298 Scientific Intelligence. 



stances. In 1873 Gatehouse proposed to replace the zinc by 

 aluminum in Fleitmann's test; and in 1878 Johnston claimed 

 that this process was successful with arsenic compounds. Fre- 

 senius confirmed this result later and concedes that the Gatehouse 

 process detects arsenie acid. This subject has now been re- 

 examined by Clark who concludes from carefully conducted 

 experiments that when arsenic acid is heated with pure aluminum 

 and potassium or sodium hydrate not a trace of arsine is evolved ; 

 the result obtained by Fresenius being due apparently either to 

 the presence of a little arsenous acid in his arsenate or to phos- 

 phorus. The author tested the suitability of this process for 

 the quantitative determination of arsenic as arsenous acid. But 

 he was unable to convert more than four-fifths of the arsenous 

 oxide into arsine owing to the formation of the solid hydride. 

 Moreover the author finds that the use of sodium amalgam in 

 place of aluminum as proposed by Davy in 1876, does not evolve 

 arsine when the arsenic present is in the form of arsenic acid. 

 Clark has also proposed an improvement in Reinsch's test for 

 arsenic. The material to be tested is boiled gently in a porcelain 

 vessel with dilute hydrochloric acid and a strip of copper foil 

 about 1 inch by £ inch. When the coating is decided the copper 

 is removed and washed with water and alcohol, and then placed 

 in a mixture of dilute potassium hydrate and hydrogen peroxide 

 and allowed to digest in the cold. The arsenic and antimony are 

 dissolved, and the copper now bright is returned to the organic 

 mixture. The alkaline liquid is heated to boiling, filtered and 

 evaporated to a small bulk. It is washed into a small flask with 

 strong hydrochloric acid, ferrous chloride is added, and the flask 

 being fitted to a worm condenser, the arsenic is distilled into 

 water. Through this distillate hydrogen sulphide is passed ; 

 when if more than o*l mgrm of arsenic be present a yellow precipi- 

 tate is obtained. This may be filtered off, washed with CS 2 , alco- 

 hol and water, dissolved off the filter with dilute ammonia, evapo- 

 rated to dryness and weighed as arsenous sulphide. According 

 to Curtmann, a yellow coloration is produced by hydrogen sul- 

 phide when only 0*001 mgriu of arsenous sulphide is present. The 

 filtrate after removal of the arsine is saturated with hydrogen 

 sulphide, the antimony sulphide filtered off, dissolved in caustic 

 soda, filtered to remove any copper sulphide, and the antimony 

 sulphide precipitated by hydrochloric acid ; the precipitate being 

 washed with alcohol, CS 2 and water and weighed as antimonous 

 sulphide. In his arsenic experiments, 0*004, 0*001 and 0*0004 

 gram of arsenic being taken, 0*0042, 0*0014 and 0*0004 gram 

 were found, the latter estimated by color. In the case of anti- 

 mony 0*004 and 0*001 being taken 0*0049 and 0*0017 were 

 found. In presence of organic matter 0*004 gram each of arsenic 

 and of antimony were taken; the quantities found being 0*0038 

 and 0*0042 gram respectively. In a second experiment where 

 0*001 gram of each was taken the arsenic found was 0*001 and 

 the antimony 0*0009. This sensitiveness appears to be all that is 



