of the experimental set-up, which requires continual attention, and 

 the length of time needed to oomplete the test limit the use of the 

 method and restrict the number of animals that can be used in bio- 

 assaying. From this point of view the carmine cone method presents 

 greater advantages, for the latter test may be completed on a larger 

 number of specimens in a relatively short time. 



The experimental set-up for recording the rate of pumping of 

 oysters by the apron method, as it has been developed in the labora- 

 tories of the Section of Shellfishery Investigations at College Park 

 and Woods Hole, has been fully described and illustrated in a previous 

 publication (Galtsoff, 1946, pages 7-11) and requires no additional 

 disctission. (See Appendix 1.) Only minor changes in the construction 

 of the set-up have been made for the present work. One of these 

 involved the use of plexiglass in place of the cellulose nitrate 

 employed previously for the construction of the apparatus. A change 

 was made also in the construction of the dumping vessel. It consisted 

 in movtntlng above the dumping chamber a small rectangular container to 

 receive the water filtered by the oyster. A glass rod inserted into a 

 drain on the bottom of the container keeps the drain open when the 

 dumping vessel is in an upright position, as shown in figure 11. When 

 the float (inside the dumping vessel) releases the catch and the vessel 

 turns over, the glass rod in the drain tube drops down and a small metal 

 ball, above the drain, closes the outlet. In this way no water is lost 

 when the dumping vessel turns over. When the vessel returns to its 

 upright position, the glass rod is pushed upward, the ball is lifted, 

 and the water frcxn the container flows into the vessel. 



In the experiments described below, crude oil was mixed with 

 carbonised sand in the ratio of 50 ml. of oil to 127 grams of sand. 

 After rinsing the mixture in sea water and wiping off excess oil, the 

 paste-like mass was placed in the two middle compartments of the mixing 

 chamber (B in figure 2, Appendix 1) through which the sea water was 

 running at a rate slightly in excess of the rate of pumping of water 

 by the oyster. The latter precaution is important since accurate results 

 cannot be obtained if the level in the tank in which the oyster is kept 

 does not remain constant (Galtsoff, 1928), 



Oysters used in the experiments were obtained either from Hadley 

 Harbor, a short distance from Woods Hole, or from Narragansett Bay, 

 Rhode Island. Prior to experimentation they were kept in the labora- 

 tory tanks for at least a week to allow for a complete adjustment to 

 the new environment. TWo oysters were used for each test. One oyster 

 received sea water flowing over a submerged layer of oil and carbonised 

 sand and tiie other, as a control, was supplied unoontaminated sea water* 



37 



