Tannin in Persimmons. 7 



the cells, the "tannin" and protoplasm will be seen to mingle 

 with the sorrounding fluid. A preparation of this kind which 

 stood for sixteen hours showed, upon the addition of ferric 

 chlorid, that the "tannin" was still to be seen as sheets and 

 strings, accompanied with protoplasm. The protoplasm, cell 

 walls and tannin-masses all showed a blue reaction, and the last 

 were coagulated. It seems therefore that the tannin in the un- 

 ripe fruit is partly separable from the mobile semi-fluid which holds 

 it in the cell, and that in this situation it is not a water-tannm 

 solution. The initial blue coloration in the coagulum fades 

 away into dirty brown, the change noted in tannin-gelatin. 



A droplet of water-egg-albumen mixture. was heated to 

 coagulation on a slide, treated with tannin, washed off thoroughly 

 and stained with ferric chlorid. A little unripe fruit juice got 

 by scraping the hard pulp was similarly treated, except that the 

 exposure to tannin was omitted. The resulting coagula were so 

 similar that the only difference observable was that the 

 albumen-tannin appeared to hold the blue color longer than 

 tannin-gelatin. 



In these experiments the similarity of artificial tannin- 

 compounds used and the tannin-mass is highly suggestive, and 

 the conclusion is hard to resist that the tannin-mass is a com- 

 pound, and not a solution in mere cell-sap. 



I have, however, sought for more direct evidence by at- 

 tempts to separate the tannin from the supposed carrier, and 

 with a measure of success. By means of treatment with dilute 

 nitric acid, the whole mass was completely dissolved in most of 

 the cells, leaving only the partly hydrolysed wall and some gran- 

 ular matter. In some of them, however, there remained a yellow, 

 hyaline or very slightly granular shell, which gave no tannin 

 reaction(figs. 12,13). The wall of this shell was about three microns 

 thick. To account for this substance aside from its origin in the 

 tannin-mass, is difficult. Its phvsical characters are like those 

 of the tannin-mass before extreme hardening. This matter needs 

 further study, but as it stands it gives rather strong support to 

 my main contention. 



The progressive dehydration above mentioned appears to 

 Howard to be connected with the fixation or rendering insoluble 

 of the tannin. I base this statement on the explanation given 

 by him as to wh}- an astringent after-taste is had on eating 



