496 



Fishery Bulletin 92(3), 1994 



Beginning in June, gill nets were fished for ap- 

 proximately four hours, once a week, until catches 

 revealed that bluefish were present in the river. The 

 nets were then fished two or three times a week. Fish- 

 ing was again reduced to one set per week in Sep- 

 tember when no further catches of bluefish were 

 made. When two weeks of fishing produced no juve- 

 nile bluefish, fishing was discontinued. Gill nets were 

 set in the Marsh River on 20, 27 June; 4, 11, 18, 25 

 July; 1, 6, 14, 16, 21, 22, 23, 27, 28, 30 August; 5, 6, 

 11, 12, 18, 20, 26 September; 3, 12 October 1990 and 

 20, 26 June; 9, 17, 22, 25, 29 July; 1, 6, 12, 15, 22, 29 

 August; 4, 5, 10, 12, 17 September; and 1, 8 October 

 1991. When bluefish were present, the net was in- 

 spected every 30 minutes without pulling and reset- 

 ting it. All fish were removed from the net, identi- 

 fied to species, and measured to the nearest mm. 



Juvenile bluefish were captured by beach seine on 

 10 August 1990 and 27 September 1991 at Kennebec 

 Point. The seine was constructed of 6.35-mm square 

 mesh and measured 1.2 m x 15.2 m. It was hauled a 

 horizontal distance of 152.4 m on each sampling date. 

 Fish were also collected at Kennebec Point on 22 

 August 1990 in a fyke net constructed of 6.35-mm 

 square mesh with two 15.2-m wings. The fyke net 

 was set in a dry channel at low tide, and a 3-hour set 

 was initiated when the incoming tide first reached 

 the hoops. Juvenile bluefish were captured at 

 Merepoint on 9 September 1991 by using the vari- 

 able mesh gill net described previously. 



Field and laboratory data recorded 



Fishing time was recorded during each sampling trip 

 to the Marsh River. The set was initiated when the 

 net was set, attached to stakes, and spread. Termi- 

 nation of the set occurred when the net was detached 

 from one stake. Temperature and salinity were re- 

 corded at the surface and bottom with a YSI Model 

 33 SCT Meter immediately following the beginning 

 of a set and just prior to the end of a set. Tide condi- 

 tion, time of slack water (if it occurred), and maxi- 

 mum depth noted from graduations on the SCT cable 

 were also recorded. 



Both individuals and schools (two or more gilled 

 in close proximity) of juvenile bluefish were captured 

 in the gill nets. The fish were stored in plastic bags 

 containing information on time of capture (to the 

 nearest 30 minutes), date of capture, and mesh size 

 of the capture panel. Fish were stored in an ice cooler 

 until returned to the laboratory. 



Within nine hours of capture, fish were wiped with 

 paper towels and weighed to the nearest 0.1 gram. 

 Both total length (TL) and fork length (FL) (Hubbs 

 and Lagler. 1958) were recorded to the nearest milli- 



meter. Stomachs were removed and frozen in small 

 scintillation vials. Heads were refrigerated until the 

 sagitta were removed. 



Daily growth rate was derived from the total in- 

 crease in length divided by the number of daily 

 growth increments counted. Total growth in length 

 is equivalent to size (FL) at capture minus the ap- 

 proximate size at hatching (2 mm, Deuel etal., 1966). 



Treatment and analysis of stomach contents 



Stomachs were thawed and submerged in 3% sea 

 water formalin for approximately 19 hours to harden 

 stomach contents. Stomachs were then rinsed sev- 

 eral times with sea water and their contents were 

 identified to the lowest taxon, counted, blotted on a 

 paper towel, and weighed to 0.0001 grams. Stomach 

 contents were then refrozen and freeze-dried ( Labconco 

 Freeze Dryer 8) for 48 hours at -40°C under a vacuum 

 of approximately 25 microns mercury (Hg). Dry weights 

 were recorded to the nearest 0.0001 gm. 



Three prey item indices were computed to reduce 

 potential bias associated with a single index ( Hynes, 

 1950; Windell, 1971; Friedland et al., 1988). We em- 

 ployed the methodology of Friedland et al. ( 1988) in 

 computing 1) the number of stomachs in which a 

 species occurred expressed as a) a percentage of the 

 total number of stomachs containing food ( F=percent 

 frequency of occurrence) and b) a percentage of the 

 total number of stomachs examined ( Fl=percent fre- 

 quency of occurrence); 2) the number of individuals 

 of each prey species expressed as a percentage of the 

 total number of food items (N=percent numerical 

 abundance); and 3) the wet weight of a species ex- 

 pressed as a percentage of the total wet weight of 

 food items (W=percent wet weight) and the dry 

 weight of a species expressed as a percentage of the 

 total dry weight of food items (Wl = percent dry 

 weight). 



Preparation and analysis of otoliths 



Saggita were removed from each fish within 24 hours 

 of capture. Excess tissue was removed with forceps. 

 Pairs of sagitta were stored in 7-mL scintillation vi- 

 als for 1-3 months prior to mounting and ageing. 

 Each sagitta was mounted concave side down on a 

 microscope slide with Epo-Kwick (Buehler Ltd., Lake 

 Bluff, Illinois). Commercial grade wet-dry sandpa- 

 per (600 grit or 24 p ) was used for initial coarse sand- 

 ing. Final sanding and polishing was performed with 

 9-p and 0.3-p wet-dry sandpaper respectively 

 (Buehler, Ltd). The addition of a small drop of glyc- 

 erine to the surface of the polished otolith assisted 

 in the resolution of daily growth rings. Three repli- 

 cate counts were performed during one day by one 



