fishes have indeterminate growth. For example, 

 Figure 3B indicates that the brain of Dover sole 

 continues to grow long after this fish begins re- 

 production (450 mm). Thus, unlike mammals and 

 invertebrates which have determinate growth, 

 the total content of extractable lipofuscin in any 

 fish organ cannot be used as a measure of age but 

 rather concentration (content per unit of weight) 

 must be used. Furthermore, even the concentra- 

 tion of lipofuscin in the brain, a slow growing 

 tissue, is a function of both the rate of lipofuscin 

 accumulation and the rate of tissue growth. It 

 follows then, that if the rate of brain growth is 

 such that it masks the actual rate of lipofuscin 

 accumulation in older organisms, lipofuscin con- 

 centration will not increase with time. 



A critical assumption underlying the use of 

 lipofuscin as a determinant of age is that 

 metabolic rate, and hence lipofuscin accumula- 

 tion rate, remains fairly constant over the portion 

 of the life history of interest. The rainbow trout 

 were reared in a hatchery where no net change in 

 the environment occurred from 1983 to 1985. 

 Dover sole, on the other hand, gradually migrate 

 into deeper, colder, and less oxygenated waters as 

 they age. Presumably the metabolism of fish 

 under these conditions would be lower as might 

 be the rate of lipofuscin accumulation in the 

 brain. 



The literature on fishes provide no conclusive 

 evidence that lipofuscin is an accurate index of 

 age over the entire lifespan. Hill and Radtke (in 

 press) reported that the extracted lipofuscin per 

 unit dry weight in the brain of the tropical fish 

 Dascillus albisella accumulates exponentially 

 with age. The relationship is driven by a single 

 point for an 11-year-old individual and it would 

 probably be linear without that single point, if 

 only fish 1-7 years old were considered. In 

 hatchery-reared Cyprinus carpio the total ex- 

 tracted lipofuscin per unit of dry weight in the 

 brain of fish of the same age (6 years) increased 

 with weight over 8-fold range in weight (Griven 

 et al.^). About 45% of the lipofuscin content could 

 be explained by difference in weight among fish. 

 Thus, in fish of the same age there was a strong 

 size effect on lipofuscin concentration. These re- 

 sults are very similar to those found in this study. 

 Aloj Totaro et al. (1985) found that lipofuscin in- 

 creased over a range of 0-2 years. But the method 



2Girven, R. J,, R. W. Gauldie, Z. Czochanska, A. D. Wool- 

 house. Manuscr. in prep. A critical tests of the lipwfuscin 

 technique of age estimation in fish. 



was different as they measured lipofuscin gran- 

 ules present in the electric lobe of Torpedo mar- 

 morata brains rather than using extractable lipo- 

 fuscin of the entire brain. Thus, all studies to date 

 seem to indicate that accumulation occurs in the 

 brains of fishes, but results are not conclusive 

 owing to small sample sizes, limited age ranges, 

 and failure to identify the effects of brain growth 

 on rates of accumulation. 



Additional research is required to evaluate 

 lipofuscin as a method of age determination in 

 fishes. The effect of brain growth on lipofuscin 

 accumulation rates must be considered in such 

 studies. A promising approach in this regard may 

 be to estimate lipofuscin on a per cell basis in- 

 stead of on a weight basis. This could be accom- 

 plished by either expressing extracted lipofuscin 

 relative to DNA concentration or by histological 

 techniques. 



Acknowledgments 



This work was funded by Sea Grant contract 

 RyNP-l-15C and by NOAA contract 43ABNF6 

 1987. The authors would like to thank A. Dizon, 

 M. M. Mullin, E. Brooks, and J. Butler for helpful 

 discussions; J. Butler, E. Lynn, and M. Draw- 

 bridge for age determination by otoliths of Dover 

 sole specimens; and M. Rowan of California Fish 

 and Game for providing the rainbow trout sam- 

 ples. 



Literature Cited 



Agius, C and S. a. Agbede. 



1984. An electron microscopical study on the genesis of 

 lipofuscin, melanin, and haemosiderin in the haemopoi- 

 etic tissues offish. J. Fish Biol. 24:471-488. 



Aloj Totaro. E., F. A Pisanti, P Russo, and P Brunetti. 



1985. Evaluation of aging parameters in Torpedo mar- 

 morata. Ann. See. R. Zool. Belg. 115(2):203-209. 



BuGH, E G . AND W J Dyer. 



1959. A rapid method of total lipid extraction and purifi- 

 cation. Can. J. Biochem. Physiol. 37:911-917. 

 BOEHLERT, G. W. 



1985. Using objective criteria and multiple regression 

 models for age determination in fishes. Fish. Bull, U.S. 

 83:103-107. 

 Dapson, R W , A T. Feldman. and G Pane. 



1980. Differential rates of ageing in natural populations 

 of old-field mice (Peromyscus polionotus). J. Gerontol. 

 35:39-44. 

 Ettershank. G. 



1984. A new approach to the assessment of longevity in 

 the Antarctic krill Euphausia superba. J. Crustacean 

 Biol. 4:295-305. 

 Ettershank, G , I MacDonnell, and R Croft. 



1983. The accumulation of the age pigment by the fleshfiy 



406 



