bisulphid of carbon, but difficult to get pure ether and alcohol, while chloroform always contains many substances mixed with it which cannot be separated. Thus the tension of its vapor as directly determined was 342 2 at 36°, while by the method of ebullition it was found to be 313-4 at the same temperature. Some liquids change their molecular structure when long boiled under high pressures, and of this change oil of turpentine furnishes a remarkable example. Other liquids appear to undergo molecular changes when left to themselves for a long time in hermetically sealed tubes: ether is a curious instance of this.

With respect to the second part of the subject, as indicated above, the author in the first place verified the law of Rudberg, that whatever be the boiling point of a saline solution the vapor has simply the temperature which it would have if disengaged from pure water. As how. ever the experiments of Rudberg were made only at ordinary atmospheric pressures, Regnault endeavored to extend our knowledge of the subject by studying the phenomena of the ebullition of saline solutions under very different pressures. The apparatus used was a copper boiler through the cover of which four closed tubes passed, two into the liquid beneath and two into the vapor above. The tubes contained a small quantity of mercury and served to contain the thermometers. A tube communicating with a refrigerator served to condense the vapor while the refrigerator itself communicated with a large reservoir of air, the pressure of which could be varied at pleasure. Concentrated solutions of chlorid of calcium were placed in the boiler and made to boil under various pressures while at the same time the temperatures of the liquid and the vapor were measured. The author gives two tables of his results, of which however our limits permit us to give but one. In this table the first column gives the pressures corresponding to the temperatures observed, the second the temperatures of the liquid, the third the temperature of the vapor and the fourth the temperature which the vapor would have had if it had been produced by distilled water boiling under the same pressure.

It will be seen from this table that the thermometer plunged into the vapor constantly indicates a temperature a little higher than that which corresponds to the vapor of pure water under the same pressure. The small difference may be attributed to radiation from the hotter liquid and to small drops of liquid projected by the boiling saline solution. The thermometers plunged into the liquid indicated abrupt variations, sometimes of several degrees. It may then be admitted that the law of Rudberg holds good for pressures much higher and much lower than the ordinary pressure of the atmosphere. By placing the bulb of the thermometer in different parts of the space above the liquid, the author found that, at a distance of 3 or 4 centimeters above the surface of the solution, the thermometer was always wetted and consequently could only indicate the temperature of the vapor of pure water. When the

bulb descends toward the surface the temperature rises but the bulb dries and this drying takes place only in the layers of vapor immediately above the liquid. This appears therefore to give the explanation of the fact observed by Rudberg; the vapor has at first the temperature of the saline solution, as it rises however it parts with its excess of heat in consequence of its low specific heat as compared with its volume, and thus at a certain height above the surface of the liquid has only the temperature which it would have if coming from pure water. As it was found impossible to determine the temperature of the ebullition of saline solutions so as to deduce from these observations certain results, the author directed his attention to the elastic forces of the vapors which these solutions emit in vacuo. These tensions may be determined with great accuracy and their observation has already led to most interesting and valuable results. Thus when any molecular change takes place in the substance dissolved, it is immediately manifested by a singular point in the curve of the elastic forces of the vapor furnished by the solution. Regnault suggests therefore that the study of the tensions of the vapor of saline solutions promises to yield results as important as those which Biot has deduced from the phenomena of circular polarization.

It is generally admitted by physicists that vapors behave in gases as in vacuo, excepting that in gases the equilibrium of tension is established slowly, while in vacuo this takes place instantaneously. The experiments of Regnault on this point confirm those of Magnus and conclusively show that elastic force of a vapor in a gas is less than in vacuo. The apparatus employed was the same as that used by the author in his former researches on the tension of the vapor of water and consisted of a vessel of 600 or 700 C. C. capacity communicating with a mercurial manometer. The whole apparatus was placed in a large vessel filled with water kept at a constant temperature. The vessel was then filled with dry air and the tension of this first determined at different temperatures. A small glass bulb previously filled with the liquid to be examined and introduced into the glass vessel was then broken by heating the vessel sufficiently; the vapor thus produced caused the mercury in the manometer to descend and its elasticity therefore could easily be measured. The following table gives the results of a single series of experiments on the tension of air containing the vapor of ether, the first column giving the temperature, the second the elastic force of the vapor in the air, the third the elastic force of the vapor in vacuo, and the fourth the difference between these two tensions.

Precisely similar results were obtained with a different form of apparatus, with different vapors in air, and with the vapor of ether in dif ferent gases. With this second form of apparatus the author also studied the influence exerted on the tension of the vapor by the total pres sure of the atmosphere which acts on the liquid and by the quan. tity of liquid in excess which moistens the sides of the containing ves

sel. The author's conclusion is that the law of Dalton on mixed gases and vapors may be regarded as a theoretical law which would proba bly be verified with all rigor in a vessel the walls of which should be formed by the volatile liquid itself of a certain thickness, but this law is inaccurate in our apparatus because the hygroscopic affinity of the matter of the vessel brings the vapor to a tension which is variable and which is always inferior to that which corresponds to saturation. Thus the table which we have given above gives the maximum tension of the vapor of ether in air at the temperatures mentioned. The tension of the vapor at the moment that dew begins to deposit on the vessel is far from being equal to that of the vapor in vacuo. If we compress the gaseous atmosphere the condensed liquid becomes more abundant and the tension of the vapor increases and approaches more and more the tension in vacuo. But it only becomes equal to this when a thick layer of liquid is found on the surface of the mercury and when the observation is made immediately after the reduction of volume.

As the result of his experiments on the tension of the vapor of mixed volatile liquids in vacuo, the author finds that two volatile liquids which are not capable of dissolving each other give in vacuo a tension of vapor equal to the sum of the tensions which these substances present separately. It is however only in this case that the law of Dalton is verified. Whenever the liquids mutually dissolve each other, the tension of the vapor is in general less than that which would be produced by the most volatile liquid separately. The difference is greater the less the proportion of the less volatile liquid.

With respect to the last of the subjects of investigation undertaken, the author found that the vapor of liquid acetic acid below the temperature of condensation possesses a higher tension than that of the solid acid at the same temperature. He attributes this however, to the presence of small quantities of foreign substances which in the case of the liquid acid are disseminated throughout the whole mass and therefore exert but a slight influence on the tension of the vapor. In the act of congelation however, the impurity separates from the mass, and therefore must exert a much greater influence on the tension.-Comptes Rendus, xxxix, 301, 391, 345, August, 1854.

2. On butylic Alcohol.-WURTZ has obtained butylic alcohol in considerable quantity by the fractionated distillation of certain varieties of fusel oil. These contained common alcohol, and the corresponding compounds of amyl and butyl, but not that of propyl which Chancel has found in the fusel oil of brandy. Perfectly pure butylic alcohol was obtained by decomposing pure iodid of butyl obtained from the raw alcohol, by means of acetate of silver, and then decomposing the acetate of butyl by caustic potash. It is a colorless liquid more fluid than amylic alcohol, and having a somewhat similar but more vinous odor. It does not deflect the plane of polarization, boils at 109° C. and has a density at 18° 5 of 0-8032. The density of its vapor is 2.589, and corresponds to 4 volumes. It takes fire readily and burns with a luminous flame; it is soluble in 10.5 times its own weight of water at 18; with potassium and sodium it forms potassic and sodic alcohol; heated with soda lime it gives butyrate of soda. Sulphuric acid forms with it sulpho-butylic acid when the mixture is kept cool, but

at common temperatures decomposes it, forming various hydrocarbons. Chlorid of zinc gives with the alcohol, butene C8H8, hydruret of butyl C8H10, and liquid carburets. By the action of potassium on iodid of butyl the author obtained butyl Cs H9, (or in Gerhardt's view, CsH9 Call); it is a colorless oily liquid. The chlorid, iodid and bromid of butyl were found by the usual processes; they are liquid and resemble corresponding compounds of amyl. The iodid of butyl readily reacts with the salts of silver, and in this manner many of the ethers of this radical may be obtained. Wurtz did not succeed in preparing the oxyd of butyl in a perfectly pure state; if appears to be a liquid having an agreeable odor and boiling between 100 and 105°. The carbonate, acetate, nitrate and formate of butyl are all liquid and do not require special notice in this place. By distilling sulpho-butylate of potash with cyanate of potash, dissolving the distillate in alcohol and then distilling this mixture with caustic potash, the author obtained butylamin. C8 H9 In a pure state this ammonia N

{

H is liquid and boils at 69°-70°;
H

it is soluble in water, alcohol and ether, and the aqueous solution is extremely caustic and precipitates most metallic solutions like ammonia itself; it dissolves alumina and even gelatinous silica. The platinum salt of this base is C8H12NCI+PICl2; the gold salt is 2C8H12NCI+ AuCls. The author points out the fact that butylamin has the same boiling point with petinin, with which it is possibly identical.— Ann. de Chemie et de Physique, xlii, 129.

3. On the volumetric determination of Copper.-DE HAEN has succeeded in applying the general method of Bunsen to the determination of copper by means of the well known reaction 2CuO, SO3+2KI= Cu2l+2KO, SO3+I. The details of the process are as follows: The copper is to be brought into the form of sulphate, the solution being either neutral or containing only a moderate quantity of free sulphuric acid. The solution is then to be diluted in a measuring flask to a definite volume, so that 100 C. C. contain 1-2 gr. of oxyd of copper; 10-20 C. C. of Bunsen's solution of KI are to be introduced into a large beaker glass; 10 C. C. of the solution of copper are to be added and the whole mixed; the solution of sulphurous acid is then to be added, and the operation finished by Bunsen's process. The presence of free nitric, chlorhydric and acetic acids, as well as of oxyd of iron and all other substances which decompose iodid of potassium must be avoided. The sulphurous acid must also be added without loss of time and the solutions must not be too dilute. If these conditions be observed a very accurate result is obtained.—Ann. der Chemie und Pharmacie, xci, 237. [It will be observed that this method fails entirely precisely in the case in which it is most desirable to have a very accurate and expeditious process for determining copper, namely, in the assay of copper ores. -W. G.]

4. On the action of iodid of amyl upon an alloy of sodium and tin.— GRIMM has studied and described an extensive series of new radicals containing tin and the elements of amyl, and obtained by the action of iodid of amyl upon an alloy of tin and sodium. As however, these

radicals do not differ essentially in properties from those already described by Löwig and other chemists, we shall content ourselves with giving their formulas and names, Am being employed as a symbol for anyl.

It will be observed that these.compounds correspond to those of ethyl and tin, but that while the amyl compound corresponding to ethstannethyl Sn4 Ams is wanting a radical is here met with to which the ethyl series affords no parallel, namely, Sn2 Am4.-Journal für practische Chemie, 62, 34.

5. New organic radicals containing arsenic.-CAHOURS and BRELIE have still further extended our knowledge of this very interesting class of bodies. By the action of iodid of methyl upon arseniuret of sodium two radicals are obtained, one corresponding to stibmethyl, the other the iodid of an arsenic ammonium which the authors term arsenmethylium. This iodid is represented by (C2H3)4 As. I, and crystallizes in magnificent brilliant tables. By double decomposition with the salts of silver other salts may be readily obtained. The same radical is also obtained when iodid of methyl is brought into contact with cacodyl, the action being represented by the equation 2C2H31+2C4H6 As (C2H3)4 As.1+C4H6 AsI. With the iodids of ethyl and amyl similar results are obtained, and the iodids of two new arsenic ammoniums are formed having the formulas (C2H3)2(C4 Hs)2As. I and (C2H3)2 (C10H11)2As. I. These results in connection with those of Landolt already mentioned in this Journal, leave no doubt as to the constitution of cacodyl which must be regarded as Dimethyl-arsenic.-Comptes Rendus, xxxix, 541.

[Note.—In connection with this subject the writer desires to draw the attention of chemists and of medical men to the probable advantages of employing the nitrogen, antimony and arsenic ammoniums in place of quinine as antiperiodic therapeutic agents. It will be remembered that the compound ammoniums of Hoffinann, tetramethyl-ammonium N(C2H3) for instance, possess in their saline compounds an intensely bitter taste. Now as quinine in combination is also an ammonium, NR4, it appears probable that the bitter taste is in some measure at least, characteristic of the type R' R4 of the ammoniums. This appears to be the case with the antimony and arsenic ammoniums so far studied, and I therefore suggest that careful experiments should be made to determine whether these compound ammoniums which can readily be prepared in the laboratory and at moderate prices may not answer in intermittent fevers, &c. as well as the expensive salts of quinine.W. G.]

6. Action of iodid of phosphorus upon glycerine.-BERTHELOT and de LUCA have observed that when crystallized iodid of phosphorus Pl2 is distilled with glycerine propylene gas is evolved, while water and iodated propylene CeHsI distill over. The proportions of these products