214 JONES AND SCHUMB. 



In his measurements at 0°C. he finds a constant potential of 0.820 

 to 0.8195 volt from 49.66% thallium down to amalgams containing 

 31.81% thallium, whereas an amalgam containing 28.44% thallium 

 shows a potential of 0.816 volt followed by a rapid drop to 0.782 volt 

 for a solution containing 21.22% thallium. 



In the interpretation of these results his reasoning is based on the 

 premise that the compound TlHg2 exists (as indicated by the results 

 of Kurnakow and Puschin). The absence of any break in the con- 

 centration-potential curve at 33-?% thallium (corresponding to TlHg2) 

 he regards as proof that all mixtures containing more than 33-g-% 

 thallium at 0° are mixtures of pure TlHg2 and pure thallium and that 

 no mixed crystals are produced, whereas TlHg2 and Hg form solid 

 solutions in all proportions. However, since his premise based upon 

 Kurnakow's result has been shown to be in error, this reasoning has 

 no force. His experimental results are, however, entirely in accord 

 with the data of Roos and of Richards and Daniels. According to 

 these experimenters the solid compound is Tl2Hg5 (28.57% Tl) which 

 can (according to Roos) form solid solutions with thallium up to the 

 limit 31.3% thallium. Similarly solid thallium can form solid solu- 

 tions with mercury containing up to 18% mercury or 82% thallium. 

 The eutectic mixture which freezes at -|-0.6°C. would therefore consist 

 of a mixture of two solid solutions containing respectively 31.3% and 

 82% of thallium. Any amalgam whose total analytical composition 

 fell between these limits would at 0° be made up of these same two 

 phases in proportions which depend upon the composition and would, 

 therefore, be expected to show a constant potential. On the other 

 hand any amalgam containing less than 31.3% thallium but more 

 than 20% thallium would be a single solid phase which would have a 

 potential dependent on the composition. The experimental results 

 of Sucheni are thus entirely in accord with the data and interpretation 

 of Roos, though not in accord with his own interpretation thereof. 



This argument will be made clear by an inspection of Figure 2. In 

 the uppermost curves are plotted the results of Kurnakow and 

 Puschin, of Roos, and of Richards and Daniels, on the freezing-points 

 of thallium amalgams. It is very plain that the results of the first 

 mentioned investigators are seriously at variance with thosp of the 

 more recent workers, and that the wholly independent determinations 

 of Roos and of Richards and Daniels agree very closely over the range 

 covered by the lattsr. 



The curve of eutectic halts, plotted from the results of Roos, shows 

 two intervals in which the halt is vanishingly small; indicating the 



