82 BULLETIN 1036, U. S. DEPARTMENT OF AGRICULTURE. 
Table 31. — Change in specific gravity of different mixtures of tar and coal-tar creosote. 
Mixture contains — 
Change 
in specific 
gravity per 
degree Fah- 
renheit. 
Tar. 
Creosote. 
Per cent. 
10 
20 
30 
40 
50 
Per cent. 
90 
80 
70 
61 
50 
0.00042 
.00042 
.00041 
.00041 
. 00040 
VOLATILITY. 
The question has been raised whether or not evaporation can be 
partially prevented by the addition of some such nonvolatile material 
as tar, the tar having some sort of binding or holding property which 
would prevent the creosote from evaporating as rapidly as if it were 
used alone. In one of their papers, Von Schrenk and Kammerer (25) 
give a number of curves showing the loss by evaporation of creosote, 
tar, and creosote-tar mixtures. The tests extended over a long 
period of time, namely, 1,000 days, or nearly 3 years. The curves 
show that the tar is less volatile than the creosote-tar mixtures, and 
that the latter are less volatile than the creosote. If, however, the 
results are plotted in the same way as those of the creosote test 
(fig. 31), the same relation between volatility and percentage of 
distillate up to 270° C. will be shown. Furthermore, the volatility 
of the creosote-tar mixture can be calculated from the known vola- 
tility of the creosote and tar. The calculation is extremely simple 
and consists in multiplying the volatility of each constituent by the 
percentage used in the mixture and taking the sum of the two products 
as the volatility of the mixtures. Calculations made in this manner 
show the results given in Table 32. 
Table 32. — Comparison of actual and calculated losses due to evaporation of tar and 
coal-tar creosote mixtures. 
Calcu- 
Loss of 
creosote. 
Loss of 
tar. 
Tar mixtures. 
lated 
loss of 
mixtures. 
Actual 
loss. 
Per cent. 
Per cent. 
Per cent. 
Per cent. 
Per cent. 
48.8 
9.o 
70 p. ct. C, 30 p. ct, T. 
37.0 
3S.0 
46.2 
9.0 
80 p. ct. 20 p. ct. 
3S.8 
40.4 
46.2 
9.0 
70 p. ct. 30 p. ct, 
35.0 
37.8 
65.0 
6.0 
80 p. ct. 20 p. ct, 
53.2 
51.8 
50.8 
5.6 
80 p. ct. 20 p. ct. 
41.8 
41.8 
59.6 
10.9 
80 p. ct. 20 p. ct. 
49.8 
48.6 
Table 32 shows that the actual loss and the calculated loss are 
practically identical. In other words, the effect of the tar is simpl}^ 
to reduce the quantity of creosote in the test, and the percentage 
