mention this in order to make up for Professor Norton's silence on this point.

The sixth and last answer by which the learned Profes sor strives to weaken the force of my objection is an explanation of the words "spherical in form" applied by him to his atom of gross matter. To say that "all bodies of matter consist of separate indivisible parts called atoms, each of which is conceived to be spherical in form," as Professor Norton says in his 3rd principle, was virtually to say that such atoms were pieces of continuous matter. Such at least was my impression. But he answers:

"The assumption that each atom is spherical in form was adopted merely as the simplest embodiment of the fundamental principles that the action of the atom was equal in all directions, and that the attractive action upon an atom of æther was neutralized at minute distances by the resistance developed at the point of contact. The existence of such a resistance necessarily implies that the elementary parts of the attractive atom, whether finite or infinite in number, act repulsively at very minute distances."

I have already allowed that, when Professor Norton himself explains the meaning attached by him to his own words, I am not entitled to contradict him. It is strange, however, that the expression "spherical in form" which is drawn from geometry, and conveys the clear notion of something geometrical only, should have been in need of an interpretation drawn from mechanical considerations. However this anomaly may be explained, let us take notice first that Professor Norton, in giving this interpretation, reveals to us a new "fundamental principle. The principle is this: "The attractive action upon an atom of æther is neutralized at minute distances by the resistance developed at the point of contact." Is this principle true? I think not.

I have rigorously proved in my 'Molecular Mechanics' (I quote my own book for the excellent reason that no other book to my knowledge has yet appeared in which the same subject has been regularly and philosophically developed) that in the true and immediate contact of matter with matter no action is possible (pp. 14, 15). So long as this theorem holds good, I cannot admit that any resistance is developed "at the point of contact" of two atoms. Moreover, if the attractive action of the so-called "gross matter upon an atom of æther is neutralized "at minute distances," surely repulsion must prevail at such minute distances: but when two atoms are at a minute distance, they are not in contact; and therefore, if the attractive action is neutralized at minute distances, the resistance developes before the two atoms reach the point of contact and

therefore the new "fundamental principle," to say the least, is incorrect.

But again, if, according to another view of the learned Professor already noticed, the intensity of action "becomes indefinitely small at indefinitely small distances," we must come to the conclusion that, according to the same view, the intensity of action at the very point of contact will become null. Therefore, if that view is adopted, no resistance will be developed at the point of contact, and the "fundamental principle" will be false, at least hypothetically. Such is the accuracy with which some physicists set down what they call "fundamental principles" and "established truths."

a

Yet, after all, if an atom of gross matter is more than a material point, the assumption that each atom is spherical in form cannot be the mere embodiment of mechanical principles. An atom which is more than a material point, and possesses size" however inappreciable in comparison with the atomic distances, must have a surface: and this surface must have a geometric form either regular or irregular. If it be spherical in form, then it would seem that Professor Norton has tried in vain to discard my geometrical interpretation of his words: whilst, if the geometrical form is not spherical, Professor Norton's own interpretation collapses; as the action of the atom cannot be conceived "equal in all directions," unless the form of the atom itself be uniformly equal all around, viz. unless the atom be a sphere. The learned author has one means only of avoiding the horns of the dilemma, viz. by allowing that his atoms are systems of discrete material points; his interpretation of the words "spherical in form " will then be substantially correct, though rather unusual; and his theory of molecular physics, disembarrassed of gross matter and its difficulties, without losing anything worth regretting, will then be able to recommend itself more strongly to a philosophical mind.

In the passage now under examination Professor Norton endeavours also to establish "a change of attractive into repulsive action at very minute distances." As I have fully refuted this view, and the arguments by which Boscovich strove to defend it, in my Molecular Mechanics,' I will now only say that the inference drawn by Professor Norton is not legitimate. The existence of a resistance between his atom of gross matter and the atom of æther does not necessarily imply that the elementary parts of the attractive atom act repulsively at very minute distances. It implies simply that the so-called gross matter is a dynamical system of elements of which some are attractive and others repulsive, the attractive always attracting, the repulsive always repelling, and the effect of their exertions being a re

sultant attractive or repulsive according as the atoms acted on are supposed to be placed beyond or within the limits of their molecular distance of equilibrium. This is the only inference that can be drawn legitimately from the impenetrability of molecules.

I might dispense with all remarks on what the writer adds about a conception which he himself as yet hesitates to adopt. The idea however is calculated by its novelty and brilliancy to fascinate a mind devoted to physical speculation, and deserves a short notice. The author says:

"But another conception may be formed of the mode of operation of an atom of gross matter, which involves no other supposition than that it acts equally outwards in all directions from a centre, and takes no account of its geometrical extent. This is that the effective attraction of the atom for the ather of space is due to the existence of a repulsion less than would be exerted by the one or more atoms of ather that would naturally occupy its place. The result would be the condensation of an atmosphere of æther around the atom, without the exertion of any direct attractive force, or of any additional force of resistance. We may conceive the molecular atmosphere of electric æther to originate in a similar way; but as the opportunity of examining and testing this idea sufficiently has not yet been obtained, I shall continue to regard the electric æther as directly attracted by the atom of gross matter, and that the antagonistic force of resistance is furnished by the repulsion of the luminiferous æther condensed around the atom."

This conception, however plausible it may be, is exposed to many serious objections, which however I am not ready to treat in this paper, as I must confine myself to the questions already raised. The least that I can say of this new theory is that it is quite unnecessary, and that, no matter how much talent may be spent in building it, it will never be more than an à priori assumption; for in the whole multitude and variety of natural facts nothing has yet been found which can serve as a basis for its future demonstration. Professor Norton himself says that the opportunity of examining and testing this idea sufficiently has not yet been obtained: I rather think that the idea has not even begun to be tested, and never will, unless the question be of testing its inadmissibility. For, though there are no facts in nature supporting the hypothesis, there are facts strongly contradicting it, as for instance molecular cohesion and gravitation. Moreover, this new hypothesis would not have for its result "the condensation of an atmosphere of æther arcund the atom without the exertion of any direct attractive force," as assumed by Professor Norton. The hypothesis that the atom of gross matter repels less than the æther which would naturally occupy its

place, would lead us to this result only, that the atoms of æther would from every side approach nearer the atom of gross matter, without however becoming closer amongst themselves, that is, without condensation. But, as it is not my intention to discuss an incidental question about which Professor Norton has not yet formed a definite opinion, I will say no more on this subject.

I trust that the reader in the preceding pages will have found sufficient evidence as to whether my criticism on Professor Norton's theory was well founded or not. It only remains for me to answer the objections which he advances against some views put forward in my Elements of Molecular Mechanics.' When this has been done, I shall consider the present controversy as closed.

4

[To be continued.]

L. On some Reactions of Hydrated Oxide of Ethylene-sodium. By J. ALFRED WANKLYN, Professor of Chemistry in the London Institution*.

AS

S described in former papers †, hydrated oxide of ethylenesodium is obtained by allowing metallic sodium to act on ten times its weight of perfectly absolute alcohol and heating the product to rather over 200° C., maintaining it at that temperature so long as alcohol distils off. In this manner a perfectly white product may be obtained having accurately the composition Na C2H5O. This substance is hydrated oxide of ethylenesodium. It is characterized by its reaction with the ethers of the fatty acids and the ether of benzoic acid, with which it gives alcohol and a salt of ethylene-sodium, thus:

It has already been explained that I do not regard sodium as mono-valent, but as tri-valent in these compounds, the radical ethylene-sodium being

((C2 H4)" Na")'.

The following reactions, which I have lately studied, tend in favour of this view of the atomicity of sodium.

Sulphuretted Hydrogen and Hydrated Oxide of Ethylene-sodium. Experiment I. 1.3815 grm. of sodium was dissolved in 15 grms. of absolute alcohol, and the product heated in the oil-bath up to * Communicated by the Author.

† Phil. Mag. vol. xxxvii. pp. 117 & 175.

above 200° C. The product was then cooled and weighed. Its weight was 4-4645 grms., being a little above the weight of the product completely freed from alcohol. (The theoretical quantity of Na C2 H5 O is 4.0844.) It was not considered necessary to drive off the last trace of alcohol in this instance. A current of dry sulphuretted hydrogen was next transmitted over the product, no external heat being applied, but considerable heat being generated by the action of the gas on the substance. After a while the passage of the gas was stopped, and the apparatus with its contents cooled and subsequently weighed. Weight of contents=5.76grms. The gas was again transmitted, and again generated heat by its action on the substance. Again the apparatus was cooled and weighed. Weight of contents =6·175 grms. Again sulphuretted hydrogen was transmitted, this time there being no generation of heat. Weighed again: weight 6.137 grms., showing that the action of the sulphuretted hydrogen was complete.

=

This experiment indicates that one molecule of sulphuretted hydrogen combines with one molecule of hydrated oxide of ethylene-sodium.

Na" C2H4

LOH+H2S=Na C2 H7 SO.

By experiment, 1:3815 grm. of sodium have yielded 6.137 grms. of product; therefore the percentage of sodium in this product equals 22:51. The theoretical percentage of sodium in Na C2 H7 SO is 22.55.

The compound Na C2 H7 S O, a nearly white solid, the formula of which I write thus,

ro C2 H5

Na"" H
SH,

is endowed with considerable stability, having during the process of its formation undergone a considerable spontaneous heating without damage. At 100° C., however, it is gradually decomposed into alcohol and sulph-hydrate of sodium, thus:

The 6-137 grms. of product, on being heated to 100° C. in the water-bath for some time, lost alcohol (2 grms. of which was condensed and weighed), 3.901 grms. of solid residue remaining. This residue was lastly heated up to 200° C. for some time, when it lost more alcohol and ultimately weighed 3.459 grms. Calculating the percentage of sodium in the 3.459 grms. of pro