ZOOLOGY AND BOTANY, MICKOSCOPY, ETC. 117 



entire inner cavity of the embryo-sac presenting the ajjpearance of being 

 instantaneously broken up by thin lamellae of protoplasm. In SymjyTiori- 

 carjms the variation in the same direction from the structure in Lo^iicera 

 is still more strongly displayed, a powerful development of protoplasm 

 taking place in the centre of the embryo-sas in the form of a mass of 

 chambers, the dividing walls of which consist of protoplasm in which septa 

 of cellulose subsequently make their appearance. The division of the 

 embryo does not begin in the CaprifoliaccEe until after the complete forma- 

 tion of this endosperm. 



Adoxa agrees essentially with Sambucus in the mode of development of 

 the endosperm, diftering altogether from that of Chrysosplenium with which 

 some authors propose to associate it. Hedera shows affinity in this respect 

 with the Caprifoliacese, especially Viburnum and Sambucus, rather than with 

 the UmbelliferEe. The ovule is monochlamydeous and bilateral, but is 

 distinguished from that of the two genera named by the presence of a 

 small quantity of nucellar tissue at the time of flowering. 



In Galium and As^erula no vacuoles are to be discovered in the parietal 

 layer of protoplasm in the embryo-sac, nor any breaking up into chambers. 

 The protoplasm subsequently breaks up into a number of polyhedral cells, 

 some of which contain two or three nuclei, though finally the endosperm 

 consists of uninucleated cells. The suspensor branches into a number of 

 haustorium-like appendages which penetrate the endosperm. 



In Borrago the author was unable to confirm the statement of Hof- 

 meister of the formation of an endosperm-tissue at the chalaza-end of the 

 embryo-sac. There is here a central band of protoplasm running in a 

 longitudinal direction through the embryo-sac, which afterwards breaks up 

 into a number of plates. Heliotro'pium had been described by Kosanoff as 

 exhibiting free- cell-formation in the upper, cell-division in the lower part 

 of the embryo-sac ; Hegelmaier was unable to confirm this ; finding, in the 

 lower part as well as the upper, the formation of septa coincident with the 

 division of the nucleus. Strictly speaking, in Heliotro'pium the embryo-sac 

 itself does not divide in the formation of endosperm, but only its contents ; 

 and this is still more strikingly the case in Specularia. 



In Atropa the endosperm is the result of true cell-division ; as is also 

 the case in Asarum. In the Labiates the embryo-sac is divided by an 

 isthmus-like neck into two parts of unequal size, and the formation of 

 endosperm takes place only in the smaller of these by cell-division. 



The author next points out differences in the mode of development of 

 the endosperm in Nujpliar and Nymphsea, although belonging to the same 

 type. 



He concludes that no sharp line of demarcation can be drawn between 

 different modes of development of the endosperm, whether you take as the 

 distinguishing character the coincidence or otherwise of the formation of 

 septa with the division of the nucleus (cell- division and free-cell-formation), 

 or the fact that in some cases it is the embryo-sac itself that divides, in 

 other cases only its protoplasmic contents. 



(4) Growth. 



Effect of Sunlight on Etiolated Seedlings.* — According to Dr. J. 

 Eeinke, when etiolated seedlings of cress were placed under a normal solar 

 spectrum, the green colouring invariably took place most rapidly on both 

 sides of the line C, about in the interval X 635-A 675 ; the curve falling 



* SB. Versamml. Deutsch. Naturf. u. Aerzte, Sept. 20, 1886. See Bot. Centralbl., 

 xxviii. (1886) p. 94. 



