514 



SCIENCE. 



[N. S. Vol. XIX. No. 482. 



operation. Platinum vessels are seriously 

 damaged, becoming brittle, and at the same 

 time increasing in weight, owing to the depo- 

 sition of iron from the iron carbonyl formed 

 in the passage of the gas through iron pipes. 

 A similar dejDosit, which can not be wiped off, 

 is formed on the bottom of porcelain crucibles,' 

 precluding the use of such crucibles in quan- 

 titative analysis. It was found that a very 

 considerable quantity of unconsumed carbon 

 monoxid escaped into the air, so that this 

 could easily be injurious to health. In a labo- 

 ratory where thirty coal-gas burners may be 

 kept going without detriment to health, hardly 

 eleven water-gas burners can be used with 

 impunity. In the ensuing discussion of the 

 paper it was suggested that where the water 

 gas is carburetted, as is usual in this country, 

 it is probable that it is less objectionable as a 

 laboratory fuel. 



YELLOW ARSENIC. 



The yellow modification of arsenic, which 

 has been observed by several chemists, has 

 been submitted to a careful examination by 

 Erdmarm and Unruh, and their results are 

 published in the Zeitschrift fur anorganische 

 Chemie. The yellow arsenic corresponds to 

 white phosphorus, and is produced by rapidly 

 cooling the vapor of arsenic. In practice the 

 best method was found to be sublimation in a 

 tube of aluminum in an atmosphere of inert 

 gas. The fumes are cooled rapidly by absorp- 

 tion in carbon bisulfid, in which yellow arsenic 

 is soluble. When a saturated solution is 

 cooled to — Y0° it deposits the yellow arsenic 

 in the form of a yellow powder, which can be 

 preserved at this temperature without change 

 if kept in the dark. When exposed to light, 

 even in solution, it is rapidly changed to or- 

 dinary arsenic. After a time a brownish-red 

 precipitate is formed in the carbon bisulfid 

 solution which seems to be a fourth modifica- • 

 tion, and reminds one of red phosphorus. The 

 molecular weight of the yellow arsenic was 

 determined, and the molecule corresponds to 

 As^. 



COPPER CYANID SOLUTIONS. 



In the same journal is a paper by F. P. 

 Treadwell and C. v. Girsewald on the colorless 



solution of copper cyanid, which is not pre- 

 cipitable by hydrogen sulfid, and which is very 

 familiar to all students of qualitative analysis. 

 The compound present in this solution is vari- 

 ously given in different text-books, some con- 

 sidering it merely a double cyanid of bivalent 

 copper and potassium, as K,Gu(CISr)„ while 

 others afiirm that the copper is present in 

 univalent form, being reduced by the potas- 

 sium cyanid. Por this the formula KCu- 

 (ON), is sometimes given. In both cases it 

 appears wholly a matter of mere conjecture. 

 The authors studied solutions containing vari- 

 ous proportions of copper and potassium cy- 

 anid, and arrived at the conclusion that the 

 salt present is KjCu.CCN),, though the salt 

 itself was not isolated. The complexity of the 

 ion which is not decomposed by hydrogen sul- 

 fid was determined to be [Cu^CCN)^'^, by the 

 use of cryoscopic methods. 



CORROSION OF IRON WATER MAINS. 



An interesting case of corrosion is reported 

 from Frankfurt in the Zeitschrift fiir ange- 

 wandte Chemie, by Martin Preund. This pipe 

 contained in places holes as large as the palm 

 of the hand. On the edges of these holes the 

 iron had been converted into a dense, dark 

 gray, soft mass resembling graphite. Analysis 

 showed the mass to be composed of ferrous 

 phosphate, ferrous silicate, carbon, and ten 

 per cent, of metallic iron. As the surround- 

 ing soil could not have supplied the phosphorus 

 or silicon, it appeared that all the materials 

 had come from the iron itself. By the action 

 of stray electric currents the phosphorus and 

 the silicon of the iron had been oxidized to 

 phosphate and silicate. In order to test this 

 supposition, Preund subjected a portion of the 

 cast iron of the pipe as anode to the action of 

 an electric current in a dilute salt or gypsum 

 solution, and found that in a short time the 

 iron became coated with a deposit containing 

 phosphate and silicate of iron, and in every 

 respect resembling the corrosion product of 

 the pipe. J. L. H. 



A QUARTERLY ISSUE OF THE • SMITHSO- 

 NIAN MISCELLANEOUS COLLECTIONS.' 

 The Smithsonian Institution has com- 

 menced the publication of a Quarterly Issue 



