At the commencement of the process of aggregation, more especially in dark red leaves, the contents of the cells often present a different appearance, as if the layer of protoplasm (primordial utricle) which lines the cells had separated itself and shrunk from the walls; an irregularly shaped purple bag being thus formed.  Other fluids, besides a solution of the carbonate, for instance an infusion of raw meat, produce this same effect.  But the appearance of the primordial utricle shrinking from the walls is certainly false;* for before giving the solution, I saw on several occasions that the walls were lined with colourless flowing protoplasm, and after the bag-like masses were formed, the protoplasm was still flowing along the walls in a conspicuous manner, even more so than before.  It appeared indeed as if the stream of protoplasm was strengthened by the action of the carbonate, but it was impossible to ascertain whether this was really the case.  The bag-like masses, when once formed, soon begin to glide slowly round the cells, sometimes sending out projections which separate into little spheres; other spheres appear in the fluid surrounding the bags, and these travel much more quickly.  That the small spheres are separate is often shown by sometimes one and then another travelling in advance, and sometimes they revolve round each other.  I have occasionally seen spheres of this kind proceeding up and down the same side of a cell, instead of round it.  The bag-like masses after a time generally divide into two rounded or oval masses, and these undergo the changes shown in figs. 7 and 8.  At other times spheres appear within the bags; and these coalesce and separate in an endless cycle of change.

After leaves have been left for several hours in a solution of the carbonate, and complete aggregation has been effected, the

* With other plants I have often seen what appears to be a true shrinking of the primordial utricle from the walls of the cells, caused by a solution of carbonate of ammonia, as likewise follows from mechanical injuries. [page 47]

stream of protoplasm on the walls of the cells ceases to be visible; I observed this fact repeatedly, but will give only one instance.  A pale purple leaf was placed in a few drops of a solution of one part to 292 of water, and in 2 hrs. some fine purple spheres were formed in the upper cells of the pedicels, the stream of protoplasm round their walls being still quite distinct; but after an additional 4 hrs., during which time many more spheres were formed, the stream was no longer distinguishable on the most careful examination; and this no doubt was due to the contained granules having become united with the spheres, so that nothing was left by which the movement of the limpid protoplasm could be perceived.  But minute free spheres still travelled up and down the cells, showing that there was still a current.  So it was next morning, after 22 hrs., by which time some