42 DROSERA ROTUNDIFOLIA. [Chap. III. 



the exterior; the latter alone being in some cases penetrated by the 

 fissures. In niiiny of the spheres the line of separation iKdween 

 the outer and inner parts was tolerably well deiined. The outer 

 parts were of exactly the same very pale purple tint, as that of 

 the last- formed smaller spheres; and these latter did not include 

 any darker central core. 



From these several facts we may conclude that, when vii^orous 

 dark-coloured leaves are subjected to the action of carbonate of 

 ammonia, the fluid within the cells of the tentacles often aggre- 

 gates exteriorly into coherent viscid matter, forming a kind of 

 bag. Small spheres sometimes appear within this bag, and the 

 whole generally soon divides into two or more spheres, which re- 

 peatedly coalesce and redividc. After a longer or shorter time the 

 granules in the colourless layer of protoplasm, which flows round 

 the walls, are drawn to and unite with the larger spheres, or form 

 small independent spheres; those latter being of a much paler 

 colour, and more brittle than the first aggregated masses. After 

 the granules of protoplasm have been thus attracted, the layer of 

 flowing protoplasm can no longer be distinguished, though a cur- 

 rent of limpid fluid still flows round the walls. 



If a leaf is immersed in a very strong, almost concentrated, solu- 

 tion of carbonate of ammonia, the glands are instantly blackened, 

 and they secrete copiously; but no movement of the tenUicles en- 

 sues. Two leaves thus treated became after 1 hr. flaccid, and 

 seem killed; all the cells in their tentacles contained spheres of 



f)rotoplasm, but these were small and discoloured. Two other ^ 

 eaves were placed in a sohition not quite so strong, and there was * 

 well-marked aggregation in 30 m. After 24 hrs. the spherical or 

 more commonly oblong masses of protoplasm became opaque and 

 granular, instead of being as usual translucent: and in the lower 

 cells there were only innumerable minute spherical granules. It 

 was evident that the strength of the solution had interfered with 

 the completion of the process, as we shall see likewise follows from 

 too great heat. 



All the foregoing observations relate to the exterior tentacles, 

 which are of a purple colour; but the green petlicels of the short 

 central tentacles are acted on by the carbonate, and by an infusion 

 of raw meat, in exactly the same manner, with the sole dilFerence 

 that the aggregate masses are of a greenish colour; m tliat the 

 process is in no way dependent on the colour of the fluid within 

 the cells. 



Finally, the most remarkable fact with respect to this salt is 

 the extraordinary small amount which suflTices to cause aggrega- 

 tion. Full details will be given in the seventh chai)ter, and here 

 it will be enough to say that with a .sensitive leaf the absorption 

 by a gland of TxAinr "f a grain (.000 tS2 mgr.) is enough to cau.se 

 in the course of one hour well-marked aggregation in the cells ira- 

 nieiliately l)eneath the gland. 



The Kffrctn of certain other Salts and Fluids. Two leaves were 

 placetl in a solutir>n of one part of acetate of ammonia to about 

 140 of water, and were acted on quite as energetically, but perhaps 



