HOBSON: FEEDING RELATIONSHIPS OF FISHES 



DISCUSSION 



The habits of fishes on Kona reefs exempHfy the 

 habits of fishes on coral reefs around the world. 

 The following discussion relates these habits to 

 the evolution of fishes on modern tropical reefs, 

 stressing the selective pressures that have shaped 

 the diverse array of forms coexisting on these reefs 

 today. I refer to some of these forms as more ad- 

 vanced, or specialized than others, even though all 

 are products of an equally long evolution, and each 

 is well adaptated to its own specific way of life. 

 Some, nevertheless, have diverged more than 

 others from the generalized carnivores that gave 

 rise to them all, and in this fact lies the basis for 

 the discussion. 



The categories erected for presentation have in- 

 distinct, overlapping limits, and some species are 

 discussed under one category, rather than 

 another, quite arbitrarily. Nevertheless, the 

 synthesis presented, though an oversimplifi- 

 cation, provides a frame of reference within 

 which new information may be assessed. Reem- 

 phasizing a point made above, this report deals 

 only with individuals of the various species that 

 behave as adults. 



Coral Reefs as a Habitat for Fishes 



Most fishes that inhabit coral reefs are among 

 the more recently evolved teleosts (Schaeffer and 

 Rosen, 1961; and others). Indeed, much of the di- 

 versity among higher teleosts expresses adapta- 

 tions to reef habitats. Of the fishes observed along 

 Kona transect lines (Table 7), 98.5% are 

 acanthopterygians.^ 



To properly appreciate the relation of modern 

 coral-reef fishes to their habitat, one should be 

 familiar with the history of tropical reefs. The 

 following outline is based on Newell (1971). 



The evolution of tropical reefs can be traced 

 through a fossil record that reaches back into the 

 Precambrian. By the Mid-Ordovician, over 400 



'Because most of the transect counts were made in daylight, 

 there is a bias toward the more advanced forms in numbers of 

 species (the greater incidence of diumally secretive habits 

 among the more primitive forms, and of diumally exposed habits 

 among the more highly evolved forms, is discussed below). Even 

 so, however, the preponderance of acanthopterygians is over- 

 whelming, especially if one also considers numbers of individu- 

 als. On Kona reefs such advanced groups as the labrids, 

 pomacentrids, and acanthurids are among the species vnth the 

 largest numbers of individuals. And although among the 

 nonacanthopterygians the numerous muraenid eels are not 

 properly represented in the counts, neither are such noctvimal 

 acanthopterygians as the numerous holocentrids. 



million years ago, animal communities had be- 

 come associated with coral-algal reefs. A succes- 

 sion of reef comftiunities then evolved during sub- 

 sequent geological history, each with its own 

 characteristic assemblage of animals, and each 

 achieved marked stability before crashing into ob- 

 livion during worldwide environmental upheav- 

 als. Between each of these periods of stable reef 

 communities, a long time passed without known 

 reefs. 



The scleractinian corals, which dominate mod- 

 ern reefs, first appeared during the Triassic, and 

 by Jurassic times, about 150 million years ago, the 

 lithothamnion-scleractinian reef community was 

 well established. Significantly, the teleostean 

 radiation also began during the Jurassic (Gosline, 

 1971), indicating that their history may closely 

 interrelate with that of the lithothamnion- 

 scleractinian reef community. But Smith and 

 Tyler (1972) suggested that the preacanthop- 

 terygian teleosts and their forebearers were 

 maladapted to reef conditions. They contended 

 that fishes entered reef habitats only upon acquir- 

 ing certain of the morphological advances that 

 marked the first appearance of acanthopterygians 

 early during the Cretaceous, over 100 million 

 years ago. Newell (1971), on the other hand, be- 

 lieved that fishes have had a much longer history 

 as reef inhabitants. He attributed their absence in 

 the fossil record of early reefs to their skeletal 

 remains having been "destroyed by scavengers 

 that abound in this strongly oxidizing environ- 

 ment." 



It is unquestioned, nevertheless, that early 

 acanthopterygian fishes — the Beryciformes 

 — were better adapted than were their predeces- 

 sors for reef habitats. Their increased success 

 probably was based mostly on increased ma- 

 neuverability and a more adaptive feeding 

 mechanism — the features which Smith and Tyler 

 (1972) felt were especially suited for coral reefs. 

 Patterson (1964) underscored this point when he 

 concluded that most skeletal differences between 

 acanthopterygians and their primitive elopidlike 

 ancestors resulted from changes that permitted 

 the fish better maneuverability: most significant, 

 the fins, given increased rigidity by replacing the 

 anterior soft rays with spines, were more effec- 

 tively positioned, and the body was shortened and 

 deepened. The advances these fishes made in their 

 feeding mechanism was especially significant, as 

 attested by Schaeffer and Rosen (1961), who 

 stated: "It is primarily the acanthopterygian 



1017 



