EMBRYOGENESIS IN FLOWERING PLANTS 311 



pea embryos, with or without the attached cotyledons, in sterile cuUure 

 media. The effects of various pre-soaicing treatments, with and without 

 growth-regulating substances, of factors in the medium, and of leaving 

 larger or smaller amounts of cotyledon attached to the growing embryo, 

 were investigated. The ratio of the amount of reserve material in the 

 cotyledons to the size of the embryo has an important effect on the 

 morphogenetic development, root and shoot growth being stronger in 

 proportion to the amount of cotyledon tissue left on the embryo. No 

 development of plumular leaves takes place in the dark, and even in 

 the light embryos deprived of cotyledons scarcely form such leaves. 

 In fact, both leaf and root formation are determined by the presence 

 of cotyledon tissue. The action of various growth substances, admini- 

 stered by pre-soaking the seeds, apparently takes place mainly by way 

 of the cotyledons. (For a survey of the physiology of seeds, see 

 Crocker and Barton, 1953.) 



NON-VIABLE EMBRYOS AND OVULAR TUMOURS 



The production of interspecific crosses affords a convenient way of 

 modifying the immediate environment of the embryo, i.e. the endo- 

 sperm. Brink and Cooper (1940, 1944, 1947) have noted that in normal 

 seed development there is a balanced or regulated growth of the embryo, 

 the endosperm and the nucellar tissue. The embryonic developments 

 in crosses between Dianthus chinensis (with 2n = 60) and D. plumarius 

 (with 2n = 90) were studied by Buell (1953), the embryogenesis in the 

 two species being closely comparable. Viable seeds are not produced 

 when D. chinensis (?) is crossed with D. plumarius {h). Young embryos, 

 however, are formed and develop to the spherical stage; but there- 

 after they cease to grow and degenerate, most of them having dis- 

 integrated 9-10 days after pollination. Various histological differences 

 were noted as between the hybrid embryo and that of D. chinensis of the 

 same age; there were also characteristic differences in the respective 

 endosperms. The reciprocal cross, D. plumarius (?) and D. chinensis 

 (c?) is also incompatible, but the embryonic development shows fewer 

 irregularities than the other. The growth of these embryos is slow as 

 compared with the normal controls (i.e. D. plumarius). Disintegration 

 of the hybrid embryos, which again attained to the spherical stage, 

 began about 4 days after pollination, and after 8 days only fragments 

 of the disorganised embryos remained. The basal cells are the first to 

 be affected. Various anomalous endosperm developments were 

 observed at an early stage, e.g. failure to digest the nucellus at the 

 normal rate. The abortion of the embryo is probably to be attributed 

 to a breakdown in chemical regulation, possibly involving growth- 

 regulating substances or enzyme balance (Sachet, 1948). Schneider 



