244 BOTANICAL GAZETTE [September 



f 



The summary of Professor Osterhout's work on spindle formation in 

 A^^aife americayia is partly as follows : During early stages in the first divis- 

 ion of the pollen mother call the spindle is enclosed in a special membrane 

 of cytoplasmic origin, which forms a complete investment around it. The 

 functions of the membrane appear to be comparable to those of the nuclear 

 wall and the limiting membrane of the cytoplasm. There is no weft of fibers 

 such as is usually described for this stage. The spindle-forming fibers are 

 radial from the beginning, and are attached both to the nuclear and spin- 

 dle walls. The second division differs radically from the first, the spindle- 

 formation resembling, in general, that described for the S{)ore mother cells of 

 Equisetum. 



The plates are in three colors, reproducing the effect of the safranin- 

 gentian violet-orange stain. — Charles J. Chamherl 



A PHYSIOLOGICAL STUDY OF THE ANTiPODALS has been untertaken by 

 T, Ikeda.^ Investigations on Tricyrtis hi?-la form the subject of the first 

 paper and it is expected that others will follow. Besides observations on 

 microtome sections prepared in the usual way, microchemical reactions were ^ 



tested on free-hand sections from fresh materials. The archesporial cell 

 becomes the megaspore mother cell directly without a previous division into 

 a primary wall cell and primary sporogenous cell ; the mother cell gives rise 

 to four megaspores, of which the one nearest the chalaza is functional. In 

 the first division of the mother cell the chromatin forms tetrads strongly 

 resembling those of such forms as Gryllotalpa, as described by zoologists. 

 Double fertilization was observed. The endosperm does not form a parietal 

 layer, but the nuclei are evenly distributed throughout the sac, A consider- 

 able amount of endosperm is formed before the division of the fertilized ^^^y 

 one of the figures showing twenty nuclei m a single section. The chief inter- 

 est in the paper lies in the microchemical investigation. In early stages no 

 starch is found in the ovule, but as development proceeds starch appears in. 

 the funiculus and outer integument, and the outer and inner surfaces of the 

 inner integument become cutinized except in the micropylar region. Starch 

 then appears in the inner integument, though not as abundantly, and dextrin 

 can be detected in the antipodal end of the sac. Just before the fusion of 

 the polar nuclei, dextrin is evident in the micropylar portion of the inner 

 integument and also in the antipodals and ^^g apparatus, while the starch 

 has disappeared from the inner integument and appeared in the nucellus. 

 After fertilization, the dextrin disappears from the integument and is pres- 

 ent only in the fertilized egg, and starch has almost entirely disappeared 

 from the integuments and is scanty in the funiculus, but after a few divisions 

 of the embryo starch becomes abundant in the integuments and in the endo- 



=^ Studies in the physiological functions of antipodals and related phenomena of fer- 

 tilization in Liliaceae. Bull. ColK Agric. Tokyo Imp. Univ. 5 : 41-72. ph. 3-4- ^902. 







