THE GENETIC AND ENVIRONMENTAL BASIS FOR EXTERNAL COLOURATION IN LAKE WHITEFISH (COREGONUS CLUPEAFORMIS (MITCHILL)) FROM SOUTHERN INDIAN LAKE, MANITOBA by C h r i s t i n e Helga Johnston A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n FACULTY OF GRADUATE STUDIES Zoology We accept t h i s t h e s i s as conforming to the r e q u i r e d standard ( THE UNIVERSITY OF BRITISH COLOMBIA June, 1984 c. C h r i s t i n e Helga Johnston, 1984 In p r e s e n t i n g 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 a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a 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 r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f 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 o f my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s understood t h a t copying o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department of ~2L00U06.S The U n i v e r s i t y o f B r i t i s h Columbia 2075 Wesbrook P l a c e Vancouver, Canada V6T 1W5 Date \ /XOg.CrSV ^€>A. DE-6 (2/79) i i ABSTRACT The purpose of this study was to investigate the nature of the colour difference between light and dark-coloured whitefish from Southern Indian Lake, Manitoba and to examine related differences between them. Subjective assessment of whitefish colour is corroborated by quantifiable differences in melanophore numbers between the two forms. Spatial distribution of l ights and darks differs both between and within regions. Lights occur throughout the lake, are sl ightly more abundant offshore than onshore and are primarily benthic in habit. Darks are restricted to certain areas of the lake, are more numerous onshore than offshore and are somewhat more pelagic than l ights . The two forms showed signif icant differences in several morphometric characters and lower g i l l raker number, but not in hemoglobin or glycerol-3-phosphate dehydrogenase a l le le frequencies. Hatchery-reared offspring of dark parents had higher mean dorsal melanophore counts than offspring of light parents. The her i tabi l i ty estimate derived from the regression of offspring (age 111 days post - fer t i l i zat ion) on male parent melanophore count was 0.10. Short and long-term experiments showed that colour of larval whitefish is subject to environmental alteration in response to l ight conditions and background colour. Short-term change is effected through redistribution of melanin granules in the melanophores, long-term change through changes in numbers of melanophores. Level of infection with Triaenophorus crassus cysts i s , on average, higher in darks than in l ights . However, the mode for both forms is 0 indicating no direct causal connection between cyst count and pigmentation. Higher mean counts of darks may be related to diet and distr ibut ion. Morphometric and meristic characters are subject to. i i i environmental modification, as shown by other studies. This plus the lack of difference between lights and darks in biochemical characteristics suggests no clear-cut separation between the two into non-interbreeding stocks. Whitefish colour is correlated with water colour and c lar i ty and may be an adaptation for concealment. iy TABLE OF CONTENTS Page ABSTRACT ii LIST OF TABLES . v vi i i LIST OF FIGURES... . . . . . x ACKNOWLEDGEMENTS xi i i INTRODUCTION . . . . . 1 STUDY AREA. . . . . . . . 4 COLLECTION OF SPECIMENS. 9 COLOUR CLASSIFICATION OF ADULT WHITEFISH Introduction 14 Methods Subjective Method : 14 Quantitative Method 15 Comparison of Subjective and Quantitative Methods. 15 Results Scale Melanophore Counts of Study Specimens 16 Var iabi l i ty in Colour C lass i f icat ion Between Observers... 16 Misclassif icat ion of Fish 26 Conclusion 26 SPATIAL DISTRIBUTION OF COLOUR CLASSES OF WHITEFISH IN SIL Introduction 29 Methods 29 Results Lake-wide Distribution Patterns of Lights and Darks 33 Areal Distribution of Lights and Darks in Region 5 33 Vertical Distribution of Lights and Darks in Region 5 33 V Temporal Changes in Distribution of Lights and Darks in Region 5 40 Colour of Young-of-the-Year Whitefish from Different Regions 40 Conclusion 46 MORPHOLOGY AND BIOCHEMISTRY OF ADULT LIGHT AND DARK WHITEFISH Introduction 47 Methods ' Morphological Analyses 47 Age and Growth 50 Biochemical Analyses Glycerol-3-Phosphate Dehydrogenase (G-3-PDH) Electrophoresis 50 Hemoglobin Electrophoresis 52 Results Morphology Differences in Morphological Characters Between Lights and Darks 53 G i l l Raker Number 53 Age and Growth 53 Biochemical Genetics Glycerol-3-Phosphate Dehydrogenase 59 Hemoglobin 59 Conclusion 59 INHERITANCE OF COLOUR v i i Methods Rearing of Fish 62 Sampling Procedure 67 Results Mean Dorsal Melanophore Counts from Different Cross Types 70 Her i tabi l i ty Estimates 70 Conclusion 78 ENVIRONMENTAL ALTERATION OF COLOUR Introduction 79 Long-term Colour Change Experiment Methods Experimental Design 79 Stat is t ica l Analysis of Data 80 Results 81 Conclusion 85 Short-term Colour Change Experiment Methods Experimental Design 85 Observations 86 Results Length of Time to Adjust Colour 86 Maximum Colour Change Achieved 86 Conclusion 87 Response of Young Whitefish to Zero Incident Light Observations 87 Conclusion 87 Short-term Colour Change in Adult Whitefish 88 vi i TRIAENOPHORUS CRASSUS CYST LEVELS OF SIL WHITEFISH Introduction 89 Methods 89 Results Regional Differences in Mean Cyst Counts 91 Differences in Mean Cyst Counts Between Light and Dark Whitefish 91 Regional Differences in Mean Cyst Count Among Light Whitefish. 96 Conclusion 96 DISCUSSION Colour Classi f icat ion of Adult Whitefish 98 Distribution of Colour Classes of Whitefish in SIL 100 Morphology and Biochemistry of Adult and Dark Whitefish 102 Inheritance of Colour 104 Environmental Alteration of Colour 108 Triaenophorus crassus Cyst Levels of SIL Whitefish 115 Functionnal Significance and Adaptive Value of Colour in Fish 119 CONCLUSION 121 LITERATURE CITED 123 v i i i LIST OF TABLES Table Page 1 Scale melanophore counts of l ight , intermediate and 25 dark lake whitefish colour c lass i f ied by 4 different groups of observers. 2 Numbers of light and dark lake whitefish misclassif ied 27 by different groups of observers. 3 Analysis of covariance on head and dorsal melanophore 45 counts of young-of-the-year lake whitefish from regions 4, 5 and 6, Southern Indian Lake. 4 Morphometric measurements and meristic counts made 48 on light and dark lake whitefish from region 5, Southern Indian Lake. * : measurements and counts selected for detailed examination in 1982 sample. 5 Age distr ibutions, numbers of males and females and 49 dates of capture of light and dark lake whitefish used for morphological analyses. 6 Analysis of covariance of morphometric characters of 54 light and dark lake whitefish from region 5, Southern Indian Lake. 7 Observed and expected (Castle-Hardy-Weinberg) BB 60 glycerol-3-phosphate dehydrogenase phenotype distributions for l ight and dark lake whitefish from region 5, Southern Indian Lake. 8 Observed and expected (Castle-Hardy-Weinberg) hemoglobin 61 phenotype distributions for l ight and dark lake white-f ish from region 5, Southern Indian Lake. 9 Ranges and means of scale melanophore counts of lake 63 whitefish used as parents in breeding experiment. 10 Mean dorsal melanophore counts and variances for 111 71 day larval lake whitefish of each cross type. 11 Results of t - tests (using the formula for unequal 72 variance) between mean dorsal melanophore counts of 111 day larval lake whitefish or different cross types. 12 Hen'tabi l i ty estimates from 4 pre-hatch lake whitefish 77 samples. 13 Long-term colour change experiment: Results of analysis 82 of covariance of regressions of larval lake whitefish melanophore counts on length at 12 days. 14 Mesh sizes of gi l lnets used by different authors for 90 col lect ing lake whitefish in Southern Indian Lake. 15 Triaenophorus crassus cyst counts of lake whitefish 92 from different regions of Southern Indian Lake. 16 Mean Triaenophorus crassus cyst counts of light and 93 dark lake whitefish in experimental catches, 1963, 1978, 1979, 1982, Southern Indian Lake. 17 Mean Triaenophorus crassus cyst counts of l ight lake 97 whitefish from regions 4, 5 and 6, Southern Indian Lake, Bodaly et a l . 1983). 18 Approximate values of the her i tabi l i ty of various 106 characters in some mammals, birds and f ishes. 19 Triaenophorus crassus cyst counts in experimental 118 catches of lake whitefish, regions 4 and 5, Southern Indian Lake, 1978-1983. X LIST OF FIGURES Figure Page 1. Map of Southern Indian Lake, Manitoba, showing 6 limnological regions, relationship to the Churchill River diversion and adjacent water bodies (Bodaly et a l . 1984). 2. A. Map of region 5, Southern Indian Lake, B. Sites 11 sampled in July 1982. 3. Map of region 5, Southern Indian Lake showing sites 13 sampled in September-October 1982. 4. Scale melanophore counts of lake whitefish from region 5 18 Southern Indian Lake by colour class as subjectively c lass i f ied by author. A. Lights, B. Intermediates, C. darks. Arrows show modes, triangles show means 5. Scale melanophore counts of total sample of lake white- 20 f ish from region 5, Southern Indian Lake. 6. Plot of scale melanophore count versus age for total 22 sample of lake whitefish from region 5, Southern Indian Lake 7. Plot of scale melanophore count versus fork length for 24 total sample of lake whitefish from region 5, Southern Indian Lake. 8. Dorsal view of lake whitefish larva showing head and 32 dorsal melanophore count areas (for counts on wild-caught young-of-the-year lake whitefish from Southern Indian Lake) 9. Proportion of l ight and dark lake whitefish in 35 experimental and commercial catches, Southern Indian x i Lake, 1963-1982. A. Sites sampled by Sunde (1963), B. Commercial catches, regions 4 and 5, 1979, 1980, 1981 (Bodaly et a l . 1984), C. Experimental catches, regions 4, 5 and 6, 1982 (Bodaly et a l . 1983). F ig . 10. Areal distr ibution of light and dark lake whitefish 39 in region 5 .Southern Indian Lake, 1982. A. Scale melanophore counts of lake whitefish in onshore sets, B. Scale melanophore counts of lake whitefish in offshore sets. F ig . 11. Vertical distr ibution of light and dark lake whitefish 42 in region 5, Southern Indian Lake, 1982. A. Scale melanophore counts of lake whitefish in top 1/3 of net, B. Scale melanophore counts of lake whitefish in middle 1/2 of net, C. Scale melanophore counts of lake whitefish in bottom 1/3 of net. F ig . 12. Scale melanophore counts of l ight and dark lake 44 whitefish captured in region 5, Southern Indian Lake, in July 1982. F ig . 13. G i l l raker count distributions of light and dark lake 56 whitefish from region 5, Southern Indian Lake. A. Upper g i l l rakers, B. Lower g i l l rakers, C. Total g i l l rakers. F ig . 14. Plots of age versus fork length for l ight and dark 58 lake whitefish from region 5, Southern Indian Lake. Open symbols represent means of samples with less than three f i sh . Closed symbols represent means of samples with three or more f i s h . Vertical bars show ranges. xrfij Fig . 15. Experimental design ut i l i zed in crosses of light and 65 dark lake whitefish. F ig . 16. Dorsal view of lake whitefish larva showing head and 69 dorsal mealanophore count areas (for counts on hatchery-reared larvae). F ig . 17. Regressions of dorsal melanophore counts of 111 day 74 lake whitefish larvae on parent scale melanophore counts. A. Offspring on mid-parent value, B. Offspring on female parent value, C. Offspring on male parent value. F ig . 18. Long-term colour change experiment: Regressions of 84 larval lake whitefish melanophore counts on total length before and after 12 day exposure to different light and background conditions. Group 1 - Pale tank, fu l l l ight . Group 2 - Dark tank, fu l l l ight . Group 3 - Dark tank, half l ight. Group 4 - Dark tank, low l ight . F ig . 19. Triaenophorus crassus cyst count distributions of 95 lake whitefish from region 5, Southern Indian Lake, 1982 (Bodaly et a l . 1983). A. Lights, B. Darks. xi i i ACKNOWLEDGMENTS I would l ike to thank my thesis supervisor Dr. C.C. Lindsey for his patience, support and stimulation. I am also indebted to Drs. R.A. Bodaly and J.W. Clayton for their help. My advisory committee, Drs. D. Brooks, J .D. McPhail and J . Myers provided helpful advice and comments on thesis and proposal drafts. My f ie ld crew humoured their inexperienced buss (most of the time) and worked very hard. For their help in this capacity I thank M. Dumas, C. Hrenchuk, R. Fudge and S. Witte. My venture into f ish culturing was ably guided by L. A l la rd , R. Fudge and D. Tretiak. Drs. G.B. Ayles and C. Wehrhahn gave suggestions concerning interpretation of the breeding experiment data. D. Hayward introduced me to HP-9836 and thus helped f a c i l i t a t e data analysis. Thanks to E.J . Hrenchuk who translated two publications from the German for me. I am very grateful to L. Wilson, D. Laroque and S. Ryland who cheerfully gave of their time and energy to put the manuscript through the word processor. I would particularly l ike to thank Dr. R.A. Bodaly, with whom the idea for the study originated, for his input and encouragement throughout. Thanks also to the SIL project staff and to Dr. J .H. Gee who helped me get started in the f i r s t place. My husband, Carl Hrenchuk, assisted in many capacities and was an unfai l ing and steady source of encouragement, enthusiasm, patience and humour. For a l l of that, I thank him. Financial support to the author was provided by a National Science and Engineering Research Council of Canada Graduate Scholarship and by x i y NSERC grants to Dr. C.C. Lindsey. F ield and laboratory work we supported by the Department of Fisheries and Oceans, Freshwater Institute, Winnipeg, Manitoba. - 1 -INTRODUCTION Colour variation is quite common within f ish species. This study concerns two dist inct ively coloured forms of lake whitefish (Coregonus clupeaformis (Mitchi11)) which occur in Southern Indian Lake, Manitoba. So-called " l ights" are si lvery overall and light green on the dorsal surface; so-cal led "darks" are darker overal l , especially above the lateral l ine , and dark brown to black dorsally. Darks are most abundant in the relatively shallow northern basin of the lake; where they occur in other basins they are confined to near-shore areas. Darks also tend to be more heavily infected with cysts of the cestode parasite Triaenophorus crassus than do l ights . The occurrence of the two forms is of economic as well as sc ien t i f i c interest. Southern Indian Lake (SIL), which is located in north-central Manitoba on the Churchill River, was subjected in 1974-1978 to flooding and major changes in flow regime as part of hydroelectric development of the Nelson River System. Before then, SIL supported northern Manitoba's largest commercial f ishery, and lake whitefish made up about 85% of the total commercial catch. Because dark whitefish have less commercial value than light whitefish (due to their less attractive colouration and higher crassus cyst levels) , fishermen tradit ional ly avoided those areas of the lake where darks were known to occur. However, since impoundment, catch per unit of effort on tradit ional f ishing grounds has declined to about one-half of pre-flooding levels (Bodaly et a l . 1984). In response, the geographic distr ibution of effort has shifted into areas formerly avoided, and consequently darks have made up an increasing proportion of the catch. In 1975, prior to impoundment, dark whitefish were not present in the - 2 -summer commercial catch; in 1981, they made up 81% of i t (Bodaly et a l . 1984). Dark lake whitefish have been noted in other lakes. Rawson (1947a) caught whitefish which he described as being darker coloured, longer and more compressed, and softer-f leshed than normal ones in the Gros Cap area of Great Slave Lake. They were relatively few in number and were confined to shallow near-shore areas. Rawson (1947b) also reported that a small proportion of the commercial lake whitefish catch from Lake Athabaska were dark, with small eyes, long shallow bodies, and elongate heads. The darks were of infer ior quality with softer f lesh , slower growth rates and heavier infection with T.. crassus cysts than typical lake whitefish. Most were caught in shallow water along the north shore of the lake (Rawson 1947b). Imhof (1977) referred to two lake whitefish colour variants in Lake Michigan - "green backs", which were commonly caught in open waters, and "brown backs", which were found in central and lower Green Bay. The basis of the colour difference between light and dark whitefish has not been investigated previously. The main questions addressed by this study were whether light and dark whitefish comprise separate subpopulations and whether the nature of the colour difference between them is genetic or environmental or both. The research tested the following six hypotheses, for which the methods and results are described separately: 1. There is no quantifiable difference in melanophore numbers to support the subjective c lass i f icat ion of SIL whitefish into "l ights" and "darks". - 3 -2. There is no difference in spatial distribution of lights and darks within the lake. 3. There are no morphological or biochemical differences between lights and darks. 4. There is no hereditary basis to colour differences between lights and darks. 5. Colour of larval whitefish is not subject to environmental al terat ion. 6. Colour differences are not correlated with infection by cysts of T. crassus. - 4 -STUDY AREA Southern Indian Lake (SIL) is located in north-central Manitoba (57°N; 99°W) on the Churchill River (Fig. 1). For experimental purposes the lake was divided into numbered regions following natural divisions into irregular basins (Fig. 1). In the Southern Indian Lake area the Precambrian Shield bedrock is overlain with glacial deposits. Sands and gravels cover the uplands surrounding region 5 and the northern half of region 4, varved si Ity clays cover the land surrounding the southern 2/3 of the lake. Permafrost is widespread except in the glaciof luvial deposits of regions 4 and 5. The Southern Indian Lake region l ies in the boreal forest zone. The climate is continental with short, cool summers and long, cold winters. The mean annual temperature at the settlement of Southern Indian Lake is -5°C with average monthly temperatures ranging from -26°C in January to +16°C in July. Mean annual precipitation is 430 mm, of which about 1/3 f a l l s as snow. The lake is ice- f ree for about 5 months from early June to late October (Newbury et a l . 1984). The lake has a surface area of 2391 km2 and a mean depth of 9.8 m. Mean depths of regions 4, 5 and 6 are 13.0 m, 5.9 m and 5.8 m, respectively (McCullough 1981). The water exchange rate for the whole lake is 0.72 year and for individual basins ranges from 0.031 (region 6) to 2.8 years (region 5). Exchange time depends on the volume of the basin and whether or not i t is in the direct flow of the Churchill River (Newbury et a l . 1984). There is no persistent thermal s t ra t i f i ca t ion , though during the warm season surface waters may be 1-2°C warmer than deeper waters. A - 5 -F ig . 1. Map of Southern Indian Lake, Manitoba, showing limnological regions, relationship to the Churchill River diversion and adjacent water bodies (Bodaly et a l . 1984 ). - 7 -south to north negative thermal gradient during the open water season, involving regions 1 to 5, is imposed by the ice-out pattern and reinforced by river diversion so that temperatures in the northern regions are 1-2°C lower than those in the southern regions. Water mass temperatures at maximum heat content for regions 1, 2, 6, 4 and 5 in 1978 were 1 4 . 8 ° , 1 4 . 2 ° , 1 4 . 1 ° , 12.0° and 13.0° C, respectively (Hecky 1984). Before impoundment less than 5% of the total shoreline length was actively eroding. The water in regions 4 and 5 was quite clear while that in region 6 was turbid. Immediately after impoundment over 80% of the shoreline was subject to erosion. An -average of 4 x 106 tonnes of mineral sediment were added annually to the lake from surrounding shorelines during the f i r s t three years after flooding (Newbury and McCuHough 1984). Suspended sediment concentrations in regions 1, 2, 4 and 6 increased markedly making the water much more turbid. Region 5 did not change signif icant ly and the water remained relatively clear. Concentrations of f i l te rab le suspended solids (that i s , those having a diameter of >1 micrometre) for regions 1, 2, 4, 6 and 5 in September 1978 were approximately 12.5, 11.9, 8 .8, 16.6 and 1.2 g*m3, respectively (Hecky and McCullough 1984). Hecky (1984) described the light regime of SIL. Average vertical extinction coefficients (k) for regions 4, 5 and 6 in 1978 were 1.11, 0.88 and 1.47-1.96, respectively. Average Secchi disc depths for regions 4, 5 and 6 were 1.4, 3.0 and 1.0 m. In regions 4 and 6 l ight scattering by suspended sediments is responsible for high values of k (Hecky 1984). - 8 -The water in poorly flushed areas, region 5 and some protected bays, is rich in dissolved humic substances which give i t a dark brown or orange colour. The water in other areas is blue/green in colour. Detailed descriptions of the geography of the SIL region and the hydraulic regime of the Churchill River and SIL are found in Newbury et a l . (1984). Hecky (1984) and Hecky and McCullough (1984) discussed aspects of the physical limnology of SIL. Primary productivity, crustacean plankton and profundal macrobenthos of SIL are described in Hecky and Guildford (1984), Patalas and Salki (1984) and Wiens and Rosenberg (1984), respectively. Common f ish species in SIL include lake whitefish (Coregonus clupeaformis), white sucker (Catostomus commersoni) and longnose sucker (C. catostomus) as the most abundant benthivores, ciscoes (Coregonus artedii and related species) as the main open-water planktivores, and northern pike (Esox lucius) , walleye (Stizostedion vitreum) and burbot (Lota lota) as the dominant piscivores. The main forage f ish are yellow perch (Perca flavescens), trout perch (Percopsis omiscomaycus), spottail shiners (Notropis hudsonis) and emerald shiners (Notropis atherinoides) (Bodaly et a l . In press). - 9 -COLLECTION OF SPECIMENS Adult lake whitefish were collected from region 5 of SIL during the periods 12-15 July and 20 September to 13 October, 1982. In July, 18 overnight sets were carried out at 8 different sites (Fig. 2) using gangs of gi l lnets with 6 panels of length 46 m (50 yd) and stretched mesh sizes of 3.8 cm (1 1/2 in) , 5.1 cm (2 in ) , 7.0 cm (2 3/4 in) , 8.9 cm (3 1/2 in) , 10.8 cm (4 1/4 in) and 13.3 cm (5 1/4 in) . Sixteen sets were on the lake bottom, 2 were f loating surface sets. On removal from the net whitefish were recorded on the basis of external colouration as l ight-coloured, dark-coloured or intermediate-coloured. Position of each whitefish in the top, middle or bottom thirds of the net was recorded. Blood samples were taken from 90 f ish for hemoglobin electrophoresis. Each whitefish captured was given a numbered tag, placed in a plast ic bag, iced immediately, frozen within 4 days and transferred to -40°C storage until examination. In September and October, 70 sets were carried out at 35 different sites (Fig. 3) using gangs of gi l lnets with 3 panels of length 46 m (50 yd) and stretched mesh sizes of 8.9 cm (3 1/2 in ) , 10.8 cm (4 1/4 in) and 13.3 cm (5 1/4 in) . On removal from the net, colour of f i s h , position in net (sets 1-21) and, in most cases, sex and spawning condition (immature, ripe, running, spent) were recorded for each whitefish. Of 354 whitefish captured, 93 were kept. Of these, 12 l ights and 11 darks were used in a breeding experiment (described below). Al l f ish kept were tagged and frozen as above. - 10 -F ig . 2. A. Map of region 5, Southern Indian Lake, B. sites sampled in July 1982. - 12 -F ig . 3. Map of region 5, Southern Indian Lake showing sites sampled in September-October 1982. - 14 -COLOUR CLASSIFICATION OF ADULT WHITEFISH Introduction Adult whitefish taken both in commercial and experimental catches at SIL have tradit ional ly been c lass i f ied subjectively according to the overall lightness or darkness of their external colouration. In order to determine whether subjective c lass i f icat ion corresponds to an objective measure of pigment, the number of melanophores was counted on a standard patch of skin. Melanophore counts were compared between f ish c lass i f ied subjectively by the author, and also between f ish c lass i f ied subjectively by different observers. Methods Subjective Method This method is used by the Freshwater Fish Marketing Corporation (FFMC) to c lassi fy commercial whitefish catches and by Fisheries and Oceans personnel to c lassi fy whitefish caught for experimental purposes. The observer assesses the lightness or darkness of external colouration of the whitefish, noting especially colour of the top of the head, dorsal and lateral body surfaces, and f ins . Whitefish c lass i f ied as lights are typical ly si lvery overall and light green dorsally. Those c lass i f ied as darks are typical ly darker overal l , particularly above the lateral l ine , and dark brown to black on the dorsal surface. A third c lass, intermediate-coloured, includes f ish with grades of colour between the two extremes. This class is not recognized by the FFMC which generally cal ls intermediate-coloured whitefish darks. - 15 -Quantitative Method Dark colouration in whitefish is due to the presence of melanophores (specialized colour cel ls which contain melanin granules) in the skin. Darkness of a f ish may be quantified by counting the number of melanophores per unit area of skin (Odiorne 1933; Summer 1939; Ahmad 1972). The exposed part of a scale is covered by a patch of skin in which melanophores are v is ib le . Because of the ease with which scales can be removed, stored, and handled and because virtual ly the same scale can be taken from each f i s h , melanophore counts were done on scales. One scale was removed from the dorso-lateral surface of each whitefish ( left side, 3 rows ventral and posterior to the dorsal insertion). Counts were done using a Bausch and Lomb projector-type scale reader which projected the scale image, magnified 45x, onto a grid. Melanophores, clearly v is ib le as dark spots, were counted within what on the scale would be a 2 mm x 1 mm area. A melanophore on a grid l ine was included in the count only i f more than half of i t was within the count area. Counts were done in the centre of the patch or in an area representative of melanophore size and density over the whole patch. Scale melanophore count distributions were determined for each colour class as well as for the total sample. Fish age and size were compared with scale melanophore count to see i f l inear relationships exist between them. Comparison of Subjective and Quantitative Methods Scale melanophore counts were done on 3 different samples of whitefish from region 5 caught at different times and subjectively - 16 -colour classed by different groups of observers (each group included 2 or more people, with some overlap between groups). Means and ranges for each colour class were compared between observer groups to determine var iabi l i ty between observers. Subjective colour c lassi f icat ions and scale melanophore counts of individual f ish were compared to determine the number of f ish misclassif ied by the different observer groups. Results Scale Melanophore Counts of Study Specimens Mean scale melanophore counts for whitefish subjectively c lass i f ied by the author as l ight , intermediate and dark-coloured are signif icantly different (light vs. intermediate, t = 3.254, p <0.005; light vs. dark, t = 16.175, p <0.001; intermediate vs. dark, t = 11.785, p <0.001), but ranges are broad and overlapping, indicating error in the subjective c lass i f icat ion (Fig. 4). The distr ibution of scale melanophore counts for the total sample is positively skewed (x2 = 66.147, 23 d . f . , p <0.005) and does not f i t a normal distr ibution (Fig. 5). There was no l inear relationship between f ish age and melanophore count or between f ish size and melanophore count ( Figs. 6 and 7). Young and small f ish tended to have low scale melanophore counts, but larger and older f ish showed a wide range of counts from low to high. Var iabi l i ty in Colour C lass i f icat ion Between Observers Ranges of scale melanophore counts of l ights , intermediates, and darks are roughly similar for the 4 different samples of whitefish (study specimens plus the 3 other samples) (Table 1). Mean melanophore - 17 -F ig . 4. Scale melanophore counts of lake whitefish from region 5, Southern Indian Lake by colour class as subjectively c lass i f ied by author. A. Lights, B. Intermediates, C. Darks. Arrows show modes, triangles show means. - 18 -15 -ID-S ' Ln u X A. Lights N=I43 Modes 3 5 - 3 9 Means 4 9 . 7 8 T r T 1 1 1 I 10-5- n B. Intermediates N=99 Mode= 4 0 - 4 4 Mean =59.29 T r 20-1 15 10" 5-C. Darks N=II2 Mode= 8 0 - 8 4 Mean=98.ll T 1~, 1 1 T——i 1 r -10 20 30 40 50 60 70 80 90 KXO 110 120 130 140 150 160 212 Number of melanophores per scale count area (2mm X I mm) - 19 -F ig . 5 Scale melanophore counts of total sample of lake whitefish from region 5, Southern Indian Lake. - 20 -30-1 25-20-O 0> X I E 3 10-4-N= 354 LnJ T 1 1 1 1 1 r— 10 20 30 40 50 60 70 80 i 1 1 1 r——i 1 X—// 90 100 110 120 130 140 150 160 212 Number of melanophores per scale count area (2mm X I mm) - 21 -F i g . 6. P l o t of s c a l e melanophore count versus age f o r t o t a l sample of lake w h i t e f i s h from region 5, Southern Indian Lake. 0 1 -> CO a CO or Number of melanophores per scale count area (2 mm X I mm) tv . £ 0 ) go Q - f\) £ 2 2 9 9 o ro O O O O O O O O O o O ' ' I I I L_ m — »m • «• • • • «••>•«•«• « r • «m» 9tm»mmm—B •• • • * M M * • «M • • m M • • • » • • • • • • • • • •••• M * • mm • <, •••>••«••«• • M M • • • •••*•<• mm m»m ••<••> M • • • ••• — • • mt • • • • • • ••• •• • • •• • m • • • ro ro O' - 23 -Fig. 7. Plot of scale melanophore count versus forklength for total sample of lake whitefish from region 5, Southern Indian Lake. - 24 -o 03 c O u o 220-200-180-*> I60H o o to l_ 0) CL IO 0) Q . O c JO E a> - O E =3 - 140-1 - 120 H x E 1 0 0 1 E CJ 80H 60" 40-20-0 .. • •••••• . • ^ • • •• • • — i — i — i — i — i — i — i — i — i — i — i i 150 200 250 300 350 400 450 — i 1 1— 500 550 Fork Length (mm) - 25 -Table 1. Scale melanophore counts of l ight , intermediate and dark lake whitefish colour c lass i f ied by 4 different groups of observers. Scale melanophore counts Lights Intermediates Darks Observer group Mean Range Mean Range Mean Range 60.15 25-136 67.85 28-120 89.57 52-157 42.91 13-74 49.81 24-88 82.42 36-140 3 4 57.62 17-148 67.73 40-117 102.86 66-157 53.58 13-101 58.24 20-135 95.26 57-152 - 26 -counts for each colour class d i f fe r between groups but each group had mean melanophore counts below 70 for their l ights and intermediates and above 80 for their darks. Misclassi f icat ion of Fish If only 2 classes of f ish are considered, lights and darks, and i f the dividing line between the 2 groups is considered to be 80 melanophores«2 mm2 (re. F ig . 5), then the number of f ish misclassif ied by the various observers can be determined. For each of the 4 samples shown in Table 2 numbers of " l ights" with counts above 80 and "darks" with counts below 80 were t a l l i e d . The percentage of f ish which were called lights but had more than 80 nielanophores*2 mm2 ranged from 0 to 21%. The percentage of f ish which were called darks but had less than 80 melanophores«2 mm2 ranged from 12 to 50% (Table 2). Sources of error in colour c lass i f icat ion by subjective assessment of external colouration are discussed below (see Environmental Alteration of Colour - Adult Whitefish; and Discussion). Conclusion Fish c lass i f ied subjectively by colour were found to have on average a corresponding difference in melanophore counts. This was true of subjective c lass i f icat ion by the author, and by 3 other sets of observers. However, agreement was imperfect between the subjective and objective methods of c lass i f i cat ion . Because most melanophore counts of whitefish c lass i f ied subjectively as intermediates f a l l within the range of melanophore Table 2. Numbers of light and dark lake whitefish misclassified by different groups of observers. Observer Total f ish Lights with Total f ish Darks with group classed as lights counts above 80 classed as darks counts below 80 1 2 3 4 27 32 39 104 5 0 8 5 23 38 42 70 9 19 5 18 ro - 28 -counts of those c lass i f ied as l ights , the rest of this paper wil l recognize only l ight and dark whitefish. A l ight whitefish is quantitatively defined as one having a scale melanophore count of <80 melanophores»2 mm2, a dark is one with a count >^ 80 melanophores»2 mm2. - 29 -SPATIAL DISTRIBUTION OF COLOUR CLASSES OF WHITEFISH IN SIL Introduction The purpose of this section is to describe the distribution of l ight and dark whitefish in SIL. Of interest are whether there is temporal consistency in their lake-wide distribution patterns, whether lights and darks show spatial or temporal segregation when they occur together, and whether young-of-the-year whitefish show the same colour and distr ibution differences as the adults. Methods Data from Sunde (1963), Bodaly et a l . (1983), and Bodaly et a l . (1984) were ut i l i zed to examine distr ibution of adult lights and darks between areas over several years. Sunde (1963) c lass i f ied whitefish as l ights or darks according to scale colour. The colour c lass i f icat ion of Bodaly et a l . (1983) was intended to follow that used by the Freshwater Fish Marketing Corporation. Data given in Bodaly et a l . (1984) are from commercial catches which were colour c lass i f ied subjectively by FFMC personnel. Areal distr ibution of l ight and dark whitefish was examined by comparing scale melanophore counts of f ish in 7 onshore and 9 offshore gi l lnet sets carried out in region 5 in July 1982 under this study (see Collection of Samples). These 16 were bottom sets, an additional 2 were f loating surface sets. Position in the net (top, middle or bottom thirds) was recorded for each whitefish captured in the bottom sets. Vertical distr ibution of l ights and darks was examined by comparing scale melanophore counts - 30 -of whitefish caught in the top, middle and bottom thirds of the net. (Seventeen f ish from the f a l l sample for which position in net was recorded are included. The 8 f ish captured in f loating sets are not included. Thus values of N (total f ish) for areal and vertical distr ibution d i f fe r . ) Two x k contingency tables were used to compare areal and vertical distr ibution of l ights and darks. To determine whether there were temporal or seasonal changes in proportions of light and dark whitefish in a given area, July and September/October, 1982 catches of l ights and darks in region 5 were compared. Scale melanophore counts were done on whitefish captured in July. As most f ish captured in September/October were subsequently released their scale melanophore counts were not known. For these f ish the comparison is based on the subjective colour c lass i f i ca t ion . Whitefish subjectively c lass i f ied as intermediates in both July and September/October catches were not included in these comparisons. Young-of-the-year (YOY) whitefish were collected by Fisheries and Oceans personnel in regions 4, 5 and 6 in July and August of 1982. The f ish were captured in a 1/2-metre net towed at the water surface for 3 minutes at a time by a boat at slow speed. Samples sizes were 9 f ish from region 4, 29 from region 5 and 18 from region 6. YOY colour was quantified by counting dorsal and'head melanophores under a binocular microscope. Dorsal counts included a l l dorsal melanophores between an imaginary l ine connecting the pectoral f in insertions and the caudal peduncle. Head counts were restricted to the cluster of melanophores on the top of the head (between the occiput and the lines joining the posterior thirds of the eyes) (Fig. 8). Analysis of covariance was used - 31 -F i g . 8. Dorsal view of lake w h i t e f i s h l a r v a showing head and dorsal melanophore count areas ( f o r counts on w i l d caught young-of-the-year lake w h i t e f i s h from Southern Indian Lake). - 3 2 -HEAD COUNT DORSAL COUNT - 33 -to compare head and dorsal melanophore counts of YOY from the 3 regions, with length (measured from the t ip of the snout to the urostyle, excluding the caudal f in) as the independent variable. Results Lake-wide Distribution Patterns of Lights and Darks Distribution of l ight and dark whitefish throughout SIL is not homogeneous. Light-coloured whitefish are found in a l l regions and predominate in regions 3, 4 and 6. Dark-coloured whitefish occur mainly in region 5 but are also found in region 4 in small concentrations, generally near shore, and in regions 1 and 2 (Fig. 9). These regional patterns are consistent between years (Fig. 9) . Areal Distribution of Lights and Darks in Region 5, 1982 Of 117 whitefish captured in onshore sets, 76 (65%) were lights and 41 (35%) were darks (Fig. 10). Of 144 whitefish captured in offshore sets, 115 (80%) were lights and 29 (20%) were darks (Fig. 10). Sl ightly more lights were taken offshore than onshore (12.8 and 10.9 f ish per set, respectively). Many more darks were taken onshore than offshore (5.9 and 3.2 f ish per set, respectively). The distributions of l ights and darks in onshore and offshore sets differed s ignif icant ly (x2 = 7.9, 1 d . f . , p <0.005). Vertical Distribution of Lights and Darks in Region 5, 1982 Vertical distr ibution of l ights in the nets was 16% in the top th i rd , 28% in the middle, and 56% in the bottom (N = 187). Of 83 darks, - 34 -F ig . 9. Proportion of l ight and dark lake -whitefish in experimental and commercial catches, Southern Indian Lake, 1963-1982. A. Sites sampled by Sunde (1963), B. Commercial catches, regions 4 and 5, 1979, 1980, 1981 (Bodaly et a l . 1984 ), C. Experimental catches, regions 4, 5 and 6, 1982 (Bodaly et a l . 1983). CP - 37 -- 38 -F ig . 10. Area! distr ibution of light and dark lake whitefish in region 5, Southern Indian Lake, 1982. A. Scale melanophore counts of lake whitefish in onshore sets, B. Scale melanophore counts of lake whitefish in offshore sets. - 39 -A.Scale melanophore counts of lake whitefish in onshore sets 10 -5 -a> N = 76 N = 4I T—I 1 | I | I | I | I | I | I | I | » » i 1 i /f~* r B. Scale melanophore counts of lake whitefish in offshore sets CD - O E 20-15 10 -5 -N= 115 N = 29 i i i i—» i • i ' i » i—i i • i—• i i i * i • i i i i i i i //1 i < 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 212 Number of melanophores per scale count area (2 mm X I mm ) - 40 -19% were in the top th i rd , 47% in the middle and 34% in the bottom (Fig. 11). These distributions d i f fe r s ignif icant ly {%2 = 12.6, 4 d . f . , p <0.025). Eight whitefish were captured in the 2 f loating sets. Two of these were lights (scale melanophore counts of 60 and 74), 6 were darks (scale melanophore counts of 81-152). Temporal Changes in Distribution of Lights and Darks in Region 5, 1982 Of 174 whitefish captured in region 5 in July and c lass i f ied on capture as l ight or dark, 117 (67%) were lights and 57 (33%) were darks (Fig. 12). Of 164 whitefish caught in the same area in September/October, 54 (33%) were lights and 110 (67%) were darks. Colour of Young-of-the-year Whitefish from Different Regions Region 5 YOY whitefish had signif icant ly higher mean head and dorsal melanophore counts than YOY from regions 4 and 6 (Table 3). Slopes of the regression lines for the various samples were not s ignif icant ly different. Because both dark and light adults occur there one might expect the colour distr ibution of region 5 YOY to be bimodal. Sample size was insuff ic ient to test th is . The contrast in appearance between YOY from regions 5 and 6 was st r ik ing . Region 5 YOY were dark-looking; large melanophores covered the entire dorsal surface, extending down the sides, and on the ventral surface followed the gut to the t a i l . There were small melanophores along and below the lateral l ine. On region 6 YOY of similar size melanophores were small, did not cover the whole head and were few ventral ly. - 4 1 -F i g . 11. V e r t i c a l d i s t r i b u t i o n of l i g h t and dark lake w h i t e f i s h i n region 5, Southern Indian Lake, 1982. A. Scale melanophore counts of lake w h i t e f i s h i n top 1/3 of net, B. Scale melanophore counts of lake w h i t e f i s h i n middle 1/3 of net, C. Scale melanophore counts of lake w h i t e f i s h i n bottom 1/3 of net. - 42 -A. Scale melanophore counts of lake whitefish in top third of net 1 0 •» N=30 I I 5 -N = 16 r u r u * I * i 1 1 ' i 1 1 1 1 * I * i * i 7hn,n. .n. B. Scale melanophore counts of lake whitefish in middle third of net 15 -i 10 5 -I N=52 1 • I < l > I ' I > I ~T~T N = 39 L n u i • i • i • 15 C. Scale melanophore counts of lake whitefish in bottom third of net N= 105 N=28 I I I 10 -5 - n L r i i * i • i • i • i • i • i i i • i • i • i • i 1 I • I ' i • i 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 Number of melanophores per scale count area (2mm X I mm) 212 - 43 -Fig . 12. Scale melanophore counts of light and dark lake whitefish captured in region 5, Southern Indian Lake, in July 1982. - 44 -2Cn a> 15-10-E 5 N = 117 n Ln N= 58 Jl T -i I i I i I • I ' I ' I ' 1 ' I ' I ' I 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 Number of melanophores per scale count area (2 mm X I mm) /A-7/ 212 - 45 -Table 3. Analysis of covariance on head and dorsal melanophore counts of young-of-the-year lake whitefish from regions 4, 5 and 6, Southern Indian Lake. F Type of Adjusted (adjusted count Comparison group means group means) d . f . Head Region 6, Region 4 367.3, 350.9 0.244 1,24 0.626 Region 6, Region 5 324.9, 549.8 107.635 1,37 0.000 Region 4, Region 5 326.6, 584.5 47.492 1,28 0.000 Dorsal Region 6, Region 4 55.7, 58.1 0.083 1,23 0.775 Region 6, Region 5 56.9, 80.5 13.676 1,43 0.001 Region 4, Region 5 60.8, 86.6 7.598 1,33 0.009 - 46 -Conclusion Distribution of light and dark-colourd whitefish (both adults and young-of-the-year) in SIL is not homogeneous. Dark whitefish are restricted to part icular areas of the lake. Where they occur together, l ights and darks appear to segregate to some degree both areally and ver t ica l ly . Lights were more abundant offshore than onshore and were most often found on the lake bottom. Darks were more numerous onshore than offshore and were more often caught off the bottom. Relative numbers of l ights and darks in region 5 changed over time in 1982 with lights predominating in the summer catch and darks predominating in the f a l l catch. Region 5, where dark adults were most numerous, also had young-of-the-year with higher melanophore counts. - 47 -MORPHOLOGY AND BIOCHEMISTRY OF ADULT LIGHT AND DARK WHITEFISH Introduction The purpose of this work was to investigate whether light and dark colouration is correlated with morphometric measurements, meristic counts or biochemical characters. Methods Morphological Analyses Fifty-two morphometric measurements and 16 meristic counts were taken on a preliminary sample of 24 light and 24 dark whitefish caught in region 5, SIL in 1979 (Table 4) . Girth is the circumference of the body immediately anterior to the dorsal f in or igin. Pectoral and pelvic f in ray counts included a l l principal rays. Regression lines of morphometric measurements on fork length and frequency distributions of meristic counts were compared between l ights and darks. Measurements and counts thought most l ikely to show differences between lights and darks were chosen for further study. Subsamples of 45 f ish of each colour were taken from the 1982 summer and f a l l samples. Fish were selected on the basis of scale melanophore count, age and sex. Only lights with scale melanophore counts <60 and darks with counts >80 were included. Al l age classes from 2 years to 16 years were represented as equally as possible and roughly equal numbers of females and males were chosen (Table 5). Twenty-two morphometric measurements, as chosen from the preliminary set (above) were taken for each f ish (Table 4) . Before examination the f ish were thawed overnight at 1°C. Al l measurements were taken on the left side of the f i s h . Measurements were straight - 4 8 -T a b l e 4 . Morphometr ic measurements and m e r i s t i c counts made on l i g h t and dark l a k e w h i t e f i s h f rom r e g i o n 5 , S o u t h e r n I n d i a n L a k e . *: measurements and counts s e l e c t e d f o r d e t a i l e d e x a m i n a t i o n In 1982 s a m p l e . Measurement o r count R e f e r e n c e Fork l e n g t h S t a n d a r d l e n g t h Body c o n t o u r Head l e n g t h * Head depth* Head w i d t h Body l e n g t h * Body w i d t h * P r e d o r s a l depth P o s t d o r s a l depth O c c i p i t a l t o d o r s a l d i s t a n c e D o r s a l t o a d i p o s e d i s t a n c e V e n t r a l t o a n a l d i s t a n c e A n a l t o cauda l d i s t a n c e A n a l depth P e d u n c l e depth P e d u n c l e l e n g t h Snout l e n g t h * P r e m a x i l l a h e i g h t * M a x i l l a l e n g t h * M a x i l l a w i d t h W i d t h of gape* P r e p o s t o r b i t a l d i s t a n c e P o s t o r b i t a l l e n g t h of head S u b o r b i t a l w i d t h I n t e r o r b i t a l w i d t h * O r b i t l e n g t h * Eye l e n g t h Eye d i a m e t e r * V e r t i c a l p u p i l d i a m e t e r * H o r i z o n t a l p u p i l d i a m e t e r H e i g h t of cheek Length o f cheek O r b i t t o a n g l e of p r e o p e r c l e O o r s a l f i n o r i g i n L e n g t h of d o r s a l base Length of depressed d o r s a l f i n * L e n g t h of l o n g e s t d o r s a l ray P e c t o r a l f i n o r i g i n L e n g t h of p e c t o r a l f i n * P e l v i c f i n o r i g i n Length of p e l v i c f i n * A d i p o s e f i n o r i g i n Length of a d i p o s e f i n * Ana l f i n o r i g i n Length of ana l base Length of depressed anal f i n * G i l l r a k e r l e n g t h * Gi11 r a k e r space* L e n g t h of lower a r c h * Average g i l l r a k e r space* Gi r t h * B r a n c h i o s t e g a l rays S t a n d a r d l a t e r a l l i n e s c a l e s T o t a l l a t e r a l l i n e s c a l e s S c a l e s above l a t e r a l l i n e S c a l e s below l a t e r a l l i n e S c a l e s b e f o r e d o r s a l f i n S u p r a p e l v i c s c a l e s Cauda l pedunc le s c a l e s C i r c u m f e r e n c e s c a l e count D o r s a l f i n rays P e c t o r a l f i n rays P e l v i c f i n rays Ana l f i n rays Upper g i l l r a k e r s * Lower g i l l r a k e r s * T o t a l g i l l r a k e r s * L l n d s e y 1963 L i n d s e y 1962 Loch 1974 L l n d s e y 1963 I b i d Hubbs and L a g l e r 1958 L i n d s e y 1963 Loch 1974 I b i d I b i d I b i d K r i s t o f f e r s o n 1978 I b i d Loch 1974 I b i d L i n d s e y 1962 Hubbs and L a g l e r 1958 I b i d Boda ly 1979 L i n d s e y 1962 I b i d Hubbs and L a g l e r 1958 L l n d s e y 1963 Hubbs and L a g l e r 1958 I b i d L i n d s e y 1962 Hubbs and L a g l e r 1958 I b i d B o d a l y 1979 I b i d I b i d Hubbs and L a g l e r 1958 I b i d I b i d K r i s t o f f e r s o n 1978 Hubbs and L a g l e r 1958 I b i d I b i d L i n d s e y 1963 I b i d I b i d K r i s t o f f e r s o n 1978 I b i d L i n d s e y 1963 K r i s t o f f e r s o n 1978 Hubbs and L a g l e r 1958 I b i d K l i e w e r 1970 I b i d B o d a l y 1979 I b i d See t e x t Hubbs and L a g l e r 1958 I b i d L i n d s e y 1962 Hubbs and L a g l e r 1958 I b i d I b i d L i n d s e y 1962 I b i d Hubbs and L a g l e r 1958 L i n d s e y 1962 See t e x t See t e x t K r i s t o f f e r s o n 1978 See t e x t See t e x t See t e x t Table 5. Age distr ibutions, numbers of males and females and dates of capture of l ight and dark lake whitefish used for morphological analyses. Colour Number of f ish and age Sex Date of capture 2 3 4 5 6 7 8 9 10 11 12 13+ F M Imm July October Light 4 4 4 4 4 4 4 4 4 2 1 6 21 22 2 26 19 Dark 1 3 1 5 5 5 5 4 4 4 4 4 1 9 25 1 27 18 - 50 -l ine and did not follow body contour. Analysis of covariance was used to discern morphometic differences between light and dark whitefish. Body length (fork length minus head length) was used as the independent variable. Three meristic counts (upper, lower and total g i l l rakers) were made on the whitefish samples using a dissecting microscope. G i l l raker number was counted on the entire f i r s t left g i l l arch after i t was removed in i ts entirety from the f i s h , including every bony rudiment. Lower g i l l raker number includes the middle g i l l raker. Differences in the distributions of meristic counts between lights and darks were tested by 2 x k contingency tables. Where necessary counts were combined so that no expected value was less than 4 .5 . Mean g i l l raker number of l ights and darks was compared using a t - tes t . Age and Growth Scales were taken from each f ish on the left side just below the dorsal f in origin. Scale ages were determined by L. Patterson, Manitoba Department of Natural Resources, Winnipeg. Scale age was plotted against fork length to provide an estimate of growth rates of light and dark whitefish from region 5. Biochemical Analysis Glycerol-3-Phosphate Dehydrogenase (G-3-PDH) Electrophoresis G-3-PDH was selected for study because Bodaly et a l . (1984) showed that there were regional differences in G-3-PDH a l le le frequencies among whitefish in SIL before flooding. - 51 -Subsamples of 36 f ish of each colour were taken from the original sample. Al l f ish used as parents in the breeding experiment, l ights with scale melanophore counts <50 and darks with counts >95 were chosen. Not a l l f ish used for morphological analysis were included. A white muscle sample was excised from the epaxial muscle bundle just below the dorsal f in on the right side of the f i s h . Tissue samples were placed in plast ic bags and frozen for later analysis. Tissue extracts were prepared by macerating approximately 1 g of white muscle tissue in a solution of 0.25 M sucrose, 300 mg«L nicotinamide adenine dinucleotide (NAD) at a 1:3 ratio of muscle t issue to solution. The extracts were centrifuged at 18 000 RPM and 1°C for 20 minutes. The clear fraction was removed by pipette and frozen unti l needed. Starch gel electrophoresis of tissue extracts was done following Tsuyuki et a l . (1966). The buffers were c i t r i c acid adjusted to pH 8.0 with t r i s (hydroxymethy1) amino methane. The gel buffer concentration was 0.002 M, the electrode buffer was 0.04 M. NAD was added to the buffers to a concentration of 100 mg«L. The samples were run at 160 V for 2 hours at a temperature of 1°C. G-3-PDH phenotypes were visualized by the methods of Clayton et a l . (1973). Genotypes of individual f ish were inferred from electrophoretic phenotypes according to an established genetic model for G-3-PDH (Clayton et a l . 1973). - 52 -Hemoglobin Electrophoresis Blood was collected from 63 light whitefish (scale melanophore counts <80) and 27 dark whitefish (scale melanophore counts >80) on capture. Blood was taken from the caudal artery (immediately behind the anal f in) of l ive f ish by vacutainer. Samples were kept in ice water and used within 6 days. The whole blood was centrigued to separate the plasma and erythrocytes. The erythrocytes were washed 3 times with a 1% NaCl solution, then lysed with 3 volumes of d i s t i l l e d water and centrifuged. The clear fraction was pipetted off and used immediately. Starch gel electrophoresis of hemoglobin extracts was done following Tsuyuki et a l . (1966). The buffers were boric acid adjusted to pH 8.5 with NaOH. The gel buffer concentration was 0.023 M, the electrode buffer was 0.3 M. EDTA was added to the gel buffer to a concentration of 0.025%. The samples were run at 160V for 2 hours at a temperature of 1°C. Hemoglobin phenotypes were visualized using ami do black stain (2.2 g ami do black, 1000 ml d i s t i l l e d water, 200 ml acetic acid, 1 000 ml methanol). Genotypes of individual f ish were inferred from electrophoretic phenotypes according to an established genetic model for hemoglobin (J.W. Clayton, pers. comm.). G-3-PDH and hemoglobin a l le le frequencies were calculated for l ight and dark samples. The Castle-Hardy-Weinberg expected distributions of phenotypes were compared with observed phenotypic distributions to test for any lack of homogeneity within samples. A l le le numbers were compared between lights and darks using 2 x k contingency tables. - 53 -Results Morphology Differences in Morphological Characters Between Lights and Darks Seven of the 21 morphometric characters, a l l head measurements, showed s ta t i s t i ca l l y signif icant differences between dark and light whitefish (Table 6). Where the difference was between adjusted group means (head depth, interorbital width, snout length and lower arch length), values were larger for l ights than for darks. Where the difference was in the slopes (snout length, eye diameter, maxilla length and orbit length), those for darks were steeper than those for l ights . For snout length and eye diameter, values for l ights were greater than those for darks at a l l sizes included in the sample. For maxilla length and orbit length, values for l ights were greater than those for darks in f ish <316 mm body length, and less than those for darks in larger f i s h . G i l l Raker Number Mean lower g i l l raker numbers for light and dark whitefish were 17.71 and 17.33, respectively. These values are s ignif icant ly different (t = 2.197, 88 d . f . , p <0.05). G i l l raker count distributions and mean upper and total counts of l ights and darks were not s ignif icant ly di f ferent , though modes were consistently lower in darks than in lights for a l l 3 types of counts (Fig. 13) Age and Growth Growth rates of l ights and darks were similar up to age 10. At that point growth of darks appears to level off , while lights continue to increase in size (Fig. 14). - 54 -Table 6. Analysis of covariance of morphometric characters of light and dark lake whitefish from region 5, Southern Indian Lake. Standard length is the independent variable. F values (and degrees of freedom) are given for tests of homogeneity of slopes and for adjusted group means. *: signif icant at p<0.05 level . Blank: not signif icant Morphometric F(d. f . ) F (d . f . ) character slopes adjusted group means Gi rth .162 (1,86) .966 (1,87) Body width .233 (1,86) .438 (1,87) Head length 2.493 (1,86) .788 (1,87) Head depth .840 (1,86) M.776 (1,87) Interorbital width .805 (1,86) *4.427 (1,87) Snout length *5.462 (1,86) *8.108 (1,87) Premaxilla height .511 (1,86) 1.016 (1,87) Maxilla length *7.799 (1,86) .005 (1,87) Gape width 2.480 (1,86) .572 (1,87) Orbit length *4.570 (1,86) .267 (1,87) Vertical pupil diameter .838 (1,86) .000 (1,87) Eye diameter *5.675 (1,86) .365 (1,87) Dorsal f in length 2.076 (1,86) 2.594 (1,87) Adipose f in length .000 (1,86) .294 (1,87) Pectoral f in length 1.373 (1,86) .906 (1,87) Pelvic f in length 2.027 (1,86) .044 (1,87) Anal f in length 2.230 (1,86) .001 (1,87) G i l l raker length .256 (1,86) • .196 (1,87) G i l l raker space .378 (1,86) .143 (1,87) Lower arch length .844 (1,86) *3.962 (1,87) Average g i l l raker space 1.376 (1,86) .193 (1,87) - 55 -F ig . 13. G i l l raker count distributions of light and dark lake whitefish from region 5, Southern Indian Lake. A. Upper g i l l rakers, B. Lower g i l l rakers, C. Total g i l l rakers. - 56 -A. Upper gill rakers 2 5 -2 0 -15 -10-5 -0 LIGHTS N = 45 X = 9.5 2 5 n DARKS N= 45 X=9.5 20-\ 15 H 101 1 , , , — 7 8 9 10 II - F = r 8 Number of upper gill rakers B. Lower gill rakers 25 _ LIGHTS ' N=45 X=I77 20 H 15 " 10 -5 -0 2 5 -2 0 -15 " 10 -16 ' 17 ' 18 ' 19 ' 20 T-0 DARKS N = 45 X= I73 16 ' 17 ' 18 1 19 Number of lower gill rakers 10 II 20 C. Total gill rakers 25 -j 20 " 15 -10 -5 " 0 -LIGHTS N=45 X"=273 24 ' 25 * 26 ' 27 ' 28 ' 29 25 n 20 15 i 10 5 i 0 DARKS N = 45 X=26.8 23 24 25 ' 26 1 27 1 28 ' 29 Number of total gill rakers - 57 -F ig . 14. Plots of age versus forklength for light and dark lake whitefish from region 5, Southern Indian Lake. Open symbols represent means of samples with less than three f i sh . Closed symbols represent means of samples with three or more f i sh . Vertical bars show ranges. - 58 -525 -500 -475 " 450-425-E 400" E xz 375 -o> c cu 350-1 325 " o L i _ 300-275-250 -225-200" 175-2 3 4 7 9 • LIGHTS • DARKS 10 12 13 Age (Years) - 59 -Biochemical Analyses Glycerol-3-Phosphate Dehydrogenase The a 2 a l le le had a frequency of 1.0 in both light and dark whitefish samples. Observed BB G-3-PDH phenotype distributions agreed with expected distributions for both light and dark samples (Table 7). A 2 x 3 contingency table comparing numbers of a l le les at b_ loci for l ights and darks showed no signif icant difference between the two groups ( x 2 = 1.05, 2 d . f . , p >0.50). Hemoglobin Observed and expected hemoglobin phenotype distributions were in agreement for both light and dark samples (Table 8). A 2 x 2 contingency table comparing numbers of a l le les for l ights and darks showed no signif icant difference between the 2 groups (x2 = 0.973, 1 d . f . , p >0.25). Conclusion The presence of signif icant differences between lights and darks in several morphometric characters and in lower g i l l raker numbers demonstrates that more is involved than simply variation in degrees of pigmentation. However, the broad overlap in characters between the two groups, and absence of signif icant differences in the two biochemical characters examined suggest no clear-cut separation into non-interbreeding stocks. Table 7. Observed and expected (Castle-Hardy-Weinberg) BB glycerol-3-phosphate dehydrogenase phenotype distributions for light and dark lake whitefish from region 5, Southern Indian Lake. Expected numbers in brackets. Colour No. of f ish and BB G-3PDH phenotypes 1,1 1,2 1,3 2,2 2,3 3,3 x 2 (4d.f.) P Light 3(3.36) 7(6.72) 9(8.56) 4(3.36) 7(8.56) 6(5.44) 0.54 p>.95 Dark 1(2.07) 8(5.36) 7(7.95) 2(3.46) 10(10.27) 8(7.62) 2.61 p>.50 Table 8. Observed and expected (Castle-Hardy-Weinberg) hemoglobin phenotype d i s t r i b u t i o n s f o r l i g h t and dark lake w h i t e f i s h from region 5, Southern Indian Lake. Expected numbers i n brackets. Colour No. of f i s h and hemoglobin phenotypes SS SF FF x 2 ( l d . f . ) p cr> i — 1 i Light 32(31.40) 25(26.15) 6(5.45) 0.118 p>.50 Dark 15(16.34) 12(9.33) 0(1.33) 2.204 p>.10 - 62 -INHERITANCE OF COLOUR Introduction To determine whether or not there is a genetic basis for observed colour differences between light and dark whitefish a breeding experiment was performed. Methods Rearing of Fish Adult whitefish were captured by g i l l netting in Area 5 in October, 1982. (See Collection of Samples). Twelve light and 11 dark parents were chosen on the basis of their external-colouration as assessed by the subjective method. These c lass i f icat ions were later verif ied by the quantitative method. Ranges and means of scale melanophore counts of parents are given in Table 9. While the ranges do not f a l l above and below 80 melanophores per count area (the dividing line between lights and darks, re. Colour Classi f icat ion of Adults) they are non-overlapping and the mean counts are s ignif icant ly different (t = 8.22, 21 d . f . , p <0.001). Four sets of crosses were done, 3 3 x 3 crosses and 1 3 x 2 cross. Experimental design is i l lust rated in F ig . 15. The crosses were made in the f ie ld using the dry fe r t i l i za t ion method. The eggs were held in screened plast ic containers in the lake (surface temperature 5 .5°-4 .5°C) before being transferred to the Freshwater Institute. The eggs were transported by truck on Oct. 20, 1982. The egg containers were carried in plast ic coolers f i l l e d with ice water at 1°C. Of a possible 33 batches (famil ies) , 7 were not completed or were lost and 26 were instal led in the hatchery (6 light x - 63 -Table 9. Ranges and means of scale melanophore counts of parents used in breeding experiment. Range of scale Mean scale Colour N melanophore counts melanophore count Light 11 17-69 45.58 Dark 12 76-126 105.64 - 64 -F i g . 15. Experimental design u t i l i z e d i n crosses of l i g h t and dark lake w h i t e f i s h . Set 1 Set 2 Sex Tag no. 15333 15337 15334 Colour D(121) L(20) D(126) Sex Tag no. Colour Family number 9 15335 D(97) 1 2 3 15338 L(17) 4 5 6 15336 D(102) 7 8 9 Set 3 Sex 9 Tag no. 15346 15347 15348 Colour L(54) D(110) L(47) Sex Tag no. Colour Family numt )er 15349 L(69) 19 20 21 15350 D(110) 22 23 24 15351 L(27) 25 26 27 Tag no. - that of individual f i sh . Colour - L = l ight ; D = dark. Number in brackets is scale melanophore count. Sex <5 Tag no. 15340 15341 15342 Colour L(63) D(105) L(65) Sex Tag no. Colour Family number 9 15343 L(44) 10 11 12 15344 D(117) 13 14 15 15345 L(64) 16 17 18 Set 4 Sex 9 Tag no. 15352 15353 Colour L(22) L(55) Sex Tag no. Colour Family number 15354 0(116) 28 29 15355 D(82) 31 32 15356 D(76) 34 35 - 66 -l ight , 15 light x dark, 5 dark x dark). In i t ia l losses were due to inabi l i ty to str ip suff icient eggs from some females and to egg mortality. The number of extant families diminished somewhat over the incubation period, primarily due to sampling, so that at hatching 20 families remained (4 light x l ight , 11 light x dark, 5 dark x dark). The eggs were incubated in screen-covered clear plast ic jars. Cooled dechlorinated water was supplied to the jars from a head tank. A constant flow was maintained through each jar at a rate just high enough to tumble the eggs very slowly. The jars stood in a cool water bath which also had a flowthrough water supply. Temperature in the head tank was 1° - 1.5°C, in the water bath was 2.5° - 2.75°C, and in the egg jars was 1.5° - 4°C at different times during incubation. Dead eggs were removed from the jars regularly. Malachite green treatments were given over the incubation period to retard growth of fungus on the eggs. Before hatching began the egg jars were transferred to f ish tanks where they were clamped above the water surface in the tanks. Flowthrough water was supplied to both jars and tanks; the latter were also provided with airstones. Hatching began in early February and continued unti l the end of March. During this period jar temperatures were gradually allowed to rise to 6° - 7°C. Hatched f ish were carried out of the jars into the tanks by the flow of water. Tank temperatures were 9° - 10°C. The lights in the hatchery were on a 12-hour daily cycle, on at 0800 hours and off at 2000 hours. The fry were i n i t i a l l y fed brine shrimp (Artemia) nauplii and ground Tetra Min, and were later switched to wild-caught Daphnia - 67 -supplemented with a mixture of ground Tetra Min and rainbow trout starter . The fry were fed 2 or 3 times dai ly . The tanks were siphoned daily to remove waste and scrubbed weekly. Sampling Procedure As early as 35 days after fe r t i l i za t ion some larvae had v is ib le melanophores, including a double dorsal l ine , a ventral l ine , and scattered melanophores on the yolk sac. Somewhat later melanophores became vis ible on the top of the head. Melanophore counts were used as a means of quantifying and comparing the colour (lightness or darkness) of progeny of different crosses. Four pre-hatch samples were taken from most families (at 85-86, 92, 111, and 119-126 days after f e r t i l i z a t i o n ) . Not al l families were sampled at each age because some egg batches were too small to allow i t . For the 111 day sample subsample size was 12 f ish from each family; for other samples smaller subsamples were sometimes used. Larvae were removed from the eggs, preserved in 5% formalin, and stored in amber-coloured jars in the dark. In i n i t i a l samples melanophore counts were done on the head, dorsal surface, yolk sac and ventral surface of a larva. The latter 2 counts proved d i f f i cu l t to do and of questionable accuracy, thus were discontinued. Dorsal and head melanophore counts were done on subsequent samples. Head counts included a l l melanophores v is ible on the dorsal surface of the head and the occiput. Dorsal counts included the double dorsal row of melanophores which extends from behind the occiput to the caudal peduncle (Fig. 16). Because dorsal counts were easier to do and more rel iable than head counts, they alone were used in most analyses. - 68 -F ig . 16. Dorsal view of lake whitefish larva showing head and dorsal melanophore count areas (for counts on hatchery-reared larvae). - 69 -HEAD COUNT DORSAL COUNT - 70 -Results The 111 day samples comprise the most complete set of counts (24 of 26 families represented) and therefore were chosen for the following analyses. Dorsal Melanophore Counts from Different Cross Types Mean dorsal melanophore counts and variances for each cross type at 111 days post - fe r t i l i za t ion are given in Table 10. The number of families and total number of f ish included in the calculations for each cross type are also given. Mean dorsal melanophore counts d i f fer in the direction expected i f colour is inherited and a l l differences, except between the means for DD and D?L 6 0 -O >» 55 -o — 50 -o 4 5 -CO c 40" O o 35-a> o ph 30-o c _o 25-a> £ 20-o in k_ o 15 -•o an 10-a> 2 0 o o 8 o • C. Offspring on male parent value • DARK X DARK 0 DARK 9 X LIGHT Cf • LIGHT 9 X DARK Cf O LIGHT X LIGHT - 1 1 1 1 1 1 1 1 1 1 1 1 1 r-20 25 30 35 40 45 50 55 60 65 70 75 80 85 1 1 1 1 1— 90 95 100 105 110 115 120 125 130 Male parent scale melanophore count - 77 -Table 12. Her i tabi l i ty estimates from 4 pre-hatch lake whitefish samples. Sample Estimate of Mid-parent h 2 from regression of offspring on -Female parent Male parent 85-86 days 0.10 0.12 0.06 92 days 0.15 0.14 0.14 111 days 0.15 0.16 0.10 119-126 days 0.12 0.16 0.0008 - 78 -s i m i l a r f o r the 4 samples, but those from o f f s p r i n g on female and male parents are q u i t e d i f f e r e n t over the 4 samples. I f the estimated h 2 from the regression of o f f s p r i n g on female parent i s g r e a t e r than that of o f f s p r i n g on male parent a maternal e f f e c t i s i n d i c a t e d (Falconer 1960). In 3 of 4 cases the h 2 estimated by the regression of o f f s p r i n g on mother i s higher than that estimated by the regression of o f f s p r i n g on f a t h e r , i n d i c a t i n g some maternal e f f e c t (Table 12). P o s s i b l e reasons f o r the very low value of h 2 estimated by o f f s p r i n g on f a t h e r regression at 119-126 days are discussed below (See D i s c u s s i o n ) . Conclusion There are s i g n i f i c a n t d i f f e r e n c e s i n c o l o u r (as measured by dorsal melanophore counts) between l a r v a l w h i t e f i s h from d i f f e r e n t c o l o u r crosses. These d i f f e r e n c e s are i n the d i r e c t i o n expected i f c o l o u r i s at l e a s t p a r t l y i n h e r i t e d . H e r i t a b i l i t y estimates show that maternal e f f e c t s account f o r part but not a l l of the observed c o l o u r d i f f e r e n c e s . That i s , there i s a measurable genetic component t o the c o l o u r d i f f e r e n c e s between dark and l i g h t w h i t e f i s h which i s independent of environmental or maternal e f f e c t s . - 79 -ENVIRONMENTAL ALTERATION OF COLOUR Introducti on Experiments were performed to test the effects both of long- and short-term exposure to different l ight and background conditions on colour of young whitefish. Some observations were made on short-term colour change in adult whitefish. Long-term Colour Change Experiment Methods Experimental Design Four plast ic 12 l i t r e dishpans (3 dark brown, 1 cream-coloured) were used as experimental tanks. Each tank had an airstone and flow-through water supply (water replacement time was 20 minutes). Fish from 1 family (X - 24 L? D<3) were divided into 4 groups. Age of the f ish was 245 days (since fe r t i l i za t ion ) at the outset and 257 days at the end of the experiment. Eight f ish from each group were preserved immediately in 5% formalin and stored in amber-coloured jars (these comprise 0 day samples 1 - 4 ) . The remaining f ish (about 32 in each group) were put in the 4 tanks. There was a control group and 3 experimental conditions. 1. cream-coloured tank, uncovered, exposed to fu l l l ight. Since this was very similar to the conditions under which other experimental f ish were reared (in l ight-coloured, uncovered tanks exposed to l i ght ) , i t was considered the control. 2. dark brown tank, uncovered, fu l l l ight . 3. dark brown tank covered with screen which excluded half the incident l ight (as measured with an incident light meter). - 80 -4. dark brown tank covered with l ight-proof black p last ic . Even though they were covered with baff les , a low level of l ight entered the tank through the drain and airstone holes. The f ish were fed daily on l ive Daphnia and dried food (ground Tetra Min and rainbow trout starter) . The tanks were cleaned dai ly . Water temperature throughout the experiment was 10° - 11°C. The tanks were l i t by a bank of cei l ing fluorescent lights which were left on for the duration of the experiment except when the covers were removed from the light exclusion tanks for cleaning and feeding. After 12 days 12 - 14 f ish remained in each group and the experiment was terminated. One or 2 f ish from each group were preserved directly from the experimental tanks to use for comparing melanophore configuration of f ish from the 4 treatments. The remaining f ish in each group were put in a white bucket for about 5 minutes (so that the melanin granules would aggregate thus f a c i l i t a t i n g counting of melanophores), then preserved. Larvae were 15.0 to 18.5 mm in length at the beginning of the experiment and 15.5 to 24.5 mm at the end of i t . Head and dorsal melanophore counts were done on the 0 day and 12 day samples. Head counts included just the melanophores on the top of the head; dorsal counts included dorsal melanophores from a line even with the pectoral insertion to the caudal peduncle. S tat is t ica l Analysis of Data Analysis of covariance was used to discern differences in melanophore counts between the 4 12 day samples with length (measured - 81 -from the t ip of the snout to the urostyle, excluding the caudal f in) as the independent variable. Since samples 1 to 4 at 0 days were random subsamples of the same population they were combined. The 0 day sample was compared with the 12 day sample from group 1 (the control) , using ANCOVA, to determine whether any signif icant gain or loss of melanophores occurred in the control group over the experimental period. Results Comparison of melanophore configuration of f ish from the 4 groups showed that in bright surroundings the melanophores appeared small and discrete, with their pigment aggregated. With increasing environmental darkness (tubs 2 - 4), melanophore pigment was increasingly dispersed. At 12 days there were signif icant differences between mean melanophore counts and slopes of regression lines of the 4 groups (Table 13). With increasing environmental darkness the f ish became darker by producing more melanophores (Fig. 18). The adjusted group mean dorsal melanophore counts at 12 days were 659, 726, 779 and 865 for groups 1 to 4 respectively. The adjusted group mean head melanophore counts at 12 days were 159, 164, 179 and 222 for groups 1 to 4 respectively. There were signif icant gains in numbers of both dorsal and head melanophores in group 1 over the experimental period, as evidenced by comparison of 0 day and group 1 12 day samples (dorsal melanophores - F (adjusted group means) = 5.524, 1,41 d . f . , p = 0.024; head melanophores - F (adjusted group means) = 23.292, 1,4 d . f . , p = 0.00002). Such gains would be expected with increased size (the f ish grew about 2.5 mm in length over the 12 days). There was no evidence of decreased melanophore number in any group. - 82 -Table 13. Long-term colour change experiment: Results of analysis of covariance of regressions of larval lake whitefish melanophore counts on length at 12 days. F i Level of Type of (adjusted d.f F 2 d . f . significance count group means) (for Fi) (slopes) (for F 2 ) (F1 and F 2 ) Dorsal 10.431 3,41 8.175 3,38 p<.001 Head 16.972 3,41 9.777 3,38 p<.001 - 83 -F ig . 18. Long-term colour change experiment: Regressions of larval lake whitefish melanophore counts on total length before and after 12 day exposure to different l ight and background conditions. Group 1 - Pale tank, fu l l l ight . Group 2 - Dark tank, fu l l l ight . Group 3 - Dark tank, half l ight . Group 4 -Dark tank, low l ight . - 84 -1350 -I 1250 H 1150 H c O 1050 • o £ 950" o sz O 850-c o « 750-E o 650-t_ o Q o X 550 H 450 H 350 A. Dorsal melanophores Group 4 (r = .9) Group 3 (r=.9) Group 2 (r=.9) Group I (r=.9) 0 days (r = .6) i 1 1 1 1 1 1 1 1 1— 14.0 15.0 16.0 170 18.0 19.0 20.0 21.0 22.0 23.0 24.0 325n 300 275 250 ._ 225 c o o a> V-O .£= Q. O C D

The ab i l i t y of f ish to change colour through movement of pigments in specialized cel ls (chromatophores) in response to environmental changes or conditions has long been of interest to researchers (for example, Cunningham and MacMunn 1893). Two general types of chromatophore response are recognized - physiological and morphological colour changes. The former are alterations in external colouration produced by changes in the distr ibution of pigment in the chromatophores. They are relatively rapid and are evoked by any one of many st imul i . The latter arise from quantitative changes in pigmentation, that i s , increases or decreases in the number of chromatophores or the amount of pigment they contain. They are gradual and inconspicuous but may f ina l l y be more marked than changes due to pigment migration. Sumner (1940) argued that the terms "physiological" and "morphological" colour change, which were introduced by S6c6rov (1909, in Sumner (1940)), were unclear as both processes are physiological. He suggested that the names "transitory" and "quantitative" were more appropriate. The nomenclature of Sumner wi l l be adopted here. - 109 -There are several kinds of chromatophores and various chromatic pigments (Fujii 1970; Fox 1957). Melanophores, which are of primary interest in this study, are brown or black and contain melanins -oxidized, polymerized end-products of tyrosine metabolism. With carotenoids, which give erythrophores and xanthophores their red and yellow colour, melanins are primarily responsible for external colouration of fishes (Fox 1957). The colour responses of lake whitefish described above involve melanophores. Early researchers noted that f ish kept for extended periods under conditions which caused transitory colour change eventually underwent quantitative colour change (Odiorne 1933; Sumner and Wells 1933; Sumner 1940; Osborn 1941a). In i t ia l theories postulated a causal relationship between the two. However, i t is now believed that the conditions which bring about transitory change w i l l , i f maintained, cause quantitative change. Quantitative colour changes arise not from transitory changes themselves, but from the operation of the agents which produce the lat ter (Odiorne 1957; Ahmad 1972, and references therein). Transitory colour change may be controlled by blood-borne hormones or nervous pathways or both. Hormonal control is phylogenetically older and is present in cyclostomes, elasmobranchs and chondrosteans (Krasnodemskaya 1978). In teleosts i t is replaced or supplemented by direct nerve stimulation of the melanophores. Pituitary hormones, possibly melanophore stimulating hormone (MSH), are thought to control melanin dispersion in f ish (Fujii 1973). A melanophore concentrating hormone (MCH) from the pituitary has been proposed, but i ts existence has not been substantiated (Fujii 1970). There is some evidence that pineal melatonin may function in melanin concentration (Hafeez 1970; Smith and Weber 1976). - 110 -The dominant control mechanism for colour change in teleosts is nervous (Scott 1965). In 1911, von Frisch (cited in Iwata and Fukuda (1973)) discovered that sympathetic neurones controlled pigment aggregation. Subsequent researchers have argued for (Parker 1943; Robertson 1951; Ahmad 1972, 1974) and against (Scott 1965) the presence of dispersing f ibres . Iwata and Fukuda (1973) traced the nervous pathways from the retina to the melanophore in the crucian carp (Carassius carassius). They found that the systems controll ing movement of melanophore pigment are mononeuronic in the peripheral system (aggregating f ibres) and dineuronic in the central system, one set of nerves exciting the peripheral aggregating motor neurones (when the f ish is on a l ight background), the other set inhibit ing them, thus allowing pigment dispersion (when the f ish is on a dark background). There is evidence that the hormonal and neural pathways involved in transitory colour change also regulate quantitative colour change (Osborn 1941b; Fuj i i 1970; Ahmad 1974). A multitude of physical, chemical and pharmacological agents have been found to affect the state of pigment dispersion in melanophores (see reviews by Fuj i i 1970, 1973). However, the most important single environmental factor influencing pigment systems of animals is light (Brown 1962). There are two types of chromatic response to l ight , primary and secondary. The former occur by direct action of light on the chromatophores or through an extraocular receptor, and predominate in embryos, larvae with underdeveloped eyes and blinded f i s h . Pigment disperses in light and concentrates in darkness. Secondary responses are controlled by way of the eyes and allow adaptation to the background. Pigment disperses on a dark background in light and - I l l -concentrates on a light background in l ight . In most adults the secondary response is dominant (Brown 1962). Apparently a f ish adjusts to the background colour by responding to the relative amounts of il lumination received from above and below, that i s , the ratio of incident to reflected light (Sumner and Keys 1929; Sumner and Doudoroff 1937; Sumner 1940). The retina is differentiated to a certain degree, the dorsal portion associated with the paling response, the ventral portion with darkening (Sumner 1933; DeGroot et a l . 1969; Iwata and Fukuda 1973). At very low light intensities (<1.75 foot-candles) f ish respond to il lumination regardless of background colour (Brown 1936); in complete darkness they become pale (Sumner 1940). The time required for transitory colour change is highly variable between species. For example, in Crenilabrus i t takes a few seconds; in Fundulus, 1 to 2 minutes; in Ameiurus, 1 to 3.5 hours; and in Angui11a, more than 20 days (Odiorne 1957). Nei l ! (1940) gave the general rule that i f the total time for transitory colour change is less than 10 minutes control is nervous; i f the time is more than 2 hours control is hormonal. Transitory colour change in response to light or background colour is fu l ly reversible (though paling and darkening responses may not occur in the same length of time (Neill 1940)), requiring only that the conditions which caused the change be reversed. Noticeable quantitative colour change can occur within 7-15 days, though i t may be 3-4 weeks before the change is complete (Odiorne 1933; Sumner and Wells 1933; Ahmad 1974). - 112 -There is some evidence that quantitative colour change may not always be entirely reversible, or at least that complete reversal may take a long time. Sumner and Wells (1933) found that guppies (Lebistes reticulatus) born and reared on a white background showed an extreme condition of depigmentation never attained by those which had been reared in a normal environment and then as adults subjected to a white background for 3 months (maximum exposure time in their experiment). Stickney and White (1975) observed that over a 3 month experimental period five of twenty moderately ambicolourate flounders (Paralichthys dentatus) entirely lost thei r ambi colouration; the rest retained i t to a l ight or moderate degree. (Ambicolouration refers to the development of pigmentation on the underside of the f i sh . ) Love (1974) held light and dark-coloured cod (Gadus morhua) from different grounds in the Atlantic together in an aquarium for 8 1/2 months. The f ish maintained their colour differences. Love did not know whether the colour difference was genetic or environmental, but thought that i t might be a response to background colour (l ight-coloured cod came from a bank where the substrate was br i l l iant white sand). The results of this study of colour change in lake whitefish may be summarized as follows -1. Larval and adult lake whitefish are capable of rapid transitory colour changes. As these occur within minutes they are probably under neural control. 2. When kept for an extended period under conditions which cause transitory darkening, larval whitefish undergo quantitative colour change (become more pigmented). 3. Larval whitefish respond to total darkness by blanching. - 113 -These results are a l l consistent with chromatic responses of other f ish species reported in the l i terature , as described above. There has been at least one other study on the chromatic behaviour of larval coregonids, that by Koller (1934). Duspiva (1931) reported that light intensity was the main stimulus for colour change for the larvae of Perca f l u v i a t i l u s , Salmo salvel inus, Abrami's brama and Leuciscus rut i lus . This prompted Kol ler (1934) to test the chromatic responses of larval Coregonus lavaretus and £ . holsatus. He found that, unlike those species investigated by Duspiva, both responded to background colour as opposed to light intensity. The colour of lake whitefish in SIL is partly under genetic control , but appears to have a large environmental component as wel l . The physical limnology of regions 4, 5 and 6 of SIL has been described above (Study Area and Distribution of Colour Classes of Whitefish in SIL). Fish respond chromatically to ambient l ight conditions and to the colour of their surroundings. Thus, the important characteristics of the water in the three basins are c lar i ty (measured by suspended sediment concentrations, vertical extinction coefficients and Secchi disc readings) and colour. The water in region 5, where there are many dark-coloured whitefish (and dark-coloured f ish of other species, for example, northern pike (Esox lucius) and walleye (Stizostedion vitreum)), is clearer and much darker in colour than the water in regions 4 and 6. A f ish in region 5, at shallow depths and swimming off the bottom, is essentially in dark surroundings with light from above. Presumably, in order to match its background i ts colour darkens. At greater depths, where light is at very low intensity or extinct, the chromatic response of the f ish would be to remain pale. - 114 -In regions 4 and 6, the water is turbid and light in colour. Fish blanch in response to l ight-coloured surroundings and under conditions of very reduced l ight , such as would be produced by high turbidity. Thus, one would expect the whitefish in these regions to be light in colour. Brown (1936) reported that f ish taken from the I l l ino is River are pale in colour when the water is siIty and dark when the water is c lear. He suggested that f ish swimming off the bottom in clear water are on the equivalent of a black background since almost no light is reflected from below, and that the s i l t in turbid water both reduces incident light and augments the reflected light entering the eyes of the f ishes, thus prompting the paling response. Knowing that f ish colour (lightness or darkness) is p last ic , the question of colour s tab i l i t y of SIL whitefish ar ises. In comparing the characteristics and habits of l ight and dark-coloured whitefish one makes an assumption about their phenotypic history - that a light f ish has always been light and a dark f ish has always been dark. Transitory colour change is completely reversible. Whitefish are capable of marked transitory colour change. Thus, for example, a l ight whitefish (one with a low scale melanaphore count) can appear either l ight or dark according to i ts surrounding at a given time. Theoretical ly, quantitative change should be reversible too, although in a much longer period of time. However evidence from studies cited above (Sumner and Wells 1933; Stickney and White 1975) suggest that i t may not be. Love (1974) observed that the colour differences between light and dark cod (Gadus morhua) were very stable. - 115 -Loss of melanophores may be temperature dependent. Odiorne (1933) found that k i l l i f i s h (Fundulus heteroclitus) kept in l ight surroundings at 10°C showed only a slight reduction in melanophores over several weeks. At 20°C melanophore degeneration set in quickly. In SIL temperatures at 5 m depth in region 5 in July 1982 averaged about 12.5°C, which may be too cool for signif icant melanophore loss to occur. According to the results of the long-term colour change experiment signif icant increases in melanophore numbers can occur at temperatures of 10°-11°C. If colour of whitefish in region 5 was total ly unstable and i f any f ish could be a quantitative l ight or dark'depending on conditions at the time, one might not expect to f ind any other differences between lights and darks. However, results of the comparison of morphological characteristics of adult l ight and dark whitefish showed that there are some morphometric and meristic differences between them. Also, the breeding experiment showed that colour is in part inherited. These results support the idea that colour of l ights and darks has some s tab i l i t y . Triaenophorus crassus Cyst Levels of SIL Whitefish The regional differences observed in lake whitefish cyst counts are correlated with differences in water depth and abundance of northern pike (Esox lucius) and cisco (Coregonus ar ted i i ) . Rawson (1947a, 1947b) reported that lake whitefish taken from shallow, near-shore areas of Great Slave Lake and Lake Athabasca had much higher X* crassus levels than did f ish from deeper, offshore waters. According to Lawler (1970) i t is a general rule that whitefish - 116 -f r o m shallow areas are more heavily infected than those from deeper water. In SIL, whitefish with the highest cyst levels came from region 5, which has a post-flooding mean depth of 5.9 m, while those with the lowest levels came from region 4, which has a post-flooding mean depth of 13.0 m. Region 6, where whitefish in experimental catches had intermediate mean cyst levels , has a post-flooding mean depth of 5.8 m (McCullough 1981). The northern pike is the adult host and the C i s c o is the preferred intermediate host of J . crassus. Catch per unit of effort s tat is t ics from 1982 (Bodaly et a l . 1983) show that pike are twice as abundant and cisco almost four times so in region 5 as in region 4. There are more pike but fewer C i s c o in region 6 than in region 4. Data from experimental f ishing show that dark whitefish generally have higher mean cyst counts than do light whitefish. There does not, however, appear to be a causal relationship between h i g h cyst count and dark colouration, for several reasons. The modal cyst count is zero for both l ight and dark whitefish and approximately three-quarters of a l l whitefish have counts of 0, 1 or 2 cysts, regardless of colour (Fig. 19). According to MacLaren (1978), in 1969 SIL was zoned into three areas based on whitefish grade as determined by cyst levels. Fish from region 6 were classed as "cutters", the lowest grade. Fish from region 4 north of Missi Rapids and region 5 were classed as "continental" or second grade. Distribution data shows that region 6 whitefish are a l l l ight in colour, as are many of those from the northern half of region 4 (Fig. 9). Furthermore, the results of this study have shown that differences in external colouration between lights and darks from SIL are partly genetically determined and partly environmentally induced. - 117 -If individual level of infection is not causally related to colour the question concerning the basis of the average correlation remains. The regional differences in cyst counts among lights suggest that mean cyst levels may be related to conditions in the particular region. Certain characteristics of region 5 are known correlates of higher cyst levels - shallow water and large pike and Cisco populations. Limnological conditions in region 5 promote dark colouration of f ish -shallow, clear, dark-coloured water. F ina l l y , the habits of dark whitefish probably make them more susceptible to infection - they prefer onshore, shallower areas, they are s l ight ly more pelagic in distr ibution than are l ights , and they have substantially more pelagic-type food items in their diets than do l ights . Whitefish do not generally accumulate 1_. crassus cysts with increasing age (past 2-3 years) because of dietary changes from planktonic to benthic (Mil ler 1952, cited in Watson (1977)). Because of th is , mean cyst count generally decreases in whitefish samples as mean size increases (MacLaren 1978). However, Petersson (1971) found that high raker whitefish (Coregonus peled and C_. oxyrhynchus) with a high frequency of plankton in the diet did accumulate X* crassus plerocercoids with age and were more heavily infected than were low raker f ish which fed mainly on the bottom. Bodaly et a l . (1984) calculated mean cyst count per pound of whitefish of round weight >1.8 lb (commercial size) captured in experimental catches in regions 4 and 5, 1978-1982 (Table 19). Mean cyst count per pound is generally low, except for the f ish captured in region 5 in 1982. A l l f ish included in this subsample were c lass i f ied as darks. The mean weight of f ish in the subsample (2.28 lb , N = 62) is higher than that for the total catch of - 118 -Table 19. Triaenophorus crassus cyst counts in experimental catches of lake whitefish, regions 4 and 5, Southern Indian Lake, 1978-1982. Fish of round weight >1.8 lbs only sampled and catches of total round weight approximately 14.5 lbs only considered (Bodaly et a l . 1984 ). Number of f ish sampled in brackets Year Region Mean cysts per lb Mean weight (lb) 1978 4 5 0.07 (64) 0.48 (20) 2.15 2.20 1979 4 5 0.32 (93) 0.46 (92) 2.11 2.39 1982 4 5 0.36 (78) 2.04 (62) 2.31 2.28 - 119 -darks in region 5 in 1982 (1.47 lb, N = 146). However the mean cyst counts per lb of the subsample and the total catch are almost equal (2.04 vs 2.16, respectively). Thus the inverse relationship between age/size and cyst count does not necessarily hold for dark whitefish. This could be because darks continue to be somewhat pelagic in,habit as they get older and thus continue to accumulate X* crassus cysts. Functional Significance and Adaptive Value of Colour Chromatic responses in animals may serve in several capacities -protective colouration, thermoregulation, mating displays and parental care, protection of the body from harmful il lumination (Fingerman 1965), communication of psychic state (for example, f r ight , aggression) to conspecifics (Lanzing and Bower 1974). Probably the most widespread function of colour change in f ish is that of concealment. There are numerous descriptions of crypsis in f i s h , but few test of the theory. A f ish may conceal i t se l f by mimicking substrate colour or pattern (Jenkins, Jr . 1969; Lanzing 1977), vegetation conspecifics (McFarland et a l . 1979), prey (Kaufman 1976), or by adjusting to ambient l ight conditions in deep water (Badcock 1969). These examples involve transitory colour changes in which a f ish may "select" from a number of available patterns that which is most appropriate to conditions at a given time. Sumner (1935) presented experimental evidence for the protective value of changeable colouration in f i sh . He exposed black and white-adapted mosquito f ish (Gambusia patruelis) in black or white tanks - 120 -to piscivorous birds, both diving and wading species. Those f ish which matched the background were less vulnerable to predation than were those which contrasted with i t . Colouration of whitefish in SIL may be an adaptation for concealment. In shallow, clear, dark-coloured water (region 5) a dark-coloured f ish would presumably be less v i s ib le , thus less vulnerable to predators. - 121 -CONCLUSION There are quantifiable differences in melanophore numbers to support the subjective c lass i f icat ion of SIL whitefish into lights and darks. Mean scale melanophore counts of f ish subjectively c lass i f ied as l ights and darks were s ignif icant ly different. Lights are quantitatively defined as having scale melanophore counts <80 melanophores-2 mm2 while darks have counts >8° melanophores«2 mm2. There are differences in spatial distr ibution of l ights and darks within SIL. Lights are found throughout the lake and predominate in regions 3, 4 and 6; darks occur mainly in region 5. Within region 5, l ights were most abundant offshore and on the lake bottom, darks were more numerous onshore than offshore and were more often caught off the bottom. Young-of-the-year from region 5 were darker in colour than those from regions 4 and 6. There were signif icant differences between lights and darks in several morphometric characters and in lower g i l l raker numbers. There were no signif icant differences between them in two biochemical characters (Hb and G-3-PDH) examined. The evidence does not suggest clear-cut separation into reproductively isolated stocks. There is some hereditary basis to colour differences between l ights and darks such that dark parents produce darker offspring than do light parents. Her i tabi l i ty estimates show that there is a genetic component to colour differences which is independent of environmental or maternal ef fects . At 111 days post - fe r t i l i za t ion h 2 measured by the regression of offspring on male parent value was comparable to estimates for reproductive t ra i ts l ike growth rate and l i t t e r s ize . - 122 -Colour of larval lake whitefish is subject to environmental al terat ion. Rapid short-term colour changes are effected through redistribution of melanin granules in the melanophores. Over the long-term colour is altered through changes in melanophore number. Adult whitefish are also capable of rapid short-term colour change. Colour differences between SIL whitefish are correlated with infection by X* crassus cysts in that darks tend to have higher average cyst levels than l ights . However, as both lights and darks have modes of 0 cysts and as there are regional differences in cyst levels of l ights , there does not appear to be a causal connection between cyst count and pigmentation. Darks, because of - the i r distr ibution patterns and dietary preferences, may be more susceptible to infection than l ights . - 123 -REFERENCES Ahmad, R.U. 1972. Quantitative ("morphological") colour changes on prolonged black & white background adaptations of the minnow Phoxinus phoxinus (L.) & the effects of chromatic spinal section on these colour changes. J . Comp. Physiol. 77: 170-189. Ahmad, R.U. 1974. Transitory colour changes in the new melanophores formed by quantitative colour change in the minnow Phoxinus phoxinus. J . Zool. Lond. 174: 285-313. Ayles, G.B. 1974. Relative importance of additive genetic and maternal sources of variation in early survival of young splake hybrids (Salvelinus font inal is x S. namaycush). J . Fish. Res. Board Can. 31: 1499-1502. Ayles, H.A. 1976. Lake whitefish (Coregonus clupeaformis (Mitchi l l ) ) in Southern Indian Lake, Manitoba. Tech. Rep. F ish. Mar. Serv. Res. Dev. 640. 34 pp. Badcock, J . 1969. Colour variation in two mesopelagic fishes and i ts correlation with ambient l ight conditions. Nature 221: 383-385. Bodaly, R.A. 1977. Evolutionary divergence between currently sympatric lake whitefish, Coregonus clupeaformis, in the Yukon Territory. Ph.D. thesis , University of Manitoba. 119 pp. Bodaly, R.A. 1979. Morphological and ecological divergence within the lake whitefish (Coregonus clupeaformis) species complex in Yukon Territory. J . Fish. Res. Board Can. 36: 1214-1222. Bodaly, R.A., C.J.Hrenchuk, and R.J.P. Fudge. 1983. Post-impoundment experimental f ish ing, Southern Indian Lake, Manitoba: Catches (1982) and Triaenophorus crassus cyst counts in lake whitefish (1978, 1979, 1982). Can. Data Rep. F ish. Aquat. Sc i . 401: vi +75 pp. - 124 -Bodaly, R.A., T.W.D. Johnson, and R.J. Fudge. 1980. Post-impoundment changes in commercial f ishing patterns and catch of lake whitefish (Coregonus clupeaformis) in Southern Indian Lake, Manitoba. Can. MS Rep. F ish . Aquat. Sc i . 1555: iv + 14 p. Bodaly, R.A., T.W.D. Johnson, R.J .P. Fudge, and J.W. Clayton. 1984. Collapse of the lake whitefish fishery in Southern Indian Lake, Manitoba, following lake impoundment and river diversion. Can. J . F ish. Aquat. S c i . 41: 692-700. Bodaly, R.A., D.M. Rosenberg, M.N. Gaboury, R.E. Hecky, R.W. Newbury, and K. Patalas. In press. Ecological effects of hydroelectric development in northern Manitoba, Canada: The Churchill-Nelson River diversion. J_n_ Sheehan, P . J . , D.R. Mi l le r , G.C. Butler and Ph. Bourdeau (eds.) Effects of Pollutants at the Ecosystem Level, SCOPE 22. John Wiley & Sons, Chichester, New York, Brisbane, Toronto. Brown, F.A. , J r . 1936. Light intensity and melanophore response in the minnow, Ericymba buccata Cope. B i o l . Bu l l . Mar. B io l . Lab. 70: 8-15. Brown, F .A. , J r . 1962. Chromatophores and color change, p. 502-537. In C.L. Prosser and F.A. Brown, J r . Comparative Animal Physiology. W.B. Saunders Co. , Philadelphia. Budd, J . 1956. Movements of tagged whitefish in northern Lake Huron and Georgian Bay. Trans. Am. Fish. Soc. 86: 128-134. Clark, F.H. 1970. Pleiotropic effects of the gene for golden color in rainbow trout. J . Hered. 61: 8-10. - 125 -Clayton, J.W., W.G. Franzin, and D.N. Tretiak. 1973. Genetics of glycerol-3-phosphate dehydrogenase isozymes in white muscle of lake whitefish (Coregonus clupeaformis). J . F ish. Res. Board Can. 30: 187-193. Cunningham, J . T . , and C A . MacMunn. 1893. On the coloration of the skins of f ishes, especially of Pleuronectidae. P h i l . Trans. Royal Soc. Lond. Ser. B. B i o l . Sc i . 184: 765-812. Duspiva, F. 1931. Beitrage sur Physilogie der Melanophoren von Fischembryonen. Akad. Wiss. Wien, Math.-naturw. k l , Sitzunber. Abt. 1, 140: 553. Falconer, D.S. 1960. Introduction to Quantitative Genetics. The Ronald Press Co. New York. 365 pp. Fingerman, M. 1965. Chromatophores. Physiol. Rev. 45: 296-339. Fox, D.L. 1957. The pigments of f ishes, p. 367-385. In_ M.E. Brown (ed.) The Physiology of Fishes, Vol . II Behavior Academic Press Inc., New York. 526 p. von Fr isch, K. 1911. Beitrage zur Physiologie der Pigmentzellen in der Fischhaut. Pflugers Arch Ges Physiol. 138: 319-387. F u j i i , R. 1970. Chromatophores and pigments, p. 307-353. In W.S. Hoar and D.J. Randall (ed.) Fish Physiology. Vol . III. Academic Press, New York. F u j i i , R. 1973. Responses of f ish pigment cel ls to environmental changes, p. 342-362. _In_ W. Chavin (ed.) Responses of f ish to environmental changes. Proceedings of a United States-Japan seminar. Tokyo, Japan. Nov. 23-26, 1970. Charles C. Thomas, Springf ield, ILL., U.S.A. Gjedrem, T. 1975. Poss ib i l i t ies for genetic gain in salmonids. Aquaculture 6: 23-29. - 126 -Graves, J . E . , and R.H. Rosenblatt. 1980. Genetic relationships of the color morphs of the serranid f ish Hypoplectrus unicolor. Evol. 34(2): 240-245. deGroot, S . J . , R. Norde, and F .J . Verheijen. 1969. Retinal stimulation and pattern formation in the common sole Solea solea (L.) (Pisces: Soleidae). Neth. J . Sea Res. 4(3): 339-349. Hafeez, M.A. 1970. Effect of melatonin on body coloration and spontaneous swimming act iv i ty in rainbow trout, Salmo gairner i . Comp. Biochem. Physiol. 36: 639-656. Hecky, R.E. 1984. The thermal and optical characteristics of Southern Indian Lake before, during and after impoundment and Churchill River diversion. Can. J . F ish. Aquat. S c i . 41: 579-590. Hecky, R .E . , and S . J . Guildford. 1980. The primary productivity of Southern Indian Lake before, during and after impoundment and r iver diversion. Can. J . F ish. Aquat. S c i . 41: 591-604. Hecky, R .E . , and G.K. McCullough. 1984. The effect of impoundment on the sediment budget and nearshore sedimentation of Southern Indian Lake. Can. J . F ish. Aquat. S c i . 41: 567-578. Hubbs, C .L . , and K.F. Lagler. 1958. Fishes of the Great Lakes Region. University of Michigan Press, Ann Arbor, Michigan. 213 pp. Imhof, M.A. 1977. Population genetic structure of lake whitefish, (Coregonus clupeaformis) in Green Bay and northern Lake Michigan, as assessed by electrophoresis of lactate, glycerol-3-phosphate, and malate dehydrogenase isoenzymes. M.Sc. thesis , University of Wisconsin, Stevens Point, WI. 44pp. Itzkovich, J . , S. Rothbard, and G. Hulata. 1981. Inheritance of pink body colouration in Cichlasoma nigrofasciatum Gunther (Pisces, Cichl idae). Genetica 55(1): 15-16. (Abstract only seen) - 127 -Iwata, K.S. , and H. Fukuda. 1973. Central control of color changes in f i s h , p. 316-341. h i W» Chavin (ed.) Responses of Fish to Environmental Changes. Proceedings of a United States-Japan seminar. Tokyo, Japan. Nov. 23-26, 1970. Charles C. Thomas, Spr ingf ield, ILL., U.S.A. Jenkins, T .M. , J r . 1969. Observations of colour changes of brown and rainbow trout (Salma trutta and S. gai rdneri) in stream habitats, with description of an unusual color pattern in brown trout. Trans. Am. Fish. Soc. 98: 517-519. Kaufman, L. 1976. Feeding behaviour and functional coloration of the Atlantic trumpet f i s h , Aulostomus maculatus. Copeia 1976(2); 377-378. Kincaid, H.L. 1975. Iridescent metall ic blue color variant in rainbow trout. J . Hered. 66: 100-101. Kliewer, E.V. 1970. G i l l raker variation and diet in lake whitefish Coregonus clupeaformis in northern Manitoba, p. 147-165. In C.C. Lindsey and C.S. Woods (ed.). Biology of Coregonid Fishes. University of Manitoba Press, Winnipeg, Manitoba. 560 pp. Klupp, R., and F. Kaufman. 1979. The heredity of colour in rainbow trout. F isch. Teichwirt. 30(2): 19-20. (Abstract only seen). Kol ler , G. 1934. Uber den Farbwechsel von Coregonenlarven. B io l . Zent ra lb l . , 54: 419. Kozikowska, Z. 1961. Influence of the habitat on the morphology and biology of f i s h . Small whitefish, perch, selected elements. Ekol. Polsk. A, 9(27): 541-678. Krasnodembskaya, K.D. 1978. Formation of pigment reactions and thei r regulatory mechanism in chondrostean f ishes. J . Ichthyol. 18(1): 120-131. - 128 -Kristofferson, A.H. 1978. Evidence for the existence of subpopulations of lake whitefish, Coregonus clupeaformis (Mitchil l ) in Lake Winnipeg. M.Sc. Thesis. University of Manitoba, Winnipeg, 94 pp. Lanzing, W.J.R. 1977. Reassessment of chromatophore pattern regulation in two species of f l a t f i sh (Scophthalmus maximus; Pleuronectes platessa). Neth. J . Sea Res. 11(2): 213-222. Lanzing, W.J.R., and C C . Bower. 1974. Development of colour patterns in relation to behaviour in T i lapia mossambica (Peters). J . F ish. B i o l . 6: 29-41. Lawler, G.H. 1970. Parasites of coregonid f ishes, p 279-309. In C C . Lindsey and C S . Woods (ed.). Biology of Coregonid Fishes. University of Manitoba Press, Winnipeg, Manitoba. 560 pp. Lindsey, C C 1962. Distinctions between the broad whitefish, Coregonus nasus, and other North American whitefishes. J . F ish . Res. Board Can. 19: 687-714. Lindsey, C C 1963. Sympatric occurrence of two species of humpback whitefish in Squanga Lake, Yukon Territory. J . F ish. Res. Board Can. 20: 749-767. Lindsey, C C 1981. Stocks are chameleons: p last ic i ty in g i l l rakers of coregonid f ishes. Can. J . F ish. Aquat. Sc i . 38: 1497-1506. Loch, J .S . 1974. Phenotypic variation in the lake whitefish, Coregonus clupeaformis, induced by introduction into a new environment. J . F ish. Res. Board Can. 31: 55-62. Love, R.M. 1974. Colour s tab i l i t y in cod (Gadus morhua L.) from different grounds. J . Cons. Int. Explor. Mer 35(2): 207-209. McLaren, J . F . , Ltd. 1978. Report to Manitoba Hydro on changes in quality of whitefish fishery on Southern Indian Lake. May 1978. - 129 -McCullough, G.K. 1981. Water budgets for Southern Indian Lake, before and after impoundment and Churchill River diversion, 1972-79. Can. MS. Rep. Fish. Aquat. Sc i . 1620, Winnipeg, 22 p. McFarland, W.N., J .C . Ogden and J .N. Lythgoe. 1979. The influence of l ight on the twilight migration of grunts. Env. B io l . Fish. 4(1): 9-22. McTavish, W.B. 1952. A biological survey of Southern Indian Lake. Report for the Manitoba Department of Mines and Natural Resources. M i l l e r , R.B. 1952. A review of the Triaenophorus problem in Canadian lakes. B u l l . F ish. Res. Board Can. No. 95. Nayudu, P.L. 1979. Genetic studies of melanic color patterns and atypical sex determination in the guppy, Poecil ia ret iculata. Copeia 1979 (2): 225-231. N e i l l , R.M. 1940. On the existence of two types of chromatic behaviour in teleostean f ishes. J . Exp. B io l . 17: 74-98. Newbury, R.W., and G.K. McCullough. 1984. Shoreline erosion and restabi l ization in Southern Indian Lake reservoir. Can. J . Fish. Aquat. S c i . 41: 558-565. Newbury, R.W., G.K. McCullough, and R.E. Hecky. 1984. The Southern Indian Lake impoundment and Churchill River diversion. Can. J . F ish. Aquat. Sc i . 41: 548-557. Odiorne, J .M. 1933. Degeneration of melanophores in Fundulus. Proc. Nat. Acad. Sc i . Wash. 19: 329-332. Odiorne, J .M. 1967. Color changes, p. 387-401. J_n M.E. Brown (ed.) The Physiology of Fishes, Vol . II Behavior. Academic Press Inc., New York. 526 pp. - 130 -Osborn, C M . 1941a. Studies on the growth of integumentary pigment in the lower vertebrates. I. The origin of a r t i f i c i a l l y developed melanophores on the normally unpigmented vertral surface of the summer flounder (Paralichthys dentatus). B i o l . B u l l . 81: 352-363. Osborn, C M . 1941b. Studies on the growth of integumentary pigment in the lower vertebrates. II. The role of the hypophysis in melanogenesis in the common catfish (Amieurus melas). B io l . Bu l l . 81: 352-363. Parker, G.H. 1943. Animal color changes and their neurohumours. Quart. Rev. B i o l . 18: 205-227. Patalas, K., and A. Sa lk i . 1984. Effect of impoundment and diversion on the crustacean plankton of Southern Indian Lake. Can. J . F ish. Aquat. Sc i . 41: 613-637. Petersson, A. 1971. The effects of lake regulation on populations of Cestodan parasites of Swedish whitefish Coregonus. Oikos 22: 74-83. Qadri, S.U. 1968. Growth and reproduction of the lake whitefish, Coregonus clupeaformis, in Lac la Ronge, Saskatchewan. J . F ish . Res. Board Can. 25(10): 2091-2100. Rawson, D.S. 1947a. Great Slave Lake, p. 45-68. ]m North West Canadian Fisheries Surveys in 1944-1945. F ish. Res. Board Can. B u l l . No. 72. Rawson, D.S. 1947b. Lake Athabasca, p. 69-85. h± North West Canadian Fisheries in 1944-45. F ish. Res. Board Can. B u l l . No. 72. Robertson, O.H. 1951. Factors influencing the state of dispersion of the dermal melanophores in rainbow trout. Physiol. Zool. 24: 309-323. - 131 -Scott, G.T. 1965. Physiology and pharmacology of color change in the sand flounder Scopthalmus aquosus. Limnol. Oceanogr. 10: R230-R246. S6c6rov, S. 1909. Farbenwechselversuche an der Bartgrundel (Nemachilus barbatula L.) . Arch. EntwMech. Org. 28: 629-660. Smith, J . R . , and L .J . Weber. 1976. Alterations in diurnal pineal hydroxyindole-o-methyltransferase (HIOMT) act iv i ty in steelhead trout (Salmo gairdneri) associated with changes in environmental background color. Comp. Biochem. Physiol. 53C:33-35. Stickney, R.R., and D.B. White. 1975. Ambicol oration in tank cultured flounder, Paralichthys dentatus. Trans. Am. Fish. Soc. 104(1): 158-160. Sumner, F.B. 1933. The di f fer ing effects of different parts of the visual f i e ld upon the chromatophone responses of f ishes. B i o l . Bu l l . 65: 266-282. Sumner, F.B. 1935. Evidence for the protective value of changeable coloration in f ishes. Am. Nat. 69: 245-266. Sumner, F.B. 1939. Quantitative effects of visual stimuli upon pigmentation. Am. Nat. 73: 219-234. Sumner, F.B. 1940. Quantitative changes in pigmentation, resulting from visual stimuli in fishes and Amphibia. B i o l . Revs. Cambridge Ph i l . Soc. 15: 351-375. Sumner, F .B . , and P. Doudoroff. 1937. Some quantitative relations between visual stimuli and the production or destruction of melanin in f ishes. Proc. Nat. Acad. S c i . , Wash. 23: 211-219. Sumner, F .B . , and A.B. Keys. 1929. The effects of differences in the apparent source of illumination upon the shade assumed by a f l a t f i s h on a given background. Physiol. Zool. 2: 495-504. - 132 -Sumner, F .B . , and N.A. Wells. 1933. The effects of optic stimuli upon the formation and destruction of melanin pigment in f ishes. J . Exp. Zool. 64(3): 377-403. Sunde, L.A. 1963. South Indian Lake whitefish infestation test , March 5 to 7, 1963. Manitoba Department of Mines and Natural Resources, Fisheries Branch MS Report. Svardson, G. 1952. The coregonid problem. IV. The significance of scale and g i l l - rakers . Rep. Inst. Fw. Res. Drott. 33: 204-232. Svardson, G. 1965. The coregonid problem. VII. The isolat ing mechanisms in sympatric species. Rep. Inst. Fw. Res. Drott. No. 46. Svardson, G. 1970. Significance of introgression in coregonid evolution, p. 33-59. J_n C.C. Lindsey and C.S. Woods (ed.). Biology of Coregonid f ishes. University of Manitoba Press, Winnipeg, Manitoba. 560 pp. Takeda, K., S. Onodera, and K. Yoshiyasu. 1978. Salvelinus leucomaenis with atypical color patterns, collected in the Mogami and Nikko river systems, Yamagata Prefecture. Jap. J . Ichthyol. 25(1): 58-64. (Abstract only seen.) Tsuyuki, H., E. Roberts, R.H. Kerr, and A.P. Ronald. 1966. Micro starch gel electrophoresis. J . F ish. Res. Board Can. 23: 929-933. Watson, R.A. 1977. Metazoan parasites from whitefish, cisco and pike from Southern Indian Lake, Manitoba: a preimpoundment and diversion analysis. M.Sc. Thesis, University of Manitoba, xi + 197 pp. Weins, A . P . , and D.M. Rosenberg. 1984. The effect of impoundment and diversion of profundal macrobenthos of Southern Indian Lake, Canada. Can. J . Fish. Aquat. Sc i . 41: 638-648.