BY MOLECULAR COALESCENCE. 147 
be constantly added to the contents of the capsule, the 
globules must press one against another, and thus their 
globular form is seen to become destroyed, and to be re- 
placed by the polyhedral form. Besides this, it can be 
seen by imspection of fig. 10 that they are smaller the 
nearer they are to the centre of the lens, and appear to 
have suffered compression, especially in the direction from 
without to within, and therefore they must have lost some 
part of their contents. Now the fact of these cells be- 
coming so rapidly enlarged by the imbibition of water, 
proves that they retain the albumen in solution, and only 
lose the water; consequently, as their sides in their 
natural state are brought nearer together, the fluid within 
them must become more and more inspissated, until 
nothing remains but their dense exterior and the portion 
of albumen which was in solution in the original albumi- 
nous globule. This portion of albumen now blending 
with that which before formed the wall of the globule in 
which it is contaimed, a permanent addition is made to 
the radiating fibre of which the individual globules are 
parts. The close apposition of the walls of a multitude of 
such globules, with the inspissated solution of albumen 
in their interior, constitutes the beautiful transverse 
marking represented in fig. 10 c. And the fact of these 
becoming separated by the imbibition of water is an 
experimental demonstration, that the cause of the collapse 
of these cells is the reverse of the effect produced upon 
them by the action of the water as applied in the experi- 
ment, namely, the loss of their fluid, probably by exosmose. 
Whilst the transverse marking exists in the radiat- 
ing fibres, no defined serration of their edges is ap- 
parent, but as the transverse lines disappear in consequence 
of the perfect blending together of the contiguous particles 
