ZOOLOGY AND BOTANY, MICROSCOPY, ETC 5. r >7 



apparatus differ in the three genera but are fixed for each species ; (6) starch 

 fills the sac in Medicago but is confined to the niicropylar end of the 

 nucellus and the inner integument in the other genera ; (7) in Trifolium 

 the pro-embryo is short and massive and more or less ill-defined ; (8) in 

 Medicago and Vicia the suspensors are well-defined with multinucleate 

 cells ; in the former it is filamentous, in the latter it is composed of 

 two superimposed pairs of cells ; (9) sterilization is most pronounced in 

 T. pratense. 



Embryo-sac of the Convallariacese.* — F. McAllister has studied 

 the development of the embryo-sac in the Convallariacaj, and finds that 

 in Polygonatum it is formed from one reduction-nucleus, in Smilacina 

 racemosa, S. amplexicaidis, and Streptopus roseus from two such nuclei, 

 and from four in Smilacina stellata, S. sessifolia, Maianthemum canadense, 

 and Medeola virginica. Wherever more than one nucleus takes part in the 

 formation of the embryo-sac, the nuclei are at first separated by cell-walls, 

 varying from split cell-plates in Smilacina stellata to temporary cell-plates 

 in Medeola virginica. This separation does not affect the morphology of 

 the cells, which are all megaspores. Thus it appears that all reduction- 

 nuclei formed from the nucleus of the megaspore-mother-cell, whether 

 separated for a time or not at all, are to be regarded as megaspore -nuclei. 

 In six of the eight species studied two megaspore-mother-cells were 

 sometimes found. 



Sexuality of Humulus.f — J. Tournois has studied the problems 

 connected with sexual reproduction in H. Lupulus and H. japonicus. 

 The time of the appearance of the flowers is closely connected with 

 external factors, e.g. variations in illumination induce corresponding 

 variations in the duration of the purely vegetative period, and in 

 H. japonicus flower-formation has been considerably hastened by reduction 

 of the daily time of exposure to sunlight. Both species of Humulus are 

 normally dioecious, the sexes being confined to different plants, but it is 

 not unusual to find both sexes on the same plant ; when this is the 

 case, however, with H. Lupulus, the male and female flowers are rarely 

 fertile at the same time. In H. japonicus, monoecious forms of two 

 kinds have been observed, viz. some of spontaneous origin and others 

 induced by well-defined experimental conditions. The most remarkable 

 results were obtained with plants grown in winter, in which progenetic 

 inflorescences were produced ; in this case the male branches were 

 sometimes transformed in such a manner as to function as monoecious, 

 or even as almost exclusively female plants. All the monoecious inflor- 

 escences of both species of Humulus were of the same morphological 

 type, viz. a biparous cyme of male flowers, the branches of which terminated 

 in clusters of female flowers. The same conditions which in H. japonicus 

 induce the transformation of male into monoecious plants, can also. cause 

 the production of stigmata or sterile carpels in male flowers, either by 

 substitution for or superposition to the male organs. Diminution of 

 transpiration is a most important factor in producing such trausforma- 



* Bot. Gaz., Mii. (1914^ pp. 137-53 (2 pis.). 



+ Ann. Sci. Nat., xix. (1914) pp. 49-191 (5 pis. and 23 figs.). 



