understood that the methods which have answered in the instances already known are not likely to answer in the present case. Kakodyle was obtained from chloride of kakodyle by driving out kakodyle by means of zinc. Supposing, however, that I had got the chloride of ethylene-sodium, what metal could I expect to be capable of driving out ethylene-sodium?

Zincethyle was obtained by the distillation of the double zinccompound of iodine and ethyle, when, as is well-known, zincethyle distils over. The non-volatility of ethylene-sodium is a bar to the establishment of a parallel process.

The only hope which I have at present of isolating the new radical is the hope of finding it among the products of the destructive distillation of the double compound of sodium-ethyle and zinc-ethyle, not, however, in the distillate, but in the residue, along with the finely divided zinc and sodium which results from the destructive distillation in question.

With respect to the chemical constitution of ethylene-sodium, there are two modes of representation which will commend themselves to the chemical mind, viz. :—

((C2 H4)" Na')', and ((C2 H4)" Na"")'.

Against the first, and in favour of the second formula (that to which I have given the preference) may be urged the fact that sodium appears to be a trivalent metal; and also that the first method of representation would necessitate the assumption that in the hydrated oxide there existed oxygen not in direct union with sodium, whilst the second formula represents the sodium as directly combined with oxygen. Another reason for giving the preference to the second formula will be mentioned on a future

occasion.

In support of the statement that sodium is a tri-valent metal, a statement which will be looked upon as a chemical heresy in certain quarters, I would bring forward the cases of the double zinc-sodium-ethyle and of sodium-triacetyle.

The analysis of the first of these compounds led to the empirical formula (Na C2 H5+Zn (C2H5)2), as will be seen on reference to my paper on the subject. It will also be remembered that I altogether failed to effect a separation between the sodium-ethyle and the zinc-ethyle. The real constitution of the compound is :

Na""-C2H5

C2 H5
Zn"-C2 H5

* Na=23, Zn=65.

Sodium-triacetyle, as I pointed out at the last Meeting of the British Association (vide also the January Number of Liebig's Annalen), is obtained by the action of sodium on acetic ether. Although its formula may be represented in another way, still the most elegant representation is Na""(C2 H3 O).

In the common sodium-salts I regard the sodium as having united with itself; thus common salt is looked upon as being

Na""-Cl

Na"-Cl

In fine, I regard sodium as being an analogue of nitrogen and arsenic rather than of hydrogen.

Hydrated Oxide of Ethylene-sodium.

This compound was described in my last paper under the name of the absolute ethylate of sodium, and is obtained by heating the well-known crystals which are the product of the action of sodium on alcohol. It is also formed by the action of sodium on the ethyle-ethers of the fatty acids.

It is a snow-white amorphous solid, non-fusible, and of remarkably low specific gravity. There are difficulties in the way of taking its specific gravity with great precision. It appears to be lighter than ether, in which it swims. There is just the possibility that this extreme lowness of specific gravity may be to some extent simulated, and that the floating in ether may be due to adherent gas (hydrogen).

Whether or not the specific gravity is lower than that of ether must be determined by further experiment; but that the specific gravity does not exceed that of water has been shown by a determination.

As has already been described, this substance possesses the property of withstanding a very high temperature without decomposition. It will bear being heated to 290° C.; possibly it will bear a much higher temperature; but, as might have been expected, it will not bear a low red heat without carbonizing."

In contact with excess of water, it gives caustic soda and alcohol, the latter, as I showed in my last paper, being obtained in the theoretical quantity from product which had undergone heating to 200° C. Heated with an insufficient quantity of water to convert all of it into caustic soda and alcohol, the rcaction is still of the same kind, a quantity of water liberating an equivalent of alcohol, thus:

2.

C2 H4 Na
H

>O +2 H2 O = 2 C2 Ho O + Na2 (HO)2.

Phil. Mag. S. 4. Vol. 37. No. 248. Mar. 1869.

N

The numbers given on experiment were :—

Quantity of water added to a weighed
quantity of the substance

=1.00 grm.

=2.309 grms.

Therefore by experiment, ratio of water to alcohol produced

Theory requires

=1.00: 2·309.

1.00 2.55.

This experiment shows very plainly that there is no formation of anhydrous soda. If Na2O were formed, double the quantity of alcohol should have been given. Furthermore, experiment with alcohol slightly moist, i. e. containing 3 per cent. of water, has shown that the new compound is very sensitive to the presence of traces of water, the smallest quantity of water effecting an equivalent decomposition according to the equation just given.

With iodide of ethyle:-It is well known that an alcoholic solution of potash or soda, as also the crystals containing alcohol and ethylate of sodium, effects an easy decomposition of iodide of ethyle into ether and iodide of the alkali-metal. The hydrate of ethylene-sodium behaves in a very different manner. I took 5.849 grms. of the hydrate and 10 cubic centims. of iodide of ethyle and heated very gradually up to 210° C. in a small retort open to the air; 9 cubic centims. of iodide of ethyle distilled over, and were condensed and measured. The product was subsequently weighed and found to be 5.948 grms., showing that there had been no sensible action, the small difference of 0.1 grm. being only that which would be occasioned by the difference of weight between the apparatus partially filled with vapour of iodide of ethyle and the apparatus filled with air. In this experiment, although the temperature of the oil-bath was 210° C., still the temperature of the interior of the retort must have kept very low, owing to the evaporation of the iodide of ethyle until near the end of the experiment. The conclusion to be drawn is that there is no rapid action between iodide of ethyle and hydrated oxide of ethylene-sodium at moderate temperatures. Under similar circumstances, either the crystals or alcoholic solution of potash or soda would have decomposed a very considerable quantity of iodide of ethyle.

When iodide of ethyle and the hydrated oxide of ethylenesodium are sealed up together and heated up to between 120° C. and 150° C. for an hour or two, there is very complete decomposition.

1.25 grm. sodium was dissolved in alcohol, and the product dried at 196° C., 3 cubic centims. of iodide of ethyle added to it and sealed up with it. The whole was then heated to from 120° C. to 150° C. for about two hours. On opening the tube there was a slight escape of gas. Water was then added to the product, whereupon about 2 cubic centims. of light oil separated. This oil proved to be common ether, boiling at 36° C., and boiling to dryness at 40° C. The aqueous liquid was distilled, and the distillate redistilled. To the second distillate a little dry carbonate of potash was added, whereupon there formed two layers, viz. an aqueous layer of carbonate and an oil. This oil had the smell of the secondary alcohols, and was, I suppose, secondary butylic alcohol*.

With the ethers of the fatty acids there is produced alcohol and a salt of ethylene-sodium, e. g.

A reaction of this kind has been observed with acetate of ethyle, acetate of amyle, valerianate of ethyle, and benzoate of ethyle. An analogous reaction has also been observed between the hydrated oxide of amylene-sodium and valerianate of amyle. There is every reason to believe that this reaction is perfectly general, applicable to the hydrated oxides of all the olefine compounds of sodium or potassium, and all the fatty and aromatic ethers.

Acetate of Ethylene-sodium.

As just mentioned, this salt is produced by the action of acetic ether on the hydrated oxide of ethylene-sodium. The reaction appears to take place slowly even at 100° C. At higher

Should further experiment confirm the indication here given, a very important step in organic synthesis will have been taken-from ethylic to isobutylic alcohol; and in like manner a regular ascent of the series should be practicable.

f Owing to the occurrence of this reaction at 100° C., the attempts made by various chemists to render acetic ether free from any alcohol with which it might be contaminated by distilling it off sodium, have had the reverse result, and, continually producing alcohol, have rendered the ether more impure. By the action of sodium on acetic ether there is produced sodium-triacetyle and hydrated oxide of ethylene-sodium, which latter attacks excess of acetic ether slowly at water-bath temperatures. A sample of acetic ether which, from the mode of its preparation, must have been originally almost, if not quite, free from alcohol, became, after several distillations off a small quantity of metallic sodium, charged with alcohol to the extent of 15 per cent. This circumstance, which really no one can be blamed for not expecting and providing against, has probably led Geuther, Frankland, and Duppa astray.

temperatures it proceeds more rapidly. At temperatures approaching to 200° C. it is rapid.

The most satisfactory way of preparing the compound appears to consist in sealing up hydrated oxide of ethylene-sodium with twice its weight of pure acetic ether, and heating for some time to 150° C. (at which temperature the materials form a clear fluid), and afterwards opening the tube and distilling off the alcohol and excess of acetic ether in the oil-bath.

It is very easy to effect a partial transformation of the hydrated oxide into the acetate, but not easy to complete the

reaction.

Acetate of ethylene-sodium is a white solid, non-fusible at 200° C., but apparently readily soluble at 150° C. in acetic ether. Its characteristic reaction is the giving of alcohol and acetate of soda* by action on water.

(C2 H+ Na) } O+
(CHINA)}O

C2

= Na2 (C2 H3 O2)2+2 C2 H6 0.

It is isomeric with butyrate of soda. The circumstance of its furnishing alcohol and acetate of soda with water shows that the ethylene is associated with the metal and not with the acid part of the salt.

With iodide of ethyle it reacts in a very interesting manner. It combines with the iodide in the first instance, forming a solid. On the addition of water to this solid (which dissolves in the water, forming a solution of iodide of sodium), there separates a considerable quantity of oily liquid. This liquid does not contain any sensible quantity of common ether. I am engaged in the investigation of it.

Valerianate of Ethylene-sodium.

Preparation.-Hydrated oxide of ethylene-sodium and pure valerianate of ethyle are heated in a retort placed in the oil-bath. Owing to the high boiling-point of valerianate of ethyle, it is unnecessary to make a preliminary digestion in sealed tubes, as in the instance of the acetate just described. The temperature of the oil-bath is to be gradually raised up to the boilingpoint of valerianate of ethyle, and then taken up gradually to 200° C. During this operation abundance of alcohol has been

* The fact that this substance really does give alcohol and an acetate, and not the salt of any higher fatty acid, was established by treating some of it with water, and then distilling off the alcohol and estimating it in the distillate, and subsequently rendering the residue acid with dilute sulphuric acid and distilling. The distillate was then saturated with carbonate of baryta, and the resulting baryta-salt_analyzed. It contained 53.34 per cent. of barium. Acetate contains 53.73.