TABLE I.-Name of sample, with name of retailer, manufacturer, price, and form in which received--Continued.

20499

20.500

METHODS OF ANALYSIS AND ANALYSES OF SAMPLES.

PARIS GREENS.

In analyzing the samples of Paris Green the following determinations were made: Moisture, sand, sodium sulphate, total arsenious oxid, copper oxid, soluble arsenious oxid, and soluble copper oxid. The methods follow.

METHODS OF ANALYSIS AND DISCUSSION.

Moisture.-Dry 1 to 2 grams for eight to ten hours at 105° to 110° C., and calculate loss as moisture.

Sand. Dissolve the sample used for moisture determination in hydrochloric acid, filter, wash, dry, and finally burn filter, calculating the residue as sand.

Sodium sulphate.-Treat the boiling filtrate from the determination of sand with a boiling solution of barium chlorid, allow to stand until the precipitate settles, leaving a clear solution; filter, wash, dry, and burn, with the usual precautions used in determining barium sulphate; calculate the barium sulphate found to sodium sulphate, since it is in this form that sulphuric acid is supposed to be present.

Total arsenious oxid.-For this determination a standard iodin solution is needed, which is prepared in the following manner: “ Dissolve 12.7 grams of powdered iodin in about 250 cc of water to which has been added 18 to 25 grams of chemically pure potassium iodid, and make the whole up to a volume of 1 liter. To standardize this solution, weigh out 1 gram of dry, chemically pure arsenious oxid, transfer to a 250 cc flask by means of about 100 cc of a solution containing 2 grams of sodium hydrate in each 100 cc, and boil until all arsenious oxid goes in solution; cool, make to a volume of 250 cc, and use 50 cc for analysis.

This 50 cc portion is concentrated by boiling in a 250 ce flask to half its volume and allowed to cool to 80° C. An equal volume of concentrated hydrochloric acid is now added, accompanied by 3 grams of potassium iodid, mixed, and the whole allowed to stand for ten minutes (to reduce the arsenic oxid formed on boiling an alkaline arsenite to arsenious oxid). The brown solution is then diluted with cold water, and an approximately N/10 solution of sodium thiosulphate added, drop by drop, until the solution becomes exactly colorless. (This end-point is easy to read without the aid of starch.) This solution is then made slightly alkaline with dry sodium carbonate (using a drop of methyl orange to read the change), then made slightly

a See article by Haywood, Jour. of the Amer. Chem. Soc., Vol. XXII, No. 9, September, 1900.

acid with hydrochloric acid, taking care that all lumps of sodium carbonate on the bottom are acted on by the hydrochloric acid. Sodium bicarbonate is now added in excess and the solution of iodin run in, drop by drop, using starch to read the end reaction. (Sometimes the solution gets dark toward the end of the titration. This must not be confused with the final dark-blue color given by the iodin and starch.) From the number of cc of iodin solution used and the weight of arsenious oxid taken, the value of each cc of iodin in arsenious oxid can be determined.

Two grams of Paris Green are weighed out and transferred to a 250 cc flask and about 100 cc of water and 2 grams of sodium hydrate added. This solution is boiled for 5 to 10 minutes, or until all of the green particles have changed to red cuprous oxid. It is then cooled to room temperature and the volume made to 250 cc. The well-shaken liquid is filtered through a dry filter and 50 cc taken for analysis. The analysis is carried out from this point forward the same as when we standardize the iodin solution.

Copper oxid. The cuprous oxid obtained in the above method by boiling the Paris Green with sodium hydroxid is poured on the filter (after taking an aliquot portion of the solution for the determination of arsenious oxid) and well washed with hot water. It is then dissolved with hot dilute nitric acid and made to a volume of 250 cc, one-fifth being taken for analysis. The copper in solution is determined either by means of the galvanic current or, when that is not handy, in the following manner: The nitric-acid solution is made alkaline with sodium carbonate, then made slightly acid with acetic acid, and about ten times the weight of the copper in potassium iodid added. When it is all dissolved, the free iodin is titrated with a standard solution of sodium thiosulphate, using starch as indicator."

Soluble arsenious oxid.-In making the determination of this constituent two methods of analysis were used, namely, the water extraction method and the sodium acetate extraction method."

The first of these methods is carried out as follows: Treat 1 gram of Paris Green with 1 liter of water (previously boiled to get rid of CO, and again cooled to room temperature) in a large flask. The flask is stoppered and shaken five times each day for 10 days. At the end of this time the solution is filtered off through a dry filter. Two hundred cc of this is treated with sodium bicarbonate and titrated with standard iodin. This gives the amount of arsenious oxid dissolved. Because of the fact that water breaks up Paris Green on standing in contact with it, some copper oxid always goes into solution at the same time as the arsenious oxid does. This was also determined

a For the details of this method, see Sutton's Volumetric Analysis, 3d ed., p. 133. See report of the associate referee in the Proceedings of the Association of Official Agricultural Chemists for 1901.

by the following method in the hope that at some future time the relation between the copper oxid dissolved and the Paris Green broken up might be determined: Another 200 cc of the above solution is treated with 5 cc of hydrochloric acid, heated to 70° C., and a current of hydrogen sulphid passed through until all arsenic and copper are precipitated. It is then filtered and washed quickly with hot water. The filter and its contents are burned in a porcelain crucible with powdered sulphur and weighed. In this way a mixture of CuO and Cu„S is obtained. To calculate the amount of CuO dissolved, consider the residue as entirely made up of CuO. This is correct, since the percentage of copper in CuO and Cu,S is the same.

The second or sodium-acetate-extraction method for determining the free arsenious oxid in Paris Green is made as follows: Digest over the flame 1 gram of Paris Green for about 5 minutes with 25 cc. of a solution of sodium acetate containing 12.5 grams of the crystallized salt. The solution is then cooled, made up to 100 cc. and 50 cc. filtered off and titrated with standard iodin in the usual way."

It will not be out of place at this point to discuss briefly these two methods of analysis. It appears from the work that has been done on the subject that the sodium-acetate method gives more closely the true percentage of free arsenious oxid in the sample of Paris Green, while the water-extraction method gives the percentage of free arsenious oxid in the green plus some arsenious oxid obtained by the decomposing action of the water on the Paris Green. In some very coarse samples of Paris Green, which are supposed not to be made as well as the finer samples, the author found recently that the sodium-acetateextraction method gave very low figures for free arsenious oxid (showing that only a small amount of free arsenious oxid as such was present), while the water-extraction method gave very high figures (seeming to show that although only a small amount of arsenious oxid was present in the free state, there was present a certain portion of the green, which was in a very loose combination, and consequently was easily broken up). That portion of the green which was in such a loose combination would, in all likelihood, when applied to the plant soon break up and scorch the foliage. It therefore appears that while the sodiumacetate-extraction method, as before mentioned, gives more closely the actual percentage of free arsenious oxid present in the green, the waterextraction method gives some idea of its stability, and consequently will express more closely the value of the compound in actual orchard practice. It has recently been called to the author's attention by Mr. C. L. Marlatt, Assistant Chief of the Division of Entomology, that samples of Paris Green which do not scorch may be made to do so by grinding them up to a very fine powder and then applying them to

a See paper of Avery and Beans, Jour. of the Amer. Chem. Society, Vol. XXIII, No. 2.

the tree. This might be expected from the work done by Avery and Beans, which shows that for any particular green the finer the green the more arsenious oxid goes into solution in water in a given length of time. It is, therefore, apparent that the water-extraction method for determining free arsenious oxid does not distinguish between the arsenious oxid set free from a badly made sample of Paris Green and the arsenious oxid set free from a very fine sample. Too much arsenious oxid from either of these causes, however, is objectionable, although, perhaps, not equally so. The finely made greens stay in suspension in water much better than the coarser greens, and are, therefore, better adapted for spraying purposes, although, of course, there is a safety limit, above which the arsenious oxid extracted by water should not run for these as well as for the coarser greens. What constitutes the best Paris Green, then, for spraying purposes is a very fine sample which will stay in suspension in water but will not break up to a great extent upon standing in contact with it. Such greens as are mentioned can be made, since a number of those above analyzed were very fine, yet gave a very low figure for soluble arsenious oxid by the water-extraction method.

The amount of free arsenious oxid allowable in Paris greens varies somewhat in different climates, but in the East, generally, and in California 4 per cent is considered the maximum amount allowable. In Idaho 6 per cent has been adopted as the maximum amount. In all cases the 7-day or 10-day water-extraction method is the one to be employed in making determinations of this constituent. In this connection it may be stated that 4 per cent was adopted in the East as the maximum amount of free arsenious oxid allowable in Paris Green when the extraction with water was limited to 24 to 48 hours, but recent work has shown that all of the free arsenious oxid does not go into solution in this time, and that very likely where the old method gave from 3 to 4 per cent the new method would give at least 5 to 6 per cent. It is the opinion of the writer, then, that more work should be done upon this subject, using more modern methods of analysis, and that until such work is done the 6 per cent adopted by the Idaho station gives more closely the safety limit of the green (when a 10-day extraction is used for analysis and when lime is mixed with the green before using) than does the 4 per cent used in California and the East. The climatic conditions in different parts of the United States would most likely change this figure somewhat for different localities.

Again, while considering this point it seems, at any rate from a theoretical standpoint, that in preventing the scorching of foliage by adding lime to Paris Green it would be much better to mix the green, in suspension in water, with the lime about 10 days before use and stir the mixture occasionally, since the free arsenious oxid does not go

a See Bul. No. 25, Idaho Agricultural Experiment Station.