72 Scientific Intelligence. 



amount of permanganate. The reaction of permanganate with 

 the eobalti-nitrite precipate is represented by a rather complex 

 equation where 22KMn0 4 are equivalent to 10K 2 NaCo(NOo) 8 . 

 The authors derive their factor from the calcium oxide factor 

 used in the titration of calcium oxalate with permanganate, but 

 perhaps they would have made the matter clearer to chemists 

 generally if they had stated that the normal quantity in the 

 reaction is 1/11 of the formula weight of K„0, or Pe factor X 

 01532 = K 2 factor.— Jour. Indust. Eng. Chem., 9, 785. 



H. L. W. 



3. The Estimation of Arsenic in Organic Compounds. — A. J. 

 Ewins has applied for this purpose a modification of the method 

 of Norton and Koch, which was devised for the determination of 

 arsenic in animal tissues. To apply the method to organic com- 

 pounds, such as salvarsan, for example, a portion of from 0-1 

 to 02 g is weighed and washed into a long-necked Kjeldahl flask 

 by means of a little water, 10 g of potassium sulphate, 0-2 to 0-3 

 g of starch and 20 cc of concentrated sulphuric acid are added, 

 and the flask is heated as in the ordinary nitrogen determination 

 until the liquid is colorless or nearly so. This usually requires 

 about 4 hours. After cooling, the residue is taken up with 

 water, transferred to a 350 cc flask, a solution of sodium hydrox- 

 ide is then added from a burette until the liquid is just distinctly 

 alkaline to litmus paper, a piece of which may be put into the 

 solution. The contents of the flask are then cooled to about 

 30 or 40° and a few drops of concentrated sulphuric acid added 

 until the liquid is again distinctly acid. A cold-saturated solu- 

 tion of hydrogen sodium carbonate is then added from a burette 

 until an excess of 5 or 10 cc of the reagent is present. The 

 arsenic is now titrated by means of A/20 iodine solution in the 

 presence of starch solution as an indicator. The test-analyses 

 carried out by the author upon a large number of complex 

 arsenic compounds show excellent results. — J. Chem. Soc, 109, 

 1355. h. l. w. 



4. A Calorimeter for Measuring Heats of Dilution. — D. A. 

 MacInnes and J. M. Bbaham have devised a calorimeter which 

 seems to be worthy of special mention on account of its remark- 

 able delicacy and accuracy. It is a modification of the adiabatic 

 calorimeter of T. W. Richards. In the Richards instrument a 

 bath surrounding the calorimeter is kept throughout the deter- 

 mination at the same temperature as that in the vessel in which 

 the heat is being evolved. This arrangement eliminates the 

 radiation and conduction of heat from the calorimeter, but a 

 correction must be made for the heat effect due to stirring. In 

 the new modification the outer bath is kept at a temperature 

 slightly lower (about 0-15°) than that of the calorimeter. This 

 difference of temperature is the one at which a thermometer 

 inserted in the calorimeter remains constant indefinitely. This 

 modification of the adiabatic method makes the use of platinum 



