ALBERT F. BLAKESLEE 65 



beek and Conklin, when an auxin such as naphthaleneacetic acid is 

 injected into young castrated ovaries. It appears that the endothelial 

 layer has a lower sensitivity threshold to certain chemicals than the 

 other tissues of the ovule. We are attempting to induce the endothelial 

 tumors to grow in tissue cultures but so far without success. Why certain 

 tissues are easily cultivated and others are difficult is not clear. Also we 

 do not know why a fertilized 7.n egg in a diploid ovary develops readily 

 to a mature seed but an unfertilized i.n ^gg in a tetraploid ovary usually 

 fails to develop. The obvious objective in our endothelium problem is 

 to find an anti-inhibitor which will neutralize or prevent the activity 

 of the embryo tumors. 



Now how can these facts be put to work? Drs. van Overbeek and 

 Marie Conklin several years ago developed special methods for the 

 culture of excised normal embryos of Datura stramonium on artificial 

 media. It was only the larger embryos which could thus be induced to 

 develop further under the conditions used. The fact that sometimes 

 haploid (177) seedlings have come from twin embryos within the same 

 seed coats suggested that there might be some stimulating substance 

 in the normal embryo sac which induced the development of the twinned 

 177 embryo. Following this suggestion, coconut milk, which is a natural 

 endosperm, was tested and found to be an effective embryo stimulator 

 which incited growth in excised embryos as small as o.i mm. in diameter. 

 Later, on coming to Smith College, we used the embryo culture tech- 

 nique to secure hybrids in wide species crosses. We have found malt 

 extract more convenient than coconut milk as an embryo stimulator. 

 Since when autoclaved it develops toxic substances, however, we have 

 had to sterilize it with a Seitz filter. 



The embryo technique we have put to work in getting hybrids with 

 which to determine the interchanges which have taken place in the 

 evolution of our ten species of Datura. It also has given us information 

 regarding important barriers to crossability. We have already said that 

 of the 90 species' combinations only 19 gave viable seeds capable of 

 germinating. In addition we have secured dissectablc embryos from 45 

 species' combinations. Of these 31 have been grown to a stage of maturity 

 at which it could be certain they were the result of hybridization. The 

 others may have been hybrids, but it sometimes happens that the 

 dissected embryo turns out to be a haploid (177) derived from the 

 development of an unfertilized egg, and sometimes an embryo is due to 



