Midgut Cell Height Defines Nutritional Status of 



Laboratory Raised Larval Northern Anchovy, 



Engraulis mordax 



Gail H. Theilacker and Yoshiro Watanabe 



ABSTRACT: The height of the midgut mucosal 

 cells was developed as a diagnostic index to assess 

 the past feeding history of larval northern anchovy 

 and to estimate starvation mortality of larval fishes 

 in the sea. The index was compared to traditional 

 histological indices and tested for field use. The 

 height of the mucosal cells was a sensitive index, 

 yielding reliable estimates of larval condition. Be- 

 cause it was sensitive, easy to measure and depend- 

 able in formalin-fixed tissue, this diagnostic index 

 should be useful and practical for field work. Addi- 

 tionally, unlike other diagnostic criteria, it resisted 

 the effects of autolysis (withstood prolonged time 

 in the collecting net). Also included is a discussion 

 of the growth characteristics of northern anchovy 

 larvae that experienced a delay in feeding. 



Larval mortality may determine recruitment of 

 young fish to a fish stock. A major cause of larval 

 mortality is starvation (Hunter 1976a; Lasker 

 1981). Several techniques have been recently 

 developed to estimate the proportion of natural 

 mortality caused by starvation. A histopatho- 

 logical technique that uses cellular criteria to 

 identify larval nutritional condition was cali- 

 brated in the laboratory (O'Connell 1976; 

 Theilacker 1978; Martin and Malloy 1980), ap- 

 plied to field studies to yield assessments of 

 larval nutritional condition (O'Connell 1980; 

 Theilacker 1986; Margulies 1986; Setzler-Hamil- 

 ton et al. 1987), and successfully used to estimate 

 rates of starvation-induced mortality (Theilacker 

 1986). Another technique that employs a mor- 

 phometric approach generated good predictions 

 of nutritional condition for several larval fish 

 species (Theilacker 1978, 1986; Martin and Mal- 

 loy 1980; Martin and Wright 1987; Setzler- 

 Hamilton et al. 1987) as did the use of an 

 RNA/DNA index (Buckley 1979; Wright and 

 Martin 1985; Clemmesen 1987; Buckley and 

 Lough 1987; Setzler-Hamilton et al. 1987). 

 Each of these diagnostic techniques suffers 



Gail H. Theilacker, Southwest Fisheries Center La Jolla 

 Laboratory, National Marine Fisheries Service, NOAA, La 

 Jolla, CA 92038 



Yoshiro Watanabe, Tohoku Regional Fisheries Research 

 Laboratory', Fisheries Agency, Shiogama, Miyagi 985, Japan. 



Manuscript accepted March 1989. 

 Fishery Bulletin, U.S. 87: 457-469. 



from various constraints. To develop histological 

 criteria, an accurate representation of the struc- 

 ture of live tissue is needed for identifying larval 

 fish condition. The usefulness of these criteria 

 depends on the tissue quality of the field-col- 

 lected specimens. For many fishes (northern 

 anchovy and striped bass, Morone saxatilis 

 (O'Connell 1980; Setzler-Hamilton et al. 1987)), 

 autolytic tissue decomposition occurs within 2-3 

 minutes after sampling. Because routine 

 ichthyoplankton collections usually take 21 min- 

 utes (Smith and Richardson 1977), special plank- 

 ton collections are required for histopathology. 

 Additionally the routine solution (3-5% for- 

 malin) used to preserve plankton does not ade- 

 quately preserve cellular structure of larval 

 fishes and special solutions (Bouin's fixative) 

 must be used (O'Connell 1976; Theilacker 1978). 

 The morphometric analysis requires extensive 

 calibration studies to estimate the effect of the 

 net and preservatives on shrinkage of body parts 

 because the morphometric indices are very sen- 

 sitive to shrinkage (Theilacker 1980; 1986). The 

 RNA/DNA index must be calibrated for tem- 

 perature effects and animal age (Ota and Landry 

 1984; Buckley 1984; Buckley and Lough 1987); 

 however, there are other considerations that 

 may limit its application. Furthermore, 

 RNA/DNA is generally regarded as an index of 

 potential growth (protein synthesis rates) not 

 starvation mortality. 



In March of 1985, we anticipated applying the 

 histopathological technique developed by O'Con- 

 nell (1976) to estimate starvation rates of north- 

 ern anchovy collected at two sites off the coast of 

 California (see Owen et al., 1989). However, 

 when aboard ship, we found that it was impossi- 

 ble to preserve the larvae within the required 

 2-3 min time period (needed to maintain tissue 

 quality) and take a sample that was representa- 

 tive of the zone inhabited by larval anchovy. We 

 were unable to use the established histological 

 criteria because it took us 5 or more minutes to 

 sample larval anchovy from 50 m to the surface 

 and process our collection. Thus, to estimate 

 anchovy starvation rates, we needed to develop 



457 



