Nov. 18, 1886] 



NATURE 



65 



nature of the other radicle with which it is combined : the 

 radicles behave independently, and as if there were no combi- 

 nation between them. 



Nor i'i it only from a study of the volumes of salts in the 

 dissolved state that such results are obtained. Numerous deter- 

 minations of the extent to which the vapour-pressure, the freezing- 

 point, and the temperature of maximum density of water is 

 influenced by the presence of various salts in it, have been made 

 by Wiillner, Blagden, Dufour, Depretz, Riidorff, and De Coppet,' 

 with the general results that certain hydrates of the salt are in 

 some cases present, and in others the salt is anhydrous ; but 

 these conclusions, which would tell more against the hydrate 

 theory than for it, are eminently unsatisfactory. The whole 

 question, however, has been re-opened by Raoult (Ann. Chitn. 

 Fhys. (5), 28,133 ; (6), 2, 66, 4, 401) ; and by an exhaustive ex- 

 tension of the work, and by including solvents other than water, 

 and solids other than salts, he has thrown a new light on the 

 subject. Not only does the salt, in its influence on the freezing- 

 point, show no signs of the presence of combined water, but it 

 shows no signs ot itself being a smgle compound ; each of the 

 radicles contained in it acts independently of the other, and in 

 precisely the same manner as s^molecuU of a non-saline substance 

 (see especially loc. cit. 4, 426). Precisely similar conclusions as 

 to the apparent non-combined state of radicles in a dissolved 

 salt were arrived at by Valson in his work on capillarity {iliid. 

 1870), and by Hugo and Vrie (ibid. 1883) in their examination 

 of the effect of membranes on salt-solutions. Other instances 

 of a similar nature, physical and thermo-chemical, might be 

 quoted. 



That atoms or molecules which are undoubtedly united may 

 retain their individuality so far as to act towards certain agents 

 as if they were free, is surely not surprising ; anl from such 

 methods as would lead us to conclude that the very radicles 

 composing a salt are uncombined, it would be useless to look for 

 evidence of the more feeble combination of the salt with its 

 water, and inconsistent to argue, from the absence of such 

 evidence, that no combined water is present. 



Although I am n jt inclined to attribute any weight to these 

 special experiments brought forward against the hydrate theory, 

 it is otherwise with more general considerations. 



The formation of hydrates cannot explain the absorption of 

 heat which in many cases accompanies dissolution. The pheno- 

 mena of solution are too universal to permit of imagining the 

 existence of some definite compound of the dissolved substance 

 with the solvent in every case. There is a continuous influence 

 exerted by the salt on its solvent too extensive to be accounted 

 for by the effect of mass on partially dissociated hydrates ; there 

 is a continuity between the fused and dissolved states in many 

 cases, and a regularity in the variation of solubility with change 

 of temperature, &c., which cannot be thoroughly explained on 

 the hydrate theory. 



However undeniable the existence of these compounds may 

 be in many cases, they do not give an adequate explanation of 

 all the facts of dissolution. 



The hydrate theory can be neither rejected nor accepted. 



The explanation of this contradiction is not, I think, very 

 difficult to find. Wc are talking about molecules of solids and 

 liquids, not as they exist, but as they do n t exist. Our chemical 

 formulas for them represent but the results of analysis, or, at the 

 most, the constitution of the substance in that transitory state of 

 simplification which immediately precedes entire decomposition ; 

 what their composition may be when in the free state, and 

 removed from all decomposing forces, we know not ; all we do 

 know or believe ah jut them is, that they are then far more com- 

 plex than chemical formul.T; represent. 



Crystalline form alone would show that a number, probably a 

 very great number, of our so-called molecules combine together, 

 bear certain definite relations and hold certain definite positions 

 towards each other, producing a molecular aggregate or physical 

 unit, which alone should receive the name of molecule. - 



Just as a number of similar particles unite to form an aggregate 



.ind discussion of the results from the point of 

 De Coppet, .-inn. Chint. Phys. (4), 23. 3C6 ; 25, 



^ For a general sumn 

 \iew of these physicists, 

 502 ; and 26, 98. 



^ In a Paper read before this Section last year (Report, p. 989), I argued 

 that our formulas adequately represented the molecules of solids and liquids 

 with which chemical reactions deal, although I fully recognised the existence 

 of far more complex aggregates ; my opinions have so far altered that at 

 present i consider these aggregates to be recognisable in many operations 

 which must be termed chemical, although in the great bulk of ordinary reac- 

 tions the simpler or ultimate molecules need al ^ne be considered. 



or true molecule of any simple substance, so will dissimilar 

 particles unite to form aggregates of a more complex nature. 



It is but natural that our prejudices in favour of the " laws " 

 of chemical combination and atomic valency, to which we owe 

 so much, should lead us to attribute the variable composition of 

 certaiu substances to our imperfect means of investigation rather 

 than to the nonconformity of these substances to our laws. 

 Whether we be right or not in our explanation, we must acknow- 

 ledge that apparent ///constancy in composition is one of the 

 most marked features of immense classes of sub .tances which 

 cannot be termed other than chemical compounds. 



The varied composition of minerals is said to receive an ex- 

 planation in the statement that isomorphous substances may 

 displace each other in definite proportions, but to an indefinite 

 extent. This is undoubtedly true, but it does not obviate the 

 necessity of recognising the existence of some form of attraction 

 between these isomorphous substances. No purely mechanical 

 or physical cause can explain this phenomenon ; mere similarity 

 of crystalline form has been proved to be incompetent to produce 

 such results. A selective power is exhibited by the substances 

 which thus unite, ^ as well defined as that selective power which 

 in the case of simpler substances has received the name of 

 chemical affinity, and the resulting compounds are characterised 

 by the same uniformity in composition and physical properties - 

 which is the attribute of ackno%vledged chemical compounds. 



Nor is it with minerals and artificial crystals only that we 

 find ourselves in what would appear to be a wide border-land 

 between chemical compounds and mixtures. Whether we study 

 the formation of alloys, the occlusion of gases by solids, ranging 

 from the most mechanical action by insensible gradations to the 

 formation of a substance having every appearance of a defiriite 

 compound, or the decomposition of some of the firmest chemical 

 bodies by so called mechanical means (filtration), or the constant 

 change in composition of many basic salts with change in the 

 circumstances of their formation, — we are forced to admit that 

 the definiteness which characterises the combination of atoms 

 may be absent from, or at any rate unrecognisable in, the 

 combination of our so-called molecules to form complex 

 aggregates. 



When we examine the constituents of these apparently in- 

 definite compounds, it becomes clear that it is only substances 

 which resemble each other which can combine in this manner ; 

 and one of the most striking features of dissolution offers such 

 a strict parallel to this, that its meaning can scarcely be 

 mistaken. 



A certaui degree of similarity in nature between the solvent 

 and the substance dissolved is the invariable accompaniment of 

 dissolution. 



Dissolution, I believe, is but one of the many results of 

 apparently indefinite chemical combination. 



We cannot obtain a satisfactory explanation of the compo- 

 sition of mmerals by admitting the existence of definite double 

 salts only, nor can we explain the phenomena of dissolution 

 by confining our attention to definite hydrates only. These may, 

 and in all probability do, exist in solution, but they are only 

 small circles within the larger ones ; their successive formation 

 and decomposition would give rise to irregularities and effects 

 such as those which are observed in some cases ; but these ir- 

 regularities would fcrm but ripples on the more regular changes 

 which would accompany the variations in the molecular aggre- 

 gates,— variations which, as in the case of minerals, would be so 

 dependent on physical conditions as to obliterate their chemical 

 nature when examined from many points of view. 



The evolution of heat accompanying dissolution will still be 

 attributable, as on the ordinary hydrate theory, to the formation 

 of chemical comoounds, but the far greater complexity, and 

 consequently, instability of the-e, than of atomic hydrates, if I 

 may so call them, will remove all difficulty in comprehending the 

 continuous effect of the mass of the solvent upon them, even 

 when the latter exceeds that of the salt many hundred-fold ; 

 where heat is evolved, therefore, the evolution will be increased, 

 though at a diminishing rate, by dilution. 



The rapid increase in the heat of dissolution, produced by a 

 rise of temperature, is but a necessary consequence of the 

 formation of a chemical compound possessing a specific heat less 



' A power or "affinity" so strong that it will sometimes induce a sah to 

 separate out in a crystalline form and with a proportion of water foreign to 

 its nature, as well as from a solution too weak to yield it of its own accord 

 (Aston and Pickering, " Multiple Sulphates," Chem. Soc. Trans-, 1886). 



=■ J. M. Thomson, on the "Double Sulphates of Nickel and Cobalt "(Sr//. 

 Assoc. Rep. 1877.209)^ 



