92 
BULLETIN OF THE BUREAU OF FISHERIES 
oxygenated by running a continuous stream of air through it. Each day the drills 
were examined, and the number of dead in each of the jars noted. In all, 8 series of 
experiments were run. The detailed procedure for each series is given in the following 
paragraphs. 
In Series I, II, III, and IY the salinities used ranged from 9 to 34 parts per mille 
at intervals of less than 2 parts per mille. In all, these involved the use of 1,350 
animals. The results may be summarized as follows: Below a salinity of 15 parts per 
mille over 50 per cent of the drills were killed. The optimum range seemed to be 
between 17 to 22 parts per mille where the number of deaths remained less than 15 
per cent. Above a salinity of 23 parts per mille the number of deaths again increased 
to over 50 per cent. These results seemed peculiar in view of the fact that at Beau- 
fort where these drills were collected the salinity only rarely fell below 20 parts per 
mille and usually remained above 30 parts per mille. This indicated that other factors 
besides salinity must be involved in the deaths at salinities above 23 parts per mille. 
One of the factors that might contribute to the death of the drills at salinities 
above 23 parts per mille is the rapid reproduction of plankton and its subsequent 
death and putrefaction. Below 23 the low salinity might act as a retarding factor 
in the reproduction of the plankton. Thus it might be that the putrefaction of a large 
number of plankton which might obtain at the higher and more nearly normal salini- 
ties might so foul the water as to kill the drills. 
In order to test this, two types of experiments were run. The first involved the 
use of sterile glassware, boiled sea water, and drills washed in sterile sea water. 
Everything was done to keep the initial plankton count to a minimum. If the putre- 
faction of the plankton were a contributing factor to the death of the drills at the 
higher salinities, we should find that with the above precautions the number of deaths 
at salinities above 23 parts per mille would be greatly reduced. Such indeed was the 
case. In Series V (see Table 4) where this was done, deaths to number of 50 per cent 
occurred only at salinities below 15 parts per mille. At the higher salinities the per- 
centage of deaths only rarely exceeded 15 per cent. 
Another way to test the above assumption is to change the sea water (of the 
proper salinity) in each jar daily. In Series VI and YII (1,200 animals) this was done, 
and again only salinities below 15 parts per mille were fatal. (Table 4.) 
The observations made during 1929 on the drills from Hampton Hoads showed 
that the lower limit for survival of the snail is approximately 12.5 parts per mille, a 
result that is similar to that one obtained during the summer of 1927. (Table 2.) 
Table 4. — Effects of varying salinity on XJrosalpinx cinerea from Beaufort, N. C., and Hampton 
Roads, Va., summer of 1929 
