461 



Using bone measurements to estimate the 

 original sizes of bluefish (Pomatomus saltatrix) 

 from digested remains 



Anthony D. Wood 



Box 200 



Graduate School of Oceanography 



University of Rhode Island Bay Campus 



South Ferry Rd. 



Narragansett, Rhode Island 02882 



E-mail address: awoodigigso un edu 



The ability to estimate the original 

 size of an ingested prey item is an 

 important step in understanding the 

 community and population structure 

 of piscivorous predators (Scharf et al., 

 1998). More specifically, knowledge 

 of original prey size is essential for 

 deriving important biological infor- 

 mation, such as predator consumption 

 rates, biomass of the prey consumed, 

 and selectivity of a predator towards 

 a specific size class of prey (Hansel et 

 al., 1988; Scharf et al., 1997; Radke 

 et al., 2000). To accurately assess 

 the overall "top-down" pressure a 

 predator may exert on prey commu- 

 nity structure, prey size is crucial. 

 However, such information is often 

 difficult to collect in the field (Trippel 

 and Beamish, 1987). Stomach-con- 

 tent analyses are the most common 

 methods for examining the diets of 

 piscivorous fish, but the prey items 

 found are often thoroughly digested 

 and sometimes unidentifiable. As a 

 result, obtaining a direct measure- 

 ment of prey items is frequently 

 impossible. 



Because of the problems of recon- 

 structing original prey size directly 

 from prey remains, numerous meth- 

 ods involving correlations between 

 measurements of specific morphologi- 

 cal features of the prey and prey size 

 (length) have been devised. External 

 body measures such as eye diameter, 

 and caudal peduncle depth (Crane et. 

 al., 1987; Serafy et. al., 1996; Scharf, 

 et. al., 1997), as well as numerous in- 

 ternal measures such as pharyngeal 

 arch length (Fickling and Lee, 1981; 

 Mclntyre and Ward, 1986; Radke et. 



al., 2000), vertebral diameter (Pikhu 

 and Pikhu, 1970; Feltham and Mar- 

 quiss, 1989), and a variety of skeletal 

 bones (Newsome, 1977; Hansel et. al., 

 1988; Scharf et. al., 1998) have been 

 used to generate models for predict- 

 ing original prey size. 



The bluefish (Pomatomus salta- 

 trix) is a voracious piscivore and is 

 among the top predatory fish species 

 in the western North Atlantic Ocean 

 (Buckel et. al., 1999). Bluefish are an 

 important fish both commercially and 

 recreationally, and over the past two 

 decades stocks off the eastern coast 

 of the United States have experi- 

 enced a dramatic decline. From 1978 

 through 1996, the commercial land- 

 ings and spawning stock biomass of 

 bluefish declined by over 60'~A (Fahay 

 et. al. 1 ). A variety of mechanisms 

 have been proposed to explain this 

 dramatic decline, including intense 

 predation by large apex predators. 

 It is known that bluefish act as an 

 important prey species for a num- 

 ber of apex predators in the North 

 Atlantic, most notably the shortfin 

 mako (Isurus oxyrinchus). Stillwell 

 and Kohler (1982) sampled 399 ma- 

 kos from 1972-79 and found that 

 bluefish made up 85% of the diet by 

 volume. The mako diet has recently 

 been reviewed and it appears that 

 the incidence of bluefish in the diet 

 has increased (assume 1 mL = l g for 

 flesh) to 94% of their diet by weight 

 (Wood et al. 2 ). Bluefish have also 

 been found to be important in the di- 

 et of bluefin tuna (Thunnus thynnus) 

 (Chase, 2002), swordfish (Xiphias 

 gladius) (Stillwell and Kolhler, 1985), 



blue shark (Prionace glauca) (Kohler, 

 1989), and the thresher shark (Alo- 

 pias vulpinis) (Kohler 3 ). 



The motivation for this study came 

 from field sampling shortfin mako 

 (Isurus oxyrinchus) stomach contents 

 where it was observed that bluefish 

 jaw bones and various other skull 

 bones were often intact, even if the 

 rest of the prey fish was digested. 

 To generate accurate estimates of 

 the original prey size a series of pre- 

 dictive equations was generated by 

 regressing bluefish skull bone mea- 

 surements with the fork length (FL) 

 and total length (TL) of the fish. Five 

 skull bones were chosen to obtain 

 measurements for the relationships: 

 the dentary, maxilla, premaxilla, 

 opercle, and cleithrum (Fig. 1). These 

 five bones were chosen because they 

 are strong bones (with the exception 

 of the opercle), covered by extensive 

 musculature, and assumed to be re- 

 silient to digestion. 



Materials and methods 



During June-September of 2000 and 

 2001, bluefish were collected by rod 

 and reel and by otter trawl in Narra- 

 gansett Bay, RI, and at bluefish fish- 

 ing tournaments along the northeast 

 coast of the United States from Ocean 



1 Fahay. M. P.. P. L. Berrien, D. L. John- 

 son, and W. W. Morse. 1999. Essential 

 fish habitat source document: Bluefish, 

 Pomatomus saltatrix, life history and 

 habitat characteristics. NOAA Tech. 

 Memo. NMFS-NE-144, 68 p. U.S. 

 Department of Commerce, NOAA, 

 NMFS-NEFSC. Woods Hole, MA. 



2 Wood, A. D., B. Wetherbee, N. E. Kohler. 

 F. Juanes and C. Wilga. 2004. In 

 prep. Predator prey interaction between 

 the shortfin mako (Isurus oxyrinchus) 

 and bluefish (Pomatomus saltatrix). 



3 Kohler, N. E. 2001. Personal commun. 

 NMFS Narragansett lab, 28 Tarzwell 

 Drive, Narragansett, RI 02882. 



Manuscript submitted 4 February 2004 

 to the Scientific Editor's Office. 



Manuscript approved for publication 



21 December 2004 by the Scientific Editor. 



Fish. Bull. 103:461-466 (2005). 



