Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An analysis of morphological variation within and between stream populations of Gasterosteus aculeatus… Shaw, Kate 1985

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

Item Metadata

Download

Media
831-UBC_1985_A6_7 S51_8.pdf [ 6.82MB ]
Metadata
JSON: 831-1.0096279.json
JSON-LD: 831-1.0096279-ld.json
RDF/XML (Pretty): 831-1.0096279-rdf.xml
RDF/JSON: 831-1.0096279-rdf.json
Turtle: 831-1.0096279-turtle.txt
N-Triples: 831-1.0096279-rdf-ntriples.txt
Original Record: 831-1.0096279-source.json
Full Text
831-1.0096279-fulltext.txt
Citation
831-1.0096279.ris

Full Text

AN ANALYSIS OF MORPHOLOGICAL VARIATION WITHIN AND BETWEEN STREAM POPULATIONS OF GASTEROSTEUS ACULEATUS LINNAEUS. by KATHLEEN ANNE SHAW B.Sc, U n i v e r s i t y of Alberta, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES (Department of Zoology) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1985 (c) Kathleen Anne Shaw, 1985 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available f o r 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 or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of ~^00i~06gY  The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date Z 3 JVLY M&S ABSTRACT Two small streams on Vancouver Island, B r i t i s h Columbia, were ex-amined for patterns of morphological v a r i a t i o n i n Gasterosteus  aculeatus. A progressive analysis beginning with P r i n c i p l e Components Analysis, followed by Nested and P a r t i a l l y Nested Multiple Analysis of Variance and then Duncan's Mult i p l e Range Test was used for pattern determination. This new technique allows the researcher to sequentially i s o l a t e the pattern of v a r i a t i o n at d i f f e r e n t l e v e l s of generality from species to i n d i v i d u a l organisms. The pattern of v a r i a t i o n for G_. aculeatus i n Bonsall Creek and Nunns Creek can be summarized as follows: The largest amount of v a r i a t i o n accounted for by the analysis i s i n t e r -preted as i n d i v i d u a l v a r i a t i o n . Populations also account f or a large amount of v a r i a t i o n and show consistent, f u l l y nested patterns of v a r i a -t i o n at each of the analysed geographic and microgeographic l e v e l s . These populations are probably genealogical u n i t s . The so-called " l e i u r u s " and "trachurus" forms on the P a c i f i c coast of North America do not appear to be evolutionary e n t i t i e s , but to be h i s t o r i c a l ar-t i f a c t s that are best viewed as labels for the extremes of a continuum of v a r i a t i o n . In areas where d i s t i n c t populations meet, d i f f e r e n t c l i n e s are documented i n the two stream systems. In Nunns Creek there i s a smooth c l i n e between populations, whereas i n Bonsall Creek there i s a step c l i n e . i i i TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS i i i LIST OF TABLES v LIST OF FIGURES v i ACKNOWLEDGMENTS i x CHAPTER 1. INTRODUCTION 1 1.1. Preamble 1 1.2. H i s t o r i c a l Review 2 CHAPTER 2. MATERIALS AND METHODS 9 2.1. C o l l e c t i o n L o c a l i t i e s and Specimens 9 2.2. Data C o l l e c t i o n 16 2.3. Analysis 20 CHAPTER 3. RESULTS 32 3.1. Data Set A l l 32 3.1.1. P r i n c i p l e Components Analysis (PCA) 32 3.1.2. Nested Mul t i p l e Analysis of Variance (ANOVA) 38 3.1.3. Duncan's Multiple Range Test (DUNCAN'S) 43 3.2. Data Set West 51 3.2.1. P r i n c i p l e Components Analysis (PCA) 51 3.2.2. Nested Mul t i p l e Analysis of Variance (ANOVA) 57 3.2.3. Duncan's M u l t i p l e Range Test (DUNCAN'S) 63 CHAPTER 4. DISCUSSION I l l 4.1. Technique I l l 4.2. Variation 113 4.2.1. Variation at the Species Level 114 4.2.2. Variation at the Population Level 117 4.2.3. "Leiurus" and "Trachurus" Forms 124 4.2.4. Clines 127 4.2.5. Individual Variation 129 4.3. Avenues for Future Study 130 LITERATURE CITED 132 Appendix 1. Description of Sites and Intersites Bonsall Creek 141 Appendix 2. Description of Sites Nunns Creek 157 Appendix 3. Character Descriptions 167 Appendix 4. Identifications of Parasites (Characters 95—101) ... 176 V LIST OF TABLES Table i . Results of PCA of data set A l l for the f i r s t six axes 33 Table i i . Percent of component variation explained by strata in the ANOVA analysis of data set A l l 39 Table i i i . Results of PCA of data set West for the f i r s t five axes 52 Table iv. Percent of component variation explained by strata in the ANOVA analysis of data set West 58 v i LIST OF FIGURES Figure 1. Map of Bonsall Creek showing c o l l e c t i o n s i t e s 11 Figure 2. Map of Nunns Creek showing c o l l e c t i o n s i t e s 13 Figure 3. G^  aculeatus showing placement of characters as described i n Appendix 3 17 Figure 4. Amount of v a r i a t i o n i n data set A l l accounted f o r by the f i r s t ten PCA axes 23 Figure 5. Amount of v a r i a t i o n i n data set West accounted for by the f i r s t ten PCA axes 26 Figure 6. Component loadings for eastern and western s t r a t a for each PCA axis ( Y l — Y 6 ) for data set A l l 44 Figure 7. Component loadings for G_. aculeatus (west), (2. aculeatus (east) and G_. wheatlandi f o r each PCA axis ( Y l — Y 6 ) f o r data set A l l 48 Figure 8. Component loadings for Bonsall Creek and Nunns Creek for each PCA axis ( Y l — Y 6 ) f o r data set West 64 Figure 9. Component loadings for regional habitat s t r a t a f o r PCA axis Yl f o r data set West 68 Figure 10. Component loadings for regional habitat s t r a t a f o r PCA axis Y2 for data set West 71 Figure 11. Component loadings for regional habitat s t r a t a for PCA axis Y3 for data set West 74 Figure 12. Component loadings f o r regional habitat s t r a t a for PCA axis Y4 for data set West 76 v i i Figure 13. Component loadings for regional habitat s t r a t a f o r PCA axis Y5 f o r data set West .79 Figure 14. Component loadings f o r s i t e s t r a t a f o r PCA axis Y l for data set West 82 Figure 15. D i s t r i b u t i o n of character 1 (standard length) for each s i t e stratum for data set West 85 Figure 16. D i s t r i b u t i o n of character 23 (depth from DII to ECTO) for each s i t e stratum f o r data set West 87 Figure 17. D i s t r i b u t i o n of character 21 (depth from DSO to VI) for each s i t e stratum for data set West 89 Figure 18. Component loadings for s i t e s t r a t a f o r PCA axis Y2 for data set West 91 Figure 19. D i s t r i b u t i o n of character 67 (number of l a t e r a l plates on r i g h t side) for each s i t e stratum f o r data set West 94 Figure 20. D i s t r i b u t i o n of character 71 (number of l a t e r a l plates between ascending process and l a s t precaudal vertebra) for each s i t e stratum for data set West 96 Figure 21. D i s t r i b u t i o n of character 73 (number of l a t e r a l plates on caudal peduncle) for each s i t e stratum f o r data set West 98 Figure 22. Component loadings for s i t e s t r a t a f o r PCA axis Y3 for data set West 100 Figure 23. Component loadings for s i t e s t r a t a for PCA axis Y4 for data set West • 1°3 Figure 24. Component loadings f o r s i t e s t r a t a for PCA axis Y5 for data set West ., 1 0 5 v i i i Figure 25. Component loadings for time strata for a l l PCA axes (Yl—Y5) for data set West 107 ACKNOWLEDGMENTS I wish to thank Dr.B.W.Coad (NMC) and Dr.J.S.Nelson (UA) f o r loaning specimens under t h e i r care. Many people were coerced into dragging f i e l d equipment over f i e l d s , through forests and up streams. In 1981 Clyde Murray and Marvin Rosenau took part i n this endeavor. In 1982 Lynne Quarmby, Marvin Rosenau, Dave Glen, Richard O'Grady, Maggie Hampong, Doug Begle and Patrick Lavin were kind enough to accompany me. I thank them a l l for t h e i r time and e f -f o r t s . A l l parasite specimens were examined and i d e n t i f i e d by Richard O'Grady. Steve Campana took the time to teach me his technique for grinding and reading o t o l i t h s . D.G. Smart of 3M Canada Ltd. provided the lapping f i l m s , free of charge, which were necessary for grinding the o t o l i t h s . Advice on d r a f t i n g , presentation and computing matters were provided by Jack Scott (UA) and Marc Majka (UBC). Dr.J.D. McPhail not only provided f i n a n c i a l support for t h i s project but, most importantly, gave me the time and space to pursue many of the topics i n biology which i n t e r e s t me. For t h i s , I am very grate-f u l . I would also l i k e to thank him for the thousands of stimulating discussions and ideas on v a r i a t i o n and v a r i a b i l i t y of sticklebacks. I am indebted to Dr.Jack Maze for his d i r e c t i o n and help i n the develop-ment of an appropriate analysis technique. The development of my ideas about biology would never have been possible without the friendship and c r i t i c a l discussions I had with Dr. Dan Brooks. Much of what I learned from him was f r e e l y given i n numerous impromptu discussions over a cof-fee, at a party or during a walk. I cannot thank him enough for his friendship and support throughout my stay at UBC. Doug Begle and Andrew Simons also provided much needed friendship and encouragement during most of this study. I thank them for always being willing to talk over a problem, a paper or an idea; as well as for enriching my views of systematics, evolution, and biology in general. Two people deserve thanks for encouraging me to undertake this enterprise Dr.Steve Ashe (MNH-Chicago) and Dr.John Addicott (UA). I w i l l always be grateful for their inspiration. Many people reviewed drafts of this thesis. I believe that their comments and criticism improved the fi n a l form. The views presented herein however, are my own. Donna Chin typed the i n i t i a l draft of this thesis. Finally I would like to thank a l l my friends and my parents, who have dealt with my excitement and frustration while remaining sup-portive . 1 CHAPTER 1. INTRODUCTION 1.1. Preamble A l l known species vary, although not a l l species vary to the same extent or in the same way. The ultimate source of this variation and variability l i e s in the genetic information coded on the DNA of each c e l l . Not a l l of the possible variation and variability is expressed or observed in the natural world. This i s because the expression of genetic information is constrained in three ways: by epigenetic i n -teractions within the nucleus and between the nucleus and the cytoplasm, by the regulation imposed by developmental programs and pathways, and by environmentally mediated processes such as selection. This variation and variability is essential for every species since i t provides the raw material for evolution. Evolutionary history unifies a l l of l i f e . Only natural groups (rather than a r t i f i c i a l ones, such as grades) can take part in evolutionary processes (Wiley, 1981). These natural groups are real and independent of our perceptions. Given that these natural groups exist, i t i s necessary for any biologist interested in evolutionary processes to f i r s t discover the natural groups involved. Natural groups can best be detected on the basis of patterns of variation (Wiley,1981). Phylogenetic systematists have developed techniques of pattern analysis which allow them to detect a variety of natural groups, from kingdoms to species. It is conceivable though, that there may exist natural groups below the level of species. 2 Populations could be natural groups in many circumstances; for ex-ample, during the formation of new species. Pattern analysis could potentially be used in these cases to detect the differentiating popula-tions. Once the pattern of variation within and between these popula-tions is known, i t may then be possible to more closely examine the role of these populations (as natural groups) in evolutionary processes such as speciation. This study attempts to discover and describe in detail the pattern or patterns of morphological variation within and between populations of Gasterosteus aculeatus Linnaeus (the threespine stickleback) in two natural stream systems. The pattern of variation from the level of species to individuals w i l l be considered. Only the species G_. aculeatus w i l l be referred to as the "stickleback". 1.2. Historical Review Gasterosteus aculeatus has been of interest to naturalists and professional biologists alike, at least since i t s description by Lin-naeus in 1758. This species has been widely studied by behaviorists, physiologists, morphologists and evolutionary biologists. Despite the tremendous number of published works on their variation and taxonomy, the problems of their geographic variation and speciation have not been solved. Current investigations are not yet satisfactory and their postulated evolutionary histories are often contradictory. The va r i a b i l i t y within the genus Gasterosteus has been discussed for almost 200 years. Cuvier and Valenciennes (1829) recognized ten species within the genus Gasterosteus, of which G_. aculeatus, G_. leiurus 3 and G_. trachurus were but three. By 1925, there were over forty species assigned to the genus. This splitting of the genus Gasterosteus was largely due to the philosophical background of the time. Researchers at this time considered each form to be an individually created im-mutable type. As reasonable naturalists, they were documenting the diversity of God's creation. In 1925 however, Bertin (1925) revised this early taxonomy; placing a l l of these species within a single highly variable species G_. aculeatus. He fe l t that the variation should best be described as a small number of morphotypes. In addition, Bertin believed that the number of lateral plates increased in direct propor-tion to the salinity of their environment. Heuts (1946,1947a,b,1949) convinced many scientists that he had a better explanation for the variability of European sticklebacks. He believed that Bertin's morphotypes ("leiurus" and "trachurus") were better viewed as two adaptive phenotypes which were environmentally i n -duced. "Trachurus" individuals were produced in saline waters and "leiurus" individuals were produced in freshwaters. In addition, Heuts f e l t that "semi-armatus" individuals represented hybrids between the other two types and that these were found in intermediate habitats, where "trachurus" and "leiurus" individuals could both be found. This explanation was in concordance with the "new synthesis" of evolutionary theory made at this time by the combination of population genetics and Darwinian evolution. Largely due to the acceptance of Heuts' work, research on European G. aculeatus is now focused on ecological and physiological questions. Two researchers however, continued to document the patterns of variation 4 for this species in Europe Miinzing (1959,1962,1963,1972) and Penczak (1960,1962,1965,1966). Their results only partly concur with the pat-tern of variation described by Heuts, since the number of exceptions has increased with almost every paper published. It is possible that as more detailed examinations of populations throughout Europe are com-pleted, that the weight of the accumulated exceptions w i l l lead to the rejection of Heuts' conclusions. Wootton (1976) has provided investigators interested in the biology of G_. aculeatus (and other gasterosteids) with a rather complete summary of the research published before 1974. There have also been a large number of published works on G_. aculeatus from the Atlantic and Pacific coasts of North America. Many of the studies of Atlantic North American populations have been concerned with the reproductive isolation of (J. aculeatus from the gasterosteid species G_. wheatlandi, Apeltes quadracus and Pungitius pungitius (Audet et a l . , 1985; Craig and FitzGerald,1982; Reisman,1968; Rowland,1983a,b; Wilz,1970; Worgan and F i t z -Gerald,1981a,b). Others have been concerned with the morphological variation of G_. aculeatus (Bell and Baumgartner,1984; Coad,1974; Coad and Power,1973,1974; Garside and Hamor,1973; Gross and Anderson,1984; Hagen and Moodie,1982; Perlmutter, 1963). However, most of these mor-phological studies have focused on meristic variation and have not been concerned with either geographic variation or evolutionary history. For a study of the pattern of variation of G_. aculeatus on the Pacific coast of North America (such as the present study), this i s not a limitation for two major reasons. F i r s t , the populations of sticklebacks in Europe and Atlantic North America have been isolated from those in Pacific 5 North America for a very long time and therefore are l i k e l y to be e v o l u t i o n a r i l y d i s t i n c t . Second there have been a tremendous number of studies concerned with the geographic v a r i a t i o n and evolutionary h i s t o r y of P a c i f i c North American sticklebacks. The patterns of v a r i a t i o n and explanations of these patterns have been investigated by many researchers on the P a c i f i c coast. The f i r s t to comment on th i s v a r i a b i l i t y was Rutter i n 1896. In the l a s t f i f t e e n years, however, i n t e r e s t i n t h i s problem has increased and focused around three i n d i v i d u a l s . Dr.Michael B e l l has published many papers ranging from f o s s i l systematics to morphological v a r i a t i o n (for example; B e l l , 1973, 1977, 1979, and B e l l et a l . , 1985). Dr.J.D.McPhail and many of h i s graduate students have studied v a r i a t i o n i n r e l a t i o n to questions of speciation (for example; McPhail, 1969, 1977, 1984; McPhail and Hay, 1983; Ridgway and McPhail, 1984; MacLean, 1980). Dr.D.W. Hagen and h i s associates have been interested i n questions pertaining to morphological differences between populations (for example; Hagen and G i l b e r t -son,1973a,b; Hagen, Moodie and Moodie,1980; Kynard,1978, 1979a,b; Moodie,1972b; Reimchen,1980; Semler,1971). The published works, ideas and opinions of these i n d i v i d u a l s and the i r associates are just beginning to provide the kind of data base and s o p h i s t i c a t i o n necessary for a b i o l o g i c a l l y reasonable d e s c r i p t i o n and explanation of the pattern of v a r i a t i o n f or G. aculeatus populations on the P a c i f i c coast of North America; which i s necessary for detecting natural groups or evolutionary e n t i t i e s . Lentic and l o t i c populations however, have not come under equal scrutiny. Populations within streams have r a r e l y been sampled on a 6 microgeographic scale nor have they been given intensive examination. Most studies including stream populations represent these populations with fewer than f i v e l o c a l i t i e s per stream and often only one or two (Avise,1976; Bell,1981, 1982b; Hagen,1973; Moodie and Reimchen,1976; Ross,1973, Withler, 1980). I intend to show that sampling on t h i s gross a scale i s not s u f f i c i e n t f o r an understanding of the pattern of v a r i a -t i o n of stream populations of G_. aculeatus. There are only f i v e published studies of v a r i a t i o n within and between stream populations of G_. aculeatus on the P a c i f i c coast of North America that have been sampled on a microgeographic scale. One study i s concerned with only a single character breeding colours of sexually mature males (Hagen and Moodie,1979). Another examines the v a r i a t i o n of only two characters l a t e r a l plate morph (sensu Hagen and G i l b e r t -son,1973a) and standard length (Baumgartner and Bell,1984). A t h i r d ex-amines the v a r i a t i o n of three independent characters number of l a t e r a l plates (and l a t e r a l plate morph), standard length and number of g i l l rakers (Bell,1982a). It should be clear that the r e s o l u t i o n of patterns 'of v a r i a t i o n within such a p l a s t i c species as G_. aculeatus cannot be achieved on the basis of so few characters, except i n very s p e c i a l cases. The two remaining studies also s u f f e r from the above short-comings, but to a lesser degree. Hagen (1967) examined s i x morphological characters number of l a t e r a l plates, number of g i l l rakers, standard length, number of dorsal rays, number of anal rays and body depth divided by standard length; and one electrophoretic locus. In addition to using a small number of characters, Hagen (1967) sampled at s i t e s 7 which averaged three-quarters of a mile apart. This is l i k e l y too coarse a scale for distinguishing between a l l possible populations. Hagen was however, more interested in behavioural and ecological i s o l a -tion between what he believed to be separate species. It would seem that a more complete and exact study could have been made i f the popula-tions had been delineated prior to the study; with special attention paid to the large number of ephemeral headwater streams which become isolated most summers. Hagen did find step-clinal variation where his two types of sticklebacks overlapped in the L i t t l e Campbell River. Bell and Richkind (1981) published a study that included only two independent morphological characters number of lateral plates (and lateral plate morph) and standard length; but also included twelve elec-trophoretic l o c i . In this study the closest sites were about 1 km apart and in many cases much further. This is probably too coarse a scale for distinguishing between a l l possible populations, although the total number of characters should allow better resolution than any of the other studies. Bell and Richkind (1981) also found c l i n a l variation within the stream system (Ventura River) they examined. A number of researchers have also attempted to place the reported patterns of variation (albeit imperfectly known) into an evolutionary framework. Miller and Hubbs (1969) explained the va r i a b i l i t y within (2. aculeatus as a result of intergradation and introgression during the secondary contact of three subspecies. In many ways, this explanation is very similar to Heuts' (1947b) explanation of variation in European sticklebacks. Hagen and McPhail (1970) while allowing for the pos-s i b i l i t y of occassional secondary contact, f e l t that local natural 8 s e l e c t i o n was responsible for the observed v a r i a t i o n . Two years l a t e r , Hagen and Gilbertson (1972) s t i l l concluded that s e l e c t i o n maintained the observed v a r i a t i o n ; although the emphasis was s l i g h t l y d i f f e r e n t . B e l l (1976) developed an i n t e r e s t i n g scenario for the evolution of G_. aculeatus i n the P a c i f i c Northwest. Unfortunately, there i s l i t t l e d i r e c t evidence to support or contradict his scenario. I t therefore remains an i n t e r e s t i n g but untestable argument. The f i n a l to the explanation of the pattern of v a r i a t i o n and therefore the evolutionary h i s t o r y of G_. aculeatus has not yet been achieved. It was my i n t e n t i o n that t h i s study would provide a b i o l o g i c a l l y reasonable d e s c r i p t i o n and explanation of the pattern of morphological v a r i a t i o n of t h i s species, i n two separate stream systems and that i t would be more complete than the studies to date. In t h i s way we can continue to b u i l d better data bases for future research. 9 CHAPTER 2. MATERIALS AND METHODS 2.1. Collection Localities and Specimens During the summer of 1981, more than forty-five locations on thirty-five streams were visited on Vancouver Island and the lower main-land near Vancouver, British Columbia. At each location an attempt was made to collect sticklebacks. From this survey, (based on the charac-teristics of the surrounding habitat such as, the presence of rooted vegetation, the flow rate, the bottom type and the presence of shade during the afternoon) i t was generally possible to predict whether or not sticklebacks would be present at a previously uncollected location. Four c r i t e r i a were used to choose the two creeks (Bonsall Creek and Nunns Creek) examined in depth by this study. (1) Each creek must have few or no tributaries or feeder streams thus reducing the complexity of the system. (2) Each creek must have continuous or nearly continuous habitat for sticklebacks. (3) Both creeks must be well matched in terms of regional habitats, temperature, bottom type, fauna and flora. (4) Both creeks must be reasonably accessible throughout their length. Bonsall Creek is approximately seven miles (11 km) long and located near the town of Duncan on Vancouver Island at a longitude of 123°43'W and latitude of 48°52'N. Near the headwaters, one branch of the creek flows along the side of Big and L i t t l e Sicker Mountains (and is un-suitable for sticklebacks) but the other branch flows through small wooded ranches and is prime habitat for sticklebacks. After a stretch of r i f f l e s and pools, the creek then slows and broadens as i t flows through farmland, once again prime stickleback habitat. The creek then 10 flows through the H a l a l t Indian Reserve, an area which has had a few of the trees removed through s e l e c t i v e logging. After flowing past a f i s h -processing plant, the creek then meanders between wooded banks to the estuary. Within the estuary, which i t shares with the Chemainus River, the channels cut by Bonsall Creek at low ti d e run out to the Shoal I s -lands i n Stuart Channel near the Crofton Pulp M i l l . Nunns Creek i s approximately four and a half miles (7 km) long and located near the town of Campbell River on Vancouver Island at a longitude of 125°15'W and l a t i t u d e of 50°02'N. The headwaters of the creek flow through a wooded ravine made up almost e n t i r e l y of pools and r i f f l e s . The creek then slows, deepens and broadens, due to a decrease i n slope and to the a c t i v i t y of beaver (Castor canadensis). From t h i s area behind the l o c a l rodeo grounds, the creek then runs through the woods of Nunns Creek Park, and then through the Campbell River Indian Reserve. This area has many trees that gradually t h i n out and are replaced by scrub and brush towards Discovery Passage. The mouth of the creek empties into the in s i d e of Tyee Spit. From the area disturbed by beavers downstream the creek i s prime stickleback habitat. Once the creeks were chosen, the en t i r e length of each creek was walked and surveyed. Id e a l l y , the sampling s i t e s were to be 100 feet i n length with 500 feet between s i t e s . In practice however, a s i t e was as close to 100 feet as possible and the next su i t a b l e l o c a t i o n more than 500 feet upstream was designated the next s i t e . Occasionally, there was f a r more than 500 feet between s i t e s (see F i g s . 1, 2; Appen-dices 1, 2). Bonsall Creek had a t o t a l of f o r t y s i t e s ( F i g . 1), Nunns Creek had a t o t a l of twelve s i t e s ( F i g . 2). See Appendix 1 for a 11 Figure 1. Map of Bonsall Creek showing collection sites. 12 13 Figure 2. Map of Nunns Creek showing collection sites. 14 15 d e s c r i p t i o n of the s i t e s on Bonsall Creek and Appendix 2 for a descrip-t i o n of the s i t e s on Nunns Creek. Between A p r i l and September of 1982, each s i t e was sampled f i v e times at approximately one month i n t e r v a l s . The greatest number of specimens from the largest number of s i t e s was c o l l e c t e d i n early June. The next largest c o l l e c t i o n was made i n l a t e July. For l o g i s t i c reasons, only the two largest c o l l e c t i o n s were examined i n d e t a i l . A two-man, two meter pole seine with a mesh s i z e of approximately 1 cm was used to c o l l e c t the specimens from almost a l l s i t e s , except at s i t e s 33 to 35 on Bonsall Creek and at s i t e s 11 and 12 on Nunns Creek where min-now traps were used. At each s i t e an e f f o r t was made to c o l l e c t f i f t y specimens of greater than 30mm standard length, but at times t h i s was not possible. Total numbers of specimens c o l l e c t e d at each s i t e are therefore a crude measure of abundance, from common (a f u l l sample of f i f t y specimens) to uncommon (a sample of few or no specimens). Once captured the l i v e specimens were quickly frozen i n water on dry i c e and kept frozen u n t i l examined i n the lab. Specimens of G_. aculeatus and i t s nominal s i s t e r taxon G_. wheatlan- d i from eastern Canada were obtained through loans from the National Museum of Canada (NMC). These specimens were included i n t h i s study as a guide to the nature and magnitude of differences between nominal s i s t e r taxa, at an i n t e r s p e c i f i c l e v e l . For each species, a series of f i v e specimens was examined from each of three regional habitats; f r e s h -water portions of streams, estuaries and sea bays. ((J. aculeatus: NMC 77-1012 5 QUEBEC; Matamec River at second f a l l s ; NMC 76- 0167 5 QUEBEC; Matamec River mouth; NMC 68-389 5 NOVA SCOTIA; Pictou Harbour. (G. 16 wheatlandi: NMC 66-0180 5 NEWFOUNDLAND; Western Arm Brook; NMC 66-0190 5 NEWFOUNDLAND; S.E. Conception Bay at mouth of brook; NMC 68-389 5 NOVA SCOTIA; Pictou Harbour.) 2.2. Data C o l l e c t i o n A "character" i s defined here as any measurable or qua n t i f i a b l e aspect of an i n d i v i d u a l which i s suspected to vary within the genus Gasterosteus. For each specimen examined, values for 101 characters were recorded. These characters include f i v e types of variables (Appen-dix 3) (Figure 3). (1) Measures of length, depth or width; which may contribute to differences perceived as size or shape. This includes characters 1—39, 42—50. (2) Placement and counts of s e r i a l l y repeated parts; which may be described as meristic characters. This includes characters 51—73, 76—89, 91—92. (3) Presence (or absence) of par-t i c u l a r a t t r i b u t e s ; which were scored as l=present, 0=absent. This i n -cludes characters 74—75, 93—101. (4) An estimate of age; character 90. (5) Codes for sex and maturity; which were scored as l=male, 2=female (character 41) and l=immature, 2=mature, 3=unripe, 4=ripe (character 40). If i n the lab, samples had been examined s e r i a l l y along the length of e i t h e r creek, the re s u l t s could be biased due to growing expecta-t i o n s . To eliminate t h i s bias, each s i t e was assigned a number and the order of examination of samples from s i t e s was then determined from a table of 14,000 random u n i t s . Each sample was then thawed and ( i f the remaining sample was greater than or equal to seven i n d i v i d u a l s ) a l l i n -d i v i d u a l s less than 30mm were refrozen. Otherwise a l l i n d i v i d u a l s less 17 Figure 3. (Z. aculeatus showing placement of characters as described in Appendix 3. a. Lateral view showing characters 9—13, 16—21, 24—26, 28—30, 32, 35 and 36. b. Lateral view showing characters 22, 23, 27, 31, 34, 37, 38 and 50. c. Ventral view showing characters 48 and 49. d. Isolated f i r s t dorsal spine showing character 43. 18 19 than 20mm were refrozen. Each thawed specimen was then assigned a number and (to eliminate any bias i n the choice of specimens) s i x i n -di v i d u a l s were chosen using a table of 14,000 random u n i t s . The remaining specimens were refrozen. In t h i s way, a maximum of s i x ran-domly chosen specimens from each s i t e were examined. The thawed, unpreserved specimens were then measured using vernier c a l i p e r s f o r the f i r s t 39 characters. Each measurement was taken to the nearest 0.1mm; except for character 15 (width of caudal peduncle) which was measured to the nearest 0.01mm. Once a l l 39 measures were taken, the pair of s a g i t t a l o t o l i t h s were removed through a single cut approx-imately between the dermosupraoccipital and f r o n t a l bones of the cranium and set aside. Each specimen was then scored for characters 93 (presence or absence of red throat pigmentation), 40 (maturity) and 41 (sex). The specimens were then eviscerated, the eyes removed and any macroscopic parasites set aside. The specimens from each s i t e were then fixed for three days i n 10% formalin, followed by 3 days of r i n s i n g i n water. They were then stained for c a r t i l a g e (using a l c i a n blue) and for bone (using a l i z a r i n red) and cleared. The double staini n g and cl e a r i n g procedure was the same as that of Dingerkus and Uhler (1977) except for the following: (1) the specimens were incubated (at about 40°C) during digestion by tr y p s i n to speed up th i s process to approximately three weeks, (2) the specimens were placed i n d i r e c t sunlight for the f i r s t two K0H:glycerine series (3:1,1:1) to speed up and improve the bleaching and cl e a r i n g of remaining muscle t i s s u e , and (3) the f i r s t bath of the KOH:glycerine serie s (3:1) consisted of up to 5ml of 3% hydrogen peroxide per 100 ml 20 to improve the bleaching of melanophores. The specimens were f i n a l l y placed i n 100% glycerine with a few c r y s t a l s of thymol added to reduce fungal and b a c t e r i a l growth. Once the specimens were double stained and cleared, further counts and measures for characters 42—89, 91—92 and 94 were taken using a Wild M5 microscope. Measurements were taken using an o p t i c a l g r a t i c u l e and l a t e r converted to the nearest 0.001 mm. The pair of s a g i t t a l o t o l i t h s which had been e a r l i e r set aside were dried and glued to precleaned sides using Krazy Glue(R). Once dry they were ground and buffed using 30mm to 0.3mm s i l i c o n e carbide and aluminum oxide 3M lapping f i l m s . The technique was e s s e n t i a l l y that of Campana (1983). A comparative scale of age (rather than an absolute scale) was developed by scoring each o t o l i t h u n t i l , through repeated examination, the same score was obtained each time (character 90). The parasites were placed i n 70% ethanol and l a t e r i d e n t i f i e d as f u l l y as possible by R. T. O'Grady. For a l i s t of these i d e n t i f i c a -t i o n s , see Appendix 4. F i n a l l y , each i n d i v i d u a l was scored for the presence or absence of each type of parasite (characters 95—101). A t o t a l of 566 in d i v i d u a l s were examined. 2.3. Analysis Not a l l of the specimens examined were included i n the analyses. Some specimens were excluded because the data for those specimens were incomplete. Incomplete data entries were generally the r e s u l t of loss of specimens or parts of specimens during staining and c l e a r i n g . This reduced the t o t a l number of specimens from 566 to 519. 21 In addition, a l l specimens collected from Bonsall and Nunns Creeks during the original stream survey (1981) were eliminated from the analyses. These were the f i r s t specimens processed for a l l 101 characters and therefore they potentially contained the largest measure-ment error. This reduced the total number of specimens from 519 to 489. A l l or some of these 489 specimens were used for the following two data sets. The f i r s t data set ( A l l ) , was used for comparisons between eastern and western Canadian populations as well as comparisons between G_. aculeatus and G_. wheatlandi. This data set consisted of a l l the specimens collected in 1982 for which there were complete data entries plus a l l the specimens of both species from eastern Canada. These specimens totalled 489. The second data set (West) was a subset of A l l and was used for analysis of specimens from Bonsall and Nunns Creeks. This subset con-sisted of a l l the specimens collected in 1982 from Bonsall and Nunns Creeks for which there were complete data entries. This data set con-tained 459 individuals. Not a l l of the characters originally assayed were used in the f i n a l analyses. A l l unvarying and redundant characters were eliminated. Three characters: 55 (total number of principle caudal rays), 60 (number of branched pectoral rays) and 84 (placement of most anterior epipleural rib) were eliminated because they did not vary. Four characters: 65 (total number of vertebrae), 66 (number of lateral plates on lef t side), 68 (total number of lateral plates) and 76 (total number of dorsal plates) were eliminated because they were split into other characters 22 and therefore the v a r i a t i o n which they might exhibit was already represented. It was necessary to eliminate a l l unvarying and redundant characters to s a t i s f y the assumptions of the analyses. This reduced the t o t a l number of characters from 101 to 94. These 94 characters were then considered for i n c l u s i o n i n the following character sets. The f i r s t character set (used with data set A l l ) does not include character 90 ( o t o l i t h count) because the sagittae from specimens on loan from the NMC had deteriorated. Another ten characters were excluded from t h i s character set because i n a preliminary P r i n c i p l e Components Analysis (PCA), they did not contribute s u b s t a n t i a l l y to any of the f i r s t s i x axes. Only s i x PCA axes were considered i n t h i s study for the following reason. Since each PCA axis summarizes a smaller and smaller portion of the o r i g i n a l v a r i a t i o n i n the data, some c r i t e r i o n must be used for choosing the number of axes to be examined. The usual r u l e i s to consider a l l axes that summarize more than one percent of the o r i g i n a l v a r i a t i o n . A PCA analysis of as large a data matrix as was produced for t h i s study, would r e s u l t i n about twenty axes that each summarize more than one percent of the o r i g i n a l v a r i a t i o n . Figure 4 shows the amount of v a r i a t i o n accounted for by the f i r s t ten PCA axes. From Figure 4 i t i s clear that a f t e r axis 6 the curve has reached an asymptote and therefore only the s i x axes before the i n f l e c t i o n point i n the l i n e are included i n t h i s study. The ten characters excluded i n this way are: 53 (number of branched caudal rays), 54 (number of un-branched caudal rays), 56 (number of dorsal rudimentary caudal rays), 61 (number of unbranched pectoral rays), 82 (placement of most anterior p l e u r a l r i b ) , 86 (number of branchiostegal rays), 96 (presence or ab-23 Figure 4. Amount of variation in data set A l l accounted for by the f i r s t ten PCA axes. P R I N C I P L E C O M P O N E N T A X I S 25 sence of Schistocephalus), 97 (presence or absence of cestode "type 2"), 98 (presence or absence of Proteocephalus) and 100 (presence or absence of Eustrongylides). The t o t a l number of characters remaining i n t h i s character set (used with data set A l l ) was 83. The second character set (used with data set West) does not include characters 59 (number of ve n t r a l rays), 74 (presence or absence of post-temporal) and 75 (presence or absence of supracleithrum). These characters were excluded because within t h i s data set they do not vary. Another ten characters were excluded from t h i s character set because i n a preliminary PCA analysis they did not contribute s u b s t a n t i a l l y to any of the f i r s t f i v e axes. Only the f i r s t f i v e axes were considered because axes s i x to ten were a f t e r the i n f l e c t i o n point i n Figure 5 and therefore contributed l i t t l e to the a n a l y s i s . The ten characters ex-cluded i n t h i s way are: 54 (number of unbranched caudal rays), 57 (number of ve n t r a l rudimentary caudal rays), 61 (number of unbranched pectoral rays), 81 (number of p l e u r a l r i b s ) , 86 (number of bran-chiostegal rays), 96 (presence or absence of Schistocephalus), 97 (presence or absence of cestode "type 2"), 98 (presence or absence of Proteocephalus), 99 (presence or absence of F i l o c a p s u l a r i i n i n e nematodes) and 100 (presence or absence of Eustrongylides). The t o t a l number of characters remaining i n t h i s character set i s 81. The elimination of the above characters from consideration within the second character set was independent of those eliminated from the f i r s t character set and v i c e versa. Six d i f f e r e n t s t r a t i f i c a t i o n s were used i n the following analyses: geography, taxon, stream, region, s i t e and c o l l e c t i o n period. Two Figure 5. Amount of variation in data set West accounted for by the f i r s t ten PCA axes. P R I N C I P L E C O M P O N E N T A X I S 28 geographic s t r a t a were s p e c i f i e d when data set A l l was analysed. These were east (N=30) and west (N=459). These s t r a t a were used to examine po t e n t i a l differences based on geographic l o c a t i o n . Geographic d i f -ferences were suspected since Gasterosteus i s found on both sides of the continent but the eastern and western populations have had no oppor-tunit y to meet and interbreed, since at least before the l a s t g l a c i a -t i o n . Two taxonomic s t r a t a were used when data set A l l was analysed. These were G_. wheatlandi (N=15) and G_. aculeatus (N=474). These s t r a t a were used to examine the nature and magnitude of species s p e c i f i c d i f -ferences. They provided a "yardstick" for comparisons between popula-tions that may or may not be d i s t i n c t species. When data set West was analysed, two stream s t r a t a were s p e c i f i e d . These were Bonsall Creek (N=321) and Nunns Creek (N=138). These s t r a t a were used to examine the v a r i a t i o n associated with d i f f e r e n t streams. Four regional s t r a t a within each stream were used when data set West was analysed (see Appendices 1 and 2). These were: Estuarine (1); Bonsall Creek s i t e s V to Z, 1 to 8 (N=68) and Nunns Creek s i t e s 1 to 4 (N=42). Marine algae was present i n a l l of the s i t e s included i n t h i s region. Lower reaches (2); Bonsall Creek s i t e s 9 to 20 (N=113) and Nunns Creek s i t e s 5 to 8 (N=48). The s i t e s included i n t h i s region were portions of the creeks with r e l a t i v e l y undisturbed courses and well treed margins. Mid reaches (3); Bonsall Creek s i t e s 21 to 30 (N=106) and Nunns Creek s i t e s 9 and 10 (N=24). The s i t e s i n t h i s region were r e l a t i v e l y open and surrounded by pastures and f i e l d s . Headwaters (4); Bonsall Creek s i t e s 31 to 35 (N=34) and Nunns Creek s i t e s 11 and 12 29 (N=24). Just beyond the sites in this region are the upstream limits beyond which i t is impossible to collect sticklebacks. These four strata were used to examine the nature and magnitude of variation as-sociated with different regional habitats of both streams. Forty-three sites were used as strata in the analysis of data set West. Essentially, each stratum i s a collection site except for a few sites that were grouped together to achieve a minimum sample size of five specimens. These grouped strata were: (5) Bonsall Creek sites Z, 1, 2, 4 (N=7), (6) Bonsall Creek sites 5, 6 (N=14), (8) Bonsall Creek sites 9, 10 (N=10), (30) Bonsall Creek sites 32, 33 (N=15) and (31) Bon-s a l l Creek sites 34, 35 (N=13). These strata were used to examine the nature and magnitude of site specific variation. Two collection period strata were also used in the analysis of data set West. These were early June (N=228) and late July (N=231). These strata were used to determine the extent and nature of the varia-tion between collection periods. Three types of s t a t i s t i c a l analyses were used: Principle Compo-nents Analysis (PCA), Nested Multiple Analysis of Variance (ANOVA) and Duncan's Multiple Range Test (DUNCAN'S). The PCA i s part of the MTS MIDAS computing package, whereas the ANOVA and DUNCAN'S are part of the MTS ANOVAR computing package. PCA was used to summarize variation in the characters within the data sets; by isolating character systems, or PCA axes, which best ex-plain the largest portions of the variation in the data matrices. Each character system is s t a t i s t i c a l l y independent of a l l other character systems (Pimentel,1979). The characters that contribute most to any 30 p a r t i c u l a r axis can also be said to be best summarized by that axis. In t h i s way those characters which vary together, as evidenced by t h e i r common major contribution to a p a r t i c u l a r PCA axis, are grouped together in t o a character system. For l o g i s t i c reasons, only the f i r s t s i x PCA were examined for data set A l l and only the f i r s t f i v e PCA axes were ex-amined for data set West. Two models were used with the Nested Multiple Analysis of Variance (ANOVA), one with data set A l l and one with data set West. In each case the "sum of squares" was apportioned; but s t a t i s t i c a l s i g n i f i c a n c e was not imformative, because a l l comparisons between s t r a t a were s t a t i s t i c a l l y s i g n i f i c a n t due to the size of the data set. This stage of the analysis allows an understanding of the character systems produced as PCA axes by examining the amount and type of the v a r i a t i o n dependent on the various s t r a t a . The ANOVAs were the f i r s t step i n ex-amining hypotheses about evolution. The f i r s t model: Y l , Y2, Y3, Y4, Y5, Y6 = A + B(A) + E where Y l — Y 6 = component loadings f o r the characters i n PCA axes 1—6 A = geography B = taxon (which i s dependent on geography) E = r e s i d u a l was used with data set A l l . This model was used to examine the r e l a t i v e contribution of the st r a t a to o v e r a l l v a r i a t i o n of eastern and western Canadian populations of G_. aculeatus and G_. wheatlandi. 31 The second model: Y l , Y2, Y3, Y4, Y5 = A + B(A) + C(AB) + D + E where Y l — Y 5 = component loadings for characters i n PCA axes 1-5 A = stream B = region (which i s stream dependent) C = s i t e (which i s stream and region dependent) D = c o l l e c t i o n period E = re s i d u a l was used with data set West. This model was used to examine the r e l a -t i v e contribution of each stratum to the o v e r a l l v a r i a t i o n . Duncan's Mult i p l e Range Test (DUNCAN'S) was used to determine which s t r a t a (except taxon s t r a t a and s i t e strata) were s t a t i s t i c a l l y d i f -ferent. Taxon s t r a t a could not be examined using DUNCAN'S because a l -though the taxon G!. aculeatus could be s p l i t into eastern and western groups, the taxon (J. wheatlandi i s only an eastern group. This asym-metry i s not accepted by the MTS DUNCAN'S computing program. For s i t e s t r a t a DUNCAN'S provided means and standard deviations f or each s i t e stratum but there were too many means (43) to perform a range t e s t . DUNCAN'S was used i n th i s study because i t a test that w i l l accept s t r a t i f i e d data sets of PCA component loadings. Once s t a t i s t i c a l l y d i s t i n c t s t r a t a were i d e n t i f i e d , i t was possible to examine the data set for patterns of v a r i a t i o n (such as c l i n e s ) . This was the second step i n examining evolutionary hypotheses. CHAPTER 3. RESULTS 3.1. Data Set A l l 3.1.1. Principle Components Analysis (PCA) In combination, the f i r s t six PCA axes account for 68.77% of the total variation in data set A l l (Table i ) . In the following section, the major characters contributing to each PCA axis (Yl—Y6) are used to characterize each axis. The f i r s t axis (Yl) accounts for 43.37% of the total variation in data set A l l . The major contributors to this axis (characters 1—31, 34—39, 42—50) are morphometric characters. Only two measured characters are not major contributors to this axis character 32 (length from last D ray to base of caudal) and character 33 (length of caudal peduncle). A l l major contributors to this axis have positive component loadings and therefore covary. This axis summarizes the variation in "size" of each of the characters and w i l l be referred to in further discussions as the "size constellation". Typically, the f i r s t axis in PCA analyses describes variation in size (Pimentel, 1979) The second axis (Y2) accounts for a further 9.07% of the variation in data set A l l . The ten characters that contribute most to this axis are (in order of magnitude): 67 (number of lateral plates on right side), 71 (number of lateral plates between ascending process and last precaudal vertebra), 73 (number of lateral plates on caudal peduncle), 72 (number of lateral plates between f i r s t caudal vertebra and last ana pterygiophore), 32 (length from last D ray to base of caudal), 33 (length of caudal peduncle), 80 (number of g i l l rakers), 69 (number of 33 Table i . Results of PCA of data set A l l for the f i r s t six axes. 34 /. TOTAL V A R I A N C E 4 3 . 37 9 . 0 7 1 . V 1 . 163 14 - . 3 8 7 0 9 - 1 2 . V 2 . 1 5 1 0 9 - . 1 1 2 7 2 3 . V 3 . 1 4 4 7 8 - . 6 7 3 3 6 - 1 4 . V4 . 1 4 0 9 8 - . 8 7 9 6 2 - 1 5 . V 5 . 1 5 5 9 8 . 5 4 2 1 2 - 3 6 . V6 . 1 5 3 1 2 . 5 2 7 8 7 - 1 7 . V7 . 1 3 7 7 2 . 9 8 5 0 9 - 1 8 . V8 . 1 3 7 2 9 • . 8 1 9 2 5 - 1 9 . V9 . 1 5 2 9 7 • . 7 4 7 2 9 - 1 1 0 . V 1 0 . 1 2 S 6 1 • . 1 1 8 3 3 1 1 . V 1 1 . 1 4 4 1 2 . 9 7 0 2 8 - 1 1 2 . V 1 2 . 1 3 2 0 3 . 1 1 2 7 1 1 3 . V 1 3 . 1 5 2 5 4 . 3 5 5 6 2 - 1 1 4 . V 1 4 . 1 3 0 0 0 - . 1 1 0 2 8 1 S . V 1 5 . 1 2 2 5 7 . 1 5 1 7 0 1 6 . V 1 6 . 1 4 5 6 3 . 9 3 5 5 7 - 2 1 7 . V 1 7 . 1 4 1 9 4 . 2 2 0 9 2 - 1 1 8 . V 1 8 . 1 5 6 3 3 . 6 7 3 6 3 - 1 1 9 . V 1 9 . 1 1 5 15 - . 1 6 9 3 2 2 0 . V 2 0 . 1 5 5 1 1 • . 8 4 8 3 5 - 1 2 1 . V 2 1 . 1 5 9 6 3 - . 5 7 8 9 1 - 1 2 2 . V 2 2 . 1 6 0 0 0 - . 6 2 6 6 9 - 1 2 3 . V 2 3 . 1 6 2 7 8 . 2 9 9 4 2 - 2 2 4 . V 2 4 . 1 5 6 1 5 - . 5 2 8 9 9 - 1 2 5 . V 2 5 . 1 5 1 5 2 - . 6 2 9 6 7 - 1 2 6 . V 2 S . 1 5 4 6 4 - . 2 1 1 1 8 - 1 2 7 . V 2 7 . 1 5 4 8 7 . 8 5 8 9 6 - 3 2 8 . V 2 8 . 1 5 3 6 0 - . 8 1 1 3 7 - 1 2 9 . V 2 9 . 1 5 8 3 5 - . 4 9 7 3 4 - 2 3 0 . V 3 0 . 9 4 8 5 3 - 1 - . 174 10 3 1 . V 3 1 . 1 5 1 4 2 - . 2 5 7 7 1 - 1 3 2 . V 3 2 . 5 4 5 6 9 - 1 - . 2 4 8 2 7 3 3 . V 3 3 . 4 6 5 9 1 - 1 - . 2 4 8 1 1 3 4 . V 3 4 . 1 3 0 2 7 . 1 6 1 12 - 1 3 5 . V 3 5 . 1 4 8 4 6 - . 1 0 7 8 3 3 6 . V 3 6 . 1 3 S 8 2 - . 1 3 4 5 2 3 7 . V 3 7 . 1 5 8 6 5 . 6 7 3 6 3 - 1 3 8 . V 3 8 . 1 3 6 9 0 . 2 1 9 7 8 - 1 3 9 . V 3 9 . 1 4 5 0 1 - . 1 2 7 5 5 - 1 4 0 . V 4 0 . 8 2 5 2 8 - 1 . 6 5 8 7 8 - 1 4 1 . V 4 1 . 3 6 6 9 0 - 2 . 8 5 3 0 0 - 1 4 2 . V 4 2 . 1 3 3 5 7 . 9 7 3 9 4 - 1 4 3 . V 4 3 . 1 1 9 6 6 . 1 5 8 2 6 4 4 . V 4 4 . 1 2 2 7 6 . 1 6 5 3 4 4 5 . V 4 5 . 1 0 3 0 1 . 1 4 7 2 5 4 6 . V 4 6 . 9 1 8 9 4 - 1 . 1 3 5 4 1 4 7 . V 4 7 . 125 18 . 1 8 3 8 8 4 8 . V 4 8 . 1 4 9 9 6 . 7 6 8 2 7 - 1 4 9 . V 4 9 1 3 9 5 6 - . 8 7 6 0 9 - 1 5 0 . V 5 0 . 1 4 3 0 5 - . 5 1 3 0 8 - 1 5 1 . V 5 1 . 7 9 3 4 4 - 2 - . 1 7 0 1 8 - 1 5 2 . V 5 2 . 5 6 3 10 - 1 . 1 0 9 6 0 5 7 . V 5 7 . 1 0 7 6 0 - 1 . 8 4 8 9 2 - 1 5 8 . V 5 8 . 3 9 5 9 6 - 1 . 7 7 7 4 4 - 1 5 9 . V 5 9 - . 4 5 3 9 0 - 1 . 1 1 6 9 4 6 2 . V 6 2 . 1 4 9 7 5 - 1 . 8 9 0 6 7 - 1 6 3 . V 6 3 . 3 7 2 8 0 - 1 . 5 7 8 1 8 - 1 6 4 . V 6 4 . 7 9 9 5 5 - 2 - . 1 2 2 0 7 6 7 . V 6 7 . 8 0 4 1 1 - 1 . 2 6 8 1 1 6 9 . V 6 9 4 3 5 8 1 - 1 . 192 17 7 0 . V 7 0 . 1 3 3 7 2 - 1 . 4 8 2 9 4 - 1 7 1 . V 7 1 . 7 2 3 7 4 - 1 . 2 6 4 8 5 7 2 . V 7 2 . 7 9 1 7 0 - 1 . 2 5 134 7 3 . V 7 3 . 8 2 7 1 5 - 1 . 2 5 8 0 6 7 4 . V 7 4 . 4 6 7 16 - 1 - . 1 2 0 4 4 7 5 . V 7 5 . 4 5 1 3 4 - 1 - . 1 1 4 5 3 7 7 . V 7 7 - . 9 0 4 1 1 - 2 . 2 3 7 2 9 - 1 7 8 . V 7 8 . 5 5 6 2 9 - 2 - . 4 2 8 7 4 - 1 7 9 . V 7 9 - . 1 2 8 0 7 - 1 . 6 9 2 7 9 - 2 8 0 . V 8 0 . 6 5 1 12 - 1 . 1 9 9 0 9 8 1 . V 8 1 . 3 0 1 3 3 - 1 . 7 4 3 7 9 - 1 8 3 . V 8 3 . 5 4 0 0 4 - 1 . 7 5 0 0 3 - 2 8 5 . V 8 5 - . 1 6 5 9 1 - 1 - . 5 2 8 1 3 - 1 8 7 . V 8 7 . 1 1 8 5 6 - 1 - . 2 7 0 6 6 - 1 8 8 . V 8 8 . 5 0 1 9 5 - 1 . 4 6 5 9 0 - 2 8 9 . V 8 9 . 4 1 4 1 1 - 1 - . 1 9 0 4 7 - 1 9 1 . V 9 1 - . 8 5 6 7 0 - 2 - . 4 5 3 4 0 - 1 9 2 . V 9 2 - . 5 9 2 7 4 - 2 - . 4 3 9 2 2 - 2 9 3 . V 9 3 . 4 6 8 7 8 - 1 - . 2 3 3 6 7 - 1 9 4 . V 9 4 . 8 9 0 9 8 - 2 . 9 0 3 7 1 - 2 9 5 . V 9 5 - . 1 5 0 2 0 - 1 - . 1 7 3 2 5 - 1 9 9 . V 9 9 . 8 8 0 0 4 - 2 . 3 5 8 4 4 - 1 1 0 1 . V 1 0 1 . 4 0 5 0 4 - 2 - . 3 5 7 2 4 - 1 ( 3 ) ( 4 ) ( 5 I ( 6 1 6 . 48 4 . 0 2 3 . 0 5 2 . 7 9 1 5 6 9 6 - 2 . 3 7 1 8 7 - 1 . 3 3 6 1 0 - 2 . 8 9 2 5 9 - 2 1 3 9 9 4 - 1 . 1 1 5 6 1 . 2 2 2 2 5 - 1 . . 5 7 0 6 6 - 2 1 0 3 6 5 - 1 . 1 8 1 4 6 . 9 1 0 7 2 - 2 . 7 9 3 4 2 - 2 4 2 4 7 3 - 1 . 5 5 0 3 3 - 1 . 3 7 2 1 7 - 1 . 1 6 3 5 8 - 1 4 3 5 8 1 - 2 . 1 1 6 7 5 . 1 4 3 7 7 - 1 . 2 7 5 9 6 - 1 7 7 8 9 8 - 2 . 1 0 0 2 6 . 1 5 5 7 5 - 1 . 5 4 1 2 2 - 1 1 3 4 4 0 - 1 . 2 1 0 5 7 . 2 2 0 0 8 - 1 . 5 1 3 2 9 - 1 1 0 4 8 3 - 1 . 1 2 2 7 6 . 5 1 7 0 8 - 2 . 2 9 8 5 3 - 1 1 8 6 0 0 - 1 . 1 1 3 3 8 . 3 8 4 5 4 - 1 . 3 4 3 7 5 - 1 4 6 9 4 1 - 1 . 8 3 5 5 0 - 1 . 2 6 2 7 7 - 2 . 1 0 6 7 6 7 8 3 5 7 - 1 . 1 2 9 6 1 - 1 . 1 9 2 4 0 - 1 . 6 2 8 1 3 - 1 1 0 0 3 4 . 1 3 0 9 6 . 1 1 2 3 2 - 1 . 2 4 6 3 5 - 1 5 6 0 1 7 - 1 . 9 5 9 4 0 - 1 . 2 5 0 1 4 - 1 . 6 7 9 9 9 - 1 4 9 0 1 8 - 1 . 1 4 6 3 1 . 1 9 8 4 5 - 1 . 2 1 9 0 7 - 1 6 1 9 7 6 - 1 . 1 0 2 8 1 - . 4 4 2 9 6 - 1 . 7 1 8 6 4 - 1 6 5 2 14 - 1 . 1 0 6 5 0 - 1 . 2 7 8 2 2 - 1 . 8 9 2 0 5 - 1 77 124 - 1 . 2 1 6 6 0 - . 1 1 7 9 7 - 1 . 4 2 0 3 3 - 2 3 3 1 4 9 - 1 - . 4 9 6 4 8 - 1 . 1 2 3 5 9 - 1 . 1 6 7 4 3 - 1 3 7 2 9 7 - 1 . 9 7 2 3 2 - 1 . 1 4 8 8 2 - 1 . 1 5 8 8 4 - 1 1 7 6 0 8 - 1 . 6 5 0 5 6 - 1 . 2 4 9 5 6 - 1 . 2 2 2 18 - 1 1 9 0 0 3 - 1 - . 6 9 8 2 7 - 1 . 1 5 2 2 3 - 1 . 8 2 1 0 2 - 2 3 4 9 9 3 - 1 . 1 5 3 3 7 - 1 . 1 9 1 9 7 - 1 . 1 5 0 7 9 - 2 5 2 2 2 7 - 1 . 2 6 7 2 9 - 1 . 9 1 9 2 3 - 2 . 7 0 3 3 7 - 2 4 8 4 5 0 - 1 - . 8 5 3 5 6 - 1 . 1 0 5 5 9 - 1 . 4 9 2 5 7 - 1 6 7 8 6 2 - 1 - . 1 0 7 2 7 . 7 2 0 7 4 - 2 . 5 6 1 4 7 - 1 8 8 7 7 0 - 1 - . 1 1 6 4 2 - . 9 8 3 0 6 - 2 . 3 1 0 3 5 - 1 6 1 4 0 9 - 1 - . 1 4 9 9 0 . 1 4 8 4 0 - 1 . 2 2 4 6 6 - 1 4 3 0 0 4 - 1 - . 1 0 0 3 0 . 9 9 8 7 0 - 2 . 2 1 7 7 2 - 1 2 5 8 8 0 - 1 - . 6 7 4 3 8 - 1 . 5 7 2 9 3 - 2 . 4 3 8 9 5 - 1 2 4 5 0 4 - 1 . 1 5 6 4 5 - . 5 9 0 1 2 - 1 . 1 1 9 4 0 4 6 3 6 8 - 1 . 8 1 5 4 7 - 1 . 2 7 3 8 3 - 1 . 3 6 4 3 9 - 1 3 8 7 7 7 - 2 . 1 0 4 5 5 - 1 - . 1 1 3 4 7 - . 2 1 5 4 6 2 6 9 9 2 - 1 . 4 4 4 9 2 - 1 - . 7 5 2 7 6 - 1 - . 2 1 5 3 6 1 1 9 4 2 - T . 1 5 2 1 0 . 6 5 6 7 4 - 2 . 8 0 1 8 4 - 2 2 0 8 7 9 - 1 - . 2 9 5 0 2 - 1 - . 4 1 4 4 3 - 3 . 1 1.420 - 1 3 2 5 7 8 - 1 - . 1 1 9 8 0 . 1 9 7 6 3 - 1 . 1 3 6 2 0 - 1 1 3 6 9 2 - 1 . 3 0 9 2 9 - 1 . 1 7 6 1 1 - 1 . 3 6 8 0 7 - 2 6 8 3 5 7 - 1 - . 2 0 1 8 7 - . 4 4 2 9 4 - 2 . 4 5 3 0 4 - 1 6 9 7 6 2 - 1 - . 1 5 2 6 7 . 2 1 6 3 2 - 1 . 6 4 1 9 3 - 1 1 7 9 0 4 - . 2 6 8 6 1 . 3 8 3 9 1 - 2 ' . 1 1 8 3 9 1 0 4 3 6 - . 3 6 7 5 9 - . 1 1 5 3 1 . 1 3 3 0 2 3 9 2 4 2 - 2 . 1 3 1 6 1 . 5 1 3 6 0 - 2 - . 4 2 4 7 7 - 1 1 3 2 3 1 - 1 . 9 1 6 9 2 - 1 - . 2 6 4 17 - 1 - . 1 2 0 1 7 1 3 6 6 5 - 1 . 6 5 5 9 4 - 1 - . 2 8 3 7 2 - 1 - . 1 3 2 6 0 2 0 7 5 9 - 1 . 9 1 8 0 9 - 1 - . 8 7 9 9 5 - 1 - . 1 9 3 7 8 4 3 0 8 8 - 1 . 1 0 5 6 1 - . 1 0 1 3 1 - . 2 0 1 2 4 1 9 1 2 4 - 1 . 9 8 8 0 7 - 1 - . 1 9 1 1 2 - 1 - . 6 8 0 9 9 - 1 9 8 2 5 0 - 1 - . 4 2 3 2 6 - 1 . 1 3 8 7 8 - 1 . 1 9 7 7 9 - 2 8 9 7 4 1 - 1 - . 9 5 4 4 5 - 1 . 7 2 0 5 8 - 2 . 3 2 0 6 0 - 1 6 7 4 1 2 - 1 . 1 3 5 0 8 - 1 . 1 0 8 9 4 - 1 . 4 0 9 6 9 - 1 9 0 1 8 0 - 2 . 2 1 6 1 0 - 1 - . 5 6 3 7 9 . 1 0 6 2 9 2 2 6 7 3 - . 2 0 5 2 6 - 1 . 5 0 8 9 8 - 2 . 1 6 9 3 4 1 1 6 9 7 - 1 - . 3 3 3 2 0 - 1 - . 1 6 1 4 8 - 1 . 1 0 8 0 0 2 1 8 8 9 . 4 9 4 3 6 - 1 - . 2 0 0 2 7 - 2 . 1 8 2 0 5 3 3 1 15 . 1 0 3 2 6 . 1 4 3 2 8 - 1 . 1 7 2 6 6 5 1 4 3 1 - 1 . 1 5 1 3 8 . 6 8 6 8 4 - 2 . 1 6 6 3 1 1 5 2 7 6 - . 1 3 6 3 8 - . 7 3 2 6 7 - 2 . 9 0 6 2 4 - 1 2 3 2 3 7 . 5 8 5 0 7 - 1 - . 1 2 3 0 6 - 1 . 4 0 3 8 2 - 1 1 0 1 2 7 . 2 8 8 1 5 - 1 - . 2 9 4 4 2 - 1 - . 6 2 7 8 B - 1 8 4 2 7 7 - 1 - . 7 6 2 8 8 - 2 . 5 2 5 8 2 - 2 - . 6 7 6 3 4 - 1 1 4 2 7 1 - 1 . 1 2 7 7 3 . 1 8 5 7 8 - 1 . 185 14 9 S 9 2 8 - 1 . 1 4 7 0 3 - 1 - . 3 7 6 16 - 1 . 6 1 1 2 3 - 1 9 5 6 14 - 1 . 1 8 0 8 7 - 1 - 2 9 9 4 9 - 1 . 6 3 3 2 7 - 1 I 0 S 10 . 4 7 9 4 8 - 1 - . 2 8 1 9 8 - t . 6 6 8 17 - 1 3 3 8 7 0 - . 1 0 4 3 6 - . 1 3 3 8 5 - t . 1 7 3 3 2 3 3 4 6 5 - . 1 0 1 1 2 - . 1 4 3 6 9 - 1 . 17 174 1 1 9 6 3 - 1 . 2 6 1 1 2 - 1 . 1 5 2 8 5 . 1 2 2 4 9 1 1 4 9 3 - 2 . 2 6 7 8 3 - 1 - . 4 4 9 6 5 . 1 0 7 1 5 1 3 5 3 1 . 4 1 6 8 5 - 1 - . 1 1 3 4 8 . 4 3 7 9 1 - 1 116 14 . 2 2 4 2 3 - 1 . 1 9 3 7 6 - 1 . 3 3 3 2 8 - 2 1 7 2 5 5 - 1 . 9 3 7 3 1 - 2 . 6 4 5 5 0 - 2 . 1 8 4 9 5 2 1 5 5 4 - . 1 3 3 6 7 - . 3 3 3 8 2 - 1 . 5 5 2 4 3 - 1 7 1 2 7 6 - 1 - . 1 4 0 1 3 - 1 . 1 7 5 9 4 - 1 . 3 1 5 9 0 7 8 18 1 - 2 . 3 4 8 0 1 - 1 - . 5 6 7 8 9 . 1 1 8 1 6 3 1 2 4 9 - . 9 6 9 9 4 - 1 - . 6 9 4 15 - 2 . 8 3 4 7 2 - 1 2 9 2 9 0 - . 9 9 4 0 8 - 1 - . 1 6 9 1 3 - 1 . 1 3 7 6 3 6 1 1 7 2 - 1 - . 2 1 0 0 7 - 1 - . 2 0 9 9 0 - 1 . 2 6 0 6 2 5 3 3 5 1 - 1 - . 5 8 6 6 3 - 1 . 7 3 0 7 7 - 1 . 1 8 4 7 1 6 7 4 2 6 - 1 . 2 6 3 7 3 . 1 5 18 1 . 2 0 7 8 9 6 8 4 0 8 - 3 . 7 7 2 9 4 - 1 - . 7 7 2 8 5 - 1 . 1 0 4 3 2 5 9 6 7 4 - 1 - . 5 5 6 7 6 - 1 . 7 1 6 6 1 - 1 . 1 4 8 6 4 4 9 4 6 8 - 1 - . 6 2 123 - 1 . 5 8 7 4 6 - 1 . 1 4 3 7 7 3 5 3 2 9 - 2 - . 7 2 5 0 4 - 1 . 3 8 5 6 5 - 1 . 9 8 0 1 8 - 1 35 l a t e r a l plates preceding ascending process), 47 (length of VI) and 30 (depth from l a s t D ray to l a s t A ray). Characters 30, 32 and 33 have negative component loadings (the others are p o s i t i v e ) and therefore vary inversely to the others. Most of these characters (47, 67, 71, 72, 73 and 80) have been t r a d i t i o n a l l y used to d i s t i n g u i s h between the "trachurus" and " l e i u r u s " forms of G_. aculeatus (Bertin,1925; Heuts,1947; Miinzing,1959; Wootton,1976). This axis summarizes the v a r i a t i o n i n "form" of each of the characters and i n further discussions w i l l be referred to as the "form c o n s t e l l a t i o n " . This c h a r a c t e r i z a t i o n i s strengthened by the fact that the remaining major contributors (characters 30, 32, 33 and 69) are consistent with the c h a r a c t e r i z a t i o n that t h i s second axis represents differences i n form. I t i s a r e l a -t i v e l y common r e s u l t i n PCA studies for the second axis to describe v a r i a t i o n i n shape (Bookstein,1978; Humphries ejt al_., 1981; Pimentel, 1979). The t h i r d axis (Y3) accounts for a further 6.48% of the v a r i a t i o n i n data set A l l . The ten characters that contribute most to t h i s axis are ( i n order of magnitude): 74 (presence or absence of posttemporal), 75 (presence or absence of supracleithrum), 59 (number of v e n t r a l rays), 88 (number of pterygiophores supporting D rays), 89 (number of ptery-giophores supporting A), 64 (number of caudal vertebrae), 52 (number of dorsal rays), 58 (number of anal rays), 83 (number of e p i p l e u r a l r i b s ) and 40 (maturity). Characters 40 and 59 have negative component loadings (the others are p o s i t i v e ) and therefore vary i n inverse propor-t i o n to the others. Most of these characters (52, 58, 59, 64, 74 and 75 are known to d i f f e r between G. aculeatus and G. wheatlandi 36 (Hubbs,1929; Scott and Crossman,1973; Wootton,1976). This axis sum-marizes the "species" v a r i a t i o n i n each of the characters and i n further discussions be referred to as the "species c o n s t e l l a t i o n " . Species s p e c i f i c differences were an expected r e s u l t . Characters 83, 88 and 89 are consistent with the view that t h i s t h i r d axis represents species d i f f e r e n c e s . Character 40 however, was not expected to be a major con-t r i b u t o r to t h i s a xis, but may indicate differences i n age or i n the numbers of ripe and unripe females between the species. I n t e r e s t i n g l y , the number of pterygiophores supporting the dorsal rays (88) and anal f i n (89) are more i n d i c a t i v e of differences between the species than the number of dorsal (52) or anal rays (58). In taxonomic studies usually only the rays are counted. Note that the magnitude of the v a r i a t i o n ex-plained as differences between G_. aculeatus and G_. wheatlandi (Y3) i s less than that between forms. This may be a r e s u l t however, of the small number (N=15) of G_. wheatlandi specimens i n t h i s study. The fourth axis (Y4) accounts for a further 4.02% of the v a r i a t i o n i n data set A l l . The ten characters that contribute most to t h i s axis are ( i n order of magnitude): 41 (sex), 40 (maturity), 93 (presence or absence of red throat pigmentation), 17 (length from A l to VI), 7 (length of upper jaw), 38 (length from l a s t P ray to VI), 3 (snout length), 30 (depth from l a s t D ray to l a s t A ray), 39 (width at ascending process) and 34 (length of pectoral base). Characters 17, 38, 39, 40 and 41 have negative component loadings and therefore vary i n -versely to the others. Most of these characters (3, 7, 17, 34, 39, 40, 41 and 93) are known or suspected to vary between the sexes (Woot-ton, 1976; McPhail pers. comm.). This axis then, summarizes the v a r i a -37 tion in "sex" of each of the characters in the data matrix and in later discussions i s referred to as the "sexual constellation". Characters 30 and 38, while not previously known to show sexual dimorphism, are not inconsistent with this characterization. The f i f t h axis (Y5) accounts for a further 3.05% of the variation in data set A l l . The ten characters that contribute most to this axis are (in order of magnitude): 87 (number of pterygiophores supporting D spines), 51 (number of dorsal spines), 78 (number of dorsal plates between DI and DII), 77 (number of dorsal plates preceding DI), 93 (presence or absence of red throat pigmentation), 41 (sex), 79 (number of dorsal plates between DII and D i l i ) , 32 (length from last D ray to base of caudal), 46 (length of Al) and 45 (length of D i l i ) . Of these ten, only characters 77 and 93 have positive component loadings. The characters with the highest component loadings (41, 51, 77, 78, 79, 87 and 93 have two features in common a polymorphism in dorsal plates and the associated spines, and sex. This axis then appears to summarize the "dorsal plate" variation and is more common in males (hence the con-tribution of characters 41 and 93) and in is later referred to as the "dorsal plate constellation". Characters 32, 45 and 46 are consistent with this hypothesis, although there was no a priori reason for ex-pecting them to be major contributors to this axis. The sixth axis (Y6) accounts for a further 2.79% of the variation in data set A l l . The ten characters that contribute most to this axis are (in order of magnitude): 85 (number of epurals), 91 (number of epurals fused together), 32 (length from last D ray to base of caudal), 33 (length of caudal peduncle), 93 (presence or absence of red throat pigmentation), 46 (length of A l ) , 45 (length of D i l i ) , 70 (number of l a t e r a l plates touching ascending process), 81 (number of p l e u r a l r i b s ) and 92 (number of epurals fused to penultimate vertebra). Characters 32, 33, 34 and 46 have negative component loadings and therefore vary inversely to the others. The two characters with the highest component loadings (85 and 91) deal with the v a r i a t i o n i n number and placement of the epurals. This axis may summarize, i n part, the "epural" v a r i a t i o n and i s l a t e r r eferred to as the "epural c o n s t e l l a t i o n " . Character 92 i s , as expected, consistent with this hypothesis. Characters 83, 33, 45, 46, 70, 81 and 91 are not inconsistent with t h i s hypothesis, but there was no a p r i o r i reason for expecting them to be major contributors to t h i s a x i s. 3.1.2. Nested Mu l t i p l e Analysis of Variance (ANOVA) This analysis examines i n d e t a i l the apportionment of the 68.77% of the v a r i a t i o n i n data set A l l accounted for by the f i r s t s i x PCA axes. This apportionment i s according to the r e l a t i v e contributions of the geographic and taxonomic s t r a t a and a re s i d u a l to each axis (Table i i ) . In t h i s analysis the re s i d u a l includes the contribution of a l l possible s t r a t a other than geography or taxon (such as region or time, i n d i v i d u a l v a r i a t i o n and any measurement e r r o r ) . For the f i r s t axis ( Y l ) , the re s i d u a l accounts for almost 88% of the v a r i a t i o n . This means that the largest amount of v a r i a t i o n i n the "size c o n s t e l l a t i o n " i s unaccounted for by the model. Geographic s t r a t a account for less than 1% of the v a r i a t i o n , i n d i c a t i n g that v a r i a t i o n i n the "size c o n s t e l l a t i o n " of the eastern and western s t r a t a i s s i m i l a r . 39 Table i i . Percent of component v a r i a t i o n explained by s t r a t a i n the ANOVA analysis of data set A l l . Yl Y2 Y3 Y4 Y5 Y6 total geography • - A 0.19 17.91 31.54 3-22 0.01 0-92 3.90 taxon • B(A) 12.11 0.25 30.21 0.84 0.14 7.53 7.48 residual - E 87.70 81.84 38.25 95-95 99.86 91-55 57-39 68-77 o Taxonomic s t r a t a account for about 12% of the v a r i a t i o n , i n d i c a t i n g that G. aculeatus and wheatlandi d i f f e r i n the magnitude of the characters contributing to the "size c o n s t e l l a t i o n " . This i s not s u r p r i s i n g since G_. aculeatus i s , on the average, much larger than G_. wheatlandi. For the second axis (Y2), the r e s i d u a l again accounts for the largest portion of the v a r i a t i o n almost 82%. This indicates that most of the v a r i a t i o n i n the "form c o n s t e l l a t i o n " i s unaccounted for by the model. Geographic s t r a t a account for about 18% of the v a r i a t i o n , i n -d i c a t i n g that eastern and western s t r a t a d i f f e r i n the magnitude of the characters contributing to the "form c o n s t e l l a t i o n " . This was an unex-pected r e s u l t . Taxonomic s t r a t a account for less than 1% of the v a r i a -t i o n , i n d i c a t i n g that the v a r i a t i o n i n the "form c o n s t e l l a t i o n " i s not a l l i e d to taxonomic dif f e r e n c e s . Again, t h i s was an unexpected r e s u l t . For the t h i r d axis (Y3), the residual accounts for about a t h i r d of the v a r i a t i o n 38%. This indicates that about 62% of the v a r i a t i o n i n the "species c o n s t e l l a t i o n " i s accounted for by the model. Geographic s t r a t a account for about 31% of the v a r i a t i o n , i n d i c a t i n g that eastern and western s t r a t a d i f f e r i n the magnitude of a l l the characters con-t r i b u t i n g to species s p e c i f i c differences. This was expected since one species, G_. wheatlandi, i s only present i n the east. Not s u r p r i s i n g l y , taxonomic s t r a t a account for about 30% of the v a r i a t i o n , i n d i c a t i n g that G. aculeatus and G. wheatlandi d i f f e r i n the magnitude of a l l the characters contributing to the "species c o n s t e l l a t i o n " . This r e s u l t i s less extreme than one might expect, since both the s t r a t a and the axis are concerned with species s p e c i f i c d i f f e r e n c e s . 42 For the fourth axis (Y4), the r e s i d u a l accounts for about 96% of the v a r i a t i o n . Almost a l l of the v a r i a t i o n i n the "sexual c o n s t e l l a -t i o n " i s therefore unaccounted for by the model. Geographic s t r a t a ac-count for only 3% of the v a r i a t i o n , i n d i c a t i n g that eastern and western s t r a t a d i f f e r s l i g h t l y i n the "sexual c o n s t e l l a t i o n " . Taxonomic s t r a t a account for less than 1% of the v a r i a t i o n , i n d i c a t i n g that the magnitude of the "sexual c o n s t e l l a t i o n " i s not markedly d i f f e r e n t between G_. aculeatus and G_. wheatlandi. For the f i f t h axis (Y5), the r e s i d u a l accounts for almost 100% of the v a r i a t i o n . E s s e n t i a l l y , a l l of the v a r i a t i o n i n the "dorsal plate c o n s t e l l a t i o n " i s therefore unaccounted for by the model. Geographic and taxonimic s t r a t a account for less than 1% of the v a r i a t i o n . This indicates that v a r i a t i o n i n the "dorsal plate c o n s t e l l a t i o n " i s s i m i l a r for eastern and western s t r a t a , as well as for both species of Gasterosteus. For the s i x t h axis (Y6), the r e s i d u a l accounts for about 92% of the v a r i a t i o n . Thus almost a l l of the v a r i a t i o n i n the "epural c o n s t e l l a -t i o n " i s unaccounted f o r by the model. Geographic st r a t a account for less than 1% of the v a r i a t i o n , i n d i c a t i n g that eastern and western s t r a t a exhibit s i m i l a r v a r i a t i o n i n the "epural c o n s t e l l a t i o n " . Tax-onomic s t r a t a account for almost 8% of the v a r i a t i o n . This indicates that G. aculeatus and G_. wheatlandi d i f f e r i n the characters con-t r i b u t i n g to the "epural c o n s t e l l a t i o n " . For a l l s i x axes ( Y l — Y 6 ) , the PCA accounts f o r 68.77% of the v a r i a t i o n , but of that 68.77% almost a l l (57.39%) i s not accounted for by the s t r a t a . This means that most of the v a r i a t i o n accounted for i n 43 data set A l l i s not explained by my model. Geographic s t r a t a did however, account for almost 4% of the v a r i a t i o n and taxonomic s t r a t a ac-count for almost 7.5% of the v a r i a t i o n i n the data set. 3.1.3. Duncan's Multiple Range Test (DUNCAN'S) This analysis i d e n t i f i e s s t r a t a that are d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . As input i t uses the PCA scores and s t r a t a from the previous analyses. I n i t i a l l y , only the re s u l t s for the geographic s t r a t a w i l l be examined ( F i g . 6). For the f i r s t axis ( Y l ) , the eastern and western s t r a t a are not s i g n i f i c a n t l y d i f f e r e n t . This indicates that the characters con-t r i b u t i n g to the "si z e c o n s t e l l a t i o n " are s i m i l a r for both s t r a t a . The eastern stratum does however, show a greater v a r i a b i l i t y i n component loadings. This i s expected since the eastern stratum includes both G_. aculeatus and G^. wheatlandi, and since G_. aculeatus grows much larger than G. wheatlandi. For the second axis (Y2), the eastern and western s t r a t a are d i f -ferent. This means that the characters contributing to the "form con-s t e l l a t i o n " are d i f f e r e n t between s t r a t a . The eastern stratum i s more "trachurus"-like and has on the average more l a t e r a l plates, more g i l l rakers, narrower caudal peduncles and longer ventral spines than the western stratum. This may be because eastern G. wheatlandi and G_. aculeatus are more "trachurus"-like than " l e i u r u s " - l i k e . Also, the sample size of the eastern stratum i s small (N=30). The western stratum however, shows greater v a r i a b i l i t y i n component loadings. 44 Figure 6. Component loadings f o r eastern and western s t r a t a f o r each PCA axis ( Y l — Y 6 ) for data set A l l (means, standard deviations and s i g -n i f i c a n c e at p=0.05 in d i c a t e d ) . COMPONENT LOADING J . I I I I < I I I , 1 1 1 1 1 1 1 1 1 1 1 1 L . 1 _ 1 l 1 1 l 1 1 - c + -1 > X 2 < ° a> 51 -< OI — — -< 4>-46 For the third axis (Y3), the eastern and western strata are d i f -ferent. This indicates that the characters contributing to the "species constellation" are different between strata. This means that on the average the eastern stratum i s less G_. aculeatus- like than the western stratum. The eastern stratum also shows a greater variability in compo-nent loadings than the western stratum, although there is a slight over-lap between the strata. This result, that the eastern stratum is more variable and less G_. aculeatus- like than the western stratum, was ex-pected since the eastern stratum includes both G. aculeatus and G_. wheatlandi. For the fourth axis (Y4), the eastern and western strata are d i f -ferent. The characters contributing to the "sexual constellation" are therefore significantly different between strata. The eastern stratum is more "female"-like and includes more mature females with longer, deeper abdominal dimensions, shorter jaws and snouts, less deep caudal peduncles, wider bodies and shorter pectoral bases than the western stratum. Both strata however, are equally variable in the observed range of component loadings. This result is probably a function of the skewed sex ratio in the eastern sample (7 males: 23 females). For the f i f t h axis (Y5) the eastern and western strata are not dif -ferent. This indicates that the characters contributing to the "dorsal plate constellation" are similar for both strata. The western stratum does however, show greater v a r i b i l i t y in component loadings. This is probably result of the larger sample size of the western stratum (N=459). 47 For the sixth axis (Y6), the eastern and western strata are d i f -ferent. The characters contributing to the "epural constellation" are therefore significantly different between strata. This indicates that on the average the eastern stratum has a higher number of epurals, shorter caudal peduncles, shorter anal and third dorsal spines, a higher number of lateral plates touching the ascending process and more pleural ribs in males than the western stratum. Both strata however, are equally variable in component loadings. For the geographic strata, due to the asymmetry in the distribution of G_. wheatlandi, DUNCAN'S provided only the means and standard devia-tions for each stratum. These values are shown i n Fig. 7. Sta t i s t i c a l significance could not be established. For the f i r s t axis (Yl), there is overlap between the observed ranges of component loadings for both _G. aculeatus strata but they are separate from the observed range of component loadings for G_. wheatlan- di . Thus G. wheatlandi exhibits the smallest values for the characters contributing to the "size constellation", G_. aculeatus (west) has inter-mediate values and G_. aculeatus (east) exhibits the largest values. This result suggests the possibility of character displacement in the "size constellation" between the eastern species. For the second axis (Y2), there i s overlap between the observed ranges of component loadings for both eastern strata, but they are separate from the observed range of component loadings for western G_. aculeatus. This could be viewed as eastern strata appearing more "trachurus"-like in the characters contributing to the "form constella-tion" (as discussed earlier), while the western G. aculeatus exhibits Figure 7. Component loadings for G. aculeatus (west), G. aculeatus (east) and G. wheatlandi for each PCA axis (Yl—Y6) for data set A l l (means and standard deviations indicated). 1 3 -1 2 -II -10 -9 -8 -? -6 -5 -4 -3 -2 --7 --8 -- 9 --10 --II -- 1 2 -- 1 3 -G. aculeatus (west) 6. aculeatus (east) G. wheatlandi PCA Axis 50 a broader range including "trachurus"-like, " l e i u r u s " - l i k e and i n t e r -mediates. Although western G_. aculeatus was expected to show wide v a r i a t i o n i n the "form c o n s t e l l a t i o n " , i t was not expected that both eastern G_. aculeatus and e s p e c i a l l y G_. wheat land i would appear more "trachurus"-like i n the "form c o n s t e l l a t i o n " . For the t h i r d .axis (Y3), there i s almost complete overlap between the observed ranges of component loadings f o r both eastern and western aculeatus but they are quite separate from the observed range of com-ponent loadings for (5. wheatlandi. A l l s t r a t a show si m i l a r amounts of v a r i a b i l i t y . This was an expected r e s u l t since both the axis, "species c o n s t e l l a t i o n " , and the s t r a t a r e f l e c t species d i f f e r e n c e s . For the fourth axis (Y4), there i s overlap between the observed ranges of component loadings for a l l three s t r a t a . This v a r i a t i o n i n the "sexual c o n s t e l l a t i o n " may be viewed as a r e s u l t of the sex r a t i o of males to females for each stratum. Western (J. aculeatus has a sex r a t i o very near 1:1, although i t has a greater v a r i a b i l i t y i n the characters contributing to the "sexual c o n s t e l l a t i o n " . Eastern G. aculeatus has a sex r a t i o skewed towards females while for (J. wheatlandi i t i s skewed further towards females. For both eastern s t r a t a , the deviation from a 1:1 sex r a t i o may be a r e s u l t of small sample siz e (N=15 for each stratum). For (2. wheatlandi however, the deviation from a 1:1 sex r a t i o may also be a function of the larger si z e of females than males since the largest specimens were chosen for analysis ( i . e . , those greater than 30mm standard length when p o s s i b l e ) . For the f i f t h axis (Y5), there i s overlap between the observed ranges of component loadings of a l l three s t r a t a . The v a r i a t i o n i n the 51 "dorsal plate c o n s t e l l a t i o n " indicates that the western G_. aculeatus shows the greatest range of v a r i a b i l i t y , but that eastern G. aculeatus shows less than the average magnitude for characters contributing to thi s a xis, while the G_. wheatlandi stratum has a tendency to show a higher than average magnitude. Since the eastern s t r a t a have small variances and overlap only a small amount, there i s a s l i g h t p o s s i b i l i t y that t h i s character c o n s t e l l a t i o n e x hibits character displacement. For the s i x t h axis (Y6), there i s overlap between the observed ranges of component loadings for both G_. aculeatus s t r a t a , but they are separate from the observed range of component loadings for G_. wheatlan- d i . G_. wheatlandi e x h i b i t s the highest values for the characters con-t r i b u t i n g to the "epural c o n s t e l l a t i o n " . Western G. aculeatus shows s l i g h t l y higher v a r i a b i l i t y , but o v e r a l l has intermediate values. Eastern G. aculeatus shows the lowest values for the "epural c o n s t e l l a -t i o n " . This r e s u l t suggests the p o s s i b i l i t y of character displacement i n the "epural c o n s t e l l a t i o n " between the eastern species. 3.2. Data Set West 3.2.1. P r i n c i p l e Components Analysis (PCA) The f i r s t f i v e PCA axes together account f or 65.16% of the t o t a l v a r i a t i o n i n data set West (Table i i i ) . In the following section, the characters contributing most to each PCA axis ( Y l — Y 5 ) w i l l be used to characterize each axis. The f i r s t axis (Yl) accounts for 44.19% of the v a r i a t i o n i n data set West. The major contributors to t h i s axis (characters 1—31,34—40, 42—50) are a l l characters which were measured, not counted nor scored 52 Table i l l . Results of PCA of data set West for the f i r s t five axes. 7, T O T A L V A R I A N C E 4 4 . 19 9 . 4 9 4 . 7 1 3 . 6 1 3 . 1 7 1 . V I . 1 6 3 5 0 . 3 6 1 12 - 1 - . 3 3 2 5 4 - 1 . 1 8 8 9 7 - 1 . 1 8 6 0 3 - 2 2 . V 2 . 1 4 9 9 6 . 1 1 0 1 2 . 1 2 4 8 1 . 1 9 5 3 0 - 1 . 2 7 4 9 5 - 1 3 . V 3 . 1 4 2 2 7 . 7 2 1 0 3 - 1 . 1 7 9 1 8 . 3 7 9 3 3 - 1 . 1 4 1 0 0 - 1 4 . V 4 . 1 3 9 0 1 . 7 2 4 4 6 - 1 . 7 9 5 2 2 - 1 . 1 7 7 7 3 - 1 . 4 0 3 7 2 - 1 5 . V 5 . 1 5 5 8 6 - . 1 3 1 6 9 - 3 . 1 1 0 3 9 . 1 3 1 4 9 - 1 . 1 5 1 2 9 - 1 S . V S . 1 5 2 1 4 . 5 5 2 8 7 - 1 . 1 0 4 8 1 . 2 3 4 5 3 - 1 . 1 1 1 0 0 - 1 7 . V 7 . 1 3 6 6 9 . 1 0 4 7 9 . 1 9 7 8 7 . 1 7 0 0 1 - 1 . 2 3 3 2 8 - 1 8 . V 8 . 1 3 8 1 5 . 6 3 4 4 3 - 1 . 1 2 5 6 4 . 2 6 5 8 5 - 1 - . 6 6 9 3 5 - 2 9 . V 9 . 1 5 1 8 1 . 7 1 7 9 8 - 1 . 1 2 4 6 4 . 1 3 6 4 2 - 1 . 4 0 3 9 0 - 1 1 0 . V 1 0 . 1 2 2 3 9 . 1 1 8 9 7 - . 1 0 5 4 9 . 6 4 0 4 9 - 1 . 7 8 6 6 6 - 2 1 1 . V 1 1 . 1 5 0 8 3 - . 6 4 3 7 9 - 1 - . 3 4 2 5 6 - 1 . 2 7 0 5 1 - 1 - . 2 6 7 0 1 - 1 1 2 . V 1 2 . 1 3 6 7 7 - . 7 8 6 3 7 - 1 - . 1 6 0 0 3 . 1 5 1 8 6 - 1 . 9 0 7 6 8 - 2 1 3 . V 1 3 . 1 5 0 4 2 - . 5 8 8 7 1 - 1 - . 6 4 7 8 9 - 1 . 1 3 2 9 8 . 1 4 5 5 4 - 1 1 4 . V 1 4 . 1 3 0 6 1 . 1 1 1 0 1 . 1 1 7 8 8 . 7 7 1 7 2 - 1 . 2 6 2 3 3 - 1 1 5 . V 1 5 . 1 1 7 9 1 - . 1 6 5 9 5 . 9 2 3 9 2 - 1 . 9 9 0 9 6 - 1 - . 3 3 9 1 3 - 1 1 6 . V 1 6 . 1 4 3 6 8 - . 3 5 1 9 4 - 1 . 2 8 3 4 4 - 1 . 1 3 8 8 7 . 1 5 6 9 6 - 1 1 7 . V 1 7 . 1 4 1 7 0 - . 1 2 9 7 2 - 1 - . 2 2 2 8 8 . 2 4 6 9 3 - 1 - . 1 8 0 5 9 - 1 1 8 . V 1 8 . 1 5 6 8 1 - . 5 6 9 8 4 - 1 - . 6 8 7 6 9 - 1 - . 1 4 7 7 5 - 1 . 141 10 - 1 1 9 . V 1 9 . 1 1 5 1 9 . 1 4 8 1 1 . 1 2 1 3 3 . 3 7 6 4 6 - 1 . 1 7 3 4 8 - 1 2 0 . V 2 0 . 1 5 5 5 S . 8 2 9 0 7 - 1 6 6 B 8 9 -1 - 4 6 9 9 5 - 3 . 2 3 7 4 3 - 1 2 1 . V 2 1 . 1 G 0 5 5 . 5 0 7 7 9 - 1 - . 6 3 3 7 3 - r . 2 7 0 3 8 -1 . 1 1 2 0 3 - 1 2 2 . V 2 2 . 1 5 9 9 S . 6 7 4 8 8 - 1 - . 2 3 9 5 0 - 1 . 6 1 0 0 5 - 2 . 1 9 7 5 1 - 1 2 3 . V 2 3 . 1 6 3 8 8 . 1 0 8 7 1 - 1 - . 4 1 5 7 1 -1 . 1 8 4 4 7 - 2 . 8 0 7 2 3 - 2 2 4 . V 2 4 . 1 5 6 6 3 . 5 7 5 0 5 - 1 - 7 8 8 S 6 - 1 . 5 8 8 3 4 - 1 . 1 0 7 0 0 - 2 2 5 . V 2 5 . 1 5 2 8 3 . 6 8 8 1 9 - 1 - . 1 0 1 7 9 . 6 5 1 3 1 -1 - . 3 9 8 0 3 - 2 2 6 . V 2 6 . 1 5 6 7 2 . 3 6 3 2 4 - 1 - . 1 2 4 6 1 3 1 9 2 5 - 1 - . 1 8 0 4 7 - 1 2 7 . V 2 7 . 1 5 4 7 6 . 1 0 1 7 7 - 1 - . 1 5 5 0 3 . 4 0 2 6 9 - 1 . 9 7 8 2 1 - 2 2 8 . V 2 B . 1 5 3 5 4 . 8 2 9 6 3 - 1 - . 9 6 9 5 0 -1 . 4 5 3 7 5 -1 . 3 8 8 7 3 - 2 2 9 . V 2 9 . 1 5 7 7 7 - . 7 4 0 9 0 - 2 - . 4 9 7 8 6 -1 . 9 1 1 9 0 -1 - . 1 8 8 8 8 - 2 3 0 . V 3 0 . 9 4 9 6 3 - 1 . 1 6 0 5 5 . 1 2 8 6 3 - . 1 4 6 0 0 - . 4 7 5 4 2 - 1 3 1 . V 3 1 . 1 5 0 3 8 . 8 5 6 3 6 - 2 . 1 0 1 8 3 . 5 4 1 4 5 -1 . 2 3 1 0 4 - 1 3 2 . V 3 2 . 4 2 2 3 6 -1 . 2 4 8 0 2 . 3 9 6 6 8 - 2 . 1 3 7 0 9 - . 1 0 8 6 3 3 3 . V 3 3 . 3 0 6 5 9 - 1 . 2 4 9 0 8 . 5 2 5 7 0 - 1 - . 1 3 1 5 1 - . 6 5 5 2 2 - 1 3 4 . V 3 4 . 1 2 9 2 3 . 3 3 1 4 3 - 2 . 1 3 6 0 6 . 2 6 9 3 7 -1 . 1 2 0 8 1 - i 3 5 . V 3 5 . 1 5 1 2 6 . 8 6 5 9 8 - 1 - . 1 1 6 8 7 -1 . 1 0 0 8 5 - 1 - . 2 3 3 3 5 - 2 3 6 . V 3 6 . 1 3 7 9 8 . 1 3 0 2 4 - . 1 0 9 8 5 . 5 8 1 1 3 - 1 . 1 4 3 5 5 - i 3 7 . V 3 7 . 1 5 9 4 2 - . 6 1 5 1 9 - 1 . 1 9 1 3 8 - 1 - . 8 8 4 8 8 - 2 . 1 8 6 6 1 - 1 3 8 . V 3 8 . 1 3 7 6 7 - . 1 3 3 6 3 - 1 - ' . 1 9 4 6 0 . 5 6 9 4 1 - 1 - . 1 4 8 4 4 - 1 3 9 . V 3 9 . 1 4 8 2 0 . 1 5 7 8 9 - 1 - . 1 4 5 3 6 . 5 8 1 9 9 - 1 . 1 1 4 2 9 - 1 4 0 . V 4 0 . 9 7 2 6 7 - 1 - . 1 3 8 1 5 - 1 - . 2 8 8 3 9 . 8 7 3 0 4 - 1 - . 1 1 5 5 9 - 1 4 1 . V 4 1 . 3 1 9 1 3 - 2 - . 6 2 3 0 9 - 1 - . 3 7 5 6 5 . 6 3 6 8 0 - 1 - . 1 0 9 9 1 4 2 . V 4 2 . 1 3 2 6 5 - . 8 0 9 4 1 - 1 . 1 1 1 7 8 . 7 6 7 2 2 - 1 . 1 5 4 0 4 - 1 4 3 . V 4 3 . 1 1 9 4 9 - . 1 4 3 2 7 . 6 2 2 1 2 -1 . 1 2 9 9 9 - . 1 4 7 3 5 - 1 4 4 . V 4 4 . 1 2 1 6 4 - . 1 5 2 8 2 . 3 3 9 7 0 - 1 . 1 4 6 9 3 - . 1 3 6 4 0 - 1 4 5 . V 4 5 . 9 8 9 6 8 - 1 - . 1 2 9 2 8 . 4 5 9 4 5 - 1 . 2 1 8 2 0 - 7 2 9 4 6 - 1 4 6 . V 4 6 . 8 9 1 3 0 - 1 - . 1 1 3 0 8 . 4 8 8 4 8 - 1 . 2 4 4 9 3 - . 8 0 6 6 0 - 1 4 7 . V 4 7 . 1 2 7 8 9 - . 1 6 8 5 6 . 6 9 0 9 6 - 1 . 1 0 2 5 3 - . 1 0 0 8 5 - 1 4 8 . V 4 8 . 1 5 6 0 9 - . 4 3 5 7 1 - 1 - . 7 4 6 6 7 - 1 - . 1 3 3 2 7 - 1 . 1 2 7 4 6 - 1 4 9 . V 4 9 . 1 4 0 8 6 . 1 0 1 6 2 - . 1 0 2 8 1 . 2 3 7 9 0 - 1 . 1 4 8 5 1 - 2 5 0 . V 5 0 . 1 4 5 3 9 . 6 2 0 2 0 - 1 . 3 0 4 5 4 - 2 . 1 6 4 6 0 - 1 . 4 7 2 8 2 - 2 5 1 . V 5 1 . 8 2 4 6 3 - 2 . 1 3 2 6 4 - 1 . 3 6 0 3 0 - 1 . 1 4 5 3 8 - 1 - . 5 6 3 2 7 5 2 . V 5 2 . 4 5 0 5 7 - 1 - . 1 7 5 6 5 . 8 7 3 4 8 - 1 . 2 7 5 7 2 - . 2 5 7 7 8 - 1 5 3 . V 5 3 . 4 4 8 3 1 - 1 . 3 1 4 2 6 - 1 . 6 4 0 1 9 - 1 . 2 4 1 6 8 - 1 . 1 2 5 3 6 - 1 5 6 . V 5 6 - . 1 6 1 2 9 - 1 . 6 2 1 0 7 - 1 . 1 1 5 4 5 . 9 1 9 3 2 - 3 - . 3 6 0 4 6 - 1 5 8 . V 5 8 . 2 8 8 6 3 - 1 - . 1 3 3 8 1 . 1 6 1 3 0 . 2 9 4 7 7 - . 3 7 2 6 7 - 1 6 2 . V 6 2 . 2 8 5 2 9 - 1 - . 6 1 6 2 8 - 1 . 1 2 3 6 8 . 1 9 9 4 6 - 1 - . 4 8 9 4 6 - 2 6 3 . V 6 3 . 1 9 5 9 8 - 1 - . 1 1 6 7 4 - . 1 1 4 0 3 . 3 0 6 9 1 - 1 . 7 9 9 7 2 - 2 6 4 . V 6 4 - . 1 5 2 9 1 - 1 . 5 0 6 0 2 - 1 . 1 8 9 8 8 . 2 2 7 2 5 - . 4 1 4 2 7 - 1 6 7 . V 6 7 . 7 5 0 3 9 - 1 - . 2 8 3 6 5 . 3 3 1 6 9 - 1 - . 9 0 1 3 7 - 1 - . 2 1 8 2 8 - 1 6 9 . V 6 9 . 3 6 9 3 2 - 1 - . 2 0 1 2 4 . 4 0 2 7 6 - 2 - . 4 9 8 6 6 - 1 . 1 7 6 8 8 - 1 7 O . V 7 0 . 1 9 7 4 1 - 1 - . 3 5 7 5 8 - 1 . 1 2 3 4 7 . 5 0 8 1 3 - 1 . 4 0 4 7 4 - 2 7 1 . V 7 1 . 6 6 1 4 0 - 1 - . 2 7 8 9 4 . 1 8 116 - 1 - . 7 8 9 9 9 - 1 - . 2 9 4 0 4 - 1 7 2 . V 7 2 . 7 6 1 0 1 - 1 - . 2 6 9 1 7 . 2 5 2 4 4 - 1 - . 8 9 7 5 4 - 1 - . 2 4 6 4 2 - 1 7 3 . V 7 3 . 7 5 6 6 5 - 1 - . 2 7 5 8 4 . 5 0 1 5 2 - 1 - . 8 7 4 7 3 - 1 - . 186 18 - 1 7 7 . V 7 7 - . 9 9 2 9 6 - 2 . 2 2 9 7 4 - 1 . 1 9 6 5 2 - 1 - . 5 6 6 2 5 - 1 . 1 6 4 4 5 7 8 . V 7 8 . 5 8 6 9 4 - 2 . 3 8 9 6 0 - 1 . 3 2 6 3 6 - 1 . 1 9 3 6 9 - 1 - . 4 S 0 5 7 7 9 . V 7 9 - . 5 3 9 0 4 - 2 . 5 0 3 4 7 - 1 - . 1 5 1 5 9 - 1 - . 1 2 6 5 7 - 1 - . 1 2 9 2 6 8 0 . V 8 0 . 5 9 6 3 6 - 1 - . 2 2 0 5 3 . 4 6 4 2 5 - 1 - . 1 7 9 9 3 - 1 . 2 5 183 - 1 8 2 . V 8 2 . 2 4 9 8 1 - 1 . 3 7 8 9 0 - 1 . 8 8 4 8 7 - i - . 3 1 5 4 0 - 1 - . 3 7 9 6 5 - 1 8 3 . V 8 3 . 4 1 4 8 2 - 1 - . 9 7 4 5 2 - 1 . 3 6 8 3 9 - 1 . 1 2 3 8 9 - . 5 2 3 1 7 - 1 8 5 . V 8 5 - . 1 7 5 3 4 - 1 . 3 0 7 4 5 - 1 . 3 5 9 6 8 - 1 . 1 9 0 2 0 - . 9 1 2 9 1 - 2 8 7 . V 8 7 . 1 2 4 0 6 - 1 . 2 2 2 7 9 - 1 . 4 8 1 17 - 1 . 1 5 7 2 9 - 1 - 5 6 8 6 3 8 8 . V 8 8 . 2 8 6 6 4 - 1 . 1 2 7 2 7 . 5 4 131 - 1 . 3 5 0 3 4 - . 4 1 3 3 5 - 1 8 9 . V 8 9 . 2 1 4 7 5 - 1 . 8 6 8 12 - 1 . 6 0 9 8 0 - 1 . 3 9 3 9 2 - . 6 2 1 5 1 - 1 9 0 . V 9 0 . 1 0 5 8 8 . 9 0 7 8 7 - 1 . 4 8 4 7 9 - 1 . 7 6 0 9 2 - 1 . 1 8 5 5 1 - 1 9 1 . V 9 1 - . 104 1 1 - 1 . 2 1 0 8 1 - 1 . 2 4 1 5 7 - 1 . 1 6 0 6 8 - . 4 6 3 5 5 - 1 9 2 . V 9 2 - . 8 2 5 2 2 - 2 . 5 8 4 2 9 - 2 . 1 8 2 9 2 - 1 . 1 3 4 2 6 . 5 6 8 8 1 - 1 9 3 . V 9 3 . 5 0 1 2 2 - 1 . 7 3 5 5 4 - 3 . 2 7 8 2 4 . 7 6 3 8 8 - 1 . 1 3 9 6 5 9 4 . V 9 4 . 1 S 5 5 6 - 1 . 5 2 0 3 7 - 2 . 8 7 8 5 1 - 1 : 1 6 5 1 2 - 1 - . 9 7 5 7 9 - 1 9 S . V 9 5 - . 1 7 5 3 5 - 1 . 2 7 7 2 1 - 2 . 3 7 6 0 3 - 2 . 1 2 5 6 1 . 5 4 7 6 7 - 1 1 0 1 . V 1 0 1 . 4 0 4 5 1 - 2 . 3 1 3 0 7 - 1 . 5 1 8 7 6 - 1 . 9 7 6 6 4 - 1 . 2 8 3 2 9 - 1 54 (except for 40 maturity). Only two measured characters are not major contributors to t h i s axis character 32 (length from l a s t D ray to base of caudal) and character 33 (length of caudal peduncle). A l l major contributors to t h i s axis have p o s i t i v e component loadings and therefore covary. This axis i s almost i d e n t i c a l to the f i r s t PCA axis produced by the analysis of data set A l l . These axes d i f f e r only by the i n c l u -sion of character 40 i n the PCA of data set West. The i n c l u s i o n of character 40 (maturity) i s consistent with the hypothesis that t h i s axis summarizes the v a r i a t i o n i n " s i z e " . This group of characters w i l l be referred to as the "size c o n s t e l l a t i o n " . The second axis (Y2) accounts for a further 9.49% of the v a r i a t i o n i n data set West. The ten characters that contribute most to t h i s axis are ( i n order of magnitude): 67 (number of l a t e r a l plates on rig h t s i d e ) , 71 (number of l a t e r a l plates between ascending process and l a s t precaudal vertebra), 73 (number of l a t e r a l plates on caudal peduncle), 72 (number of l a t e r a l plates between f i r s t caudal vertebra and l a s t anal pterygiophore), 33 (length of caudal peduncle), 32 (length from l a s t D ray to base of caudal), 80 (number of g i l l rakers), 69 (number of l a t e r a l plates preceding ascending process), 52 (number of dorsal rays) and 47 (length of VI). Characters 32 and 33 have p o s i t i v e component loadings and therefore vary inversely to the others. This axis i s a l -most i d e n t i c a l to the second PCA axis produced by the analysis of data set A l l . These axes d i f f e r i n that (1) character 52 i s included as a major contributor i n the PCA of data set West, (2) the order of impor-tance as contributors of characters 32 and 33 i s reversed and (3) the sign for each component loading ( e i t h e r p o s i t i v e or negative) i s also 55 reversed. As with the second PCA axis for data set A l l , t h i s axis may be described as summarizing the v a r i a t i o n i n "form" and t h i s group of characters i s l a t e r referred to as the "form c o n s t e l l a t i o n " . This c h a r a c t e r i z a t i o n i s strengthened by the f a c t that the remaining major contributors (characters 32, 33, 52 and 69) are not inconsistent with t h i s c h a r a c t e r i z a t i o n . The t h i r d axis (Y3) accounts for a further 4.71% of the v a r i a t i o n i n data set West. The ten characters that contribute most to t h i s axis are ( i n order of magnitude): 41 (sex), 40 (maturity), 93 (presence or absence of red throat pigmentation), 17 (length from A l to VI), 7 (length of upper jaw), 38 (length from l a s t P ray to VI), 64 (number of caudal vertebrae), 3 (snout length), 58 (number of anal rays) and 12 (length from DII to D i l i ) . Characters 12, 17, 38, 40, 41, 58 and 64 have negative component loadings and therefore vary inversely to the others. This axis (Y3) i s very s i m i l a r to the fourth PCA axis produced by the analysis of data set A l l . These axes d i f f e r i n that (1) characters 12, 58, 64 replace 30, 34 and 39 as major contributors i n the PCA of data set West and (2) the order of importance of character 3 as a major contributing character i s reduced by one. As with the fourth PCA axis for data set A l l , t h i s axis may be described as summarizing the v a r i a t i o n i n "sex" and t h i s group of characters i s l a t e r referred to as the "sexual c o n s t e l l a t i o n " . Characters 12, 38, 58 and 64, while not previously known to show sexual dimorphism, are not inconsistent with t h i s c h a r a c t e r i z a t i o n . The fourth axis (Y4) accounts for a further 3.61% of the v a r i a t i o n i n data set West. The ten characters that contribute most to t h i s axis 56 are (in order of magnitude): 89 (number of pterygiophores supporting A), 88 (number of pterygiophores supporting D rays), 58 (number of anal rays), 52 (number of dorsal rays), 46 (length of Al), 64 (number of caudal vertebrae), 45 (length of D i l i ) , 85 (number of epurals), 91 (number of epurals fused together) and 44 (length of DII). Characters 44, 45 and 46 have negative component loadings and therefore vary i n -versely to the others. This axis is unique to this analysis of data set West, since i t shares only 2 and 3 major characters with the f i f t h and sixth axes (respectively) produced by the analysis of data set A l l and in neither case are the relative magnitudes similar. Most of these characters (52, 58, 64, 85, 88, and 89) are known or suspected to ex-hibit meristic variation (Coad and Power, 1974; Lindsey, 1972; Penczak, 1965; Rutter, 1896; Wootton, 1976). This axis may be described as sum-marizing the variation in "meristics" of each of the characters in the data matrix, and in later discussions is referred to as the "meristic constellation". Characters 44, 45, 46 and 91, do not vary in number but may increase in size with growth, and are not inconsistent with this characterization. The f i f t h axis (Y5) accounts for a further 3.17% of the variation in data set West. The ten characters that contribute most to this axis are (in order of magnitude): 87 (number of pterygiophores supporting D spines), 51 (number of dorsal spines), 78 (number of dorsal plates between DI and DII), 77(number of dorsal plates preceding DI), 93 (presence or absence of red pigmentation), 79 (number of dorsal plates between DII and D i l i ) , 41 (sex), 32 (length from last D ray to base of caudal), 94 (presence or absence of deformed vertebrae) and 46 (length 57 of Al). Characters 77 and 93 have positive component loadings (the others are negative). This axis is almost identical to the f i f t h PCA axis produced by the analysis of data set A l l . These axes differ in that (1) the order of importance as contributors of characters 41 and 79 is reversed, (2) character 94 is included as a major contributor in the PCA of data set West and (3) the order of importance of character 46 is reduced by one. As with the f i f t h PCA axis for data set A l l , this axis summarizes the "dorsal plate" variation and this group is later referred to as the "dorsal plate constellation". Characters 32, 46 and 94 are not inconsistent with this hypothesis. 3.2.2. Nested Multiple Analysis of Variance (ANOVA) This analysis examines in detail the apportionment of the 65.16% of the variation in data set West accounted for by the f i r s t five PCA axes. This apportionment is according to the relative contributions of the stream, region, site and collection period strata and a residual to each axis (Table i v ) . In this analysis the residual includes the contribu-tion of individual variation, any measurement error and any other possible strata. For the f i r s t axis (Yl), the residual accounts for about 59% of the variation. This indicates that over half of the variation in the "size constellation" i s unaccounted for by the model. Stream strata account for less than 1% of the variation, indicating that the variation in the "size constellation" of Bonsall Creek and Nunns Creek is similar. Re-gion strata account for almost 23% of the variation, indicating that the specimens collected in the four regional habitats differ in the "size Table iv. Percent of component variation explained by strata in the ANOVA analysis of data set West. Yl Y2 Y3 Y4 Y5 total stream - A 0.74 6.77 2.01 9 .42 0 .47 1.42 habitat -B(A) 2 2 . 8 4 55.1 1 1 .73 14.88 1 .31 15.98 site -C(AB) 16.80 8 .43 1 2 .30 7.53 1 1 .82 9.45 time -D 0.27 0.87 8 .43 4.27 0.02 0.75 residual -E 59 .35 28 .82 75 .53 63.91 86.38 3 7 . 5 6 65 .16 constellation". This i s not surprising, since the specimens found in marine areas are, on the average, larger than those found in freshwater areas (Hagen,1967; Penczak,1965). Site strata account for almost 17% of the variation, indicating that this data set shows site specific variation in the "size constellation". Time strata account for less than 1% of the variation, indicating that the variation in the "size constellation" of the two collection periods is similar. This suggests that the populations included in data set West exhibit a stable age distribution. For the second axis (Y2), the residual accounts for about a third of the variation almost 29%. This indicates that about 71% of the variation in the "form constellation" i s accounted for by the model. Stream strata account for almost 7% of the variation. Bonsall Creek and Nunns Creek therefore differ in the "form constellation". Prior to analysis, this was an unexpected result. Region strata account for over half of the variation in this axis about 55%. This means that there are marked regional habitat specific differences in the "form constella-tion". Based on the tremendous literature on the "trachurus" and "leiurus" forms, this result was expected (for example, Bertin,1925; Heuts,1947; Miinzing,1959; Wootton,1976). Site strata account for a further 8% of the variation, suggesting that the "form constellation" of data set West shows some site specific variation. Time strata ac-count for less than 1% of the variation, indicating that the variation in the "form constellation" of the two collection periods is similar. This was an expected result, since the "trachurus" and "leiurus" forms are now believed to be largely genetically determined and therefore the 61 v a r i a t i o n i n form should not be time dependent (Hagen,1973; Hagen and Gilbertson,1973). For the t h i r d axis (Y3), the r e s i d u a l accounts for three-quarters of the v a r i a t i o n about 75%. This means that most of the v a r i a t i o n i n the "sexual c o n s t e l l a t i o n " i s unaccounted f o r by the model. Stream s t r a t a account for 2% of the v a r i a t i o n , i n d i c a t i n g that there are s l i g h t differences i n the "sexual c o n s t e l l a t i o n " between Bonsall Creek and Nunns Creek. Region s t r a t a also account for about 2% of the v a r i a t i o n , suggesting that the regional habitats d i f f e r s l i g h t l y i n the "sexual c o n s t e l l a t i o n " . Site s t r a t a account for a further 12% of the v a r i a t i o n , i n d i c a t i n g that the "sexual c o n s t e l l a t i o n " exhibits s i t e s p e c i f i c v a r i a -t i o n . Time s t r a t a account for a further 8% of the v a r i a t i o n . This means that the "sexual c o n s t e l l a t i o n " d i f f e r e d between the two c o l l e c -t i o n periods. This was an expected r e s u l t , since i n d i v i d u a l s of both sexes are more mature and exhibit a greater range of secondary sexual c h a r a c t e r i s t i c s l a t e r i n the summer. For the fourth axis (Y4), the r e s i d u a l accounts for almost 64% of the v a r i a t i o n . Most of the v a r i a t i o n i n the "meristics c o n s t e l l a t i o n " i s therefore unaccounted for by the model. Stream s t r a t a account for about 9% of the v a r i a t i o n , i n d i c a t i n g that Bonsall Creek and Nunns Creek d i f f e r somewhat i n the "meristic c o n s t e l l a t i o n " . Region s t r a t a account for almost 15% of the v a r i a t i o n i n t h i s axis. The "meristic c o n s t e l l a -t i o n " therefore exhibits regional habitat s p e c i f i c v a r i a t i o n . Site s t r a t a account for almost 8% of the v a r i a t i o n , i n d i c a t i n g that the "meristic c o n s t e l l a t i o n " exhibits some s i t e s p e c i f i c v a r i a t i o n . Time s t r a t a account for a further 4% of the v a r i a t i o n , i n d i c a t i n g that the 62 two c o l l e c t i o n periods d i f f e r i n the "meristic c o n s t e l l a t i o n " . This r e s u l t was unexpected. For the f i f t h axis (Y5), the r e s i d u a l accounts for 86% of the v a r i a t i o n . Most of the v a r i a t i o n i n the "dorsal plate c o n s t e l l a t i o n " i s therefore unaccounted f o r by the model. Stream s t r a t a account for less than 1% of the v a r i a t i o n , i n d i c a t i n g that the "dorsal plate con-s t e l l a t i o n " i s s i m i l a r for both Nunns Creek and Bonsall Creek. Region s t r a t a also account for about 1% of the v a r i a t i o n , i n d i c a t i n g that there i s l i t t l e regional habitat s p e c i f i c v a r i a t i o n within the "dorsal plate c o n s t e l l a t i o n " . S i t e s t r a t a however, account for almost 12% of the v a r i a t i o n . The "dorsal plate c o n s t e l l a t i o n " therefore shows s i t e s p e c i f i c v a r i a t i o n . Time s t r a t a account for less than one percent of the v a r i a t i o n i n t h i s axis. Therefore the "dorsal plate c o n s t e l l a t i o n " i s not dependent on c o l l e c t i o n period. This r e s u l t was expected, since most of the major contributing characters to t h i s axis are presumed to be g e n e t i c a l l y determined and therefore the v a r i a t i o n i n the "dorsal plate c o n s t e l l a t i o n " should not be time dependent. For a l l f i v e axes ( Y l — Y 5 ) , the PCA accounts for 65.16% of the v a r i a t i o n ; but of that 65.16% over h a l f (37.56%) i s not accounted for by the s t r a t a . This means that half of the v a r i a t i o n accounted for by the PCA of data set West i s not explained by my model. Region s t r a t a did however, account for almost 16% of the v a r i a t i o n and s i t e s t r a t a ac-counted for about 9% of the v a r i a t i o n i n the data matrix. Stream s t r a t a only accounted for about 1% of the v a r i a t i o n and time s t r a t a accounted for less than 1%. 63 3.2.3. Duncan's Multiple Range Test (DUNCAN'S) This analysis identifies strata that are different at the 5% probability level. As input i t uses the PCA scores and strata from the previous analyses. I n i t i a l l y , only the results for the stream strata w i l l be examined (Fig. 8). For the f i r s t axis (Yl), stream strata are different. This in-dicates that the characters contributing to the "size constellation" are different between the strata. Bonsall Creek individuals tend to be slightly larger in "size" on the average than those from Nunns Creek. Bonsall Creek also exhibits a wider range in "size" than Nunns Creek. Although one might expect Bonsall Creek to show a greater variability in the "size constellation" due to a larger sample size (Bonsall Creek N=321, Nunns Creek N=138), the significant difference in the values for the "size constellation" between the creeks was not expected. For the second axis (Y2), stream strata are different. This i n -dicates that the characters contributing to the "form constellation" are different between the strata. Nunns Creek on the average has more "leiurus"-like individuals, tending to have fewer lateral plates, fewer g i l l rakers, deeper but shorter caudal peduncles, fewer dorsal rays and shorter ventral spines than those of Bonsall Creek. This may be due in part to the smaller amount of marine regional habitat present at the mouth of Nunns Creek, but since almost one third (N=42 of 138) of the Nunns Creek specimens were from the marine region, this cannot be the only explanation. Bonsall Creek exhibits a wider range of variation in the "form constellation", possibly due to a larger sample size (Bonsall Creek N=321, Nunns Creek=138). 64 Figure 8. Component loadings for Bonsall Creek and Nunns Creek for each PCA axis (Yl—Y6) for data set West (means, standard deviations and sig-nificance at p=0.05 indicated). ? -6 -5 -4 -3 -2 -I -0 --| - - -O 0. -2 -SE O O - 3 --4 -- 5 -- 6 -- ? -I I I I + i i i i I Yl Y3. Y2 -I = Y4 I +_ I Y5 Bonsall Creek Nunns Creek PCA Axis 6 6 For the t h i r d axis (Y3), stream strata are d i f f e r e n t . The characters contributing to the "sexual constellation" therefore d i f f e r between the strata. This means that Bonsall Creek specimens on the average tend to be more "male"-like, with more individuals showing red throat pigmentation, shorter less deep abdominal dimensions, longer jaws and snouts, fewer caudal vertebrae, fewer anal rays, and less deep bodies than Nunns Creek specimens. Both strata however, are equally variable i n the observed range of component loadings. This may be a function of a s l i g h t l y skewed sex r a t i o or of more obvious sexual dimor-phism (seen as more pronounced male secondary sexual characteristics) i n Bonsall Creek than Nunns Creek. For the fourth axis (Y4), stream strata are d i f f e r e n t . The characters contributing to the "meristic constellation" therefore d i f f e r between the strata. Nunns Creek specimens on the average tend to ex-h i b i t more dorsal and anal pterygiophores and rays, more caudal ver-tebrae, more epurals and shorter anal and dorsal spines than Bonsall Creek specimens. This may be a function of many factors including h i s t o r i c a l constraints, selection, d r i f t and environmentally induced v a r i a t i o n , such as temperature. Both strata are equally variable i n the observed range of component loadings. For the f i f t h axis (Y5), stream strata are not s i g n i f i c a n t l y d i f -ferent. This indicates that the "dorsal plate constellations" are similar for both Bonsall Creek and Nunns Creek. Bonsall Creek does however, show a greater v a r i a b i l i t y i n component loadings probably as a result of the larger sample size (Bonsall Creek N=321, Nunns Creek N=138). 67 For the region s t r a t a , r e c a l l that there were four types of regional habitats within each stream. The following portion of the analysis i d e n t i f i e s s t r a t a within each stream or groups of s t r a t a that are d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . As input i t uses the PCA scores and s t r a t a from the previous analyses (Figs. 9-13). For the f i r s t axis (Yl) ( F i g . 9), there are three d i s t i n c t groups of region s t r a t a . They are: (A) Bonsall Creek regional habitat 1 (BI), (B) Bonsall Creek regional habitat 3 plus Nunns Creek regional habitats 1, 3, 4 (B3, Nl, N3, N4) and (C) Bonsall Creek regional habitats 2, 4 plus Nunns Creek regional habitats 1—4 (B2, B4, Nl, N2, N3, N4). This indicates that the " s i z e c o n s t e l l a t i o n " d i f f e r s between each of these three groups. The f i r s t group, BI, i s c l e a r l y larger i n " s i z e " than eit h e r of the other two groups. The i n t e r p r e t a t i o n of the other two groups i s more d i f f i c u l t , since they share some regional habitats i n common (Nl, N3, N4). This could be interpreted as the specimens from regional habitats Nl, N3 and N4 being s p l i t into two classes i n terms of t h e i r respective values for the "size c o n s t e l l a t i o n " . The second group (B3, Nl, N3, N4), then, contains a l l the i n d i v i d u a l s which tend to be smaller i n " s i z e " . The t h i r d group (B2, B4, Nl, N2, N3, N4) i s intermediate i n " s i z e " when contrasted with the f i r s t two. Regional habitats Bl and B4 exhibit the greatest v a r i a b i l i t y i n the component loadings for t h i s axis.. Although one might expect those i n d i v i d u a l s found i n estuarine regional habitats (Bl and Nl) to be larger than those found i n freshwaters (B2;—B4, N2—N4), (since they would generally be described as "trachurus"); only Bonsall Creek regional habitat 1 specimens were s i g n i f i c a n t l y l a r g e r . 68 Figure 9. Component loadings for regional habitat strata for PCA axis Yl for data set West (means and standard deviations indicated). 69 o 5 < o U l o a. 2 O o 14 -13 -12 -II -10 -9 -8 -7 -6 -5 -4 -3 -2 -I -0 - -- I --2 -- 3 --4 --5 --6 --? --8 -BI N4 N2 B2 Nl B3 N3 B4 R E G I O N A L H A B I T A T S T R A T U M 70 For the second axis (Y2) ( F i g . 10), there are four d i s t i n c t groups of region s t r a t a . They are: (A) Bonsall Creek regional habitat 1 ( B l ) , (B) Nunns Creek regional habitat 1 (Nl ) , (C) Nunns Creek regional habitats 2—4 (N2, N3, N4) and (D) Bonsall Creek regional habitats 2—4 (B2, B3, B4). This suggests that the "form c o n s t e l l a t i o n " d i f f e r s between each of these four groups. The f i r s t group (Bl) i s c l e a r l y the most "trachurus"-like of a l l the groups and the second group (Nl) i s s l i g h t l y less "trachurus"-like. Although Bl and Nl might be expected to be equally "trachurus"-like, since they are both estuarine regional habitats, the r e s u l t that Bl specimens are d i s t i n c t from Nl specimens requires another explanation. It i s possible that there i s environmen-t a l influence on the "form" that the Nl specimens are generally less "trachurus"-like because the region they occupy i s smaller and more con-tained by the stream banks (even at high t i d e ) and i s therefore less estuarine than the Bl regional habitat; or more l i k e l y , that they are separate populations. Of the remaining groups, specimens from B2, B3 and B4 are the most " l e i u r u s " - l i k e i n the "form c o n s t e l l a t i o n " . Again, although B2—B4 and N2—N4 could be expected to be equally " l e i u r u s -l i k e , since they are freshwater regional habitats, the r e s u l t that specimens from B2—B4 are d i s t i n c t from N2—N4 specimens requires another explanation. It i s again possible that there i s environmental induction of "form" that the B2—B4 specimens are more " l e i u r u s " - l i k e because the region they occupy i s larger and often more pond-like than the N2 and N3 regional habitats, but the N4 regional habitat i s very much " l e i u r u s " habitat a large beaver pond. An h i s t o r i c a l explanation for t h i s e n t i r e pattern might be that each creek has i t s own d i s t i n c t i v e 71 Figure 10. Component loadings for regional habitat strata for PCA axis Y2 for data set West (means and standard deviations indicated). R E G I O N A L H A B I T A T S T R A T U M 73 form and that both of the so-called " l e i u r u s " and "trachurus" forms are d i s t i n c t i n each creek because they arose from d i f f e r e n t stocks, or at d i f f e r e n t times. The variances are s i m i l a r f o r a l l region s t r a t a f o r th i s a x i s . For the t h i r d axis (Y3) ( F i g . 11), there are two d i s t i n c t groups of region s t r a t a . They are: (A) Bonsall Creek regional habitats 3 and 4 plus Nunns Creek regional habitats 1—4 (B3, B4, Nl, N2, N3, N4) and (B) Bonsall Creek regional habitats 1—4 plus Nunns Creek regional habitats 2 and 4 (BI, B2, B3, B4, N2, N4). This indicates that the "sexual con-s t e l l a t i o n " d i f f e r s between these two groups. The i n t e r p r e t a t i o n of t h i s result i s d i f f i c u l t , since they share groups of region s t r a t a . This might be interpreted as the specimens from regional habitats B3, B4, N2 and N4 being s p l i t i nto two classes i n terms of t h e i r respective values for the "sexual c o n s t e l l a t i o n " . The second group (BI, B2, B3, B4, N2, N4) contain the i n d i v i d u a l s which appear to be more "male"-like than the f i r s t group (B3, B4, Nl, N2, N3, N4). This r e s u l t may be a function of a s l i g h t l y skewed sex r a t i o or the r e s u l t of more obvious sexual dimorphism i n the secondary sexual c h a r a c t e r i s t i c s of the second group, but the pattern i s unclear. The variances of a l l the region s t r a t a are s i m i l a r for t h i s a x is. For the fourth axis (Y4) ( F i g . 12), there are four d i s t i n c t groups of region s t r a t a . They are: (A) Bonsall Creek regional habitats 1—3 (BI, B2, B3), (B) Nunns Creek regional habitats 1, 2, and 4 (Nl, N2, N4), (C) Nunns Creek regional habitats 2—4 (N2, N3, N4) and (D) Bonsall Creek regional habitat 4 plus Nunns Creek regional habitats 3 and 4 (B4, N3, N4). The "meristic c o n s t e l l a t i o n " therefore d i f f e r s between these 74 Figure 11. Component loadings for regional habitat strata for PCA axis Y3 for data set West (means and standard deviations indicated). B2 B 3 B4 Nl R E G I O N A L H A B I T A T S T R A T U M 76 Figure 12. Component loadings for regional habitat strata for PCA axis Y4 for data set West (means and standard deviations indicated). REGIONAL HABITAT S T R A T U M 78 four groups. The f i r s t group; B l , B2 and B3, consists of those specimens with the lowest values on the average f or the "meristic c o n s t e l l a t i o n " fewer dorsal and anal pterygiophores and rays, fewer caudal vertebrae, fewer epurals and longer dorsal and anal spines than the specimens from the other regions. The i n t e r p r e t a t i o n of the other three groups i s more d i f f i c u l t , since a l l three share N4 i n common, while two share N2, and another pair share N3. This could be i n t e r -preted as the specimens from regional habitat N4 being s p l i t i n t o three classes and N2 and N3 being s p l i t into two classes, i n terms of t h e i r respective values for the "meristic c o n s t e l l a t i o n " . The second (Nl, N2, N4) and t h i r d (N2, N3, N4) groups, then contain the in d i v i d u a l s which tend to exhibit just lower and just higher than the mean for the "meristic c o n s t e l l a t i o n " they are both intermediate groups. The fourth group (B4, N3, N4) consists of a l l the in d i v i d u a l s which tend to show the highest values for the "meristic c o n s t e l l a t i o n " . The variances for a l l of the region s t r a t a are s i m i l a r for t h i s a xis. I t i s important to note that B l , B2 and B3 are not only d i s t i n c t from a l l of the Nunns Creek regional habitats (Nl, N2, N3, N4), but are d i s t i n c t from Bonsall Creek regional habitat 4 (B4). That B4 i s more s i m i l a r to the Nunns Creek regional habitats may be a re s u l t of d r i f t , environmental induc-t i o n , l o c a l adaptation ( s e l e c t i o n ) , or a common hist o r y ; although the l a s t p o s s i b i l i t y i s the least l i k e l y . For the f i f t h axis (Y5) ( F i g . 13), none of the region s t r a t a or groups of region s t r a t a are s t a t i s t i c a l l y d i s t i n c t . This indicates that the magnitudes of the characters contributing to the "dorsal plate con-s t e l l a t i o n " are s i m i l a r for a l l regions. Some regions ( B l , B2, B4 and 79 Figure 13. Component loadings for regional habitat s t r a t a for PCA axis Y5 for data set West (means and standard deviations i n d i c a t e d ) . 80 R E G I O N A L H A B I T A T S T R A T U M 81 N3) however, do show a greater variability in component loadings indicating that some areas of both creeks have far more variation in the "dorsal plate constellaton" than others. This may indicate potentially discrete populations or demes within each stream. For the site strata, recall that there are 43 strata Bonsall Creek has 31, Nunns Creek has 12. Also recall that DUNCAN'S provided only the means and standard deviations for each stratum since there were too many means to do a range test. These values are shown in Figs. 14, 18, 22, 23 and 24; but s t a t i s t i c a l significance could not be established. For the f i r s t axis (Yl), there seems to be three patterns of note (Fig. 14). F i r s t , for the Bonsall Creek Sites (1—31) there appears to be a general decrease and then leveling out of the mean component loadings for the "size constellation", with a slight increase in mean component loadings in the headwater portions of the stream. This pat-tern may indicate c l i n a l variation in characters contributing to the "size constellation". Second, Bonsall Creek site strata 1, 3, 14, 29, 30 and 31 show the greatest amount of variation in this axis. This result, that some of the estuarine sites and the headwater sites exhibit the largest range of variation, was unexpected. Third, the variation in the "size constellation" for Nunns Creek appears to be similar throughout i t s length, with the possible exception of site strata 32, 33 and 41. It is surprising that the most estuarine site strata of Nunns Creek (32 and 33) should show the lowest average component loadings for this axis (especially when the more estuarine sites of Bon-s a l l Creek show the highest average component loadings). 82 Figure 14. Component loadings f o r s i t e s t r a t a f o r PCA axis Y l for data set West (means and standard deviations i n d i c a t e d ) . COMPONENT LOADING o o o a ) ^ r i o r o A o > o D o r o * ( ^ I I I l I I I I 1 I I I 1 I 84 The p o s s i b i l i t y of c l i n a l v a r i a t i o n i n the f i r s t axis (Yl) was ex-amined i n greater d e t a i l . The three characters which were the greatest contributors to t h i s axis (characters 1, 23, 21) should show c l i n a l v a r i a t i o n i f t h i s pattern i s indeed present. For each s i t e stratum the mean, standard deviation and range of characters 1, 23 and 21 are shown (Figs. 15-17). For character 1 (standard length); there remains a s l i g h t p o s s i b i l i t y of c l i n a l v a r i a t i o n , although the pattern i s far less obvious than f o r the e n t i r e "size c o n s t e l l a t i o n " . The same i s true f o r characters 23 (depth from DII to ECTO) and 21 (depth from DSO to VI). The l i k e l i h o o d of the f i r s t axis (Yl) showing c l i n a l v a r i a t i o n i s reduced, since those characters which contributed the most to t h i s axis do not obviously show c l i n a l v a r i a t i o n . For the second axis (Y2), there seems to be three patterns of note ( F i g . 18). F i r s t , for both Bonsall and Nunns Creeks there appears to be a general increase and then l e v e l i n g out of the mean component loadings for the "form c o n s t e l l a t i o n " . This may in d i c a t e c l i n a l v a r i a -t i o n i n the magnitude of the characters contributing to "form c o n s t e l l a -t i o n " . Second, Bonsall Creek s i t e s t r a t a 12—14, exhibit higher compo-nent loadings than the s i t e s t r a t a before or a f t e r them. This "over-shooting" of the mean component loading may indicate character displace-ment or a reverse step c l i n e . Third, the range of v a r i a t i o n exhibited by Nunns Creek i s smaller than that of Bonsall Creek and tends to be lower i n value (or more "trachurus"-like) o v e r a l l . The p o s s i b i l i t y of c l i n a l v a r i a t i o n i n the second axis (Y2) was also examined i n greater d e t a i l . Again, the three characters which were the greatest contributors to t h i s axis (characters 67, 71, 73) should Figure 15. Distribution of character 1 (standard length) for each site stratum for data set West (means, standard deviations and ranges i n -dicated). STANDARD LENGTH (CHARACTER I) A 98 Figure 16. Distribution of character 23 (depth from DII to ECTO) for each site stratum for data set West (means, standard deviations and ranges indicated). D I I TO ECTO (CHARACTER 2 3 ) 88 Figure 17. Distribution of character 21 (depth from DSO to VI) for each site stratum for data set West (means, standard deviations and ranges indicated). 06 91 Figure 18. Component loadings for site strata for PCA axis Y2 for data set West (means and standard deviations indicated). COMPONENT LOADING l _ _ L _ CJ1 _J_ _1_ _ l _ ro — o — ro _l l_ 01 I I U l CD I I oo H m w CD ro 0 0 H TO > ro 01 ro |_ CD 93 show c l i n a l variation i f this pattern i s indeed present. For each site stratum the mean, standard deviation and range of characters 67, 71 and 73 are shown (Figs. 19-21). For character 67 (number of lateral plates on right side) step-clinal variation is very apparent for Bonsall Creek site strata; while for Nunns Creek, smooth cl i n a l variation appears to be present. That i s , for Bonsall Creek site strata, although the range for any particular site stratum may be broad; the mean number of lateral plates on the right side i s either high (greater than 28) or low (less than 11) with only one site stratum (8) exhibiting intermediate values. For Nunns Creek however, most of the means for character 67 are inter-mediate and they generally decrease in value. For character 71 (number of lateral plates between ascending process and last precaudal vertebra) and character 73 (number of lateral plates on caudal peduncle) the above pattern holds in both cases. The hypothesis that the second axis (Y2) shows c l i n a l variation i s strengthened, since those characters which contribute most to this axis obviously show c l i n a l variation. Further, note that the pattern of an "overshooting" of the average value for site strata 9—13 is again present for character 67 and to a lesser extent for character 71. It should also be noted that the pattern of c l i n a l variation i s strikingly different between Nunns Creek and Bonsall Creek. For the third axis (Y3), there appears to be no obvious patterns (Fig. 22). A l l but two site strata (2 and 8) are very close to the average component loadings for the "sexual constellation". This may i n -dicate that while each site stratum may vary somewhat from a 1:1 ratio, few deviate strikingly. Figure 19. Distribution of character 67 (number of lateral plates on right side) for each site stratum for data set West (means, standard deviations and ranges indicated). NUMBER OF P L A T E S (CHARACTER 6 7 ) S6 Figure 20. Distribution of character 71 (number of lateral plates between ascending process and last precaudal vertebra) for each site stratum for data set West (means, standard deviations and ranges i n -dicated) . 98 Figure 21. Distribution of character 73 (number of lateral plates on caudal peduncle) for each site stratum for data set West (means, stan-dard deviations and ranges indicated). NUMBER OF P L A T E S (CHARACTER 7 3 ) 100 Figure 22. Component loadings for site strata for PCA axis Y3 for data set West (means and standard deviations indicated). COMPONENT LOADING ' I I I _ * w ro — o I l I I — ro 01 ^ ui cn J I I L I I cs 1 CD ro cn H m io -ro 00 H TO ro ex CD Ol ro 1 (0 TT T CD u CM ' T Ol A. T Ol <0 ro i _ z i t 01 TOT 102 For the fourth axis (Y4), there seems to be two patterns of note ( F i g . 23). F i r s t , f o r Bonsall Creek' there are two areas with higher than average mean component loadings f o r the "meristic c o n s t e l l a t i o n " s i t e s t r a t a 10 plus 11 and s i t e s t r a t a 29 plus 30 and 31. This may indi c a t e that there are d i s c r e t e demes or populations on a scale of two to three s i t e s t r a t a . Second, f o r Nunns Creek the compo-nent loadings for the "meristic c o n s t e l l a t i o n " are generally con-siderably larger than those for Bonsall Creek. This implies that Nunns Creek and Bonsall Creek are discrete systems i n terms of v a r i a t i o n i n the characters contributing to the "meristic c o n s t e l l a t i o n " . For the f i f t h axis (Y5), there seems to be only a single pattern of note ( F i g . 24). That i s ; while most s i t e s t r a t a vary l i t t l e i n the "dorsal plate c o n s t e l l a t i o n " , a few vary a great deal s i t e s t r a t a 6, 8, 14, 15, 30 and 40. Note that t h i s v a r i a b i l i t y i s present i n both streams, but that i t i s only once found i n adjacent s i t e s t r a t a . This may indicate that there are d i s c r e t e demes or populations on a scale of one to two s i t e s t r a t a . For the c o l l e c t i o n period s t r a t a , r e c a l l that there were two c o l -l e c t i o n periods f o r each stream. The following portion of the DUNCAN'S analysis i d e n t i f i e s s t r a t a that are s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . As input i t uses the PCA scores and s t r a t a from the previous analyses ( F i g . 25). For the f i r s t axis ( Y l ) , c o l l e c t i o n period s t r a t a are not s i g -n i f i c a n t l y d i f f e r e n t . This indicates that the characters contributing to the "size c o n s t e l l a t i o n " are s i m i l a r for both the e a r l y June and.late J u l y c o l l e c t i o n periods. The variances for both c o l l e c t i o n periods are 103 Figure 23. Component loadings for site strata for PCA axis Y4 for data set West (means and standard deviations indicated). 104 to ro T CM Z J. T z J . I I IO IO I I IO 0 0 0 ) ' CM I O CQ 1 CM | — 1 — 8 i 1 1 IO 2 r -CM m UJ CO r "~r~ - i 1 r — o — - r CM 1 ^ r o 1 ^ CM r o ONIQVOn lN3N0dW0Q 105 Figure 24. Component loadings for site strata for PCA axis Y5 for data set West (means and standard deviations indicated). 9NIQV01 !N3NOdWO0 107 Figure 25. Component loadings for time strata for a l l PCA axes (Yl—Y5) for data set West (means, standard deviations and significance at p=0.05 indicated). 108 ? - i 6 -5 -4 -3 -o 5 2-Q < °. I -0 ---I --2 --3 --4 --5 --6 --? -Ui O Q. 2 O O I I I I I I I T i i i i i i Y. I I I -T-I Y2 I I + " i # Y3 Y4 I Y5 early June late July PCA Axis 109 also similar. This may mean that there is a stable age distribution for these populations, since (3. aculeatus tends to show indeterminate growth. The slightly lower mean component loadings for the late July collection may reflect the recruitment of young of the year. For the second axis (Y2), collection period strata are different. This indicates that there are differences in the "form constellation" between early June and late July. The variances for both collection periods however, are similar. Individuals caught in late July may be more "leiurus"-like in form due to the oceanward movement of the "trachurus"-like individuals in mid-summer (Hagen,1967; McPhail, pers. comm.). For the third axis (Y3), collection period strata are different. This indicates that there are differences in the "sexual constellation" between early June and late July. The variances for both collection periods are similar. This indicates that the specimens collected in late July are more "male"- like in appearance. Most likely, this is due to more males becoming reproductively mature in mid-summer. For the fourth axis (Y4), collection period strata are different. This indicates that there are differences in the "meristic constella-tion" between early June and late July. The variances for both collec-tion periods are similar. This result, that the individuals of the early June collection on the average tend to have more dorsal and anal pterygiophores and rays, more caudal vertebrae, more epurals and shorter anal and dorsal spines than the late July collection may be due to developmental differences between adults and young of the year, recruited by late July. 110 For the f i f t h axis (Y5), collection period strata are not sig-nificantly different. This indicates that the magnitude of the characters contributing to the "dorsal plate constellation" are similar for both the early June and late July collection periods. The variances for both collection periods are also similar. I l l CHAPTER 4. DISCUSSION The merit of th i s thesis i s twofold; i t includes a technique for pattern determination an examination of the pattern of v a r i a t i o n of sticklebacks i n two natural streams. 4.1. Technique The technique for pattern determination used i n t h i s thesis con-s i s t e d of the following: f i r s t ; a P r i n c i p l e Components Analysis (PCA) was used to group together the characters (or a t t r i b u t e s of the or-ganisms, not necessa r i l y morphological) which covary (whether d i r e c t l y or inversely) along a single a x i s . This group of characters, or character c o n s t e l l a t i o n , can be seen as a summary of the v a r i a t i o n most important for that a x i s . Each axis i s s t a t i s t i c a l l y independent from a l l other possible axes and therefore allows (and may i n f a c t , require) a m u l t i p l i c i t y of patterns. When used i n th i s manner, as a c l u s t e r i n g technique and not f o r phylogenetic or other evolutionary i n t e r p r e t a -t i o n s , PCA i s e s s e n t i a l l y free of expectations and biases i t simply summarizes the v a r i a t i o n i n the o r i g i n a l data matrix along a var i e t y of independent axes. Second; a Multiple Nested Analysis of Variance (ANOVA) was used to apportion the v a r i a t i o n summarized by each PCA axis according to a_ p r i o r i groupings (or strata) expected on some b i o l o g i c a l grounds to a f f e c t the pattern of v a r i a t i o n i n the o r i g i n a l data matrix ( s t a t i s t i c a l population). In th i s way the r e l a t i v e contribution of various factors can be examined e x p l i c i t l y and therefore more c l e a r l y . At t h i s stage i n the analysis, i t becomes possible to discuss some of 112 the potential causes of the patterns of variation. Third; a Duncan's Multiple Range Test (DUNCAN'S) was used to determine which strata or groups of strata were different from one another. In this way the pat-tern of variation in the original data matrix is further dissected by examining in detail the pattern of variation between strata for each PCA axis. This stage of the analysis allows greater understanding of the patterns of variation and therefore, a greater understanding of the potential proximal causes of the perceived patterns. This technique for pattern determination could be used profitably in a variety of studies, whenever the pattern of variation or var i a b i l i t y i s unclear. This is especially true for speciose groups as well as single species that show a great deal of variation. For ex-ample, the pattern of variation within and between morphotype pairs of G_. aculeatus in many Pacific West Coast lakes is confusing at best (J.D.McPhail, pers.comm.). Use of the above technique would allow: (1) a determination of the characters whose variation is significant, (2) a determination of the characters which covary (the character constella-tions), (3) an assessment of the relative importance of habitat d i f -ferences, drainage basins, time since isolation, or other strata, to the variation in each character constellation and (4) given that any of the strata are indeed important, an assessment of the pattern of variation within those strata. This technique might also be of use in the description and detection of salmonid stocks or coregonids of the Great Lakes. Character constellations provided by the PCA could summarize the type or types of variation in the included s t a t i s t i c a l populations. ANOVA analysis could then determine the relative importance of habitat, 113 drainage basin, year of major returns or spawning, presence or absence of other species, water quality or other strata. Finally a DUNCAN'S analysis could provide an in depth analysis of potential differences within strata, such as drainage basins, etc. 4.2. Variation The pattern of variation in natural stream populations is part of a far larger pattern that of the variability and variation in a l l l i f e forms. A l l known species are variable in at least one aspect, although not a l l species vary to the same extent. The ultimate source of varia-tion and variability l i e s in the genetic information coded in the DNA. The expression of this information in natural populations however, is constrained at three levels; epigenetic interaction within the nucleus and between the nucleus and cytoplasm, regulation through developmental programs and pathways, and elimination of individuals through selection and other environmentally mediated processes. This means that the variability and variation observed in natural populations is only a por-tion of the variability and variation present for each species. It is the variation and variability present for each species which is the material basis for evolution. It should be noted that not only are variation and vari a b i l i t y qualities of a l l natural groups but also that natural groups can only be detected on the basis of the pattern of variation for each natural group (Wiley,1981). If researchers wish to document the entire spectrum of variation, (which would greatly f a c i l i t a t e our understanding of natural groups and therefore evolutionary patterns and processes) i t is necessary and most 114 informative to sequentially i s o l a t e the pattern of v a r i a t i o n at each l e v e l of generality, from a l l of l i f e to the i n d i v i d u a l . This requires the methodology and co-operation of systematists to provide phylogenetic analyses (a type of pattern analysis) from the l e v e l of kingdoms to the l e v e l of species and population b i o l o g i s t s to provide the pattern analysis from the l e v e l of subspecies or populations to the l e v e l of i n d i v i d u a l s . Many studies however, only attempt to examine the pattern of v a r i a t i o n at a single l e v e l without regard f o r any others. For example; i n a study of the v a r i a t i o n at the population l e v e l those characters which are best placed as variables i n the v a r i a t i o n at the next higher l e v e l (species) w i l l be perceived as constants, while those characters which are best placed as variables i n the v a r i a t i o n at a lower l e v e l (the deme or the i n d i v i d u a l ) w i l l most l i k e l y be perceived as "noise" or random, unstructured v a r i a t i o n . This r e s u l t s i n a loss of understanding of the pattern of v a r i a t i o n . These patterns are the only basis for detecting natural groups. Our understanding of both the proximal and ultimate causes of evolution and the nature of v a r i a t i o n ultimately depends on our a b i l i t y to detect these natural groups. 4.2.1. V a r i a t i o n at the Species Level In t h i s study, since the pattern of v a r i a t i o n from the l e v e l of the populations to the l e v e l of i n d i v i d u a l s i s of prime i n t e r e s t ; i t i s therefore necessary to take a b r i e f look at the pattern of v a r i a t i o n at the next higher l e v e l the species. In th i s way the documented pat-terns of v a r i a t i o n can be placed at the appropriate l e v e l , reducing the number of necessary explanations or generalizations for those patterns. The specific answer to the philosophical question, "What is a species?", ultimately depends on the perceived pattern of variation between species in nature. That perception however, is drastically af-fected by the philosophical background of the researcher asking the question. The discussion of what a species i s , has been long and acrimonious (for example; Ehrlich,1961; Ghiselin,1974; Hull,1976; Mayr,1963; Simpson,1961; Sokal,1973; Wiley,1978,1980). Most biologists prefer either the "biological species" concept (championed by Ernst Mayr) or the "evolutionary species" concept (championed by G.G.Simpson and later Ed Wiley). Mayr (1963) defines a species as "groups of inter-breeding populations that are reproductively isolated from other such groups" (ibid, p. 12). A researcher using this definition as a tool would detect species by looking for disjunctions in the reproductive linkages between populations; evidenced by either behavioral disjunc-tions or disjunctions in quantifiable, presumably inherited charac-teri s t i c s of the populations in question. There are numerous operational d i f f i c u l t i e s with the application of this tool to the problem of recognizing species in nature. For example; hybridization between organisms in different genera or subfamilies (Hubbs, 1955) i s an intractable problem when applying this tool. Application may also depend on arbitrary decisions of interbreeding or i n t e r s t e r i l i t y of lab or natural populations and is therefore suspect. Finally, there is a good possibility that there may be more gene flow between otherwise very distinct species than was previously believed (White,1978). Wiley (1978) defines a species as "a single lineage of ancestral descendant populations of organisms which maintains i t s identity from other such 116 lineages and which has i t s own evolutionary tendencies and h i s t o r i c a l f a t e " ( i b i d , p. 18). Operationally, a researcher using t h i s d e f i n i t i o n would detect species by looking for c h a r a c t e r i s t i c s which were unique and always present for a single lineage autapomorphies. A p p l i c a t i o n of t h i s technique then, requires only a thorough examination of the a t -t r i b u t e s of the groups i n question and the proviso that each species, to be detectable, must have at least one autapomorphic a t t r i b u t e . The evolutionary species concept i s therefore both methodologically and p h i l o s o p h i c a l l y more simple to j u s t i f y (although not without i t s c r i t i c s ( for example; B a l l , 1983; Mishler and Donoghue, 1982; W i l l i s , 1981)) and therefore a stronger t o o l . For t h i s reason, the evolutionary species concept w i l l be used i n further discussions of species. The pattern of v a r i a t i o n which i s s i g n i f i c a n t at the l e v e l of species then, has already been s p e c i f i e d by the choice of the evolutionary species concept. That i s , autapomorphic a t t r i b u t e s are the expected pattern of v a r i a t i o n for characters whose v a r i a t i o n i s impor-tant at the l e v e l of species. In t h i s study the derived states (deter-mined by outgroup comparison) of character 74 (presence or absence of posttemporal) absence of posttemporal and character 75 (presence or absence of supracleithrum) absence of supracleithrum are autapomorphic for G^. wheatlandi. Character 59 (number of v e n t r a l rays) consistently d i f f e r s between G_. aculeatus and G_. wheatlandi but p o l a r i t y could not be determined. Other characters such as 88 (number of pterygiophores supporting D rays), 89 (number of pterygiophores supporting A), and 64 (number of caudal vertebrae); which were included as major contributors to the t h i r d PCA axis (Y3) (the "species c o n s t e l l a t i o n " ) i n the analysis of data set A l l are a l l i e d to the above autapomorphic characters (74 and 75) but are not themselves autapomorphic. These " a l l i e d " characters represent v a r i a t i o n which i s nearly autapomorphic, but s t i l l vary within as well as between species. They may represent c h a r a c t e r i s t i c s which have a higher p r o b a b i l i t y of becoming autapomorphic through the course of continued evolution. 4.2.2. V a r i a t i o n at the Population Level Again the s p e c i f i c answer to the philosophical question, "What i s a population?", u l t i m a t e l y depends on the perceived pattern of v a r i a t i o n between populations i n nature. The p h i l o s o p h i c a l o r i e n t a t i o n of the researcher w i l l d r a s t i c a l l y a f f e c t the perceived pattern of v a r i a t i o n . Populations have been defined i n a v a r i e t y of ways. A S t a t i s t i c a l Population i s "the t o t a l i t y of i n d i v i d u a l observations about which i n -ferences are to be made, ex i s t i n g anywhere i n the world or at least within a d e f i n i t e l y s p e c i f i e d sampling area l i m i t e d i n space and time" (Sokal and Rohlf, 1969, p. 7). If a b i o l o g i s t were to use t h i s c r i t e r i o n , populations would be seen as a r t i f i c i a l l y defined groups. Although a r t i f i c i a l groups can be useful for some purposes, a r e a l un-derstanding of natural patterns and processes can only be achieved f o r natural groups. A B i o l o g i c a l Population has been defined as "a group of conspecific organisms that occupy a more or less well defined geographic region and exhibit reproductive continuity from generation to generation; i t i s generally presumed that e c o l o g i c a l and reproductive interactions are more frequent among these i n d i v i d u a l s than between them and the members 118 of other populations of the same species" (Futuyma, 1979, p. 506). In p r a c t i c e however, most b i o l o g i s t s define the population or populations i n t h e i r studies operationally; that i s , according to habitats or sampling areas which are only presumed to be more i n t e r n a l l y cohesive than some other grouping. This practice ensures that popula-tions are unitary, such that no i n d i v i d u a l i s a member of a two or more groups or populations. Populations however, are probably composed of a hierarchy of nested and p a r t i a l l y nested groups. Behavioral studies commonly use a v a r i e t y of groupings which may be both nested and p a r t i a l l y nested. For ex-ample; an i n d i v i d u a l could conceivably be a member of a p a i r , a family, a male dominance hierarchy, a s o c i a l group and a foraging group. These groupings are determined by the pattern of s o c i a l i n t e r a c t i o n s which are of i n t e r e s t i n behavioral studies. If populations are delineated as genealogical rather than behavioral groups, the expected pattern of groups i s s l i g h t l y d i f f e r e n t . If populations are genealogical groups, one would expect a h i e r a r -chy of f u l l y nested groups but no p a r t i a l l y nested groups. This i s because no i n d i v i d u a l can be simultaneously part of two or more disjunct genealogies. Any i n d i v i d u a l ' s smallest genealogical group may however, be part of a larger, more i n c l u s i v e genealogical group. This may be the best way to envision the pattern of v a r i a t i o n at the l e v e l of demes, subpopulations, populations and subspecies. A deme has been defined as "a population that i s homogeneous with respect to the mixing of genes, but with frequency-dependent f i t n e s s values" (Wilson, 1980, p. 20). A researcher keeping t h i s d e f i n i t i o n i n mind would be forced to undertake 119 a rather extensive examination of the patterns of inheritance and quan-titative genetics of the groups involved. In practice however, most non-theoretical biologists would take a deme to be a somewhat differen-tiated sub-population. This however, is a rather vague use of the term. If populations are hierarchically arranged as genealogical groups, the term deme would take on a more specific operational meaning. A deme could be perceived as the smallest genealogical group showing population level variation (rather than at the level of species or individuals). This hierarchy would lend support to models and experiments concerning such topics as levels of selection. It is necessary however, to deter-mine the pattern or patterns of variation for each level before processes such as selection can be effectively studied. To document the presence of populations and/or subpopulations and/or demes requires a method that would allow the genealogical relationships to be inferred. Observable evidence including actual i n -terbreeding, behavior and f e r t i l i t y , would seem to be the obvious choice to detect discrete populations. This type of evidence however, is ex-pensive and time consuming to collect, often equivocal at best and only worthwhile where other evidence already points to the possibility that the groups in question are not freely interbreeding. In most cases the evidence which points to the possibility of discrete groups is the pat-tern of character variation. The patterns of character variation therefore must r e a l i s t i c a l l y be the f i r s t indicators of populations. The use of characters to infer genealogy requires bridge principles (sensu Hempel, 1965), to link patterns of character variation to pat-terns of genealogy. These bridge principles can be summarized as f o l -120 lows; (1) characters are heritable and (2) the patterns of inheritance are reflected ( a l b e i t imperfectly) i n patterns of character va r i a t i o n . The above principles are almost universally accepted for most mor-phological studies. The expected pattern of variation at the l e v e l of populations i s therefore not one of autapomorphies (which i s indicative of variation at the l e v e l of species) but one of r e l a t i v e l y consistent, discrete v a r i a t i o n , present for a variety of inclusive groups smaller i n geographic range than species. In this project, independent, discrete groups of characters which covary (character constellations) were determined through PCA analysis. These discrete groupings of characters potentially indicate the ex-istence of discrete genealogical groups given the principles l i n k i n g characters to genealogy, above. Since each character constellation i s s t a t i s t i c a l y independent from a l l other character constellations, not a l l character constellations necessarily need to r e f l e c t the same genealogical pattern; i n fact, one might expect a variety of nested pat-terns. I t i s therefore possible to document the pattern of variation at a number of levels of populations, subpopulations and demes; using discrete groups of characters that covary over a variety of geographic areas. The strata (geographic location, stream and regional habitat) used i n the ANOVA and DUNCAN'S analyses are a l l environmentally discrete or r e l a t i v e l y discrete areas. The s i t e strata used i n these analyses however, were operationally defined microgeographic areas. A l l of these strata are pot e n t i a l l y important as markers for the boundaries between populations, but since populations could conceivably be distributed over 121 other gradients than space (for example; behavioral or temporal gradients) they are by no means the sole conceivable markers for popula-tion boundaries. Geographic areas and microgeographic areas however, can be more easily and reasonably described and delimited than portions of other possible gradients. In addition, i f selection is dependent on the environment, then geographic and microgeographic strata could poten-t i a l l y be important in the distribution of populations. Populations could conceivably be detected at the level of geographic location, stream, regional habitat, and/or site as well as groups of streams, groups of regional habitats and/or groups of sites. The analysis of data set A l l revealed that within the species (3. aculeatus, there are distinct populations on both sides of the conti-nent. Eastern and western populations differ, in the magnitudes of the characters contributing to the "size constellation" and "form constella-tion". Due to the asymmetry in the distribution of _G. wheatlandi; i t was not possible to detect either distinct populations within the species G_. wheatlandi, or to determine at what significance level, or for how many axes eastern and western populations of G. aculeatus are distinct. It remains true however, that there are distinct eastern and western populations of (J. aculeatus. This is not a surprising conclu-sion since the distribution of G_. aculeatus includes totally disjunct eastern and western Canadian areas. The analysis of data set West revealed that within the western population of G. aculeatus, the two streams included in this study are distinct populations. Bonsall Creek and Nunns Creek differ in the mag-nitudes of the characters contributing to a l l of the f i r s t four PCA 122 axes. Therefore there are differences i n the "size c o n s t e l l a t i o n " , the "form c o n s t e l l a t i o n " , the "sexual c o n s t e l l a t i o n " and the "meristics con-s t e l l a t i o n " between the two creeks. The populations within each creek should therefore be considered separately. This i s not a s u r p r i s i n g conclusion, since most researchers consider the i n d i v i d u a l s from each drainage basin as p o t e n t i a l l y d i s t i n c t populations. Further analysis of data set West revealed that within the popula-t i o n i n Bonsall Creek, there are at le a s t three d i s t i n c t subpopulations. Within the population i n Nunns Creek, the analysis distinguished between two subpopulations. Almost 16% of the t o t a l v a r i a t i o n i n data set West i s accounted for by these populations, d i s t i n c t at the l e v e l of regional habitats. Within Bonsall Creek, the estuarine regional habitat, B l ; was d i s t i n c t from other regional habitats (B2, B3, and B4) on the basis of the "size c o n s t e l l a t i o n " and the "form c o n s t e l l a t i o n " . The headwaters regional habitat, B4; was d i s t i n c t from the other regional habitats ( B l , B2 and B3) on the basis of the "meristics c o n s t e l l a t i o n " . Within Nunns Creek, the estuarine regional habitat (Nl) was also d i s t i n c t from the other regional habitats (N2, N3 and N4) on the basis of the "form con-s t e l l a t i o n " . This r e s u l t , that the populations occupying estuarine habitats are d i s t i n c t from the populations found i n t h e i r respective drainage basins, was expected based on the extensive l i t e r a t u r e of the so-c a l l e d " l e i u r u s " and "trachurus" forms. However, the conclusion that the populations occupying estuarine habitats d i f f e r between the creeks, was unexpected since the so-called "trachurus" form i s generally believed to be conservative and panmictic throughout i t s range (Bell,1976; Withler,1980; McPhail, pers. comm.). The conclusion that 123 the population found in the headwaters of Bonsall Creek is distinct was suspected prior to analysis, because the culverts through which the creek flows in this region could potentially isolate populations up-stream from those below. It is interesting to note that the culvert between sites 30 and 31 and the culvert between sites 32 and 33 were a l -most certainly installed after the turn of the century. This means that differentiation between these populations may have taken less than 85 years. The analysis of data set West revealed that specific site strata and small groups of site strata show a great deal more variation than the site strata adjacent to them. This site specific variation is representative of about 9% of the total variation in data set West and may be indicative of subpopulations or demes at the level of sites or small microgeographic areas. For the "meristic constellation", site strata 10 plus 11, 29 plus 30 plus 31 and site stratum 33 are distinct. For the "dorsal plate constellation", site strata 6, 8, 14 plus 15, 30 and 40 are distinct. One might conclude from this result that site strata are potentially indicative of variation at the level of demes, since a number of sites are well within the vagility of a single i n -dividual and yet specific sites show site specific variation. This means that there may be significant variation over distances of less than 1/4 mile within regional habitats within creek systems. The im-plications of this result are that studies involving creek populations demand more in-depth sampling when the studies depend on populations and/or the patterns of variation. 124 In sum, populations do show r e l a t i v e l y consistent patterns of v a r i a t i o n detectable at a va r i e t y of l e v e l s . For each of these l e v e l s , the populations involved showed patterns of v a r i a t i o n which were s i m i l a r i n a l l aspects except geographic and microgeographic area. In each case the detected populations were a l l parts of f u l l y nested groups. For th i s reason, these populations are probably genealogical u n i t s . It should also be noted that at each of the analysed geographic and microgeographic l e v e l s , at le a s t one population or subpopulation was d i s t i n c t i v e from broad geographic areas to s i t e s f i v e hundred feet apart. 4.2.3. "Leiurus"and "Trachurus" Forms The patterns of v a r i a t i o n within the genus Gasterosteus have been of i n t e r e s t to b i o l o g i s t s for almost 200 years. I n i t i a l l y , the " l e i u r u s " and "trachurus" forms were considered d i s t i n c t species. Cuvier and Valenciennes i n 1829 described them as two of the ten species comprising Gasterosteus (Bertin, 1925). In 1925, Bertin attempted to summarize a l l the known v a r i a t i o n within the family Gasterosteidae. Un-t i l h i s r e v i s i o n , within the genus Gasterosteus there were over f o r t y species, of which the " l e i u r u s " and "trachurus" forms were but two. Be r t i n considered considered a l l of these to be various forms of a single species. Heuts, i n 1947, was one of the f i r s t researchers to a t -tempt to in t e r p r e t the pattern of v a r i a t i o n for the " l e i u r u s " and "trachurus" forms as part of a single variable species, G_. aculeatus. He suggested that the body forms of in d i v i d u a l s of the species G_. aculeatus were environmentally produced, such that G. aculeatus of the 125 "trachurus" form were produced by the effect of the salt water environ-ment on the expression of the genome. Similarly, G_. aculeatus "leiurus" were produced by the effect of the freshwater environment on the expres-sion of the genome. Although since Heuts' time researchers have discarded the idea that the body forms of (J. aculeatus are environmen-tal l y produced and have discarded the idea that there are more than two living species of Gasterosteus (G. aculeatus and G_. wheatlandi), the idea that within the species aculeatus there are two distinct forms ("leiurus" and "trachurus") has been perpetuated. An important question to ask then, i s : Are the "leiurus" and "trachurus" forms evolutionary entities or are they rather convenienent names l e f t over as an artifact from earlier typological thinking? The pattern of variation within the species G. aculeatus is generally viewed as less complex in Europe and eastern North America than in western North America. Throughout Europe, the pattern does ap-pear to be that the "trachurus" form is usually found in salt or brackish water, while the "leiurus" form is usually found in freshwater. There are a few striking exceptions (Bell,1979; Kynard and Curry,1976; Miinzing,1962,1971,1972; Penczak,1965; McPhail, pers. comm.), but the pattern appears to be quite consistent. In Europe then, the "trachurus" and "leiurus" forms may potentially be evolutionary entities although this may not necessarily be the case in eastern or western North America. The pattern of variation on the eastern coast of North America has not been extensively documented. Interestingly, eastern Canadian populations of £. aculeatus of the "trachurus" form and populations of G. wheatlandi potentially show character displacement and may be a 126 species p a i r . Resolution of t h i s question w i l l await phylogenetic and genetic analyses. In eastern North America then, i t remains a pos-s i b i l i t y that the "trachurus" and " l e i u r u s " forms may be evolutionary e n t i t i e s , but the pattern of v a r i a t i o n i s unclear. The pattern of v a r i a t i o n within G_. aculeatus on the western coast of North America i s far more complex. The general scenario for these populations i s that the marine and brackish water "trachurus" form i s a r e l a t i v e l y conservative, non-variable form comprised of r e l a t i v e l y few widespread panmictic populations and that the " l e i u r u s " form was repeatedly independently derived from the "trachurus" form. One f a i l i n g of t h i s scenario i s that, as seen i n t h i s study, the populations from estuarine regional habitats show the greatest range of v a r i a b i l i t y rather than the l e a s t . Other f a i l i n g s of t h i s scenario include land-locked, freshwater "trachurus"-like populations and the fac t that almost a l l ponds, lakes and streams have t h e i r own d i s t i n c t i v e population or populations which would be considered to be " l e i u r u s " - l i k e . Although most populations could be described as more or less "trachurus"-like or " l e i u r u s " - l i k e , there are p o t e n t i a l l y thousands of d i f f e r e n t i a t e d populations on the western coast of North America. Some of these populations may a c t u a l l y possess true autapomorphies and therefore be d i s t i n c t species Mayer Lake CJ. aculeatus (Moodie,1972; Moodie and Reimchen,1973,1976) may be an example of t h i s . In neither t h i s study nor any other study of v a r i a t i o n i n stream populations of sticklebacks, were populations found which exhibited autapomorphies. For this reason, the groups found i n studies of v a r i a t i o n i n streams are best viewed as v a r i a t i o n at or below the populational l e v e l . On the basis of i n f e r r e d 127 assortative mating, Hagen (1967) did however consider the " l e i u r u s " and "trachurus" forms to be separate species. It seems more reasonable to say that the estuarine population i n the L i t t l e Campbell River was d i s t i n c t from the population or populations upstream. To brand these as G_. trachurus and G. l e i u r u s may lead to confusion over the extent of t h i s l o c a l pattern and i s perhaps excessive. Since both the " l e i u r u s " and "trachurus" populations were found to d i f f e r between streams, t h e i r usefulness as names for evolutionary groups i s severely l i m i t e d . Fur-thermore, i n one area (Bonsall Creek), the " l e i u r u s " form probably con-s i s t s of two d i s t i n c t populations. These terms ("leiurus" and "trachurus") should then be reserved as labels for two extremes of one type of v a r i a t i o n that of body form. In t h i s way they may r e t a i n some meaning for western North American stream populations, since they are not even d i s t i n c t i v e labels f or separate populations. In western North America then, the " l e i u r u s " and "trachurus" forms are not evolutionary e n t i t i e s , but labels for the extremes of a continuum of v a r i a t i o n . 4.2.4. Clines Given that there are two or more populations within a single stream system, there are three p o t e n t i a l patterns of v a r i a t i o n f or those areas where they meet. One p o t e n t i a l pattern i s a smooth c l i n e . This could be the r e s u l t of free s p a t i a l intermixing and/or free interbreeding. In both cases although the means for zones of intermixing or interbreeding for any p a r t i c u l a r c h a r a c t e r i s t i c would be intermediate, the actual com-po s i t i o n of the populations would d i f f e r . If the populations are only s p a t i a l l y intermixing no i n d i v i d u a l s should a c t u a l l y exhibit i n t e r -128 mediate values. If however, the populations are interbreeding, then some individuals w i l l actually exhibit intermediate values for the characteristics differing between the populations. Another potential pattern is a step-cline. This could be the result of contact between populations which interbreed and/or intermix on a very limited scale. It could also be the result of intense selection against intermediates. The f i n a l potential pattern is not c l i n a l , but one of s t r i c t parapatric populations. That i s , populations which meet but neither intermix nor interbreed. In this study, the pattern of variation in each creek differs. In Nunns Creek, the pattern is most easily interpreted as a smooth cline where the populations intermix and interbreed. Individuals with inter-mediate values for the characteristics differing between the populations were collected at most sites. In Bonsall Creek however, the pattern is most easily interpreted as a step cline where the populations intermix over a limited area. There may be a small amount of interbreeding in this area since a very few individuals with intermediate values for characteristics differing between the populations were collected. It should be noted that the population densities are far less in the areas of potential contact between the estuarine and freshwater populations (as estimated by the # specimens caught values in Appendix 1). This may serve to reduce contact and interbreeding by these populations. The pattern of variation in areas of contact is markedly different between the streams studied. 129 4.2.5. Individual V a r i a t i o n Darwin (1859) considered i n d i v i d u a l v a r i a t i o n to be of prime impor-tance i n the course of evolutionary change. Population g e n e t i c i s t s s i m i l a r l y consider i n d i v i d u a l v a r i a t i o n to be important. In most recent studies however, the only i n d i v i d u a l v a r i a t i o n which i s generally recog-nized i s v a r i a t i o n of genes or proteins. There are very few studies of i n d i v i d u a l v a r i a t i o n of morphological c h a r a c t e r i s t i c s . These few studies are those of M. Soule' and his students (Soule', 1967; Soule'eJL a l . , 1973) on asymmetry i n natural populations of normally b i l a t e r a l l y symmetrical species. In his studies, the intent i s not to document the patterns of v a r i a t i o n , but rather to investigate the underlying processes. The one study of the patterns of v a r i a t i o n which includes an analysis for i n d i v i d u a l v a r i a t i o n i s that of Maze (1984 and pers. coram.) a study of v a r i a t i o n i n Douglas f i r . In t h i s study, of the 65.16% of the v a r i a t i o n accounted for by the ANOVA model for data set West, over half (37.56%) of the v a r i a t i o n acounted for was apportioned to the r e s i d u a l . For t h i s model, the ef-fects of v a r i a t i o n associated with streams, regional habitats, s i t e s and time periods are a l l s p e c i f i c a l l y accounted f o r . This leaves the r e s i d u a l to account for the measurement error, the v a r i a t i o n within and between i n d i v i d u a l s ( i n d i v i d u a l v a r i a t i o n ) and any other possible s t r a t a . I t would be reasonable to assume that the measurement error could not be more than 10% of the v a r i a t i o n . This leaves at least 27% of the v a r i a t i o n unaccounted f o r . Since I can not conceive of any other reasonable, possible but not included s t r a t a ; I am forced to i n t e r p r e t t h i s 27% of the t o t a l v a r i a t i o n as i n d i v i d u a l v a r i a t i o n . This can only 130 be a tentative i n t e r p r e t a t i o n since i t was not s p e c i f i c a l l y tested for by my model. To s p e c i f i c a l l y include i n d i v i d u a l v a r i a t i o n i n the ANOVA model i t would be necessary to document the v a r i a t i o n i n measures or counts of repeated parts. This i s somewhat d i f f i c u l t f o r organisms which are b i l a t e r a l l y symmetrical, since the v a r i a t i o n from side to side i s constrained by a v a r i e t y of developmental and functional n e c e s s i t i e s and therefore not always free to express v a r i a t i o n . In plants however, i t can be r e l a t i v e l y simple to document the v a r i a t i o n i n measures or counts between leaves or branches of the same i n d i v i d u a l . This was true for the study by Maze (1984 and pers. comm.) where v a r i a t i o n between i n -d i v i d u a l s accounted for 28.1% of the t o t a l v a r i a t i o n and v a r i a t i o n within i n d i v i d u a l s accounted for 10% of the t o t a l v a r i a t i o n a t o t a l amount s i m i l a r to that for G. aculeatus on the P a c i f i c West Coast. 4.3. Avenues for Future Study The documentation of the patterns of v a r i a t i o n within these two stream systems could provide the groundwork for a v a r i e t y of studies. It i s necessary to document the l o c a t i o n and v a r i a t i o n of populations (as natural groups the e n t i t i e s which take part i n evolutionary processes) before the evolutionary processes producing them can be best investigated. In t h i s way, i t i s possible to i s o l a t e the smallest area necessary and s u f f i c i e n t for i n v e s t i g a t i o n of evolutionary processes. The appropriate areas, locations and characters have been shown for v a r i a t i o n at the l e v e l of populations for Bonsall Creek and Nunns Creek. This could be useful i n many studies, for example, a test of assortative mating i n areas of overlap, evaluation of the stepping stone model of 131 gene flow, or levels of selection acting on specific groups of popula-tions or characters. Another area of research which could profitably be examined is the pattern of variation in more complex stream or river systems. This study could then be seen as a "pilot study" for more com-plex systems. 132 LITERATURE CITED Audet,C; FitzGerald,G.J. and Guderley,H.1985. S a l i n i t y preferences of four sympatric species of sticklebacks (PiscesrGasterosteidae) during t h e i r reproductive season. Copeia 1985:209—213. Avise,J.C.1976. Genetics of plate morphology i n an unusual population of threespine sticklebacks (Gasterosteus aculeatus). Genet.Res.27:33—46. Ball,I.R.1983. On groups, existence and the ordering of nature. Syst.Zool. 32:446—451. Baumgartner,J.V. and Bell,M.A.1984. L a t e r a l plate morph v a r i a t i o n i n the C a l i f o r n i a populations of the threespine stickleback, Gasterosteus aculeatus. Evol.38:665—674. Bell,M.A.1973. Pleistocene threespine sticklebacks, Gasterosteus  aculeatus, (Pisces) from southern C a l i f o r n i a . J.Paleontol.47:479—483. Bell,M.A.1976. Evolution of phenotypic d i v e r s i t y i n Gasterosteus  aculeatus superspecies on the P a c i f i c coast of North America. Syst.Zool.25:211—227. Bell,M.A.1977. A l a t e Miocene marine threespine stickleback, Gasterosteus aculeatus aculeatus, and i t s zoogeographic and evolutionary s i g n i f i c a n c e . Copeia 1977:277—282. Bell,M.A.1979. Low-plate morph of the threespine stickleback breeding i n s a l t water. Copeia 1979:529—533. Bell,M.A.1981. L a t e r a l plate polymorphism and ontogeny of the complete plate morph of threespine sticklebacks (Gasterosteus aculeatus). Evol.35:67—74. Bell,M.A.1982a. D i f f e r e n t i a t i o n of adjacent stream populations of threespine sticklebacks. Evol.36:189—199. Bell,M.A.1982b. Melanism i n a high elevation population of Gasterosteus  aculeatus. Copeia 1982:829—835. 133 Bell,M.A. and Baumgartner,J.V.1984. An unusual population of Gasterosteus aculeatus from Boston, Massachusetts. Copeia 1984:258—262. Bell,M.A. and Richkind,K.E.1981. Clinal variation of lateral plates in threespine stickleback fish. Amer.Nat.117:113—132. Bell,M.A.; Francis,R.C. and Havens,A.C.1985. Pelvic reduction and i t s directional asymmetry in threespine sticklebacks from the Cook In-let Region, Alaska. Copeia 1985:437—444. Bertin,L.1925. Recherches bionomiques, biome<triques et systematiques sur les Epinoches (Gasterosteides). Ann.Inst.Oceanogr.,Monaco.2:1—204. Bookstein,F.L.1978. The measurement of biological shape and shape change. Lecture Notes in Biomathematics 24, Springer-Verlag, New York. 191pp. Campana,S.1983. Factors affecting the production of daily growth incre-ments in the otoliths of fishes. Unpubl. Ph.D. Dissertation, University of British Columbia, Vancouver. Coad,B.W.1974. Vertebral frequencies with notes on anomalies in samples of threespine sticklebacks (Gasterosteus aculeatus L.) from eastern North America. Can.Field Nat.88:220—223. Coad,B.W. and Power,G.1973. Observations on the ecology and phenotypic variation of the threespine stickleback, Gasterosteus aculeatus L., 1758, and the blackspotted stickleback, G.wheatlandi Putnam, 1867, (Osteichthyes: Gasterosteidae) in Amory Cove, Quebec. Can.Field Nat.87:113—122. Coad,B.W. and Power,G.1974. Meristic variation in the threespine stick-leback, Gasterosteus aculeatus, in the Matamek River system, Quebec. J.Fish.Res.Board, Canada.31:1155—1157. Craig,D. and FitzGerald,G.J.1982. Reproductive tactics of four sym-patric sticklebacks (Gasterosteidae). Envir.Biol.Fishes 7:369—375. Cuvier,G. and Valenciennes,A.1829. Histoire Naturelle des Poissons. Paris. 4:459—514. 134 Darwin,CR. 1859. On the Origin of Species by Means of Natural Selec- tion: or the preservation of favoured races in the struggle for l i f e . London, reprinted by Macmillan Publ.Co.Inc., New York. 1962. 512pp. Dingerkus,G. and Uhler,L.D.1977. Enzyme clearing of alcian blue stained whole small vertebrates for demonstration of cartilage. Stain Technology 52:229—232. Ehrlich.P.R.1961. Has the biological species concept outlived i t s use-fulness? Syst.Zool.10:167—176. Futuyma,D.J.1979. Evolutionary Biology. Sinauer Associates, Inc., Sun-derland, Mass. x+565pp. Garside.E.T. and Hamor,T.1973. Meristic variation in the threespined stickleback, Gasterosteus aculeatus L., in eastern Canadian waters. Can.J.Zool.51.547—551. Ghiselin,M.T.1974. A radical solution to the species problem. Syst.Zool. 23:536—544. Gross,H.P. and Anderson,J.M.1984. Geographic variation i n the g i l l -rakers and diet of european threespined sticklebacks, Gasterosteus  aculeatus. Copeia 1984:87—97. Hagen,D.W.1967. Isolating mechanisms in threespine sticklebacks, (Gasterosteus). J.Fish.Res.Board, Canada.24:1637—1692. Hagen,D.W.1973. Inheritance of numbers of lateral plates and g i l l rakers in Gasterosteus aculeatus. Heredity 30:303—312. Hagen,D.W. and Gilbertson,L.G.1972. Geographic variation and environ-mental selection in Gasterosteus aculeatus L. in the Pacific Northwest, America. Evol.26:32—51. Hagen,D.W. and Gilbertson,L.G.1973a. The genetics of plate morphs in freshwater threespine sticklebacks. Heredity 31:75—84. Hagen,D.W. and Gilbertson,L.G.1973b. Selective predation and the inten-sity of selection acting upon the lateral plates of the threespine stickleback. Heredity 30:273—287. 135 Hagen,D.W. and McPhail,J.D.1970. The species problem within Gasterosteus aculeatus on the Pacific coast of North America. J.Fish.Res.Board, Canada.27:147—155. Hagen,D.W. and Moodie,G.E.E.1979. Polymorphism for breeding colors in Gasterosteus aculeatus. I. Their genetics and geographic distribution. Evol.33:641—648. Hagen,D.W. and Moodie,G.E.E.1982. Polymorphism for plate morphs in Gasterosteus aculeatus on the east coast of Canada and an hypothesis for their global distribution. Can.J.Zool.60:1032—1042. Hagen,D.W.; Moodie,G.E.E. and Moodie,P.F.1980. Polymorphism for breeding colors in Gasterosteus aculeatus: (2) Reproductive suc-cess as a result of convergence for threat display. Evol.34:1050—1059. Hempel,C.G.1965. A s p e c t s o f S c i e n t i f i c E x p l a n a t i o n . F r e e P r e s s , New York. Heuts,M.J.1946. Physiological isolating mechanisms and selection within the species Gasterosteus aculeatus L. Nature, London.158:839—840. Heuts,M.J.1947a. Experimental studies on adaptive evolution in Gasterosteus aculeatus L. Evol.1:89—102. Heuts,M.J.1947b. The phenotypical variability of Gasterosteus aculeatus (L.) populations in Belgium. Its bearing on the general geographical va r i a b i l i t y of the species. Verhandelingen van de Koninklijke Vlaamse Academie voor Wetenschappen Letteren et Schone Kunston van Belgie Klasse der Wetenschappen 9:1—63. Heuts,M.J.1949. Racial divergence in f i n ray variation patterns in Gasterosteus aculeatus. J.Genet.49:183—191. Hubbs,C.L.1929. The Atlantic american species of the fish genus Gasterosteus. Occ.Pap.Mus.Zool.Univ.Michigan.200:1—14. Hubbs,C.L.1955. Hybridization between fish species in nature. Syst.Zool.4:1—20. 136 Hubbs,C.L. and Lagler,K.F.1958. Fishes of the Great Lakes Region. Cranbrook Institute of Science, Bulletin 26, xi+213pp. Hull,D.L.1976. Are species really individuals? Syst.Zool.25:174—191. Humphries,J.M.; Bookstein,F.L.; Chernoff.B.; Smith,G.R.; Elder,R.L. and Poss,S.G.1981. Multivariate discrimination by shape in relation to size. Syst.Zool.30:291—308. Kynard,B.E.1978. Breeding behavior of a lacustrine population of threespine sticklebacks (Gasterosteus aculeatus L. ). Behaviour 67:178—207. Kynard,B.E.1979a. Nest habitat preference of low plate number morphs in threespine sticklebacks (Gasterosteus aculeatus). Copeia 1979:525—528. Kynard.B.E.1979b. Population decline and change in frequencies of lateral plates in threespine sticklebacks (Gasterosteus aculeatus). Copeia 1979:635—638. Kynard,B.E. and Curry,K.1976. Meristic variation in the threespine stickleback, Gasterosteus aculeatus, from Auke Lake, Alaska. Copeia 1976:811—813. Lindsey,C.C.1962. Experimental study of meristic variation in a popula-tion of threespine sticklebacks, Gasterosteus aculeatus. Can.J.Zool.40:271—312. Linnaeus,C.1758. Systema Nature. Regnum animale. ed.10 Upsala. 295—297. MacLean,J.1980. Ecological genetics of threespine sticklebacks in Heisholt Lake. Can.J.Zool.58:2026—2039. Mayr,E.1963. Populations, Species and Evolution. Belknap Press, Har-vard University, Cambridge, Mass. xviii+453pp. Maze,J.1984. A study of the factors affecting morphological differen-tiation in a population of Pinus ponderosa. Can.J.Bot.62:403—407. McPhail,J.D.1969. Predation and the evolution of a stickleback (Gasterosteus). J.Fish.Res.Board, Canada.26:3183—3208. McPhail,J.D.1977. Inherited interpopulation differences i n s i z e at f i r s t reproduction i n threespine stickleback, Gasterosteus  aculeatus L. Heredity 38:53—60. McPhail,J.D.1984. Ecology and evolution of sympatric sticklebacks (Gasterosteus): morphological and genetic evidence for a species pair i n Enos Lake, B r i t i s h Columbia. Can.J.Zool.62:1042—1048. McPhail,J.D. and Hay,D.E.1983. Differences i n male courtship i n fresh water and marine sticklebacks (Gasterosteus aculeatus). Can.J.Zool.61: 292—297. Miller,R.R. and Hubbs,C.L.1969. Systematics of Gasterosteus aculeatus with p a r t i c u l a r reference to intergradation and introgression along the P a c i f i c coast of North America: a commentary on a recent con-t r i b u t i o n . Copeia 1969:52—69. Mishler,B.D. and Donoghue.M.J.1982. Species concepts: a case for pluralism. Syst.Zool.31:491—503. Moodie,G.E.E.1972a. Morphology, l i f e h i s t o r y and ecology of an unusua stickleback (Gasterosteus aculeatus) i n the Queen Charlotte I s -lands, Canada. Can.J.Zool.50:721—732. Moodie,G.E.E.1972b. Predation, natural s e l e c t i o n and adaptation i n an unusual threespine stickleback. Heredity 28:155—167. Moodie,G.E.E. and Reimchen,T.E.1973. Endemism and conservation of sticklebacks i n the Queen Charlotte Islands. Can. F i e l d Nat.87:173—175. Moodie,G.E.E. and Reimchen,T.E.1976. Phenetic v a r i a t i o n and habitat differences i n Gasterosteus populations of the Queen Charlotte Is-lands. Syst.Zool.25:49—61. Munzing,J.1959. Bi o l o g i e , v a r i a b i l i t a t und genetik von Gasterosteus  aculeatus L. ( P i s c e s ) . Untersuchungen im Elbegebeit. Int.Rev.Ges.Hydrobiol.44: 317—382. 138 Miinzing,J.1962. E i n neuer semiarmatus-Typ von Gasterosteus aculeatus L. (Pisces) aus dem Izniksee. Mitt.Zool.Inst.Hamb.60:181—194. Miinzing,J.1963. The evolution of v a r i a t i o n and d i s t r i b u t i o n a l patterns i n european populations of the threespine stickleback, Gasterosteus aculeatus. Evol.17:320—332. Miinzing,J.1972. Polymorphic Populationen von Gasterosteus aculeatus L. (Pisces, Gasterosteidae) i n sekundaren Intergradionszonen der Deut-schen Bucht und benachbarter Gebiete. Die geographische V a r i a t i o n der Lateralbeschildung. Faunistisch—onkol.Mitt.4:69—84. Penczak.T.1960. New cases of v a r i a b i l i t y of the stickleback Gasterosteus aculeatus L. Fragm.Faun.,Warsaw.8:403—410. Penczak,T.1962. Morphometry of the stickleback, Gasterosteus aculeatus L., from the B a l t i c Sea. Zoologica Poloniae (Archivum S o c i e t a t i s Zoologorum Poloniae) 12:239—246. Penczak,T.1965. Morphological v a r i a t i o n of the stickleback (Gasterosteus aculeatus L.) i n Poland. Zoologica Poloniae (Ar-chivum S o c i e t a t i s Zoologorum Poloniae) 15:3—49. Penczak,T.1966. Comments on the taxonomy of the threespine stickleback, Gasterosteus aculeatus Linnaeus. Ohio J.Sci.66:81—87. Perlmutter,A.1963. Observations on fishes of the genus Gasterosteus i n the waters of Long Island, New York. Copeia 1963:168—173. Piraentel,R.A.1979. Morphometries, the multivariate analysis of  b i o l o g i c a l data. Kendall/Hunt Publ.Co., Dubuque, Iowa. Reimchen,T.E.1980. Spine deficiency and polymorphism i n a population of Gasterosteus aculeatus: An adaptation to predators? Can.J.Zool.58:1232—1244. Reisman,H.M.1968. Reproductive i s o l a t i n g mechanisms of the blackspotted stickleback, Gasterosteus wheatlandi. J.Fish.Res.Board, Canada. 25:2703—2706. Ridgway,M.S. and McPhail,J.D.1984. Ecology and evolution of sympatric sticklebacks (Gasterosteus): mate choice and reproductive i s o l a -t i o n i n the Enos Lake species p a i r . Can.J.Zool.62:1813—1818. Ross,S.T.1973. The systematics of Gasterosteus aculeatus (Pisces: Gasterosteidae) i n c e n t r a l and southern C a l i f o r n i a . Nat.Hist.Mus. Los.Ang.Co.Contrib.Sci.243:1—20. Rowland,VI.J.1983a. I n t e r s p e c i f i c aggression and dominance i n Gasterosteus. Env.Biol.Fish.8:269—277. Rowland,W.J.1983b. I n t e r s p e c i f i c aggression i n sticklebacks: Gasterosteus aculeatus displaces Apeltes quadracus. Copeia 1983:541—544. Rutter,C.1896. Notes on the freshwater f i s h e s of the P a c i f i c slope of North America. I. On the sticklebacks of C a l i f o r n i a with s p e c i a l reference to i n d i v i d u a l v a r i a t i o n s . Proc.Calif.Acad.Sci.,Ser.2, 6:245—254. Scott,W.B. and Crossman,E.J.1973. Freshwater Fishes of Canada. Fish. Res.Board, Canada.Bull.184. xiv+966pp. Semler,D.E.1971. Some aspects of adaptation i n a polymorphism for breeding colors i n the threespine stickleback (Gasterosteus aculeatus) . J.Zool. 165:291—302. Simpson,G.G.1961. P r i n c i p l e s of Animal Taxonomy. Columbia University Press, New York. xiv+247pp. Sokal,R.R.1973. The species problem reconsidered. Syst.Zool. 22:360—374. Sokal,R.R. and Rohlf,F.J.1969. Biometry: the p r i n c i p l e s and practice  of s t a t i s t i c s i n b i o l o g i c a l research. W.H.Freeman and Co. San Francisco. xxii+776pp. Soule^M.1967. Phenetics of natural populations. I I . Asymmetry and evolution i n a l i z a r d . Amer.Nat.101:141—160. Soule',M.; Yang,S.Y.; Weiler,M.G.W. and Gorman,G.C.1973. Island l i z a r d s the genetic-phenetic v a r i a t i o n c o r r e l a t i o n . Nature.242: 191—193 White,M.J.D.1978. Modes of Speciation. W.H.Freeman and Co., San Fran-c i s c o . 455pp. 140 Wiley,E.0.1978. The evolutionary species concept reconsidered. Syst.Zool. 27:17—26. Wiley,E.0.1980. Is the evolutionary species f i c t i o n consideration of classes, i n d i v i d u a l s and h i s t o r i c a l e n t i t i e s . Syst.Zool.29:76—80. Wiley,E.0.1981. Phylogenetics: the theory and practice of phylogenetic  systematics. John Wiley and Sons, Inc., New York. xv+439pp. Willis,E.0.1981. Is a species an interbreeding u n i t , or an i n t e r n a l l y s i m i l a r part of a phylogenetic tree? Syst.Zool.30:84—85. Wilson,D.S.1980. The Natural S e l e c t i o n of Populations and Communities. Benjamin/Cummings Publ.Co.Inc., Menlo Park, C a l i f . xviii+186pp. Wilz,K.J.1970. Reproductive i s o l a t i o n i n two species of stickleback (Gasterosteidae). Copeia 1970:587—590. Withler,R.E.1980. Genetic r e l a t i o n s h i p s among threespine sticklebacks Gasterosteus aculeatus. Unpubl. MSc t h e s i s . U n i v e r s i t y of B r i t i s h Columbia, Vancouver. 136pp. Wootton,R.J.1976. The Biology of the Sticklebacks. Academic Press, London. x+387pp. Worgan,J.P. and FitzGerald.G.J.1981. D i e l a c t i v i t y and d i e t of three sympatric sticklebacks i n t i d a l s a l t marsh pools. Can.J.Zool. 59:2375—2379. Worgan,J.P. and FitzGerald.G.J.1981. Habitat segregation i n a s a l t marsh among adult sticklebacks (Gasterosteidae). Envir. B i o l . F i s h . 6:105—109. 141 APPENDIX 1. DESCRIPTION OF SITES AND INTERSITES BONSALL CREEK The tables included in this section (Tables 1, 2, 3) summarize the physical characteristics, collection information, and the flora and fauna of each site on Bonsall Creek. The following text includes a short description of each site and major intersite (see Fig. 1). The specific identifications included in this section, especially those of flora are tentative. Site _V i s located on the most northern tip of one of the outer Shoal Islands. At low tide, the tide channel of Bonsall Creek meets the waters of Stuart Channel, at this surfgrass (Phyllospadix scouleri) bed. At high tide, this area is completely inundated. Like sites W to Z and 1 to 6, site V can best be reached from the parking lot at the Crofton Pulp M i l l picnic grounds. Site W is a surfgrass bed located on the mudflats between the outer and inner Shoal Islands. At high tide, the tide channel at this site is also completely covered by salt water. Site X. is located on the eastern side of one of the inner Shoal Islands. This surfgrass bed is at the mouth of a small side channel which flows into the tide channel of Bonsall Creek. At low tide the tide channel is about forty feet wide. At high tide this area is com-pletely covered by salt water. Site _Y is located on the western side of one of the inner Shoal Inlands. The large rocks in this site have a dense growth of algae, especially Ulva sp. and Enteromorpha intestinalis. Like a l l other sites which are situated in tide channels of the estuary, this site is com-142 Table 1. Physical characteristics and collection information for each site of Bonsall Creek. V W X Y Z 1 2 3 4 5 6 7 8 habitat stratum s i t e stratum c o l l e c t i o n JUNE date JULY water temp. JUNE C JULY # specimens JUNE caught JULY # specimens JUNE examined JULY s i t e length ( f t ) s i t e width ( f t ) next s i t e ( f t ) upstream depth (cm) (low tide) bottom type Bl Bl Bl Bl Bl Bl Bl Bl Bl Bl Bl 1 2 3 4 5 5 5 5 6 6 7 6 6 6 6 6 6 6 6 6 6 6 10 25 25 25 25 25 26 26 26 26 26 26 26 26 15 15 15 14 20 20 20 20 19 19 19 13 22 24 23 24 25 25 24 24 23 19 21 12 12 54 18 40 31 2 1 1 0 1 1 4 10 42 8 18 0 0 0 0 0 2 3 14 1 23 6 6 6 6 2 1 1 0 1 1 4 6 6 0 5 0 0 0 0 0 2 3 6 1 6 100 100 50 100 100 100 100 100 100 100 100 100 80 — 40 40 50 60 60 60 40 30 30 60 45 40 700 500 150 500 500 500 500 645 600 500 2525 480 340 75 75 30 40 30 30 20 30 40 75 50 60 35 S+M S+M S GSM G G+M G+M G+M G+M S+M G+M G G+M S+M = sand and mud, S = sand, GSM = gravel, sand and mud, G+M = gravel and mud, G = gravel habitat stratum s i t e stratum c o l l e c t i o n JUNE date JULY water temp. JUNE C JULY # specimens JUNE caught JULY # specimens JUNE examined JULY s i t e length ( f t ) s i t e width ( f t ) next s i t e ( f t ) upstream depth (cm) (low tide) bottom type 9 10 11 12 13 14 15 16 17 18 19 20 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 B2 8 8 9 10 11 12 13 14 15 16 17 18 9 9 9 9 9 9 7 7 7 7 26 26 26 25 25 25 25 25 23 23 23 23 9 9 9 8 8 7 9 9 10 11 10 9 9 9 9 8 8 8 8 8 8 9 60 42 19 0 53 24 51 53 4 30 16 12 51 15 — 5 28 46 50 51 17 50 6 6 6 0 6 6 6 6 2 6 6 4 6 6 - 5 6 6 6 6 6 6 100 85 100 100 95 120 115 120 100 80 65 100 45 25 30 45 25 30 25 20 15 20 20 30 325 525 515 420 740 545 710 1675 630 435 850 2420 60 100 50 50 75 60 80 30 60 75 75 75 GSM GSM G GSM G+M GSM GSM G G+M M M+S S+M GSM = gravel, sand and mud, G = gravel, G+M = gravel and mud, M = mud, M+S = mud and s t i c k s , S+M = sand and mud 21 22 23 24 25 26 27 28 29 30 habitat stratum site stratum B3 19 B3 20 collection date JUNE JULY 8 23 8 24 water temp. C JUNE JULY 11 11 14 16 # specimens caught JUNE JULY 53 50 50 4 # specimens examined JUNE JULY 6 12 6 4 site length (ft) site width (ft) next site (ft) upstream 65 20 850 100 20 1900 depth (cm) 120 75 bottom type M M B3 B3 B3 B3 B3 B3 B3 B3 21 22 23 24 25 26 27 28 8 8 8 8 8 8 8 8 24 24 24 24 24 24 24 24 13 14 13 12 11 10 10 9 15 17 16 18 17 16 14 14 97 52 46 37 1 5 3 76 51 50 51 45 16 37 52 5 6 6 6 6 1 5 3 6 6 6 12 6 6 6 6 5 100 100 100 100 100 115 105 95 15 15 15 15 10 10 10 15 565 520 630 500 600 480 635 4875 75 50 60 15 25 25 15 30 M M G G G G G M M = mud, G = gravel 143c 31 32 33 34 35 habitat stratum B4 B4 B4 B4 B4 s i t e stratum 29 30 30 31 31 c o l l e c t i o n JUNE 10 10 7 7 7 date JULY 23 23 23 23 23 water temp. JUNE 9 9 11 9 7 C JULY 11 11 13 13 — # specimens JUNE 2 53 3 20 1 caught JULY 4 50 0 14 0 # specimens JUNE 2 6 3 6 1 examined JULY 4 6 0 12 0 si t e length ( f t ) 100 100 100 100 100 s i t e width i ( f t ) 5 6 8 10 4 next s i t e ( f t ) 500 2510 600 975 upstream depth (cm) 50 50 50 50 15 bottom type GSM GSM M M+S M+S GSM = gravel, sand and mud, M = mud, M+S = mud and st i c k s 144 Table 2. Aquatic vertebrate fauna and flora found at each site of Bonsall Creek. FISH Syngnathus griseolineatus  Cymatogaster aggregata  Leptocottus armatus  Oligocottus maculosus  Oligocottus snyderi  Platichthys s t e l l a t u s  Anoplarchus sp. Onchorhynchus keta  Onchorhynchus kisutch  Porichthys notatus  Clevlandla ios  Cottus asper  Salmo sp. Lampetra sp. HERPS Ambystoma g r a c i l e  Taricha granulosa  Rana aurora  Thamnophis sp. PLANTS Phyllospadix s c o u l e r i  Gigartlna sp. Fucus sp. Ulva sp. Enteromorpha i n t e s t i n a l i s unident. green alga grass Lemna minor  Typha l a t i f o l i a  Lysichitum americanum  Potamogeton sp. Elodea sp. Oenanthe sarmentosa 9 10 11 12 13 14 15 16 17 18 19 20 FISH Syngnathus grlseolineatus  Cymatogaster aggregata  Leptocottus armatus  Oligocottus maculosus  Oligocottus snyderi  Platichthys S t e l l a t u s  Anoplarchus sp. Onchorhynchus keta  Onchorhynchus kisutch  Porichthys notatus  Clevlandla ios  Cottus asper  Salmo sp. Lampetra sp. HERPS Ambystoma g r a c i l e  Taricha granulosa Rana aurora * * * * * * * * * * * * * * * * * * * * * * Thamnophls sp. PLANTS Phyllospadix s c o u l e r i  Gigartina sp. Fucus sp. Ulva sp. Enteromorpha i n t e s t i n a l l s unident. green alga grass Lemna minor  Typha l a t i f o l i a  Lysichitum americanum  Potamogeton sp. Elodea sp. Oenanthe sarmentosa * * * * * * * * * * * * * * 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 FISH Syngnathus griseolineatus  Cymatogaster aggregata  Leptocottus armatus  Oligocottus maculosus  Oligocottus snyderi  Platichthys s t e l l a t u s  Anoplarchus sp. Onchorhynchus keta  Onchorhynchus kisutch  Porichthys notatus  Clevlandia ios Cottus asper * * * Salmo sp. * * * * Lampetra sp. * * * * * * * * * * * * * * * * HERPS Ambystoma g r a c i l e * Taricha granulosa * Rana aurora * * * * * * * Thamnophis sp. * PLANTS Phyllospadix s c o u l e r i  Gigartina sp. Fucus sp. Ulva sp. Enteromorpha i n t e s t i n a l i s unident. green alga grass * * * * * * * * * * * * * * Lemna minor * * * * * * * * * * * Typha l a t i f o l i a Lysichitum americanum * * * * * * Potamogeton sp. Elodea sp. * * * * * * * * * * Oenanthe sarmentosa * * * * * 146 Table 3. Aquatic invertebrate fauna found at each site of Bonsall Creek. Cnidaria Ctenophora Mollusca Physidae Lymnaeidae Planorbidae Unionidae Crassostrea gigas Crustacea Balanus glandula Amphipoda Isopoda Malacostracan shrimp Upogebia pugetensis Hemigrapsus sp. Pagurus sp. Astracus sp. Insecta Ephemeroptera Anisoptera Zygoptera Plecoptera Gerridae Corixidae Notonectidae Lethocerus americanus Ranatra sp. Dytiscus sp. Colymbetes sp. Gyrinus sp. Hydrophilidae Trichoptera Diptera Annelida Hirudinea 9 10 11 12 13 14 15 16 17 18 19 20 Cnidaria Ctenophora Mollusca P ^ y g 2_ d £ t g * * * * * * * * * * Lymnaeidae * * * * * * * * * Planorbidae * * * * Unionidae Crassostrea gigas Crustacea Balanus glandula Amphipoda * Isopoda Malacostracan shrimp Upogebia pugetensis  Hemigrapsus sp. Pagurus sp. Astracus sp. Insecta Ephemeroptera * * * * * Anisoptera * Zygoptera * Plecoptera Gerridae * * * * * * * Corixidae * Notonectidae * * * * * * Lethocerus americanus  Ranatra sp. Dytiscus sp. * * Colymbetes sp. * Gyrinus sp. * Hydrophilidae * * * Trichoptera * * * * * * * * * Diptera * Annelida Hirudinea * * 147b 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Cnidaria Ctenophora Mollusca Physidae Lymnaeidae Planorbidae Unionidae Crassostrea gigas Crustacea Balanus glandula Amphipoda Isopoda Malacostracan shrimp Upogebia pugetensis Hemigrapsus sp. Pagurus sp. Astracus sp. Insecta Ephemeroptera Anisoptera Zygoptera Plecoptera Gerridae Corixidae Notonectidae Lethocerus americanus Ranatra sp. Dytiscus sp. Colymbetes sp. Gyrinus sp. Hydrophilidae Trichoptera Diptera Annelida Hirudinea * A * * * * A A A * * A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A 148 p l e t e l y inundated at high t i d e . S i t e Z i s part of a t i d e channel, distinguished by a large gravel bar surrounded by Ulva sp. and Enteromorpha i n t e s t i n a l i s . Site 1 i s located i n the t i d a l mud f l a t s , where the edges of the tide channel support a dense growth of Ulva sp. and Enteromorpha i n - t e s t i n a l i s . S i t e 2 i s located i n the t i d a l mud f l a t s , where the southern edge of the tide channel supports a dense growth of Ulva sp. and En- teromorpha i n t e s t i n a l i s . S i t e _3 i s located at a Y-junction of the tide channels i n the estuary. Ulva sp. and Enteromorpha i n t e s t i n a l i s grow throughout the s i t e . S i t e 4- i s part of a t i d e channel i n the estuary. This i s the f i r s t s i t e where the surrounding mudflats have a sparce cover of sedges and prickleweed ( S a l i c o r n i a v i r g i n i c a ) . Site _5 i s located i n the estuary and i s the f i r s t s i t e where Bonsall Creek runs between mud banks. The banks are about three feet high and on the f l a t s above them there i s a dense cover of sedges and prickleweed. This s i t e had a l o t of Ulva sp. and Enteromorpha i n -t e s t i n a l i s , but these algae were fragmented and not attached to the gravel. S i t e 6_ i s located i n the estuary and l i k e s i t e 5 has three foot mud banks topped by sedges and prickleweed. The Ulva sp. and Enteromor- pha i n t e s t i n a l i s were attached to the cobble and gravel. The creek at t h i s s i t e i s about s i x t y feet wide. Intersite the 2525 feet (about half a mile) separating sites 6 and 7 do not appear to be prime stickleback habitat. In fact, judging by the small number of fi s h caught between sites Z and 7, i t might be said that much of the estuary i s not prime habitat for sticklebacks. The creek, for the half mile between sites 6 and 7, flows between three to six foot banks topped by grasses and sedges. The creek i t s e l f is deep and appears to have l i t t l e or no attached vegetation throughout this stretch. Site 7_ in this area, the creek is bounded by grass and willows Salix spp.) atop three foot banks. Much of the site i s a combination of r i f f l e s and backwaters. Enteromorpha intestinalis occurs here at-tached to the gravel. Access to sites 7—11 is best from a farm road which parallels the creek. The entrance to this road is on the south side of the f i r s t bridge crossing Bonsall Creek. Site 8^ is the last site where Enteromorpha intestinalis is found. For this reason, i t is the last site included in the f i r s t regional habitat. The banks are about a foot high and covered with grass. The southern side of the stream supports a dense growth of cat-t a i l s (Typha l a t i f o l i a ) . Site 9 i s bounded on one side by large trees: western red cedar (Thuja plicata), bigleaf maple (Acer macrophyllum), sitka spruce (Picea sitchensis), and willows (Salix spp.). The other side is a grassy, un-dercut bank, where most of the specimens were captured. Site 10 is bounded on one side by large, overhanging, western red cedar. The opposite bank is grassy, with willows and sitka spruce further back from the stream. Along the grassy bank, Elodea sp. is 150 rooted in water about two feet deep and this was where most of the specimens were caught. The rest of the site is over three feet deep. Site 11 is crossed by the f i r s t bridge over Bonsall Creek on the Crofton Road. The site i t s e l f i s about thirty-five feet wide and is quite shady due to the presence on both banks of large western red cedar, bigleaf maples and and willows. The best stickleback habitat is in large patches of Elodea sp. and Potamogeton sp. Site 12 is in a small valley shaded by a number of large bigleaf maples and in places has a sandy bottom. The rooted aquatic vegetation occurs in a number of small patches throughout the more shallow parts of the site. Sites 12 to 16 can best be reached by following the r a i l -way tracks that run near the creek and cross i t in three places. Site 13 this site i s , according to a local farmer, very close to the limit of t i d a l influence for water level of the creek. There are a few large trees, but the site is quite open and there is a luxuriant growth of rooted aquatic plants throughout the site. Both banks are covered with blackberry (Rubus discolor), nettles (Urtica dioica) and skunk cabbage (Lysichtum americanum). Site 14 is a very dark site due to the many large western red cedar and bigleaf maples which are present on both sides of the creek. There are-so many-trees that there is very l i t t l e underbrush; The site i t s e l f has no rooted aquatic vegetation except for a few skunk cabbage. Site 15 is in a small valley and shaded by a couple of large apple (Malus diversifolia?) and bigleaf maple trees. Potamogeton sp. is found in large patches in the downstream half of the site. 151 Site 16 is lined on the south side by a grove of willows and young bigleaf maples. The other bank is open and grassy. Potamogeton sp. is very dense throughout the entire site. Intersite the 1675 feet (about a quarter mile) separating sites 16 and 17 is mostly very deep (more than five feet) with numerous log jams and overhanging brush and trees. Much of this i s pools and bars with l i t t l e rooted aquatic vegetation and a large cobble bottom. Within this area is a fish processing from which effluent may occasionally escape or seep into the creek. Site 17 is in a small valley with a few large western red cedar and bigleaf maples on the valley sides. The edges of the creek are grassy and grass is also growing in the stream. Between the grass and the trees, there are a number of red alders (Alnus rubra) growing on the shore. Sites 17 to 20 can best be reached by parking at the nearby fish processing plant and walking through the back of the Halalt Indian Reserve on a foot t r a i l . Site 18 is surrounded, well back from the edge of the creek, by many large western red cedar. The banks of the creek at this site are grassy and skunk cabbage is abundant. The creek at this site is about twenty feet wide. The water is just over knee deep, but in one spot i s over ten feet deep. The shallower portions of the site are f i l l e d with grass and Potamogeton sp. Site 19 is in a small valley and bordered by western red cedar, willows, devil's club (Opolopanax horridum) and thimbleberry (Rubus par- viflorus). The small amount of stickleback habitat is made up of grass, skunk cabbage and Elodea sp. 152 Site 20 i s in a small valley and bordered by western red cedar, willows, red alders, twinflower (Linnaea borealis), thimbleberry and devil's club. Sticklebacks were largely collected in and near the aquatic plants Potamogeton sp. and grass. Intersite the 2420 feet (about half a mile) separating sites 20 and 21 is quite deep (in places more than ten feet) and the mud bottom and sides are extremely slippery. Beaver (Castor canadensis) have been quite active in this area and local people have destroyed at least two dams, although two beaver lodges are intact. Although the flow rate is quite slow, the lack of rooted aquatic vegetation and aquatic inver-tebrates indicates that this i s not prime stickleback habitat. Site 21 is bounded on one side by a steep bank with willows and other brush which overhang the creek. The other side is a gentle bank with grass edging a f i e l d . The stickleback habitat in this site is primarily under the grassy bank and in the grass rooted in the stream. The second bridge crossing Bonsall Creek (Westholme Bridge) marks the upstream end of this site. Site 22 is just upstream from the junction of Whitehouse Creek with Bonsall Creek. One side of the site is a grassy bank on the edge of a potato f i e l d . The other bank (twenty feet away) is covered by a dense growth of willows and other brush. The entire site i s a dense patch of Elodea sp. growing to a height of almost three feet in the deeper areas. Sites 22 to 30 can best be accessed by walking up the creek and/or through the bordering pastures and fi e l d s . Intersite the 1900 feet separating sites 22 and 23 is quite deep (more than fifteen feet) because i t has been dredged for use as an i r -153 r i g a t i o n canal. Both banks are covered with mats of c a t - t a i l s and grass which extend over the water surface. Although sticklebacks could possibly be found under these mats, to c o l l e c t them would have been too dangerous to make i t worthwhile. Site 23 i s bordered by red alders on one side and a cow pasture on the other. There are clumps of grass and Elodea sp. rooted i n the mud that provide habitat for the sticklebacks. Site 24 i s bordered by willows on one side and a cow pasture on the other and i s very similar to s i t e 23. Site 25 i s bordered on both sides by pastures; there i s a single large bigleaf maple on the northern bank. There are also patches of Elodea sp. and skunk cabbage along the northern bank where most of the sticklebacks were captured. Site 26 i s a very open s i t e with pastures on either side. Most sticklebacks were caught i n the deeper areas where the bank was s l i g h t l y undercut and the pasture grass was dragging i n the water. The s i t e i s about f i f t e e n feet wide. Site 27 i s centered i n a pasture where one side of the creek i s edged by a s i x foot bank, while the other i s only about three feet high and a very gentle slope. The only habitat for sticklebacks was a small amount of Elodea sp. and grass i n one of the s l i g h t l y deeper areas. Site 28 i s i n a pasture, but with blackberry thickets on both sides. The gravel bottom does not support any rooted aquatic vegeta-t i o n , although the s o i l from the banks provides a substrate for the small amount of grass and Elodea sp. edging this s i t e . 154 Site 29 is banked on one side by a western red cedar forest, while the other side opens onto a pasture. A small amount of Elodea sp. and grass were found on the sides of the creek in this s i t e . This site was severely disturbed during the summer of 1982 by the removal of a great deal of the gravel bottom, probably by one of the local farmers. Site 30 is essentially a swamp draining into Bonsall Creek. The water i s choked with grass, Elodea sp., water parsley (Oenathe sarmen- tosa) and skunk cabbage. By the end of the summer (September) the con-nection between this swamp and Bonsall Creek was dry, although stick-lebacks were s t i l l present in the swamp. Intersite the 4875 feet (about three-quarters of a mile) separating sites 30 and 31 consists of beautiful pool and r i f f l e habitat with a large cobble and gravel bottom but devoid of rooted aquatic vegetation. The Trans-Canada Highway crosses the creek 3000 feet up-stream from site 30. This crossing has diverted the creek through a large culvert with concrete baffles and then the creek drops about a foot to the creek bed. This small waterfall and the culvert are probably a significant barrier to upstream movement of sticklebacks; especially since there is not suitable habitat for sticklebacks above or below the culvert for a considerable distance. Site 31 is edged by red alders, birch (Betula) sp., western red cedar and pastures on both sides. The water course is choked by Elodea sp., water parsley, skunk cabbage and grass. Sites 31 and 32 can best be accessed from Somenos Road and then walking from the road down a sloped pasture. 155 Site 32 is bordered by willows on the edges of the pastures through which the creek runs. Sticklebacks are found under the undercut banks and in the small patches of Elodea sp. and water parsley. The creek is now only about six feet wide. Intersite the 2510 feet (about half a mile) separating sites 32 and 33 i s heavily overgrown by willows and blackberries, and is essen-t i a l l y impassible. This area has almost no rooted aquatic vegetation. Somenos Road crosses the stream 2200 feet upstream from site 32. In this case, the creek passes through a small culvert under the road and then drops more than four feet from the end of the culvert to the water surface. This would be a very effective barrier against upstream move-ment, although i t would potentially allow downstream movement by stick-lebacks. It should be noted that no other fish species are found up-stream of this barrier. Site 33 is bordered by willows and skunk cabbage on one side and by a grassy pasture on the other side. The site i s choked by grass, skunk cabbage, water parsley and Elodea sp. By the end of the summer (September) the creek no longer flowed in this area; i t was reduced to a series of potholes, many containing sticklebacks. Sites 33 to 35 are best accessed from the nearby railway tracks and then by a cow path along the edge of the creek. Site 34 i s in a rather open red alder and western red cedar woods, used as a cow pasture. The creek is f u l l of skunk cabbage, water parsley, grass and Elodea sp. Like site 33, this site by the end of the summer is dry except for potholes, many containing sticklebacks. Site 35 is totally overgrown by willows and f i l l e d with wate parsley. By the end of the summer this site was dry. The creek bed only about four feet wide at this site. 157 APPENDIX 2. DESCRIPTION OF SITES NUNNS CREEK The tables included in this section (Tables 4, 5, 6) summarize the physical characteristics, collection information, and the flora and fauna of each site on Nunns Creek. The following text includes a short description of each site. The specific identifications included in this section, especially those of flora are tentative. Site 1_ is located on the inside of Tyee Spit where Nunns Creek meets the waters of the inlet. At low tide, this site i s like a trough, the deepest areas f u l l of Fucus sp., Gigartina sp., Ulva sp. and other marine algae. At high tide, the creek is s t i l l held within a channel, although the water may rise to a depth of over ten feet. The banks are grassy and covered with sedges with a few scrubby willows (Salix spp.) above the high tide mark. Sites 1 to 5 can best be reached by driving to the tip of Tyee Spit and then walking up the creek. Site 2 i s located on a gravel f l a t s , which at low tide i s barely covered by water. At the edges of the site are found Fucus sp., Gigar- tina sp., Ulva sp., sedges and grasses. Most sticklebacks were captured in this vegetation. Above the high tide mark are scrubby willows. The creek at this site is about thirty feet wide. Site 3 is also located on the gravel f l a t s , although at low tide i t is covered with about ten inches of water. There is some Fucus sp. present in this site, but the most abundant types of vegetation are sedges and grass growing into the creek from the edges. Up above on the banks are scrubby willows. 158 Table 4. Physical characteristics and collection information for each site of Nunns Creek. 1 2 3 4 5 6 7 8 9 10 11 12 habitat stratum Nl Nl Nl Nl N2 N2 N2 N2 N3 N3 N4 N4 s i t e stratum 32 33 34 35 36 37 38 39 40 41 42 43 c o l l e c t i o n JUNE 4 4 4 4 4 4 4 4 4 4 4 4 date JULY 27 27 27 27 28 28 28 28 28 27 27 27 water temp. JUNE 15 16 16 17 17 15 14 13 11 11 12 14 C JULY 17 22 22 21 16 15 15 15 15 14 18 18 # specimens JUNE 40 47 50 50 49 50 49 50 19 42 51 60 caught JULY 7 23 44 59 52 54 50 51 41 43 22 51 # specimens JUNE 6 6 6 6 6 6 6 6 6 6 6 6 examined JULY 6 2 6 6 12 12 6 6 12 12 6 6 s i t e length ( f t ) 100 100 100 100 100 100 100 100 100 100 100 100 s i t e width ( f t ) 35 30 20 20 35 25 30 15 15 10 — — next s i t e ( f t ) 500 500 500 500 500 460 650 515 490 1720 400 — upstream depth (cm) 75 5 20 30 50 50 30 50 40 20 75 75 (low tide) bottom type G G G+M G+M G+M G+M M M+S M+S M+S M M G = gravel, G+M = gravel and mud, M = mud, M+S = mud and s t i c k s Table 5. Aquatic vertebrate fauna and flora found at each site of Nunns Creek. FISH Leptocottus armatus  Oligocottus maculosus Platichthys s t e l l a t u s Anoplarchus sp. Oncorhynchus keta Oncorhynchus kisutch Cottus asper  Salmo sp. Lampetra sp. HERPS Ambystoma g r a c i l e  Taricha granulosa Rana aurora PLANTS Fucus sp. Gigartina sp. Ulva sp. grass Typha l a t i f o l i a  Lysichitum americanum  Potamogeton sp. Elodea sp. Alisma plantago-aquatica Oenanthe sarmentosa 162 Table 6. Aquatic invertebrate fauna found at each site of Nunns Creek. Mollusca Physidae Planorbidae Annelida Hirudinea Crustacea Amphipoda Isopoda Malacostracan shrimp Insecta Ephemeroptera Anisoptera Zygoptera Gerridae Lethocerus americanus  Ranatra sp. Dytiscus sp. Colymbetes sp. Hydrophilidae Trichoptera 164 S i t e 4_ i s also located on the gravel f l a t s with edges of sedges and grasses and a small amount of Fucus sp. and Ulva sp. The banks are about three feet high with scrubby willows and a few s i t k a spruce (Picea  s i t c h e n s i s ) growing on them. Si t e _5 i s bounded by three to s i x foot mud banks about t h i r t y -f i v e feet apart, which on one side are covered i n sedges, grasses and a few willows and on the other side close to the creek are covered by grasses and above the high t i d e mark by willows, red alders (Alnus  rubra) and a f ew s i t k a spruce. The sticklebacks were pr i m a r i l y captured i n the patches of grass, Elodea sp. and water plantain (Alisma plantago- aquatica). Site 6 i s located just downstream of the f i r s t bridge crossing Nunns Creek on the Tyee Spit road. Like s i t e 5, t h i s s i t e i s bounded by three to s i x foot mud banks, cloaked i n sedges and grass close to the stream. The west bank i s also covered with willows, while the east bank i s covered with willows and thimbleberry (Rubus p a r v i f l o r u s ) . S i t e _7 i s located just upstream of the f i r s t bridge crossing Nunns Creek on the Tyee Spit road. The mud banks are about three feet high and on the east, covered with c a t - t a i l s (Typha l a t i f o l i a ) and on the west close to the creek are covered with sedges, grasses, willows and twinflower (Linnaea bo r e a l i s) and higher on the bank by s i t k a spruce, western red cedar (Thuja p l i c a t a ) and red alders. The aquatic vegetation consists of a very dense growth of water plantain with a l i t t l e Elodea sp. and grasses. This was prime stickleback habitat. Sites 7 to 9 are most e a s i l y accessed by walking upstream from the Tyee Spit road. 165 Site 8_ i s bounded by two foot mud banks topped by c a t - t a i l s and water plan t a i n . The western bank also has a number of s i t k a spruce . The water plantain and skunk cabbage (Lysichtum americanum) grew i n small patches on the western h a l f of the s i t e . At t h i s s i t e , the creek i s only about f i f t e e n feet wide. Si t e 9_ i s bounded by two foot banks topped with patches of cat-t a i l s and patches of twinflower, thimbleberry and blackberry (Rubus  d i s c o l o r ) . Further up the banks on both sides are red alders, willows and s i t k a spruce. The sticklebacks were c o l l e c t e d i n small patches of water parsley (Oenanthe sarmentosa), skunk cabbage and c a t - t a i l s . This s i t e i s 115 feet downstream from the culvert over which passes Highway 16. The s i t e i s probably the l a s t s i t e where the water l e v e l i s i n -fluenced by the t i d e . Sites 1 to 9 are a l l i n the Campbell River Indian Reserve. S i t e 10 i s almost p a r a l l e l to 16th Avenue i n Campbell River, just downstream of the culvert over which passes 16th Avenue. The southern bank i s covered with c a t - t a i l s and willows. The northern bank i s covered with b i g l e a f maple (Acer macrophyllum) and red alders, with an understory of thimbleberry, twinflower and red huckleberry (Vaccinium  parvifolium). The aquatic v e g i t a t i o n consists of two large patches of Potamogeton sp., Elodea sp. and waterleaf. At t h i s s i t e , the creek i s about ten feet wide. I n t e r s i t e the 1720 feet separating s i t e s 10 and 11 does not ap-pear to be prime stickleback habitat. Both banks are l i n e d by trees mostly red alders and bi g l e a f maples, but there i s e s s e n t i a l l y no rooted aquatic vegetation nor aquatic invertebrates. This area appears to be 166 quite barren. Site 11 is located on the northeastern edge of the pond created by beaver (Castor canadensis) activity. The bank vegetation is primarily red alders and blackberry. The aquatic vegetation consists of mats of floating grass and skunk cabbage. Most of the sticklebacks were trapped near the mats of floating grass. The best access to this site i s via the parking lot of the Campbell River Rodeo Grounds. Site 12 is located on the southwestern edge of the pond created by beaver activity. The bank vegetation is primarily sitka spruce. The site is f i l l e d with dead snags and skunk cabbage, with a few mats of floating grass. Most of the sticklebacks were trapped near the snags and skunk cabbage. Access to this site i s obtained by crossing the beaver pond in a southwesterly direction. Intersite upstream of site 12, Nunns Creek is made up of a com-bination of small pools and r i f f l e s flowing through a wooded ravine for about two and a half miles. This part of the stream has quite a high gradient compared to the lower reaches the stream. There is also l i t t l e or no aquatic vegetation and is therefore not prime habitat for stick-lebacks. 167 APPENDIX 3. CHARACTER DESCRIPTIONS 1. Standard Length as i n Hubbs and Lagler (1958) 2. Head Length as i n Hubbs and Lagler (1958) 3. Snout Length as i n Hubbs and Lagler (1958) 4. Length of Orbit as i n Hubbs and Lagler (1958) 5. I n t e r o r b i t a l Width (least bony width) as i n Hubbs and Lagler (1958) 6. Depth of Head as i n Hubbs and Lagler (1958) 7. Length of Upper Jaw as i n Hubbs and Lagler (1958) 8. Length of Mandible as i n Hubbs and Lagler (1958) 9. Length from Snout to DSO ( F i g . 3, character 9) i s the distance from the most anterior point on the upper l i p to the most posterior t i p of the dermosupraoccipital. 10. Length from DSO to DI ( F i g . 3, character 10) i s the distance from the most posterior t i p of the dermosupraoccipital to the base of the f i r s t dorsal spine. 11. Length from DI to DII ( F i g . 3, character 11) i s the distance from the base of the f i r s t dorsal spine to the base of the second dorsal spine. 12. Length from DII to D i l i ( F i g . 3, character 12) i s the distance from the base of the second dorsal spine to the base of the t h i r d dorsal spine. 13. Length of Dorsal Base ( F i g . 3, character 13) i s the distance from the base of the t h i r d dorsal spine to the base of the l a s t dorsal ray. 168 14. Depth of Caudal Peduncle as in Hubbs and Lagler (1958) 15. Width of Caudal Peduncle is the least width of the caudal peduncle, at the midline. 16. Length of Anal Base (Fig. 3, character 16) is the distance from the base of the anal spine to the base of the last anal ray. 17. Length from Al to VI (Fig. 3, character 17) is the distance from the base of the anal spine to the base of the ventral spine. 18. Length from VI to ECTO (Fig. 3, character 18) is the distance from the base of the ventral spine to the most anterior tip of the ec-tocoracoid. 19. Length from ECTO to snout (Fig. 3, character 19) is the distance from the most anterior tip of the ectocoracoid to the most anterior point on the upper l i p . 20. Depth from DSO to ECTO (Fig. 3, character 20) is the distance from the most posterior tip of the dermosupraoccipital to the most an-terior tip of the ectocoracoid. 21. Depth from DSO to VI (Fig. 3, character 21) is the distance from the most posterior tip of the dermosupraoccipital to the base of the ventral spine. 22. Depth from DI to ECTO (Fig. 3, character 22) is the distance from the base of the f i r s t dorsal spine to the most anterior tip of the ectocoracoid. 23. Depth from DII to ECTO (Fig. 3, character 23) is the distance from the base of the second dorsal spine to the most anterior tip of the ectocoracoid. 169 24. Depth from DI to VI (Fig. 3, character 24) i s the distance from the base of the f i r s t dorsal spine to the base of the ventral spine. 25. Depth from DII to VI (Fig. 3, character 25) i s the distance from the base of the second dorsal spine to the base of the ventral spine. 26. Depth from D i l i to VI (Fig. 3, character 26) i s the distance from the base of the t h i r d dorsal spine to the base of the ventral spine. 27. Depth from DII to A l (Fig. 3, character 27) i s the distance from the base of the second dorsal spine to the base of the anal spine. 28. Depth from D i l i to A l (Fig. 3, character 28) i s the distance from the base of the t h i r d dorsal spine to the base of the anal spine. 29. Depth from D i l l to Last A Ray (Fig. 3, character 29) i s the distance from the base of the t h i r d dorsal spine to the base of the la s t anal ray. 30. Depth from Last D Ray to Last A Ray (Fig. 3, character 30) i s the distance from the base of the l a s t dorsal ray to the base of the la s t anal ray. 31. Depth from Last D Ray to A l (Fig. 3, character 31) i s the distance from the base of the last dorsal ray to the base of the anal spine. 32. Length from Last D Ray to Base of Caudal (Fig. 3, character 32) i s the distance from the base of the l a s t dorsal ray to the base of the middle caudal ray. 33. Length of Caudal Peduncle as i n Hubbs and Lagler (1958) 170 34. Length of Pectoral Base ( F i g . 3, character 34) i s the distance from the base of the most dorsal pectoral ray to the base of the most v e n t r a l pectoral ray. 35. Depth from DSO to dPl ( F i g . 3, character 35) i s the distance from the most posterior t i p of the dermosupraoccipital to the base of the most dorsal pectoral ray. 36. Depth from DI to dPl ( F i g . 3, character 36) i s the distance from the base of the f i r s t dorsal spine to the base of the most dorsal pectoral ray. 37. Length from Last P Ray to ECTO ( F i g . 3, character 37) i s the distance from the base of the most ve n t r a l pectoral ray to the most anterior t i p of the ectocoracoid. 38. Length from Last P Ray to VI ( F i g . 3, character 38) i s the distance from the base of the most ventral pectoral ray to the base of the ventral spine. 39. Width at Ascending Process i s the maximum width at the midline taken at the ascending process of the p e l v i c skeleton. 40. Maturity i s scored from an examination of the gonads. 41. Sex i s scored from an examination of the gonads. 42. Length of G i l l Raker i s taken from the f i r s t g i l l raker on the ceratobranchial of the f i r s t arch. 43. Length of DI ( F i g . 3, character 43) i s the distance from the l a t e r a l flange at the base of the d i s a r t i c u l a t e d spine to the t i p of the spine. 44. Length of DII i s the distance from the l a t e r a l flange at the base of the d i s a r t i c u l a t e d spine to the t i p of the spine. 171 45. Length of D i l i i s the distance from the l a t e r a l flange at the base of the d i s a r t i c u l a t e d spine to the t i p of the spine. 46. Length of A l i s the distance from the l a t e r a l flange at the base of the d i s a r t i c u l a t e d spine to the t i p of the spine. 47. Length of VI i s the distance from the l a t e r a l flange at the base of the d i s a r t i c u l a t e d spine to the t i p of the spine. 48. Length of Ventral Shield ( F i g . 3, character 48) i s the maximum distance from the anterior process of the p e l v i c skeleton to the t i p of the posterior process of the pe l v i c skeleton. 49. Width of Ventral Shield ( F i g . 3, character 49) i s the maximum width of the p e l v i c skeleton. 50. Length of Ascending Process ( F i g . 3, character 50) i s the distance from the a r t i c u l a t i o n of the ventral spine to the t i p of the ascending process of the pel v i c skeleton. 51. Number of Dorsal Spines. 52. Number of Dorsal Rays includes a l l unbranched and branched dorsal rays, does not include dorsal spines. 53. Number of Branched Caudal Rays. 54. Number of Unbranched Caudal Rays includes unbranched p r i n c i p l e caudal rays only, does not include rudimentary rays. 55. Total Number of P r i n c i p l e Caudal Rays a t o t a l of values for characters 53 plus 54. 56. Number of Dorsal Rudimentary Caudal Rays. 57. Number of Ventral Rudimentary Caudal Rays. 58. Number of Anal Rays includes a l l unbranched and branched anal rays, does not include anal spine. 172 59. Number of Ventral Rays includes a l l unbranched ventral rays, does not include v e n t r a l spine. 60. Number of Branched Pectoral Rays. 61. Number of Unbranched Pectoral Rays. 62. Number of Branches of Ascending Process includes a l l branches of the dorsal t i p of the ascending process of the p e l v i c skeleton which are at least as wide as t a l l . 63. Number of Precaudal Vertebrae includes a l l vertebrae without com-plete haemal spines. 64. Number of Caudal Vertebrae includes a l l vertebrae with complete or nearly complete haemal spines plus the hypural plate. 65. Total Number of Vertebrae a t o t a l of values for characters 63 plus 64. 66. Number of L a t e r a l Plates on Le f t Side a t o t a l of values f o r characters 69, 70, 71, 72 plus 73. 67. Number of L a t e r a l Plates on Right Side. 68. Total Number of L a t e r a l Plates a t o t a l of values for characters 66 plus 67. 69. Number of L a t e r a l Plates Preceding Ascending Process includes those plates on the l e f t side which precede but do not touch the ascending process of the p e l v i c skeleton. 70. Number of L a t e r a l Plates Touching Ascending Process includes only those plates on the l e f t side which touch the ascending process of the p e l v i c skeleton. 71. Number of L a t e r a l Plates Between Ascending Process and Last Precaudal Vertebra includes those plates on the l e f t side 173 posterior to the ascending process of the pelvic skeleton and an-terior to caudal vertebrae. 72. Number of Lateral Plates Between First Caudal Vertebra and Last Anal Pterygiophore includes those plates on the left side posterior to the last precaudal vertebra and anterior to the last anal ptery-giophore; unkeeled plates. 73. Number of Lateral Plates on Caudal Peduncle includes those plates on the l e f t side posterior to last anal pterygiophore; keeled plates. 74. Presence or absence of Posttemporal. 75. Presence or absence of Supracleithrum. 76. Total Number of Dorsal Plates a total of values for characters 77, 78, plus 79. 77. Number of Dorsal Plates Preceding DI includes those dorsal plates posterior to the tip of the dermosupraoccipital up to and including the dorsal plate supporting the f i r s t dorsal spine. 78. Number of Dorsal Plates Between DI and DII includes only those dorsal plates posterior to the dorsal plate supporting the f i r s t dorsal spine up to and including the dorsal plate supporting the second dorsal spine. 79. Number of Dorsal Plates Between DII and D i l i includes only those dorsal plates posterior to the dorsal plate supporting the second dorsal spine up to and including the dorsal plate supporting the third dorsal spine. 80. Number of G i l l Rakers includes a l l principle and rudimentary g i l l rakers on both the epibranchial and ceratobranchial of the f i r s t 174 arch on the le f t side. 81. Number of Pleural Ribs. 82. Placement of Most Anterior Pleural Rib scored as the number of the most anterior vertebra with an associated pleural rib. 83. Number of Epipleural Ribs. 84. Placement of Most Anterior Epipleural Rib scored as the number of the most anterior vertebra with an associated epipleural r i b . 85. Number of Epurals. 86. Number of Branchiostegal Rays. 87. Number of Pterygiophores Supporting D Spines. 88. Number of Pterygiophores Supporting D Rays does not include those pterygiophores supporting dorsal spines. 89. Number of Pterygiophores Supporting A includes those ptery-giophores supporting both the anal spine and a l l anal rays. 90. Otolith Count i s scored from the ground and polished sagittae. 91. Number of Epurals Fused Together a subset of the value of character 85. 92. Number of Epurals Fused to Penultimate Vertebra a subset of the value of character 85, including only those epurals fused to the neural arch of the last vertebra to precede the hypural plate. 93. Presence or Absence of Red Throat Pigmentation. 94. Presence or Absence of Deformed Vertebrae includes both deformed precaudal and caudal vertebrae. 95. Presence or Absence of Trematode Metacercariae i s scored from an external examination for large melanized black dots on any body surface. 175 96. Presence or Absence of Schistocephalus is scored from an ex-aminaton of the contents of the digestive tract and body cavity. 97. Presence or Absence of Cestode "type 2" Procercoids i s scored from an examination of the contents of the digestive tract and body cavity. 98. Presence or Absence of Proteocephalus is scored from an examina-tion of the contents of the digestive tract and body cavity. 99. Presence or Absence of Filocapsulariinine Nematodes is scored from an examination of the digestive tract, body cavity and muscle t i s -sue. 100. Presence or Absence of Eustrongylides is scored from an examina-tion of the digestive tract, body cavity and muscle tissue. 101. Presence or Absence of Neoechinorhynchus is scored from an ex-aination of the contents of the digestive tract. 176 APPENDIX 4. IDENTIFICATIONS OF PARASITES (CHARACTERS 95—101) Character 95 metacercariae of an unidentified trematode. Character 96 plerocercoid of a cestode; order Pseudophyllidea, family Diphyllobothriidae, Schistocephalus sp., probably ^. solidus. Character 97 procercoid of an unidentified cestode (cestode "type 2"). Character 98 adult cestode; order Proteocephalidea, family Proteocephalidae, probably Proteocephalus sp. Character 99 nematode; order Ascarididea, family Heterocheilidae, sub-family Filocapsulariinae. Character 100 nematode; order Dioctophymidea, family Dioctophymida, subfamily Hystrichinae, probably Eustrongylides sp. Character 101 acanthocephalan; order Neoechinorhynchidea, family Neoechinorhynchidae, subfamily Neoechinorhynchinae, probably Neoechinorhynchus sp. 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096279/manifest

Comment

Related Items