INTRODUCTION. 13 



are disputed points. The same is true also of the effects of lime and 

 magnesia on the burning qualities. The two facts which have not 

 been contradicted are insufficient in themselves to explain the burn- 

 ing qualities of different samples of tobacco, for it rarely happens 

 that tobacco contains enough chlorin to produce any injurious effects, 

 and it frequently happens that samples very rich in potash have a 

 poor burn, while others comparatively poor in this constituent show 

 excellent burning qualities. 



It seemed quite possible that some further light might be thrown 

 on the subject by extracting different samples of tobacco with various 

 solvents and noting the effect on the burning qualities. It should be 

 stated here that all samples of tobacco used in the experiments de- 

 scribed in this paper had been thoroughly fermented, and the results 

 are not intended to be applied to unfermented tobacco. Extraction 

 of various samples with petroleum ether and with ordinary ethyl or 

 sulphuric ether did not appreciably affect the burning qualities. 

 When strong alcohol was used as the solvent the same result was 

 obtained, except in a single .case, where the fire-holding capacity was 

 considerably improved by the extraction. 



More striking results are obtained when tobacco is extracted with 

 water. The following simple experiment is very interesting and 

 instructive: A leaf of tobacco having a good glowing capacity is 

 divided along the midrib into two parts, one of which is extracted 

 for forty-eight hours with a comparatively large volume of distilled 

 water. After being dried the extracted portion of the leaf will be 

 found to have entirely lost its glowing capacity. Now, if the extract 

 be evaporated down to a very small volume and a bit of the extracted 

 leaf saturated therewith and dried, it will once more show a good 

 fire-holding capacity. Whether this is less or greater than the orig- 

 inal leaf possessed will depend, of course, on the concentration of the 

 extract. This extract will further impart a good burn to filter paper 

 and to other samples of tobacco showing poor burning qualities. 

 This simple experiment seems to prove conclusively that the active 

 principle or principles in imparting to the tobacco leaf its capacity 

 for holding fire can be extracted with water. 



The problem, then, is to determine the composition of the extract 

 and to - discover which of its constituents contribute to the burning 

 qualities of the tobacco. One hundred grams of tobacco having a 

 good burn were extracted with 1 liter of distilled water ; the extract 

 was poured off and the tobacco again extracted with the same quan- 

 tity of water for twenty-four hours longer. The extracts thus ob- 

 tained were combined, filtered, and evaporated to about 150 c. c. 

 During the evaporation a considerable quantity of calcium citrate 

 separated out, which was removed by filtration. The filtrate was 



105 



