The Bulletin. 9 



METHODS OF ANALYSIS. 



The continuous distillation of 100 cc. of oil was carried out in an 

 Engler flask electrically heated to give two drops of distillate per 

 second. The flask was protected by asbestos and no correction was 

 made for the thennonieter stem. Repeated comparisons have shown 

 that the results are substantially the same as with the present legal 

 method in which the temperature correction is required. 



The viscosity was taken at 08" F. with an Engler-Ubbelohde vis- 

 cosimeter standai'dized by the German Rcichsanstalt. 



The photometric method was similar to that recommended by the 

 International Committee. Glass lamps were used. The reservoirs 

 were cylindrical with flat bottoms and held about 325 cc. The 

 initial oil level was i> cm. below the top of the wick tube and the 

 drop in oil level was usually about 40 mm. (1.6 in.) during the total 

 burning period of 7} hours. A No. 1 ''Model" burner and Macbeth 

 chimney No. 502 were used. New American wicks, recently dried 

 for one hour at 110" C. were used each time. The lamps were al- 

 lowed to stand over night after filling and trimming. 



The illuminating power was determined after burning ^ hour 

 and again 7 hours later. During the first ^ hour after lighting the 

 flames were turned up to the highest safe limit and were not again 

 disturbed. The oil was kept at a constant temperature of 80" to 85" 

 F. by immersing in running water. Usually about 40 cc. of oil 

 remained in the lamp at the last measurement. The measurements 

 were made with a Reichsanstalt photometer using a standardized 

 Hefner lamp. The Hefner unit was taken as equal to .90 candle 

 power and never varied more than 0.5 per cent on account of 

 humidity. Each reading was made five times. Many of the photo- 

 metric tests were made in duplicate. 



OILS TESTED. 



The oils analyzed were chosen on account of some special feature, 

 such as marked color, high or low viscosity, high specific gravity, etc., 

 or, as in a number of cases, simply to get a sample of as many 

 brands as possible. 



Of the samples shown in the following tables, only 17 per cent of 

 those passing the flash test gave any distillate below 150" C. 



vSeven oils gave less than 15 per cent distillate below 200" C: No. 

 987 (9.1 per cent) ; 1189 (6.1 per cent) ; 1396 (12.8 per cent^ ; 

 1766 (4.1 per cent); 1803 (12.3 per cent); 2161 (8.5 per cent); 

 2478 (0.5 per cent). Six others (Nos. 611, 1515, 1749, 1763, 1765 

 and 1975) gave from 15 per cent to 19 per cent distillate below 

 200" C. 



