NOVEJIBER 14, 19l)2.] 



SCIENCE. 



797 



lebores and pyrethrums were generally found 

 of good quality. The roach powders, some of 

 which are also sold for killing fleas, ants and 

 other insects, were found to consist generally 

 of borax or pyrethrum, or both. The price at 

 which these are sold should yield a satisfac- 

 tory profit to the manufacturer, one specimen 

 of borax containing nearly seven per cent, of 

 impurity, selling at one dollar per pound. 

 Roach pastes contain one per cent, more or 

 less of phosphorus, mixed with flour, meal, 

 sugar, molasses or lard. Several bug poisons 

 consist of gasoline or of turpentine. ' Rough 

 on Rats ' is a mixture of arsenious oxid with 

 barium carbonate. Altogether the report 

 makes very interesting reading and is valu- 

 able for reference. A point of particular in- 

 terest to chemists is the description of the 

 methods of analysis used in each case. 



From his investigations of the metallic car- 

 bids Moissan has recently drawn several geo- 

 logical conclusions and developed a new theory 

 regarding the origin of petroleum. Accord- 

 ing to his views, in the early periods of the 

 earth's history almost the entire quantity of 

 carbon was combined with metals. Later, 

 water reacting with these carbids formed 

 hydrocarbons, and from these carbon dioxid 

 was formed by oxidation. The origin of nat- 

 ural gas is the action of water upon aluminum 

 carbid, by which methane is evolved. From 

 other metallic carbids liquid carbon com- 

 pounds have been formed, although a different 

 origin is possible for some petroleums. Cer- 

 tain volcanic phenomena may be caused by 

 the action of water upon easily decomposable 

 carbids, while in other cases a similar action 

 might give rise to earthquakes. After all, 

 Moissan's theory seems to be an extreme devel- 

 opment of that put forth a number of years 

 ago by Mendeleef, and which has been fur- 

 nished very strong support by the experimental 

 work of Moissan. 



Moissan has continued his researches upon 

 liquid silicon hydrid, and finds from its vapor 

 density that its formula is Si,H„. It ignites 

 on contact with the air, and when dried by 

 sulfviric acid it explodes still more violently 

 on coming to the air. When the electric spark 

 is passed through the vapor under reduced 



pressure it is completely decomposed and 

 amorphous silicon obtained in long filaments. 

 This amorphous silicon, probably thus for the 

 first time obtained in a pure condition, pos- 

 sesses remarkable reducing properties. Potas- 

 sium permanganate is slowly reduced in the 

 cold, copper sulfate and gold chlorid are re- 

 duced to the metal on boiling, and mercuric 

 chlorid is reduced to calomel. 



The deterioration of platinum crucibles, 

 even when carefully used, is well known and 

 has been generally attributed to the action of 

 the carbon of the flame, though other explana- 

 tions have not been wanting. The matter has 

 been taken up recently by W. Rosenhain, and 

 his results have lately appeared in the Pro- 

 ceedings of the Royal Society. He attributes 

 the cause of the brittleness of platinum which 

 has been used for a long time to a recrystalli- 

 zation of the metal, which has taken place 

 at a temperature far below the fusing-point. 

 Similar changes are known to take place in 

 the case of lead, tin, zinc and cadmium. 

 When platinum is heated to a high tempera- 

 ture, even in the presence of a large excess of 

 oxygen, it gradually becomes quite brittle, so 

 that when at a low red heat it is easily shat- 

 tered by a blow. An examination of the frag- 

 ments reveals a crystalline structure through- 

 out the metal. Under such circumstances the 

 surface assumes a crystalline appearance, and 

 this crystalline structure is not merely super- 

 ficial, but penetrates the metal. Indeed the 

 external appearance is probably due to an etch- 

 ing action of the gases of the flame. 



J. L. H. 



THE COMET B, 1902, AND THE MASS OF 

 MERCURY. 

 Mr. F. E. Seageave calls attention to the 

 close approach of Comet b, 1902, to Mercury 

 on 1902, November 29<i 171i. The heliocentric 

 coordinates of Merctiry at that time are 

 ;. = 225°6', /9=+0°14'0", log. r=9.65328 and 

 of the comet, according to the elements given 

 in the Lick Bulletin ISTo. 25, ;.=225°18', 

 ;?=— 1°50'40", log. r= 9.63581. The loga- 

 rithm of the least distance will, therefore, be 

 0.0177, corresponding to a distance of 1.644,000 

 miles. 



