celite (1 part residue:2.1 parts celite, w/w). The dried residue-celite 
mixture was packed in a chromatographic column and successively 
percolated with petroleum ether, ether, acetone, and methanol. In each 
case, fresh solvent was continually passed through the column until the 
weight of material removed from the column became negligible. This 
entire procedure permitted gross separation of £3 on the basis of relative 
solubilities. The petroleum ether (E7) and ether (E8) percolates were 
then extensively studied after initial separation of the residues of each 
percolate into neutral and acidic and basic substances. The methods of 
isolation and identification were similar to those described previously (2, 
9, 10). The residues from the acetone (E9) and methanol (E10) percolates 
were found to contain mostly water-soluble materials, which were not of 
interest in this study, and extensive attempts to identify the relatively 
small amounts of water-insoluble substances were not made. 
The lower layer (E4) was diluted fourfold with water. After 
refrigeration at 5° C. for 1 month, the mixture was filtered through celite 
and gave a filtrate (E12) and a precipitate (Ell) on celite. The precipitate- 
celite mixture was exhaustively extracted with methanol and the methanol 
extracts were evaporated to a residue. Extraction of the residue with 
water gave water-soluble (E13) and water-insoluble (E14) fractions. 
Table 1 shows the weights and composition of the water-insoluble 
substances found in the fractions. Table 2 gives the total amounts of 
water-soluble and water-insoluble substances in the ethanol extractives. 
The weights and composition of the ethanol extractives are given in table 3. 
More than 92 percent of the ethanol extractives consisted of water-soluble 
material, 
The final compilation of the composition of total n-hexane and ethanol 
extractives is shown in table 4. The values are given to the third decimal 
place to show relative quantities isolated, and no statistical significance 
is intended. Actually all values should be considered as semiquantitative 
estimates, since the isolation methods are not analytical procedures. 
The nature of the paraffins, neophytadiene, polyene, phthalates, sterols, 
sterol glycosides, tocopherols, fatty acid salts, higher fatty acids, neutral 
resins, and substances not eluted from silicic acid or alumina columns 
with methanol have been reported (2, 9, 10, 1l, 12, 13). Solanesol (3, 5) 
and scopoletin (6) are well-known components of tobacco. The ''Fatty 
Acid Esters" in the hexane extract (10, 11) and, in part, the ethanol 
extract (9) have also been reported. The ''Unidentified Acids" consisted 
mainly of either thick, highly colored resins or black, amorphous tars, 
which showed infrared spectra, with strong absorption for the carboxyl 
and carbonyl groups. Perhaps much of this material could be 
described as acidic resin (4). 
Comparison of these data with theresultsof recent quantitative studies 
on tobacco constituents is of interest. Analytical methods for the 
determination of paraffinic hydrocarbons (8), sterols (7), higher fatty 
5 
