Beer, Development of tlie pollen grain and anther of some Onagraceae. 311 
stage botli mitotic and amitotic divisions of the nuclei take 
place but in older antliers tlie tapetal nuclei divide exclusively 
by amitosis. Side by side with this continuous multiplication of 
nuclei a nuclear degeneration can be observed in which the 
nuclear membrane and nucleolus are resolved into a group of 
deeply staining hbres or narrow laminae. 
These üb res are no doubt identical with those observed by 
Meves in the tapetum of Nympliaea alba. These hbres have 
been compared by Meves to the chromidial structures found in 
certain animal cells and this comparison has a great interest in 
the light of G-oldschmidt's recent work. 
The hbres in the tapetum of Oenothera are found to be to 
a large extent derivatives of the chromatin of the nucleus and 
tliey increase in number as the cell advances in age and its 
nuclei continue to divide (fragment) and to break down. 
The interesting conclusion is, therefore, reached that a large 
portion of the material which replenishes the exhausted pollen¬ 
protoplast has at one time or another entered into the cornpo- 
sition of a tapetal nucleus. 
9. The hitine is a continuous membrane lining the entire 
inner surface of the pollen grain. It first appears and reaches 
its greatest development at the bases of the three interstitial 
bodies whilst over the rpst of the pollen grain it extends as an 
exceedingly delicate layer. _ Düring the development of the 
intine three thick strands of cytoplasm connect the centrally 
placed nucleus with the :spots beneath the interstitial bodies 
where tlie membrane is growing most vigoouslv. In Oenothera 
the intine gives the reactions of a pectic body but httle or no 
cellulose can be detected in it. In Gaura Lindheimeri and Epi- 
lobium tetragonum both pectic bodies and cellulose occur in the intine. 
10. The Perforation of the closing disc of the interstitial 
bodv and the disintegration of tlie laminae of that body liave 
been most closely followed in the pollen grains of Gaura Lind¬ 
heimeri. The closing disc is perforated and the interstitial 
laminae are u eaten away“ in advance of the growing intine in 
a manner which suggests the action of a solvent, probably an 
enzyme. 
11. In Oenothera longiflora , even when growing under the 
most favourable conditions, many of the pollen grains becoine 
arrested in their development. They all seem to advance until 
their protoplast is completely exhausted by secretion but the 
tapetal material is insufficient to allow all of them to carry their 
development further. 
12. The mature pollen grains are surrounded and bound 
together by “fibrils“ which are derived from the special-mother- 
cell wall. When the special-mother-cell wall breaks down it 
forms at first a structureless mucilage which no longer gives 
any of the reactions of callose. Later this mucilage becomes 
drawn out into “fibrils“ and these are very resistant to solvents. 
Rudolf Beer. 
21 * 
