Fishery Bulletin 95( 1 ), 1997 
1 12 
that the lowest frequencies are usually associated 
with increased stress, such as poorer quality habi- 
tat, food shortage, or a shorter growing season (Bull 
and Shine, 1979). Nonspawning condition has been 
induced experimentally for several species by reduc- 
ing their food supply (e.g. haddock [Hislop et al., 
1978]; Newfoundland winter flounder [Burton and 
Idler, 1987]; and plaice [Norwood et al., 1989]). 
Nutrients are in good supply and do not limit pri- 
mary production on the Campbell Plateau (which 
forms a major portion of the Southern Plateau sur- 
vey area); however, chlorophyll concentrations over 
depths of 450 m or greater are generally low (Heath 
and Bradford, 1980). Heath and Bradford suggest 
that because of this and other characteristics of the 
area, there never will be a well-developed zooplank- 
ton community with a high biomass on the Campbell 
Plateau. Areas of higher productivity are found on 
the island shelves and shallow rises in the area, or 
downstream from the Campbell Plateau itself. The 
energetics of the food chain in the study area are not 
known. It is possible that the lack of high primary 
and secondary productivity in the area contributes 
toward nonreproduction in some hoki from year to 
year. 
Whatever the cause, nonspawning among adult hoki 
has important implications for stock assessment and 
risk estimation in the management of New Zealand 
hoki stocks. The proportion of fish that migrate to 
spawn is a scaling factor that relates the number of 
fish observed during the spawning season (using tools 
such as CPUE and acoustics) to the total population. 
It also provides a buffer between the total stock and 
the population vulnerable in any one year to the 
greatest fishing effort that is applied during the 
spawning season. Further, it reduces the stock size, 
which is needed for calculating the stock-recruit rela- 
tionship used in predicting future recruitment. 
The effect of these factors may be minimal if the 
level of nonspawning fish is constant. If, however, 
the proportion spawning varies from year to year, as 
suggested by our study, the implications for model- 
ling may be both complex and important. 
It is likely that there are other species not neces- 
sarily related to hoki that could also have signifi- 
cant and variable proportions of nonspawning fish. 
There may therefore be major implications for the 
stock assessment of those species as well. Any stock 
assessment tool that is used to obtain an estimate of 
absolute abundance from a spawning population (e.g. 
acoustics, egg-production method) should take 
nonspawning into account. It is also important that 
the effect of nonspawning on any stock-recruitment 
relationship (assumed or measured) be taken into 
account because one of the more serious difficulties 
in determining the stock-recruitment relationship of 
any species is obtaining a reliable measure of the 
spawning stock size (Hilborn and Walters, 1992). 
Further, the stock-recruitment dynamics of a popula- 
tion could be masked, particularly if the level of 
nonspawning is correlated to some environmental fac- 
tor or autocorrelated because of some inherent life strat- 
egy, such as improved longevity or increased egg size. 
We have shown that nonspawning among adult 
hoki is substantial, and it has important conse- 
quences for stock assessment. For these reasons, it 
is clear that a better understanding of its variabil- 
ity, and how widespread its occurrence might be 
among other species, would be useful for fisheries 
management worldwide. 
Acknowledgments 
We thank Peter Horn for ageing the samples; Kevin 
Sullivan, Patrick Cordue, and Len Tong for useful 
discussions; the scientific staff who participated in 
the trawl surveys; the officers and crew of GRV 
Tangaroa\ and the scientific observers for collecting 
samples from commercial vessels. We also thank 
three anonymous reviewers for their comments and 
suggestions. 
Literature cited 
Bell, J. D., J. M. Lyle, C. M. Bulman, K. J. Graham, 
G. M. Newton, and D. C. Smith. 
1992 . Spatial variation in reproduction and occurrence of 
non-reproductive adults in orange roughy, Hoplostethus 
atlanticus Collett (Trachichthyidae), from south-eastern 
Australia. J. Fish Biol. 40:107—122. 
Bull J. J., and R. Shine. 
1979 . Iteroparous animals that skip opportunities for 
reproduction. Am. Nat. 114(2):296-316. 
Burton, M. P., and D. R. Idler. 
1987 . An experimental investigation on the non-reproduc- 
tive, post-mature state in winter flounder. J. Fish Biol. 
30:643-650. 
Chatterton, T. D., and S. M. Hanchet. 
1994 . Trawl survey of hoki and associated species in the 
Southland and Sub-Antarctic areas, November-December 
1991 (TAN9105). N.Z. Fish. Data Rep. 41, 55 p. 
Foucher, R. P., and R. J. Beamish. 
1980. Production of non-viable oocytes by Pacific hake 
(Merluccius productus). Can. J. Fish. Aquat. Sci. 37:41-48. 
Francis, R. I. C. C. 
1984 . An adaptive strategy for stratified random trawl 
surveys. N.Z. J. Mar. Freshwater Res. 18:59-71. 
Gunn, J. S., B. D. Bruce, D. M. Furlani, R. E. Thresher, 
and S. J. M. Blaber. 
1989 . Timing and location of spawning of blue grenadier 
Macruronus novaezelandiae (Teleosti: Merlucciidae) in 
Australian coastal waters. Aust. J. Mar. Freshwater Res. 
40:97-112. 
