ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 53 



monocotyledons and dicotyledons, while in Nyniphaeacese and others 

 there is a gradual transition from petals to filaments by narrowing of the 

 former. Again, the petal and the filament have essentially the same 

 anatomical organisation, while that of the petiole is quite different. The 

 connective is merely the continuation of the filament, and it is the con- 

 nective, and not the anther-cells, which frequently become petaloid. The 

 anther-cell, like the ovule, is a structure sui generis. In very rare in- 

 stances sepals, petals, or carpels are supported on a true petiole quite 

 distinct from the blade in the first two cases. 



Development of Pollen in Asclepiads.* — Paul Dop has studied the 

 development of pollen in species of Asclepias, Vincetoxicum, Gom- 

 phocarpus, Marsdenia, and other genera, and finds throughout results 

 comparable to those obtained by Chauveaud in Vincetoxicum officinale. 

 The pollen-mother-cells arise by division of subepidermal cells, and 

 give rise to the pollen-grains directly by division into four. The 

 nutritive layer formed of one or several layers of cells, secretes the 

 waxy envelope around the pollinium, while the caudicles and retinacula 

 are secreted by the epidermal cells of the stigma. After the dehiscence 

 of the pollen-sac, the pollinium escapes from the sac and attaches itself 

 to the caudicles. 



Germination of Pollen-Grains.f — P. P. Richer finds that the pollen 

 of certain species which will not germinate in pure water, will germinate 

 if a stigma of the same species or an allied species be placed in the 

 water. On the other hand, it germinates less successfully, or may even 

 fail to germinate, in the presence of a stigma of a very different plant. 

 He concludes that there exist in the stigma special substances which 

 encourage the germination of the pollen of the same plant, while 

 inhibiting the germination of the pollen of a strange plant. 



Double Fertilisation in Crucifers.J — L. Guignard has been able to 

 follow all stages of this phenomenon in Capsella Bursa-pastoris and 

 Lepidium sativum. The sexual apparatus has the normal structure and 

 arrangement ; the two polar nuclei fuse only a short time before fertili- 

 sation, and the large secondary nucleus lies close to the oosphere. The 

 double fertilisation follows the usual course. The male nuclei reach 

 almost simultaneously the nucleus of the oosphere, and the secondary 

 nucleus of the embryo-sac, but the process of fusion is completed 

 earlier in the case of the latter, and the division of the resulting cell 

 precedes that of the egg. In the course of development of the embryo 

 the albumen gradually disappears, with the exception of the peripheral 

 layer. This, which the author has previously styled the proteid layer, 

 persists in the ripe grain in all Crueifers, as it does also in almost all 

 families the seed of which is described as exalbuminous. 



Recent Investigations in the Embryo-sac of Angiosperms. 

 1). H. Campbell gives a resume of recent work on the course of events 

 in the embryo-sac of Angiosperms. He regards Peperomia as the most 

 primitive form yet described, basing his conclusions on the absence of a 



* Comptes Rendus, cxxxv. (1902) pp. 710-2. 



t Tom. cit., pp. 634-6. % Tom. cit., pp. 497-9. 



§ Amer. Natural., xxxvi. (1902) pp. 777-86 (5 figs, in text). 



