Chemistry and Physics. 451 



SCIENTIFIC INTELLIGENCE. 



I. Chemistry and Physics. 



1. New Experiments in Preparing Diamonds. — In his exam- 

 ination of the Canyon Diablo meteorite, Moissan noticed that 

 the diamonds found in it occurred in fissures containing sulphide of 

 iron, and it seemed possible that sulphur might have had an influ- 

 ence in the formation of these crystals. He has, therefore, made 

 some new experiments upon the formation of diamonds by crys- 

 tallization from iron saturated with carbon in the electrical fur- 

 nace and rapidly cooled with water, and has modified his 

 experiments of several years ago by additions of sulphide of 

 iron, silicide of iron, or phosphide of iron to the fused metal 

 before cooling. The results showed that the production of dia- 

 monds was somewhat better in the presence of sulphur than had 

 formerly been the case without this addition, and silicon appeared 

 also to exert a favorable influence, but in this case more difficulty 

 was encountered in the separation of the diamonds from the 

 ingot, on account of the presence of carbide of silicon. No dia- 

 monds were obtained in two experiments where phosphide of 

 iron was added. Some of the diamonds produced in the presence 

 of sulphur were large enough to be separated with the naked eye 

 by means of a steel point, while the largest of those made in the 

 presence of silicon had a length of *75 mm . Their size is of the 

 same order as that of those from the meteorite, and they are 

 practically microscopic objects. Moissan has confirmed his previ- 

 ous conclusion that the artificial substance is actually diamond, 

 and he finds that, like the natural substance, it often shows weak 

 double refraction. He regards the diamond as a form of carbon 

 which has been liquefied under high pressure, since he has shown 

 previously that at ordinary pressure all forms of carbon vaporize 

 without becoming liquid, and all produce graphite. — Comptes 

 Pendus, cxl, 277. • h. l. w. 



2. Atomic Weights of Sodium and Chlorine. — T. W. Rich- 

 ards and R. C. Wells have made a very elaborate investigation 

 in revising these important atomic weights. As is usual with the 

 work of Richards, the greatest precautions have been taken in 

 purifying materials, detecting sources of error and avoiding or 

 making accurate allowances for them, and in the use of varied 

 methods and materials. The description of the work inspires the 

 utmost confidence in it, as well as admiration for the skill and 

 patience displayed in carrying it out. As a result of their 

 research it appears that the work of Stas on the atomic weights 

 in question was slightly erroneous. From a study of the ratios 

 AgCl:NaCl, Ag:NaCl, and Ag : AgCl, they find that when the 

 atomic weight of silver is taken as 107*930, sodium is 23*008, and 

 chlorine is 35*473, whereas the previously accepted numbers are 

 23-05 and 35*45. Many other atomic weights are affected, in 



