This extraction with chloroform is repeated a number of times, thus extracting most of the alkaloids of hellebore, either partially or wholly. This solution, which we will call A, is put aside, and the contents of the separatory funnel is made alkaline with ammonia and extracted with petroleum ether to obtain any alkaloids that are left, particularly veratrine, making solution B.

Portions of solution A are placed in watch glasses, allowed to evaporate spontaneously, and finally tested with concentrated sulphuric acid. On employing two samples known to be pure, the following results are obtained: The residue left is amorphous. On treating it with a drop of sulphuric acid a yellow solution is first formed, changing to brown, then purplish red on the edges, and finally, after standing about half an hour, purplish red throughout the whole mass. Portions of B are now treated in the same way, and on two pure samples the following results are obtained: The residue is amorphous. On being treated with concentrated sulphuric acid a yellow solution first forms, which turns dark yellow, chestnut, and finally brown red, with sometimes a greenish color around the edges of the drop."

Of course, there are some substances that might give the same play of colors with sulphuric acid as those described, but probably not any that would likely be used to cheapen the product.

ANALYSES AND DISCUSSION.

The results obtained by the above methods of analysis are given in the following table:

a According to Allen's Commercial Organic Analysis, the pure alkaloids should give the following play of colors with concentrated sulphuric acid:

Jervine: Yellow, brown yellow, bright green.

Pseudojervine: Yellow, bright green.

Rubijervine: Yellow, orange, dark red.

Veratralbine: Yellow, orange red, blood red with green fluorescence.

Veratrine: Exactly like cevadine except that the red solutiou is not fluorescent.

Cevadine: Yellow, brown red, blood red with greenish fluorescence; if allowed to stand in air long enough, becomes purple.

A glance at the moisture figures in Table V shows that the variation in the 9 samples is not very great, with a minimum of 6.23 per cent and a maximum of 8.87 per cent.

As to the percentage of ash, it must be remembered that these preparations are the powdered root of a plant. We would therefore expect to find a rather high ash content, which would be much increased if the roots had not been properly cleansed. The figures show that there is at any rate no reason that the ash content should be above 15 per cent, and evidently, with proper handling, this can be very much reduced. Adopting, then, 15 per cent as the maximum amount allowable, it will be seen that samples 19836 and 19837 are very much above this limit, and 19558 slightly above.

Since the alkaloidal principles in the above 9 samples, on being treated with concentrated sulphuric acid, act exactly similar to the alkaloidal principles of 2 samples known to be pure, and also follow fairly closely the color changes as indicated for the alkaloids of the hellebores in chemical literature, these samples appear to belong to the class of hellebores. The color changes seem to follow most closely those indicated for cevadine, which might be expected if the above samples are roots of Veratrum viride, since this alkaloid appears in quantities greater than the sum of all the other alkaloids.

PYRETHRUMS.

The next group of substances to be taken up are the pyrethrums, which include Pyrethrum roseum, or Persian Insect Powder, and Pyrethrum cinerariæ folium, or Dalmatian Insect Powder, and Buhach.

The ground flower heads of these plants are used to kill insects. Such preparations contain as their active principle a volatile oil, which quickly oxidizes on exposure to the air to an inactive resin." This oil may be determined by extracting with ether and drying at a low heat. Four samples examined by Hilgard gave the following results:

Such powders sometimes contain either lead chromate, barium chromate, or tumeric, to give them a bright yellow color or to palm off some entirely different powder as one of the pyrethrums.

METHODS OF ANALYSIS AND DISCUSSION.

In making an analysis of these products the following determinations were made: Moisture, ash (a large amount of which indicates the

a Hilgard. Report of the College of Agri. Univ. of California, 1879, p. 68.

presence of barium or lead chromate), ether extract, and qualitative tests for lead, barium, chromium, and tumeric.

Moisture. This is determined by drying a weighed quantity of the powder for from 12 to 14 hours in hydrogen at the temperature of boiling water, and reporting the loss as moisture.

Ash.-Burn the dried, weighed portion from the moisture determination at a low red heat until a white ash is obtained. The residue. is reported as ash.

Ether extract.-Extract a weighed portion of about 2 grams of the powder for 16 hours with ether in any of the ordinary extraction apparatus. Drive off the ether on the steam bath and dry at a temperature of about 70° C., not leaving the extraction flask in the oven longer than necessary. The residue is reported as ether extract.

Lead, barium, and chromium are tested for in a nitric acid solution of the ash by the ordinary methods employed, and chromium by the borax bead test, using a small quantity of the ash.

Turmeric.-Treat a portion of the powder with alcohol and filter off the extract. Saturate a piece of filter paper with the extract and dry on a watch glass over the steam bath. Repeat three or four times. Now dip the paper in a solution of borax slightly acidified with hydrochloric acid. Dry on a watch glass over the steam bath, and if tumeric is present a cherry-red color appears.

ANALYSES AND DISCUSSION.

In Table VI, following, are the analyses of 10 samples of pyrethrums or imitations thereof.

19528. Black, tarry oil; strong smell of pyrethrum.
19529. Brown oil; strong smell of pyrethrum.

19530. Greenish-brown oil; not very strong smell of pyrethrum.
19559. Greenish-brown oil; not very strong smell of pyrethrum.
19677. Black, tarry oil; strong smell of pyrethrum.

20371. Green oil; some smell of pyrethrum.

20483. Dark green, tarry oil; strong smell of pyrethrum.

20485. Light green oil; strong smell of pyrethrum.

20486. Dark green, tarry oil; strong smell of pyrethrum.
20487, Dark green, tarry oil; some smell of pyrethrum.
b See correspondence with manufacturers, page 60.

Table VI shows that the percentages of both moisture and ash are about normal and fairly close together in samples 19528, 19529, 19559, 19677, 20371, 20483, 20485, 20486, and 20487, with the exception of 20371, in which the ash is rather high. This high ash content might easily occur, however, in a sample that had not been tampered with in any way. The ether extract figures for the above numbered samples are much higher than those found by Hilgard for Buhach, and the residue has a strong characteristic smell, indicating that the samples are high in that active principle which kills insects. On the whole, all of the above-mentioned samples appear to be the true powdered pyrethrum flower.

Sample 19530 is a bright yellow powder, which has a higher percentage of ash and a lower percentage of ether extract than the average. The oil obtained by extracting with ether has not a very strong pyrethrum odor, and the presence of lead and chromium in the ash of this sample shows that lead chromate has been added to give the powder a brighter yellow color. This sample appears to be either only partially composed of the flower of pyrethrum, or composed of flowers of an inferior quality or of flowers mixed with stems. The use of lead chromate in powders of this class, which are to be blown out in the air of a room, does not appear to us as commendable, since they are, to a certain extent, breathed by human beings, and the lead chromate contained might give rise to troubles more or less serious.

MIXTURES CONTAINING BORAX.

Practically all of these mixtures are used for the destruction of cockroaches and water bugs, and some are further advertised to kill fleas, moths, ants, lice, etc. The substances contained in these compounds are so many in number and the methods of analysis so diverse that only the methods that are used for the determination of borax will be given. Where there is no interfering substance the following method a is used:

METHODS OF ANALYSIS AND DISCUSSION.

Weigh 2 grams of the sample out into a 200 cc flask and shake well with water till all borax is dissolved. Make up to the mark, filter through a dry filter, throwing away the first 5 cc, and take aliquot portions of 50 cc for analysis. Add methyl orange and then dilute sulphuric acid drop by drop to the appearance of a pink color. Boil to get rid of carbon dioxide, cool, add a little more methyl orange and titrate back with fifth normal sodium hydroxid until the pink color just changes to yellow. All boric acid is now in the free state. Add

a Thomson. Sutton's Volumetric Analysis.

enough glycerin to amount to at least 30 per cent after the titration is finished; then add phenol phthalein and titrate with N/5 sodium hydroxid to the appearance of a permanent pink color. From the number of cc of soda used in the last titration the borax may be calculated. The writer has found during the course of this work that the best way to standardize the sodium-hydroxid solution is against a weighed. quantity of chemically pure Na,B,O,10H,O, the titrations to be carried

through in exactly the same manner as described above. Sometimes the borax can not be dissolved directly from the sample and titrated because of interfering substances. In such a case as this the method following, as used in the food laboratory, Bureau of Chemistry, is applied.