C8H1NO4, HCI+PtCl. The formation of these bases is indicated by the equations

3C4H4O2+NH3=(C4H4O2)3H3N-C12H16NO6

2C4H4O2+NH3 (C4H4O2)2H3N-C8H11NO4.

These results may perhaps lead to the discovery of the true constitution of the complex organic alkaloids containing oxygen. They appear to show that Berzelius' view that the alkaloids are conjugates of ammonia may be true in some cases at least.-Comptes Rendus, xlix, 898.

7. On a new series of Alcohols.-WURTZ has also found that oxyd of ethylene unites with water to form new alcohols which he terms diethylene and triethylene alcohols. The reactions involved are represented simply by the equations

€4H4O2+2HO=C4H604. Glycol.

C4H4O2+2HO=C8H1006. Diethylene-alcohol.

3C4H4O2+2HO=C12H14O8. Triethylene-alcohol.

Oxyd of ethylene also unites directly with glycol so as to form the diethylene and triethylene alcohols. The equations are

2C4H4O2+C4H604 C12H14O8.

C4H4O2+C4H604C8H1006.

All these substances behave like alcohols. The diethylene alcohol was also discovered by Lourenço and called by him the intermediate ether of glycol.-Comptes Rendus, xlix, 813.

8. Researches on the Platinum metals; from a letter of Dr. WOLCOTT GIBBS to one of the editors.-"The completion of my researches on the platinum metals has been delayed much longer than I expected. This has arisen partly from the intrinsic difficulty of the subject and partly from its expansion in particular directions in a very unexpected manner. A brief sketch of the results already obtained-imperfect as they aremay perhaps interest you.

The osmium-base of which Dr. Genth and myself published a brief notice about two years since in the Journal, proves to be the type of a very extensive series of compounds which promise to be of much theoretic interest. You will remember that the chlorid of that base is produced when osmite of potash, KO, OsOз, is added to a solution of salammoniac. I now find that new complex bases are formed when the osmite is added to solutions of the chlorids of narcotin-ammonium, cinchonin-ammonium, &c., &c.; in short, almost all the complex alkaloids which I have yet tried give analogous bases containing osmium in the radical. The new bases are very easily decomposed with evolution of osmic acid. They are more stable in the presence of an excess of chlorhydric acid and give crystalline double salts with the chlorids of gold and platinum.

These however are not the only or even the most remarkable basic compounds which I have discovered. Many of the ammonia-metal bases already described are capable of forming new bases into which osmium enters either as a conjugate body or as replacing hydrogen. When, for instance, osmite of potash is added to a solution of the chlorid of pallad-diamin, 2NH3. PdCl, a yellowish brown solution is formed

which on addition of chlorhydric acid gives a beautiful yellow crystalline precipitate insoluble in cold water and containing palladium, osmium and the elements of ammonia.

When osmite of potash is added to a solution of chlorid or sulphate of luteocobalt, 6NH3. Co2Cl3 or 6NH3. Co2O3. 3SO3, a buff-yellow precipitate is thrown down, which on addition of HCl gives a wine-yellow solution. This solution after a short time deposits beautiful crystals of the chlorid of a new base containing osmium, cobalt, and the elements of ammonia. The chlorid gives well crystallized salts with the chlorids of platinum, gold and mercury. Its solution is decomposed by gentle heating, osmic acid being evolved while a black powder is thrown down.

The other ammonia-cobalt bases give analogous compounds which however are decomposed almost as soon as formed. It is my intention if possible to examine the relations of osmite of potash to one or two of the arsenic and antimony bases, as for example to the chlorids having the formulas (C4H5)+ASCI and (C4H5)4SbCl.

When ammonia is added to a solution of osmite of potash, the red color of the latter passes immediately to wine-yellow. Fremy supposes that an osmiamid is formed here having the formula OsO2. H2N. I find that the product is a new osmium base, the chlorid of which is formed at once by neutralizing the yellow solution with HCl. This chlorid has probably the formula UNCI, though it may be

H3 NCI.

In like manner I am still doubtful whether the true forOsO2 mula of the other ammonia-osmium base is 2NH3. OsO2. O or 2NH3. OsO+2HO. Experiments now making will decide this point. Meantime I may say that I do not agree with Claus in considering the formula 2NH3. ÖSO2. O improbable since we have many analogous cases, as for instance in xanthocobalt and flavocobalt, the formulas of the chlorids of which are, you will remember,

NO2. 5NH3. Co2 O. Cl2 and 2NO2.5NH3. Co2 O. Cl2.

The chlorid 2NH3. OsO2. Cl or 2NH3. OsCl+2HO is decomposed by boiling with chlorhydric acid, giving sal-ammoniac and a new chlorid which is perhaps identical with that just mentioned. This is easily explained by the equation

2NH3. OsCl+HCl=NH3. OsCl+NH4Cl,

and the decomposition would then be exactly analogous to that of pallad-diamin under the same circumstances

2NH3. PdCl+HCl=NH3. PdCl+NH4 Cl.

I hope soon to decide these questions by experiment and analysis, but the analyses are very difficult and tedious, and I have to feel my way and find out new methods for almost every determination.

You will see from the above that osmium is likely to become one of the most interesting of the elements, and that it is capable of forming an extraordinary number and variety of compounds. I am also very busy with the remarkab e class of double nitrites which I described at the meeting of the Am. Association for the Advancement of Science in August last. There are still some difficulties to be overcome, but I am

confident that I shall be able to effect a perfect separation of all the metals of the platinum group. The white iridium salt appears to have the formula Ir2O3, 2NO3+3KO. NO3, but it contains a small quantity of chlorine which may be essential. Its stability and insolubility are very remarkable, and will I think prove of great value in separating iridium from the other metals. The corresponding salts of the other metals of the group appear to be all soluble. The ruthenium salt is soluble even in alcohol and ether, and gives with sulphid of ammonium a magnificent red solution. This is by far the most delicate test for ruthenium yet discovered as the reaction is peculiar to that metal. Claus' beautiful reactions with sulphocyanid of potassium and sulphydric acid are much inferior for qualitative purposes."

New York, March 30, 1860.

TECHNICAL CHEMISTRY.

9. Solution of Cellulose in Ammonio-oxyd of copper.-Some time since ERDMANN, in his Journal für praktische Chemie (lxxvi, 386) expressed the opinion that cellulose is not really dissolved by cuprate of ammonia, as stated by SCHWEIZER (ibid, lxxii, 109), but only swollen to a sort of thin mucilage like the well known limpid "solution" of starch.

This view was based upon the fact that when a clear solution of cellulose in NH3 CuO is diluted with a large excess of water, the cellulose separates entirely in the course of a few days.

In defense of his original statement Schweizer now urges that the cellulose must be really dissolved by ammonio-oxyd of copper: since its fibres are unquestionably destroyed when this reagent comes in contact with them-as may be distinctly seen with the microscope; and since the cellulose precipitated from a solution of cotton in the above mentioned reagent no longer exhibits any trace of definite structure.

A solution of cotton in NH3 CuO may also be filtered perfectly clear--although this operation is somewhat difficult when large quantities of the liquid are operated upon. The solution is moreover capable of passing through the cell membranes of plants as shown by CRAMER (ibid, lxxiii, 6).

The destruction of the solvent power of the cuprate of ammonia by dilution appears to depend upon alterations which this compound itself undergoes under certain circumstances. It often happens that a solution of ammonio-oxyd of copper which at first dissolved cotton with the greatest ease gradually loses this power even when kept in carefully closed vessels completely filled with the liquid. It is moreover well known to chemists that solutions of ammonio-oxyd of copper, and of the ammoniocopper salts, undergo decomposition when diluted with large quantities of water;-hydrate of copper being precipitated.

This decomposing influence which water exerts upon solutions of the compound of ammonia and oxyd of copper is in the opinion of Schweizer the cause of the gradual precipitation of cellulose from such solutions when these are largely diluted.-Journal für praktische Chemie, lxxviii, 370; compare also CRAMER, ibid. lxxiii, 1, et seq.

10. Decoloration of Indigo by Sesquioxyd of Iron.-According to KUHLMANN, when a solution of blue indigo is acted upon at the temperature of 150° (C.) by hydrated oxyd of iron its color is, almost immediately, completely destroyed. The same thing occurs with a number of other coloring matters.—In noticing this fact BARRESWIL suggests that persul

phate of iron may perhaps be applied in calico-printing as a discharge for indigo and also in bleaching blue rags for paper making.-Répertoire de Chimie Appliquée, Oct. 1859, p. 420.

[The observation that salts of the sesquioxyd of iron have the power of bleaching indigo and other organic coloring matters was first made by Prof. H. WURTZ of Washington and published two years since in this Journal (vols. xxv, 378, and xxvi, 52)-also in the Proc. of Amer. Assoc. 1858 and in several foreign journals-to him unquestionably belongs whatever credit may attach to the discovery-F. H. s.]

11. Aluminum Leaf-A Parisian gold beater, DEGOUSSE, has succeeded in obtaining leaves of aluminum as thin as those from gold and silver. The aluminum must be reheated repeatedly over a chafing dish during the process of beating. This leaf is less brilliant than that of silver but it is not so easily tarnished as the latter. It is easily combustible, taking fire when held in the flame of a candle and burning with an exceedingly intense white flame.

According to FABIAN, (Dingler's polyt. Journal, cliv, 438,) the chemical lecturer will find aluminum leaf to be well adapted, for exhibiting the characteristic properties of the metal. It dissolves, for example, with surprising rapidity in a solution of caustic alkali.

[A specimen of this leaf accompanies the description of it in Répertoire de Chimie Appliquée, Oct. 1859, p. 435, also Nov., p. 488.]

12. Critical and Experimental Contribution to the Theory of DyeingUnder this title a somewhat extended treatise by Prof. BOLLEY of Zurich has appeared in the L. E. and D. Philosophical Mag. [4] xviii, 481, Supplement to Dec. 1859.

Two questions have long been agitated among chemists interested in the theory of dyeing. (1.) In what part of the colored fibre is the coloring matter situated? Does it merely adhere to the surface, or does it penetrate the entire substance of the cell-walls of such fibres as cotton and flax? Or lastly, in the case of such fibres is it stored up in the interior of the cells? (2.) What is the nature of the union between the dye and the fibre? Is it a chemical combination, or is it due to mere surface attraction? After comparing the various theories which have been advanced during the last century and discussing the merits of each, the author records the results of his own experiments, from which it appears that wool and silk in all cases where they have not been dyed with colors in a mere state of suspension seem to be impregnated with the dye throughout their entire mass; while in the case of cotton, by far the larger portion of the coloring matter adheres to the surface of the fibre, the penetration of the cell-walls by the dye being either very slight or altogether wanting.

That the theory of W. Crum (L. E. and D. Phil. Mag,, April, 1844,- · compare this Journal, [2], xxviii, 125), in accordance with which the tubular form of the cotton fibres is an essential condition to their taking a dye, is unfounded, appears from the fact that the amorphous cotton-gelatine precipitated from its solution in cuprate of ammonia (see this Journal [2], xxvii, 118) may be mordanted and dyed like ordinary cotton. In like

* In which case the coloring matter only adheres as a crust to the surface of the fibre.

manner sulphate of baryta and other pulverulent mineral bodies may be mordanted and dyed with decoctions of dyewoods.

With regard to the nature of the force which binds the coloring matter to the fibre-whether or no it be chemical attraction? Bolley concludes that there is no sufficient reason for accepting the view, principally developed by Chevreul [and by Kuhlmann, Comptes Rendus. Tomes xlii, xliii et xliv], that dyeing is a direct consequence of chemical affinity. He believes that the power possessed by fibres of attracting certain bodieswhether salts or coloring matters or both-from their solutions, belongs to that class of phenomena which results from the action of finely divided mineral or organic bodies (charcoal or bone black for example,) on such solutions. The distinction between the action of charcoal and of fibres in thus removing saline matters, or dyes, from their solutions is one of degree only, the nature of the operation being identical in either case.

A given weight of well prepared animal charcoal can, as a rule, deprive a larger quantity of liquid of its color than an equal weight of wool or silk. Neither wool or silk can remove all the color from a solution as charcoal can, their effect extending only to a certain degree of dilution beyond which the particles of coloring matter resist their attraction. Dyes which may have been taken up without a mordant by wool or, especially, by silk may be removed again by long washing in water, a fact which is not true in the case of charcoal, or only to a very slight extent. The attraction of coloring matters for water is therefore more completely overcome by charcoal than by animal fibre; but even the cleanest vegetable fibres, as, unmordanted and completely bleached cotton, possess a certain power of attracting coloring matter. That cotton should have less effect in this matter than wool or silk is not surprising in view of the great difference in the structure of cotton fibre as compared with that of the two substances last mentioned. It is well known that wool and silk in consequence of their physical constitution belong to the class of strongly absorbent or hygroscopic substances, i. e. in consequence of a certain porosity or looseness of their particles they swell up when moist and become easily penetrated by a liquid throughout their entire mass; on the other hand the cell-walls of cotton fibres are denser, less penetrable and at the same time thinner and therefore unable to contain the same quantity of liquid.

It has been often urged that since fibres, especially those of animal origin, not only exert an attraction for salts &c. but also possess the power of decomposing some of them, their action must be chemical But in this respect the behavior of charcoal is similar to that of the fibres. So too with regard to the increased attraction for color exhibited by mordanted cotton which is on a par with the fact observed by Stenhouse that the decolorizing power of wood charcoal is considerably increased by precipitating alumina upon it.

According to the Author mordants act by producing insoluble colors (lakes). Their behavior towards coloring matters in solution must be ascribed to chemical affinity, with which however the fibres themselves have nothing to do.

The so-called substantive dyes become insoluble from some other cause than the addition of a mordant, for example oxydation of protoxyd of iron, or of white indigo.