28 



EXPERIMENTS ON PROCESSING PERSIMMONS. 



To the other tube a simple mercury trap was attached so that gases 

 could readily find their way out but could not enter the desiccator. 

 In no case, however, were any changes of pressure observed. 



After five days samples of the gases were drawn from the top and 

 bottom of each desiccator. The analyses of the gases were made by 

 the usual Hempel methods, taking due precautions to avoid con- 

 tamination with the outside air. The analyses are given in the fol- 

 lowing table: 



Analyses of gases from desiccators filled with persimmons, after standing five 





Description. 



Analyses of gases. 



Lot. 



Carbon 

 dioxid 

 (C0 2 ). 



Oxygen 

 (62). 



Nitrogen 

 by differ- 

 ence (N2). 



A 



Nothing added; top 



45.7 

 46.4 

 52.6 

 54.6 

 47.8 

 54.4 

 84.1 

 89.6 



0.8 

 1.0 



.6 



.0 

 2.7 



.3 

 1.1 



.0 



53.5 







52.6 



B 



Starch added; top 



46.8 



C 



Starch added; bottom 



Sand added; top *- . 



45.4 

 49.5 





Sand added; bottom 



45.3 



D 



Carbon dioxid added: top 



14.8 





Carbon dioxid added; bottom 



10.4 









They show that a fairly complete absorption of oxygen occurred in 

 lots A, B, and C. In lot C the sample of gas withdrawn from the top 

 of the desiccator was from above the surface of the sand next the 

 cover. The oxygen was here much higher than in the other samples. 

 As is usual in such experiments, the oxygen was found to have been 

 displaced by much more than an equal volume of carbon dioxid, an 

 illustration of the well-known fact that carbon dioxid evolution con- 

 tinues when anaerobic conditions are imposed. Each desiccator con- 

 tained some of all of the iiye varieties. In the desiccator containing 

 the fruit alone, Tsuru was found to be softening slightly, but the 

 fruits were extremely astringent. The two Hachiya specimens were 

 both softening and were very astringent. Of the three Tane-nashi 

 fruits used, two were firm and one was softening. The softening fruit 

 was but very slightly astringent and processing was perceptible. 

 Eight specimens of Triumph were used and all were firm save one, 

 which was just commencing to soften and was distinctly astringent. 

 The variety Zengi was firm and nonastringent. 



The five specimens of Tsuru from the desiccator containing starch 

 were all firm and very astringent. Of the five Hachiya fruits em- 

 ployed, two were beginning to soften and were very astringent. The 

 two fruits of Tane-nashi were both firm. One was nonastringent, 

 while the other contained enough tannin to be inedible. Two of the 

 nine Triumph fruits employed were soft and free from soluble tannin; 

 and the firm fruits seemed rather less astringent than normal per- 



