No erosion of the flat was evident during-; the tests at the two 

 lower current velocities, an elapsed time of at least 8 hours. However, 

 after about it hours' experimentation at a current velocity of 25 cm/sec 

 a very small accumulation of sand particles was noted on the wooden flume 

 bottom just downstream from the "flat". The flat itself appeared to be 

 unchanged, Hjulstrom's (1939) curve for erosion and deposition of uniform 

 material shows that the velocitj'^ required to erode sediments is high for 

 coarse particles, lowest for medium sand (about 20 cm/sec), and rises 

 again as the particles decrease in diameter. This apparent contradiction 

 is due to the fact that the curve of Hjulstrom refers to uniform sediment 

 and hence does not a:,-'ply to the flat in the present experiment. The large 

 proportion of silt and very fine sand, together with organic detritus, 

 Drobably gave the latter sediment sufficient cohesiveness to withstand 

 erosion. The relatively high velocity of current required to erode clay is 

 a comparable examole. 



The results of the 28 runs are given below in h sections labeled 

 with the appropriate current velocities. In the graphs (Figs.2-U) the clams 

 of similar behavior have been placed in four size groups instead of seven, 

 for clarity and convenience. The results are oresented in greater detail 

 in tables 1 and 2, 



Mean current velocity of 13 cm. per second 



Clams of all seven size grouns were moved to a similar extent by a 

 current flowing 13 cm, per second. A fe\^ of each size group were dispersed 

 downstream a few inches at a time, most proceeding no farther than half the 

 length of the flat. The predominant type of movement was slow and intermit- 

 tent rolling . Sliding occurred rarely and never more than one-half inch. 

 At the end of 30 minutes only 2.3?o of the seven groups of clams were in the 

 sediment trap. The smallest clams appeared to be moved most raoidly as the 

 percentage of them landing in the trap would indicate (fig, 2). 



In comparing the burrowing rates of the various size groups, it was 

 noted that the smaller the clams, the faster they were able to burrow as 

 shown by the curves in figure 2. Of the 2-7 mm, clams.. 20^ were burrowed 

 in 5 minutes, ^9% in 10 minutes, and 8h% at the end of 30 minutes. In con- 

 trast are the 20-22 mm, clams, only 2% of which burrowed in l5 minutes and 

 only IiO^ in 30 minutes- The intermediate size groups conformed to this 

 general pattern, the 8-13 mm, clams starting slowly but equalling the 2-7 mm. 

 clams after 30 minutes, while the lii-19 mm,, clams started more slowly and 

 66;? burrowed in 30 minutes (fig» 2), 



Of the seven size groups of clams combined, 72,5^ were able to burrow 

 in 30 minutes, while 2,3^ were washed into the sediment trap and 25.2^ 

 remained inactive on the surface of the flat. 



