COAL-TAR AND WATER-GAS TAR CREOSOTES. 
99 
Out of 28 tar solutions whose content of water-gas tar ranged from 
10 to 45 per cent, 16 were rejected because both fractions were low in 
gravity, and 7 because one fraction was low; 5 passed both tests. 
Of these last, 1 was eliminated on account of both float and coke 
tests, 1 failed to pass the float test, and 1 failed in the coke test; but 
2 passed all tests. Of these last, 1 contained 65 per cent of creosote, 
20 per cent of coal tar, and 15 per cent of water-gas tar; and the other 
contained 65 per cent of creosote and 35 per cent of water-gas tar. 
The method of determining specific gravity of the fractions rec- 
ommended by the American Society for Testing Materials, the 
American Wood Preservers' Association, and the American Railway 
Engineering Association is as follows: 
As specific gravity is an absolute physical determina- 
tion, any recognized method which can be applied to 
the quantity and quality of material at hand to be 
tested must be considered satisfactory. The following 
methods are convenient and accurate means for the rel- 
atively small amounts of oil available in determining 
specific gravity of fractions to be tested. 
Liquid fractions. — The Westphal balance may be used. 
If the fraction to be tested is liquid at a temperature 
not exceeding 60° C, the Westphal balance can be 
- used with convenience and rapidity. A special type A 7 . « 
of Westphal balance is obtainable, designed for testing A / j *v * 
very small quantities. However, the ordinary type of ^-h 2 cm -»! I 
Westphal balance can be adapted to testing small frac- j« 2:5cm H 
tions by the use of a special plummet. When using 
the ordinary balance with the special plummet, extra 
care is needed that the adjustment of the balance be 
accurately made. The plummet can readily be made 
in the laboratory from a piece of ordinary glass tubing 7 h%ujsm- 
mm. outside diameter, sealed at the end, and by melting 
into the glass where sealed, a short platinum wire. Fig. 47.-Pan used for determining 
. , . i • •, ft , 1n , , , , , specific gravity of solid or semi- 
After cooling, place 9 to 10 g. of mercury m the tube, so iid fractions of creosotes, 
making a column 35 to 40 mm. high. Seal off the tube 
within 20 mm. of the top of the mercury column with blow pipe flame. The plum- 
met shall have a length of about 55 to 60 mm. over all, and shall weigh between 10 
and 12 g. 
Solid and semi-solid fractions . — A pan of the form shown in figure 47 having the fol- 
lowing approximate dimensions, may be used: 
Diameter of base 20 mm. 
Diameter of top 25 mm. 
Depth 12 mm. 
Diameter of wire 1 mm. 
Total weight < 7 g. 
The pan and wires are made of platinum or nickel. 
Solid or semisolid fractions of oil which can not be readily liquified can be rapidly 
and accurately tested in this apparatus by the usual method of weighing in air and in 
