80 REPORT — 1871. 



family group of the heterologues of tlie fatty alcohols, seems, from its want of 

 stability, very proue to merge into the isomeric and far more permanent modifica- 

 tion of the glycolite of water. In this remarkable metamorphosis the double 

 carbon adjunct of the principal water-base becomes first of all converted into an 

 acid twin carbon-nucleus, which reunites under this new form with the old water- 

 base, whereupon, under the combined influence of base and acid, the remaining 

 water-molecule becomes decomposed, so as to surrender its oxygen to the envelope 

 of the acid twin carbon-nucleus, while the hydrogen connects itself with the same 

 nucleus under the tj-pical form of a hydrocarbon adjunct. It is worthy of note 

 that in this singular and characteristic rean-angemeut of the constituent elements 

 the organic molecule has, without loss of substance and without loss of satu- 

 rating capacity, passed at one bound from the categorj' of a true and genuine al- 

 cohol into the categorj' of a true and genuine water- salt. As regards the two 

 remaining heterologues of glycolate of water and oxyglycolate of water, you can- 

 not but see that their formation is due to the successive absoi-ption of two mole- 

 cules of oxygen by the envelope of the glycolous acid constituent, and that they 

 differ from each other in this respect only, that the former contains for its prin- 

 cipal constituent formate of watei", while the latter contains instead of it oxy- 

 formate of water. This oxyformate is a highly interesting isomeric modification 

 of the neutral carbonate of water, which, on account of its excessive want of 

 stability, cannot be obtained in a state of isolation. The compound before us 

 differs from the isomeric neutral carbonate in being decidedly monobasic, while 

 the latter is as decidedly bibasic. The cause of this apparent anomaly becomes 

 now fully revealed ; for it is plain that one of the two hydrogen-molecules, which 

 in the ordinary carbonate of water are both of them readily displaceable by metals, 

 has assumed the hydrocarbon form of gi-ouping, in consequence of which it will 

 cease to play the part of a basic nucleus ; and although it may become eliminated 

 or exchanged in obedience to other modes of substitution, it is certain that the 

 ordinary process of double decomposition has no control over it. 



The Molecular arrangement of the Alloy of Silver and Copiier employed for 



the British Silver Coinage. By William Chandler Robeets, Chemist of 



the Mint. 



Experiments have demonstrated that when a molten alloy of silver and copper 

 is allowed to cool, the composition of the resulting metal is not uniform, the cooling 

 being attended with a remarkable molecular rearrangement, in virtue of which 

 certain constituents of the molten alloy become segregated from the mass, the homo- 

 geneous character of which is thereby destroyed. 



Thus, to take an extreme case, an alloy containing 77'33 per cent, of silver and 

 22'67 per cent, of copper was cast in a cubical mould of 42 millimetres. A portion 

 cut from the centre of the mass gave on assay 78-318 per cent, of silver, while a 

 portion cut from one of the angles was found to contain only 77"01o per cent, of 

 silver, showing a ditference of 13-03 milliemes. 



Level proved that the alloy containing 71'89 per cent, of silver is homogeneous, 

 and in all alloys containing more silver than this amouut the centre of the soli- 

 dified mass is richer than the exterior ; on the other hand, in alloys of fineness lower 

 than 71 "89, the centre contains less silver than the external portions. 



The alloy employed for the British silver coinage contains 92o parts of silver and 

 75 parts of copper in 1000 parts of alloy. The metals are melted together and cast 

 into bars 18 inches long and 1 inch thick ; these bars are subsequently rolled into 

 strips or ribands, and from these ribands the disks of metal to form the coins are 

 cut. 



Experiments conducted in the most careful manner proved that the centre of 

 the riband contained more silver by two parts in the thousand than the external 

 edges. The increase in richness from one edge of the riband to the centre, and 

 the coiTesponding decrease in riclmess from the centre to the opposite edge, was 

 extremely regular, as was shown by the curve or graphic representation of the 

 results by which the paper was illustrated. 



