14 



BOTANICAL GAZETTE [july 



(5) It is conceivable that the triple fusion nucleus in double fer- 

 tilization, instead of the egg^ might form the embryo. The only 

 known case of the endosperm developing anything resembling an 



r 



embryo is Balanophora. Treub'^ studied B. elongaia and the 



same 



conditions were afterward found in B. globosa by Lotsy.^^ Fertili- 

 zation does not take place in these forms, and the egg and synergids 



crm 



divides, forming a mass of endosperm tissue. In the midst of the 

 endosperm a group of smaller cells is cut off, forming a '^pseud- 

 embr}'o," which was followed to the five or ten cell-stage. Tetrad 

 formation apparently does not take place, so that there is probably no 



E 



reduction of chromosomes, the '^pseudembr}^o" having the 2X 

 number. 



All of these possibilities, therefore, are without parallel in our 

 present knowledge, but it seems most probable that the source of the 

 twenty-one chromosomes is to be sought in some such situation during 

 fertilization, or the genesis of the individual having this number of 

 chromosomes. It is hoped that further investigation of megaspore 

 and embryo sac development and fertilization in O. lala will lead to the 

 explanation of this remarkable situation, though of course the real 

 explanation may be quite different from any of the possibilities here 

 suggested. \ • 



On any of these assumptions the number of chromosomes is 

 twenty-one. If it is twenty, however, another explanation must be 

 sought. The correctness of the latter number is supported by the 

 fact that I have counted many telophases of the heterotypic mitosis, 

 in which the number of chromosomes was ten in each daughter 

 nucleus. Also if twenty-one chromosomes originated from the fusion 

 of three nuclei, two of which represented one parent and one the other 

 parent, it might be expected that in synapsis and the reduction 

 mitosis a segregation of the chromosomes would indicate this origin. 

 One might perhaps expect a pairing of seven chromosomes (maternal) 



+ 



with seven others (paternal), leaving the other seven unpaired to pass 



*6 Treub, M., L'organe femelle et rapogamie du Balanophora elongaia Bl. Ann. 

 Jard. Bot. Buitenzorg 15:1-22. pis, 1-8. 1898. 



*7 LoTSY, J. P,, BalanopJwra globosa Jungh. Eine wenigstens ortlich-verwittwete 

 Pflanze. Ann, Jard. Bot. Buitenzorg II. 1:174-186. pis. 26-2g. 1899. 



