efficient than the nitrate, and of potassium salts the sulphate than the chloride.

V. H. V.

Egg Albumin and Albumoses. By R. H. CHITTENDEN and P. R. BOLTON (Studies from Lab. Physiol. Chem., Yale Univ., 2, 126-155). -These experiments were designed to contrast the products of digestion of egg albumin with those obtained by Kühne and Chittenden from fibrin. Four samples of albumin were prepared; in some cases it was separated from globulin by saturation with magnesium sulphate, in others by dilution and the subsequent addition of acetic acid. An elementary analysis of these four samples gave the following average percentages:-C, 52:18; H, 6·93; N, 1581; S, 187; 0, 23-21. Further, coagulated products did not differ in composition from non-coagulated albumin. These results do not agree with any of the formulæ ascribed to albumin by previous observers.

Peptic digestion of the albumin resulted in the formation of albuminoses of which the percentage composition and reactions were determined. In composition they were found to differ from each other somewhat more than the albumoses from fibrin; collectively, however, there is less difference in composition between the albumoses and the egg albumin from which they were formed than in the case of the albumoses from fibrin. The following table gives the final results:

In their reactions the different albumoses (proto-, deutero-, hetero-, and dys-albumose) obtained from egg albumin do not differ essentially from those obtained from fibrin.

W. D. H.

Metallic Compounds of Albumin and Myosin. By R. H. CHITTENDEN and H. H. WHITEHOUSE (Studies from Lab. Physiol. Chem., Yale Univ., 95—125).—Many researches on the subject of the metallic compounds of albumin have been carried out since Lieberkühn attempted to establish the molecular weight of albumin by the analysis of copper albuminate. The more recent work of Harnack (Abstr., 1882, 747) showed that two compounds of albumin (from white of egg) with copper occur, one containing 1.35 per cent., and the other 2.64 per cent. of copper. In the present research, albumin was freed from globulin by the use of dilute acetic acid, and both the

acetate and sulphate of copper were used in the preparation of the albuminate. The precipitate was well washed with water, powdered, and dried. The percentage of copper was first determined as cupric oxide by ignition and weighing: the oxide was then dissolved in dilute nitric acid, treated with hydrogen sulphide, and the amount of cuprous sulphide obtained weighed. By the former method, in 15 preparations the average result was 117 per cent. of copper; and by the latter method 0.94 per cent.; the preparations, therefore, contain 0.23 per cent. of ash. In order to obtain less ash, Harnack dissolved the albuminate in sodium carbonate and reprecipitated it by the careful addition of acid; this process was repeated several times. This treatment certainly increases the percentage of copper, but is a source of error, as the sodium carbonate withdraws a portion of the albumin. Long-continued washing with water also causes partial dissociation of the compound. The results obtained correspond with the formula (C72H112N18ŠO22)4 + Cu · H2.

A normal lead salt causes a small precipitate when added to albumin, whilst with basic lead acetate the albumin is completely precipitated. This confirms the previous statement of Berzelius (Lehrbuch der Chemie, 9, 29). The preparations were well washed from both lead and albumin, dried, and the lead was determined, first by simple. ignition, and then obtained as sulphate, which was ignited. The results indicate that more than one compound of lead is formed, that made by the addition of a large excess of the basic acetate, containing about five times as much lead as the ordinary basic lead compounds. An iron compound which was found to be more stable than the copper albuminate, and corresponded fairly well with the formula (C72H112N18SO22) + Fe - H3, and a zinc compound, (C72H112N18SO22)4 + ZnH2; acid compounds with uranium, (CH112N18SO22)3 + U — H2; with mercury, (C2H112N18SO22) + Hg - H2; and with silver, (CH112N18SO22)3 + Aga - H2, were also prepared and analysed. Much stress is not laid on such formulæ, as it seems possible to form a large variety of compounds by simply modifying the conditions of precipitations. This, with the undoubted tendency of the compounds to dissociation, may account for the lack of agreement in the results of different workers. Similar compounds prepared from myosin obtained by extracting ox flesh with 15 per cent. ammonium chloride were prepared, and the percentage results show that these two forms of proteïd matter do not form corresponding compounds with the metallic salts used. This is illustrated by the following table:

Caseïn and Caseoses. By R. H. CHITTENDEN and H. M. PAINTER (Studies from Lab. Physiol. Chem., Yale Univ., 2, 156-199).Danilewsky (Zeit. physiol. Chem., 7, 433) has asserted that caseïn is a mixture of two proteïds, caseoprotalbin, partially soluble, and caseoalbumin, insoluble in hot 50 per cent. alcohol. Hammarsten (ibid., 7, 227) has shown that the peculiar behaviour of Danilewsky's casein is due to its containing calcium phosphate, the presence of which impurity depends on the use of hydrochloric acid in the precipitation of the caseïn, as this acid does not favour the removal of the salt as well as acetic acid. He also considers that caseïn is a single proteïd. In the present research, seven distinct preparations of casein were made. Elementary analyses show a close agreement throughout, and the results mcreover accord closely with those of Hammarsten.

In the digestion of caseïn with hydrochloric acid, peptones are ultimately formed, and the name caseose is given to the intermediate products. These were separated by the methods of Kühne and Chittenden into proto-, hetero-, and deutero-caseose, which correspond with the albumoses with similar names. The quantity of heterocascose obtained was usually very small. The reactions characteristic of albumoses apply generally to the caseoses. Unlike proto-albumose however, protocaseose is precipitated from aqueous solutions by acetic acid. The average of the analyses of 10 preparations of protocaseose gives the following percentage results :

Deuterocaseose contains a smaller percentage of carbon than protocascose, and heterocaseose contains fully as much carbon as caseïn itself.

An insoluble, semi-gelatinous substance which separates in the first stage digestion has not yet been investigated. Weyl's commercial" caseïn-peptone" contains large quantities of caseoses.

W. D. H.

Animal Tannin. By M. VILLON (Chem. News, 56, 175).-Corn weevils (Calandra granaria) were killed, ground in a mortar, and digested for one hour in boiling 90 per cent. alcohol. The residue from the evaporation of the extract is taken up with ethyl acetate at 50°, and precipitated by means of ammoniacal zinc acetate. The precipitate is decomposed with oxalic acid, and the solution evaporated in a vacuum. In this way 3 per cent. of a substance having all the general properties of tannin is obtained from the weevils. This animal tannin, fracticornitannin, forms small reddish-yellow scales.

D. A. L.

Physiological Chemistry.

Influence of some Organic and Inorganic Substances on Gas Metabolism. By R. H. CHITTENDEN and G. W. CUMMINS (Studies from Lab. Physiol. Chem., Yale Univ., 2, 200-236).-The question investigated was the consumption of oxygen and the elimination of carbonic anhydride. The animal, a rabbit, was confined in a large bell-jar, through which air was drawn by means of three aspirators, which were arranged to work evenly, and emptied themselves in half an hour. Two-fifths of the mixed air was drawn successively through three absorption-tubes, each of which contained 100 c.c. of standard baryta. The carbonic anhydride was estimated by titration with standard oxalic acid. The results obtained are not absolute, but express quite well the comparative action of the various substances experimented with. It was found necessary to have the animal in a state of hunger during an experiment. Accordingly it was deprived of food for three days previous to the experiment, as well as during the three days for which the experiment lasted. During the last two of the three days, the substance experimented with was given in small and oft-repeated doses. Every day numerous determinations of carbonic anhydride were made, and comparison could then be made between the normal excretion of that substance and that during the administration of the drug. Observations were also made on the body temperature. The following are briefly the results obtained with the following substances:-Uranyl nitrate acts slowly, but when taken in sufficient quantity tends to raise materially the body temperature, and increase very noticeably the excretion of carbonic anhydride. Copper sulphate shows a marked influence in depressing body temperature, and a still greater influence in diminishing the production of carbonic anhydride. Arsenious oxide

given in non-toxic doses also diminishes the excretion of carbonic anhydride, presumably through its action on the metabolic activity of the tissue cells. Tartar emetic acts similarly. Morphine sulphate has but little action; a slight fall in the excretion of carbonic anhydride immediately after a dose must be attributed to the semisomnolent condition of the animal. Quinine sulphate, which is so important as an antipyretic, is found in a healthy hungry rabbit

to have only a very slight depressing influence on body temperature, and a minimum effect on the production of carbonic anhy dride. Cinchonidine sulphate produced a slight rise of temperature and a gradual diminution in the amount of carbonic anhydride excreted; it also caused tetanic convulsions and finally death. Antipyrine had little effect, the only change being a fall in temperature just before death. W. D. H.

Action of Uranium Salts on Digestive Ferments. By R. H. CHITTENDEN and M. T. HUTCHINSON (Studies from Lab. Physiol. Chem., Yale Univ., 2, 55-67).—Very small quantities of uranium salts promote the activity of saliva; larger amounts hinder and finally stop its amylolytic action. 100 c.c. of digestive mixture contained 2 c.c. of neutralised saliva, and 1 gram of pure potato-starch. The mixture was warmed at 40° for 30 minutes and then boiled to prevent further action; the amount of sugar present was estimated as dextrose by Fehling's solution. This was compared with other digestive mixtures to which a certain percentage of a uranium salt was added. The following table illustrates the results obtained with uranyl nitrate:—

Uranyl acetate was more inhibitory in its action, due possibly to its greater acidity; 0.0003 per cent. increases, 0'003 per cent. stops amylolytic action. Experiments with similar results were obtained with ammonio-uranous sulphate, sodio-uranic sulphate, potassio-uranic oxychloride, and ammonio-uranic citrate. The action of the salts varies with different specimens of saliva, according to the amount of proteïd present which is precipitated by them. Loss of amylolytic power is due in part to partial direct destruction of the ferment as well as to change in reaction of the fluid. There must, however, be something in addition to the mere presence of these salts dependent on chemical constitution that controls the action of the ferment.

The same salts were also investigated in regard to their action on the proteolytic action of pepsin-hydrochloric acid. Similar series of experiments were made, and the results compared by estimating the amount of fibrin left undigested after an hour's action. The potassium uranic oxychloride was the only salt that produced initial stimulating action on the ferment; the others all retarded its activity. The difference in action of the various salts seems to depend on the