78 REPORT— 1901. 



below the curve the saturation has broken down, and the solid solution 

 has separated into two solid phases. Just as would be the case with a 

 freezing-point curve, the phase which first crystallises on the descending 

 branch IC' is copper-rich, while that of the ascending branch X E' is tin- 

 rich. Moreover, when the temperature falls to the eutectic angle C or X, 

 the residual matter breaks up into the solid eutectic, apparently common 

 to all the alloys from B to D. 



The solid at D' is practically homogeneous even after the transforma- 

 tion of the lower curve has taken place ; that is, the slowly cooled alloy 

 here contains one phase : this may be the compound Cu4Sn. The slow 

 cooled alloy at E is also homogeneous, although when barely solid it is far 

 from being so. There can be hardly any doubt that this alloy when 

 slowly cooled or chilled below E' is the pure compound CujSn ; but be- 

 tween the temperatures E' and e this body may possibly not exist, and 

 above e it certainly does not. This decomposition of the CujSn at or 

 even before melting explains why the freezing-point curve has no summit 

 corresponding to a body which almost certainly exists in the slowly 

 cooled alloys. It would be worth while to examine the changes in the 

 electrical resistance of these alloys when chilled. 



Alloys containing somewhat more tin than CugSn go through similar 

 changes as they cool. They solidify completely at temperatures that are 

 not more than 30 or 40 degrees below their freezing point, the first 

 matter solidifying being richer in copper than the alloy as a whole. When 

 just solid the alloys appear to be uniform, and they remain so until their 

 temperature falls to Roberts- Austen and Stansfield's curve, at which point 

 a solid, that may be Cu3Sn, crystallises out of the solid solution in long bars. 

 These bars do not entirely fill the alloy, but are surrounded by mother 

 substance which grows in bulk with increasing percentage of tin. 



The structure of the chilled alloys shows many other interesting 

 features which the authors hope to discuss at a future time. 



Isomorplious Derivatives of Benzene. — Second Reiwrt of the Committee, 

 consisting o/ Professor H. A, Miers (Chairman), Dr. W. P. Wynne, 

 and Dr. H. E. Armstrong (Secretary). (Drawn up by the Secre- 

 tary.) 



The investigation of the 1:3:5 series of sulphonic chlorides and bromides 

 derived from 1 : 3 dichloro-, dibromo- and chlorobromo-benzene has been 

 continued during the past year and is almost completed. The results 

 confirm and extend those previously arrived at, but also show that it will 

 be necessary to study very carefully the dependence of the crystalline 

 form on temperature and solvent. Progress has been made in preparing 

 material for the examination of the 1:2:3 series, the third set to which 

 the 1 : 3 di-derivatives can give rise ; and the sulphonic derivatives of the 

 1 : 2 dichloro-, dibromo- and bromochloro-benzenes are also under inves- 

 tigation. 



The crystallographical relationship of corresponding methyl-, ethyl-, 

 propyl- and butyl-benzene sulphonic derivatives is also being made the 

 subject of study, with a view to determine the alteration in crystalline 

 form produced on introducing homologous hydrocarbon radicles into 

 benzenesulphonic acid. The results thus far obtained show that a very 

 thorough examination of the series will be required to bring to light the 



