gests the association has a selective advantage. This 

 behavior was not, however, a mechanism for the 

 association, because fishes remained near the object 

 in the absence of predators. 



Use of the object as a spawning substrate would 

 apply only to adult fishes. Adults, however, repre- 

 sented only a small portion of the total individuals 

 present. Adults of two species, Euthynnus lineatus 

 and Canthidermis maculatus, were in reproductive 

 condition. Euthynnus does not have attached eggs, 

 however (Calkins and Klawe, 1963), and no eggs 

 were found on any of the floating materials. 



The cleaning-station hypothesis suggested by 

 Gooding and Magnuson (see footnote 1 ) could apply 

 only to some of the fishes. Canthidermis maculatus 

 alone showed behavior that could possibly be inter- 

 preted as cleaning. This species was taken in only 

 14 of the 70 collections. Except in the collection 

 made farthest from shore, where Canthidermis and 

 Balistes polylepis were the only species present, no 

 differences in the species composition were evident in 

 the collections that contained Canthidermis. Thus, 

 if Canthidermis regularly consumes parasites of other 

 fishes the activity does not appear to influence the 

 presence of these fishes. 



Artificial reefs established in sandy locations 

 rapidly attract groups of fishes that would not other- 

 wise inhabit these areas (Carlisle et al., 1964). The 

 artificial reef provides the habitat requirements for 

 certain fishes in an otlierwise unsuitable area. Simi- 

 larly, a drifting object may provide a suitable habitat 

 for inshore fishes that have pelagic juvenile stages 

 or that have become displaced from shore. This 

 explanation seems to be plausible for the presence of 

 Abudefduf saxatilis, Blenniolus breripinnis, Batistes 

 polylepis, and other species. The very restricted 

 size range of Chromts atrilohata and Pseud upcneus 

 grandisquamis suggests that these species are avail- 

 able for association during a limited period. Many 

 of the Pseudupencus were near the size at which 

 metamorphosis takes place. Approaching metamor- 

 phosis was indicated by the slight color changes in 

 some of the individuals and by pronounced changes 

 in coloration after the fish were kept in a shipboard 

 aquarium 34 hours. Possibly large premetamorphic 

 juveniles maj' be attracted to objects because of 

 changes associated with metamorphosis; for these 

 fish the object may represent an inshore or non- 

 pelagic habitat. 



In summary, we found little evidence to support 

 the mechanisms postulated by other authors. We 



have suggested two mechanisms: (1) fishes are at- 

 tracted to drifting materials because the object 

 functions as a schooling companion, and (2) for 

 species not adapted to a pelagic life and others under- 

 going a change from a pelagic to other modes of 

 existence, drifting materials may function as a sub- 

 stitute for a reef or other substrate. In both situa- 

 tions the object may have the same function, that is, 

 provide a visual stimulus in an optical void. 



The occurrence of juvenile fishes beneath flotsam 

 was much more frequent than that of adults. That 

 some species, as Chromis atrilohata, Pseudupeneus 

 grandisquamis, and Agonostomus monticola, are pela- 

 gic only as juveniles can explain the absence of the 

 adults. Of the species that are pelagic as juveniles 

 and adults, the juveniles were in tlie vicinity of an 

 object for longer periods and remained closer to the 

 object than did the adults. Owing to their larger 

 size and faster swimming speed, adults are probably 

 less susceptible to predation. Thus, for adult fishes 

 the selective advantage of maintaining a close associ- 

 ation with a drifting object may be small. It is also 

 possible that development is accompanied by an 

 increase in the specificity of the responses of school- 

 ing fishes to other schooling companions. The val- 

 ence of flotsam as a schooling object would then be 

 lowered and intermittent association with drifting 

 objects might be expected. 



ACKNOWLEDGMENTS 



Richard R. Whitney gave editorial and statistical 

 assistance. Frank J. Hester navigated our vessel 

 during the fall cruise, assisted in making the collec- 

 tions, and took the underwater motion pictures. 

 Donald C. Aasted de\eloped the procedure for han- 

 dling the small purse seine and assisted in making the 

 collections. Frederick H. Berry, Bureau of Com- 

 mercial Fisheries Biological Laboratory, Brunswick, 

 Ga., and Richard H. Rcsenblatt, Scripps Institution 

 of Oceanography, La JoUa, Calif, helped identify the 

 fishes. Rosenblatt also read the manuscript and 

 offered suggestions. 



LITERATURE CITED 



Alverson, Dayton L., and Norman J. Wilimov.sky. 



196.3. Catching fish— 1990 style. Fish. News 19(10): 

 1, 5, 9, and 12. 

 Atz, Jambs W. 



1953. Orientation in schooling fishes, hi Proc. Conf. 

 Orientation Anim., Off. Nav. Res., W.ashington, D.C., 

 Section 2: 103-130. 



FLOTSAM IN OFFSHORE WATERS 



27 



