UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Colour pigments in the penpoint gunnel Apodichthys flavidus and their ecological significance Wilkie, Donald Walter 1966

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

Item Metadata

Download

Media
831-UBC_1966_A6_7 W54.pdf [ 5.55MB ]
Metadata
JSON: 831-1.0104805.json
JSON-LD: 831-1.0104805-ld.json
RDF/XML (Pretty): 831-1.0104805-rdf.xml
RDF/JSON: 831-1.0104805-rdf.json
Turtle: 831-1.0104805-turtle.txt
N-Triples: 831-1.0104805-rdf-ntriples.txt
Original Record: 831-1.0104805-source.json
Full Text
831-1.0104805-fulltext.txt
Citation
831-1.0104805.ris

Full Text

COLOUR PIGMENTS IN THE PENPOINT GUNNEL APODICHTHYS FLAVIDUS AND THEIR ECOLOGICAL SIGNIFICANCE by DONALD WALTER WILKIE B.A., University of Br i t i sh Columbia, I960 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of ZOOLOGY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July, 1966 In presenting t h i s thesis i n p a r t i a l f u lfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission., for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by his representatives„ I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my wri t t e n permission. Department of ~Zoo<~o<^Y The Uni v e r s i t y of B r i t i s h Columbia Vancouver 8, Canada 1*1 &c ABSTRACT Field and laboratory studies were undertaken to examine the ecological role of colouration in the penpoint gunnel Apodichthys  flavidus, i ts structural basis and possible origin. A. flavidus was found to vary in colour from green through brown to red. The vast majority of f i sh collected matched at least one type of vegetation from their habitat. Those observed directly within vegetation were of the same colour as the vegetation. In habitat selection experiments A. flavidus was found to prefer cover under rocks to that within vegetation, but when provided with vegetation alone chose that which i t matched. The colour phases observed in A. flavidus were found to be de-termined directly by the pigments they contained not by differences in stages of chromatophore expansion. Green f i sh owe their colour p r i -marily to ester i f ied dihydroxy e carotene conjugated with a protein and dispersed throughout the integument. The colour of red f i sh results primarily from esters of astaxanthin contained in erythrophores. Brown f ish incorporate the colouration systems of both the red and green phases, but the modifications involved have not been fu l ly worked out. Colour change experiments showed that A. flavidus cannot undergo complete changes of colour phase in response to environment alone. Diet has an influence on colour, but complete colour changes were not produced experimentally. Larvae were reared from the eggs of green and brown individuals. A l l developed colouration more similar to that of the Artemia upon i i i which they were fed than to their parental type. This evidence is discussed in terms of a possible dietary origin of colour variation and weighed against polymorphism. It is suggested that the colouration of A. flavidus has a cryptic function which is of importance primarily during food seeking. It is hypothesized that the vegetation upon which A. flavidus larvae settle in conjunction with early diet primarily determines the colouration of individuals. iv TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS iv LIST OF TABLES v i i LIST OF FIGURES x i i ACKNOWLEDGEMENTS xiv INTRODUCTION 1 FIELD OBSERVATIONS 3 Methods and materials 3 Colour notation 4 Descriptions of stations 6 Results and discussion 9 Correlation between f ish colour and habitat . . . 10 Direct observations 13 Seasonal abundance 1^  Vert ical distribution 15 Diurnal act iv ity 15 Form of vegetation . 15 Summary 19 HABITAT SELECTION EXPERIMENTS 20 Experiment 1 - Reflected light 2 0 Experiment 2 - Transmitted light 22 Experiment 3 - Algae and rocks 25 Experiment 4 - Algae 29 General conclusions 30 V Page DESCRIPTION OF SKIN 32 DESCRIPTION OF PIGMENTS 36 Green phase 36 Red phase 44 Brown phase . . . . . . . . . . ^7 FIRST STAGE COLOUR CHANGE EXPERIMENTS 51 Conclusions 52 SECOND STAGE COLOUR CHANGE EXPERIMENTS 5 3 Methods and material 53 Results 59 Discussion ^0 Red phase 6 0 Green phase ^ Brown phase D Conclusions ^5 REARING EXPERIMENTS 6 7 Methods and material ^7 Results 6 8 Discussion 74 Summary 76 FINAL DISCUSSION 78 SUMMARY 8 2 LITERATURE CITED 8 5 APPENDIX 8 7 8 7 Part I - Description of stations ' v i Page Part II - Tables of field collections 104 Part III - Results of colour change experiments . . . . 122 v i i LIST OF TABLES Page 1. Gross comparison of principal collecting stations . . 8 2. Chi-square analysis of the colour frequencies of A. flavidus at the three regular collecting stations 12 3. Analysis of first choices made by A. flavidus in reflected light experiment 21 4. First choices made by A. flavidus in transmitted light experiment 23 5. Observations of green A. flavidus kept in tank with transmitted green light in one end and transmitted red light in the other 24 6. First choices of habitat made by A. flavidus when given alternatives of rocks covered with Ulva (green) and Porphyra (red) 26 7. Chi-square analyses of 10 day habitat observations in rock and algae experiment. Analysis arranged to test for selection of matching algae 27 8. Chi-square analyses of 10 day habitat observations in rock and algae experiment. Analysis arranged to test the possibility that space under one of the rocks in each tank was preferred 29 9. Distribution of A. flavidus in an aquarium containing bunches of red and green algae 30 v i i i Page 10. Comparison of a xanthophyll suspected of being 3,3' dihydroxy e carotene with samples of known carotenoids (adapted from Crozier and Wilkie, 1966) 42 11. Summary of colour change experiments indicating general trends 61 12. Development of pigmentation in reared A. flavidus . . . 70 13. Description of skin of reared A. flavidus at the last development stage attained (dark orange, 25-29 mm T.L.). Dispersion indexes are adapted from Hogben and Slome, (1931) 72 14. Description of skin from side of captured (wild) A. flavidus 20-25 mm T.L. Dispersion indexes are adapted from Hogben and Slome, (1931) . 75 ix LIST OF APPENDIX TABLES Part II - F ield collections Page 1. Lumberman's Arch seine collections of f ish greater than 50 mm T.L. . . , 105 2. Lumberman's Arch collections of f i sh less than 50 mm T.L 108 3. Second Narrows shore collections 109 4. Second Narrows seine and SCUBA collections of A. flavidus I l l 5. Jordan River seine collections of f ish greater than 50 mm T.L H2 6. Agate Beach seine collections 113 7. Agate Beach Rotenone collections 116 8. Muir Point collections 117 9. Harling Point collections of A. flavidus 118 10. West Vancouver and Howe Sound Rotenone collections . . ng 11. Summary of Lumberman's Arch SCUBA diving observations . 120 12. Jordan River collections of A. flavidus under 50 mm . . 121 Part III - Results of colour change experiments 13. Results of holding A. flavidus under white daylight over a white background. Colour notation in this and subsequent tables are from Munsell, 1929 123 14. Results of holding A. flavidus under white daylight over a black gravel background . . . . . . . . . . . . 124 Page 15. Results of holding A. flavidus under white daylight over a neutral gray background [N/6(MunselD ] 126 16. Gross histology and pigment analysis of f i s h from neutral gray background 128 17. Results of holding A. flavidus under white daylight over green gravel background, [5G6/6 (Munsell)] . . . . 129 18. Results of holding A. flavidus i n white daylight over a red gravel background [5R5/12 (Munsell)] . . . . 130 19. Results of holding A. flavidus i n darkness 131 20. Results of holding A. flavidus under transmitted green l i g h t 132 21. Gross histology and pigment analysis of f i s h from transmitted green l i g h t experiment 134 22. Results of holding A. flavidus under transmitted red l i g h t 135 23- Gross histology and pigment analysis of f i s h from transmitted red l i g h t experiment 137 24. Results of holding A. flavidus over neutral gray background and feeding them red gammarids containing "astaxanthin." •' 138 25. Results of holding A. flavidus i n brown algae under white daylight 139 xi Page 26. Gross h i s to logy and pigment ana lys i s of f i s h with undetermined h i s t o r y from ho ld ing tanks and of C a l i f o r n i a specimens 140 27. Results of ho ld ing A. f l a v i d u s in red algae under white day l igh t 141 28. Gross h i s t o l o g y and pigment ana l ys i s of f i s h from red algae experiment 143 x i i LIST OF FIGURES Page 1. Apodichthys flavidus collection stations 7 2. A brown A. flavidus with a piece of Laminaria 11 3. A green A. flavidus shown with Ulva 11 4. Length-frequency plot of brown and green A. flavidus from Lumberman's Arch 16 5. Length-frequency plot of A. flavidus from Jordan River . 18 6. Top view of apparatus used in transmitted light preference experiment 22 7. Melanophore index (from Hogben and Slome, 1931) . . . . 3 2 8. Analysis of skin pigment from green A. flavidus (56Y5/8 Munsell) 37 9. Absorption spectra of pigment from green A. flavidus showing the crude pigment extract in comparison with fraction 2 following chromatography 38 10. Absorption spectrum of pigment fraction 1 from a green A. flavidus compared with the spectrum of ft carotene . . 4 0 11. Analysis of pigment from red A. flavidus (5R3/8 Munsell) 45 12. Absorption spectrum of principal pigment fraction from red A. flavidus 46 13. Analysis of skin pigment from a red-brown A. flavidus (2.5YR3/4 Munsell) 48 x i i i LIST OF APPENDIX FIGURES Page 1. Profile of Lumberman's Arch collecting station M-10 showing general floral zonation in seining area during the summer of 1962 90 2. Profile of Second Narrows collecting station M-12 illustrating zonation 92 3. Profile near centre of collecting area at Jordan River, Station VI-5 • 96 4. Profile of Agate Beach collecting station VI-8 100 xiv ACKNOWLEDGEMENTS To the many people who helped in so many ways i t is d i f f i cu l t to give adequate thanks. Drs. John Mclnerney, Kenneth Stewart and Michael Smith, Messrs. Terry McLeod, Evan Fagan, Vince Penfold, Robert Kiwala, John Rawle and many graduate students provided invaluable assistance with f i e ld col lections. Dr. Earl Herald through Walt Schneebli and Dave Powell made specimens available from the San Francisco area. Drs. Dennis L. Fox and Michael Smithy MessrsGeorge Crozier.and Tom Hopkins provided advice and assistance with the pigment analyses. Dr. Tony Perks read and offered valuable cr i t ic ism of the pigment section of the manuscript. Dr. Murray A. Newman provided laboratory space and collecting equipment and on many occasions the generous assistance of his staff. Dr. C. C. Lindsey brought the problem to my attention and acted as supervisor. Although most of the work was done while we were not in contact, I always fe l t his influence upon my undertakings. Others who acted on the committee were Drs. Paul Dehnel, P. A. Larkin, T. G. Northcote, R. F. Scagel, G. G. E. Scudder and N. J . Wilimovsky. Dr. R. H. Rosenblatt kindly read the manuscript and offered many helpful suggestions. Particular thanks are given to Drs. T. G. Northcote and R. J . Krejsa who provided assistance in many ways as well as constant encouragement, and to my wife Beth who has lived through i t a l l with patience and affection. Mrs. Lorayne D. Buck typed the manuscript with utmost competence. To a l l of you: my deepest thanks. 1 INTRODUCTION Since the publication of Cott's (1940) book i t has been generally accepted that colouration can have several adaptive roles. These may include concealment, mimicry, advertisement, courtship display and protection from solar radiation. The diverse array of colours and patterns which animals display results from an equally diverse number of pigments and integumentary structures. Many animals are capable of concentrating or dispersing the pigments within c e l l s or organs i n order to effect colour changes and further increase adaptability. Among the vertebrates, fishes are unexcelled i n the u t i l i z a t i o n of colour. This study was undertaken to examine the role of colouration i n the penpoint gunnel Apodichthys flavidus Girard 1854. This i s a species common along the shores of B r i t i s h Columbia and known to occur from Alaska to Point Conception (Clemens and Wilby, 1961). I t i s an e e l - l i k e blennioid f i s h belonging to the family Pholidae. I t s body i s elongate and l a t e r a l l y compressed. A maximum length of 18 inches may be attained, although none approaching t h i s size was found in the present study. The diet consists c h i e f l y of small crustaceans and molluscs. A wide variety i n colouration i s reported by Clemens and Wilby (1961), including green, yellow, orange, brown, and red. The green and yellow-green phases blend well with the eel-grass habitat. Jordan (1925) remarked on the colouration of A. flavidus, stating that i t s colouration apparently depends upon the colour of vegetation i t inhabits, but he cited no evidence. A s i m i l a r l y undocumented 2 statement i s made by Bigelow and Schroeder (1953) regarding the colouration of the closely related Pholis gunnellus. They state that the ground t i n t of the f i s h matches the vegetation or bottom. The remarks of the above authors suggest that they consider the role of colour i n these fishes to be concealment. The present study was designed to determine whether the colour-ation of A. flavidus has a structural or a pigment basis, what pigments are involved, to what extent these are dependent on en-vironment or are hereditary, and how they relate to the ecology and behaviour of the species. 3 FIELD OBSERVATIONS Methods and materials Preliminary seine hauls were made at several locations between Squamish and White Rock on the B r i t i s h Columbia mainland, and between Nanaimo and Jordan River on Vancouver Island, i n order to locate areas of abundance i n different habitats. Following these collections regular stations were established. The most effective c o l l e c t i n g method proved to be beach seining. Because of the variety of condi-tions at the stations several different seines were used. These included: a 10 f t wide "common sense" seine, a 12 f t wide bag seine, 50 and 75 f t haul seines with 150 ft, ropes.. A dip-net was used to co l l e c t small specimens from s p e c i f i c vegetation. G i l l nets, traps, and angling were found to be of l i t t l e value. Shore collections were made i n the i n t e r t i d a l zone during low-tide, usually with the aid of small hand nets. Emulsified rotenone (Chem-Fish Special) was used i n i n t e r t i d a l pools and i n the subtidal zone. In the l a t t e r zone collections were made with the aid of SCUBA. One quart of rotenone mixed with 3 quarts of water was dispensed over each standard ten foot square section of bottom to be poisoned. In areas where current was a pro-blem larger quantities were employed. SCUBA diving was used in several other aspects of t h i s study. Observations were made of the f l o r a and physical habitats i n areas seined and also i n other areas to observe gunnels i n thei r natural habitat. Collections were made by using hand nets and, towards the 4 end of the study, with quinaldine. When quinaldine was used, one part was mixed with 10 parts acetone and squirted under rocks where f i s h might be hiding. The f i s h usually darted out from under the rocks upon contact with the chemical but frequently collapsed a few feet away where they could be picked up with nets. Recovery was rapid and complete. Colour notation The Munsell Color System, (Munsell, 1929) was used to describe the colours of the f i s h during most phases of the study. The colour standards of the Munsell Color Charts have been deposited with the National Bureau of Standards and are cross referenced to the ISCC-NBS Method of Designating Colours and Dictionary of Colour Names (1955). Munsell colour notation i s easily transferred to that of other common systems. Each colour i s given a numerical notation for i t s hue, brightness, and saturation. In the Munsell systems these become hue, value and chroma (H V/C). For example, a sp e c i f i c moderate red has a notation of 5R 5/10 (H = 5R; V = 5; C = 10). The "Pocket" and "Student" editions of the Munsell book contain 40 constant hue charts, containing a t o t a l of 956 colour standards. The student edition was used primarily i n t h i s study. Fish were rated according to the general colouration of their sides, since this colour was representative of that displayed by the f i s h i n their natural attitude. Later in the study i t was realized that b e l l y colour provided a better indication of pigment content, because t h i s area was free of melanophores. 5 In the field, colour evaluations were made where possible with the living fish lying in a V-trough lined with white vinyl. The colour of the fish was determined by scanning the standards and choosing the best match. For maximum accuracy masks should be used which expose "equal areas of the sample" and standard and cover the remainder of the standards and sample. This procedure was not practical with wet, wriggling fish. The same trough was used for laboratory evaluations, but was illuminated with Luxo fluorescent desk lamp 14 inches above the trough. Criticism - Since the fish were wet and shiny they were difficult to compare with flat colours of the standards. Furthermore, slight changes in the position of the fish with regard to light source and the observer could change its apparent colouration. Nevertheless a test of the system under ideal viewing conditions in which 10 freshly captured fish were evaluated twice, resulted -in a range of differences not exceeding one Munsell value and two Munsell chroma. Readings were more difficult with fish from the colour change experiments which tended to become faded and muddy coloured with two visible layers of colours; for example, a red tinge might overlie a muddy green. No single notation could adequately describe these fish. The case was similar with splotchy fish, which could only be evaluated on the basis of average ground colour. The green colour of rich Ulva-green specimens was more saturated than any Munsell value, therefore, they could not be scored accurately. They were given a value of 5GY5/8. 6 The colour notation may be misleading and imply quantification. It must be emphasized that neither the magnitude of colour changes nor the difference in colour between two specimens could be determined from the notation alone. The Munsell: book must be referred to in order to ascertain such differences. In spite of these shortcomings the Munsell colour system appears to be the best for this type of work. The sequence of the standards is more natural and convenient than that used in the Ridgeway charts. The differences between standards is small enough to determine a l l significant colour differences, whereas the ISCC-NBS centroid colour charts are too few for adequate colour descriptions. Description of stations Collecting stations were established at 8 sites, (Fig. 1). Detailed descriptions are given in the Appendix..' The salient features of the 3 monthly stations and 2 most important supplemental stations are summarized in Table 1. 7 Table 1. Gross comparison of principal collecting stations. Jordan River VI-5 Agate Beach VI-8 Lumbermen1 M-10 s Arch M-11 2nd Narrows M-12 Temperatures 40-58 40-52 36-68 36-68 42-55 Recorded (°F) Salinities 28.6-31.1 27-3-32.5 17.1-27.0 17.1-27.0 16.2-24.7 Recorded (per mil) Mean Tidal 6.6 6.2 10.3 10.3 10.3 Amplitude (ft) Intertidal Boulders Loose Sand Packed Boulders Zone Gravel Boulders Subtidal Mud and Gravel Sand and Mixed Gravel and Seining Limits Sand Rock Boulders Exposure - Semi- Semi- Protected Protected Protected General Area protected exposed Inner Bay Inner Bay Strait Plant cover in Luxuriant. No inter- Mixed green Brown dom- Moderate area of efficient Red algae tidal cover. and brown. inant . Some cover. Lower collecting - June abundant, Subtidal Red algae red and intertidal overlain by cover mostly be- green. mostly brown, brown. Some luxuriant, yond effi- some green. patches of mostly brown. cient sein- Red algae Phyllospadix. Some Phyllos- ing. dominant in padix and red upper sub-algae. tidal . 9 Results and discussion Details of collections are given in the ..Appendix. Seining efficiency must be considered in the examination of the field collections. This was studied by means of SCUBA observations. Both the operation of the seine and the behavior of the fish were found to affect seining efficiency. A. flavidus apparently spends part of its time browsing in the vegetation above the bottom, and part hiding within the rocks of the bottom where i t is unlikely to be seined. Some specimens hiding in vegetation were observed to escape from the seine by diving into bottom cover as the seine approached them. Others were seen to plunge under the lead line after having been within the net. It was not possible to determine the percentage which escaped. Although efforts were made to weight the nets properly, the larger seines could not be adjusted to drag the bottom evenly through-out the haul. In water beyond the depth of the seine the floats exerted a force on the lead line, which they did not exert in shallow water. If enough weight was added to compensate for this force, the seine dragged too heavily in the shallower water and became fouled with gravel or mud. With the most suitable weighting for this station, seining efficiency f e l l off rapidly beyond about 75 ft. This meant that the chances of catching a fish in the red algae zone were small and even smaller was the chance of keeping it in the seine all the way to the beach. In spite of this, most seine hauls were begun 150 ft offshore in the hope of acquiring some red specimens. The vegetation itself affected the efficiency of seining. Most algae tended to l i f t the net off the bottom. Zostera was particularly troublesome, not only 10 lifting the lead line, but frequently causing the netting to wrap around i t . Thick beds of Zostera were impossible to seine. The results of the seining collections can only be considered approxima-tions of the relative abundance of A. flavidus. Correlation between fish colour and habitat Three colour phases of A. flavidus were collected corresponding with the red, green and brown colours of vegetation. The range of colours of fish within each colour phase was such, that an almost complete continuum was formed, from green through olive to brown, and brown through red brown to red. Some fish had up to 30 dark spots along the dorsal surface of the sides. A few had these spots or bars along the side also. This condition occurred in fish of all colours. The majority of fish collected in each colour phase were such a close match in colouration to vegetation in their habitat that the colours could not be distinguished visually. This is illustrated for the green and brown phases in Figure 2 and Figure 3. The colours shown are the most typical ones in each phase. Monthly collections at the 3 regular stations have been summar-ized by colour phase in Table 2. Chi-square analysis indicates that the overall proportions of colour phases of A. flavidus collected were roughly similar to the proportions of colour phases of vegetation in the collecting areas. A good general correlation was also found in the supplemental seine collections at Second Narrows (M-12). Data were too few from the other supplemental seine and rotenone collections to permit meaningful comparisons. 11 Figure 3. A green A. flavidus shown with Ulva. 12 Table 2. Chi-square analysis of the colour frequencies of A. flavidus at the three regular collecting stations. A. Chi squares Lumbermen's Arch (M-10) Colour Green Brown Red Observed 119 36 5 Expected 74.1 57-9 28.0 Chi Square 27-3 ** 8.3 * 18.9 ** Agate Beach (VI-8) Green Brown Red 46 59 4 50.4 39.5 14.1 0.4 NS 9.6 ** 11.9 ** Jordan River (VI-5) Green Brown Red 65 85 78 105.5 82.6 39.9 15.5 ** 0.1 NS 36.4 ** B. Additive test for non-significant values beginning with smallest  deviation: 0.1 = 0.1 NS 0.1 +0.4 s 0.5 NS 0.1 +0.4 +8.3 = 8.8 * 0.1 +0.4 +8.3 +9 .6 a 18.4 ** 0.1 +0.4 +8.3 +9 .6 +11.9 , a 30.3 ** All other values highly significant. C. Summary Lumbermen's Arch - more green specimens collected than expected by chance alone, but fewer brown and red. Jordan River - more red and fewer green. Agate Beach - more brown and fewer red. Note 1. Calculated from overall totals of colour phases collected. 13 Direct observations It was seldom possible during seine and rotenone collections to know the species or colour of the vegetation from which a fish was collected; but in those cases in which such observations were possible, the colour of the fish correlated well with that of the vegetation in which i t had been hiding. For example, at Jordan River, specimens shaken from the kelp Egregia were invariably a matching brown. Direct observations were made with SCUBA. A. flavidus was usually difficult to locate even in areas in which i t was known to occur. None were seen in 3 out of the 4 dives made at Lumbermen's Arch. On the one dive, in which they were encountered, about 12 were seen hiding in brown or green algae. In each case the fish appeared to be of the same colour as the alga. Both plant and rock cover were used for escape after detection. One fish, found hiding in Ulva, was pursued cautiously for several minutes and eventually captured. This fish confined its movements to the Ulva, the colour of which i t matched, moving from one patch to another, each time i t was disturbed. It did not attempt to hide among the rocks even though the bottom was bouldery and provided with many good hiding places. Diving observations were made on two occasions at Second Narrows (St. M-12). Several red A. flavidus were seen in the red algae zone, both under rocks covered with red laver and within the red laver i t -self. Two apparently brown specimens were seen in Laminaria. No specimens were seen which appeared to be in a non-matching habitat. No red specimens were seen during any of the shore collections. This would be expected i f colour is used cryptically since the shore 14 collections were above the red algae zone. A single brown A. flavidus was encountered at Saxe Pt., Victoria. This specimen was swimming amongst Nereocystis near the surface in about 20 ft of water. When the fish saw me i t turned around abruptly and headed through the kelp towards the bottom. I attempted to follow, but could not. Approximately 12 other observational dives were made in areas where A. flavidus was known to occur including the Agate Beach collecting station (VI-8), but no A. flavidus were observed. This emphasizes the effectiveness of its cryptic nature. Direct observations of A. flavidus in association with matching vegetation have been related to me by Mr. M. Varga of Undersea Gardens at Oak Bay, B.C. The Underseas Gardens are a novel aquarium in which the visitors descend into a barge-like vessel and view the specimens in an outside wire enclosure surrounding the vessel. The "Gardens" float and are several feet above the bottom during low tide. Drifting and voluntary algae gradually accumulate on the wire screen and must be removed periodically. A. flavidus is said to become a frequent inhabitant of the algae as the cover becomes abundant. Seasonal abundance A. flavidus reached maximum abundance in the seining areas during the summer, the period when vegetation was most abundant. It declined in fa l l with the decrease in vegetation and was rarely seined during the winter, the period of coldest water and spawning activity as well as least vegetation. 15 Vertical distribution Shore and seine collections, diving observations and rotenone collections suggest that A. flavidus is most common in the lower inter-tidal zone and upper subtidal zone. It was seldom found above the water line except during the spawning season and at exceptional low tides. It was never found below the 20 ft tidal level even though a number of efficient rotenone collections were made between the 20 and 50 ft levels. Diurnal activity A. flavidus appears to be more active in the vegetation during the day than at night. At Sooke during the period of April through August the average number of individuals captured at night was 0.7 per seine haul as compared with 5.9 per day haul. A similar situation is indicated at Lumbermen's Arch. Form of vegetation An examination of the length-frequency plot of the Lumbermen's Arch collections (Fig. 4) indicates different distributions for the brown and green phases. This plot combined with the field observa-tions suggests that certain types of algae are more effective for different lengths of fish. The young-of-the-year fish appear to find excellent cover in the Sargassum where their body size correlates well with the thallus size. On the other hand the Sargassum probably does not provide good cover for fish large enough to prey upon the YOTY. Likewise, Ulva, Zostera and similar types probably provide good cover for medium sized fish, but not for large fish. For 30 Total length (class midpoint) - mm Figure 4. Length-frequency plot of brown and green A. flavidus from Lumberman's Arch. 17 example, a f i s h of 300 mm would obviously contrast i n size with a blade of Zostera, and would generally be too large to hide in the small, patchily distributed Ulva, but i t would find adequate cover in the large folds of a laminarian. A similar relationship between the form of vegetation and size of A. flavidus i s suggested by the Jordan River data. The length-frequency curves (Fig. 5) reveal a f a i r l y even d i s t r i b u t i o n of lengths within a l l three colour phases, but with larger numbers of brown spe-cimens below 70 mm T.L. and above 250 mm T.L. The largest f i s h c o l -lected were also brown. Fi e l d observations suggest that the small f i s h at t h i s station may find favorable cover i n Egregia and tend to choose i t or take on i t s colouration. This would account for the high pro-portion of small brown specimens. The laminarians, being more stable seasonally and of larger size, may provide more favourable cover for large A. flavidus. 15 5$ roo 150 200 250 3013 rfo Total length (class midpoint) - mm Figure 5. Length-frequency plot of A. flavidus from Jordan River. 19 Summary 1. The proportions of colour phases of A. flavidus collected were roughly similar to the proportions of colour phases of vegetation present in the collecting areas. 2. Diving observations and seine haul data provide evidence that A. flavidus utilizes colouration for concealment within vegetation, amongst which it appears to wander fairly extensively. 3. A. flavidus utilizes hiding places under rocks during the nesting season and apparently when resting or threatened. It may utilize plant cover primarily while feeding. 4. A. flavidus appears to be more active by day than by night. 5. There is a seasonal decline in numbers seined, which corresponds with changes in abundance of plant cover, but also with the period of colder water, and in part with the period of breeding activity. 6. The length-frequency distributions and collecting observations, suggest that the form of a plant may make i t more suitable for one size range of A. flavidus than another, and that there may be changes of habitat and colour with increase in size. 20 HABITAT SELECTION EXPERIMENTS I t appears from the evidence of the f i e l d studies that A. flavidus wanders considerably within the vegetation and must make frequent choices of habitat. Several laboratory studies were carried out i n order to gain information about this process. Experiment 1 - Reflected l i g h t Hypothesis - Given a choice of two gravel backgrounds of different colour A. flavidus w i l l choose the one which r e f l e c t s l i g h t with a wave length most similar to i t s skin colour. Method and materials - The experiment was carried out in a 20 gallon aquarium provided with running sea water from the same system which supplied the specimen holding tanks. The flow was stopped and the i n l e t hose removed just prior to each part of the experiment so that there were no currents or obstructions during the t r i a l s . The outsides of the aquarium were covered with white translucent p l a s t i c . Illumination was solely from a fluorescent l i g h t c e i l i n g f i x t u r e . Non-toxic coloured aquarium gravel was used to provide the coloured backgrounds. Each half of the aquarium bottom was completely covered with the chosen colour. P r i o r to t h i s and the following experiments f i s h were held in bare, 40 gallon plywood holding tanks for periods ranging from a few days to a month. A single f i s h was used for each t r i a l . I t was introduced to the aquarium from a glass j a r (with a glass plate cover). The j a r contain-ing the f i s h was put into the centre of the aquarium upside down. After permitting the f i s h about 10 seconds to observe the two colours 21 of the aquarium bottom, the l i d was removed and the jar was l i f ted out of the tank, leaving the f ish behind. The f i r s t choice that each f ish made was scored. Results - The results of the experiment are summarized in Table 3. Table 3. Analysis of f i r s t choices made by A. flavidus in reflected light experiment. Fish colour Bottom Green colour chosen red Chi-square Green 2 4 0.667 NS Green 6 9 0.400 NS Red 2 4 0.667 NS Fish colour Bottom Brown colour chosen Green Chi-square Brown 6 4 0.667 NS Brown 6 6 0.0 NS Some observations were made by leaving f ish in the tank after they had made their f i r s t choice. Most moved about the tank intermittently and randomly for several hours without showing a definite preference. During the periods when the f ish were settled on the bottom most of them rested against the corners. If a rock or shell was placed in the tank the f ish almost invariably used i t for cover no matter which end i t was moved to. When several f i sh were placed in the aquarium without 22 protective cover they tended to intertwine themselves into a b a l l with-out apparent regard for l i g h t colour. The intertwining behaviour i s common when th i s species i s kept i n bare tanks. Conclusions - The A. flavidus did not show a sign i f i c a n t preference for the gravel bottom which most nearly matched t h e i r colouration. Experiment 2 - Transmitted l i g h t Hypothesis - Given a choice of two areas illuminated by trans-mitted l i g h t of diff e r e n t wave lengths A. flavidus w i l l choose the area which i s illuminated by the wave length most similar to i t s own skin colour. Method - Except for illumination t h i s experiment was carried out in the same manner as the previous one, (Fig. 6 ) . Aquarium filter filter 25 watt lamp Divider 25 watt lamp Figure 6. Top view of apparatus used i n transmitted l i g h t preference experiment. 23 Gelatine filters were cut to f i t the halves of one side of the aquarium. Twenty-five watt incandescent lamps were used as light sources. The intensities of the light transmitted through the filters were balanced by varying their distance from the tank. A Norwood direct reading light meter, indicated that 16 foot-candles was passing through each f i l t e r into the aquarium. The wave lengths of the transmitted light were measured with a Browning hand spectroscope. Approximately 90% of the light passing through the green f i l t e r was between wave lengths of 490-580 mji. Maximum transmission was near 540 mjj. Approximately 90% of the red light exceeded 590 mp. Maximum transmission appeared to be near 700 mjj. Results - First choices are given in Table 4. Table 4. First choices made by A. flavidus in transmitted light experiment. Fish colour Light colour chosen Chi-square Red Green Red 5 5 0.0 NS Green 12 8 0.8 NS Six fish were observed continually for one hour periods and then intermittently for periods up to 2 days (Table 5). When fish were-not moving around they usually settled in the corners of the tank. Some fish wrapped themselves around the intake hose if it was left lowered into the tank. 24 Table 5. Observations of green A. flavidus kept in tank with trans-mitted green light in one end and transmitted red light in the other. Fish First First 60 Observations Minutes No. Colour Choice Red Green 1 2.5GY6/4 Red 20 40 Fish inactive; stayed on bottom in red zone for first 20 min. Chased to green side where i t re-mained unmoved for rest of hour. 2 5GY7/9 Red 36 24 Actively swam against glass at red light source during first 6 min. but made trips of a few sec. into green zone. Inter-mittent observations during next two days in-dicated that fish was s t i l l unsettled. 3 10GY5/6 Green 6 54 Fish settled after about 25 min. No observation recorded after first hour 4 2.5GY6/8 Red 39 21 Remained in red for first 65 sec. Spent rest of hour exploring inter-mittently. After 6 hours fish s t i l l not settled. 5 5GY5/9 Red 46 14 Active for first 30 min., but remained s t i l l in red zone for last 30 minutes. After 11 hours seemed to have settled in red. 6 5GY5/8 Green 22 38 Changed ends every 2-3 min. in first 5 min. St i l l changing ends after 20 hours. 25 Conclusions - The A. flavidus did not show a significant preference for the colour of transmitted light most similar to their own colour-ation. Experiment 3 - Algae and rocks Hypothesis - If rocks, covered with different colours of algae, are placed in the opposite ends of an aquarium A. flavidus will select the end which contains the alga most similar in colour to its own skin. Method - Two 20 gallon aquaria were set up in the laboratory. Each aquarium had a stand-pipe outlet in the centre and a water inlet at one end. The bottom was covered with granite pea gravel. A flattened rock approximately 6 inches in diameter and covered with Ulva (5GY5/8) was placed in one end of each aquarium. A similar rock covered with the red laver Porphyra was placed in the other. Fish were introduced by use of a glass jar as in the previous experiments, and their first choices recorded. In the second part of the experiment two red fish and two green fish were introduced into each tank one at a time and left for 10 days. Their positions were noted 3 times per day. Results - In this experiment the fish usually tried to get under the two large rocks in the tank immediately after being introduced. They randomly explored the rocks until they could find an opening. They did not always completely explore one rock before looking for an entrance under the adjacent rock; but usually entered the first open-ing they chanced upon. Their choice did not appear to be influenced 26 by the presence or absence of another fish under the rock. The fish did not hide within the algae. It was thought that after a few days they might move to the algae at least in search of food but this did not appear to happen. The first choice (Table 6) was taken to be the colour of algae that the fish headed towards after leaving the introduction jar. Table 6. First choices of habitat made by A. flavidus when given alternatives of rocks covered with Ulva (green) and rocks covered with Porphyra (red). Fish colour Alga colour Chi square Red Green  Red 5 2 1 . 2 9 NS Green 6 1 0 1 . 0 0 NS Daily distributions of the fish in the 1 0 day multiple fish experiment are given in Table 7. The scores refer to the colour of algae on the rocks under which the fish were hiding. In order to make the daily observations, i t was necessary to l i f t the rocks slightly. The fish typically reacted in one of three ways to this disturbance. 1 ) Raised their heads and looked around but stayed under the rock. 2 ) Swam quickly out from under the rock and sought other cover, usually under the adjacent rock. 3) Swam quickly out from under the rock and then randomly about the tank but returned to their original position within a short time after the rock was replaced. 27 Table 7. Chi-square analyses of 10 day habitat observations in rock and algae experiment. Analysis arranged to test for selection of matching algae. Fish Colour Habitat Colour Times Fish Present Chi Square Percent Probability Aquarium 1 Red Red Red Green 36 24 2.60 > 1 0 Green Green Red Green 49 11 13.70** < 1 Aquarium 2 Red Red 25 1.67 < 20 Red Green 35 Green Red 21 5•40 > 5 Green Green 39 Combined Totals Red Red 61 0.03 > 90 Red Green 59 Green Red 70 3.33 > 5 Green Green 50 28 Most habitat changes took place during or immediately after the rock lifting. The fish became tamer as the experiment progressed and were less likely to flee when the rocks were lifted. All of the fish came at least part way out of their hiding place during feeding by the time the experiment was complete. Some of them hit the food almost as soon as it was placed in the water. Following feeding they nearly always returned to the hiding place they had occupied before feeding. Discussion - In the first choice experiment A. flavidus more fre-quently chose the space under the rock with matching algae; but a statistically significant preference is not indicated. Chi-square analyses of the 10 day part of the experiment are given in Table 7. Conclusions from these analyses must be drawn cautiously since the disturbing of the fish during observations influenced the fish's distribution. Further, the number of observa-tions is not a measure of the number of responses and is therefore only a rough measure of preference. However, if there is a strong preference these factors should not obscure i t . It can be seen from the data that the green A. flavidus in aquarium 1 showed a significant preference for the space under the red-algae rock. This space was also chosen more frequently by the red fish, but the preference was not significant. These results suggested that the space under the rock might be more important than the algae on i t . If the data are rearranged to test this possibility (Table 8) then this does appear to be the case. 29 Table 8. Chi square analysis of 10 day habitat observations in rocks and algae experiment. Analysis arranged to test the possi-bility that space under one of the rocks in each tank was preferred. Habitat Times chosen Times not chosen Chi square Tank 1 Rock with red alga 35 9.12** Tank 2 Rock with green alga 74 46 6.53* Conclusions -1. A. flavidus chose to hide under the algae-covered rocks rather than within the algae on i t . This coincides with the observations made in the field which suggest that under stress A. flavidus seeks cover under the rocks rather than within the vegetation. 2. A. flavidus did not show a statistically significant pre-ference for a rock covered with matching algae. 3. In the 10 day experiment the space under one of the two rocks in each aquarium was preferred by A. flavidus without regard to algae on i t . Experiment 4 - Algae Hypothesis - Given a choice of no other cover but algae, A. flavidus will choose the alga most similar in colour to its own skin. Method - An aquarium was set up with a bunch of Ulva in one end and a 30 bunch of Porphyra in the other. A small amount of solder wire was used to hold the bunches together and weight them down. The f i sh were introduced as in the previous three experiments. They were left in the aquarium for ten days and their positions recorded once per day. They could usually be observed without frightening them out of the algae they occupied. Results - The results of the experiment are given in Table 9. The f i sh did show a s ta t i s t i ca l ly significant preference for matching algae. The f i sh spent nearly a l l their time hiding under the algae but some of them were observed to leave i t during feeding time after the f i r s t few days. Table 9. Distribution of A. flavidus in an aquarium containing bunches of red and green algae. Colour of alga Chi square Red Green Total choices 17 3 by red f i sh 17.00 * * Total choices 4 16 by green f i sh General conclusions - The completely unnatural state under which the f i sh were kept in experiments 1 and 2 did not demonstrate any colour preference that the f ish might make in nature. The third experiment strongly suggests that the primary need of the f ish is cover in form of a solid hiding place. Preference of matching vegetation was ind i -31 cated in experiment 4, where the only cover the fish had was in algae. From this experiment and the field observations i t appears possible that in nature the fish might seek cover under the rocks in times of known danger as well as during the reproductive period and utilize cryptic colouration only at certain times, such as during feeding. The choice of a matching plant may be made on the basis of its form or odour as well as its colour. However, since A. flavidus appears to wander within its habitat amongst a variety of algae of the same colour, a l l which have different odours and form it appears that the choice would be made primarily on the basis of colour. According to Walls (1942) "No reasonable student of the problem any longer doubts that fishes - a l l duplex teleosts at least - can experience hue as a sensation apart from brightness." 1 know of no work since then which would lead one to believe that A. flavidus does not have colour vision. 32 DESCRIPTION OF SKIN The skin of A. flavidus is smooth and slippery. Although i t contains scales, they are small and not visible to the naked eye. Wet mounts were made of strips of skin from freshly caught spe-cimens and examined under the light microscope and under the phase microscope. Marked histological differences were apparent between the three colour phases. Green fish (5GY5/8 (Munsell)) (strong yellow-green (ISCC NBS)) These fish are typically of the same colour as the green alga Ulva. Their colouration is caused by a yellow-green pigment which appears to be dispersed continuously throughout the dermis rather than within discrete chromatophores. Melanophores, in punctate and semi-punctate condition, Figure 7, are distributed relatively uniformly over the body but become absent on the belly. In this state of dispersion they play l i t t l e role in the colouration or pattern other than rendering the fish a darker green than would result from the yellow-green pigment alone. The flesh underlying the skin has a pale green cast which renders the fish greener than it would be i f the flesh were white; but which other than this does not contribute to the colouration. 1 2 3 4 5 Figure 7. Melanophore index (from Hogben and Slome, 193 1) . 33 The yellow-green pigment appeared continuously dispersed in a l l f i sh examined over 100 mm, except for one laboratory specimen. In this specimen, which had been maintained on a pale background for several weeks, the pigment was in star-shaped masses about the same size as the melanophores. These structures were paler than the pigment in the skin of "normal" specimens and thus did not appear to have resulted from aggregation of the pigment, but rather from loss of pigment. Other discrete yellow pigment masses were observed in larval A. flavidus. Those in the very early stages of larvae which had been hatched and reared in the aquarium, had the appearance of typical punctate xantho-phores. In wild larvae, the pigment had a dispersed appearance which seemed to result from a greater abundance of pigment. These discrete pigment masses may have been centres of pigment distribution. However, regardless of their exact nature the lack of definit ive xanthophores in normal post-larval specimens probably precludes physiological colour changes involving the yellow-green pigment. Dark-green f ish (7.5GY4/6 (Munsell)(moderate olive-green (ISCC-NBS) -These f i sh had evenly dispersed melanophores with a H.S. index of 3 to 3 . 5 . In some f ish there appeared to be more melanophores than in Ulva-green f i sh ; but the greater dispersion of melanin was the primary cause of the darkening. Under the microscope the yellow-green pig-mentation appeared the same as in the Ulva-green f i sh . Green f ish with spotted-pattern - These f ish were characterized by the presence of one or more rows of squarish spots along the back and sides. The spots resulted from a combination of more melanophores and 34 greater dispersion of melanin (H.S. index 4 to 5) than occurred in uniform specimens. Red fish - The general colouration of red A. flavidus is caused by numerous erythrophores containing a granular pigment. (The granules have a diameter of about 1^.) In dark red specimens, e.g. 5R3/6 (Munsell), erythrophores are dispersed to the extent that they appear to form a complete network. In less red specimens the erythrophores appear as discrete units, which would at first appear to be the result of pigment aggregation, but which probably results primarily from the loss of pigment. This is discussed further under color change experi-ments. The state of erythrophore dispersion f i t the Hogben-Slome index quite well i f the "network stage" was given an H.S. index of 5 . This index was utilized to describe the condition of the erythrophores in subsequent experiments. Melanophores appear to play the same roles as they do in the green fish. Brown fish - Pigment structures similar or identical to those of both the red and green fish appear to be utilized in the brown fish. A typical kelp coloured A. flavidus (2 .5Y5/6 (Munsell); light olive brown, (ISCC-NBS) was found to be covered with orange-red erythrophores in punctate and semi-punctate condition and, between them, an evenly dispersed yellow pigment. However, the inter-erythrophore pigment varied in colour from yellow-green to orange-red depending upon the shade of brown apparent to the naked eye. In the more red-brown specimens it became indistinguishable from that of the erythrophore. It was not possible to tell whether this color variation was the result of combinations of the basic red and yellow-green pigments or the presence of different pigments. The melanophores, as in the green and red fish, provide an over-all darkening, and are responsible for the dark spots when they occur. 36 DESCRIPTION OF PIGMENTS The nature of the skin pigments of A. flavidus was investigated in 1963 at the Fisheries Research Board of Canada, Fisheries Tech-nological Station, on the University of Brit ish Columbia campus, with the assistance of Dr. Michael Smith. The procedures and information from this study were then used in the analysis of the colour change experiments. Further studies, resulting in the more rigorous identif icat ion of the pigments, were carried out at the Scripps Institution of Ocean-ography in 1965, with the help of Dr. Denis Fox. Fish for these studies were obtained from Moss Beach, San Mateo County, Cal ifornia. Green phase The scheme of analysis used in the Scripps study is outlined in Figure 8. The skin was stripped from the sides of frozen specimens, weighed, then extracted in pure ethanol in a tissue grinder. A Bausch and Lomb Model 505 recording spectrophotometer was used to measure spectral absorptions. The pigment extracts were kept under nitrogen as much as possible during the analysis in order to prevent oxidation. The spectrum of the crude pigment extract i s shown in Figure 9. Two very dist inct pigment fractions were separated by column chromotography ( s i l i c a gel -cel i te, 111). The f i r s t fraction moved rapidly through the column in pure hexane. The second fraction did not move in pure hexane and moved only very slowly when developed with 2.5% methanol in hexane; but was eluted through the column with 5% 37 Whole skin - ground and extracted with ethanol I Fi ltered through ce l i te I Washed into hexane with salt water Spectrum ("crude") Figure 9 I Chromatographed on s i l i c a ge l -ce l i te column. Developed with increasing concentration of methanol in hexane I Fraction 1 (undefined) I eluted with hexane I Spectrum Figure 10 Partitioned (epiphasic) I Original pigment probably lso-j3 carotene 1 Fraction 2 (yellow-green band) I eluted with 5% methanol in hexane I Spectrum Figure 9 Partitioned (epiphasic) I Hydrolyzed in ethanol ION NaOH Partitioned hexane methanol 20 80 Original pigment probably dihydroxy e carotene Figure 8. Analysis of skin pigments from green A. flavidus (5GY5/8 Munsell). W a v e l e n g t h Figure 9. Absorption spectra of pigment from green A. flavidus showing the crude pigment extract in comparison with fraction 2 following chromatography. 39 methanol in hexane. Fraction 1 - The absorption curve in hexane is shown in Figure 10 along with the spectrum of a sample of a pure j3 carotene obtained from Dr. L. Zechmeister. The two curves are similar in shape. One of the principal characteristics of j3 carotene curves, distinguishing them from other carotenes, is the presence of an inflection instead of a maximum at position I. The A. flavidus pigment has lower maxima values than the pure ft carotene, but they compare well with those reported by Lee (1965) for isomerized ft carotene from the isopod Idothea montereyensisI A. flavidus ( ~ 418) 444.0 470 1_. montereyensis ( ~ 419) 444.5 472 It appears from the absorption spectrum that the pigment in this fraction (1) may be isomerized ft carotene. Fraction 2 - The absorption curve in hexane is shown in Figure 9. It is almost identical with the crude spectrum before chromatography and the resultant removal of fraction 1. From its spectrum and partition behaviour, fraction 2 appeared to be an undescribed zanthophyhl ester, perhaps derived from epsilon carotene. Known samples of lutein, zeaxanthin, isozeaxanthin and epsilon carotene were compared with the A. flavidus pigment, by thin layer co-chromatography and chemical tests as well as absorption spectra, (Crozier and Wilkie, 1966). From these experiments i t was suggested that i t was an undescribed pigment, an esterified 3,3' dihydroxy xanthophyll of e carotene. This analysis is 41 summarized in Table 10. The relative proportions of pigments found in skin extract of the green phase are given below: Optical Density % X Vol. j3 carotene - like fraction 0.8 7.5 dihydroxy e carotene 9.9 92.5 10.7 100.0 The absorption maxima of crude skin extracts of green A. flavidus which were obtained in initial experiments at the Fisheries Technolo-gical Center with B. C. fish are compared below with those obtained at Scripps with California fish. The solvent in each case was acetone. The shape of both curves is similar; but the maxima differ as can be seen below: Maxima in Acetone B. C. fish 418 442 472 California fish 415 436 466 Difference 3 6 6 These differences are almost certainly due to differences in the techniques and solvents used in the two different studies, since sub-sequent examination of 3 green B. C. specimens has yielded the same results as were obtained with California specimens. The acetone and ethanol extractions of the pigments from the green fish were more yellow than the pigments appeared to be while in the skin of the fish. It was found that the pigment could be extracted, Table 10. Comparison of a xanthophyll suspected of being 3,3* dihydroxy e carotene with samples of known carotenoids (adapted from Crozier and Wilkie, 1966). Compound Spectral Maxima Benzene (minimum) m^  Hexane Parti-tion ratio P. R. after Methy-lation Hydroxyl position* a na Epoxide test Relative** position TLC e - carotene 446-7(464)477-8 435(454)466 100:00 100:00 0 0 neg. • 3,3' dihydroxy e - carotene 447 (464)477-8 436(454)466 20:80 88:12 2 0 neg. 2 Isozeaxanthin (4,4' dihydroxy j3 - carotene) 447(463)474 22:78 88:12 2 0 neg. 1 Lutein (3,3' dihydroxy a - carotene) 453 (470)482 441(458)470 12:88 60140 1 1 neg. 3 Zeaxanthin (3,3' dihydroxy j3 - carotene) 447(464)475 10:90 10:90 0 2 neg. * a - allylic na - non-allylic ** The number 1 indicates least polar, etc. 43 unaltered in colour, through the use of an aqueous solution, e.g., buffered phosphate solution (pH 6.7). Treatment of the aqueous solution with an equal volume of acetone or ethanol plus heat caused i t to become yellow. Analysis of the yellow solution yielded the same results as that obtained when the pigment was extracted directly with ethanol, namely dihydroxy e carotene except that the ft carotene-like fraction was not recovered. However, i ts presence in the aqueous extract cannot be precluded because the quantities of pigments treated in this manner may have been too small to permit i ts detection. The shift in colour and solubi l i ty of the pigment suggests that the carotenoid may be in association with a protein or lipoprotein. Carotenoproteins have been reported in plants and animals by a number of workers and have been discussed in reviews by Fox (1953) and Goodwin (1952). Recently Cheesman (1966) has reported an ester i f ied carotenoprotein in the hermit crab Eupagurus. Further work is being done to examine this poss ib i l i ty in A. flavidus since the process which produces the shift in colour towards green plays a significant role in i t s adaptive colouration. Conclusion The colouration of green A. flavidus appears to result primarily from ester i f ied 3,3' dihydroxy e carotene, which is rendered more green by i ts association with a protein or lipoprotein. A second carotenoid which may be isomerized fj carotene is present in small quantities; but appears to have l i t t l e effect on the macro-scopic appearance of the f i sh even though i t is yellow in colour. It is not known whether this component forms the same association with 44 protein that the dihydroxy e carotene appears to form. Red phase The scheme of analysis used in the Scripps study is outlined in Figure 11. The procedures were basically the same as those used for the green fish. Chromatography on a 1:1 silica gel-celite column yielded four fractions, each epiphasic in hexane over 95% methanol. Fraction 1 - This fraction yielded the same absorption spectrum and partition ratio after hydrolysis as had been found in the green fish for dihydroxy e carotene and was probably the same pigment. Fraction 2 - The absorption spectrum yielded by this fraction is similar to that of a ketone, but not enough was present to permit identifica-tion. Fractions 3 and 4 - These were the major fractions present. Their spectra (Fig. 12) and chemical behaviour suggest that they were two esters of astaxanthin. The relative proportions of the pigments found in red A. flavidus are given below: Optical Density % X Vol. dihydroxy e carotene 3.2 2.9 unidentified carotenoid 4.2 3.8 astaxanthin 101.4 _93.3 108.8 100.0 45 Ground whole skin extracted with ethanol Fi l tered through ce l i te I Washed into hexane I Spectrum run ("crude") Figure 12 Chromatographed on S i l i ca ge l -ce l i te column 1:1 0.1% methanol in hexane Fraction 1 Fractions eluted with 5% methanol in hexane T 3 4 washed into hexane I Spectrum I Partitioned (epiphasic) I Hydrolyzed '('ION NaOH) I Spectrum washed into hexane I Spectrum I Partitioned (epiphasic) Reduced (Na BH4) I Not recovered washed into hexane I Spectrum Partitioned (epiphasic) I Hydrolyzed (ION NaOH) I Purple crystals washed into hexane I Spectrum Partitioned (epiphasic) Hydrolyzed (ION NaOH) I Purple crystals Partitioned (19:81) Original pigment was astaxanthin (esterified) Original pigment was dihydroxy e carotene (esterified) Figure 11. Analysis of skin pigments from red A. flavidus (5R3/8 • Munsell). 1 400 420 440 460 480 500 Wavelength m^  gure 12. Absorption spectrum of principal pigment fraction from red A. flavidus. 47 The crude pigment was rerun in acetone for comparison with the U.B.C. experiments. It yielded an identical spectrum, but with the single maximum about 5 mp, lower than that found at U.B.C. However, red B.C. fish examined at Scripps produced the same absorption spectrum as found with the California fish. Thus the original differences are probably due to differences in the analyses not differences in pig-ments . Conclusions The general colouration of red A. flavidus results primarily from the presence of a red pigment which appears to be esterified astaxan-thin (3,3' -dihydroxy-4: 4' -diketo-/J-carotene). Brown phase In initial experiments at the Fisheries Technological Station absorption spectra were obtained from A. flavidus which had the appearance of being composites of the spectra of the red and yellow-green pigments. In order to test this possibility, acetone extrac-tions were made from "pure" red and "pure" green specimens. The ab-sorption spectrum of a mixture containing equal quantities of the two pigments was indeed similar to the spectra obtained with pigment from brown specimens. Investigation at Scripps of the pigments from a moderate reddish brown specimen (ISCC-NBS designation; 2.5YR3/4 (Munsell), yielded two pigments apparently identical to the major components found in red •and green specimens, namely esterified astaxanthin and esterified dihydroxy e carotene. Details of analysis are given in Figure 13. 48 Ground whole skin extracted with ethanol I Filtered through celite I Washed into hexane I Spectrum ("crude") Chromatographed on celite-silica gel 1:1 Fraction 1 Fraction 2 Washed into hexane Washed into hexane Spectrum 414 436 466 Spectrum Partitioned (epiphasic) Partitioned (epiphasic) Hydrolyzed ION NaOH Hydrolyzed (ION NaOH) Partitioned 20:80 Purple crystals Original pigment was dihydroxy e carotene Original pigment was astaxanthin Figure 13. Analysis of skin pigments from a red-brown A. flavidus (2.5 YR3/4 Munsell). 49 There was no detectable ft carotene or the unidentified orange xantho-phyll. The mixture experiment at U.B.C. and the more detailed analysis at Scripps suggest, that a combination of the major pigment from green fish, with the major pigment from a red fish can result in the brown phase of A. flavidus. This does not preclude the possibility that other pigments can play a major role in the colouration of other brown specimens as perhaps is suggested by the variation discovered in colouration of the inter-erythrophore pigments (p. 34)., nor does i t re-veal the role of, or extent of, protein association with the pigments. A similar colouration system has been reported in the isopod Idothea montereyensis by Lee (1965). He shows that two pigments are primarily responsible for all three colour phases: red, green and brown. Lutein (3,3'-dihydroxy-a carotene) conjugated with protein and not contained in chromatophores causes the colouration of the green phase. Canthaxanthin in erythrophores and conjugated with a protein causes the red phase. Varying proportions of these two pigments account for the colouration of brown specimens. Conclusions 1) The colouration of red A. flavidus is caused by a red pigment contained in erythrophores. The pigment is probably esterified as-taxanthin. 2) The colouration of green specimens of A. flavidus, appears to result primarily from the presence of esterified di-hydroxy e carotene in association with a protein or lipoprotein and which does not appear 50 to be contained in chromatophores. A second pigment which may be isomerized [} carotene is present in small quantities and probably does not contribute significantly to the colouration. 3) The colouration of brown A. flavidus appears to result from combinations of the major pigments found in red and green phases. Whether there are only two major components involved in a l l brown specimens is not known. The role of protein association with the e xanthophyll probably varies with the hue of brown. 51 FIRST STAGE COLOUR CHANGE EXPERIMENTS These preliminary experiments were carried out in an attempt to induce physiological colour changes. They were performed in the assistant curator's laboratory at the Vancouver Public Aquarium. This room has no windows and therefore received no natural daylight. Experiment 1 - Hue changes in response to background could result from the stimulation of dorsal retinal elements which have the capacity to differentiate between reflected lights of different wave lengths. This possibility was tested by holding fish in aquaria with different coloured bottoms. Standard 10 gallon stainless steel and glass aquaria were set up with running sea water. Galvanized iron sheets painted with red, green, yellow or white paint were used under the glass bottoms to provide the different reflecting backgrounds. Standard aquarium lights and reflectors provided the illumination. Several green and brown A. flavidus were put in each aquarium and kept for 30 days. No gross changes towards a matching background occurred. Experiment 2 - Since A. flavidus lives within vegetation it was thought that stimulating the basal retinal elements with the desired hue of reflected light might result in a colour change. In order to accomplish this the sides and ends of the aquaria were covered with painted sheets of galvanized iron in addition to those on the bottom. The above experiment was then repeated with a new series of A. flavidus. Again no gross colour changes were observed. 52 Experiment 3 - Even though the first two experiments did not induce apparent physiological colour changes i t seemed possible that trans-mitted light, rather than reflected light might be effective, since A. flavidus, when hiding in vegetation, could be subjected to more transmitted than reflected light. Red, green and yellow incandescent lights were added to the aquaria and another series of green and brown fish were held for 30 days. There was no apparent difference in these results from those of the first two experiments. Conclusions The only apparent colour change involving the chromatic pigments was a general fading. This was more noticeable in green than brown fish. Although the experimental conditions differed considerably from those encountered in the field, they did suggest that physiological colour changes alone, that is, colour changes resulting from movement of pigment in chromatophores, could not accout for the variety of colour phases in A. flavidus. This is to be expected since i t has already been shown by pigment analysis and histology that the colour phases are morphologically different. The experiments do not preclude the possibility of comprehensive morphological colour changes with prolonged exposure to these or other conditions. 53 SECOND STAGE COLOUR CHANGE EXPERIMENTS The primary purpose of these experiments was to investigate the possibility that morphological colour changes accounted for the various colours of A. flavidus. Methods and materials The experimental conditions were similar to those in the pre-liminary experiments; but steps were taken to reduce extraneous light-ing to a minimum- All-glass, ten gallon aquaria were constructed with glued corners and no supporting frames. Each aquarium was placed in a separate compartment covered with white translucent vinyl plastic. The principal light source was the daylight from the windows which shone through the plastic. This was supplemented with the fluorescent room lights. Illumination was very similar for each compartment. The aquaria were provided with running sea water from the main aquarium water system. Temperatures ranged from 10.3 to I7.I 0 C during the experiments. Salinities varied between approximately l7°/„ and The fish were thus subjected to considerable variation in water quality. During and prior to the experiments the fish were kept on a standard diet which consisted of chopped, peeled and boiled shrimp (Pandulus sp.); chopped, white lingcod flesh (Ophiodon elongatus), and frozen adult brine shrimp (Artemia salina). This diet provided small amounts of several carotenoids; but total carotenoid was pro-bably considerably less than occurs in the natural diet. On the basis of a report by Gilchrist and Green (1960) i t was believed that adult 54 Artemia salina contained small quantities of astaxanthin, which in turn was being provided to the fish. However, i t has subsequently been suggested that they do not; but contain instead canthaxanthin, {$ carotene, and echinenone (Davis et al., (1965) with canthaxanthin the major end product. Krinsky (1964) has found that over 95% of the total carotenoid found in the nauplii is canthaxanthin, and concludes that "the major carotenoid pigment of Artemia salina is canthaxanthin." Several species of Pandulus have been analyzed for pigments and generally have been reported to contain astaxanthin and carotene (Goodwin, 1952). In peeled, boiled shrimp most of the carotenoid pigments, however, have been removed. White Ophiodon flesh is low in carotenoids. The degree of colour change which occurred in the experiments was determined by using the Munsell colour scale plus histological and pigment analyses of representative fish. As a means of comparing the pigments formulae were derived for approximating the relative contributions made by the red and green pigment complexes. An examination of absorption spectra of the pigments from pure red and pure green fish revealed that the optical density of the red pigment complex was 1/2 of its total at 520 m^  while the green pigment complex was almost zero. This led to the trial formulae: a) Relative quantity of red pigments = 2 x abs^Q b) Relative quantity of green pigments = a^'s^-j2 " ^ a^S520 (The proportion of pigment components = , .) 55 The formula was tested by mixing aliquots of pigments from a red and green fish and seeing if their relative concentrations could be compared with those determined by the formulae. The optical densities of pigments from the individual fish were: Red 0.42 Green 0.70 In a 50:50 mixture these values would be halved, i.e., 0.21 and 0.35. The resultant curve had an O.D. of 0.11 at 520 m^  and 0.45 at 472 m^ . Substituting these values into the formulae? Red component = 2 x 0.11 = 0.22 Green component = 0.45 - 0.11 = 0.34 Therefore, the formula provided a good approximation of the pigment components. For comparative purposes the absorption curve of the pigment extract from freshly captured Ulva green fish was rated as 100% green  pigment complex and that from the freshly captured red fish as 100% red  pigment complex. Since the value of the green pigment complex was 0.02 at 520 m^ j instead of zero, a correction factor was needed for optical densities between this value and 0.11 (already known to be correct from the mixture experiment). The following values were assigned on the assumption of a geometric progression: 56 Optical Density Correction 0.02 -0.020 0.03 -0.01 0.04 -0.005 (negligible) -0.00 It must be emphasized that the formulae only gives an approxima-tion of the pigment components. They will not detect small quantities of major pigment complexes. Furthermore, they are based on two un-proven assumptions! 1) That the major pigment components in all fish are the same (astaxanthin and dihydroxy e carotene). 2) That the minor pigments are never present in sufficient concentra-tion to influence the shape of the curve. For maximum accuracy all pigment extractions should be spectro-graphed at the same optical densities as those in the derivation experiment. This was not always possible in practice because some fish were low in total pigment and therefore not enough was available to obtain this concentration. Four general types of experiments were conducted: coloured trans-mitted light, coloured background, algae-habitat, and feeding. The last two types of experiments were undertaken when i t appeared that the fish in the first experiments were not acquiring new colour. Unfortunately, there was not sufficient time to conduct the latter types of experiments as long as desired. 57 1. Transmitted light experiments a) Green - A l l six surfaces of the aquarium were covered with f i l t e r s made from green gelatine, sandwiched between sheets of glass. Ninety percent of the v i s ib le light passing through the f i l t e r s had a wave length between 490 and 580 mjj,. Maximum transmission was near 540 imj. A white disc held inside the aquarium reflected light with a Munsell value of 5GY5/6. This is a l i t t l e further toward the blue and a l i t t l e further from the yellow than is Ulva green. b) Red - F i l ters were constructed as above. Ninety percent of the v i s ib le light passing through the f i l t e r s , exceeded 590 mjj, with the maximum near 700 m .^ The Munsell value for reflected light was 5R4/10. This is less purple than is the typical red phase of A. flavidus. c) Darkness - A l l sides were covered with dark opaque plast ic. d) White light - The aquarium was left uncovered and was illuminated by the light which passed through the translu-cent vinyl compartment covering. 2. Coloured background experiments The background colour was provided by coloured sand placed inside the tank. This was used, instead of placing colour on the outside of the glass, in order to prevent the reflection of white light from the glass bottom. Background colours ut i l i zed were: a) white b) black c) neutral grey 58 d) green (Munsell 5G6/6) e) red (Munsell 5R5/12) 3 . A l g a l - h a b i t a t experiments Since algae i s the natura l hab i ta t of A . f l a v i d u s during much of i t s l i f e , i t was thought the co lour response of f i s h i n algae might be d i f f e r e n t from that obtained under less natura l c o n d i t i o n s . One type of a lga was used i n each aquarium and renewed as often as required to keep i t f r e s h . In the red a lga experiment Porphyra was used and in the brown a lga experiment Laminaria was used. 4. Feeding experiment Only one feeding experiment was undertaken. The general problem with feeding experiments of th i s type, i s prov id ing the cor rec t food. If the carotenoid i s not the r ight one, or in the r ight form, i t may not be a s s i m i l a t e d . For t h i s experiment, I was able to obtain f rozen plankton which contained a good quanti ty of b r ight red gammarids. The pigment extracted from the gammarids was found to have an absorpt ion curve i d e n t i c a l with that of red A. f l a v i d u s . Disease - Disease was a ser ious problem. Heavy i n f e s t a t i o n s of the trematode Gyrodactylus were common during the spr ing and summer. Whether Gyrodactylus was the primary pathogen, or an oppor tun is t , i s not known; but i t was almost c e r t a i n l y the f i n a l cause of death of many f i s h . Poor water qua l i t y was probably a cont r ibu t ing f a c t o r . Temperatures were h i g h , and s a l i n i t i e s low, during the summer and furthermore, the aquarium water was being drawn from a sewage enriched a rea . Several species of f resh ly caught f i s h from the Lumberman's 59 Arch area were examined, and found with a few Gyrodactylus attached. Undoubtedly many could enter the aquarium water system on newly introduced f i s h . Others probably enter i n the water supply. No satisfactory cure was found for the disease. Treatment dosages strong enough to control the Gyrodactylus, usually k i l l e d the f i s h also. A variety of treatments was t r i e d , including baths of: forma-l i n , copper sulphate, d i s t i l l e d water, phenoxyethanol, merthiolate, methylene blue, b r i l l i a n t green, malachite green and pyridylmecuric acetate (PMA). PMA had recently been reported as a successful cure of this disease i n freshwater f i s h (Geibel and Murray, 1961) and appeared to be successful i n this instance since i t k i l l e d the organisms with l i t t l e apparent harmful effect. However, a pattern of d e b i l i t a t i o n , followed by death within a few weeks, was associated with otherwise successful PMA treatments. Results Changes involving the red pigment complex were r e l a t i v e l y easy to determine v i s u a l l y and microscopically, since the pigment was enclosed i n s u p e r f i c i a l chromatophores. Both chromatophore counts and disper-sion indices could be used as indications of pigmentation. On f i s h i n i t i a l l y green, newly developed erythophores gave the f i s h a fa i n t red tinge which was v i s i b l e even with the naked eye. Changes i n the green pigment complex were more d i f f i c u l t to determine since the pigment was dispersed freely i n the integument and no dispersion values could be assigned. Furthermore, i t was beneath the s u p e r f i c i a l erythrophores. 60 The results of the experiments are presented in detai l in the Appendix. The general trends are summarized in Table 11. The chromatic colour changes were supplemented by achromatic (melanophore) responses of two types: a) Physiological changes - These resulted in a general lighten-ing or darkening of the f i sh , and appeared to depend upon the albedo, and the light intensity, in the manner common to many fishes. b) Morphological changes - These changes, in conjunction with physiological changes, resulted in two types of disruptive patterns. On dark backgrounds rows of squarish black spots developed which contained up to twice as many melanophores as the surrounding tissue. On light non-matching back-grounds, rows of light areas occurred ("splotchy pattern") in some, but not a l l specimens, which resulted from loss of melanophores; and in the case of red and brown specimens, loss of erythrophores also. In a few cases both lightening and darkening processes occurred in the same f i sh. . ' . . Discussion Red phase A l l three colour phases had a general tendency to lose pigment in captivity, but this was most marked in the red phase. The loss was readily apparent on a neutral-grey sand background within 1-2 days. It was most rapid on well illuminated light-coloured backgrounds and least in transmitted red light and red algae. The loss was so 61 Table 11. Summary of colour change experiments indicating general trends Experiment Original Fish Colour Trend White Background Green Lighter. Moderate loss of green pigment. Red Lighter. Rapid loss of red pigment. Brown Darker. Possible loss of green pigment. Black Green Darker. Possible loss of green pigment. Dark Background disruptive spots form. Red Moderate loss of red pigment. Dark disruptive spots form. Brown Darker. Moderate loss of red pigment. Grey Background Green Red Brown Lighter. Some i n i t i a l loss of green pigment. Developed disruptive spots. Moderate loss of red pigment. Developed disruptive spots ( l ight) . Lighter. Slight change. Green Green Background Red Brown Slight loss of green pigment. Rapid loss of red pigment. Possible increase in green pigment. Developed splotchy pattern. L i t t l e pigment change. Red Background Green Moderate loss of green pigment. Red Moderate loss of red pigment. Brown Lighter. Table 11 (continued) 62 Experiment Original Fish Colour Trend Darkness Green Darker. Little pigment loss. Dark disruptive spots form. Red Splotchy. Moderate loss of red pigment. Brown Darker Transmitted Green Light Green Red Brown Slight loss of green pigment Rapid loss of red pigment. Possible acquisition of green pigment. Greener appearance. Transmitted Red Light Green Moderate loss of green pigment. acquisition of red pigment. Red Moderate loss of red pigment. Brown Acquisition of red pigment. Some Red Algae Habitat Green Darker.. Acquisition of red pigment. Red Some loss of red pigment. Brown Apparent increase in red pigment. Gammarid Feeding (17 days) Green Rapid acquisition of small amounts of red pigment. 63 rapid that i t overshadowed any physiological colour changes which may have occurred. The rate of loss appeared to be affected by the chro-matic quality of the environment as well as light intensity as indi-cated by the more rapid loss of red pigment in darkness than in red surroundings. The loss of red pigment in red fish under al l conditions coupled with the rapid appearance of erythrophores on green fish which were fed astaxanthin-rich gammarids suggests that, even in nature, red specimens require a constant dietary source of pigment in order to maintain their rich red colour. The faded colour of nesting red specimens (which are probably not feeding) appears to be a parallel situation. As the red fish faded, a green colouration appeared. This occurred first between the fin rays and on the chin. It resulted at least in part from unmasking of pigment already present (as shown by pigment analysis p.44). However, there are suggestions of green pigment gain in some of the experiments. Such gains, if real, were slight. No fish previously red developed any of the green colours observed in nature. The most changed specimen had been held for 104 days in transmitted green light and could at best be described as muddy green. In these experiments no dietary dihydroxy e carotene or closely related carotenoid is known to have been provided. Any integumentary gain in the basic green pigment would have to result either from redeposition of pigment stored elsewhere, or modification of other dietary carotenoids. Neither of these processes seems of significance in A. flavidus. It has not been determined whether the fish could become greener if a rich dietary source of the green pigment was 64 supplied. Green phase Green fish showed less tendency to lose pigment than red fish under comparable conditions, but more than brown fish. On a neutral-grey sand background initial loss was observable after 1-2 weeks. Sub-sequent loss was more gradual and even after several months on this background fish were recognizably green. Pigment loss was most rapid on light coloured backgrounds and exceedingly slow in transmitted green light, after an ini t i a l loss. Faded green fish appeared yellower than they were originally. This may have been merely a matter of pigment concentration, but could involve chemical changes, such as disassocia-tion of protein from the carotenoid. Green fish under two types of conditions developed visible ery-throphores. a) Standard diet and red environment. b) Neutral-grey background and astaxanthin-rich diet (gammarids). Although light quality influenced pigment deposition, diet is of the utmost importance. It appears unlikely that green fish on the standard diet would ever develop sufficient pigmentation to change into the red phase regardless of environment. The development of erythrophores was much more rapid in the 17 day gammarid feeding experiment even though the environment did not favour astaxanthin deposition. The extent of changes which might be possible over a longer period of time is unknown, but i t seems clear that i f a complete change into the red phase were possible, 65 an astaxanthin-rich diet would be required. Brown phase The pigmentation of brown A. flavidus was more stable than that of red or green specimens under comparable conditions. Loss was greatest on l i g h t backgrounds, only s l i g h t on neutral and dark backgrounds, and imperceptible i n the two f i s h held i n brown algae. Brown specimens held i n red or green l i g h t tended to s h i f t i n hue toward that of the l i g h t . Pigment analysis indicated a corresponding s h i f t i n pigment proportions; but there may also have been a change in the association of the carotenoid with the protein ( t h i s was not detectable with the procedures used). A complete change of phase would involve s i g n i f i c a n t l y more loss o± "non-matching" pigment and gain of "matching" pigment than occurred i n any experiment. I t cannot be concluded whether such changes would occur with the provision of a proper diet and environment. Conclusions 1) Physiological colour changes certainly do not account for the occurrence of the 3 colour phases of A. flavidus. Physiological changes involving the melanophores do occur i n a manner similar to that reported for most fishes and can provide some adjustment toward the background within each colour phase. 2 ) Morphological changes occurred which involved the gain and loss of integumentary carotenoids. 66 3) The rates of these changes were influenced by illumination but not completely controlled by i t . 4) The influence of diet is not known with certainty; but red pigment was deposited in the integument of two fish maintained on a neutral background. It appears likely that for complete retention of colouration, even in matching algae or transmitted light, dietary input of appropriate carotenoids is needed. Furthermore if com-plete colour changes from one colour phase to another can occur, i t appears that a diet rich in the appropriate carotenoids must be supplied. Redeposition of pigments stored elsewhere cannot produce such changes, nor does A. flavidus appear capable of comprehensive modification of different dietary carotenoids into integumentary pigments. 67 REARING EXPERIMENTS It was originally hoped to raise A. flavidus from eggs of known parents under a variety of dietary and lighting conditions, and perhaps even to back-cross offspring with parents. However, pairs with eggs proved difficult to find, the induction of spawning in captivity was equally difficult, and the raising of larvae to adulthood has not yet been possible. Nevertheless, larvae were kept alive long enough for chromatic pigments to develop and thus enable comparison with equiva-lent sized wild specimens. Methods and materials Several areas were sampled, including the regular collecting sta-tions. Since the fish do not appear to spawn much above lower low water, there were only a few opportunities to gather samples during the nesting season. The spawning period appears to extend from December through March. A. flavidus which were found with egg masses were placed in separate 20 to 40 gallon aquaria containing gravel, and supplied with running sea water. Artificial nests were built with rocks and bricks. Egg masses collected without parents were placed in gallon jars or floating crinoline baskets. They were aerated vigorously enough to gently tumble the eggs, in order to keep them free of dirt, and detritus and provide an adequate supply of oxygen. A variety of artificial habitats was utilized in attempts to in-duce spawning of adult fish, collected without eggs. 68 Considerable trial and error was involved in getting the newly hatched larvae to eat; but eventually a satisfactory procedure was developed. The yolk of hard boiled eggs was gently crushed in crinoline pouches and repeatedly dipped in a beaker of sea water in order to obtain a suspension of particulate yolk and sea water. This mixture was fed by means of a drip feeder, as the first food of the larvae. Slowly settling particles of yolk were provided for at least 16 hours during the first day. Newly hatched Artemia salina were introduced during the second day along with the yolk and usually were the preferred food by the end of the day. Once the fish were eating Artemia, the egg yolk was discontinued. Although the larvae switched very readily to Artemia from the yolk, they did not appear to readily accept Artemia as their first food. Results Larvae were hatched from several egg masses. Two batches were maintained long enough to develop significant chromatic pigmentation. Their development is summarized for comparative purposes in Table 12. Details are given below! Collection W-43-3 - February 3, 1962 at Second Narrows A pair of A. flavidus with an egg mass was found under a large asphalt slab at approximately the 0.0 ft tidal level: a) 5GY6/8; 296 mm T. L. b) 7-5Y7/10; 318 mm T. L. 6 9 During the portion of the nesting period that the f ish were in captivity (February 3 thru March 1), both parents tended the eggs, usually singly, but sometimes joint ly. One f ish would co i l around the egg mass and fan i t continuously with the posterior half of i ts body while the other lay nearby. If the egg mass escaped from under the cover, one or both f ish would immediately retrieve i t , usually by coi l ing around i t and worming back under cover. Neither parent would feed during this period, nor did they later appear to make any attempts to eat the newly hatched larvae. The larvae averaged about 13 mm upon hatching. A row of melano-phores along each side of the gut tube was the only evidence of pigment other than the b i le in the gal l bladder. The stages of pigment development are described in Table 12. It is important to note that the erythrophore pigment of the larvae was the same colour as that in the Artemia naupli i . The xanthophore-like masses were never abundant enough to contribute significantly to the colouration of the larvae. In the later larval stages there was a suggestion of a faint yellowish tinge on the head, but no xanthophores could be seen. The f ina l development stage is described in more detai l in Table 13. Artemia was provided on the f i r s t day, but not eaten. On the second day egg yolk and Artemia were both provided. Shortly after i t s ' introduction, egg yolk was seen in the stomach of many larvae; but, within 24 hours most were eating Artemia by preference. The presence of extoparasitic trematodes (Gyrodactylus sp.) was f i r s t discovered on day 33. The larvae were divided into several batches for treatment. Untreated larvae brist led with Gyrodactylus 70 Table 12. Development of pigmentation in reared A. flavidus. Col lection W-43-3 W-125-11 Egg Source Second Narrows Jordan River (laid in aquarium) Date February 3, 1962 January 9, 1963 Parents 5GY6/8 x 7-5Y7/10 5GY6/8 x unknown Hatching Date March 1, 1962 March 29 , 1963 Age (Days) Average Length (mm) Development Average Length (mm) Development 0 13 Transparent with 13 large melano-phores along gut tube. Transparent with large melanophores along gut tube. 5 Small melano-phores appear on occiput and pre-anal fin base. 10 Melanophores appear on dor-sal and anal fin bases. Melanophores appear along dorsal, pre-anal fin bases and on occiput. 14 14 Xanthophores appear on top of head (few 10-15 15 Xanthophores appear on head (few). 30 17 Pre-anal fin nearly resorbed. 38 19 Erythrophores uni-formly cover entire body. 50 25 Melanophores in-creasing over body except ventrally. Larvae settling to bottom. Head, dorsal and lateral surfaces covered with melano-phores . 7 1 Table 12 (continued). Age (Days) Average Length (mm) Development Average Length (mm) Development 68 26 71 72 81 28 30 Little change. 25 Little change. 29 Terminated. Erythrophores uniform over entire body. Pale yellow diffuse tinge on head. Melanophores in-creasing . Terminated. 72 Table 13. Description of skin from reared A. flavidus at the last developmental stage attained (dark orange, 25 - 29 mm T. L.), Dispersion indexes are adapted from Hogben and Slome, (1931) Locality Melanophores Erythrophores "Xanthopores" Guanine 2 2 No/mm Index No/mm Index Side 250 3 400 1.5 0 Yes Dorsal 450 4 600 1 0 Yes Dark Spots Dorsal 85 3 600 1 0 Yes Light Spots Silver 0 - 0 - 0 Closely Eye packed Streak crystals Black Contin- 3-4 6 1 0 Not Eye uous visible Streak 73 within a few days and died. Various treatments were utilized on the remaining larvae as mentioned on page 59. Freshwater and formalin baths were the most successful, but in each case resistant populations developed following initial successes and eventually overcame the remaining larvae. Pyridylmercuric acetate (PMA) was not utilized In any treatments. (Its use was unknown to me at this time.) Collection W-125-11 January 9, 1963 at Jordan River. A faded Ulva green specimen, 250 mm T. L. was collected along with 18 others under flat rocks in the lower intertidal zone (approx. 1.0 ft level). It was held in a large holding tank along with several other brown and green adults. Gravel, flat rocks, and bricks were provided to encourage nesting and spawning. The specimen in question spawned successfully with an unknown mate on, or just prior to February 8. The parent and egg mass were removed and provided with cover, in a 20 gallon wooden tank. Hatching occurred on March 27. The larvae were fed entirely on egg yolk for the first 3 days, then converted to Artemia naupli.i .-Pigmentation followed the same pattern of development as that observed for W-43-3 larvae except that the development of erythrophores and subsequent darkening was delayed. Gyrodactylus occurred as i t had done the previous year. However, since the pyridylmecuric acetate treatment was now known, i t was tried on small batches of A. flavidus larvae. It appeared to be very effec-tive, since i t killed nearly all the Gyrodactylus with no apparent loss of A. flavidus. These experiments were misleading, however. Further work with adult A. flavidus strongly suggests that PMA causes a severe 74 delayed mortality. This would appear to explain the mysterious death of several batches of larvae including W-125-11. It is also a possible cause of the slower development of pigmentation in this batch. Discussion The pattern of pigment development was similar for all larvae of both batches and resulted in a single colour phenotype. No "wild" red specimens of this size were available for comparison, but the reared larvae can be compared with wild kelp-yellow and Ulva-green specimens of similar size, (Table 14). Both colour phases of the wild larvae had abundant xanthophore-like masses, which were not apparent on the reared larvae. Although the kelp coloured larva had more erythrophores, than the reared specimens, they were less expanded and did not mask the underlying yellow-green pigment. The wild larvae were much more similar to the parental type of the reared fish than they were themselves. Based on the similarities between the wild larvae and comparable wild adults, i t would appear the development of chromatic pigments in the reared larvae followed a pattern determined more by diet than genotype. The Artemia pigment was identical in colouration to the erythrophores of the larvae, and was most likely its source. As previously stated, this pigment is most likely canthaxanthin. The A. flavidus larvae could have deposited unmodified canthaxanthin or could quite conceivably have converted i t by oxidation to astaxanthin prior to deposition, since canthaxanthin is considered to be a pre-cursor to astaxanthin (Lee, 1965). The yellow pigment of the larvae could easily be one of several Table 14. Description of skin from side of captured (wild) A. flavidus 20 - 25 mm T. L. Dispersion indexes are adapted from Hogben and Slome, (1931). Colour Melanophores Erythrophores "Xanthophores" 2 2 2 No/mm Index No/mm Index No/mm Index Kelp Yellow 300 1 1200 1 1000 3.5 Ulva Green 300 1 0 - 1000 3.5 76 carotenoids commonly found in egg yolk, and known also to occur in fish, or perhaps less likely, could have been derived from the Artemia. The phototactic behaviour of the A. flavidus larvae suggests an early li f e history similar to that reported for the gunnel Pholis  gunnellus (Linnaeus) by Bigelow and Schroeder (1953). P. gunnellus larvae, after hatching from the eggs, emerge from the nesting site and drift near the surface until they are 30-40 mm long at which time they sink to the bottom. During the plankton stage they are transparent except for a row of dorsal fin spots which develops when they are 25-30 mm in length. Summary 1) Larvae were hatched in aquaria from two separate batches of eggs and maintained for long enough to follow the early stages of pigment development. 2) The larvae from both batches developed almost identical pigmentation. The predominant chromatic pigment was contained in erythrophores which were the same colour as Artemia nauplii (orange). 3) They differed significantly in pigmentation from the parental types, which were kelp-yellow X Ulva-green and Ulva-green X unknown. 4) Kelp yellow and Ulva green larvae were found which were the same size as the reared larvae, but which were far more similar in colouration to the parents of the reared larvae than were the reared larvae themselves. 5) The larvae, almost transparant upon hatching, were positively 77 phototactic. Shortly after the formation of erythrophores began, the larvae settled to the bottom and underwent rapid development of melanophores and additional erythrophores. 78 FINAL DISCUSSION The evidence from field and laboratory studies suggests that the colouration of A. flavidus is utilized to provide concealment within the vegetation of its habitat. The colours of A. flavidus were usually found to be almost identical to that of plant cover available in its habitat, and the proportion of each colour phase collected varied with the colour of available plant cover. When i t was possible to observe A. flavidus directly in vegetation the fish and plant were found to match. In the laboratory when A. flavidus was given a choice of two colours of algae i t showed a preference for the matching one. A. flavidus appears to be well adapted in form, colour and behaviour for its existence within the vegetation. A. flavidus hides under rocks when nesting and apparently when resting at night. Individuals detected in plants during SCUBA obser-vations also tended to take cover under rocks. In the aquarium A. flavidus were found to prefer hiding places under rocks rather than within vegetation on rocks. It seems likely that the less secure cover of vegetation is utilized only during specific times and may be used primarily when A. flavidus is seeking food. Clearly physiological colour changes cannot account for the three colour phases of A. flavidus since each phase is morphologically different from the others in terms of its skin pigments. Physiological colour changes involving the melanophores did result in lightening or darkening and contributed to the formation of disruptive patterns. These changes supplement the adaptive colouration provided by the basic colouration. Although there were no complete colour changes from one 79 phase to another, adaptive morphological colour changes did occur involving the loss of non-matching pigment and the gain of matching pigment. Loss of pigment appeared to be influenced by the hue of environment as well as light intensity and albedo. Gain of a pigment appears to be favoured by a matching environment, but requires dietary input of the appropriate carotenoid. Feeding experiments were not extensive enough to determine the changes possible, but i f complete changes can occur i t is almost certain that a rich dietary source of pigments is required. A. flavidus demonstrated no ability to change colour by the redeposition of pigments stored elsewhere in its body, or to modify the dietary carotenoids not closely related to e carotene into integumentary pigments. It would appear that i f morphological colour changes are possible they would be slow and that non-matching individuals would occur in the field. This may account for the unusual red-grey specimens collected at Jordan River in which the underlying red colour was masked by expanded melanophores. The ability that A. flavidus demon-strated to produce disruptive patterns would also be of adaptive value in protecting imperfectly matched individuals. Such individuals might also make greater utilization of rock cover. A modifiable triphasic colouration system has recently been described in the intertidal isopod Idothea by Lee (1965). The three colour phases were caused by two basic pigments as in A. flavidus; but Lee was able to show that complete colour changes took place. A similar ability cannot be completely precluded in A. flavidus. Since complete colour changes were not demonstrated in A. flavidus 8 0 the possibility of genetic polymorphism must be considered. This can occur where two or more distinct environments are present at the same time and same place. Such a situation may be provided by the patches of red, green and brown algae in the habitat of A. flavidus. Cepaea nemoralis is frequently discussed as an example of such poly-morphism; but a situation more similar to that of A. flavidus is exemplified by Papilio machaon and Papilio polyxenes which develop brown or green chrysalia depending upon the colouration of the leaves and stems upon which pupation occurs (Sheppard, 1959). The fact that the colours of A. flavidus tend to form a continuum, rather than distinct phases, suggests that polymorphism may not be involved. Furthermore, the larvae reared from both batches of eggs were all the same and were more similar in colour to the Artemia nauplii than to the parental types and gave a definite impression of dietary rather than genetic influence. Wild larvae were captured which were more like green and brown adults than were the reared larvae. It may be that i t is the early feeding habits of the larvae which primarily determine the colour of A. flavidus. On the basis of this possibility and the information now available, the following life history is hypothesized. Young A. flavidus hatch and rise immediately to the surface where they remain as part of the plankton, until they are about 25 mm long. At this time they become negatively photo-actic, and settle onto benthic vegetation, where they begin feeding on colour adapted crustaceans. The pigments from the crustaceans either unchanged, or moderately altered, are deposited in the integu-ment of A. flavidus and become its initial colouration. A. flavidus 81 then prefers vegetation similar in colour to Itself. This hypothesis accounts for the colour phases of A. flavidus without invoking polymorphism or the ability to make complete changes from one colour phase to another. It does not rule out the possi-bility that colour changes can occur. SUMMARY Field and laboratory studies were undertaken to examine the basis and role of colouration in the penpoint gunnel Apodichthys  flavidus. The three colour phases of A. flavidus showed a general correlation with that of the vegetation. At Lumbermen's Arch, green specimens were the most common, while at Sooke and Jordan River, brown specimens dominated. By far the greatest proportion of red specimens were found at Jordan River where red algae were in far greater abundance than at the other 3 stations. Direct observations revealed that those A. flavidus utilizing vegetation for cover matched i t in colour. The suitability of a plant as cover for A. flavidus appeared to depend upon its size and form and to vary with the size of A. flavidus. For example Sargassum was found to harbour an abundance of post larvae, but not any larger specimens. A. flavidus was found to utilize cover under rocks when nesting and threatened and apparently during night and other periods of inactivity. It appears to use vegetation only at specific times such as while actively seeking food. In laboratory experiments A. flavidus showed an ability to select vegetation of matching colour, but preferred to hide under rocks rather than in algae if rocks were also available. 83 7« The colour differences observed i n A. flavidus were determined d i r e c t l y by differences i n pigmentation and not by different states of chromatophore expansion. Red specimens owed their colour to an erythrophore system containing primarily astaxanthin. The major pigment of green specimens was 3 , 3 ' dihydroxy epsilon carotene which was bound to a protein and freely dispersed within the integument. The brown f i s h examined contained a mixture of the two basic carotenoids found in red and green specimens. 8. Because of the morphological differences between colour phases, physiological colour changes cannot account for the colour variation in A. flavidus. 9. Morphological colour changes resulting from redeposition of pig-ments stored elsewhere were discounted as a possible cause as a result of the colour change experiments. 10. Change of colour phase with proper diet and environment may be possible, but A. flavidus probably does not have the a b i l i t y to r a d i c a l l y modify dietary carotenoids. 11. A l l the larvae reared from two batches of eggs from different parents developed the same pigmentation. Their colouration was i d e n t i c a l with that of the Artemia upon which they fed and different from that of the parents. On the other hand, larvae of the same size were captured i n the f i e l d with colouration similar to that of the parents of the larvae. 84 12. Polymorphism was considered unlikely since the colour of A. flavidus formed a continuum rather than distinct phases and because of the evidence provided by the larvae rearing experiment. 13. It is hypothesized that the colours of A. flavidus are initially determined by the early diet of newly settled larvae. 85 LITERATURE CITED Bascom, W. 1960. Beaches. Sci. Amer., 203 (2): 81-94. Bigelow, H. B., and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. Fish. Bull. 74, U. S. Fish and Wildl. 53: 1-577-Cheesman, D. F., and J. Preeble. 1966. Astaxanthin ester as a prosthetic groupJ a carotenoprotein from the hermit crab. Comp. Biochem. Physiol., 17: 929-935. Clemens, W. A., and G. V. Wilby. 1961. Fishes of the Pacific Coast of Canada. Bull., Fish. Res. Bed. Canada, No. 68, 1-443. Cott, H. B. 1940. Adaptive coloration in animals. Methuen, London, 1-508. Crozier, G. F., and D. W. Wilkie. 1966. Occurrence of a dihydroxy e carotene in a fish. Comp.. Biochem. and Physiol. (in press). Davies, B. H., Wan-Jean Hsu and C. 0. Chichester. 1965. The metabolism of carotenoids in the brine shrimp Artemia salina. Biochem. J., 94 (2): 26. Fox, D. L. 1953. Animal biochromes and structural colours. Cambridge, London, 1-378. Geibel, G. E., and P. J. Murray. 1961. Channel catfish culture in California. Prog. Fish-Cult., 23 (3): 99-105-Gilchrist, B. M., and J. Green. 1960. The pigments of artemia. Proc. Roy. Soc, London, Ser. B, 152: 118. Goodwin, T. W. 1952. The comparative biochemistry of the carotenoids. Chapman and Hall, London, 1-356. Hogben, L., and D. Slome. 1931. The pigmentary effector system VI. The dual character of endocrine co-ordination in amphibian colour change. Proc. Roy. Soc, London, Ser. B, 108: 10-53. Jordan, D. S. 1925- Fishes. Appelton, New York, 1-773. Kelly, K. L., and D. B. Judd. 1955. The ISCC-NBS method of designating colors and a dictionary of color names. NBS Circular 553-Krinsky, N. I. 1964. Canthaxanthin, the major carotenoid of the crustacean Artemia salina. Abstracts, VI Int. Cong. Biochem. I.U.B. 32: 582. 86 Lee, W. L. 1965. Pigments and color change, and their role in the ecology of natural populations of the marine isopod Idothea (P entidotea) montereyensis Maloney, 1933. Unpub. Ph D thesis Stanford, 1-191. Munsell, A. H. 1929. A color notation. Munsell, Baltimore, 1-74. Sheppard, P. M. 1959. Natural selection and heredity. Harper and Brothers, New York, 1-209. Walls, G. L. 1942. The vertebrate eye and its adaptive radiation. Bull. 19, Cranbrook Inst, of Science, Bloomfield Hills, Mich., 1-785. 87 APPENDIX PART I Description of Stations 88 APPENDIX Part I Description of stations St. M-10 Lumberman's Arch, Stanley Park, B. C. St. M-10 is a shallow bay surrounding Lumberman's Arch pool in Burrard Inlet, 1 mile South-East of First Narrows Bridge. Strong tidal currents up to 5-1/2 knots pass through the narrows, by-passing this bay for the most part, but causing counter-currents of various velocities. The area is well protected from the wave action of the open sea. Wash from the almost constant heavy shipping may be bene-ficia l to plants and animals remaining above the tide, especially during the very low midday summer tides. Tidal ranges taken from Pacific Coast Tide and Current Tables (1962) are given below. Mean tidal amplitude is given as 10.3 feet. Higher H W Lower L W Recorded Extremes Mean Large Mean Large Highest Lowest Tides Tides Tides Tides H W L W 12.4 14.2 2.1 -1.7 16.0 -2.6 Surface temperatures taken at the time of collections over a period of 20 months ranged from 36° F to 68° F. Surface salinities are variable. Greatest dilution occurs during early summer when the Fraser River reaches its maximum discharge. Many other streams and rivers emptying into the inlet contribute to the dilution. Salinities taken during collections ranged from 17-l^ to 89 27.0^. Turbidity was variable depending upon tide and season. Most periods of poor visibility occurred during the summer. Transparency ranged from 2 inches vertically to about 20 ft horizontally at a depth of 10 ft. Raw sewage from one of the main domestic sewers of Vancouver enters the inlet at Brockton Point, 5/8 of a mile east of the station. Material from i t was constantly encountered in the collecting area. Although i t may have been beneficial to the flora and fauna, i t discouraged the use of SCUBA for making observations and collections. The beach is mixed sand and boulders. Several inches of mud overlie,:, the sand in lower intertidal zone and upper subtidal zone during summer, but tends to be carried away during the winter. The bouldery nature of the sub-tidal zone continues offshore well beyond the collecting area. A light coating of si l t usually covers the entire substrate beyond the intertidal zone. Because of the boulders, the use of the larger seines was limited to an area opposite the west end of the pool. A profile of this region is given in Figure 1. The general plant zonation indicated in Figure 1 varies with sub-strate and season. Ulva is generally absent from the main seining area; while Zostera and Laminaria dominate. The Zostera becomes dirty in appearance and covered with a red-brown epiphyte during the summer. Although the general zonation pattern can be made out, the zones are not homogenous. About October of both 1961 and 1962 the vegetation began to decline. Moderate cover remained in most of the collecting areas throughout the winter of 1961-62. Vegetation became sparse during the winter of 1962-63 following severe autumn storms. The most Ulva , Sargassum , Laminaria-Alaria • ( Zostera i • Red alaae Mean high tide 12 Drain l l - 6 0 6 I depth - Tidal 12 i i i • J 80 40 0 40 i Distance in feet from zero tide level 30 120 160 Scale Vertical 1" =20' Horizontal 1" =40 Appendix Figure 1. Profile of Lumberman's Arch collecting station M-10 showing general floral zonation during summer of 1962. 91 damaging was a destructive gale on October 12 which had winds with velocities up to 70 m.p.h. New algae formed good cover by the end of March. Maximum abun-dance appears to be reached by July. St. M-11 Lumberman's Arch, Stanley Park, B. C. Station M-11 is a bouldery point, exposed during low tide, at the eastern end of the same bay which encloses M-10. The bouldery flat of the intertidal zone drops quickly from the -1 ft level to the muddy bottom of the bay. This station is more exposed to the main tidal currents than station M-10, especially on the north face of the drop-off. Laminarians become very abundant and dominate the lower intertidal flat. Nereocystis, Ulva and red algae also occur. Zostera is absent. Mixed green, brown and red algae occur down the boulder slope to its base, a depth generally not exceeding 6 feet at zero tide. St. M-12 Second Narrows, Vancouver, B. C. This collecting station is located on the north shore of the narrows in the region of the "Old Second Narrows" Bridge, just west of the mouth of Seymour River. In this area, wave action is slight, but tidal currents are strong, attaining velocities up to 6-1/2 knots during flood tide and up to 5-1/2 knots during ebb tides. (Pacific Coast Tide and Current Tables [1962]). A profile Is-'given in Figure 2. i '. :v • ? , •• J The substrate of the intertidal flat is gravel and boulders over fine material. Slabs of asphalt up to 10 square feet l i e in the lower Zones 1 2 3 4 .Ulva. , Brown algae Red algae i Mean high tide 12 6 0 - 6 12 18 i i • i i 24 40 20 0 20 40 Distance in feet from zero tide level 60 Scale 1" - 20' Appendix Figure 2 . Profile of Second Narrows collecting station M - 1 2 , illustrating zonation. 93 intertidal zone near the bridge, modifying the cover. The sub-tidal slope is covered with gravel and boulders up to 6 inches, and scattered larger rocks seldom exceeding 12 inches. The substrate in general is quite stable and unmoved by tidal currents or wave action. Floods of the Seymour River undoubtedly result in periodic distur-bance of habitat. However, in the area near the bridge the habitat remained relatively unchanged during the period of study. Salinities and temperatures vary with tides and season. Fresh water sweeps over the collecting area during the outgoing tide especially when the river is in flood. The extent of the dilution is not known since only a few observations were made at this station. The temperatures and surface salinities taken under the bridge during collections are given below. Date Tide Temperature ° F Salinity-, cy Feb. 2, 1962 0.4 ft 44 24.7 Feb. 3, 1962 -0.2 ft 43 24.4 Mar. 4, 1962 0.7 ft 42 23.8 Sept . 23, 1962 2.0 ft 55 21.8 Oct. 8, 1962 2.8 ft 52 23.7 Nov. 16, 1962 0.5 ft 47 16.2 Dec. 10, 1962 -0.6 ft 44 Mar. 9, 1963 6.5 ft 49 . The plant zonation observed during October is given in Figure 2. It was essentially the same during subtidal collections in March, 94 September and November. The most salient feature of the sub-tidal zone at this station is abundance of red algae and its proximity to shore at low tide. The genera present were identified as: Iridaea, Gigartina and a red laver, probably Porphyra, with the red laver dominant. St. VI-5 Jordan River, Vancouver Island, B. C. Jordan River Is located 30 miles east of Cape Flattery on the Vancouver Island side of Juan de Fuca Strait. The strait is about 15 miles wide and runs easterly from the Pacific Ocean. The northern shore is predominately steep and rocky with gravel beaches in semi-exposed areas and a few muddy or sandy bays in more protected areas. Being close to the open ocean and in an area where the bottom slope is gradual, the exposed beaches in the Jordan River area are subject to strong and almost constant wave action. A half-mile east of the river mouth lies a shallow mud and sand bay which is partially protected by a man-made boulder breakwater along the western shore. The breakwater is a continuation of the foreshore intertidal flat and is covered by water during high tide. The area along the bayward edge of the breakwater was the principal collecting region. Inter-tidal pools were also sampled. Collections could not be made along the exposed southern foreshore because of its bouldery nature and almost constant wave action. The bay is subject to dilution from surface run-off and river water. Salinities taken during collections.ranged from 28.6^ , to 31.1%0; temperatures from 40 to 58° F. Tidal ranges are given below. 95 H H W L L W Extremes Mean Tides Large Tides Mean Tides Large Tides Highest H W Lowest L W 9.8 11.5 3.1 0.9 12.4 0.0 Turbidity was variable, with visibility ranging from essentially zero, to several feet horizontally, depending upon the tide and wind. The vegetative pattern at this collecting station can be seen from the profile in Figure 3. Green algae (Enteromorpha and Ulva) are found on the loose boulders of the mid-intertidal zone. The bulk of vegetation grows below this zone on a band of boulders packed with sand which extends to the muddy bottom of the bay. The packed nature of this bottom provides few hiding spaces under the rocks. At the centre of the seining station this band lay between the 1.5 and 3.5 foot tidal levels, during 1961 and 1962. Plants living in this collecting area are subject to catastrophic destruction by storms; and dessication during low tides, particularly on hot summer days. The floral life span and distribution patterns likely vary considerably from year to year. In October of 1961, the first time this station was sampled, vegetation was scarce in the collecting areas except for the con-sicerable quantities washed up on the beach. A few patches of Phyllospadix occurred toward the mouth of the bay and sporadically along the breakwater. During the winter and early spring algae remained sparse. By April of 1962 Laminaria and Egregia were abundant and • Vegetation Zone . (Mixed) i Mean_ Wg_h_tide_ • Boulder jetty-Packed Boulders —112 9 6 3 xi +-> a cu "O •—i ro T3 Mud 20 10 0 10 20 30 Distance in feet from low tide intercept 40 Scale !":= 10' Appendix Figure 3. Profile near centre of collecting area at Jordan River, Station VI-5. 97 dominated the lower intertidal band. Ulva was present here and in the loose boulder zone above. Phyllospadix continued to be patchy. Red algae were present mostly along the upper margin of this band but were not well developed. Vegetation had increased by the end of May. Red algae were now abundant throughout lower intertidal band. An approximation of the percentage of ground covered by the various species at this time is shown below. The dominants are indicated by asterisks. Red Algae Brown Algae Green Algae Phyllospadix Pornhyra* Laminaria* Ulva Rhodvmenia Egregia Iridaea 257o Alaria 40% 57c 307c Agardhiella Nereocystis Gigartina Little change in vegetation proportions was noticed in June. Collections were made on July 1 8 , the day of the lowest predicted tide of the year, ( 1 . 2 ft at Sooke). As low tide approached i t appeared that the vegetation had remained essentially unchanged from June. However, with prolonged exposure to the hot sun, the plants became dessicated and it appeared that many might not recover. Red algae, Egregia and Phyllospadix seemed to be the most adversely affected. Subsequent observations during September revealed considerable decrease in Egregia and shallow red algae with a lesser decrease in laminarians. 98 Phyllospadix and Ulva seemed to have recovered almost completely from the effects of exposure. Vegetation was much sparser in October, approaching the condition of the preceding October. Station VI - 8 Agate Beach, Vancouver Island Agate Beach is located five miles west of Sooke and 40 miles east of Cape Flattery on Juan de Fuca Strait. The collecting station lies in an indentation between two rock outcroppings approximately 250 yds. apart. The beach, which is covered with coarse gravel, is partially protected from wind and wave action during the prevailing westerly winds by headlands to the west and a small island (Agate Rock), 125 yards offshore. However, i t is subject to heavy wave action during storms, particularly those blowing from the southeast, and is devoid of plant cover. Pacific Coast Tide and Current Tables (1962) indicate higher high water mean tides of 9.3 ft and large tides of 10.9 ft with highest high water recorded at 12.2 ft. Lower low water mean tides of 3.1 ft are indicated, with large tides of 0.8 ft and lowest low water recorded is -0.2 ft. Mean tidal amplitude is 6.2 ft as opposed to 10.3 ft at Lumberman's Arch. Surface temperatures, recorded at the time of collections ranged from 40 °F to 52°F. Temperatures recorded at the 30 ft level ranged between 42° and 52o. Surface salinities varied from 27.3^0to 32.5%0. Clarity of the water was not measured, but was generally good. Hori-zontal visibility at the fifteen foot level was never observed to be less than fifteen feet. Because of its gravelly nature the profile of the beach is 99 constantly changing. In general, i t follows the pattern described by Bascom (I960). During summer the berm is built up and the prof i le steepens. In winter the heavier surf removes material from the berm and deposits i t at the base of the intert idal zone. A prof i le of the beach in intermediate state is shown in Figure 4, The instabi l i ty of the intert idal zone at Agate Beach prevents the growth of algae except for a slimy growth in the lower intert idal zone during summer. The gravel of the subtidal zone is generally coarser than that of the intert idal zone and contains a few boulders and occasional out-croppings of base rock. A layer of s i l t , thickening into mud with increasing depth, builds up into the lower intert idal zone during f a i r weather and recedes during stormy weather. The general f l o ra l zonation pattern can be seen in Figure 4, Alar ia and Laminaria dominate the subtidal zone. Nereocystis is abundant wherever rock outcroppings occur. Other brown algae include Egregia, Sargassum, Graci lar ia and Costaria. Ulva becomes common at the upper limit of the subtidal zone during summer, and forms a margin with a species of red alga possibly Iridaea. Their occurrences were short lived in this very unstable region. Some red algae are found in the Laminaria zone, usually attached to the rock outcroppings and larger boulders. Phyllospadix occurs in patches, and may occupy up to 57a of the outer half of the algae zone. In the summer the f lo ra l layer may reach 3 feet in thickness. During the f a l l and winter of 1961-62 the f lora was gradually reduced to about 25% of the peak crop. The growth during the summer of 1962 was similar to the previous year; but a severe storm during October 1962 left the bottom almost devoid . P h v l l o s D a d i x I Brown a l g a e • G r e e n a l g a e Red a l g a e - v . M e a n h i g h t i d e - 12 6 Ldal depth 0 6 Ldal depth 12 H 1 • i i • • 18 80 40 0 40 80 120 D i s t a n c e i n feet from z e r o t i d e l e v e l 160 S c a l e V e r t i c a l 1" = 2 0 ' H o r i z o n t a l 1" = 4 0 ' Appendix Figure 4. Profile of Agate Beach collecting station Vl-8. 101 of flora, a condition in which it virtually remained until the follow-spring. Obviously this beach differs markedly from Lumberman's Arch Beach (Station M-10) in several aspects, the most apparent being exposure, and the absence of estuarine conditions. Collections and observations were also made at Agate Rock. This jagged outcropping lies in water ranging from a minimum of 20 ft on the inner side to a maximum of 55 ft on the outer side as measured from tide level zero. Pieces of broken rock lie at the base of its precipitous sides and are built up to form slopes in some areas. Codium fragile and Phyllospadix occur on the flatter slopes of the lower intertidal zone and a short distance below. Patches of Halosaccion occur here also. Coralline algae are abundant in the subtidal zone down to a depth of about 15 feet. Crustose and small foliose red algae continue to depths of approximately 50 feet. Laminarians are abundant in some areas between the 5 and 15 foot levels, particularly on the eastern boulder slopes. A rich bed of Nereocystis surrounds the island especially on the offshore sides during much of the year. Most of the holdfasts are attached between the 20 and 50 foot levels. The bottom surrounding the island is muddy. Supplemental collecting stations St. VI - 9 Muir Point, Vancouver Island, B. C. Muir Point lies at the mouth of Sooke Harbour and is 2 miles east of Station VI-8. The beach is solid rock, covered in places with a 102 mixed layer of gravel and sand. Boulders up to several feet in diameter overly the beach, which slopes gently off-shore. A combi-nation of good substrate and moderately heavy wave action promotes a luxuriant and varied growth of algae, which extends over 200 feet off-shore during the summer. A natural spit (Whiffenspit) of mixed sand, gravel and rock has built up at the mouth of the harbour at the east end of the beach. Tides, temperatures and salinities are similar to those given for Agate Beach. Since this area was unseinable shore collections were made with a dip-net and with rotenone. Off-shore collections were made with rotenone and the aid of SCUBA. St. VI-19 Saxe Point, Vancouver Island, B. C. Saxe Point is located in Esquimalt near the entrance to Victoria harbour. A steep rocky shore runs north from the point and drops 10 to 20 feet, to a muddy bottom. Brown algae forms good cover along the base of the rock wall. A large bed of Nereocystis remains during the spring, summer and early f a l l at the south face of the Point. Subtidal observations and rotenone collections were made by using SCUBA at this station. St. VI-21 Harling Point, Victoria, B. C. Harling Point lies 5 miles east of Saxe Point. It is a low pro-tuberance of solid rock fully exposed to the wave action of Juan de Fuca Strait. Collections were made in tide pools and rocky channels between the 0.0 and 3 ft tide levels. The tide pools and channels were generally free of loose substrate except for occasional rocks and pockets of broken shell. Mean tidal amplitude at Victoria is 103 5.7 ft. Tide levels (1962, Tide and Current Tables) are given below. HHW LLW Recorded Extremes Mean Large Mean Large Tides Tides Tides Tides HHW LLW 8.5 9.6 2.8 0.3 12.1 -1.5 Currents up to 3 knots sweep by the point. Temperatures and salini-ties were similar to Agate Beach. Phyllospadix, laminarians and Nereocystis were particularly abundant in the collecting areas during the spring and summer collec-tions. No underwater observations were made in this area because of the occurrence of a raw sewage outlet in the immediate area. 104 APPENDIX PART II F i e l d Collections Appendix Table 1. Lumberman's Arch seine collections of fish greater than 50 mm T.L. Collection Date Net Temp. Approximate (hauls) (ft) (F) tidal depth (ft) Green Colour Olive Brown Red 63.1 (3) W-27.1 (4) W-133 (4) W-54 (3) W-137 (3) W-141 (12) W-64 (7) W-65 (6) W-66 (3) W-67 (2) W-68 (10) W-142 (8) 62-40 (2) 1:5:63 1:8:62 Night 3:12:63 3:16:62 Night 3:22:63 Night 4:27:63 5:5:62 5:6:62 5:10:62 5:13:62 5:19:62 5:24:63 6:3:62 50 10 50 50 75 10 10 10 10 50 Dip 10 75 47 45 48 49 52 51 51 60 62 55 -2 to +8 + 1 to +4 -6 to +4 -8 to +2 -4 to -1 -3 to 0 -3 to +1 -1.5 to +1 -4 to +4 0 to +3 -4 to -1 -8.5 to -0.5 0 0 0 0 3 2 0 4 4 1 3 2 0 0 0 0 3 1 1 2 1 0 2 4 0 0 1 0 4 1 1 1 2 1 0 3 0 0 0 0 0 0 1 0 0 0 0 0 Appendix Table 1 (Continued). Collection Date Net Temp. Approximate Colour (hauls) (ft) (F) tidal depth Green Olive Brown Red (ft) W-79 (6) 6:16:62 10 57 -1 to +1 0 0 0 0 62-41 (3) 6:21:62 50 61 -9 to 0 4 0 2 0 L-l (2) 6:30:62 50 - -2 to +9 1 0 - 0 0 Night W-144 (15) 7:8:63 10 - -2.8 to -0.8 4 0 1 0 W-145.1 (9) 7:21:63 10 - -2.3 to -0.3 2 0 1 0 W-145.2 (2) 7:21:63 50 - -9 to 0 4 0 3 1 W-146 (6) 7:20:63 10 - -3 to -0.8 5 0 1 0 62-45 (3) 7:14:62 75 59 -5 to +5 1 0 3 0 W-86 (3) 7:22:62 . 1 0 62 +5 to +7 0 0 0 0 W-87 (3)* 7:28:62 10 59 -2 to -0.5 16 0 2 0 W-95 8:8:62 75 60 -5 to +5 0 0 0 0 Night 62-50 (2) 8:18:62 50 - -1.3 to -9.3 5 3 1 0 W-96 (3) 9:9:62 50 58 -7 to +3 8 5 2 1 Appendix Table 1 (Continued). Collection Date Net Temp. Approximate Colour (hauls) (ft) (F) tidal depth Green Olive Brown Red (ft) W-103 (5) 9:15:62 10 61 -1 to +2 7 1 0 0 W-106 (3) 10:7:62 50 53 -7 to +3 11 0 3 2 W-108 (3) 10:12:62 75 53 -8 to +2 0 0 0 0 Night W-10 (4) 11:5:61 50 49 -4 to +6 4 2 0 0 62-63 11:10:62 50 52 -0 to +6 0 0 1 0 W-15 (2) 11:19:61 50 48 -2 to +10 0 0 0 0 Totals 94 25 36 5 Appendix Table 2. Lumberman's Arch collections of fish less than 50 mm T.L. Collection Hauls Date Net Colour of Fish Green Yellow Brown green*  W-64 W-64 W-65 W-66 W-66 W-68 W-68 W-68 W-68 W-79 W-79 6 6 10 10 10 10 2 2 5:5:62 5:5:62 5:6:62 5:10:62 5:10:62 5:19:62 5:19:62 5:19:62 5:19:62 6:16:62 6:16:62 10' 10' Dip Dip Dip Dip Dip Dip Dip 10' 10' 2 0 1 0 0 0 3 1 2 0 1 1 0 1 6 0 1 4 0 0 0 1 0 3 7 0 0 5 0 1 Habitat 70% Zostera 30% Sargassum and Laminarians 90% Zostera 10% Browns Sargassum 90% Zostera 50% Zostera, 50% Browns Sargassum Laminarians 90% Zostera Mixed 90% Zostera Mixed Totals 10 14 19 * Yellow green fish appeared "kelp" coloured when on Sargassum. Appendix Table 3. Second Narrows shore collections. Collection Date Species Seen Captured Distribution (ft) Remarks Total length (mm) W-41 W-42 W-43 W-48 W-53 2:2:62 Anoplarchus Many 0 Pholis Several 0 2:2:62 Anoplarchus Many Pholis Pholis ca. 12 2:3:62 Anoplarchus Many Pholis ca. 12 A. flavidus 7 2:6:62 Anoplarchus Many ca. 20 3:4:62 Anoplarchus Many Pholis 1 A. flavidus 1 0 0 0 5 7 13 0 1 1 0.4 to +1 0.4 to ? 0.4 to +1.5 0.4 to ? -0.2 to +1 -0.2 to +0.5 -0.2 to +o.5 One pair with 170, 250 Not as high as Anoplarchus Not as high as Anoplarchus + 0.5 eggs. Higher than Pholis. Three Pholis 174, 250 223, 309, 324 egg masses. 0.7 to +2.0 30-50 with eggs < 0.7 With eggs < 0.7 170 Appendix Table 3 (Continued). Collection Date Species Seen Captured Distribution Remarks Total (ft) length (mm)  W-115 11:16:62 A. flavidus 2 pair 2 pair 0.4 Other spp. not 272, 323, recorded 310, 325 W-119 12:10:62 A. flavidus 3 3 -0.6 to 0.0 Other spp. not 154, 200, recorded 210 Appendix Table 4. Second Narrows seine and SCUBA collections of A. flavidus. Collection Date Net Temperature Tidal depth (ft) Colour Green Brown Red Remarks W-104 (1) W-104 (1) W-104 (1) W-115 (1) W-131 9:23:62 50 9:23:62 50 9:23:62 50 11:16:62 10 3:9:63 Hand 55 55 55 47 49 -10 to +2 0 0 0 -10 to +2 0 1 2 -20 to +2 0 0 2 -2.5 to +0.5 2 7 0 -16 0 0 2 Net fouled Not out to red algae SCUBA and quinaldine Totals Appendix Table 5. Jordan River seine collections of fish greater than 50 mm T.L. Collection Date Net Temp. Approximate (hauls) (ft) (F) tidal depth (ft) Green Olive- Brown Red -Red-gray W-35 (2) W-36 (6) W-60 (8) W-69 (8) W-70 (3) W-74 (4) W-82 (6) W-83 W-99 (10) W-5 (4) W-110 (9) 1:23:62 1:23:62 4:26:62 5:23:62 5:23:62 6:20:62 7: 18:62 7:18:62 9:12:62 10:5:61 10:28:62 50 10 10 50 50 10 10 Rotenone 10 52 10 51 10 51 40 40 50 52 50 57 58 -1 to +8 + 5 to +8 + 1 to +3 +1 to +3 +1 to +4 +1 to +3 +1 to +3 Tide pool +1 to +3 +3 to +6 + 1 to +4 0 0 7 13 1 3 1 2 15 0 0 0 0 6 11 2 1 0 0 3 0 0 0 0 6 31 1 13 21 0 13 0 0 0 0 3 25 2 22 3 7 5 0 0 0 0 0 11 0 0 0 0 0 0 0 Totals 42 23 85 67 11 Appendix Table 6. Agate Beach seine collections. Collections (hauls) Date Net Temp. Approximate (ft) (F) tidal depth (ft) Green Colour Olive Brown Red W-6 (2) W-7 (1) W-8 (2) W-lll (2) W-112 (2) W-13 (3) W-25 (1) W-26 (2) W-27 (1) W-38 (3) W-40 (1) W-51 (3) W-52 (1) W-55 (5) 10:5:61 10:6;61 10:6;61 10:28:62 10:29:62 11:9:61 12:20:61 12:20:61 12:20:61 1:23:62 Night 1:24:62 2:28:62 3: 1:62 3:21:62 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 51 50 51 45 43 43 40 44 40 47 •12 to +6 •13 to +5 •12 to +7 •11 to +6 •11 to +8 •11 to +9 •11 to +8 • 4 to +7 •13 to +3 • 14 to +4 •12 to +7 •12 to +5 • 6 to +4 •12 to +7 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 5 0 8 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 Appendix Table 6 (Continued). Collection (hauls) Date Net Temp. Approximate (ft) (F) tidal depth (ft) Green Colour Olive Brown Red W-57 (1) W-58 (2) W-59 (1) W-61 (1) W-71 (2) W-72 (2) W-73 (1) W-75 (3) W-76 (3) W-81 (1) W-85 (2) W-88 (3) 4:25:62 4:25:62 4:25:62 Night 4:27:62 5:23:62 Night 5:24:62 5:24:62 6:20:62 Night 6:21:62 7:17:62 Night 7:20:62 8:21:62 Night 50 50 50 50 50 50 50 50 50 50 50 50 48 48 48 50 50 50 50 49 -14 to +3 -12 to +5 -12 to +7 - 4 to +5 -11 to +7 -15 to +2 -12 to +6 -11 to +8 - 6 to +4 -12 to +7 •15 to +2 •11 to +8 0 1 1 3 0 2 8 1 3 0 3 1 0 0 0 1 0 3 0 0 0 0 1 1 4 4 1 3 0 6 7 0 7 0 5 1 0 0 0 0 0 1 0 0 0 0 2 0 Appendix Table 6 (Continued). Collection (hauls) Date Net (ft) Temp. (F) Approximate tidal depths (ft) Green Colour Olive Brown Red W-89 (1) 8:22:62 50 49 -12 to +6 2 3 4 0 W-98 (2) 9:11:62 50 52 -12 to + 4 9 0 1 0 Totals 14 9 59 4 Appendix Table 7. Agate Beach Rotenone collections. Collection Date Location Tidal depth (ft) Habitat Rotenone Significant specimens W-89 W-90 W-124 !:22;62 Agate Rock -35 to -45 8:22:62 Seining Site - 4 to 0 W-113 10:24:62 Agate Rock ca. -30 1:9:63 Agate Rock ca. -40 W-126 1:10:63 Agate Rock ca. -25 Boulders 1/2 gal Little vegetation Ulva 1/2 gal Iridaea Laminarians Red lavers 1/2 gal Boulders Red lavers 1/2 gal Boulders Red lavers 1/3 gal Corallines Boulders 3 Anoplarchus 1 green A. flavidus 5 Anoplarchus 6 Anoplarchus 2 Anoplarchus Appendix Table 8. Muir Point collections. Collections Date Method Tidal depth Habitat A. flavidus (ft) G 0 B R W-12 11:8:61 Shore +2.5 to 3.5 Rocks, mixed 0 0 0 0 flora. W-17 12:6:61 Shore +2.5 to 3.5 Rocks, mixed 0 0 0 0 flora. W-23 12:19:61 Rotenone +2 to 3 Rocky tide 0 0 0 0 pool. W-136 3:17:63 Rotenone -2 to +1 Rocks. Flora 0 0 4 7 rich; red and brown. Appendix Table 9. Harling Point collections of A. flavidus. Collection Date Method Tidal depth (ft) Colour G 0 B R Habitat Comment W-21 12:18:61 Rotenone 2.5 to 3.5 0 0 0 0 Mixed red, brown and green flora. Abundant. W-80 7:17:62 Seine -1.5 to 0.5 0 0 2 0 Mostly green and brown flora. Abundant. Seining inefficient. W-80 W-139 7:17:62 Rotenone 0 to 1.5 0 0 10 1 3:30:63 Rotenone 1 to 3 3 0 0 1 Red, brown, green flora. Abundant Red, brown, green flora. Abundant. Connected tide-pool. 1 green with overlaying red tinge. Totals 3 0 12 2 Appendix Table 10. West Vancouver and Howe Sound Rotenone collections. Collection Date Tidal Depth Flora A. flavidus Pholis Anoplarchus (ft) BC 62-317 8: 62 0 0 6 W-94 9:5:62 -12 to -17 Red lavers; 0 1 1 moderate. W-109 10:20:62 - 9 to -12 Red lavers; 0 0 18 l ight. W-114 11:8:62 ca. -21 Red lavers; 0 0 12 l ight. W-118 12:6:62 ca. -20 Red lavers; 0 0 13 light W-128 3:2:63 ca. -13 Laminarians, 0 0 3 red lavers; moderate W-140 4:20:63 -15 to -25 Browns and 0 0 0 reds; abundant. Totals 0 1 53 Appendix Table 11. Summary of Lumberman's Arch SCUBA Diving observations. Collection Date Area Habitat Tidal depth (ft) Remarks W-18 W-14.5 11:25:61 12:16:61 Offshore from Station M-10 Station M-10 Good cover. Rocky; -14 to -11 red, green and brown algae present. Cover light. -10 to -8 Sandy; some rock. Mostly brown algae. No A. flavidus seen. No A. flavidus o Pholis seen. W-29 1:13:62 Station M-11 Rocky. Moderate to light algal cover on rocks. •15 to 0 Six Pholis seen under rocks. On captured. W-56 3:30:62 Station M-11 Bouldery slope. Good algal cover, mixed. •6 to 0 About 12 A. flavidus seen in algae; one cap-tured, (green). One Pholis seen under rocks. 121 Appendix Table 12. Jordan River collections of A . flavidus under 50 mm. Collection Date Colour (hauls) Green Brown Red W-69 (8) 5:23:62 0 6 0 W-70 (3) 5:23:62 0 1 0 W-74 (4) 6:20:62 2 5 1 W-82 (6) 7:18:62 0 4 1 W-83 (0) 7:18:62 1 0 2 Totals 3 16 4 122 APPENDIX PART III Results of co lour change experiments Appendix Table 13. Results of holding A. flavidus under white daylight over a white background. Colour notation in this and subsequent tables are from Munsell,3 1929. Numb er Size (mm) Original Colour Final Colour Days Visual Interpretation 3 4 164 189 101 103 215 169 5GY5/8 2.5GY4/4 5R2/2 7-5R4/3 2.5Y6/6 2.5Y5/6 2.5GY7/6 2.5GY8/10 2.5YR6/8 7.5R6/10 Splotchy 2.5Y7/8 2.5Y7/8 48 24 48 48 24 48 Loss of green pigment. Lighter Lighter, probably l i t t l e green pigment change. Loss of red pigment. Loss of red pigment. Lighter, probably l i t t l e pigment change. Lighter, probably a l i t t l e pigment change. Appendix Table 14. Results of holding A. flavidus under white daylight over a black gravel background. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 138 5GY5/8 107 5GY5/8 189 5GY5/8 142 5GY5/8 126 5GY5/8 5GY4/6 26 1 row moderate dorsal dark spots. 5GY4/6 26 2 rows moderate dark spots • 5GY5/6 26 1 row moderate dorsal dark.spots. 2.5GY4/4 24 2 rows dark spots. 2.5GY4/4 24 2 rows dark spots. Possible loss of green pigment but generally darker i n appearance because of increase i n dark spots. Possible loss of green pigment but generally darker i n appearance because of increase in dark spots Possible loss of green pigment but generally darker i n appearance because of increase in dark spots. Possible loss of green pigment but generally darker i n appearance because of increase in dark spots. Possible loss of green pigment but generally darker i n appearance because of increase i n dark spots. Appendix Table 14 (Continued). Number Size (mm) Original Colour Final Colour Days Visual Interpretation 107 95 164 5GY5/8 5GY5/8 5R3/8 2.5GY4/4 24 1 row dorsal dark spots. 2.5GY4/4 Uniform 24 5R5/4 26 1 row dorsal dark spots. Possible loss of green pigment but generally darker in appearance because of increase in dark spots. Possible loss of green pigment but generally darker in appearance because of increase in dark spots. Loss of red pigment. 258 2.5Y5/6 10Y5/6 18 Loss of red pigment. Appendix Table 15. Results of holding A. flavidus under white daylight over a neutral gray background [N/6(Munsell)]. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 105 92 255 168 126 97 290 5GY5/8 5GY5/8 2.5GY5/6 5GY6/4 Splotchy 5R3/.6 5Y5/6 1 row dorsal dark spots 5Y5/6 2.5GY6/6 100 2 rows of dark spots 2.5GY6/8 62 2 rows of dark spots 5GY6/4* 30 10Y6/6 49 Splotchy 2 rows of dark spots 5R5/4 60 Splotchy 1 row dorsal dark spots 5Y6/8 30 1 row dorsal dark spots 2.5Y5/6 100 Lighter and yellower. Lighter and yellower. Lighter and yellower. Loss of green pigment. Possibly some loss of green pigment. Loss of red pigment. Insignificant change. Became slightly yellower. 178 7-5Y5/6 2.5Y6/8* 100 Slightly yellower. Appendix Table 15 (Continued). Number Size Original Colour Final Colour Days Visual Interpretation (mm)  9 247 7-5YR5/6 7-5YR5/6* 81 No significant change. 2 rows of dark spots 2 rows dark spots 10 206 10YR4/4 10YR4/4* 6 No significant change. * Pigment analyzed. Appendix Table 16. Gross histology and pigment analysis of fish from neutral gray background. Fish Number Size (mm) Colour Melanophores Index No/mm^  Erythrophores Index No/mm^  Green Pigment Distribution Pigment Analysis % Red % Green 3 255 5GY6/4 3.5 50 - - Continuous 0 100 8 173 2.5Y6/8 3.5 90 2 60 Continuous 30 70 9 247 7.5YR5/6 4 110 2 700 Continuous 45 55 10 206 10YR4/4 3.5 100 4 300 Continuous 43 57 0 0 Appendix Table 17• Results of holding A. flavidus under white daylight over green gravel background, [5G6/6 (Munsell)]. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 5 6 287 236 187 187 221 239 166 172 5GY5/8 2.5GY6/8 2.5GY5/6 2.5GY6/8 5GY5/8 2.5GY6/8 7-5GY6/8 5GY5/6 1 row dorsal dark spots 5GY7/9 2.5GY6/6 5Y5/6 10Y5/6 10Y5/6 10Y6/6 5R3/8 Green tinge under 5R6/4 Splotchy 49 49 49 49 61 75 49 49 186 5R3/6 5R4/4 Splotchy 14 Probably slight loss of green pigment. Possibly slight loss of green pigment. Probably slight loss of green pigment. Insignificant change in green pigment. No significant change. Possibly an increase in green pigment. Lighter. Probably no pigment change. Loss of red pigment. Possibly slight acquisition of green pigment. Loss of red pigment. Appendix Table 18. Results of holding A. flavidus in white daylight over a red gravel background, [5R5/12 (Munsell)]. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 1 2 3 4 6 7 187 5GY5/8 186 5GY5/8 Dorsal dark spots 195 5GY5/8 181 2.5GY5/6 Dorsal dark spots 191 5GY5/8 184 10R4/8 166 7-5R3/8 10Y6/6 2.5GY7/6 10Y6/6 7-5Y7/8 2.5GY7/8 Splotchy 5YR6/8 5R5/4 Splotchy 77 22 100 77 23 23 60 Loss of green pigment. Loss of green pigment. Loss of green pigment. Loss of green pigment. Loss of green pigment. Loss of red pigment. Loss of red pigment. 99 5Y5/6 5Y6/8 60 Lighter. Appendix Table 19. Results of holding A. flavidus in darkness. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 6 7 295 145 246 112 152 298 115 166 5GY5/8 5GY7/8 2.5GY5/6 7.5GY6/10 5R4/4 5YR4/4 2.5Y5V6 2.5GY7/4 5GY4/5 10Y5/6 1 row dorsal dark spots 2.5GY4/4 3 rows dark spots 5GY6/8 2 rows dark spots 2.5R3/2 Splotchy 2.5Y4/4 10Y4/6 1 row dorsal dark spots 10Y3/2 (very dark) 67 24 22 67 67 67 67 24 Darker but l i t t l e or no loss of pigment. Darker but l i t t l e or no loss of pigment. Darker but l i t t l e of no loss of pigment. Darker but l i t t l e or no loss of pigment. Became very splotchy because of light and dark spots. Loss of red. Darker. Originally deep red, but had been captive for 3 months. Belly"remained-pale red, but rest of fish dark gray. Appendix Table 20. Results of holding A. flavidus under transmitted green light. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 101 96 135 168 160 140 75 5GY5/8 5GY5/8 5R3/6 5R3/6 5R3/6 5R3/6 2.5YR5/6 5GY5/6 5GY6/5 1 row dorsal dark spots Red tinge over 7-5GY6/2 Splotchy Green tinge under 2.5R6/4 Green tinge under 2.5YR6/4* Green tinge under 10R5/4 Green tinge under 10R6/4 Splotchy 28 62 62 62 194 28 28 Possibly some loss of green pigments. Slight loss of green colouration. Loss of red pigment. Possible acquisition of green pigment. Became very splotchy. Loss of red pigment. Possible acquisition of green pigment. General lack of pigment. Loss of red pigment. Possible acquisition of green pigment. Became splotchy. Loss of red pigment. Possible acquisition of green pigment. Loss of red pigment. Possible increase in green pigment. Appendix Table 20 (Continued). Number Size Original Colour Final Colour Days Visual Interpretation (mm)  8 160 5Y4/4 7.5Y6/8 62 Possible increase in green pigment. 9 192 5Y6/8 2.5GY5/6 28 Appears greener in 1 row dorsal dark spots- colour. 10 250 7-5Y6/8 2.5GY5/5 28 Appears greener in Splotchy with dark spots colour. * Pigment analyzed. Appendix Table 21. Gross histology and pigment experiment. analysis of fish from transmitted green light Fish Number Size (mm) Colour Melanophores Index No/mm Erythrophores Index No/mm^  Green Pigment Distribution Pigment Analysis % Red % Green 5 160 Green under 2.5YR6/4 2 90 1-2 400 Continuous 6 7 3 3 GO 4>-Appendix Table 22. Results of holding A. flavidus under transmitted red light. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 4 5 170 185 142 82 80 80 111 159 5GY5/8 5GY5/8 5GY7/10 7.5GY6/6 1 row dorsal dark Splotchy spots 1 row of dorsal dark spots 5GY7/10 Red tinge over* 1 row dorsal dark 2.5GY7/8 spots 5GY5/8 5GY5/8 5GY5/8 2.5R3/8 5R4/4 5GY5/8 Red tinge over 2.5GY6/8 Red tinge over 2.5GY6/6 5YR5/11 5YR4/8 24 63 105 28 49 63 92 85 No change detectable. Probably some loss of green pigment. Became very splotchy. Possible loss of green pigment. Acquisition of red pigment. No change detectable. Acquisition of red pigment. Possibly some loss of green pigment. Acquisition of red pigment. Loss of green pigment. Moderate loss of red pigment. Moderate loss of red pigment. Appendix Table 22 (Continued). Number Size Original Colour Final Colour Days Visual Interpretation (mm)  9 200 5Y5/6 5YR5/8* 63 Acquisition of red \ pigment. 10 164 2.5Y5/6 7.5YR5/8 49 Acquisition of red pigment. * Pigment analyzed. Appendix Table 23. Gross histology and pigment analysis of fish from transmitted red light experiment. Fish Size Colour Melanophores Erythrophores Green Pigment Pigment Analysis Number (mm) Index No/mm2 Index No/mm^  Distribution % Red % Green 3 142 Red tinge 2.5 50 1 300 Continuous 12 88 over 2.5GY7/8 9 200 5YR5/8 2 100 1 600 Continuous 36 64 Appendix Table 24. Results of holding A. flavidus over neutral gray background and feeding them red gammarids containing "astaxanthin." Number Size Original Colour Final Colour Days Visual Interpretation (mm)  1 164 5GY5/8 Red tinge over 17 Acquisition of red 2.5GY5/4 pigment. 2 112 5GY7/4 Muddy green YR 17 Acquisition of red pigment. Appendix Table 25. Results of holding A. flavidus in brown algae under white daylight. Number Size Original Colour Final Colour Days Visual Interpretation (mm) 1 119 2.5Y4/4 5Y5/6 45 Lighter, probably l i t t l e change in pigment. 2 115 2.5Y4/4 5Y5/6 45 Lighter, probably l i t t l e change in pigment. Appendix Table 26. Gross histology and pigment analysis of fish with undetermined history from holding tanks and of California specimens. Fish Number Size (mm) Colour Melanophores Index No/mm2 Erythrophores Index No/mm2 Green Pigment Distribution Pigment % Red Analysis % Green 1 152 2.5GY5/6 1-2 55 None None Continuous 0 100 2 136 2.5GY5/6 2 50 None None Continuous 0 100 3 173 2.5GY5/6 2.5 65 None None Continuous 0 100 4 187 5GY6/4 2-2.5 60 None None Continuous 0 100 5 111 7.5R4/3 2 65 Continuous Network None 100 0 6 193 10YR4/6 3 70 2 300 Continuous 43 57 Appendix Table 27. Results of holding A. flavidus in red algae under white daylight. Number Size (mm) Original Colour Final Colour Days Visual Interpretation 179 184 108 122 173 164 153 5GY5/8 1 row dorsal dark spots 5GY5/8 1 row dorsal dark spots 5GY5/8 (No dorsal dark spots) 5GY5/8 5GY6/8 5R3/6 100 5R4/6 Splotchy 5R3/6 Red tinge over 7.5Y5/6 1 row dorsal dark spots 50 5GY4/6 11 1 row dorsal dark spots 5GY4/6 11 (No dorsal dark spots) Red tinge over 5GY5/4* Red tinge over 5GY7/8 5R4/8 10R4/6 Uniform 26 26 14 42 5R4/6* 11 Loss of green pigment. Acquisition of red pigment Darker but probably only slight loss of green pigment. Darker but probably only slight loss of green pigment. Loss of green pigment. Increase of red pigment. Acquisition of red pigment. Some loss of red pigment. From gray background. Original colour was 10R3/4, therefore, moderate loss of red pigment. Some loss of red pigment. Appendix Table 27 (Continued). Number Size Original Colour Final Colour Days Visual Interpretation (mm)  9 105 10 172 11 172 7.5Y5/6 2 rows 2.5YR4/4 5Y6/6 Splotchy 5YR5/6* 50 of dark spots 10R6/6* 50 Splotchy 7.5YR4/6 26 Splotchy Increase in red pigment. Increase in red pigment. Increase of red pigment. Loss of green pigment but green s t i l l underlying. * Pigment analysis. Appendix Table 28. Gross histology and pigment analysis of fish from red algae experiment. Fish Size Colour Melanophores Erythrophores Green Pigment Pigment Analysis Number (mm) Index No/mm2 Index No/mm2 Distribution % Red % Green 4 122 Red tinge 5GY5/4 3.5 60 1 350 Continuous 0 100 8 100 5R4/6 3 50 4 700 None 100 0 9 105 5YR5/6 3 80 5 600 Not evident 76 24 10 172 10R6/6 3.5 60 3-4 600 Not evident 83 17 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0104805/manifest

Comment

Related Items