UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Reproductive and population biology of Pacific ocean perch (Sebastes alutus (Gilbert)) Leaman, Bruce Michael 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1988_A1 L42.pdf [ 15.82MB ]
Metadata
JSON: 831-1.0098067.json
JSON-LD: 831-1.0098067-ld.json
RDF/XML (Pretty): 831-1.0098067-rdf.xml
RDF/JSON: 831-1.0098067-rdf.json
Turtle: 831-1.0098067-turtle.txt
N-Triples: 831-1.0098067-rdf-ntriples.txt
Original Record: 831-1.0098067-source.json
Full Text
831-1.0098067-fulltext.txt
Citation
831-1.0098067.ris

Full Text

REPRODUCTIVE AND POPULATION BIOLOGY OF PACIFIC OCEAN PERCH (Sebastes a lutus (Gi lbert ) ) By BRUCE MICHAEL LEAMAN B . S c , Simon Fraser Un ive rs i t y , 1972 Sc, The Un ivers i t y of B r i t i s h Columbia, 197 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (DEPARTMENT OF ZOOLOGY) We accept t h i s thes i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA A p r i l 1988 © Bruce Michael Leaman; 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) ABSTRACT This study examines the reproductive and population biology of a long-lived (80-y l i f e span) f i s h , Sebastes alutus ( G i l b e r t ) . The objectives of the study were (i) to e s t a b l i s h whether groups of adult S. alutus delineated by e x p l o i t a t i o n h i s t o r i e s could be o b j e c t i v e l y i d e n t i f i e d as separate sub-populations; ( i i ) to i d e n t i f y the e f f e c t s of density-independent mortality caused by e x p l o i t a t i o n , and r e l a t e them to predictions of l i f e h i s t o r y theory ( i i i ) to examine the long-term implications of t h i s mortality pressure on the dynamics of the species; and (iv) to determine the contribution of these r e s u l t s to the development of management programs for t h i s species. A lernaepodid copepod g i l l parasite (Neobrachiella robusta (Wilson 1912)) was used for the f i r s t time as a b i o l o g i c a l tag to delineate separate sub-populations (stocks) of a commercial f i s h . Intensity of i n f e c t i o n and mean c h a r a c t e r i s t i c s of the parasite population per f i s h achieved complete separation between stocks indistinguishable with morphological features. Discriminant analysis showed the parasite also functioned moderately well (34-76% correct c l a s s i f i c a t i o n ) as a stock discriminator of i n d i -v idual f i s h . Use of a u x i l l i a r y information on the stock i d e n t i t y of hosts improved the c l a s s i f i c a t i o n power of the discriminant function. The density-independent mortality of the commercial fishery on S. a l u t u s has s e v e r e l y t r u n c a t e d the age spectrum of some s t o c k s . Some compensatory growth changes f o r f i s h i n the s t o c k s under the s t r o n g e s t s e l e c t i o n are e v i d e n t . Examination of the v a r i a n c e s t r u c t u r e of female l e n g t h a t age suggests an i n v e r s e r e l a t i o n of m o r t a l i t y and growth r a t e g i v i n g r i s e t o s m a l l e r , o l d e r f i s h . S m a ller, o l d e r f i s h can be accounted f o r u s i n g the same growth f u n c t i o n as f o r the l a r g e r f i s h seen a t younger ages, and these s i m i l a r growth forms can g i v e an aggregate appearance o f q u a d r a t i c growth. Growth changes do not y i e l d s i g n i f i c a n t d i f f e r e n c e s i n s i z e a t m a t u r i t y , although age a t m a t u r i t y changes, i m p l y i n g developmental or environmental c o n s t r a i n t s on m a t u r a t i o n . F e c u n d i t y e s t i m a t i o n methodology was e v a l u a t e d and the v o l u m e t r i c method p r e v i o u s l y used f o r t h i s s p e c i e s found t o be i n f e r i o r t o a g r a v i m e t r i c method. S i g n i f i c a n t d i f f e r e n c e s i n f e c u n d i t y as a f u n c t i o n of body v a r i a b l e s were found among e x p l o i t a t i o n groups. S i g n i f i c a n t d i f f e r e n c e s i n oocyte c h a r a c t e r i s t i c s among s t o c k s were found, w i t h s i z e and age shown t o have s e p a r a b l e e f f e c t s . L i g h t l y e x p l o i t e d s t o c k s had s i g n i f i c a n t l y h i g h e r oocyte q u a l i t y (as expressed i n oocyte weight), e f f e c t e d through d i f f e r e n c e s i n the oocyte diameter-oocyte weight r e l a t i o n s h i p . H i s t o l o g i c a l examination d e t a i l s the developmental sequence of oocytes and e s t a b l i s h e s the maturation p e r i o d o f oocytes and i v f i s h . Northern s t o c k s were shown t o have s i g n i f i c a n t l y l a r g e r o o c y t e s . F o l l i c u l a r a t r e s i a i s suggested as an a l t e r n a t i v e energy source t o the embryo death which has been p r e s e n t e d as the source f o r matrotrophy i n t h i s genus. Complete a t r e s i a o f a r i p e oocyte complement was i d e n t i f i e d i n t h i s genus f o r the f i r s t time. No evidence of r e p r o d u c t i v e senescence was found. The h y p o t h e s i s of i n c r e a s i n g r e p r o d u c t i v e e f f o r t w i t h age, i n c l u d i n g the independence of age and s i z e e f f e c t s , was con-f i r m e d . No evidence of r e p r o d u c t i v e c o s t c o u l d be found. The h y p o t h e s i s o f i n c r e a s e d r e p r o d u c t i v e e f f o r t e a r l i e r i n l i f e as a mechanism t o o f f s e t i n c r e a s e d a d u l t m o r t a l i t y was supported. However, th e p o t e n t i a l of t h i s i n c r e a s e , a c h i e v e d by growth r a t e i n c r e a s e s , i s much l e s s than i s needed t o compensate f o r the r e d u c t i o n i n l i f e t i m e r e p r o d u c t i v e e f f o r t caused by h i g h f i s h i n g m o r t a l i t y . The e f f e c t s of changing m o r t a l i t y r a t e s on s e v e r a l reproduc-t i v e v a l u e i n d i c e s was examined w i t h d e t e r m i n i s t i c and s t o c h a s t i c s i m u l a t i o n models. Cohort r e p r o d u c t i v e v a l u e i s the most s e n s i -t i v e o f the i n d i c e s examined and may be the most r o b u s t t o measurement e r r o r . Reproductive v a l u e i s a more s e n s i t i v e index of p o p u l a t i o n s t a t e than o t h e r i n d i c e s i n use and may p l a y a r o l e i n d e t e r m i n a t i o n and e v a l u a t i o n of optimal h a r v e s t p o l i c i e s . However, an experimental approach t o i t s use w i l l be r e q u i r e d . V TABLE OF CONTENTS ABSTRACT . i i ACKNOWLEDGEMENTS xiv I. INTRODUCTION 1 Age-specific reproduction 5 Reproductive value 12 Predictions of l i f e h i s t o r y theory 15 L i f e h i s t o r y theory and harvested populations . . . . 19 I I . GENERAL METHODS AND MATERIALS 25 1. Reproductive biology sampling 25 2. Parasite sampling 31 3. Fecundity and oocyte c h a r a c t e r i s t i c s 32 4. Size and weight at age 38 5. Modelling studies 39 I I I . HISTORY OF EXPLOITATION AND IDENTIFICATION OF STOCKS . 40 1. History of expl o i t a t i o n 40 2. Previous evidence for stock i d e n t i f i c a t i o n 45 3. Parasites as i d e n t i f i e r s of S. alutus stocks . . . . 54 4. Discussion 62 5. Conclusions 68 v i IV. REPRODUCTIVE BIOLOGY AND RELATION TO LIFE HISTORY THEORY 70 1. Growth 70 2. Reproductive biology 83 3. Discussion 133 4. Conclusions 145 V. REPRODUCTIVE VALUE MODELLING AND IMPLICATIONS FOR MANAGEMENT 151 1. Introduction 151 2. Model description and parameter estimation 156 2. Results 164 3. Discussion 176 VI. SUMMARY 181 VII. LITERATURE CITED 184 v i i LIST OF TABLES Table 1. Description of S. alutus maturity stages 27 Table 2. Gravimetric method of fecundity estimation and oocyte c h a r a c t e r i s t i c s for S. alutus 35 Table 3. Estimated landings of P a c i f i c ocean perch (Sebastes  alutus), by stock, 1959-1985 43 Table 4. Estimated biomass (t) of the f i v e P a c i f i c ocean perch (Sebastes alutus) stocks p r i o r to major f i s h e r i e s and at present 45 Table 5. L o c a l i t y , depth, and number of Sebastes alutus sampled for parasites, by area 55 Table 6. Population structure of female Neobrachiella robusta taken from the g i l l s of Sebastes alutus. by area . . . 56 Table 7. S t a t i s t i c a l t ests of d i f f e r e n t i a t i o n based on mean values of parasite (Neobrachiella robusta) c h a r a c t e r i s t i c s between Sebastes alutus stock pairs 60 Table 8. S t a t i s t i c a l tests of d i f f e r e n t i a t i o n based on mean values of parasite (Neobrachiella robusta) c h a r a c t e r i s t i c s between Sebastes alutus samples from d i f f e r e n t depths . 61 Table 9. Percent of Sebastes alutus c l a s s i f i e d to stock by d i s -criminant analysis 63 Table 10. Residual sum of squares and mean square error f o r quadratic (Q) and asymptotic (A) estimation of length and round weight (RDWT) at age for f i v e stocks of S. alutus 71 Table 11. Regression s t a t i s t i c s f or weight (g) rel a t i o n s h i p s of S. alutus females, by stock 76 Table 12. Estimated s i z e (cm) and age at 50% maturity f o r female S. alutus c o l l e c t e d i n 1982, by stock 84 Table 13. C o e f f i c i e n t of v a r i a t i o n of eggs/mL for volumetric subsampling of S. alutus ovaries, by sample type . . . 89 Table 14. Analysis of variance of mean eggs/mL f o r volumetric sampling of S. alutus fecundity, by subsampling volume 90 Table 15. Analysis of variance of eggs/mL as a function of sampling p o s i t i o n within an egg suspension f o r 2 and 5mL subsampling volumes 92 Table 16. Sampling s t a t i s t i c s f or gravimetric estimation of S. v i i i alutus fecundity; fixed and variable subsample weights 93 Table 17. Dry weight of S. alutus oocyte samples i n drying oven at 50° C, at 12 h in t e r v a l s 93 Table 18. Regression s t a t i s t i c s of fecundity estimation for S. alutus, by stock 95 Table 19. Multiple regressions of gonadal index (GIS) and tes t s of si g n i f i c a n c e among exp l o i t a t i o n groups of S. alutus 99 Table 20. Multiple regressions of oocyte c h a r a c t e r i s t i c s and te s t s of si g n i f i c a n c e among ex p l o i t a t i o n groups of S. alutus. Data from l i b e r a t e d oocytes 101 Table 21. Relationship of egg weight (EGWT) and egg diameter (EGDIA) between exp l o i t a t i o n groups, for S. alutus . . 104 Table 22. ANOVA for mean numbers of a t r e t i c to non-atretic f o l l i c l e s among stocks and expl o i t a t i o n groups of S. alutus. Data from h i s t o l o g i c a l samples 117 Table 23. ANOVA of oocyte sizes by stock and e x p l o i t a t i o n group f o r S. alutus. Data from h i s t o l o g i c a l samples . 119 Table 24. Percent composition of ovarian contents f o r post-f e r t i l i z e d S. alutus. by category, for the Rennell Sound and Goose Island Gully stocks 122 Table 25. Notation and sources for parameters used i n simulation models 157 Table 27. Changes i n cohort reproductive c h a r a c t e r i s t i c s with F=0.0 and F=0.05, by stock, for S. alutus with quadratic growth 166 Table 28. Reproductive value indices, biomass, y i e l d , and mean age changes with f i s h i n g mortality (F), over 200 y horizon, for a simulated S. alutus population 175 Table 29. Reproductive value indices at age 12 for S. alutus as a function of f i s h i n g mortality (F) , averaged over the l a s t 10 years of 200-yr simulations 175 i x LIST OF FIGURES Figure 1. Sebastes alutus stocks o f f B r i t i s h Columbia . . . 41 Figure 2. Catch his t o r y of S. alutus stocks o f f B r i t i s h Columbia 42 Figure 3. Age composition of S. alutus o f f southwest Vancouver Island i n 1982 48 Figure 4. Age composition of S. alutus i n Goose Island Gully i n 1982 49 Figure 5. Age composition of S. alutus i n Moresby Gully i n 1982. 50 Figure 6. Age composition of S. alutus o f f Rennell Sound i n 1982. 51 Figure 7. Age composition of S. alutus i n the Langara Spit area i n 1982 52 Figure 8. Length at age d i s t r i b u t i o n for female S. alutus i n Moresby Gully, together with a f i t t e d von Bertalanffy growth curve 72 Figure 9. D i s t r i b u t i o n of residual error between estimated and observed length at age for Moresby Gully S. alutus females 73 Figure 10. Mean length at age d i s t r i b u t i o n for Vancouver Island S. alutus females, 1965-1982 78 Figure 11. Mean length at age d i s t r i b u t i o n for Goose Island Gully S. alutus females, 1966-1982 79 Figure 12. Mean length at age d i s t r i b u t i o n for Moresby Gully S. alutus females 1974-1982 80 Figure 13. Mean length at age d i s t r i b u t i o n for Langara Spit S. alutus females, 1966-1982 81 Figure 14. Length-maturity ogive for Vancouver Island S. alutus. 1982 86 Figure 15. Length-maturity ogive for Moresby Gully S. alutus females, 1982. 87 Figure 16. Mean S. alutus oocyte sizes obtained by sampling volume and pipette type 91 Figure 17. S p e c i f i c fecundity (eggs/gm somatic weight) at age fo r unexploited stocks of S. alutus 96 Figure 18. S p e c i f i c fecundity (eggs/cm) at age for unexploited stocks of S. alutus X 98 Figure 19. Section of ovary from a 'maturing* S. alutus showing the small (50-100 nm) and intermediate (200-250 jum) s i z e classes of oocytes 105 Figure 20. Section of ovary from a mature S. alutus female showing the small (50-100 jtzm) and large (600 /xm) s i z e classes of oocytes 106 Figure 21. Section of S. alutus ovary showing the concentration of immature oocytes near the branch of the ovarian lamellae 108 Figure 22. Oogonial nest (ON) with immature oocytes from a S. alutus ovary 109 Figure 23. Early perinucleolar stage of S. alutus oocyte development. [Nucleus (N) , n u c l e o l i (NU) ] I l l Figure 24. Perinucleolar stage of S. alutus oocyte showing development of yolk globules (YG) at the periphery of the oocyte 112 Figure 25. F u l l y mature S. alutus oocyte showing large l i p o i d a l v e s s i c l e s (LV) 113 Figure 26. F u l l y mature S. alutus oocyte. The nucleus has l o s t i t s i n t e g r i t y and no n u c l e o l i are present 114 Figure 27. A t r e t i c f o l l i c l e i n normally developing S. alutus female, i l l u s t r a t i n g hypertrophied f o l l i c u l a r c e l l s . . 116 Figure 28. Ruptured f o l l i c l e s (RF) i n a post-spawned S. alutus ovary from the Rennell Sound stock 123 Figure 29. Residual, ruptured chorion and embryo remaining i n the ovary of a post-spawned S. alutus from the Rennell Sound stock 124 Figure 30. A t r e t i c , u n f e r t i l i z e d f o l l i c l e s remaining i n the ovary of a S. alutus from the Rennell Sound stock. . . 125 Figure 31. Embryonic stages i n the ovary of a S. alutus from the Goose Island Gully stock 126 Figure 32. Pre-ovulatory, non-atretic f o l l i c l e s remaining i n the ovary of a S. alutus from the Goose Island Gully stock 127 Figure 33. Catch h i s t o r i e s of several P a c i f i c and A t l a n t i c r o c k f i s h f i s h e r i e s 152 x i Figure 34. Cohort fecundity and reproductive values at age for two l e v e l s of f i s h i n g mortality, simulated using the population parameters of the Langara Spit S. alutus. N 0 i s the i n i t i a l s i z e of the cohort 165 Figure 35. S e n s i t i v i t y of predicted stock y i e l d to perturba-tions i n the input parameters for the multiple cohort simulation model, a f t e r a period of 100 y 168 Figure 36. S e n s i t i v i t y of predicted stock biomass to perturba-tions i n the input parameters for the multiple cohort simulation model, a f t e r a period of 100 y 169 Figure 37. Response of several reproductive value (RV) indices (at age 12) to f i s h i n g mortality, for a composite stock of S. alutus. Values averaged over the l a s t ten years of 200-y simulations. See text for d e f i n i t i o n of RV indices 173 x i i ACKNOWLEDGEMENTS I wish t o express my thanks t o my s u p e r v i s o r y committee, Drs. J . D. McP h a i l , H. D. F i s h e r , C. C. Lindsey, C. J . Walters, D. M. Ware, and e s p e c i a l l y t o my r e s e a r c h s u p e r v i s o r , Dr. N. J . Wilimovsky, f o r t h e i r a d v i c e on the d e s i g n o f t h i s p r o j e c t and t h e i r t h o u g h t f u l comments on a d r a f t o f the t h e s i s . I am a l s o g r a t e f u l t o Dr. R. J . Beamish, D i r e c t o r o f the P a c i f i c B i o l o g i c a l S t a t i o n (Department o f F i s h e r i e s and Oceans) f o r encouraging me t o i n i t i a t e t h i s p r o j e c t , and f o r h i s support throughout i t s e x e c u t i o n . Two c o l l e a g u e s a t the P a c i f i c B i o l o g i c a l S t a t i o n , Drs. Z. Kabata and T. J . M u l l i g a n , p r o v i d e d much a d v i c e on p a r i s i t o l o g y and a n a l y t i c methods, r e s p e c t i v e l y , which l e d t o co-authored man u s c r i p t s . Dr. Kabata k i n d l y screened and staged the p a r a s i t e samples used i n t h i s study, and Dr. M u l l i g a n a l s o reviewed a d r a f t o f the t h e s i s . I thank them f o r t h e i r p a t i e n t and thorough c o u n s e l on these t o p i c s , as w e l l as t h e i r encouragement. Numerous i n d i v i d u a l s a s s i s t e d me i n the f i e l d and l a b o r a -t o r y , p a r t i c u l a r l y w i t h the t e d i o u s t a s k s o f f e c u n d i t y e s t i m a t i o n and h i s t o l o g i c a l s c r e e n i n g . I am e s p e c i a l l y g r a t e f u l t o Dick Nagtegaal, V a l e r i e Bowie, L o u i s e F o c k l e r , Graham G i l l e s p i e , A n i t a O t t e r d y k s , Ted C a r t e r , Greg Workman, and Deborah Colmer. John Bagshaw prepared the h i s t o l o g i c a l s e c t i o n s and o f f e r e d a d v i c e on oocyte l i b e r a t i o n media. Joan Nott c h e e r f u l l y typed some of the x i i i f i r s t d r a f t o f the t h e s i s . T h i s p r o j e c t was c o m p l e t e l y funded by the Canadian Department o f F i s h e r i e s and Oceans. I r e s e r v e a s p e c i a l g r a t i t u d e f o r R i c k S t a n l e y . He has p r o v i d e d the e s s e n t i a l q u a l i t i e s o f a t r u e f r i e n d throughout t h i s study; an honest and thorough c r i t i c , as w e l l as a c o n t i n u a l s u p p o r t e r when most needed. My f a m i l y , J a n i c e and Ryan, have supported me and put up w i t h my absences w i t h admirable p a t i e n c e . Words are not adequate t o express my g r a t i t u d e t o them. 1 I. INTRODUCTION Evolutionary persistence of a species implies a range of mechanisms to cope with challenges to eithe r i n d i v i d u a l s u r v i v a l or reproduction. An important component of a species success i s the manner of reproduction and elaboration of i t among genera-t i o n s . Major evolutionary advances are often associated with changes i n reproductive modes (Keeton 1980). Reproductive mode and l i f e span may therefore play central roles i n evolutionary success and a wide spectrum of these two features i s evident. The extremes of t h i s continuum for sexually reproducing species are t y p i f i e d by short-lived, highly-fecund invertebrates or plants and long-lived, low-fecundity mammals. The former group has received considerably more attention from evolutionary b i o l o g i s t s than the l a t t e r . In part, the shortage of research on long-lived animals i s due to the s i m i l a r i t y i n l i f e spans between investigators and subjects, making observations of the s e l e c t i v e process d i f f i c u l t . Nonetheless, long-lived organisms exhibit some unique c h a r a c t e r i s t i c s , the understanding of which could provide insight into evolutionary processes. I argue that the focus on sh o r t e r - l i v e d animals, while allowing more opportunity for experimental manipulation, may have precluded some understanding of the hierarchy of processes which lead to evolutionary persistence. While l i f e h i s t o r y t h e o r i s t s have postulated many mechanisms for adaptive response to pres-sures of demographic change (Gadgil and Bossert 1970, Stearns 2 1976), the theory cannot be validated i f v a r i a b i l i t y i n those parameters contributing to the response mechanisms i s i n s u f f i -c i e n t f o r s e l e c t i v e advantage. For species with longevity (long l i f e ) , we need to observe populations under strong s e l e c t i n g pressure i n the natural environment to comprehend which com-ponents of the l i f e h i s t o r y are s u f f i c i e n t l y l a b i l e or variable to confer s e l e c t i v e advantage. This should enable us to deter-mine i f the basic material for evolutionary processes, upon which l i f e h i s t o r y theory rests, can be found i n a natural system. For long- l i v e d organisms such studies w i l l be protracted, but we may s t i l l gain insight from the examination of short-term mortality e f f e c t s . In choosing to examine a ro c k f i s h with a l i f e span of over 80 years I was influenced by several factors. F i r s t l y , species of t h i s type have not been adequately investigated from the perspectives of eithe r experimental or evolutionary biology. Secondly, groups of t h i s species which had been subjected to large demographic a l t e r a t i o n s were available f o r study. Lastly, the species not only has longevity but also highly determinate (asymptotic) growth, allowing me to examine the reproductive aspects of l i f e h i s t o r y ( i . e . , some of the basic features of s e l e c t i v e advantage) while separating the e f f e c t s of age and s i z e . Much of the l i f e h i s t o r y l i t e r a t u r e confounds these two variables and I suggest i t also clouds our i n t e r p r e t a t i o n of evolutionary mechanisms. This d i s t i n c t i o n can best be appre-3 c i a t e d against the background of studies examining the evolution of l i f e h i s t o r i e s . In the continuum of reproductive modes, longevity represents an extreme expression of i n d i v i d u a l s u r v i v a l over a 'breed-and-die' a l t e r n a t i v e . The advantages conferred by longevity are manifold but one of the most important i s the p o t e n t i a l to optimize reproductive contribution through temporal and s p a t i a l (in the case of mobile organisms) spreading of the p o t e n t i a l mortality r i s k s associated with reproductive uncertainty (den Boer 1968, L u c k i n b i l l and Clare 1985). By d e f i n i t i o n , such organisms w i l l have multiple age groups i n t h e i r populations and generally be iteroparous (have multiple reproductive episodes). I f the l a t t e r i s true, the population w i l l also be composed of overlapping generations and the genome w i l l be buffered against catastrophic reproductive f a i l u r e (Roughgarden 1979). Most long-l i v e d organisms w i l l also have l a t e maturity and determinate growth, since the advantage of large s i z e can be discounted by longevity (Charlesworth 1980). A discussion of the r e l a t i v e merits of apparent strategies for persistence must include some objective measure of success, or f i t n e s s . A widely accepted d e f i n i t i o n for f i t n e s s i s the contribution that an i n d i v i d u a l makes to future gene pools (Calow 1979, Roughgarden 1979). However, genotypes i n sexually reprodu-cing species are not transmitted i n t a c t and the consequent 4 d i f f i c u l t i e s i n the detection of causal genetic mechanisms have led to considerable controversy over a measurable index of f i t n e s s . Perhaps the majority view favours measurement of the rate of increase (r) of a p a r t i c u l a r genotype or a l l e l i c frequen-cy as the best index (Crow and Kimura 1970, Charlesworth 1972, 1980). The minority view questions whether t h i s r represents an accurate measure of f i t n e s s because i t assumes a stable age d i s t r i b u t i o n , which implies s t a b i l i t y i n environmentally-induced mortality. This group (Williams 1966a, Schaffer and Tamarin 1973, Schaffer 1974b, G i l l e s p i e 1977, Goodman 1978, 1979, Bulmer 1985) suggests the average (or geometric) rate of population increase i n heterogeneous environments, or simply the a b i l i t y to p e r s i s t i n temporally varying environments (Giesel 1976), are more r e a l i s t i c measures of f i t n e s s . However, the l a t t e r measure i s l i m i t e d by an i n a b i l i t y to q u a n t i t a t i v e l y assess f i t n e s s under coincident but numerically unequal persistence. For example, which of insects or vertebrate carnivores, two species groups with long evolutionary h i s t o r i e s but grossly d i f f e r e n t abundances and population growth rates, i s more f i t ? In part, the divergence of population ecologists and population g e n e t i c i s t s rests on the adoption of rate of increase as an index by both groups although the indices are measured phenotypically by the former and genotypically by the l a t t e r . A c r i t i c i s m often voiced i s that the phenotypic index implies a genetic mechanism that i s not generally ( i f ever) demonstrated 5 (Reznick 1985). This reproach i s directed to the school of thought on population increase o r i g i n a t i n g with Lotka (1913), Lewis (1942), and L e s l i e (1945). This c r i t i c i s m by population g e n e t i c i s t s i s not e n t i r e l y f a i r , since molecular g e n e t i c i s t s have i n turn questioned the measure of a l l e l i c frequencies; the n e u t r a l i s t - s e l e c t i o n i s t controversy amply att e s t s to t h i s d i s a -greement (reviewed by Roughgarden 1979). The r e s u l t has been a p a r t i a l synonomy of the phenotypic ('population') and genotypic ( a l l e l i c frequency) rates of increase i n the l i t e r a t u r e . No c l e a r resolution of the merits of rate of increase vs. persistence e x i s t s . Giesel (1976:58) noted that the r e l a t i o n s h i p between them i s not c l e a r since we do not know with certainty whether an observed high rate of increase always confers "evolu-tionary staying power". The persistence viewpoint i s not widely addressed i n the l i t e r a t u r e , presumably because i t i s nested within the modified rate of increase concept (e.g. geometric r ) . However, Slobodkin (1964), Giesel (1976), and Lewontin (1965) have argued that e i t h e r persistence, longevity or the p r o b a b i l i t y that a population w i l l not go extinct are the best measures of f i t n e s s . Age-specific reproduction The substantial number of papers concerned with reproduction and l i f e h i s t o r y theory (Stearns 1980) attests to the basic r o l e that reproduction i s believed to play i n determining f i t n e s s . 6 Much of t h i s l i t e r a t u r e has involved the apparent dichotomy of semelparity (single l i f e t i m e reproductive episode) vs. iteropa-r i t y as 1 s t r a t e g i e s 1 1 and i t i s widely acknowledged that Cole (1954) generated the concept that only minor modifications of semelparous l i t t e r s i z e were necessary to equate the r e s u l t s of the two strategies. The general merit of Cole's paper and the subtlety of h i s assumption generated a lengthy debate i n the l i t e r a t u r e (Cody 1966, Harper 1967, Gadgil and Bossert 1970, Bryant 1971, Murdoch 1966a), culminating i n a rigorous treatment of the paradox by Charnov and Schaffer (1973). Resolution required recognition of the variable d i s t r i b u t i o n of mortality with age and the probable s u r v i v a l cost of reproduction; facts overlooked by Cole. Recently, further commentary on the sexual reproduction aspects of Cole's r e s u l t has also been supplied (Charnov et a l . 1981, Waller and Green 1981). Perhaps fortunate-l y , Cole's r e s u l t was questioned by many (including Cole himself) under the reasonable tenet that since i t e r o p a r i t y was widespread i n nature, i t s merits were not completely understood. It e r o p a r i t y i s assumed to be an adaptive strategy when the p r o b a b i l i t y of successful reproduction i s low or pre-reproductive s u r v i v a l i s eithe r lower or more variable than that of adults 1 The terms 'strategies* and ' t a c t i c s ' have gained wide-spread use i n the l i f e h i s t o r y l i t e r a t u r e , although they are inappropriate by t h e i r implication of cognitive processes. I use the terms throughout the thesis only to describe evolved proces-ses and place my re s u l t s i n the context of current l i t e r a t u r e , rather than i n t h e i r precise dictionary context. 7 (Holgate 1967, Murphy 1968, Gadgil and Bossert 1970). Organisms with iteroparous l i f e h i s t o r i e s must evolve a procedure for d i s t r i b u t i n g obtained energy among reproduction, maintenance and growth fo r each of the periods where reproduction i s possible. I t i s t h i s a l l o c a t i o n procedure which has the major impact upon f i t n e s s , and which has occupied much of the l i f e h i s t o r y l i t e r a -ture i n the past two decades. A measure related to f i t n e s s which has enjoyed an active commentary i n the l i t e r a t u r e i s reproductive value. Defined o r i g i n a l l y by Fisher (1930), i t i s the average number of young that a female of a r b i t r a r y age i n a stable age d i s t r i b u t i o n can expect to produce at that age and over the remainder of her l i f e , r e l a t i v e to a female at b i r t h . This i s an important concept since i t stresses the variable d i s t r i b u t i o n of reproductive p o t e n t i a l with age. Subsequent variants of reproductive value were: t o t a l reproductive value (Caughley 1970, Charlesworth 1980); eventual reproductive value (Goodman 1967) ; modified reproductive value (Schaffer 1974a); time/age-specific reproduc-t i v e value (Vandermeer 1968); and, residual reproductive value (Williams 1966b). These measures were a l l attempts to deal with v a r i a b l e age d i s t r i b u t i o n or temporal e f f e c t s on age-specific values. The d i s t r i b u t i o n of age-specific reproductive a t t r i b u t e s involves several components: siz e or age at f i r s t maturity, s i z e 8 or age-specific fecundity, reproductive e f f o r t and reproductive l i f e span. With the possible exception of s p e c i f i c fecundity each of these components has received considerable attention (Stearns 1976, 1977, 1980, Charlesworth 1980). Cole's (1954) early treatment of reproduction has been described as 'cost-free' (Cody 1966, Charnov and Schaffer 1973, Giesel 1976) since i t assumed that reproductive output at any age could be increased without any acknowledged e f f e c t on other l i f e h i s t o r y parameters. Williams (1966b) noted the flaw i n t h i s argument and several subsequent studies have attempted to demonstrate the cost of reproduction empirically (Tinkle 1969, Love 1970, Avery 1975, Tinkle and Hadley 1975, Calow 1979, B e l l 1980, H i r s h f i e l d 1980, Minchella and Loverde 1981, Partridge and Farquhar 1981, Clutton-Brock et a l . 1982). Given that a cost to reproduction may e x i s t (in the form of mortality or decreased subsequent fecundity), the c r i t i c a l question becomes: what i s the optimal way to a l l o c a t e age-speci-f i c reproductive e f f o r t to solve s p e c i f i c s u r v i v a l problems? A d d i t i o n a l l y , i s there a set of equivalent solutions comprised of d i f f e r e n t values of the components noted above? These questions encompass research that treats l i f e h i s t o r i e s as e i t h e r deter-m i n i s t i c (fixed energy a v a i l a b i l i t y , stable age d i s t r i b u t i o n , constant reproductive e f f o r t , time-invariant mortalities) i n which density-dependence i s the prime control v a r i a b l e ; or s t o c h a s t i c , i n which variable, density-independent factors 9 control abundance. To solve the deterministic question authors have i n v e s t i -gated the age-specific d i s t r i b u t i o n of reproductive e f f o r t . That there are energetic l i m i t a t i o n s on t h i s e f f o r t i s i n t r i n s i c i n Lotka's (1913) and Fisher's (1930) works and i s presented e x p l i c i t l y by Lack (1947), Cody (1966), Williams (1966b), and Skutch (1967) among others. Reproductive e f f o r t was o r i g i n a l l y defined as the proportion of assimilated energy allocated to reproduction (Williams 1966b), including nongonadal expenditures (e.g. migrations, courtships). Concerning the gonadal products alone, several general conclusions have been drawn. Calow (1979) reviewed s e v e r a l empirical studies which indicated strong negative c o r r e l a t i o n s between reproductive e f f o r t (fecundity) at one age and subsequent su r v i v a l or fecundity of that i n d i v i d u a l . I f fecundity i s an index of reproductive e f f o r t then i t should increase with age (Williams 1966b, Gadgil and Bossert 1970), although Giesel (1974, 1976) noted that the effectiveness of i n -creased age-specific fecundity depended on the congruency of the age and fecundity d i s t r i b u t i o n s , i n other words "... a population fecundity d i s t r i b u t i o n should mirror i t s usual age d i s t r i b u t i o n . " (Giesel 1976:63). For example, exponentially increasing fecun-d i t y with age may have l i t t l e benefit where mortality rates preclude s i g n i f i c a n t numbers of animals from reaching older ages. However, Fagen (1972) provided a numerical counter-example and suggested a r e s t r i c t e d number of organisms (toothed whales) to 10 which i t might apply. Charlesworth and Leon (1976), i n applying Schaffer's (1974a) reproductive e f f o r t model, also demonstrated that reproductive e f f o r t should increase with age but noted that i f the s e n s i t i v i t y of adult s u r v i v a l to reproductive expenditures increases with age, a negative r e l a t i o n s h i p between the two may be adaptive. Both that paper and Charlesworth (1973) pointed out the assump-t i o n of an e v o l u t i o n a r i l y stable strategy (after Maynard Smith and Price 1973, Maynard Smith 1978b) upon which those conclusions r e s t . Charlesworth and Leon (1976:456) elaborated that "Selec-t i o n w i l l thus tend to improve s u r v i v a l and fecundity more ra p i d l y early i n the l i f e h i s t o r y and may even depress them l a t e r on so that the form of the l i f e h i s t o r y . . . r e f l e c t s the more rapid improvement of the e a r l i e r period ...", or that the early stages of l i f e contribute more to t h i s e v o l u t i o n a r i l y stable strategy. In p a r t i a l contrast to these t h e o r e t i c a l treatments, f i e l d studies (Emlen 1970, Giesel 1976) indicate that reproductive e f f o r t does not r i s e continuously with age but instead reaches a maximum early i n reproductive l i f e and decreases thereafter. Giesel (1976) explained t h i s phenomenon by noting that natural mortality may r e s t r i c t the number animals surviving to particpate i n l a t e reproduction, and departing from a deterministic ap-proach, that decreased reproductive e f f o r t l a t e r i n l i f e may 11 enhance s u r v i v a l as a hedge against reproductive uncertainty (also i n Tinkle 1969, Bulmer 1985). The energetics of reproductive e f f o r t have been de t a i l e d for l i z a r d s ( F i t z p a t r i c k 1973, Avery 1975), plants (Harper and Ogden 1970, Ogden 1974, Hickman 1975) and fishes (Wootton 1979). Ener-g e t i c s of f i s h reproduction have received considerable attention due to t h e i r economic importance. These studies have helped to supply a more sound basis for conceptual models of energy a l l o c a t i o n (e.g. Ware 1982). While several studies of the genetic aspects of age-specific reproductive e f f o r t , i n d i c a t i n g the h e r i t a b i l i t y of fecundity c h a r a c t e r i s t i c s , have been published (Perrins and Jones 1974, Charlesworth 1980, Rose and Charlesworth 1981a,b), there has been no demonstration of the genetic mechanism involved. However, Bradshaw (1973), Giesel (1974), and Rose and Charlesworth (1981a) have demonstrated increased f i t n e s s for polymorphic fecundity and hatching t r a i t s , and increasing age-specific s e n s i t i v i t y of fecundity with increasing environmental variance, respectively. In contrast, Giesel and Z e t t l e r (1980) and Giesel et a l . (1982) showed a p o s i t i v e genetic c o r r e l a t i o n between reproductive e f f o r t and s u r v i v a l . The deterministic approach to reproductive e f f o r t a l l o c a t i o n has allowed more precise description of f i t n e s s a t t r i b u t e s but as 12 Giesel (1976:70) commented, " . . . p r a c t i c a l l y every vertebrate that has been investigated has demonstrated the a b i l i t y to regulate i t s l e v e l of reproduction r e l a t i v e to the amount of energy i t has a v a i l a b l e . " Therefore, research into the optimization of repro-ductive t r a i t s i n response to variable reproductive success and resource a v a i l a b i l i t y holds the most promise f o r explaining observed l i f e h i s t o r i e s . Reproductive value E a r l i e r , I noted the major role that the concept of repro-ductive value (after Fisher 1930) and i t s derivatives have played i n the evaluation of f i t n e s s . Reproductive e f f o r t and reproduc-t i v e value are c l o s e l y linked i n that the summation of the former over reproductive l i f e constitutes the raw material of the l a t t e r . Various authors have used a measure of reproductive value i n attempts to determine optimal l i f e h i s t o r i e s , and i n p a r t i c u l a r how age-specific reproductive e f f o r t patterns c o n t r i -bute to optimization. Fisher's reproductive value was previously defined; subsequent variants and t h e i r a t t r i b u t e s are: (i) Total reproductive value (Charlesworth 1980). I t i s the sum over a l l ages of reproductive value at age. I t has been used both to provide a measure of t o t a l reproductive p o t e n t i a l of an age-structured population at any time, and as a measure of the r e l a t i v e merits of sub-populations (genotypes) within the population. A p a r t i c u l a r merit of t h i s measure i s the relaxation of the requirement f o r stable age structure; an important c h a r a c t e r i s t i c when age-specific mortality rates change. ( i i ) Eventual reproductive value (Goodman 1967). This index was defined to y i e l d a measure of the contribution of i n d i v i -duals of a given age to the abundance of the population at a future time, when i t i s i n approximately stable age d i s t r i -bution. I t i s the measure which i s more appropriate for the contribution of mixed-age reproducers, since i t incorporates the generation time. The l a t t e r quantity was defined as the mean age of the parents (female) of a cohort. This measure of reproductive value i s more r e a l i s t i c yet both b i o l o g i -c a l l y and a n a l y t i c a l l y imprecise for mixed-age, broadcast reproducers because paternal influence i s unknown and assumed i n s i g n i f i c a n t . ( i i i ) Modified reproductive value (Schaffer 1974a). This index was introduced to "... disentangle from the reproductive value... the e f f e c t s on future fecundity of varying E i [reproductive e f f o r t ] . " This measure was an attempt to incorporate a variable schedule of incremental fecundity changes such that the incremental changes were independent of previous reproductive e f f o r t , rather they were dependent s o l e l y on the growth of the organism (hence reproductive potential) since the previous reproductive episode. (iv) Age/time s p e c i f i c reproductive value (Vandermeer 1968). Assuming an approximately stable age d i s t r i b u t i o n , Vander-meer 's value i s the t o t a l number of b i r t h s by animals of age 14 i and older, per animal aged i . He noted the major d i f f e -rence between h i s index and Fisher's being that h i s value fluctuated with the age d i s t r i b u t i o n whereas Fishers' was constant. When the age d i s t r i b u t i o n i s stable the two values are equivalent; when the age d i s t r i b u t i o n i s dynamic Vandermeer's index represents the present value of future o f f s p r i n g for that s p e c i f i c age d i s t r i b u t i o n , (v) Residual reproductive value (Williams 1966b). Williams, with others, was concerned with d i s t i n g u i s h i n g whether the annual a l l o c a t i o n of reproductive e f f o r t was always the maximum possible under each environmental s i t u a t i o n , or whether the a l l o c a t i o n was a constant such that surplus energy i n good years might be allocated to the soma (hence survival) for future reproduction. To quantify t h i s future p o t e n t i a l he defined the residual reproductive value as the reproductive value f o r a given age minus that a c t u a l l y committed at that reproduction. Each of these indices of reproductive value was formulated fo r use i n a s p e c i f i c argument so i t i s not s u r p r i s i n g that they d i f f e r i n when and how they can be applied. A measure of t h i s s p e c i f i c i t y i s evident where two authors (Schaffer and Williams) used t h e i r index i n t r y i n g to demonstrate e n t i r e l y d i f f e r e n t concepts of l i f e h i s t o r y ; maximization of age-specific reproduc-t i v e e f f o r t vs. maximizing long-term reproductive p r o f i t on investment, respectively. Each believed that maximizing r was 15 the object but as S l a t k i n (1974) and G i l l e s p i e (1977) noted, t h i s maximization may be achieved by d i f f e r e n t age-specific s t r a t e -gies . The i n v e s t i g a t i o n of optimality under fi x e d reproductive schedules has been generally confined to matrix formulations or other l i f e table analyses (after Lewis 1942 and L e s l i e 1945, 1948). I f organisms are adapting to v a r i a b l e environmental pressures then such fixed-value approaches cannot be e n t i r e l y r e a l i s t i c , although several studies (Schaffer 1974a,b and Taylor et a l . 1974) have shown that maximization of reproductive value at each age i s sometimes analogous to t h i s approach, i n a stable age d i s t r i b u t i o n . Predictions of l i f e h i s t o r y theory L i f e h i s t o r y models incorporating a l t e r n a t i v e a l l o c a t i o n s of energy to reproduction and s u r v i v a l have provided the most ins i g h t into understanding population responses to variable s e l e c t i o n pressures. Murphy (1968), Gadgil and Bossert (1970), Fagen (1972), Charlesworth and Leon (1976), Schaffer and Rosen-zweig (1977), Michod (1979), Schaffer (1979b) and R i c k l e f s (1981a) are noteworthy for exploring r e s u l t s of a l t e r n a t i v e strategies of a l l o c a t i o n , p a r t i c u l a r l y with regard to age at f i r s t reproduction r e l a t i v e to age-specific mortality. Giesel (1976) reviewed the general predictions of these models as follows: 16 (i) When the costs of reproduction increase at a greater rate than the benefits, single-stage or more chronologically l i m i t e d reproduction w i l l evolve. ( i i ) Increased mortality of adults should r e s u l t i n lowered age at f i r s t maturity and increased reproductive e f f o r t p r i o r to ages where mortality increases. ( i i i ) I f costs of reproduction increase continuously with age, reproductive e f f o r t should also increase with age. (iv) Changes i n mortality a f f e c t i n g a l l ages uniformly should e l i c i t no a l t e r a t i o n of the reproductive e f f o r t schedule. (v) Under conditions of chronic resource l i m i t a t i o n , increased resource a v a i l a b i l i t y should y i e l d o v e r a l l increases i n reproductive e f f o r t . (vi) I f mortality i s absorbed primarily by pre-reproductive i n -d i v i d u a l s or the expectation of successful reproduction i s low, then delayed maturity and increased l i f e span w i l l be adaptive. ( v i i ) Conversely to ( v i ) , i f mortality of adults i s r e l a t i v e l y higher, early maturity and increased early reproductive e f f o r t w i l l be favoured. I t i s obvious that some of these predictions may involve responses i n more than one l i f e h i story c h a r a c t e r i s t i c and some authors have been tempted to speculate that a broad set of co-evolved l i f e h i s t o r y c h a r a c t e r i s t i c s or ' t a c t i c s 1 e x i s t (reviewed by Stearns 1976, 1977, also Silvertown 1981). There i s reason to 17 question t h i s and Stearns (1980), a f t e r conducting experimental work i n t e s t of h i s own conjecture, was drawn to comment that • t a c t i c s 1 as such may not ex i s t . He noted that supposedly co-adapted t r a i t s were more often noted i n surveys of broad taxono-mic groups than within species. One c r i t i c i s m that can thus be l e v e l l e d at many of the conclusions drawn from l i f e h i s t o r y models i s that they represent a synthesis of ideas drawn from examination of d i f f e r e n t processes i n d i f f e r e n t species groups. In addition, the system constraints assumed i n some maximization treatments (Schaffer 1974a, Taylor et a l . 1974) have been c r i t i c i z e d as u n r e a l i s t i c (Caswell 1980, R i c k l e f s 1981b); an active debate continues (Schaffer 1981, Yodzis 1981, Bulmer 1985) . While there i s no shortage of l i f e h i s t o r y observations and studies i n the l i t e r a t u r e the majority of these as noted are eit h e r surveys among related taxa or t h e o r e t i c a l studies without supporting f i e l d t e s t s . There are r e l a t i v e l y few studies which have investigated the predictions of l i f e h i s t o r y theory under f i e l d experimental conditions. Solbrig and Simpson (1974) inves-ti g a t e d the e f f e c t s of disturbance i n s e l e c t i n g for dandelion biotypes with d i f f e r e n t a l l o c a t i o n s of energy to reproductive and vegetative growth. They found that increased density-indepen-dent, v a r i a b l e mortality at one s i t e yielded increased reproduc-t i v e e f f o r t r e l a t i v e to a control s i t e . While these r e s u l t s conform to t h e o r e t i c a l predictions, Charlesworth (1980) noted 18 that age-specific data were not c o l l e c t e d . In a s i m i l a r experi-ment Law et a l . (1977) and Law (1979a) examined two types of meadow grass, one c h a r a c t e r i s t i c of colonizing s i t u a t i o n s and one of more stable pasture areas. They found, as predicted, higher reproductive e f f o r t i n the opportunist form i n i t s f i r s t repro-ductive season. Doyle and Hunte (1981) applied a r t i f i c i a l s e l e c t i o n to amphipod cultures and obtained higher r values, as well as lower age and variance of age at f i r s t maturity. They deduced that the primary increase i n f i t n e s s was due not so much to increased f e r t i l i t y as to the decrease i n the number of i n f e r t i l e adult females. Ekman and Askenmo (1986) also showed reproductive cost, as interpreted from higher s u r v i v a l of non-breeding male willow t i t s . Several a r t i f i c i a l a l t e r a t i o n s of reproductive success i n birds have been conducted. Askenmo (1979) increased the brood s i z e of flycatchers and induced greater mortality of the adults, while Kluyver (1971) was able to increase adult s u r v i v a l of great t i t s by decreasing the brood s i z e . With the exception of some studies noted i n the following section, there have been few experimental studies of strongly iteroparous species. The i n t e r -pretation of other studies may be confounded by environmental e f f e c t s (Stearns 1977, Reznick 1985), or the lack of d e t a i l on age-specific e f f e c t s has not provided conclusive t e s t s of theory (Sokal 1970). A detailed, though subjective, review of the empirical evidence for reproductive cost has been presented by 19 Reznick (1985). Both that study and Tuomi et a l . (1983) stressed that environmental e f f e c t s and physiological buffering may act to mask the detection of reproductive costs i n nature. L i f e h i s t o r y theory and harvested populations Unexploited, iteroparous species generally have r e l a t i v e l y s t a t i c mortality schedules or p r o b a b i l i t i e s of mortality i n the adult stage, and expl o i t a t i o n of such populations has been widespread among fishes. Unfortunately, i t i s only for a very few populations that l i f e h istory data have been c o l l e c t e d throughout the expl o i t a t i o n history. Daan (1975), i n reviewing the dynamics of North Sea cod, noted a smaller s i z e at maturity following f i s h i n g but concluded that i t provided l i t t l e compensa-t i o n for the e f f e c t s of age-group reduction. He also suggested that recent increases i n juveniles were related to decreased adult depensatory mortality. Garrod and Knights (1979) i n v e s t i -gated several systematic groups over broad geographic areas and d e r i v e d a generalized stock-recruit r e l a t i o n s h i p based on a g e - s p e c i f i c reproduction. Their conclusion that delayed maturity yielded decreased eggs/recruit i s unique although i t may have resulted from the weighting procedure used. Boyce (1981) examined unharvested and harvested beaver populations and found that the harvested group had a lower age at f i r s t maturity but no change i n growth rate. The l a t t e r i s unusual, as compensatory growth rate increases i n exploited 20 populations are common (e.g. Ricker 1975), and comprise the major component of changes i n age at f i r s t maturity. However, the inverse r e l a t i o n of variance i n s i z e at age and age i n Boyce's data suggests possible errors i n ageing, and h i s r e s u l t s should be interpreted cautiously. Highly iteroparous species have evolved d i s t i n c t i v e l i f e h i s t o r i e s which are presumably adaptive to the environmental challenge of uncertain reproductive success. Demersal f i s h species include some of the most extreme of t h i s group. The genus Sebastes i s one of the most speciose o f f the P a c i f i c coast of North America ( M i l l e r and Lea 1972, Chen 1986). P a c i f i c ocean perch (S. alutus (Gilbert)) i s a widespread species of the genus, ranging from northern C a l i f o r n i a to the Bering Sea and i n depth from 40 to at l e a s t 600 m (Hart 1973) . Its colour, keeping q u a l i t i e s and o v e r a l l abundance have combined to render i t the most sought a f t e r and valuable of the Sebastes species i n the eastern P a c i f i c Ocean. However, both h i s t o r i c a l and recent investigations have indicated that e x p l o i t a t i o n of t h i s species has proceeded without reference to i t s evolved l i f e h i s t o r y c h a r a c t e r i s t i c s (Leaman and Beamish 1984, Leaman 1987a). In p a r t i c u l a r , the age-structured nature of i t s populations has been ignored. Ages of S. alutus have t r a d i t i o n a l l y been estimated from surface readings of o t o l i t h s (Westrheim 1973) or scales ( G r i t -senko 1963, Chikuni and Wakabayashi 1970) but the a p p l i c a t i o n of t h i n sectioning (Beamish 1979) or breaking and burning (Archibald et a l . 1981) techniques has revised estimates of maximum age from a 40y range to an 80y range. L i f e h i s t o r y theory suggests that such a species w i l l have evolved c h a r a c t e r i s t i c s adapted for longevity. I t i s the p l a s t i c i t y of some of these c h a r a c t e r i s t i c s under the influence of altered demography and density that t h i s project investigated. S p e c i f i c a l l y , I was concerned with deter-mining whether age/size s p e c i f i c fecundity, reproductive e f f o r t , reproductive value, growth rates, and size/age at f i r s t maturity were s u f f i c i e n t l y p l a s t i c under ex p l o i t a t i o n to provide some compensatory response to the e f f e c t s of age-group reduction and loss of t o t a l reproductive value. Groups of S. alutus which have been subjected to substan-t i a l l y d i f f e r e n t e x p l o i t a t i o n h i s t o r i e s provided an opportunity for examination of population responses to al t e r e d demography, and whether the responses were of magnitude s u f f i c i e n t to compensate for reduction i n adult s u r v i v a l . The e x p l o i t a t i o n h i s t o r y of these groups suggests some segregation at the adult stage and they can be considered as stocks i n the f i s h e r y context (MacLean and Evans 1981). However, the degree of b i o l o g i c a l segregation i s not c l e a r . The hi s t o r y of these stocks i n B.C. waters has been detailed by Westrheim et a l . (1972), Gunderson (1977), Gunderson et a l . (1977), Ketchen (1980, 1981), Archibald et a l . (1983), Leaman and Nagtegaal (1982) and Leaman (1987a). B r i e f l y , they consist of three highly exploited stocks (west Vancouver Is., Queen Charlotte Sound, and Langara S p i t ) , one with a short but intensive ex p l o i t a t i o n h i s t o r y (west Queen Charlotte Is.) and one with only recent e x p l o i t a t i o n (southern Hecate S t r a i t ) . Fishing e f f o r t has thus conducted a fortuitous (from a s c i e n t i f i c standpoint) experiment on the demography and abundance of these stocks. The three groups of stocks are approximately 15%, 60%, and 90+% of t h e i r unexploited biomass, respectively (Westrheim 1980, Stocker 1981, Leaman 1985) and provided a system for examining how the responses to t h i s density-independent adult mortality may contribute to some mechanisms postulated by l i f e h i s t o r y t h e o r i s t s . L i f e h i s t o r y theory suggests several c h a r a c t e r i s t i c s as p o t e n t i a l indicators or determinants of f i t n e s s : population growth rate, l i f e span, mortality and n a t a l i t y schedules, i n d i v i d u a l growth rate, age (size) at f i r s t maturity, reproduc-t i v e value (or residual reproductive value), reproductive e f f o r t , t o t a l stock fecundity, reproductive l i f e span, breeding frequency and recruitment v a r i a b i l i t y . Much of the l i f e h i s t o r y l i t e r a t u r e has been concerned with the pot e n t i a l linkages (or at least covariance) of these parameters, t h e i r mean values and t h e i r variances. This study concentrated on a smaller subset of t h i s group. The l i f e span (80+y) and generation time (30+y)of P a c i f i c 23 ocean perch prohibited t e s t i n g of responses i n a single popula-t i o n . However, the d i f f e r e n t stocks noted above and the general-l y conservative biology of the species permit examination of population responses by analogy. The l i f e h i s t o r y hypotheses examined were: 1. In a strongly iteroparous organism does an increase i n adult mortality r e s u l t i n a lowering of the age at f i r s t maturity ( t m ) ? Associated with the lowered t m i s there an increase i n reproductive e f f o r t of younger f i s h , r e l a t i v e to the unexploited state? Increased reproductive e f f o r t may be through increased fecundity, egg (offspring) s i z e or spawning frequency (Gadgil and Bossert 1970, Schaffer 1974b, R i c k l e f s 1977, among others). 2. I f (1) i s true, does increased reproductive e f f o r t at younger ages decrease the rate of su r v i v a l to older ages (Tinkle 1969, Bostock 1978, Law 1979a,b, Mann and M i l l s 1979, B e l l 1980)? 3. Does adaptation act so as to maximize residual reproductive value, rather than reproductive e f f o r t , i n any given year (Williams 1966a; Schaffer 1979a,b)? The objectives of t h i s study were: (i) to e s t a b l i s h whether groups of S. alutus previously delineated by e x p l o i t a t i o n h i s t o r i e s could be objectively i d e n t i f i e d as separate stocks ( i i ) to i d e n t i f y the e f f e c t s of density-independent mortality caused by e x p l o i t a t i o n ; ( i i i ) to examine some of the predictions of l i f e h i s t o r y theory i n a natural system, for a long-lived iteroparous animal; (iv) to examine the long-term implications of t h i s mortality schedule or s e l e c t i v e pressure on the dynamics of these sub-populations; and, (v) to determine the contribution of these r e s u l t s i n the evaluation of present l i f e h i s t o r y theory. The t h e s i s i s organized into three broad categories, which follow a presentation of general methodology. F i r s t l y , the existence of separable sub-populations or stocks of S. alutus i s examined. Secondly, the differences i n the physical and repro-ductive c h a r a c t e r i s t i c s of f i s h i n these stocks, generated by varying l e v e l s of density-independent mortality, are determined. F i n a l l y , I present the re s u l t s of simulation modelling of the dynamics of these stocks and discuss t h e i r relevance to general l i f e h i s t o r y theory and fishery management. 25 I I . GENERAL METHODS AND MATERIALS 1. Reproductive biology sampling The majority of samples were c o l l e c t e d from the R/V G.B. REED, a 53.9 m side trawler, using a Nor'Eastern bottom trawl with 8.9 cm mesh codend. This trawl had a 32 m footrope equipped with 46 cm bobbins on the groundline, and was rigged with three b r i d l e s to permit a large v e r t i c a l opening («9.4 m). Hauls were made at locations where S. alutus was known to occur from previous research and commercial trawling (Lapi and Richards 1981, Leaman and Nagtegaal 1982). The Sebastes spp. have in t e r n a l f e r t i l i z a t i o n with a delay between insemination and f e r t i l i z a t i o n . In the waters o f f B r i t i s h Columbia, insemination generally occurs i n September and f e r t i l i z a t i o n i n December. Col l e c t i o n s were therefore made during November, when oocytes would be at t h e i r maximum, u n f e r t i l i z e d s i z e and confusion of immature and mature oocytes would be minimized. Sampling occurred at deeper depths (250-400 m) than normally sampled because of the seasonal bathymetric migration of the species (Gunderson 1971, Ketchen 1981, Leaman 1985) . A l l f i s h were sampled from catches <1 t . Larger catches were subsampled by the method of Westrheim (1967, 1976), except that the subsample sizes were increased (four tubs vs. two) to accommodate the larger number of specimens required. The s a g i t t a l o t o l i t h s were extracted and stored i n a glycerine/-water/thymol solution for subsequent reading ashore. Fork length, sex, maturity, round weight, and ovary weight data were obtained f o r a l l sampled f i s h . Maturity state was assessed from external appearance of the gonads according to the c r i t e r i a i n Table 1 (Leaman et a l . 1985). Weights of round f i s h and ovaries were obtained with an e l e c t r o n i c balance (K-TRON DP-1) to ± 0.1 g-While the influence of maternal s i z e (length or weight) on reproductive c h a r a c t e r i s t i c s i s well known, the contribution of maternal age has seldom been investigated f o r iteroparous species. Unfortunately, age-specific sampling was not p r a c t i c a l since the time required to age the specimens i n the f i e l d would have been p r o h i b i t i v e . Body length was therefore used as a sampling correlate of age. The determinate growth of S. alutus (Archibald et a l . 1983, Leaman 1987a) meant that a length-s t r a t i f i e d sampling scheme was required i n order to have adequate representation of older f i s h . The only previous data upon which to base sample si z e estimates were those contained i n Gunderson (1976). His data showed an average c o e f f i c i e n t of v a r i a t i o n i n fecundity at length of 0.218. From Cochran (1977), i f the CV i s known, the number of samples necessary for the desired l e v e l of p r e c i s i o n i s : where C = (CV) 2 S = standard deviation from preliminary sampling Y = mean from preliminary sampling. The appropriate sample sizes for p r e c i s i o n of 10% and 5% with error p r o b a b i l i t y or=0.05 would be 5 and 19, respectively. Table 1. Description of S. alutus maturity stages. Code External gonad condition and colour Females 1 Immature (translucent, yellow) 2 Maturing (small, yellow eggs; translucent or opaque) 3 Mature (large, yellow eggs; opaque) 4 F e r t i l i z e d (large, orange-yellow eggs; trans-lucent) 5 Embryos or larvae (includes eyed eggs) 6 Spent (large, f l a c c i d , red ovaries; residual larvae and eggs may be present) 7 Resting (moderate siz e , form, red-grey colour) Males 1 Immature (translucent, s t r i n g - l i k e ) 2 Maturing (swelling, brown-white) 3 Mature (large, white; e a s i l y broken) 4 Ripe (sperm i n duct or running) 5 Spent ( f l a c c i d , red) 6 Resting (ribbon-like; small, brown) While sample si z e of 5 f i s h per cm i n t e r v a l might therefore have been appropriate there were additional variance considera-tions about the data presented by Gunderson (1976). In part, the v a r i a t i o n i n h i s data arose through the log-normal d i s t r i b u t i o n of CV with s i z e . This may have been because h i s fecundity estimation procedure did not employ a target CV and there was 28 only one f i s h per s i z e i n t e r v a l (mm) . This high variance i n measurements meant that the CV may not have been close to the true value. In such instances the more involved formula of Cochran (1977) must be used: 4 n. = t C Yx2 S\ 2 1 + 8C + n i V n i where: S;L and 7 are the mean and variance of the f i r s t sample; C i s the desired (CV) 2; i s the i n i t i a l sample number; and n-j- i s the f i n a l sample number. The r e s u l t s of t h i s a p p l i c a t i o n by s i z e i n t e r v a l to Gunder-son's data vary considerably, e.g. at 3 9 cm, n^=5 and nt=13 at 37 cm, n;i=6 and n--=41, for desired CV=0.10. The high variance among sizes i n these data gave r i s e to the range i n sample si z e s . I therefore elected to use the sample si z e for the length i n t e r v a l s where CV was well determined (5 samples per cm) as guideline, but to modify the subsampling to account for determinate growth and the eventual need to determine reproductive characters by age. I t was also apparent that sub-sampling the ovaries for fecundity estimation should be based on a target p r e c i s i o n . A sample of f i v e f i s h per age group required s t r a t i f i e d 29 sampling by length. Previous age-length matrices for t h i s species (Shaw and Archibald 1981) were used to estimate the age composition by length i n t e r v a l and the sample a l l o c a t i o n was thus: <39 cm - 5 samples per cm 40-43 cm - 10 samples per cm >43 cm - 20 samples per cm. This s t r a t i f i c a t i o n was employed a f t e r a random sample of 100 o t o l i t h s had been c o l l e c t e d for determination of age and sex composition. Ovaries retained for laboratory processing were stored i n e i t h e r modified Gilson's f l u i d (fecundity estimation) or Smith's formal dichromate (histology) (Gray 1954). The composition of these solutions was: Gilson's (1897) F l u i d Equal parts of: g l a c i a l a c e t i c acid chloroform 60% ethyl alcohol Smith's formal dichromate (acetic) Equal parts of: Solution A: 8% KCr20 7 i n H 20 and Solution B: 9% formaldehyde (40%) 5% g l a c i a l a c e t i c acid 86% d i s t i l l e d H 20 (after 48 h transfer to 3% formaldehyde f o r storage) 30 Ovaries i n Gilson's f l u i d were vigorously agitated every other week for three months, to a s s i s t i n the d i s s o l u t i o n of connective t i s s u e and l i b e r a t i o n of the oocytes. For those ovaries where oocytes were more d i f f i c u l t to l i b e r a t e , the Gilson's f l u i d was changed a f t e r two months. Additional samples of p o s t - f e r t i l i z a t i o n and post-spawning f i s h were obtained i n February, from commercial landings of S. alutus. These samples were used s o l e l y for h i s t o l o g i c a l pur-poses, since the handling practices aboard vessels and i n processing plants give r i s e to extrusion of the embryos and introduce bias i n fecundity estimation. Ages were determined from the o t o l i t h s by the Ageing Labora-tory at the P a c i f i c B i o l o g i c a l Station, using the break-and-burn technique. In t h i s technique the o t o l i t h i s broken across the focus, the cross-section burned over an alcohol flame, coated with cooking o i l to enhance contrast, and the annular zones enumerated under r e f l e c t e d l i g h t , using f i b r e optics and a Wild M8 d i s s e c t i n g microscope (Chilton and Beamish 1982). Annular zones consisted of alternating opaque and hyaline zones and were distinguished from checks by t h e i r continuity from eith e r the dorsal or ventral t i p to the edge of the sulcus acousticus. Burning of the cross-section also i n t e n s i f i e s the difference between true annuli and checks. 31 2. Parasite sampling A l l parasite (Neobrachiella robusta (Wilson)) samples used i n t h i s analysis were obtained within a two week period, and where possible, from two d i f f e r e n t depths at each s i t e . Expenda-ble bathythermograph (XBT) readings of surface and bottom temper-atures at each s i t e were also made. The g i l l s were removed from the f i s h and placed i n i n d i v i d u a l j a r s containing 10% formalin. The g i l l s were subsequently examined i n the laboratory under a dis s e c t i n g microscope. The prevalence and i n t e n s i t y of i n f e c t i o n with N. robusta was determined and the copepods sorted to sex and stage, a f t e r Kabata (1987). Parasites were divided into the following categories: (i) Early stages. This category contained stages from copepodid to chalimus IV; ( i i ) Preadults. Designated a l l those parasites that had changed from the f r o n t a l filaments to the b u l l a as the organ of attachment, but had yet to a t t a i n the d e f i n i -t i v e s i z e and shape; ( i i i ) Non-ovigerous females. Those specimens a t t a i n i n g f u l l s i z e but without egg sacs; and (iv) Ovigerous females. Adult parasites with egg sacs. D i f f e r e n t i a t i o n of stock units was examined by comparison of mean values per f i s h f or prevalence and i n t e n s i t y of i n f e c t i o n , as well as numbers of juvenile, adult, ovigerous, and non-32 ovigerous females, for those stock units that could not be d i f f e r e n t i a t e d on the basis of host s i z e . Similar t e s t s were also conducted on the basis of depth s t r a t i f i c a t i o n within putative stock u n i t s . The discriminating power of using the parasite to c l a s s i f y i n d i v i d u a l f i s h to stock was examined using discriminant function analysis (Rao 1973, SAS 1982) on subsets of samples grouped by the s i z e composition of the host. 3. Fecundity and oocyte c h a r a c t e r i s t i c s The mortality e f f e c t s of e x p l o i t a t i o n were examined through t e s t s of reproductive c h a r a c t e r i s t i c s among stock units and e x p l o i t a t i o n groups ( l i g h t l y exploited and heavily exploited). (i) Fecundity estimation. The most recent work concerning r o c k f i s h fecundity (Gunderson 1976, 1977; Boehlert et a l . 1983) employed a volumetric method of estimation (Bagenal and Braum 1968). In t h i s method the l i b e r a t e d oocytes suspended i n 2000 mL of water are s t i r r e d and 2 or 5 mL subsamples (4-6) taken. In early work a 2 mL Stempel pipette was used fo r sub-sampling (Raitt 1933, Simpson 1951), however Kandler and Pirwitz (1957) found that the Stempel pipette did not y i e l d a representative subsample. In spite of t h i s finding, some workers (Mason et a l . 1983, Mason 1984) continue to use a Stempel pipette, although t o t a l volume containing the oocytes i n suspension i s reported as 10-15000 mL from which 55 subsamples are taken. Error s t a t i s t i c s are not always included i n published studies, however the CV for volumetric subsample means was reported by Boehlert et a l . (1983) as 0.6-10.2% with mean of 4.2% (2 mL subsamples). Gunderson (1976) reported mean CV of 11.5% and 10.5% and ranges of 4.0-22.0% and 2.3-24.0% (5 mL subsamples), for Vancouver Island and Queen Charlotte Sound samples, respectively. Mason et a l . 1983 and Mason (1984) did not report s t a t i s t i c s f o r i n d i v i d u a l subsamples but t h e i r data imply a range i n CV of 13-70%. I performed comparative tests (analysis of variance, ANOVA) of these methods using the same oocyte t e s t sample, with replace-ment. In addition to the estimation of fecundity I also examined the properties of the methods i n r e l a t i o n to the s i z e d i s t r i b u -t i o n of the oocytes extracted with each technique. The t e s t oocyte sample was mixed i n 3500 mL of water and samples of 2, 5, and 10 mL were extracted with a regular bulb pipette, and 2 mL samples using a Stempel pipette. In addition to the t e s t s of subsample s i z e I also examined whether sampling l o c a t i o n within a suspension of oocytes could be a source of error. Several authors have noted the need for complete mixing of the suspension because of the variable s i z e and/or s p e c i f i c gravity of the oocytes (Bagenal and Braum 1968, Gunderson 1976). To t e s t the influence of p o s i t i o n within the suspension from which the sample was taken and the absolute accuracy of the estimators I sampled a f i x e d volume containing 2000 oocytes f i v e times from f i v e 34 locations with both 2 and 5 mL subsamples. The r e s u l t s of the volumetric subsampling led me to i n v e s t i -gate a l t e r n a t i v e methods of estimation. Gravimetric subsampling has been reported to have lower variance than volumetric subsam-p l i n g for some species (McGregor 1922, Wolfert 1969, Bagenal 1978) and I examined t h i s for S. alutus ovaries. The method used i s described i n Table 2. B r i e f l y , approximately equal-weight subsamples (± .01 gm) were taken from f i l t e r e d oocytes, counted twice i n a counting chamber, r e f i l t e r e d , dessicated f o r 24h at 50° C and weighed to ± 0.1 mg. Subsampling was continued to a target p r e c i s i o n of 5% CV i n oocytes/gm of ovary weight. The appropriate time for desiccation to constant weight was examined with s i x subsamples of oocytes. The r e p e a t a b i l i t y of these methods was examined on a subsample of ovaries, a f t e r they had been returned to storage for s i x months. In addition to fecundity several other measures of reproduc-t i v e e f f o r t were examined for t h e i r s e n s i t i v i t y to population changes. These measures were s p e c i f i c fecundity (fecundity/-somatic weight) and two gonadal indices, GIS (ovary weight/soma-t i c weight) and GIR (ovary weight/round weight). A l l were investigated because consideration of fecundity alone y i e l d s no understanding of how the apportioning of t o t a l energy varies with body or age c h a r a c t e r i s t i c s . Multiple regressions of fecundity and the gonadal indices against s i g n i f i c a n t body variables and age were compared among stocks and e x p l o i t a t i o n units. Table 2. Gravimetric method of fecundity estimation and oocyte c h a r a c t e r i s t i c s for S. alutus. 1. Digested oocyte suspension drained and f i l t e r e d through stacked 100-750 Lira sieves. 2. Oocytes transferred to pre-weighed p l a s t i c M i l l i p o r e f i l t e r s . 3. Oocytes suctioned for 3 min to remove excess l i q u i d . 4. Oocytes and f i l t e r weighed to ±0.01 g. 5. Four subsamples (approx. 0.20 g) taken with spatula and transferred to pre-weighed p l a s t i c containers, weighed to 0.0001 g, placed i n covered p e t r i dishes. 6. Subsamples placed i n counting chamber with small amount of water and counted twice. Extrapolation based on subsample wt.:total wt. r a t i o y i e l d s fecundity. 7. Subsample re-suctioned and placed back into o r i g i n a l weighed container. 8. Subsamples placed i n dessicator and dried f o r 24 h at 50° C. 9. Dry weight of subsamples obtained to 0.0001 g. For the estimation of oocyte diameters 1. Subsample of approximately 1000 oocytes taken from f i l t e r e d oocytes and placed i n covered p e t r i dish with g r i d pattern etched on bottom. 2. Using a c a l i b r a t e d ocular micrometer, each oocyte i n two rows of the g r i d measured to ±70 Lim to obtain oocyte diameters. 3. Irregular oocytes measured by length and width. ( i i ) Oocyte c h a r a c t e r i s t i c s . Changes i n reproductive e f f o r t can occur through changes i n either or both of oocyte numbers and in d i v i d u a l oocyte c h a r a c t e r i s t i c s . I examined the minimum, maximum, and mean oocyte diameters, as well as oocyte dry weight, i n r e l a t i o n to body weight and age by stock and ex p l o i t a t i o n group, to t e s t whether ex p l o i t a t i o n i s associated with any differences i n these characters. For each ovary a subsample of approximately 1000 oocytes from the f i l t e r e d sample was placed i n a covered p e t r i dish, which had a g r i d pattern etched into the bottom of the dish. Using a c a l i b r a t e d ocular micrometer i n a di s s e c t i n g microscope (Wild M5) , the diameter of each oocyte i n two rows of the g r i d was measured to ± 70 jum. Diameters of i r r e g u l a r oocytes were estimated as the average of the lengths of the major and minor axes. Oocyte dry weight was estimated as the mean of the e s t i -mates from the fecundity subsamples. Multiple regressions of oocyte sizes (diameter) and oocyte weight against body variables, fecundity and age were compared among stocks and expl o i t a t i o n units. I.also examined the r e l a -t ionship of oocyte weight and oocyte diameter among stocks through comparison of t h e i r regression l i n e s . Differences among regressions were tested with an F-test (Zar 1984). ( i i i ) H i s t o l o g i c a l studies. H i s t o l o g i c a l sections were taken from s i x locations within each ovary to determine i f assessment of maturational state or oocyte sizes was influenced by place of o r i g i n within the ovary, and i f there was any gradient of development within the ovary. Samples were imbedded, sectioned, mounted, stained with Harris* hematoxylin and counterstained with a l c o h o l i c eosin (Gray 1954). Rectangular axes were marked on a l l sections and diameters of 20 mature and 20 immature oocytes were measured (± 10 Lira) along the axes with a c a l i b r a t e d ocular micrometer, i n a compound microscope (Olympus BHA). Measurements were taken only of those oocytes sectioned through the nucleus, to avoid the p o s i t i v e bias i n mean diameter that would otherwise be generated by the higher encounter p r o b a b i l i t y of larger oocytes. Previous work (Foucher and Beamish 1980) with oocytes of s i m i l a r s i z e and nucleoplasm/cyto-plasm r a t i o (Merluccius productus Ayres) showed the average error i n estimation of true oocyte diameter, when measuring any oocyte sectioned through the nucleus, to be 5.4%. Photographs of repre-sentative and unusual features were made with a Wild MPS12 camera on the same microscope. H i s t o l o g i c a l sections of both pre- and p o s t - f e r t i l i z e d ovaries were examined to investigate several features of matura-t i o n and reproduction: maturation of i n d i v i d u a l f i s h ; maturation of oocytes; f o l l i c u l a r a t r e s i a ; oocyte sizes within the ovary; f e r t i l i z a t i o n of oocytes; range of developmental stages; and 38 multiple spawning. The same tests of oocyte diameter differences between ex p l o i t a t i o n groups, as were conducted f o r oocyte diameter from the fecundity sampling, were also conducted with data from the h i s t o l o g i c a l sampling. The oocyte diameters from these two data sets are not i d e n t i c a l because the oocytes imbibe some of the Gilson's f l u i d and water during the fecundity estimation process. I also examined the r a t i o of a t r e t i c to non-atretic f o l l i c l e s among e x p l o i t a t i o n groups with ANOVA. 4. Size and weight at age A d e t a i l e d examination of si z e and weight at age, and implied growth rates, among stocks and e x p l o i t a t i o n groups was undertaken. I f i t t e d von Bertalanffy (a type of negative-exponential function) and quadratic curves to lengths and weights at age using nonlinear, least-squares c r i t e r i a with commercial software (SAS, SYSTAT). Both the standard von Bertalanffy function (Ricker 1975) and the more robust form suggested by Schnute (1981) were employed. Adequacy of f i t and i n t e r p r e t a t i o n of growth was made through examination of the d i s t r i b u t i o n and autocorrelation of residual error. The r e s u l t s of t h i s work led to c o l l aboration with a colleague (Dr. T. J . Mulligan) on the in v e s t i g a t i o n of model ambiguity i n the i n t e r p r e t a t i o n of growth through the examination of length-at-age data (Mulligan and Leaman, i n prep.). Some aspects of t h i s work are reported here. 39 5. Modelling studies A simulation model was constructed i n stages to: examine the reproductive performance of a cohort throughout i t s l i f e ; deter-mine population responses to both e x p l o i t a t i o n pressure and recruitment uncertainty; and to evaluate several indices of reproductive value as measures of stock condition. The model was b u i l t i n stages both to obtain cohort s p e c i f i c reproductive information and to permit v a l i d a t i o n of the components. In i t s ultimate form i t i s structured around the standard exponential mortality equations and a stochastic stock-recruitment r e l a t i o n -ship. I t i s composed of simple growth, mortality and recruitment modules, driven by a cycle and r e p l i c a t e c o n t r o l l e r . The major control variables are f i s h i n g and natural mortality rates together with the variance of the stock-recruit r e l a t i o n s h i p . A description of the key processes and the estimation of t h e i r parameters i s included i n Chapter VI. 40 I I I . HISTORY OF EXPLOITATION AND IDENTIFICATION OF STOCKS 1. History of e x p l o i t a t i o n The presumed stocks of Sebastes alutus are located o f f southwest Vancouver Island, i n Queen Charlotte Sound (Goose Island G u l l y ) , i n Hecate S t r a i t (Moresby G u l l y ) , o f f the west coast of the Queen Charlotte Islands (Rennell Sound), and i n Dixon Entrance (Langara Spit) (Fig. 1). Demographic and density manipulations have been substantial (Fig. 2) and they range from severe (Langara Spit, Goose Island Gully, Vancouver Island) through short-term intensive (Rennell Sound) to short-term moderate (Moresby G u l l y ) . The catch data for the Langara Spit, Goose Island Gully and Vancouver Island stocks i n F i g . 2 repre-sent the best resolution of both domestic and foreign f i s h e r i e s s t a t i s t i c s (Ketchen 1980, Leaman 1985). The major p a r t i c i p a n t s i n these f i s h e r i e s were the USSR, Japan, the USA and Canada (Table 3) . S t a t i s t i c s of the Soviet fishery i n p a r t i c u l a r are very poor. Documentation of the Japanese fi s h e r y i s considerably better concerning area and gross quantity of catch but the problem of accurate species composition i s as great as with the Soviet data. The l a t t e r condition e x i s t s because a l l red r o c k f i s h were reported as 'ocean perch' and included any or a l l of S. aleutianus (Jordan and Evermann), S. alutus, S. babcocki (Thompson) , S. proricrer (Jordan and Gilbert) , S. reedi (Westrheim and Tsuyuki) , S. variectatus Quast, and S. zacentrus (Gilbert) . Ketchen (1980) examined these data i n d e t a i l and derived h i s t o r i -c a l catch s e r i e s for the Vancouver Island, Goose Island Gully and 41 134° 132° 130° 128° 126° 124° F i g . 1. S e b a s t e s a l u t u s s t o c k s ; A -I s l a n d G u l l y , C - Moresby G u l l y , D -Southwes t Vancouve r I s l a n d , B - Goose R e n n e l l Sound, E - L anga r a S p i t . 42 o - o RENNELL SOUND • — • MORESBY GULLY YEAR LANGARA SPIT QUEEN CHARLOTTE SND. VANCOUVER ISLAND 1964 YEAR Figure 2. Catch history of S. alutus stocks o f f B r i t i s h Columbia 43 Table 3. Estimated landings of P a c i f i c ocean perch (Sebastes  a l u t u s ) r by stock, 1959-1985. Year Vancouver 3 Island Goose Is. Gully Moresby Gully R e n n e l l b Sound Langara' Spit 1959 968 1890 _ 1960 1575 1679 - - — 1961 2485 1199 - - — 1962 3857 1838 - — — 1963 3769 3712 - - — 1964 2095 3450 - - — 1965 3468 7478 - - 24740 1966 16286 20752 - - 16196 1967 13468 12119 - - 8163 1968 10392 10213 - - 9096 1969 3357 6872 - - 4328 1970 3847 6489 - - 1671 1971 3595 3455 2 . - 3033 1972 2567 5645 - - 4469 1973 3787 3755 - - 3514 1974 1444 7269 10 - 2442 1975 818 4209 97 - 1833 1976 1251 2442 43 79 1992 1977 913 1693 41 1549 2822 1978 1014 865 162 2414 22 1979 741 951 225 839 227 1980 835 1226 2433 877 85 1981 790 801 2166 599 109 1982 830 570 3562 614 342 1983 1147 1215 2204 835 292 1984 1291 841 2042 841 2174 1985 843 759 1939 830 1938 aINPFC Vancouver Area (47°30'-50 o30'N). Includes catches from 52°00 1-54°00'N. c1965-1976 data from 54 000 1-54°30'; 1979-1985 data from 54°00'-pr o v i s i o n a l U.S.-Canada boundary area (approx. 54°25 ,N). 44 Langara Spit stocks based on observer reports, surveillance f l i g h t s , p a t r o l boat sightings, and s t a t i s t i c s of performance by vessel class for the Soviet f l e e t . I t i s doubtful that further improvements i n resolution of these catch data are possible; F i g . 2 contains h i s 'middle' estimates for the Soviet and Japanese f i s h e r i e s . Catch data for the Rennell Sound and Moresby Gully stocks were derived from the monitoring program of the Department of F i s h e r i e s and Oceans (B i o l o g i c a l Sciences Branch). E x p l o i t a t i o n of these two stocks began i n l a t e 1976 and 1979, respectively. A l l stocks have been under quota management since 1977 although management has been i n e f f e c t u a l for the Goose Island Gully stock and suffered a notable lapse for the Moresby Gully stock i n 1982 (Leaman 1985). Present and past biomass estimates were derived with a v a r i e t y of methods including biomass surveys, regression of f i s h e r i e s s t a t i s t i c s and a n a l y t i c models, and are reviewed i n Leaman (1985). Table 4 presents these estimates and the e s t i -mated percentage reduction of each stock from i t s unexploited or l i g h t l y exploited state. Clearly, the Vancouver Island, Goose Island Gully and Langara Spit stocks have suffered major reduc-tions i n standing biomass, p a r t i c u l a r l y over the 1965-1970 period. Catch rates (CPUE, t/h) of the commercial f i s h e r i e s for most stocks are well below peak l e v e l s and for stocks which have a longer time-series of data, declines were commensurate with those i n catch over the 1965-1975 period (Archibald et a l . 1983; Leaman 1985, 1987b). High l e v e l s of instantaneous f i s h i n g mor-t a l i t y rate (F) have been maintained by even the r e l a t i v e l y low l e v e l of e f f o r t from the domestic f l e e t i n recent years (Archi-bald et a l . 1983) . Table 4. Estimated biomass (t) of the f i v e P a c i f i c ocean perch (Sebastes alutus) stocks p r i o r to major f i s h e r i e s and at present. Estimated biomass Stock I n i t i a l Present % reduction Vancouver Island 53,000-81,000 1,850 97-98 Goose Is. Gully 82,000 6,700 92 Moresby Gully 35,500 32,000-34,000 4-10 Rennell Sound 14,000-22,000 12,000-20,000 14-45 Langara Spit 104,000 2,800 97 2. Previous evidence for stock i d e n t i f i c a t i o n P a c i f i c ocean perch has not been successfully tagged with conventional external tags because the mortality associated with t h e i r capture i s almost t o t a l . The species i s physoclistous and the barotrauma (major i n t e r n a l i n j u r i e s r e s u l t i n g from the expansion of gases i n the swim bladder and c r a n i a l sinuses) of r e t r i e v i n g the animals from 200-400 m depths i s consistently 46 f a t a l . D e f i n i t i o n s of stocks must therefore be based on i n d i r e c t evidence. The majority of e f f o r t concerning delineation of S. alutus stocks has pertained to the Goose Island Gully and Vancouver Island stocks. Stock i d e n t i f i c a t i o n studies have examined genetic v a r i a t i o n and size-age composition differences. There has been only l i m i t e d previous study of whether the Moresby Gully, Rennell Sound, or Langara Spit stocks are unique. (i) Genetic v a r i a t i o n An early study (Tsuyuki et a l . 19 68) found no evidence for genetic i s o l a t i o n of S. alutus stocks i n B.C. waters, based on haemoglobin c h a r a c t e r i s t i c s . Johnson et a l . (1970a,b, 1971, 1972, 1973) found a low degree of protein polymorphism (8% of 25 l o c i ) i n S. alutus o f f the coast of Washington and noted that deeper water specimens were more heterozygous than those i n the shallows. In addition, there were s i g n i f i c a n t non-random associations of genotypes between the sexes, but the authors could not rule out sampling variance as a cause for t h e i r findings. More recently, Wishard et a l . (1980) and Seeb (1986) examined a l l e l i c frequencies of protein isozymes and found s i g n i -f i c a n t differences between the Gulf of Alaska and the Washington-Oregon region. However, the li m i t e d sampling of the B.C. coast makes i t d i f f i c u l t to determine i f the observed differences represent the extremes of a continuum or true segregation. I t was only i n the Gulf of Alaska that any g e n e t i c a l l y d i s t i n c t , contiguous populations were found; a separate group was i d e n t i -f i e d i n Prince William Sound. Seeb's evidence supports a hypothesis of two broad categories of S. alutus, a Gulf type and a B.C.-Oregon type. ( i i ) Size-age composition Evidence for stock segregation of S. alutus i n B.C. waters also comes from age composition differences among areas, associ-ated with d i f f e r e n t e x p l o i t a t i o n h i s t o r i e s (Leaman 1985, Figs 3-7). Because of the e x p l o i t a t i o n context, these groups would constitute d i f f e r e n t stocks (MacLean and Evans 1981). This i s most apparent for the stocks around Queen Charlotte Sound (Figs. 4 and 5). Clearly, these groups have d i f f e r e n t age compositions which have also persisted over time (Archibald et a l . 1981). The differences i n e x p l o i t a t i o n documented i n Table 3 would give r i s e to exactly these types of differences. That the differences p e r s i s t over time implies l i t t l e mixing of f i s h from separate stocks, once they have recruited to the fishery. I t i s possible that mixing occurs p r i o r to recruitment, but i f so, i t has not been of magnitude s u f f i c i e n t to counter the major e f f e c t s of e x p l o i t a t i o n . 40 i Vancouver Is. S. alutus - 1982 Female Male 03 o CM C\2 O CO Age Figure 3. Age composition of S. alutus o f f southwest Vancouver Island i n 1982. Goose Is. Gully S. alutus - 1982 Female Male Age Figure 4. Age composition of S. alutus i n Goose Island Gully i n 1982. Figure 5. Age composition of S. alutus i n Moresby Gully i n 1982. Figure 7. Age composition of S. alutus i n the Langara Spit in 1982. Westrh.eim (1973) suggested d i f f e r e n t deep and shallow water stocks of S. alutus o f f southwest Vancouver Island, and t h e i r existence elsewhere. This hypothesis was based on the observed differences i n siz e at age between shallow (larger) and deep (smaller) water f i s h . However, available evidence suggests that the deep and shallow water f i s h were from the same stock. Younger, faster-growing f i s h are more abundant i n shallower water and must pass through a f i s h i n g zone to occupy a deeper water habitat. In the middle depth range an intermediate s i z e at age was observed, rather than a mixture of large and small sizes as the hypothesis would predict. Recent studies (Beamish 1979, Archibald et a l . 1981) also indicate that some of the ages assigned to f i s h i n Westrheim's study would have been inaccurate. In addition, h i s comparisons were based on f i s h from d i f f e r e n t cohorts and I w i l l show that there can be strong v a r i a t i o n i n si z e at age among d i f f e r e n t cohorts at the same depth. An a l t e r n a t i v e hypothesis i s that faster-growing f i s h of a cohort (hence the largest at age) r e c r u i t to the fi s h e r y f i r s t , so that those members of a cohort eventually making i t from shallow water, through the fishery, to deeper water would be smaller at age than t h e i r shallow-water companions. Johnson et a l . (1971) found f i s h i n deeper water to have higher l e v e l s of heterozygosity. Since deep water f i s h are also older, some support for a longevity-heterozygosity r e l a t i o n s h i p i s also evident. 54 3. Parasites as i d e n t i f i e r s of S. alutus stocks 2 The only previous study of S. alutus parasites i s that of Sekerak (1975), who examined the suite of parasites (internal and external) of several Sebastes spp. o f f B.C. While he noted some geographic differences, he did not census a l l l i f e stages of the parasites found on each f i s h . He showed s i g n i f i c a n t differences i n the i n t e n s i t y of i n f e c t i o n between samples from the Queen Charlotte Sound and Vancouver Island areas as well as a s i g n i -f i c a n t r e l a t i o n s h i p between i n t e n s i t y of i n f e c t i o n of B r a c h i e l l a  robusta (=Neobrachiella robusta (Wilson)) and length of the host. Neobrachiella robusta (Wilson) i s a lernaepodid copepod common on Sebastes spp. of the northeast P a c i f i c (Kabata 1970, Sekerak 1975). There i s no known intermediate host and the developmental stages have only recently been described (Kabata 1987). The primary s i t e of attachment i s the rakers of the g i l l arches, with most specimens (97+%) being found on the f i r s t arch and generally on the longer rakers (Kabata 1987). In the samples c o l l e c t e d for t h i s study (Table 5) I d i s t i n -guished nonovigerous and ovigerous females but that does not necessarily denote age d i s t i n c t i o n s . Adult females produce several p a i r s of egg sacs during the course of t h e i r l i v e s . Consequently, a nonovigerous female might be an old i n d i v i d u a l ^Portions of t h i s section have been published as Leaman and Kabata (1987). 55 Table 5. L o c a l i t y , depth, and number of Sebastes alutus sampled for parasites, by area. Sample Depth No. of no. L o c a l i t y (m) f i s h 1 Southwest Vancouver Is. 201-225 46 2 Southwest Vancouver Is. 214-280 48 3 Triangle Is. 258-282 48 4 Southeast edge, Goose Is. Bank 208-221 48 5 Moresby Gully 283-302 48 6 Moresby Gully 336-357 48 7 Rennell Sound 260-272 48 8 Rennell Sound 271-313 48 9 Langara Spit (outside) 245-293 48 10 Langara Spit (outside) 245-293 48 past i t s egg-laying stage, or an egg-laying female during a pause between episodes of ov i p o s i t i o n . Hence, an ovigerous female could be a young specimen carrying i t s f i r s t batch of eggs and possibly be younger than some nonovigerous females. These two groups are therefore considered together as adults. The samples of N. robusta c o l l e c t e d from the Vancouver Island S. alutus area consisted of 528 females, with r e l a t i v e l y few males (sex r a t i o s of 1:18.5 and 1:11.2) and a high proportion of adults (Table 6). No early developmental stages were present and immature copepods were <2 0% of the t o t a l i n both samples. Surface and bottom temperatures nearest to the source of these two samples were 11.5°C and 7.2°C, respectively. The copepod samples taken from Goose Island Gully also indicated a high i n f e c t i o n rate, with a t o t a l of 454 females and 56 Table 6. Population structure of female Neobrachiella robusta taken from the g i l l s of Sebastes alutus, by area. % of population Stock/ PREV INT PARTOT JUV NON OV Sex Area (%) (range) r a t i o Vancouver Island 1 95.6 5. 9 (1-15) 259 14.7 17.3 68. 0 1:18.5 2 91.7 6. 1 (1-15) 269 17.5 23.4 59. 1 1:11.2 Goose Island Gully 3 93.7 4. 4 (1-17) 199 84.4 5.0 10. 6 1:18.1 4 95.8 5. 6 (1-15) 255 55.6 15.4 29. 0 1:5.2 Moresby Gullv 5 47.9 2. 2 (1-6) 50 68.0 14.0 18. 0 1:6.2 6 45.8 1. 4 (1-3) 30 82.7 — 17. 3 1:7.5 Rennell Sound 7 29.2 1. 3 (1-2) 18 44.4 22.2 33 . 4 no males 8 29.2 1. 4 (1-3) 20 45.0 40. 0 15. 0 1:20.0 Lancrara Spit 9 27.1 1. 2 (1-2) 15 60. 0 6.7 33. 3 1:7.5 10 33.3 1. 4 (1-4) 22 63 . 6 22.7 13. 7 1:5.5 PREV = prevalence of infected f i s h INT = mean i n t e n s i t y of i n f e c t i o n per f i s h PARTOT = t o t a l number of parasites i n sample JUV = mean number of juvenile stages per f i s h NON = mean number of non-ovigerous adults per f i s h OV = mean number of ovigerous adults per f i s h s i m i l a r sex r a t i o s to the Vancouver Island samples (Table 6) . However, the demographic composition of the samples was d i f -ferent. Immature specimens comprised 84.4 and 55.6% of the two samples, respectively, and early stages were abundant (75.9 and 20.1%). Surface and bottom temperatures from the XBT casts nearest the sampling stations were 9.3°C and 6.2°C, respectively. 57 Moresby Gully, although nearer to Goose Island Gully than the l a t t e r i s to Vancouver Island, had a copepod population greatly d i f f e r e n t from that of the Goose Island Gully (Table 6). I t was much less abundant, consisting of only 80 females. Males were r e l a t i v e l y well represented (sex r a t i o s 1:6.2 and 1:7.5). The population was predominantly immature, adult females c o n s t i -t u t i n g only 32.0 and 17.3% of the two samples, respectively. Surface and bottom temperatures from XBT stations near the sampling l o c a l i t i e s were very s i m i l a r (9.4°C and 6.4°C) to those for the Goose Island Gully stations. Fish from the fourth presumed stock (Rennell Sound) ca r r i e d even fewer copepods than those from Moresby Gully (Table 6). The t o t a l number of females from both samples was 38. Males were scarce; absent from one sample and represented by a single specimen i n the other. Immature specimens comprised le s s than h a l f of the indivi d u a l s i n both samples. Surface (9.0°C) and bottom (5.6°C) temperatures were lower than those at the more southerly stations. The copepods from the Langara Spit samples consisted of only 37 females (Table 6). Males were better represented than i n the samples from Rennell Sound (sex r a t i o s of the samples were 1:7.5 and 1:5.5, re s p e c t i v e l y ) . The samples also had a higher number of immature copepods than Rennell Sound, with immature specimens 58 comprising 60.0 and 63.6%, respectively. The values f o r surface and bottom temperatures were the same as f o r the Rennell Sound stations. Samples were segregated into l a r g e - f i s h (Moresby Gully, Rennell Sound) and s m a l l - f i s h (Vancouver Island, Goose Island Gully, Langara Spit) groupings based on the consistent d i f f e -rences i n the samples from these geographic l o c a l i t i e s over time (Leaman 1985, 1987b) . The hypotheses of a unit stock f o r the Vancouver Island, Goose Island Gully, and Langara Spit samples, and the Moresby Gully and Rennell Sound samples were tested on the basis of mean parasite c h a r a c t e r i s t i c s per f i s h (Table 7) . Among the samples of smaller f i s h (Vancouver Island, Goose Island Gully, Langara S p i t ) , the Vancouver Island and Goose Island Gully samples were indistinguishable on the basis of i n t e n s i t y of i n f e c t i o n (INT, p=0.087) but could be e a s i l y separated by the composition of t h e i r parasite populations (JUV, NON, OV, AD, p<0.001). The Langara Spit stock could be distinguished from ei t h e r Vancouver Island or Goose Island Gully i n a l l parasite c h a r a c t e r i s t i c s (p<0.001). D i f f e r e n t i a t i o n of samples of larger f i s h (Moresby Gully, Rennell Sound) was also possible on the basis of i n t e n s i t y of i n f e c t i o n (INT, p<0.01) and numbers of juvenile parasites present (JUV, p<0.001), but the differences between areas were not as comprehensive (NON, OV, and AD were non-significant) as those f o r 59 the other sample grouping. Only samples from the Vancouver Island, Goose Island Gully and Moresby Gully stocks had s u f f i c i e n t bathymetric separation (Table 8) for examination of depth e f f e c t s on parasite charac-t e r i s t i c s . No consistent or comprehensive differences between shallow and deep samples were apparent. Intensity of i n f e c t i o n and mean number of juvenile parasites were not s i g n i f i c a n t l y d i f f e r e n t between depths within any stock. S i g n i f i c a n t d i f f e -rences between depths were noted only for mean numbers of reproductive females (OV) i n the Goose Island Gully stock. Parasite load (INT) was also examined as a function of host length (Sekerak (1975). A regression of INT against LEN for a l l stocks combined was non-significant (p>.06) and highly variable (r 2=.005). The highest parasite load was found i n the 35-40 cm range, with lower values at the extremes of the length d i s t r i b u -t i o n . Only one stock (Vancouver Island) showed a s i g n i f i c a n t regression of these two variables however the re l a t i o n s h i p was also highly variable (r^=.074); regressions for a l l other stocks were non-significant. Observations at lower length i n t e r v a l s have high leverage i n these regressions. 60 Table 7. S t a t i s t i c a l t e sts of d i f f e r e n t i a t i o n based on mean values of p a r a s i t e (Neobrachiella robusta) c h a r a c t e r i s t i c s between Sebastes alutus stock pairs indistinguishable by host c h a r a c t e r i s t i c s . VI=Vancouver Island, GIG=Goose Island Gully, MOR=Moresby Gully, RSD=Rennell Sound, LA=Langara Spi t . Mean / Standard error H 0: - X 2 = 0 Variable VI GIG t Prob. > t INT 5.63/0.397 4.74/0.330 1.7197 0.0872 JUV 0.92/0.119 3.21/0.252 8.2285 0.0001 NON 1.16/0.145 0.53/0.100 3.5706 0.0005 OV 3.55/0.322 1.00/0.195 6.7770 0.0001 AD 4.71/0.353 1.53/0.241 7.4386 0.0001 MOR RSD INT 0.84/0.126 0.39/0.086 3.1974 0.0017 JUV 0.61/0.098 0.17/0.051 4.0771 0.0001 NON 0.07/0.031 0.13/0.037 1.0857 0.2791 OV 0.16/0.045 0.09/0.030 1.1527 0.2507 AD 0.23/0.058 0.22/0.047 0.1388 0.8898 VI LA INT 5.63/0.397 0.40/0.072 12.9676 0.0001 JUV 0.91/0.119 0.25/0.053 5.1059 0.0001 NON 1.16/0.144 0.06/0.029 7.4432 0.0001 OV 3.55/0.322 0.08/0.028 10.7295 0.0001 AD 4.71/0.353 0.15/0.039 12.8476 0.0001 GIG LA INT 4.74/0.330 0.40/0.072 12.8542 0.0001 JUV 3.21/0.252 0.25/0.053 11.4780 0.0001 NON 0.53/0.100 0.06/0.029 4.4870 0.0001 OV 1.00/0.195 0.08/0.028 4 . 6440 0.0001 AD 1.53/0.241 0.15/0.047 5.6733 0.0001 INT = mean number of parasites per f i s h JUV = mean number of juvenile stages per f i s h NON = mean number of non-ovigerous adults per f i s h OV = mean number of ovigerous adults per f i s h AD = mean number of adults per f i s h 61 Table 8. S t a t i s t i c a l tests of d i f f e r e n t i a t i o n based on mean valu e s of p a r a s i t e (Neobrachiella robusta) c h a r a c t e r i s t i c s between Sebastes alutus samples from d i f f e r e n t depths. Mean / Standard error H 0: X i - X 2 = 0 Variable Deep Shallow t Prob. > t Vancouver Island INT 5.64/0.549 5.61/0.580 0. 0465 0.9630 JUV 1.00/0.188 0.83/0.143 0. 7346 0.4645 NON 1.33/0.205 0.98/0.203 1. 2324 0.2210 OV 3.31/0.427 4.78/0.486 0. 7597 0.4454 AD 4.65/0.471 4.78/0.534 0. 1521 0.8478 Goose Island Gully INT 5.31/0.451 4.17/0.473 1. 7358 0.0827 JUV 2.92/0.330 3.50/0.380 1. 1588 0.2495 NON 0.83/0.156 0.23/0.112 3. 1471 0.0022 OV 1.56/0.292 0.44/0.236 2 . 9982 0.0035 AD 2.39/0.340 0.67/0.298 3 . 8371 0.0002 Moresby Gully INT 1.04/0.217 0.65/0.125 1. 5839 0.1174 JUV 0.71/0.171 0.52/0.094 0. 9609 0.3398 NON 0.15/0.412 0 / 0 2 . 4522 0.0180 OV 0.19/0.077 0.13/0.048 0. 6890 0.4928 AD 0.33/0.105 0.13/0.048 1. 8091 0.0750 INT = mean number of parasites per f i s h JUV = mean number of juvenile stages per f i s h NON = mean number of non-ovigerous adults per f i s h OV = mean number of ovigerous adults per f i s h AD = mean number of adults per f i s h 62 While aggregate parasite differences delineated stocks well, c h a r a c t e r i s t i c s for i n d i v i d u a l f i s h were le s s u s e f u l . Discrimi-nant analysis indicated a range of 34.4-76.0% correct c l a s s i f i c a -t i o n of i n d i v i d u a l s to stock on the basis of numbers of juvenile, nonovigerous, ovigerous parasites (by f i s h ) , and host length, for a l l samples combined (Table 9). Addition of t o t a l parasites per f i s h to the discriminant function did not improve i t s power. One-half of the samples from the l a r g e - f i s h and s m a l l - f i s h e x p l o i t a t i o n groups noted e a r l i e r were used as a 'learning' sample, to generate a discriminant function with which to c l a s s i f y the other h a l f of the samples. C l a s s i f i c a t i o n power improved s u b s t a n t i a l l y . Correct c l a s s i f i c a t i o n of i n d i v i d u a l s from s m a l l - f i s h t e s t samples ranged from 73.4-87.5% and for l a r g e - f i s h samples from 42.7-82.3%. The Moresby Gully stock had the poorest correct c l a s s i f i c a t i o n and the Langara Spi t stock the best. I t appears that although the parasite d i s t r i b u t i o n among stocks i s not of the present-vs.-absent type, i t may s t i l l have considerable value as a stock discriminator for i n d i v i d u a l f i s h , i f a u x i l i a r y information on pote n t i a l stock i d e n t i t y i s also a v a i l a b l e . 4. Discussion On the basis of these coincident samples, i t i s c l e a r that separate groups of P a c i f i c ocean perch can be delineated by populations of t h e i r g i l l parasite, N. robusta, where they are 63 Table 9. Percent of Sebastes alutus c l a s s i f i e d to stock by discriminant analysis. See text for d e t a i l s of variables used. LA=Langara Spit, RSD=Rennell Sound, MOR=Moresby Gully, GIG=Goose Island Gully, VI=Vancouver Island. A l l stocks Stock VI GIG MOR RSD LA VI 69.2 16.0 6.4 6.4 2.1 GIG 11.5 59.4 19.8 4.2 5.2 MOR 2.1 4.2 34.4 24. 0 35.4 RSD 2.1 2.1 10.4 38.5 46.9 LA 0.0 1.0 11.5 11.5 76.0 Small-- f i s h stocks Stock VI GIG LA VI 73.4 18.1 8.5 GIG 15. 6 71.0 10.4 LA 4.1 8.3 87.5 Large-fish stocks Stock MOR RSD MOR 42.7 57. 3 RSD 17.7 82.3 otherwise indistinguishable. Although the copepod i t s e l f shows no s i g n i f i c a n t morphological differences among host populations (cf. Kabata 1970), i t d i f f e r s i n i t s population structure, as well as i n prevalence and i n t e n s i t y of i n f e c t i o n . The s i g n i f i -cant differences i n aggregate parasite c h a r a c t e r i s t i c s for f i s h of the same s i z e provide powerful discrimination i n l i g h t of a po t e n t i a l c o r r e l a t i o n between host length and i n t e n s i t y of i n f e c -t i o n . The value of t h i s parasite as a b i o l o g i c a l tag i s enhanced 64 by a u x i l i a r y information on potential stock a f f i n i t i e s e s t a b l i s -hed by the s i z e frequencies of the host. The l a t t e r have resulted from d i f f e r e n t i a l e x p l o i t a t i o n h i s t o r i e s among stocks (Fig. 2). Sekerak (1975) also examined parasites as stock i d e n t i f i e r s f o r S. alutus. While he found the same decrease i n the i n t e n s i t y of i n f e c t i o n of N. robusta with increasing l a t i t u d e noted here, he did not examine the demographic c h a r a c t e r i s t i c s of the parasites on each host. In contrast to t h i s study, Sekerak noted a s i g n i f i c a n t difference i n the i n t e n s i t y of i n f e c t i o n between samples from stocks equivalent to my Vancouver Island and Goose Island Gully designations. However, the lower value he found for the Goose Island Gully stock may be because he d i d not sample immature stages of the parasite, which account f o r the majority of the i n f e c t i o n i n my Goose Island Gully samples. I f immature stages are ignored, my conclusions would mirror those of Sekerak. I believe my sampling describes the parasite populations of these two stocks more accurately. Sekerak also noted a p o s i t i v e r e l a t i o n s h i p between s i z e of host and i n t e n s i t y of i n f e c t i o n , however I could not confirm that fi n d i n g as general. The r e l a t i o n s h i p for a l l stocks was highly v a r i a b l e and non-significant. There may be several explanations fo r t h i s difference. Sekerak's conclusion was based on a compa-r i s o n of i n t e n s i t y of i n f e c t i o n vs. mean s i z e of host i n h i s samples, rather than on a comparison by s i z e i n t e r v a l within or among stocks. As such, h i s comparisons may have been i n s e n s i t i v e to the d i s t r i b u t i o n of sizes contributing to the means. The areas he examined had already been subjected to high f i s h i n g pressure and the s i z e range of the hosts was not as extensive as that i n t h i s study. Therefore, he did not have opportunity to examine t h i s r e l a t i o n s h i p within the large numbers of 40+ cm f i s h reported here. Lastly, there i s l i t t l e doubt that the smallest f i s h (<30 cm) i n the samples harbour fewer parasites and i t i s these f i s h which contribute the majority of the s i g n i f i c a n c e i n Sekerak 1s finding. Because these f i s h are only p a r t i a l l y r e c r u i -ted to the f i s h i n g grounds (Archibald et a l . 1983) they are an uncertain r e f l e c t i o n of the i n t e n s i t y of i n f e c t i o n f o r t h i s s i z e group, i n the population. I f these f i s h are representative then the r e l a t i o n s h i p of i n t e n s i t y of i n f e c t i o n and host s i z e may not be continuous, rather there may be a threshold s i z e beyond which i n t e n s i t y of i n f e c t i o n i s uncorrelated with host s i z e . The influence of f i s h age on i n t e n s i t y of i n f e c t i o n i s obviously of minor importance i n comparison with those features giv i n g r i s e to the strong l a t i t u d i n a l differences i n i n f e c t i o n f o r S. alutus stocks. Whether these features govern the parasite d i r e c t l y , or i n d i r e c t l y through the biology/physiology of the host, i s uncertain. P a r a s i t i c copepods of the family Lernaepodidae have seldom been used as b i o l o g i c a l tags. The main reason f o r t h i s lack of i n t e r e s t i s the fac t that they are, as a rule, attached by t h e i r bullae to the s u p e r f i c i a l t i s s u e layers of t h e i r hosts and are e a s i l y l o s t when the host i s being handled. This danger i s p a r t i c u l a r l y great when the parasites are attached to the exposed outer surfaces of the f i s h . Neobrachiella robusta does not su f f e r from t h i s handicap, because i t i s attached to bony structures ( g i l l rakers) confined within the protected space of the opercular c a v i t i e s . An unspecified Clavellodes was used by Vooren and Tracy (1976) as a possible tag for three putative stocks of Cheilodac- t y l u s macropterus (Bloch and Schneider) and was found on one stock only. Pal (in press) suggested the use of C l a v e l l i s a  i l i s h a e P i l l a i (=C. h i l s a e P i l l a i ) as an in d i c a t o r of upstream migrations of i t s clupeoid host, Hilsae i l i s h a (Hamilton). The suggestion, however, was not put to use. F i n a l l y , Siegel (1980) attempted to use Eubrachiella antarctica (Quidor), to d i s t i n g u i s h between stocks of three species of channichthyid f i s h e s . The si g n i f i c a n c e of h i s findings was questioned by MacKenzie (1983) because, among other things, the copepod was attached to the outer surfaces, including the d e l i c a t e t i s s u e of the f i n s . The re s u l t s presented here are the f i r s t attempt at i d e n t i f y i n g stocks of a commercial f i s h species, using a lernaepodid copepod as a tag. 67 In the A t l a n t i c Ocean, there has been some a c t i v i t y concer-ning two non-lernaepodid p a r a s i t i c copepods, Sphyrion lumpi Kr^yer and Chondracanthus nodosus Muller. MacKenzie (1983) reviews several studies supporting stock segregation of Sebastes  marinus i n the northwest A t l a n t i c and the Gulf of Maine based on the i n t e n s i t y of i n f e c t i o n with Sphyrion lumpi. S. lumpi i s attached to the exposed body surfaces and i s therefore less desirable as a stock i d e n t i f i c a t i o n tag, although the cephalotho-rax may remain i n the host tissues a f t e r the adult parasite i s dislodged. In contrast, C. nodosus i s anchored i n the opercular c a v i t i e s (as N. robusta) of several A t l a n t i c species, including Sebastes marinus and S. mentella. Studies i n d i c a t i n g p o t e n t i a l u t i l i t y of t h i s copepod for stock segregation of both Sebastes spp. are c i t e d by MacKenzie. While I am encouraged by the pot e n t i a l to d i s t i n g u i s h stock units by the c h a r a c t e r i s t i c s of t h e i r i n f e c t i o n with t h i s para-s i t e , several cautionary comments must also be made. Some of the discriminating power of the parasite i s based upon the composi-t i o n of i t s population, which may i t s e l f be seasonally dynamic. Therefore, i t i s important to determine whether these differences are consistent throughout the year, although Sekerak (1975) noted that seasonal e f f e c t s were not s i g n i f i c a n t for parasites of S. alutus. In addition, there are l a t i t u d i n a l c l i n e s i n both the i n t e n s i t y of i n f e c t i o n and the proportions of juvenile and adult parasites; gradients which may be a temporal a r t i f a c t of the 68 l a t i t u d i n a l progression of the seasons. However, t h i s argument can be countered by consideration of the gross differences i n the parasite populations between the Goose Island Gully and Moresby Gully stocks, i n the absence of s i g n i f i c a n t temperature d i f f e -rences between t h e i r environments. Observed differences may also be smaller for l e s s widely spaced samples but the large d i f f e -rences between the two most proximal (<100 km) stocks (Goose Island Gully and Moresby Gully) provide encouraging support for N. robusta as a useful tag. 5. Conclusions Geographic groups of s i m i l a r adult S. alutus can be d i s t i n -guished on the basis of t h e i r g i l l parasite, N. robusta. This implies i s o l a t i o n of stocks at the adult stage but does not carry implications of genetic segregation. Separate genetic studies have shown that S. alutus stocks from at l e a s t southeastern Alaska to Oregon are g e n e t i c a l l y s i m i l a r (Seeb 1986). Genetic exchange between these groups must be maintained p r i o r to recruitment to adult stocks. The release of larvae i n offshore locations (Leaman 1985) provides s u f f i c i e n t opportunity f o r t h i s exchange. The r e l a t i o n s h i p between these geographically segrega-ted adult stocks and recruitment to them, remains to be deter-mined. However, the examination of f i s h e r y e f f e c t s on the b i o l o g i c a l c h a r a c t e r i s t i c s of these stocks can be undertaken with 69 a confidence that c h a r a c t e r i s t i c s of adjacent stocks w i l l not contaminate these measurements or determinations. The following chapter i s devoted to t h i s examination. 70 IV. REPRODUCTIVE BIOLOGY AND RELATION TO LIFE HISTORY THEORY 1. Growth The major differences among stocks are i n t h e i r age composi-tions (Figs. 3 to 7) and the resultant s i z e composition. Sexual dimorphism r e s u l t s i n greater female length and weight over that of males for the same ages. The form of growth also varies with sex and stock, p a r t i c u l a r l y f or females. Comparison of residuals for observed vs. predicted values of length (Table 10) indicates male growth as the t r a d i t i o n a l negative exponential form (e.g. von Bertalanffy, VB) whereas observed female s i z e shows a systematic negative trend i n the residuals from a VB f i t for older f i s h (e.g. Figs. 8-9). This d i s t r i b u t i o n suggested that the addition of a quadratic term would reduce the res i d u a l error, since the maxima occurred i n the middle (25-35 y) of the age compositions with shorter lengths at older ages. The standard negative exponential f i t to these data both underestimates the lengths of female f i s h i n the 25-35 y range and overestimates those of the older age groups. Differences i n res i d u a l sums of squares among stocks for quadratic vs. VB functions were larger for length estimation than for round weight. While description of growth as a quadratic function led to a lower residual error, the f i t to the data was less s a t i s f a c t o r y i n terms of residual d i s t r i b u t i o n at the extremes of the age range, due to the symmetry of the function. Subsequent to t h i s examination, I have worked with a colleague (Dr. T. J . Mulligan) to develop growth models which 71 Table 10. Residual sum of squares and mean square error for quad-r a t i c (Q) and asymptotic (A) estimation of length and round weight (RDWT) at age for f i v e stocks of S. alutus. VI=Vancouver Island, GIG=Goose Island Gully, MOR=Moresby Gully, RSD=Rennell Sound, LA=Langara Spit. Residual sum of squares Mean square error Best Stock Variable Quadratic Asymptotic Quadratic Asymptotic f i t Females Length LA 317.5 487.9 4.071 4.396 Q RSD 550.5 588.1 4.626 4.324 Q MOR 267.8 1169.3 3.938 4.955 Q GIG 382 . 6 887.4 3.788 3.892 Q VI 421.1 1119.5 3 .074 4.889 Q Males Length LA •• 354 . 2 318.0 3 . 220 2.891 A RSD 704.9 236.2 7.050 2 .362 A MOR 1615.9 352.2 6.191 2 .994 A GIG 1237.6 284.4 5.838 2 . 609 A VI 445.2 419.9 4.281 4.241 A Females RDWT (X10 6) (xlO 6) (xlO 6) (xlO 6) LA 1.9415 2.6471 2.4891 2.3848 Q RSD 8.0681 9.1716 6.7800 6.7438 Q MOR 2.7189 6.5234 3.9983 3.9062 Q GIG 2.3717 1.9277 2.3482 2.0291 A VI 2.3645 4.6251 1.7259 2.4089 Q Males RDWT (X10 6) (XlO 6) (XlO 6) (xlO 6) LA 0.0917 1.3442 0.8337 1.2220 Q RSD 1.6238 1.0653 1.6238 1.0653 A MOR 2.0799 1.2856 1.7333 1.0713 A GIG 1.0679 0.9741 0.9797 0.8937 A VI 0.7916 0.7853 0.7995 0.7933 A 50 i Moresby Gully females 45 40 -35 0 20 40 Age 60 80 to Figure 8. Length at age d i s t r i b u t i o n for female S. alutus in Moresby Gully, together with a f i t t e d von Bertalanffy growth curve. .15 i .1 -H .05 0 -.05 -0 Moresby Gully females •8* #0* *3> 20 40 Age i 60 * * $, * # 80 Figure 9. Distribution of residual error between estimated and observed length at age for Moresby Gully S. alutus females. u> examine both an inverse r e l a t i o n of growth and mortality rates, and time-dependence of growth rate (Mulligan and Leaman, i n prep.)« We believe that an asymptotic function i s appropriate to describe the growth of i n d i v i d u a l f i s h (Brett and Groves 1979) but we also believe that smaller sizes of older f i s h are repre-sentative observations. We have examined t h i s problem with simulated data sets and with a larger data set (N=753) from the Moresby Gully stock. We constructed two models based on the competing hypotheses of (i) growth rate-dependent mortality, and ( i i ) time-dependent growth with growth rate-independent morta-l i t y , designated G(n) and L ( t ) , respectively. For the G(n) model, a l t e r n a t i v e growth and mortality rates are correlated scalars of the mean rates. Both models c o r r e c t l y i d e n t i f i e d the appropriate growth process when tested against data sets con-structed with the a l t e r n a t i v e hypotheses. Testing both models against the Moresby Gully data set suggests that the G(n) model provided a s l i g h t l y better description of the s i z e frequency d i s t r i b u t i o n . This means that an inverse r e l a t i o n of growth and mortality rates, rather than a long-term increase i n growth rate, i s the more appropriate explanation for the smaller, older f i s h i n t h i s stock. Growth differences among stocks were examined by t e s t i n g for s t a t i s t i c a l separation of the log-transformed regressions of length and weight on age. Regressions of both length and weight were s i g n i f i c a n t l y d i f f e r e n t among stocks (p<.01) and the maximum differences involved the two most exploited stocks, Langara Spit and Vancouver Island (Table 11). The r e l a t i o n s h i p of length and weight i s approximately cubic ( i . e . , b«3 i n the r e l a t i o n s h i p , Weight = a (Length) , however there are s i g n i f i c a n t differences among stocks (F=9.051, p<.001). A l l exploited stocks have lower values of b and higher values of a than unexploited stocks (Table 11) . The net e f f e c t of these differences i s that f i s h from exploited stocks are heavier at smaller lengths than those from unexploited stocks, but t h e i r ultimate weights at the asymptotic lengths are lower. The weights for the Goose Island Gully f i s h from which fecundity samples were obtained had to be estimated because of a balance malfunction. These weights were obtained using a length-weight regression from a sample of 210 f i s h , taken one day a f t e r the c o l l e c t i o n of the fecundity samples. This r e l a t i o n s h i p of weight and length was used to calculate the weights of the Goose Island Gully f i s h i n Table 11. The two stocks e x h i b i t i n g the greatest departures from the general r e l a t i o n s h i p s of body parameters were the Vancouver Island and Langara Spit stocks, yet t h e i r c o r r e l a t i o n s with e x p l o i t a t i o n differences were not unequivocal. In part, t h i s may be due to the general changes i n stocks since the early 1960's. Figures 10-13 present sizes at age for the Vancouver Island, Goose Island Gully, Moresby Gully and Langara Sp i t stocks, ob-76 Table 11. Regression s t a t i s t i c s f or weight (g) relat i o n s h i p s of S. alutus females, by stock. RDWT=round weight, OVWT=ovary weight, SOMWT=somatic weight, LEN=length(cm). Y = a * b (X) Stock Variables/ Langara Rennell Moresby Goose Is. Vancouver s t a t i s t i c s Spit Sound Gully Gully Is. Log(RDWT)-Log(LEN) a -1.9899 b 3.0963 RMS 0.00095 ad j . N 0.911 81 -3.0865 3.7720 0.00152 0.841 122 -2.2039 3.2432 0.00093 0.928 71 -1.9053 3.0460 c c 104 -1.9022 3.0493 0.00075 0.938 140 Log(SOMWT)-Log(LEN) a -1.7671 b 2.9455 RMS 0.00091 ad j . N 0.907 81 -3.0289 3.7215 0.00142 0.847 122 •2.0850 3.1588 0.00093 0.924 71 •1.7721 2.9536 c c 104 -1.8632 3.0139 0.00072 0.939 140 Log(OVWT)-Log(LEN) a b RMS ad j . N -8.8034 6.4779 0.01004 0.811 81 •6.4498 5.0520 0.01491 0.49 122 -7.3195 5.5288 0.01637 0.678 71 -7.9438 5.8780 c c 104 -4.9428 4.0650 0.11500 0.636 140 SOMWT-RDWT a 47.130 15.416 22.664 29.964 9.4687 b 0.9063 0.9310 0.9369 0.9341 0.9491 RMS 83.56 208.02 291.75 C 120.57 adj. r 2 0.998 0.997 0.997 c 0.998 N 81 122 71 10 4 140 OVWT-RDWT a -47.130 -15.417 -22.663 -30.965 -9.4689 b 0.0937 0.0690 0.0631 0.0659 0.0509 RMS 83.56 208.00 291.75 c 120.53 adj. r 2 0.871 0.665 0.586 c 0.621 N 81 122 71 104 140 77 OVWT-SOMWT a -50.287 -12.816 -19.882 -33.077 -7.8079 b 0.10165 0.07111 0.06397 0.0706 0.5181 RMS 101.57 239.29 331.32 C 133.57 adj. r 2 0.843 0.611 0.530 c 0.580 N 81 122 71 104 140 c - weights calculated for Goose Island Gully tained from research sampling. No h i s t o r i c a l data are available fo r the Rennell Sound stock and those f o r the Langara Spit and Moresby Gully stocks are limited, however some trends are suggested. Sizes at age are s l i g h t l y larger i n more recent years regardless of ex p l o i t a t i o n history, and differences are largest f o r the most heavily exploited stocks. The conjunction of these observations suggests not only that there i s a density-mediated response to expl o i t a t i o n , but also that there may be a broader-scale e f f e c t . The ex p l o i t a t i o n h i s t o r y of P a c i f i c ocean perch throughout the northeast P a c i f i c may be the p r i n c i p a l cause of such changes (Fig. 2). Clearly, the magnitude of catch reduction throughout the Alaska-Oregon region and the amount of e f f o r t expended i n these unrestricted f i s h e r i e s indicates major deple-tions of standing stocks. A general elevation of growth i n survivors and subsequent r e c r u i t s i s consistent with such a catch hi s t o r y , and the evidence for bathymetric migration with ontogeny by the species (Gunderson 1977, Leaman 1985). While sizes at age have increased s l i g h t l y f or exploited 78 Southwest Vancouver Island females Figure 10. Mean length at age d i s t r i b u t i o n for Vancouver Island S. alutus females, 1965-1982. 79 Goose Island Gully females Figure 11. Mean length at age d i s t r i b u t i o n for Goose Island Gully S. alutus females, 1966-1982. Figure 12. Mean length at age d i s t r i b u t i o n for Moresby Gully alutus females 1974-1982. Figure 13. Mean length at age d i s t r i b u t i o n for Langara Spit alutus females, 1966-1982. 82 stocks, the comparison with previous years (Figs. 10-13) also indicates c o h o r t - s p e c i f i c growth features. Mean s i z e at age may vary by 1-3 cm among cohorts and measurements among years may not be consistent. In addition, the sizes at age for f i s h <12y should be viewed with caution because these f i s h are only p a r t i a l l y r e c r u i t e d to the f i s h i n g grounds (Archibald et a l . 1983) . Fish sampled at ages younger than 12y are l i k e l y to be larger than the mean siz e at age, for that cohort. The conver-gence of estimated sizes at older ages among samples suggests that p a r t i a l v u l n e r a b i l i t y (or a v a i l a b i l i t y ) changes, as well as growth changes, may be operating. A further caution i s that h i s t o r i c a l age data were estimated from surface readings of o t o l i t h s , which have been demonstrated to underestimate ages beginning near age 15 (Beamish 1979, Archibald et a l . 1981). Recent work (Stanley 1987) also suggests a very small p o s i t i v e bias ( i . e . , over-ageing) for the break/burn technique at ages <6y. Unfortunately, i t i s not possible to re-estimate previous ages because the o t o l i t h s were not i n d i v i d u a l l y i d e n t i f i e d i n h i s t o r i c a l samples, rather they were grouped by s i z e i n t e r v a l . I t i s likewise impossible to resolve correct ages from surface readings because a discrete surface estimate (e.g 20y) i s associ-ated with a d i s t r i b u t i o n of ages (e.g. 20-40y) from break/burn or section techniques. These observations suggest that there has been a compensatory increase i n growth rate since the mid-1960s, but the magnitude of the change may be only s l i g h t l y greater than the normal inter-cohort v a r i a t i o n . 83 The technique of breaking and burning the o t o l i t h to estimate the age of S. alutus has not been validated, i . e . shown to produce accurate ages, because of the d i f f i c u l t y i n tagging the animals. However, I have validated t h i s technique for a congener, S. flavidus (Ayres), using a tag-recapture study i n c o n j u n c t i o n with i n j e c t i o n of oxy t e t r a c y c l i n e (Leaman and Nagtegaal 1987). While t h i s does not v a l i d a t e the technique for S. alutus. the close taxonomic l i n k s within t h i s genus (Chen 1986) suggest that the technique may also be v a l i d f o r other Sebastes spp. Further support comes from the strong presence of the 1952 cohort i n the age composition of the Moresby Gully and Rennell Sound stocks (at age 30y). This cohort was previously i d e n t i f i e d as extremely strong at young ages (6-10 y) , using surface ageing, i n other stocks from Alaska to Oregon (Westrheim et a l . 1972). The i d e n t i f i c a t i o n of t h i s strong cohort i n the l i g h t l y exploited Rennell Sound and Moresby Gully stocks at age 3 0 y implies v a l i d i t y of the break/burn technique to at l e a s t that age. 2. Reproductive biology (i) Size and age at maturity Table 12 presents sizes and ages at maturity f o r female S. alutus from the f i v e stocks i n November, 1982. Size of f i s h at f i r s t maturity was estimated as the length at which 50% of the 84 Table 12. Estimated s i z e (cm) and age at 50% maturity for female S. alutus c o l l e c t e d i n 1982, by stock, and estimates from previous research (R) and commercial (C) sampling of these stocks. Stock Year N Month L.50 A.50 Vancouver Island 1982 232 Nov 33.7 (R) 7.2 (R) 1970 127 Apr 34.6 (R) 1969 1151 Feb 34.5 (R) 1968 1377 Apr-Jun 34.6 (R) 1967 1029 Feb-Apr 34. 6 (R) Goose Island Gully 1982 231 Nov 33.5 (R) 8.0 (R) 1981 268 Jan-Dec 39.8 (C) 1980 750 Jan-Dec 37.0 (C) 1978 1215 Jan-Dec 38.6 (C) 1977 205 Jan-Dec 40.3 (C) Moresby Gully 1982 1981 1980 1979 239 233 639 633 557 Nov June June Jan-Dec Jan-Dec (R) (R) 33.2 37.8 36.6 (R) 38.0 (C) 38.4 (C) 7.9 (R) Rennell Sound 1982 139 Nov 38.5 (R) 10.2 (R) 1981 247 Jan-Dec 42.7 (C) 1980 222 Jan-Dec 35.6 (C) 1979 1519 Jan-Dec 36.0 (C) 8 . 6 (C) Spit 1982 117 Nov 34.6 (R) 7.9 (R) 1970 178 May 35.6 (R) 85 f i s h sampled were mature ( L > 5 0 ) (Ricker 1975) and was taken from the length-maturity ogives (e.g. Figs. 14 and 15) . The corres-ponding ages at f i r s t maturity were estimated from the asymptotic age-length function. Length at maturity i s generally lower when estimated from research samples than when commercial samples are used. This difference r e f l e c t s the s i z e - s e l e c t i v e grading pra c t i c e s of the commercial fishery. With the exception of the Rennell Sound stock, L > 5 0 varies l i t t l e among stocks (X=33.8 cm, S.E.=0.301). This indicates some developmental constraint on the siz e at sexual maturity for the species, since the faster-growing f i s h i n the exploited stocks did not mature at a smaller s i z e than those i n the unexploited stocks. The value for the Rennell Sound stock i s poorly deter-mined due to the low proportion of immature f i s h ; such f i s h are r a r e l y encountered i n t h i s stock (Leaman and Nagtegaal 1982). Ages at 50% maturity ( A > 5 0 ) corresponding to these L > 5 0 r e f l e c t growth differences among the stocks whether the A # 5 0 are deter-mined d i r e c t l y or estimated from the age-length r e l a t i o n s h i p s . The form of the maturity ogive varies l i t t l e among stocks, i . e . maturation rate with s i z e i s r e l a t i v e l y consistent, although there i s sc a l i n g of the maturity ogives along the abscissa. Maturation appears to occur r a p i d l y with s i z e . The range of siz e over which maturation occurs i s narrow (2-3 cm) and when growth rate i s considered, suggests that the process may be completed within one year. 87 L E N G T H ( c m ) Figure 15. Length-maturity ogive for Moresby Gully S. alutus females, 1982. 88 ( i i ) Fecundity estimation I could not achieve the p r e c i s i o n reported i n the l i t e r a t u r e f o r the volumetric subsampling methods (Table 13) . For a l l methods, the c o e f f i c i e n t of v a r i a t i o n of the subsample mean s t a b i l i z e d with approximately seven subsamples, however there was a d i r e c t r e l a t i o n between subsample volume (mL) and the c o e f f i -c i e n t of v a r i a t i o n C r i t i c i s m of the Stempel pipette (Kandler and Pirwitz 1957) on the basis of the sampling variance alone appears unwarranted, since the c o e f f i c i e n t of v a r i a t i o n of the Stempel subsamples was lower than that of the 2 mL standard subsamples, and approximated that of the 5 mL subsample. There was no cl e a r r e l a t i o n between sampling method and estimated mean oocytes/mL. The lowest estimate (25.1/mL) was from the 5 mL standard sub-sample and the highest (34.3/mL) from the 10 mL standard (Table 14). The Stempel mean (31.4/mL) was higher than the 2 mL standard subsample mean (28.7/mL). Since fecundity i s l i n e a r l y r elated to oocytes/mL, the r e s u l t i n g fecundity estimates have the same d i s t r i b u t i o n . While mean oocyte sizes were only s l i g h t l y d i f f e r e n t between subsample methods (Figure 16), the large sample sizes (1000-2900 oocytes) render even these small differences s i g n i f i c a n t . For equal numbers of subsamples the 10 mL subsampling technique was superior i n terms of the c o e f f i c i e n t of v a r i a t i o n and standard error of the estimated oocytes/mL (Table 14); differences i n the means were s i g n i f i c a n t (F=3.101, p<.05). No 89 Table 13. C o e f f i c i e n t of v a r i a t i o n of eggs/mL for volumetric subsampling of S. alutus ovaries, by sample type. C o e f f i c i e n t of v a r i a t i o n - eggs/mL Subsample 2 mL bulb 5 mL bulb 10 mL bulb 2 mL Stempel 1 2 . 14 . 10 .09 .01 3 .40 .08 .09 .02 4 .38 .13 .09 . 02 5 .33 .14 .08 .23 6 .30 .18 . 09 . 22 7 .30 .22 . 12 .21 8 .30 .23 .12 .25 9 .33 .21 .11 .25 10 .31 .22 .11 .24 e f f e c t of subsampling location within the suspension was apparent (Table 15). However, subsample volume was related to v a r i a b i l i t y i n estimated oocytes/mL. Means were s i g n i f i c a n t l y d i f f e r e n t (F=3.862, p<.02) for the f i v e sets of 2 mL samples but not for the 5 mL samples (F=1.489, p>.20) (Table 15). Both methods of estimating the t o t a l number of oocytes i n suspension performed poorly. The 5 mL grand mean yielded an estimated t o t a l of 13 64 ± 195 while the 2 mL estimates ranged from 150 ± 1707 to 1200 ± 915. While t h i s experiment was of necessarily smaller scale than the regular volumetric sampling (500 mL vs. 3500 mL) the proportion of the t o t a l volume sampled was increased (5% vs. 1%) to o f f s e t p o t e n t i a l e f f e c t s of scale. The p r e c i s i o n of the gravimetric estimator was considerably 90 Table 14. Analysis of variance of mean eggs/mL for volumetric sampling of S. alutus fecundity, by subsampling volume. Subsample number 2 mL 5 mL 10 mL 2 mL Stempel 1 28.0 21.8 30.2 29.5 2 37.0 17.8 36.5 29.0 3 12.0 20.6 37.2 28. 0 4 40.5 25.4 31.8 47.0 5 35.0 21.0 34.5 27.5 6 38.5 30.2 38.6 33 . 0 7 26.0 34.4 26.9 18.0 8 23.5 19.4 38.3 27.5 9 17.0 27.2 37.6 39 . 0 10 29.0 32.8 31.4 35. 0 X 28.7 25.1 34.3 31.4 S.E. 2.97 1.86 1.27 2.47 H 0: Ml=M2=M3=M4 a = 0. 05 Hi- Mi7*M27*M37^4 SS DF MS F TOTAL Groups 464.362 3 154. 787 3.101 (P<0.05) Error 1796.954 36 49. 915 greater than that of the volumetric (Table 16). A 5% c o e f f i c i e n t of v a r i a t i o n i n oocytes/gm could be consistently achieved within four subsamples and processing time was reduced. Extending the time of oocyte drying from 2 4 to 48 h yielded an average decrease of only 0.22% for the entire subsample (Table 17). Since each subsample consisted of approximately 1000 oocytes, t h i s gain was deemed outside the bounds of measurement p r e c i s i o n for in d i v i d u a l oocytes. The mean difference between r e p l i c a t e counts of a 16 92 Table 15. Analysis of variance of eggs/mL as a function of sampling p o s i t i o n within an egg suspension f o r 2 and 5mL subsampling volumes. Total count of 2000 eggs i n 500 mL of water. 2 mL 5mL Number of Positio n subsamples egg/mL S.E. egg/mL S.E. Top; side 5 2.2 0.583 15.6 3.203 Middle; side 5 3.0 0.707 12.0 1. 308 Bottom, side 5 4.2 0.970 13.4 1.503 Middle, centre 5 0.6 0.245 10.6 2.088 Bottom, centre 5 4.8 1.319 16.6 1.435 F=3.8 65 Test H 0: (P<0.02) Mi=M2=M3=M4: F=1.489 (P>0.2) Grand mean 13.64 egg/mL Estimated t o t a l eggs 1364 ± 195. subsample of ovaries (3 6) , made a f t e r they had been returned to storage for s i x months, was 1.69% (expressed as a percentage of the mean for each p a i r ) , with standard error of 0.372. The method i s therefore robust to long-term storage and was adopted for a l l subsequent estimation of fecundity and oocyte charac-t e r i s t i c s . In a l l cases round weight was the most precise s i n g l e - v a r i -able predictor of fecundity but age was an additional s i g n i f i c a n t 93 Table 16. Sampling s t a t i s t i c s for gravimetric estimation of S. alutus fecundity; fixed and variable subsample weights. Subsample Egg Mean Cum. Eggs per Fecundity C V. wt. count mean gram (%) .2651 1445 1451 5473 235515 1457 .2668 1295 1301 1376 4876 209823 8.15 1307 .2620 1380 1384 1379 5282 227298 5.90 1387 .2616 1359 1358 1374 5191 223369 4.80 1356 .2653 1450 1448 1388 5458 234851 4.60 1447 .2331 1240 1235 5298 227974 _ 1231 .2468 1328 1326 1281 5372 231185 1.00 1325 . 1138 644 642 1068 5641 242747 3.30 640 . 1954 1110 1110 1078 5681 244433 3 .50 Table 17. Dry weight of S. alutus oocyte samples i n drying oven at 50° C, at 12 h i n t e r v a l s . Sample weight (gm) S e r i a l Sample % change no. no. Oh 12h 24h 36h 48h 24h -> 48h 15262 1 1.0288 0.1825 0.1824 0.1813 0.1811 -0.71 2 0.8468 0.1559 0.1542 0.1547 0.1545 +0.19 15227 1 0.8972 0.1427 0.1406 0.1423 0.1411 +0.35 2 1.0525 0.1906 0.1898 0.1889 0.1893 -0.26 15417 1 0.8485 0.1797 0.1768 0.1764 0.1761 -0.39 2 0.7897 0.1628 0.1592 0.1585 0.1584 -0.50 X = -0.22% 94 predictor for a l l l i g h t l y exploited stocks. Predictive regres-sions of fecundity were s i g n i f i c a n t l y d i f f e r e n t (F=3.531, p<.01) among stocks but not (p=.08) between stocks grouped by e x p l o i t a -t i o n h i s t o r y (Vancouver Island/Goose Island Gully/Langara Spit vs. Moresby Gully/Rennell Sound) (Table 18). However, the Goose Island Gully p r e d i c t i v e regression i s a constrained estimation because the estimates of weight were generated, i . e . they have no variance. Therefore, the p o t e n t i a l influence of e x p l o i t a t i o n was tested on a subset of these exploited stocks (LA+VI). In t h i s case exploited stocks were shown to have s i g n i f i c a n t l y d i f f e r e n t (F=31.066, p<.0001) pred i c t i v e fecundity r e l a t i o n s h i p s from unexploited stocks. The general e f f e c t of t h i s difference i s a lower fecundity for f i s h i n exploited stocks r e l a t i v e to a s i m i l a r f i s h i n unexploited stocks. While fecundity at age has changed with e x p l o i t a t i o n due to compensatory growth changes, i t appears that these growth changes may have consumed some energy resources which might otherwise have been allocated to the gonads. S p e c i f i c fecundity (oocytes/gm somatic weight) showed l i t t l e v a r i a t i o n among stocks or e x p l o i t a t i o n groups. Its r e l a t i o n s h i p to age was generally asymptotic, although unexploited stocks showed the same quadratic r e l a t i o n s h i p associated with lower weights at age i n the oldest f i s h (Fig. 17). S p e c i f i c fecundity displayed a s i m i l a r r e l a t i o n s h i p to both somatic and round weights i n a l l stocks. Incremental changes i n weight were 95 Table 18. Regression s t a t i s t i c s of fecundity estimation for S. alutus. by stock. Log F = a + brjlog LEN + b ^ o g RDWT + b 2 l o g AGE Stock a *>o b l *>2 MSE R 2 (adj) LA -1. 62101 -2 . 14299 1. 15219 -0. 08343 0.00823 0.783 RSD 1. 23846 -0. 38785 1. 50075 0. 07873 0.00962 0. 673 MOR 1. 13838 0. 66549 0. 88026 0. 28419 0.01257 0.712 GIG 2. 00000 -1. 68915 1. 92963 0. 10823 0.01206 1.000 VI 2. 14857 0. 93589 0. 45558 0. 19249 0.00931 0.577 UNX 1. 21959 -0. 27089 1. 41023 0. 14882 0.01086 0. 687 EXP 0. 64703 1. 09089 0. 96870 0. 06709 0.01096 0.641 TOTAL 1. 23873 0. 07851 1. 23665 0. 11162 0.01101 0.683 EXPSUB 0. 84529 0. 93315 0. 93449 0. 05159 0.01051 0. 698 Test H 0: a l l stock regressions estimate the same population. H-^ : a l l stock regressions do not estimate the same popula-t i o n . F = 0.0360/0.0102 = 3.5312 F . 01(1) ,16,498 = 2 - 0 4 r e j e c t H 0 (p<0.01). Test H 0: UNX and EXPSUB estimate the same regression. H-^ : UNX and EXPSUB do not estimate the same regression. F = 0.33163/0.01067 = 31.0659 F.01(1),4,406 = 3 - 3 7 r e j e c t H 0 (p<0.0001). 500 Specific fecundity - unexploited stocks so \ cn 400 300 g 200 O <U 100 OH 0 0 20 40 Age 60 Figure 17. Spec i f i c fecundity (eggs/gm somatic weight) at age for unexploited stocks of S. alutus. 97 mirrored c l o s e l y by those i n fecundity, beyond a threshold value. Variance i n s p e c i f i c fecundity was s u b s t a n t i a l l y lower (residual mean square was 30% less) i n the exploited group r e l a t i v e to the unexploited. Bagenal (1978) suggested using oocytes/cm as an a l t e r n a t i v e measure of s p e c i f i c fecundity because of the seasonal v a r i a t i o n i n weight c h a r a c t e r i s t i c s , which are not related to reproductive status. This length-based s p e c i f i c fecundity did not display any more structured r e l a t i o n s h i p with age i n e i t h e r e x p l o i t a t i o n group and r e f l e c t e d primarily the d i s t r i b u t i o n of length with age (Fig. 18). The r e l a t i o n s h i p does not appear to be more constant than the weight-based index (cf. F i g . 17) . The lower v a r i a t i o n i n t h i s measure of s p e c i f i c fecundity for exploited stocks was s i m i l a r to that noted for the weight-based index. The two gonadal indices, GIS and GIR, provided a contrast to s p e c i f i c fecundity i n t h e i r relationships with weight and age. As indices they are i n s e n s i t i v e to the form of energy packaging (e.g. oocyte size) but do give better representation of t o t a l reproductive e f f o r t than s p e c i f i c fecundity. Both GIS and GIR were highly variable but were p o s i t i v e l y correlated with age, round and somatic weights. S i g n i f i c a n t differences (F=7.847, p<.001) were evident i n the r e l a t i o n s h i p of GIS to age and somatic weight between expl o i t a t i o n groups (Table 19). The major source of the difference was the non-significance of age as a 15000 - i CO so £f 10000 O CM ft CO 5000 0 0 Specific fecundity - unexploited stocks •A* 20 40 Age i — 60 **. ~ i 80 Figure 18. Spec i f i c fecundity (eggs/cm) at age f o r unexploited stocks of S. alutus. 99 Table 19. Multiple regressions of gonadal index (GIS) and tests of s i g n i f i c a n c e among expl o i t a t i o n groups of S. alutus. Group a bAGE bSOMWT df MSE R 2 (adj) UNX 2.9605E-2 4.050E-4 9.439E-6 190 1.9582E-4 .2107 EXP 2.2483E-2 9.702E-5 1.756E-5 322 1.0000E-4 .1966 TOT 2.4430E-2 4.032E-4 1.160E-5 512 1.4276E-4 .2947 Test H 0: Regressions estimate the same r e l a t i o n s h i p H]_: Regressions do not estimate the same r e l a t i o n s h i p F = 0.001071/0.000137 = 7.8469 F = 3.78 p<.001 regressor i n the exploited stocks. In general, GIS i s a more powerful in d i c a t o r of reproductive e f f o r t than GIR, i . e . has a higher c o e f f i c i e n t of determination (R 2) than GIR. In addition to the differences i n GIS between e x p l o i t a t i o n groups based on age, there was a s i g n i f i c a n t difference based on somatic weight alone. The exploited stocks had lower GIS values (F=29.9, p<.001) at the same somatic weight than unexploited stocks. These differences i n reproductive e f f o r t at age and weight, as measured by the gonadal index, mirrored the lower values of fecundity and s p e c i f i c fecundity for f i s h of s i m i l a r s i z e i n exploited stocks noted e a r l i e r . 100 ( i i i ) Oocyte c h a r a c t e r i s t i c s In r e l a t i o n to age or weight measures alone, both oocyte diameter and oocyte weight were highly variable within a given stock (R2=0.01-0.36). The l i g h t l y exploited Moresby Gully and Rennell Sound stocks showed low but negative c o r r e l a t i o n of oocyte weight and somatic weight while exploited stocks showed p o s i t i v e c o r r e l a t i o n . However, the magnitude of t h i s d i s t i n c t i o n diminished (R2=0.04 -> 0.01) when stocks were examined as e x p l o i -t a t i o n aggregates (Table 20). In general, oocyte c h a r a c t e r i s t i c s were poorly correlated with body parameters. The exception was the Langara Spit stock where oocyte diameter showed a strong p o s i t i v e c o r r e l a t i o n with body weight (somatic weight and round weight); elsewhere, whole-body variables had low c o r r e l a t i o n with oocyte features. Multiple regressions of oocyte diameter, maximum oocyte diameter, minimum oocyte diameter, oocytes/gm ovary weight, and oocyte weight on somatic weight, ovary weight, fecundity and age were used to t e s t for differences correlated with e x p l o i t a t i o n (Table 20) . A l l variables were s i g n i f i c a n t regressors of oocyte characters for f i s h from unexploited stocks. By comparison, age was not a s i g n i f i c a n t regressor of oocyte characters i n the exploited stocks. The net e f f e c t i s to have an additional increase i n oocyte weight and oocyte diameter for older f i s h of the same fecundity and weight as younger f i s h . In a l l cases regressions of oocyte characters were s i g n i f i c a n t l y d i f f e r e n t 101 Table 20. Multiple regressions of oocyte c h a r a c t e r i s t i c s and tes t s of si g n i f i c a n c e among exp l o i t a t i o n groups of S. alutus. Data from l i b e r a t e d oocytes. Variable a b0VWT bAGE bFEC df MSE R 2 (adj) UNX MAXD 773. 5 3.0506 2.1434 -0.0006 255 10,659 0 .465 MIND 479. 9 2.3663 1.4956 -0.0005 255 8, 385 0 . 372 EGDIA 637. 1 2.4408 1.4605 -0.0005 255 7,074 0 .444 EGGM 3781. 0 -8.1728 -14.28 0.0046 225 1.068E6 0 . 099 EGWT 547. 2 2.6879 2.6519 -0.0008 255 15,168 0 .319 EXP MAXD 813 . 5 2.1647 -1.0374 -0.0002 359 8,801 0 . 167 MIND 466. 0 1.5136 -0.2397 -0.0002 359 5, 603 0 .119 EGDIA 661. 5 1.6749 -0.6294 -0.0002 359 5,524 0 . 158 EGGM 5541. 4 -18.1857 -10.943 -0.0015 359 1.236E6 0 . 129 EGWT 482. 1 2.5426 0.8339 -0.005 359 11,849 0 . 182 TOTAL MAXD 799. 1 2.6032 1.0811 -0.0004 614 9,924 0 .311 MIND 458. 4 1.9435 1.3916 -0.0004 614 7,374 0 .267 EGDIA 646. 9 2.0738 0.8911 -0.0003 614 6, 359 0 .310 EGGM 4896. 2 -10.4722 -25.308 -0.0031 614 1.282E6 0 . 128 EGWT 492 . 2 2.5117 2.9691 -0.007 614 13,707 0 . 281 F-tests of oocyte regressions among ex p l o i t a t i o n groups: H 0: Regressions estimate same relati o n s h i p s . H]_: Regressions do not estimate same rel a t i o n s h i p s . F.05(2),4,610 = 3 - 7 4 MAXD MIND EGDIA EGGM EGWT F = 7.2735** Therefore r e j e c t H 0 F = 15.088** " 11 " F = 5.777** " " " F = 16.1425** " " " F = 6.5894** " 11 11 102 (p<.001) between ex p l o i t a t i o n groups. The expression of t o t a l reproductive e f f o r t was mediated through an inverse r e l a t i o n of oocyte s i z e and number, independent of e x p l o i t a t i o n h i s t o r y . There was also a general increase i n oocyte diameter with ovary weight, although v a r i a t i o n was large. Oocyte dry weight i s often used as an analogue of oocyte q u a l i t y (Blaxter and Hempel 1963, Bagenal 1978) and although that r e l a t i o n i s imperfect, oocyte weight has been causally linked with l a r v a l success. Therefore, the r e l a t i o n s h i p of oocyte weight to various ovarian and body c h a r a c t e r i s t i c s was examined more thoroughly between exp l o i t a t i o n groups. Stepwise regres-sions of oocyte weight on oocyte diameter, fecundity, ovary weight, somatic weight and age were performed to i d e n t i f y predictor variables. Standard multiple regressions were then performed on a l l variables i d e n t i f i e d with the stepwise proce-dure. The prime determinant of oocyte weight across a l l stocks was oocyte diameter with somatic weight, ovary weight, and fecundity being s i g n i f i c a n t additional regressors. While oocyte diameter accounted for most of the variance i n oocyte weight, tests among ex p l o i t a t i o n groups used a l l s i g n i f i c a n t v a r i a b l e s . As with other oocyte c h a r a c t e r i s t i c s , s i g n i f i c a n t differences (F=31.754, p<.001) between ex p l o i t a t i o n groups were found, with the t o t a l e f f e c t being a heavier oocyte i n the unexploited stocks r e l a t i v e 103 to the weight of the same siz e oocyte i n the exploited stocks. The difference between ex p l o i t a t i o n groups appears to have a ph y s i o l o g i c a l basis. While oocyte diameter varied s i g n i f i c a n t l y between the groups (Table 20) as a function of the same body variables, there was also a difference (p<.001) i n the oocyte diameter-oocyte weight r e l a t i o n s h i p (Table 21), hence the organic content of the oocytes (oocyte weight i s dry weight). I f dry oocyte weight i s an indicator of oocyte q u a l i t y then f i s h from the unexploited stocks produced a higher q u a l i t y oocyte. (iv) H i s t o l o g i c a l studies a. Maturation of in d i v i d u a l f i s h . Sections of ovaries from f i s h assessed as immature by the gross morphology of the ovary (Table 2) indicated that maturation of an i n d i v i d u a l f i s h takes at lea s t one year to complete. An intermediate stage between immature and mature, designated as •maturing* by Westrheim (1958), can be found i n S. alutus throughout the year. I n i t i a l l y , Westrheim thought t h i s stage was a precursor to maturation i n the year of census but examination of ovaries throughout the year (Westrheim 1975) suggested that t h i s was u n l i k e l y . The answer to t h i s question requires h i s t o -l o g i c a l examination of these 'maturing' ovaries at the time of maximum oocyte development for mature ovaries. Figure 19 provides the h i s t o l o g i c a l evidence that 'maturing' 104 Table 21. Relationship of egg weight (EGWT) and egg diameter (EGDIA) between ex p l o i t a t i o n groups, for S. alutus. Group C o e f f i c i e n t Intercept RSS df R 2 (adj) UNX 1.0048 -128.437 2410428 255 0.5726 EXP 1.1361 -272.554* 2166953 359 0.5832 TOTAL 1.0887 -218.246 5020836 614 0.5722 Test H 0: Regressions estimate same r e l a t i o n s h i p . H]_: Regressions do not estimate same r e l a t i o n s h i p . (5050836-4577381)/ 4577381 F = = 31.7541** (1+1) (2-1) / 614 ovaries did not contain oocytes developed to a stage compatible with f e r t i l i z a t i o n . V i t e l l o g e n e s i s was s t i l l i n the early stages, the nucleus occupied a large portion of the c e l l volume and was i n the perinucleolar stage. By comparison, mature oocytes (Fig. 20) were 3-10 times larger with well developed yolk globules, numerous l i p o i d v e s i c l e s , and no longer showed the perinucleolar n u c l e o l i . The d i f f i c u l t y of determining the exact duration of t h i s 'maturing' stage for an i n d i v i d u a l f i s h i s obvious, however i n d i r e c t evidence for a one year duration can be seen i n comparisons of the three maturational stages. In mature f i s h (Fig. 20) there were only two d i s t i n c t s i z e modes of oocytes, very small («100 M^O primary and large («600 /xm) secondary oocytes. In 'maturing' f i s h there were also two siz e modes, the small primary stage seen i n mature f i s h and a s l i g h t l y 105 Figure 19. Section of ovary from a 'maturing* S. alutus showing the small (50-100 /im) and intermediate (200-250 /nm) s i z e classes of oocytes. 106 Figure 20. Section of ovary from a mature S. alutus female showing the small (50-100 Lira) and large (600 Lira) s i z e classes of oocytes. 107 larger («250 jum) secondary group that was unique to t h i s maturity stage (Fig. 19); the larger mature oocytes were not found. That mature ovaries did not contain t h i s smaller, secondary s i z e mode indicates both that the maturation of oocytes must be complete within a year and that maturation of in d i v i d u a l f i s h must not exceed one year. For the l a t t e r to exceed one year would require evidence of a t h i r d s i z e mode of secondary oocytes, at the time of maximum oocyte development for f e r t i l i z a t i o n . b. Maturation of in d i v i d u a l oocytes. The maturity stages present i n the h i s t o l o g i c a l samples permit the description of the complete maturation process for S. alutus oocytes. Primary oocytes a r i s e from oogonial nests which are present throughout the ovarian tiss u e but are concentrated near the branches of the ovarian lamellae (Fig. 21). Oogonia i n these nests are d i f f i c u l t to recognize (Fig. 22) and no synaptic n u c l e i have been p o s i t i v e l y i d e n t i f i e d . F o l l i c u l o g e n e s i s appears to begin i n the oogonial nests. As the f o l l i c l e s move away from the oogonial nests development proceeds through three recogni-zable stages (after Tokarz 1978): chromatin-nucleolus, early perinucleolus and l a t e perinucleolus. In the chromatin-nucleolus stage the oocytes are quite small («3 0 /xm) with prominent n u c l e o l i and r e t i c u l a t e chromatin present i n the nucleus. The ooplasm i s dense and intensely basophilic. As the oocyte enters the early perinucleolar stage i t 108 Figure 21. Section of S. alutus ovary showing the concentration of immature oocytes near the branch of the ovarian lamellae. 109 Figure 2 2 . Oogonial nest (ON) with immature oocytes from a S. alutus ovary. 110 increases i n s i z e («150 /zm) and the nuclear region enlarges. Nu c l e o l i multiply and assume positions around the perimeter of the nucleus ('perinucleolar'). The ooplasm becomes more acido-p h i l i c as v i t e l l o g e n e s i s begins and small v e s s i c l e s appear near the nucleus (Fig. 23) . Yolk globules begin to form at the periphery of the ooplasm and formation proceeds c e n t r a l l y (Fig. 24) . The l a t e perinucleolar (diplotene) stage i s of longest duration and marked by the largest changes. During t h i s stage the oocyte enlarges 2-10 times as v i t e l l o g e n e s i s i s completed, with most of the increase accounted for i n the ooplasm. Yolk globules enlarge and the number of l i p o i d a l v e s s i c l e s increases dramatically. As the oocyte attains f u l l maturity (Figs. 25 and 26) the l i p o i d a l v e s s i c l e s increase i n s i z e up to «30 nm and are 2-5 times as large as the largest yolk globules. The n u c l e o l i lose t h e i r perinucleolar orientation, the nucleus diminishes and the nuclear membrane loses i t s i n t e g r i t y . At f u l l maturity a d i s t i n c t nucleus i s not apparent. Mean diameters of f u l l y mature oocytes range from 530-63 0 jum i n h i s t o l o g i c a l section, while those of the immature oocytes are from 97-103 /L im. Relative s i z e d i s t r i b u t i o n within the ovary i s treated i n section (d). c. F o l l i c u l a r a t r e s i a I do not have s u f f i c i e n t samples to q u a n t i t a t i v e l y assess a t r e s i a over the complete developmental sequence, however I l l Figure 23. Early perinucleolar stage of S. alutus oocyte develop-ment. [Nucleus (N), n u c l e o l i (NU)]. 112 Figure 24. Perinucleolar stage of S. alutus oocyte showing development of yolk globules (YG) at the periphery of the oocyte. 113 Figure 25. F u l l y mature S. alutus oocyte showing large l i p o i d a l v e s s i c l e s (LV). Figure 26. F u l l y mature S. alutus oocyte. The nucleus has l o s t i t s i n t e g r i t y and no n u c l e o l i are present. 115 some inferences can be made. In mature ovaries at the stage of maximum oocyte development the r a t i o of a t r e t i c f o l l i c l e s (Fig. 27) to normal mature f o l l i c l e s ranges from 0.27-0.50, by stock (Table 22) . Differences among stocks are highly s i g n i f i c a n t (F=7.97, p<.001) but only marginally so (0.01<p<0.05) between ex p l o i t a t i o n groups. The stocks at the geographic extremes (Vancouver Island and Langara Spit) exhibit the largest depar-tures from the grand mean, hence the s i g n i f i c a n c e of e x p l o i t a t i o n group differences may be a geographic a r t i f a c t . That both of these stocks are highly exploited yet t h e i r departures are of opposite sign renders t h i s inference more p l a u s i b l e . A geogra-phic c l i n e i n the proportion of f o l l i c l e s undergoing a t r e s i a might be expected i f temperature were a c o n t r o l l i n g v a r i a b l e i n f o l l i c u l a r development, although most evidence suggests hormonal (Wasserman and Smith 1978) or n u t r i t i o n a l (Dunn and Tyler 1969, Dunn 1970) influences as the major factors c o n t r o l l i n g f o l l i c u l a r maturation. A t r e s i a i n p o s t - f e r t i l i z e d or post-mature oocytes i s discussed i n section (e) below. d. Oocyte sizes H i s t o l o g i c a l sections were taken from s i x locations within each ovary to determine i f assessment of maturational state or oocyte c h a r a c t e r i s t i c s was influenced by place of o r i g i n 116 Figure 27. A t r e t i c f o l l i c l e i n normally developing S. alutus female, i l l u s t r a t i n g hypertrophied f o l l i c u l a r c e l l s . 117 Table 22. ANOVA for mean numbers of a t r e t i c to non-atretic f o l l i c l e s among stocks and exp l o i t a t i o n groups of S. alutus. Data from h i s t o l o g i c a l samples. W=within sample, A=among samples. Stock or group N Mean r a t i o A t r e t i c / n o n - a t r e t i c Standard error of r a t i o Langara Spi t 13 Rennell Sound 7 Moresby Gully 15 Vancouver Is. 51 Test H, 2 _ 2 W A Unexploited 64 Exploited 22 Test H 0 : CTW 0.267 0.397 0.295 0.498 F = 7.97** (p<0.001) 0.454 0.337 F = 5.986* (p<0.02) 0. 041 0. 059 0. 037 0. 030 0. 218 0.151 within the ovary. Within stocks there was a consistent d i f -ference i n the diameters of mature oocytes between the central and peripheral regions of the ovary, peripheral oocytes being larger. Though consistent, the differences were not s t a t i s t i c a l -l y s i g n i f i c a n t i n the aggregate. The smallest mature oocytes were consistently present at the centre of the ovary i n compari-son with anterior and posterior sections along the midline. There was also an anterior-posterior c l i n e of increasing s i z e of mature oocytes at the periphery of the ovary. Medial sections of the ovary show l i t t l e difference between anterior and posterior but the smaller s i z e was evident i n the central section. Immature oocytes showed trends s i m i l a r to mature oocytes 118 with regard to diameters at central and peripheral sections of the ovary. Centrally located immature oocytes also tended to be smaller than eit h e r anterior or posterior oocytes. However, there was no consistent evidence of an anterior-posterior gradient f o r peripheral immature oocytes, as there was for mature oocytes. Comparisons among stocks revealed s i g n i f i c a n t differences (p<.001) i n mature and immature oocyte s i z e s , however these differences were not associated with e x p l o i t a t i o n e f f e c t s (Table 23) . The o r i g i n of the differences among stocks appeared to be geographic since the two northern stocks, which have d i f f e r e n t e x p l o i t a t i o n h i s t o r i e s , had larger mature oocytes than the two more southerly stocks, which were also at d i f f e r e n t e x p l o i t a t i o n states. While the Rennell Sound stock had s i g n i f i c a n t l y larger mature oocytes than the three other stocks (F=14.330, p<.002) the geographical differences among stocks of s i m i l a r e x p l o i t a t i o n h i s t o r y (Vancouver Island vs. Langara Spit and Moresby.Gully vs. Rennell Sound) were also s i g n i f i c a n t (p<.002) (Table 23), in d i c a t i n g s i z e and age of f i s h were not overriding factors. Unfortunately, sample sizes within stocks f o r s i z e and age-speci-f i c examinations were too small for meaningful comparisons. Estimation of mean oocyte s i z e can be influenced by the phy s i o l o g i c a l stage or 'age' of the oocytes. I t could be argued that the larger oocytes i n northern stocks r e f l e c t a l a t e r stage 119 Table 23. ANOVA of oocyte sizes by stock and e x p l o i t a t i o n group for S. alutus. Data from h i s t o l o g i c a l samples. Mature oocyte diameters. Source df SS MS F H 0: m = M 2 = M 3 = M 4 Stock 3 226.910 75.636 23.620** Error 20 64.043 3.202 Total 23 290.953 prob. F <0.0001 r e j e c t H Q  H 0 : M E = Group 1 0.013 0.013 0.0009 Error 22 291.066 13.230 Total 23 291.073 prob. F >0.99 accept H 0  H 0 : MRSD = MMOR+VI+LA Group 1 114.804 114.804 14.330** Error 22 176.241 8.010 Total 23 291.045 prob. F <0.001 r e j e c t H 0  H 0 : MRSD = M^OR Stock 1 174.360 174.360 45.892** Error 10 37.992 3.799 Total 11 212.352 prob. F <0.001 r e j e c t H Q  H 0 : MLA = Mvi Stock 1 52.639 52.639 20.215** Error 10 26.039 2.603 Total 11 78.678 prob. F <0.002 r e j e c t H 0 Immature oocyte diameters H 0: Mi = M 2 = M 3 = M4 Stock 3 5.659 2.492 8.820** Error 20 4.277 0.213 Total 23 9.936 prob. F <0.001 r e j e c t H 0 120 of development, rather than larger ultimate s i z e . Countering t h i s argument i s the fact that r e s i d u a l , u n f e r t i l i z e d oocytes i n spawning and post-spawning f i s h (next section) show the same differences noted i n the pre-spawners. Temperature influence on developmental rates would also, on the basis of the colder temperatures i n the area of the northern stocks, lead to exactly the opposite trend i n oocyte si z e as that observed. In addition, these differences among stocks were noted i n those oocytes having undergone nuclear d i s s o l u t i o n , which i s the stage immediately p r i o r to f e r t i l i z a t i o n and beyond which there i s no further oocyte growth (Moser 1967). Lastly, the Goose Island Gully sample discussed i n the following section, although s l i g h t l y lagged on the spawning period of the Rennell Sound sample, displayed a l l of the same stages and differences noted above. e. F e r t i l i z a t i o n The fecundity estimates calculated i n t h i s study are for p r e - f e r t i l i z e d females and therefore are an uncertain r e f l e c -t i o n of f e r t i l i t y , i . e . the number of larvae ultimately released. However, censusing f e r t i l i t y i s d i f f i c u l t i n t h i s genus because the p r o b a b i l i t y of extrusion of larvae increases with proximity to p a r t u r i t i o n (release of larvae). I f the only source of samples i s from commercial landings, t h i s p r o b a b i l i t y increases due to the storage and handling practices of vessels and proces-sing plants. Although some authors (Boehlert et a l . 1983) have showed decreasing fecundity as p a r t u r i t i o n approaches, t h e i r 121 samples were from commercial landings and quantitative con-clusions should be viewed cautiously. However, a t r e s i a of ripe f o l l i c l e s i s a common feature of vertebrate ovaries (Byskov 1978) and the general conclusions of these authors are undoubtedly correct. My estimates of fecundity represent the maximum pot e n t i a l fecundity and should not be interpreted as f e r t i l i t y (number of young born). Females were examined near the time of p a r t u r i t i o n (Feb.-Mar.) to determine whether a l l ripe oocytes were f e r t i l i z e d and i f p a r t i c u l a r ages, sizes, or stocks were associated with incomplete f e r t i l i z a t i o n . Samples could be obtained from only two stocks (Goose Island Gully and Rennell Sound) but they were stocks of opposite e x p l o i t a t i o n h i s t o r i e s and had maximum p r o b a b i l i t y of manifesting such e f f e c t s . Sample sizes from the two stocks were small but s i m i l a r (Goose Island Gully - 69) and (Rennell Sound - 71) and most f i s h were spawning or spent. The Rennell Sound sample was characterized p r i m a r i l y by ruptured f o l l i c l e s (corpora a t r e t i c a , 77%) with some a t r e t i c , unovulated oocytes (18%) and some residual embryos (5%) (Table 24) . Three f i s h lagged the general population and showed l a t e embryonic development. An additional three f i s h (4.2% of tota l ) had not ovulated and the complete complement of f o l l i c l e s was undergoing the early stages of a t r e s i a . Figures 28-30 i l l u s t r a t e t h i s s u i t e of conditions for the Rennell Sound sample. 122 Table 24. Percent composition of ovarian contents for post-f e r t i l i z e d S. alutus, by category, for the Rennell Sound and Goose Island Gully stocks. A t r e t i c Ruptured Ruptured Early Late Not f o l l i c l e s chorion embryos embryos ovulated Rennell Sound fN=71) Mean 17.8 76.9 0.3 3.7 1.3 0.0 S.E. 2.3 3.0 0.3 1.8 0.7 0.0 Max. 100.0 98.2 20.0 88.9 44.8 0.0 Min. 0.0 0.0 0.0 0.0 0.0 0.0 Goose Island Gully (N=69) Mean 5.8 18.6 0.9 30.8 7.4 36.4 S.E. 1.0 2.7 0.7 3.5 1.7 4.1 Max. 88.5 100.0 95.0 100.0 90. 0 100.0 Min. 0.0 0.0 0.0 0.0 0.0 0.0 S.E. = standard error of mean Max. = maximum value Min. = minimum value The Goose Island Gully sample was c o l l e c t e d one week a f t e r the Rennell Sound sample but showed a broader range of develop-mental stages (Table 24) and suggested a more extended period of p a r t u r i t i o n compared with the Rennell Sound stock. Embryonic stages were more common as were pre-ovulatory, non-atretic f o l l i -c l e s . No d e f i n i t e examples of completely a t r e t i c oocyte comple-ments were found i n t h i s stock. Figures 31-32 i l l u s t r a t e these stages for the Goose Island Gully stock. These data suggest both temporal v a r i a t i o n i n the duration of reproduction and d i f f e r e n t i a l success i n female insemination 123 Figure 28. Ruptured f o l l i c l e s (RF) i n a post-spawned S. alutus ovary from the Rennell Sound stock. 124 Figure 29. Residual, ruptured chorion and embryo remaining i n the ovary of a post-spawned S. alutus from the Rennell Sound stock. 125 Figure 30. A t r e t i c , u n f e r t i l i z e d f o l l i c l e s remaining i n the ovary of a S. alutus from the Rennell Sound stock. 126 Figure 31. Embryonic stages i n the ovary of a S. alutus from the Goose Island Gully stock. 127 Figure 32. Pre-ovulatory, non-atretic f o l l i c l e s remaining i n the ovary of a S. alutus from the Goose Island Gully stock. 128 or f e r t i l i z a t i o n among stocks. That the l a t t e r was associated with a l i g h t l y exploited stock might suggest reproductive senescence, but the age range (9-41 y) of those u n f e r t i l i z e d i n d i v i d u a l s argues against t h i s explanation. A t r e s i a of a f u l l y mature oocyte complement has not previously been recorded for t h i s genus, although a t r e s i a of sub-mature oocytes has been noted i n i t and others (Byskov 1978, Foucher and Beamish 1980). This f i n d i n g merits future examination of po t e n t i a l causes, i n p a r t i c u l a r to determine i f behavioural rather than environmental factors are involved. f. Range of developmental stages A l l developmental stages from pre-ovulatory, r i p e f o l l i c l e s to hatched larvae were found i n the Goose Island Gully females. In contrast, the Rennell Sound sample showed a much higher degree of synchrony, with 90% of the f i s h having only ruptured and a t r e t i c f o l l i c l e s present i n the ovary. A cause for the greater synchrony has not yet been determined but may be related to the much smaller area occupied by t h i s stock. However, synchronous oocyte development i s the common condition i n vertebrate ovaries (Tokarz 1978). The h i s t o l o g i c a l samples indicate that p a r t u r i t i o n may be protracted for some stocks. A recent study o f f C a l i f o r n i a (Echeverria 1987) suggest t h i s i s the normal condition and that examples such as the Rennell Sound synchrony are exceptional. 129 Detailed data on the gestation length for i n d i v i d u a l females have not been c o l l e c t e d f or offshore Sebastes spp., however such data are a v a i l a b l e for inshore forms. Boehlert and Yoklavich (1984) estimated a gestation length of 37 days for S. melanops o f f Oregon while Dygert (1986) estimated 41.5 days f o r S. caurinus o f f Washington. Gestation might take longer i n colder, offshore waters but the mean of these two studies (39) days may be used as an estimate for gestation of S. alutus embryos. I f t h i s i s true, and the early embryos observed are eventually released, then p a r t u r i t i o n o f f Rennell Sound may have occurred as early as February, while i n Goose Island Gully i t could s t i l l be occurring well into A p r i l . The eventual fate of the unovulated oocytes i n the Goose Island Gully sample i s d i f f i c u l t to determine. Since many oocytes i n the same f i s h had already ovulated and been f e r t i -l i z e d , these unovulated oocytes may represent eit h e r i n s u f f i c i e n t insemination (an abnormal s i t u a t i o n for in t e r n a l f e r t i l i z a t i o n ) , an i n t e r n a l f a i l u r e i n the process of ovulation, or an extremely protracted period of p a r t u r i t i o n . Both recent studies o f f C a l i f o r n i a (Echeverria 1987) in d i c a t i n g such protraction, and the r e l a t i v e bathymetry of Goose Island Gully vs. Rennell Sound (broad coastal shelf vs. narrow offshore bank, respectively) favour the l a t t e r explanation. A broad coastal shelf may present more s p a t i a l opportunity for successful recruitment into a stock than a smaller 'target', such as the Rennell Sound bank. 130 g. Multiple spawning Several Sebastes spp. o f f C a l i f o r n i a have been reported to spawn more than once per year (MacGregor 197 0) although a recent review of 34 species (Echeverria 1987) indicated that, while some species may have multiple spawning, the normal con-d i t i o n i s one of protracted spawning. Some species may have a p a r t u r i t i o n period of up to nine months o f f C a l i f o r n i a although the precise duration may be less because Echeverria assumed p a r t u r i t i o n within a month i f eyed larvae were present i n the ovary. Multiple spawning for S. alutus would be suggested i f there were more than two s i z e modes of oocytes present i n the ovary at any point i n the reproductive cycle. None of my samples showed such a s i z e d i s t r i b u t i o n of oocytes, i n f e r r i n g a single spawning episode per year. Although he conducted no h i s t o l o g i c a l examina-tions Westrheim (1975) reached the same conclusion based on a review of the seasonality of maturity states of S. alutus from research vessel samples. I also examined the data from the l i g h t l y exploited stocks (Rennell Sound, Moresby Gully) for evidence of other elements regarded as important components or mechanisms of l i f e h i s t o r y theory: reproductive cost and senescence. 131 (i) Reproductive cost. A cost of reproduction might be advanced as an explanation for the smaller s i z e of older f i s h r e l a t i v e to middle-aged f i s h . Reproductive cost might be ex-pressed as e i t h e r a d i r e c t s u r v i v a l function or some non-fatal, d e b i l i t a t i n g function r e s u l t i n g i n shrinkage or lack of growth. An increase i n mortality i s very d i f f i c u l t to demonstrate without following i n d i v i d u a l s for whom reproductive e f f o r t i s known; something not possible with my data and perhaps impossible i n t h i s genus. Age compositions were examined fo r evidence that females, with t h e i r greater energetic demands during reproduc-t i o n , suffered higher mortality than males, f o r f u l l y r e c ruited f i s h . Random samples by sex and stock (Fig. 3 to 7 and Archibald et a l . 1981) did not indicate a higher rate of mortality on females, rather they suggested the opposite. In a l l samples there was a s l i g h t l y higher abundance of females i n the older age groups, i n d i c a t i n g either higher mortality f o r males or d i f f e r e n -t i a l a v a i l a b i l i t y with age. The l a t t e r i s possible, considering the seasonal bathymetric migrations of t h i s species and the dates fo r these c o l l e c t i o n s . However, these differences were neither major nor generally found i n the more extensive samples from commercial f i s h e r i e s on these stocks (Archibald et a l . 1981). A non-fatal, physiological cost of reproduction was i n v e s t i -gated using the gonadal indices previously defined (GIS, GIR) . I f a cost to reproduction exists then i t should be r e f l e c t e d i n 132 changes of these r a t i o s with age, assuming that p h y s i o l o g i c a l processes i n these older f i s h would be mirrored by somatic and gonadal weights. A reproductive cost hypothesis implies either increased mortality from reproductive a c t i v i t y , or maintenance of reproductive e f f o r t at the expense of the somatic t i s s u e . Increa-sing gonadal indices with age would support such a hypothesis. Although variable, the two gonadal indices r e f l e c t only d i f f e r e n -ces i n s i z e with age, even i n view of the smaller, older f i s h noted e a r l i e r . For these older f i s h , the two measures of reproductive e f f o r t are appropriate or s l i g h t l y higher for f i s h of that s i z e . The r e l a t i v e constancy of these relationships argues that somatic resources are not being diverted to maintain gonadal expenditures, appropriate to the somatic weight p r i o r to reproduction. However, i t i s possible that t r a n s f e r of somatic resources r e s u l t s i n a decreased gonadal output, appropriate to the al t e r e d somatic weight, hence no change i n the gonadal indices. I t does not seem possible to resolve t h i s dichotomy without h i s t o r i c a l data on reproductive e f f o r t by i n d i v i d u a l s , but the enhanced oocyte c h a r a c t e r i s t i c s of these older f i s h suggests that the former explanation (no reallocation) i s more probable. ( i i ) Senescence. Neither oocyte nor gross gonadal charac-t e r i s t i c s showed any evidence of reproductive senescence of older f i s h i n l i g h t l y exploited stocks. While older f i s h did have lower oocyte weights, they can be accounted for e n t i r e l y by lower 133 body weights at these older ages. H i s t o l o g i c a l data showed increasing numbers of a t r e t i c f o l l i c l e s with age but there were very few samples >40 y. More importantly, the number of mature f o l l i c l e s increased concomitantly, and although v a r i a b l e , the proportion of a t r e t i c f o l l i c l e s showed no trend. 3. Discussion The lower lengths and weights at older ages merit detailed consideration. Certainly, older f i s h of smaller s i z e or weight than younger f i s h have been observed and commented on previously, and i n harvested f i s h populations they are generally thought to be an a r t i f a c t of s e l e c t i v e f i s h i n g mortality on faster-growing f i s h (Ricker 1969). The presence of smaller, older i n d i v i d u a l s i n the l i g h t l y - e x p l o i t e d Rennell Sound and Moresby Gully stocks argues against t h i s explanation for S. alutus. A l t e r n a t i v e l y , i t can be argued that either: (i) there i s a threshold or cumulative cost to reproduction; ( i i ) senescence manifests i t s e l f through ph y s i o l o g i c a l costs leading to shrinkage; ( i i i ) growth rate within a cohort i s inversely correlated with mortality rate ; or (iv) there has been a long-term increase i n S. alutus growth rate. Empirical evidence supporting any of these hypotheses i s scanty, p a r t i c u l a r l y for iteroparous species with many reproduc-t i v e episodes. Indeed, most studies of the costs of reproduction i n natural populations (Browne 1982, B e l l 1984 a,b) have not 134 demonstrated the theorized costs, but have shown the opposite. Reznick (1985) has detailed the empirical evidence f o r reproduc-t i v e cost and found most studies to be inconclusive because of a f a i l u r e to e s t a b l i s h mechanisms for the process described. However, fo r the few studies he regarded as well designed, the support for a reproductive cost argument was strong. Ekman and Askenmo (1986), i n a long-term study of b i r d populations have suggested that reproductive cost e x i s t s by showing lower s u r v i v a l fo r breeding vs. non-breeding males. Unfortunately, that study did not e s t a b l i s h a continuous function of reproductive e f f o r t and s u r v i v a l , nor did i t census maternal s u r v i v a l r e l a t i v e to c l u t c h s i z e . Many experimental manipulations of reproductive success have eith e r questionable a p p l i c a t i o n to natural systems (e.g. Askenmo 1979) or have f a i l e d to demonstrate reproductive costs consistent with the hypotheses ( L u c k i n b i l l 1984) . The argument for a senescence function leading to smaller, ( i . e . decreased) sizes of older f i s h i s not supported by repro-ductive c h a r a c t e r i s t i c s (Section 2 ( i i i ) ) , however i t i s quite possible that reproduction would have the highest p r i o r i t y of energy a l l o c a t i o n for these older f i s h (Wootton 1977). Another a l t e r n a t i v e for smaller sizes of older f i s h , an inverse r e l a t i o n -ship of growth rate and mortality, appears to have the best support from experimental studies (Brett 1974, 1979) although f i e l d studies are l i m i t e d for p o s t - l a r v a l f i s h (Gerking 1959). Recently, Mann et a l . (1984) presented evidence f o r such a trade-135 o f f i n two species of freshwater f i s h although the e f f e c t s of density and r a t i o n were uncontrolled. I am unaware of s i m i l a r studies f o r marine f i s h . The a l t e r n a t i v e of long-term changes i n growth rate, i s very d i f f i c u l t to v e r i f y . Although Boehlert and Yoklavich (1987) suggest t h i s process i n S. pinniqer and S. diploproa. t h e i r evidence i s based on back-calculated sizes at age. Estimated sizes for f i s h i n early years were therefore based on older f i s h than those for more recent years and the p e r i l s of Lee's phenomenon loom large (Lee 1912, Ricker 1975). The contention by some authors (Knight 1978, Roff 1980) that asymptotic growth i n general, and use of the von Bertalanffy function i n p a r t i c u l a r , i s inappropriate and misleading may be a t t r i b u t e d to the persistent observations of these smaller and larger than expected f i s h , throughout the age range. Unfortu-nately, the problem i s generally addressed from the perspective of c u r v e - f i t t i n g , rather than from that of describing the under-l y i n g growth process(es). I t i s therefore not s u r p r i s i n g that the use of s i n g l e functions to describe e n t i r e size-at-age d i s t r i b u t i o n s has been unsatisfactory. Boehlert and Yoklavich (1984b) and Theilacker (1987) have presented r e s u l t s which suggest a possible p h y s i o l o g i c a l mecha-nism f o r the growth-mortality inverse r e l a t i o n s h i p . Their studies show two d i f f e r e n t modes of growth i n l a r v a l clupeoids, e i t h e r high growth rate or high e f f i c i e n c y of conversion, but not 136 both. I f the processes they observed i n clupeoids are general, then the small, older S. alutus may be the i n d i v i d u a l s with high conversion e f f i c i e n c i e s . Westrheim's (1973) observation of smaller s i z e for older S. alutus i n deeper water, where there i s presumably le s s available food, lends support f o r such a mecha-nism. The inverse r e l a t i o n s h i p of growth and mortality rates I suggest (Mulligan and Leaman, i n prep.) argues that the form of growth for i n d i v i d u a l f i s h could be consistent (e.g. a l l f i s h may have a simple monotonic growth h i s t o r y ) , yet v a r i a t i o n s within i t could s t i l l give r i s e to d i s t r i b u t i o n s which imply combinations of d i f f e r e n t growth forms. This inference i s encouraging because the a l t e r n a t i v e of maintaining genetic machinery f o r d i s t i n c t growth forms i s e v o l u t i o n a r i l y expensive and has strong implica-tions about the nature of the s e l e c t i n g forces at work. However, a major conclusion of our work i s that models of competing hypotheses can be ambiguous unless size-at-age data are c o l l e c t e d throughout the growth period of a cohort. Resolution of compe-t i n g hypotheses about growth without such data i s d i f f i c u l t or impossible. The problems of such data c o l l e c t i o n f o r long-lived species (e.g. S. alutus) are obvious. The larger oocytes i n the northern stocks conform with previous i n t r a - s p e c i f i c relationships noted for marine f i s h (Mann and M i l l s (1979). The general explanation for t h i s phenomenon i s 137 that northern waters are less productive, hence larvae require greater resources to make a successful t r a n s i t i o n to exogenous feeding. However, further research on the two northern stocks of P a c i f i c ocean perch may be required as Ware (1975) has noted the tendency for e a r l i e r spawners to have larger eggs. Since partu-r i t i o n i n the Rennell Sound stock preceded that of the majority of the Goose Island Gully stock, the p o s s i b i l i t y of the mechanism noted by Ware must be acknowledged. Resolution w i l l require samples at exactly the same phys i o l o g i c a l stage of development, and blastodisc-stage embryos would be the best the choice. E a r l i e r - s t a g e oocytes are d i f f i c u l t to stage with the necessary p r e c i s i o n and later-stage embryos would d i s t o r t measurements. Additional study of inter-annual v a r i a t i o n of oocyte si z e i n stocks under d i f f e r e n t temperature regimes might also provide i n s i g h t into t h i s process. Recent work on external n u t r i t i o n for embryos (matrotrophy) i n several Sebastes spp. (Boehlert and Yoklavich 1984a, Boehlert et a l . 1987) bears on my r e s u l t s about a t r e s i a of mature oocytes. Those authors concluded, on the basis of respiratory and c a l o r i c measurements, that developing Sebastes embryos required almost 1.5 times more energy for development than was contained i n the yolk at f e r t i l i z a t i o n , and that there must be a source of n u t r i t i o n external to the yolk reserves. This f i n d i n g was also supported by h i s t o l o g i c a l evidence of absorbtion of yolk proteins i n the hindgut of embryos, when such proteins were not present i n 138 the maternal plasma. They concluded that embryo death and subsequent release of proteinaceous materials was the source of additi o n a l energy for the developing embryos. My findings of a t r e t i c f o l l i c l e s i n both developing, p o s t - f e r t i l i z e d , and post-p a r t u r i t i o n females suggest f o l l i c u l a r a t r e s i a as an alt e r n a t i v e candidate f o r t h i s energy source. They also underscore Wourms1 (1981) contention that c l a s s i f i c a t i o n of reproduction and embryonic n u t r i t i o n into discrete categories (e.g ovoviviparous, viviparous, l e c i t h o t r o p h i c , matrotrophic, etc.) i s inappropriate when dealing with such a continuum of processes. A high proportion of mature f o l l i c l e s undergoing a t r e s i a i s the common s i t u a t i o n i n higher vertebrates (75-99% i n mammals) but there are divergent views on the normal value i n fi s h e s . The general view (Rastogi 1966, Vladykov 1956) i s that t h i s propor-t i o n i s approximately 30-40% i n tel e o s t s , although Henderson (1963) concluded only 3-5% of mature oocytes become a t r e t i c . Perhaps s i g n i f i c a n t l y , Vladykov considered wild f i s h while Henderson examined hatchery-reared f i s h . Observations on v i v i p a -rous or ovoviviparous fishes are sparse and q u a l i t a t i v e (Magnus-son 1955, Moser 1967). No conclusive evidence of a t r e s i a i n immature oocytes was found, however t h i s does not indicate the extent of a t r e s i a i n small oocytes, because they disappear very quickly and leave no traces. Small f o l l i c l e s may vanish within several hours (Byskov 139 1978) and neither thecal nor granulosa c e l l s are reported to hypertrophy i n most teleosts (Rajalakshmi 1966). Lehri (1968) suggested that only a few f o l l i c l e s become a t r e t i c p r i o r to v i t e l l o g e n e s i s and Moser (1967) indicated a t r e s i a i s uncommon i n f o l l i c l e s <400 jum for S. paucispinis. The Sebastes spp. may contain exceptions as I found hypertrophied granulosa c e l l s i n S. alutus (Fig. 27) and Bowers (pers. comm.) has found s i m i l a r evidence for S. flavidus o f f C a l i f o r n i a . I do not have an explanation for the more synchronous and s l i g h t l y e a r l i e r p a r t u r i t i o n of the Rennell Sound stock compared with the Goose Island Gully stock. Boehlert et a l . (1987) suggest that larger oocytes are associated with a longer gesta-t i o n period and the larger oocytes of the Rennell Sound stock, combined with a lower temperature regime, might be expected to delay p a r t u r i t i o n r e l a t i v e to more southerly stocks. A possible cause for the e a r l i e r Rennell Sound p a r t u r i t i o n i s that the stock i s p r i m a r i l y large, old f i s h and such f i s h have been observed to spawn before younger and smaller f i s h i n several species (Bagenal 1978), although that finding i s not universal. I t also notewor-thy that the Goose Island Gully stock contained i n d i v i d u a l s who had released larvae at approximately the same time as those i n the Rennell Sound stock, although the mean time of p a r t u r i t i o n was l a t e r . A more detailed study of the time of p a r t u r i t i o n , p a r t i c u l a r l y for stocks i n s i m i l a r l o c a l i t i e s and having s i m i l a r demographic structure, may provide the answer. 140 My r e s u l t s can also be used to examine the modifications and trade-offs predicted by l i f e h i s t o r y theory. This study permits a more d e t a i l e d treatment of these predictions because the longe-v i t y and determinate growth form allow age and s i z e to be exa-mined as independent variables; often confounded i n l i f e - h i s t o r y studies. The predictions concerning unmodified populations are la r g e l y r e a l i z e d . T h e o r e t i c a l l y : a) Reproductive e f f o r t should increase with age. Reproduc-t i v e e f f o r t as measured by fecundity or ovary weight does i n -crease with age, although there i s confounding of age and weight throughout some of the l i f e s p a n . As for other poikilotherms, reproductive e f f o r t i n S. alutus increases with s i z e , however where the age range was extensive, ovary weight and fecundity of older f i s h were greater than for younger f i s h of the same somatic weight. In t h i s sense the prediction i s true. I t i s also true i n a much broader sense since older i n d i v i d u a l s continue to reproduce when somatic growth diminishes or ceases. At those ages a l l energy surplus to maintenance i s channelled into reproduction (including non-gonadal expenditures such as spawning migrations), hence reproductive e f f o r t as a proportion of t o t a l energy a l l o c a t i o n increases s u b s t a n t i a l l y with age. This may be p a r t i c u l a r l y so, as suggested by Clarke (1987), i f l i f e t i m e reproductive output i s fixed and reproductive e f f o r t must be increased i n higher l a t i t u d e s due to temperature e f f e c t s . The larger oocyte sizes i n the more northerly S. alutus stocks 141 support t h i s contention. b) Late maturity. The age frequencies of l i g h t l y exploited stocks indicate that mortality i s r e l a t i v e l y constant and low (M=0.05, Archibald et a l . 1981, Leaman 1987a) throughout adult l i f e . I f age at maturity i s considered, i t i s obvious that mortality i n t h i s species i s absorbed p r i m a r i l y by p r e - r e c r u i t and pre-reproductive i n d i v i d u a l s . Under such mortality schedules the optimum l i f e h i story (Gadgil and Bossert 1970) i s to delay maturity u n t i l a f t e r t h i s period has passed. The l a t e age at maturity (7-9 y) of S. alutus c l e a r l y supports t h i s p r e d i c t i o n . c) Interaction of age and fecundity d i s t r i b u t i o n . Giesel (1974) suggested that long-term rates of population increase would be highest i f the d i s t r i b u t i o n s of age and fecundity were congruent. The maxima of i n d i v i d u a l fecundity (43-48 y) and the mean age i n l i g h t l y exploited stocks («32 y) argue that t h i s congruency i s not r e a l i z e d . While the mean age of 32 y i s less than that of maximum fecundity, the long-term average age i n unexploited stocks i s even lower (24 y; Leaman 1987a); the present means are greatly influenced by the strong 1952 cohort. Such strong, infrequent cohorts are c h a r a c t e r i s t i c of r o c k f i s h stocks and bias the mean age during t h e i r residence. In addi-t i o n , the quadratic growth form noted e a r l i e r tends to decrease the mean age of maximum weight (hence fecundity). The net e f f e c t of these factors i s to decrease the congruency of age and 142 fecundity d i s t r i b u t i o n . These differences c l e a r l y run counter to Giesel's suggestion. A l t e r n a t i v e l y , i t can be argued that the mean age of maximum in d i v i d u a l fecundity w i l l be l e s s important than that of maximum cohort fecundity. Since the l a t t e r occurs at ages <20 y, the v a r i a t i o n i n cohort strength w i l l act, on average, to make t h i s age somewhat older and the d i s t r i b u t i o n of fecundity and age w i l l be more s i m i l a r . L i f e h i s t o r y l i t e r a t u r e hypothesizes that adaptations to a l t e r e d mortality schedules i n iteroparous species w i l l be effected p r imarily through a l t e r a t i o n s i n the age-specific a l l o c a t i o n of reproductive e f f o r t (Gadgil and Bossert 1971). In the present context, the higher mortality rate on adults caused by f i s h i n g pressure confers s e l e c t i v e advantage to increased reproductive e f f o r t e a r l i e r i n l i f e . Is there evidence that t h i s may occur? My r e s u l t s indicate support for the hypothesis of increased reproductive e f f o r t e a r l i e r i n l i f e as a mechanism to counteract increased adult S. alutus mortality. While the time since major perturbation i n the exploited stocks has been i n s u f f i c i e n t to demonstrate s e l e c t i o n i t i s c l e a r that younger f i s h i n these stocks do have higher reproductive e f f o r t r e l a t i v e to unexploited f i s h of the same age, i . e . there are conditions f o r s e l e c t i v e advantage based on reproductive e f f o r t . However, i t i s also evident that there are s i g n i f i c a n t differences i n the a l l o c a t i o n 143 of obtained energy between s i m i l a r f i s h from the two e x p l o i t a t i o n groups. I have shown that sizes at age are greater f o r f i s h from exploited stocks but that these f i s h had lower reproductive e f f o r t than unexploited f i s h of the same s i z e . This d i s p a r i t y i s evident i n the s i g n i f i c a n t differences (p<0.01) of the GIS index as a function of age and somatic weight between groups, which i s i n turn based on the s i g n i f i c a n t l y d i f f e r e n t (p<0.01) r e l a t i o n -ship of ovary weight to both somatic weight and somatic weight-age. I t i s of i n t e r e s t that the difference i n these regressions was not one of slope but of elevation. This would be predicted i f the groups used the same physiological process to transform obtained energy to gonadal products but d i f f e r e d i n the l e v e l of basal energy. While separate physiological processes between the stocks cannot be ruled out, previous genetic studies of the species (Tsuyuki et a l . 1968, Wishard et a l . 1980, Seeb 1986) showed no d i f f e r e n t i a t i o n of enzymes over very broad areas (Oregon-Alaska), and separate processes therefore seem u n l i k e l y . I f we accept that lowered density i n the exploited stocks y i e l d s increased energy a v a i l a b i l i t y for i n d i v i d u a l s then there appears to be a hierarchy of energy a l l o c a t i o n favouring the soma over the gonads. This suggests that for S. alutus the evolutio-nary advantage of a t t a i n i n g larger s i z e exceeds that for i n -creasing reproductive e f f o r t , e a r l i e r i n l i f e . Such a hierarchy i s common where increased s i z e confers e i t h e r a s u r v i v a l advan-tage or increased reproductive success (mating, o f f s p r i n g 144 qu a l i t y , f e r t i l i t y ) , or where reproduction c a r r i e s a s i g n i f i c a n t cost. Neither reproductive cost nor enhanced o f f s p r i n g q u a l i t y appear supported by t h i s study. Offspring q u a l i t y (as r e f l e c t e d i n oocyte weight) i s a l i n e a r function of ovary weight and these faster-growing f i s h have lower ovary weight, hence l i g h t e r oocytes. S i m i l a r l y , female mortality i s s l i g h t l y lower than for males (Archibald et a l . 1981). This leaves a s u r v i v a l advantage and/or increased reproductive success as the probable benefits of f a s t e r growth. I noted that longevity was associated with smaller ultimate s i z e i n unexploited stocks and our analysis of s i z e at age (Mul-l i g a n and Leaman, i n prep.) suggests that rates of mortality and growth are inversely correlated. I f t h i s inverse r e l a t i o n s h i p holds, then these faster-growing f i s h incur higher mortality and an ultimate s u r v i v a l advantage would not be r e a l i z e d . They are more fecund than f i s h of the same age due to some compensatory growth changes, and therefore do provide conditions f o r s e l e c t i v e advantage. Advantage may also take the form of increased s u r v i v a l by rapid passage through a window of s i z e - s p e c i f i c mortality, or of e a r l i e r attainment of the s i z e at which succes-s f u l reproduction i s enhanced (e.g. behavioural aspects of breeding). A l t e r n a t i v e l y , an ultimate s u r v i v a l advantage may be gained through the physiological mechanisms of maximizing growth rate or conversion e f f i c i e n c y (Boehlert and Yoklavich 1984b, Theilacker 1987). In the case of the exploited stocks, increased 145 food a v a i l a b i l i t y associated with decreased density may provide advantage to the growth rate maximizers and may also explain the lower reproductive e f f o r t on the part of these faster-growing f i s h , r e l a t i v e to other f i s h of the same s i z e . 4. Conclusions The conclusions of t h i s chapter with respect to the ques-tions posed i n the Introduction are as follows: (i) In a strongly iteroparous organism, does an increase i n adult mortality r e s u l t i n a lowering of the age at f i r s t maturity? The compensatory growth of younger f i s h i n the exploited populations, presumably i n response to lowered density and increased food a v a i l a b i l i t y , confirms the l a b i l i t y of age at f i r s t maturity. The difference i n age at f i r s t maturity for f i s h i n exploited stocks ranges from 1-3 y younger than those i n unexploited stocks. While t h i s provides opportunity for adaptive response to mortality e f f e c t s , the magnitude of the change i s only s l i g h t l y greater than the normal inter-cohort v a r i a t i o n i n s i z e at age. Resultant increases i n fecundity at age (10-15%) are small compared with the major decreases i n l i f e t i m e reproductive e f f o r t , associated with the truncation of the average l i f e span (25-35 y vs. 14-16 y) caused by t h i s mortality increase. The maximum compensatory response to e x p l o i t a t i o n w i l l therefore be very l i m i t e d . I t should be noted that s i z e at maturity i s e i t h e r developmentally or environmentally constrained, as the lowering 146 of age at maturity did not r e s u l t i n any change i n the s i z e at maturity. ( i i ) Associated with the lowered age at f i r s t maturity, w i l l there be an increase i n reproductive e f f o r t of younger f i s h , r e l a t i v e to the unexploited state? Reproductive e f f o r t for f i s h at age 12 averages only 3.3% more for f i s h from exploited stocks compared with those from unexploited. The maximum difference between age 12 f i s h i n the two groups i s 29%. However, the fecundity for a given s i z e f i s h i s lower for f i s h i n exploited stocks r e l a t i v e to those of the same s i z e from unexploited stocks. This difference relates to the a l l o c a t i o n of obtained energy between the gonads and the soma, and w i l l be treated i n point (iv) below. ( i i i ) Does increased reproductive e f f o r t at younger ages decrease the rate of s u r v i v a l to older ages? I was unable to examine t h i s question. I could not derive a method of determining previous reproductive e f f o r t and was therefore unable to associate i t with age-specific s u r v i v a l . Determination of such a r e l a t i o n s h i p i n t h i s species may not be possible. (iv) Does adaptation act so as to maximize residual reproductive value, rather than reproductive e f f o r t , i n any given year? The answer to t h i s question i s a q u a l i f i e d one. In a s t r i c t sense, I cannot answer any questions about adaptation with t h i s study, because I have not examined an evolutionary process. However, the responses to mortality do provide indications of what adaptations may be possible. The compensatory growth changes 147 associated with the density-independent f i s h i n g mortality on adults argue i n favour of maximizing reproductive e f f o r t . The adaptive response to such mortality schedules w i l l be to increase reproductive e f f o r t e a r l i e r i n l i f e ; the increased growth and e a r l i e r maturity of f i s h from exploited stocks c l e a r l y support t h i s process. However the discovery that f i s h of the same siz e are l e s s fecund i n exploited stocks suggests that growth i n younger f i s h may have a higher p r i o r i t y of energy a l l o c a t i o n than reproduction. I f t h i s i s true, then there i s a measurable residual reproductive value, and reproductive e f f o r t f o r a given s i z e i s not at a maximum. The r e l a t i o n s h i p of these two v a r i a -bles among years, for a given s i z e , remains to be determined. In addition to these findings concerning the questions posed i n i t i a l l y , I have also made several other conclusions concerning the reproductive biology of Sebastes alutus: (v) The form of growth d i f f e r s between the sexes with males ex h i b i t i n g a standard asymptotic form, while older (>40 y) females were generally shorter than f i s h aged 20-40 y . Examina-t i o n of the variance structure of female length at age suggests an inverse r e l a t i o n s h i p of growth and mortality rate. These smaller f i s h can be accounted for using the same growth form as the larger f i s h seen at younger ages (20-40 y), yet give r i s e to an appearance of a quadratic growth function. Determination of accurate growth relationships for such long-lived f i s h cannot be accomplished without size-at-age observations f o r s i n g l e cohorts 148 throughout t h e i r growth hist o r y . (vi) The volumetric method of fecundity estimation previously used for rockfishes i s i n f e r i o r to gravimetric methods, i n terms of the p r e c i s i o n of the estimate. The volumetric methods were highly s e n s i t i v e to subsample volume and were negatively biased. ( v i i ) S i g n i f i c a n t l y d i f f e r e n t relationships between fecundity and body variables among stocks and between e x p l o i t a t i o n groups e x i s t . ( v i i ) Oocyte c h a r a c t e r i s t i c s within stocks were highly variable, with low corr e l a t i o n s to single body variables. Oocyte diameter was inversely related to fecundity, independent of ex p l o i t a t i o n h i s t o r y . Oocyte diameter increased, but was highly v a r i a b l e with ovary weight. Oocyte qua l i t y (as expressed i n oocyte weight) varied s i g n i f i c a n t l y between ex p l o i t a t i o n groups. Fish from unexploited stocks had heavier oocytes than s i m i l a r f i s h from exploited stocks, effected through differences i n the oocyte diameter-oocyte weight re l a t i o n s h i p between groups. ( v i i i ) Maturation of i n d i v i d u a l f i s h takes approximately one year, and a stage designated as "maturing" i n the l i t e r a t u r e i s fu n c t i o n a l l y immature. Size frequency d i s t r i b u t i o n of oocytes i n the ovary showed that oocytes take one year to mature and that there i s no multiple spawning ( i . e . batch fecundity) of S. alutus i n B.C. waters. (ix) F o l l i c u l a r a t r e s i a was common i n mature ovaries; the r a t i o of atretic/mature ranged from 0.27-0.50, by stock. Differences among stocks were highly s i g n i f i c a n t but only marginally so among 149 e x p l o i t a t i o n groups. The major source of differences was a geo-graphic c l i n e , increasing north to south. A t r e t i c f o l l i c l e s are also suggested as an a l t e r n a t i v e energy source to embryo death, which has been presented as the source of matrotrophy for Sebastes spp. i n the l i t e r a t u r e . (x) Northern stocks had s i g n i f i c a n t l y larger mature oocytes than southern stocks, an e f f e c t not related to e x p l o i t a t i o n or f i s h s i z e . (xi) Ovaries sampled during the period of p a r t u r i t i o n provided the f i r s t evidence of complete a t r e s i a f o r the mature oocyte complement i n t h i s genus. This condition was noted only i n ovaries from an unexploited stock. Fish with normal embryonic development also showed evidence of a t r e t i c f o l l i c l e s (3-30% of t o t a l ) . The Rennell Sound stock showed the t y p i c a l vertebrate synchrony of oocyte development but development of oocytes i n the Goose Island Gully stock was more asynchronous. ( x i i ) The hypothesis of increased reproductive e f f o r t with age was confirmed, including the independence of s i z e and age e f f e c t s . ( x i i i ) The hypothesis of congruent i n d i v i d u a l fecundity and age d i s t r i b u t i o n s was not supported by t h i s study. However, t h i s lack of congruency i s less evident when cohort fecundity i s considered. Strong v a r i a t i o n i n cohort strength w i l l act to make average age and cohort fecundity d i s t r i b u t i o n s more congruent, over long time periods. (xiv) The hypothesis of increasing reproductive cost with age was 150 examined through scrutiny of mortality rates and gonadal indices. The age composition for females did not indicate higher mortality rates than f o r males. The lack of detectable change i n the two gonadal indices with age argues that gonads are not developed at the expense of somatic resources. (xv) Neither oocyte nor gross gonadal c h a r a c t e r i s t i c s showed any evidence of reproductive senescence. 151 V. REPRODUCTIVE VALUE MODELLING AND IMPLICATIONS FOR MANAGEMENT3 1. Introduction Trawl f i s h e r i e s for demersal f i s h have been a component of world f i s h e r i e s for over 100 y but trawl catches of Sebastes spp. are a r e l a t i v e l y recent phenomenon. Most major f i s h e r i e s for these species i n the northern hemisphere have had t h e i r genesis within the past 30 y, some only within the l a s t decade. While these f i s h e r i e s have been based on several d i f f e r e n t Sebastes spp., t h e i r h i s t o r i e s have been remarkably s i m i l a r . In every instance of a major ro c k f i s h trawl fishery, the progress of the unrestrained fishery has been several years of r e l a t i v e l y high catches, based on accumulated biomass, preceding a rapid decline and subsequent years of very low catches (Fig. 33, Leaman 1987a). In the Gulf of Alaska for example, the peak catch of Sebastes alutus was 348,000 t i n 1965, subsequently declined to only 45,000 t by 1970, and i s now less than 2% (approximately 5,000 t) of the peak l e v e l (Balsiger et a l . 1985). In many instances the o r i g i n a l p a r t i c i p a n t s i n these f i s h e r i e s abandoned them when catch rates f e l l to an uneconomic l e v e l . The unquestioned cause of these declines was f i s h i n g mortality f a r i n excess of l e v e l s which might have produced sustainable f i s h e r i e s . In a l l cases present catches are less than 10% of maximum l e v e l s . 3 P o r t i o n s of t h i s section have been published as Leaman (1987a). 152 , , ! ! f I960 1965 1970 1975 1980 1985 Y E A R Figure 3 3 . Catch h i s t o r i e s of several P a c i f i c and A t l a n t i c rockfish f i s h e r i e s . 153 I have recently argued that there has never been a succes-s f u l management program for a major r o c k f i s h stock, and i n p a r t i c u l a r no examples of successful recovery from overexploi-t a t i o n (Leaman 1987a). Where we see some evidence of increased catches a f t e r major declines (e.g. Gulf of St. Lawrence r e d f i s h ) , i t i s not the r e s u l t of a directed management action so much as the appearance of the progeny of cohorts that were present i n the fishery, p r i o r to management action and major increases i n f i s h i n g mortality. In most instances these cohorts were subse-quently eliminated by the fishery. In some cases, t h i s lack of management success may be a t t r i -buted to a h i s t o r i c a l absence of management authority i n the areas of the f i s h e r i e s . However, a change i n our understanding of the underlying biology of these species has also contributed to a d i f f e r e n t perception of appropriate management (Beamish 1979, Archibald et a l . 1981). Their l i f e span, growth rate and l a t e recruitment («12 y) to the fishery combine to create long periods over which large changes may go undetected, as well as long response times to management measures. Some of the reason for f a i l u r e may also be that the indices normally monitored as indicators of stock status (e.g biomass) are i n s e n s i t i v e to those features governing the evolutionary and commercial persistence of these species. The major r e s u l t s of Archibald et a l . (1981) were decreases 154 of over 50% in.estimated natural mortality rates, and a doubling of estimated l i f e spans. Another r e s u l t was that the change i n ageing technique produced almost no change i n the estimated parameters of growth. This occurred because the change i n age composition affected primarily those i n d i v i d u a l s who had already reached a large proportion of t h e i r ultimate s i z e , even when estimated using previous ageing techniques. Archibald et a l . considered t h i s finding an almost minor r e s u l t , yet i n some ways i t i s equally as dramatic as the mortality rate changes, for i t implies that many species of r o c k f i s h grow very l i t t l e over almost h a l f of t h e i r l i f e span, or more. More importantly, because the f i s h reproduce throughout t h e i r adult l i v e s , i t changes the dominance of the l i f e cycle from somatic to gonadal production. This change should therefore require a change i n the a n a l y t i c a l approach to r o c k f i s h population dynamics, and implies that consideration of l i f e t i m e reproductive e f f o r t should occupy a more prominent p o s i t i o n . The general approach to incorporating reproduction into population dynamics models has been to formulate a r e l a t i o n s h i p between adult spawners and reproduction. This may happen eith e r d i r e c t l y , where adult spawners (biomass or numbers) are linked to reproductive output, or i n d i r e c t l y where spawners are linked to r e c r u i t s . The attractiveness of such stock-recruitment r e l a t i o n -ships i s that they permit prediction of future values for recruitment. While there are a number of technical problems with 155 such re l a t i o n s h i p s (e.g. density e f f e c t s , measurement pr e c i s i o n ) , they can often be overcome (Cushing 1977, Gulland 1983), although not always (Walters 1986). What cannot be overcome or ignored i s that the basis for such relationships i s neither more nor less than a c o r r e l a t i o n analysis between, c e r t a i n l y i n the case of rockfishes, two highly variable quantities. Its v a l i d i t y rests on the accuracy, robustness and contrast i n the observations from which i t i s constructed. In almost every instance f o r r o c k f i s h stocks, these observations are from an extremely rapid, one-way trend i n stock s i z e of adults, where there are few r e p l i c a t e observations, except at low abundance. Incorporating reproductive considerations into models of semelparous species has been r e l a t i v e l y straightforward and has led to the concept 'replacement stock 1 (e.g. Ricker 1975). This concept implies a management p o l i c y of ensuring s u f f i c i e n t spawning biomass or numbers for the cohort to reproduce i t s e l f i n the next generation. For iteroparous species, t h i s concept has no analogue within any given year. Instead, these species have evolved to take advantage of repeated spawnings to achieve the same objective. One component of t h e i r 'replacement stock' i s therefore t h e i r t o t a l l i f e t i m e reproductive output. The other component i s the fact that multiple cohorts spawn together, so that t o t a l reproductive output i n any year i s the sum of variable reproductive e f f o r t from animals of d i f f e r e n t ages (hence sizes) and reproductive capacity. Clearly, both i n d i v i d u a l and demogra-156 phic c h a r a c t e r i s t i c s contribute to t h i s replacement concept for iteroparous species. Ideally, one would l i k e to incorporate reproduction by employing an index that r e f l e c t s a combination of age-specific reproductive e f f o r t and i t s contribution to o v e r a l l stock reproduction, i . e . i t s reproductive value (RV). My approach to investigating t h i s problem was to examine a l t e r n a t i v e RV indices with a simulation model of stock dynamics and f i s h i n g mortality. The r e s u l t s of the previous two chapters concerning the stock i d e n t i t y and reproductive biology of Sebastes alutus were employed i n t h i s examination. 2. Model description and parameter estimation A simulation model was constructed to: examine the reproduc-t i v e performance of a cohort throughout i t s l i f e ; determine population responses to both e x p l o i t a t i o n pressure and r e c r u i t -ment uncertainty; and, evaluate several indices of reproductive value as measures of stock condition. In i t s ultimate form i t i s structured around the standard exponential mortality equations and a stochastic stock-recruitment r e l a t i o n s h i p . I t i s composed of simple growth, mortality and recruitment modules, driven by a cycle and loop c o n t r o l l e r . The major control variables are f i s h i n g and natural mortality rates together with the variance of the stock-recruitment re l a t i o n s h i p . Table 25 contains a l i s t of notation as well ,as the source of the parameters and r e l a t i o n -ships employed i n the model. A description of the key processes 157 Table 25. Notation and sources for parameters used i n simulation models. Index values a age at which reproductive value calculated i year index j age index t time index Assumed values k Brody growth c o e f f i c i e n t LQO Maximum asymptotic length M instantaneous natural mortality PVj p a r t i a l v u l n e r a b i l i t y of age j f i s h to f i s h i n g mortality RDWTj round weight at age j t 0 t h e o r e t i c a l time at which length i s zero V variance of stock-recruitment a,j3 parameters of Ricker stock-recruitment r e l a t i o n s h i p Source t h i s study t h i s study Archibald et a l . (1981) Archibald et a l . (1983) t h i s study t h i s study Archibald et a l . (1983) Archibald et a l . (1983) Calculated values C-LJ catch of age j f i s h i n year i Ej fecundity of age j f i s h FECi t o t a l stock fecundity i n year i N-^ j number of age j f i s h i n population i n year i RV reproductive value Z instantaneous t o t a l mortality Control v a r i a b l e F instantaneous f i s h i n g mortality 158 and the estimation of t h e i r parameters follows: a) Growth. Size at age t i s described by (i) the t r a d i t i o -nal von Bertalanffy negative exponential, and ( i i ) f o r a l i m i t e d number of cases, a quadratic function. (i) L t = L*, * (l-EXP(-K(t-t 0))) ( i i ) L t = A + Bt + C t 2 The parameters were estimated by non-linear regression of lengths-at-age for each stock, using Schnute's generalized formula i n the case of the von Bertalanffy function. Although the i n v e s t i g a t i o n of growth has been more comprehensive than these simple p r e d i c t i v e functions (Mulligan and Leaman, i n prep.) the von Bertalanffy function was used i n the model because I believe i t i s appropriate for the growth of i n d i v i d u a l s . Growth was assumed to be density-independent i n the model, although the s e n s i t i v i t y of the r e s u l t s to growth changes was investigated. b) Instantaneous natural mortality (M) . The problems of following i n d i v i d u a l s of t h i s deep-water species over time make the estimation of natural mortality imprecise. Also contributing to t h i s imprecision i s the v a r i a b i l i t y i n age estimation and the f a c t that the ageing technique i s unvalidated. Two approaches to the estimation of M were taken: (i) estimating M as minimum Z (instantaneous t o t a l mortality) from analyses of catch curves from l i g h t l y exploited stocks (Archibald et a l . 1981) and ( i i ) estimating M a n a l y t i c a l l y from a detailed catch-at-age analysis (Archibald et a l . 1983). The f i r s t method provided a mean estimate of M=0.05; the second showed l i t t l e s e n s i t i v i t y to the estimate of M as long as i t was <0.1. Balsiger et a l . (1985) and Kimura (pers. comm.) have provided additional support for an estimate of M=0.05. The incorporation of M into t o t a l mortality i s described below. 159 c) Fishing mortality. Fishing mortality i s the primary control v a r i a b l e i n the model and i s incorporated v i a the Baranov catch equation (Ricker 1975): where i and j are year and age indices respectively, Cj^j i s the catch of f i s h aged j i n year i , N^j i s the t o t a l number of f i s h aged j i n year i , F^ i s f i s h i n g mortality i n year i , PVj i s p a r t i a l a v a i l a b i l i t y of age group j to F^, and Z-^  i s t o t a l mortality i n year i . PV and M are treated as time-invariant. Total catch within year i s simply the summation over j of the above; catch i n weight i s calculated by multiplying C-^ j by the appropriate weight at age j . Annual changes i n the cohort s i z e or number are described as i f f i s h i n g and natural mortality occur throughout the year: i+l'j+1 l j e = N * e ( - F i * P V ^ - M > > ID Population changes are the summation of the above over j . A var i a b l e i n i t i a l period of F^=0 can be incorporated into the model. d) Recruitment. Recruitment i s driven by a Ricker stock-recruitment r e l a t i o n s h i p with stock expressed as t o t a l fecundity 160 of spawners and recruitment as the number of age 6 f i s h , s i x years l a t e r . The r e l a t i o n s h i p i s s l i g h t l y modified to express fecundity as r e l a t i v e to age 8 f i s h , thus: FEC.= J (N^MEj/Eg)) The Ej values were obtained through multiple l o g - l i n e a r regression of round weight (RDWT) and age, with weight being generated from the length-weight rela t i o n s h i p , thus: log (E.) = a + b log(RDWT.) + b log(j) J i J ^ fo r each stock. The stock-recruitment r e l a t i o n s h i p was further modified to incorporate a m u l t i p l i c a t i v e , log-normal (0, a 2) error term which serves to emulate the observed sequence of v a r i a t i o n i n cohort strength. Therefore: -fi*FECi V N. - = a * FEC. * e p * e 1+6 l The Ricker parameters a and p as well as the variance (V) of the stock-recruitment r e l a t i o n s h i p were estimated for the Goose Island Gully stock using the catch-at-age model of Archibald et a l . (1983), and these values used for a l l stocks. The sequence of i n d i v i d u a l values of V was generated with the International Mathematics and S t a t i s t i c s Library random number generator, GGNML. 161 e) Reproductive value: Two measures of reproductive value were investigated, age-specific reproductive value (RV) (after Vandermeer 1968) and eventual reproductive value (ERV) (after Goodman 1967). The f i r s t quantity i s a measure of the expecta-t i o n of future young, at a given age, for the e x i s t i n g mortality schedule. For t h i s model i t was calculated as: m RV = 2 (N.*E.) j=a J J where a = age for which reproductive value estimated j = age index m = maximum age * The value a was chosen as age 12 so that only cohorts for which PVj>0.5 were included. Eventual reproductive value was presented by Goodman (19 67) as a more appropriate measure of reproductive value for mixed-age spawners, since the expected future reproduction i s normalized by the r e c i p r o c a l of the generation time. The generation time i s taken to be the average age of spawning females j . m RV a = (S (N,*E.)/(n) j=a J J m = (2 (N.*E.)/((SN.*j)/2N.)) j=a J J j J j J Schaffer (1974a) introduced the concept of modified repro-ductive value i n an attempt to deal with the p o s s i b i l i t y of reproductive cost. He established schedules of incremental 162 fecundity at age wherein the increments were e x p l i c i t l y indepen-dent of previous reproductive e f f o r t . Most studies e i t h e r ignore p o t e n t i a l reproductive cost or i m p l i c i t l y assume i t i s zero. For iteroparous species t h i s assumption i s often a pragmatic neces-s i t y when reproductive e f f o r t for i n d i v i d u a l s cannot be assessed at each reproductive episode. Fecundity schedules calculated as I have done, from simultaneous samples of d i f f e r e n t age groups, incorporate e f f e c t s of p r i o r reproductive expenditures whose influence ( i f any) w i l l be unknown. The indices I examined make no e x p l i c i t accounting for pot e n t i a l reproductive costs, however any e f f e c t s w i l l be i m p l i c i t i n the fecundity schedules esta-blished. f) Dimensions, r e p l i c a t i o n , c y c l i n g , and s e n s i t i v i t y : T h e model population i s presently dimensioned to 80 age groups, a f t e r which, i n terms of contributions to eit h e r the f i s h e r y or popula-t i o n , i n d i v i d u a l s pass into an absorbtion c e l l . The basic dura-t i o n of the simulation i s 100 y and can be incremented i n blocks of 100 y. Calculations are r e p l i c a t e d 30 times to incorporate v a r i a t i o n i n the stock-recruitment r e l a t i o n s h i p assigned by the random number generator, within boundaries of the user-specified V (the variance of the stock-recruitment r e l a t i o n s h i p ) . To estimate the r e s u l t s of fixed f i s h i n g mortality p o l i c i e s , the simulations were continued u n t i l the population s t a b i l i z e d , usually taking 80-100 y. 163 The s e n s i t i v i t y of the model r e s u l t s to changes i n the input parameters was examined with a f i n i t e - d i f f e r e n c e c a l c u l a t i o n , using 0.1 as the perturbation (5). The r e l a t i v e s e n s i t i v i t y of the r e s u l t function, f, (biomass or yield) to v a r i a t i o n s i n parameter bj[ was estimated as: f(bi+«bi) - f( b i ) Si= fS or i n the case of vector inputs (e.g. p a r t i a l recruitments, or number at age N(j)) f (e 1+6e 1,e 2+«Se 2. . .) - f ( e 1 , e 2 . . . ) Si= f<S where 6" i s constant over the range e^ ... e n (after Rivard 1982). These s e n s i t i v i t y c o e f f i c i e n t s are approximations of the p a r t i a l derivatives of the y i e l d and biomass functions, with respect to the input parameters. They measure the r e l a t i v e change i n estimated y i e l d and biomass associated with given changes i n the values of the input parameters. A s e n s i t i v i t y c o e f f i c i e n t of 1.0 means that the r e s u l t function changes by the i d e n t i c a l proportion as that of the input parameter. S i m i l a r l y , a value of -0.5 means that the r e s u l t function decreases by one-ha l f of the proportion that input parameter was increased by. The c o e f f i c i e n t s are d i r e c t l y comparable i n that they represent the same r e l a t i v e change for the various inputs. 164 2. Results a) Single cohort version E x p l o i t a t i o n e f f e c t s on cohort reproductive characters were investigated for an a r b i t r a r y cohort strength (10 6 f i s h ) . Figure 34 d e t a i l s these e f f e c t s for the von Bertalanffy growth function. The growth characterization, while s a t i s f a c t o r y for in d i v i d u a l s , i s not completely s a t i s f a c t o r y for the cohort because of the d i s t r i b u t i o n of residual error i n describing the data. The net e f f e c t , because of the d i s t r i b u t i o n of abundance with age, w i l l be to underestimate cohort reproduction with t h i s uniform growth function. In opposition to t h i s bias, decrements i n reproductive characters associated with increases i n F w i l l be s l i g h t l y overestimated. The exact magnitude of these e f f e c t s i s d i f f i c u l t to estimate because the previous growth h i s t o r y (or growth •type 1) of in d i v i d u a l s i n the fecundity samples i s unknown. Table 27 i l l u s t r a t e s the s h i f t s i n reproductive characters of an a r b i t r a r y cohort with (F=0.05) and without f i s h i n g mor-t a l i t y , for the asymptotic growth form, by stock. In the unex-p l o i t e d state cohort weight maximizes at approximately 14 y and cohort fecundity approximately two years l a t e r . Application of even t h i s low l e v e l of F decrements the ages of these maxima by 4.5 y and 3.4 y, respectively. Maximum cohort weight and fecun-d i t y decrease by an average 16 and 23%, with the greatest changes being recorded for the l i g h t l y exploited stocks. 165 5. alutus LANGARA AGE Figure 3 4 . Cohort fecundity and reproductive values at age for two l e v e l s of f i s h i n g mortality, simulated using the population parameters of the Langara Spit S. alutus. NQ i s the i n i t i a l size of the cohort. 166 Table 27. Changes i n cohort reproductive c h a r a c t e r i s t i c s with F=0.0 and F=0.05, by stock, for S. alutus with quadratic growth. Stocks are Langara Spit (LA), Rennell Sound (RSD), Moresby Gully (MOR), Goose Island Gully (GIG), and Vancouver Island (VI). Stock F Max. Age at RV Max. Age at RV at Age of Fecundity (1) at COFEC (2) (2) Max. COWT (1) (1) (2) LA 0. 00 326857 46 619 99. 936 13 3173. 5 13/14 0. 05 326857 46 7 96. 877 7 1304. 6 10 RSD 0. 00 294227 46 580 117. 567 7 3808. 2 14 0. 05 294227 46 7 117. 191 6 1549. 8 11 MOR 0. 00 464022 49 760 136. 961 15 4190. 3 14 0. 05 464022 49 5 126. 098 9 1518. 5 10 GIG 0. 00 301030 44 632 99 . 416 13 3030. 4 14 0. 05 301030 44 9 93. 928 9 1189. 2 10 VI 0. 00 293224 45 823 115. 649 12 3238. 7 14 0. 05 293224 45 25 109. 467 9 1243. 8 10 COFEC = cohort fecundity x 10 6 eggs COWT = cohort weight RV = reproductive value x 10 6 eggs 167 Changes i n reproductive value at age are f a r more dramatic. Introduction of 5% f i s h i n g mortality causes mean reproductive value reductions of: 96% at the age of maximum in d i v i d u a l fecundity, 68% at the age of maximum cohort fecundity, and 67% at the age of maximum cohort weight. For comparison only, the decrements of reproductive value at the ages of maximum in d i v i d u a l fecundity and maximum cohort fecundity averaged 98% and 61%, respectively, when a quadratic growth function was used. The ages at which these maxima occur are 2-9 y below those for the asymptotic growth function. The r e s u l t s f o r the Rennell Sound stock best i l l u s t r a t e the problems of the quadratic growth characterization, since the ages of the maxima are c l e a r l y u n r e a l i s t i c . These r e s u l t s led me to abandon further use of t h i s growth characterization. Only a small set of evaluations were conducted with the sing l e cohort version of the model and most e f f o r t was devoted to the multiple cohort version. b) Multiple cohort version: Figures 35 and 3 6 afford some understanding of the r e l a t i v e s e n s i t i v i t i e s of the simulations, a f t e r a 100 y period, to perturbations of the input variables. Fishing mortality was the control v a riable, at l e v e l s of F=0.01-0.10. The inherent b i o l o g i c a l structure and time lag of t h i s simulation model mimic Sensitivity coefficients I. Yield lOOy horizon .01 .02 .03 .04 .05 .06 .07 .08 F level Weight Fecundity Partial recruitment Stock recruitment Initial recruitment Recruitment variance Initial N(j) j Natural mortality Figure 35. S e n s i t i v i t y of predicted stock y i e l d to perturbations i n the input parameters for the multiple cohort simulation model, af t e r a period of 100 y. t horizon Weight Fecundity I Partial recruitment i Stock recruitment Initial recruitmeni Recruitment variance Initial N(j) Natural mortality -5 H 1 1 1 1 1 1 1 i i .01 .02 .03 .04 .05 .06 .07 .08 .09 .1 F level Figure 36. S e n s i t i v i t y of predicted stock biomass to perturba-tions i n the input parameters for the multiple cohort simulation model, afte r a period of 100 y. Sensitivity coefficients - 100^ II. Biomass 10 i 170 that of the natural population. Since the f i r s t age at which p a r t i a l recruitment i s >0.5 i s 12 y, there i s l i m i t e d opportunity for the feedback control mechanisms of the model to assert themselves over time periods of <30 y. Over a 10 y period, i n i t i a l values have l i t t l e influence at F >0.3 and above F=0.6 no parameters have s e n s i t i v i t i e s >1.0. The i n i t i a l age vector has almost no influence on y i e l d . At periods >20 y the s e n s i t i v i t i e s of biomass and y i e l d to perturba-tions of the input parameters begin to overcome transient e f f e c t s and show more stable patterns, with stock-recruitment and fecundity showing more p o s i t i v e influence on y i e l d and biomass. P a r t i a l recruitment and natural mortality have negative sensi-t i v i t y c o e f f i c i e n t s for y i e l d and biomass, and are approximately twice the input perturbation over a 100 y period (Figs. 35 and 3 6) . Changes i n variance of recruitment and weight at age are roughly equivalent i n t h e i r e f f e c t on r e s u l t s and close to 1.0. The s e n s i t i v i t y of biomass to perturbations of fecundity at age and natural mortality increases with increasing F, while the s e n s i t i v i t y to other parameters i s l a r g e l y independent of F. The s e n s i t i v i t y of biomass and y i e l d to the parameters of the stock-r e c r u i t r e l a t i o n s h i p (a,/3) f i r s t increases, then decreases at l e v e l s of F>0.08. Forward simulations are most se n s i t i v e to estimates of fecundity at age, stock-recruitment parameters, the p a r t i a l 171 recruitment vector and natural mortality. The e f f e c t s of the f i r s t two are d i r e c t l y proportional and the l a s t two inversely proportional. Y i e l d estimates are sharply increased by increases i n fecundity at age and the parameters of the stock recruitment r e l a t i o n s h i p , p a r t i c u l a r l y at higher F values (Fig. 35) . In contrast, increases i n p a r t i a l recruitment values and natural mortality decrease estimates of available y i e l d and biomass. The e f f e c t of increasing weight at age (e.g. the e f f e c t of compen-satory growth) on y i e l d or biomass i s s t r i c t l y proportional to the change i n the input. This e f f e c t i s quite small compared with increasing fecundity at age, for example, where the sensi-t i v i t y can be up to f i v e times the input at higher f i s h i n g m o r t a l i t i e s . In summary, only four of the input parameters (or vectors) have appreciable influence on estimated biomass or y i e l d . Of these four, I have reasonable confidence i n three: natural mor-t a l i t y , p a r t i a l recruitment and fecundity at age. The parameters of the fourth, the stock recruitment r e l a t i o n s h i p , are not well determined because the parent stock does not have a strong r e l a t i o n s h i p to subsequent recruitment. Increases i n these parameters w i l l therefore have disproportionate influence since they act to strengthen the re l a t i o n s h i p . I n i t i a l recruitment, numbers at age, and recruitment variance have r e l a t i v e l y l i t t l e influence on estimates of long-term (>100 y) y i e l d and biomass. 172 The f u l l model with stochastic v a r i a t i o n s i n the stock-recruitment r e l a t i o n s h i p was used to examine y i e l d and reproduc-t i v e value responses to a suite of F values by stock. Figures 35-37 i l l u s t r a t e some of the y i e l d , biomass and RV responses to f i s h i n g mortality l e v e l s at the equilibrium l e v e l s (200 y duration). Stock s i z e at F=0.0 varies from approximately 73,000-93,000 t among stocks, with Rennell Sound having the lowest and Vancouver Island the highest estimated unfished biomass. The difference i n these maxima i s consistent with the avai l a b l e habitat i n each l o c a l i t y , the Rennell Sound stock inhabiting a narrow offshore bank while the Vancouver Island stock occupies a broad coastal slope. Maximum equilibrium (200 y) y i e l d was achieved at F=0.06 for a l l stocks except Moresby Gully, where the maximum occurred at F=0.07. These maxima varied between 1050-1620 t among stocks and, with the exception of the Moresby Gully stock, followed the same pattern as that of the equilibrium unexploited biomass. Changes i n RV i n response to F mirror those of biomass (r2=0.99) but are consistently greater (X=10.1%, S.E.=0.59) at each l e v e l of F. The buffering e f f e c t of cohort v a r i a t i o n i s c l e a r l y evident when comparing the single and multiple cohort versions of the model. For the former, introduction of f i s h i n g mortality at F=0.05 reduces RV by 72% from the unexploited value. However, the multiple cohort version predicts a decrease of only 42% for the same F. These values are RV for the e n t i r e cohort at age 12, Reproductive value 200-y stochastic simulations G - RV12 ERV12/10 IRV12 * - IERV12*10 ( ! ! ( ! , 0 .02 .04 .06 .08 .1 .12 Fishing mortalit)7 Figure 37. Response of several reproductive value (RV) indices (at age 12) to fis h i n g mortality, for a composite stock of S. alutus. Values averaged over the l a s t ten years of 200-y simula-tions. See text for d e f i n i t i o n of RV indices. 174 not those for individuals, and r e f l e c t equilibrium mortality schedules. Several other indices of reproductive value were examined for t h e i r s e n s i t i v i t y to population changes: RV/individual (IRV); cohort eventual RV (ERV); and eventual RV/individual (IERV). The eventual reproductive value indices are more appropriate for mixed-age spawners (Goodman 1967). Cohort EVRV also declines with increasing F although the r e l a t i v e magnitude of the decline i s less than that of cohort RV fo r the same F (Table 28, F i g . 37). Both measures of i n d i v i d u a l RV and ERV are s e n s i t i v e to the stochastic v a r i a t i o n i n cohort strength. The values plotted i n F i g . 37 are for the l a s t year of 2 00-yr simulations and r e f l e c t the v a r i a t i o n of the stochastic process. When indices are averaged over several years of the simulations they exhibit more stable patterns (e.g. Table 29) . The i n d i v i d u a l reproductive value indices, while showing more contrast between each other than the cohort indices, are less s e n s i t i v e to the changes induced by F. The hierarchy of sensi-t i v i t y i s thus RV, ERV, IRV, and IERV. These indices are not Fisher's t r a d i t i o n a l reproductive value index, which i s decreased by the p r o b a b i l i t y of s u r v i v a l to each age. That index responds i n a s i m i l a r fashion but due to the low natural mortality rate, i s a s l i g h t l y convex function of 175 Table 28. Reproductive value indices, biomass, y i e l d , and mean age changes with f i s h i n g mortality (F), over 200 y horizon, for a simulated S. alutus population. F Biomass Y i e l d Mean RV 1 2 ERV 1 2 I R V i 2 IERV 1 2 (t) (t) age 0. 00 95520 0 24. 01 1. 40620 58. 5713 5. 8815 0. 2449 0. 02 74110 750 19. 63 0. 97715 49. 7763 3. 8914 0. 1982 0. 03 62600 1010 18. 54 0. 79714 42. 9876 4. 0823 0. 2201 0. 04 58250 1290 17. 23 0. 70960 44. 1939 3. 1665 0. 1837 0. 05 47600 1260 16. 22 0. 54336 33. 5050 2. 7606 0. 1701 0. 06 41580 1330 15. 49 0. 46437 29. 9870 3. 1147 0. 2010 0. 07 35760 1300 14. 73 0. 37836 25. 6826 2. 3062 0. 1565 0. 08 29020 1150 14. 17 0. 29137 20. 5625 2. 3654 0. 1669 0. 10 20420 930 13. 16 0. 18699 14. 2106 1. 9453 0. 1478 0. 12 13510 680 12. 56 0. 11679 9. 3001 2. 0825 0. 1658 RV = cohort reproductive value ERV = cohort eventual reproductive value IRV = i n d i v i d u a l reproductive value IERV = i n d i v i d u a l eventual reproductive value Table 29. Reproductive value indices at age 12 for S. alutus as a function of f i s h i n g mortality (F) , averaged over the l a s t 10 years of 200-yr simulations. RV 1 2 ERV 1 2 IRV i 2 IERV 1 2 0. 00 1. 4074 58. 8031 5. 8509 0.2436 0. 02 0. 9667 49. 0812 4. 4152 0.2249 0. 03 0. 8023 43 . 5920 3. 7185 0.2005 0. 04 0. 7145 41. 9375 3. 3448 0.1941 0. 05 0. 5550 34 . 0392 2 . 9950 0.1846 0. 06 0. 4652 30. 2995 2. 7085 0.1748 0. 07 0. 3776 25. 7908 2 . 9378 0.1589 0. 08 0. 2981 21. 0985 2. 3157 0.1633 0. 10 0. 1892 14. 2698 1. 9688 0.1495 0. 12 0. 1180 9. 3824 1. 6914 0.1346 176 age, with a maximum at age 34 for F=0.0. The index value at the age of f u l l recruitment (in the context of t h i s study, 12 y) i s 83% of the maximum. With F=0.05 the maximum of the discounted index occurs at age 30 and i s reduced by 60% from the unexploited maximum, thus mirroring the changes i n F i g . 37. 3. Discussion E a r l i e r I noted the poor h i s t o r i c a l record of managing ro c k f i s h f i s h e r i e s . Several authors (Francis 1986, Leaman and Beamish 1984, Getz et a l . 1987) have argued that the most d i f f i c u l t problem i n managing f i s h e r i e s based on such age-structured species i s determining the r e l a t i o n s h i p between short-term y i e l d and long-term v i a b i l i t y . The source and s i g n i f i c a n c e of the enormous v a r i a t i o n i n cohort strength within these species i s s t i l l very much uncertain. Some authors take the view that a stock can sustain higher e x p l o i t a t i o n rates when these strong cohorts appear i n the fishery (McFarlane et a l . 1985) while others suggest that they should be treated as normal and quite necessary features of the 'replacement stock 1 concept, and not be subjected to higher than average e x p l o i t a t i o n rates (Hightower and Grossman 1986, 1987, Leaman 1987a). Getz et a l . (1987) have noted that the p o l i c y of removing more of the stock when biomass increases through recruitment v a r i a t i o n ('exploitative' policy) tended to reduce biomass more rapidl y than other p o l i c i e s and also increased the r i s k of fishery collapse. The presence of the 177 1952 cohort as the dominant cohort i n the Moresby Gully stock emphasizes the contribution to population c h a r a c t e r i s t i c s made by these occasional strong cohorts. The y i e l d p o t e n t i a l and r e s i l i -ence of these stocks to f i s h i n g mortality i s proportional to t h i s v a r i a t i o n i n cohort strength (Archibald et a l . 1983, Hightower and Grossman 1986). Most studies which examine the problem of r e h a b i l i t a t i o n of depleted stocks have suggested that target harvest p o l i c i e s , rather than target spawning stock or target biomass, w i l l be economically optimal when recruitment v a r i a b i l i t y i s large, unless c a p i t a l can be rapidly transferred into other revenue-producing a c t i v i t i e s (Getz et a l . 1987, Hightower and Grossman 1987). The l a t t e r i s u n l i k e l y for r o c k f i s h f i s h e r i e s because the c a p i t a l investment i s large ($1-2 million) and tends to highly s p e c i a l i z e d ( i . e . non-mobile). Unfortunately, the recovery time fo r depleted stocks i s also longest with a target harvest p o l i c y , because i t w i l l not permit closure of f i s h e r i e s when recruitment i s poor. Recovery i s most rapid when a target spawning stock i s s p e c i f i e d and the fishery closed u n t i l t h i s target i s reached (Archibald et a l . 1983, Hightower and Grossman 1987). A weakness of most studies, my own included, on the r e l a -t i o n s h i p of long and short-term productivity i s that the model-l i n g of recruitment v a r i a t i o n can only be done on the basis of the r e l a t i o n s h i p as perceived from h i s t o r i c a l data. The v a r i -178 ation i s generally described by some stochastic stock-recruitment function based on observations of these two variables, and either e x p l i c i t information or i n t u i t i o n about other environmental or b i o l o g i c a l features influencing the re l a t i o n s h i p . While the h i s t o r i c a l recruitment series may be reproduced with some pre c i s i o n , most r o c k f i s h stocks are now at quite d i f f e r e n t l e v e l s of abundance and demographic composition than those which gave r i s e to the time s e r i e s . I know of no studies f o r such species where a s u f f i c i e n t time series of stock-recruitment data i s availa b l e , p a r t i c u l a r l y involving r e p l i c a t e observations at the extremes of stock l e v e l s , to permit v a l i d a t i o n of the postulated r e l a t i o n s h i p . How w i l l t h i s problem be solved? Unfortunately, there does not appear to be a simple solution. The p r i n c i p a l d i f f i c u l t y i s that to which I referred i n the Introduction to the t h e s i s ; the s i m i l a r i t y i n the l i f e spans of these f i s h and the b i o l o g i s t s who study them. This s i m i l a r i t y leads to a lengthy period over which understanding w i l l be gained. A necessary component of t h i s learning process w i l l be to monitor stock status with indices which r e f l e c t the contribution of ind i v i d u a l s to long-term stock v i a b i l i t y . The work I have presented on reproductive value indices may help i n t h i s development. I t i s possible that manipulative experiments examining p a r t i c u l a r segments of the stock-recruitment r e l a t i o n s h i p may 179 provide some of the answers (Walters 1986). I have generated such experiments involving pulsed-overfishing and open-fishing on the Langara Spit and Vancouver Island stocks (Leaman 1985, 1987b), however there can be severe d i f f i c u l t i e s when t h e i r execution and inter p r e t a t i o n depends on complete cooperation from a user group, who may not perceive the experiment as being i n t h e i r best i n t e r e s t s (Leaman and Stanley, unpubl.). Data qua l i t y and quantity can often be highly variable and al t e r n a t i v e , compulsory measures carry t h e i r own set of data q u a l i t y problems. Large-scale manipulative experiments remain an area of active i n t e r e s t and t h e i r p o t e n t i a l has not yet been f u l l y explored. From a management perspective, the routine monitoring of reproductive value indices affords an opportunity to obtain more se n s i t i v e feedback than more commonly used indices, such as CPUE. For example, both here and elsewhere (Archibald et a l . 1983) I have estimated the l e v e l of biomass generating long-term maximum y i e l d as «47% of the unfished biomass. The cohort reproductive value index would have declined to approximately 3 3% of the unfished value, for the same l e v e l of biomass. I t i s u n l i k e l y that CPUE would decline i n s t r i c t proportion with the decline i n biomass ( p a r t i c u l a r l y at low stock levels) due to the aggregating nature of t h i s species and consequent v u l n e r a b i l i t y to trawling (Leaman 1987b), whereas reproductive value indices w i l l continue to r e f l e c t such stock changes. I t i s important to employ such indices because of the long response time which i s necessary for 180 correction of errors i n management p o l i c y . In t h i s regard, the cohort-based reproductive value indices w i l l be most useful because they should be more robust to errors i n estimation than the individual-based indices. Accurate estimation of the absolute abundance of individuals i n current time i s extremely d i f f i c u l t for most marine species, and may only be possible using stock assessment models which estimate abundance a f t e r the fact (e.g. cohort a n a l y s i s ) . The evolved l i f e h i s t o r y and demographic structure of ro c k f i s h populations, when combined with the f i s h e r i e s concept of 'replacement stock,' suggest that consideration of reproductive value should be a strong management focus (Leaman 1987a). I t i s cl e a r that the incorporation of reproductive value into manage-ment w i l l need to be an active and experimental process of some duration. A f i r s t step may be the reconstruction of stock h i s t o r i e s and examination of recruitment patterns i n r e l a t i o n to reproductive value. While t h i s w i l l obviously take some time, i t should d i r e c t our management models more toward b i o l o g i c a l p r i n c i p l e s and away from an assumption that relationships producing past recruitment patterns w i l l r e f l e c t present proces-ses, i n spi t e of major fishery e f f e c t s on stock c h a r a c t e r i s t i c s . 181 VI. SUMMARY In t h i s study I examined the population and reproductive biology of a long-lived f i s h (Sebastes alutus). Through the use of a copepod g i l l parasite (Neobrachiella robusta), I established a basis for population sub-structure (stocks). This work i s the f i r s t use of a lernaepodid copepod for stock i d e n t i f i c a t i o n . Stock i d e n t i t y was based on the i n f e c t i o n rate and demographic structure of the parasite complement of the host f i s h . Samples of f i s h which were morphologically indistinguishable were completely separated on the basis of t h e i r parasites. These r e s u l t s permitted examination of the b i o l o g i c a l c h a r a c t e r i s t i c s of f i s h from each stock, with confidence that adjacent stocks would not introduce contamination of the c h a r a c t e r i s t i c s . The r e s u l t s of the parasite analysis do not imply genetic i s o l a t i o n , and even though adults are separated into d i s c r e t e units, other studies show that there i s genetic exchange among them. The reproductive biology of these stocks was examined i n r e l a t i o n to the high mortality rates imposed by commercial f i s h e r i e s and the pote n t i a l compensatory mechanisms postulated i n l i f e h i s t o r y theory. Physical c h a r a c t e r i s t i c s (size and age composition) show major e f f e c t s of the high mortality rates. There appears to be an inverse r e l a t i o n s h i p of growth and natural mortality rates, such that older f i s h (40-80 y) are smaller than many younger (3 0-40 y) f i s h . In a more detailed study, with a colleague, I have shown that the smaller f i s h observed at older ages can be accounted for using the same type of growth form as 182 the larger f i s h seen at younger ages, and that these s i m i l a r growth forms can give r i s e to an appearance of a quadratic growth function. This work emphasizes the need to c o l l e c t cohort-s p e c i f i c growth data to e s t a b l i s h accurate growth r e l a t i o n s h i p s . Compensatory responses to e x p l o i t a t i o n have taken the form of an apparent density-mediated growth increase for f i s h r e c r u i -t i n g into, and presently i n , heavily exploited stocks. While lengths at age for these f i s h are 5-15% greater than f o r f i s h from l i g h t l y exploited stocks, normal inter-cohort v a r i a t i o n i s of almost the same magnitude. This compensatory growth increase gives r i s e to changes i n age at maturity, although s i z e s at maturity d i f f e r only s l i g h t l y among stocks. Developmental or environmental constraints on maturation are implied by the s t a b i l i t y of the s i z e at maturity. The reproductive e f f o r t changes accompanying t h i s growth increase constitute the major compensatory mechanism for higher mortality rates. However, some evidence of a hierarchy for a l l o c a t i o n of increased energy resources was found. The f a s t e r growth rate a r i s i n g from mortality e f f e c t s led to s l i g h t l y lower s i z e - s p e c i f i c fecundity, r e l a t i v e to f i s h from unexploited stocks. Growth may therefore have a higher p r i o r i t y than reproduction i n t h i s species, although t h i s e f f e c t i s not large. The e f f e c t of changing mortality rates on the reproductive c h a r a c t e r i s t i c s of t h i s species was examined with a stochastic 183 simulation model. Results suggest that reproductive value ( l i f e t i m e reproductive output) can provide an a l t e r n a t i v e and more s e n s i t i v e index of population condition than some indices presently used i n f i s h e r i e s assessment (e.g. catch r a t e ) . S e n s i t i v i t y analysis of the population model parameters indicates that fecundity at age, natural mortality rate, the p a r t i a l recruitment vector, and the stock-recruitment r e l a t i o n s h i p may be the most c r i t i c a l elements i n understanding the long-term dynamics of the stocks. In order to obtain the data for the population model, a d e t a i l e d examination of the reproductive biology of S. alutus was undertaken. The volumetric method of fecundity estimation previously used was shown to be unreliable and a gravimetric method was developed for t h i s species. The developmental sequence for f i s h and i n d i v i d u a l oocytes was described and the f i r s t examples of complete a t r e s i a of a ripe oocyte complement i n t h i s genus was discovered. F o l l i c u l a r a t r e s i a was common i n developing oocytes and was suggested as an a l t e r n a t i v e source of embryo n u t r i t i o n , to the embryo death and d i s s o l u t i o n reported i n the l i t e r a t u r e . No evidence of reproductive senescence was found, even though f i s h up to 80 y of age were examined. This study indicates that an understanding of l i f e t i m e reproductive e f f o r t can be a valuable addition to the methods of stock management presently used. 184 VII. LITERATURE CITED Archibald, C. P., D. A. Fournier, and B. M. Leaman. 1983. Recon-s t r u c t i o n of stock h i s t o r y and development of r e h a b i l i t a t i o n strategies for P a c i f i c ocean perch i n Queen Charlotte Sound, Canada. N. Amer. J . Fish. Mgmt. 3:283-294. Archibald, C. P., W. Shaw, and B. M. Leaman. 1981. Growth and mortality estimates of rockfishes (Scorpaenidae) from B. C. coastal waters, 1977-1979. Can. Tech. Rep. Fish . Aquat. S c i . 1048. 57p. Askenmo, C. 1979. Reproductive e f f o r t and return rate of male pied flycatchers. Am. Nat. 114:748-752. Avery, B. A. 1975. Clutch s i z e and reproductive e f f o r t i n the l i z a r d Lacerta v i v i p e r a Jacquin. Oecologia 19:165-170. Bagenal, T. B. 1973. Fish fecundity and i t s r e l a t i o n s with stock and recruitment. Rapp. P-V. Reun. Cons. Int. Explor. Mer. 1973:186-198. Bagenal, T. B. 1978. Aspects of f i s h fecundity, p. 75-101 i n Gerking, S. B. (ed.). Ecology of Fish Production i n Fresh Water. Blackwell S c i . Publ., Oxford. Bagenal, T. B. and E. Braum. 1968. Eggs and early l i f e h istory, i n W.E. Ricker (ed.) Methods for assessment of f i s h produc-t i o n i n fresh waters. IBP Handbook No. 3:159-181. Blackwell S c i . Publ., Oxford, x i i i + 313p. Balsiger, J . W., D. H. Ito, D. K. Kimura, D. A. Somerton, and J . M. Terry. 1985. B i o l o g i c a l and economic assessment of P a c i f i c ocean perch (Sebastes alutus) i n waters o f f Alaska. U. S. Dept. Commerce, NOAA Tech. Memor. NMFS F/NWC-72. 210p. Beamish, R. J . 1979. New information on the longevity of P a c i f i c ocean perch (Sebastes alutus). J . Fish. Res. Board Can. 36:1395-1400. B e l l , G. 1980. The costs of reproduction and t h e i r consequences. Am. Nat. 116:45-76. B e l l , G. 1984a. Measuring the cost of reproduction I. The co r r e l a t i o n structure of the l i f e table of a planktonic r o t i f e r . Evolution 38:300-313. B e l l , G. 1984b. Measuring the cost of reproduction I I . The co r r e l a t i o n structure of the l i f e tables of f i v e freshwater invertebrates. Evolution 38:314-326. Blaxter, J . H. S. and G. Hempel. 1963. The influence of egg size on herring larvae (Clupea harenqus L.). J . Cons. perm. i n t . 185 Explor. Mer. 28:211-293. Boehlert, G. W. , W. H. Barss, and P. B. Lamberson. 1983. Fecun-d i t y of the widow ro c k f i s h (Sebastes entomelas) o f f the coast of Oregon. Fish. B u l l . 80:881-884. Boehlert, G. W., M. Kusakari, and J . Yamada. 1987. Reproductive mode and energy costs of reproduction i n the genus Sebastes. p. 143-152 i n Proceedings of the International Rockfish Symposium. Univ. of Alaska Sea Grant Rep. 87-2. Boehlert, G. W. and M. M. Yoklavich. 1984a. Reproduction, embryo-nic energetics, and the maternal-fetal r e l a t i o n s h i p i n the viviparous genus Sebastes (Pisces:Scorpaenidae). B i o l . B u l l . 167:354-370. Boehlert, G. W. and M. M. Yoklavich. 1984b. Carbon a s s i m i l a t i o n as a function of ingestion rate i n l a r v a l P a c i f i c herring, Clupea harenqus p a l l a s i . J . Exp. Mar. B i o l . Ecol. 79:251-262. Boehlert, G. W. and M. M. Yoklavich. 1987. Long term cycles of growth i n Sebastes: extracting information from o t o l i t h s , p. 197-207 in. Proceedings of the International Rockfish Symposium. Univ. of Alaska Sea Grant Rep. 87-2. Bostock, S. J . 1978. Seed germination strategies of f i v e peren-n i a l weeds. Oecologia (Berl.) 36:113-126. Boyce, M. S. 1981. Beaver l i f e h i s t o r y responses to e x p l o i t a t i o n . J . appl. Ecol. 18:749-753. Bradshaw, W. E. 1973. Homeostasis and polymorphism i n vernal development of Chaoborus americanus. Ecology 54:1247-1259. Brett, J . R. 1974. Tank experiments on the culture of pan-size sockeye (Oncorhynchus nerka) and pink salmon (0. qorbuscha) using environmental control. Aquaculture 4:341-352. Brett, J . R. 1979. Environmental factors and growth, p. 599-565 in W. S. Hoar, D. J . Randall and J . R. Brett (eds.). Fish Physiology. Vol. VIII. Academic Press, New York. Brett, J . R. and T. D. D. Groves. 1979. Physiological energetics, p. 279-352 i n W. S. Hoar, D. J . Randall and J . R. Brett (eds.). Fish Physiology. Vol. VIII. Academic Press, New York. Browne, R. A. 1982. The costs of reproduction i n brine shrimp. Ecology 63:43-47. Bryant, E. 1971. L i f e - h i s t o r y consequences of natural s e l e c t i o n : 186 Cole's r e s u l t . Am. Nat. 105:75-76. Bulmer, W. G. 1985. Selection for i t e r o p a r i t y i n a variable environment. Am. Nat. 126:63-71. Byskov, A. G. 1978. F o l l i c u l a r a t r e s i a , p. 533-562 i n Jones, R. E. (ed.). The Vertebrate Ovary: Comparative Biology and Evolution. Plenum Press. London. Calow, P. 1979. The cost of reproduction - A physiological approach. B i o l . Rev. 54:23-40. Caswell, H. 1980. On the equivalence of maximizing reproductive value and f i t n e s s . Ecology 61:19-24. Caughley, G. 1970. A comment on Vandermeer's pseudo-reproductive value. Am. Nat. 104:214-215. Charlesworth, B. 1972. Selection i n populations with overlapping generations. I I I . Conditions for genetic equilibrium. Theor. Popul. B i o l . 3:377-395. Charlesworth, B. 1973. Selection i n populations with overlapping generations. V. Natural s e l e c t i o n and l i f e h i s t o r i e s . Am. Nat. 107:303-311. Charlesworth, B. 1980. Evolution i n Age-structured Populations. Cambridge Univ. Press, Cambridge, x i i i + 30Op. Charlesworth, B. and J . A. Leon. 1976. The r e l a t i o n of reproduc-t i v e e f f o r t to age. Am. Nat. 110:449-459. Charnov, E. L., J . J . B u l l , and S. T. Mitchell-Olds. 1981. A note on sex and l i f e h i s t o r i e s . Am. Nat. 117: 814-818. Charnov, E. L. and J . R. Krebs. 1974. On cl u t c h - s i z e and f i t n e s s . I b i s 116:217-219. Charnov, E. L. and W. M. Schaffer. 1973. L i f e h i s t o r y consequen-ces of natural s e l e c t i o n : Cole's r e s u l t r e v i s i t e d . Am. Nat. 107:791-793. Chen, L. 1986. M e r i s t i c v a r i a t i o n i n Sebastes (Scorpaenidae), with an analysis of character association and b i l a t e r a l pattern and t h e i r s i g n i f i c a n c e i n species separation. NOAA Tech. Rep. NMFS 45. 17p. Chikuni, S. and K. Wakabayashi. 1970. On the scale characters of the P a c i f i c ocean perch i n the Bering Sea. I I I . O b j e c t i v i t y and accuracy of age determination by scale reading. B u l l . Far Seas Fish. Res. Lab. (Shimizu). 3:205-214. 187 Chilton, D. E. and R. J . Beamish. 1982. Age determination methods for fishes studied by the Groundfish Program at the P a c i f i c B i o l o g i c a l Station. Can. Spec. Publ. Fish. Aquat. S c i . 60. 102p. Clarke, A. 1987. Temperature, l a t i t u d e and reproductive e f f o r t . Mar. Ecol. (Prog. Ser.). 38:88-99. Clutton-Brock, T. H., E. F. Guiness, and S. D. Albon. 1982. Red Deer - Behavior and Ecology of Two Sexes. Univ. Chicago Press, Chicago. Cochran, W. G. 1977. Sampling Techniques. 3rd E d i t i o n . John Wiley and Sons, New York. Cody, M. 1966. A general theory of clutch s i z e . Evolution 20:174-184. Cole, L. C. 1954. The population consequences of l i f e h i s t o r y phenomena. Q. Rev. B i o l . 29:103-137. Crow, J . F. and M. Kimura. 1970. An Introduction to Population Genetics Theory. Harper and Row, New York, xiv + 59lp. Daan, N. 1975. Consumption and production i n the North Sea cod (Gadus morhua): An assessment of the e c o l o g i c a l status of the stock. Neth. J . Sea Res. 9:24-55. den Boer, P. J . 1968. Spreading of r i s k and s t a b i l i z a t i o n of animal numbers. Acta Biotheoretica 18:165-194. Doyle, R. W. and W. Hunte. 1981. Demography of an estuarine amphipod (Gammarus lawrencianus) experimentally selected for high • r 1 . A model of the genetic e f f e c t s of environmental change. Can. J . Fish. Aquat. S c i . 38:1120-1127. Dunn, R. S. 1970. Further evidence for a three-year maturation time i n the winter flounder (Pseudopleuronectes americanus). J . Fish. Res. Board Can. 27:957-960. Dunn, R. S. and A. V. Tyler. 1969. Aspects of the anatomy of the winter flounder with hypotheses on oocyte maturation time. J . Fish. Res. Board Can. 26:1943-1947. Dygert, P. H. 1986. Management implications of v a r i a b i l i t y i n reproduction and growth i n commercial marine f i s h e s . Ph. D. Thesis, Univ. Wash., 275p. Echeverria, T. Wyllie. 1987. Thirty-four species of C a l i f o r n i a r o c k f i s h : maturity and seasonality of reproduction. Fish. B u l l . 85:229-250. 188 Ekman, J . and C. Askenmo. 1986. Reproductive cost, age-specific s u r v i v a l and a comparison of the reproductive strategy i n two European t i t s (genus Parvus). Evolution 40:159-168. Emlen, J . M. 1970. Age s p e c i f i c i t y and ecological theory. Ecology 51:588-601. E v e r i t t , B. S. and D. J . Hand. 1981. F i n i t e Mixture D i s t r i b u -t i o n s . Chapman and H a l l , London. 143p. Fagen, R. M. 1972. An optimal l i f e h i s t o r y i n which reproductive e f f o r t decreases with age. Am. Nat. 106:258-261. Fisher, R. A. 1930. The Genetical Theory of Natural Selection. Oxford. Clarendon Press. F i t z p a t r i c k , L. C. 1973. Energy a l l o c a t i o n i n the Allegheny Mountain salamander Desmoqnathus ochraphaeus. Ecol. Monog. 43:43-58. Foucher, R. P. and R. J . Beamish. 1980. Production of nonviable oocytes by P a c i f i c hake (Merluccius productus). Can. J . Fish. Aquat. S c i . 37:41-48. Francis, R. C. 198 6. Two f i s h e r i e s biology problems i n west coast groundfish management. N. Am. J . Fish. Mgmt. 6:453-462. Gadgil, M. and W. H. Bossert. 1970. L i f e h i s t o r y consequences of natural s e l e c t i o n . Am. Nat. 102: 52-64. Garrod, D. J . and B. J . Knights. 1979. Fish stocks: t h e i r l i f e h i s t o r y c h a r a c t e r i s t i c s and response to e x p l o i t a t i o n . Symp. Zool. Soc. Lond. 44:361-382. Gerking, S. D. 1959. Physiological changes accompanying ageing i n f i s h e s . CIBA Foundation Colloquia on Ageing. 5:181-207. Getz, W. M. , R. C. Francis, and G. L. Swartzman. 1987. On managing variable marine f i s h e r i e s . Can. J . Fish. Aquat. S c i . 44:1370-1375. Giesel, J . T. 1974. Fitness and polymorphism for net fecundity d i s t r i b u t i o n i n iteroparous populations. Am. Nat. 108:321-331. Giesel, J . T. 1976. Reproductive strategies as adaptations to l i f e i n temporally heterogeneous environments. Ann. Rev. Ecol. Syst. 7:57-79. Gies e l , J . T. and E. Z e t t l e r . 1980. Genetic c o r r e l a t i o n of l i f e h i s t o r i c a l parameters on c e r t a i n f i t n e s s indices i n Drosoph- i l a melanogaster: rm, rs, d i e t breadth. Oecologia (Berl.) 189 47:299-302. Gies e l , J . T., P. A. Murphy, and M. N. Manlove. 1982. The influence of temperature on genetic i n t e r r e l a t i o n s h i p s of l i f e h i s t o r y t r a i t s i n a population of Drosophila melanogas- t e r : what tangled data sets we weave. Am. Nat. 119:464-479. G i l l e s p i e , J . H. 1977. Natural s e l e c t i o n for variances i n o f f s p r i n g numbers: A new evolutionary p r i n c i p l e . Am. Nat. Ill:1010-1014. Goodman, D. 1978. Management implications of the mathematical demography of long-lived animals. Mar. Mamm. Comm. Doc. MM8D-008. i i + 80p. Goodman D. 1979. Regulating reproductive e f f o r t i n a changing environment. Am. Nat. 113:735-748. Goodman, L. A. 1967. On the r e c o n c i l i a t i o n of mathematical theories of population growth. J . Roy. Stat. Soc. 13 0:541-543. Gray, P. 1954. The Microtomist 1s Formulary and Guide. Blakiston. New York. 794p. Gritsenko, O. F. 1963. Age and growth of P a c i f i c r o c k f i s h of the Bering Sea. i n Soviet f i s h e r i e s investigations i n the northeast P a c i f i c . Part I: 328-331. (Translated from Russian by Clearinghouse Fed. S c i . Tech. Inform., S p r i n g f i e l d , V i r g i n i a , TT67-51203) . Gunderson, D. R. 1971. Reproductive patterns of P a c i f i c ocean perch (Sebastes alutus) o f f Washington and B r i t i s h Columbia and t h e i r r e l a t i o n to bathymetric d i s t r i b u t i o n and seasonal abundance. J . Fish. Res. Board Can. 28: 417-425. Gunderson, D. R. 1976. Population biology of P a c i f i c ocean perch (Sebastes alutus) stocks i n the Washington-Queen Charlotte Sound region and t h e i r response to f i s h i n g . Ph. D. thesis, Univ. Wash., Seattle, Wa. 14Op. Gunderson, D. R. 1977. Population biology of P a c i f i c ocean perch, Sebastes alutus. stocks i n the Washington - Queen Charlotte Sound region, and t h e i r response to f i s h i n g . F i s h . B u l l . 75: 369-403. Gunderson, D. R. , S. J . Westrheim, R. L. Demory and M. E. Fraidenburg. 1977. The status of P a c i f i c ocean perch (Sebastes alutus) stocks o f f B r i t i s h Columbia, Washington, and Oregon i n 1974. Fish. Mar. Serv. Tech. Rep. No. 690. 63p. 190 Harper, J . L. 1967. A Darwinian approach to plant ecology. J . Ecol. 55:247-270. Harper, J . L. and J . Ogden. 1970. The reproductive strategy of higher plants. I. The conception of strategy with special reference to Senecio v u l g a r i s . J . Ecol. 58:681-698. Hart, J . L. 1973. P a c i f i c fishes of Canada. Fish. Res. Board Can. B u l l . 180. 740p. Henderson, N. E. 1963. Extent of a t r e s i a i n maturing ovaries of the eastern brook trout, Salvelinus f o n t i n a l i s (Mitchell) . J. Fish. Res. Board Can. 20:899-908. Hickman, J . C. 1975. Environmental unpredictabilityand the p l a s t i c energy a l l o c a t i o n strategies i n the annual Polygonum  cascadense (Polygonaceae). J . Ecol. 63:689-702. Hightower, J . E. and G. D. Grossman. 1986. Comparison of constant e f f o r t harvest p o l i c i e s f o r f i s h stocks with variable recruitment. Can. J . Fish. Aquat. S c i . 42:982-988. Hightower, J . E. and G. D. Grossman. 1987. Optimal p o l i c i e s f or r e h a b i l i t a t i o n of overexploited f i s h stocks using a deter-m i n i s t i c model. Can. J . Fish. Aquat. S c i . 44:803-810. H i r s h f i e l d , M. F. 1980. An experimental analysis of reproductive e f f o r t and cost i n the Japanese Medaka Or y z i a l l a t i p e s . Ecology 61:285-292. Holgate, P. 1967. Population s u r v i v a l and l i f e h i s t o r y phenomena. J. Theoret. B i o l . 14:1-10. Johnson, A. G. , F. M. Utter, and H. 0. Hodgins. 1970a. E l e c t r o -phoretic variants of L-alpha-glycerophosphate dehydrogenase i n P a c i f i c ocean perch (Sebastes alutus). J . Fish. Res. Board Can. 27:943-945. Johnson, A. G. , F. M. Utter, and H. 0. Hodgins. 1970b. Inters-p e c i f i c v a r i a t i o n of tetrazolium oxidase i n Sebastodes (ro c k f i s h ) . Comp. Biochem. Physiol. 37:281-285. Johnson, A. G., F. M. Utter, and H. 0. Hodgins. 1971. Phosphoglu-comutase polymorphism i n P a c i f i c ocean perch (Sebastes  alutus). Comp. Biochem. Physiol. 39B:285-290. Johnson, A. G., F. M. Utter, and H. 0. Hodgins. 1972. Electropho-r e t i c i nvestigation of the Family Scorpaenidae. Fis h . B u l l . 70:402-413. Johnson, A. G., F. M. Utter, and H. 0. Hodgins. 1973. Estimate of genetic polymorphism and heterozygosity i n three species of 191 r o c k f i s h (genus Sebastes). Comp. Biochem. Physiol. 44:397-406. Kabata, Z. 1970. Some Lernaepodidae (Copepoda) from fishes of B r i t i s h Columbia. J . Fish. Res. Board Can. 27: 865-885. Kabata, Z. 1987. The developmental stages of Neobrachiella  robusta (Wilson, 1912), a p a r a s i t i c copepod of Sebastes (Teleostei: Scorpaeniformes). Can. J . Zool. 65:1331-1336. Kandler, R. and W. Pirwitz. 1957. Uber die Fruchbarkeit der P i a t t f i s c h e i n Nordsee-Ostsee-Raum. K i e l e r Meeresforsch. 13:11-34. Keeton, W. T. 1980. B i o l o g i c a l Science (3rd Ed.). W. W. Norton & Co., New York. 113lp. Ketchen, K. S. 1980. Reconstruction of P a c i f i c ocean perch (Sebastes alutus) stock h i s t o r y i n Queen Charlotte Sound. Part I. Estimation of foreign catches, 1965 - 1976. Can. MS Rep. Fish. Aquat. S c i . 1570. 46p. Ketchen, K. S. 1981. Reconstruction of P a c i f i c ocean perch (Sebastes alutus) stock h i s t o r y i n Queen Charlotte Sound. Part I I . Catch per unit e f f o r t as a measure of abundance, 1959 - 1979. Can. MS Rep. Fish. Aquat. S c i . 1599. 72p. Kluyver, H. N. 1971. Regulation of numbers i n populations of great t i t s (Parus m. major) . i n P. J . den Boer and G. R. Gradwell (ed.). Proc. Adv. Study Inst. Dynamics Numbers Popul., Pudoc, Waageningen. Knight, W. 1968. Asymptotic growth: an example of nonsense disguised as mathematics. J . Fish. Res. Board Can. 25:1303-1307. Lack, D. 1947. The si g n i f i c a n c e of clutch s i z e . I b i s 89:302-352. Lapi, L. A. and J. E. Richards. 1981. Data c o l l e c t e d during rock-f i s h (Sebastes spp.) assessments of the west Langara Is. and west coast Vancouver Is. f i s h i n g grounds i n 1979. Can. Data Rep. Fish. Aquat. S c i . 286. 113p. Law, R. 1979a. The costs of reproduction i n annual meadow grass. Am. Nat. 113: 3-16. Law, R. 1979b. Harvest optimization i n populations with age d i s t r i b u t i o n s . Am. Nat. 114: 250-259. Law, R. , A. D. Bradshaw, and P. D. Putswain. 1977. L i f e h i s t o r y v a r i a t i o n i n Poa annua. Evolution 31: 233-246. 192 Leaman, B. M. 1985. Slope r o c k f i s h . i n Tyler, A. V. and G. A. McFarlane (eds.). Groundfish stock assessments f o r the west coast of Canada i n 1984 and recommended y i e l d options for 1985. Can. MS. Rep. Fish. Aquat. S c i . 1813:247-342. Leaman, B. M. 1987a. Incorporating reproductive value into P a c i f i c ocean perch management, p. 355-368 i n Proceedings of the International Rockfish Symposium. Univ. of Alaska Sea Grant Rep. 87-2. Leaman, B. M. 1987b. Slope r o c k f i s h . i n Tyler, A. V. and M. Saun-ders (eds.). Groundfish stock assessments f o r the west coast of Canada i n 1986 and recommended y i e l d options f o r 1987. Can. MS. Rep. Fish. Aquat. S c i . 1930:105-140. Leaman, B. M. and R. J . Beamish. 1984. E c o l o g i c a l and management implications of longevity i n some northeast P a c i f i c ground-fi s h e s . Int. North Pac. Fish. Comm. B u l l . 42:85-97. Leaman, B. M. and Z. Kabata. 1987. Neobrachiella robusta (Wilson 1912) (Copepoda:Lernaepodidae) as a tag f o r i d e n t i f i c a t i o n of stocks of i t s host, Sebastes alutus (Gilbert 1890) (Pisces:Teleostei). Can. J . Zool. 65:2579-2582. Leaman, B. M. and D. A. Nagtegaal. 1982. Biomass estimation of r o c k f i s h stocks o f f the west coast of the Queen Charlotte Islands during 1978 and 1979. Can. MS Rep. Fish. Aquat. S c i . 1652. 46p. Leaman, B. M. and D. A. Nagtegaal. 1987. Age v a l i d a t i o n and revised natural mortality rate for y e l l o w t a i l r o c k f i s h . Trans. Am. Fish. Soc. 116:171-175. Leaman, B. M., D. A. Nagtegaal, R. D. Stanley, and E. W. Carter. 1985. Rockfish fecundity cruise (GBR-R82-1), November 8-28, 1982. Can. Data Rep. Fish. Aquat S c i . 526. 64p. Leaman, B. M. and R. D. Stanley. Unpubl. Ms. From n o b i l i t y to nightmares: some object lessons i n the planning and conduct of f i s h e r y harvest experiments. Ms. presented at the 1987 Western Groundfish Conference, Gleneden, Oregon. Lee, R. M. 1912. An investigation into the methods of growth determination i n fishes. Cons. Explor. Mer, Publ. de Circonstance 63:35p. Lehri, G. K. 1968. C y c l i c a l changes i n the ovary of the c a t f i s h C l a r i a s batrachus (Linn.). Acta Anat. 69:105-124. L e s l i e , P. H. 1945. On the use of matrices i n c e r t a i n population mathematics. Biometrika 33:183-212. 193 L e s l i e , P. H. 1948. Some further notes on the use of matrices i n population mathematics. Biometrika 35:213-245. Lewis, E. G. 1942. On the generation and growth of a population. Sankhya 6:93-96. Lewontin, R. C. 1965. Selection for colonizing a b i l i t y . p . 74-91 i n H.G. Baker and G.L. Stebbin (eds.). The Genetics of Colonizing Species. Academic Press, New York. Lotka, A. J . 1913. A natural population norm. I I . J . Wash. Acad. S c i . 3:289-293. Love, R. M. 1970. The Chemical Biology of Fishes. Academic Press, New York. 547p. L u c k i n b i l l , L. S. 1984. An experimental analysis of l i f e h i s t o r y theory. Ecology 65:1170-1184. L u c k i n b i l l , L. S. and M. J . Clare. 1985. Selection f o r l i f e span i n Drosophila melanocfaster. Heridity 55:9-18. MacGregor, J . S. 1970. Fecundity, multiple spawning, and descrip-t i o n of gonads i n Sebastodes. U.S. Fish. W i l d l f . Serv., Spec. S c i . Rep. Fish. 596. 12p. MacKenzie, K. 1983. Parasites as b i o l o g i c a l tags i n f i s h popula-t i o n studies. Adv. Appl. B i o l . 7: 251-331. MacLean, J . A. and D. O. Evans. 1981. The stock concept, d i c r e t e -ness of f i s h stocks, and f i s h e r i e s management. Can. J . Fish. Aquat. S c i . 38:1889-1898. Magnusson, J . 1955. Mikroscopisch-anatomische untersuchungen zur fortpfanzungsbiologie des rotbarsches (Sebastes marinus Linne). Z e i t s c h r i f t fur Zellforschung, Bd. 43:121-167. Mann, R. H. K. and C. A. M i l l s . 1979. Demographic aspects of f i s h fecundity. Symp. Zool. Soc. London 44: 161-177. Mann, R. H. K., C. A. M i l l s , and D. T. Crisp. 1984. Geographical v a r i a t i o n i n the l i f e - h i s t o r y t a c t i c s of some species of freshwater f i s h . p. 171-186 i n Potts, G. W. and R. J . Wootton (eds.). Fish reproduction: strategies and t a c t i c s . Academic Press. London. Mason, J . C. 1984. On the fecundity of the s a b l e f i s h (Anoplopoma  fimbria) i n Canadian waters. Can. Tech. Rep. Fish. Aquat. S c i . 1290. 17p. Mason, J . C. , R. J . Beamish, and G. A. McFarlane. 1983. Sexual maturity, fecundity, spawning, and early l i f e h i s t o r y of 194 s a b l e f i s h fAnoplopoma fimbria) o f f the P A c i f i c coast of Canada. Can. J . Fish. Aquat. S c i . 40:2126-2134. Maynard Smith J . 1978b. Optimization theory i n evolution. Ann. Rev. Ecol. Syst. 9:31-56. Maynard Smith, J . and G. R. Price. 1973. The l o g i c of animal con-f l i c t . Nature. London. 246:15-18. McFarlane, G. A., W. Shaw, and A. V. Tyler. 1985. Sablefish. p. 163-186 i n A. V. Tyler and G. A. McFarlane (eds.). Ground-f i s h stock assessments o f f the west coast of Canada i n 1984 and recommended y i e l d options for 1985. Can. MS Rep. Fish. Aquat. S c i . 1813:353p. Michod, R. E. 1979. Evolution of l i f e h i s t o r i e s i n response to age-specific mortality factors. Am. Nat. 113:531-550. M i l l e r , D. J . and R. N. Lea. 1972. Guide to the coastal marine fishes of C a l i f o r n i a . C a l i f . Dept. Fish Game, Fish. B u l l . 157. 235p. Minchella, D. J . and P. T. Loverde. 1981. A cost of increased early reproductive e f f o r t i n the s n a i l Biomphalaria glabra- t a . Am. Nat. 118:876-881. Moser, H. G. 19 67. Seasonal h i s t o l o g i c a l changes i n the gonads of Sebastes p a u c i s p i n i s Ayres, an ovoviviparous t e l e o s t (Scorpaenidae). J . Morph. 123:329-354. Murdoch, W. W. 1966a. Population s t a b i l i t y and l i f e h i s t o r y phenomena. Am. Nat. 100:5-11. Murphy, G. I. 1968. Pattern i n l i f e h i s t o r y and the environment. Am. Nat. 102: 52-64. Ogden, J . 1974. The reproductive strategy of higher plants. I I . The reproductive strategy of Tussilago f a r f a r a L. J . Ecol. 62:291-324. Pal, R.N. (In press). Parasites as the indicators of migration of H i l s a i l i s h a (Hamilton) i n the Hugli estuary. J . Inland Fish. Soc. India 12: (in press). Partridge, L. and M. Farquhar. 1981. Sexual a c t i v i t y reduces l i f e span of male f r u i t f l i e s . Nature 294:580-582. Perrins, C. M. and P. J . Jones. 1974. The inheritance of clutch s i z e i n the great t i t (Parus manor L.). Condor 76:225-229. Ra i t t , D. S. 1933. The fecundity of the haddock. Scient. Invest. Fishery Bd. S c o t l . No. 1:1-42. 195 Rajalakshmi, R. 1966. Atres i a of oocytes and ruptured f o l l i c l e s i n Gobius q i u r i s (Hamilton-Buchanan). Gen. Comp. Endocrin. 6:378-385. Rao, C. R. 1973. Linear S t a t i s t i c a l Inference and I t s Applica-tions, 2nd E d i t i o n . John Wiley and Sons, New York. 625p. Rastogi, R. K. 1966. The occurence and s i g n i f i c a n c e of f o l l i c u l a r a t r e s i a i n the c a t f i s h , Mystus tenqara (Ham.)(Teleostei). Acta. Zool. 49:307-319. Reznick, D. 1985. Costs of reproduction: and evaluation of the empirical evidence. Oikos 44:257-267. Ricker, W. E. 1969. E f f e c t s of s i z e - s e l e c t i v e mortality and sampling bias on estimates of growth, mortality, production, and y i e l d . J . Fish. Res. Board Can. 26:479-541. Ricker, W. E. 1975. The computation and i n t e r p r e t a t i o n of the b i o l o g i c a l s t a t i s t i c s of f i s h populations. Fi s h . Res. Board Can. B u l l 191. 382p. R i c k l e f s , R. E. 1977. On the evolution of reproductive strategies i n birds: reproductive e f f o r t . Am. Nat. 111:453-478. R i c k l e f s , R. E. 1981a. The optimization of l i f e h i s t o r y patterns under density dependence. Am. Nat. 117:403-408. R i c k l e f s , R. E. 1981b. Fitness, reproductive value, age structure and optimization of l i f e - h i s t o r y patterns. Am. Nat. 117: 819-825. Rivard, D. 1982. APL programs for stock assessment (revised). Can. Tech. Rep. Fish. Aquat. S c i . 1091:146p. Roff, D. A. 1980. A motion for the retirement of the von Berta-l a n f f y function. Can. J . Fish. Aquat. S c i . 37:127-129. Rose, M. R. and B. Charlesworth. 1981a. Genetics of l i f e h i s t o r y i n Drosophila melanoqaster. I. Sib analysis of adult females. Genetics 97:173-186. Rose, M. R. and B. Charlesworth. 1981b. Genetics of l i f e h i s t o r y i n D r o s o p h i l a melanoqaster. I I . Exploratory s e l e c t i o n experiments. Genetics 97:187-196. Roughgarden, J . 1979. Theory of Population Genetics and Evolu-tionary Ecology: An Introduction. Macmillan, New York. 634p. SAS I n s t i t u t e . 1982. SAS User's Guide: S t a t i s t i c s . SAS I n s t i t u t e , Cary, North Carolina. 584p. 196 Schaffer, W. M. 1974a. Selection for optimal l i f e h i s t o r i e s : the ef f e c t s of age structure. Ecology 55: 291-303. Schaffer, W. M. 1974b. Optimal reproductive e f f o r t i n fl u c t u a t i n g environments. Am. Nat. 108:783-790. Schaffer, W. M. 1979a. Equivalence of maximizing reproductive value and f i t n e s s i n the case of reproductive strategies. Proc. Natl. Acad. S c i . (USA) 76: 3567-3569. Schaffer, W. M. and M. L. Rosenzweig. 1977. Selection f o r optimal l i f e h i s t o r i e s I I . Multiple e q u i l i b r i a and the evolution of alternate reproductive strategies. Ecology 58:60-72. Schaffer, W. M. and R. H. Tamarin. 1973. Changing reproductive rates and population cycles i n lemmings and voles. Evolution 27:114-125. Schnute, J . 1981. A v e r s a t i l e growth model with s t a t i s t i c a l l y stable parameters. Can. J . Fish. Aquat. S c i . 38:1128-1140. Seeb, L. W. 1986. Biochemical sytematics and evolution of the Scorpaenid genus Sebastes. Ph. D. thesis, Univ. Wash., Seattle, Wa. x + 176p. Sekerak, A. D. 1975. Parasites as indicators of populations and species of rockfishes (Sebastes: Scorpaenidae) of the northeastern P a c i f i c Ocean. Ph. D. thesis, Univ. Calgary. Calgary, Alberta. 251p. Shaw, W. and C. P. Archibald. 1981. Length and age frequencies of r o c k f i s h e s (Scorpaenidae) from B. C. coastal waters, 1977-1979. Can. Data Rep. Fish. Aquat. S c i . No. 289. 119p. Siegel, V. 1980. Parasite tags for some Ant a r c t i c channichthyid f i s h . Arch. Fish. Wiss. 31: 97-103. Silvertown, J . W. 1981. Seed s i z e , l i f e span and germination date as coadapted features of plant l i f e h i s t o r i e s . Am. Nat. 118:860-864. Simpson, A.C. 1951. The fecundity of p l a i c e (Pleuronectes  platessa L.). Fishery Invest., London (2)17:1-27. Skutch, A. F. 1967. Adaptive l i m i t a t i o n of the reproductive rate of b i r d s . Ibis 109:579-599. Sl a t k i n , M. 1974. Hedging one's evolutionary bets. Nature 250:704-705. Slobodkin, L. B. 1964. Experimental populations of Hydrida. J . 197 Anim. Ecol. 33 (Suppl.):131-148. Sokal, R. R. 1970. Senescence and genetic load: evidence from Tribolium. Science 167:1733-1734. Solbrig, O. T. and B. B. Simpson. 1974. Components of regulation of a population of dandelions i n Michigan. J . Ecol. 62: 473-486. Stanley, R. D. 1987. A comparison of age estimates derived from the surface and cross-section methods of o t o l i t h reading reading for P a c i f i c ocean perch (Sebastes alutus). p. 187-196 i n Proceedings of the International Rockfish Symposium. Univ. of Alaska Sea Grant Rep. 87-2. Stearns, S. C. 1976. L i f e h i s t o r y t a c t i c s : a review of ideas. Q. Rev. B i o l . 51: 3-47. Stearns, S. C. 1977. The evolution of l i f e h i s t o r y t r a i t s : a c r i t i q u e of theory and a review of the data. Ann. Rev. Ecol. Syst. 8: 145-171. Stearns, S. C. 1980. A new view of l i f e h i s t o r y evolution. Oikos 35:266-281. Stocker, M. (ed.) 1981. Groundfish stock assessments o f f the west coast of Canada i n 1981 and recommended t o t a l allowable catches for 1982. Can. MS Rep. Fish. Aquat. S c i . 162 6. xxx + 282p. Summerfelt, R. C. and G. E. H a l l (eds.). 1987. Age and Growth of Fish. Iowa State Univ. Press. 544p. Taylor, H. M. , R. S. Gourley, C. E. Lawrence and R. S. Kaplan. 1974. Natural s e l e c t i o n of l i f e h i s t o r y a t t r i b u t e s : an a n a l y t i c a l approach. Theor. Popul. B i o l . 5:104-122. Theilacker, G. H. 1987. Feeding ecology and growth energetics of l a r v a l northern anchovy, Engraulis mordax. Fish. B u l l . 85:213-228. Tinkle, D. W. 1969. The concept of reproductive e f f o r t and i t s r e l a t i o n to the evolution of l i f e h i s t o r i e s of l i z a r d s . Am. Nat. 103: 501-516. Tinkle, D. W. and N. F. Hadley. 1975. Lizard reproductive e f f o r t : c a l o r i c estimates and comments on i t s evolution. Ecology 56:427-434. Tokarz, R. R. 1978. Oogonial p r o l i f e r a t i o n , oogenesis, and f o l l i c u l o g e n e s i s i n nonmammalian vertebrates, p. 145-180 i n Jones, R. E. (ed.). The Vertebrate Ovary: Comparative 198 Biology and Evolution. Plenum Press. London. Tsuyuki, H. , E. Roberts, R. H. Lowes, W. Hadaway, and S. J . Westrheim. 1968. Contribution of protein electrophoresis to r o c k f i s h (Scorpaenidae) systematics. J . Fi s h . Res. Board Can. 25: 2477-2501. Tuomi, J . , T. Hakala, and E. Haukioja. 1983. A l t e r n a t i v e concepts of reproductive e f f o r t s , costs of reproduction, and selec-t i o n i n l i f e h i s t o r y evolution. Am. Zool. 23:25-34. Vandermeer, J . H. 1968. Reproductive value i n a population of a r b i t r a r y age d i s t r i b u t i o n . Am. Nat. 102:586-589. Vladykov, V. D. 1956. Fecundity of wild speckled trout (Salve- l i n u s f o n t i n a l i s ) i n Quebec lakes. J . Fish. Res. Board Can. 13:799-841. Vooren, C. M. and Tracy, D. 1976. Parasites i n t a r a k i h i (Pisces: Cheilodactylidae) from the three areas around New Zealand. N. Z. J . Mar. Freshw. Res. 10: 499-509. Waller, D. M. and D. Green. 1981. Implications of sex for the analyses of l i f e h i s t o r i e s . Am. Nat. 117: 810-813. Ware, D. M. 1982. Power and evolutionary f i t n e s s i n t e l e o s t s . Can. J . Fish. Aquat. S c i . 39:3-13 Wasserman, W. J . and L. D. Smith. 1978. Oocyte maturation i n nonmammalian vertebrates, p. 443-468 i n Jones, R. E. (ed.). The Vertebrate Ovary: Comparative Biology and Evolution. Plenum Press. London. Westrheim, S. J . 1958. On the biology of P a c i f i c ocean perch, Sebastes alutus ( G i l b e r t ) . M. Sc. Thesis, Univ. Wash., Seattle. 106p. Westrheim, S. J . 1967. Sampling research trawl catches at sea. J . Fish. Res. Board Can. 24:1187-1202. Westrheim, S. J . 1973. Age determination and growth of P a c i f i c ocean perch (Sebastes alutus) i n the northeast P a c i f i c Ocean. J . Fish. Res. Board Can. 30: 235-247. Westrheim, S. J . 1975. Reproduction, maturation, and i d e n t i f i c a -t i o n of larvae of some Sebastes (Scorpaenidae) species i n the northeast P a c i f i c Ocean. J . Fish. Res. Board Can. 32:2399-2411. Westrheim, S. J . 1976. A further note on sampling researach trawl catches at sea. J . Fish. Res. Board Can. 33:1196. 199 Westrheim, S. J . (ed.) 1980. Assessment of groundfish stocks o f f the west coast of Canada i n 1979 and recommended t o t a l allowable catches for 1980. Can. Data Rep. Fish. Aquat. S c i . 208. 265p. Westrheim, S. J . , D. R. Gunderson, and J . M. Meehan. 1972. On the status of P a c i f i c ocean perch (Sebastes alutus) stocks o f f B r i t i s h Columbia, Washington, and Oregon i n 1970. Fish. Res. Board Can. Tech. Rep. 326. 48p. Wilkinson, L. 1986. SYSTAT: The system f o r s t a t i s t i c s . Systat Inc., Evanston, 111. Williams, G. C. 1966a. Adaption and Natural Selection. Princeton Univ. Press, Princeton, New Jersy. Williams, G. C. 1966b. Natural selection, the costs of reproduc-t i o n and a refinement of Lack's p r i n c i p l e . Am. Nat. 100: 687-690. Wishard, L. N., F. M. Utter, and D. R. Gunderson. 1980. Stock s e g r e g a t i o n of f i v e r o c k f i s h species using na t u r a l l y occurring biochemical genetic markers. Mar. Fish. Rev. 42:64-73. Wolfert, D. R. 1969. Maturity and fecundity of walleyes from the eastern and western basins of Lake E r i e . J . Fish. Res. Board Can. 26:1877-1888. Wootton, R. J . 1977. E f f e c t s of food l i m i t a t i o n during the breeding season on the size , body components and egg production of female sticklebacks (Gasterosteus acculeatus). J . Anim. Ecology. 46:823-834. Wootton, R. J . 1979. Energy costs of egg production and environ-mental determinants of fecundity i n t e l e o s t f i s h e s . Symp. Zool. Soc. Lond. 44:133-159. Wourms, J . P. 1981. V i v i p a r i t y : the maternal-fetal r e l a t i o n s h i p i n f i s h e s . Am. Zool. 21:473-515. Yodzis, P. 1981. Concerning the sense i n which maximizing f i t n e s s i s equivalent to maximizing reproductive value. Ecology 62(6):1681-1682. Zar, J.H. 1984. B i o s t a t i s t i c a l Analysis, 2nd. E d i t i o n . Prentice-H a l l Inc., Englewood C l i f f s , New Jersey. 718p. Publications: Bruce Michael Leaman Leaman, B. M., 1976. An inexpensive tag for short-term v i s u a l tracking studies. J . Fish. Res. Board Can. 33:1628-1629. Leaman, B. M. 1981. A b r i e f review of survey methodology with regard to groundfish stock assessment, i n Doubleday, W. G. and D. Rivard (eds.). Bottom trawl surveys. Can. Spec. Publ. Fis h . Aquat. S c i . 58:113-123. Archibald, C. P., D. Fournier, and B. M. Leaman. 1983. Recon-s t r u c t i o n of stock hi s t o r y and development of r e h a b i l i t a t i o n strategies for P a c i f i c ocean perch i n Queen Charlotte Sound, Canada. N. Am. J . Fish. Mgmt. 3:283-294. Leaman, B. M. and R. J . Beamish. 1984. Ecolog i c a l and management implications of longevity i n some northeast P a c i f i c ground-fis h e s . Int. North Pac. Fish. Comm. B u l l . 42:85-97. Leaman, B. M. 1987. Incorporating reproductive value into P a c i f i c ocean perch management, pp. 355-3 68 i n Proceedings of the International Rockfish Symposium. Univ. of Alaska Sea Grant Rep. 87-2. Leaman, B. M. and Z. Kabata. 1987. Neobrachiella robusta (Wilson 1912) (Copepoda:Lernaepodidae) as a tag for i d e n t i f i c a t i o n of stocks of i t s host, Sebastes alutus (Gilbert 1890) (Pisces:Teleostei). Can. J . Zool. 65:2579-2582. Leaman, B. M. and D. A. Nagtegaal. 1987a. I d e n t i f i c a t i o n of species assemblages and r e s u l t s of management applications for shelf and slope rockfishes o f f B r i t i s h Columbia, pp. 3 09-3 2 8 i n Proceedings of the International Rockfish Symposium. Univ. of Alaska Sea Grant Rep. 87-2. Leaman, B. M. and D. A. Nagtegaal. 1987b. Age v a l i d a t i o n and revised natural mortality rate for y e l l o w t a i l r o c k f i s h . Trans. Am. Fish. Soc. 116:171-175. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098067/manifest

Comment

Related Items