igii] Francis Erncst Lloyd 15 



The form o£ the central cavity with its flattened arms is very simi- 

 lar to the branching arms of dough formed by the coalescing of 

 bubbles of gas in rising bread, many of which are preserved in form 

 by the baking, and are therefore seen on slicing the bread. 



In Order to explain, on the same grounds, the remaining details, 

 especially those in definitively matured tannin-cells, we must invoke 

 the behavior of gelatin and similar colloid masses during coagula- 

 tion and shrinkage. Those who have iised glycerol-jelly as a 

 mounting medium to any considerable extent, are familiär with 

 the Gurions tunnelling and cavity formation which frequently 

 ensues upon the drying out of the preparation. The resemblance 

 to these of the spheroidal and hemispheroidal cavities, and of the 

 numerous blind canals which often occur, is close enough to 

 Warrant us in advancing this purely physical explanation of the 

 appearances in the tannin-mass, especially if we may add that the 

 soft and yielding elements of which it is composed may suffer 

 displacements. 



On this view, then, the canals are merely the Spaces at the angles 

 of intercepting planes of contact of adjoining masses of the jelly- 

 like bodies which compose the tannin-mass, and are analogous 

 therefore to the intercellular spaces between the walls of paren- 

 chyma cells. Their irregularity results from the irregularity of 

 the component elements of the tannin-mass, aggravated by prob- 

 able displacements. Shrinkage, or coagulation, after the fashion 

 of cellulose hydrates and jellies as above indicated, will account 

 for secondary changes leading to the formation of cavities of 

 various shapes. If this shrinkage is not symmetrical, its lack of 

 symmetry may be due to the uneven structure of the masses. For 

 the purpose of the account which follows, I shall assume the correct- 

 ness of this view, but while I believe in its essential truth, I do not 

 overlook the possibility that further study of younger material 

 than I have yet examined may make some modification of it 

 necessary. 



We may next ask concerning the contents of the canals and 

 cavities within the tannin-mass. We can dispose of the originally 

 superficial, usually hemispherical cavities mentioned by Howard 

 (1906) and myself, as I have already done, since they are found 



