CHAP. III. THE PROCESS OF AGGREGATION. 61 



under a high power, is the appearance of the finest 

 granules in the fluid within the cells, making it 

 slightly cloudy. These granules soon aggregate into 

 small globular masses. I have seen a cloud of this 

 kind appear in 10 s. after a drop of a solution of car- 

 bonate of ammonia had been given to a gland. With 

 dark red leaves the first visible change often is the 

 conversion of the outer layer of the fluid within the 

 cells into bag-like masses. The aggregated masses, 

 however they may have been developed, incessantly 

 change their forms and positions. They are not filled 

 with fluid, but are solid to their centres. Ultimately 

 the colourless granules in the protoplasm which flows 

 round the walls coalesce with the central spheres or 

 masses ; but there is still a current of limpid fluid 

 flowing within the cells. As soon as the tentacles 

 fully re-expand, the aggregated masses are redis- 

 solved, and the cells become filled with homogeneous 

 purple fluid, as they were at first. The process of re- 

 dissolution commences at the bases of the tentacles, 

 thence proceeding upwards to the glands ; and, there- 

 fore, in a reversed direction to that of aggregation. 



Aggregation is excited by the most diversified 

 causes, by the glands being several times touched, 

 by the pressure of particles of any kind, and as these 

 are supported by the dense secretion, they can hardly 

 press on the glands with the weight of a millionth of 

 a grain,* by the tentacles being cut off close beneath 



* According to Hofmeister (as gation is a different phenomenon, 



quoted by Sachs, ' Traite de Bot." as it relates to the contents of the 



1874, p. 958), very slight pres- cells, and only secondarily to the 



sure on the cell-membrane arrests layer of protoplasm which flows 



immediately the movements of along the walls ; though no doubt 



the protoplasm, and even deter- the effects of pressure or of :i 



mines its sepatation from the touch on the outside must be 



walls. But the process of aggre- transmitted through this layer. 



