134 MISSOURI BOTANICAL GARDEN. 



it migrates into the surrounding integuments. The facts 

 described above do not consist, I believe, with the latter alter- 

 native, but are quite applicable to the former. 



The appearance of tannin simultaneously with the begin- 

 ning of secondary thickening of the endosperm cell walls has 

 a good deal about it to warrant the belief that the tannin 

 plays some role during the formation of reserve cellulose. 

 The reaction is not in the cell lumen, but definitely in the cell 

 wall, and, in the more deeply reacting cells, in the primary 

 membranes. This appears not to result from displacement 

 of tannin 22 as the reaction appears at first in the cell wall. 

 In copper acetate material, in which we may imagine dis- 

 placement of tannin from the lumen into the wall to have oc- 

 curred, the tannin reactions in other tissues (e. g. the raphe) 

 are in the lumina. Ethyl nitrite material gave identical 

 results in the clearest and most convincing fashion. The 

 endosperm, as stated elsewhere, did not show coloration, but, 

 upon the application of iron acetate, gave a prompt, iron- 

 blue reaction in the cell-walls, but not in the lumina. In 

 the integuments and raphe, a similar reaction was given 

 in the lumina of the tannin cells. 



This parallelism between the distribution of tannin and 

 the thickening of the cell walls, and the ultimate disappear- 

 ance of the tannin when the endosperm is fully developed, 

 indicate an analogy between the tannin in question and 

 other nutrients in transitory form. For this view speaks 

 the fact that the tannin in the endosperm is throughout 

 diffusible, since, in contrast to the tannin in specialized cells, 

 it is found permeating the cell walls. This conclusion is 

 justified from another point of view. Strecker, according to 

 J. Reynolds Green 23 has shown that a "tannin" is a glucoside, 

 yielding gallic acid and glucose by hydrolysis. The fer- 

 mentation of infusions of certain galls, yielding, with the dis- 

 appearance of tannin, gallic and ellagic acids, (also pointed 



22 See above, p. 132. 



23 The Soluble Ferments and Fermentation, p. 160. See also 

 Dekker, 1906. 



