Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An analysis of aggressive behavior, growth, and competition for food and space in medaka (Oryzias latipes)… Magnuson, John Joseph 1961

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

Item Metadata

Download

Media
831-UBC_1961_A1 M2 A6.pdf [ 6.82MB ]
Metadata
JSON: 831-1.0105779.json
JSON-LD: 831-1.0105779-ld.json
RDF/XML (Pretty): 831-1.0105779-rdf.xml
RDF/JSON: 831-1.0105779-rdf.json
Turtle: 831-1.0105779-turtle.txt
N-Triples: 831-1.0105779-rdf-ntriples.txt
Original Record: 831-1.0105779-source.json
Full Text
831-1.0105779-fulltext.txt
Citation
831-1.0105779.ris

Full Text

AN ANALYSIS OF AGGRESSIVE BEHAVIOR, GROWTH, AND COMPETITION FOR FOOD AND SPACE IN. MEDAKA.- (QRYZIAS LATIPES) - PISCES,. CYPRINODONTIDAE by JOHN. JOSEPH MAGNUSON B.S., The U n i v e r s i t y of Minnesota,. M.S., The-University of Minnesota, 1958 A Thesis Submitted i n P a r t i a l F u lfilment of The Requirements f o r the Degree of Doctor of Philosophy i n the. Department of Zoology We accept t h i s ' t h e s i s as conforming, to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Junej 1961 4 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree tha t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference, and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s representatives. It i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Zoology  The U n i v e r s i t y of B r i t i s h Columbia, Vancouver 8, Canada. Date May 25, 1961 GRADUATE STUDIES Field of Study: Zoology Population Dynamics P. A. Larkin Fisheries Seminar - Staff Marine Field Course J. R. Adams, B. McK. Bary Comparative Physiology W. S. Hoar Theoretical Population Dynamics P. A. Larkin, N. J. Wilimovsky Other Studies: Fisheries Law G. F. Curtis Fisheries Economics A. D. Scott Introduction to Synoptic Oceanography G. L. Pickardj Introduction to Dynamic Oceanography G. L. Packard! Philosophical Problems , B. Savery ®fye Pmtarsttg of JUnitgir CoLuntbra FACULTY OF G R A D U A T E STUDIES PROGRAMME OF T H E F I N A L O R A L E X A M I N A T I O N FOR THE DEGREE OF D O C T O R O F P H I L O S O P H Y of JOHN JOSEPH MAGNUSON B.S. University of Minnesota M.S. University of Minnesota IN ROOM 3332, BIOLOGICAL SCIENCES BUILDING FRIDAY, JUNE 9, 1961 at 9:30 a.m. COMMITTEE IN CHARGE DEAN W. H. GAGE, Chairman P. A. LARKIN G. L. PICKARD I. McT. COWAN B. SAVERY P. A. DEHNEL A. D. SCOTT J. R. ADAMS M. ADASKIN External Examiner: DR. G. H. ORIANS University of Washington AN ANALYSIS OF AGGRESSIVE BEHAVIOR, GROWTH, AND COMPETITION FOR FOOD AND SPACE IN MEDAKA (ORYZIAS LATIPES)— PISCES, CYPRINODONTIDAE ABSTRACT The role and consequences of aggressive behavior in competi-tion for food and space were studied among laboratory populations of juvenile medaka. Growth rate and relative condition were used to measure the success of an individual fish in different competitive situations. Both were measured relative to sibs of the same age and size raised in isolation under the same conditions. Temperature, day length, and light intensity were held constant, and fresh water was circulated. All fish were raised in nylon baskets (30 meshes/cm) suspended into a common water bath. Length or weight, or both, of each fish was measured every 6 days for at least 24 days between 0 and 66 days after hatching. Quantitative records of aggression, activity and location preference were taken throughout the day. Paramecium, Artemia salina nauplii and pellets (diameter = 0. 25 mm to 0.5 mm) were used as food. A. salina were asumed to. be "in excess" if active nauplii were present at all times. Limited food was 10 pellets per fish per day (0.68 mg per fish per day). Growth was followed for 648 fish in populations of 1, 2, 4, 8, or 16 fish in 1, 4, or 8 liter baskets. No difference in average growth was observed at different densities, and growth depensation was no greater than would be expected from genetic differences in growth potential, as long as food was supplied "in excess" and the accumulation of waste products was prevented. Aggressiveness was at a low level, and both large and small fish were equally aggressive. Spatially localizing excess food did not alter the relationship. When food supply was limited, a social hierarchy developed in which large fish were socially dominant, chased small fish away from food, and grew faster than small fish. Aggressive actions increased in frequency just after limited food was presented. If food was localized spatially the social hierarchical society changed into a territorial society in which the dominant defended the food area, and the dominant's competitive advantage increased. Ag-gressive behavior was initiated by an internal state of "hunger" and the presence of food stimuli and smaller medaka. Visual isolation between competitors increased the dominant's advantage if food was contagiously distributed, but decreased it if food was evenly distributted. When food was evenly distributed and the environment had a semi-isolated subsection for each fish, both dominant and subordinate grew equally well. If population size was large the dominant could not chase all subordinates from the food area, and consequently the growth advantage of social dominance was in part lost. In addition fre-quency of aggressive actions by the dominant decreased. Aggressive behavior only dispersed medaka through the habitat if food was evenly distributed. Small fish could not eat pellets as fast as large fish and if all fish had equal access to the limited food supply the rate at which they ate was important in determining their growth rate. Action of aggressive behavior as a competitive mechanism for space or Lebensraum and the influence of environment on both the expression of aggressive behavior and the extent to which it reserves the food supply is discussed. Applicability of these findings to field situations and other species of fish is also considered. ABSTRACT The r o l e and consequences.of aggressive behavior i n competition, f o r food and space were studied among laboratory populations of juvenile medaka. Growth rate and r e l a t i v e condition were used to measure the success of an i n d i v i d u a l f i s h i n d i f f e r e n t competitive s i t u a t i o n s . Both were measured r e l a t i v e to sibs of the same age and si z e raised i n i s o l a t i o n under the same conditions. Temperature, day length, and l i g h t i n t e n s i t y were held constant, and f r e s h water was c i r c u l a t e d . A l l f i s h were -raised in. nylon baskets (30 meshes/cm) suspended i n t o a common water bath. Length or weight, or both, of each f i s h was measured every 6 days, f o r at l e a s t 24 days, between 0 and 66 days a f t e r hatching. . Quantitative records of aggression, a c t i v i t y , and l o c a t i o n preference were taken throughout the. day. Paramecium, Artemia  sa-lina n a u p l i i and p e l l e t s (diameter = 0.25 mm.to 0 .5 mm) were used as food. A. s a l i n a were assumed to be- " i n excess" i f active n a u p l i i were present at a l l times. Limited food was 10 p e l l e t s per f i s h per day (0.68 mg per f i s h per day). Growth was followed f o r 648 f i s h i n populations of 1, 2, 4, 8, or 16 f i s h i n 1, 4, or 8 l i t e r baskets. No difference i n average growth was observed at d i f f e r e n t d e n s i t i e s , and growth depensation was no greater, than would be expected from, genetic d i f -ferences i n growth p o t e n t i a l , as long, as food was. supplied "i n . excess" and the- accumulation of waste products was prevented. Aggressiveness was at a low l e v e l , and both large and small, f i s h were equally, aggressive. S p a t i a l l y l o c a l i z i n g excess food did. not a l t e r the r e l a t i o n s h i p . When.food supply was. l i m i t e d a s o c i a l hierarchy developed i n which large - i i i -f i s h were-socially dominant, chased small f i s h away from food; and grew f a s t e r than:.small f i s h . Aggressive, actions increased in. frequency just a f t e r l i m i t e d food was presented-. I f food was l o c a l i z e d s p a t i a l l y the s o c i a l h i e r a r c h i c a l , society changed: in t o a t e r r i t o r i a l society i n which the dominant defended the food area; and the-dominant 1s competitive, advantage, increased. Aggressive behavior was i n i t i a t e d - b y i n t e r n a l state-of "hunger" and the. presence of food stimuli, and'smaller, medaka. Vi s u a l i s o l a t i o n , between, competitors increased:,the. dominant's advantage i f food was. contagiously d i s t r i b u t e d , but decreased:it i f food was evenly d i s t r i b u t e d . When-food was. evenly d i s t r i b u t e d and. the environment had a semi-isolated subsection f o r each f i s h both.dominant and subordinate grew equally w e l l . I f population si z e was large the dominant could not chase-all sub-ordinates from the food area, and consequently the growth advantage of s o c i a l dominance, was i n - p a r t l o s t . In, addition frequency of aggressive actions by the dominant decreased. Aggressive behavior only dispersed medaka through the-habitat i f food was evenly d i s t r i b u t e d . Small, f i s h could not eat p e l l e t s as f a s t as large, f i s h and i f a l l f i s h had equal access to the. l i m i t e d food supply the rate at which they ate was important i n determining t h e i r growth rate. Action of aggressive behavior as>a competitive mechanism f o r space or Lebensraum and the influence of environment on both the expression of aggres-sive, behavior- and the extent to which i t reserves the food supply i s discussed-. A p p l i c a b i l i t y of these f i n d i n g s to f i e l d s i t u a t i o n s and other species of f i s h i s also considered. ACKNOWLEDGMENTS The author wishes to express h i s gratitude to Dr. Peter A. Larkin, who offered valuable counsel and stimulation.throughout the. study and during..the preparation of the manuscript. Special thanks are due. to Dr. William. S. Hoar and Dr. Casimir C. Lindsey. f o r t h e i r i n t e r e s t i n the study and t h e i r c r i t i c i s m of the manuscript; Thanks are also due to Dr. Stanley. W. Nash and Dr. Lorraine Schwartz f o r advice on s t a t i s t i c a l methods,, to Dr. C y r i l V. Finnegan and Dr. A. J . Wood f o r c r i t i c a l l y reading the manuscript, to Mr. Taizo Miura f o r t r a n s l a t i n g Japanese papers and communicating with Japanese s c i e n t i s t s , to Mr. John Geoffrey Eales.and.Mr. Taizo. Miura f o r assisting, with laboratory observations, to Mr. Md. Youssouf A l i f o r a s s i s t i n g i n c u l t u r i n g medaka, and to Dr.. John C. Briggs f o r suggesting the experi-mental animal. The project was financed by the National Research Board of Canada,,, and research f a c i l i t i e s were provided by the Vancouver Public Aquarium Association.. The assistance and cooperation of the s t a f f of the public aquarium was g r e a t l y appreciated. TABLE OF CONTENTS Page INTRODUCTION. 1 MATERIALS. AND METHODS 4 Experimental Animal , . • 4 Laboratory I n s t a l l a t i o n s 4 Physical Constants 6 Breeding and Hatching 6 Foods and Feeding 7 Length and Weight. Measurements - 7 Q u a n t i f i c a t i o n of Behavior 8 COMPETITION FOR SPACE ,(EXPERIMENT I) 10 Introduction 10 Description of Experiment , 10 Results 11 Growth v a r i a b i l i t y among i s o l a t e s 11 Growth v a r i a b i l i t y in.populations r e l a t i v e to control . . . 15 Average growth at d i f f e r e n t d e n s i t i e s 17 Aggressive behavior 18 Summary of Results 19 COMPETITION FOR LIMITED FOOD (EXPERIMENT II) . . 21 Introduction 21 Description of Experiment 21 Results 24 Growth rates 24 Condition (Relative weight) . 29 - V -Relative condition and growth . . . . . J2 Aggressive behavior and a c t i v i t y comparisons J-4 Diurnal rhythms, i n behavior 40 Relation between growth and behavior 47 Summary of Results 5$ COMPETITION FOR EXCESS FOOD (EXPERIMENT I I I ) 60 Introduction and Description of Experiment. 60 Results and Conclusions 60 Summary of. Results 61 COMPETITION. FOR LIMITED FOOD. IN LARGER. POPULATIONS (EXPERIMENT. IV) . . 62 Introduction 62 Description of Experiment 62 Results 65 Growth rates 65 Behavior . 67 Relation between growth and behavior . . . . . 72 Summary of Results ' 75 DISCUSSION 77 SUMMARY OF RESULTS .. 95 LITERATURE CITED. 98 APPENDIX 103 FIGURES Page Figure 1. A water b a t h w i t h nylon baskets (above) and with baskets, and thermostat removed (below). See overlay f o r i n d i v i d u a l items. . . . . Figure- 2. Growth depensation among i s o l a t e d medaka sibs grown under " i d e n t i c a l " environmental conditions, compared, from date of f e r t i l i z a -t i o n . S o l i d l i n e = male, broken, line. = female. . . . . . 13 Figure 3- Growth depensation among isolated, medaka sibs grown under " i d e n t i c a l " environmental conditions, compared- from date of hatching. S o l i d l i n e = male, broken l i n e = female. 14 Figure 4. Diagram of. treatments used i n experiment II showing l o c a t i o n and amount of food, siz e and number of f i s h , and. size and. topography of basket.. 23 Figure 5• . M u l t i p l e comparisons.of s i z e - s p e c i f i c growth rates of large f i s h , small f i s h , and both f o r each treatment, except XF, r e l a t i v e to the s i z e - s p e c i f i c growth rates of controls ( AW - &W)j (any two means not enclosed by the same bracket are d i f f e r e n t , p £ 0 .05) . • =» large f i s h , o = small f i s h , — = mean f o r whole treatment. 28 Figure 6. M u l t i p l e comparisons of r e l a t i v e condition' of large f i s h i small f i s h , and both f o r each treatment* except XF~, as measured, as a deviai-t i p n from the weight on length regression of controls (W-$); (any two means not enclosed by the same.bracket, are d i f f e r e n t , p * 0 .05) . • = large f i s h , o = small, f i s h , . — = mean f o r whole treatment. 31. Figure 7. Relation between mean r e l a t i v e growth and" mean r e l a t i v e condition of large and small f i s h i n each treatment, except XF. o = small f i s h , • = large f i s h . 33 - v i i -Figur.e 8. M u l t i p l e comparisons, of aggressiveness of large f i s h ; small f i s h , and both f o r each treatment as calculated from days 27;21, 59}53> and 5i}^5 (any two means not enclosed by the same- bracket are d i f f e r e n t , p 0 . 0 5 ) . • m large f i s h , O = small f i s h , — = mean for whole treatment. 37 Figure 9. M u l t i p l e comparisons o f a c t i v i t y of large f i s h , .small f i s h , and.both f o r each t r e a t -ment as calculated from days 27}21, 39}33» and 51>4-5 ( a n y two.means not enclosed by the same• bracket are d i f f e r e n t , p € 0 .05) . • = large f i s h , O = small, f i s h , — = mean for whole treatment. 39 Figure 1G. Diurnal changes i n a c t i v i t y and aggressiveness of large and small f i s h i n the. i s o l a t e controls (CL1, OST), i n p l a i n 1 - l i t e r l i m i t e d food-treatment (L1 ), and i n the excess, food treatment (XF). • = large f i s h , o = small f i s h . 41 Figure 11. Diurnal changes -in a c t i v i t y and aggressiveness of large and: small f i s h i n the no-food, t r e a t -ment (NF), and a c t i v i t y , aggressiveness, and l o c a t i o n preference of large and small f i s h i n the l i m i t e d food l o c a l i z e d on bottom treatment (LB). • = large f i s h , o e> small f i s h . . 42 Figure 12. Diurnal changes i n a c t i v i t y , aggressiveness, and l o c a t i o n preference of large and small f i s h " i n treatments with a p a r t i a l p a r t i t i o n across the basket- which were fed on one side (liP.1) and both sides (LP2) of the p a r t i t i o n . • = large f i s h , o = small f i s h i 4-3 Figure 13- Diurnal changes i n a c t i v i t y and aggressiveness of large and small f i s h in.- shallow l i m i t e d food treatment (LSH.), 4 - l i t e r i s o l a t e t r e a t -ment (0L4, CS4), and p l a i n 4 - l i t e r l i m i t e d food treatment (L4). • » large f i s h , o = small, fish.. . . . . . . . . . . . . . . 44 Figure 14. Differences i n the growth rates of large and small fis h - i n each treatment p l o t t e d against the differences i n their, aggressiveness, during the 2?5 hours a f t e r food. was. presented. 55 - v i i i -Figure 15. E f f i c i e n c y of aggression to the large dominant f i s h as influenced-by (a) amount of food, (b) l o c a l i z a t i o n of food supply, and. (c) size of - environment. E f f i c i e n c y = (GTj-Gs)/(Aggres-sive actions by large f i s h i n 2.5 hours a f t e r food was-supplied) 57 Figure 16. Diagram showing p a r t i t i o n s and food l o c a t i o n s of the 8 - l i t e r baskets used i n experiment IV. (top view) 64 f, 2 v3 _ , 2 Figure 17. Relation between growth depensation lSs t+6' ^ s and (a) the s p a t i a l d i s t r i b u t i o n of l i m i t e d food i n . a subdivided habitat and (b) the extent to which the habitat i s subdivided. (Brackets enclose means which do not d i f f e r at p i 0.05) 66 Figure 18. Relation .between, frequency of aggressive actions i n populations of 8 f i s h . a n d ( a ) the s p a t i a l d i s t r i b u t i o n of l i m i t e d food i n a subdivided habitat and (b) the extent to which the habitat i s subdivided i n evenly d i s t r i b u t e d food treatments. (Brackets enclose those-means which are not d i f f e r e n t at p £ 0.05) 69 Figure 19. Average*number o f . f i s h in. subsections, during the 5 minutes a f t e r food was introduced. 71 TABLES Page Table 1. Design of experiment I showing treatments and number of r e p l i c a t i o n s . 11 Table; 2. Mean 12-day length increments., (mm), of male and female sibs grown i n i s o l a t i o n under " i d e n t i c a l " conditions calculated f o r f i s h . 10 mm, 15 mm, 20 mm, and 25 mm long. 12 Table 3. Design, of experiment II showing treatments, and number of replications,, with.treatment codes and de s c r i p t i o n s . . . « 22 Table 4. Linear regressions.of. (W-t+6 - Wt) on W^. i n milligrams f o r each 6-day i n t e r v a l f o r large and small i s o l a t e d f i s h i n 1-liter, baskets, 0L1 and GS1, while food supply was: l i m i t e d (y * W t + 6 - Wt , x = Wf). 25 Table 5* Average s i z e - s p e c i f i c growth rates ( AW - £W; and rela t i v e , conditions (W - ft) of large and small f i s h and of the whole treatment r e l a t i v e to the growth and condition of control f i s h of the same s i z e , presented f o r each treatment, except XF. . 26 Table 6. Average aggressiveness:and a c t i v i t y during the daylight hours f o r large and small f i s h in-each treatment and.' f o r the whole treatment as calculated from days 27;21, 39;33, and51;45. 36 Table 7- Differences i n average growth, condition, aggressiveness, and a c t i v i t y between—large, and small f i s h i n each treatment averaged f o r days. 27;21, 59;33, and 51 j45 52 Table 8. Simple, p a r t i a l , and multiple: c o r r e l a t i o n c o e f f i c i e n t s f o r y-| = (GT J -GS ) * Y2 " (%-^s)> *1 - (AggTy-Aggs) and x 2 - ( A c t L - A c t s ) . (See text f o r explanation.) 5^  Table 9. Design of experiment IV showing treatments and number of r e p l i c a t i o n s with treatment codes and de s c r i p t i o n s . . . . . . . 63 - X -Table 10. The average increase i n v a r i a b i l i t y of weight-frequency d i s t r i b u t i o n s during,.a 6-day period due to differences i n growth rate i n populations of 8 f i s h 65 Table-11. Frequency of aggressive actions/2.5 min by 8 f i s h and the c o e f f i c i e n t of v a r i a t i o n (%) r e s u l t i n g from differences i n frequency of aggression on d i f f e r e n t days, of. observation, during the 5 minutes a f t e r food was provided. 67 Appendix Table. Mean.values of r e l a t i v e condition f o r large and small f i s h i n each, treatment, except XF, expressed, as (W-W) and (log 1 0W. - l^ gToW) 105 INTRODUCTION Although growth depensation—' i s a common phenomenon i n f i s h , i t i s d i f f i c u l t to determine whether i n t r a s p e c i f i c competition i s one of i t s causes. Hubbs and Cooper (1935) observed growth depensation i n two centrar-chid species and postulated that e i t h e r : ( i ) l a r g e r f i s h had a competitive advantage over smaller f i s h f o r food; ( i i ) ; t h e r e was a consistent difference i n the h a b i t a t of f i s h j ( i i i ) f a s t growth i n one year had a p h y s i o l o g i c a l e f f e c t upon growth rate i n the following year; or ( i v ) there were genetic dif f e r e n c e s i n growth p o t e n t i a l s within the populations. Lar k i n et a l . (195^) observed a negative as s o c i a t i o n between average annual growth increments and the annual change i n v a r i a b i l i t y of size f r e -quency d i s t r i b u t i o n s within year classes of rainbow trout, Salmo gairdneri Richardson (Salmo gairdneri kamloops Jordan); and concluded that both the mean size and the changes i n v a r i a b i l i t y are measures of competitive i n t e n s i t y . Brown (1946, 1951> 1957) observed growth depensation i n laboratory populations of young-of.-the-year brown trout, Salmo t r u t t a Linnaeus, and postulated p h y s i o l o g i c a l 11 stress" as a mechanism which res u l t e d i n poor growth among the smaller f i s h due to the presence of larger f i s h even when food was " i n excess." The influence of a water-borne growth i n h i b i t i n g agent described by Richards (1958), Rose ( i 9 6 0 ) , and West ( i960) also r e s u l t s i n a reduced growth rate under more crowded conditions and i n growth depensation 1/ Growth depensation r e f e r s to the increase i n variance of a size-frequency d i s t r i b u t i o n with time due to differences i n growth rates (Ricker, 1958)• -2-among Rana pipiens tadpoles. Aggressive behavior expressed as s o c i a l hierarchy or t e r r i t o r i a l i t y has often been postulated as a mechanism of competition f o r food and space which, benefits the larger animals and therefore r e s u l t s i n growth depensation ( A l l e e et a l . 1948; C o l l i a s , 1944; Kalleberg, 1958; Kawanabe, 1958; M. Newman, 1956; Noble, 1939; Noble and Borne, 1938). Carpenter (1958) l i s t e d increased access to food and to space* per se, as two possible func-tions of t e r r i t o r i a l i t y , but stated that these functions are ascribed "mainly on i n f e r e n t i a l basis., since c o n t r o l l e d experimentation has not yet c l e a r l y defined and delineated the area of i t s ^ t e r r i t o r i a l i t y ' s ^ possible e f f e c t s . " Aggressive behavior, the mechanism studied i n t h i s paper, has not been i s o l a t e d previously as a competitive mechanism which r e s u l t s i n growth depensation (see discussion) because: ( i ) genetic differences i n growth p o t e n t i a l have been assumed n e g l i g i b l e ; ( i i ) some of the- associations between growth rate, r e l a t i v e s i z e , and appetite would e x i s t i n the absence of aggressive behavior; ( i i i ) s o c i a l rank i s l a r g e l y determined by r e l a t i v e s i z e , and conclusions that s o c i a l rank determines r e l a t i v e size may be a m i s i n t e r p r e t a t i o n of the c o r r e l a t i o n ; ( i v ) supplying food " i n excess" i s a r e l a t i v e condition and may be misleading i f considered as an absolute. I f aggressive behavior i s an important competitive mechanism, i t s action should be i s o l a t e d under d i f f e r e n t environmental conditions, and i t s u t i l i t y to the animal should be measured i n the quest of s p e c i f i c resources. I f growth depensation i s to be used, as a measure of competitive i n t e n s i t y i t i s necessary to remove the extraneous influence of other f a c t o r s , such as genetics and conditioning of the water. In the following experiments, the r o l e and consequences of aggressive -5-behavior as a mechanism of i n t r a s p e c i f i c competition f o r food and space are inve s t i g a t e d . Attention i s given to the influence of such f a c t o r s as the r e l a t i v e size of a f i s h , the amount and the s p a t i a l d i s t r i b u t i o n of the food, the amount of s p a t i a l i s o l a t i o n between competitors, population density, and population s i z e . Growth rate and r e l a t i v e condition are used to measure the success of i n d i v i d u a l f i s h i n d i f f e r e n t competitive s i t u a t i o n s . In experiment I the e f f e c t s of crowding on growth and growth depensa-t i o n are inve s t i g a t e d when food i s supplied " i n excess." In experiment II the e f f e c t s of l i m i t e d food are studied. Modifying f a c t o r s such as the topography of the habitat and the amount and s p a t i a l d i s t r i b u t i o n of the food are considered. In experiment I II the consequences of l o c a l i z e d supplies of excess food are investigated. In experiment IV the action and consequences of aggressive behavior are examined among larger populations i n more complicated environments. In a l l cases b i o l o g i c a l conditioning of the water i s removed as a-factor, genetic differences are accounted f o r , and aggressive behavior i s studied as a possible competitive mechanism. MATERIALS AND METHODS Experimental Animal A domestic stock of "golden" medaka, Oryzias l a t i p e s (Temminck and Schlegel), a small cyprinodont f i s h , was used as an experimental animal. Briggs and Egami (1959) outline some features which make i t a useful laboratory animal. Medaka are e a s i l y r a i s e d , mature at lengths, near 27 mm, and breed r e a d i l y on successive days f o r several months. Juveniles e x h i b i t a g o n i s t i c behavior, are t o l e r a n t to starvation, and can be weighed and measured a l i v e with l i t t l e handling m o r t a l i t y . Laboratory I n s t a l l a t i o n s Controlled environment apparatus was b u i l t i n the Vancouver Public Aquarium, Vancouver, B.C., i n the summer, 1959» and the experiments were con-ducted p e r i o d i c a l l y from August, 1959> through March, 1961. The laboratory contained six controlled-temperature water baths (213.4 l i t e r s ) with adjust-able water inflows (Figure 1). .The medaka populations were placed i n 1- or 4 - l i t e r baskets made from nylon "horsehair" c r i n o l i n e (10 meshes/cm) with nylon c h i f f o n l i n e r s (50 meshes/cm). For some treatments the 1 l i t e r baskets were divided i n t o two equal sections by a p a r t i a l p a r t i t i o n made from nylon "horsehair" c r i n o l i n e with a 5 x 4 cm hole i n i t . A l l baskets were suspended in t o the water baths. Each water bath held either eight 4 - l i t e r baskets (20 x 20 x 10 cm) plus two 1 - l i t e r baskets (10 x 10 x 10 cm), or t h i r t y - e i g h t 1 - l i t e r baskets. Movable opaque b a f f l e s penetrated the upper 10 cm of water and prevented f i s h v i s i o n between baskets. Glass and wax paper covers, placed over each basket, prevented undue interference- from the experimenter. BIMETAL THERMOSTAT / ' 4-LITER BASKET 1-LITER BASKET • / WATER INLET / STAND PIPE DRAIN / STAINLESS STEEL AIR STONE . • . HEATING ELEMENT / , / • • MOVABLEOPAQUE BAFFLES Figure 1. A water bath with nylon baskets (above) and with baskets and thermostat removed (below). See overlay f o r i n d i v i d u a l items. -6-Fluorescent l i g h t s were arranged evenly over the water baths, and a time clock c o n t r o l l e d day length. A behavior observation water bath (90 l i t e r s ) was b u i l t with an i n c l i n e d mirror above i t . This bath held either two 4-liter baskets and two 1 - l i t e r baskets or ten 1 - l i t e r baskets, and had the same l i g h t and temperature c h a r a c t e r i s t i c s as the s i x mentioned above. Behavior was observed through a narrow s l i t i n a masking screen. Physical Constants A l l experiments were conducted i n fresh dechlorinated water from the Vancouver c i t y water supply, at a 16 hour daylength, and at a l i g h t i n t e n s i t y (water surface) which, slowly decreased from 19-23 f t - c at the beginning of the experiments to 13-18 f t - c at the end due to exhaustion of the fluorescent tubes. Water temperature was held at 24°C with a maximum deviation between days and baths of ±0.4 degrees and a usual deviation of < ±0.1 degrees. Between basket locati o n s i n a bath the range was < 0.1 degrees at any one time. A water flow of 4.5 l i t e r s per hour provided a minimum volume exchange of 9y% every 4 days. Each basket was washed under a faucet every 6 days when the f i s h were removed f o r measurements. Three a i r stones created a s l i g h t current i n the baskets and e s p e c i a l l y i n the small baskets near the center of the bath (Figure 1). Breeding and Hatching The brood stock was held at 24°C. One pair was used i n experiment I, three p a i r s i n I I , and one pair i n I I I , and many pa i r s i n experiment IV. Breeding occurred soon a f t e r the l i g h t s went on, and the eggs adhered to the female's abdomen. Eggs were removed immediately by the experimenter, and each egg was separated. Malachite green (1:100,000) was used p e r i o d i c a l l y to i n h i b i t fungus during incubation, and egg mo r t a l i t y was l e s s than. ^ >%. Hatch-ing occurred n a t u r a l l y i n experiment I, but was.induced 23 days a f t e r f e r t i -l i z a t i o n i n I I , and 11 days a f t e r f e r t i l i z a t i o n i n I I I , and 10 days a f t e r f e r t i l i z a t i o n i n IV, by f l u c t u a t i n g the water temperature between 24 and 28°C. The l a t t e r procedure; r e s u l t s in. a uniform hatching date. Foods and Feeding In a l l experiments, l i v i n g Paramecium sp. were fed to the newly-hatched medaka three times each day f o r at l e a s t 7 days. After the t h i r d day, l i v i n g brine shrimp n a u p l i i , Artemia s a l i n a , were also supplied three times each day eith e r u n t i l the end of the experiment or u n t i l a small p e l l e t e d food was substituted. I f active, n a u p l i i were s t i l l present i n the basket before each feeding, food was assumed to be i n excess. The p e l l e t e d food was prepared 2/ from "Dina-fish; super-fry r a t i o n " — by grinding and sieving u n t i l a homo-geneous c o l l e c t i o n of small sized p a r t i c l e s was obtained which would pass through a 0 .5 mm but not a 0.25 mm'.soil sieve. A l l f i s h i n l i m i t e d food treatments were fed 10 p e l l e t s per f i s h per day (0 .68 mg per f i s h per day). This amount was eaten i n l e s s than f i v e minutes. P e l l e t s f l o a t e d u n t i l they were f o r c e f u l l y nudged by a f i s h at which time they sank i f not eaten. The f i s h swam to the bottom and ate the p e l l e t s which had sunk a f t e r the surface food was eaten. I f a bottom p e l l e t was desired, the p e l l e t s were dampened before being placed i n the basket. Length and Weight Measurements Small f i s h were handled e n t i r e l y with eyedroppers. As the f i s h grew 2/ Manufactured by W i l l i s H. Small Feed Co., Eugene, Oregon. l a r g e r , p l a s t i c tubes with larger bores and suction bulbs were used. Total f i s h length was measured under magnification with a c a l i b r a t e d movable substage. A f i s h was placed on i t s side i n a V-shaped groove between two p a r a l l e l glass tubes l y i n g on the bottom of a small glass p e t r i dish. The water i n the dish just covered the f i s h , and surface tension held the f i s h i n place. No anesthetic was necessary. The t i p of the f i s h ' s snout was aligned at a f i x e d point i n the f i e l d , and the displacement of the movable substage which was necessary to a l i g n the t i p of the f i s h ' s t a i l at the f i x e d point i n the f i e l d was the measure of t o t a l f i s h length. C a l i b r a t i o n s on the substage vernier were to 0.1 mm, and when care was taken to o r i e n t the f i s h i n a con-s i s t e n t manner, measurements of t h i s accuracy could be r e p l i c a t e d exactly. Total weight of an i n d i v i d u a l f i s h was measured on a Mettler H-5 balance. A f i s h was placed i n a p l a s t i c cylinder (8 mm i n diameter and 12 mm high) covered with s i l k b o l t i n g c l o t h on the bottom. Water was blo t t e d out through the b o l t i n g c l o t h onto paper toweling and adhering droplets were removed with a small paper b l o t t e r . The f i s h plus cylinder were weighed, and f i s h weight was determined by subtraction. Weights were made to the nearest 0.1 mg and could be r e p l i c a t e d within ±0.6 mg with 95% confidence. Each f i s h was measured twice to reduce measurement er r o r . A f i s h could be weighed and measured i n a 5-minute period during which i t was out of the water f o r about JO seconds (time on the balance pan). When the f i s h was placed back i n i t s basket i t appeared to behave normally and would take food within 5 minutes. Quantification, of Behavior In experiment I and IV observations were made with the baskets i n s i t u , but i n experiment II selected baskets with t h e i r f i s h were removed from t h e i r bath and placed i n the observation bath 8 hours p r i o r to observation. An observation consisted of a 2 . 5-iainute period and was immediately r e p l i c a t e d . A g r i d of strings ( 5 cm i n t e r v a l s and p a r a l l e l to sides of the baskets) was placed over each basket. A c t i v i t y was recorded as the number of times a f i s h swam under the s t r i n g s . Agonistic behavior was recorded as a t o t a l count of aggressive actions by each f i s h (nips, chases, f r o n t a l and l a t e r a l t h r e a t s ) . Location preference was measured by accumulating the time a f i s h spent, i n s p e c i f i c l o c a t i o n s and by expressing t h i s as a percentage of the observation period. In experiment I the records f o r d i f f e r e n t f i s h i n the same basket were taken i n consecutive time periods, but i n experiment II. a l l records f o r a l l f i s h i n a basket were taken synchronously. COMPETITION FOR SPACE (EXPERIMENT I) Introduction Aggressive behavior expressed as s o c i a l hierarchy or t e r r i t o r i a l i t y i s often postulated as a mechanism used i n competition f o r space, per se ( i . e . where space i s a vague and undefined something, not i n c l u d i n g any s p e c i f i e d resources i n l i m i t e d supply, but which animals nevertheless compete f o r with density dependent consequences). This hypothesis was explored by comparing v a r i a b i l i t y i n growth within populations of d i f f e r e n t d e n s i t i e s a f t e r removing the v a r i a b i l i t y which resulted- from genetic differences within the populations and. by comparing average growth rates at d i f f e r e n t d e n s i t i e s . Aggressive behavior and s o c i a l h i e r a r c h i e s were investigated to determine whether the postulated mechanism was present. Description of Experiment Medaka were reared at 1, 4, and 16 f i s h per l i t e r , and i s o l a t e s were raised as controls to estimate genetic v a r i a b i l i t y of the stock (Table 1). E f f e c t s of b i o l o g i c a l conditioning were removed not only by. c i r c u l a t i n g new water, but also by allowing the water i n each bath to c i r c u l a t e - f r e e l y through baskets containing a l l . population d e n s i t i e s and i s o l a t e d controls. V a r i a -b i l i t y i n size-frequency d i s t r i b u t i o n s occurring i n the populations and exceeding the genetic base was considered to r e s u l t from i n t e r a c t i o n s between f i s h i n the population. Eggs from a single p a i r on 10 consecutive spawning days were used. Each day's eggs were randomly a l l o c a t e d to treatments with the r e s t r i c t i o n s that a single r e p l i c a t e only contained eggs from the same spawning day, and that -11-Table 1. Design of experiment I showing treatments and number of r e p l i c a t i o n s . Size of Population Size Baskets 1 4 16 1 l i t e r No. f i s h No. baskets Density 18 18 control f i s h 48 12 4/1iter 96 6 16/liter 4 l i t e r s No., f i s h No. baskets Density 18 18 control f i s h 48 12 l / l i t e r 96 6 4 / l i t e r Total f i s h Total baskets 324 72 eggs from each spawning day were d i s t r i b u t e d proportionately to a l l t r e a t -ments. Treatments were randomly al l o c a t e d to p o s i t i o n s i n water baths with the r e s t r i c t i o n that equal proportions of a treatment's r e p l i c a t e s were placed i n each of the s i x water baths. Total f i s h length was measured every 6 days f o r 66 days a f t e r hatching, except f o r populations of 16 which were measured every 6 days f o r the f i r s t 24 days, and every 12 days thereafter. Quantitative observations of a c t i v i t y and aggressive behavior were made at 1-hour i n t e r v a l s and just a f t e r each feeding approximately 30 & n d 60 days a f t e r hatching. One basket from each treatment was observed. In populations of 4 or 16 f i s h , records were kept f o r a large, medium, and small sized f i s h . Results Growth V a r i a b i l i t y Among Isolates Even medaka of the same age and parentage varied i n size when grown - 1 2 -under environmental conditions as s i m i l a r as possible with the f a c i l i t i e s used (Figure 2 ) . Much: of t h i s v a r i a t i o n i n size was a r e s u l t of variable hatching dates among eggs f e r t i l i z e d on the same day. Yet when f i s h were compared, using hatching date as age zero, v a r i a t i o n s i n si z e s t i l l resulted from genetic differences i n growth p o t e n t i a l (Figure- 5 ) - Each f i s h also tended to maintain i t s r e l a t i v e p o s i t i o n i n the size d i s t r i b u t i o n of i s o l a t e s graphed together f o r comparison. Some v a r i a t i o n i n growth might have been due to a difference i n growth 3 / p o t e n t i a l between the two sexes. The 12-day length increments— of a l l i s o l a t e d males were compared with the increments of a l l i s o l a t e d females when the f i s h were 10 mm, 15 mm, 20 mm, and 25 mm long (Table 2 ) . No d i f f e r -ence existed i n 12-day length increments between the sexes at 10, 15i and 20 mm lengths (sign t e s t , n = 16, p = ca.O .50. i n each case), but females grew Table 2 . Mean 12-day length increments (mm) of male and female sibs grown i n i s o l a t i o n under " i d e n t i c a l " conditions calculated f o r f i s h 10 mm, 15 mm> 20 mm, and 25 mm long. Sex n i Length of f i s h (mm) 10 15 20 25 M 16 6.7 6.1 5.1 3 . 2 F 16 6.7 6 . 0 5-5 4 . 0 Mean 6.7 6 . 0 5-2 3 .6 3/ A '12-day increment f o r each f i s h at each of these sizes was determined g r a p h i c a l l y by pl o t t i n g . t h e length of a f i s h at age t + 12 days against the length of the same f i s h at age t (Waiford p l o t ; Ricker, 1958). The points were connected by s t r a i g h t l i n e s , and the distance between t h i s broken l i n e and a 45° l i n e passing through the o r i g i n ( l i n e of no growth) i s a 12-day increment of a f i s h at any p a r t i c u l a r length at age t . -13-40 Figure 2 . Growth depensation among i s o l a t e d medaka sibs grown under " i d e n t i c a l " environmental conditions, compared from date of f e r t i l i z a t i o n . S o l i d l i n e = male, broken l i n e = female. -14-Figure 5- Growth depensation among i s o l a t e d medaka sibs grown under " i d e n t i c a l " environmental conditions, compared from date of hatching. S o l i d l i n e = male, broken l i n e = female. f a s t e r than males at 25 mm lengths (sign t e s t , n = 16, p = 0.04). When females reached approximately 28 mm i n length, they began to l a y eggs, and the reduced growth rate among the males occurred at the onset of sexual a c t i v i t y . As males and females grew larger, the successive 12-day length i n c r e -ments decreased (Table 2). The 12-day increments were l e s s a t 15 mm than at 10 mm, l e s s at 20 mm than at 15 mm, and less at 25 mm than at 20 mm. This was true of both sexes f o r each of the above comparisons (sign, t e s t , n = 16, p < 0.02 i n each case). Growth V a r i a b i l i t y i n Populations Relative to Control Growth rates could be traced f o r i n d i v i d u a l isolated, control f i s h , but not f o r each f i s h i n populations, because the l a t t e r could not be i n d i v i d u a l l y i d e n t i f i e d at successive measuring days. Since d i r e c t growth comparisons of large or small i s o l a t e d control f i s h could not be made with large or small f i s h i n the treatment populations, a measure of growth depensation was developed to compare the treatment populations with the controls. This measure of divergence among treatment populations, when compared to the con-t r o l , was i n d i c a t i v e of the v a r i a b i l i t y i n growth rates exceeding that expected from genetic v a r i a b i l i t y i n the stock. A l l comparisons were made between groups of immature f i s h growing from 10 mm. to 20 mm i n length, and extraneous sources of v a r i a t i o n such as hatching dates were removed by regression analysis.^ The above was accomplished by making comparisons from a regression of the v a r i a t i o n i n time, a f t e r f e r t i l i z a t i o n f o r a group of f i s h to reach a length of 20 mm (s + ) on the v a r i a t i o n i n time they took to t 2 0  reach a length of 10 mm (s . ). The r e s u l t i n g regressions were l i n e a r i f x10 the cube roots of each variance were p l o t t e d . Isolate control f i s h were divided, randomly i n t o groups of four, the cube roots of the above variances -16-were computed f o r each group, and the l i n e a r regression, of y on x was cal c u l a t e d . The control regression i s : y = 0.46066 + 0.85859 x, where 1_ 1 y = ( s 2 , 9, and x = ( s 2 ) 5 (9596 0.1., n = 8, b = 0.85859 1 0.2415 *20 10 and y = 2.7135 * 0.3624; s x = 1.6051, s y x = 0.4197, s y = 1.4318). Values of y and x were also calculated f o r each population of 4 and 16 f i s h , and the average deviation of y from the control regression was determined. A s i g n i -f i c a n t p o s i t i v e deviation would i n d i c a t e a more rapid growth rate among larger f i s h i n a population r e l a t i v e to the smaller f i s h than would be expected from genetic differences alone. The mean deviation from the control regression was +0.33 C^ g = °'24, p > 0.25) f o r populations of. 4 in. 4-liter baskets, -0.047 (tg = 0.33, P > 0'25) for populations of 4 i n 1 - l i t e r baskets, +0.065 ( t ^ = 0.79, P > 0.25) f o r populations of 16 i n 4-liter baskets, and +0.183 ( t ^ = 1.81, p = 0.13) f o r populations of 16 i n 1-liter baskets. None of these mean deviations were d i f f e r e n t from zero and populations of high density were no more v a r i a b l e than would be expected from genetic differences i n growth p o t e n t i a l . This conclusion was reached even though the t t e s t used 4/ was biased towards decisions of s i g n i f i c a n t d ifferences where none existed.— The above analysis demonstrated that growth depensation was no larger or smaller i n populations of high density with le s s space per f i s h than among i s o l a t e d control f i s h . Larger f i s h had no competitive advantage f o r space, per se, when.food was provided i n excess, and smaller f i s h did as well i n groups as i n i s o l a t i o n . 4/ The control regression, was. assumed to be- a parameter, and average deviation from regression f o r each treatment was tested against zero using a t t e s t . The estimate of standard error of these means was calculated from the deviations of the treatment populations from the control regression. Average Growth at D i f f e r e n t Densities A general reduction i n growth i s often used to measure the consequences of competition when a s p e c i f i c resource i s l i m i t e d . Mean growth rates' i n mm/day f o r f i s h growing between 10 mm and. 20 mm were calculated f o r the con-t r o l s and f o r each population i n each treatment. Growth rates of the i s o -l a t e d controls were averaged by fours to get a range of v a r i a t i o n s i m i l a r to the v a r i a t i o n s i n mean growth between populations within a treatment. The data were analyzed i n f a c t o r i a l design with 2 l e v e l s of basket si z e and 3 l e v e l s of population s i z e . The 0.52 mm/day growth rate of a l l f i s h i n 1-liter baskets was not d i f f e r e n t from the 0.53 mm/day growth rate of f i s h i n 4 - l i t e r baskets (^ 1,32 = 0*826; p = 0.40), but the 0.53 mm/day growth f o r a l l i s o l a t e s and 0.53 nam/day f o r populations of 4 f i s h were d i f f e r e n t from the 0.51 mm/day f o r populations of 16 f i s h (Fg ^ = P = 0.04). Interaction between basket size and population size was not s i g n i f i c a n t (F2 ^ ~ 0.0\Q); p > 0.25). Only populations of 16 were growing slower than the controls. Decreasing the amount of space by 4 times did not a f f e c t the growth rate of equal sized populations, but incr e a s i n g the population's size by 4 times, even when space was increased proportionately, reduced the growth r a t e . I t appears that space., per se, i s not the i n f l u e n t i a l f a c t o r . The reason that f i s h i n populations of 16 grew more slowly was not determined i n t h i s experiment, but perhaps i t was a r e s u l t of a s c a r c i t y of food. At the beginning of the experiment the amount of brine shrimp, so l u t i o n was fed i n proportion to the basket size rather than i n proportion to population s i z e . A f t e r about 2 weeks, populations of 16 were able to eat most of the food provided whereas large numbers of dead brine shrimp c o l l e c t e d i n baskets with 1 or 4 f i s h . . The amount of food given to the populations of 16 was doubled, -18-but a f t e r about 1 week a depletion was noticed again. F i n a l l y each popula-t i o n was fed i n proportion to i t s s i z e , and there was no further evidence of food depletion. By t h i s time, however, the f i s h had reached a size of approximately 16 mm. Aggressive- Behavior During experiment I aggressive behavior was observed i n three d i f f e r e n t s i t u a t i o n s . F i r s t , aggressive actions appeared to be more common just before and a f t e r feeding or near l o c a l concentrations of food. Second, medaka satiated with food would l i e motionless at the surface and chase or threaten those f i s h which approached too c l o s e l y . Third, sexually mature males fought vigorously with each other and would often nip females which did not respond during courtship. Behavior patterns used i n these experiments as measures of aggressive behavior are f a m i l i a r aspects of f i s h ethology and are not de-scribed here i n d e t a i l . The r e l a t i o n s h i p s between environment and behavior presented below are from preliminary data, and the same r e l a t i o n s h i p s are demonstrated i n d e t a i l i n experiments II through IV. Aggressive behavior included chases, hips, f r o n t a l threats, and l a t e r a l t h r eats. S o c i a l h i e r a r c h i e s were established in. which larger medaka were most aggressive. Both males and females were able to assume s o c i a l dominance u n t i l the onset of reproductive behavior. Then a new s o c i a l order was formed i n which males dominated females regardless of body s i z e . Subordinate f i s h did o c c a s i o n a l l y nip or threaten a dominant f i s h . A d a i l y rhythm i n behavior was associated with the feeding schedule. A c t i v i t y increased immediately a f t e r food was introduced (sign t e s t , n = 14, p = 0.006) but one hour l a t e r decreased to a l e v e l even lower than before food was introduced (sign t e s t , n = 10, p = 0.001). The medaka were satiated and l a y almost motionless near the surface. Subordinates were oc c a s i o n a l l y -19-chased from the surface by the dominant f i s h , and the smallest f i s h often set up residence near the bottom. Two to three hours a f t e r the f i s h were fed, a c t i v i t y slowly increased as f i s h began to search f o r more food. At the next feeding time there was again a burst of feeding a c t i v i t y , and the pattern was repeated. Aggressive actions followed a diurnal pattern s i m i l a r ,to general a c t i v i t y , but were often more frequent just before a feeding time. A comparison between treatments demonstrated no increase i n aggressive actions when space was more l i m i t e d (populations of higher density). The lowest frequency of aggressive actions, 0.5 per f i s h per 2.5 minutes, was observed at the greatest population density or at 16 f i s h per l i t e r (sign t e s t comparison to next lowest, n = 12, p = O.OOJ). Summary of Results Limited space had no measurable e f f e c t on growth rates of medaka when the influence of b i o l o g i c a l conditioning of the water was eliminated and food was supplied " i n excess." D e f i n i t e s o c i a l h i e r a r c h i e s were established i n which l a r g e r f i s h were dominant, but large dominant f i s h had no negative influence on the growth of small subordinates.. Differences i n size within a population could be accounted f o r by v a r i a t i o n s i n hatching date and genetic v a r i a b i l i t y i n the growth p o t e n t i a l of the stock. Growth depensation occurred among sibs grown i n i s o l a t i o n under " i d e n t i c a l " environmental, condi-t i o n s . There was no i n t e r a c t i o n between these control f i s h , yet each tended to maintain i t s r e l a t i v e p o s i t i o n i n a composite size-frequency d i s t r i b u t i o n . Aggressive actions were not more frequent when space was more l i m i t e d , but occurred at a l l d e n s i t i e s apparently i n response to food. A f t e r food, was presented, medaka fed to s a t i a t i o n , but three or four hours l a t e r a food - 2 0 -stimulus again i n i t i a t e d intense feeding behavior. Apparently, the f i s h were fed i n excess only i n r e l a t i v e terms, f o r ostensibly i f they were fed more frequently they would have eaten more. In summary, aggressive behavior did not appear to be a mechanism used i n competition f o r space, per se. Also l i m i t e d space had no detrimental e f f e c t s upon any member of a- population i f other f a c t o r s u s u a l l y associated with l i m i t e d space were e i t h e r supplied i n excess (food) or eliminated ( b i o l o g i c a l conditioning of water). COMPETITION FOR LIMITED FOOD (EXPERIMENT II) Introduction The purpose of experiment II was to determine-whether large f i s h had a competitive advantage over small f i s h f o r a l i m i t e d food supply when space, per se, was eliminated as a f a c t o r . S p a t i a l d i s t r i b u t i o n of food, amount of food, and degree of environmental i s o l a t i o n between competitors were varied to determine the influence of these modifying f a c t o r s . Aggressive behavior was investigated as a competitive mechanism, and e f f e c t s of b i o l o g i c a l con-d i t i o n i n g were removed as i n experiment I. Description o f Experiment Each population was composed of two sibs with a 6-day difference i n age, and p a i r s of i s o l a t e d controls with 6-day age differences were maintained. A l l f i s h were fed i n excess on brine shrimp n a u p l i i u n t i l 24 days a f t e r the older f i s h hatched or 18 days a f t e r the younger f i s h hatched (day 24;18). At t h i s date older f i s h averaged about 15 nan i n t o t a l length or 5 mm longer than younger f i s h . This size difference persisted u n t i l the end of the experi-ment, and i n d i v i d u a l f i s h , could e a s i l y be i d e n t i f i e d by t h e i r r e l a t i v e s i z e . On day 24;18 a l i m i t e d food supply of 10 p e l l e t s per f i s h per day was im-posed. I f ei t h e r f i s h i n a population had a competitive feeding advantage i t would get 10 p e l l e t s plus some p e l l e t s provided f o r the other f i s h . This s i t u a t i o n could be detected by comparing the growth of each with the grov/th of i s o l a t e d controls of the same age and s i z e . A description of each t r e a t -ment and i t s l e t t e r code are given i n Table 5 and Figure 4. Eggs were used from three sets of spawning days from each of three -22-Table J . Design of experiment II showing treatments and number of r e p l i c a t i o n s , with treatment codes and de s c r i p t i o n s . Treat-ment Code No. F i s h No. of Per Basket Basket  R e p l i -cations Small Large Size Features ( l i t e r s ) Food Total Amount Per Day Location 0L1 CS1 L1 XF NF LB LP1 LP2 LSH CL4 CS4 L4 9 9 9 9 9 6 9 9 9 0 .25 4 4 4 p l a i n p l a i n p l a i n p l a i n p l a i n 50 mm diam. p e t r i dish on bottom incomplete p a r t i t i o n across basket incomplete p a r t i t i o n across basket very shallow p l a i n p l a i n p l a i n 10 p e l l e t s surface 10 p e l l e t s surface 20 p e l l e t s surface excess scattered Artemia s a l i n a 0 20 p e l l e t s wet i n bottom dish 20 p e l l e t s surface, one side of p a r t i t i o n 20 p e l l e t s surface, both sides of p a r t i -t i o n 20 pellets, surface 10 p e l l e t s surface 10 p e l l e t s surface 20 p e l l e t s surface -25-TO P S I D E C L I CS LI N F TOP S IDE L P I L P 2 L S H T O P S I D E L A R G E F I S H S M A L L F I S H F O O D W A T E R L E V E L C L 4 CS4 L 4 Figure 4. Diagram of treatments used i n experiment II showing l o c a t i o n and amount of food, size and number of f i s h , and size and topography of basket. -24-spawning p a i r s . A set of spawning days consisted of 2 spawning days which occurred 6 days apart. Hatching was induced 23 days a f t e r f e r t i l i z a t i o n , and newly-hatched f i s h were all o c a t e d to treatments so that each had a matched p a i r i n every other treatment i n terms of spawning day and parentage. Those which hatched 6 days l a t e r were a l l o c a t e d i n the same way except that they were also matched with the older f i s h by time of hatching. Treatments were randomly a l l o c a t e d to water bath posi t i o n s with r e s t r i c t i o n s that each bath contained an equal proportion of r e p l i c a t e s from, each treatment, and no two r e p l i c a t e s of a treatment occupied the same p o s i t i o n i n d i f f e r e n t baths. Each f i s h was weighed and measured every 6 days from day 18;12 through day 48;42. Eight p a i r s of observations on aggressive behavior, a c t i v i t y , and l o c a t i o n preference were made through the day at 2-hour i n t e r v a l s , and one p a i r of observations was made a f t e r each feeding. This sequence of observations was repeated at 12-day i n t e r v a l s from approximately day 15,9 through day 51}45. Three randomly chosen r e p l i c a t e s were observed from each treatment at each date. Results Growth Rates Growth rate was used to measure the r e l a t i v e success of a f i s h i n d i f f e r e n t competitive s i t u a t i o n s . Isolates reared i n 1-liter baskets served as controls from which the growth rates i n a l l other treatments were com-pared. The regressions, (Wt+(^ - W+^ = a + bW^ , where Wt = weight i n milligrams on day t and (W+^g - W^) = the six-day weight increment, were calculated f o r large and small i s o l a t e s during the f i r s t three 6-day -25-i n t e r v a l s a f t e r l i m i t e d feeding was begun (Table 4).-2' Deviations of the s i z e - s p e c i f i c 6-day weight increments (AW - AW) were calculated from the appropriate control regression f o r each treatment and control f i s h (Table 5)> where AW = the 6-day weight increment of a f i s h of a s p e c i f i c weight, and AW = the expected 6-day weight increment of f i s h of the same weight as determined from the controls. By d e f i n i t i o n , the average growth rate of control f i s h , CL1 and CS1, would be zero. I f there were no competitive i n t e r -actions within treatment populations, average growth rate of f i s h i n any treatment r e l a t i v e to the controls would also be zero. I f a large f i s h i n a population had a competitive advantage, i t s growth rate r e l a t i v e to the Table 4. Linear regressions of (W^ +g - W^.) on Wt i n milligrams f o r each 6-day i n t e r v a l f o r large and small i s o l a t e d f i s h i n 1 - l i t e r baskets, CL1 and CS1, while food supply was l i m i t e d (y - w t + 6 - w t > x - wt)« F i s h Size Interval A f t e r L i m i t i n g Food t = Age / Days \ I A f t e r j \ Hatching/ n i a i n t e r c e p t b slope y s x s y Large 1st 24 9 0.339 0.165 2.933 6.768 2.510 2.601 Small 1st 18 9 1.551 0.218 3-189 5-784 1 .042 1.277 Large 2nd 50 9 0.128 0.120 2.367 8.229 1.415 1.764 Small 2nd 24 9 0.761 0.185 2.911 4.730 1.943 2.018 Large 3rd 56 9 2.179 ; 0.001 2.200 9.311 1.044 0.977 Small 3rd 50 9 0.700 0.146 2.822 5-896 1.246 1.451 5/ The si x regressions i n Table 4 could not be adequately described by a single regression equation because the pooled within groups regression c o e f f i c i e n t (bw) was not equal to the l i n e a r regression c o e f f i c i e n t (bm) of the group means (F-| ^ 47 = 5.98, p = 0.01) (Dixon and Massey, 1956). Even though slopes of the 6 regressions are not s i g n i f i c a n t l y d i f f e r e n t (F5,42 = 0-914, p > 0.25), i n d i v i d u a l slope estimates were maintained because the slopes appeared to be decreasing i n a regular manner as the mean size of the f i s h increased. -26-Table 5' Average s i z e - s p e c i f i c growth rates ( A W - A W ) and r e l a t i v e conditions ( W - V/) of large and small f i s h and of the whole treatment r e l a t i v e to the growth and condition of control f i s h ' of the same si z e , presented f o r each treatment, except XF. Treat- Growth Condition ment Code Fi s h Size ( A W - AW) i n mg d.f. =(^-6)1/ ( W - W ) i n mg n i d.f. = (^-2)-/ CL1 Large 0.00 27 0.00 27 CS1 Small 0.00 27 0.00 26 Mean 0.00 0.00 L1 Large -0.60 20 +0.93 17 Small -2.23 20 -0.70 17 Mean -1 .42 +0.12 XF compared by another method N P Large -2.87 27 -0.33 27 Small ' -2.94 25 -0.79 25 Mean -2.90 -0.66 L B Large -0.27 16 +0.08 16 Small -2.74 16 +O.35 16 Mean -1.50 +0.22 LP1 Large +0.45 18 + 1.15 17 Small -2.38 18 -0.52 17 Mean -O.96 +0.32 LP2 Large -0.21 18 -0.03 18 Small -0.89 18 +0.46 18 Mean -0.55 +0.22 LSH Large -0.33 16 -0.08 16 Small -2.97 16 -0.86 16 Mean -1.65 -0.47 CL4 Large +1 .43 27 +2.52 27 CS4 Small +1.80 27 +2.59 26 Mean +1.62 +2.56 L4 Large +1-37 27 +1.86 27 Small -1.08 27 +0.15 27 Mean +0.14 +1.00 \J Two degrees of freedom were l o s t from each of the 3 regressions used f o r each mean deviation. 2/ Two degrees of freedom were l o s t because a l l values are deviations from 1 regression l i n e . -27-controls would be greater than a small f i s h ' s . Comparisons made i n the above manner are independent of genetic differences, and of p h y s i o l o g i c a l d i f f e r -ences e x i s t i n g between f i s h of d i f f e r e n t ages and s i z e s . Comparisons of average growth rates f o r large f i s h and f o r small f i s h i n each treatment, except XF, were made by multiple comparison techniques (Duncan, 1955? Krammer, 195^)- * n Figure 5 average growth rates r e l a t i v e to the controls are plotted f o r the large f i s h i n a treatment, the small f i s h i n a treatment, and f o r the mean of the large and small f i s h combined. The v e r t i c a l l i n e connects these three values i n a given treatment. Brackets on the l e f t side of the f i g u r e enclose means of large or of small f i s h f o r within or among treatment comparisons which can not be distinguished at the 0.05 p r o b a b i l i t y l e v e l . For example, the small f i s h i n L4 was not growing slower than the large f i s h i n LP2 (p > 0.05)> but was growing slower than the large f i s h i n i t s own treatment (p < O.O5). The brackets on the r i g h t side of the f i g u r e enclose the means f o r a whole treatment which can not be distinguished from the mean f o r another whole treatment at p 6 0.05. For example, the mean growth of f i s h i n LP2 was not le s s than the controls, CL1 and CS1 (p > 0.05), but the mean growth of f i s h i n LP1 was l e s s than the growth of the controls (p < 0.05). Average growth of both f i s h i n a treatment was u s u a l l y l e s s than growth of controls; f o r example, as shown on the r i g h t side of Figure 5, f i s h i n LP1, L1 , LB, and LSH were a l l growing slower than controls CL1 and CS1. Apparently a disoperative i n t e r a c t i o n which depressed t h e i r r e l a t i v e growth rates was occurring between the two f i s h i n most treatment, populations. _ ^ 6/ Pooled variance f o r these comparisons equals 4.548 with d.f. = 2 = 141. Degrees of freedom were divided by two i n appraising tabled p r o b a b i l i t y l e v e l s (Snedecor, 1956) because the pooled variance of small f i s h 3.367 was not equal to the pooled variance f o r large f i s h 5.715. -28-E + 2 + I -0 - I ^ -2 --3 T R E A T M E N T C O D E Figure 5 - Multiple comparisons of s i z e - s p e c i f i c growth rates of large f i s h , small f i s h , and both f o r each treatment, except XF, r e l a t i v e to the s i z e - s p e c i f i c growth rates of controls ( A t f - A W ) ; (any two means not enclosed by the same bracket are d i f f e r e n t , p £ 0 . 0 5 ) . 9 •= large f i s h , O = small f i s h , — = mean for whole treatment. -29-Small f i s h i n a treatment population were u s u a l l y more adversely influenced by the disoperative i n t e r a c t i o n ; f o r example, as shown on the l e f t side of Figure 5> small f i s h were growing slower than large f i s h i n LP1, LB, LSH, L1, and L4. Excluding L4, these small f i s h were growing as poorly as those f i s h which were not fed, as i n NF. There were several notable exceptions to the above generalizations; f o r example, the f i s h i n LP2 grew as well as controls, and small f i s h did not grow slower than large f i s h . Isolated f i s h i n 4-lite r baskets grew f a s t e r than i s o l a t e d f i s h i n 1 - l i t e r baskets. Detailed analyses of these data are given following presentation of behavior data. F i s h i n XF were not comparable to l i m i t e d food i s o l a t e s . Since both large and small f i s h i n XF were fed i n excess throughout the experiment, i t was assumed that they would have equal s i z e - s p e c i f i c growth rates i f there were no i n t e r a c t i o n between them. The technique used f o r comparison was developed by Parker and Larkin (1959) who state that growth can be described c c by an equation comparable i n form to B = W +^^  - W^  , where c i s a constant which changes the slope of the Walford l i n e to +1 (c = 1-x i n Parker and Larkin ) , and B i s the adjusted s i z e - s p e c i f i c growth increment. I f two groups of f i s h have equal c values, the B of each group i s a comparable measure of growth rate which i s independent of f i s h s i z e . Large and small f i s h had a common c of O.38 and the B = 0.868 f o r small f i s h was not d i f f e r e n t from the B = 0.826 f o r large f i s h ( t ^ g = 0.168, p > 0.25). When food was supplied i n excess i n d i v i d u a l growth rates of large and small f i s h were the same, and large f i s h had no competitive advantage. Condition (Relative Weight) Condition or r e l a t i v e weight of a f i s h might be expected to respond to competitive i n t e r a c t i o n s between f i s h . As with growth comparisons, i s o l a t e d f i s h i n 1 - l i t e r baskets, CL1 and CS1, served as controls from which the -50-condition of a l l other f i s h was compared. The r e l a t i o n between weight and length f o r a l l control f i s h - ^ was log 1 QW t = -2.6329 + 3-4455 l o g 1 Q / t , where log^W^ = y = logarithm of weight i n milligrams on day t, and l o g ^ Q ^ = x = logarithm of length i n millimeters on day t (95% C.I., n = 64, b = 3.4455 ±0.4742, y = 1.2119 ±0.0295; s Y = 0.06204, s„ Y = 0.1179, A yA S y = 0.24554). Relative condition of each control and treatment f i s h was measured as (W - W) where W i s the actual weight of a f i s h of a s p e c i f i c A length, and W i s the expected weight of a f i s h of that length according to the control regression. These measures of r e l a t i v e condition (Table 5) w i l l be p o s i t i v e i f f i s h are i n better condition than controls and negative i f they are i n poorer condition. By d e f i n i t i o n the mean r e l a t i v e condition of control f i s h i s zero. LeCren (1950 developed and used a d i f f e r e n t measure of r e l a t i v e condition (x ) . A comparable measure to LeCren's r e l a t i v e condi-V W / tion was t r i e d i n the present study but was rejected. These values and the reasons f o r not using them are presented i n the appendix. Relative conditions of large and small f i s h i n each treatment, except XF, were compared by multiple comparison techniques i n Figure 6 (Duncan, 1955; Krammer, 1956).— No treatment with a l i m i t e d food supply had a mean r e l a -t i v e condition l e s s than the con t r o l s . Yet r e l a t i v e conditions of small f i s h i n treatment populations were often l e s s than the r e l a t i v e condition of large 7/ Length-weight data f o r both large and small control f i s h from day 24;18 through day 42;36 were analyzed by covariance and found to be adequately described by a single regression of the logarithm of weight on the logarithm of length (?2,60 = °'0if» P > 0 ' 2 5 ) . 8/ Pooled variance equalled 1.940, d.f. = 271/2 f o r comparisons between 1 - l i t e r basket treatments; equalled 6.778, d.f. = 73/2 for comparisons between 4 - l i t e r basket treatments; and 2.9668, d.f. = 344/2 f o r compari-sons of 1 - l i t e r versus 4 - l i t e r basket treatments. Degrees of freedom were divided by two i n appraising tabled p r o b a b i l i t y l e v e l s (Snedecor, 1956) because the pooled variance among large f i s h was not equal to the pooled variance among small f i s h . -51-+ 3 i s + 2 r o + | o < r r 2 0 r i Q L_ O 6 1 1 J - I L 6 _ i _ _ J 6 ^ L4 LPI LB LP2 LI LSH NF T R E A T M E N T C O D E Figure 6. M u l t i p l e comparisons of r e l a t i v e condition of large f i s h , small f i s h , and both f o r each treatment, except XF, as measured as a^ deviation from the weight on length regression of controls (W-W); (any two means not enclosed by the same bracket are d i f f e r e n t , p £ 0.05). ® = large f i s h , O = small f i s h , — = mean for whole treatment. -32-f i s h ; f o r example, as shown on the l e f t side of Figure 6 , small f i s h had a lower r e l a t i v e condition i n L4, L1, and LP1. I f there were competitive i n t e r a c t i o n s i n the populations, they did not r e s u l t i n a general lowering of r e l a t i v e condition, but resulted i n a lower condition only among small f i s h i n c e r t a i n cases. When f i s h were fed i n excess, as i n treatment XF, regressions of weight on length f o r large and f o r small f i s h were the same ( F 2 = 2 . 0 2 , p > 0 .10) and large f i s h had no competitive advantage. The combined regression f o r large and small f i s h i n XF, l o g 1 0 W t = -2.7189 +3-5844 l o g 1 0 ^ t (95% 0 . 1 . , n = 87, b = 3.5844 ±0.1271, y = 1.6113 ±0.0143; s x = 0.11302, S y x = 0.0666, S y = 0 .41050) , was compared to the control regression above. The intercepts are not d i f f e r e n t but the regression c o e f f i c i e n t was greater f o r the XF treatment than f o r CL1 and CS1 controls ( t ^ y = 3*28, p < 0 .01) which demon-strates that f i s h fed i n excess were i n better condition than l i m i t e d food controls over the whole range of observed s i z e s . Relative Condition and Growth Mean r e l a t i v e growth of f i s h - s i z e treatment combinations was p o s i t i v e l y associated With mean r e l a t i v e condition (r = 0.84, n = 18, p < 0 . 0 1). This a s s o c i a t i o n also.existed among f i s h within a treatment. The c o r r e l a t i o n c o e f f i c i e n t among f i s h i n CS1 and CL1 was 0.68 (n = 18, p < 0 .01) and among f i s h i n L4 was 0.82 (n = 18, p < 0 . 0 1). Both r e l a t i v e condition and r e l a t i v e growth were measuring the same response but r e l a t i v e growth was more sens i -t i v e to differences i n competitive s i t u a t i o n s . Since both variables were measures of the same response the influence of d i f f e r e n t treatments on large and small f i s h was emphasized i f both were con-sidered at the same time (Figure 7 ) . Each point i n Figure J represents the average of large or small f i s h i n a treatment. In terms of growth and Figure 7. Relation between mean r e l a t i v e growth and mean r e l a t i v e condition of large and small f i s h i n each treatment, except XF. O = small f i s h , 6 = large f i s h . -5*-condition, the f i s h represented by points i n the center of the f i g u r e were comparable to the i s o l a t e d f i s h i n 1 - l i t e r baskets fed 10 pellets/day. Those i n the lower l e f t were smaller and le s s robust than controls while those i n the upper r i g h t were larger and i n better condition than the controls. F i s h fed i n excess (XF) were not p l o t t e d but would be i n the upper r i g h t . The poorest environment, as judged by growth and condition of f i s h , was NF with no food, and the best was XF with excess, food. A l l other baskets received, a l i m i t e d amount of food. Among these, larger 4 - l i t e r environments were more favorable than smaller 1 - l i t e r environments. The large f i s h was one component of a small f i s h ' s environment and likewise the small f i s h was a component of a large f i s h ' s environment. In general the presence of a large f i s h i n l i m i t e d food treatments resulted i n a poorer environment f o r a small f i s h (L1, LB, LSH, and LP1), whereas the presence of a smaller f i s h did not r e s u l t i n a poorer environment for the large f i s h . In one treatment (LP2) the presence of the other f i s h did not reduce the q u a l i t y of the environment fo r e i t h e r the large or the small f i s h . Aggressive Behavior and A c t i v i t y Comparisons Aggressive actions, a c t i v i t y counts, and l o c a t i o n preferences were recorded from two 2.5-minute periods at 2-hour i n t e r v a l s during daylight hours f o r large and small f i s h . Three baskets from each treatment were observed. A c t i v i t y counts were recorded f o r each treatment, aggressive actions f o r a l l except the i s o l a t e s , and l o c a t i o n preferences only i n LB, LP1, and LP2. A d e s c r i p t i o n of each treatment with i t s code number was given i n Table 2 and Figure 4. Average number of aggressive actions/2 . 5 min during the day was c a l -culated f o r the large and the small f i s h i n each treatment on days 27;21, 39;33> a n d 51>^5 by averaging the eighteen 2.5-minute observations made on -35-each f i s h i n a day (Table 6 ) . A mean value f o r the whole treatment was calculated by averaging the large f i s h with the small f i s h f o r a treatment. Large and small f i s h i n each treatment were compared by multiple comparison techniques i n Figure 8 (Duncan, 1955; Krammer, 1956).—' Large f i s h were more aggressive than small f i s h (Figure 8) i n a l l l i m i t e d food treatments (L1, LB, LP1, LP2, LSH, and L4), and i n the no-food treatment (NF), but large and small f i s h were equally aggressive i n the excess food treatment (XF). Large f i s h i n XF were less, aggressive than large f i s h i n a l l other treatments (Figure 8), but were not more aggressive than small f i s h i n the other treatments. A l o c a l i z a t i o n of the food i n a basket resulted i n higher l e v e l of aggressiveness f o r large f i s h . For example, average aggressiveness f o r large f i s h i n LP1 and LB, which had l o c a l i z e d food supplies, was 2.87 aggressive actions/2.5 min while i n L.1, L4, LSH, and LP2, which had dispersed food supplies, i t was only 1.82 aggressive actions/2.5 min (multiple comparisons, n = 15, 45; p £0.05). Another u s e f u l comparison was made by grouping treatments on the basis of previously observed growth rates (Table 5> Figure 5)« A l l treatments i n which large f i s h grew f a s t e r than small f i s h were c a l l e d "competition" treatments (L1, LB, LP1, LSH, and L4), while those i n which both f i s h grew equally well (excluding the i s o l a t e s ) were c a l l e d "no competition" treatments (XF, NF, and LP2). Average aggres-siveness was at a higher l e v e l i n "competition" treatments, 1.24 aggressive actions/2.5 min, than i n "no competition" treatments, 0.71 aggressive 9/ Pooled estimates of variance among f i s h treated a l i k e and associated degrees of freedom used f o r comparisons among small f i s h were s 2 x = 0.132, d.f. = 60; among large f i s h were s 2 x = 1.92, d.f. = 60; between large and small f i s h were s 2 x = 0.972, d.f. = 120/2; and between the means of the large and small f i s h combined were s 2 x = 0.972, d.f. = 120. The degrees of freedom f o r comparisons between large and small f i s h were divided by two i n appraising the tabled p r o b a b i l i t y l e v e l s (Snedecor, 1956) because the variances of small and large f i s h were not equal. -36-Table 6. Average aggressiveness and a c t i v i t y during the daylight hours f o r large and small f i s h i n each treatment and f o r the whole treatment as calculated from days 27;21, 39;33» and y1;hy. Treatment F i s h Aggressiveness A c t i v i t y Code Size n^ (Aggressive Actions / 2 . 5 min) (Counts/2.5 min) CL1 Large 9 - 12.4 CS1 Small 9 _ 12.5 Mean 18 - 12.4 L1 Large 9. 2.12 15.8 Small 9 0.12 10.7 Mean 18 1.12 13.2 XF Large 9 0.44 8.2 Small 9 0.55 8.5 Mean 18 0.59 8.3 NF Large 9 1.44 11.0 Small 9 0.30 9.7 Mean 18 0.87 10.3 LB Large 9 2.42 16.5 Small 9 0.16 12.4 Mean 18 1.28 14.4 LP1 Large 6 3.54 15.5 Small 6 0.24 10.2 Mean 12 1.88 12.8 LP2 Large 9 1.54 9.1 Small 9 0.17 8.1 Mean 18 O.85 8.6 LSH Large 9 1.58 11 .0 Small 9 0.21 8.6 Mean 18 0.89 9.8 CL4 Large 9 24.4 CS4 Small 9 — 24.2 Mean 18 - 24.3 L4 Large 9 2.06 28 .0 Small 9 0.07 28 .4 Mean 18 1.06 28.2 -37-E 3.0 L4 LSH NF LP2 XF TREATMENT CODE Figure 8. Multiple comparisons of aggressiveness of large f i s h , small f i s h , and both f o r each treatment as calculated from days 27;21, 39J33> and 51;45 (any two means not enclosed by the same bracket are d i f f e r e n t , p £ 0 .05) . • = large f i s h , O = small f i s h , — = mean for whole treatment. -58-actions / 2 . 5 min, (multiple comparison, n = 84,; 54; p £ 0 .05) . Large f i s h were more aggressive and were s o c i a l l y dominant, while the small f i s h behaved as subordinates. The s o c i a l dominance of the large f i s h was presumably determined more (on the average) by r e l a t i v e size than geno-type because the si z e difference was i n i t i a t e d by rearing sibs of d i f f e r e n t ages. Aggressiveness of the large s o c i a l l y dominant f i s h was influenced by environmental f a c t o r s and i n general increased when food was l i m i t e d and I l o c a l l y concentrated. I f food was i n excess the large f i s h was no more aggressive than the small f i s h . Average d a i l y a c t i v i t y counts/ 2 . 5 min of the large and the small f i s h i n each treatment were calculated f or days 27;21, 39;53J and 51;45 by averaging the eighteen 2.5-minute observations made on a f i s h i n one day (Table 6 ) . In addition the small f i s h was averaged with the large f i s h to get a mean value f o r the whole treatment. Mu l t i p l e comparison techniques (Duncan, 1955; 10/ Krammer, 1956) were used i n the analysis of the data (Figure 9 ) • — F i s h i n 4 - l i t e r baskets apparently were more act i v e than f i s h i n 1 - l i t e r baskets (Figure 9)> but t h i s was at l e a s t i n part a bias i n the counting techniques. I f a f i s h swam across a 4-li.ter basket, i t passed under J strings and moved 20 cm (1.5 a c t i v i t y counts/10 cm of movement); but i f a f i s h swam across a 1 - l i t e r basket, i t passed under 1 s t r i n g and moved 10 cm (1 a c t i v i t y count/10 cm of movement). At t h i s rate a f i s h swimming 120 cm back and f o r t h i n a 4 - l i t e r basket would be recorded as 18 counts and i n a 1 - l i t e r basket as 10 counts. F i s h did not always swim back and f o r t h however, and i t was not possible to c a l c u l a t e a factor to e q u i l i b r a t e the a c t i v i t y data f o r 1 - l i t e r versus 4 - l i t e r comparisons. 10/ Pooled estimate of variance among f i s h treated a l i k e was s 2 x = 602.1 and degrees of freedom d.f. = 152. L4 NF LSH LP2 XF C S 4 LB LI LPI C L I CL4 L O L l L K I CSI Figure 9. Multiple comparisons of a c t i v i t y of large f i s h , small f i s h , and both f o r each treatment as calculated from days 27;21, 29;33, and 51j45 (any two means not enclosed by the same bracket are d i f f e r e n t , p i O . O 5 ) . • = large f i s h , O = small f i s h , — = mean fo r whole treatment. -40-There were no s i g n i f i c a n t d i f f e r e n c e s between a c t i v i t i e s of a large and a small f i s h i n the same treatment (Figure 9) , and there were not many d i f f e r ences among average a c t i v i t i e s of whole treatments. (Figure 9 ) . Average a c t i v i t y i n 1 - l i t e r treatments was never greater than a c t i v i t y of co n t r o l s . Only f i s h i n treatment XF were l e s s a c t i v e than i s o l a t e c o n t r o l s . Some us e f u l comparisons were made by grouping treatments on the basis of growth rates i n t o "competition" treatments (L1, LB, LPT', -LSH, and L4) and "no competition" treatments (0L1, 0S1; XF; NF; LP2; and Glib-, CS4). Average a c t i v i t y i n "competition" treatments, 15-7 counts/ 2 . 5 min, was greater than averag a c t i v i t y i n "no competition" treatments, 12.8 counts/2 .5 min (multiple, compar ison , n = 84, 90; p £ 0 .05) . In "no competition" treatments, a c t i v i t y of large f i s h , 13.0 counts/2 .5 min, was not greater than a c t i v i t y of small f i s h , 12.7 counts/ 2 . 5 min (multiple comparison, n •» 45 , 45; p > O .05); but i n "com-p e t i t i o n " treatments, large f i s h , 17•5 counts/ 2 . 5 min, were more ac t i v e than small, f i s h , 13.0 counts/ 2 . 5 min (multiple comparisons, n = 42, 42; p ~ 0 .05) . Increased a c t i v i t y i n "competition" treatments was p r i m a r i l y a r e s u l t of i n -creased a c t i v i t y of large f i s h . Even though a c t i v i t y counts of small f i s h were the- same i n both "competition" and "no competition" treatments, there was a difference, i n what they were doing-while accumulating a c t i v i t y counts. Much of a small f i s h ' s a c t i v i t y i n competition treatments consisted of escaping, aggressive actions of the large f i s h , while i n some of the "no competition." treatments (0L1, CS1 and 0L4, 0S4) there was none of t h i s a c t i v i t y and i n others, such as XF, l i t t l e of t h i s a c t i v i t y . Diurnal Rhythms i n Behavior Behavior observations from each treatment on day 59;55 and 51J 45 were averaged and graphed i n Figures 10—15 f o r each time period. Each point represents twelve 2.5-minute periods of observation. Lights went on at 0 hours and o f f at 16 hours.. In l i m i t e d food treatments the single feeding -41-ACTI VIT Y/2.5 mm. io -2 0 ACTIV I TY / 2.5 min io AGGRESSIVE ACTIONS/2.5min. J L 0 I 3 5 7 9 FED ~ - - o -J 1_ L II 13 15 16 2 0 ACTIVITY/2 5min. AGGRESSIVE ACTIONS/2.5min. 1 \ ~ o ~ o o-— • • . A i 1 1 1 1 o • 0 1 3 5 7 9 II 13 15 1 XF F E D FED FED HOURS A F T E R L I G H T S W E N T ON Figure 10. Diurnal changes i n a c t i v i t y and aggressiveness of large and small f i s h i n the i s o l a t e controls (CL1, CS1), i n p l a i n 1 - l i t e r l i m i t e d food treatment (L1), and i n the excess food treatment (XF). • = large f i s h , o = small f i s h . -42-ACTIVIT Y /2.5min. 2 5 _ AGGRESSIVE ACTIONS/2.5min. o i NF 5 16 ACTIVITY/2.5min. 2 5 AGGRESIVE ACTIONS/2.5min. % T I ME. I N • BOTTOM DISH 5 0 0 I 7 - • -/ - ' ^ -o - • • 1 1 1 1 1 1 1 1 1 1 1 L B 15 16 F E D HOURS A F T E R L I G H T S WENT ON Figure 11. Diurnal changes i n a c t i v i t y and aggressiveness of large and small f i s h i n the no-food treatment (NF), and a c t i v i t y , aggressiveness, and l o c a t i o n preference of large and small f i s h i n the l i m i t e d food l o c a l i z e d on bottom treatment (LB). • = large f i s h , o = small f i s h . ACTIVITY/2.5 min. 2 5 0 AGGRESSIVE ACTIONS/2.5 min. 5 o 1 0 0 % TIME ON RIGHT SIDE OF PARTITION 5 0 O- • • ' - o o I I I i i i i i U • u o y ——— I I I y- ~_ _ - ° - " i i i i 0 l 3 5 7 9 II 13 15 I L P 2 FED HOURS A F T E R L I G H T S W E N T ON Figure 12. Diurnal changes in a c t i v i t y , aggressiveness, and l o c a t i o n preference of large and small f i s h i n treatments with a p a r t i a l p a r t i t i o n across the basket which were fed on one side (LP1) and both sides (LP2) of the p a r t i t i o n . • = large f i s h , o = small f i s h . -44-ACTIVITY/2.5min. 2 5 AGGRESSIVE ACTIONS/2.5min. o I o o-mi <yf " - o 1 i i 1 1 1 1 1 5 FED LSH 13 15 16 ACTIVITY/2.5 min. AGGRESSIVE ACTIONS/2.5 min L4 I 13 15 16 ACTIVITY/2 5mi n 2 5 o I 15 16 FED HOURS A F T E R L I G H T S W E N T ON Figure 13. Diurnal changes i n a c t i v i t y and aggressiveness of large and small f i s h . i n shallow l i m i t e d food treatment (LSH), 4 - l i t e r i s o l a t e treatment (0L4, GS4), and p l a i n 4 - l i t e r l i m i t e d food treatment (L.4). « = large f i s h , o = small f i s h . - 4 5 -was at 5 ' 2 5 hours, and i n XF the three feedings were at 1.25, 5«25J a n < i 9.25 hours a f t e r the l i g h t s went on. A c t i v i t y of both large and small f i s h increased immediately a f t e r food was introduced i n a l l l i m i t e d food treatments (L1, LB, LP1, LP2, LSH, and L4) (Figures 10-13); and i n the i s o l a t e s (CL1, CS1, CL4, and CS4) (Figures 10, 13)... A c t i v i t y did not increase i n the excess, food treatment (XF) at feeding time (Figure 10), and i n the no-food treatment (NF) a c t i v i t y was steady throughout the day (Figure 11). A c t i v i t y i n the form of general a p p e t i t i v e behavior was at a steady l e v e l among i s o l a t e s during periods when no food was i n the basket. At feeding time the f i s h swam r a p i d l y back and f o r t h among the food p a r t i c l e s u n t i l a l l p e l l e t s had been eaten. The steady l e v e l of food search-ing behavior was then resumed. The same general pattern was observed i n the li m i t e d food populations, but was furth e r complicated by s o c i a l i n t e r a c t i o n s between the large and small f i s h . High a c t i v i t y at hour 1 i n XF (Figure 10) was the consequence of sexual behavior. The f i s h i n XF were the only ones which had grown f a s t enough to approach maturity by day 5"l>45. Aggressiveness of the large f i s h increased within one or two minutes a f t e r food was presented i n the l i m i t e d food treatments (L1, LB, LP1, LP2, and L4) (Figures 10-13) and. several hours a f t e r food was presented i n the li m i t e d food treatment (LSH.) (Figure 13). In excess food treatment (XF) no increase i n aggressiveness occurred a f t e r food was introduced (Figure 10). When no food was provi ded (NF) the 1 evel of aggressiveness was steady throughout the. day (Figure 11). Aggressiveness of small f i s h remained steady or decreased at feeding time i n l i m i t e d food treatments (L1, LB, LSH, and L4) (Figures 10, 11, 13) but increased i n the two l i m i t e d food treatments which had p a r t i a l p a r t i t i o n s across the baskets (LP1 and LP2) (Figure 12). No increase i n aggressiveness -46-occurred i n XF when food was supplied (Figure 10), and i n NF. aggressive actions were at steady l e v e l throughout the day (Figure 11). In treatments L1 and L4 the l i m i t e d food was scattered evenly on the surface of the water. Neither the large nor the small f i s h defended any s p e c i f i c area. When food was present the large f i s h fed r a p i d l y but stopped feeding at frequent i n t e r v a l s and swam around i n the basket. While swimming throughout the basket the large f i s h often encountered the small f i s h , nipped i t , and chased i t i n t o a corner or to the bottom. The large f i s h resumed feeding, only to repeat the sequence 10-20 seconds l a t e r . The small f i s h began feeding during the periods when the large f i s h was feeding, but was us u a l l y disturbed too frequently to eat many p e l l e t s . In LB the l o c a l i z e d food supply was placed i n a dish on the bottom. The large f i s h defended the food dish only when food was present (Figure 11). During t h i s period the small f i s h would c o n t i n u a l l y attempt to enter the dish but would be chased away by the large f i s h . This continued f o r 2 or 3 hours a f t e r food was introduced, even though a l l the food appeared to be gone 5 °r 10 minutes a f t e r i t was supplied. The large f i s h stayed i n the dish except when chasing the small f i s h away. About 3 hours a f t e r a feeding time the large f i s h l e f t the dish, and the small f i s h moved in t o the dish and searched f o r food (Figure 11). About 2 hours l a t e r the small f i s h also l e f t the dish, and neither f i s h occupied i t u n t i l food was presented on the following day (Figure 11). In LP1 the food was concentrated on the r i g h t side of a p a r t i a l p a r t i -t i o n and f l o a t e d on the surface. The large f i s h tended to inhabit the h a l f of the basket i n which the food was placed during the 5 hours before and the 3 hours a f t e r food was supplied (Figure 12). Aft e r discovering the food the large f i s h remained i n the food area except when chasing away the small f i s h -47-which c o n t i n u a l l y re-entered the food area. The small f i s h also was aggres-sive and attempted to chase the large f i s h from the food area but these attempts ended i n f a i l u r e . As i n LB above, the large f i s h l e f t the feeding area about 3 hours a f t e r food was introduced and the small f i s h then remained i n the food area (Figure 12). In LP2 the food was simultaneously placed on both sides of a p a r t i a l p a r t i t i o n . Both f i s h were aggressive but did not defend s p e c i f i c areas. The large f i s h moved f r e e l y from one side of the p a r t i t i o n to the other, but each time the large f i s h changed sides the small f i s h immediately swam to the opposite side (Figure 12). At feeding time t h i s behavior l e f t each f i s h alone with one-half of the food supply even though they alternated sides many times while food was present. In LSH the basket was only 2.5 cm deep and the two f i s h seemed to i n t e r -fere with each other. Both the large and small f i s h were i n a c t i v e , and when they did swim i t was only for short distances. The large f i s h did not respond to the food with a sudden increase of feeding a c t i v i t y and aggression, but rather fed slowly f o r several hours a f t e r food was introduced (Figure 13). By t h i s time most of the p e l l e t s had sunk, and the large f i s h tended to defend the p e l l e t s f o r long periods without eating any. Relations Between Growth and Behavior Aggressive behavior provided large f i s h with a competitive advantage when food supply was l i m i t e d , but not when food was absent or i n excess. In NF the large f i s h was s o c i a l l y dominant (Figure 8) but no food was present, and the dominant and subordinate f i s h l o s t weight at the same rates (Figure 5)- The large f i s h was more aggressive when a l i m i t e d food supply was added (L1) and grew better than the subordinate (Figure 5)- Aggressive actions by the large f i s h were most frequent immediately a f t e r food was added -48-and u s u a l l y prevented the subordinate.from eating many p e l l e t s . With the introduction of excess food (XF) the s o c i a l hierarchy disappeared, neither the large nor the small f i s h was aggressive when food was added, and they grew equally well (Figures 5> 8)» Aggressiveness was associated with the i n t e r n a l state of "hunger," and the external f a c t o r s of food s t i m u l i and other medaka. The aggressiveness of a p a r t i c u l a r medaka was high i f the other f i s h was smaller, but was low i f the other f i s h was la r g e r . Aggressive actions among immature medaka were most frequent when the f i s h had a l i m i t e d food supply, food s t i m u l i were present, and another smaller medaka was near; f o r example, the period just a f t e r food was supplied i n l i m i t e d food treatments. Aggressive actions were moderately frequent when no food s t i m u l i were present even though the f i s h had not been fed f o r some time and a smaller f i s h was present (NF). I f the f i s h were fed i n excess (XF) the frequency of aggressive actions was only 7.5% of the highest frequency even though food s t i m u l i and smaller medaka were present. Aggressive behavior was a mechanism i n i t i a t e d by the i n t e r n a l state of "hunger," a feeding stimulus, and smaller medaka, which gave a compe-t i t i v e advantage to large f i s h when food supply was l i m i t e d . Evidence i n d i c a t i n g that aggressiveness was a competitive mechanism f o r food and not f o r space, per se, was provided by a comparison of treatments L1, L4, and XF. Even though L1 environments were one-fourth the size of L4 environments, large f i s h i n L1 and L4 were equally aggressive (Figure 8), and the differences between the growth rates of large and small f i s h were the same i n each sized environment (Figure !?)• When the amount of space was the same but the amount of food was l i m i t e d (L1) rather than i n excess (XF) the large" f i s h was more aggressive and the difference i n growth rates between large and small f i s h was greater. The difference i n aggressiveness and .-49-growth was evidently due to l i m i t e d food, not the amount of space. Even though some of the fa c t o r s associated with space, such as b i o l o g i -cal conditioning and abundance of food, were removed i n these experiments, there were some fa c t o r s which remained to produce r e s i d u a l unexplained e f f e c t s associated with space. The average growth among f i s h fed 10 p e l l e t s per f i s h per day i n 4 - l i t e r baskets ( C L 4 , CS4) was s i g n i f i c a n t l y greater than among f i s h fed the same amount i n 1 - l i t e r baskets ( C L 1 , C S 1) (Figure 5)• The same was true f o r the average growth of L4 and L 1, although the dominant f i s h had an equal advantage r e l a t i v e to the subordinate i n both sized baskets. Factors associated, with space which would produce these e f f e c t s were not i s o l a t e d i n the present experiment. Other possible explanations which suggest themselves are that a microfauna was on the nylon l i n e r s which provided more food i n larger baskets; the f i s h found p e l l e t s which sank to the bottom more e a s i l y i n large baskets because the water column was not as narrow; a water-borne growth i n h i b i t o r (Richards, 1958) w a s a p a r t i c l e and was too large to pass out through the f i n e meshes of the baskets with the c i r c u l a t i n g water and would be more concentrated i n smaller baskets; or the amount of current flow was l e s s on the average through large baskets than small baskets and l e s s energy was used i n swimming. I t was apparently not associated with any psychological phenomenon, mediated v i s u a l l y , because differences i n density had no e f f e c t s i f food was supplied i n excess. I t i s conceivable that when f i s h were fed i n excess they did not eat any of the f e c a l material i n the container, whereas i n the l i m i t e d food treatments more f e c a l material would have been eaten. Whatever the cause i t was not measured i n the experiment. S p a t i a l d i s t r i b u t i o n of the l i m i t e d food supply influenced the conse-quences of competition. When food was l o c a l i z e d i n one part of the environ-ment as i n LP1 and LB, aggressiveness of the large f i s h was greater than when -50-food was evenly d i s t r i b u t e d i n the environment as i n L1 and LP2 (Figure 8). Likewise the increase i n aggressiveness just a f t e r food was supplied appeared to be greater i n l o c a l i z e d food treatments (LP1 and LB) than i n dispersed food treatments (L1 and LP2) (Figures 10, 11, 12). There was an i n d i c a t i o n that large f i s h had a greater competitive advantage f o r food i n LB than i n L1 (Figure 5) J and the large f i s h d e f i n i t e l y had a greater competitive advantage i n LP1 than i n LP2. L o c a l i z i n g the l i m i t e d food supply increased both the aggressiveness and the competitive advantage of the large s o c i a l l y dominant f i s h . I f the l i m i t e d food supply was l o c a l i z e d (LB or LP1) the aggressive behavior took the form of t e r r i t o r i a l i t y . Whenever the small f i s h approached the food area, the large f i s h chased i t away. Defense of a s p e c i f i c area disappeared when a l l the food was eaten. Although aggressive behavior appeared to serve i n the defense of an area or space i t was a c t u a l l y func-t i o n i n g as a competitive mechanism f o r food, and as i n the cases l i s t e d above was i n i t i a t e d by the i n t e r n a l state of "hunger," other smaller medaka, and the presence of food s t i m u l i . The amount of s p a t i a l i s o l a t i o n between competitors when food was evenly d i s t r i b u t e d influenced the consequences of competition. Control f i s h (CL1, CS1) represented complete i s o l a t i o n and the removal of a l l i n t e r a c t i o n s between competitors; LP2 represented a p a r t i a l i s o l a t i o n , i n the form of either distance or obstructions i n the environment; and L1 represented no i s o l a t i o n between competitors. A l l were given the same amount of food, but even so, as the amount of i s o l a t i o n decreased, there was a decrease i n average growth rate, an increase i n aggressive interchanges between compe-t i t o r s , and an increase i n competitive advantage to the large s o c i a l l y dominant f i s h (Figures 5, 8). Segregation of the environment or s p a t i a l -51-i s o l a t i o n between competitors reduced the influence of competition and decreased the advantage of the large dominant f i s h i f food was evenly d i s t r i b u t e d . Aggressiveness not only served as a competitive mechanism f o r food, but also tended to disperse f i s h throughout the environment i f food was evenly d i s t r i b u t e d . For example, the f i s h i n LP2 were u s u a l l y on opposite sides of the p a r t i a l p a r t i t i o n (Figure 12) because the small subordinate always moved to the side f a r t h e s t from the large dominant f i s h . This occurred only when food was evenly d i s t r i b u t e d i n the environment. In LP1, which had a l l the food on the r i g h t side of the p a r t i a l p a r t i t i o n , the subordinate f i s h con-t i n u a l l y re-entered the side containing both the dominant f i s h and the food. By c o n t i n u a l l y re-entering t h i s area the subordinate was exposed to more aggressive actions than the subordinate i n LP2 which avoided the side con-t a i n i n g the dominant f i s h . Aggressive behavior only resulted i n a dispersed d i s t r i b u t i o n of medaka i f the n e c e s s i t i e s of the subordinate f i s h were found in. a l l subsections of the h a b i t a t . In these experiments, growth and condition were measured to demonstrate the consequences of competition under d i f f e r e n t environmental conditions, and a c t i v i t y and aggressiveness were measured as p o t e n t i a l mechanisms which might mediate the consequences of competition. The association between these two response variables and two p o t e n t i a l mediating variables were measured by multiple and p a r t i a l c o r r e l a t i o n and regression techniques. Data f o r days 27;21, 39;33> and 51>4-5 were averaged for each treatment. Levels of a c t i v i t y and aggressiveness were averaged only f o r the 5'25 and 7-00 hours, because the two hours a f t e r food was presented appeared to be most important i n determining the consequences of competition. The average difference between the large and the small f i s h i n each treatment was used f o r analysis (Table 7)> -52-Table 7. Differences i n average growth, condition, aggressiveness, and a c t i v i t y between large and small f i s h i n each treatment averaged f o r days 27;21, 59;53> and 51;45. ment y1 y2 Code (G L-G S) (0h-0s) Average Aggressiveness During 2.5 Hours After Feeding  Large Small 1 F i s h F i s h (Agg L-Agg g) Average A c t i v i t y During 2.5 Hours Af t e r Feeding  Large Small 2 Fish F i s h (Act. - A c O CL1 CS1 0.00 0.00 L1 +1.65 +1.63 XF 0.00 0.00 NF +0.07 +0.26 LB +2.47 -0.27 LP1 +2.83 +1.67 LP2 +0.68 -0.49 LSH +2.64 +0.78 CL4 CS4 0.00 0.00 L4 +2.45 +1.71 0.00 0.00 0.00 6.37 0.33 +6.04 0.23 0.27 -0.04 2.98 0.88 +2.10 8.43 0.10 +8.33 9.13 2.58 +6.55 5.45 1.82 +3.63 5.50 0.48 +5.02 0.00 0.00 0.00 9.32 0.37 +8.95 15-9 17.4 .-1-5 22.6 13.2 +9.4 8.3 7.0 +1.5 10.5 10.4 +0.1 21.7 17.5 +4.2 15.2 9.0 +6.2 13.8 10.3 +3.5 14.0 14.1 -0.1 15.9 17.4 -1.5 58.4 40 .8 -2.4 -53-where y.| = (G^-Gg) or the growth of the large minus the growth of the small f i s h ; - (Ojj-Cs) o r condition of large minus condition of small f i s h ; x.j = (Agg^-Aggg) or aggressiveness of large minus aggressiveness of small f i s h at 5-25 and 7.00 hours; and X£ = (Act^-Actg) or the a c t i v i t y of the large minus the a c t i v i t y of the small f i s h at 5*25 and 7-00 hours. Simple, p a r t i a l , and multiple c o r r e l a t i o n c o e f f i c i e n t s f o r these data are presented i n Table 8. Growth dif f e r e n c e s (Gjj-Gg) were associated with differences i n condition (Ojj-Cg). Differences i n growth and condition between large and small f i s h are two ways of measuring the same response. Growth differences (G-jj-kg) w e r e associated with differences i n aggression (Agg^-Aggg) but not with differences i n a c t i v i t y (Actjj-Actg) (Table 8). The simple c o r r e l a t i o n c o e f f i c i e n t between y-\ and x^ (0.910), the p a r t i a l c o r r e l a t i o n c o e f f i c i e n t which removed the influence of a c t i v i t y (0.891), and the multiple c o r r e l a t i o n c o e f f i c i e n t which takes into account the influence of a c t i v i t y (0.911), were a l l v i r -t u a l l y the same. Therefore p r e d i c t i o n of the growth differences was not improved by considering differences i n a c t i v i t y ; differences i n a c t i v i t y had no influence on the consequences of competition but differences i n aggression were h i g h l y associated with the growth consequences of competition. None of the associations between condition differences (O^-Cg) and aggression or a c t i v i t y differences were s i g n i f i c a n t , but the pattern of associations was the same as f o r the growth differences (Table 8). Condition was not as s e n s i t i v e a measure of the e f f e c t s of competition as was growth. The r e l a t i o n between growth differences (G^-Gg) and aggressive d i f f e r -ences (Aggjj-Aggg) was adequately described by the l i n e a r regression: (G L-G S) = -0.145-+ 0.341 (Agg L-Agg s) and was presented i n g r a p h i c a l l y i n Figure 14 (95% C.I., n = 10, -54-Table 8. Simple, p a r t i a l , and multiple c o r r e l a t i o n c o e f f i c i e n t s f o r y 1 = ( G L - G S ) , y 2 = ( O L - ° S ) > x1 = (Agg L-Aggg) and x 2 = ( A c t L - A c t s ) . (See text f o r explanation.) Co r r e l a t i o n P r o b a b i l i t y C o e f f i c i e n t s Level P = Simple C o r r e l a t i o n C o e f f i c i e n t s Growth x Condition ry^2 = 0.622 = 0.05 Growth x Aggression Growth x A c t i v i t y Condition x Aggression Condition x A c t i v i t y T y i X 1 v 2 r y 2 X 1 y 2 x 2 = 0.910 0.414 0.576 0.317 <0.01 > 0.05 >o.05 ?o.05 Aggression x A c t i v i t y r x 1 x 2 = 0.425 >0.05 P a r t i a l C o r r e l a t i o n C o e f f i c i e n t s Growth x Aggression ( A c t i v i t y ) Growth x A c t i v i t y (Aggression) r y - i x r x 2 r y 1 x 2 - x 1 = 0.891 0.075 < 0.01 >0.05 Condition x Aggression ( A c t i v i t y ) r y 2 x r x 2 = 0.514 >o.05 Condition x A c t i v i t y (Aggression) r y 2 x 2 * x i = 0.098 > 0.05 M u l t i p l e , C o r r e l a t i o n C o e f f i c i e n t s Growth x Aggression and A c t i v i t y Condition x Aggression and A c t i v i t y R y l X l x 2 R y 2 x i x 2 — 0.911 0.581 < 0.01 ^ 0 . 0 5 - 5 5 -+ 4 Figure 14. Differences i n the growth rates of large and small f i s h i n each treatment plotted against the differences i n t h e i r aggressiveness during tVie 2 . 5 hours a f t e r food was presented. -56-b = 0.341 ±0.117, and y 1 = 1.240 +0.551; s ^ = 3-4479, s x = O.5686, s v = 1.2896). The difference i n aggressiveness between the large and small f i s h was important as a mediating f a c t o r or as a mechanism by which the e f f e c t s of competition were unequally passed on to the large and the small f i s h . These differences i n aggressiveness of the large and small f i s h varied from treatment to treatment as induced by amount of food, s p a t i a l d i s t r i b u -t i o n of food, and topography of the h a b i t a t . The variables of the environ-ment influenced the growth consequences of competition, but did so through the action of the aggressive behavior i n various environments. Detailed analysis of the behavior data discussed previously also supported these conclusions. Even though differences i n aggressiveness between large and small f i s h had a great influence upon the growth consequences of competition, an aggressive action by the large f i s h was not equally e f f i c i e n t i n d i f f e r e n t environmental s i t u a t i o n s . E f f i c i e n c y of aggression f o r the large f i s h was defined as the number of milligrams, per aggressive action, by which the growth of large f i s h exceeded the growth of small f i s h , and was calculated by d i v i d i n g column 1 by column 3 i n Table 7« E f f i c i e n c y of aggression was zero when no food or excess food was provided but was high at intermediate l e v e l s of food abundance (Figure 15a). In l i m i t e d food environments the e f f i c i e n c y of aggression increased as the food became more l o c a l i z e d i n i t s s p a t i a l d i s t r i b u t i o n (Figure 15b)• E f f i c i e n c y of aggression decreased i f the habitat had a dispersed food supply and was subdivided by p a r t i a l p a r t i -tions (Figure 15b). E f f i c i e n c y of aggression decreased as the size of the environment or the amount of space per f i s h increased (Figure 15c). Environ-mental features such as the amount of food, s p a t i a l d i s t r i b u t i o n of the food, and the topography of the h a b i t a t influenced the consequences of competition, -57-Figure 15. E f f i c i e n c y of aggression to the large dominant f i s h as influenced by (a) amount of food, (b) l o c a l i z a t i o n of food supply, and (c) size of environment. E f f i c i e n c y = (G L-Gg)/(Aggressive actions by large f i s h i n 2.5 hours a f t e r food was supplied) -58-not only by a l t e r i n g the frequency of aggressive actions, but also by a l t e r i n g the e f f i c i e n c y of an aggressive action by the large f i s h i n terms of growth. Summary of Results Each medaka had the same chance of having genetic p o t e n t i a l f o r becoming an aggressive s o c i a l dominant, but the large f i s h always dominated the small f i s h . The difference between the aggressiveness of a large and small f i s h was influenced by many environmental f a c t o r s . An excess food supply resulted i n a d i s i n t e g r a t i o n of the s o c i a l hierarchy and a low l e v e l of aggression. In l i m i t e d food treatments the difference between the aggressiveness of large and small f i s h was greater just a f t e r food was presented, increased i f the food was s p a t i a l l y concentrated, and decreased i f the amount of i s o l a t i o n between competitors was increased. I f food was l o c a l i z e d , aggressive behavior took the form of t e r r i t o r i a l i t y , but the defense of l o c a l i z e d areas d i s -appeared several hours a f t e r the l i m i t e d food supply was completely eaten. Two f i s h competing f o r a: l i m i t e d food supply grew slower than two f i s h grown i n i s o l a t i o n on the same amount of food. Differences between the growth of large and small f i s h from treatment to treatment were c l o s e l y associated with environmentally induced differences i n aggressiveness. The more aggressive the large f i s h i n a treatment the better t h e i r growth rates r e l a t i v e to subordinate f i s h . Aggressive behavior was the mechanism through which the consequences of competition (poor growth) were unequally d i s t r i -buted between large and small f i s h . Dominant f i s h had a competitive advan-tage over small f i s h i n l i m i t e d food environments unless the environment provided both p a r t i a l i s o l a t i o n for the subordinate and food i n a l l subsec-tions of the h a b i t a t . S o c i a l dominance did not confer a competitive -59-advantage when the food was absent or i n excess. Aggressive actions by the dominant large f i s h were not equally e f f i c i e n t i n terms of the competitive advantage they provided as the environmental factors were changed. E f f i c i e n c y of aggression was zero i f there was no food or excess food i n the habitat. When food was present but l i m i t e d i n supply the e f f i c i e n c y of aggression decreased as the size of the environment increased, and increased when food was more l o c a l i z e d . COMPETITION FOR EXCESS FOOD (EXPERIMENT I I I ) Introduction and Description of Experiment The purpose of experiment III was to determine whether large f i s h had a competitive advantage f o r an excess food supply which was s p a t i a l l y l o c a l i z e d , and whether they would defend the l o c a l i z e d area which contained an excess food supply. Each treatment population was composed of two f i s h with a 4-day age difference, reared i n 4 - l i t e r baskets. A l l f i s h were fed i n excess with brine shrimp n a u p l i i u n t i l day 32; 28 and thereafter were fed an excess of p e l l e t s placed i n p e t r i dishes on the bottom. P e t r i dishes were cleaned and r e f i l l e d d a i l y . In the dispersed excess food treatment (XD) 2 p e t r i dishes were placed i n opposite corners of the basket, and i n the l o c a l i z e d excess food treatment (XL) 1 p e t r i dish was placed i n a corner. Any advantage i n competition f o r the excess s p a t i a l l y l o c a l i z e d food could be determined by comparing the- growth of large f i s h i n XD and XL and by comparing the growth of small f i s h i n XD and XL. I f there were no competitive advantage, the large f i s h i n each treatment would grow at the same rates and the small f i s h i n each treatment would grow at the same rat e s . Hatching was induced 11 days a f t e r f e r t i l i z a t i o n , and the f i s h from each spawning day were raised as a group u n t i l day 52}28 at which time they were randomly al l o c a t e d to the treatments. Each f i s h was weighed every 6 days from day 32;28 through 62;58. Both XL and XD were r e p l i c a t e d 6 times. Results and Conclusions S i z e - s p e c i f i c 6-day weight increments were determined g r a p h i c a l l y f o r -61-each f i s h from a Walford p l o t (Ricker, 1958) when the small f i s h were 20 and 30 mg i n weight and when the large f i s h were 25, 35 > a n d 4? mg i n weight. Values f o r the small f i s h were averaged f o r 20 and 30 nig and f o r the large f i s h they were averaged f o r 25> 35> and ^5 mg. Growth of large f i s h i n XD with a dispersed excess food supply, 8.9 mg/6 days, was not d i f f e r e n t from growth of large f i s h i n XL with a l o c a l i z e d excess food supply, 8.9 mg/6 days ( t 2 ^ = 0.02, p > 0 .25) . Small f i s h also grew equally well i n the two t r e a t -ments, 7.6 mg/6 days i n XD and 8.0 mg/6 days i n XL (t^y = 0.72, p > 0.25). Large f i s h did not have a competitive advantage f o r excess food even though i t was l o c a l l y concentrated. Although no detai l e d behavior observations were made on these f i s h , observations were made on a d d i t i o n a l baskets with larger populations. As many as 20 medaka, 10—15 mm.in t o t a l length, would crowd in t o the 50 mm food dish at one time as long as the amount of food i n the dish was maintained i n excess. Few aggressive interchanges were observed among them, and large f i s h did not chase small f i s h away. As i n experiment I I , medaka were not aggres-sive i n the presence of food s t i m u l i and smaller medaka as long as food was i n excess. L o c a l i z i n g the excess food did not a l t e r t h i s behavior. Summary of Results Large medaka did not defend l o c a l i z e d feeding areas i f food was provided i n excess. Large and small medaka grew equally well whether the excess food had a dispersed or a l o c a l i z e d d i s t r i b u t i o n . COMPETITION FOR LIMITED FOOD IN LARGER POPULATIONS (EXPERIMENT IV) Introduction The purpose of experiment IV was to examine the action and consequences of aggressive behavior i n larger populations i n more complicated environments when food was l i m i t e d i n supply. S p a t i a l d i s t r i b u t i o n of the food and the number of subdivisions i n the habitat were varied to c l a r i f y the influence of these modifying f a c t o r s . The influence of b i o l o g i c a l conditioning of the water was removed as i n previous experiments. Experimental animals were selected from many d i f f e r e n t spawning p a i r s . Description of Experiment Each population was composed of 8 f i s h raised i n 8 - l i t e r baskets, and food was supplied at a rate of 10 p e l l e t s per f i s h per day. Individual f i s h could not be i d e n t i f i e d , and differences i n growth rates of large and small f i s h were studied by comparing the v a r i a t i o n i n sizes observed i n d i f f e r e n t treatments. A d e s c r i p t i o n and code number f o r each treatment are presented i n Table 9 and Figure 16. P a r t i a l p a r t i t i o n s which subdivided some baskets were made from nylon "horsehair" c r i n o l i n e (10 meshes/cm), and holes through p a r t i t i o n s were 3 cm wide and 5 cm deep and were placed with the top edge 3 cm below the water surface. In those treatments not receiving food i n every subsection of the habitat, the subsections i n which food was placed were evenly spaced a f t e r one food l o c a t i o n had been chosen at random. Food was placed i n the same locatio n s throughout the experiment. Eggs were selected from many spawning p a i r s , but were f e r t i l i z e d on the same day. Temperature was varied 10, 11, 12, and 13 days a f t e r f e r t i l i z a t i o n -63-Table 9. Design of experiment IV showing treatments and number of r e p l i c a t i o n s with treatment codes and descriptions. Basket Pood Treatment Code No. of Replications Population Size Size ( l i t e r s ) No. of ' Subsections No. of P e l l e t s Per Day Location on Surface 1E 3 8 8 1 80 even 4E 3 8 8 4 80 even 8E 3 8 8 8 80 even 8C4 3 8 8 8 80 i n 4 sub-sections 8C2 3 8 8 8 80 i n 2 sub-sections 8C1 3 8 8 8 80 i n 1 sub-section to induce hatching, and approximately the same number hatched on each of the four days. They were fed " i n excess" on brine shrimp n a u p l i i f o r 15 days a f t e r hatching was induced, at which time they were randomly a l l o c a t e d to the treatments under the r e s t r i c t i o n that each population contained 2 f i s h from each hatching day. Treatments were randomly a l l o c a t e d to water bath p o s i -t i o n s under the r e s t r i c t i o n s that each bath contained only a single r e p l i c a t e from a given treatment, and that no two r e p l i c a t e s of a treatment occupied the same p o s i t i o n i n d i f f e r e n t baths. Each f i s h was weighed every 6 days from 15 through 59 days a f t e r hatch-ing was induced. The posit i o n s of f i s h i n the habitat and the t o t a l number of aggressive actions were recorded f o r two 2.5-minute periods a f t e r food was presented on days 30» 32, 36, and 38. Two r e p l i c a t e s from each t r e a t -ment were observed on each date. -64-FOOD PELLET HOLE IN PARTITION. 8 C 4 4 E 8 C 2 8 E 8 C I Figure 16. Diagram showing p a r t i t i o n s and food locations of the 8 - l i t e r baskets used i n experiment IV. (top view) -65-Results Growth Rates Growth depensation i n populations of 8 f i s h was measured by the increase i n the variance of size-frequency d i s t r i b u t i o n s during 6-day i n t e r v a l s . Variances of weight d i s t r i b u t i o n s within each population were calculated f o r days 15, 21, 27, 33, & n d 39, but day 59 was omitted because deaths of small f i s h were b i a s i n g the estimates. A cube root transformation of each variance was used to convert the variances i n t o a normally d i s t r i b u t e d v a r i a b l e . Increase i n variance due to growth differences within a population was expressed as the 6-day increment i n the cube root of the variances r i IT • L ( s ^ t + 6 ^ - ( s ^ ) ^ J . This measure of growth depensation increases i n value as differences between growth rates of large and small f i s h increase. Growth depensation estimates f o r the 15-21, 21-27, and 27-59 day i n t e r v a l s were averaged f o r each treatment (Table 10), and the averages were compared by 11/ multiple comparison techniques i n Figure 17 (Duncan, 1955? Krammer, 1956).— Growth depensation was greater when the l i m i t e d food supply was Table 10.. The average increase i n v a r i a b i l i t y of weight-frequency d i s t r i b u t i o n s during a 6-day period due to differences i n growth rate i n populations of 8 f i s h . Treatment Code 1E 4E 8E 8C4 8C2 801 Mean increase i n v a r i a b i l i t y +0.40 +0.44 +0.24 +0.31 +0.42 +0.49 11/ Pooled variance used i n multiple comparisons was equal to 0.2213 with 46 degrees of freedom. Time i n t e r v a l s were used as a blocking v a r i a b l e . 0.0 Q UJ CO CO UJ a: a. x LU < CO LU CL UJ Q O 0.4 0.2 0.0 o u i o f f i 2 our of 8 4 O U T 0 f 8 8 O J r of 8 NUMBER OF S U B S E C T I O N S C O N T A I N I N G THE FOOD NUMBER OF SUBSECTIONS IN H A B I T A T Figure 17« Relation between growth depensation ( 2 J - (s +.) and (a) the s p a t i a l d i s t r i b u t i o n of l i m i t e d food i n a subdivided habitat and (b) the extent to which the habitat i s subdivided. (Brackets enclose means which do not d i f f e r at p £ 0.05) - 6 7 -s p a t i a l l y concentrated than when i t was evenly d i s t r i b u t e d (Figure 17a). Much of the growth depensation i n the s p a t i a l l y concentrated food treatments resulted from slow growth of one or two i n d i v i d u a l s rather than from extremely f a s t growth of one or two i n d i v i d u a l s . In treatments with an evenly d i s t r i b u t e d food supply, growth depensation appeared to be l e a s t i f there was one subdivision per f i s h , intermediate i f there were no subdivisions, and greatest i f there was one subdivision f o r every 2 f i sh (Figure 17b). Behavior Average number of aggressive a c t i o n s / 2 . 5 min during the 5-minute- period a f t e r food was presented was calculated f o r each treatment (Table 11). Each mean i s based on sixteen 2 .5-minute observations and i s the t o t a l number of aggressive actions by a l l 8 f i s h during the 2 .5-minute period. These Table 11. Frequency of aggressive actions/2 . 5 m i n by 8 f i s h and the c o e f f i c i e n t of v a r i a t i o n -(%) r e s u l t i n g from differences i n frequency of aggression on d i f f e r e n t days of observation during the 5 minutes a f t e r food was provided. Treatment Code 1E 4E 8E 8C4 8C2 8C1 Frequency of Aggressive Acts/ 9 . 0 • 43 .9 51 .4 81.8 36 .9 24.8 2 . 5 Minutes C o e f f i c i e n t of V a r i a t i o n 7 6 . 2 - 3 7 . 4 113.0 56.3 6 2 . 7 3 9 . 8 -68-data are compared by multiple comparisons i n Figure 18 (Duncan, 1955; 12/ Krammer, 1956).— The highest l e v e l of aggression occurred when food was placed i n one out of every two subsections of the habitat (8C4) (Figure 18a). Aggressiveness was l e s s i n treatments with a greater l o c a l i z a t i o n of the food supply and appeared to be les s i n treatment 8E with an evenly d i s t r i b u t e d food supply. Frequency of aggressive actions was more variable from day to day i n 8E which had 10 p e l l e t s of food i n each of the 8 subsections ( c o e f f i c i e n t of v a r i a -t i o n = 113%) than i n treatments with contagiously d i s t r i b u t e d (bunched, clustered or s p a t i a l l y l o c a l i z e d ) food (8C4, 8C2, and 8C1) ( c o e f f i c i e n t of v a r i a t i o n = 56, 63, and k0% r e s p e c t i v e l y ) (Table 11). The frequency of aggressive actions varied between 2 and 198 per 2 .5 min i n 8E. In the case with 198 aggressive actions, 190 were from the aggressive interchanges between a single p a i r which happened to be i n the same subsection at feeding time. I f f i s h i n 8E were evenly d i s t r i b u t e d at feeding time, the l e v e l of aggres-sion was very low. In treatments with an even d i s t r i b u t i o n of food, aggressive actions were more frequent i f the habitat had more subdivisions (Figure 18b). The l e a s t number of aggressive actions was observed when there were no p a r t i a l p a r t i -t i o n s (1E) i n the 8 - l i t e r baskets. The aggressiveness i n 1E u s u a l l y occurred 4-5 minutes a f t e r the food was introduced at which time almost a l l the food had been eaten, and f i s h were concentrated near single p e l l e t s which had sunk to the bottom. As mentioned above, the high l e v e l of aggression i n 8E which 12/ The l o g i o transformation was used to achieve homogeneous variance. Pooled variance of the logarithms of aggressiveness used f o r multiple comparisons between 1E and 8E versus any other treatment was 0.08152 with 42 degrees of freedom, and for comparisons among 4E, 8C4, 802, and 8C1 was 0.04084 with 28 degrees of freedom. X 6 0 CO U_ co >-m 3 0 ES UT z : in 0 CM cc UJ CL CO o u < UJ > CO 6 0 CO UJ cr o < la-CO 3 0 en UJ CD 2 I out of 8 2ou t of 8 4 out of 8 8 out of 8 N U M B E R O F S U B S E C T I O N S C O N T A I N I N G T H E F O O D N U M 8 E R O F S U B S E C T I O N S I N H A B I T A T Figure 18. Relation between frequency of aggressive actions i n populations of 8 f i s h and (a) the s p a t i a l d i s t r i b u t i o n of l i m i t e d food i n a subdivided habitat and (b) the extent to which the habitat i s subdivided i n evenly d i s t r i b u t e d food treatments. (Brackets enclose those means which are not d i f f e r e n t at p i 0 . 0 5 ) -70-had 8 subsections resulted from the chance occurrence of more than one f i s h i n a subsection at the time of feeding. While food was present i n the environment, the d i s t r i b u t i o n of f i s h was p r i m a r i l y determined by the l o c a t i o n of the food (Figure 19)- In the t r e a t -ments with an even s p a t i a l d i s t r i b u t i o n of food (1E, 4E, and 8E) the f i s h were dispersed over the whole environment, but i n treatments with contagiously d i s t r i b u t e d food (804, 802, and 801) f i s h were concentrated i n feeding areas. The number of f i s h moving-into a single subsection increased as the contagion of the food supply increased; f o r example, when food was placed i n one-half the subsections (804), approximately 2 f i s h crowded in t o each food area, but when food was placed i n one-eighth of the subsections ( 8 d ) , 4.8 f i s h crowded i n t o the food subsection to feed. Even though concentrations of 7 f i s h were occas i o n a l l y found i n a single subsection, the average percentage of f i s h i n subsections with food decreased as the food supply was more l o c a l i z e d . One hundred percent of the medaka were i n subsections with food when i t was evenly d i s t r i b u t e d (8E), 90% when i t was i n one-half of the subsections (804), 70% when i t was i n one-fourth of the subsections (802), and 60% when i t was i n one-eighth of the subsections (8C1). Often one, two, or three f i s h found the food subsection a minute or more before the remainder did, and much of the food was eaten by these few i n d i -v i d u a l s . Large f i s h did not appear to set up t e r r i t o r i e s i n the food sec-t i o n s , but oc c a s i o n a l l y a f i s h would be i n the container when the food was introduced. As the contagion of the food supply increased, the contagion of the f i s h d i s t r i b u t i o n also increased a f t e r the food was introduced. Yet the contagion of the f i s h d i s t r i b u t i o n was not proportional to that of the food, and a lower percentage of f i s h were i n food subsections i n those treatments with a greater l o c a l i z a t i o n of food. - 7 1 -© 0.2 © 0. I 0.3 © 0.3 © I E 8 C 4 H O L E IN PARTITION 0 ^© © © 4 E © © ('•') © (0~6) (09) © 8 E FOOD AREA \ 0 . 7 © 0 . 3 0 . 2 0 . 4 0 4 © 0 . 8 8 C 2 0 . 5 0 . 6 0 . 8 0 . 4 0 . 4 0 . 2 0 . 3 8 C I Figure 1 9 . Average number of f i s h i n subsections during the 5 minutes a f t e r food was introduced. -72-F i s h learned to recognize the single subsection which received food i n 801, but i n treatments receiving food i n 2 (802) or 4 (8C4) subsections, they did not learn to recognize subsections which received food. Before food was introduced the f i s h i n 802 and 8C4 were swimming throughout the basket. When food was introduced they swam from section to section u n t i l they located the food. Locating the subsections containing food appeared to be a matter of chance i n the l a t t e r 2 treatments. Often 4 or 5 f i s h would concentrate i n a single subsection while another containing, food was empty. A dominant f i s h would often chase a smaller f i s h out of one subsection and then not return to the subsection which contained the food. Relation between Growth and Behavior The modifying influence of the s p a t i a l d i s t r i b u t i o n of food was studied i n those treatments which were divided i n t o 8 subsections by p a r t i a l p a r t i -t i o n s . Growth depensation was greatest when a l l the food was placed i n one subsection, and i t was l e a s t when the food was evenly d i s t r i b u t e d (Figure 17a). Although dominant large f i s h apparently had a greater competitive advantage when food was s p a t i a l l y concentrated, the data indicated that other f a c t o r s were also involved. F i r s t , frequency of aggression was at a low l e v e l when food was placed i n only one subsection, while i t was at a higher l e v e l when food was placed i n every second subsection or even i n every subsection. As many as 6 f i s h would move in t o a single subsection to feed when food was con-t a g i o u s l y d i s t r i b u t e d and only 10-40 aggressive actions/2 . 5 min resulted, while i n other treatments the presence of two f i s h i n one subsection resulted i n as many as 190 aggressive actions/2 . 5 min. A large f i s h was not able to defend the food a^.ea from the other 7 f i s h , and the presence of more than 1 smaller f i s h lowered the aggressiveness of large f i s h . Second, l o c a t i n g the food section was p r i m a r i l y a matter of chance. In treatments which had only -75-1 or 2 subsections with food, a f i s h not f i n d i n g a food section i n the f i r s t 2-5 days was l e f t f a r behind those which located and learned to feed on the p e l l e t e d food at the s t a r t of the experiment. Af t e r the food was found, i t was r a p i d l y eaten, and the time during which i t was a v a i l a b l e was short. At the end of the experiments, populations i n treatments with a contagious food d i s t r i b u t i o n (802 and 801) had one or two very small f i s h which did not even enter the food subsections, while the other f i s h entered a food subsection as soon as they found i t even i f there were larger f i s h there f i r s t . These small f i s h may have been chased from the food section at the s t a r t of the experi-ment, or perhaps they never learned to search f o r the food immediately a f t e r i t was introduced. In conclusion, i f food was l o c a l i z e d i n only a few subsections of the habitat, high growth depensation resulted e i t h e r because the smallest f i s h were chased from the food areas e a r l y i n the experiment, the small f i s h , due to chance d i s t r i b u t i o n , never had the opportunity to learn to locate and feed on the l o c a l i z e d food p e l l e t s , or because smaller f i s h were slower and had l e s s chance of l o c a t i n g the food subsection during the short periods while food was present. When food was placed i n every second subsection (802) aggressiveness was high and f i g h t i n g occurred between the two f i s h which entered each subsection containing food. Growth depensation i n t h i s treatment was more l i k e l y a product of the s o c i a l hierarchy than chance. F i s h seldom f a i l e d to f i n d a subsection containing food, but often chased, a second f i s h out of a food area. Aggressive behavior did not r e s u l t i n a dispersed d i s t r i b u t i o n of medaka unless food was present i n every subsection of the h a b i t a t . In t h i s case (8E) f i g h t i n g was very intense i f a f i s h happened to f i n d i t s e l f i n a sub-section with both food and another medaka. -74-E f f e c t s of p a r t i a l i s o l a t i o n between competitors were studied i n t r e a t -ments which received an even s p a t i a l d i s t r i b u t i o n of food but had no sub-sections (1E), 4 subsections (4E), or 8 subsections (8E) separated by p a r t i a l p a r t i t i o n s . The data indicated that growth depensation decreased when the environment was subdivided so that each f i s h had a subsection containing food (Figure 17a). I t i s doubtful that aggressive behavior provided the large f i s h with a s i g n i f i c a n t competitive advantage when no subsections were i n the environment (1E). Aggressive actions were l e a s t frequent i n the treatment with no p a r t i a l p a r t i t i o n s , although d e f i n i t e s o c i a l h i e r a r c h i e s were estab-l i s h e d . Large f i s h appeared to eat more r a p i d l y and ate more because they could graze e f f e c t i v e l y over the whole environment. Food p a r t i c l e s on the f a r side of the environment were never temporarily inaccessable due to a maze of subsections such as were present i n other treatments. The advantage to the large f i s h i n terms of rate of feeding was noted e s p e c i a l l y i n the consumption of larger food p a r t i c l e s . Large f i s h swallowed these p a r t i c l e s with apparent ease, while smaller f i s h u s u a l l y made several nips at the larger p a r t i c l e s and then rejected them i n preference to a smaller p a r t i c l e nearby. P a r t i a l p a r t i t i o n s provided p a r t i a l i s o l a t i o n between the competi-t o r s . In treatments with subsections, the food supply of the small f i s h was protected,and the rate of feeding was not as important a fa c t o r i n deter-mining the t o t a l amount of food eaten by a f i s h . In the environment with 4 subsections or 2 f i s h per subsection (4E), aggressive behavior again appeared-to be more important i n determining which f i s h ate the p e l l e t s . Aggressive behavior apparently served better as a competitive mechanism i n those s i t u a t i o n s i n which competition was p r i m a r i l y i s o l a t e d as an i n t e r a c t i o n between two f i s h . When more than two f i s h were involved a medaka was not able to defend the food area from a l l intruders, and both the frequency of aggression and the growth advantage i t gave the dominant appeared to decrease. Summary of Results Growth depensation increased i f the food was more contagiously d i s t r i -buted, but aggressiveness was most intense when a subsection of the environ-ment contained food f o r every two f i s h . I f food was concentrated i n only one-eighth or one-fourth of the habitat, growth depensation resulted apparently because some f i s h , due to lack of opportunity, never learned to feed on the p e l l e t s . They did not learn to eat the p e l l e t s e i t h e r because the large f i s h chased them from the food areas before small f i s h became conditioned to the p e l l e t s or because chance phenomenon e a r l y i n the experiment resulted i n some f i s h not f i n d i n g the food during the short time i t was a v a i l a b l e . A large medaka was not able to defend the food areas from the 4 to 5 other f i s h which also entered and fed on the p e l l e t s , but a smaller proportion of f i s h were i n food areas when food was contagiously d i s t r i b u t e d . The high l e v e l of aggres-sion i n a treatment with food d i s t r i b u t e d i n a l l subsections resulted from f i g h t s between two f i s h which were situated by chance i n the same subsection at feeding time. A dispersed d i s t r i b u t i o n of medaka due to aggressive behavior was not observed unless n e c e s s i t i e s of the subordinates were found i n a l l subsections of the h a b i t a t . I f food was evenly d i s t r i b u t e d i n the environment, growth depensation was le s s i f there was one subsection i n the habitat f o r each f i s h . Growth depensation i n treatments with a subsection f o r every two f i s h resulted from the consequences of aggressive behavior. When there were no subsections, growth depensation resulted because large f i s h could eat the larger food p a r t i c l e s more r a p i d l y than small f i s h could, and the rate of feeding during -76-the short time i n which food was present determined the amount of food eaten per f i s h . Aggressive behavior i n the l a t t e r treatment was at a low l e v e l . As mentioned above, aggressiveness was high when there was 1 subsection f o r each f i s h due to the chance d i s t r i b u t i o n of more than one f i s h i n a subsec-t i o n at feeding time. Aggressive behavior appeared to be more fun c t i o n a l as a competitive mechanism f o r l i m i t e d food when competition was p r i m a r i l y i s o l a t e d to i n t e r -actions between two f i s h . I f population size,increased or i f more than two f i s h competed f o r the food i n one subsection, a d d i t i o n a l f a c t o r s such as rate of feeding, chance, learning, etc. had a great influence upon the growth consequences of competition. When food was evenly d i s t r i b u t e d and there was a subsection of the habitat f o r each f i s h , aggressive behavior dispersed the f i s h . In t h i s case the subordinate was no longer at a competitive disadvantage. DISCUSSION Competing animals can, influence each other, i n at l e a s t two ways i n terms of growth consequences.of competition., F i r s t , i f they share-a l i m i t e d resource i n such a way that no genotypic or pheno.typic character provides one f i s h with a p r i o r i t y f o r the resource, poor growth consequencesof competi-t i o n w i l l be equally d i s t r i b u t e d amongrall members of the, population. Second, i f any f i s h has.a c h a r a c t e r i s t i c which gives, i t f i r s t choice or access to the resource, detrimental, growth consequences- of. competition.will be unequally d i s t r i b u t e d among, members-of the population. A gradation might exist,, from equal, sharing of. the. resource to complete-possession of the resource by one. or few i n d i v i d u a l s . Situations i n which.the resource i s not shared; have been c a l l e d "contest," and si t u a t i o n s i n which the. resource i s shared have been c a l l e d "scramble" .by Nicholson; (1954). Aggressive behavior i s a mechanism.which might, provide, c e r t a i n animals (those which win) with.a competitive advantage, and they would get more than t h e i r share of the. resource i n question.. I f the a c q u i s i t i o n of the resource had any e f f e c t upon the growth of. the animal, competitive situations, in.which aggressive behavior was important might be characterized by a wide v a r i a t i o n i n growth rates among members of. the population. Many other f a c t o r s pro-ducing the- same* effects, must be accounted f o r or eliminated before the r o l e of an aggressive-behavior mechanism and growth consequences of competition can be determined. Even. i f . animals, shared the resource with no c l a s s d i s t i n c t i o n , genetic differences, i n the growth p o t e n t i a l would r e s u l t in. growth depensation. (an expanding size d i s t r i b u t i o n due to differences in. growth r a t e ) . Some, workers -78-(Brown, 1946; A l l e e et a l . 1948) neglected to compare, the v a r i a b i l i t y i n growth among f i s h . i n populations with the v a r i a b i l i t y among.those which were rais e d i n i s o l a t i o n . By doing t h i s , these i n v e s t i g a t o r s assume.that none of the genetic, differences i n growth rate were large enough, to bias t h e i r con-c l u s i o n s . A considerable amount of growth depensation, however, i s observed among the o f f spring,..of a single p a i r of medaka from a h i g h l y inbred domestic stock even when they are r a i s e d i n i s o l a t i o n under " i d e n t i c a l " environmental conditions. A l l e e et a l . (1948) publish.data, which, indi c a t e that genetic differences should not have been neglected. They raised:immature, green sunfish both i n i s o l a t i o n and i n populations of, 4 and. noted: that growth of individual, f i s h i n populations was-positively associated with social, rank of the f i s h . . The i m p l i c a t i o n i s that s o c i a l rank resulted i n observed differences i n growth ra t e . Yet i n their. Table 2 (page 7) the v a r i a b i l i t y i n growth rates observed i n populations of 4 f i s h i s i d e n t i c a l to the v a r i a b i l i t y observed among the same, number of i s o l a t e d f i s h . This- demonstrates that- the- same v a r i a b i l i t y existed even-in the: absence of s o c i a l i n t e r a c t i o n s . Probably the r e l a t i v e size of each.fish was associated with i t s rate of growth,, and i n addition the r e l a t i v e s i z e determined the p o s i t i o n of. the f i s h i n the. social, hierarchy, but there was.no causal r e l a t i o n s h i p between rate of growth, and p o s i t i o n i n the s o c i a l hierarchy; Greenberg (1947) demonstrated, t h a t ' r e l a t i v e size i s important in. determining, the s o c i a l rank of an i n d i v i d u a l .green sunfish. A number-of. other factors, known to r e s u l t i n growth, depensation must also be removed to i s o l a t e , e f f e c t s of aggressive behavior i n competition. The accumulation,of excreted substances r e s u l t s i n growth i n h i b i t i o n . Rose (1959> 1960) demonstrates, that these water-borne i n h i b i t o r s have a greater influence on smaller members -of a population;and r e s u l t i n growth depensation. - 7 9 -In a d d i t i o n growth depensation occurs among immature carp, Oyprinus. carpio, i f the size: of. the; food p a r t i c l e i s large and i t s abundance i s low (Nakamura and Kasahara, 195^). Extreme growth depensation, was, observed among young-of-the-year smallmouth. bass,' Micropterus- dolomieui Lacepede; resulting, from cannibalism of. the smallest by the l a r g e s t members of. the population (Langlois, 1956). To the best o f the author's knowledge a l l of the above f a c t o r s r e s u l t i n g i n growth depensation have been removed or accounted f o r i n the design and a n a l y s i s of the present experiments (see materials>and methods, and descrip-t i o n of experiments) though, there may be other unknown f a c t o r s which have been neglected.' To study competition, the; p a r t i c u l a r resource f o r which competition i s occurring should be determined, and.its abundance or a v a i l a b i l i t y should, be va r i e d . The two resources of the environment considered i n these studies were food, and space. Space i s a more tenuous."resource" than food. Livings space, or Lebensraum ha-s=.long, been considered important as an e c o l o g i c a l f a c t o r (see A l l e e e t a l , 1949, p . 2 2 ; Larkin,, 195°"), and many hypotheses, were, put forward to explain slower growth of aquatic vertebrates and invertebrates which were reared i n smaller volumes of water or at higher population.densities. Among Rana pipiens:tadpoles, the, hypotheses that surface area, r e s t r i c t e d movement, or c o l l i s i o n s : between i n d i v i d u a l s , inhibited, growth at higher-population de n s i t i e s were found to be unnecessary by Richards: (1958) . She,was-able.to explain most of these r e s u l t s by the presence: of a growth i n h i b i t o r y substance i n the feces which accumulated at higher concentrations, i n smaller more crowded containers. Growth i n h i b i t i o n r e s u l t i n g from ammonia excretions was also demonstrated i n t r o u t (Brockway, 195°) and i n carp. (Kawamoto, 1961). The -80-need for space, per se, in, aquatic animals i s primarily a. consequence, of accumulating, waste-products i f food i s supplied i n excess. Whether, space, i n this.context can be'considered as a resource i s a matter of opinion and definition. Other aspects of space as i t i s involved in competition w i l l be discussed in conjunction with, competition for food. Food i s easily visualized as a resource, but some confusion arises when attempts, are. made, to, supply i t "in excess." Food can be limited i n amount, not only spatially, but also temporally; In addition some, foods are more stimulating and are eaten, in, greater amounts. If a food i s present "in excess" for only short periods of time, fish, w i l l not always be feeding to satiation. Brown: (1946, 1951* 1957)t for example, fed an excess.of minced li v e r twice a day to brown trout fry, but observed that the l i v e r was only eaten while suspended in the water. Consequently the f i s h would have no food available as soon as a l l particles f e l l to the bottom.. Medaka feed to satia-tion on brine shrimp nauplii, but within one or two hours begin to search for food again. Two meals.per day can not be considered as "excess"; instead palatable food should be present at a l l times-. If food i s not highly stimu-lating ...to f i s h , i t i s debatable whether fis h can. be fed in.excess except i n relative terms;. Medaka.would feed on living brine shrimp nauplii after they ceased, to feed on<pellets or-frozen brine shrimp,. . Oare: must be taken, i n interpreting results when i t i s assumed..that food i s supplied in excess, especially when the excess i s present only during short periods of time or the food i s not very stimulating to the f i s h i n terms of i n i t i a t i n g feeding behavior. When, the abundance of individual resources, i s varied, the. action of aggressive behavior-is observed, and. care i s taken to remove-extraneous factors influencing growth, then both the growth consequences of competition -81-f o r a s p e c i f i c resource and the .action of the aggressive behavior mechanism can be studied. In the .present experiments the amount of space was-varied by r a i s i n g equal numbers of, f i s h , i n containers of d i f f e r e n t sizes, and. by r a i s i n g d i f f e r -ent numbers of f i s h i n containers of the same: sizes;. In both cases food was provided " i n excess," and accumulation of water-borne growth i n h i b i t o r s was prevented. In these s i t u a t i o n s , l i m i t i n g the amount of space d i d not reduce growth,rates nor cause growth depensation. Aggressive behavior was not more common; when space-was most l i m i t e d ; The lowest frequency of aggression was observed when; the-least space was supplied. Aggressive behavior did not.in t h i s case function as a competitive mechanism, f o r space>.and, i n f a c t , compe-t i t i o n f o r space- was, apparently, not taking place i n medaka at de n s i t i e s up to 16 f i s h per l i t e r . Neither a general depression i n growth, rate nor growth.depensation was observed i n medaka. populations, r e l a t i v e to i s o l a t e s i f food was supplied " i n excess." The presence of one animal did not lower the. growth rate of a l l other members of the-population, nor did any f i s h have a competitive advan-tage f o r food over any other f i s h ; . Aggressive behavior was: at a very low l e v e l , and both large and, small members' of: the: population were equally aggressive; The resource was present " i n excess" amounts, both s p a t i a l l y and temporally, the aggressive- behavior mechanism was not operating, and evidently there was no competition of any. sort occurring i n these-popula-t i o n s . Yet when food was l i m i t e d i n supply a s o c i a l hierarchy, was.-estab-l i s h e d i n which, large f i s h dominated small f i s h . In,addition, average growth rates of a l l f i s h reared i n populations were less, than among i s o l a t e d f i s h fed- the same amount, of food, and growth depensation. occurred. Small f i s h d i d not show any reluctance:to feed, but did not get many, food p a r t i c l e s , because - 8 2 -large dominant f i s h kept chasing, them, away from food. Competition f o r food was taking,place, and. the aggressive'behavior mechanism, was operating i n a way which resulted i n the l a r g e . f i s h g e t t i n g a greater share of the l i m i t e d supply of food. Nakamura and Kasahara (1956, 1957) observed that the amount of growth depensation i n carp decreased when more food was supplied and that l i t t l e growth depensation occurred among.fish, grown i n . i s o l a t i o n . Immature carp were not observed to be aggressive..12/ Brown (19^6, 1951> 1957) observed that l a r g e r s o c i a l l y dominant brown trout i n h i b i t e d the growth of smaller ones even when food was supplied " i n excess" and a good c i r c u l a t i o n of water was provided. I t i s d i f f i c u l t to evaluate her conclusions; because she assumes that the f i s h , were; fed- " i n excess" even though, they were, given minced l i v e r only twice each day. Since Brown assumed that food was,'supplied " i n excess," the s o c i a l h i e r a r c h i e s which she observed could not be considered i n terms of competition f o r food. She. postulated that the very presence of the large f i s h . in. some way, p o s s i b l y " s t r e s s , " resulted i n slower growth among small brown t r o u t . In medaka, small f i s h , in. the presence of large f i s h grew as well as i s o l a t e d controls when food was " i n excess." Also when medaka were not fed at a l l , b o t h.fish did equally poorly even though the large f i s h was quite aggressive towards the small f i s h . . Only when food supply was l i m i t e d did dominant, large f i s h , have* a- competitive advantage over small f i s h in. terms of growth. Evidently any "stress." provided by the presence of large f i s h was-not important, but access: to. l i m i t e d food aug-mented by social, dominance was-important. Aggressiveness of medaka was i n part i n i t i a t e d by food s t i m u l i . Fre-quency of aggressive actions increased when food was added i n a l l l i m i t e d . 15/ Behavior observations, reported to the author i n a l e t t e r dated May. 18, 1961, to Mr. Taizo Miura from Mr. Kenji Chiba, Freshwater Fishery Research Laboratory, Miya, Hino Machi, Minamitamagun., Tokyo, Japan. - 8 3 -food treatments, but did not increase i n excess food treatments. An increase i n aggression a f t e r food i s presented has been observed among.the f i s h e s i n immature Do l l y Varden.;. ( Salvelinus,: malma fWalbauml) , cutthroat t r o u t (Salmo  c l a r k i Richardson), coho salmon. (Oncorhynchus. kis u t c h CWalbaum]) (M. Nevmian, 1960); rainbox-f t r o u t (Salmo..gairdneri Richardson) and brook trout. (Salvelinus  f o n t i n a l i s [ M i t c h e l l ] ) (M. Newman, 1956); brown trou t and. A t l a n t i c salmon (Salmo.. salar Linnaeus) (Kalleberg* 1958); Iowa;, darter (Etheostoma.exile [Girard) ) and f a n t a i l darter- (Etheostoma..flabellare Rafinesque) (Winn, 1958); and in.mature white cloud mountainfish. (Tanichthyss albonubes Lin) (unpub-l i s h e d observations by the author). This, r e l a t i o n between food and aggres- • siveness i s probably a common phenomenon i n many species of f i s h and, as i n medaka, could function, to reserve a larger portion of. food f o r dominant f i s h . Aggressive- behavior of. immature medaka:-can. be thought of as a mechanism.which p o t e n t i a l l y provides, dominant f i s h with a competitive advantage and which i s i n i t i a t e d by the i n t e r n a l state of. "hunger," and the: presence of food s t i m u l i and smaller medaka. Addi t i o n a l f a c t o r s other than.the amount, of food and. a f i s h ' s p o s i t i o n i n the s o c i a l hierarchy influence growth consequences: of competition f o r food. The-extent to which .the food supply i s , shared and also, the extent to which the o v e r a l l growth rate i s depressed-due to l i m i t e d food supply depends i n part upon modifying influences of many environmental f a c t o r s as-discussed below. F i r s t , the modifying influence of the s p a t i a l d i s t r i b u t i o n cf food i s considered. Medaka concentrate i n the parts of the environment containing food. In populations of. two f i s h the s o c i a l l y dominant medaka. defends and occupies, the food area while food i s present and f o r several hours a f t e r i t i s a l l eaten. Aggressive actions i n t h i s case are- more e f f i c i e n t i n -84-i n reserving the, food f o r the dominant than when food i s scattered through the environment. I f excess food i s s p a t i a l l y l o c a l i z e d in,the environment, however, the two f i s h share the food, with no d i f f e r e n t i a l access. The same conditions ( i n t e r n a l state of "hunger," food s t i m u l i , and smaller medaka.) i n i t i a t e aggression when food i s either. s p a t i a l l y concen-trated or evenly d i s t r i b u t e d , but t h e - l o c a l i z a t i o n of food added s i t e attach-ment to. the behavior pattern. Noble (1939) defined a: t e r r i t o r y as "any defended area," and Tinbergen (1957) defined i t as a combination, of i n t r a -s p e c i f i c h o s t i l i t y and s i t e attachment. In terms of these two general d e f i n i t i o n s , the l o c a l i z a t i o n of the food supply was. s u f f i c i e n t to change a s o c i a l h i e r a r c h i c a l society into a t e r r i t o r i a l society i n the. medaka. Both Hinde . ( l 9 5 6 ) a n < i Tinbergen (1957) have discussed the. v a r i e t y of objects which, animals ( p r i m a r i l y b i r d s ) defend i n t e r r i t o r i a l behavior. Tinbergen points out that " h o s t i l i t y without s i t e attachment may. serve the purpose-of reserving v i t a l objects just as. well as the two tendencies combined — i t a l l depends upon the nature of.the. defended, object." The present study i n d i c a t e s that ( i ) the nature of the defended object ( i t s s p a t i a l d i s t r i b u t i o n and m o b i l i t y ) brings out the. expression of s i t e attach-ment, ( i i ) the defense of. the o b j e c t . i s more e f f i c i e n t , i f . i t i s s p a t i a l l y l o c a l i z e d , and ( i i i ) aggression only functions, to reserve the food i f the supply i s l i m i t e d . Dependence, of s i t e attachment upon the nature of the resource or object defended i s probably quite common, i n fishes-. The. stationary nests b u i l t by male, ten-spined; sticklebacks, Pygosteus; pungitius Linnaeus. (Morris, 1958), and the p i t s dug by some c i c h l i d fishes, during the reproductive period (Baerends- and Baerends-van Roon, 195°) form the center, of their, t e r r i t o r i e s . Morris (1958) argues that the center of the t e r r i t o r y i s more stable than the periphery. In the b i t t e r l i n g , . Rhodeus;amarus Linnaeus, i f the "nest s i t e " moves, the male f i s h ' s t e r r i t o r y moves- also (Boeseman. et a l . 1938j as: c i t e d i n .Tinbergen, 1951). B i t t e r l i n g s . deposit, their, eggs i n . the. mantle c a v i t y of a freshwater clam,, and. the. male, defends the, .clam as-a moving t e r r i t o r y . Winn (1958) observed that c e r t a i n darters, subfamily, Etheostomatinae, defended a reproductive t e r r i t o r y i f eggs were on the rock, but l o s t i n t e r e s t i n the rock i f eggs:were removed. Likewise, the male moved from trock to rock de-fending each f o r only a short time i f no.eggs were l a i d on a rock which i t defended. Other species-of darters have a t e r r i t o r y which moves about with the female; Winn. (1958) also observed that•those- species of darters which l a i d t h e i r eggs i n one spot and.had well developed t e r r i t o r i a l defense-were l e s s fecund than darters which did not have t h i s behavior as h i g h l y developed. Evide n t l y egg defense, in. t e r r i t o r i a l s i t u a t i o n s i s . more, e f f i c i e n t . Iri summary, t h e r e - i s often, some-object i n the t e r r i t o r y about which defense i s centered. In immature medaka. this, object is, food. S p a t i a l d i s -t r i b u t i o n of the food determines:-not only whether t e r r i t o r i e s or s o c i a l h i e r a r c h i e s w i l l be formed, but also the:'efficiency of - aggressive, action i n terms, of the. competitive advantage to the dominant. A second p e c u l i a r i t y of the environment i n f l u e n c i n g the action of the aggressive behavior mechanism and the growth consequences, of competition i s the amount o f i s o l a t i o n between..competitors. In.the present experiments p a r t i t i o n s with: and without holes through them-,were, used to. p r o v i d e - p a r t i a l and complete i s o l a t i o n between-competitors. The-partitions, can be thought of as a s t y l i z e d form of obstruction provided by rocks,, vegetation, t u r b i d i t y or distance i n nature. This aspect of the environment must always be considered i n conjunction with the s p a t i a l d i s t r i b u t i o n of food. I f food i s evenly d i s t r i b u t e d , complete i s o l a t i o n prevents.all -86-i n t e r a c t i o n between the competitors, and the two f i s h , share the food resource equally... Sharing i n t h i s case i s induced entirely..by c h a r a c t e r i s t i c s of the h a b i t a t . As the amount of i s o l a t i o n decreases* aggressive i n t e r a c t i o n s between, competitors increase* and the dominant gets an. increasing proportion of the l i m i t e d food. In a complicated environment, however, the dominant.can defend and reserve the food i n one area, but the-subordinate i s l e f t to feed unmolested i n another part of the environment. This occurs only when there i s a p a r t i a l l y i s o l a t e d area of the environment f o r each f i s h . This p a r t i a l i s o l a t i o n need- only be present while food i s a c t u a l l y in.the environment. Medaka-move from one area to another* but the dominant can only be e f f e c t i v e i n one area at a time. The small f i s h avoids the area i n which the dominant i s feeding and i s f o r the most part at no competitive disadvantage even though i t i s s o c i a l l y subordinate and food supply i s l i m i t e d . When a l l i s o l a t i n g b a r r i e r s are removed from the environment, the dominant medaka i s able to prevent the subordinate from feeding by chasing i t away from:food. Among other, species-which, defend more permanent, t e r r i t o r i e s than: medaka, there are many cases i n which topography of the habitat either, influences the siz e or the borders of the t e r r i t o r y . Kalleberg (195$) observed i n A t l a n t i c salmon f r y that t e r r i t o r i e s were more c l o s e l y packed i f the topography of the substrate was interrupted by l a r g e r rocks, and that t e r r i t o r i e s were smaller i n t u r b i d water. Greenberg (1947) noted that more t e r r i t o r i e s were formed among green.sunfish i f p a r t i a l p a r t i t i o n s were placed i n the aquaria. Vegetation can be used to demarcate and to increase the p o t e n t i a l number of breeding t e r r i t o r i e s i n male- O o l i s a labia,. ( F o r s e l i u s , 1957), and F a b r i c i u s (1950) demonstrated that the size-and. shape ;of reproductive t e r r i t o r i e s were i n part determined by the substrate and the presence, and density of vegeta-t i o n i n male white cloud, mountainfish and i n the bream,. Abramis brama. In summary, the topography of the habitat or the v i s u a l i s o l a t i o n between- competitors- determines both the. number of f i s h that can occupy a given area and. the extent to which food, d i s t r i b u t e d evenly i n that area w i l l be shared. The more v i s u a l i s o l a t i o n i n the environment the..smaller e f f e c t aggressive behavior mechanism, has. i n reserving a larger proportion of the food supply f o r the. dominant. The above i s true only when food i s evenly d i s t r i b u t e d . I f food i s con-ta g i o u s l y d i s t r i b u t e d aggressive behavior does not disperse the medaka throughout the habitat.. They concentrate, i n the food areas. In t h i s s i t u a -t i o n the amount of v i s u a l i s o l a t i o n a lso influences the number of f i s h i n a given, area and the extent to which the food is- shared. In contrast to the evenly distributed, food, i f , food i s . contagiously d i s t r i b u t e d the more v i s u a l i s o l a t i o n . i n . the environment the-greater e f f e c t aggressive behavior mechanism has in. reserving a larger proportion of the food supply f o r the. dominant. A t h i r d complication develops i n r e l a t i o n .to s p a t i a l d i s t r i b u t i o n of food and, topography of the habitat, i f populations are large. Aggressive behavior i s most e f f e c t i v e i n the defense- of a resource i f competition i s occurring between, only two f i s h at one time; A large dominant medaka. could defend the food from one smaller medaka but not from several, smaller medaka at the same time. The same-phenomenon, was-observed. (Swingle, and,,Smith, 1943) i n largemouth bass., Micropterus? saimoides, (Lacepede), while they.def end t h e i r nest from egg pr.edation by b l u e g i l l s , Lepomis macrochirus Rafinesque. Bass egg, s u r v i v a l was very low at high de n s i t i e s of b l u e g i l l s , because-while the male bass chased, one b l u e g i l l from the.nest area, several, others would dart i n from the other side. There i s at l e a s t one case i n f i s h e s (Berwein, 1941) i n which, groups of Phoxinus a c t i v e l y drive o f f smaller, or larger i n d i v i d u a l s or another group. These, s i t u a t i o n s may be rare among the f i s h e s , but even -88-wherethey do occur i t i s l i k e l y that present considerations would, apply because; the. group at l e a s t temporarily would-be acting, as. an. i n d i v i d u a l . In. medaka,. not only was aggression, i n e f f e c t u a l , i n reserving, the-resource when, there were many subordinates,, but also aggressive actions of the domi-nant, were-less-frequent. Among;.green sunfish the. highest, frequency of aggression occurs, at intermediate d e n s i t i e s (Greenberg, 1947). Evidently at low densities, the f i s h do not come; i n t o contact. ( s p a t i a l , i s o l a t i o n ) as. f r e -quently, and at high densities, .aggressiveness- i s i n some:..way i n h i b i t e d . I t i s . doubtful that the dominant medaka "decided" i t would not be able to chase a l l . t h e - f i s h away and had better eat as. much as. possible,. Instead the presence of multiple s t i m u l i probably had an i n h i b i t i n g or confusing e f f e c t on the directed attacks, of the dominant. Ayu, Plecoglossus: a l t i v e l i s Temminck and-Schlegel, defend f e e d i n g - t e r r i -t o r i e s i f population density i s low, and those:; defending . . t e r r i t o r i e s seem to grow f a s t e r than those which are not able to maintain t e r r i t o r i e s - a n d form schools. (Kawanabe, 1958). At- high, densities/ none- of. the f i s h are: able to defend; t e r r i t o r i e s , and, t e r r i t o r i a l , behavior breaks up,into a. schooling.soci-ety. Kalleberg-(1958) observed that i n large-populations of young.trout i n hatchery troughs no t e r r i t o r i e s are* established! at intermediate d e n s i t i e s the population forms; two f a c t i o n s , t e r r i t o r y holders and.non-territory holders, and at lower-densities, s i t e attachment i s maintained by a l l . i n d i v i d u a l s . Anabantid f i s h e s ( F o r s e l i u s , 1957) would not form reproductive t e r r i t o r i e s at high d e n s i t i e s , but. when- the. density was reduced would almost immediately begin.to set up breeding, t e r r i t o r i e s . In a, t h e o r e t i c a l mathematical a n a l y s i s of social, h i e r a r c h i e s , Landau. (1951) demonstrated that a s o c i a l bias ( f o r example; r e l a t i v e - s i z e , p r i o r residence,; etc.) is. l e s s e f f e c t i v e i n establishing, h i e r a r c h i e s ' i n large populations than, in. small populations. -89-In summaryi aggressive behavior even with s i t e attachment i s only e f f e c -t i v e as a competitive mechanism.to the dominant i f the. number, of subordinates contacted at any one time i s small. I t i s a competitive mechanism-function-ing primarily, at the l e v e l of the i n d i v i d u a l . A fourth consideration i s that not only the abundance of food, but also the r e l a t i o n between size of food p a r t i c l e and si z e of a f i s h ' s mouth i s important i n determining the amount of food a v a i l a b l e to a p a r t i c u l a r f i s h . Even i f the food i s small enough to be eaten by the smallest f i s h the size of the food p a r t i c l e i s important i n that i t determines^the rate at which a f i s h can eat-the food. Smaller, medaka can not eat the l a r g e r p a r t i c l e s as f a s t as the large f i s h can. In a "scramble" type of competition i n which a l l f i s h have equal, access to the food, the t i m e r involved i n eating an i n d i v i d u a l food item i s important. I f food i s present only f o r a short time the small f i s h does, not get as much food as the large f i s h . Even i f the food were a v a i l a b l e f o r long periods of time but scarce and widely scattered, the time spent eating each p a r t i c l e a f i s h found would reduce the amount of time, i t had f o r searching. The importance of t h i s l a t t e r aspect i s emphasized: by H o l l i n g (1959) who develops a mathematical model f o r predation i n which a decreased searching time i s a v a i l a b l e f o r a predator at high, prey densities due to the increased proportion of time spent a c t u a l l y devouring the prey. Hartman (1958) observed that rainbow trout could swallow- smaller, stone-f l y nymphs (Plecoptera) but often, rejected larger nymphs even though, they f i t t e d i n t o the mouth because the stonefly would anchor i t s e l f to the nose- of the f i s h and attempt to crawl out. He observed also that smaller f r y rejected smaller s t o n e f l y nymphs than, did larger trout; The same r e l a t i o n -ship was observed i f . caddis f l y larvae (Trichoptera) with the cases removed were used f o r food. This time spent attempting to eat an i n s e c t but f i n a l l y -90-r e j e c t i n g i t would reduce hunting, time. The. studies, by Nakamur.a and Kasahara (1956) demonstrate that growth depensation i s greater i f . the food p a r t i c l e s i z e i s larger but decreases i f increased amounts of t h i s food i s provided. These data a l l i n d i c a t e that i f food i s l i m i t e d i n supply e i t h e r temporally or s p a t i a l l y , the rate of feeding may be an important f a c t o r determing the growth consequences of competition. In t h i s respect a species feeding on. small food p a r t i c l e s would be more l i k e l y to share the food than?fish feeding upon large food, p a r t i c l e s . I f there are 20 small, p a r t i c l e s two f i s h would "scramble" f o r the resource, but i f there i s one large p a r t i c l e i t would be-impossible to share the resource i n the f i s h world. Assuming that t h e - p a r t i c l e could be swallowed i n t a c t , only one f i s h could eat i t . In summary, the-size of the food p a r t i c l e i s important i n determining the extent to which l i m i t e d food i s shared because the size determines the s c a r c i t y of food p a r t i c l e s , the rate, a f i s h can eat the-food, and whether the p a r t i c l e can be eaten..at a l l . The action and consequences; of aggressive behavior described i n the present i n v e s t i g a t i o n function i n laboratory populations of immature medaka, but i t i s d i f f i c u l t to determine to what extent the findings can be generalized. Some; f i s h e s have not been observed to demonstrate aggressive behavior, and among these f i s h e s the mechanism would not f u n c t i o n . Other f i s h are only aggressive i n the breeding season and are t e r r i t o r i a l only i n terms of pro-curing a mate or defending the eggs and young. Yet as observations on behavior of immature, f i s h accumulate i n the l i t e r a t u r e i t becomes, i n c r e a s i n g l y obvious, that many species defend t e r r i t o r i e s or are aggressive during, the sexually immature period of t h e i r l i f e . Aggressive behavior among, juveniles -91-has been observed i n green sunfish,. (Greenberg,- 1947); medaka; ayu, (Kawanabe, 1958), i n niany of the- salmonids, (Hoar,. 1954; Stringer and Hoar, 1955; Kalleberg, 1958; M. Newman,- 195&; Lindroth, 1955, and.others); and i n several^ marine species as. c i t e d by Kalleberg (1958). In, many, of these species and i n the adults of additional, species aggressive behavior i s associated with feeding, as-mentioned previously. Evidently the mechanism described here i s p o t e n t i a l l y a v a i l a b l e f o r observation in, many other species. Evidence i n d i c a t e s that the aggressive mechanism, .would, function f o r a wide, v a r i e t y of species under the r i g h t environmental conditions.. Medaka (Kawabata, 195^, I960), f o r example, tend to be a schooling f i s h i n nature but i n confined conditions demonstrate? both social, hierarchy and. t e r r i t o r -i a l i t y . Young sockeye salmon also school in. nature but i n the-laboratory form h i e r a r c h i e s (H. Newman, 1959). I t can be argued, that aggressive behavior may be only a laboratory phenomenon-which would, not function i n nature* Individuals of a species demonstrating, t h i s behavior i n confinement, however, do possess, the genetic potential, to respond as-a s o c i a l dominant. Those species, demonstrating aggressive behavior i n the laboratory would also be expected to demonstrate the same type. of. behavior i f s i m i l a r conditions were found i n nature. Care, must be taken i n any attempt at g e n e r a l i z a t i o n from one species to another because aggressive behavior d i f f e r s even i n c l o s e l y related.species. Winn, (1958) organized the phylogeny of the Etheostomatinae i n part on the differences in. t h e i r reproductive t e r r i t o r i a l i t y ; Some species were not t e r r i t o r i a l but demonstrated h i e r a r c h i e s , others demonstrated moving, t e r r i -t o r i e s , and others, had d e f i n i t e t e r r i t o r i e s with s i t e attachment. The same could be expected i n food competition among immature f i s h e s . Lake trout, Salvelinus namaycush (Walbaum), are not aggressive-as. immatures. even at -92-feeding time. .(M. Newman, .1960), even though many of i t s close r e l a t i v e s are aggressive; I t i s also d i f f i c u l t to generalize even among f i s h , that are known to be aggressive because there are a large number;of. a l t e r n a t i v e functions which aggressive behavior might, serve i n d i f f e r e n t species. For example,, Carpenter (195$) l i s t s 32 functions of t e r r i t o r i a l behavior which have been postulated f o r animals;;. . C i c h l i d fishes, i n a t e r r i t o r i a l mosaic pay no attention to t e r r i t o r i e s i f food i s placed i n the environment,.and they a l l rush, i n t o the food area (Baerends and Baerends-van. Roon,. 1950) • Ka-lleberg (1958) demonstrates that. the. feeding t e r r i t o r i e s of immature A t l a n t i c salmon break up i n t o a schooling society i f the water v e l o c i t y i s - reduced to zero. Diebschlag (1941) observed i n pigeons .that one* chased other pigeons, from: i t s t e r r i t o r y i n which food was placed only a f t e r i t had fed to s a t i a t i o n . Medaka, on- the other hand, only chased, f i s h from the food area i f they were not s a t i a t e d . Apparently aggressive behavior and t e r r i t o r i a l i t y have many d i f f e r e n t functions which operate in. some, cases quite d i s t i n c t l y . The func-t i o n i n food competition,, as depicted by these experiments, i s just: one of the functions aggressive behavior might be-expected to serve i n other species. As pointed, out i n the, present study, a d e t a i l e d consideration of. the environment i s necessary.to determine the action of aggressive behavior and the growth consequences.of competition f o r food. The aggressive-behavior mechanism..is not r i g i d , and stereotyped; rather i t i s within l i m i t s adaptable to the environmental s i t u a t i o n . In f i s h e s t h i s a d a p t a b i l i t y would, appear to a r i s e from a p l a s t i c expression of. the genetic tendency to aggressive behavior in. d i f f e r e n t s i t u a t i o n s . Although learning can not be disregarded i t i s probably of minor importance. Aggressive behavior expressed as s o c i a l hierarchy or t e r r i t o r i a l i t y i s a competitive mechanism which p o t e n t i a l l y provides a competitive advantage to -93-the dominant, by augmenting access to the food supply. I t i s by i t s nature a mechanism which would be expected to r e s u l t i n an unequal d i s t r i b u t i o n of the food resource among-the population. Even though, social, dominance provides a p o t e n t i a l advantage, the extent to which food i s shared depends upon, ( i ) s p a t i a l d i s t r i b u t i o n of. the. food, ( i i ) visual, i s o l a t i o n provided by the habitat, ( i i i ) population density, ( i v ) size of environment, (v) abundance of food, ( v i ) si z e of the-food p a r t i c l e s . The f l u c t u a t i o n of these environ-mental f a c t o r s can a l t e r the competitive, advantage- derived from s o c i a l dominance from complete, sharing to complete hoarding, of the. food. The d i s -t i n c t i o n , made by Nicholson. (1954) and B i r c h (1957)» that "contest" occurs for-resources, which ean; not be consumed and that "scramble" occurs f o r resources which are consumed seems: unwarranted. Competition, f o r food among medaka: can. lead e i t h e r to a- "contest" or to &. "scramble:." Which ..one occurs depends upon environmental f a c t o r s . Competition i n the form of "scramble" would be expected to r e s u l t i n . large f l u c t u a t i o n s .in population size- (Nicholson, 195^), whereas i n the form of "contest" would be expected to r e s u l t i n small fluctuations, i n population s i z e . I f aggression or t e r r i t o r i a l i t y i s to dampen f l u c t u a t i o n s in..popula-tion., s i z e i t should prevent sharing of food when food i s abundant and when population size i s large. In. medaka as. the abundance of food increases-or as population density increases, aggressiveness and. t e r r i t o r i a l i t y are l e s s e f f i c i e n t i n reserving food f o r the dominant. The "contest" occurring at low population densities- or at low food abundance, changes to "scramble" at high d e n s i t i e s : o r high food abundance. The aggressive behavior mechanism is. too f l e x i b l e to s t a b i l i z e fluctuations, i n medaka populations, and natural popula-tions of medaka would be expected to demonstrate large f l u c t u a t i o n s associated with any f l u c t u a t i o n s ' i n food abundance. Aggressive behavior i n -94-juvenile medaka. would be expected to help, maintain, a portion of the popula-t i o n i n periods, o f low food abundance., but would not check.rapid population growth under more-favorable food conditions. Territories-may be more r i g i d , in. other, species. For example, F o r s e l i u s (1957) observed that Anabantid f i s h e s would not set up reproductive t e r r i -t o r i e s at high d e n s i t i e s , and M i l l e r (1958) observed that t r o u t introduced i n t o an unfamiliar section of stream inhabited by other, t r o u t were displaced downstream; by the- t e r r i t o r i a l residents* Reproductive t e r r i t o r i e s of birds are probably more stable and f i x e d i n size than feeding t e r r i t o r i e s : of medaka and may l i m i t the number of reproducing p a i r s i n a.given area. (Hinde, 1957; Lack, 195^; MacArthur, 1958)' However, the aggressive behavior mechanism i n juvenile medaka and perhaps i n juvenile ayu (Kawanabe, 195$) i s too f l e x i b l e to l i m i t density. In general., aggressive .behavior i s a competitive mechanism .in immature medaka which can provide the; dominant animal with a competitive food advan-tage when food i s . l i m i t e d i n supply. The mechanism, i s more e f f e c t i v e i n reserving food f o r the dominant i f the food i s s p a t i a l l y l o c a l i z e d ; the number of challengers i s low, and the environment, i s small. I f food is-evenly d i s -t r i b u t e d in. space, increasing..the v i s u a l i s o l a t i o n i n the environment reduces the e f f e c t of the. mechanism;, but i f food i s contagiously distributed,, increasing, the v i s u a l i s o l a t i o n i n the environment increases the e f f e c t of the mechanism. Aggressive behavior w i l l disperse- the competitors throughout the h a b i t a t only i f food, i s found in. a l l areas. The phenomena described i n the present study would be expected to occur among f i s h which exhibit, aggres-sive behavior i n connection-with, food, i n habitats, containing contagiously d i s t r i b u t e d food l i m i t e d i n supply, and among, fishes-which l i v e near the substrate or among aquatic vegetation. SUMMARY OF RESULTS 1. Growth depensation occurs among medaka. sibs grown i n i s o l a t i o n under the same environmental conditions. 2 . Medaka raised i n smaller containers or at higher population de n s i t i e s do not grow slower than medaka with more space per f i s h , provided that other f a c t o r s u s u a l l y associated with space are supplied i n excess (food) or ^ eliminated (conditioning, of environment). J . Growth depensation i s no greater i n populations at de n s i t i e s up to 16 f i s h per l i t e r than i t i s among, sibs raised i n i s o l a t i o n under, the same conditions (food i n excess, conditioning of environment removed). 4. Aggressive behavior i s not greatest at the highest population.densities, but seems to be highest at an intermediate density. 5' I f food i s l i m i t e d i n supply, a. s o c i a l hierarchy develops i n which larger medaka are dominant and grow f a s t e r than subordinates; the dominant has no advantage i f no food or excess, food i s supplied. 6. Aggressiveness i s i n i t i a t e d i n juvenile medaka by the i n t e r n a l state of "hunger" and the presence of food s t i m u l i and. smaller.medaka; frequency of aggressive actions i s highest just a f t e r l i m i t e d food i s placed i n the environment, i s intermediate i f the environment contains no food, and lowest i f the environment contains excess food. 7. When l i m i t e d food i s s p a t i a l l y l o c a l i z e d , the dominant defends the food area as a t e r r i t o r y and the growth advantage of s o c i a l dominance i s higher than i f food, i s evenly d i s t r i b u t e d . 8. I f an excess amount of food i s lo c a l i z e d , the f i s h share i t with no d i f f e r e n t i a l access. -96-9. In a large population the dominant can not defend the concentrations, of l i m i t e d food from a l l subordinates,, and both the.growth advantage, of s o c i a l dominance and the frequency of aggression, by the dominant f i s h decrease. 10. As the amount of environmental i s o l a t i o n between competitors i s decreased, aggressive interactions, between them, increase, general growth rate of the. population .decreases even though the, same .amount of food i s provided, and the;, competitive advantage of s o c i a l dominance, increases. 11. I f there is.one semi-isolated subsection i n the. environment f o r each f i s h and l i m i t e d food i s evenly d i s t r i b u t e d , both the dominant and subordinate-grow equally w e l l . 12. Aggressive behavior disperses medaka throughout the environment i f food i s evenly d i s t r i b u t e d , but does not r e s u l t i n a dispersed d i s t r i b u t i o n of medaka i f food i s contagiously d i s t r i b u t e d . 13. Increasing the visual, i s o l a t i o n i n the environment increases the competi-t i v e advantage to the dominant i f food i s contagiously d i s t r i b u t e d , but decreases the advantage, i f food i s evenly d i s t r i b u t e d . 14. I f the, population: i s very small and food i s evenly d i s t r i b u t e d , the advantage of s o c i a l dominance increases i f the s i z e - o f the environment i s decreased. 15. I f a l l . f i s h have, equal, access to a l i m i t e d food supply, the rate at which.they can eat i s important i n determining t h e i r growth rates; small medaka can not eat p e l l e t s as r a p i d l y as can large medaka. 16. In medaka aggressive behavior i s evidently not a mechanism used i n competition f o r space, per se, but i s a mechanism;which reserves a greater, portion of a l i m i t e d food supply f o r the dominant under c e r t a i n environmental conditions. -97-Oompetition f o r food mediated: by aggressive behavior can. be-altered from "scramble" to "contest" and-all gradations, between by changing, environ-mental f a c t o r s such as. ( i ) amount, of food, ( i i ) si z e of food p a r t i c l e , ( i i i ) s p a t i a l d i s t r i b u t i o n . o f food, ( i v ) v i s u a l i s o l a t i o n i n the environment, (v) population- s i z e , and ( v i ) population, density. LITERATURE CITED A l l e e , W. C , Alfred.E.. Emerson, Orlando Park, Thomas Park; and K a r l P. Schmidtv 1949* P r i n c i p l e s of animal, ecology. W. Saunders Co., P h i l a d e l p h i a . 837 p. Allee,. W-. C , Bernard Greenberg, G. M. Rosenthal, and Peter Frank* 1948. Some: e f f e c t s of s o c i a l organization on growth i n the green sunfish, Lepomis cyanellus. J . E x p t l . Zool,. 108:1 —19« Baerends, G. P., and J . M. Baerends-van Roon. 1950« An in t r o d u c t i o n to the study of the ethology of C i c h l i d f i s h e s . Behaviour, Suppl. 1, 243 p. Berwein, Maria. 1941. Beobachtungen.und. Versuche uber das g e s e l l i g e Leben von E l r i t z e n . Z. Vergl... Physiol.. .28:402-420. Birch, L. G. 1957. The meanings, of competition. Am. N a t u r a l i s t 91:5-18' Boeseman, M. J . , J . van der D r i f t ; J . M. van Roon, N. Tinbergen, and J . J . t e r Pelkwi.jk. 1958; De bittervoorns en hun mossels. De Levende Natuur 43:129-136. (not seen) Brockwa-y, Donald R. 1950« Metabolic products and t h e i r e f f e c t s . Progr. F i s h - C u l t u r i s t 12:127-129. Brown, M. E. 1957. Experimental studies on growth, p. 561-400. In Margaret E. Brown,, t e d . ] , The, physiology of f i s h e s , v o l . 1, Academic. Press Inc., New York. Brown,. Margaret E. . 1946. The growth of brown trout (Salmo t r u t t a Linn.) I . Factors influencing, the growth of trout f r y . J . E x p t l . B i o l . 22:118-129. Brown, Margaret E. 1951• The-growth of brown t r o u t (Salmo t r u t t a Linn.) IV. The e f f e c t of food and temperature on the s u r v i v a l and growth of f r y . J . E x p t l . B i o l . 28:473-491. Briggs, John C , and Nobuo Egami. 1959> The medaka. (Oryzias l a t i p e s ) . A commentary and a bibliography. J . F i s h . Res. Bd; Canada .16: 363-380. Carpenter, C. R. 1958. T e r r i t o r i a l i t y : A review of concepts and problems, p. 224-250. In Anne Roe and George Gaylord Simpson, Led.], Yale Univ. Press, New Haven, Conn. C o l l i a s , Nicholas E. 1944. Aggressive behavior among vertebrate animals. Physiol.. Zool. 17:83-125. - 9 9 -Diebschlag, E. 1941. Psychologische Beobachtungen fiber die Rangordnung bei der Haustaube* Z. Tierpsychol. 4:173-188. Dixon, V / i l f r i d J . , and Frank J . Massey, Jr.. 1957. Introduction to, s t a t i s t i c a l a n a l y s i s . 2nd ed. McGraw-Hill Book Go. Inc., New York. 488 p. Duncan, David B. 1955* M u l t i p l e range and multiple F t e s t s . Biometrics 11: 1-42. F a b r i c i u s , E r i c . 1951. The topography of the spawning bottom: as. a f a c t o r influencing, the s i z e of the t e r r i t o r y i n some species of f i s h . . Rept. Inst. Freshwater. Res., Drottningholm 3 2 : 4 3 - 4 9 . F o r s e l i u s , Sten. 1957. Studies of Anabantid f i s h e s ; I. A q u a l i t a t i v e d e s c r i p t i o n of the reproductive behaviour i n t e r r i t o r i a l species investigated under laboratory conditions with s p e c i a l regard to the genus O o l i s a . An i n t r o d u c t i o n . Zool. Bidrae Uppsala, 32: 93-301 . Greenberg, Bernard. 1947- Some r e l a t i o n s between t e r r i t o r y , s o c i a l hierarchy, and leadership, in. the green, sunfish (Lepomis cyanellus). P h y s i o l . Zool. 20:267-299. Hartman, G; F. 1958; Mouth siz e and food size i n young rainbow trout, Salmo g a i r d n e r i . Copeia 1958:233-234. Hinde, R. A. 1956. The b i o l o g i c a l , s i g n i f i c a n c e of. the t e r r i t o r i e s of b i r d s . I b i s 9 8 : 3 4 0 - 3 6 9 . Hoar, William.S. 1954. The-behaviour of juvenile P a c i f i c , salmon j with p a r t i c u l a r reference to.the sockeye (Oncorhynchus-nerka). J . F i s h . Res. Bd. Canada 11:69-97. H o l l i n g , C. S. 1959. Some c h a r a c t e r i s t i c s of simple types of predation, and parasitism.. Can. Entomol. 91J385-398. Hubbs, Carl L., and Gerald P. Cooper. 1935- Age-and growth of the long-eared and green sunfishes i n Michigan. Papers-Mich. Acad. S c i . 20: 669-696. Kalleberg, Harry. 1958. Observations, i n a stream tank of t e r r i t o r i a l i t y and. competition in. juvenile salmon, and.trout (Salmo.salar L. and S. t r u t t a L . ) . Rept. Inst. Freshwater Res., Drottningholm 3 9 : 5 5 - 9 8 . Kawabata, Masakazu. 1954. Socio-ecological studies, on•• the k i l l i - f i s h , Aplocheilas l a t i p e s . I. General, remarks on the s o c i a l behavior. CIn Japanese, English summary]; Jap. J . E c o l . 4:109-113. Kawabata, Masakazu. 1960. E f f e c t of environment,, p. 61-64. In K i n j i Imanishi, Ced."], Animal society and i n d i v i d u a l s , Iwanami Press, Tokyo. C In Japanese.1 -100-Kawamoto, N. Y. 1961. The influence of excretory substances of f i s h e s on t h e i r own. growth. Progr. P i s h - O u l t u r i s t 25:70-75. Kawanabe, Hiroya. 195$. On the s i g n i f i c a n c e of the social, structure f o r the mode of density e f f e c t i n a salmon-like f i s h , "Ayu", Plecoglossus  a l t i v e l i s Temminck..et Schlegel. Memi C o l l . S c i . , Univ. Kyoto, Ser.B, 25 (Biol.)s .171-I80. Kramer, Clyde Young. 195°". Extension of multiple range t e s t s to group means with unequal numbers of r e p l i c a t i o n s . Biometrics 12:307-510. Lack, David. 1954. The natural regulation of animal numbers. Clarendon Press, Oxford. 545 p. Landau, H. G. 1951• On dominance, r e l a t i o n s and the structure of animal s o c i e t i e s : I I . Some e f f e c t s of possible s o c i a l f a c t o r s . B u l l . Math. Biophys. 13:245-262. Langlois, T. H. 1936. A study of. the small-mouth bass, Micropterus dolomieu (Lacepede) i n rearing ponds, i n Ohio. Ohio B i o l . Surv. 6:191-225. Larkin, P. A. 1956. I n t e r s p e c i f i c competition and population control i n freshwater f i s h . J . F i s h . Res. Bd. Canada 13:327-542. Larkin, P. A., G. C. Anderson; W. A. Clemens., and D. C. G. Mackay. 1950. The production of Kamloops•trout (Salmo g a i r d n e r i i kamloops, Jordan) i n Paul Lake, British.Columbia.' S c i . Publ. B.. C. Game Dept., No. 1, 37 p. LeCren, E. D. 1951. The. length-weight relationship, and seasonal cycle i n • gonad.weight and condition i n the perch (Perca f l u v i a t i l i s ) . J . Animal E c o l . 20:201-219. Lindroth, Arne. 1955* D i s t r i b u t i o n , t e r r i t o r i a l behaviour and movements of sea trout f r y i n the r i v e r Indalsfilven. Rept. Inst. Freshwater Res., Drottningholm 56:104-119. MacArthur, Robert H. 1958. Population ecology of some-warblers of northeastern coniferous forests.. Ecology 39:599-619. M i l l e r , Richard B. 1958. The ro l e of competition in. the m o r t a l i t y of hatchery t r o u t . J . F i s h . Res. Bd. Canada. 15:27-45. Morris, Desmond. 1958. The reproductive, behaviour of the ten-spined stickleback (Pygosteus pungitius L . ) . Behaviour, Suppl. 6, 154 p. Nakamura, Nakaroku, and Shogor-o Kasahara. 1956. A study on the phenomenon of the Tobi-Koi or shoot carp - I I . On the e f f e c t of p a r t i c l e "size and quantity of. the food. C In. Japanese, English summary]; B u l l . Jap. Soc. S c i . F i s h . 21:1022-1024. - 1 0 1 -Nakamur.a, Nakaroku, and Shogoro Kasahara. 1957- A study on the phenomenon of the Tobi-Koi or shoot carp - III.. On the r e s u l t of culturxng the modal group and the. growth of carp f r y reared i n d i v i d u a l l y . C In Japanese, English summary}) B u l l . Jap. Soc. S c i . Fish.. 2 2 : 6 7 4 - 6 7 8 . Newman, H. William. 1959- Observations on aggressiveness i n sockeye. salmon smolts. Copeia 19 5 9 * 7 6 . Newman, Murray A. 1 9 5 6 . S o c i a l behavior and i n t e r s p e c i f i c competition i n two trout species. P h y s i o l . Zool. 29*64-81. Newmanj Murray Arthur. 1 9 6 0 . A comparative study of the resi d e n t i a l , behavior of juvenile salmonids.. Ph.D. Thesis. Univ. B r i t i s h Columbia. 2 9 5 P. Nicholson, A.. J . 195^- An out l i n e of the dynamics of animal populations. Austral.. J . Zool. 2 : 9 - 6 5 . Noble, G. K. 1 9 3 9 . The experimental animal from the n a t u r a l i s t ' s point of view. Am. N a t u r a l i s t 7 5 ' 1 1 5 - 1 2 6 . Noble,^Gv K., and Ray Borne. 1 9 5 8 . The s o c i a l hierarchy in. Xiphophorus and other fishes•. B u l l . E c o l . Soc. Am. 1-9(2):14. Parker, Robert R., and Peter A. Larkin. 1959- A concept of. growth i n f i s h e s . J . F i s h . Res. Bd. Canada 1 6 : 7 2 1 - 7 4 5 . Richards, C h r i s t i n a MacGregor. 1 9 5 8 . The i n h i b i t i o n , of growth, i n crowded Rana pipiens tadpoles. P h y s i o l . Zool; 3 1 * 1 5 8 - 1 5 1 . Ricker, W. E. 1 9 5 8 . Handbook of computations, f o r biological,, s t a t i s t i c s of f i s h populations. F i s h . Res. Bd. Canada B u l l . 119, 500 p. Rose; S. Meryl.. 1959^ Failure-of. survival, of slowly growing.,members of a population. Science 1 2 9 : 1 0 2 6 . Rose, S; Meryl. 1 9 6 0 . A feedback mechanism of growth control i n tadpoles. Ecology 41:188-199. Snedecor, George W. 1 9 5 6 . S t a t i s t i c a l methods applied to experiments^ i n agriculture- and biology. 5"th ed.* Iowa State C o l l . Press, Ames. 5 5 4 p. Stringer., G. E., and W. S. Hoar. 1955* Aggressive behavior, of underyearling Kamloops t r o u t . Can. J . Zool. 33*148 -160. Swingle., H. S., and E. V. Smith. 1 9 4 3 . Factors affecting, the reproduction of b l u e g i l l . bream, and. largemouth black bass i n ponds. Agr... E x p t l . Sta. A l a . Polytech. Inst. C i r c . 8 7 , 8 p. Tinbergen, N. 1951- The study of i n s t i n c t . Clarendon. Press, Oxford. 228 p. -102-Tinbergen, N. 1957. The functions of t e r r i t o r y . B i r d Study 4:14-27. West, Lorna B. I960. The nature of growth, i n h i b i t o r y material from crowded Rana pipiens tadpoles. P h y s i o l . Zool. 33*252-239. Winn> Howard E. 1958. Comparative reproductive behavior and. ecology of fourteen species of darters-(Pisces-Percidae). E c o l . Monographs 28:155-191. APPENDIX In experiment I I r e l a t i v e c o n d i t i o n .was expressed as the difference between, weight of a f i s h , and weight of an i s o l a t e control f i s h of the same length (W-W)* This measure of condition was 2.45 times-more va r i a b l e f o r large f i s h (pooled s 2 x = 3.985, d.f. = 183) than i t was f o r small f i s h (pooled s 2 x = 1.627, d.f. = 179). I f t h i s i s true among: i n d i v i d u a l f i s h , i t i s possible that i t would also, be- true f o r the: mean .effects of various, t r e a t -ments; the expectation, i s that large f i s h , i n thei same environmental s i t u a t i o n as small f i s h would have a. greater mean, deviation from, the controls. I d e a l l y , the variance^ i n the condition values should be homogeneous f o r comparisons: of large f i s h with small, f i s h . A logarithmic transformation (log.jQ w - log-jQW) was. used i n an attempt to achieve homogeneity. Yet i n t h i s case the measure of condition was 2.58 times more var i a b l e f o r small f i s h (pooled s 2 x = 0.0029071, d.f. => 161) than f o r large f i s h (pooled s 2 x = 0.0011247, d.f. = 165). The difference between the variances was just as large f o r the (W-W) and ( l o g ^ f - log-jQW) values, but was i n the reverse d i r e c t i o n . Since neither- method resulted i n homogeneous variance another c r i t e r i o n was. used f o r s e l e c t i o n . The-large and small controls (CL1, 0S1) by d e f i n i -t i o n have the same•value of condition (0.00) which would be expected since t -. • . . . they were raised i n i s o l a t i o n and no growth differences due to competition would occur. Likewise both the large and small, f i s h r a i s e d i n i s o l a t i o n i n 4 - l i t e r baskets, (0L4, 0S4) should have the same condition, when measured, as deviations from the 0L1 and 0S1 controls since they also, had no competitive i n t e r a c t i o n s . In the transformed data, difference between-the i s o l a t e s i n -104-4 - l i t e r containers (CL4, 0S4) was 0.0215 or 34% of the, greatest mean d i f -ference (0.0634) between-large and small f i s h i n a treatment observed i n L1. On the other hand, i n the untransformed data the difference between large and small, i s o l a t e s i n 4 - l i t e r baskets was 0.07 o r 4% of the greatest mean difference (1.71) between large and. small f i s h , i n a treatment observed i n L4. These comparisons indicate, that' the log transformations were not as v a l i d as the untransformed. data, f o r comparing large, and. small, f i s h which were i n the same environmental s i t u a t i o n . In general, the log transformation resu l t e d i n a greater difference.between the small f i s h i n a treatment and the i s o l a t e controls of the same length than did the.untransformed data. This occurred-both when the, small f i s h i n a treatment had a p o s i t i v e condition (CS4) and, when the- small f i s h had a negative condition (NF and L1). Untransformed data (W-W) were chosen f o r presentation p r i m a r i l y because large and/small f i s h grown: i n i s o l a t i o n , i n 4 - l i t e r baskets had more s i m i l a r values of. (W-W) than they did, values of (log 1 QW - logTtw). The mean values of (W-W) and (log 1 QW - f o r each treatment are presented i n the appendix t a b l e . -105-Appendix Table. Mean values of r e l a t i v e condition f o r large and small f i s h i n each treatment, except XF, expressed as (W-W) and (log 1 0W - log 1 0W). Treatment Gode F i s h Size (W-W) (log 1 0W - log 1 0W) CL1 Large 0.00 0.0000 0S1 Small 0.00 0.0000 Difference 0.00 0.0000 L1 Large +0.95 +0.0118 Small -0.70 -0.0516 Difference +1.65 +0.0634 NF Large -0.55 -0.0154 Small -0 .79 -0.0436 Difference +0.26 -0 .0282 LB Large +0.08 +0.0023 Small +0.35 +0.0067 Difference -0.27 -0.0044 LP1 Large +1.15 +0.0200 Small -0.52 -0.0173 Difference +1.67 +0.0373 LP2 Large -0 .05 +0.0042 Small +0.46 +0.0174 Difference -0 .49 -0.0132 LSH Large -0.08 -0.0016 Small -0 .86 -0.0276 Difference +0.78 -0 .0260 CL4 Large +2.52 +0.0300 CS4 Small +2.59 +0.0515 Difference -0 .07 -0.0215 L4 Large +1.86 +0.0222 Small +0.15 +0.0001 Difference +1.71 +0.0221 C L I C S I L I X F IM F T O P S I D E L B L PI L P 2 L S H T O P S I D E L A R G E F I S H S M A L L F I S H F O O D W A T E R L E V E L C L 4 C S 4 L 4 I. s i C O N D I T I O N R E L A T I V E T O C O N T R O L S + ( W . - W H n mg ro + OJ r i S AMONG LARGE AND SMALL FISH FOR I LITER VS. 4 LITER : 1 BASKETS 1 —| COMPARISONS AMONG LARGE AND SMALL FISH IN LITER- OR IN 4 LITER BASKETS COMPARISON AMONG MEANS FOR I LITER VS. 4 LITER BASKETS _| COMP. AMONG MEANS FOR I LITER BASKETS '3T e 1 0 . 2 1 ( L 8 ) 1 0.1 0 .3 1 (2^0) | 0 . 3 0 I E 8 C 4 H O L E IN P A R T I T I O N A C T I V I T Y / 2 . 5 m i n 2 5 _ A G G R E S S I V E A C T I 0 N S / 2 . 5 m i n N F 13 15 16 A C T I V I T Y / 2 . 5 m i n 2 5 A G G R E S I V E A C T I O N S / 2 . 5 m i n % T I M E IN B O T T O M D I S H 50 i i i i i i i i i 1 1 f >§L N 0 1 3 5 7 9 II 13 15 1 L B F E D H O U R S A F T E R L I G H T S W E N T O N I if A C T I V I T Y / 2 . 5 m i n 10 2 0 ACTIV I T Y / 2 . 5 min 10 -A G G R E S S I V E A C T I O N S / 2 . 5 m i n 0 1 3 5 7 F E D 1 1 1 1 I I 1 1 LI 13 15 16 A C T I V I T Y / 2 . 5 min 2 0 10 A G G R E S S I V E A C T I 0 N S / 2 . 5 m i n - v ^*^~-~ r i i i i 1 I 1 o rv - -inn r - - • o XF 0 1 3 5 7 9 II F E D F E D F E D H O U R S A F T E R L I G H T S W E N T O N 13 15 16 7 * feus A C T I V I T Y / 2 . 5 m i n 2 5 A G G R E S S I V E ACTIONS/2 .5mfn 0 I 5 7 F E D -- o 1 1 1 1 1 1 1 1 L S H II 13 15 16 A C T I V I T Y / 2 . 5 mm A G G R E S S I V E ACT I O N S / 2 . 5 mfn L 4 II 13 15 16 A C T I V I T Y / 2 . 5 m i n 15 16 H O U R S A F T E R L I G H T S W E N T O N i F O O D P E L L E T H O L E I N P A R T I T I O N . I E 8 C 4 _I_J 1 4 E 8 C 2 .'•u'VU'.i.'V:-•;\\.-;. .• i l f t f * • • • -8 E 8 C I I A C T I V I T Y / 2 . 5 m i n 2 5 -A G G R E S S I V E ACTIONS/2 .5 min 5 -0 1 0 0 % TIME ON FOOD SIDE OF PARTITION' .50 0 I LPI 15 16 A C T I V I T Y / 2 . 5 min 2 5 A G G R E S S I V E ACTIONS/2 .5min 5 o 1 0 0 % TIME O N RIGHT SIDE OF PARTITION 5 0 J j _ 11 i 1 _ _ - » t 0 0 o 1 1 1 1 1 1 1 1 P - - - - O . y — ~ ~ ° ^ r ^ : - — n — ^ 1 1 1 r ^ _ - o - — 1 1 1 1 - LP2 o I 13 15 16 3 5 7 9 II F E D H O U R S A F T E R L I G H T S W E N T O N \ ^ O N N m > m — i o o o m AGGRESSIVENESS o b ( Agg. Act. / 2.5 min ) b r o-- o - 0-ro x _ T T T COMPARISONS AMONG LARGE AND SMALL FISH IN ALL TREATMENTS r ~ i r r o - o-o f . 9 "~l i . i _ _ • COMP. AMONG MEANS FOR EACH TREATMENT ( L A R G E + SMALL| I : I 0~ GO o o r to CD O O co r CO X "D ro x -r ACTIV ITY (COUNTS/2.5 min ) IV) CD "T" COMPARISON AMONG LARGE AND SMALL FISH IN ALL TREATMENTS r I I . ~ i i i i i : i , LARGE+ SMALL . COMPARISON AMONG MEANS FOR EACH TREATMENT ( 5 ;— J - L _L ro CT) ro oo S I Z E - S P E C I F I C G R O W T H R E L A T I V E TO CONTROLS ( A W - A W)4n ; I I I + ro — o — oo t~ in m > oo —\ r to m ro o -m [~ CD X + ro T r C O M P A R I S O N S AMONG L A R G E A N D SM 1 r r i i A L L F I SH IN A L L T R E A T M E N T S n r ~i i i i C O M P A R I S O N S AMONG M E A N S FOR '1 r i / L A R G E + SMALL\ E'AC H i T R E A T M E N T I 1 

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.831.1-0105779/manifest

Comment

Related Items