\m\ 



680 



BEPORT— 1884. 



object of this research is to iletormine tin* curves of expansion of solutions wliicli 

 ronluin, in equal voliiMK's, wei^'htsof tlie salts proportioiml to llic molecular wejirlits. 

 Tlie ex|)orinients made show the feasibility of einployiiijj; sinipliT ujijiaratus thun 

 that liitherto used in (hiterinininijr the coellicient.s of expansion of saline solutions. 

 The apparatus is a <,'ra(lnated bulh-tubo with a bore lariro euoUL'h to achnit a .small 

 capillary tulx> with wliicli the solutions are run into the Ijulh-tuhc 'I'lie solutioiiH 

 are hoiled a few seconds, wliile still hot poured into t ho bulb-tube, and are tlitn 

 closed offfroTU the nir by a mercury thread. The l)ath ia a iar},'!' vessel of water 

 stirred constantly, Two fovivn of readinfrs are taken, one with llie teniperutun! 

 risill^• and one with the temperature falliufr. 'J'he experiment is rejected uiilc-s 

 these two sets give the same results. The experiments recordeil in our ])aper wciv 

 maile witli an ordinary tliei'mometer prraduate<i in single deirrt'es and do not jrivu 

 .sTitliciently accurate results, but we hope with more delicate thermometers to very 

 much increase the accuracy. 



8. On the ProiJnction of Permanent (la.'^/roni ParaJJlii Oils, 

 llij Dr. Stevenson ,M.vcad\m, F.Ii.S.l'J. 



The paraflln oils employed in the investigation were r/v/r/c ;w/y/^« oi7, heiiif; 

 the oil obtained direct from the destructive distillation of ahale in retorts; i/rcfit 

 2^nr(iJJin oil, which is obtained by distilling or re-nmniiig the crudt» jiaraflln oil nml 

 removing the lighter or more intlammahle oil by fraotional distillatitm and freeziiij; 

 out the solid paratlin ; and Idne paraffin ail, which is the ^iroduct obtained by 

 rectifying tlie parallin oil by sulphuric ac-id and soda, and distilling off the ])iuai!lii 

 spirit, burning oil and intermediate oil, and freeiiiiig out the solid paratlin ns 

 parallin .scale. After many trials it was found that the best results were ohtaineil 

 by the simple arrangement of employing iron retorts at a good cherry-red heat, ami 

 running the oil as a thin stream direct into the retorts, so that it qiuckly inipi";rol 

 upon the red-hot metal, and without the intervention of any coke or other mat eriiil 

 in the retort. The ioUowing table gives the condensed results of one serieti of 

 trials : — 



Crude Panillin Oil Green Paratlin Oil 



Specific gravity of oil, "I j 

 (water =1000) . ,/ j 



Flashing point in close \ 

 tester . . . j 



Gas from gallon of oil . [ 



Oas from tun of oil . . 



Candle power of gas 



Light value of one cubic j 

 foot of gas in sperm > 

 candles . . J 



Light value of g.is fromS 

 gallon of oil in sperm '• 

 candles . . j 



Liglit value of gas from 

 tun of oil in .S2ierm 

 candles 



Liglit value of the- oils as 

 oils, determined from 

 previous experimental 

 data . , . 



Percentage of oil light ^ 

 obtained in gas light/ 



850 



92° F. 



98-70 cub. ft, 



2(;,02(; cub. ft. 



COoG candles 



1208-04 grain.s 



17-052 lbs. 



4494 lbs. 



5565 lbs. 

 8075 



884 



105° F. 



102-52 cui>, ft. 



25,977 cuh. ft. 



5;i-24 candles 



1277-70 grain.s 

 18-718 lbs. 

 4741 lbs. 



5936 lbs. 



79-S7 



nine Piiraflin Oil 



878 



I 



1 03° F. j 



127-42 cuh. ft, I 



32,492 cul). ft. I 



54-28 candlus ' 



1302-72 grains 



23-704 lbs. 

 6047 lbs. 



7420 lbs. 

 81-41) 



In these trials, therefore, the gas obtained from the oils is equal in light power 

 to 80 per cent, or four-fifths of the total light of the oils when such are consumed as 



