26 



SCIENCE 



[N. S. Vol. XLA'II. No. 1201 



The effect of pressure upon the formation of 

 bensene and toluene from gas oil: Gdstav Eolofp. 

 Gas oil derived from a Pennsylvania crude petro- 

 leum was subjected to pressures of one, eleven and 

 eighteen atmospheres at constant temperature of 

 700° C. to form benzene and toluene. The follow- 

 ing table tabulates the analytical data: 



An experi/ment m scrubbing carburetted water 

 gas from recovering airomatics: Eobekt J. Moore 

 and Gdstav Egloff. The following tables cover 

 the percentages of benzene, toluene and xylene ob- 

 tained by scrubbing one thousand cubic feet of 

 carburetted water gas, using a paraffin ' ' straw ' ' 

 oil as absorbent medium. 



PERCENTAGES OP BENZENE, TOLUENE, XYLENE AND 

 UNSATURATEDS IN LIGHT OIL OBTAINED 



AROMATICS RECOVERABLE PROM 1,000 CU. FT. OP GAS 

 Liters Gallons 



Benzene 0.267 0.0707 



Toluene 0.242 0.0640 



Xylene 0.062 0.0163 



In view of the faot that the annual production of 

 carburetted water gas is in the neighborhood of 

 150 billion cubic feet the above figures assume an 

 added significance. 



Deposition of silver films on glMSs: Alexander 

 Silver and Eaymond M. Howe. A paper dealing 

 with the study of various reactions involved in 

 depositing silver films on glass from ammoniacal 

 silver solutions by the use of aldehydes in the pres- 

 ence of alcohols and sugars. By modifying the 

 old Liebig method it is found possible to produce 

 perfect silver films on glass through the introduc- 



tion of alcohols and sugars, the mirrors forming 

 in the cold. The rate of deposition is controllable. 

 The cost of mirror production is lowered consider- 

 ably and the efiiciency of the process as developed 

 is higher than that of any of the older methods in 

 use. 



Some notes on chars and other solid decolorizing 

 agents: Charles E. Wood. The decolorizing ef- 

 fects on sugar solutions of bone char, animal and 

 vegetable chars and several other kinds of carbon, 

 including lamp black and aquadag are compared 

 with one another and with fuller's earth and clay. 

 A brief statement as to optimum conditions of 

 manufacture and use is given and a few theoret- 

 ical considerations are brought together. 



Comparative tests of porcelain laboratory ware: 

 C. E. Waters. Eive kinds, which were all that 

 could be obtained, were tested: two American, two 

 German and one Japanese. One German and the 

 two American wares did not stand up well when 

 heated to 225° C. and quickly cooled by floating 

 on water at room temperature, or when suddenly 

 heated in the flame of a Fletcher burner. There 

 was not only actual breakage, but in many cases 

 the glaze cracked under the tongs when a hot 

 piece was picked up. The solubilities in hot so- 

 lutions of sodium hydroxide and carbonate, in 

 nearly boiling concentrated sulphuric acid, in 

 fused sodium nitrate and in a fused mixture of 

 this salt and the carbonate, were comparatively 

 slight. In this respect there is little choice be- 

 tween the five brands. Ferric oxide, when ignited 

 in a thin layer over the bottom of the dish, 

 stained the glaze, but was readily removed by di- 

 gestion with hydrochloric acid. The loss in 

 weight under this treatment was so small that there 

 could have been little or no formation of an 

 easily soluble silicate. 



Comparative tests of chemical glassware: Percy 

 H. Walker and F. W. Smither. Composition, co- 

 efiicient of expansion, refractive indices, condition 

 of strain, effect of heat and mechanical shock and 

 resistance to various chemical reagents were deter- 

 mined on seven kinds of glassware, which bear 

 permanent manufacturers' trade-marks, and which 

 are offered for sale on the American market. Two 

 of these kinds of glass were of foreign manufac- 

 ture. The tests, which were intended to furnish 

 information as to the relative values of the dif- 

 ferent makes of glass for laboratory operations, 

 show that the five kinds of American-made ware 

 are distinctly superior to one of the foreign brands 

 and at least equal to the other foreign brand. 



