96 R. A. BRINK 



higher than earlier believed. Various investigators have expressed the opinion 

 that the mere presence of growing pollen tubes in the style causes enlarge- 

 ment of the ovules. This view now appears to be incorrect. 



On the other hand, there is a steadily increasing amount of evidence to 

 show that the incipient growth of the ovules, following many interspecific 

 matings which do not yield functional seeds, is a response to fertilization. 

 That is to say, the block in the reproductive cycle which was assumed to 

 intervene prior to fertilization actually occurs following syngamy. Embrycs 

 are formed in these cases, but they perish when the young seed fails to de- 

 velop. Some rather extreme examples of this phenomenon which have been 

 observed in our laboratory include Nicotiana ghitinosa X Petunia violacea, 

 N. glntinosa X Lycopersicon esculentum, and Medicago saliva X M. scutellata. 



It is not to be inferred that all hybrid embryos of this general class are 

 capable of growing into mature plants. The fact that the seeds containing 

 them collapse is not proof, however, of intrinsic inviability. An unknown but 

 probably significant proportion of these novel zygotic combinations are po- 

 tentially propagable. The problem is to discover the means by which they 

 may be reared. This brings us to the second point — the nature of the problem 

 to be faced in growing very small excised embryos. 



With few exceptions, the embryos which have been successfully culti- 

 vated artificially have been removed from the seed at rather advanced stages 

 of development. Unless they are multicellular and differentiation has at least 

 begun, the embryos usually do not grow on the media which thus far have 

 been devised. There are reasons for thinking that the nutritional require- 

 ments of these older embryos are simpler than those in a juvenile condition. 

 Histological evidence shows that at the early stages of seed development the 

 embryo is enclosed, or nearly enclosed, in the highly active, young endo- 

 sperm. The endosperm cells adjacent to the proembryo and the very young 

 embryo remain intact. A little later, as the embryo enlarges, these cells 

 begin to break down and their contents disappear. Eventually all the endo- 

 sperm tissue is consumed in most species. 



One may infer from these facts that the embryo is dependent upon the 

 endosperm for certain metabolites which initially the embryo is quite in- 

 capable of synthesizing. The endosperm may be pictured as secreting the 

 needed materials at the early post-fertilization stage, and yielding them 

 later in a more passive fashion as the tissue becomes lysed. Meanwhile the 

 embryo becomes progressively less dependent upon the endosperm by acquir- 

 ing for itself the synthetic capabilities previously limited to the nurse tissue. 

 On this view the very young embryo is an obligate parasite on the endo- 

 sperm. Once past the state of obligate parasitism, growth of the embryo may 

 be effectively supported by comparatively simple nutrients such as may be 

 provided in artificial culture media. 



Visualized in those terms, the problem of cultivating very young, excised 



