Cytologie und Befruchtung. 22 L 



die zwei Polarkerne niemals inmitten des Embryosacks, son- 

 dern stets am antipodalen Ende zusammentreffen. 



Die befruchtete Eizelle theilt sich zunächst durch eine Quer- 

 wand in eine auffallend grosse und bläschenförmige Suspensor- 

 zelle und die erste Zelle des Embryos. Letztere theilt sich 

 durch Querwände in drei Zellen und in der untersten dieser 

 Zellen gehen weitere Theilungen durch zwei rechtwinklig auf- 

 einander stehende Wände vor sich. Fritsch (London). 



Stevens, F. L., Gametogenesis and Fertilization irh 

 Albiigo. (Botanical Gazette. XXXIl. 1901. p. 77—98, 157 

 — 169, 238—261. pl. 1—4.) 



The earliest investigations of fertilization among the Phyco- 

 mycetes published since the development of modern methods 

 of cytological technique were those of Wager, Istvänffi, 

 and Trow, but it was not until the paper by Stevens, 

 published in the Botanical Gazette, XXVIII, 1899, p. 149—176, 

 225 — 245, pl. 11 — 15, that attention was called to the peculiar 

 method of fertilization in Albiigo Bliti. It was thought at the 

 time that this might prove to be peculiar to this species, and 

 therefore an unrelated phenomenon. The present paper not 

 merely confirms the earlier observations, but shows the relation 

 between ordinary and multiple fertilization by a series of forms 

 thought to illustrate the direction of evolution. 



Gametogenesis in A. Poriiilacae, A. Tragopogonis and 

 A. Candida is described in detail, and the facts are discussed 

 in relation to previous work. In A. Portulacae multiple 

 fertilization is very similar to that in A. Bliti, except that the 

 number of nuclei involved is far greater. The results are 

 diametrically opposed at almost every point to those of 

 Berlese, although in accord with those of Istvänffi. The 

 ,,receptive papilla" is more prominent in this species than in 

 any other yet known. 



The Chief interest of the paper, however, is in connection 

 with A. Tragopogonis, in which the oosphere is originally 

 multinucleate, developing as in A. Bliti and A. Portulacae ; but 

 as it matures, one or a few only of the nuclei become attached 

 to the coenocentrum, the others degenerating in the ooplasm. 

 Later all of the nuclei attached to the coenocentrum, except 

 one, degenerate, leaving the oosphere uninucleate. The anthe- 

 ridial tube is usually uninucleate, rarely possessing two or three 

 nuclei. 



This fact, supported by collateral evidence from the mode 

 of zonation, the formation of the receptive papilla, and the 

 prominence and functional perfection of the coenocentrum, seems 

 to indicate that the primitive oosphere among the Peronosporeae 

 was multinucleate, and that the uninucleate condition is a 

 derived one. This necessitates some change in the current 

 view as to the origin of this group and the nature of the 

 oosphere. The author holds that if the Phycomycetes are related 



