is at least deep enough to allow the pump intake 

 hose to be submerged. The device may also be 

 operated on exposed tidal flats if a water supply is 

 within reach of the intake hose. The small screen 

 over the end of the intake hose prevents debris 

 from entering the water pumping system. 



The boat is anchored and the sampling device 

 lowered over the side and pushed several centi- 

 meters into the sediment prior to starting the 

 pump. The operator may choose to stand in the 

 water with waders in areas where currents and 

 wind cause the boat to move a great deal resulting 

 in an unstable work platform. 



When the pump is activated, water is forced 

 down the space between the inner and outer PVC 

 pipes, forcing a suspension of sediment and all 

 associated organisms up the smaller inner pipe 

 and into the attached sieve basket. The operator 

 forces the device down into the sediment to the 

 desired depth, then the pump is shut off. At this 

 point, the sieve basket contains all organisms and 

 objects too large to pass through the screen. Most 

 sediment washes easily through. 



Discussion 



After collecting more than 2,000 samples over a 

 1-yr period, no problems have been encountered 

 with the system. A 2.5 mm mesh screen was used 

 on the sieve basket. As the device was designed 

 specifically to take quantitative samples of the 

 deep-burrowing, stout razor clam, Tagelus ple- 

 beius Solander, for a secondary production study, 

 the 2.5 mm mesh was adequate for sampling small 

 bivalves and adults. Collections were made from 

 small aluminum or fiber glass boats (3.8-4.9 m 

 long). The system was also tested at a beach site 

 where the water pump was placed on the shore 

 while the operator carried the sampler into adja- 

 cent intertidal waters. Though possible, this 

 method lacks the mobility of sampling from a boat. 



Most bivalves were recovered in an undamaged 

 condition (Table 2). In no case was a bivalve 

 recovered with a shell broken such that shell 

 length could not be measured, though in some 

 specimens, small breaks in the shell allowed body 

 fluids to escape resulting in a specimen unsuitable 

 for an accurate biomass determination. The soft 

 shell clam, Mya arenaria, and the stout razor 

 clam have thin gaping shells that are much 

 more susceptible to damage than the hard clam, 

 Mercenaria mercenaria. Van Arkel and Mulder 

 (1975) reported that damage to small polychaete 



Table 2. — Condition of bivalve.? recovered in 1,575 separate 

 samples by the flushing-coring device. 



'Any break in the shell, no matter how small, that might allow loss of body 

 fluids. 



worms was no more than if a hand corer and 

 ordinary sieve had been used. Polychaetes were 

 not specifically examined for damage in the 

 present study. Many were collected in the sieve 

 basket and though some were obviously broken, 

 most were not. 



The maximum shell lengths (largest external 

 dimension of the shell) of the collected speci- 

 mens were 10.4, 9.15, 10.2, and 10.8 cm for 

 M. mercenaria, T. plebeius, Mya arenaria, and 

 Ensis directus, respectively. The heaviest bivalve, 

 Mercenaria mercenaria, was estimated to weigh 

 200-300 g. Several of the large M. mercenaria 

 were placed in the sediment and repeatedly 

 "sampled" from the sand flat to the sieve basket 

 to test the lifting capability of the system. The 

 test specimens were always easily raised. 



Direct visual observations in shallow water 

 indicate that no water or sediment escapes under 

 the outer pipe when in operation. The outer pipe 

 has a large diameter ( 16 cm ), thus the surface area 

 of a single sample (0.02 m^) is quite large. 



The main advantages of this system are its 

 mobility (as mobile as a small, shallow-draft boat) 

 combined with the fact that deep cores can be 

 taken and sieved in one fast process. Samples can 

 routinely be taken to a depth of 65-70 cm below the 

 sediment surface. Organisms can be immediately 

 placed in storage containers, ready to be processed 

 upon returning to the laboratory. The use of PVC 

 parts and removable sieve basket screens allows 

 the investigator to build and modify the device 

 to one's own specifications depending on the par- 

 ticular goals of the study. 



Acknowledgments 



I wish to thank Czeslaw A. Bartusiak for 

 providing assistance and a workshop in which 

 to construct the device. This project was sup- 

 ported in part by an Old Dominion University, 

 Institute of Oceanography (Norfolk, Va.). Grad- 

 uate Fellowship. 



384 



