resinous; streak gray; color pitch-black: opaque; fracture uneven-subconchoidal; brittle.

B.B. it fuses with intumescence to a black slightly magnetic glass. Dissolves easily in hydrochloric acid.

The analyses gave the following results.

b. Allanite from near Eckhardt's Furnace, Berks county, Pa.In color, general appearance and its blowpipe reactions, resembles very much the allanite from Orange county.

H.=6; Sp. gr. (at 27° Cels.) = 3.825-3.831.

The analyses gave the following results:

It is found abundantly near Eckhardt's Furnace, Berks County, Pa., associated with quartz, zircon, mica and titaniferous magnetite.

c. Allanite from Bethlehem, Northampton county, Pa.-Massive; H.-5; Sp. grav. (at 16° Cels.) = 3.491; lustre resinous; color brownish black; streak gray; opaque; fracture subconchoidal.

B.B. it decrepitates slightly and fuses with intumescence to an iron black magnetic slag. Hydrochloric acid dissolves it readily. Occurs in a decomposed granite in flat pieces of not more than half an inch in thickness; their surfaces are covered with a crust of hydrated sesquioxyds of iron and cerium, etc., resulting from the decomposition of the allanite by the action of atmospheric

water.

=

In the allanite from Orange county, the ratio of oxygen of RO: R2O3: SiO3, is equal to 1:08:18, which is 5:4:9, corresponding with the formula: 5R,Si +4 Si, the allanites of Reading and Bethlehem give very near the ratio 1:1:2, corresponding with the formula R,Si+Si.

3

=

The slight variation in the Orange county allanites of which the ratio of oxygen was found to be 5:4:9 may be owing to small impurities of this mineral; I believe that, when pure, it has, like the two others, and like all allanites which have been examined with regard to the state of oxydation of the iron in them, the ratio 1:1:2, or the constitution of garnet. The quality of water in the allanites, resulting in all probability from a change, beginning in the composition of this mineral, as previously suggested by Prof. Rammelsberg, was found to vary from 1.19 per cent. to 3.01 per cent.; but if we take it into consideration, the composition of the Orange county allanite may be expressed by the formula 2(R ̧ Si+i)+; that of the allanites from Bethlehem and Eckhardt's Furnace by (R,Si+Si)+H.

3

13. Tungstates in North Carolina.

Tungstates having been found only at two or three localities in the United States, it was interesting to find them in North Carolina at two localities, viz: Dr. Cosby's mine, near Pioneer Mills, Cabarrus county, (a, b and c,) and at the Washington mine, Davidson county, (d).

a. Wolfram occurs in irregular lamellar masses in brown hematite and in a hydrate of the sesquioxyds of iron and manganese, which appears to be a result of the decomposition of spathic iron. It is associated with scheelite, tungstate of copper, and sulphate of baryta.

b. Scheelite is found in white, yellowish white and brownish crystals and crystalline masses. The crystals are rough and very imperfect, but on one I could observe the planes P. OP. P∞. They are sometimes more than half an inch in length. It occurs also in granular and compact masses.

c. Tungstate of Copper (? and Lime); a new mineral.?Amorphous; massive and pulverulent; sometimes (if massive) of the lustre of wax, but usually dull; color between siskin and pistachio green.

B.B. in a tube gives water and blackens; on charcoal it fuses with intumescence easily to an iron-black slag, containing globules of metallic copper; with fluxes it gives the reactions of copper and tungstic acid. Soluble in hydrochloric acid with separation of tungstic acid; the solution contains oxyd of copper and lime. I believe that the lime belongs to the constitution of this mineral, and is not owing to an admixture of Scheelite, with which it is associated, and that its composition is analogous to that of Volborthite, which it somewhat resembles, or a hydrated tungstate of copper and lime.

d. Scheeletine.-Only one lump of quartz, which had a few crystals of this very rare mineral upon it, was found at the Washington Mine, Davidson county, N. C.

The crystals are quadratic octahedra P, some also in combination with the plane o P. The planes sometimes being curved, give the crystals a barley shaped appearance. Color lavender blue and yellowish white; lustre pearly-subadamantine; brittle. B.B. with microcosmic salt in the reducing flame gives an azure-blue glass; with carbonate of soda upon charcoal metallic lead. It is associated with pyromorphite, brown blende, iron pyrites, etc.

14. Scorodite.

I am not aware that another locality has been observed in the U. S., except Edenville, N. Y. It occurs also, coating the cavities of quartz and brown hematite, associated with gray copper, copper and iron pyrites at Geo. Luderick's farm, Cabarrus county, N. C.

It is found there in aggregations of greenish white, brownish and leek-green crystals. Only rarely they are large enough to distinguish their form, which is a combination of the planes P.∞ P2.∞ P co.

15. Wavellite.

This is another mineral of which only a few localities are recorded in the United States. I found it at the Washington Mine, Davidson county, N. C., in a talcose slate in globular concretions with a radiated structure, associated with actinolite, galena, blende, iron pyrites, silver, etc.

CORRECTION. In the article, Owenite identical with Thuringite, this Journal, last volume, p. 411, for Schmiedeberg read Schmiedefeld.

Philadelphia, Oct. 4, 1854.

ART. IV. On the Diamagnetic Force; by Prof. TYNDALL.*

WITH regard to the character of diamagnetic force great diversity of opinion prevails. In Germany we have Weber affirming that diamagnetic bodies possess a polarity opposed to that of iron. Weber's countryman, Von Feilitsch, combats this opinion in a series of Memoirs recently published in Poggendorff's 'Annalen.' He affirms that diamagnetic bodies possess a polarity the same as that of iron; and endeavors to bring the phenomena into harmony with this view. In this country, on the contrary, we have Prof. Faraday, and it was believed, Prof. Thomson, neither of whom are prepared to admit the existence of any polarity whatever on the part of diamagnetic bodies. . These divergences were a sufficient proof of the difficulty of the subject, and the necessity of caution in dealing with it; the author, therefore, thought it well to commence with the fundamental phenomena, and ascending from them to the more complicated, to endeavor to obtain, by strict adherence to experiment, a clear insight as to the real nature of that force by which certain bodies are repelled by the poles of a magnet.

From an extensive series of experiments made with different bodies, and under the most diverse circumstances, the author selected a few which clearly exhibited the law according to which the repulsive force augments when the strength of the repelling magnet is increased. Were the repulsion of a diamagnetic body dependent on any constant property of the mass, then its repulsion must be simply proportional to the strength of the magnet; but it is proved by the concurrent testimony of experiments carried on in Germany, France, and England, that, for a wide range of magnetic power, the repulsive force increases as the square of the strength of the influencing magnet. This leads inevitably to the conclusion, that the repulsion of a diamagnetic body depends, not alone on the magnet operating upon it, but upon the joint action of the magnet and diamagnet. A piece of bismuth, for example, in presence of the magnet is thrown by the latter into a state of excitement, which varies as the magnetic strength varies, and in virtue of which the substance is repelled. The next question to be decided is, whether the state of excitement evoked by one pole, in a diamagnetic body, enables a pole of an opposite quality to repel it.

To decide this, two cores of soft iron were so bent, that the two semi-cylindrical ends of the cores could be placed close together, so as to form a single cylinder of the same diameter as that of the straight portions of the cores. The cores being placed in suitable helices could be so excited that the contiguous poles

* Proceedings of the British Association, 1854, Athenæum, Oct. 7.

were of the same or of opposite names. A bar of bismuth was freely suspended, so that both poles could act upon it simultaneously. When the cores were excited, so that the poles were alike, the bismuth was repelled; when the poles were of different names, the bismuth bar remained motionless; all action upon it was annulled. This experiment confirms those of Reich, and proves that the condition, whatever it may be, which is evoked by one magnetic pole is neutralized by the other, that each particular pole evokes a condition peculiar to itself;-and here we obtain the first glimpse of the dual nature of the force under consideration.

po

The next portion of the inquiry treated of the deportment of diamagnetic bodies when acted upon, first, by the magnet alone; secondly, by the electric current alone; and, thirdly, by the current and the magnet combined. When we speak of the deportment of bismuth in any one of the cases mentioned, no exact meaning can be attached to the phrase unless it be first strictly defined in what direction, as to the planes of crystallization, the mass has been cut. A bar of bismuth, in which the planes of principal cleavage are parallel to the length of the bar, and acted upon by the voltaic current alone, will set itself parallel to the current's direction. A bar, on the contrary, in which the planes of cleavage are transverse, will set itself at right angles to the current's direction, The former bar Prof. Tyndall calls a normal diamagnetic bar; the latter an abnormal one. The most perfect antithesis is observed in all cases between the deportment of the normal diamagnetic bar and a bar of soft iron; the forces which cause a deflexion of the former from right to left produce a deflexion of the latter from left to right. If the former take up sition of equilibrium from southwest to northeast the position taken by the latter will be from southeast to northwest; and throughout all the experiments the same opposition of action is exhibited. By mechanical means, an abnormal magnetic bar was obtained a bar which set its length at right angles to the line joining the poles. The abnormal diamagnetic bar shows throughout a deportment precisely antithetical to that of the normal magnetic one; but when we compare the normal magnetic with the abnormal diamagnetic, or the normal diamagnetic with the abnormal magnetic, the deportment is in all cases perfectly alike. It is evident, therefore, that unless the influence of structure be attended to, the greatest errors and the most inaccurate conclusions may be founded on the deportment both of magnetic and diamagnetic bodies in the magnetic field; but the thing which chiefly concerns us is the strong presumption which the experiments justify, that whatever be the nature of the influences evoked in magnetic bodies by the action of currents, or magnets, or of both combined, to an influence, of the same nature but anSECOND SERIES, Vol. XIX, No. 55.-Jan., 1855.

4