have here put down the lowest point at which each of the salts tried has actually been observed to remain liquid without any special precautions being taken to retard the crystallization. With some few of the nitrates it makes a difference in this respect whether they are moderately or strongly heated. If any one of them is heated but little more than is necessary to effect its fusion, it will begin to solidify before it gets many degrees below the melting point. But when the nitrates of lime, iron and chrome are brought nearly up to the boiling point, they can be cooled very low before they begin to shoot. Considering the difference in color between the solid and the liquid nitrates of iron, chrome and cobalt, there seems to be no special absurdity in supposing that some strongly heated nitrates may have to overcome a reluctance to change of state as well as one to change of form, and are therefore slower in beginning to show crystals.

The composition has in each instance been determined anew, either by simple ignition or by drying down with an excess of sulphuric acid, and so there is little room for error. It appears that a greater uniformity prevails among the nitrates than among any other salts. In all but four of those examined there are either six or three equivalents of water to each equivalent of nitric acid. In only two cases, has the same base been found capable of forming two different crystallized nitrates.

SEXHYDRATES.

Nitrate of Magnesia, Mg,.-This salt melts at 194° F. The liquid has been cooled to 188°. It boils at 290°.

When the heating is continued, the salt remains liquid and clear till about five equivalents of water and a little of the acid are expelled. The residue is not entirely soluble. It becomes hot in recombining with water.

Nitrate of Zinc, Zn H.-Melts at 971° F. It has been cooled in the liquid form to 87°. It boils at 268°.

Some of the melted crystals, on continued boiling, remained thin and clear till 42 p. c. of the weight was gone. The residue hardened to a vitreous mass on cooling, which had a composition not far from Zn, 3H,. This substance did not heat much when treated with water; but when some crystals were boiled till about four equivalents of water passed off, the residue evolved considerable heat in recombining with water.

Nitrate of zinc cannot be heated long without becoming basic and partially insoluble in water.

Nitrate of Manganese, Mn NH.-Melts at 781° F. Some dry crystals liquefied in a stoppered bottle during the hot weather of June and remained melted till September, though the temperature was sometimes as low as 60°. It boils at 265°.

If the boiling is continued, decomposition soon commences and black oxyd of manganese is precipitated. This gradual formation of peroxyd is also effected by a long continued steam heat.

Some liquid nitrate at 70° F. was found to have a density of 1.8104, while the solid salt at 70° had the specific gravity 1.8199. Nitrate of Nickel, Ni.N.H.-Melts at 134° F. The liquid has been cooled to 115°. It boils at 278°.

When the boiling is continued the liquid remains clear till three equivalents of water are expelled. It then begins to thicken and parts with acid.

Nitrate of Cobalt, Co....-I had too little of this to determine accurately the melting and boiling points, but they differ little from those of the nickel salt.

Pernitrate of Iron, Fe+ 3NH.-Melts at 117° F. May remain liquid at 70°, after being strongly heated. It boils at 257° F.

The specific gravity of some in the liquid state at 70° F., was found to be 1.6712, while the same solidified and cooled to 70°, had a density of 1.6835.

Nitrate of Chrome, er+зH.-This salt melts at about 98° F. It has been cooled to 68°. It boils at 258°.

6.

Nitrate of Alumina, Al+3NH..-Melts at 163° F., can be cooled to 1471°, and boils at 273°.

Nitrate of Uranium, NH..-This beautiful salt melts at 139°. It may remain liquid at 115°. It begins to boil at 245°.

When the boiling was continued, the stuff remained thin and clear till about four equivalents of water and a little of the acid passed off. The residue gave with water a solution which was turbid at first but soon became clear. Some heat was evolved during the solution.

Nitrate of Copper.-When nitrate of copper crystallizes at a low temperature, it forms a pale blue salt having the composition Ou H. These crystals are not permanent in hot weather, for at 791° F. they break up into a liquid and crystals of the trihydrate. To make the whole liquid requires a heat above 100° F., and so the pale crystals have no definite solidifying point.*

TRIHYDRATES.

Nitrate of Copper, Cu H.-This is the formula of the crystals which form above 79° F. They have nearly the same shape as the sexhydrate sometimes assumes, but are deep blue and are permanent in every state of the air. The composition is erroneously given in some books as Cu,,-probably because the analysts took no pains to ascertain the dividing limit between the two salts, and tried a mixture.

A solution of nitrate of copper is sometimes sold, standing at 55° B. As a solution saturated at 50° F. has just this strength, it is not strange that the maker often finds his returned carboys broken by huge masses of pale crystals.

SECOND SERIES, VOL. XXVII, No. 79.—JAN., 1859.

The trihydrate melts at 238°. It has been cooled down to 224° before beginning to shoot. It boils at 338°.

If the boiling is continued, nitric acid immediately begins to pass off, and a green basic nitrate is deposited.

Nitrate of Lanthanum, La N,.-This was found to melt at 104° F., and was cooled to 70° without crystallizing immediately. It boiled at 258°. These figures, however, cannot be considered as exact, for the salt used for trial amounted to but 32 grams, and was not absolutely free from didymium and cerium.

Nitrate of Glucina, Be+ 3 NH3.-Melts at 140° F. and may be cooled as low as 85° before it begins to fix. It boils at 285°. Some boiled till the thermometer rose to 320°, gave off acid, but remained perfectly clear. When this residue was cooled to 61°, a crystal did not cause it to solidify, because it was too basic. But the addition of strong nitric acid, induced a rapid crystallization, the temperature rising to 142°.

When the salt was boiled not quite so long, the product could be made to solidify, but the resulting temperature was considerably lower. Dilution with a basic salt, has therefore the same effect on the melting point as dilution with water.

TETRAHYDRATES.

Nitrate of Strontia, Sr N.-Unlike any other hydrated nitrate, this salt crystallizes in the monometric system.

The composition of hydrated nitrate of strontia is always laid down in the books as Sr,. But this formula has no analogy in its favor, and having repeatedly tried good crystals formed at a low temperature I have invariably found but four equivalents of water. The nitrate crystallized above 75° F., is generally anhydrous, and that formed below 60° is hydrated, but between these temperatures there is no certainty. Thus a solution saturated at 84° F., while cooling down to 62°, deposited nothing but anhydrous crystals; and a solution saturated at 71°, by standing some hours where the thermometer did not get below 70°, gave only fully hydrated crystals.

The hydrated salt is resolved by heat into a liquid and the anhydrous nitrate. Even the hot weather of summer causes it to sweat, if kept in a close vessel. In dry air, it loses all its water by efflorescence.

Nitrate of Lime, Ĉa H.-Melts at 111° F. Some that was heated only to 124°, began to crystallize when it had cooled to 96°. After being heated to 153°, it remained liquid over night and got down to 574°. This salt boils at 270°. When the boiling is continued, the mass remains liquid and clear till about one third of the water passes off. Farther heating renders it anhydrous, with scarcely any loss of acid. This dry residue evolves a strong heat in recombining with water.

Nitrate of Cadmium, ċa N,.-This salt melts at 139° F. It has been cooled to 91° before beginning to crystallize. It boils at about 270°. On continued boiling it continues clear and thin till nearly three equivalents of water are gone. When all the water has passed off, a small portion of the remaining dry mass is insoluble.

3

Nitrate of Bismuth, which was formerly supposed to be a trihydrate, has been found more recently to have the anomalous composition Bi N, 10. In several trials of a pure nitrate dried over sulphuric acid, I have obtained, by ignition, 48 per cent of oxyd. This would make the quantity of water as near eleven as ten equivalents. It is barely possible then that nitrate of bismuth may be a combination of a trihydrate with a tetrahydrate and have for its true formula Bi N, H, +Bi N3 H12.

3

9

This salt is not deliquescent and does not effloresce, even when kept for a long time over sulphuric acid. It is insoluble, as a whole, in water, and does not melt clear. At 1631 F., it resolves itself into a clear liquid and one opaque solid. The mixture has been cooled to 155°, but on stirring it solidified again while the temperature rose to 1634°. Some of the liquid part decanted clear, formed on cooling a mass of crystals quite wet with acid and having altogether a composition not far from +2NH..

3 11

38

3

ART. IV. Further Observations on the Allotropic Modifications of Oxygen, and on the Compound Nature of Chlorine, Bromine, &c.; by Professor SCHÖNBEIN.*

THESE last six months I have been rather busily working on oxygen, and flatter myself not to have quite in vain maltreated my favorite; for I think I can now prove the correctness of that old idea of mine, according to which there are two kinds or allotropic modifications of active oxygen, standing to each other in the relation of to, i. e. that there is a positively-active and a negatively-active oxygen,—an ozone and an ant-ozone, which on being brought together neutralize each other into common or inactive oxygen according to the equation (+Ổ)+ (-8)=0.

The space allotted to a letter being so small, I cannot enter into the details of my late researches, and must confine myself to some general statements, which I hope, however, will give a clear notion of the nature of my recent doings. A paper will

* Addressed as a letter to Prof. Faraday, and communicated by him to the L., E. and D. Phil. Mag., xvi, 178.

before long be published in the Transactions of the Academy of Munich.

Ozonized oxygen, as produced from common oxygen by the electrical spark or phosphorus, is identical with that contained in a number of oxy-compounds, the principal ones of which are the oxyds of the precious metals, the peroxyds of manganese, lead, cobalt, nickel and bismuth,-permanganic, chromic and vanadic acids; and even the peroxyds of iron and copper may be numbered among them.

The whole of the oxygen of the oxyds of the precious metals exists in the ozonic state, whilst in the rest of the oxy-compounds named, only part of their oxygen is in that condition. I call that oxygen negatively-active, or ozone par excellence, and give it the sign -O on account of its electromotive bearing. Though generally disinclined to coin new terms, I think it convenient to denominate the whole class of the oxy-compounds containing -Ŏ"ozonids." There is another less numerous series of oxycompounds in which part of their oxygen exists in an opposite active state, i. e. as +0 or antozone, wherefore I have christened them "antozonids." This class is composed of the peroxyds of hydrogen, barium, strontium, and the rest of the alkaline metals; and on this occasion I must not omit to add, that what I have hitherto called ozonized oil of turpentine, ether, &c., contain their active oxygen in the +Ŏ state, and belong therefore to the class of the "antozonids."

о

Now, on bringing together (under proper circumstances) any ozonid with any antozonid, reciprocal catalysis results, the -Ŏ of the one and the +0 of the other neutralizing each other into O, which, as such, cannot be retained by the substances with which it had been previously associated in the -O or +Ŏ condition. The proximate cause of the mutual catalysis of so many oxy-compounds depends therefore upon the opposite states of the active oxygen contained in those compounds.

I will now give some details on the subject.

1. Free ozonized oxygen =(-O), and peroxyd of hydrogen =HO+(+Ỏ), or peroxyd of barium = gen BaO+(+0) (the latter suspended in water), on being shaken together destroy each other, HO+(+Ỏ) or BaO+(+Å) being reduced to HO or BaO, and +8 and -8 transformed into O.

=

2. Aqueous permanganic acid Mn2O2+5(-O), or a solution of permanganate of potash mixed with some dilute nitric acid is almost instantaneously discolored by peroxyd of hydrogen or