150 
BULLETIN OF THE BUREAU OF FISHERIES 
of the liquor. (Table 3.) As has been pointed out, the oyster which stays open 
least during treatment lives longest. Such a composite figure, as stated above, 
assumes that the fact that an oyster remains closed results in protection. That the 
assumption is justified will appear on examination of the records of individual oysters. 
The graph (fig. 33) shows more clearly the decrease in effectiveness of the liquor as 
the concentration is lowered, and gives a picture of the results not otherwise obtain- 
able. There is, certainly, a fairly wide distribution of the points, but most of them 
fall into a definite curve. 
Table 3. — Data concerning the specimens which died as a result of treatment with sulphite liquor 
solutions. These data are shown graphically in Figures 32 and S3 
As stated above, death occurred in some of the specimens in concentrations as low 
as 0.67 parts per thousand. To this concentration two specimens were subjected 
(experiment No. 23, fig. 18), but only one died during the time the experiment was 
continued. The other oyster was still living at the end of the test, but the activity 
of this specimen could certainly not be called normal. During the entire experimental 
period this oyster averaged only 14.5 hours per day open, while the two control 
specimens averaged 22.9 and 21 .8 hours per day open, respectively. This experimental 
oyster was open 35 per cent less time than the controls. Then this oyster was under 
a disadvantage in that it could take in no more than 65 per cent as much food as 
normal oysters. This would be the case even if the liquor exerts no direct lethal, or 
toxic, effect upon the oyster in other respects. In experiment No. 22 one specimen 
died while one continued to live during 29 days of treatment with 1 part per thou- 
sand. However, the latter, during that period, remained open only 6.74 hours per day 
on the average. It was able to feed a maximum of only 30 per cent as much as the 
control oysters. 
In experiment No. 2 (0.5 part per thousand) neither specimen died during the 31 
days that the test was continued, but both together averaged only 6.9 hours per day 
open, as contrasted with the controls, which remained open an average of 18.15 hours 
per day. Feeding time of the experimental oysters was thereby reduced by about 
62 per cent. 
