The organic portion of this substance consists of dextros and dextrin, showing that glucose sirup has been added to cause the lead arsenate to adhere to the foliage.

This compound is the least soluble of all the arsenical insecticides tested, and consequently is not liable to scorch the foliage. It has been used as an insecticide with good results.

Large amount of barium salts also dissolved by cold water.

The two compounds which give this substance any value as an insecticide are the arsenious and arsenic acids. Since, however, all of the arsenic is present in a soluble condition, an application of this mixture to the foliage would burn it badly, consequently the use of this insecticide can not be recommended. It is understood that the results from this mixture have been so unsatisfactory that it has been withdrawn from the market. One point in connection with White Arsenoid which has not been mentioned in any of the State experiment station reports on the subject is the probable toxic effect of barium in a soluble condition on plants sprayed with it.

a The methods for determining PbO and As̟,05 give rather low results.
See correspondence with manufacturers, page 59.

This is an excellent full strength sample of White Arsenic, and could be used with good effect along with lime in preparing homemade calcium arsenite.

This mixture contains as its active poisonous constituents arsenious oxid and barium carbonate, and will be apt to serve the purposes for which it is intended. Great care should be taken not to leave any of this compound carelessly around a house, since it might be stirred up with the dust, breathed by persons, and finally cause arsenic poisoning; also, in the presence of organic matter, the same action might take place, as has been noted in the case of arsenical wall papers, resulting in the formation of a very poisonous gas, arsine.

The remainder appears to be Persian Insect Powder or pyre

thrum.

This mixture may be effective in killing and driving away roaches, ants, etc., but the advisability of scattering it around indiscriminately, as is usually done with roach powders, is very doubtful, since it contains arsenic, which might rise with the dust of the room, and

a See correspondence with manufacturers, page 62.
Corresponding to about 12 to 13 per cent London Purple.

being breathed, cause arsenic poisoning; also, as in the case of Rough on Rats, the arsenic might under certain conditions be converted to arsine.

This is another compound which may be effective in killing and driving away fleas, lice, croton bugs, etc., but as to the advisability of using it around human habitations the same may be said of it that has been said of Royal Roach Powder.

This sample was originally prepared by using approximately equal portions of CuSO,5H2O and unslaked lime. The reasons why such a compound can hardly be successful have already been stated (p. 27).

A considerable amount of carbon dioxid is also present. This solution was evidently made from strong ammonia, copper sulphate, and copper carbonate. The value of a mixture of this kind as a fungicide. is well known and requires no further discussion.

❝See correspondence with manufacturers, page 61.

This compound is an excellent sample of CuSO, 5H„O.

This substance contains sulphur and copper oxid probably in the form of copper sulphate, which is of value in treating surface mildew on the vine. The amount of tobacco is too small to be of any value. The gypsum adds weight, and serves as an agent to mitigate the action of the sulphur and copper sulphate.

SOAPS.

In making an analysis of this class of compounds, the following determinations were made: Moisture, total fatty matter, total alkali, free alkali, and in certain cases calcium oxid, neutral fat + unsaponifiable matter, and resin. The methods follow:

METHODS OF ANALYSIS AND DISCUSSION.

Moisture.-Weigh a 100 cc beaker containing about one-half inch of recently ignited sand and a small glass rod. Add about 5 grams of the sample, and weigh again. Dissolve the soap by means of 25 or

@See correspondence with manufacturers, page 60.

Most of the methods are from "Oils, Fats, and Waxes," by Benedikt and Lewkowitsch, with slight modifications.

more cc of strong alcohol on the water bath. If the soap is very hard it may be as well to leave it in contact with the alcohol overnight, so as to soften it. Evaporate off the alcohol and dry in the oven at 110° C., either to constant weight or to the point where it only gains or loses a little during one-half hour extra drying.

Total fatty matter.-Five to 10 grams of the soap is dissolved in hot water in a beaker by means of gentle heat and constant stirring. When the liquid finally begins to boil, a few drops of methyl orange are added and dilute hydrochloric acid until the indicator shows an acid. reaction. Continue the boiling and stirring until all fatty matter has collected in drops. Add about 5 grams of dry beeswax, accurately weighed on a weighed watch glass. Boil again till the fat collects on the top as a clear layer free of specks. Rinse off the rod and boil till the fat has again collected. Allow the contents of the beaker to cool until the fatty layer has entirely solidified. Remove this layer by means of a platinum spatula, wash with cold water, dry between filter papers, and transfer to the weighed watch glass previously spoken of. Gather together all particles of fat left behind by means of the platinum spatula, and when this is not possible by means of small weighed pieces of filter paper, add to the watch glass and contents, dry in a desiccator overnight, and weigh. The total weight thus obtained, minus the combined weights of the watch glass, beeswax, and pieces of filter paper, gives the weight of the total fatty matter. In the absence of resin, neutral fat, and unsaponifiable matter, all of this fatty matter is reported as fatty acids, but if any of the other three are present they are subtracted and the remainder reported as fatty acids, which multiplied by 0.965 gives fatty acid anhydrid.

Total alkali.-A weighed quantity of the soap is decomposed by hydrochloric acid and the water is filtered off from the fat, which is washed. Both potassium and sodium in the filtrate are first determined as the mixed chlorids in the ordinary manner, and the potassium then determined by means of platinum chlorid solution.

Free caustic alkali. -Dissolve 30 grams of the soap in strong or absolute alcohol. Filter rapidly by means of a hot-water funnel into a narrow-necked flask. Wash the filter with alcohol till clean. The filtrate is now titrated with N 10 hydrochloric acids, using phenol phthalein as indicator. In the case of all whale-oil soaps examined, an acid reaction was obtained at this point denoting free fatty acids.

Calcium oxid.-This was found to be present in one sample of soap and was determined by first acidifying a weighed quantity of soap in water solution with hydrochloric acid, just as in the determination of 1301-No. 68-02-3