If it

with cold water, then dried at gentle heat and weighed. be desired to estimate the alkaloids of this compound, one gram "kolanin," one gram calcium hydroxide, and three grams chalk with a little seventy per cent. alcohol are mixed together and evaporated on the water-bath to about six grams, and extracted with alcoholic chloroform in the manner already described.

The objections to this method are in part among those already mentioned in the discussion of Jean's method. The use of lime or other alkalies in the assay of a caffein-bearing drug is to be deprecated. The solvent used is not a proper one for the reason that sufficient alcohol is present to extract other constituents in addition to the alkaloids, which are not removed from the caffein during the subsequent treatment of the residue, and which, when weighed with the caffein, lead to erroneous results. Moreover, the manner of applying the menstruum is inconvenient, does not insure complete extraction, and is in no way preferable to the ordinary extraction by the use of Soxhlet's apparatus. The addition of the sulphuric acid is unnecessary and does not add to the purity of the final product, which is dark-colored and very plainly impure. The objection to weighing a final residue as caffein, finds especial application in this method.

As the properties of caffein kolatannate had not been made known very generally at the time of publication of these methods, there is some excuse for the assumption of both these writers that it is wholly insoluble. Carles has proceeded on this hypothesis in directing the drug to be extracted with cold water to remove the water-soluble constituents before exhausting it with alcohol to remove the caffein compound, but inasmuch as caffein kolatannate is not only somewhat soluble in water, but considerably more soluble in solutions of caffein and of tannin, the extraction of kola by water will remove a considerable amount of it. The same is to be said of the final washing of the caffein compound with water, which is quite inadmissible in quantitative work. Carles seems to have recognized the uncertain value of gravimetric determinations of caffein kolatannate, and is to be commended for offering an alternate method providing for its valuation according to the amount of its alkaloids.

At our request, Mr. James W. Cobb, Ph.C., has assayed a

sample of dried kola by each of the foregoing methods and by the method adopted by us last year,' after thoroughly familiarizing himself with them by preliminary work.

It will be seen that the impurity of caffein separated by Jean's method amounts to 4.84 per cent., and that of the caffein by Carles' method amounts to 14.25 per cent., which was determined by titrating the caffein with Wagner's reagent, after Gomberg's method.

Both give very low results as compared with those obtained by our own. They are tedious and otherwise unsatisfactory, in addition.

Further notes on the assay of kola are in preparation.

All the work reported upon in this paper has been done under the provision of the Stearns Fellowship of the University of Michigan.

ANN ARBOR, MICH., AUGUST, 1897.

NOTE.

A Correction. In the September number of this Journal the author, together with Mr. F. Thompson, published an article entitled "A Preliminary Thermochemical Study of Iron and Steel,'' in which they give the results obtained on a number of samples of iron and steel of different composition and which had been subjected to varying heat treatments. On taking up the work again this fall and reviewing critically the results of last year's work, I find that there are at least two serious errors in the results that I wish to explain now in order to prevent anyone else from being misled by the conclusions drawn from the results as stated in the article. The statement that the loss by radia1 Knox and Prescott, 1896: This Journal, 19, 73.

tion was only 0.24° in one hour was a mistake of observation. Careful determinations on this point, made recently, show that it must have been 1.24° instead of the former figure. The error from not taking into account the loss by radiation will amount to from about fifteen to forty calories per gram of metal, the loss depending on the time of solution. Another source of uncertainty, and one for which the author alone is responsible, is the fact that the solution used for dissolving the samples contained about four per cent. of free hydrochloric acid, and consequently the solution of the iron may not in all cases have proceeded according to the equation, Fe + CuCl, = FeCl, + Cu, and this copper redissolved according to the equation Cu + CuCl, = Cu ̧Ci,. Part of the iron may have dissolved directly in the free hydrochloric acid, and, since the heat absorbed in the decomposition of 2HCl is less than that necessary for the reduction of 2CuCl, to Cu,C,, more heat would be rendered sensible when a good deal of iron was dissolved directly in hydrochloric acid than when solution was effected through cuprous chloride.

The work is being gone over again very carefully with view to eliminating any errors that may exist in the previous work, and I therefore request that judgment be suspended on the previous work until further results may enable us to form a more reliable conclusion than could be drawn from our former data. E. D. CAMPBELL.

ANN ARBOR, MICH., DECEMBER 14, 1897.

ERRATUM.

On page 942, line 5 (Vol. 19), for "antimony" read "phosphorus."

BOOKS RECEIVED.

A Practical Treatise on Mineral Oils and their By-Products, including a Short History of the Scotch Shale Oil Industry, the Geological and Geographical Distribution of Scotch Shales, Recovery of Acid and Soda used in Oil Refining and a List of Patents Relating to Apparatus and Pro cesses for Obtaining and Refining Mineral Oils. By İltyd I. Redwood. London: E. & F. N. Spon, Limited. New York: Spon & Chamberlain. 1897. xiv +336 pp. Price $6.00.

The Early History of Chlorine. Alembic Club Reprints, No. 13. Papers by Carl Wilhelm Scheele (1774), C. L. Berthollet (1785), Guyton de

Morveau (1787), and J. L. Gay-Lussac and L. J. Thenard (1809). Edinburgh: William F. Clay, 1897. 48 pp. Cloth. Price Is. 6d.

Researches on the Molecular Asymmetry of Natural Organic Products. Alembic Club Reprints, No. 14. By Louis Pasteur, membre de la Société Chimique de Paris (1860). Edinburgh: William F. Clay. 1897. 46 pp. Cloth. Price 1s. 6d.

Ornithology of North Carolina.

A List of the Birds of North Carolina, with Notes of Each Species. Issued by The North Carolina Agricultural Experiment Station, Raleigh, N. C. October 30, 1897. 36 pp.

Current Thought. New Series. Vol. 1, No. 1. January, 1898. C. Elton Blanchard, 802 Ansel Ave., Cleveland, Ohio. 21 pp.

The Mechanics of Soil Moisture. By Lyman J. Briggs, Physicist of the Division of Soils. Bulletin No. 10. U. S. Department of Agriculture, Division of Soils, Washington, D. C. 24 pp. Second Report on Food Products. Twenty-first Annual Report of the Connecticut Agricultural Experiment Station for 1897. Part I. New Haven: The Tuttle, Morehouse and Taylor Press. 1897. xvi+64 pp.

Montana Swine Feeding. Bulletin No. 14. Montana Agricultural Experiment Station, Bozeman, Montana. April, 1897. 36 pp. 4 half-tone plates.

THE JOURNAL

OF THE

AMERICAN CHEMICAL SOCIETY.

I

THE DIGNITY OF ANALYTICAL WORK.'

By C. B. Dudley.

T will doubtless be conceded by all, that in the choice of the field to which one proposes to devote his life-work, a number of things should be consulted. Among these may be mentioned not only mental capacity and the opportunities for training by courses of study, which may be available to him, but also what may be termed natural inclination or love for the work. Just how much weight should be given to each of these elements, is a query not easily answered, but few will deny that genuine interest in or real love for the field of work chosen, should be allowed as great sway as possible. Those of us who have gotten far enough along in our life-work, to be able to look back somewhat, and to see and differentiate the causes that have shaped our line of effort, know full well that circumstances beyond our control, rather than our inclinations and desires, have in many cases determined our course, but the fact nevertheless. remains, that for the best results, for the attainment of even moderate success, one's efforts must be in a field agreeable to him, and his heart must be in his work. Fortunate is the man for whom circumstances so shape themselves, that he is able to pass his years in the field of his choice, and spend and be spent in work that is congenial to him.

1 Presidential Address delivered at the Washington Meeting of the American Chemical Society, December 29, 1897.