FISHERY BULLETIN VOL 77. NO 3 



bergen il960), there may be an obvious delay in 

 the attack on a potential prey if it is new to the 

 predator. No such delay was seen in these experi- 

 ments using semistarved fish. It is unlikely that P. 

 mucosa has developed a defense mechanism which 

 is specific in its action only to marine fish. The 

 ability of F. heteroclitus to consistently consume 

 P. mucosa probably reflects the predaceous mum- 

 michog's lack of sensitivity or its ability to over- 

 come the irritation or unpalatability of this worm. 



Cypnnodon variegatus always rejected the phyl- 

 lodocids and rarely investigated the worm without 

 attempted ingestion. The small, terminal mouth 

 of this fish, with its large tricuspid teeth and pro- 

 tractile premaxillaries, was quite efficient at 

 quickly devouring all nonphyllodocid polychaetes 

 offered to the fish during these experiments. 



Cynoscion regalis is primarily an active pelagic 

 predator (Table 1). There has been some research 

 concerning the feeding habits of juvenile sciaenids 

 (Thomas 1977; Chao and Musick 1977) which 

 found that small C. regalis feed mainly on mysids, 

 copepods, and small fish. Annelids do form, how- 

 ever, a portion of the weakfish's diet (Table 1). 

 Bigelow and Schroeder (1953) noted that the diet 

 of C. regalis varies with locality and availability of 

 prey. Small weakfish reject P. mucosa as a food 

 item. Cynoscion regalis showed active investiga- 

 tory behavior which usually consisted of tapping 

 or bumping the sinking worm with its snout. Di- 

 rect contact often resulted in a rapid shunning of 

 the worm by the fish. This may indicate the pres- 

 ence of sensitive nares chemoreceptors. The 

 juvenile fish would not be a major threat to these 

 worms even were they readily available. 



Preliminary work indicates that antipredation 

 responses are active in P. maculata.E. sanguinea, 

 Phascoleopsis gouldi. Stylochus zebra, and Lineus 

 ruber. All these organisms, except the sipunculid, 

 secrete large quantities of mucus. The epidermis 

 of many sipunculids is densely packed with gland 

 cells (Tetry 1959) and some secretion from these 

 glands may serve to protect the animal from pre- 

 dation. Stylochus zebra, is a commensal of pagund 

 crabs; abundant production of mucus by this worm 

 may tend to keep this relationship commensal. 



The role of secretory defense mechanisms is well 

 established in many species of marine animals 

 (Graham 1957; Thompson 1960, 1969; Bakus 

 1 968 ), but many questions concerning the broader 

 aspects of antipredational responses remain un- 

 answered. Are antipredation responses reflected 

 in the composition of marine benthic com- 



munities? How does effectiveness of antipredation 

 mechanisms vary with size of predators or degree 

 of predator satiation? In similar-sized predators, 

 what differences allow one species to prey on a 

 given organism and not on the other? What 

 changes in diet would be found if feeding studies 

 involving analyses of stomach contents typically 

 were extended to include all size classes offish? 

 Does previous exposure to an antipredation 

 mechanism produce "learning" in potential 

 marine predators? 



In responding to slow-moving predators many 

 potential prey species have evolved escape reac- 

 tions (Doering 1976). In dealing with highly 

 mobile fish predators, many species of potential 

 prey have developed such defense mechanisms as 

 protective secretions. Lagler et al. (1977:142) 

 stated, "In general the esophagus (offish) is so 

 distensible that it can accommodate anything that 

 the fish can get into its mouth . . . . " With the dis- 

 covery of the repulsive characteristics of certain 

 phyllodocids, the indications of antipredation 

 mechanisms in a sipunculid and turbellarian re- 

 ported here, added to what is known of nemerteans 

 and opisthobranchs, it is clear that a closer exami- 

 nation must be made of interspecific molecular 

 interactions which occur within marine com- 

 munities. 



ACKNOWLEDGMENTS 



I gratefully acknowledge the many helpful 

 suggestions concerning this manuscript which 

 were received from M. R. Carriker, F. C. Daiber, B. 

 Brown, R. Palmer, and L. Williams. I also thank L. 

 Watling and G. Entrot for constructive criticisms 

 of an earlier draft of this paper. Harlan Dean, G. 

 Entrot, S. Howe, P. Nimeskern, F. Prezant, J. 

 Vargas, and W. Wehling helped in the collection of 

 animals used in this study, and I thank them for 

 their efforts. The use of the RV Clione and 

 facilities at the Marine Science Institute, Nahant, 

 Mass., were kindly supplied by N. W. Riser. 

 Thanks also to P. Savage for typing the manus- 

 cript. 



LITERATURE CITED 



B.AKL'S. G, J. 



1966 Some relationships of fishes to benthic organisms on 



coral reefs. Nature iLond.l 210:280-284, 

 1968, Defensive mechanisms and ecology of some tropical 

 holothurians Mar Biol (Berl I 2:23-32. 



614 



