334 



Fishery Bulletin 93(2), 1995 



as food is withheld. A decrease in the thickness of the 

 intestine during starvation has been noted for other 

 fishes (Kostomarova, 1962; Nakai et al., 1969; Umeda 

 and Ochiai, 1975; Ehrlich et al., 1976; O'Connell, 1976; 

 Theilacker, 1978, 1986; Kashuba and Matthews, 1984; 

 Boulhic and Gabaudan, 1989; Oozeki et al., 1989; 

 Theilacker and Watanabe, 1989; McFadzen et al., 1994), 

 including the Altantic cod, Gadus morhua L. (Yin and 

 Blaxter, 1987), which is closely related to walleye pol- 

 lock. In this study we describe a midgut cell height in- 

 dex for laboratory-reared walleye pollock. 



Shrinkage of larval fish captured in a net and pre- 

 served aboard ship differs from that observed in the 

 laboratory (Blaxter, 1971; Theilacker, 1980, 1986; 

 Hay, 1981; McGurk, 1985; Jennings, 1991). The 

 amount a larva shrinks may be dependent on size, 

 the duration of handling (time in the net and how 

 soon after death it is preserved), and the type of pre- 

 servative used. Preservative and gear- related (net) 

 shrinkage have been examined for larvae of a vari- 

 ety of fish species (Theilacker, 1980, 1986; Hay, 1981, 

 1982; Fowler and Smith, 1983; Tucker and Chester, 

 1984; McGurk, 1985; Radtke, 1989; Kruse and Dalley, 

 1990; Jennings, 1991; Hjorleifsson and Klein- 

 MacPhee, 1992). Jennings (1991) found that the 

 magnitude of shrinkage differed among species and 

 concluded that a correction factor for each species must 

 be determined. To relate our laboratory observations 

 to the field, we derived shrinkage indices for larval 

 walleye pollock subject to net treatment and several 

 preservatives. To determine the utility of the midgut 

 cell height index in the field, larval walleye pollock were 

 collected in Shelikof Strait, and their nutritional con- 

 dition was assessed from experimental results. 



Methods 



Laboratory rearing 



Adult walleye pollock were collected from Shelikof 

 Strait, Gulf of Alaska, in April of 1990 to 1993. The 

 fish were spawned aboard ship and the fertilized eggs 

 flown to Friday Harbor Laboratories, University of 

 Washington, in 1991 and to the Alaska Fisheries 

 Science Center in 1990, 1992, and 1993. We raised 

 the larvae in 120-L black fiberglass circular tanks 

 with clear plastic covers and used a 16-h daylight 

 cycle. Seawater temperatures were maintained at 

 6°C which are typical in May in Shelikof Strait when 

 larvae initiate feeding (Kendall et al., 1987). Each 

 year there were two treatments: one tank contained 

 larvae into which prey were added (fed tank), the 

 second contained larvae that were never offered prey 

 (starved tank). Prey consisted of rotifers, Brachionus 



plicatilis, at 10/mL and copepod nauplii, Acartia sp., 

 at a minimum of 1-2/mL. Rotifers were raised on 

 algal diets of Isochrysis galbana and Pavlova lutheri, 

 which are high in unsaturated fatty acids (Nichols 

 et al., 1989). The dinoflagellate, Katodinium rotun- 

 datum, was also added as prey for the rotifers and 

 copepods. Ammonia levels were low for both treat- 

 ments (<0.4 ppm). We sampled larvae from both the 

 fed and starved tanks every day or every other day. 



Calibration of midgut cell height 



For the histological analysis, walleye pollock larvae 

 were preserved in either Bouin's solution which was 

 replaced with 70% ethanol 24 to 48 h later or in Z- 

 Fix (solution of 10% formalin with zinc and buffers 

 added 1 ). Larvae were processed with standard his- 

 tological procedures; they were dehydrated in a bu- 

 tyl alcohol series, embedded in paraffin wax, seri- 

 ally sectioned at 6 pm in the sagittal plane, and 

 stained with hematoxylin and eosin. We measured 

 the mucosal cell height of the anterior dorsal por- 

 tion of the midgut at 400x magnification (Fig. 1 ). This 

 area was chosen because it exhibits little gut folding 

 which can make the midgut cell height measurement 

 too variable to be useful. We measured three to six 

 neighboring cells (with clearly defined nuclei, base- 

 ment membrane, and microvilli) from the top of the 

 basement membrane to the top of the microvilli and 

 recorded the average height. 



Fixative effects on larval length 



To determine the shrinkage of larvae placed directly 

 into preservative (laboratory shrinkage), we mea- 

 sured the standard length (SL, tip of upper jaw to end 

 of notochord, to nearest 0.08 mm) of live larvae sampled 

 from the fed tank and placed them individually into 

 Bouin's solution, 5% formalin, Z-Fix, or 95% ethanol. 

 Bouin's solution was changed to 70% ethanol 24 to 48 

 h later. Final size of larvae was determined one year 

 after preservation. Final size of ethanol-fixed larvae 

 was measured in distilled water; larvae preserved in 

 the other fixatives were measured in the fixative. 



Net-treatment and subsequent fixative 

 effects on larval length 



To examine the effect of net treatment on larval length, 

 a larva was sampled from the fed tank and its stan- 

 dard length was measured. It was then placed in a small 

 net and submerged in a tank of 6°C seawater that re- 

 circulated through the net to simulate a towed sam- 



Anatech, Ltd., Battle Creek, MI. 



