The style of the book is concise and a mathematical treatment predominates. The historical development of any subject is avoided so that the reader hardly knows to whom he may attribute certain views advanced, by no means current among physical chemists. This avoidance of the historical treatment detracts much from the value of the book, for in such a subject where there still exists a conflict of views on many points, the only satisfactory treatment is the historical.

The statement of the author (p. iv) that "whoever works through" the book "conscientiously, paying particular attention to the problems, will get a very fair idea of physical chemistry," is, in the opinion of the reviewer, unwarranted. What the reader will get is a knowledge of a lot of subtle distinctions about energy, some practice in the handling of mathematical equations, and an incomplete acquaintance with some physico-chemical laws. But a "fair" idea of physical chemistry he certainly will C. E. LINEBarger.

not get.

BOOKS RECEIVED.

Spectrum Analysis. By John Landauer, LL.D. Authorized English edition by J. Bishop Tingle, Ph.D., F.C.S. New York: John Wiley & Sons. 1898. x+239 pp. Price $3.00.

The Sugar Beet in Illinois. Bulletin No. 49. University of Illinois, Agricultural Experiment Station, Urbana, Ill. 1898. 52 pp.

The Cost of Production of Corn and Oats in Illinois in 1896. Bulletin No. 50. University of Illinois, Agricultural Experiment Station, Urbana, Ill. 24 pp. An Ephemeris of Materia Medica, Pharmacy, Therapeutics and Collateral Information. Vol. V, No. 1. January, 1898. By Edward R. Squibb, Edward H. Squibb, and Charles F. Squibb. Brooklyn, N. Y. 1898. 176 pp. Preparation and Application of Fungicides. Bulletin No. 25. Connecticut Agricultural Experiment Station, New Haven, Conn. 1898. 16 pp. On the Crystalline Structure of Gold and Platinum Nuggets and Gold Nuggets. By A. Liversidge, LL.D., F.R.S., Professor of Chemistry, University of Sydney, N. S. Wales. 10 pp. 16 plates.

On Some New South Wales and Other Minerals. On the Amount of Gold and Silver in Sea-water. The Removal of Gold and Silver from Sea-water by Muntz Metal Sheathing. By A. Liversidge, M.A., F.R.S., Professor of Chemistry, University of Sydney, N. S. Wales. 42 pp.

Abbreviated Names for Certain Crystal Forms. Models to Show the Axes of Crystals. By A. Liveridge, M.A., F.R.S., Professor of Chemistry, University of Sydney, N. S. Wales. 10 pp.

PROTEIDS OF THE HORSE BEAN (Vicia faba).'

BY THOMAS B. OSBORNE AND GEORGE F. CAMPBELL.

Received March 29, 1898.

HE only references which we have found to investigations

THE of the proteids of this seed that at present have any

importance are contained in papers by Ritthausen."

In his earlier work' Ritthausen described as legumin the proteid extracted from the horse bean by potash water. Later, he gave the results of treating the earlier preparations with salt solution and also of extracting the seeds with brine and precipitating the dissolved proteid by dilution or by acids, and concluded that the preparations which he obtained were mixtures of two proteids which could only be separated by dissolving in acid or alkali, precipitating by neutralization and extracting the precipitate with brine; the part dissolving he regarded as conglutin; that undissolved as legumin. A careful study of Ritthausen's work leaves us in much doubt respecting the nature of the proteids of this seed, and we have therefore undertaken the present investigation.

In order to avoid contaminating our preparations with tannin, which Ritthausen found to be present in considerable quan1 Reprinted from advance sheets of the Report of the Connecticut Agricultural Experiment Station for 1897. Communicated by the authors.

2 Early references to legumin have been already noticed in our paper on "Legumin of the Pea and Vetch," Report of this Station for 1895, p. 262. This Journal, 18, 583. 8 Die Eiweisskorper, etc., Bonn, 1872, p. 170.

4 J. prakt. Chem., 26, 504, and 29, 448.

tity in the skin of these beans, we removed the greater part of the outer coating from the coarsely broken seeds by a current of air and the remainder by hand-picking. In this way it was possible to separate the brown outer seed-coat completely and by then grinding the broken beans to obtain a flour free from tannin.

A preliminary extraction made with this flour gave us a large yield of proteid which, in general, had nearly the properties and composition of the products similarly obtained from the pea and lentil, but differed throughout in containing less carbon than the corresponding substances from these other seeds, and in being, to a considerable extent, soluble in water. As these aqueous

solutions reacted strongly acid with litmus and when neutralized gave precipitates soluble in brine which were reprecipitated by dilution, we were led to believe that the differences which we had found between the proteids of this seed and those of the pea, lentil, and vetch were due to a combination of the proteids with the acid of the seed. As this view appeared to be confirmed, we omit further details of our first extraction and proceed to an account of the second.

Two kilograms of the bean flour were treated with ten liters of ten per cent. salt solution, protected with thymol, and left over night in a cool place. The extract was strained through fine bolting-cloth and allowed to deposit suspended matter during two hours. The turbid liquid was siphoned off, centrifugated, and filtered nearly clear. The extract was then saturated with ammonium sulphate, the precipitate produced filtered out, removed from the paper, suspended in a little water and dialyzed for twenty-four hours, whereby so much of the sulphate was removed that the proteid dissolved. The solution so obtained was then filtered perfectly clear through a pulp filter and dialyzed for forty-four hours. The precipitate which had formed at first separated in spheroids, but these on settling united to a coherent mass, A, from which the solution B was decanted.

A portion of the precipitate A, weighing twenty-eight grams when air-dry, was washed thoroughly with water (in which it partly dissolved) and then with alcohol and dried at 110° for analysis.

This required 0.36

The remainder of precipitate A was dissolved in brine and the resulting solution, having an acid reaction, was neutralized to litmus by adding very dilute potash water. gram of potassium hydroxide. The slightly turbid solution was filtered absolutely clear and dialyzed for forty-eight hours. The large precipitate which resulted was found to dissolve completely in brine and to yield a perfectly neutral solution, which, when heated in a boiling water-bath, gave some coagulum. A portion of this precipitate was washed with water and alcohol, dried at 110°, and analyzed.

The remainder of this substance was dissolved in 250 cc. of three per cent. brine and the solution diluted to 600 cc. A rapidly settling precipitate resulted which formed a fluid deposit from which the somewhat turbid mother-liquor was soon decanted; the proteid was washed thoroughly with water and alcohol, yielding when dry 19.13 grams of preparation 78.

1 Numbered consecutively with the preparations of the proteids of the lentil. This Journal, 20, 362.

The solution decanted from 78 was diluted with 150 cc. of water and a second precipitate obtained, wholly like the first, which when dried weighed eleven grams and had the following composition:

The solution from which 79 had separated was further diluted with 500 cc. of water and allowed to stand over night in a cool room. The next morning the solution was decanted from a small precipitate that had settled out. This, after drying, weighed five grams and had the following composition :

The solution decanted from 80 was dialyzed for four days but only a trace of proteid separated. Preparations 78, 79, and 80 contained no trace of coagulable proteid and may therefore be considered to be pure legumin. Since, however, 77, of which these were fractions, was shown by this treatment to contain only a very little coagulable matter, it too is essentially pure legumin as indicated by the analysis.

Solution B, described on page 394, was further dialyzed for twenty-four hours and the resulting precipitate C filtered out and the filtrate D treated as described on page 397.

The precipitate C was dissolved in ten per cent. brine, and the resulting solution diluted until it contained one per cent. of salt. The precipitate that formed was filtered out and the filtrate added to solution D. The precipitate was again dissolved in