STUDIES ON THE CUEING OF LEAF TOBACCO. 25 



In the experiments with picked leaves any apparent differences in 

 the ash content- of the whole leaf as between the green and cured leaves 

 are, of course, due to slight differences in the composition of the dupli- 

 cate samples and to experimental errors. As between the two leaf 

 parts, however, it is to be observed that in every instance the leaf 

 web appears to gam in ash content at the expense of the midrib dur- 

 ing the curing. There is no doubt that this movement of mineral 

 matter from stem to web takes place during the later stages of dry- 

 ing, and tobacco growers are familiar with the discoloration fre- 

 quently caused thereby in the leaf in the vicinity of the stem. The 

 movement is doubtless due simply to diffusion and is not to be con- 

 nected with the reverse movement due to physiological translocation. 

 As is well known, the leaf web normally dies and dries long before the 

 stem, and the latter finally collapses rather suddenly. The remain- 

 ing cell sap then oozes out into the leaf web, causing the discoloration 

 referred to. This movement of materials, occurring after the death 

 of the protoplasm, is not large and is localized. In the experiments 

 in stalk curing there is a loss of ash in curing in the leaf as a whole, 

 which, of course, can not be due to respiration and can only be 

 accounted for by the assumption that a portion of the ash has passed 

 into the stalk. 



CHANGES IN THE CARBOHYDRATES. 



In both methods of curing, nearly all the starch disappears both 

 from the leaf web and from the midrib during the process. We have 

 been unable to obtain a reaction for starch by the iodin test in the 

 cured samples. This is in accordance with the results of Mliller- 

 Thurgau already referred to. The more or less complete disappear- 

 ance of starch is one of the most characteristic changes involved in 

 curing, and the relative freedom from starch of the cured leaf is a 

 measure of the completeness of the curing process. It will be seen 

 also that in complete curing by either method practically all of the 

 reducing sugars disappear. The incompletely cured samples of 1908 

 reveal the fact, however, that the disappearance of starch precedes 

 that of the sugars, as is to be expected, and, furthermore, that there 

 is a temporary accumulation of sugars in the midrib, undoubtedly 

 derived from the starch of the leaf web. It has been pointed out 

 that analyses of flue-cured leaf have shown that under that system 

 of rapid curing the leaf contains high percentages of reducing sugars, 

 which are doubtless derived from the splitting up of starch, and the 

 premature killing of the protoplasm prevents the oxidation of the 

 sugars by respiration . 



Our results as a whole indicate that the pentosans, which, gener- 

 ally speaking, are not physiologically plastic, undergo but little change 

 in the leaf web, but in the stems there is a more decided decrease of 

 these constituents. The crude-fiber content undergoes little or no 

 change, except that in complete curing there appears to be a slight 



