This is a complex mixture produced to the extent of a few tenths per cent. in ordinary alcoholic fermentation. Thus from four patent-still distilleries the following quantities of fusel oil were produced for every 100 gallons of absolute alcohol manufactured, the materials fermented being grain cereals: - l

1 Appendix Q, Royal Commission on Whisky Report, 1909.

Fusel oil consists chiefly of four higher alcohols, namely, normal propyl, isobutyl, active amyl, and inactive amyl alcohols. The last is the largest constituent. Small proportions of esters and free fatty acids are also present, with more or less ethyl alcohol and water.

In addition to the foregoing, other alcohols of the butyl and amyl series, especially normal butyl, have been noted as occurring in fusel oil, and also small quantities of hexyl and heptyl alcohols, furfural, terpenes, and basic substances.

By the use of special ferments, starch mashes may be directly fermented to acetone and fusel oil, with little or no ethyl alcohol (Fernbach). The fusel oil thus obtained contains a large proportion of butyl alcohol. The acetone is separated from the oil by fractional distillation.

As regards the source of the higher alcohols in ordinary fusel oils, Ehrlich1 has shown that they are formed by the action of yeast upon the amino-acids which result from the decomposition of proteid bodies present in the mash and also in the yeast itself. Leucine, (CH3)2CHCH1CH(NH1)COOH, yields inactive amyl alcohol. Similarly, isoleucine yields the active amyl alcohol.

In order that this may take place, the nitrogenous residues of the amino-acids must be split off in the form of ammonia, but this substance cannot be detected in the liquid. Ehrlich's conclusion, therefore, is that the ammonia is assimilated as fast as it is formed, and is utilised in the production of the nitrogenous constituents of the growing yeast. The production of fusel oil depends upon whether the yeast derives its nitrogenous nutriment from amino-acids, or from other sources which yield assimilable nitrogen more readily - as, for example, ammonium salts. When these are present, the amino-acids are not attacked, and little or no fusel oil is produced. In a solution of pure sugar containing no other more readily-assimilated nitrogenous nutriment, the yield of fusel oil can be increased either by adding amino-acids or by so arranging the conditions that they are unfavourable to the growth of the yeast, and thus inducing it to excrete a part of its own nitrogen in the form of amino-acids. In the ordinary fermentation of pure sugar by yeast, the percentage of fusel oil calculated on the total yield of alcohol ranges from about 04 to 07 per cent. When leucine is added to the solution, the production of fusel oil may rise to as much as 3 per cent., the maximum yield being obtained when the proportion of yeast to sugar is one to five.1

1 Ber., 1907, 40,1027.

In connection with Ehrlich's view that under certain conditions the yeast itself may yield up some of its nitrogen, it may be mentioned that J. R. Carracido had previously put forward a somewhat similar suggestion in discussing the mechanism of the production of glycerol during alcoholic fermentation.2 He attributed the formation of glycerol, not to the decomposition of sugar by yeast, but to an autolytic destruction of the yeast itself, probably owing to the action of an enzyme on the protein material of the yeast-cells.3