thought that, notwithstanding the removal of certain landmarks (a fence and tree), in the field, he would still be able to locate it very closely. This he subsequently undertook to do by trenching, but was unsuccessful in finding the mass. I was, myself, present and assisted in a further search for it in September last, using a surveyor's magnetic needle, with the hopes of the same being attracted to the mass and discovering it, but all to no purpose. Mr. Kyser seems to feel very confident of his knowledge of the immediate vicinity of the mass where he buried it 22 years ago, but is unable to prove its presence by re-discovery. Nor has he the aid of another eye-witness, his father having died soon after the original finding and burying as above mentioned.

The smaller piece which was, as before said, found in 1883,

1.

was presented by Mr. Kyser, Sr., to Mr. W. S. Adams who, at that time. kept a plow factory in the city of Plymouth. It was retained in their family until last November, when it was brought to Ward's Natural Science Establishment in Rochester, N. Y. by Mrs. Adams from whom I procured it.

The mass, as represented in the accompanying cut (fig. 1) is a lengthened, tongue-like form, not unlike a rude Mound-builder's axe. Its greatest length is 12 inches; its width 79 inches; its thickness in the middle about two inches,

Its surface is deeply eroded by oxidization, so that, although sound and free from scales, it shows no signs of an original crust. The characteristic pittings of meteorites are also by the same cause rendered somewhat feeble, although still quite clearly visible. We have cut a number of thin slices from the mass. These etched in dilute nitric acid give very clear Widmanstätten figures, which are well shown in the accompanying cut (fig. 2.) There are, further, several small nodules of

A careful analysis of this iron has been very kindly made. for me by Mr. J. M. Davison of the Reynolds Laboratory of the University of Rochester, and I give the same below.

This iron, herein briefly noticed, is interesting in many ways, and it is much to be regretted that the large mass, of which the record seems to me to be entirely reliable, cannot be re-discovered.

2

SCIENTIFIC INTELLIGENCE.

I. CHEMISTRY AND PHYSICS.

2

1. On Diammonium.-The suggestion of CURTIUS, the discoverer of hydrazine or diamide, that there should be a hypothetical radical diammonium which bears the same relation to diamide that ammonium does to ammonia, has been verified by this chemist in connection with SCHRADER; they having prepared a large number of double salts containing this diammonium radical. Diamide itself HN-NH, is extremely unstable and its separate existence is yet somewhat uncertain; while its hydrate NH. HO is very permanent. Moreover, diamide is also unlike ammonia in the fact that it is a diacid base while ammonia is monacid. The normal hydrazine chloride is CIH,N-NH,Cl and the sulphate is (H,N-NH,)"SO,. The analogy thus shown between diammonium and the bivalent metals of the alkali-earths, is further strengthened by the sparing solubility of its sulphate and its inability to form alums with the sulphates of the alumina group. On the other hand, however, certain properties of diammonium show that it resembles closely the alkali metals. Thus its hydrate generally acts as a monacid base. Its chloride. NII, Cl, is decomposed below 100° into hydrogen chloride and NH. HCl, which cannot be made to lose more hydrogen chloride without destruction of the base. The hydrate N,H,.(HO), can exist only in solution. On evaporation it passes into the hydrate NH. HO, which boils without decomposing. Ammonia gas displaces only half the acid of the sulphate NH,. H,SO, leaving the stable sulphate (N,H,),H,SO,. Moreover only one nitrate NH.HNO, appears to exist and only one thiocyanate N,H ̧. SONII. The authors conclude that diammonium may act as a univalent radical (N,H)" and also as a bivalent one (NH)", the salts of the former being the more stable. Double salts of diammonium sulphate with the sulphates of copper, nickel, cobalt iron, manganese, cadmium and zinc have been obtained, all of which are anhydrous. They are readily thrown down as precipitates on mixing strong solutions of the metallic sulphate and diammonium sulphate. The latter may be either the sulphate NHSO, or (NH) SO,. While the former of these sulphates is difficultly soluble the latter is deliquescent; yet the sparingly soluble double salts always contain the latter sulphate. Moreover salts of the type RSO,. (N ̧H ̧), and R”SO ̧ . (NH), have been obtained, the former containing zinc or cadmium, the latter nickel or cobalt. All attempts to obtain alums containing the diammonium sulphate N,H,. SO, have been unsuccessful.-J. prakt. Ch., II, 1, 311, September, 1894.

2

4

G. F. B.

2. On Nitrogen Trioxide.-The actual existence of the trioxide of nitrogen or nitrous anhydride N,O,, appears to have been established by LUNGE and PORSCHNEW. Although the oxides

NO, and NO, show scarcely any tendency to unite at the ordinary temperature, the authors find that at -21°, these oxides combine in exact molecular proportions to form NO, which condenses as an indigo-blue liquid; 98.3 per cent of the trioxide having been obtained in one experiment. It is perfectly stable at and below this temperature; but at a slightly higher temperature, even under pressure, it begins to decompose, the dissociation becoming almost complete on conversion of the liquid into gas. Hence it would seem that the trioxide is not capable of existing in the gaseous state; although certain facts observed in the investigation seem to indicate that a residue of NO, molecules may escape dissociation and may exist side by side with the molecules of N,O, and N,O, into which the trioxide is decomposed.-Zeitschr. anorg. Chem., vii, 209, September, 1894.

2

4

G. F. B.

3. On the Physical Properties of Nitrogen Monoxide.-The physical properties of carefully purified nitrogen monoxide have been studied by VILLARD. The gas was obtained pure either by the decomposition of its hydrate or by fractioning the liquid oxide; the gas in the latter case being passed through suitable purifying and drying agents. It was then liquefied and allowed to boil to expel the dissolved gases. As thus obtained, it is free from the less liquefiable gases, its maximum vapor pressure is independent of the volume of the vapor, and a small increase of pressure causes complete liquefaction. The densities of the gas and of the liquid at various temperatures are given as follows:

Temperature ..
Density of liquid..
Density of gas....

0° 5°

10° 17.5° 26.5° 32.9° 34.9° 36.3° 0.9105 0885 0.856 0.804 0.720 0640 0-605 0.572 0-0870 0:099 0114 0146 0207 0.275 0·305 0:338

The critical temperature of pure N,O is 38.8°, the critical volume is 0.00436, the critical density 0-454 and the critical pressure 77.5 atmospheres.-C. R., exviii, 1096, May, 1894.

G. F. B.

4. On the use of the Refractive Index for determining Critical Temperatures.-By a careful observation of the interference bands, CHAPPUIS has been able to note the changes which take place in the index of refraction of a liquefied gas in the vicinity of its critical temperature. For this purpose the liquid was contained in a cylindrical cavity in a steel prism, having apertures closed by optically plane and parallel glass plates; the whole being immersed in a liquid whose temperature could be maintained constant. By means of a pair of Jamin mirrors, two beams of light, starting from a Billet compensator, pass through the liquid in the prism, traversing in their course the enveloping bath, the sides of which also are made of plane parallel glass. When liquid carbon dioxide is used, the refractive index is constant and the bands remain stationary; beyond this temperature the refractive index increases rapidly and the bands fall. At 31-61° the curve of the index shows a vertical tangent; and the intersection of this curve with the straight line which represents the index above this temperature, is the critical point of the

index. The uncorrected results vary between 31.60° and 31·62° only, and the corrected value for the critical point is given as 3140°; in close agreement with Amagat's value 31·35°. — C. R., exviii, 976, April, 1894.

G. F. B.

5. On the Constants of Refraction of Carbonyl Compounds. -In order to ascertain the influence exerted by the presence of a number of carbonyl groups in a molecule upon its molecular refraction, NASINI and ANDERLINI have determined the constants of refraction of several carbonyl compounds. Mond and Nasini had attributed the exceptionally high molecular refraction of nickel tetracarbonyl to the fact that in this substance nickel is an octad. But Gladstone had expressed the opinion that the equally high molecular refraction of ferropentacarbonyl, is due rather to the peculiar arrangement of the carbonyl groups than to the presence of iron as a decad. The substances examined were quinone, diacetyl, dipropionyl, tetrachlorotetraketohexamethylene, dibromodichlorotetraketohexamethylene, leuconic acid and potassium croconate; the last two being examined in solution in water, and the others (excepting the second and third) in solution in benzene. The measurements were made for the line Ha and the results calculated both for the formula M (μ-1)/6 and the formula M(-1)/(-2)6. It was observed that the experimental values agreed well with the calculated ones. If, however, the atomic refraction of potassium be taken as 81 as given by Gladstone, the observed molecular refraction of potassium croconate becomes 58.20 for the first formula given above; so that if the ordinary constant be assigned to the carbonyl groups, the atomic refraction of potassium must be taken as 22.5 for the first, and 12:40 for the second, of these formulas. Since the measurements made on the other carbonyl compounds show that the presence of several carbonyl groups causes no abnormal increase in the molecular refraction, the authors attribute the abnormally high molecular refraction of this substance to the metal alone; though they admit that the anomalous results sometimes obtained with such compounds may be due in part to the fact that they were made in solution.-Gazzetta Chim. Ital., xxiv, i, 157; J. Chem. Soc., lxvi, ii, 301, August, 1894.

G. F. B.

6. On the Electrolysis of Copper Sulphate in Vacuo.—It is a well recognized fact that the electrolytic deposition of copper from a solution of its sulphate does not conform rigorously to Faraday's law; Gray having shown that the deposit of metal is heavier the higher the current density and the lower the temperature. This appears to be due to the fact that copper is slightly soluble in a copper sulphate solution. Since Schuster had suggested that this solution of the copper is due to the oxygen present in the copper sulphate solution, GANNON has made a comparison between the masses of copper deposited in two voltameters in series, from one of which the air was exhausted. His results show that with a neutral solution, the deposit of copper in the vacuum tube is higher than that in the one under the atmos