ZOOLOGY AND BOTANY, MICROSCOPY, ETC. L'71 



the opening takes place as before. Finally, it is proved that Steinbrinck's 

 vacuum method, and also the method involving the use of solutions for 

 removing water, are both unsuitable for testing the force of cohesion. 

 The second section deals with the hygroscopic mechanism for opening 

 the anthers. It is shown that when the fibres are dried a contraction 

 takes place in both length and breadth, the contraction being greatest in 

 a direction perpendicular to the thin membranes ; the latter are consider- 

 ably more active than the fibres in bringing about the opening, which is 

 thus primarily due to them, while the fibres form a support to the inner 

 walls. Both the structure of the individual cells and the arrangement 

 of the fibres favour the opening of the valves. It is thus possible to 

 explain the dehiscence of the anthers where the fibres are arranged so as 

 to form a support on the inner walls of the cells ; but further investiga- 

 tions are needed to discover the cause where this is not the case, e.g. 

 where they are arranged in an annular or spiral manner. 



Opening of Achenes in the LiguliflorEe.*- — P. Lebard publishes 

 the results of his work upon the opening of the achenes of the Liguli- 

 floroe at the time of germination. These achenes consist of two parts, 

 viz. the true achene which encloses the seed, and the podocarp or 

 podogyne ; the latter organ is most conspicuous in Podospermum, less so 

 in Tragopogon, and by examination of these two genera it is shown to be 

 the morphological equivalent of the peduncle ; the structure forming its 

 axis lines a hollow cavity continuous with that of the achene, and 

 through the walls the fibro-vascular bundles pass from the receptacle to 

 the pericarp. In the achene itself there are several longitudinal fibrous 

 bands alternating with bands of parenchymatous or lacunar tissue. The 

 fibrous bands coalesce towards the apex, until in Tragopogon pratensis 

 they form a continuous layer rendering dehiscence impossible in that 

 region. There are two chief types of dehiscence. In a few genera, 

 e.g. Tragopogon, Podospermum, etc., several slits occur at the base of 

 the achene, frequently extending into the podogyne ; in all other 

 genera, a few slits are formed, and owing to the thrust of the radicle and 

 the expansion of the cotyledons, one of these slits is prolonged either 

 halfway or the entire length of the achene ; dehiscence then results 

 from the pressure of the embryo and its increase in volume consecjuent 

 upon increased hygrometric condition of the atmosphere. Podospermum 

 laciniatum is an exception in that the radicle has no part in the splitting 

 of the fruit-wall, which is due solely to the action of humidity on the 

 fibrous bands ; thus, this achene forms a transition stage between 

 dehiscent and indehiscent fruits. Finally, the form of the cotyledon 

 varies with the method of dehiscence. Where there are several basal 

 slits, the cotyledons are needle-like ; where there is one main slit they 

 have a broad lamina ; and where the slits extend only half-way up the 

 pericarp the cotyledons assume an intermediate form. 



Development of Xyris.f — S. Weinzieher has studied the life-history 

 of the development of Xgris indica, with the following results. The 



* Rev. Gen. Bot., xxv. (1913) pp. 432-42 (1 pi. and 15 figs.), 

 t Flora, vi. (1914) pp. 393-432 (2 pis. and 10 figs.). 



