6 THE ROYAL SOCIETY OF CANADA 



Either there may be numerous or several vacuoles, or there may be one large one. 

 Sooner or later a small vacuole, or the large one, may burst and the expelled fluid 

 sets as a red ppt. which takes a definite form if not too copious. A single zonation 

 is clearly defined, beyond which a loose zone of precipitate is to be seen. Essentially 

 the same takes place with a half-saturated solution, but more rapidly. By means 

 of a mixture of glycerin and the reagent, the sudden expulsion of fluids may be 

 avoided, and instead they may be expelled slowly, forming precipitation tubes and 

 tube-like structures. 



If a very weak solution of the reagent is used the tannin mass breaks up into 

 minute globules (not in a granular precipitate) which are colourless at first, but 

 soon take on the characteristic reaction colour. 



When the reagent reaches the dry tannin flakes there is an immediate slight 

 solution with colour. The reagent attacks the flakes slowly, etching the surface 

 irregularly. As the etching proceeds, a coarse amorphous precipitate is formed 

 which does not hang together or show any zonation. 



If the weak solution is used no colour is seen at all, but there is gradual solution 

 of the flakes, occupying a few minutes. No precipitate is to be seen. There is no 

 swelling or imbibition previous to dissolving. 



It is to be concluded that in the tannin mass there is another substance present 

 which interferes with the course of the reaction. 



Cells in which the tannin occurs, in association with an easily 

 water-soluble second substance, stand in contrast with those which 

 were described originally by Flueckiger in 1867, and later and more 

 fully by Tichomirow (1884, 1905). I refer to the so-called "in- 

 clusions" observed by these observers in the pericarps of certain 

 fruits {Phoenix, Ceratonia,- Diospyros, Acras, etc.), and in some leaves. 

 It is unnecessary for me to repeat what can be found in that admir- 

 able work "Lehrbuch der Pharmacologie" by Dr. A. Tschirch (1909- 

 12) more than to repeat what that author said in 1912 concerning 

 the tannin masses, namely, that "what they consist of is unknown;" 

 nor is it necessary here to recount many of the details concerning 

 these structures already presented by myself (Lloyd, 1916 and earlier). 

 My purpose now is solely to adduce additional evidence of the presence 

 of a second substance associated with the tannin and of the specific 

 nature of this association: namely, that it consists in the adsorption 

 of the tannin on the associated substance. This has in the case of 

 the persimmon, as I have already said, been identified as a cellulose- 

 like body (Clark I.e.). It is, at all events, a carbohydrate. 



When the tannin cell reaches maturity this substance presents a 

 sort of structure referable to that fact that it is emulsoidal. It is 

 traversed by canals spanning stretches between more or less fusiform 

 cavities, all extensions of superficial crevices, and the whole suggesting 

 its origin as several separate gelatinous masses, later compacted 

 together (Fig. 2). If the cell dies before maturation of the fruit, 

 oxidation of the tannin content intervenes and causes red or red- 



