(about 0.25 to 0.40 m^) surrounded by sand and 

 coral rubble (one at Midway, one at Kure). 



It was not our intent to characterize the mero- 

 plankton from each substrate — the collections 

 were too few for this; rather, we sought only a 

 general understanding of the types and numbers 

 of organisms that emerge from the lagoon floor. 



To begin each set of collections, we placed our 

 trap ( Figures 1, 2) in position between sunrise and 

 0800 h. First we buried the lower portion of the 

 metal frame in the sand and secured it with soil 

 anchors. A tight seal around the base of the trap 

 was judged critical to prevent entry by organisms 

 from the surrounding water. Next we attached the 

 net ( which had a 0.333-mm mesh) to this base and 

 allowed it to remain in position throughout the 

 day. We retrieved the net between 1730 h and 

 sunset, washed all materials into the cod end, then 

 removed the materials, and placed them in K/i 

 Formalin. The net, with an empty cod end in place, 

 was then reattached to the frame and left in place 



FiGliRE 2. — The meroplankton trap m place to sample or- 

 ganisms that emerge from sand in the lagoon of Midway Atoll. If 

 the trap had not been designed to exclude holoplankters, we 

 believe the collections would have included, among other holo- 

 plankters, calanoid copepods iAcartia sp.i that swarmed close to 

 the adjacent reefs by day. including at their bases, and dispersed 

 throughout the area at night. 



throughout the night. The following morning, 

 again between sunrise and 0800, the entire 

 trap — base as well as net — was retrieved, and the 

 collected organisms placed in preservative as be- 

 fore. Having thus completed one set of collections, 

 we moved to another site and repeated the proce- 

 dure. (We would have reversed the order of collec- 

 tions in some sets, e.g., nighttime first, if appreci- 

 able numbers of organisms had been taken by day; 

 as it turned out, however, essentially all or- 

 ganisms were taken in the nighttime samples, as 

 detailed in the Results.) 



Our trap worked as follows: Organisms rising 

 from the substrate inside the trap swam upward 

 through the small upper opening of the inner cone 

 and entered the space within the larger outer cone 

 (Figure lA). Some may have continued up into the 

 cod end, which floated above, but this had no bear- 

 ing on the collections. When the organisms re- 

 turned toward the sea floor all except those that 

 happened to descend through the small orifice of 

 the inner cone were trapped where the two cones 

 converged at their common base. In retrieving the 

 net, we reached in under the edge attached to the 

 metal frame and grasped the inner cone around its 

 smaller orifice, thus closing it. We then pulled this 

 out, thus everting the inner cone and producing a 

 diamond-shaped bag (Figure IB) with the orifice 

 closed in our grasp at one end and the cod end at 

 the other. We then towed the net back to the boat, 

 still enclosing the smaller orifice in our grasp, so 

 that, as we swam, all materials inside were swept 

 back into the trailing cod end. 



Results 



The organisms collected by our trap, day and 

 night, are listed in Table 1. The general absence of 

 organisms in the daytime collections was predic- 

 table, based on the many reports which have con- 

 cluded that the diel emergence of such forms is 

 primarily a nocturnal phenomenon (see references 

 listed above). Among organisms we observed 

 swarming close to reef structures in the vicinity of 

 our trap during the day were calanoid copepods 

 (most of them Acartia sp. I, mysids, and larval 

 fishes. Although such forms disperse in the water 

 column at night (Emery 1968; Hobson and Chess 

 1976, 1978), their absence from our trap collec- 

 tions is consistent with the contention that the 

 holoplanktonic forms associated in varying degree 

 with the reef are distinct from those organisms 

 that live by day in or on the substrate. 



277 



