bg 
1897] CURRENT LITERATURE 385 
by plasmolysis, and from one cell to another, provided only that protoplasmic 
continuity through living fibers is maintained. The author believes that Palla’s 
failure to observe such fine connecting fibers between plasmolyzed masses in 
pollen tubes led to his erroneous conclusion that the nucleus is not necessary 
to cell wall formation. Very interesting is the fact that mere contact between 
two plasma masses is not sufficient for the transmission of the formative influ- 
ence of the nucleus from one to the other. Just what this réle of the nucleus 
is in cell wall formation the author does not attempt to state.—R. A. H. 
Dr. C. ISHIKAWA of the Imperial University, Tokyo, Japan, has published 
a third paper” in his series of “Studies of reproductive elements.” It treats 
of the pollen grains of A//ium fistulosum, with special reference to the pro- 
cess of chromosome reduction. In this process he finds some important 
deviations from what has hitherto been regarded as typical of flowering 
plants. Although he did not succeed in observing the original delimitation 
of the sporogenous region from the y tissue of the anthers, he obtained 
very young sporogenous cells in stamens which were just making their appear- 
ance as protuberances on the flower disk. In these early sporogeneous cells 
the nuclei were already -distinguishable by poverty of chromatin, and in 
division showed the reduced number (eight) of chromosomes, the division 
being described as heterotypic. The chromosomes split longitudinally, and 
the segments during metakinesis remain attached by their ends so as to form 
rings. In the pollen mother cell the spirem is apparently one continuous 
ribbon, which, after a synapsis stage, becomes divided into segments having 
the chromatin granules arranged in double rows. The segments shorten and 
become Sharply bent at the middle. Longitudinal fission is now complete, 
and the bent segments attaching themselves together in pairs at the point of 
bending give rise to eight X-shaped figures which resemble the tetrads of 
Some animals. Metakinesis proceeds according to the heterotypic method. 
Arriving at the poles of the spindle, the eight V-shaped daughter chromo- 
Somes break transversely at the angle, and form sixteen rod-shaped grand- 
daughterchromosomes. No lete resti dition ensues, but division of 
the pollen daughter cell into the pollen grains follows at once, according to 
the heterotypic method. The next stage, the formation of the vegetative 
and generative nuclei, presents no unusual phenomena. It may be added 
that in the first division of the pollen mother cell bodies resembling centro- 
somes were frequently observed. 
_ Doubtless the most interesting observation in this account of spermato- 
_ 8enesis is the early reduction of the number of chromosomes. In all other 
flowering plants in which chromosome reduction has been investigated it is 
* Die Entwicklung der Pollen-Kérner von Adium fistulosum L., ein Beitrag zur 
romosomen Reduktion im Pflanzenreiche. Jour. Coll. Sci. (Tokyo) 10:—- 
