STUDIES ON THE CUEING OF LEAF TOBACCO. 17 



Maryland export leaf reported is of interest chiefly because of its 

 unusually high content of cellulose, since this type is characterized 

 by its "chaffiness." The analyses of cigar-wrapper leaf reported by 

 Jenkins and Winton are fairly comparable with our results, shown 

 in Table III, since the material was grown in the same locality and the 

 results obtained in the two cases are not essentially different. 



CHANGES IN COMPOSITION OF LEAF TOBACCO IN AIR CURING. 



While the data presented in Table III show the comparative 

 composition of cured and uncured leaf tobacco in the form there 

 presented, they do not properly bring out the changes in composition 

 which take place during the curing. As shown in Table I, there is 

 a marked but variable loss of dry w T eight in curing, and it is evident 

 that the simple comparison of the percentage composition of cured 

 and uncured leaves does not give a correct index as to the actual 

 changes in content of the various constituents involved in the curing. 

 As pointed out by Mohr, 1 nearly all early investigators have made 

 this error in collecting their data, with the result that many of their 

 conclusions are entirely erroneous. There are two ways in which 

 this difficulty can be overcome. One method is to base all compari- 

 sons on unit areas of the leaf rather than on weights, and this is 

 undoubtedly the most satisfactory method in dealing with a small 

 number of leaves, but where larger quantities of material are 

 required the procedure requires much time and labor. A second 

 method, which has been followed in our experiments, is to collect 

 the leaves in pairs, so that each leaf which is cured shall have a dupli- 

 cate which is quickly dried and weighed, as w T as described on page 4. 

 In this way the total loss in weight in curing is obtained simply 

 by comparing the weights of the cured and uncured leaves, and it 

 is only necessary to correct the results of the analysis of the cured 

 leaves for the total loss of weight in curing. There are necessarily 

 some individual variations in the leaf "pairs," so that this method is 

 not entirely reliable when only a few leaves are used, but if the 

 composite samples include a hundred or more leaves the individual 

 variations are of little or no consequence. 



In Table IV we have corrected the results of the analyses of the 

 cured leaf as shown in Table III for the respective loss in weight in the 

 curing of each sample as shown in Table I, so that all results are cal- 

 culated on the basis of the uncured leaf. We have added data show- 

 ing the gain or loss of each constituent for the two leaf parts and for 

 the whole leaf in each case. Table V presents in a similar manner 

 the data for the total nitrogen and nitrogen in the various forms of 

 combination found in the tobacco. 



1 Mohr, E. C. J. Gepfliickter und am Stamme getrockneter Tabak. Die Landwirtschaftliehen Ver- 

 suchs-Stationen, Bd. 59, Heft 3/4, p. 256, 1903. 



29731°— 14 3 



