506 SUMMARY OF CURRENT RESEARCHES RELATING TO 



narrower than that of D. longifolia. In the division of the pollen- 

 mother-cells the number of chromosomes was found to vary, for beside 

 the expected number of 15, the numbers 10 and 20, corresponding to 

 those of the parents, were also found, and in fact all three numbers in 

 the same anther. The pollen-grains themselves were also either of 

 hybrid type or else exactly like those of one or other parent. The de- 

 scendants of these hybrids were not investigated, so that the exact 

 meaning of these important results must at present remain obscure. 



Behaviour of Nuclei in Plant Hybrids.* — W. A. Cotton has 

 studied the behaviour of the nuclei in the pollen-development of an 

 artificially produced hybrid cotton, Gossypium barbadense and G. herba- 

 ceum. The very early stages were not investigated, but the pollen- 

 mother-cells were seen in the synapsis stage, and the chromatin segments 

 when first observed were in long delicate loops much twisted and bent. 

 The loops thicken to form rings, and then by further condensation lose 

 their ring character, but in the metaphase of the first division this 

 character reappears. In any one nucleus the loops and rings are of 

 uniform size, and the two sizes of rings found in the hybrid pigeon and in 

 some pure forms were not observed. The first division was normally 

 heterotypic, but it could not be determined whether the split at the 

 metaphase separated chromosomes which correspond to the halves of the 

 heterotype ring or whether the split followed the second cleavage. The 

 second division was normally homotypic but could not be studied in 

 detail. Besides the normal divisions, a certain number of abnormal 

 ones were observed, of which details are given. The author believes, 

 however, that the purity of the sex-cells, which is apparently one of the 

 corollaries of Mendel's law, is brought about by the normal tetrad 

 divisions. This point, together with the question of the probable result 

 of hybridity of Cryptogams, is treated of in the earlier part of the paper. 



Non-Sexual Nuclear Fusions.! — Nemec discusses the question of 

 these fusions in relation to the work of his pupil, Blazek, who found 

 that when the roots of seedlings of Pisum sativum and Viola Faba were 

 exposed to vapour of benzene or to a 1 per cent, solution of copper 

 sulphate, multinucleate cells were produced which afterwards became 

 uninucleate by nuclear fusion. These fusions are, of course, not con- 

 nected with sex (neither are the fusions found in the endosperm of 

 Corydalis and Tulipa) and Nemec believes that they are brought about 

 by an auto-regulating mechanism of the cell which reduces typical cells 

 to the uninucleate state when by any means they have become multi- 

 nucleate. The author suggests that the morphological characteristic of 

 sex lies rather in cell fusion than in nuclear fusion since the latter power 

 is possessed also by vegetative cells. 



Relationship of the Nuclear Membrane to the Protoplast.:}: — 

 A. A. Lawson has studied the formation of the nuclear membrane 

 especially in the spore-mother-cells of Passiflora cmrulea and the arche- 

 sporial cells of Equisetum limosum. His results, which were confirmed 



* Bull. Torrey Bot. Club, xxx. (1903) pp. 133-72 (2 pis.). 

 t S.B. k. Bohm. Ges. Wisseusch., 1902 (1903) pp. 59 (6 pis.). 

 $ Bot. Gazette, xxxv. (1903) pp. 305-19 (1 pi.). 



