action in this direction rather than in the direction of chemical affinity or chemical action; the latter notion is so vague.

Granting this independence of the constituents, the notion of a diaphragm permeable to one constituent but not to the other, becomes clear enough.

In the arrangement considered above, the diaphragm is permeable to the constituent present in larger quantities and therefore generally called the solvent. In this case ether is the solvent. As the diaphragm is impermeable to the solute, should we not claim that it receives the full pressure of the solute, not allowing any activity on the part of the solute to pass through the diaphragm? How can we claim anything else? But then when we do this, how can we claim that the hydrostatic pressure preventing the entrance of the ether, comes from anything else than the column of pure ether in the solution, which column of pure ether has the height of the column of solution? We cannot, consistently, with what we have assumed regarding the independence of the constituents of the liquid mixture. For we are not to look upon a molecule of ether locked to a molecule of solute, so that whither the ether molecule goes, the other molecule must go. Not at all, not even for an instant. And yet, how can we mean otherwise when we say that s in equation (1) shall refer to the solution as a whole and not to the solvent in it.

Now in the paper of Noyes and Abbot' from which paper the above calculations have been taken, there is a collection of data which will be useful in testing this view.

In that paper they describe an arrangement which we must look upon as purely imaginary; the arrangement cannot be looked upon as possible in fact, though perhaps interesting from a theoretical point of view. Instead of putting the diaphragm at the bottom of the tube, they put it at the top of it and assume that the pure solvent rises in the tube until its hydrostatic pressure balances the osmotic pressure of the solute. But in their development of the necessary formulas, they do not take into account the external air pressure needed to force up the column of liquid supposing the arrangement to be subject to the air-pressure, or if in a partial vacuum such as we have assumed in our arrangement, they do not show how such a column of

1 Loc. cit.

liquid could possibly rise in the tube, for the difference in pressure, p1, is insignificant compared with 7. In either case, so far as their results depend upon their theoretical deductions, they are valueless, but considered by themselves, the data are very valuable, and will serve us a good turn.

Substituting in (2) the value for s for pure ether, given by Noyes and Abbott, equal to 0.7206, and passing to Briggs' logarithms, we get

On the other hand, knowing the concentration, we can calculate by Avogadro's law. The two values should agree. In the following table, largely from Noyes and Abbot, c is the quantity of solute, in one series naphthalene, in the other azobenzene, in one part of ether at 12.9°, p is the vaporpressure of the solution in cm. of mercury, s is its density, π, is the osmotic pressure calculated from (3), 7, is the osmotic pressure calculated from (3) but substituting the density of the solution for the density of the pure solvent, 7, is the osmotic pressure calculated according to Avogadro's law, 47, is the percentage variation of 7,, compared with 7, as the standard, ▲π, is the percentage variation of 7,, compared with π, as the standard.

0

2

0

Judging from these figures, there can hardly be any question that the density of the solvent and not that of the solution is to be used in computing 7 by the hydrostatic method.

RUTGERS COLLEGE, May, 1898.

METHOD OF PREPARING A STRICTLY NEUTRAL AMMONIUM CITRATE SOLUTION.

FOR

BY A. D. COOK.

Received June 8, 1898.

'OR the benefit of many analytical chemists who are engaged in fertilizer work and for the purpose of securing uniformity in results, I respectfully submit the following pertaining to the neutrality and preparation of the chemical reagent "ammonium citrate."

This reagent has caused more trouble than all the other reagents required in fertilizer work, and yet it is an extremely easy matter to get a strictly neutral reaction.

The method adopted by the Association of Official Agricultural Chemists does not state the most essential fact in the preparation of this reagent, and the one which, if universally adopted, would overcome many obstacles in its preparation.

The failure to obtain a strictly neutral solution of ammonium citrate has caused great discrepancy in results among chemists who have analyzed the same material. I refer more particularly to the analysis of concentrated phosphates where there is a large per cent. of available phosphoric acid. The total phosphoric acid running as high as fifty per cent., the insoluble eleven per cent., making the available thirty-nine per cent. In the chemical laboratory at this station, where from 800 to 1000 samples of commercial fertilizers are analyzed annually, I have had ample opportunity to try different methods and to compare results obtained by chemists working with different solutions of ammonium citrate. It has been pointed out by fellow-workers in this field that a strictly neutral solution may be obtained by allowing the solution to stand after ammonia has been added to the citric acid and the proper dilution made. I have found that the practicability of this procedure depends upon the temperature of the solution. If vigorous stirring is neglected the solution will be slightly alkaline. Vigorous stirring, thus causing

heat by chemical action, will generate sufficient heat to drive off the excess of ammonia, and this is the main point to observe in securing its neutrality.

This reagent is made up as follows in this laboratory: 740 grams of commercial citric acid are carefully weighed out and placed in a four-liter graduate containing 1900 cc. of ten per cent. ammonium hydroxide. With a suitable glass rod the citric acid is thoroughly and vigorously stirred until the citric acid has all dissolved. Distilled water is now added until the meniscus reads 4000 cc. The solution is again stirred and carefully transferred to a large porcelain evaporating dish. The solution is allowed to stand over night and in the morning large oval crystals are noticeable on the sides of the four-liter graduate, and invariably the solution when tested for neutrality will be found strictly neutral. If the solution is not vigorously stirred sufficient heat will not be evolved to drive off the excess of ammonia, and when tested will be found to be slightly alkaline, but by resorting to vigorous stirring, a strictly neutral reaction will be obtained. The solution, after being transferred to the reagent bottle, is brought to the required temperature, 20° C., and distilled water added until the specific gravity is 1.09. On testing the neutrality of this solution, both with coralline and cochineal as indicators, it will be found unnecessary to alter its neutrality in the least degree, the solution being strictly neutral.

THE

THE ASSAY OF TELLURIDE ORES.

BY CHARLES H. FULTON.

Received June 8, 1898.

HE growing importance of telluride ores and the fact of their relative difficulty of assay in contrast with other ores, has led to this work. The object of the work was to determine where the difficulty and losses lay, and if possible to remedy these defects of the assay by proper methods and precautions. Ore No. 1 is a telluride ore, having a gangue, mainly of quartz. A very small amount of pyrite is also present.

Ore No. 2 is a very rich telluride from Cripple Creek, Colorado, containing considerable pyrite. The gangue is siliceous.