chlorophyl, which has been even in a feeble degree subjected to acids, can be considered as a constituent of chlorophyl.

Comptes Rendus, lxi, 436. M. A. Trécul observes what he thinks to be naturally crystallized chlorophyl, in cells in the bark of Lactuca Altissima.

Liebig & Will's Jahresbericht, 1863, p. 561. Stein remarks that the red spots in the flowers of Aesculus Hippocastanum, also the flowers of Aesculus Pavia, which are first yellow then red, are turned green by alcoholic solution of soda.' In other red flowers the red is turned green by alkalies, but is turned blue by the acetates of alumina, magnesia, or protoxyd of manganese. This red coloring matter appears to be the same as that of such blue flowers as Hyacinthus botryoides and Centaurea cyanus, which is reddened by alcoholic hydrochloric acid, and then again turned green by alkalies, and blue by the above named acetates. Stein supposes the blue coloring matter to be formed by combination of the red coloring matter with a base, namely lime, since the ashes of cockle contain much lime. (Centaurea cyanus, immersed directly in ammonia, becomes green, apparently by optical combination of yellow produced by the alkali with the blue remaining unchanged.-J. W.)

Jahresbericht, 1865, p. 628. F. V. Jodin finds that green leaves killed by alcohol, or by heating in closed vessels to 100° C., are rapidly bleached when exposed to light and air, but retain their color if kept in the dark.

Journal für praktische Chemie, xcv, 219. Vohl took, in 1856 horse chestnut leaves, which had been killed by a strong night frost, and laid them away in a close stoneware jar with water, so as to allow them to rot in the dark. In 1865, nine years later, he took out those leaves, and by treatment, first with ether, then with alcoholic ether, extracted vegetable wax and chlorophyl, thus proving chlorophyl to be a tolerably permanent body, when not exposed to light or air.

ART. XXII.-On a Modified Form of the Nitrate of Silver Test for Arsenic Acid; by CHARLES E. AVERY, Student in the Massachusetts Institute of Technology.

SINCE arsenate of silver is slightly soluble in an aqueous solution of nitrate of ammonium, and readily soluble both in ammonia and dilute nitric acid, it is not easy to detect small quantities of arsenic by means of nitrate of silver, as usually employed, unless the test be applied with extreme care.

It

is evident, however, that if the liquid to be tested for arsenic acid, as for instance the nitric acid solution of a spot or mirror of metallic arsenic,-could be charged with a salt, or acidulated with an acid incapable of dissolving arsenate of silver, it would be possible to test for arsenic without special precautions. At the suggestion of Prof. F. H. Storer, I have tested this idea by experiment.

All acids which I have tested in this regard exert some solvent power; but those having very little solvent action may be used almost as though they had none. I find, in fact, by experiment, that the addition either of acetate of sodium, acetate of ammonium, or Rochelle salt, to a mixed solution of arsenic and nitric acids, is sufficient to ensure the immediate precipitation of arsenate of silver, when ammonio-nitrate of silver is introduced.

By placing a small quantity of a nitric acid solution of arsenic acid upon a watch glass, stirring into it a few drops of a strong solution of either of the alkaline acetates or of Rochelle salt, and then adding a drop or two of ammonio-nitrate of silver, the characteristic brown-red precipitate of arsenate of silver is at once thrown down, even when the solution under examination contains comparatively little arsenic.

The acetates are to be preferred to the double tartrate; for, unless there be nitric acid enough in the liquor tested to set free all the tartaric acid, white bitartrate of potassium separates on agitation, and obscures the reaction.

Instead of the acetates or tartrate, recently precipitated carbonate of silver may be employed to neutralize free nitric acid in testing for arsenic acid. If the nitric acid solution be poured upon an excess of freshly precipitated carbonate of silver, the red arsenate, instantly formed, shows clearly on the ground of snow-white carbonate; this is a striking reaction, and therefore a delicate test. Oxyd of silver, when tried in a like manner, gave no results of value.

Experiments made with sulphate, succinate, and nitrate of ammonium, served merely to establish the superiority of the acetates and carbonate, as above described; and to show that arsenate of silver is less readily precipitated in the presence of sulphate, and especially nitrate of ammonium, than from solutions of either of the other salts employed in my experiments.

When present in relatively large quantity, arsenic acid readily precipitates silver from a solution of nitrate of ammonium and ammonio-nitrate of silver, but the color is uncertain; the same objection applies to the succinate of ammonium.

The experiments were varied by changing the proportions of

the reagents so that the several results stated above are deduced from many trials. The same tests being tried, with phosphoric in place of arsenic acid, the alkaline acetates gave the best results.

The following quantitative experiments were then tried, to determine the relative solubility of arsenate of silver in solutions of acetates of sodium, Rochelle salt, and nitrate of ammonium. In experiment No. 1, 20 grms. of crystallized acetate of sodium were dissolved in a mixture of 100 c. c. of water and 60 drops of a saturated solution of arsenic acid. In No. 2, 20 grms. of Rochelle salt and in No. 3 a like quantity of nitrate of ammonium, were dissolved in similar mixtures.

Each of the solutions was nearly but not quite neutralized by adding carbonate of sodium; and ammonio-nitrate of silver was then added to them from a dropping flask. A slight milkiness, due probably to the presence of chlorine in the carbonate of sodium, was produced by the first drop in each of the three solutions; but this cloudiness had a distinct red tinge in he tartrate solution, (No. 2,) while it was bluish white in the acetate (No. 1) and the nitrate (No. 3). When twelve drops of the ammonio-nitrate had been added, a distinct red precipitate had formed in No 1, (acetate,) and a very strong one in No. 2, (tartrate); but no change appeared in No. 3, (nitrate). The precipitate in No. 2, showed white streaks, or layers, at the moment of its formation (probably of bitartrate of potassium), which soon disappeared.

Adding more of the ammonio-nitrate to No. 3, with occasional drops of arsenic acid solution, to keep the mixture acid, a purple cloudiness appeared, which increased up to the sixtyfifth drop, but no precipitate fell.

In a repetition of No. 3, (nitrate,) no precipitate fell until 80 drops of the ammonio-nitrate of silver had been added; when it reached 85 drops, the precipitate was exceedingly voluminous. In three other experiments, the liquids were acidulated with tartaric acid. In experiment A, 75 drops of a strong solution of arsenic acid, 125 c. c. of water, and 25 grms. of crystallized acetate of sodium, were mixed, neutralized with carbonate of sodium, and tartaric acid added. In B and C, 25 grms. of Rochelle salt and of nitrate of ammonium were respectively substituted for acetate of sodium.

Seven drops of ammonio-nitrate were then added to each of the solutions. In A and B, the red-brown arsenate precipitate appeared; but in C, no precipitate was formed except the light cloudiness due to impurities.

Boston, Dec. 1, 1868.

ART. XXIII.-Notices of papers in Physiological ChemistryNo. II; by GEORGE F. BARKER, M. D.

5. On the formation of Sugar in the liver.*

[Continued from page 32.]

A

(43.)† In a paper on the origin of the sugar of the chyle, published June 28, 1858, COLIN calls attention to the large amount of saccharine matter produced in the intestines. horse, for example, consuming daily 5 kilograms of hay, as much straw, and 3.6 kilos. of oats, obtains from this food, according to Boussingault, 6196 grams of sugar, of starch and of other analogous principles, capable in great measure, of being converted into glucose and dextrin. A fraction of this mass enters the portal vein, goes to the liver, and thus reaches the general circulation; another fraction, mixed with the chyle is absorbed by the lacteals, and poured into the blood; so that this fluid finally receives the whole of the absorbed products. Is it a matter of surprise, therefore, that sugar should be found in the chyle? and can its intestinal origin be doubted? Poisseuille and Lefort have asserted that the sugar found in herbivorous chyle is brought by the lymphatics and the arteries from the liver; though they have given no proof of this singular assertion. To account for the presence of sugar in the chyle they contend, 1st, that it is carried there by the blood and lymph; 2d, that the lacteals cannot absorb it, even when ready formed; 3d, that the chyle is simply an intestinal lymph, to which fatty matters have been added; and 4th, that the glucose found there is in small quantity, much inferior to that existing in other lymphatic vessels. As to the first point, the liquid taken from the great chyliferous trunks passing to the receptaculum chyli of a carnivore, as also the fluid drawn from the large lacteal vessels which accompany the mesenteric arteries of ruminants fed on meat, is evidently pure chyle unmixed with lymph; and, as it contains sugar, this sugar must have come from the intestine. To controvert the second assertion, nothing is easier than to show that the lacteals absorb saccharine substances with great facility; since the liquid

From this point, the range of the discussion widens to include the amyloid function not only of liver tissue, but also of other tissues. The obvious bearing of these facts on the glycogenic theory, is a sufficient reason for noticing them in this connection.

Through an oversight, paragraphs (43) and (44) in the January number, were prematurely inserted. They should be numbered (55) and (56), and placed in their proper order.

C. R., xlvi, 1264.

taken from a thoracic fistula after a saccharine food, shows a gradually increasing proportion of sugar as the absorption goes on. In the third place, physiologists generally agree in considering the chyle as the product of an absorption effected by the intestinal villa; they think with reason that the chyle comes from the food because of its fibrin and albumin, its fatty matters and salts. Why does it not derive its sugar also from this source? But lastly, it is not true that chyle contains less glucose than lymph. The authors quoted have compared the one liquid taken from a herbivore, with the other from a carnivore; the chyle from a mutilated and dying cow with the lymph taken from a dog. Colin, on the other hand, as a result of experiments on more than 30 animals, cows, bulls, rams, pigs, and dogs-the two fluids being collected at the same timefinds that the amount of sugar is sometimes equal in both, sometimes unequal; but that in the latter case, the difference is always in favor of the chyle. The absolute quantity of glucose in the chyle is somewhat variable; it is less with herbivora than with flesh-eaters, since in the former the chyle is largely diluted. In the solipeds and in ruminants fed on hay and straw, it oscillates from 130 to 160 milligrams to the 100 grams of liquid; in the carnivora fed exclusively on meat, from 120 to 140. It increases rapidly when the food is rich in sugar; with a dog previously fed on meat, it rose from 137 to 205 milligrams within two hours after the ingestion of a liter of milk containing 40 grams of glucose, and then returned to the normal quantity. In the case of a horse whose chyle, when fed on hay and straw, contained 150 milligrams of sugar in 100 grams, it rose to 214 milligrams in one hour after giving the animals 200 grams glucose in several liters of water; and to 259 milligrams two hours later. Moreover, the activity of digestion affects the amount of sugar. The chyle of a bull, having a thoracic fistula from the mesenteric lacteals, and another from the lymphatics of the neck, showed at first 104 to 110 milligrams of sugar; as the animal became feeble it fell to 84, then to 66, and at death only traces could be detected. Lastly, compared with the lymph, the quantity of sugar varies very little; in a bull. 100 grams of chyle contained 106, of lymph 102 milligrams of sugar; horse A, chyle 149, lymph 123 milligrams; horse B, chyle 141, lymph 112; mare, both 158 milligrams; a dog, chyle 128, lymph 152; a second dog, both 135.

(44.) POISSEUILLE and LEFORT replied to Colin, July 19,* stating that he had evidently confounded the results of separate experiments made by them. His statements refer solely to

C. R., xlvii, 112.