On the other hand, A. tonsa typically occurs in the open and less 

 turbid waters in the Chesapeake Bay system. Apparently, this species 

 is not stimulated to increase feeding activity by low particle con- 

 centrations which may be uncharacteristic of its natural habitat 

 (Herman, Mihursky, and McErlean, 1968). Because these species are 

 nonselective suspension feeders, the much reduced uptake of radio- 

 active phytoplankton observed with increasing concentrations of sus- 

 pended solids can be accounted for simply by the ingestion of increas- 

 ing numbers of unlabeled particles, since the gut of both species 

 was full during all experimental treatments. 



Our results are not directly comparable with those of Wilson 

 (1973). The plastic beads he fed to A. tonsa were optically sized, 

 and particle uptake was measured according to size-frequency distri- 

 bution. The median size (diameter) of food (M. lutheri) particles 

 used in our experiments was 8 to 10 micrometers (optically measured) . 

 This size is below the minimum bead size (13.9 micrometers) which 

 Wilson found that A, tonsa ingested. However, as Wilson noted, the 

 artificiality of his feeding study with plastic beads may have 

 eliminated the variables of particle shape and palatability. 



Wilson (1973) demonstrated that the passive filtering behavior of 

 A. tonsa could not account for the selection of a narrow, variable 

 size range of plastic beads. However, passive filtering did exist 

 because the minimum selected food size did not change with body size. 

 This is typical of a process in which particles are not grasped 

 individually, but are concentrated against closely spaced setae and 

 are handled collectively. Bead sizes outside the selected range 

 were ingested in the same proportion as they were present in the 

 water column. Therefore, both selective grasping and nonselective 

 filtering (taking all particles indiscriminately) were operating at 

 the same time. 



The median sizes (Stokes' diameters) of sediments used are 

 reported by weight and not by optical size-frequency distributions. 

 These were: Si02 (17 micrometers), <15 micrometers Si02 (6.2 micro- 

 meters), natural sediment (<0.8 micrometer), and Fuller's earth 

 (<0.5 micrometer). These diameters are equivalent to those of a 

 sphere with the same specific gravity which has settled through a 

 column of water at some specified temperature. Since the Fuller's 

 earth particles and those of the natural sediment were not spherical, 

 there may have been a significant departure between values given as 

 equivalent diameter (Stokes) and actual dimensions. In fact, the 

 natural particle-size distribution of the Patuxent River sediment was 

 destroyed before size analysis by the analytical procedure required 

 to remove the large quantity of organic matter (approximate specific 

 gravity 1.1) from the inorganic mineral particles (approximate 

 specific gravity 2.6). In addition, both the natural sediments and 

 the Fuller's earth formed fairly large agglomerates when suspended in 

 saline water (microscopic observation) . 



31 



