TOXICITY OF COAL TAR CREOSOTE FOR LIMNORIA LIGNORTJM. 
225 
oil or olive oil. So experiments were made using the concentrated light oils. In 
the first experiments of this series the animals, in a Stender dish, were freed from 
water as far as possible with a capillary pipette. The animals were then separated 
from each other and covered with 25 cc. of the oil. At the end of the period the 
animals were very thoroughly washed with repeated changes of fresh sea water 
before final transfer to sea water. The results are given in experiments 64, 68, 
and 62 — the top row of Figure 3. It will be seen that the toxicity diminishes from 
benzol to xylol. 
The degree of “ drying” — that is, the thickness of the film of seawater surround- 
ing each animal — was found to have a striking effect on the velocity of poisoning 
by the light oils. Experiments 46?/, 3 47 y, and 48 y (second row, fig. 3) illustrate 
this point. The method of drying the Limnoria for these experiments was as follows : 
A number of animals in a single drop of sea water were dropped upon a sheet of 
filter paper. When the water had been thoroughly absorbed, the Limnoria were 
dropped onto another filter paper, and from that onto a small scoop of very fine 
copper screen. The animals could thus be instantaneously immersed in or removed 
from the light oil, which was contained in a petri dish. That the period of drying — 
a maximum of five minutes — was not injurious to the animals was shown by control 
experiments. 
A comparison of the results given in the top row, Figure 3 (in which the water 
was taken up with a pipette), with those in the second row (where drying with 
filter papers was used) shows the effect of more thorough drying. This shortening 
of the time required to effect the same grade of poisoning — virtually from minutes 
to seconds — was brought about by an average reduction of less than 0.5 mm. in 
the thickness of the water film surrounding each animal. 
Experiments 55 and 65 (bottom row, fig. 3) illustrate, respectively, the poison- 
ing of Limnoria by saturated sea-water solutions of benzol and xylol. The poisoning 
here may be compared with that produced by a 0.1 per cent solution of phenol 
(experiment 33). The diminution of toxicity of the light oils with rise in boiling 
point is probably largely a function of their respective solubilities in sea water, benzol 
being the most and the xylols the least soluble. 
TAR ACIDS AND BASES. 
TAR ACIDS. 
The experiments under this head cover phenol, orthocresol, metacresol, 
paracresol, alphanaphthol, and betanaphthol. The results are given in the upper 
two rows of Figure 4. Here, in contrast with the fractions and the light oils, the 
toxicity of the tar acids rises with rise of boiling point — from phenol to the naphthols. 
It may be pointed out here, however, that toxicity experiments with a given 
concentration of any purified constituent of a creosote oil will yield widely different 
results, depending upon the medium — sea water or neutral oil— in which the toxic 
substance is dispersed. Fraction IV, for instance, very probably contained high- 
boiling and highly toxic tar acids. But Fraction IV was only slightly toxic, due 
3 Experiments numbered with a y were made in 1917; those numbered % in 1916. All other experiments were made in 1915. 
