TEANSACTIONS OF SECTION B. 637 



tesimal composition of the dilute alcohol or acid. The author noted farther the 

 confirmation of these views hy the subsequent studies of Crompton on the electrical 

 conductivity of diluted sulphuric and other acids, and of solutions of the hydroxyda 

 of potassium and sodium, as well as to the able discussion of these results lately 

 published by Professor H. E. Armstrong. 



He then proceeded to recall his own contributions to the theory of solution, 

 dating from 185.3, in which he has affirmed the chemical nature of solution, 

 and especially his essay entitled Thoughts on Solution and the Chemical Process, 

 in the ' Chemical Gazette ' for 1855. It was therein maintained not only that 

 solution is a chemical process, but that 'all chemical union is nothing else 

 than solution. The uniting species are, as it were, dissolved in each other, for 

 solution is mutual ;' and subsequently that 'the type of the chemical process is 

 found in solutions, from which it is possible, under changed physical conditions, 

 to regenerate the original species.' After giving an analysis of the paper of 18.jo, 

 its conclusions are resumed under the followmg heads: 1. The conception that 

 in the process of aqueous solution there are formed definite compounds with 

 water, accompanied by all the phenomena of chemical union. 2. That in these 

 compounds of a single proportion of a salt or other species with many proportions 

 of water there are clearly defined limits beyond which the furtber addition of 

 water gives rise (A) to decomposition involving a new arrangement of the elements 

 of the previously united bodies (double decomposition), or (B) to simple admixtures 

 of one definite solution, or liquid species, with another less dense solution, or with 

 water. 3. That these compounds are separable in a solid state by changes of tem- 

 perature and (theoretically at least) in a liquid state, by the influence of gravity. 

 4. That liquidity is but an accident of solution, since it is a state depending on 

 temperature and on pressure, which state may be assumed by all species, whether 

 elemental or compound. 



The author then proceeded to notice his later treatment of the subject in his 

 'New Basis for Chemistry,' and especially in the second (revised and augmented) 

 edition (Boston, 1888). Therein are considered the cryohydrates of Guthrie, and 

 the remarkable liquid compounds of salts with small quantities of water, got by 

 Tilden and Shenstone at high temperatures under pressure. He further notes the 

 evidences of the integral vaporisation of solutions above the critical point, under 

 pressure, observed by Hannay and Hogarth, and the previous separation of some 

 of these at high temperatures into lighter and denser layers. Earlier observations 

 of similar separations, and the recent studies of Traube and Neuberg on the dif- 

 ferentiation by gravity of various saline solutions in dilute alcohol, are also noticed. 

 All of these facts are considered as illustrations of the theory of solution so long 

 maintained by the author, which, at least so far as aqueous solutions are concerned, 

 is now adopted by Mendel^ef. 



4. The Comjposition of Copper-Tin Alloys. By A. P. Latjeie. 



Experiments by Lodge, Roberts-Austen, &c., have shown that the physical pro- 

 perties of these alloys point to the existence of two compounds, CUjSn and Cu^Sn. 



In testing the E.M.F. of these alloys in constant voltaic cells the author finds a 

 change of E.M.F. corresponding to the formula CuSn, and by suitable arrange- 

 ments has succeeded in eating out the excess of free tin in an alloy, leaving the 

 compound Cu3Sn behind. 



5. The Composition of ike ancient Roman Mortar from the London Wall. 

 By John Spillek, F.C.S. 



With the view of collecting fresh evidence as to the chemical union of lime 

 and silica by long contact, the author had examined a sample of ancient Roman 

 mortar taken from the section of old London Wall recently laid bare in St. 

 Martin's-le-Grand. 



From the fact that a large quantity (11 per cent.) of silica could be dissolved 

 out readily by caustic soda, and there being a deficiency of carbonic acid without 



