550 Scientific Intelligence. 



a considerable amount of helium and also a little neon. These 

 gases amounted to 0*17 per cent by volume of the original gas, or 

 nearly one per cent of the residue, so that it was estimated that 

 at least 12 cubic meters of helium and neon had escaped from the 

 crack in the carnallite during the 4-J- years of the flow of gas. 

 It appears that Strutt had already detected a small quantity of 

 helium in carnallite by dissolving the mineral in water. 



In speculating in regard to the origin of the hydrogen the 

 author is inclined to discard an ingenious theory advanced by 

 Precht, that this gas arises from the reaction of ferrous chloride 

 upon the water of crystallization of the mineral with the forma- 

 tion of hematite, a well-known constituent of the deposits. In 

 fact, by heating appropriate mixtures in evacuated sealed tubes 

 at 100-110° C. for 160 hours, no evidence of the formation of 

 hydrogen or ferric oxide could be obtained. This experimental 

 result, and the fact of the presence of helium, have led the author 

 to advance the view that radio-activity has been the cause of the 

 formation of hydrogen as well as of helium. Radio-active sub- 

 stances are supposed to have decomposed water into its elements, 

 then the oxygen has reacted with ferrous chloride to form hematite, 

 leaving the hydrogen. The radio-active bodies are supposed to 

 have disappeared from the deposit, on account of being of shorter 

 duration than uranium, which seems to be absent, and which has 

 so long a life that it could not be expected to disappear in the 

 time that has elapsed since the deposits were formed. 



The author doubts that the feeble radio-activity of potassium 

 is the source of the helium, although Strutt is inclined to take this 

 view of the matter, for he argues that the helium is by no means 

 uniformly distributed among the potassium salts, and, besides, 

 there is no ground at present for assuming the production of 

 helium from potassium. — Berichte, xliii, 111. h. l. w. 



2. Detection of Methyl Alcohol. — A rapid and accurate method 

 for the detection of this substance, particularly in the presence of 

 ethyl alcohol, has been devised by G. Deniges. It is based upon 

 the conversion of the alcohols into their corresponding aldehydes 

 by the action of potassium permanganate under definite condi- 

 tions, and the testing for formaldehyde by means of bisulphite- 

 fuchsine under conditions that the author has worked out. He 

 takes 0-1 cc of the alcohol to be tested in a test-tube, adds 5 CC of 

 1 per cent KMn0 4 solution, then 0'2 CC (not more) of pure H 2 S0 4 , 

 and mixes. After 2 or 3 minutes l cc of 8 per cent oxalic acid is 

 added, and the mixture is agitated. It rapidly becomes decol- 

 orized, and when a yellowish wine color is reached, l cc of pure 

 H 2 S0 4 is added and it is agitated again, when it becomes com- 

 pletely decolorized. Then 5 CC of bisulphited fuchsine solution are 

 added, the liquid is mixed and allowed to stand, when a violet 

 color, more or less intense according to the amount of methyl 

 alcohol originally present, appears. The color usually reaches a 

 maximum in about 15 minutes, it is very strong with 1 per cent 

 of methyl alcohol, and is appreciable at a dilution of one-thou- 



