spinosa which was never seen with more than its head protruding from 

 the abandoned worm tube in which it lived. The grunts Haemulon 

 aurolineatum and H. f lavolineatum leave the reef at night and 

 apparently range over a wide area of the adjacent sand plain. 



Activity cycles (table 1, column 6) also provide a means by which 

 more species can use a limited volume of habitat. Some species are 

 nocturnal, moving off the reef just after sunset and returning to it 

 just before daybreak; others are diurnal. We did not detect any 

 species that were truly crepuscular, i.e. active only during daylight 

 and dusk, although some diurnal species seemed to be more active 

 shortly before sunset. It was difficult to determine whether some 

 species were diurnal or nocturnal. For example, Hypoplectrus puella 

 hovers around the reef during the daytime but it was not seen at night 

 and we could not tell whether it takes refuge in holes at night or 

 moves off the reef to feed. At or immediately preceding the departure 

 of night active species, the daytime species retire to their nocturnal 

 retreats and in the morning the process is reversed. The changeover 

 periods are short and definite and generally span less than fifteen 

 minutes (for details of the sequence during the changeover see Talbot 

 and Collette, this volume). 



The inactive periods during which the fishes are not actively feeding 

 or engaging in any other special activity must be considered in any 

 discussion of space sharing. There are a variety of ways in which 

 fish spend their resting hours. Some such as the species of Chromis 

 insinuate themselves into the shelter of crevices or holes in the 

 reef - others such as the cardinalf ishes and squirrelf ishes gather 

 into small groups in caves or around the bases of coral colonies. 

 Apogon conklini takes refuge in the spines of a sea urchin. Some 

 wrasses burrow into the sand and so on. During the inactive periods 

 the fish relies heavily on structural and behavioral adaptations for 

 defense against predators. When the fish are actively feeding they 

 are alert and prepared to flee or seek shelter to avoid danger but 

 in their resting places they are unprepared for flight and hence depend 

 on camouflage and remaining motionless so that predators do not locate 

 them. Many species school during the day and disperse to feed at 

 night. Schools in this case appear to be the quiescent refuge. It 

 appears that defense mechanisms are generally more closely related to 

 the inactive periods than to the active periods. 



It is a common impression that tropical species have no true annual 

 seasons but there is a growing body of evidence that this is not true. 

 Although the seasons may not be well-marked, they are none the less 

 present. During our two week study of the observation reef we found 

 no evidence of spawning activity in any species although we suspected 

 that one or more cardinal fishes were brooding eggs in their mouths 



VI~2fi7 



