88 BULLETIN 1036, U. S. DEPARTMENT OF AGRICULTURE. 
neering Association, and the American Wood Preservers' Association 
are as follows: 
Specific Gravity. 
(a) Hydrometer. — The hydrometer shall be of the form shown in figure 40. It shall 
have the following dimensions: 
Length of stem 125 mm. ; permissible variation 12. 5 mm. 
Length of bulb 105 mm. ; permissible variation 10. 5 mm. 
Length of scale 80 mm. ; permissible variation 8.0 mm. 
Diameter of stem 6 mm. ; permissible variation 0. 5 mm. 
Diameter of bulb 22 mm. ; permissible variation 2. mm. 
A set of two hydrometers with ranges 1.00 to 1.08 and 1.07 to 1.15 will suffice. 
(b) Cylinder. — The cylinder shall be of the form shown in figure 40. It shall have 
the following dimensions: 
Length 300 mm. ; permissible variation 30 mm. 
Diameter 32 mm. ; permissible variation 3 mm. 
The oil shall be brought to a temperature of 38° C. (100° F.), and the determination 
shall be made at the temperature unless the oil is not entirely liquid at 38° C. In 
case the oil requires to be brought to a higher temperature than 38° C. in order to 
render it completely fluid, it shall be tested at the lowest temperature at which it is 
completely fluid, and a correction made by adding 0.0008 to the observed specific 
gravity for each degree centigrade above 38° C. at which the test is made. This cor- 
rection factor does not apply with equal accuracy to all oils, but serious error due to 
its use will be avoided if the foregoing precaution is observed with respect to avoiding 
unnecessarily high temperature. Before taking the specific gravity the oil in the 
cylinder should be stirred thoroughly with a glass rod, and this rod when withdrawn 
from the liquid should show no solid particles at the instant of withdrawal. Care 
should be taken that the hydrometer does not touch the sides or bottom of the cylinder 
when the reading is taken, and that the oil surface is free from froth and bubbles. 
FREE-CARBON TEST. 
The determination of free carbon or insoluble matter was supposed 
to be a measure of the amount of tar that might be present. When 
tar was obtained chiefly from gas houses this test did, in a rough way, 
give an indication of the amount of coal tar present in creosote. If 
5 per cent of the tar containing 20 per cent of the free carbon were 
mixed with a pure distilled creosote, then the mixture would contain 
about 1 per cent of free carbon. In modern coke-oven practice tars 
may run as low as 4 per cent free carbon. Twenty-five per cent of 
such tar could be mixed with creosote and the mixture would still 
have only 1 per cent of free carbon. Furthermore, water-gas tar 
containing no free carbon could be added in any proportion without 
being identified by the free-carbon test. 
The free-carbon test has been conducted by a number of different 
methods. In general it depends on the solubility of the oils in some 
