63 REPORT— 1876. 



the smaller one being replenished with weak liquor. A limited supply of air may 

 be introduced into the second pan, and evaporation proceeds very rapidly. 



The liquor in the pans need never lose its level, because, as salts pass from below 

 and steam from above continuously, it is continuously replaced by liquor flowing in ; 

 the air-pipes may therefore be only two or three inches below the surface of the fluid. 

 The pressure thus being not great, an ordinary fan will be sufiicient to force the air 

 through for evaporation. 



On Sewage Purification and Utilization. By J. Banks. 



On a new Voltaic Battery. By H. W. Biggs. 



On the Action of Pentacliloricle of Phosj^horus on Turpentine, 

 By Prof. Ceum Beown. 



Note on Anthracene-testing. By Jas. T. Bkoavn. 



In the earlier days of anthracene manufacture, when it was obtained solely from 

 the last runnings of oil, and when the distillation was stopped comparatively early 

 for the double reason of saving the bottoms of the stills and producing a good 

 marketable pitch, the principal solid impurities were naphthalene, phenanthrene, 

 and paraffin. With samples of this description the method of testing by agitation, 

 after washing with petroleum spirit, with a limited quantity of bisulphide of carbon 

 gives approximate and practically useful results. When, however, the demand for 

 anthracene inci-eased, the tar-distillers found it more advantageous to carr}' on the 

 distillation as far as possible, only stopping just before the point at which coking 

 conmienced. This method of working gives an entirely difl'erent variety of crude 

 anthracene, viz. one in which the principal solid impurities have higher boiling- 

 points than anthracene. These bisulphide of carbon fails to remove ; that test 

 therefore, with these samples, ceases to give true indications of their commercial 

 value. To correct this the authrakinone test was introduced, and was, judging fi'om 

 the terms in which it was proposed, looked upon as applicable to all commercial 

 anthracenes. The appendix to the paper soon followed, and showed that experience 

 had not confirmed those anticipations ; and now the kinone produced requires to be 

 tested as to its purity, as the result is by no means deflnite. In applying the kinone 

 test to commercial samples various minor difficulties occur, one of which is that 

 damp samples of anthracene are apt to lose moisture during the time that is occu- 

 pied in reducing them to a sufficient degi-ee of fineness to Jillow the small quantity 

 of I gramme to be a correct sample of the bulk ; and another and more serious one 

 is the uncertainty caused by the occasional occurrence of accidental impurities in 

 the quantity weighed out. To remedy these defects and facilitate the testing, the 

 author proposes the following modification : — 



W^eigh out 50 grammes of the crude anthracene, and measm'e out 250 cubic 

 centims. of petroleum spirit; triturate the anthracene in a mortar with a sufficient 

 quantity of the spirit to form a thin cream, and pour it on a weighed filter (taking 

 care at the same time to leave in the mortar any grit or sand which may be present) ; 

 rinse on to the filter any anthracene which may be round the sides of the mortar, 

 and employ the remainder of the spirit in washing the filter and its contents. 

 Allow it to drain, then fold it carefully, press between bibulous paper, dry at about 

 (50° -80° C, and weigh. Crush to a fine powder the contents of the filter, and from 

 that quantity weigh out the gramme required for the kinone test; then proceed in 

 the usual manner. In calculating the result allowance must of course be made for 

 the diminution in weight caused by washing the crude sample with petroleum 

 spirit. 



The method proposed in the foregoing short note does not claim for itself theo- 

 retica accuracy, but it claims the following advantages : — ^ 



