METHODS 



Larval Settlement Plates 



The settlement substrates were made of Lexan, which is a non- 

 toxic plastic used in many kinds of experiments on chemically 

 sensitive marine organisms, such as larvae and phy top lank ton. 

 Disk-shaped plates (24 cm in diameter) were used in order to 

 present a similar leading edge to all oncoming flows. Four 

 different types of plates were created: thick-elevated, thin- 

 elevated, thick-benthic, and thick-elevated-manganese. The thick 

 plates were 1-cm thick and the thin plates were 1.5-mm thick. 

 Elevated plates were supported 6 cm above the seafloor on three 

 legs. Thick-benthic plates were set directly on the seafloor. 

 Thick-manganese plates were covered with a fine powder (silt-size 

 particles) of f erromanganese crust that had been collected 

 (unpreserved) previously at a nearby site on Cross Seamount. The 

 powder was affixed to the plates with Tile Clad, a non-toxic 

 adhesive in general use for marine aquaria. 



The two plate thicknesses were selected on the basis of 

 laboratory studies, performed in the 20-m Flume at Woods Hole 

 Oceanographic Institution (see Butman, Chapman, Geyer, and 

 Trowbridge, in press, for description of flume). Flow over a 

 plate was visualized by releasing dye upstream from its leading 

 edge. Visualizations were connected at boundary shear velocities 

 (u*) between 0.3 and 1.0 cm s -1 which are within the range of those 

 found in the deepsea (u* is commonly less than 1 cm s -1 in deep-sea 

 boundary layers; Chriss and Caldwell, 1982; Grant, Williams and 

 Gross, 1985). In this range of shear velocities, the flow 

 separated and formed a recirculating eddy at the leading edge of a 

 1-cm-thick plate supported 6 cm above the flume bottom. No flow 

 separation occurred over the 1.5-mm-thick plate, and a well-behaved 

 boundary layer developed downstream from the leading edge. The 

 eddy over the thick plate was very small at low shear velocities, 

 but extended up to 3 cm downstream from the leading edge at higher 

 values of u*. The boundary layer reattached downstream from the 

 eddy and attained a thickness similar to the boundary layer over 

 the thin plate. Eddies forming at the edge of benthic-thick plates 

 (those sitting on the flume bottom) were smaller than the eddies 

 over elevated-thick plates (supported 6 cm above the flume bottom) 

 at corresponding shear velocities. 



If larvae do respond to the flow regime over substrates, then 

 two settlement patterns can be predicted from these flume 

 observations. Larvae would be expected to settle in densities that 

 vary as a function of distance from the leading edge of a plate, 

 due to the growth of the boundary layer downstream. They also may 

 settle in different densities at the leading edge of a thick plate 

 (due to the presence of an eddy) than at the leading edge of a thin 

 plate. 



254 



