporophyte is male or fe- 

 male for its lifetime, but the determination 

 here is of a rather indefinite sort, since a 

 large proportion of such dioecious plants 

 have been reported as occasionally pro- 

 ducing a few hermaphroditic flowers. The 

 gametophytes in all seed plants are strictly 

 and unchangeably unisexual. 



The flowers of hermaphrodite or perfect 

 shrubs and trees offer a most interesting 

 case of successively repeated sex determina- 

 tions in the life of the same individual. 

 With each succeeding blooming period the 

 stamens and pistils are differentiated anew 

 in the flower buds out of what must be 

 regarded as the strictly non-sexual new 

 cell growth of the year. Such hermaphro- 

 ditism, requiring the operation of sex-de- 

 termining factors with each successive sea- 

 son, is sharply in contrast with that in 

 animals in which the sex glands are fixed 

 for life. 



. Polygamous and polygamodioecious seed 

 plants complicate the conditions still fur- 

 ther, but it is sufficiently plain that no 

 simple differential distribution of chromo- 

 somes on Mendelian principles at the pe- 

 riod of chromosome reduction could in any 



