Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The influence of light intensities and durations during early development on meristic variation in some… Canagaratnam, Pascarapathy 1959

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

Item Metadata

Download

Media
831-UBC_1959_A1 C2 I6.pdf [ 9.55MB ]
Metadata
JSON: 831-1.0106469.json
JSON-LD: 831-1.0106469-ld.json
RDF/XML (Pretty): 831-1.0106469-rdf.xml
RDF/JSON: 831-1.0106469-rdf.json
Turtle: 831-1.0106469-turtle.txt
N-Triples: 831-1.0106469-rdf-ntriples.txt
Original Record: 831-1.0106469-source.json
Full Text
831-1.0106469-fulltext.txt
Citation
831-1.0106469.ris

Full Text

THE INFLUENCE OF LIGHT INTENSITIES AND DURATIONS DURING EARLY DEVELOPMENT ON MERISTIC VARIATION IN SOME SALMONIDS by PASCARAPATHY CANAGARATNAM B,Sc., University of Ceylon, 1951 M.A., University of Brit i s h Columbia, 1957 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Zoology We accept this thesis as conforming to the required standard The University of Brit i s h Columbia May 1959 ABSTRACT Experiments were designed to study the effects of various intensities and durations of light during early development on meristic variation in kokanee and sockeye salmon (Oncorhynchus nerka) f pink salmon (0. gorbuscha) and in rainbow trout (Salmo gairdneri). An experiment with sockeye was designed exclusively to test the period of f i x a -tion of vertebrae. Among the salmon species rates of hatching and yolk-sac absorption were fastest under the longer light durations and higher intensities. In sockeye and pink an increasing rate of yolk-sac absorption was correlated with increasing amount of light. In trout the rate of yolk-sac absorption showed the opposite results. Mortalities were high under a l l the experimental conditions. Although meristic v a r i a b i l i t y observed was deemed to be phenotypic yet the affects of selective mortality could not be entirely excluded. Abnormalities in the verte-bral column were prevalent in the *pre-urostylic' region. The occurrence of abnormal vertebrae was correlated with light only in sockeye. The activ i t i e s of the pituitary and thyroid glands of trout showed a positive correlation with higher amounts of ligh t . Sockeye scale counts, along the lateral line and on the oblique rows from the origin of dorsal and anal fins to lateral l i n e , were lowest under conditions of higher light and longer durations. Fin ray numbers in a l l species were the lowest at high light intensities and longer dura-tions. Vertebral counts were lowest at high light inten-s i t i e s and long durations in pink salmon and rainbow trout, but higher in sockeye. At lower light intensities and dura-tions results were variable. Vertebral counts of sockeye increased with increasing light at temperatures of both 8°C and 12°C. Differences among lots at 8°C were significant but those among comparable lots at 12°C were not. The action of light on meristic v a r i a b i l i t y was weak at the higher temperature. It was found in sockeye that vertebral numbers were not fixed before 142 D° and that the period of sensit-i v i t y was prolonged (142 D° to 300 D°). In presenting t h i s thesis i n p a r t i a l fulfilment of the requirements fo r 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. It i s understood that copying or publication of this thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. The University of B r i t i s n Columbia, Vancouver $, Canada. Department i TABLE OF CONTENTS page ACKNOWLEDGEMENTS v i i i INTRODUCTION . 1 MATERIALS AND METHODS 4 Experimental Design . . . . . . . . . . . 4 Experiment A . . . . . . . . . . . . . 4 Water supply and aeration . . . . . 6 Temperature control . . . . . . . . 7 Control of Light Durations and Intensities 8 Fungus Control and Cleaning of Tanks 9 Feeding 10 Preservation and Clearing of Fish for Meristic Study. . . . 10 X-Ray Method for Sockeye in Experiment A 11 Other Meristic Counts in X-rayed Sockeye and a l l Other Lots 12 Experiment B 13 Experiment C • • 14 Method of Transfer. 16 Day degrees 16 Stage of Development at 1st, 2nd and 3rd Transfers 17 RESULTS 25 A. Hatching time and Yolk-sac Absorption. 25 Experiment A 25 Experiment B 27 Hatching times 28 Yolk-sac absorption 30 i i page B. Mortalities . . . . . . 31 v. Experiments A and B . . . . . . . . 32 Experiment C . . . . . . 35 C. Abnormalities in the Vertebral Column . 39 Experiment A . , 42 Experiment B . . . . . . . 44 Experiment C 44 D. Meristic Variations 50 Correlation of Meristic Counts with Length of Fish. 50 Method of Analyses 50 Experiment A - Sockeye Scales. . . . 51 Lateral Line scales . . 51 Scales on oblique row from origin of dorsal f i n to lateral line 51 Scales on oblique row from origin of anal f i n to lateral l i n e . . . . . . . . . . 52 Sockeye secondary caudal rays . . 52 Total vertebrae (kokanee, sockeye and pink) . . . . . . . . 58 Abdominal and Caudal vertebrae (kokanee and p i n k ) . . . . . . . 61 Comparison of Lateral Line Scales and Total Vertebral Counts of Sockeye. . . . . . . 62 Dorsal and Anal Fin Rays (ko-kanee, sockeye and pink) . . . 62 Experiment B - Rainbow trout . . . . 70 Vertebrae 70 Dorsal and Anal Fin Rays 70 E. Histological .Study of Pituitary and Thyroid Glands of Rainbow Trout (Experiment B). 78 Method . . . . . . . . . . . . . . . 78 Pituitary Gland 78 Analyses . . . . . . . . . . . . . . 79 Thyroid Gland. . . . . . . . . . . . 80 Analyses 81 i i i page Size of Pituitary Epithelial Components . . in a l l Lots 81 Activity of Thyroid Gland 81 Area of Pituitary Epithelium and Thyroid Activity 82 F. Experiment C - Period of Fixation of Vertebrae in Sockeye 87 Vertebral Counts of Controls in Sections I-VI 87 Transfers Between 8-hour and 16-hour High Light Intensity. . . . . 89 Other Transfers. 91 G. Vertebrae of Sockeye, Experiments A and C. 99 DISCUSSION 102 SUMMARY. 120 LITERATURE CITED 124 XV LIST OF FIGURES Figure Page 1 Arrangement of tanks and light hoods in Experiment A and B. 19 2 Instrument board and light hood of one unit 19 3 Dorsal f i n rays of sockeye to show the rays involved in the count. 20 4 Anal f i n rays of sockeye. 20 5. Mean daily temperature record for experiment C. 21 6 Light hood and trough of 16-hour light dura-tion of experiment C. 22 7 Two light hoods and troughs with time switches 22 8 Arrangement of baskets in a section or compart-ment of experiment C. 23 9a Arrangement of baskets at beginning of experi-ment C. 24 9b Arrangement of baskets after the three trans-fers in experiment C. 24 10 Five sockeye selected from experiment C section 1 basket No. 36 to il l u s t r a t e the different sizes in a lot and that the degree of vertebral ossification depends on size. 43 11 Vertebral synostoses (fused and complex) in some of the experimental lots, 47 12 Experiment A. Histograms of percentage ab-normal vertebrae in kokanee, sockeye and pink salmon. 48 14 Experiment C. Histograms of percentage ab-normal vertebrae in sockeye control l o t s . 48 13 Experiment B. Histograms of abnormal verte-brae in rainbow trout. 49 V Figure Page 15 Influence of light on lateral line scales of sockeye in experiment A. 53 16 Influence of light on scales from origin of dorsal f i n to lateral line of sockeye in experiment A. 54 17 Influence of light on scales from origin of anal f i n to lateral line of sockeye in experiment A. 55 18 Influence of light on secondary caudal rays of sockeye in experiment A. 57 19 Influence of light on total vertebral of kokanee, sockeye and pink in experiment A. 65 20 Influence of light on abdominal vertebrae of kokanee and pink in experiment A. 66 21 Influence of light on caudal vertebrae of kokanee and pink in experiment A. 67 22 Influence of light on dorsal f i n rays of kokanee, sockeye and pink in experiment A. 68 23 Influence of light on anal f i n rays of koka-nee, sockeye and pink in experiment A. 69 24 Influence of light on total vertebrae of rainbow trout in experiment B» 72 25 Influence of light on abdominal vertebrae of rainbow trout in experiment B. 73 26 Influence of light on caudal vertebrae of rainbow trout in experiment B. 74 27 Influence of light on dorsal and anal f i n rays of rainbow trout in experiment B. 75 28 Median longitudinal sections of trout p i t u i -tary, experiment B. 84 29 Transverse sections of thyroid f o l l i c l e s in trout, experiment B. 85 v i Figure Page 30 Influence of light on area of epithelial components of pituitary gland of rainbow trout in experiment B. 86 31 Influence of light on total vertebrae of sockeye control lots in experiment C. 88 32 Percentage frequency polygons of total verte-bral counts in controls and transfers effected at various times after f e r t i l i z a t i o n in ex-periment C. 95 33 Percentage frequency polygons of abdominal vertebrae of sockeye in experiment C. 96 34 Percentage frequency polygons of caudal vertebrae of sockeye in experiment C. 97 35a and b Percentage frequency polygons of total vertebrae of sockeye in experiment C. 98 36 Comparison of mean vertebral counts (total) of sockeye reared at 8°C in experiment C, and in experiment A at 12°C. 101 LIST OF TABLES Table I Record of mortalities in experiments A and B. 33 II Record of the source of eggs, hatching and yolk-sac absorption times and the time of preservation for meristic study of the various species used in experiments A and B. 34 III Record of mortalities in experiment C. 38 IV Occurrence of vertebral abnormalities in experiments A, B and C. Only control lots of experiment C were included. 46 Va Frequency distribution of lateral line scales of sockeye in experiment A. 56 v i i Table Pag© Vb Frequency distribution of scale counts from origin of dorsal f i n to lateral line along the oblique row of sockeye in experiment A. 56 Vc Frequency distribution of scale counts from origin of anal f i n to lateral line along the oblique row of sockeye in experiment A. 56 Via Frequency distribution of upper secondary caudal f i n ray counts in sockeye, experiment A. 59 VIb Frequency distribution of lower secondary caudal f i n ray counts in sockeye, experiment A. 59 VII Frequency distributions of total vertebral counts in experiments A and B. 60 VIII Frequency distributions of abdominal vertebral counts in experiment A and B. 63 IX Frequency distribution of caudal vertebral counts in experiment A and B. 64 X Frequency distribution of dorsal f i n ray counts in experiments A and B. 76 XI Frequency distribution of anal f i n ray counts in experiments A and B. 77 XIla Mean glandular area of pituitary of rainbow trout in experiment B. 83 Xllb Histological study of thyroid gland in rain-bow trout of experiment B. 83 XIII Frequency distribution of total vertebral counts in experiment C, 92 XIV Frequency distribution of abdominal vertebral counts in experiment C. 93 XV Frequency distribution of caudal vertebral counts in experiment C. 94 v i i i ACKNOWLEDGEMENTS This work was done under the supervision of Dr. C. C. Lindsey, Department of Zoology, University of Br i t i s h Columbia. It is a pleasure to acknowledge his guidance and assistance during a l l phases of the experimental work and writing of the thesis. Sincere thanks are due to Dr. P. A. Larkin, Direc-tor of the Institute of Fisheries, University of Brit i s h Columbia, for advice on s t a t i s t i c a l procedures and for the many valuable criticisms and comments in the preparation of the thesis. I also thank Dr. W. A. Clemens for c r i t i c a l l y reading the thesis. I appreciate the advice and assistance given by Dr. P. Ford - staining techniques, Dr. K. Graham and Mr. T. Pletcher - photomicrography, and Dr. W. S. Hoar - endocrine study. Miss Sachi Tabata helped with some of the histologi-cal preparations. Dr. S. M. Friedman, Department of Anatomy, made available the X-ray machine and Miss Betty Barrett, University Health Service, helped with the X-ray photography. Mr. G. Campbell, Agricultural Mechanics, constructed the two light hoods used in experiment C. The eggs used in the ex-periments were supplied by the following:- Mr. P. Gilhousen, International Salmon Commission, and Messrs. J. Cartwright, G. F. Hartman, F. P. Maher and T.G. Northcote a l l of the ix B r i t i s h Columbia Game Department. To a l l these persons I owe my thanks. I am very grateful to Dr. I. McT. Cowan, Head of the Department of Zoology, for granting permission to work on this problem and for his unstinted kindness and encourage-ment throughout. I also thank the Director of Fisheries, Colombo, Ceylon, for extending my study leave to complete this work. Financial aid from the following is gratefully acknowledged:- Grants from the National Research Council of Canada, Colombo Plan Scholarship from the Technical Co-opera-tion Service, Department of Trade and Commerce, Ottawa, on an application made by the Government of Ceylon, and funds from the Institute of Fisheries, University of Bri t i s h Columbia. INTRODUCTION Meristic characters play an important role in sys-tematic ichthyology and in racial studies associated with fisheries biology and management. Classification of fishes, particularly to the lower categories such as species and subspecies, may require counts of several meristic series. Among those commonly used are scales, vertebrae, g i l l rakers, f i n rays and spines. Unlike body proportions which may change with size, meristic counts are fixed in early devel-opment and are not alterable with growth and maturity. Meristic counts commonly show wide va r i a b i l i t y between individuals of a species. This v a r i a b i l i t y , particu-l a r l y in vertebral numbers among natural populations of f i s h , was attributed to genetic differences by Heincke (1898) and Schnackenbeck (1931). On the other hand many workers have considered that differences were phenotypic and attributable to environmental conditions. In f i e l d observations, meristic characters can be correlated with environmental conditions at the time of fer-t i l i z a t i o n and during early development. This does not necessarily preclude the possibility that there i s some genetically fixed connection between the meristic series and observed environmental conditions. Many meristic characters that are used to distinguish geographical races are inherited, -2-but prevailing environmental factors certainly influence the f i n a l expression. However, the interplay of genetic and environmental factors makes segregation of their effects d i f -f i c u l t (Gordon, 1957). Field observations are thus inade-quate in providing explanation of the role of various factors in producing va r i a b i l i t y in meristic counts. In the experimental f i e l d , many workers have shown that in a great majority of cases environmental factors, one or more, induce changes in one or more meristic series, to such a degree that the experimentally raised stock is s i g n i -ficantly distinct from the parents (literature was reviewed by TSning (1952)). Of the environmental factors manipulated experimentally, temperature, s a l i n i t y , oxygen, carbon-dioxide, light, and gravel cover, in that order of frequency, have been used with a limited number of species. There have been many inconsistencies in the results obtained for different species of f i s h . Apparently for none of the species has there been an adequate description of the embryonic mechanisms underlying meristic variations. The importance of light as an environmental variable producing meristic changes in fishes has been experimentally tested by McHugh (1954) who demonstrated that light inten-sity during development could influence vertebral counts of grunion, Leuresthes tenuis. Lindsey (1958), demonstrated that light duration could affect vertebral and anal f i n ray counts in kokanee, Oncorhynchus nerka. The present experiments were conducted to explore the degree and direction of variation of various meristic series which could be produced by light of different inten-s i t i e s and durations, for some species of salmonids, and to attempt to learn something of the mechanisms involved in these changes. A major experiment was set up to test the effect of light upon the period of vertebral fixation in sockeye salmon by a method of transfers of embryos to d i f f e r -ent environmental conditions at various developmental stages. Another object of this research:was to test the effect of two temperatures on vertebral counts in sockeye and to infer the combined action of light and temperature. Meristic characters studied were scales and secondary caudal f i n rays in sockeye; vertebrae, dorsal and anal f i n rays in salmon (kokanee, sockeye, pink), and rainbow trout. Many competent authorities in fi s h physiology have stated that light influences the activity of the pituitary and thyroid glands. Some histological work on the trout groups was carried out to test i f there was a correlation between endocrine activity and meristic variation. -4-MATERIALS AND METHODS Experimental Design Three experiments A, B and C, were designed in early f a l l of 1957 to study the influence of light durations and intensities on meristic v a r i a b i l i t y in three species of salmonidae, v i z . Oncorhvnchus nerka.(sockeye salmon (sea run form) and kokanee, (landlocked form)), 0. gorbuscha (pink salmon) and Salmo gairdneri (rainbow trout). In a l l cases eggs were stripped from several females and f e r t i l i z e d with the milt from one or two males. Stripping and f e r t i l i z a t i o n of eggs were conducted in the f i e l d , and eggs were transported to the laboratory in the University of British Columbia. Fertilized eggs were placed in trays and brought gradually (usually 3 to 4 hours) to the desired temperature before being set in the experimental tanks. Time elapsed between f e r t i l i -zation and establishment of the experimental conditions never exceeded 12 hours. Experiment A In experiment A, kokanee, pink and sockeye salmon eggs were hatched and reared at a constant temperature of 12°C. Fertilized eggs of the above species were obtained at different times as indicated in Table I. The total number -5-of eggs of each species was divided into 5 lots by volume and placed in perforated, rectangular aluminum trays. The number in each lot is lis t e d in Table I. The main purpose of experiment A was to determine the effect of light on meristic counts. The five lots were subjected to varying durations and light intensities. One lot was kept in total darkness, two lots at 16 hours light duration with one at high and the other at low intensity, and the remaining two lots at 8 hours light duration and inten-sity treatment comparable to the 16 hour lots. Five separ-ate tanks were required for this experiment. A l l tanks were of the same size (54 cm x 54 cm x 38 cm) and were constructed with 3/4" plywood painted inside with black neoprene, a water-proof non-toxic paint. The vol -ume of water held by each tank at the outflow level (which was 7 cm. from the top) was approximately 70 l i t r e s . The perforated, aluminum trays containing the eggs were placed on a frame (aluminum) such that there was approxi-mately 7 cm. of water above the layer of eggs. About 3 cm. of the side walls of the trays were above waterlevel to pre-vent the young fry from mixing with the other species and also to confine them to the same area where a known intensity of light obtained. The relative position of trays on the frames in the five tanks was kept the same for each species. -6-The area of trays was not uniformly similar for a l l species, but was almost the same for the five lots of each species so as to avoid any problem of the effects of "space factor" on development (Brown, 1946 and Comfort, 1956). Water supply and aeration Fresh water from the University mains was dechlor-inated and f i l t e r e d before being taken into the laboratory pipes. Tygon tubing of 6 mm bore and glass T-tubes were used to provide each tank with a continuous supply of water. Water entered the tanks at or near the bottom and opposite to the outlet end which was 7 cm below the top of tank (Fig-ure 1). Two gas diffuser stones, with one near the point of entry of water, provided a constant supply of compressed a i r . This method of aeration served to keep the oxygen con-tent high and to mix the incoming cold water. The amount of water supplied to each tank was regulated by means of clamps on the tygon tubing and aeration too was maintained approximately uniform. The volume of water entering each tank was measured and the flow maintained throughout the experimental period i n a l l the tanks, at approximately 20 l i t r e s per hour. The rate of renewal was measured by using a strong solution of malachite green before the eggs were placed in the tanks. The entire 70 l i t r e s of water was dyed; after maintaining the -7-flow of water at 20 l i t r e s per hour with the above aeration:, the outflow water did not show any visib l e tinge of green colouration after 8 hours. The water in the tanks was thus probably completely renewed two or three times a day. Temperature Control Water temperature was maintained at 12°C t 0.1°. This temperature for experiment was selected after examina-tion of temperature records of the laboratory water supply. No cooling f a c i l i t i e s were available and hence i t was impera-tive that the incoming cold water be heated to a temperature that was not lethal for the species under consideration and which could be maintained for the entire experimental period. Flexible, stainless steel heaters of 500 watt capacity with 70 cm effective heating surface, one to each tank, were bent to form a loop and f i t t e d on to an instrument board on top. The effective heating surface of the heater was placed close to the water source and aerators, but away from the region of the frame and egg tray. Temperature was set by means of a "Quickset" bimetal thermoregulator (Aminco product) and the heating was controlled by means of a super-sensitive relay (Aminco product). A small neon lamp was fi t t e d on the instrument board (Figure 2) of each tank and connected to heaters to f a c i l i t a t e i n i t i a l temperature setting -8-and thereafter acted as an indicator when the heating ele-ment was in operation. Temperature was checked two or three times a day by means of a thermometer, sensitive to 0.01°C, through a hole (which was plugged when not in use) on the instrument board. Even with a flow of cold water (sometimes at 6-7°C) at 20 l i t r e s / h r . the temperature was usually maintained with-in ± 0.1°C. It very seldom varied by ±0.2°C. Control of Light Durations and Intensities Light hoods (Figure 1) were made to f i t the top of each tank except the area occupied by the instrument board. The edges of the hood were lined with 1 cm thick dark sponge rubber bands so that when the hoods were placed on the tanks the weight of the former gave a good lightproof f i t a l l along the edges. The inside centre of the hood was provided with an electric lamp holder into which was f i t t e d a frosted glass filament bulb (G.E.). The two light durations of 16-hours and 8-hours were controlled by means of el e c t r i c a l l y operated time switches. Light intensity of 5.9 ft . / c d l s . was obtained by using a 7.5 watt filament lamp. The lower light intensity of 1 f t . / c d l . was obtained by coating a 7.5 watt filament lamp with white paint. Light intensities were measured with a Photovolt foot-candle meter. Most of the lamps used lasted the entire period of the experiment and showed no change in intensity readings. When lamps burned out they were replaced and intensities remained constant. The foot candle measurement recorded for each intensity i s the mean for readings obtained at various points on the trays. The total dark tank had the hood f i t t e d without a lamp. In this experiment the hoods were not l i f t e d except to check the lamps daily for a brief moment. Any obser-vation, cleaning or picking dead eggs was done during the time the light was on. In the dark tank similar operations were performed in the night with dim reflected light from another room. The number of dead eggs, larvae, fry and, in the case of sockeye, dead fingerlings were recorded to ascer-tain f i n a l survival up to the time of preservation. Samples removed for other study were accounted for in the f i n a l anal-ysis of mortality. Hatching time (50%) and rate of yolk absorption were also recorded. Fungus Control and Cleaning of Tanks Saprolegnia. a fungus which grows very rapidly on eggs held in this laboratory, especially on dead eggs, was kept under f a i r l y good control by using daily a 150 ml 1:200,000 solution of malachite green (Barrett and Hum 1954) in a l l tanks. The solution was introduced through the hole on the -10-ins trument board used to insert the thermometer. Since the trays were set on a frame the process of cleaning was easy. A seagull feather was used to brush off any sediment and slime on sides and bottoms of trays. The bottom of tank was kept clean by siphoning off the deposits. Feeding In a l l cases f i s h were kept under the same condi-tions for some time after yolk absorption and hence i t was necessary to feed them. Live brine shrimp nauplii (hatched from dry eggs in laboratory) and Daphnia (cultured in 10 large aquaria) were given in excess three times daily. Preservation and Clearing of Fish for Meristic Study Lots within each species were k i l l e d at the same time and preserved in 8% neutral formalin. Fish were kept in formalin for about three weeks before being stained and cleared by the Alizarin - KOH - ultraviolet light method of Hollister (1934). The Hollister method was modified by varying the strength of KOH and the time of the different clearing stages in accordance with the size of the f i s h . -11-X-Rav Method for Sockeye in Experiment A Sockeye were transferred on April 17, 1958, five months after f e r t i l i z a t i o n , to another set of similar tanks with appropriate controls. These f i s h were transferred to make room for the trout eggs of experiment B where light con-ditions were varied, and to l e t them grow to a size s u f f i c i -ent to enable counting of scales. Sockeye lots were k i l l e d on June 12, 1958, and pre-served in 10% formalin for two weeks. They were marked and preserved i n 35% iso-propyl alcohol before being X-rayed for vertebral counts. A soft ray machine, in the Department of Anatomy, was used. Exposure recommended by Miller (1957) was found to be most satisfactory. A l l lots were X-rayed on non screen IIford films using an exposure of 5 milliamperes and 26 kilovolts for 6 seconds at a distance of 32 inches. Ver-tebrae between the basioccipital and urostyle were counted easily by placing the films on a ground glass illuminated screen. The last vertebra with urostyle was omitted in counts in these lots and in a l l the other experiments, be-cause the centrum in this region was not well ossified, and stained poorly. Because the position of the f i r s t caudal vertebra was most d i f f i c u l t to distinguish in X-rayed specimens, only total counts of vertebrae of this species were taken. -12-Clothier (1950) states that in the "genus Oncorhynchus the f i r s t haemal spine i s inde-terminate because i t is only a minute process and succeeding spines become gradually longer. The vertebra on which these short haemal spines suddenly increase in length and curve backward was selected as the diagnostic character for these species..." In X-ray films sudden increase in length was very often ob-scured by the overlapping r i b s . In other species, which were cleared and stained, the f i r s t caudal vertebra was very easy to distinguish. The vertebra on which the haemal arches were joined to a median spine was taken as the f i r s t caudal vertebra. Other Meristic Counts in X-rayed Sockeye and a l l Other Lots Scales, dorsal, anal and caudal secondary rays (upper and lower) were counted under a binocular microscope f i t t e d with a cross hair and vernier stage, after specimens were soaked in 1% KOH and a l i z a r i n stain for a week, washed in water and kept in 35% iso-propyl alcohol. The dye remains on scales and rays for a sufficiently long time to permit counts to be made conveniently. The fork length of each f i s h was also measured. A l l other species were stained and cleared as described earlier. In dorsal and anal f i n ray counts a l l elements were well stained and clearly v i s i b l e (Figure 3 and 4) and hence a l l rays were counted, the last two being considered one -13-(Taning 1944, Seymour 1946). Although secondary caudal ray counts were made i t was observed in trout, kokanee and pink that there was much variation with size of f i s h . The secondary caudal ray counts could thus be biased i f specimens were examined in which development of these elements was only p a r t i a l . A l l lots of sockeye were allowed to grow for a longer time to attain a larger size, and to thus allow complete formation of rays, and in this group the secondary caudal ray counts did not show any correlation with size. In subsequent anal-ysis of secondary caudal ray counts only sockeye were used. Experiment B. Rainbow trout eggs from Cultus Lake hatchery were reared under identical conditions as in experiment A but the light intensities and durations were varied. There were six lots in each of which approximately 1600 eggs were used (Table I.) Two tanks had continuous light durations (24 hours) with light intensities in one at 20 ft . / c d l s . and the other at 0.3 ft. / c d l s . Another two tanks had 4 hours light duration with light intensities in one at 36 f t . / c d l s . and the other at 0.4 ft. / c d l s . There was one set for a 12-hour light duration with 19 ft. / c d l s . and the sixth was in total darkness. Samples were periodically fixed for histological -14-Experiment C The main object of this experiment was to deter-mine the time of vertebral fixation in developing sockeye salmon. The procedure adopted was very similar to that of Taning (1944) in experiments with sea trout. Transfers in this case were effected from one light duration to another and also between durations and intensities. The temperature was not controlled at a sustained level but a l l groups were subjected/to the prevailing temperature during the period. Temperatures were recorded at three different times of the day and a mean obtained for each day (Figure 5). Sockeye eggs (about 12,000) from the same stock as the ones used in Experiment A were divided into 36 lots and each lot placed in a basket (15 cm x 10 cm x 11 cm) made from fibre glass and plastic coated netting. The baskets were placed in two metal troughs (220 cm x 80 cm x 23 cm) coated with non-toxic paint. The outlet tube was placed such that the level of water in the trough was maintained at a depth of 12.5 cm. Each trough was covered by a light hood made of plywood. Each light hood and trough was divided i n -to three sections by means of wooden partitions. The p a r t i -tion in the trough was provided with five holes (1 cm diameter) to permit flow of water from one section to another. The source of water was at one end of the trough, flowed through the three sections and drained off at the exit pipe situated -15-in the third compartment. The partitions were f i t t e d to the sides of the trough and hood and sealed along the edges with plasticine to make a light seal. The holes in the trough partitions were guarded on either side by means of two wooden shields to prevent light diffusing from one section to another. The two light hoods were provided with electric lamp holders in the middle of each section. One side of each hood was provided with a sliding door for each section (Figure 6). One trough was maintained at 16-hours light duration with three intensities (20 f t . / c d l . Section I, 7 f t . / c d l . Section II, and 1 f t . / c d l . Section III). In the other trough 2 sections were at 8-hour light duration (20 f t . / c d l . Section VI; 7 f t . / c d l . Section V, and the third section was in total darkness Section IV). Durations were e l e c t r i -cally controlled.by means of two time switches (Figure 7). Six baskets were numbered and placed in each com-partment (Figure 8). The baskets rested on inverted trout hatchery trays to permit free flow of water on a l l sides and also to raise the upper edges of the baskets about 3 cm above water level to prevent any mixing of fry. The capacity of each trough with the exit level at 12.5 cm was approximately 100 l i t r e s . Water flow was main-tained on the average at about 200 litres/hour. This ensured complete renewal of water every half hour. -16-Method of Transfers Of the six baskets in each section one was kept as a control and the others were transferred to other sec-tions during three stages of development. The f i r s t transfer was made after 16 days from f e r t i l i z a t i o n or 142 D° (Day degrees), the second after 24 days or 206 D° and third after 36 days or 300 D°. Since transfers at a l l three stages could not be made to and from a l l sections because of space limitations a l l three transfers were effected between section I and section VI(16-hour and 8-hour high light intensity). One or two transfers at the f i r s t , second or third stages were made in a l l sections. The arrangement of baskets in the two troughs at the beginning and at each of the three trans-fers i s shown in Figures 9 a and b. After the third transfer the positions of baskets remained the same until the end of experiment. Sections were cleaned, fed, and mortality records were kept, as in experiment A. A l l lots were fixed for ver-tebral study at the same time (222 days after f e r t i l i z a t i o n or 1576 D°). The only exception was the control of the total dark section where because of high mortality counts were also made of the preserved dead f i s h (189 days after f e r t i l i -zation or 1245 D°). Pav degrees Apstein (1909) suggested a temperature unit system -17-called "Tagesgrade" which i s a product of degrees of tempera-ture in centigrade units above a threshold and the number of days involved. Threshold temperature was recognised by Ap-stein as the lowest point at which development could take place, rather than 0° C, the freezing point of water. The calculation of day degrees in the present experiments was es-sentially similar to that of Taning (1944). As the absolute zero of biological processes of salmonid eggs is not known i t is assumed that 0° C i s very close to biological zero. Stage of Development at 1st. 2nd and 3rd Transfers The stages of embryological development, at each of the three transfer times, were examined, f i r s t l y to ascer-tain whether there was any variation in control and transfer-red lots, and secondly to know to what degree development had taken place. Eggs preserved in Bouin*s f l u i d , from controls and transferred baskets, were stained in Borax carmine and whole mounts prepared. A careful study of these revealed that the developmental stage at time of transfer was essentially similar in a l l cases. A brief description of the embryos at each transfer time i s given below. -18-Staee at 1st Transfer (16 days after f e r t i l i z a t i o n or 142 D°) Embryos measured from 4.5 to 6 mm and the blastopore was closed. Head and caudal regions were firmly attached to the yolk mass. In the head region the optic cup was v i -sible and the outline of the otic capsule was distinct. The neural groove was clear a l l along the dorso-median length of the embryo to the t a i l end. The somites along the trunk were marked out with lateral grooves but were not distinguish-able in the t a i l end region. Stage at 2nd Transfer (24 days after f e r t i l i z a t i o n or 206 D°) Embryos measured from 8 to 9 mm. The head and ti p of t a i l region were free from the yolk mass and the pec-toral f i n bud was just v i s i b l e . The gut and anus were dis-tinguishable. There was a membranous fold along both dorsal and ventral margins of the length of the body. Somites in the trunk and t a i l regions were very clear but were indistinct at the very end of the latter region (about 1 or 2 mm). Stage at 3rd Transfer (31 days after F e r t i l i z a t i o n or 300 D°) Embryos measured 11 to 13 mm. Head and t a i l ends were clearly off the yolk mass. Eyes and the membranous fold along the body were quite prominent. Somites were not quite distinct in either trunk or t a i l . The notochord was bent up at the t a i l end and somites in this region were indistinct. -19-Figure 1 Arrangement of tanks and light hoods in Experi-ment A and R. . A-Aminco relay, B-tank, C-Light hood, D-outlet pipe. Figure 2 Instrument board and light hood of one unit. A-Air valve, B-clamp on inlet water tube, C-light hood, D-thermoregulator, E-heater ter-minals, F-neon lamp. -20-X 6 0 Figure 3. Dorsal f i n rays of sockeye to show the rays involved i n the count. 1 - f i r s t ray 15 - l a s t doubleray counted as one. (Specimen treated with KOH-Alizarin). X 60 Figure 4. Anal f i n rays of sockeye. 1 - f i r s t ray 18 - l a s t double ray counted as one. (Specimen treated with KOH-Alizarin). Figure 5> Mean daily temperature record for experiment C -22-2 0 Figure 6. Light hood and trough of 16-hour l i g h t duration of experiment C Sections I, II and III marked on raised doors. P a r t i t i o n between Section II and III c l e a r l y v i s i b l e and made up of A -p a r t i t i o n i n hood and B - p a r t i t i o n i n trough C. Figure 7. Two l i g h t hoods and troughs with time switches. -23-Figure 8. Arrangement of baskets i n a section or compart-ment of experiment C. Baskets made of f i b r e glass p l a s t i c coated netting and marked with metal tags on the top corner of each basket above water l e v e l . -24-I I I IV VI SECT. CO cc o I I II I I I IV VI INTENSITY 2 0 F T / C D L 7 FT/CDL I F T / C D L DARK 7 F T / C D L 2 0 FT/CDL Figure 9a. Arrangement of baskets at beginning of experiment C, in IV V CONTROL 1ST TRANSFER 16 DAYS 142 D° 2ND TRANSFER 2 4 DAYS 2 0 6 0 ° "3RD TRANSFER ' 36 DAYS 3 0 0 D ° Figure 9b. Arrangement of baskets after the three transfers in experiment C. -25-RESULTS A. Hatching Time and Yolk-sac Absorption As a possible clue to mechanisms underlying l i g h t -induced meristic, variation, records were kept of hatching time of eggs reared under various light conditions. Hatching time was taken as that time when 50% of the total number had hatched, including dead larvae, in each l o t . Hatching in salmonids takes place over a considerably long period, es-pecially at low temperatures. Although experiments A and B conducted in this project were maintained at 12°C "t 0.1*C, the period of hatching within a lot lasted 15-20 days in the case of kokanee and sockeye, 13-17 days for pink salmon and from 8-10 days for rainbow trout. The time of yolk-sac absorption (up to the so-called "buttoning-up" stage) was reckoned from hatching time. Times (in days and day degrees) for 50% hatching and yolk-sac ab-sorption, together with time of fixation of young for mer-i s t i c study in experiments A and B are given in Table II. Experiment A !  Kokanee Hatching time was fastest (58 days) in the higher light intensities of both 16-hour and 8-hour durations. In -26-the longer (16-hour) duration of lower intensity hatching time was only a day earlier than in the shorter (8-hour) duration of a similar intensity. The lot in total darkness took the longest time (63 days) for 50% hatching. Differ-ence between high light intensity lots and the dark lot was exactly five days. There was no regularity in the times for yolk-sac absorption. The lot kept under 8-hour high intensity died in consequence of a laboratory accident. Sockeye Although the dark lot took the longest time (68 days) as in kokanee, there was no consistency in hatching times in the other lo t s . In yolk-sac absorption the dark lot again took the longest time. Sockeye groups showed a good correlation between light durations and intensities of the various lots and yolk-sac absorption times. Pink This group of salmon hatched out in about half the time taken by sockeye or kokanee. Here again the lot reared in the dark hatched out last (31 days). The one subjected to an 8-hour low intensity took the same time as dark l o t . A l l the other lots had a hatching time of 29 days. In yolk -27-absorption the pattern was similar to that of sockeye. The dark lot took the longest time (15 days), the 8-hour low intensity lot 13 days, and the remaining lots took 12 days. Although pink salmon under natural conditions mi-grate downstream soon after emergence from the gravel, the lots reared in the laboratory were fed as outlined earlier and successfully kept for nearly 100 days before being k i l l e d and fixed for meristic study. Experiment B Rainbow trout hatching times were shortest in lots reared at 24-hour and 12-hour high light intensities (25 days). The longest time was in the 4-hour low intensity lot (27 days). Yolk-sac absorption times show an opposite trend to that just seen in salmon. The 4-hour low intensity lot (33 days), dark lot (34 days), and 24-hour low intensity lot (34 days) took the shortest periods. Lots in higher intensities of the 4, 12, and 24-hour durations took 36, 37 and 40 days res-pectively. Hatching times were short compared to salmon except pink, but yolk-sac absorption took a much longer time than any of the salmon groups. Samples taken from the six lots, 62 days after f e r t i l i z a t i o n , showed that a few from the higher light duration lots s t i l l contained considerable quantity of yolk. -28-Hatching times The characteristic pattern of hatching observed in these salmonids suggested that the time at which 50% of the eggs had hatched was the best measure of rate of develop-ment of the group. Approximately 80% of the eggs would hatch within a relatively short period of time, seldom ex-ceeding twelve days. A small fraction of the eggs hatch either precociously or substantially later than the main group. The mean hatching time would be biased by these abe-rrant individuals to a greater extent than the median time of hatching. This method was also used by Seymour (1956), for chinook salmon, Oncorhynchus tshawytscha. In the salmon groups, lots reared in darkness usually hatched later than those reared under the various conditions of l i g h t . For the trout the reverse was true, those reared in darkness hatched earlier than those under l i g h t . The number of days taken for 50% hatching time among the species was very markedly different. This was evidently an inherent difference. There are several alternative explanations for the differences in hatching time among the lots within a species. The use of 50% hatching time as a criterion of rate of develop-ment may account for small apparent differences. The effect of the various light treatments seems the most l i k e l y explana-tion for the larger differences in hatching time. The -29-activity of the embryo is probably of great importance for hatching. In the medaka Pryzias latipes y Ishida (1944) re-ported that the i n i t i a t i o n of respiratory movements, parti-cularly of the operculum, seemed to rupture the hatching glands in the pharynx. The enzymes digested the shell and hatching followed shortly after. Trifonova (1937) observed that hatching time of salmon eggs could be advanced from 5 to 7 days i f the eggs were subjected to about 6 hours of as-phyxia by bubbling hydrogen through the hatching water. Hatching of Fundulus eggs could be delayed indefinitely by maintaining a high oxygen tension in the sea water in which they were kept (Milkman 1954). A l l these evidences indicate that activity of the embryo is of prime importance in the hatching operation. Activity of embryos under the various experimental conditions was not observed. However, i t is reasonable to assume that light stimulates activity. This could possibly account for the differences recorded in hat-ching times. Haempel and Lechler (1931) found that ultra violet light of very low intensity accelerates hatching in salmonids. The light source used in these experiments produced almost a negligible quantity of ultra violet light but i t may have been enough to produce similar effects. Lyubitskaya (1956), found in a number of species of f i s h , including Salmo, that in a majority of cases the beginning of hatching occurred -30-earlier in white than in monochromatic l i g h t . The above evidences strongly support the view that salmonid hatching times are accelerated (period reduced) by li g h t . The source of light, i t s durations and intensities are also important• Yolk-sac absorption The number of days taken for yolk-sac absorption was reckoned from 50% hatching time to the stage when the yolk-sac was faintly v i s i b l e . This was by no means easy since the alevins were in varying stages of development within a l o t . The time recorded in Table II was only a rough e s t i -mate. Smith (1957) states "judging when the external yolk-sac i s no longer visi b l e i s rather a matter of aesthetics of larval embonpoint". Lafon's (1947) work on the u t i l i z a t i o n of yolk in trout involves analyses up to the stage when the yolk-sac was half consumed. Even this stage was d i f f i c u l t to judge in these observations. However, the results i n -dicated some measure of difference between species. In sockeye and pink salmon dark lots took the longest time. Activity of alevins in the lots under light could be a possible explanation for rapid u t i l i z a t i o n of yolk material. There i s also the possibility that greater endocrine secretions, stimulated by li g h t , enhanced growth. -31-This w i l l be dealt with in greater detail in the section on the relationship of pituitary and thyroid in the trout l o t s . Since pink salmon has a shorter l i f e cycle i t i s evident from the times given in Table II that a l l metabolic processes are faster. In kokanee and trout the dark lots took the short-est time. Light durations and intensities are very d i f -ferent in the case of trout and results may be due to this treatment as well as a function of the developmental processes of the species. It i s apparent in the experimental obser-vations that longer light durations and high intensities slow the rate of yolk absorption in trout. B. Mortalities The production of meristic differences between d i f -ferent lots of experimentally reared f i s h might be inter-preted as the result of selective mortality on different genotypes, rather than as direct phenotypic modification. Mortality in each experiment i s given in Tables I and II. The total mortality recorded i s the cumulative sum of a l l dead eggs, larvae and fry up to the end of the experiment. When eggs or fry were removed for other studies they were accounted for in the total mortality. If a certain number of eggs or fry (x) was removed at time t, knowing the total mortality of eggs or fry at that time, then the expected -32-mortality in the sample removed was calculated and added to the total mortality at time t. Experiments A and B Table I. Total mortality of kokanee expressed as a percentage of the f e r t i l i z e d eggs at the beginning of the experiment, was high in a l l lots, particularly in the larval and fry stages. Mortality of sockeye was slightly lower than in kokanee, and was more severe i n the egg stage. Percentage survival of pink was much higher than in sockeye or kokanee although mortality on the whole was s t i l l high. Mortalities in a l l lots of rainbow trout were very high particular in the 24 and 12-hour intensity l o t s . In these lots total mortalities were 96.7% and 93.0% respectively. Table I. Record of mortalities in experiments A and B Species Light dura- No. of fer- No. No. of No. of Total As % of No. As % of tion and i n - t i l i z e d hatched eggs larvae dead f e r t i l i - survi- f e r t i l -tensity eggs dead and fry zed eggs ving ized hrs.ft./cdl. dead eggs kokanee 16 5.9 844 654 190 542 732 86.7 112 13.3 Sept. 26, 8 5.9 1062 750 312 742 - - 8 -1957 to 16 1.0 878 623 255 530 785 89.4 93 10.6 Mar. 16, 8 1.0 912 632 280 538 818 89.4 94 10.4 1958 dark 1010 568 442 459 901 89.2 109 10.8 sockeye 16 5.9 717 200 517 111 628 87.5 89 12.5 Nov. 16, 8 5.9 528 127 401 32 433 82.0 95 18.0 1957 to 16 1.0 685 139 456 99 555 81.0 130 19.0 June 12, 8 1.0 507 110 397 31 428 84.4 79 15.6 1958 dark 684 243 441 116 557 81.4 127 18.6 pink 16 5.9 469 294 175 201 376 80.2 93 19.8 Oct. 12, 8 5.9 447 318 129 196 325 72.7 122 27.3 1957 to 16 1.0 402 240 162 175 337 83.8 65 16.2 Mar. 16, 8 1.0 432 243 189 123 312 72.2 120 27.8 1958. dark 508 358 150 252 402 79.1 106 20.9 trout 24 20.0 1754 179 1566 122 1688 96.7 57 3.3 Apr. 17, 12 19.0 1667 625 1042 509 1551 93.0 116 7.0 1957 to 24 0.3 1573 988 585 729 1314 83.5 259 16.5 July 31, 4 36.0 1647 1043 604 753 1357 82.4 290 17.6 1958. 4 0.4 1639 938 701 680 1381 84.2 258 15.8 dark 1703 860 843 658 1501 88.1 202 11.9 -34-Table II. Record of the source of eggs, hatching and yolk-sac absorption times and the time of preservation for meristic study of the various species used in ex-periments A and B. Species Source of Light dura- Hatch- Yolk sac Fixed for eggs and tion and ing absorption meristic date of intensity time (nearly study af-f e r t i l i z - (50%) complete) ter f e r t i -ation Time after l i z a t i o n 50% hatch-ing hrs.ft./cdl.Days D° Days D° Days D° Kootenay 16 5.9 58 693 26 312 169 2025 Lake, B.C. 8 5.9 58 693 not rec K it tt kokanee 16 1.0 59 705 28 336 tt it Sept. 26, 8 1.0 60 717 27 324 tt tt 1957. dark 63 753 26 312 it it Cultus 16 5.9 66 792 21 252 209 2508 Lake, B.C. 8 5.9 63 756 22 264 tt tt sockeye 16 1.0 63 756 25 300 tt it Nov. 16, 8 1.0 64 768 23 276 it tt 1957. dark 68 816 28 336 tt it Chehalis 16 5.9 29 348 12 144 156 1872 Creek, B.C. 8 5.9 29 348 12 144 tt tt pink 16 1.0 29 348 12 144 tt ti Oct. 11, 8 1.0 31 372 13 156 tt tt 1957. dark 31 372 15 180 it it Cultus 24 20.0 25 300 40 480 105 1260 La ke, B.C. 12 19.0 25 300 37 444 •« 1 1 trout 24 0.3 26 312 34 408 »« •» Apr. 17, 4 36.0 26 312 36 432 '» 1957. 4 0.4 27 324 33 396 '» dark 25 312 34 408 tt ft recorded Wc\t pttuierstlg of ^rtifeh (Eolumhta Faculty of Graduate Studies - P R O G R A M M E O F T H E FINAL ORAL EXAMINATION FOR THE DEGREE OF DOCTOR OF PHILOSOPHY of PASCARAPATHY CANAGARATNAM ' B. Sc., Ceylon, 1951 M. A., British Columbia, 1957 IN ROOM 187A BIOLOGICAL SCIENCES BUILDING FRIDAY, MAY 22, 1959 AT 10:30 A. M. C O M M I T T E E I N C H A R G E DEAN G. M. SHRUM: Chairman I. McT. COWAN W . N . H O L M E S C C. LINDSEY - P. FORD P. A. LARKIN H, B. HAWTHORN. R, F. SCAGEL . P. CONSTANTINIDES External Examiner: Dr. Ferris Neave Biological Station, Nanaimo THE INFLUENCE OF LIGHT INTENSITIES AND DURATIONS DURING EARLY DEVELOPMENT ON MERISTIC VARIATION IN SOME SALMON IDS ABSTRACT Experiments were designed to study the effects of various intensi-ties and durations of light during early development on meristic variation in kokanee and sockeye salmon (Oncorhvnchus nerka). pink salmon (O. gorbuscha) and in rainbow trout (Salmo gairdneri). An experiment with sockeye was designed exclusively to test the period of fixation of vertebrae. Among the salmon species rates of hatching and yolk-sac absorp-tion were fastest under the longer light durations and higher intensities. In sockeye and pink an increasing rate of yolk-sac absorption was cor-related with increasing amount of light. In trout the rate of yolk-sac absorption showed the opposite results. Mortalities were high under all the experimental conditions. Although meristic variability observed was deemed to be phenotypic yet the effects of selective mortality could not be entirely excluded. Abnormalities in the vertebral column were prevalent in the "pre-urostylic' region. The occurrence of abnormal vertebrae was correlated with light only in sockeye. The activities of the pituitary and thyroid glands of trout showed a positive correlation with higher amounts of light. Sockeye scale counts, along the lateral line and on the oblique rows from origin of dorsal and anal fins to lateral line, were lowest under conditions of higher light and longer durations. Fin ray numbers in all species were the lowest at high light intensities and longer durations. Vertebral counts were lowest at high light intensities and long durations in pink salmon and rainbow trout, but higher in sockeye. At lower light intensities and durations results were variable. Vertebral counts of sockeye increased with increasing light at temperatures of both 8 ° C and 12° C . Differences among lots at 8 ° C were significant but those among compar-able lots at 12° C were not. The action of light on meristic variability was weak at the higher temperature. It was found in sockeye that vertebral numbers were not fixed before 142 D ° and that the period of sensitivity was prolonged (142 D ° to 300 D ° ) . GRADUATE STUDIES Field of Study: Zoology Population Dynamics. P. A . Larkin Fisheries Biology and Management P. A . Larkin Comparative Ethology W. S. Hoar and M . D. F . Udvardy Other Studies: Fisheries Law G. F. Curtis Fisheries Economics '. . . . . . ; . . A . D. Scott Introduction to Dynamic Oceanography.. . G . L. Pickard Philosophical Problems . . . . . . . . . . . . . . . . B. Savery PUBLICATIONS Canagaratnam, P . , andJ.C. Medcof. 1955. Ceylon beach seine fishery. Fish. Res. Stn., Dept. Fish. Ceylon, Prog. Rep. 1: 17-23. Canagaratnam, P . , andJ.C. Medcof. 1956. Ceylon's beach seine fishery. Fish. Res. Stn., Dept. Fish. Ceylon. Bull. No. 4. 32 pp. Canagaratnam, P. 1959. Growth of fishes in different salinities. J. Fish. Res. Bd. Canada. 16(1): 121 - 130. -35-Experiment C Table III Mortalities ranged from 68% to 100%. Generally, the mortalities were much lower in comparison with the other experimental lots reared at 12°C. The high mortality was very unfortunate in a study of meristic v a r i a b i l i t y . A number of reasons could be given, as possible causes for the high mortalities. Morris (1956) found in a large number of rearing t r i a l s that mor-ta l i t y was not clearly associated with any specific symptom. The water supply in the laboratory was found to have very low total dissolved solids, v i z . 7-20/million, and the f i l t e r i n g system may have been imperfect. Hobbs (1937) stated that high mortality occurred in hatcheries supplied by artesian water. These mortalities were due to blue-sac disease or torsion of the vertebral column. Although malachite green solution was used regularly in a l l experiments fungus attack was not completely eliminated. Vibert (1949) experimentally demonstrated that mortality due to blue-sac disease in alevins hatched under gravel was appreciably lower than in alevins reared in hatchery troughs. Blue-sac disease was observed in only a few instances in the present experiments. Some dead alevins and fry had physical defects and a few were due to torsion of the vertebral column. The number of eggs used in each lot may have been -36-too great and alevins that hatched out were confined to a limited space. An important consideration i s the effect of hatching enzymes on unhatched eggs and embryos. Wintre-bert (1912) stated that enzymes released by the f i r s t eggs to hatch may attack the capsules of unhatched eggs i f they are held in close confinement. In these experiments eggs were rather crowded, but only in a single layer, and although water was renewed the possibility of the above effects can-not be ruled out completely. Hayes (1930) showed that the enzymes may attack not only the capsules of unhatched eggs, but the developing embryos as well. Many precocious hat-chings seen in these experiments may be due to these enzy-matic actions. The controlled temperature of 12°C was much higher than the temperatures at which these f i s h develop under natural conditions (fluctuating between 10°C and almost free-zing). Smith (1957) mentions that temperatures should not be too high or too low at the beginning of development. The salmon eggs were obtained from f i s h i n waters where tempera-tures were around 9°C or 10°C. In the case of rainbow trout (a spring spawner) temperatures in nature fluctuate, but are generally lower than the experimental one. The sockeye in experiment C were reared at temperatures comparable with that in nature, but the mortalities were relatively high. If temperature was high for early development of eggs, -37-mortalities should have been high before hatching. This cannot be demonstrated. -38-Table III. Record of mortalities in experiment C Basket Light No. of No. Total As % of No. As % of No. duration f e r t i - hatched dead f e r t i - survi- f e r t i -and intensity lized lized ving lized hrs. f t . / c d l . eggs eggs eggs 1 375 315 265 70.7 110 29.3 11 397 204 318 80.1 79 19.9 :.20 16 20 325 146 268 82.5 57 17.5 36 340 187 255 75.0 85 25.0 35 325 34 308 94.7 17 5.3 32 311 94 259 83.2 52 16.8 31 383 242 277 72.3 106 27.7 16 327 226 225 68.8 102 31.2 28 386 139 309 80.0 77 20.0 2 8 20 391 170 342 87.5 49 12.5 6 435 323 337 77.4 98 22.6 5 462 292 345 74.8 117 25.2 7 386 255 275 71.0 111 29.0 30 354 119 299 84.4 55 15.6 22 347 165 281 80.9 66 19.1 29 16 7 453 145 385 84.9 68 15.1 14 318 119 250 78.6 68 21.4 17 335 170 264 78.8 71 21.2 25 418 150 364 87.0 54 13.0 8 356 163 259 72.0 97 27.3 4 398 126 319 80.2 79 19.8 12 8 7 339 158 261 76.9 78 23.1 21 322 107 250 77.6 72 22.4 33 342 219 256 74.8 86 25.2 13 393 176 309 78.6 84 21.4 24 326 37 326 100.0 0 0 23 364 101 323 88.7 41 11.3 34 16 1 323 112 284 87.9 39 12.1 3 372 9 372 100.0 0 0 10 353 72 313 88.6 40 11.4 19 433 144 433 100.0 0 0. 18 352 160 284 80.7 68 19.3 15 295 107 339 81.0 56 19.0 26 dark 378 189 301 79.6 77 20.4 9 363 194 296 81.5 67 18.5 27 316 84 283 89.5 33 10.5 -39-The possibility that light treatments caused mortality was also considered. The only instances where mortality figures were obviously higher in illuminated than in dark treated lots were in the lots exposed to 12 and 24-hour durations with high intensities of experiment B. The proportion of dead eggs was also very high and i t seems as though high light intensities, like high temperature, may produce injurious effects in early development. There was no indication, except in the above mentioned cases, that light durations or intensities influenced mortalities. C. Abnormalities in the Vertebral Column The environment may modify not only the total num-ber of parts in meristic series, but also the number and nature of members of the series. Malformations in the skele-ton of teleosts, particularly in the vertebral column, are commonly encountered. Schmidt (1921), was f i r s t to dis-cover and describe vertebral abnormalities in sea-trout. Eandler (1935) in the plaice, Ford (1937) in the herring, o Taning (1944) in sea-trout, Ford (1947) in the order Isos-pondyli and Molander and Swedmark (1957) in the plaice are some of the other publications in which vertebral malforma-tions are mentioned. Many abnormalities are generally regarded as patho-logical, some congenital and others, as in the case of teleosts, -40-as caused by environmental factors especially temperature o (Taning 1944, Molander and Swedmark 1957). Abnormalities in teleostean vertebral columns are termed synostoses and classified according to comparable o structures in higher vertebrates. Taning (1944) catalogued vertebral abnormalities systematically. No attempt was made to categorize the abnormalities encountered in the present study, except to distinguish between true fusions and complex vertebrae (Kandler 1932, 1935). True fusions according to Kandler are characterised by vertebrae having double neural and haemal arches and also sutures or traces of sutures in the body (centrum) of the vertebra. Such vertebrae were considered as having arisen from two originally separate elements and therefore he counted them as two vertebrae. In the present study, whenever ver-tebrae of the above type were encountered they were counted as two. Schmidt (1921) expressed the opinion that, in the case of sea-trout, fused vertebrae have a value of 1.5 verte-o brae. Taning (1944) found that the 5th vertebra from the urostyle was very frequently abnormal. The centrum was from normal breadth to double breadth, with or without any trace of fusion. When there was no evidence of double origin in the centrum he examined the nature of the hypaxial element. He states, -41-"provided this element or the centrum i t s e l f i s of a distinctly double nature the 5th ver-tebral segment is considered composed of 2 vertebrae and is accordingly counted as 2." Owing to the inevitable establishment of size hierarchies within lots confined to a limited space (Brown 1946, Stringer and Hoar 1955),,the degree of ossification varied depending to a large extent on size (Figure 10). Therefore, lots in each experiment were reared for as long a time as possible. In a l l these experiments the various lots of any one species were k i l l e d and preserved at the same time. The number of days or day degrees for the different species used varied. Complex vertebrae according to Kandler (1935) are those vertebrae having accessory neural and haemal arches and where the centra of the vertebrae are, in most cases, longer than normal vertebrae. Such vertebrae were found in many speciments especially in the trout lots of experiment B. A l l complex vertebra were counted as one unless there was clear indication of fusion. Counts were always made on the l e f t side of f i s h . Ford (1947) found in order Isospondyli, to which salmonids belong, that vertebral variations on the two sides were in most cases large enough to give significant differences in s t a t i s t i c a l analyses intended to distinguish between species or between biological groups of the same species. This -42-source of error would not apply to comparisons of two experi-mental groups chosen at random from a common stock, except in the unlikely event that the two experimental treatments influenced different sides of the f i s h . It does however apply to comparisons between species in the present experi-ments. Most of the fusions or complexities were noted in the pre-urostyle region, and in or around the same position. Abnormalities in the abdominal region were few. Some of the typical vertebral abnormalities are illustrated i n Figure 11 selected from kokanee of experiment A, sockeye of experi-ment C, and trout experiment B. The frequency of occurrence of abnormalities in the various experiments were tabulated, classifi e d as abdominal or caudal and the total expressed as a percentage of the sample size of each l o t . (Table IV). Experiment A Abdominal malformations were few and occurred only in one instance in sockeye and in four of the five lots in kokanee. Caudal (or Taning*s "pre-urostyle") synostoses occurred in f a i r proportion except i n pink salmon where i t occurred only in the 8-hour lot s . The percentage frequency histograms (Figure 12) clearly show that a l l three groups (kokanee, sockeye and pink) in the 8-hour light durations have high vertebral malformations. The kokanee in 8-hour -43-X4 Figure 10. Five sockeye selected from experiment C section 1 basket No. 36 to ill u s t r a t e the different sizes in a lot and that the degree of vertebral ossification depends on size. Fertilized on 16th November 1957, k i l l e d and preserved on 26th June 1958. -44-high intensity was based on only 8 specimens (with one ab-normal) and cannot be considered a sufficiently large sample. However, sockeye and kokanee in the 16-hour and 8-hour dura-tions in both high and low intensities have a f a i r l y high percentage of abnormal vertebrae. Pink and sockeye in total darkness did not have any abnormal vertebrae. Kokanee showed a 3.8% abnormality in total darkness, much lower figure than in most of the other treatments. Experiment B A l l lots in trout showed a high percentage of ab-normalities (Figure 13). A great majority of these were classed as complex vertebrae and true fusions were few. Un-like the other experiments, the dark lot showed a high per-centage of malformations. Experiment C Only the control lots of each of the sections I-VI were taken for vertebral malformation study (Figure 14). These lots were a l l subjected to varying temperatures as indicated earlier. Of the six lots, abnormalities were found only in the caudal region of four l o t s . The total dark lot was free of abnormalities and although this lot -45-comprised only dead f i s h preserved about a month earlier than the other lot s , there was no trace of any abnormality in centra, neural or haemal arches or spines. The other lot that was devoid of malformations was the one in 16-hour high intensity. Analysis of vertebral abnormalities in experiments A and C shows that lots subjected to light were not influenced by the environment except in the case of sockeye. If f i s h with abnormal vertebrae in the dark lot were k i l l e d selec-tively then some abnormalities would have been expected in the dead f i s h in experiment C but this was not the case. Again, experiment B supports the view that selection against abnormalities did not take place in the dark lot (28.4%). The differences produced by intensities or durations of light are clear in the sockeye, but for a l l lots in the three experiments i t i s very d i f f i c u l t to generalize. If salmon and trout are treated separately, light appears to influence the frequency of abnormalities (phenotypic) in the former. In the case of trout the high percentage of verte-bral abnormalities i s probably due to several environmental and perhaps genetic conditions, and the role of light in effecting these abnormalities has been obscured. -46-Table IV. Occurrence of vertebral abnormalities in experi-ments A, B and C. Only control lots of experiment C were included. Light duration No. of vertebral No. of % of Species and intensity abnormalities sample sample hrs. f t . / c d l . Abdominal Caudal examined kokanee 16 5.9 1 4 80 6.3 sockeye - 2 45 4.4 pink — — 64 0 J S X D . A kokanee 1 8 12.5 sockeye 8 5.9 3 2 55 9.1 pink • — 3 80 3.8 E X T J . A kokanee 1 2 58 5.2 sockeye 16 1.0 - 3 95 3.2 pink - - 35 0 Exp. A kokanee 5 80 6.3 sockeye 8 1.0 - 7 59 11.9 pink - 1 80 1.3 Exp. A kokanee 2 1 80 3.8 sockeye dark - - 86 0 pink - - 36 0 Exp. A 24 20.0 4 30 13.3 12 19.0 - 26 68 38.2 trout 24 0.3 - 28 80 35.0 Exp. B 4 36.0 - 24 80 30.0 4 0.4 - 24 80 30.0 dark — 21 74 28.4 16 20.0 _ 93 0 8 20.0 - 11 90 12.2 sockeye 16 7.0 - 8 94 8.5 Exp. C 8 7.0 — 3 49 6.1 16 1.0 - 8 72 11.1 dark - - 50 0 -47-1 Figure 11. Vertebral synostoses (fused and complex) in some of the experimental lo t s . Pre-urostyle or caudal ab-normalities are referred to below from the last vertebra. 1. 2. and 3. are from sockeye in experiment C. l.-4th caudal with two neural arches and spines. 2.-4th caudal with 2 neural arches and one spine, 5th caudal 2 haemal arches and one haemal spine. 3.-4th 2 neural arches 1 spine; 5th 2 ele ments of centrum v i s i b l e , 2 haemal spines - counted as 2 vertebrae. 4. Trout - 5th caudal counted as 2 vertebrae, 2 elements in centrum, 2 neural spines and 2 haemal spines 5. Kokanee - 9th and 10th abdominal vertebrae fused. -48-10 CO U) O m < 0 1 OARK HIGH LOW 8 HOURS 23 K O K A N E E ^ S O C K E Y E PINK HIGH LOW 16 HOURS Figure 12. Experiment A. Histograms of percentage abnormal vertebrae i n kokanee, sockeye and pink salmon. CO UJ 10 ir o z CD < 0 DARK 2 0 f t / c d l . + 7ft/cdl . 8 HOURS 20ft/cdl.~r» 7 —*• I ft/cdl. 16 HOURS Figure 14. Experiment C. Histograms of percentage abnormal vertebrae in sockeye control l o t s . 4 0 -49-35 3 0 25 co UJ 5 2 0 cr o z m < 10 0 DARK H L 4 HOURS 12 HOURS H — J — L 2 4 HOURS Figure 13. Experiment B. Histograms of abnormal vertebrae i n rainbow trout. -50-Meristic Variations Meristic counts of the various series are given in summarized form in Tables V to XI. Correlation of Meristic Counts with Length of Fish A l l series of meristic counts within l o t s , used in the analyses, were tested for correlation with length of f i s h . This was done by plotting individual counts against the length of f i s h . Although inspection of many plots showed no correlation, Chi-square tests were performed to eliminate doubt. Since meristic counts were not correlated with length of f i s h , i t was evident that the number of parts in the various series was determined at an early stage of development. Method of Analyses Mean counts were plotted against durations, the two intensities forming separate series with the mean of the total dark lot as a common reference point. When d i f f e r -ences produced by light durations or intensities were doubt-f u l , s t a t i s t i c a l analyses were carried out. Tests of s i g -nificance for these analyses were based on " t " and"F" distribu-tions (Snedecor, 1957). Significance was taken at the -51-0.01 level. Experiment A - Sockeye Scales (i) Lateral line scales. Table Va. Figure 15 shows the mean values for the two intensities i n both durations. In the 8-hour durations the lower intensity had a lower mean count than the higher intensity, whereas in the 16-hour duration the order was reversed. Differences between the dark, 8-hour low intensity and 16-hour high intensity were s i g -nificant. Means of both intensities in the two durations when analysed separately showed significant differences. High intensity of 16-hours duration produced the lowest mean count while the 8-hour duration of same intensity produced the highest mean count. The difference bet-ween these two means was also significant. Although light durations as well as intensities produced s i g n i -ficant differences in the mean counts, there was no apparent pattern in the distribution of means in the high and low intensities. ( i i ) Scales on oblique row from origin of dorsal f i n to l a t -eral l i n e . Table Vb. Figure 16 shows that the pattern of the mean -52-counts for the high and low intensities i s somewhat similar to the means of lateral line scales. High intensity of the longer duration produced the lowest mean count. Differences between means of dark and 16-hour high intensity, between the two intensities of the 16-hour lots, and between high intensity lots of 8 and 16-hour durations were a l l significant. ( i i i ) Scales on oblique row from origin of anal f i n to lateral l i n e . Table Vc. The pattern of the low intensity means (Fig-ure 17) departed from the series considered above. Differences between means of both high intensity dura-tions and dark lots were significant. Difference be-tween means of the two intensities in the 8-hour dura-tion was also significant. Sockeye secondary caudal ravs A l l caudal rays were not included in analyses because va r i a b i l i t y was only observed in the unjointed, un-branched rays in upper and lower folds of the caudal f i n anterior to the caudal fan. Primary rays constituting the caudal fan were counted and found to be constant (19) i n a l l lots. Frequency distribution (Table VI a and b) in both 53-131.5 128.0 1 • 1— : I 0 8 |6 DURATION (HOURS) Figure 15. Influence of light on lateral line scales of sockeye in experiment A. 54 19.75 18.50 1 1 : 1 0 8 16 DURATION (HOURS) Figure 16. Influence of light on scales from origin of dorsal f i n to lateral line of sockeye in experiment A. -55* 18.0 17.0 0 8 DURATION ( H O U R S ) Figure 17. Influence of light on scales from origin of anal f i n to lateral line of sockeye in experiment A. Table V a. Frequency distribution of lateral line scales of sockeye in experiment A Light duration and intensity Lateral line scales hrs. f t . / c d l . 125 26 27 28 29 130 31 32 33 34 35 36 37 38 39 140 41 Number Mean 16 5.9 8 12 6 4 3 3 2 2 2 1 1 1 - _ 45 128.11 8 5.9 1 1 5 6 2 5 6 10 8 2 4 2 1 1 - 1 - 55 131.40 116 1.0 12 5 9 9 12 10 5 11 6 3 6 2 2 1 2 - 95 130.02 8 1.0 5 5 4 4 10 9 6 6 5 2 3 59 129.68 dark 3 6 7 10 9 9 8 7 8 6 2 2 1 5 - - 3 86 130.97 Table V b. Frequency distribution of scale counts from origin of dorsal f i n to lateral line along the oblique row of sockeye in experiment A.  Light duration and intensity Oblique row scales hrs. f t . / c d l . 17 18 19 20 21 22 Number Mean Table V c. Frequency distribution of scale counts from origin of anal f i n to lateral line along the oblique row of sockeye in experiment At  Oblique row scales 14 15 16 17 18 19 20 Number Mean 16 5.9 5 16 17 2 5 - 45 18.68 - 2 9 21 11 2 - 45 17.04 8 5.9 - 5 21 21 7 ' 1 55 19.60 - - 3 16 22 11 3 55 17.91 16 1.0 9 12 37 25 11 1 95 19.21 1 5 25 31 23 9 1 95 17.06 8 1.0 - 9 26 19 5 - 59 19.34 — - 11 17 27 3 1 59 17.42 dark - 14 26 34 11 1 86 19.52 — — 15 28 34 9 — 86 17.43 -58-upper and lower caudal f i n ray counts showed a trend towards lower counts with increasing l i g h t . Mean counts of lower secondary rays were much higher than mean counts of the upper secondary rays (Figure 18). Analyses of the means of the upper rays showed significant differences between dark and both high intensity lots and those of the intensities of 8-hour duration. The only significant difference in means of the lower rays was between dark and 16-hour high intensity. Variation in mean ray counts with light durations and intensities showed a regular pattern when compared with the mean count of the dark l o t . Longer durations or higher intensities produced lower mean counts than the mean count of the dark, although differences between many lots were not significant. Total Vertebrae (kokanee, sockeye and pink) Table VII. Mean counts of the three groups were represented together in Figure 19 to show differences among species as well as within a species. None of the differences between means of any treatment in sockeye were significant. In ko-kanee only the low intensity durations were considered in analysis (because a l l except eight f i s h were lost accidently in the 8-hour high intensity lot) and i t was found that dark and 16-hour durations were similar, but that difference bet-ween mean count of 8-hour duration and either dark or 16-hour -59-Table VI a. Frequency distribution of upper secondary caudal f i n ray counts in sockeye, experiment A. Light duration and intensity Upper rays hrs. f t . / c d l . 10 11 12 13 14 15 Number Mean 16 5.9 9 13 15 5 2 1 45 11.58 8 5.9 6 15 26 5 3 - 55 11.71 16 1.0 11 23 34 23 4 - 95 11.85 8 1.0 - 5 16 27 10 1 59 12.76 dark 1 5 25 30 21 4 86 12.89 Table VI b. Frequency distribution of lower secondary caudal f i n ray counts in sockeye, experiment A. Light duration and intensity Lower rays hrs. f t . / c d l . 12 13 14 15 16 Number Mean 16 5.9 3 17 20 3 - 45 13.60 8 5.9 - 11 41 3 - 55 13.85 16 1.0 5 27 52 11 - 95 13.73 8 1.0 - 5 29 24 1 59 14.36 dark - 4 50 26 6 86 14.40 -60-Table VII. Frequency distributions of total vertebral counts in experiments A and B. Light duration Species and intensity Hum- Mean hrs. f t . / c d l . 59 60 61 62 63 64 65 66 67 68 ber kokanee 16 5.9 - 7 35 27 11 - 80 64.53 8 5.9 - 2 4 2 - - . 8 64.00 16 1.0 - 6 15 28 5 4 58 64.76 8 1.0 4 13 24 31 7 1 80 64.34 dark 1 2 22 44 8 3 80 64.81 sockeye 16 5.9 - 4 17 21 3 45 64.51 8 5.9 - 8 20 24 3 55 64.40 16 1.0 1 11 43 37 3 95 64.32 8 1.0 1 9 26 19 4 59 64.27 dark 1 12 43 25 5 86 64.24 pink 16 5.9 . _ - - • - 5 21 24 12 2 - 64 64.76 8 5.9 15 38 19 6 2 80 65.27 16 1.0 1 - - 4 6 12 9 3 - - 35 63.89 8 1.0 - - - 1 5 24 30 19 1 - 80 65.20 dark - - - - 2 7 20 5 2 - 36 64.94 trout 24 20.0 1 8 17 4 - 30 61.80 12 19.0 - 13 32 20 3 68 62.19 24 0.3 1 4 35 35 5 80 62.49 4 36.0 - 10 33 32 5 80 62.40 4 0.4 - 6 22 46 6 80 62.65 dark - 9 30 30 5 74 62.42 -61-durations was significant. Although sockeye and kokanee are morphologically similar (Ricker 1938, 1940) the mean counts of dark lots showed significant difference. Mean counts showed inverted V-shaped curves for pink salmon and the reverse pattern in kokanee. Some of the significant differences between means of the various lots in pink are clearly v i s i b l e i n Figure 19. Abdominal and Caudal Vertebrae (kokanee and pink) Tables VIII and IX. In pink salmon, abdominal and caudal counts (Fig-ures 20 and 21) display a form similar to total counts in a l l lots except the 8-hour low intensity abdominal count and 16-hour high intensity caudal count. Significant differences between means were found in 16-hour intensities of both ab-dominal and caudal counts and also between dark and 16-hour duration low intensity in abdominal count. In kokanee the 8-hour high intensity mean count was taken from only 8 specimens (many were accidentally lost dur-ing the early fry stage). The means were used in graphical presentation but were not considered in tests of s i g n i f i -cance. The V-shaped curve of mean abdominal count at the low intensity corresponds closely to the respective mean total count curve. -62-Comparison of Lateral Line Scales and Total Vertebral Counts  of Sockeye Mean counts of lateral line scales of sockeye reared under the various light treatments showed significant differences but failed to show any distinct trend between intensities or durations. Total vertebral means of a l l lot s , though s t a t i s t i c a l l y insignificant, suggested a gradual i n -crease with increasing l i g h t . However, the scale-vertebra ratio remained in a l l cases, as shown in data contained in Foerster and Pritchard (1935a), at approximately 2:1. What was evident from the present study was that increase in mean vertebral count was not accompanied by an increase in mean lateral line scale count. Dorsal and Anal Fin Ravs (kokanee, sockeye and pink) Tables X and XI Within sockeye lots or kokanee lots differences between means were t r i v i a l , but differences between some of the comparable means of these two types were significant, particularly in anal f i n ray counts. Dorsal and anal ray counts of pink salmon showed a similar pattern, though many differences between means of both counts were small and s t a t i s t i c a l l y insignificant. Light -63-Table VIII. Frequency distributions of abdominal vertebral counts in experiment A and B. Light duration Species and intensity Num-hrs. f t . / c d l . 34 35 36 37 38 39 40 41 42 ber Mean kokanee 16 5.9 - - 6 26 41 5 2 80 37.64 8 5.9 1 4 3 - - - - 8 35.25 16 1.0 - - 2 22 24 8 2 58 37.76 8 1.0 1 3 15 26 30 5 - 80 37.20 dark - - 2 30 39 8 1 80 37.70 pink 16 5.9 8 5.9 16 1.0 8 1.0 dark - - 2 14 27 17 4 - - - 8 35 26 10 1 - 5 5 20 3 1 - •- 2 20 31 22 5 - - - 4 14 18 -- - 64 38.11 - 1 80 38.53 - - 35 37.60 - - 80 38.10 - - 36 38.39 trout 24 20.0 1 4 13 8 3 1 - 30 36.63 12 19.0 - 1 22 31 14 - 68 36.85 24 0.3 - - 10 26 34 10 - 80 37.55 4 36.0 - 12 30 30 8 - 80 37.43 4 0.4 - - 6 26 42 6 - 80 37.60 dark 5 15 42 11 1 74 37.84 -64-Table IX. Frequency distribution of caudal vertebral counts in experiment A and B. Light duration Species and intensity Caudal vertebrae Num-hrs. f t . / c d l . 22 23 24 25 26 27 28 29 ber Mean 16 5.9 1 23 40 9 80 26.80 8 5.9 - - 1 7 - 8 27.87 kokanee 16 1.0 - 17 24 17 - 58 27.00 8 1.0 1 17 38 19 - 80 27.13 dark 11 49 20 *** 80 27.11 16 5.9 _ 4 23 28 9 64 26.66 8 5.9 - 2 26 42 10 - 80 26.75 pink 16 1.0 3 4 13 10 5 - 35 26.29 8 1.0 - 5 25 40 9 1 80 26.70 dark mm 2 16 15 2 1 36 26.56 24 20.0 mm, 2 3 10 11 3 1 30 25.43 12 19.0 - 1 8 27 31 1 - 68 25.34 trout 24 0.3 - 2 25 33 16 4 - 80 24.94 4 36.0 - 2 16 44 18 - - 80 24.97 4 0.4 - - 14 49 16 1 - 80 25.05 dark 1 8 24 30 10 1 74 24.58 6 3 . 8 0 8 16 DURATION ( H O U R S ) Figure 19. Influence of light on total vertebrae of kokanee, sockeye and pink in experiment A. -66-38 .6 0 8 ' 16 DURATION (HOURS) Figure 20. Influence of light on abdominal vertebrae of kokanee and pink i n experiment A. -67-2 8 . 0 2 7 . 8 \— .6 < ^ . 4 < 3 O o < <x CD """" h -cr £ 2 7 . 0 26.2 PINK ? ONLY 8 SPECIMENS 0 . 8 16 DURATION (HOURS) Figure 21. Influence of light on caudal vertebrae of kokanee and pink in experiment A. 15.5 68-DURATION (HOURS) Figure 22. Influence of light on dorsal f i n rays of kokanee, sockeye and pink in experiment A. -69 Figure 23. Influence of light on anal! f i n rays of kokanee, sockeye and pink in experiment A. -70-durations of 8-hour and 16-hour produced significant d i f f e r -ences between means within the respective intensities. Experiment B Rainbow Trout Vertebrae Tables VII-IX In total vertebral counts differences between means, in the two intensities of 24-hour duration, dark and 24-hour high intensity, and 4-hour high and 24-hour high were s i g n i -ficant. High intensities of longer durations produced lower mean counts (Figure 24). The graph of mean abdominal vertebrae (Figure 25) showed similar curves, as in total counts, except in the case of the 4-hour duration lots in which mean counts were lower than for the dark l o t . Significant difference between means was noted to be the same as in total vertebrae. Mean caudal counts of the high intensity lots showed opposite results — longer durations and higher intensities produced higher mean counts (Figure 26). The curve for low intensity lots was similar to the corresponding curve for total vertebrae. However, significant differences between means for lots described in total vertebral counts were also significant in this case. Dorsal and Anal Fin Rays Tables X and XI Means of dorsal f i n ray counts, though distinct for -71-the two intensity treatments showed significant differences only between the two intensity lots of 24-hour duration and between dark and 24-hour high intensity. As seen earlier in vertebral counts longer durations of higher intensities produced lower mean counts. Anal f i n ray counts did not show significant d i f f e r -ence between means of any of the lots reared under the various conditions (Figure 27). 6 2 . 8 < o ZD o c_> < CE 00 L d t -cr > w 6 2 . 0 h 61.8 4. I DURATION ( H O U R S ! Figure 24. Influence of li g h t on total vettebray of rainbow trout i n experiment B. 73-Figure 25. Influence of light on abdominal vertebrae of rainbow trout i n experiment B. (sanc-H) NOiivano. \pZ 21 . i7 o I — J — ' 1 g/sz Figure 26. Influence of light on caudal vertebrae of rainbow trout in experiment B. DURATION ( H O U R S ) Figure 27. Influence of light on dorsal and anal f i n rays of rainbow trout i n experiment B. -76-Table X. Frequency distribution of dorsal f i n ray counts in experiments A and B. Light duration Species and intensity Dorsal f i n rays hrs. f t . / c d l . 13 14 15 16 17 Number Mean 16 5.9 3 40 36 1 80 14.44 8 5.9 - 1 7 - 8 14.87 kokanee 16 1.0 1 30 27 - 58 14.45 8 1.0 5 43 32 - 80 14.47 dark 3 30 44 3 80 14.59 16 5.9 7 35 3 mm 45 13 .91 8 5.9 5 33 16 1 55 14.23 sockeye 16 1.0 13 70 12 - 95 13.99 8 1.0 6 31 31 1 59 14.29 dark 6 51 28 1 86 14.28 16 5.9 24 34 6 64 14.72 8 5.9 5 45 30 - 80 15.31 pink 16 1.0 20 14 1 - 35 14.46 8 1.0 4 40 35 1 80 15.41 dark 7 17 12 36 15.14 24 20.0 12 16 2 30 14.66 12 19.0 11 42 15 - 68 15.06 trout 24 0.3 6 511 22 1 80 15.23 4 36.0 7 49 22 2 80 15.24 4 0.4 3 48 28 1 80 15.34 dark 8 37 27 2 74 15.31 -77-Table XI. Frequency distribution of anal f i n ray counts in experiments A and B. Species Light duration and intensity Anal rays Num-hrs. f t . / c d l . 12 13 14 15 16 17 18 19 ber Mean 16 5.9 I 32 45 2 80 16.60 8 5.9 - 3 5 - 8 16.62 kokanee 16 1.0 - 16 41 1 58 16.74 8 1.0 2 36 38 4 80 16.55 dark 2 24 51 3 80 16.69 16 5.9 - 11 31 3 45 16.82 8 5.9 - 10 36 9 55 16.98 sockeye 16 1.0 1 11 65 18 95 17.05 8 1.0 - 4 38 17 59 17.22 dark - 5 57 24 86 17.22 16 5.9 1 16 35 12 64 16.91 8 5.9 - 4 37 38 1 80 17.45 pink 16 1.0 1 8 21 5 - 35 16.86 8 1.0 - 1 29 48 2 80 17.64 dark - 4 17 15 36 17.31 24 20.5 3 10 13 4 30 13.60 12 19.0 - 24 33 11 68 13.81 trout 24 0.3 1 32 43 4 80 13.63 4 36.0 1 24 46 9 80 13.79 4 0.4 - 19 54 7 80 13.85 dark •a 18 49 7 74 13.85 -78-E. Histological Study of Pituitary and Thyroid Glands of Rain- bow Trout (Experiment B) The pineal organ of fishes, a recognized photo-receptor (Breder and Rasquin, 1950), was li s t e d among en-docrine tissues, at least in Lebistes (Pflugfelder, 1953 and 1954) where i t was suggested that action of pineal se-cretions was mediated through pituitary and thyroid glands. There is also evidence that the pituitary-thyroid mechanism is affected by the amount of illumination (Hoar, 1957). Therefore, these two organs were studied to discover whether or not there might be a correlation between endocrine activ-it y and meristic variations. Method Samples of eggs, alevins, and fry were taken peri-odically from a l l lots of experiment B and fixed in Bouin's f l u i d for histological examination of pituitary and thyroid. Standard procedures were used in dehydration, wax embedding and seria l sectioning. Pituitary Gland Since specimens used were small the whole heads were sectioned, 8 microns thick, in a para-sagittal plane. -79-The ribbon containing pituitary sections of each f i s h was mounted on one slide. Several staining techniques were tried to bring out the acidophil and basophil c e l l s clearly, but in every instance, the whole of the glandular or epi-t h e l i a l components appeared basophilic. Heidenhain's Azan stain as described by Pantin (1948) was found most sat i s -factory. Although sections were made of several stages in trout development the only stage in which a complete series was obtained for comparative study was in the fry stage (28-32 mm). In earlier stages (6-15 mm) the pituitary primor-dium was a tiny group of undifferentiated ce l l s and often indistinguishable from the surrounding mass of buccal epi-thelium. Analyses The para-sagittal section nearest the median line was selected for study. This section was easily recogniz-able since i t had the longest antero-posterior measurement and also the infundibular cavity opened into the third ven-t r i c l e of the brain (Figure 28). None of the c e l l types described in telostean pituitary gland (Scruggs 1939, Woodman 1939, Kerr 1940 and -80-1942, Bretschneider and Duyvene de Wit 1947) was discernible. This was not due to the method of staining but to the early undifferentiated stage studied (Woodman 1939). Demarcation of the main parts (pars anterior, pars tuberalis and pars intermedia) in the glandular or epithelial segment was also not distinct (Figure 28). It was then decided to consider the entire area of the epithelial segment. Beach (1956) used the entire basophilic area of the pituitary gland of adult goldfish to correlate seasonal changes in pituitary with development of the ovary. Camera lucida drawings of epithelial segments of the pituitary were made, and the area was measured by means of a polar planimeter. This area was divided by the c a l i -bration values of objective and eye piece used to obtain the actual area. Thyroid Gland Transverse s e r i a l sections, cut 10 microns thick, were stained with Ehrlich's haematoxylin and eosin. Sec-tions at the beginning of the second g i l l arch were found to contain the largest number of f o l l i c l e s , concentrated around the ventral aorta (Figure 29), and these were used in the analyses. -81-Analvses The number of f o l l i c l e s was counted and the size of a l l f o l l i c l e s was measured in microns, along the long axis, and epithelial height was taken at several points on each f o l l i c l e . The means for each of the above mentioned were obtained for entire samples in each l o t . The percen-tage of collo i d in the lymphatic space was estimated and the presence of peripheral vacuoles in the collo i d was expressed as a percentage of the total number of f o l l i c l e s . In addi-tion, the ratio of granular to clear col l o i d and also the ratio of eosinophilic to basophilic colloid were estimated. These measurements and observations were made to express the activity of the thyroid gland in the various l o t s . Size of Pituitary Epithelial Components in a l l Lots Table XIla shows the mean areas for the various lots together with the mean length of fish used. High light i n -tensity lots showed an increase in area of epithelium with increasing durations (Figure 30). Activity of Thyroid Gland Histological examination alone i s not the best c r i -terion of thyroid activity, but i t is generally believed that low squamous and high or folded columnar c e l l s indicate i n -activity or overactivity respectively, (Maximow and Bloom 1957). -82-It has been demonstrated experimentally that increased a c t i -vity of the thyroid gland is reflected by the presence of peripheral vacuoles (artefacts in themselves), and by more basophilic and granular colloid resulting from enzymatic hydrolysis (De Robertis, 1949). Dales and Hoar (1954) found that thyroxine and thiourea treated chum salmon showed marked changes in early development and this was correlated with activity of thyroid gland. Histological characteristics were used to evaluate activity of thyroid gland. Measurements and observations made, in the present histological study of trout thyroid gland, in each of the various experimental lots, were collated (Table XIlb) so that elucidation of activity, though complicated, was more accurate. Examination of the characteristics analysed re-vealed that the thyroids in a l l lots in high light intensi-ties were more active than those in the low intensity l o t s . The dark lot was the least active and the 24-hour high inten-sity most active. Area of Pituitary Epithelium and Thyroid Activity The data obtained from the pituitary and thyroid analyses suggested that there was some relationship between the area of the former and activity of the latt e r . This was evident in only extreme conditions, such as in dark and 24-hour high light intensity. In these two cases small area was correlated with low activity and vice versa. Table XII a. Mean glandular area of pituitary of rainbow trout in experiment B. Light Mean Mean glandular area duration length of pituitary and intensity and No. hrs. f t . / c d l . mm. ( ) mm2 24 20.0 30.5 (6) 0.1132 12 19.0 30.9 (9) 0.0952 24 0.3 29.0 (9) 0.0863 4 36.0 29.9 ( ID 0.0886 4 0.4 30.8 (9) 0.1091 dark 29.6 (7) 0.0827 00 CO Table XII b. Histological study of thyroid gland in rainbow trout of experiment B. Measure- 1 ments and observations were made in transverse sections at the beginning of the second g i l l arch and around ventral aorta. Light dura- Mean Average Average Average % Colloid Presence of Ratio of tion and i n - length No. of size of height of in lymph- peripheral Granular: Eosinophilic: tensity and No. f o l l i c l e s f o l l i c l e s epithelium atics vacuoles as clear c o l l - basophilic hrs. f t . / c d l . mm. ( ) (microns) (microns) % f o l l i c l e Nx oid. G:C colloid.E:B 24 20.0 29.5(6) 11.3 44.7 6.0 57 25 1:1 2:3 12 19.0 30.8(6) 10.7 45.2 4.6 58 22 1:4 3:2 24 0.3 29.4(6) 11.8 29.3 5.2 44 28 1:9 3:2 4 36.0 29.8(6) 11.0 40.0 4.6 56 60 1:1 1:1 4 0.4 30.5(6) 9.8 40.0 4.5 70 35 1:2 2:1 dark 28.8(6) 11.0 26.2 4.5 40 5 1:6 4:1 -84-/ 8 0 0 X g o o Figure 28. Median longitudinal sections of trout p i t u i -tary, experiment B. 1-24 hours high i n t e n s i t y ; 2-dark; 3-4 hours high i n t e n s i t y , (a) - t h i r d ventricle} (b) infundibular c a v i t y ; (c) pars nervosa; (d) - e p i t h e l i a l l a y e r . -85-Figure 29. Transverse sections of thyroid f o l l i c l e s i n trout, experimental B. 1-beginning of second g i l l arch where analyses were done; 2-4 hours high i n t e n s i t y ; 3-Dark, note photograph taken below and to l e f t of ventral aorta since f o l l i -c l e s were absent around blood v e s s e l ; 4-24 hours high i n t e n s i t y . ',» 0 . 0 8 1 ' 1 : 1 0 4 12 2 4 DURATION (HOURS) Figure 30* Influence of li g h t on area of epithelial components of pituitary gland of rainbow trout i n experiment B. -87-Experiment C - Period of Fixation of Vertebrae in Sockeye Transfers of developing sockeye eggs, kept at identical temperatures, were effected at three different embryonic stages. Emphasis was placed on light durations and hence the main results were based on transfers between 8-hour and 16-hour high intensity lots where reciprocate transfers were made at a l l three stages. Other transfers made were between intensities or durations at one or two of the three stages selected (Figure 9b). Vertebral Counts of Controls in Sections I-VI The frequency distribution of total abdominal and caudal vertebral counts for the various lots (baskets) in the different sections are given i n Tables XIII-XV. Mean total counts of control lots were compared to ascertain the extent of var i a b i l i t y between durations and intensities. The means of total vertebral counts of the control lots (sections I-VI) are represented in Figure 31. The mean counts of dark, 8-hour duration ( i f high intensity lot could be excluded) and 16-hour duration ( a l l intensities) showed a trend towards increase in vertebrae with increasing light, and a well-graded difference between intensities as well as between durations. -88-Figure 31. Influence of light on total vertebrae of sockeye control lots i n experiment C. -89-The degree of va r i a b i l i t y was very high between durations and between intensities in 16-hour duration (Fig-ure 31). Difference between mean counts of dark (61.80) and 16-hour high (63.80) or medium (62.91) intensities, was significant. The low intensity lot of 16-hour duration had a mean count of 62.09 and although not significantly d i f f e r -ent from the mean of the dark lot showed a general trend of increased numbers. The 8-hour duration lots of the high and medium intensities had the position of their mean counts in reverse to that seen in the corresponding lots of 16-hour duration. The difference between the mean counts in the lots of the 8-hour intensities was small, but that between high intensity lots of 8-hour and 16-hour durations was mark-edly distinct. Transfers Between 8-hour and 16-hour High Light Intensity Percentage frequency polygons were drawn for con-trols and each of the transferred lots at the different times. By this method i t was possible to show the difference in the modes in relation to the controls. In total vertebral counts (Figure 32) the lot transferred from 8-hour to 16-hour after 16 days (142D°) showed that the mode as well as the mean was similar to the control of 16-hour, whereas those transferred after 24 days, (206D°) and 36 days (300 D°) had modes and means which resembled that of the 8-hour control. This means -90-that after a developmental period of 206 D° no change in fre-quency or mean of vertebral counts was effected by the new environment, i.e. lig h t . Transfers from 16-hour to 8-hour showed comparable changes in the respective periods, but the percentage frequency distributions and means of the transfers at 206 D° and after were not as distinct as in the reciprocal transfers. However, the f i r s t transfer from 16-hour to 8-hour at 142 D° had a mean count and frequency distribution resembling the control of the 8-hour section. The abdominal vertebral counts of the same lots in the same transfers did not appear to show the distribution of percentage frequencies or modes as in the total counts. The difference between the means of the control lots was i n -significant, and also among controls and transferred lots, except in the lot transferred at 142 D° from 8-hour to 16-hour where the mean was greater than either control (Figure 33). The percentage frequency polygons of caudal verte-bral counts, of the controls and transferred lots, from 8-hour to 16-hour, resembled those of the total counts. In the reciprocal transfers the pattern was uniform and there was no indication of change with reference to controls, though the difference between the means of the controls was significant (Figure 34). -91-Other Transfers Lots transferred from 8-hour medium intensity to 16-hour of similar intensity and to dark at different stages revealed some interesting results. In total vertebral counts the lot transferred from 8-hour to dark at 206 D° did not show any appreciable change, but another lot transferred at 300 D° was unlike either control. A similarly aberrant result was obtained in the transfer made from 8-hour to 16-hour at 206 D° (Figure 35a). It was shown earlier that there was no consistency in abdominal and caudal vertebral counts, therefore, they were not considered in these trans-fers. One transfer from 16-hour low intensity to 8-hour high intensity was effected at 206 D°. Percentage frequency polygons showed that the transferred lot was completely different from either controls. The means of the control lots differed by only 0.17 vertebra but the transferred lot had a mean count of 63.28, an increase of over one vertebra (Figure 35b). -92-Table XIII. Frequency distribution of total vertebral counts in experiment C. Bas-Light duration Sect, ket and intensity Total vertebrae Num-No. No. hrs. f t . / c d l . 58 59 60 61 62 63 64 65 66 ber Mean 1 mm mm 3 30 38 19 3 93 63.88 11 - 1 21 27 19 3 - 71 63.03 I 20 16 20 - 8 22 16 2 1 - 49 62.31 36 - 1 8 27 31 5 2 74 63.50 35 - 5 4 4 3 - - 16 62.31 32 1 9 20 16 mm *• 46 62.10 31 10 8 33 27 11 1 90 62.26 16 - 3 14 33 36 5 91 63.28 VI 28 8 20 6 9 38 11 8 1 73 62.12 2 2 4 19 14 5 2 46 62.48 6 - 4 20 33 20 1 78 62.92 5 - 4 29 43 22 8 106 63.00 7 1 7 21 37 26 2 94 62.91 30 2 4 18 19 5 - 48 62.44 II 22 16 7 1 4 17 24 9 3 - 58 61.78 29 - 6 17 15 11 5 1 55 61.91 14 - 1 6 20 25 6 1 59 62.54 17 1 - - 3 26 24 7 1 62 62.55 25 — 2 6 15 21 5 — 49 62.43 8 - 4 i i 25 26 11 1 78 62.41 V 4 8 7 - 1 7 18 35 10 3 74 62.74 12 1 3 25 35 6 1 - 71 61.64 21 - 2 19 30 13 2 - 66 61.91 33 - 5 23 29 18 1 76 62.83 13 4 16 27 19 6 72 62.09 III 23 16 1 3 5 15 11 5 39 62.26 34 2 5 16 10 - 33 62.03 10 - 5 16 10 4 35 62.37 19 — 5 14 19 10 2 — 50 61.80 18 2 15 16 18 12 - - 63 61.37 IV 15 dark - 4 18 17 11 - 1 51 61.77 26 - - 4 28 30 6 - 68 62.56 9 - 6 18 25 6 3 - 58 61.69 27 1 8 9 9 2 1 1 31 61.33 -93-Table XIV. Frequency distribution of abdominal vertebral counts in experiment C. Sect. No. Bas-ket No. Light duration and intensity hrs. f t . / c d l . 32 33 34 35 36 37 38 39 Num-ber Mean 1 1 13 40 27 12 93 36.39 11 5 13 26 25 2 71 36.08 I 20 16 20 4 9 20 13 3 - 49 35.04 36 - - 4 18 36 16 74 36.86 35 - 1 3 3 7 2 16 36.37 32 1 7 30 8 46 35.98 31 1 5 14 32 30 8 - 90 36.21 16 - 5 20 31 34 1 91 37.06 28 8 20 - - 13 26 25 9 - 73 36.42 2 - 4 3 22 11 6C - 46 36.27 6 - 3 32 23 20 - 78 36.77 5 - 1 6 40 39 20 - 106 36.67 7 - 1 1 22 38 25 7 94 36.13 30 - 2 5 12 14 12 3 48 35.80 22 16 7 1 5 16 18 16 2 - 58 34.84 29 - 1 9 26 12 7 55 36.27 14 - 1 9 14 22 13 - 59 35.63 17 - 1 3 15 26 14 - 62 35.96 25 1 4 26 15 3 49 36.30 8 2 8 29 23 16 78 36.35 4 8 7 1 5 28 29 11 74 36.60 12 4 21 37 7 2 71 35.75 21 1 16 32 15 2 66 36.01 33 9 20 31 14 2 76 35.74 13 III 23 34 10 16 1 9 24 29 7 2 - 1 6 20 11 1 - 1 6 20 6 -- 4 14 9 8 -72 39 33 35 35.53 36.13 35.94 35.60 19 4 2 14 28 2 - 50 35.44 18 3 15 19 21 3 2 63 35.19 15 dark 4 18 27 2 - 51 35.53 26 - 3 24 24 16 1 68 35.83 9 - 7 18 22 9 2 58 35.67 27 - 7 13 9 1 1 31 35.23 -94-Table XV. Frequency distribution of caudal vertebral counts in experiment C. Bas- Light duration Sect, ket and intensity Caudal vertebrae No. No. hrs. f t . / c d l . 24 25 26 27 28 29 30 Number Mean 1 mm 9 36 41 9 - 93 27.49 11 - 1 20 33 15 2 - 71 26,96 I 20 16 20 - - 8 26 10 4 1 49 27.26 36 - 3 30 33 7 1 - 74 26.63 35 2 1 9 4 - - 16 25.94 32 1 5 27 13 46 26.13 VI 31 16 28 2 6 5 8 20 3 1 2 2 14 48 25 7 56 25 24 41 5 3 25 15 12 44 20 11 55 34 2 1 1 2 5 90 91 73 46 78 106 26.05 26.22 25.71 26.22 26.15 26.34 II 7 30 22 29 14 17 16 - 2 29 52 9 2 - 2 19 21 6 -- 1 15 31 9 2 2 23 24 5 1 -- 2 15 28 14 -- 3 23 31 5 -94 48 58 55 59 62 26.79 26.65 26.93 25.64 26.92 26.61 25 1 8 8 1 21 V 4 8 7 1 8 12 1 18 21 2 12 33 - 2 26 14 - - 49 26.08 45 10 1 - 78 25.86 44 21 - - 74 26.15 41 10 1 - 71 25.89 43 9 - - 66 25.90 11 44 16 3 76 27.09 III 13 23 34 10 16 - 3 34 26 9 2 3 23 10 1 - 7 17 8 1 - 12 16 6 72 39 33 35 26.57 26.12 26.09 26.77 19 - 8 24 12 4 2 50 26.36 18 1 9 32 20 1 - 63 26.17 IV 15 dark - 7 27 16 1 - 51 26.21 26 - 1 23 37 7 - 68 26.73 9 1 11 32 14 - - 58 26.01 27 2 5 14 8 2 - 31 26.09 95-58 62 66 58 62 6 6 Figure 32. Percentage frequency polygons of total vertebral counts in controls and transfers effected at v a r i -ous times after f e r t i l i z a t i o n i n experiment C. Day degrees and days (in parenthesis) before trans-fer shown on l e f t margin. Arrows indicate the direction of transfer. The mean count for each group i s on rishe margin. -96-Figure 33. Percentage frequency polygons of- abdominal vertebrae of sockeye in experiment C. Details as i n figure 32 Figure 34. Percentage frequency polygons of caudal vertebrae of sockeye i n experiment C. Details as in figure 32 -98-58 62 (a) X 62.91 (CONTROL) 61.91 6 2 . 4 4 6 2 . 4 3 (CONTROL) 6 2 . 5 6 61.33 61.80 (CONTROL) 6 6 2 0 6 D ° X 6 2 . 2 6 63.28 6 2 . 0 9 58 62 (b) 6 6 Figure 35a and b. Percentage frequency polygons of total vertebrae of sockeye i n experiment C. These transfers showed aberrant counts. Details as i n figure 32. -99-Vertebrae of Sockeye. Experiments A and C The sockeye eggs reared in experiments A and C were obtained from the same group of males and females and f e r t i l i z e d at the same time. In experiment A temperature was sustained at 12°C - 0.1 whereas in experiment C i t was subjected to the variable temperatures of the water supply in the laboratory. The results of transfers in experiment C indicated that mean total vertebral counts were varied, in several lot s , only during the early stage (142 D°). Aber-rant counts were obtained in later transfers (206 D° and 300 D Q) and hence the period for fixation of the number of verte-brae was not conclusive. However, i t was evident that before the 142 D° period the vertebral number could be altered and there was a prolonged period of sensitivity. During the period (142 D° - 300 D°) the temperature dropped from 8.5°C to 7.4°C (Figure 6), and the mean was approximately 8°C. Alterations in total number of vertebrae, in the controls of experiment C, could be presumed to have been effected during the sensitive period and, therefore, the temperature for this experiment, for comparison with experiment A, was considered to be 8°C. Lots with similar light conditions at the two tem-peratures (8°C and 12°C) were compared (Figure 36). At both temperatures there was a steady increase in mean total ver-tebral counts with the durations as well as intensities of -100-light . At 12°C, as seen in the results of experiment A, none of the differences between means was significant. At 8°C the difference between means of comparable durations as well as intensities was significant. The least d i f f e r -ence between means of comparable lots reared at 8°C and 12°C was 1.7 vertebrae. Figure 36. Comparison of mean vertebral counts (total) of sockeye reared at 8°C i n experiment C, and at 12*0 in experiment A. -102-DISCUSSION Selective Mortality Analyses of meristic series as obtained in the experiments showed many complicated results. In several instances significant differences were obtained, between dark and long light durations and between the intensities of the long durations. Could these differences as obtained have been the result of differential mortality, where light eliminated unsuitable genotypes? Mortalities discussed earlier were evidently not caused by the light conditions used, except in two instances in rainbow trout (12-and 24-hour high intensity). That l i g h t , in conjunction with other factors, intensified mortality is unlikely since some dark lots showed higher mortality. Some experimental work with different temperatures (Taning, 1952; Lindsey 1952, and Molander and Swedmark 1957) showed that the influence of selective mortality on meristic va r i a b i l i t y could not be excluded, except in Taning's (1944) sea-trout experiment. Heuts (1947) demonstrated that in sticklebacks selective mortality occurred before hatching. In the present experiments pre-hatching mortality was high in a l l lots but the frequency distribution of vertebral counts showed a wide range. Thus, i f selection for any particular genotype was operative the distribution should be confined -103-to a narrow range. Heuts also showed that i f selective mortality occurred, a higher standard deviation would be expected,in the distribution of lots reared under conditions having maximum survival. The standard deviations for a l l lots with markedly higher survivals did not show any cor-relation with mean vertebral counts. The occurrence of high vertebral abnormalities in lots subjected to the various light conditions (except pink) and the corresponding percen-tage survival did not indicate a selection for vertebral malformations. If selective mortality eliminated individuals pro-ducing either low or high counts in any particular meristic series then i t could be expected to operate on a l l other series in like manner. However, this has not been the case with the present results. Selection against any one unsuit-able genotype could take place under the particular rearing conditions, but i t i s not probable that several genotypes could be selected at the same time. A l l the evidence presented i s highly suggestive tat selective mortality did not cause the meristic variation obtain, but the unusually high mortality makes i t rather d i f f i c u l t to state conclusively that i t is so* Scales Scale counts in controlled experiments were studied -104-in rainbow trout by Mottley (1934), with temperatures as the variable environmental factor. He found that higher tem-peratures produced fewer oblique scale rows, but these rows were counted at several longitudinal rows above the lateral l i n e . Neave (1943) demonstrated that in taxonomic work on salmonids counting of oblique scale rows had many disadvan-tages, especially because of the more frequent branching of the oblique scale l i n e . He also stated: "Determination of the number of oblique rows apparently takes place at a different time from, or under the influence of other factors, than determination of the number of lateral line scales". In view of the above evidence i t i s irregular to compare l a -teral line scale counts used in the present study with oblique row counts used by earlier authors. Hubbs (1941) found that an increase in lateral line scales in abnormal suckers was due to delayed lepidogenesis. The formation of scales on and outside the lateral line at the caudal t i p of the body is considerably delayed (Neave, 1943). However, variation in lateral line scale counts, of young sockeye salmon about or over 2\ inches in length, was not observed by Foerster and Pritchard (1935a). In the present study a l l sockeye, in experiment A, had fork lengths ranging from 2\ to 3| inches while the majority were over 2\ inches and there was no correlation of lateral line scale count with length. -105-Consequently i t i s concluded that f i s h had attained their definitive scale counts prior to preservation in a l l lots used in the present experiment. Apparently variation in scale count produced by light has not been experimentally tested before and hence i t is not possible to compare the present results with previous studies. Nevertheless, i t is profitable to examine the way in which scales vary with temperature and lig h t . Results of lateral line scale counts in sockeye indicated, though not in a l l lot s , a lower number with long duration and high intensity. Since the action of light on metabolism and metameric segmentation is now known, an exp-lanation for the above correlation cannot be given. However, i f Hubbs* (1941) explanation for increased scale count in abnormal suckers is valid then the converse may also be ap-plicable. It has been demonstrated in this study that light speeds up hatching and yolk absorption and, therefore, i t i s reasonable to presume that growth and lepidogenesis are also speeded up. If scale foci are l a i d down at approximately constant absolute distances from one another (Hubbs 1927), and i f the ratio of scale foci distance to total length of lateral line is greater than that of a normally growing f i s h , i t i s then conceivable that a decreased scale count w i l l ensue. Scale counts on the oblique row from the origins -106-of dorsal and anal fins to the lateral line showed a similar pattern, especially the high light intensity l o t s . The ob-lique row counts on dorsal and ventral sections of the lateral line differed somewhat in details. This may be due to the relative proportion of body parts on either side of the lateral l i n e . Fin Ravs Secondary or unsegmented caudal rays of fishes in the antero-dorsal and antero-ventral flaps of the caudal fan were not considered in earlier experimental work on mer-i s t i c studies. These caudal rays of sockeye (experiment A), in the upper and lower series, showed an almost identical pattern with both durations and intensities of l i g h t . o Taning (1952) in sea trout, and Seymour (1956) in Chinook salmon, obtained maximum values for both dorsal and anal rays at intermediate temperatures. Lindsey (1958) found in kokanee, that 8-hour light duration produced more anal rays than 16-hour duration. The results obtained in the present study showed that dorsal and anal f i n ray counts of kokanee, sockeye, pink, and rainbow trout (with exceptions in kokanee) were, lower in the longer duration of l i g h t . The kokanee 8-hour duration high intensity lot (comparable to the kokanee in Lindsey (1958)) was lost accidentally, but the -107-eight survivors showed a higher mean (insignificant) than the 16-hour l o t . However, in sockeye and pink lower light conditions produced higher counts. Although mean counts in many of the dorsal and anal f i n ray series did not show significant differences, the pattern within a species (except kokanee) was almost alike and generally followed corresponding mean total vertebral counts (which w i l l be discussed l a t e r ) . It is improbable that dorsal and anal rays are determined according to the earlier established vertebral number. There is evidence that the number of axial skeletal elements and that of the median skeletal series are determined indepen-dently (Lindsey, 1954). Kokanee (landlocked form) and sockeye (sea-run form), used in experiment A, are classif i e d as one species (p_. nerka). Ricker (1940) suggested that kokanee have evolved from *residual* (Ricker, 1938) populations of sea-going sock-eye. As yet, no significant morphological differences have been shown to exist among these forms of sockeye. The range for dorsal and anal ray counts are re-ported as follows: 9 to 13, and 13 to 16 (Foerster and P r i t -chard, 1935b); 11 to 16, and 13 to 17 (Carl gt §i, 1959). Systematists have omitted the small anterior rays (usually three) which are less than one-half the length of the longest rays. In the present study, when evaluating the affect of -108-light on these series, there was no valid reason for omit-ting the clearly v i s i b l e small anterior rays. Therefore, the range for dorsal and anal ray counts in the experimental lots was 13 to 16, and 15 to 18 respectively for both forms of salmon. Nevertheless, kokanee and sockeye differed s i g n i -ficantly, i n dorsal ray counts in the longer light duration lots , and in anal ray counts in the dark lo t s . F e r t i l i z e d eggs of the above forms of salmon were obtained from areas widely separated geographically. Thus, the differences ob-served in this experiment may not occur i f the two forms of O. nerka are taken from the same area. Vertebrae Dannevig (1932) was the f i r s t to suspect that light may have been responsible for lower vertebral counts in cod. McHugh (1954) found in the grunion that the mean number of vertebrae decreased with increasing l i g h t . Total vertebral counts showed diametrically opposite results for kokanee and pink and the pattern for sockeye was quite unlike the others. Although the results obtained for sockeye were not s t a t i s t i -cally significant, there was a trend towards increase in vertebrae with increasing l i g h t . In 8-hour l i g h t , kokanee had a lower count, while pink and trout had higher counts -109-than the corresponding lots reared in the dark. In 16-hour light, kokanee and pink had lower counts than the dark, but some of the differences were not significant. These results are confusing. In abdominal and caudal counts pink seemed to follow a plan somewhat similar to that of total counts. The 8-hour lot in kokanee had a higher caudal count (sample size was small) than the 16-hour lot and is in agreement with Lindsey's (1958) results. The irregularity observed in ab-dominal and caudal counts may be due to sex differences. In sticklebacks, Lindsey (1952) showed that the position of the f i r s t haemal arch was displaced posteriorly in females re l a -tive to the whole vertebral column. It was not possible to distinguish the sexes in these young salmonids and the ratio of females to males was unknown. Since adult females in salmonids have a substantial egg capacity in the abdomen, i t is most li k e l y that increased number of vertebrae in this region is of adaptive significance. In kokanee, the sexes did not differ significantly in total vertebral counts (Ver-non 1957). Thus the posterior shifting of the haemal spine relative to the whole vertebral column is an important con-sideration. If the ratio of females to males was other than normal in these experiments then i t was not surprising to find widely discrepant abdominal and caudal counts. Total counts are more reliable (because of the position of the f i r s t haemal spine is indeterminate) and unless sexes could be -110-separated, abdominal and caudal counts could not be consid-ered without running the risk of obscuring the results. Another important factor that should be considered is the occurrence of vertebral abnormalities, described and discussed in section C. Vertebral abnormalities have been shown to be correlated with temperatures (Molander and Swed-mark, 1957). Unless fused vertebrae are recognized and counted as two, the originally differentiated vertebral seg-ments would be reduced. Many of the variations in vertebral numbers, produced by light, have been mainly through fusions in the pre-urostylic region. Taning (1952) obtained the characteristic V-shaped distribution of vertebral numbers in sea-trout and the maxi-mum count occurred at an intermediate temperature of 6°C. In Chinook salmon (0_. tshawvtscha) Seymour (1956) found a similar relationship with the lowest count at 10°C. Mean total vertebral counts of sockeye (experiments , A and C) reared at two temperatures produced some interesting results. At both temperatures (8;°C and 12°C) there was a gradual increase in means with increasing li g h t . At 12°C the differences produced by light are small but a l l lots com-pared with corresponding lots at 8°C had large differences, and significant results were obtained among lots in the lat t e r . -111-In sockeye i t was quite probable that 8°C could be consid-ered as an intermediate temperature as eggs w i l l develop well below this temperature. The lines joining 8°C and 12°C lots may represent the right limb of a V-shaped curve. Since the various degrees of light produced greater differences at 8°C i t i s evident that at higher temperatures the effects of light are negated. Theoretically, i t seems that light might not have any influence on vertebral numbers at s t i l l higher temperatures where the genetic limit for vertebrae would be reached. Whether light w i l l react in a like manner at lower and lower temperatures i s d i f f i c u l t to predict, without some knowledge of the action of light on developmental processes at different temperatures. However, in the present case i t may be suggested that the action of light is only secondary. When temperature f a i l s to give f u l l expression within the potential genetic range of an or-ganism, other environmental factors could act as auxiliary modifying agents. Several insignificant results and irregular patterns obtained in the meristic traits of sockeye and even in other species in the present study may be due to the weak action of light relative to that of the high sustained tem-perature . Endocrine Function and Meristic Variability Willier (1955) states "one of the ways whereby cel l s -112-tissues, and organs of the vertebrate organism are correlated functionally is through hormones 'sensu s t r i c t u * , .... The functional relationship between the anterior pituitary and the thyroid is one among others to be set up between endocrine glands of the developing embryo". It i s important to know the earliest stage when these two organs become functional in embryonic development. Evidence from analyses of anuran embryos (Allen 1927, Gorbman & Evans 1941), and chick embryos (Martindale 1941), indicates strongly that anterior pituitary exerts a trophic effect on the thyroid gland in early embryonic l i f e . Consequently, i t is reason-able to assume that in teleosts a functional pituitary-thyroid relationship is established in a very early stage of embryonic development. Is the period of functional pituitary-thyroid re-lationship prior to or after fixation of vertebrae and other meristic series in salmonids? Taning (1944) found that in sea-trout vertebrae were fixed before hatching, and in the present study, to be described in the next section, transfer experiments with sockeye (Experiment C) showed in a majority of cases that vertebrae were fixed long before hatching but after the appearance of optic vesicles, otic capsules and pectoral buds (6-15 mm). Some of the histological sections of the head at the 6-15 mm stages showed pituitary primorida and a few minute thyroid f o l l i c l e s devoid of c o l l o i d . In the smaller embryos (6 to 7 mm) there was no trace of thyroid -113-tissue. This finding and evidence from other animal em-bryos suggests that pituitary and thyroid are probably functional long before fixation of many meristic series other than vertebrae. Results of pituitary and thyroid analyses in this study were obtained long after the trout hatched. Hence the hypothesis that the same relationship in endocrine a c t i -vity existed between the various lots at vertebrae formation and later at the time the samples were preserved cannot be proven, since endocrine function varies with age and stage of development (Willier 1955). However, the difference between dark lots and long duration high intensity lots sug-gested that light had some influence on pituitary and conse-quently thyroid a c t i v i t y . Direct correlation of caudal ver-tebrae, or inverse correlation of abdominal and total vertebrae with area of pituitary and activity of thyroid in dark lots and high light intensity lots, might be coincidental but is certainly suggestive. Fixation-Period of Vertebrae (Sockeye, experiment C) o Taning (1952) concluded, after a series of transfer experiments (using different temperatures), that determination of vertebrae in the sea trout begins very early in develop-ment (gastrulation period), and that a supersensitive period -114-prevails from 145 D° to 165 D°. During this supersensitive period widely different changes in vertebral numbers could o be produced by 'shock* treatment (Taning, 1950). Transfer experiments conducted to study the fi x a -tion-period of vertebral numbers revealed that light could influence the fi n a l numbers. Although i t was not definitely r established that the plastic period was within the time of the three transfers, yet i t was evident, at least in total counts, that in many instances sockeye were inert to light before 142 D°. Some histological examination at or about this stage showed that the pituitary was indistinguishable and there was no indication of thyroid f o l l i c l e s or c o l l o i d . Even in later embryos, as described earlier, only a few minute thyroid f o l l i c l e s were v i s i b l e . It is thus concluded that at or before 142 D° the pituitary-thyroid mechanism was not stimulated by li g h t . The transfers made between 8-hour and 16-hour high intensities suggested that light could be effective only within the f i r s t and second period (142 D° to 165 D°). Aber-rant counts in total, abdominal and caudal vertebrae were obtained at any one of the three transfer times. This may be attributed to 'shock' during transfer. The sensitive period evidently extended from 142 D° to 300 D° in sockeye, and is beyond the range established for sea trout vertebrae (Taning, 1952). -115-Results of transfer experiments, with light as the variable factor, correspond to Taning 1s (1944) sea-trout work where temperature was variable. It was fortunate that sockeye transfers were conducted at a lower temperature such that any change produced by light was comparatively larger than that produced at 12°C. A definitive period of vertebral fixation in sockeye could be obtained by conducting o experiments with temperatures (Taning 1944 and 1952) or with light as in the present study (but with sustained tempera-ture). It is imperative that more frequent transfers should be effected befoas and after the period used here. Mechanisms of Variation Several theories have been advanced in attempts to explain the action of environmental factors on meristic t r a i t s . Within a given species meristic characters are genetically controlled but i t has been demonstrated that environmental factors determine the f i n a l numbers within the genetic range. Hubbs (1926) suggested that variation in absolute length of presumptive area at time of series formation was one method of explaining meristic variation. At low tem-perature, delay in segmentation produced a greater absolute length of f i s h and i f meristic elements are equally spaced -116-a higher number w i l l result. Spacing within series and the dependence or independence of a series on other series have also been advanced as possible explanations for mer-i s t i c differences obtained with different temperatures (Lindsey 1952). Mechanisms of meristic segmentation under the i n -fluence of various environmental factors cannot be elucidated without understanding the fundamental biophysical principles governing segmentation. Turing (1952) showed that periodic structures would be produced from a homogeneous i n i t i a l state i f certain chemical processes interacted in certain ways and i f chance perturbations occurred around the i n i t i a l steady state. Waddington and Deuchar (1953) commenting on this hypothesis state: "It appears rather unsatisfactory to appeal to such an inherently chancy mechanism as this to explain a regular and basic phenomenon of development such as meristic segmentation". These workers have experimentally demonstrated on embryos of Triturus alpestris that: "spatial dimensions other than the antero-posterior one are involved in the causation of the transverse segmentation". This work supports the view that a larger embryo (from larger egg) with a longer and thicker body w i l l have a larger number of somites. Marckmann (1954) found that sea-trout larvae reared at intermediate temperatures had lower numbers of ver-tebrae and maximum body weights, while larvae reared at lower -117-or higher temperatures produced higher numbers of vertebrae and lower body weights. His interpretation was that at intermediate temperature the larvae with maximum body weight had a low and economic metabolism per day degree such that yolk store was available for growth. Both higher and lower temperatures gave an uneconomic, i.e. higher metabolism, reflected in the lower body weight and higher number of ver-tebrae. His theory gained support from the fact that the rate of development during the incubation period reached a maximum at the intermediate temperature. Several environmental factors other than tempera-ture are known to produce meristic variations, e.g. s a l i n i t y , oxygen. To these can be added li g h t . Some previous ex-periments had suggested that both duration and intensity could control the number of skeletal parts in f i s h , but the present study is the f i r s t to examine the effect of both variables on several series in several species. In common with temperature studies, the results, rather than revealing a simple mechanism, have unfolded a further f i e l d of com-plexities. Different meristic series may respond d i f f e r -ently in one species, and the same series may respond d i f f e r -ently in different species. There seems often (but not always) to be a tendency for a high light level to reduce the number of parts, accompanied by evidence of increased activity of the pituitary and thyroid. A promising further line of research might therefore concern the endocrine system -118-since the experimental evidence for endocrine involvement in the present study i s circumstantial. The mechanism by which light durations and inten-s i t i e s produce changes in f i n a l meristic numbers may be through an acceleration of metabolism. Metabolic rate as well as the hormonal balance of endocrine secretions are controlled by temperature. In the study of the trout p i t u i -tary and thyroid there was some evidence that high light i n -creased the activity of the above glands, but i t was unknown whether this activity was the same during metamerism. How-ever, the meristic series analysed suggested that faster development produced fewer vertebrae and also fewer dorsal and perhaps anal rays. The correlation between the f i n a l dorsal rays in trout as well as dorsal and anal rays of pink salmon and vertebral numbers is evident in the results obtained. For a better understanding of endocrine control of meristic segmentation, experiments may be designed to trace the action of light (at known temperatures) on early develop-mental stages. Also knowledge is essential as to the part that endocrine secretions may play in varying the f i n a l ex-pression of meristic characters, and whether light i s the promoter. Refined techniques for analyses are available. Pituitary function may be eliminated by excision of various..; parts, and even chemical thyroidectomy may be performed in -119-an attempt to correlate endocrine activity on metamerism. Results and discussion of the present experimental study reveal that meristic v a r i a b i l i t y is a complicated process. The mechanism of the phenomenon cannot be given in any simple explanation that w i l l embrace a l l the v a r i -a b i l i t i e s of these plastic structures. -120-SUMMARY 1. a. Two experiments, A and B, were conducted, at a sus-tained water temperature of 12°C "t 0.01°, to study the influence of light durations and intensities during early development on meristic v a r i a b i l i t y in salmon (A-kokanee, sockeye and pink) and rainbow trout (B). b. Another experiment, C, was designed to test the period of vertebral fixation in sockeye salmon by transferring embryos at three developmental stages to different light conditions. 2. The rate of hatching was fastest in lots subjected to long-er light durations and high intensities. 3. In the salmon species the lots reared under light had a faster rate of yolk-sac absorption than those lots reared in the dark. In sockeye and pink lots an increasing rate of yolk-sac absorption was correlated with increasing amount of li g h t . In the trout the reverse was true, a decreasing rate of yolk-sac absorption was observed with increasing amount of l i g h t . 4. High mortalities occurred in allthe species and in a l l the lots . However, the sockeye lots in experiment C, showed relatively lower mortality. 5 . Abnormalities in the vertebral column in a l l the species -121-studied were prevalent in the pre-urostylic region. Only in sockeye was the frequency of occurrence of abnormali-ties correlated with l i g h t . In trout the exceptionally high percentage of malformations in a l l the lots seemed to suggest that the affects of other factors might have been predominant. 6. The lateral line scale counts of sockeye (experiment A) in both light durations and intensities were significantly different. The lot in the longer light duration of high intensity had the lowest count. 7. The mean scale counts on sockeye along the oblique row from the origin of the dorsal f i n to the lateral line and that from the origin of the anal f i n to the lateral line were lowest in the lots subjected to the longer light dura-tions of high intensity. Many of the differences between means were significant except that between the two inten-sity lots of the 16-hOur light duration. 8. The secondary caudal ray counts of sockeye in the lower series were greater than that in the upper series. The mean counts for the lots i n the two light intensities f o l -lowed a similar pattern in the two series, and within each series lower counts were obtained with increasing l i g h t . 9. Many of the differences among the dorsal and anal f i n ray counts within species were not significant. The dorsal -122-ray counts in kokanee and sockeye lots in the 16-hour light duration were significantly different. Again, in the anal ray counts kokanee and sockeye showed in many comparable lots significant differences. 10. The total vertebral counts among the species studied in these experiments showed different patterns. The common feature i n a l l species, except sockeye, was the decreased count obtained in the lot reared under longer light dura-tion of high intensity. Although sockeye showed increas-ing counts with increasing light the differences between lots were not significant. The sockeye reared at 8°C also showed increasing counts with increasing l i g h t , but in this case the differences between lots were significant. The minimum difference between the 8°C lots and the 12°C lots of sockeye was 1.7 vertebrae. It was suggested that the action of light was weak at the higher temperature. The abdominal and caudal vertebral counts in a l l the spec-ies were observed to be inconsistent, probably due to the indeterminate position of the f i r s t caudal vertebra. 11. It was possible to show that the vertebral numbers in sock-eye were not fixed before 142 D® and that the period of sensitivity was usually between 142 D° and 206 D°. Aber-rant counts were obtained in some of the transferred lots of sockeye at a l l the transfer times (142 D°, 206 D° and 300 D?). It was concluded that these aberrant counts -123-were caused by 'shock' during transfer, and therefore the sensitive period, as observed i n this experiment, extended from 142 D° to 300 D% 12. The area of the epithelial components of the pituitary gland and the activity of the thyroid gland of trout showed a positive correlation with increasing amount of l i g h t . Since the histological work was not coincidental with the time of vertebral fixation (or fixation of other meristic series) i t was only conjectural that endocrine activity was correlated with meristic variation. -124-LITERATURE CITED Allen, B.M. 1927. Influence of the hypophysis upon the thy roid gland in amphibian larvae. Univ. C a l i f . Pub. Zool., 31:53-78. Apstein, C. 1909. Die Bestimmung des Alters pelagisch leben- . den Fischeier. . Milleilunger des Deutschen Seefischerei-Vereins. 25 (2): 364-373. (Translated by M. C r u l l ) . Barrett, I., and D.R. Hurn. 1954. A manual of game fi s h culture for use in B.C. trout hatcheries. B r i t i s h Col-umbia Game Dept., 40 pp., mimeo. Beach, A.W. 1956. Seasonal changes in the cytology of the pituitary gland and the ovary of goldfish. M.A. Thesis Univ. British Columbia. Breder, CM., and P. Rasquin. 1950. A preliminary report on the role of the pineal organ in the control of pig-ment ce l l s and light reactions in recent teleost fishes. Science 111:10-12. Bretschneider, L.H., and J.J. Duyvene de Wit. 1947. 'Sexual Endocrinology of Non-Mammalian Vertebrates'. Elsevier, New York. Brown, M.E. 1946. The growth of brown trout (Salmo trutta Linn.). I. Factors influencing the growth of trout fry. J. Exp. B i o l . 22: 118-129. Carl, G.C., W.A. Clemens, and C.C. Lindsey. 1959. The fresh-water fishes of British Columbia. Prov. Mus., Handbook No. 5, 192 pp. Clothier, C.R. 1950. A key to some southern California fishes based on vertebral characters. State of California, Dept. of Natural Resources, Div. of Fish and Game, Bureau of Marine Fisheries, Fish. Bull. No. 79: 1-83. Comfort, A. 1956 'The Biology of Senescence'. Routledge and Kegan Paul, London. -125-Dales, S., and W«S. Hoar. 1954. Effects of thyroxine and thiourea on the early development of chum salmon (Qn- corhynchus keta). Can. J. Zool. 32: 244-251. Dannevig, A. 1932. Is the number of vertebrae in cod i n f l u -enced by light or high temperature during the early stages. J. du Conseil. 7: 60-62. De Robertis, E. 1949. Cytological and cytochemical bases of thyroid function. Ann. N.Y. Acad. Sci. 50: 317-335. Foerster, R.E., and A.L. Pritchard. 1935a. The identified tion of the young of five species of Pacific salmon. Rept. B.C. Fish.Dept., 1934: 106-116. Foerster, R.E., and A.L. Pritchard. 1935b. A study of the variation in certain meristic characters in the genus Oncorhvnchus in British Columbia. Trans. Roy. Soc. Cana-da. Section V: 85-95. Ford, E. 1937. Vertebral variation in teleostean fishes. Journ. Mar. B i o l . Assoc. 22: 1-60. Ford, E. 1947. Vertebral variation in teleostean fishes. III. Isospondyli. Journ. Mar. B i o l . Assoc. 26: 390-397. Gorbman, A., and H.M. Evans. 1941. Correlation of histologi-cal differentiation with beginning function of develop-ing thyroid gland of frog. Proc. Soc. Exp. B i o l , and Med., 47: 103-106. Gordon, M. 1957. Physiological genetics of fishes. Chapter X; Vol. II in *The physiology of fishes'; edited by M.E. Brown. Academic Press New York. Haempel, 0., and H. Lechler. 1931. Uber die wirkung von Ultravioletter bestrahlung auf Fisheier und Fischbrut. Z. vergleich. Physiol. 14: 265-272. Hayes, F.R. 1930. The metabolism of developing salmon eggs. I. The significance of hatching and the role of water in development. Biochem. Jour., 24 (3): 723-734. -126-Heincke, F. 1898. Naturgeschichte des Herlngs. T e i l I. Die Lokalformen und die Wanderungen des Herings in den europaischen Meeren. Abhandlungen des Deutschen See-fischerei - Vereins. 2 (1): 1-238. Heuts, M.J. 1947. Experimental studies on adaptive evolu-tion in Gasterosteus aculeatus L. Evolution I: 89-102. Hoar, W.S. 1957. The endocrine organs. Chapter VI; Vol. I in 'The physiology of fishes'; edited by M.E. Brown. Academic Press, New York. Hobbs, D.F. 1937. Natural reproduction of quinnat salmon, brown and rainbow trout in certain New Zealand waters. N.Z. Marine Dept., Fish. B u l l . No. 6, 104 pp. Hoilister, G. 1934. Clearing and dyeing of f i s h for bone study. Zoologica, Vol. 12 (10): 89-101. Hubbs, C.L. 1926. The structural consequences of modifica-tions of the development rate of fishes. Amer. Natura-l i s t , 60: 57-81. Hubbs, C.L. 1927. The related effects of a parasite on a f i s h . J . Parasitol., 14: 75-84. Hubbs, C.L. 1941. Increased number and delayed development of scales in abnormal suckers. Pap. Mich. Acad. Sci. 26: 229-237. Ishida, J . 1944. Hatching enzyme in the fresh water f i s h Orvzias latines. Annot. Zool. Japon 22 (137): 155-164. Kandler, R. 1932, Unsicherheiten bei Bestimmung der Wirbel-zahl infolge Verwachsungserscheinungen. Journ. du Cons. 7: 373-385. Ka'ndler, R. 1935. Rassenkundliche Untersuchungen an Platt-fischen. I. Variablitetsstudien an den Flossenstrahlen und Wirbelzahlen der Ostseeschollen. Ber. Deutsche Wiss. Koram. f. Meeresforschung. N.F. 7 (4): 381-493. -127-Kerr, T. 1940. On the histogenesis of some teleost p i t u i -taries. Proc. Roy. Soc, Edinburgh, 60: 224-240. Kerr, T. 1942. A comparative study of some teleost p i t u i -taries. Proc. Zool. Soc. London, (A) 112: 37-56. Lafon, M. 1947. Recherches sur 1'utilisation des reserves v i t e l l i n e s au cours de 1*embryogenese. (I) Donnees sur 1 * embryon de Truite (Salmo fario et Salmo irideus). Arch, intern. Physiol. 55: 123-152. Lindsey, C.C. 1952. Environmental determination of the number of teleost f i n rays. Ph.D. Thesis. Cambridge University. Lindsey, C.C. 1954. Temperature-controlled meristic variation in the paradise f i s h Macropodus opercularis (L.). Can. J. Zool. 32: 87-98. Lindsey, C.C. 1958. Modifications of meristic characters by light duration in kokanee, Oncorhynchus nerka. Copeia 1958 (2): 134-136. Lyubitskaya, A.I. 1956. Influence of different parts of the visibl e spectrum regions on the developmental stages of embryos and larvae of the fishes. Zool. Jour. 35 (12): 1873-1886. (In Russian - English abstract). Marckmann, K. 1954. Is there any correlation between meta-bolism and number of vertebrae in the sea trout (Salmo  trutta L)? Medd. f. Komm. for Danmarks Fi s k e r i - og Havunders. N.S. 1(3): 1-9. Martindale, F.M. 1941. Initiation and early development of thyrotropic function in the incubating chick. Anat. Rec, 79 : 373-393. Maximow, A.A., and W. Bloom. 1957. A Textbook of Histology. 7th. Ed. Saunders Co., Philadelphia. 628 pp. Lie Hugh, J.L. 1954. The influence of light on the number of vertebrae in the grunion, Leuresthes tenuis. Copeia 1954 (1): 23-25. -128-Milkman, R. 1954. Controlled observation of hatching in Fundulus heteroclitus. B i o l . Bull. 107: 300. Miller, R.R. 1957. Utilization of X-rays as a tool in sys-tematic zoology. Systematic Zoology, 6 (1): 29-40. Molander, A.R., and M.M. Swedmark. 1957. Experimental i n -vestigations on variation in plaice (Pleuronectes plat- essa Linne). Inst. Marine Research, Lysekil. Ser. B i o l . Rep. No. 7: 1-45. Morris, R.W. 1956. Some aspects of the problem of rearing marine fishes. B u l l . Inst. Oceanogr. Monaco. No. 1082, 61 pp. Mottley, C. McC. 1934. The effect of temperature during development on the number of scales in the Eamloops trout, Salmo kamloops Jordan. Contr. Canad. B i o l . , 8: 254-263. Neave, F. 1943. Scale pattern and scale counting methods in relation to certain trout and other salmonids. Trans. Roy. Soc. Canada, Sect. V, 37: 79-91. Pantin, C.F.A. 1948. Notes on Microscopical Techniques for Zoologists. Cambridge Univ. Press. 79 pp. Pflugfelder, 0. 1953. Wirkungen der Epiphysektomie auf die Postembryonalentwicklung von Lebistes reticulatus Peters. Wilhelm Roux* Arch. Entwicklungsmech. Organ 146: 115-136. Pflugfelder, 0. 1954. Wirkungen partieller Zerstorungen der Parietalregion von Lebistes reticulatus. Wilhelm Roux* Arch. Entwicklungsmech. Organ 147: 42-60. Ricker, W.E. 1938. 'Residual' and kokanee salmon in Cultus Lake Br i t i s h Columbia. J. B i o l . Bd. Canada, 4: 192-218. Ricker, W.E. 1940. On the origin of kokanee, a freshwater type of sockeye salmon. Trans. Roy. Soc. Canada, V, 34: 121-135. -129-Schmidt, Johs. 1921. Racial investigations. VIII. The numerical signification of fused vertebrae. C. R. Lab. Carlsberg. 14 (16): 1-5. Schnakenbeck, W. 1931. *Zum Rassenproblem bei den Fischen'. J. Cons. Int. Explor. Mer. 6: 28-40. Scruggs, W.M. 1939. The epithelial components of the teleost pituitary gland as identified by a standardized method of selective staining. J. Morph. 65: 187-213. Seymour, A.H. 1956. Effects of temperature upon Chinook salmon. Ph.D. Thesis. Univ. of Washington. Smith, S. 1957. Early development and hatching. Chapter VII; Vol. I in 'The physiology of fishes'} edited by M.E. Brown. Academic Press. New York. Snedecor, G.W. 1957. St a t i s t i c a l Methods. 7th. Ed. Iowa State College Press, Ames, Iowa. 534 pp. Stringer, G.E., and W.S. Hoar. 1955. Aggressive behavior of underyearling Kamioops trout. Can. J. Zool. 33: 148-160. Taning, A.V. 1944. Experiments on meristic and other charac-ters in fishes I. Medd. Kom. Danmarks Fisheri- og Hav-unders. Fisheri, XI (3): 1-66. Taning, A.V. 1950. Influence of the environment on number of vertebrae in teleostean fishes. Nature, Lond., 165:28. o ° Taning, A.V. 1952. Experimental study of meristic characters in fishes. B i o l . Rev. 27: 169-193. Turing, A.M. 1952. The chemical basis of morphogenesis. Philos. Trans. B. 237: 37-72. Trifonova, A.N. 1937. La physiologie de la differenciation et de la croissance. I. L'equilibre Pasteur - Meyerhof dans le developpement des poissons. Acta Zool. 18:375-445. -130-Vernon, E.H. 1957, Morphoraetric comparison of three races of kokanee (Oncorhynchus nerka) within a large B r i t i s h Columbia Lake. J. Fish. Res. Bd. Canada, 14: 573-598. Vibert, R. 1949. Du Repeuplement en truites et saumons par enfouissement de "boites d'alevinage" garnies d'oeufs dans les graviers. Bu l l . Fr. Pise., No. 153: 125-150. Waddington, C.H., and E.M. Deuchar. 1953. Studies on the mechanism of meristic segmentation. I. The dimensions of somites. J. Embryol. exp. Morph. 1: 349-356. Willier, B.H. 1955. Ontogeny of endocrine correlation. Section X in "Analysis of development"; edited by B.H. Willier; P.A. Weiss; and V. Hamburger. Saunders Co., Philadelphia. Wintrebert, P. 1912. Le mecanisme de l'eclosion chez la truite arc - en - c i e l . C.R. Soc. B i o l . , 72: 724-727. Woodman, A.S. 1939. The pituitary gland of the Atlantic salmon. J. Morph. 65: 411-435. 16 3 0 10 2 0 31 10 2 0 3 0 10 2 0 28 10 2 0 31 10 2 0 3 0 10 2 0 31 10 2 0 3 0 NOV. 57 DEC. 5 7 J A N . 5 8 F E B . 5 8 M A R C H 5 8 APRIL 5 8 MAY 5 8 J U N E 5 8 Figure 5. Mean daily temperature record for experiment C. -24-III IV VI SECT. O IV V VI INTENSITY 2 0 F T / C D L 7 F T / C D L I F T / C D L DARK 7 F T / C D L 2 0 F T / C D L Figure 9a. Arrangement of baskets at beginning of experiment C, in IV VI C O N T R O L 1ST T R A N S F E R A 16 DAYS 142 D ° 2 N D T R A N S F E R 2 4 DAYS 2 0 6 D° 3 R D T R A N S F E R 3 6 DAYS 3 0 0 D ° Figure 9b. Arrangement of baskets after the three transfers in experiment C. -48-10 -(A U J H _ J < DC O •z. m < f I 1 ^ KOKANEE ^ SOCKEYE PINK i i i Mi DARK HIGH LOW HIGH LOW 8 HOURS 16 HOURS Figure 12. Experiment A. Histograms of percentage abnormal vertebrae in kokanee, sockeye and pink salmon. 10 L d < 03 < 0 DARK J 20ft/cdl. I 7ft/cdl. 1 1 S i 8 HOURS 20ft/cdW*7—»• Ift/cdl. 16 HOURS Figure 14. Experiment C. Histograms of percentage abnormal vertebrae in sockeye control lots. -49-40 i-35 30 25 to U J 5 20 a: o z CD < 15 -10 DARK H—r— L 4 HOURS 12 HOURS H—I— L 2 4 HOURS Figure 13. Experiment B. Histograms of abnormal vertebrae in rainbow trout. -53-Figure 15. Influence of light on lateral line scales of sockeye in experiment A. -54-19.75 18.50 1 = 1  0 8 16 DURATION (HOURS) Figure 16. Influence of light on scales from origin of dorsal f i n to lateral line of sockeye in experiment A. -55-18.0 17.0 0 8 16 DURATION (HOURS) Figure 17. Influence of light on scales from origin of anal f i n to lateral line of sockeye in experiment A. -57-Figure 18. Influence of light on secondary caudal rays of sockeye in experiment A. -65-Figure 19. Influence of light on total vertebrae of kokanee, sockeye and pink in experiment A. -66-Figure 20. Influence of light on abdominal vertebrae of kokanee and pink in experiment A. -67-28.0 27.8 .6 < Q <t o o o < or CO w .4 or £ 27.0 < U J .8 26.2 ? ONLY 8 S P E C I M E N S _L 0 8 16 DURATION (HOURS) Figure 21. Influence of light on caudal vertebrae of kokanee and pink in experiment A. Figure 22. Influence of light on dorsal f i n rays of kokanee, sockeye and pink in experiment A. -69-Figure 23. Influence of light on anal f i n rays of kokanee, sockeye and pink in experiment A. 0 4 12 24 DURATION (HOURS) Figure 24. Influence of light on total vertebrae of rainbow trout in experiment B. -73-Figure 25 Influence of light on abdominal vertebrae of rainbow trout in experiment B. 0 4 12 24 DURATION (HOURS) Figure 26. Influence of light on caudal vertebrae of rainbow trout in experiment B. Figure 30. Influence of light on area of epithelial components of pituitary gland of rainbow trout in experiment B. -88-Figure 31. Influence of light on total vertebrae of sockeye control lots in experiment C. -95-Figure 32. Percentage frequency polygons of total vertebral counts in controls and transfers effected at v a r i -ous times after f e r t i l i z a t i o n in experiment C. Day degrees and days (in parenthesis) before trans-fer shown on l e f t margin. Arrows indicate the direction of transfer. The mean count for each group is on right margin. Figure 33. Percentage frequency polygons of abdominal vertebrae of sockeye in experiment C. Details as in figure 32 -97-Figure 34. Percentage frequency polygons of caudal vertebrae of sockeye in experiment C. Details as in figure 32. -98-X / \ 62.91 / 1 \ (CONTROL) 58 62 66 (a) Figure 35a and b. Percentage frequency polygons of total vertebrae of sockeye in experiment C. These transfers showed aberrant counts. Details as in figure 32. -101-61.5 _i 8°C TEMPERATURE I2°C Figure 36; Comparison of mean vertebral counts (total) of sockeye reared.at 8°C in experiment C, and at 12°C in experiment A. 

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.831.1-0106469/manifest

Comment

Related Items