76 
BULLETIN OF THE BUREAU OF FISHERIES 
parts per thousand on August 27 but no set was obtained at this time. On the other 
hand, at the same time the salinity in the bayou (fig. 10) and in West Bay (fig. 11) 
was between 24 and 30 parts per thousand, and setting was going on. 
These records are suggestive of a marked correlation between the salinity of 
the water and the setting of oyster larvse. The results obtained in Offatts Bayou 
are the most significant because the small size and relatively great depth of the 
channels prevent a rapid exchange of water, and variations in salinity are not as 
frequent or extreme as in either West Bay or East Bay. Consequently the daily 
samples of water tested gave results more representative of actual salinity conditions 
than in the case of the open bays. Although less accurate the records taken at the 
Deer Islands are also valuable for they are reasonably consistent in spite of the 
exposure of the bay to the effects of tides and winds. The results indicate that setting 
periods followed rises in salinity above approximately 20 parts per thousand, although 
it is certainly impossible to state an exact limit. 
It has long been recognized that very low salinity is unfavorable to the production 
of seed oysters. A number of investigations on the Gulf coast have indicated the 
harmful effect of freshets on oyster larvse. Moore and Pope (1910, p. 19) found that 
shells planted in the Bayou St. Denis in Louisiana “remained barren during the 
period in which the crevasse water (from the Mississippi River) was pouring over the 
beds, but after this was stopped and the water grew more salt a small set appeared 
on these shells * * although their figures of salinity during this period were 
between 13 and 14 parts per thousand. In Mississippi, Moore (1913) found heavy 
setting in water, the average salinity of which was about 22 parts per thousand, as 
contrasted to the observation of Moore and Pope (1910) that in one bay in Louisiana 
not far away there was a very slow set throughout the summer in water ranging in 
salinity up to about 13 parts per thousand, though the figures represent very few 
determinations. Churchill (1920) stated that oysters seem to thrive best in waters 
of salinity between 14 and 28 parts per thousand, and pointed out that the production 
of spat is inhibited by freshets. He considered larvae to be extremely sensitive, 
particularly to cold and rain. 
Conditions of salinity in the Gulf coast waters are entirely different from those 
in the waters of Long Island Sound where Prytherch (1929) has been able to predict 
the peak of setting to occur close to 15 days after the peak of spawning. This 
investigator found that in Milford Harbor, Conn., the salinity ranged from 25 to 28 
parts per thousand. On the other hand the salinity in Galveston Bay during the 
spring and summer varied from less than 2 to nearly 30 parts per thousand. It is 
probably this variability which is largely responsible for the discrepancy between the 
spawning and setting times in Galveston Bay and those in Milford Harbor. 
If, as appears to be the case, a salinity of above 20 parts per thousand was neces- 
sary in Galveston Bay before oyster larvse would set, the question arises as to how this 
factor produced such an effect. The effect may have been indirect, due to the 
abundance of the organisms on which oyster larvse feed which in turn may depend 
upon a certain salinity in order to attain maximum abundance. It is possible, though 
hardly probable, that oyster larvse require water of a certain osmotic pressure in 
order to mature. It is also suggested that salinity as such is secondary in promoting 
setting and that the most important factor is some specific ion or salt in the water, 
a certain concentration of which is essential to the larvse. Wheeler (1910), for 
example, found that in the vicinity of Beaufort, N. C., there was considerable dif- 
ference between samples of water with respect to the relative percentage of some of 
