FISHERY BULLETIN: VOL. 78, NO. 3 



forms. We will use the terminology proposed by 

 Batley and Florence (1976*. According to these 

 authors, labile copper, as defined by experimental 

 conditions, includes ionic, as well as some dissoci- 

 able, complexed forms; bound copper is that frac- 

 tion of the total copper which is not labile and 

 includes soluble copper-organic complexes, copper 

 bound to high molecular weight organic mate- 

 rials, and copper occluded in or adsorbed onto 

 highly dispersed colloids. Although current cop- 

 per emission standards are defined in terms of the 

 total copper concentration in the water i Anony- 

 mous 1972. 1976 1 complexation of copper has been 

 shown to reduce its toxicity to marine organisms 

 (Lewis et al. 1972, 1973: Davey et al. 1973: Sunda 

 and Guillard 1976: Harrison et al.^V Ionic copper 

 has been suggested as the form most toxic to 

 freshwater fishes iPagenkopf et al. 1974*. During 

 our testing of the early life stages of the northern 

 ancho\y. we determined both total copper concen- 

 trations and the percent copper in the labile forms. 

 Northern anchovy. Engraulis mordax, is a 

 pelagic, filter-feeding fish that spawns in upwell- 

 ing waters along the Pacific coast of the United 

 States and Mexico i Ahlstrom 1960 1. During recent 

 years, the northern ancho\y catch has been the 

 third largest commercial catch on the Pacific coast 

 (McAllister 1976: Pinkas 1977 '. Having conducted 

 earlier tests on the sensitivity to copper of Pacific 

 herring during its early life stages i Rice and Har- 

 rison 1978), we set two objectives for the present 

 study: to conduct similar tests on the northern 

 ancho\y and to compare the sensitivities of these 

 two species offish during their early life histories. 

 We continuously exposed northern anchoNy em- 

 bryos and larvae to copper over a range of concen- 

 trations and then constructed comparative toxic- 

 ity curves. 



METHODS 



Northern anchovy embryos were collected in 

 San Francisco Bay. Calif., between the Tiburon 

 Peninsula and Angel Island. Collections and tests 

 were carried out over a period of 2 \t. Collections 

 were made with a 0.5 m. 505 iim mesh, nylon 

 plankton net. towed for 2 min just below the sur- 



posium. U.S. En\-iron. Prot. Agencv EcoL Res. Ser. ER^ 600 3- 

 77-085. July 1977. 204 p. 



^Harrison. F. L., J. P. Knezovich. and J. S. Tucker. 

 1979. Sensiti\'ity of Crassostiva gigas embryos to different 

 chemical forms of copi)er. Univ. Calif.. Lawrence Livermore 

 Lab., LiN-ermore. UCRL 52725; XLTIEG CR-1088. 



face of the water. Collections from each tow were 

 placed into a plastic bag half full of seawater: the 

 bag was inflated with air and then held in an 

 insulated ice chest containing seawater from the 

 collection site. Water temperature at the collection 

 site was between 17" and 18.5' C: upon arrival at 

 La\\Tence Livermore Laboratory the temperature 

 of the water in the ice chest was always <19" C. 



The water for the bioassay system was obtained 

 from the University of California Marine Station 

 at Bodega Bay. Calif. This water is pumped from 

 the ocean off the open coast in an area that re- 

 ceives little anthropogenic input. The water con- 

 tains low levels of trace metals, dissolved organics, 

 and particulate material. The collected seawater 

 was stored in a 40.000 1 underground tank and 

 passed through a filter with 1.0 /xm openings prior 

 to experimental use. 



Two embr\-o tests were conducted during 1976 

 (test I: 7 July 1976: test II: 23 July 1976' and one 

 embryo test during 1977 (test III: 27 June 1977). 

 All larval tests were conducted during 1977 (test I: 

 6 June 1977: test II: 13 June 1977). 



All transfers of embryos or lar\'ae were carried 

 out with a large-bore, polished glass pipette. Dur- 

 ing embryo tests, healthy embryos estimated to be 

 8-10 h old ( stages IV- V, Ahlstrom 1943 ) were placed 

 either directly into exposure chambers containing 

 seawater with known concentrations of copper or 

 allowed to hatch in control seawater. Larvae used 

 during larvae tests hatched within the 12-h period 

 preceding the test: hatched larvae were placed 

 directly into exposure chambers containing sea- 

 water with known concentrations of copper. Ap- 

 proximately 50 embryos or 30 larvae were used in 

 each exposure chamber. 



Anchovy embryos and larvae were exposed to 

 copper in 500 ml clear glass, flow-through, expo- 

 sure chambers (Figure 1) which, in turn, were im- 

 mersed in a water bath (mean temperature: 

 16.8'' =1.0" C'. Seawater containing a known con- 

 centration of copper, as copper chloride, was 

 pumped into each chamber from a 19 1 plastic jug 

 at a rate of 5 ml min. About 5 h were required to 

 replace 95'~f of the water in the exposure chamber: 

 the mixture of seawater and copper in the 19 1 

 plastic jugs was prepared daily. 



The height of the water in the exposure cham- 

 bers was maintained by a constant-level, outflow 

 siphon. The mouth of the siphon, located at the 

 base of the exposure chamber, was covered with 

 nylon netting 1 265 ^tm pore size^ to prevent the loss 



676 



