of tissue in formalin-seawater did not invalidate 

 the analysis; thus, historical samples of zooplank- 

 ton (and, presumably, larval fish) appeared to be 

 usable for study of intraspecific geographic or in- 

 terannual variability. 



Although most studies of lipofuscin (especially 

 by histology) have concerned postmitotic cells, 

 such as nervous tissue, we evaluated the assay as 

 an estimate of the relative efficiency of growth of 

 larvae analyzed whole. That is, we were less in- 

 terested in lipofuscin as an indicator of age (for 

 which, as we show, other measures of mass are 

 useful) than as an indicator of health whose rela- 

 tion to mass would reflect environmental condi- 

 tions over time. We therefore investigated the im- 

 portance of interfering fluorescing pigments 

 (Csallany and Ayaz 1976), tested the effect of 

 preservation in formalin (Nicol 1987), and mea- 

 sured the accumulation of extractable lipofuscin 

 in three species of larval fish reared in the labora- 

 tory — California grunion, Leuresthes tenuis; 

 white seabass, Atractoscion nobilis; and Califor- 

 nia halibut, Paralichthys californicus (hereafter 

 referred as grunion, seabass, and halibut respec- 

 tively). 



Analytical Considerations 



We analyzed lipofuscin by a method first de- 

 scribed by Fletcher et al. (1973), as modified by 

 Ettershank (1984b). The tissue to be analyzed 

 (usually whole larvae) was frozen (-15° or 

 -70°C) and later freeze-dried, and a 1-5 mg sam- 

 ple was homogenized in at least 2 mL of 2:1 

 chloroform: methanol in a Wheaton glass tissue 

 homogenizer. After extracting for 3-4 hours at 

 4°C, 100 mM MgCl2 (25% of the solvent volume) 

 was added, and the solutions were thoroughly 

 mixed and then centrifuged for 20 minutes at 

 3000 rpm at -4°C. The lower, chloroform layer 

 was withdrawn for fluorometric analysis after 

 reaching 20°C in a water bath, and the fluores- 

 cence was measured on a Turner 111^ fluorometer 

 using a CS 7-60 filter (approximately 360 nm) for 

 excitation and a CS 47B filter (approximately 430 

 nm) for emission. A known concentration of 

 quinine sulfate was the standard, and results are 

 therefore reported as "fluorescence units" or FU/ 

 mg. 



A stock solution of quinine sulfate (2 mg/L in 1 

 N H2SO4) was stored in a light-tight reagent bot- 



tle. To compare analyses done at different times, 

 sets of standards were prepared from this stock in 

 distilled water (0.02, 0.04, 0.06, 0.08, and 0.10 

 |jLg/mL). Some of the sets prepared over a year's 

 time are shown in Figure 1; the overall reproduci- 

 bility is good. 



60 1- 



UJ 



o 



45 



UJ 



o 



CO 



UJ 30 



tr 

 o 



Z) 



15 







iReference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NCAA 



.03 .06 .09 .12 



QUININE SULPHATE (/ig-mr') 



Figure 1. — Reproducibility of different sets of quinine sulfate 

 standards prepared from a stock solution over a year. 



Precision of the method, including extraction 

 from tissue, was estimated by comparing repli- 

 cate samples of liver and white muscle from adult 

 halibut. The coefficients of variation for 5 deter- 

 minations on each of 3 sets of tissue were for mus- 

 cle, 0.2, 0.35, and 0.57, and for liver, 0.1, 0.15, and 

 0.18. The difference in variability between the 

 two kinds of tissues is probably due to the greater 

 difficulty in homogenizing muscle tissue. 



Quinine sulfate standards were compared on a 

 Farrand Spectrofluorometer, with excitation at 

 350 nm and emission at 420 nm, and on the 

 Turner 111 fluorometer. The correlation coeffi- 

 cient for measurements on the two instruments 

 was 0.97. We therefore used the Turner 111 rou- 

 tinely, so that analyses could be done easily in a 

 hood. 



Csallany and Ayaz (1976) described interfer- 

 ence by retinol in the analysis of organic-solvent- 

 soluble lipofuscin in mammalian tissues, and rec- 

 ommended a chromatographic step to remove this 

 contaminant. We extracted a variety of fish tis- 

 sues in 2:1 chloroform: methanol, and after addi- 

 tion of MgCl2 and centrifugation, dried the chlo- 



408 



