THE section (II.) that Dr. Frankland devotes to his researches in Applied Chemistry is not the least interesting of the work, though the chief topics are Gas and Water. The author's investigation of White's process for manufacturing hydrocarbon gas by passing steam over red-hot coke, and carbonising the gas in the retort, led to the clear distinction of the illuminating from the non-illuminating constituents of the hydrocarbon gas and of ordinary coal-gas. It was shown that marsh gas is valueless as a light producer during combustion, and that the luminosity of a gas flame is due to the heavier hydrocarbons present, whose illuminating value can be deduced from analysis and expressed in terms of olefiant gas. Such an indirect method of estimating the illuminating value of a sample of gas is certainly interesting, but it is clearly unsafe; for it involves the assumption that the illuminating value is directly proportional to the percentage of a hydrocarbon mixture of unknown constitution, calculated somewhat empirically into equivalents of the well-defined ethylene. Moreover, some recent experiments by Dittmar seem to show that ethylene does not contribute nearly so much to the luminosity of a hydrogen gas flame as benzole vapour. Hydrogen containing as much as 10 per cent. of ethylene gave a very feebly luminous flame, while hydrogen charged with only 3 per cent. of benzole vapour afforded a brilliant light when the gas was burned. Fortunately Dr. Frankland does not wholly rely upon the method in comparing—as he does in his introductory remarks on the gas investigations—the London supply of 1851 with that of 1876; for he has partially employed the photometer as a check. In 1851 the London gas supply contained 701 per cent. of olefiant gas, or its equivalent of other illuminating hydrocarbon, while in 1876 the percentage was 702. From these data it was concluded that there was no difference in illuminating power although the 1876 gas should be, according to the Act of Parliament, four candles better than that of 1851. Dr. Frankland says:— Experimental Researches in Pure, Applied, and Physical Chemistry. By E. Frankland, &c. (London: Van Voorst.)