An interesting case is that of chloroform. It had been found that chloroform prepared from acetone was distinctly inferior as an anaesthetic to that made from alcohol, and the most probable explanation appeared to be the presence of minute quantities of irritating products of decomposition, such as chlorine, carbonyl chloride or hydrochloric acid in the chloroform derived from acetone. The subject was investigated by Wade and Finnemore,1 who found that neither ordinary distillation nor treatment with sulphuric acid, washing, drying, and subsequent rectification through a pear still-head disclosed any difference between samples of chloroform prepared respectively from acetone and alcohol.

The problem was eventually solved by them with the aid of an evaporator still-head of five sections, which was employed in all the subsequent distillations. That chloroform and alcohol form a mixture of minimum boiling point was. shown by Thayer,2 and Wade and Finne-more isolated and examined not only this binary azeotropic mixture but also a ternary chloroform-alcohol-water mixture, boiling constantly at 55.5° and containing 4.0 per cent of alcohol and 3.5 per cent of water. Now anaesthetic chloroform always contains a little alcohol and a trace of water, and it was found that with anaesthetic chloroform prepared from acetone distillation commenced at 55.5°, and there was also distinct evidence of the formation of the binary chloroform-alcohol mixture, which boils at 59.4°. With chloroform made from alcohol, on the other hand, distillation began at 54°.

It therefore appeared that the "acetone " chloroform contained no foreign substance beyond the alcohol and water which are always present, whereas the " alcohol " chloroform contained a minute quantity of a volatile impurity. It was suspected that this might be ethyl chloride, and it was found that on distilling chloroform containing 0.5 per cent of alcohol, a trace of water, and 0.2 per cent of ethyl chloride, the liquid began to boil at 51° and the indications of the binary chloroform-alcohol mixture were obliterated. With only 04 per cent of ethyl chloride distillation commenced at 52.65°, and there was clear indication of the formation of the binary chloroform-alcohol mixture.

Finally, direct chemical proof of the presence of ethyl chloride was obtained in the first fractions, not only from the mixtures to which minute quantities of that substance had been added, but also from the chloroform prepared from alcohol. Not the slightest indication of its presence, on the other hand, could be detected in the first fraction from " acetone " chloroform.

It was thus proved that the superior anaesthetic properties of "alcohol' chloroform are due to the presence of a trace of ethyl chloride - probably about 0.05 per cent or 2 c.c. per Winchester quart1 Wade and Finnemore, " Influence of Moist Alcohol and Ethyl Chloride on the Boiling Point of Chloroform," Trans. Chem. Soc, 1904, 85, 938.

2 Thayer, " Boiling Point Curves," J. Amer. Chem. Soc, 1899, 3 36.

And it was found by Mr. Rowell at Guy's Hospital that " acetone" chloroform to which a small quantity of ethyl chloride has been added is therapeutically identical with chloroform made from alcohol. Wade and Finnemore suggest that the addition of a further quantity of ethyl chloride would be even more beneficial in decreasing the time of induction and generally facilitating anaesthesia. Loss by evaporation is trifling even when the chloroform contains as much as 1 per cent of ethyl chloride.

c. Mixtures of Constant Boiling Point. - When two or more of the components form mixtures of constant boiling point the difficulty in interpreting the experimental results is greatly increased.

Benzene In American Petroleum

For example, if we subject a quantity of American petroleum to distillation, collecting fractions from 30-60°, 60-90°, and 90-120°, and then treat each fraction with a mixture of strong nitric and sulphuric acids, we shall find that a considerable amount of dinitrobenzene will be obtained from the fraction from 60 - 90° and of dinitrotoluene from that between 90° and 120°. The obvious conclusion from these results is that benzene and toluene are present in American petroleum, and this was, in fact, stated long ago to be the case by Schorlemmer.

If, however, instead of collecting the distillate in three large fractions, we were to separate it into nine with a range of 10 degrees each, it would then be found, after repeated distillation, that on treating each fraction with the mixed acids, dinitrobenzene would be obtained almost exclusively from the distillate that came over between 60° and 70° and dinitrotoluene from that between 90° and 100°, although the boiling points of benzene and toluene are 80.2° and 110.6° respectively.

Possible Explanations Of Results

In order to explain these facts two assumptions may be made ; either (a) it is not benzene and toluene that are present in American petroleum, but some other more volatile compounds from which dinitrobenzene and dinitrotoluene respectively are formed by the action of nitric and sulphuric acids ; or (b) benzene and toluene, when mixed with the paraffins present in petroleum come over chiefly at temperatures 14° or 15° below their true boiling points.

The first assumption, so far as the conversion of a hydrocarbon other than benzene into dinitrobenzene is concerned, was actually made by one observer, but it is practically certain that the second is the correct one, for it has been found, as already stated, that when a mixture of normal hexane (b.p. 68.95°) with, say, 10 per cent of benzene, is distilled, the benzene cannot be separated but comes over with the hexane at 68.95°. Lastly, the fact that when American petroleum is distilled, the greater part of the benzene comes over below 69° (mostly at 65 - 66°) is explained by the presence of isomeric hexanes which boil at lower temperatures than normal hexane and also carry down the benzene with them.