UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phylogenetic analysis of the cottid genus Artedius (Teleostei:Scorpaeniformes) Begle, Douglas P. 1984

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

Item Metadata

Download

Media
831-UBC_1984_A6_7 B45.pdf [ 7.84MB ]
Metadata
JSON: 831-1.0095983.json
JSON-LD: 831-1.0095983-ld.json
RDF/XML (Pretty): 831-1.0095983-rdf.xml
RDF/JSON: 831-1.0095983-rdf.json
Turtle: 831-1.0095983-turtle.txt
N-Triples: 831-1.0095983-rdf-ntriples.txt
Original Record: 831-1.0095983-source.json
Full Text
831-1.0095983-fulltext.txt
Citation
831-1.0095983.ris

Full Text

C. 1 PHYLOGENETIC ANALYSIS OF THE COTTID GENUS ARTEDIUS (TELEOSTEI:SCORPAENI FORMES) by DOUGLAS P. BEGLE B.Sc, Stanford University, 1980 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in 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 February 1984 (c) Douglas P. Begle, 1984 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e a n d s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may b e g r a n t e d b y t h e h e a d o f my d e p a r t m e n t o r b y h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t b e a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . D e p a r t m e n t o f T h e U n i v e r s i t y o f B r i t i s h C o l u m b i a 1956 Main M a l l V a n c o u v e r , C a n a d a V6T 1Y3 D a t e i i ABSTRACT The genus Artedius Girard (Cottidae:Scorpaeniformes) i s revised. Artedius hankinsoni i s synonymized with Artedius  l a t e r a l i s , since the defining characters of Artedius hankinsoni (Hubbs, 1926) f a i l to separate i t from Artedius l a t e r a l i s . Evidence of the diph y l e t i c condition of Artedius Girard i s presented. Based on the character analysis in t h i s study, five of the seven nominal species are hypothesized to represent a monophyletic group (A. c o r a l l i n u s , A. l a t e r a l i s , A. harringtoni, A. f e n e s t r a l i s , and A. notospilotus) hereafter referred to as Artedius sensu s t r i c t u . Two Artedius species, Artedius  creaseri and Artedius meanyi, do not share any of the synapomorphies of Artedius sensu s t r i c t u , therefore the genus Ruscarius Jordan and Starks is resurrected to include Ruscarius  meanyi and Ruscar ius c r e a s e r i . Evidence is presented for the monophyly of Ruscarius, which i s hypothesized to be more closely related to Chitonotus and Icelinus than i t is to Artedius. Phylogenetic analysis of the adult data (including Artedius, Ruscarius, 01igocottus, Clinocottus, Chitonotus, Orthonopias, Icelinus and Hemilepidotus) yielded three cladograms. These d i f f e r only in the placement of Chitonotus. The results from this study are compared with those of Bolin (1947). A l a r v a l data matrix for Washington's (1982) study i s reconstructed and three l a r v a l cladograms are presented. F i n a l l y , an analysis of taxonomic congruence i s attempted using the consensus technique of Adams (1972). Two trees were prepared, one showing the consensus of the adult trees, one showing the consensus of the l a r v a l trees. The l a r v a l consensus tree was then compared with the adult consensus tree to give an o v e r a l l consensus tree. This f i n a l consensus tree recognizes two generic groups: one composed of Artedius, Clinocottus and Oliqocottus, and one composed of Ruscarius and Icelinus. Thus both adult and l a r v a l c l a s s i f i c a t i o n s support the removal from Artedius of creaseri and meanyi to the genus Ruscarius in a clade with Icelinus separate from the clade including Artedius, 01iqocottus and Clinocottus. iv TABLE OF CONTENTS ABSTRACT i i LIST OF FIGURES v i i ACKNOWLEDGEMENTS ix INTRODUCTION 1 HISTORICAL REVIEW 2 METHODS 9 Adult Specimens 9 Larval Data 9 Merist i c s 10 Systematic Method 10 Comparison of C l a s s i f i c a t i o n s 18 Adults 18 Larvae 18 Taxonomic Congruence 19 The Genus Artedius Girard 1856 21 Taxonomic Revision 21 A r t i f i c i a l key to Artedius 21 Diagnosis of Artedius 22 Artedius c o r a l l i n u s 24 Synonymy 24 Diagnosis 25 Specimens Examined 25 Artedius f e n e s t r a l i s 26 Synonymy 26 Diagnosis 28 Specimens Examined 28 Artedius harringtoni 29 Synonymy 29 Diagnosis 31 Specimens Examined 31 Artedius l a t e r a l i s 32 Synonymy 32 Diagnosis 35 Specimens Examined : 35 Artedius notospilotus 36 Synonymy 37 Diagnosis 38 Specimens Examined 39 The genus Ruscarius Jordan and Starks 1895 40 Taxonomic Revision 40 A r t i f i c i a l key to Ruscarius 40 Ruscar ius diagnosis 41 Ruscar ius c reaser i 42 Synonymy 42 Diagnosis 43 Specimens Examined 43 Ruscar ius meanyi 44 Synonymy 44 Diagnosis 45 Specimens Examined 45 v i PHYLOGENETIC ANALYSIS 47 L i s t of Adult Characters 47 L i s t of Larval Characters 49 Adult Character Analysis 49 . RESULTS 82 Adult Trees 82 Monophyly and Relationships of Artedius 82 Monophyly of Ruscar ius 83 Alternative trees - the relationships of Ruscarius .. 84 Larval Trees 86 COMPARISON OF CLASSIFICATIONS 88 Adults 88 Larvae 91 CONGRUENCE OF LARVAL AND ADULT CLASSIFICATIONS 96 SUGGESTIONS FOR FUTURE STUDIES 100 LITERATURE CITED 102 Appendix 1. Adult Data Matrix 115 Appendix 2. Larval Data Matrix 116 v i i LIST OF FIGURES Figure 1. Artedius c o r a l l i n u s 118 Figure 2. Artedius f e n e s t r a l i s 120 Figure 3. Artedius harringtoni 122 Figure 4. Artedius l a t e r a l i s 124 Figure 5. Artedius notospilotus 126 Figure 6. Ruscar ius creaser i 128 Figure 7. Ruscar ius meanyi 130 Figure 8. Orthonopias t r i a c i s 132 Figure 9. Side view of scale ridge (schematic) 134 Figure 10. Chin pigmentation patterns 136 Figure 11. Top view of scale ridge (schematic) 138 Figure 12. Pterotic flange 140 Figure 13. Adult Wagner tree 1 142 Figure 14. Adult Wagner tree 2 144 Figure 15. Adult Wagner tree 3 146 Figure 16. Cladogram of Artedius sensu s t r i c t u 148 Figure 17. Cladogram of Ruscar ius 150 Figure 18. Adams consensus tree for adult trees 152 Figure 19. Larval Wagner tree 1 154 Figure 20. Larval Wagner tree 2 156 Figure 21. Larval Wagner tree 3 158 Figure 22. Adams consensus tree for l a r v a l trees 160 Figure 23. Phylogenetic tree of Bolin (1947) 162 Figure 24. Wagner tree calculated in the absence of v i i i Ruscarius meanyi 164 Figure 25. Dichotomous l a r v a l tree of Washington (1982). ..166 Figure 26. Polytomous l a r v a l tree from Washington (1982). .168 Figure 27. Adams consensus tree of adult and l a r v a l Adams trees 170 ix ACKNOWLEDGEMENTS I wish to thank the following for loaning specimens under their care: Dr. W.E. Eschmeyer (CAS), Dr. R. Rosenblatt (SIO), Dr. S.H. Weitzman (USNM), Dr. R.R. M i l l e r (UM), Dr. R. Lavenberg (LACM), Dr. D.E. McAllister (NMC) and Dr. T. Pietsch (UW). Mr. Robert Carveth was especially h e l p f u l regarding the UBC f i s h c o l l e c t i o n and freel y shared much of his extensive knowledge of Cottids. This study was begun under Dr. N.J. Wilimovsky who provided p a r t i a l f i n a n c i a l support. Dr. J.D. McPhail was kind enough to oversee the completion of the project and also provided some f i n a n c i a l support. The completion of thi s thesis was largely made possible by a University Graduate Fellowship from the University of B r i t i s h Columbia (1982-1984). Two people deserve special thanks. Dr. Daniel Brooks introduced me to phylogenetic systematics and provided friendship and encouragement at c r i t i c a l times throughout the study. Much of what I learned from him I learned outside of the classroom in innumerable impromptu discussions. I cannot thank him enough for his friendship and support throughout my stay at UBC. Kate Shaw also provided much-needed friendship and encouragement throughout the course of t h i s study. I thank her for helping me understand many topics in systematics and evolutionary theory and for uncounted hours spent just t a l k i n g . I also thank her for the countless hours of c r i t i c a l discussion about thesis topics. Kate Shaw and Andrew Simons were kind enough to read X e a r l i e r drafts of thi s thesis. The cladograms were prepared by Ms. Maggie Hampong. Most of a l l I thank my mother who provided unflagging support throughout my time at U.B.C, without which I would not have begun th i s enterprise, much less finished i t . $3.21, $3.23T $SIGNOFF 1 INTRODUCTION The C o t t i d a e a r e a S c o r p a e n i f o r m f a m i l y composed of s i x t y -seven genera ( N e l s o n , 1976). F o r t y of the s i x t y - s e v e n genera are found between the A l e u t i a n I s l a n d s and Ba j a C a l i f o r n i a . The taxonomic l i m i t s of the f a m i l y a r e p o o r l y d e f i n e d and i t s m o n o p h y l e t i c s t a t u s has not been demonstrated. A major f a c t o r c o n t r i b u t i n g t o t h i s c o n f u s i o n i s the absence of sound s y s t e m a t i c s t u d i e s of the genera, i n c l u d i n g taxonomic a n a l y s e s of the a p p r o p r i a t e s p e c i e s . Even i f the C o t t i d a e i s a n a t u r a l t a x o n , i t i s l i k e l y t h a t a number of genera a r e not m o n o p h y l e t i c and t h a t many s p e c i e s w i t h i n those genera a r e i n v a l i d . G i v e n t h a t our p r e s e n t knowledge of the genera i s r e l a t i v e l y i n c o m p l e t e , t h i s study has f o u r g o a l s : (1) a taxonomic r e v i s i o n of the genus A r t e d i u s , (2) d e f i n i t i o n of the g e n e a l o g i c a l r e l a t i o n s h i p s of the s p e c i e s w i t h i n A r t e d i u s , (3) an e x a m i n a t i o n of the r e l a t i o n s h i p of A r t e d i u s t o o t h e r C o t t i d g e n e r a , and (4) a t e s t of taxonomic congruence u s i n g data from l a r v a l and a d u l t s t a g e s . The genus A r t e d i u s G i r a r d i s d i s t r i b u t e d from the A l e u t i a n s I s l a n d s t o Ba j a C a l i f o r n i a . The genus i s w e l l - r e p r e s e n t e d i n museum c o l l e c t i o n s and a l a r g e c o l l e c t i o n of l a r v a e i s a l s o a v a i l a b l e f o r study-. 2 HISTORICAL REVIEW The genus Artedius was established by Girard in 1856 to accomodate two species of scaled c o t t i d s , Artedius l a t e r a l i s (previously described as Scorpaenichthys l a t e r a l i s by Girard in 1854), and Artedius notospilotus ( Calcilepidotus l a t e r a l i s Ayres 1854 and Hemilepidotus nebulosus Girard 1856 - though the l a t t e r was Ayres' unpublished name). In 1882, Jordan and Gilbert placed both Artedius l a t e r a l i s and Artedius notospilotus in the genus Icelus Kroyer, in the family Icelidae, along with other c o t t i d species which were "more or less scaly" (p. 683). In their description of A. notospilotus, they remark that the northern specimens "represent a d i s t i n c t variety." In 1883 they recognized this "northern variety" as a d i s t i n c t species, Artedius f e n e s t r a l i s . However, due to the large number of species that had been transferred to and from Artedius since Girard erected the genus, they despaired of separating Artedius from Icelus, and deferred to their previous (1882a) decision placing a l l Artedius species in Icelus. In 1895 Jordan and Starks described Ruscarius meanyi from Puget sound. Its a f f i n i t i e s apparently lay with Chitonotus Lockington 1882, from which i t was distinguished by a s c a l i e r back and weaker preopercular armature. Jordan and'Starks (1895) also removed Artedius f e n e s t r a l i s to a new genus, Astrolytes, based on i t s rougher, more scaly cranium and stronger preopercular armature. Artedius notospilotus was subsequently 3 transferred to Astrolytes in 1896 by Jordan and Evermann. In 1896 Starks described two new species, Artedius  asperulus and Axyrias harrinqtoni, both from Puget Sound. Bean and Weed (1920) later described a new genus and species from the same region, Pteryqiocottus macouni. This they suspected was the male of Axyrias harrinqtoni Starks, with which i t was synonymized by Bolin in 1944. Regan (1913), in his treatment of the Scleroparei (those fishes with the t h i r d suborbital extended backward onto the preopercle, forming a "suborbital stay"), placed both Artedius and Astrolytes in the family Cottidae and placed the family in the d i v i s i o n Cottiformes. No mention was made of Ruscar ius, but both Icelus and Icelinus were also included in the Cottidae. Within his series Cottiformes, Jordan (1923) r e s t r i c t e d the Cottidae to those genera that lacked scales. Thus Artedius, Astrolytes, Axyrias, Ruscarius, Ruscariops, and Pterygiocottus were united with Chitonotus, Orthonopias, Stelqistrum , Icelus, Icelinus, and twenty-seven other scaled genera and placed in the Icelidae. No new Artedius-1ike species were described u n t i l 1926 when Hubbs described three species that were apparently related to Artedius. He placed each species in a monotypic genus. He considered Parartedius hankinsoni (a single specimen from Point Loma, C a l i f o r n i a ) and A l l a r t e d i u s c o r a l l i n u s (also a single specimen, from Point Lobos, Monterey County) clo s e l y related to !a. The t h i r d species, Ruscar iops creaseri (from San Diego County), was related to Ruscar ius meanyi of Jordan and Starks 4 (1895). Hubbs favored monotypic genera and in 1926 he removed A. notospilotus from Astrolytes and placed i t into a new genus: Parastrolytes. In 1940, L.S. Berg reverted to Regan's (1913) treatment of Artedius and i t s r e l a t i v e s and placed them back in the Cottidae. The family Ice.l idae he reserved for Icelus. The remaining members of Jordan's Icelidae were reassigned to the Cottidae. One year l a t e r , in 1941, Taranets completed his comprehensive treatment of the anatomy and relationships of the Cottidae. He divided the Scleroparei into twelve families and erected within the Cottidae thirteen subfamilies. His subfamily Oligocottinae contained two large generic groups: those with bony plates (scales) and lacking an anal p a p i l l a ( A r t e d i i n i ) , and those with an anal p a p i l l a and lacking bony plates ( O l i g o c o t t i n i ) . The former included Artedius, A l l a r t e d i u s , Parastrolytes, Axyr ias and Orthonopias. Ruscarius and Ruscariops were removed to a subfamily, Icelinae, along with Icelus, Icelinus and Stelgistrum. This he j u s t i f i e d by their common possession of two rows of bony plates which occur "one along the l a t e r a l l i n e , and one at the base of the dorsal f i n . In addition, plates are usually found in other parts of the body." (p.4). This description is hardly accurate, as Ruscar ius, Ruscar iops, and S t e l g i strum a l l possess rows of scales covering the area between the l a t e r a l l i n e and the dorsal f i n bases. By his own d e f i n i t i o n , these three genera belong in the subfamily Oligocottinae along with Artedius and not in the I eelinae. 5 The last species to be assigned to the genus Artedius was A. d e l a c y i , described from Alaska by Hubbs and Schultz in 1941. They c i t e d i t s thicker l i p s and steeper snout as features distinguishing in from Artedius l a t e r a l i s which was not known from Alaska at that time. Except for Artedius which contained three species (A. asperulus, A. delacyi, and A. l a t e r a l i s ) , a l l the remaining genera were monotypic and remained separate u n t i l 1944, when Bolin gathered Ruscariops, Astrolytes, Parartedius, Parastrolytes and Axyr ias into Artedius, and placed Artedius  asperulus in synonymy with Artedius f e n e s t r a l i s . He retained as subgenera Ruscariops (A. creaseri) , Astrolytes, (A. f e n e s t r a l i s and A. notospilotus), Axyr ias (A. harrinqtoni) and Artedius (A. c o r a l l i n u s , A. l a t e r a l i s and A. hankinsoni). Neither Ruscar ius meanyi nor Artedius delacyi was included in the taxonomic analysis. Bolin (1947) presented a branching diagram of the C a l i f o r n i a members of the genus Artedius, including A. d e l a c y i . This i s the only phylogenetic dendrogram based on adult characters for Artedius and i t s close r e l a t i v e s (Fig. 23) . No further specimens of Ruscar ius meanyi were coll e c t e d u n t i l 1963, when Rosenblatt and Wilkie collected nine specimens in B r i t i s h Columbia. Redescribing the species, they referred i t to Artedius Girard on the basis of Bolin's r e d e f i n i t i o n of Girard's genus. It was subsequently co l l e c t e d from Alaska (Peden and Wilson, 1976), B r i t i s h Columbia (Peden, 1972), Puget Sound (Moulton, 1977) and C a l i f o r n i a (Lea, 1974). 6 Since Hubbs and Schultz' (1941) defining characters f a i l e d to separate putative A. delacyi from Alaskan A. l a t e r a l i s , Hubbard and Reeder (1965) and Quast (1968) concluded that Artedius delacyi was not d i s t i n c t from Artedius l a t e r a l i s . Recent c l a s s i f i c a t i o n schemes have d i f f e r e d l i t l e from that of Berg (1940). Nelson (1976) and Greenwood et a l . (1966) both place Artedius in the Cottidae along with 01igocottus Girard and Clinocottus G i l l , while Icelus i s retained as the only member of the Icelidae. In recent years a great deal of information on the morphology of l a r v a l c o t t i d s has accumulated, but only two studies have attempted to use these l a r v a l data in reconstructing phylogeny. Richardson (1981) recognized six phenetically derived groups within the c o t t i d genera for which larvae are known. Her "group one" includes Artedius, Orthonopias, Clinocottus and 01igocottus. Icelus and Icelinus were included in "group two", along with Par i c e l i n u s , Triglops  and Chitonotus. These groupings were primarily based on shared states of preopercular spines, snout and body shape, pigmentation, and di f f e r e n t d i v e r t i c u l a of the gut. These character states do not persi s t in adults. Meristic characters, which do persi s t in adults, were used for the purpose of ident i f icat ion. Richardson's (1981) "group one", including Artedius, was characterized by a unique preopercular spine pattern, in a l l li k e l i h o o d synapomorphic for that group, i f the shared states are indeed homologous. However, no attempt was made to 7 e s t a b l i s h any inter-generic relationships. Washington (1982) did attempt such an analysis for Artedius, CIinocottus and Oligocottus. She suggested that, on the basis of an 1,2 pelvic state (one spine, two soft rays), A. meanyi should be removed to Icelinus, which also has an 1,2 state. A l l other Artedius have an 1,3 state. Under th i s scheme, A. creaseri would then become the s i s t e r group of an A. meanyi-Ieelinus clade and an A. meanyi-Ieelinus-A. creaseri clade would be the s i s t e r group of Artedius (less A. creaseri and A. meanyi) plus Oligocottus plus CIinocottus. Within the genus Artedius three problems have emerged. F i r s t , there has been some d i f f i c u l t y in recognizing v a l i d species. Usually t h i s has been resolved by closer study and the only species now in question i s Artedius hankinsoni. The status of t h i s species i s discussed l a t e r in this study. Second, there has been some question as to whether or not these species constitute a monophyletic group. Third, the r e l a t i o n of that group to Artedius and other scaled Cottid genera has not been resolved. The last two problems have remained unsolved primarily because researchers have attempted to j u s t i f y groups on the basis of plesiomorphic characters. Witness Bolin's (1947) l i s t of characters defining Artedius: "comparatively large head, normal structure of the pelvic f i n s and by the unadvanced anus." These characters, taken as whole, supposedly serve to d i f f e r e n t i a t e the genus at hand from other genera defined in a similar fashion. However, no one synapomorphic character has 8 ever been proposed to serve as j u s t i f i c a t i o n for the monophyly of Artedius• This omission has resulted i n : (1) the transfer of species to and from Icelus; (2) the suggestions (Bean and Weed, 1920, Howe and Richardson, 1978) that Orthonopias should ac t u a l l y be included in Artedius; and (3) the r e f e r r a l of Ruscarius meanyi to Icelinus by Washington (1982). When no synapomorphy i s recognized d i f f e r e n t groupings and schemes of relationship cannot objectively be defended (Wiley, 1981). The obvious solution i s to test the hypothesis of monophyly of Artedius by searching for j u s t i f i c a t i o n in the characters (Hennig, 1965, 1966; Rosen, 1974; Wiley, 1975, 1979, 1981, 1982). A natural group must be supported by synapomorphies at the appropriate l e v e l of u n i v e r s a l i t y . Plesiomorphic characters w i l l not do, i f the delineation of natural taxa is our goal (Nelson, 1971, 1972, 1973, 1979), as they inevitably lead to paraphyletic taxa ( F a r r i s , 1979a,b). Such taxa are classes and therefore timeless abstractions. Such taxa cannot p a r t i c i p a t e in natural processes and should be avoided by those attempting to discover the pattern of natural order (Wiley, 1977, 1979, 1981, 1982; Rosen 1974). 9 METHODS Adult Specimens A l l specimens were preserved in either isopropanol (37.5% or 50%) or ethanol (70%). Cleared and stained specimens were prepared using the method of Dingerkus and Uhler (1977). Cleared and stained specimens were stored in 100% glycerin with thymol added to retard decomposition. Photographs of cleared and stained f i s h were made through a Bausch and Lomb dissecting microscope, with a Nikon F 35 mm camera and Nikon microscope adapter. Institutions providing specimens were: C a l i f o r n i a Academy of Sciences (CAS and SU), Los Angeles County Museum of Natural History (LACM), Scripps I n s t i t u t i o n of Oceanography Fish Div i s i o n (SIO), University of B r i t i s h Columbia Fish Museum (BC), National Museum of Canada (NMC), University of Michigan Museum of Zoology (UMMZ), University of Washington Fish Collection (UW), and the United States National Museum (USNM). Larval Data Information on l a r v a l morphology was taken from Washington 10 (1982). Larval specimens were not examined in the present study. Me r i s t i c s For both preserved and cleared and stained specimens a l l elements were counted. The one exception was the caudal f i n count which included only p r i n c i p a l rays. Systematic Method "If we find that the ascertainment of the order of nature is f a c i l i t a t e d by one terminology or one set of symbols rather than another, i t is our clear duty to use the former." T.H. Huxley, quoted in Jordan and Kellogg, 1907. In constructing a general reference system (Hennig, 1966) for comparative biology (sensu Nelson, 1970) we must choose that method which best allows us to infer the h i s t o r i c a l course of evolution. Phylogeny i s the history of l i f e and that i s what we are in business to discover. This involves specifying the genealogical relationships of natural taxa. Obviously, the primary problem is j u s t i f y i n g the natural status of the groups 11 at hand. As Wiley (1979) noted, the only necessary and s u f f i c i e n t c r i t e r i o n of natural status i s unique genealogy. Characters, which are our only d i r e c t l y observable evidence, are only secondary r e f l e c t i o n s of unique h i s t o r i c a l r elationships. If we are to link characters to genealogy we need bridge p r i n c i p l e s (sensu Hempel, 1965) since characters exhibit two phenomena precluding their use as necessary and s u f f i c i e n t c r i t e r i a of a group's natural status: ontogeny and character modification during phylogenetic descent (Wiley, 1978). If a l l characters . remained unchanged throughout ontogeny and through the course of phylogenetic descent, they would constitute not only necessary but also s u f f i c i e n t evidence for the natural status of a group. Since they can be modified during the course of ontogenesis and phylogenesis we need bridge p r i n c i p l e s to move from that which we observe (characters) to that which we wish to infer (genealogy). Put simply, characters cannot be divorced from the concept of evolutionary homology (Wiley, 1981). Our method should involve a search for evolutionary novelties which serve to define natural (monophyletic) taxa (Hennig, 1966, Wiley, 1979, 1981). Such characters are termed apomorphies i f , at the level of analysis under study, they are hypothesized to be the derived state of an ancestral, r e l a t i v e l y generalized homologue. An apomorphic character found in only one taxon is termed an autapomorphy. Shared apomorphies are termed synapomorphies. Only synapomorphies may be used to delimit natural, monophyletic groups. 1 2 The r e l a t i v e l y generalized state of that homologue is termed a plesiomorphy and when shared by two or more taxa, a symplesiomorphy (Hennig 1966). Notice that a l l characters when they are evolutionary novelties, delimit a natural, monophyletic group. Thus at some l e v e l of un i v e r s a l i t y a l l character states are apomorphic, delimiting a natural taxon. They cannot be used to diagnose other taxa at a l e v e l of un i v e r s a l i t y at which they are not apomorphic (Hennig, 1966, Wiley, 1979, 1981). Note that apomorphy and plesiomorphy are r e l a t i v e terms and are meaningful only i f the l e v e l of un i v e r s a l i t y being considered is s t r i c t l y s p e c i f i e d . The c r i t i c a l part of any such analysis is the determination of hypothesized plesiomorphic and apomorphic states for each character. Hennig (1966) discusses four methods: chorological (biogeographical), holomorphological (out-group comparison), geological precedence and ontogenetic precedence. Of these four, the f i r s t and t h i r d have generally been discarded due to their reliance on a p r i o r i assumptions about the pattern of character change in time and space. These assumptions may or may not be j u s t i f i e d , depending on the pa r t i c u l a r circumstances. Because of t h i s , outgroup and ontogenetic analyses are preferred. In outgroup analysis, character states shared with the outgroup are judged to be plesiomorphic at that l e v e l , while the alternative state i s hypothesized to be apomorphic. In Nelson's (1978) formulation, the ontogenetic c r i t e r i o n follows from von Baer's law which states that r e l a t i v e l y generalized states give rise to specialized ones as ontogeny progresses. 1 3 This method has two immediate drawbacks: non-terminal addition of character states and instances of d e - d i f f e r e n t i a t i o n e f f e c t i v e l y refute von Baer's "law" making the attendant character analysis suspect (Brooks, 1984 in press; Kluge, 1984 in press). Nelson (1978) claimed, in his formulation, that ontogenetic analysis can proceed independently of outgroup comparison, but Wiley (1980) noted that in cases of non-terminal addition meaningful character analysis can only be attained by resort to outgroups. Fink (1982) pointed out that paedomorphosis can further confound character analysis by making derived characters appear primitive, again necessitating the use of outgroups as a check. Polarizing characters becomes more d i f f i c u l t when two or more apomorphic states exist at a given l e v e l of analysis. These multistate characters present special d i f f i c u l t i e s in establishing transformation series. Mickevich (1983) distinguished four types of multistate characters based upon their d i s t r i b u t i o n on a previously-derived cladogram. These are: 1. Additive characters, for which F a r r i s optimization (a parsimony technique - see F a r r i s 1974 for a description) is possible. These allow construction of a character state tree. 2. Non-additive characters, in which optimization leads to multiple p o s s i b i l i t i e s for transformation series. At best these allow resolution of subsections of a tree. 3. Convergent characters, in which i t is equally l i k e l y that one state was derived from d i f f e r e n t conditions. 1 4 These are of minimal use in the resolution of trees. 4. D i s j o i n t characters, in which no transformation can be hypothesized to link whole subsets of character states, although transformation series may be established within the individual subsets. These may allow sections of the tree to be resolved, but the relationships between these sections remain ambiguous. Once character analysis i s complete a cladogram may be constructed, l i n k i n g s i s t e r taxa (those hypothesized to share an immediate common ancestor) by means of synapomorphies. Rarely do a l l the characters indicate the same relationships due to homoplasy (non-homologous s i m i l a r i t y ) . In such an event, the cladogram with the least number of c o n f l i c t i n g characters (the fewest number of steps or character state changes) i s chosen as best representing the data; i t is the least refuted hypothesis. Parsimony considerations force us to minimize the number of c o n f l i c t s , each of which requires an ad hoc explanation. This leads to a preference for the tree with the minimum possible number of steps for a p a r t i c u l a r data set. It is precisely t h i s parsimony requirement that " c l a s s i c a l " evolutionary taxonomy of Mayr (1942, 1969) and Simpson (1945, 1961) f a i l s to s a t i s f y . Their reliance on overa l l s i m i l a r i t y and adaptive divergence results from attempts to infer a pattern (phylogeny) from a p a r t i c u l a r process (Darwinian evolution). The use of o v e r a l l s i m i l a r i t y f a i l s to separate plesiomorphic from apomorphic information and, as a result, characters are not applied s t r i c t l y at the level at which they constitute a 15 c r i t i c a l test of the hypothesis at hand (Wiley 1981). The notion of process was the mold into which the information on pattern was f i t t e d , r esulting in paraphyletic groups (grades) which are not defensible constructs in a system which attempts to reconstruct the genealogical history of monophyletic groups. We must then ask: If we are seeking to discover the order in nature, i s t h i s the appropriate methodology? This pattern/process dichotomy in systematics is r e f l e c t i v e of a higher-level structure/function controversy (see Eldredge and Cracraft, 1981). The question that arises i s , can an understanding of a p a r t i c u l a r process be attained without prior examination of the parts (patterns) involved? Churchland (1982) thought not; though his discussion centers on the study of cognition, he finds an isomorphism with natural science and in both, function and process are to be studied as emergent properties with an understanding of them contingent upon a p r i o r i s t r u c t u r a l analysis. Churchland (1982) discussed the f a i l i n g s of a f u n c t i o n a l i s t approach which he labels "top-down". He considered such an approach to be faulty for two reasons. F i r s t of a l l , i t aims at analyzing the common attributes (functions) of a group of e n t i t i e s without reference to the factors which produced them. Secondly, i t may focus our explanatory e f f o r t s on perceived groupings which may in fact be unrelated to the properties (structures, and by extension, genealogy) of their members. "In the event, and as in our own history, such inquiry 1 6 (bottom-up) deepens, and changes, our i n i t i a l conception of what i t i s we should be trying to simulate. But i f such inquiry is denied any central significance, and i f our primitive i n i t i a l conceptions are e x p l i c i t l y given the a p r i o r i status of a s t i p u l a t i v e functional s p e c i f i c a t i o n of the domain of phenomena at issue, then we threaten to make a prison of our own ignorance." (Churchland, 1982, i t a l i c s h i s ) . Within a f u n c t i o n a l i s t paradigm, systematic information cannot be used for c r i t i c a l tests of competing hypotheses of process, since these patterns already r e f l e c t the process-level bias of the systematist. C l a d i s t i c analysis, on the other hand, does not assume any p a r t i c u l a r process of change, only that change (descent with modification) occurs and that characters are passed on, modified or not (Wiley, 1981 p. 78). Characters are used to delimit natural groups only at the l e v e l at which they are apomorphic. Notions of functional e f f i c i e n c y (adaptiveness) are never used to j u s t i f y the natural status of a p a r t i c u l a r group. Another exi s t i n g method in use in systematics today has come to be known as phenetics (Sneath and Sokal, 1973). I r o n i c a l l y , since i t does not pretend (usually) to estimate phylogeny, t h i s amalgam of numerical techniques has f a l l e n into the same trap (grouping by o v e r a l l s i m i l a r i t y ) as has evolutionary taxonomy, which does claim to reconstruct phylogeny. Since phenetic techniques arose from pre-existing s t a t i s t i c a l methods never intended for use in b i o l o g i c a l 17 systematics, i t i s hardly surprising that they f a i l to separate the derived from the generalized components in their measures of ov e r a l l s i m i l a r i t y . If the derived component i s not separated, the genealogy of l i f e w i l l escape largely unnoticed (Farris 1977). Only i f there is no homoplasy and homogeneous rates of change during any given time period w i l l phenetic analysis y i e l d the same c l a s s i f i c a t i o n as a c l a d i s t i c analysis. C l a d i s t i c analysis on the other hand, i s a powerful tool for estimating phylogeny even in the presence of homoplasy and d i f f e r e n t i a l rates of change. The Wagner program •in the PHYSYS package of F a r r i s and Mickevich (1983) was used to analyze the data matrix obtained by Hennigian character analysis. It i s e n t i r e l y phylogenetic in nature and i s described in d e t a i l in F a r r i s (1970) and F a r r i s , Kluge and Eckardt (1970). In any event, the results of a Wagner analysis and hand computation are invariably the same. The program was used in th i s study in two instances: to check the cladogram derived by hand and to re-analyze the re-coded l a r v a l data of Washington (1982). These two results were then used to generate a consensus tree (sensu Adams, 1972) in an analysis of taxonomic congruence. 18 Comparison of C l a s s i f i c a t i o n s Adults A phylogenetic tree was extracted from Bolin's (1947) analysis of the Cottidae. Two measures were employed to assess the tree's v a l i d i t y as an hypothesis of phylogeny. F i r s t l y , the character data from t h i s study were mapped onto Bolin's (1947) tree and the number of steps needed to account for the data counted. Secondly, since i t i s not possible to d i r e c t l y compare a c l a s s i f i c a t i o n with a data matrix from which i t was not constructed, the data matrix from t h i s study was converted to a Manhattan distance matrix (where the the Manhattan distance between two taxa is the number of character state differences between taxa divided by the number of characters). Both my Wagner tree and Bolin's tree were assessed for goodness of f i t by c a l c u l a t i n g the cophenetic co r r e l a t i o n c o e f f i c i e n t (R). Since Bolin did not include Ruscarius meanyi in his analysis this species was deleted for the purposes of comparison from the present analysis. Larvae Before a congruence study could be attempted i t was necessary to ensure that the best tree for the l a r v a l data had 19 been calculated. Sat i s f y i n g t h i s desideratum was i n i t i a l l y d i f f i c u l t since Washington (1982) published no data matrix and one had to be reconstructed from the body of the text. A Wagner analysis was then performed on the reconstructed data matrix and the re s u l t i n g tree compared with two trees derived from Washington's (1982) study: the dichotomous cladogram which was presented and the implied polytomous tree which resulted from collapsing a l l of the unsupported branches to the node below. These trees were then compared in the same way as the adult trees ( i . e . by counting the number of steps and by f i t to the Manhattan distance matrix). Taxonomic Congruence If there i s t r u l y only one genealogy of l i f e , then congruence w i l l necessarily be a consequence of any method which sets out l o g i c a l l y to discover that history, though d i f f e r e n t data sources may allow reconstruction of that history to greater or lesser degrees. Taxonomic congruence, as defined by Mickevich (1978), provides a measure of the degree to which the " c l a s s i f i c a t i o n s of the organisms remain stable as various l i n e s of evidence are considered". This s t a b i l i t y is a consequence of c l a s s i f i c a t o r y method and i s often c i t e d as being the sole property of one or another competing algorithm ( F a r r i s , 1971, 1982; Fink 1979, Mickevich 1978, 1980). The importance of a stable c l a s s i f i c a t i o n l i e s in hypothesis testing: new sources of information should never be interpreted as refuting previous i \ 20 hypotheses of relationship unless the new information is actually contradictory (Farris 1971; Mickevich 1978, 1980). These d i f f e r e n t sources may be of the same type of data from d i f f e r e n t l i f e stages, such as larvae and adults, or from di f f e r e n t types of characters, such as behavior and morphology. Taxonomic s t a b i l i t y i s also a requisite property of methods claiming to provide the best general reference system for comparative evolutionary biology (Mickevich, 1978), one that represents a l l characters as well as possible (Fink, 1979; Mickevich, 1980). "Without congruence tests there is no way of determining whether c l a s s i f i c a t i o n represents a repeatable, natural order, or instead portrays only the idiosyncracies of one suite of observations" (Mickevich 1980). There have been a few such tests of congruence. To date, these have shown that phylogenetic systematics provides the most stable method for finding maximum congruence (Andersen, 1979; Baird and Eckhardt, 1972; Baverstock et a l . , 1979; Hood and Smith, 1982; Jensen and Barbour, 1981; Mickevich, 1978, 1980; Mickevich and F a r r i s , 1981; Mickevich and Johnson, 1976; Miyamoto, 1981; Mundinger, 1979; Schuh and F a r r i s , 1981; Shuh and Polhemus, 1980). This study attempts a congruence analysis using the l a r v a l data of Washington (1982) and the present analysis of adults. 21 THE GENUS ARTEDIUS GIRARD 1856 Taxonomic Revision Artedius Girard, 1856, p. 134. (genotype by subsequent designation of Jordan and Evermann 1896 Scorpaenichthys l a t e r a l i s Girard) Astrolytes Jordan and Starks, 1895 Axyr ias Starks, 1896 Pterygiocottus Bean and Weed, 1920 Al l a r t e d i u s Hubbs, 1926 Parartedius Hubbs, 1926 Parastrolytes Hubbs, 1926 A r t i f i c i a l key to Artedius 1a. Scales present on the occiput 2. 2a. Preorbital c i r r u s present Artedius harringtoni. 2b. No preorbital c i r r u s 3. 22 3a. Scales extend under anterior orb i t Artedius f e n e s t r a l i s . 3b. No scales under anterior orbit Artedius notospilotus. 1b. No scales on the occiput 4. 4a. C i r r i present above upper l i p Artedius c o r a l l i n u s . 4b. No c i r r i above upper l i p Artedius l a t e r a l i s . Diagnosis of Artedius Recall that Bolin characterized Artedius (the only such attempt at diagnosis on record) as follows: "comparatively large head, normal structure of the pelvic f i n s and by the unadvanced anus." (1947, p. 161) Even s l i g h t f a m i l i a r i t y with Cottids equips one to f i t l i t e r a l l y dozens of genera into the above description. In th i s l i g h t , the following i s a l i s t of diagnostic characters apomorphic for Artedius sensu s t r i c t o alone. 23 3. Scale ridge curving to become p a r a l l e l with basal plate. 4. Generally darker background pigmentation of l a t e r a l body surface interrupted by c i r c u l a r areas of l i g h t e r pigmentation (same as ventral body surface, not white spots of Orthonopias). These l i g h t e r c i r c l e s larger closer to ventral edge of dark areas. Largest, at margin, incomplete, producing a scalloped area above anal f i n . Margins of each c i r c l e of l i g h t pigmentation c r i s p and well-defined. 12. In preserved specimens, a pattern of c i r c u l a r blotches of l i g h t background pigmentation interrupting darker pigmentation of ventral surface of head, extending backwards onto branchiostegal membrane. 27. Postcleithra absent. 35. Scale ridge a small semicircle. 45. Triangular flange at posterior edge of pterotic long, extending at least to posterior edge of cranium. 24 Artedius c o r a l l i n u s Artedius c o r a l l i n u s (Hubbs 1926) (Fig. 1) Range: Orcas Island (Washington) to Baja C a l i f o r n i a . Habitat: Rocky i n t e r t i d a l areas to 70 f t . Synonymy Alla r t e d i u s c o r a l l i n u s Hubbs, 1926, p. 8; Point Lobos, Monterey County, C a l i f o r n i a . Artedius c o r a l l i n u s Bolin, 1937, p. 63 (description, f i r s t Baja record); Bolin, 1944, p. 53, f i g . 20 (in key, description, d i s t r i b u t i o n , r e l a t i o n s h i p s ) ; M i l l e r and Lea, 1972, p. 126, f i g . p. 126 (in key, also d i s t r i b u t i o n ) ; F i t c h and Lavenberg, 1975, p. 128 (name only); Howe and Richardson, 1978, p. 12 (in key, plus diagnosis, synonymy, meristic v a r i a t i o n ) ; Hubbs, F o l l e t t and Dempster, 1979, p. 18 (name only - C a l i f o r n i a c h e c k l i s t ) ; Eschmeyer, Herald and Hamman, 1983, p. 161, p i . 18. 25 Diagnosi s 20. Cardiform teeth on dentaries, premaxillae, palatines and vomer. 26. Two to sixteen simple paddle-shaped f l a t c i r r i just above upper l i p along i t s length. These located anywhere from end of maxilla forward, not di s t r i b u t e d evenly on each side of head. Usually two small c i r r i on either side of premaxillary symphysi s. 22. 43-46 rows of scales along body, extending from o r i g i n of f i r s t dorsal f i n to base of ultimate or penultimate dorsal soft ray. Spec imens Examined UMMZ 14168 1 (holotype) Monterey, Pt. Lobos; CAS 19453 1 Mexico:Baja; CAS uncat. 1 CA:Monterey; CAS W51-74 1 CA:Palos Verdes; CAS 28838 1 CA:Mendocino; CAS uncat. 1 Mexico:Baja; CAS 39853 1 CA:Monterey; CAS Acc.#1972:23 1 CA:Monterey; CAS 48958 2 CA:Orange Co.; SU 29536 1 CA:Monterey; SU 48970 7 CA:Orange County; SU 35365 1 CA:Monterey; SU 40881 1 CA:Pt. Lobos; LACM W53-395 1 Mexico:Coronado Island; LACM 1989 2 Mexico:San Martin Island; LACM W70-16 48 CA:Monterey; LACM W65-26 13 CA:Ventura Co.; LACM 9421-6 9 Mexico:San Nicolas Island; LACM 31301-1 34 CA:Diablo Cove; LACMW67-152 2 Mexico:Punta Banda; LACM 31937-8 1 CA:Humboldt Co. 26 Artedius f e n e s t r a l i s Artedius f e n e s t r a l i s (Jordan and Gi l b e r t 1882) (Fig. 2) Range: Aleutian Islands to southern C a l i f o r n i a . Habitat: I n t e r t i d a l to 180 feet. Synonymy Artedius notospilotus Jordan and Jouy, 1882, p. 6 (No. 27416); Bean, 1882a, p. 250; 1882b, p. 471 (not of Girard). Icelus notospilotus Jordan and G i l b e r t , 1882b, p. 690. ( the "northern variety", not of Girard). Icelus f e n e s t r a l i s Jordan and G i l b e r t , 1882b, p. 973. (no l o c a l i t y given) Artedius f e n e s t r a l i s Jordan and G i l b e r t , 1883, p. 577 (Commencement Bay, Washington); Jordan, 1887, p. 898 (name only); Bolin, 1944, p. 48, fig.18 (description, d i s t r i b u t i o n and r e l a t i o n s h i p s ) ; Wilimovsky, 1954, (name only - Alaska c h e c k l i s t ) ; M c A l l i s t e r , 1960, p. 41 (name only--from Clemens and Wilby 1949); Clemens and Wilby, 1961," p. 298, f i g . 186 ( l i f e history, d i s t r i b u t i o n ) ; MacPhee and Clemens, 1962, p. 33 (depth d i s t r i b u t i o n ) ; Wilimovsky, 1963, p. 182 (name only -Alaska c h e c k l i s t ) ; Delacy, M i l l e r , and Borton, 1972, p. 15 27 (name only); M i l l e r and Lea, 1972, p., 126. F i g . p. 127 (in key and d i s t r i b u t i o n ) ; Quast and H a l l , 1972, p. 20 (name only -Alaska c h e c k l i s t ) ; Blackburn, 1973, (larvae: Cottid 4); Hart, 1973, p. 478 ( l i f e history, d i s t r i b u t i o n ) ; F i t c h and Lavenberg, 1975, p. 128 (name only); Peden and Wilson, 1976, p. 235 (d i s t r i b u t i o n - B r i t i s h Columbia); Richardson and Pearcy 1977, (larvae: Artedius sp. 2); Howe and Richardson, 1978, p. 13 (in key, plus diagnosis, synonymy, and meristic v a r i a t i o n ) ; Hubbs, F o l l e t t and Dempster, 1979, p. 19 (name only - C a l i f o r n i a c h e c k l i s t ) ; Pearcy and Myers, 1979, p. 212 (larvae - Yaquina Bay, Oregon); Richardson and Washington, 1980, (larvae: Artedius sp. 2); Washington, 1982, (larvae); Eschmeyer, Herald and Hamman, 1983, p. 162, p i . 18. Astrolytes f e n e s t r a l i s Jordan and Starks, 1895, p. 807 (Puget Sound); Jordan and Evermann, 1898a, p. 1899 (in key, and d i s t r i b u t i o n ) ; Jordan and Gilbert 1899, p. 456 (name only); Gilbe r t and Thompson, 1905, p. 977 (description, synonymy); Evermannn and Goldsborough, 1907, p. 298 (d i s t r i b u t i o n Alaska); Starks, 1911, p. 188 (description, r e l a t i o n s h i p s ) ; Gilbert and Burke, 1912, p. 36; Kincaid 1919, p. 30 (in ch e c k l i s t ) ; Hubbs 1926, p. 1. Artedius asperulus Starks, 1896, p. 553, ( v i c i n i t y of Port Ludlow, Washington); Jordan and Evermann, 1898, p. 1903 Astrolytes notospilotus Jordan and Evermann, 1900, f i g . 689a. (not of Girard) 28 Diagnosis 16. Three to ten simple short c i r r i in a li n e on top of each of transverse head tubercles. 17. No c i r r i located along suborbital stay. 28. A single row of scales extending forward to a point under anterior margin of o r b i t . 30. Body scales continuing onto the dorsal surface of the caudal peduncle, covering area extending from end of second dorsal base to o r i g i n of caudal f i n . Spec imens Examined USNM 27206 1 (Holotype) WA:Puget Sound; BC 63-252 2 Alaska; BC 63-245 3 Alaska; BC 53-74 12 Burrard Inlet; BC 63-154 4 Alaska; BC 62-989 5 Alaska; BC 62-555 4 Alaska; BC 60-226 3 BC:Vancouver Id.; BC 62-291 2 BC:Gardner Canal; BC 62-589 7 Alaska; BC 61-501 1 Alaska:Lynn Canal; BC 53-232a 1 BC:Joassa Channel; BC 53-232b 1 BCrJoassa Channel; BC 59-484 2 Alaska:Zachar Bay; BC 53-210 2 BC:Gale Passage;,BC 61-497 3 Alaska:Auke Bay; BC 62-574 2 AlaskarPt. Armstrong; BC 61-495 2 Alaska:Auke Bay? BC 54-447 6 BC:Saturna Id.; BC 53-86 5 BC:English Bay; BC 55-359 2 WArSan Juan Id.; BC 62-881 6 BC:Sooke; BC 62-637 7 BC:Howe Sd.; BC 62-731 2 BC:Saxe Pt.; BC 53-156 3 BC:Echo Bay; CAS 17765 6 CA:Sonoma,Bodega Bay; SU 16676 15 CArDel Norte Co.; CAS 40354 7 29 Oregon; CAS 16662 10 CA:Del Norte Co.; CAS 40403 3 OR:Cape Perpetua; SU 40882 1 CArHumboldt Bay; LACM 3.1700-4 5 CA:Diablo Cove; LACM 4339 1 WArSkagit Co.; LACM 4391 1 WArSkagit Co.; UW 17208 4 WA:San Juan Island; UW 5450 15 WA:Edmonds; UW 980 5 WA:West Seattle. Artedius harringtoni  Artedius harrinqtoni (Starks 1896) (Fig. 3) Range: Kodiak Island to southern C a l i f o r n i a Habitat: I n t e r t i d a l to 70 feet, usually in rocky areas, esp e c i a l l y abundant in kelp beds. Synonymy Axyrias harrinqtoni Starks, 1896, p. 554, p i . 74 ( v i c i n i t y of Port Ludlow, Washington); Jordan and Evermann 1898a, p i . 904 (in key, and d i s t r i b u t i o n ) ; Kincaid, 1919, p. 30 (in c h e c k l i s t ) ; Schultz and DeLacy, 1936, p. 78 (annotated che c k l i s t ) . Axyr ias harringtoni i Bean and Weed, 1920, p. 72 30 (except mature males). Pteryqiocottus macouni Bean and Weed, 1920, p. 73, p i . 3 (Ucluelet, B r i t i s h Columbia); Jordan, 1923, p. 212 (name only). Artedius harringtoni Bolin, 1944, p. 45, f i g . 17 (description, d i s t r i b u t i o n ) ; M c A l l i s t e r , 1960, p. 41 (name only--from Clemens and Wilby 1949); Clemens and Wilby, 1961, p. 296. f i g . 185 ( d i s t r i b u t i o n , l i f e h i s t o r y ) ; MacPhee and Clemens, 1962, p. 33 (depth d i s t r i b u t i o n - Puget Sound); DeLacy, M i l l e r and Borton, 1972, p. 15 (name only); M i l l e r and Lea, 1972, p. 128, f i g . p. 128 (in key, and d i s t r i b u t i o n ) ; Quast and Hall,1972, p. 20 (name only - Alaska c h e c k l i s t ) ; Blackburn, 1973, (larvae: Cottid 6); Hart, 1973, p. 473 ( l i f e h istory, d i s t r i b u t i o n ) ; F i t c h and Lavenberg, 1975, p. 128 (name only); Peden and Wilson, 1976 p. 235 ( d i s t r i b u t i o n - B r i t i s h Columbia); Richardson, 1977, (larvae: Artedius sp.1); Richardson and Pearcy, 1977, (larvae: Artedius sp. 1); Howe and Richardson, 1978, p. 14 (in key, plus diagnosis, synonymy, and meristic v a r i a t i o n ) ; Hubbs, F o l l e t t , and Dempster, 1979, p. 19 (name only - C a l i f o r n i a c h e c k l i s t ) ; Pearcy and Myers, 1979, p. 212 (larvae - Yaquina Bay Oregon); Richardson, Laroche, and Richardson, 1980, ( l a r v a l d i s t r i b u t i o n - Oregon Coast); Richardson and Washington, 1980, (larvae); Washington, 1982, (larvae); Eschmeyer, Herald and Hamman, 1983, p. 162, p i . 18. 31 Diagnosis 20. In adult males, teeth on vomer and palatines cardiform; those on dentaries and premaxillae i r r e g u l a r l y canine, with numerous d i s t i n c t i v e l y longer canine teeth on both upper and lower jaws. 21. Seven branchiostegal rays, one additional on ceratohyal. 22. 43 to 46 scale rows above l a t e r a l l i n e . 23. Incised membrane of anal f i n convex in males, concave in females. 24. In males, anal f i n covered with an hexagonal latticework of l i g h t pigmentation on a darker background. 25. Penis present, in form of small cone. 30. Dorsal body scales continuing onto dorsal surface of caudal peduncle. 31. One very large c i r r u s anterior to o r b i t , plumose in mature ma1e s . 32. No c i r r i in the region between the pectoral f i n base and the l a t e r a l l i n e . 34. Seven to twelve simple c i r r i along preopercular margin, not confined to preopercular spines. Specimens Examined SU 5047 1 (Holotype) WA:Port Ludlow; BC 63-936 1 BCrJervis 32 Inlet; BC 65-43 16 AK:Kodiak Island; BC 61-301 3 BC:Bute Inlet; BC 62-637 6 BC:Howe Sound; BC 65-43 x AK:Kodiak Id.; CAS 29521 5 CA:Mendocino; CAS 29488 8 CA:Mendocino; CAS 25889 5 CA:Pacific Grove; LACM 31937-8 1 CArHumboldt Co., Trinidad; LACM 31938-12 8 CA:Del Norte Co.; LACM 1679 12 CA:San Luis Obispo; LACM 7909 20 CA:San Luis Obispo; UW 18020 ser. WA:San Juan Island; UW 3020 4 WArCape Johnson; UW 17208 4 WA; UW 18020 6 WA:San Juan Island; UW 14303 1 WA:San Juan Island. Artedius l a t e r a l i s  Artedius l a t e r a l i s (Girard 1854) (Fig. 4) Range: Kodiak Island to Baja C a l i f o r n i a . Habitat: I n t e r t i d a l to 45 feet. E s p e c i a l l y abundant in t idepools. Synonymy Scorpaenichthys l a t e r a l i s Girard, 1854 p. 145 (Monterey and San Luis Obispo, C a l i f o r n i a ) . Artedius l a t e r a l i s Girard, 1856, p. 134; 1857, p. 14, 33 (not plate 22a), f i g s . 5, 6; Gunther, 1860, p. 174 (diagnosis, synonymy); Jordan and Jouy, 1881, p. 6 (name only - i n ch e c k l i s t ) ; Jordan, 1887, p.898 (name only) ; Eigenmann and Eigenmann,1892, p. 355 ( d i s t r i b u t i o n ) ; Jordan and Starks, 1895, p. 807; Jordan and Evermann, 1896, p. 437 (name only, in c h e c k l i s t ) ; Jordan and Evermann, 1898, p. 1902 (in key, and d i s t r i b u t i o n ) ; Greeley, 1899, p. 19 (color description, habitat); Osgood, 1910, p. 20 ( name only); Starks, 1911, p. 190 (description, r e l a t i o n s h i p s ) ; Halkett, 1913, p.99 (name only, in c h e c k l i s t ) ; Kincaid, 1919, p. 30 (name only, in c h e c k l i s t ) ; Bean and Weed, 1920, p. 72 (not 24765); Hubbs 1926, p. 7; Budd, 1940, f i g s . 58-74 (eggs, l a r v a l development); Hubbs and Schultz, 1941, p. 4; Bolin, 1944, p. 55, f i g . 21 (in key, and d i s t r i b u t i o n ) ; M c A l l i s t e r , 1960, p. 41 (name only - from Clemens and Wilby 1949); Clemens and Wilby, 1961, p. 299, f i g . 187; MacPhee and Clemens, 1962, p. 33 (depth d i s t r i b t i o n ) ; Hubbard and Reeder, 1965, p. 507 ( d i s t r i b u t i o n , synonymy, f i r s t v e r i f i e d Alaskan record); Quast, 1968, p. 485 (description, d i s t r i b u t i o n ) ; DeLacy, M i l l e r and Borton, 1972, p. 15 (name only, in c h e c k l i s t ) ; M i l l e r and Lea, 1972, p. 126, f i g . p. 126 (in key, and d i s t r i b u t i o n ) ; Quast and H a l l , 1972, p. 20 (name only - Alaska c h e c k l i s t ) ; Hart, 1973, p. 481 (in key, also l i f e history and d i s t r i b u t i o n ) ; F i t c h and Lavenberg, 1975 p. 128 (name only); Marliave, 1975 (larvae); Peden and Wilson, 1976, p. 235 ( d i s t r i b u t i o n - B r i t i s h Columbia); Howe and Richardson, 1978, p. 14 (in key, plus diagnosis, synonymy and meristic v a r i a t i o n ) ; Hubbs, F o l l e t t , and Dempster, 1979, p. 19 34 (name only, in C a l i f o r n i a c h e c k l i s t ) ; Washington, 1982, (larvae); Eschmeyer, Herald and Hamman, 1983, p. 162, p i . 18. Icelinus l a t e r a l i s Jordan and G i l b e r t , 1883, p. 689. Parartedius hankinsoni Hubbs, 1926, p. 4 (Point Loma, C a l i f o r n i a ) . Artedius delacyi Hubbs and Schultz, 1941, p.4.; Bolin, 1947, p. 162; Wilimovsky, 1954, p. 285 (name only - Alaska c h e c k l i s t ) . Artedius hankinsoni Bolin, 1944, p. 57, f i g . 22. Comments: Artedius hankinsoni i s here formally synonymized with Artedius  l a t e r a l i s . In Hubbs' (1926) characterization of A. hankinsoni, two features were c i t e d as being diagnostic: the low scale count and the m u l t i f i d upper preopercular spine. Bolin (1944) adds no new characters, but includes the species in Artedius, which Hubbs had not. In Artedius, there i s considerable v a r i a t i o n in the upper preopercular spine within each species, indeed even between the l e f t and right spines of one i n d i v i d u a l . In the putative A. hankinsoni examined in t h i s study, none save the type specimen displayed anything worthy of being c a l l e d " m u l t i f i d " . The remainder are indistinguishable from the normal range of va r i a t i o n found in Artedius l a t e r a l i s . Apart from the spine, only the low scale count on the body remains. Examination of larger series of l a t e r a l i s in this study yielded a few additional "A. hankinsoni" types, in 35 addition to numerous specimens intermediate between A. l a t e r a l i s and A. hankinsoni. A l l of these specimens, A. hankinsoni included, are indistinguishable from A. l a t e r a l i s in every other respect. The existence of these intermediates, coupled with the r e l a t i v e s c a r c i t y of A. hankinsoni, leads to the conclusion that A. hankinsoni represent A. l a t e r a l i s in which the scale development i s incomplete. The type specimen seems to have been unique in possessing a m u l t i f i d preopercular spine, a feature not shared by other putative A. hankinsoni. Diagnosi s 22. Twenty-four to twenty-six rows of scales along dorsal body surface, extending from second or th i r d dorsal spine to penultimate or antepenultimate dorsal soft ray. 29. No scales on body behind pectoral f i n base. Spec imens Examined UMMZ 126491 1 AKtdelacyi paratype; UMMZ 126490 1 AKtdelacyi holotype; USNM 117494 1 AKtdelacyi delacyi paratype; UMMZ 55001 1 hankinsoni holotype; BC 53-295 16 CA:Morro Bay; BC 62-589 4 BCtSooke; BC 62-731 2 BC:Saxe Pt.; BC 53-38 1 BC:Nanaimo; BC 53-36 232 1 BCrJoassa Channel; BC 53-40 2 WArCape Johnson; BC 53-296 32 CA:San Luis Obispo; BC 53-88 6 BCrNanaimo; BC 54-452 2 BCrSaturna Island; BC 59-286 1 BCrBurrard Inlet; BC 62-42 1 BC:Saturna Island; BC 54-448 2 BC:Saturna Island; BC 54-96 1 BCrStanley Park; BC 53-88 8 BCrNanaimo; BC 60-238 4 Washington; CAS 50393 17 CArSan Nicholas Id.; CAS 27316 16 CA:Duxbury Reef; CAS 50391 13 CA:Carmel; CAS 50392 12 CArSanta Barbara; CAS 50388 4 OR:Cape Arago; SU 68852 4 CA:Duxbury Reef; LACM 23612 1 MexicorBaja; LACM W71-9 3 CArMonterey; LACM 21132 9 OR:Yaquina Light; LACM 8310 1 CA:Palos Verdes (hankinsoni); LACM 1679 14 CA:San Luis Obispo; LACM 33709-3 6 CA:Arena Cove; LACM 21133 2 Mexico:Baja, Santo Tomas; LACM 876 1 CA:Pacific Grove; LACM 31860-2 23 CA:San Luis Obispo; UW 3019 16 WArCape Johnson; UW 3020 4 WArCape Johnson; UW 2312 4 Cleared and stained; UW 15741 ser. AKrKodiak Island (delacyi); UW 17412 5 WArSan Juan Island. Artedius notospilotus Artedius notospilotus Girard 1856 (Fig. 5) Ranger Puget Sound (Washington) to Baja C a l i f o r n i a . Habitatr I n t e r t i d a l to 170 feet. 37 Synonymy Calcylepidotus l a t e r a l i s Ayres, 1855, p. 77 (not of Girard). Hemilepidotus nebulosus Girard, 1856, p. 134. (refers to Ayres previously unpublished name). Artedius notospilotus Girard, 1856, p. 134 (Tomales Bay, C a l i f o r n i a ) ; 1857, p. 535, p i . 24, f i g s . 5,6; 1858, p.71; G i l l 1862, p. 279 (name only); Jordan and Gi l b e r t , 1881, p. 454 (in c h e c k l i s t , and d i s t r i b u t i o n ) ; Jordan and Jouy, 1881, p. 6 (name only, in c h e c k l i s t ) ; Jordan, 1887, p. 898 (in part); Eigenmann and Eigenmann, 1892, p. 355 ( d i s t r i b u t i o n ) ; Bolin, 1944, p. 50 (in key, description, d i s t r i b u t i o n , r e l a t i o n s h i p s ) ; DeLacy, M i l l e r and Borton, 1972, p. 15 (in chec k l i s t : PUget Sound); M i l l e r and Lea, 1972, p. 127, f i g . p. 126 (in key, and d i s t r i b u t i o n ) ; F i t c h and Lavenberg, 1975, p. 128 (name only); Howe and Richardson, 1978, p. 15 (in key, plus diagnosis, synonymy, meristic v a r i a t i o n ) ; Hubbs, F o l l e t t , and Dempster, 1979, p. 19 (name only, in C a l i f o r n i a c h e c k l i s t ) ; Eschmeyer, Herald and Hamman, 1983, p. 163, p i . 18. Artedius l a t e r a l i s Girard, 1858, p. 71, p i . 22b, f i g s . 5, 6 (specimen no. 366, col l e c t e d by Ayres, figure based on t h i s specimen only); Bean and Weed, 1920, p. 72 (27645 only) . Icelus notospilotus Jordan and G i l b e r t , 1882b, p. 690 (in part). 38 Astrolytes notospilotus Jordan and Evermann, 1896, p. 436 (in key, and d i s t r i b u t i o n ) ; 1898a, p. 1899, 1900, f i g . 689 (not f i g . 689a); Jordan, 1905, p. 442, f i g . 382 (mature male, Puget Sound); Starks and Morris, 1907, p. 219 (name only, record of Jordan and G i l b e r t , 1881, p. 61); Jordan, 1925, p. 653, f i g . 551 (mature male, Puget Sound, same specimen figured in Jordan 1 905) . Parastrolytes notospilotus Hubbs, 1926, p. 2. Diagnosis 10. Additional short, rounded spinelets encrusting two or three main upper preopercular spines. 18. In adults, lower preopercular spines serrated at posterior edge, extended dorso-ventrally along preopercular margin. 19. Six to twelve i r r e g u l a r l y spaced pores in groove bounded an t e r i o r l y by ethmoid hump, po s t e r i o r l y by o r b i t a l margin, not divided evenly by midline. 39. In adults, head tubercles raised d i r s a l l y , adorned with irregular groups of knoblets. 40. Serrations present on posterior posttemporal and upper border of supracleithrum. 39 Specimens Examined USNM 329 1 (Holotype) CA:Tomales Bay; USNM 27206 1 (O r i g i n a l l y syntype of Artedius f e n e s t r a l i s ) ; USNM 26865 6 CA:Santa Barbara; CAS 27317 17 CArDuxbury Reef; CAS 50390 1 CA:San Francisco; CAS 50389 2 Mexico:Baja; CAS W53-398 1 CA:Los Angeles; CAS Acc.# 1958-VI:9 1 San Francisco; CAS Acc.# 1972-1:24 1 No Data; CAS 13469 1 CA:San Francisco; SIO 55-34 1 CA:San Luis Obispo; SIO H52-164 1 MexicotPunta Rocosa; SIO 55-105 8 CA:San Luis Obispo; SIO 'H52-168 4 MexicorPlaya Maria; SIO 74-122 1 CA:Pt. Conception; SIO 63-1052 1 Mexico:Bahia San Quintin; SIO 63-1054 2 Mexico:Bahia San Quintin; SIO 63-1056 2 Mexico:Bahia San Quintin; SIO 73-101 12 CA:E1 Segundo. 40 THE GENUS RUSCARIUS JORDAN AND STARRS 1895 Taxonomic Revision  Ruscarius Jordan and Starks, 1895. Artedius Bolin (in part), 1944. A r t i f i c i a l key to Ruscarius 1a. Scales on eye Ruscarius meanyi. 1b. No scales on eye Ruscarius c r e a s e r i . i Based on the phylogenetic analysis, meanyi and crease r i , both formerly included in Artedius, are shown to comprise a separate monophyletic lineage, as they share none of the characters which are hypothesized to be apomorphic for Artedius. Conversely, the character states uniting meanyi and creaseri are not shared with Artedius sensu s t r i c t u . Since meanyi was o r i g i n a l l y described by Jordan and Starks in 1895 in the genus Ruscarius, i t is here proposed that this genus be resurrected to include Ruscarius meanyi and Ruscarius c r e a s e r i . These two species have never been diagnosed as a monophyletic unit. Such a diagnosis i s here provided, in addition to diagnoses for each of the species. 41 Ruscarius Jordan and Starks, 1895. Ruscarius diagnosis 3. Scale ridge almost perpendicular to basal plate, c t e n i i curving very l i t t l e . 4. Lateral body surface covered with l i g h t s t i p p l i n g of darker pigmentation, interrupted by well-defined areas without any darker pigmentation. 5. A small patch of ten to twenty scales located just caudad to dorsal edge of a x i l l a . 7. Upper preopercular spine narrow and extended caudally. This spine b i f i d , forking only in posterior quarter of i t s length. 11. Adult males with dusky black pigmentation covering lower jaw, throat, ventral body surface and a l l f i n s except caudal. 6. Scattered scales on snout above upper l i p , on t i p of ethmoid hump and above ascending processes of premaxillae. 28. Single row of scales extending forward under anterior o r b i t . 31. One to four c i r r i located at upper anterior o r b i t a l margin. 42 Ruscarius creaseri  Ruscar ius creaser i (Hubbs 1926) (Fig. 6) Range: Carmel Bay (California) to central Baja C a l i f o r n i a . Habitat: I n t e r t i d a l to 90 feet. Synonymy Ruscariops creaseri Hubbs, 1926, p. 12. (Bird Rock, San Diego County. Paratypes from White Point, Los Angeles County, and Point Lobos, Monterey County, C a l i f o r n i a ) . Artedius creaseri Bolin, 1944, p. 43, f i g . 16 (description, r e l a t i o n s h i p s ) ; M i l l e r and Lea, 1972, p. 126, f i g . p. 126, (in key, and d i s t r i b u t i o n ) ; F i t c h and Lavenberg, 1975, p. 128 (name only); Howe and Richardson, 1978, p. 13 (in key, plus diagnosis, synonymy and meristic v a r i a t i o n ) ; Hubbs, F o l l e t t and Dempster, 1979, p. 18 (name only - C a l i f o r n i a c h e c k l i s t ) ; Washington, 1982, (larvae); Eschmeyer, Herald and Hamman, 1983, p. 162, p i . 18. 43 Diagnosis 9. Three to ten simple c i r r i in a cluster at dorsalmost posterior edge of opercle, just anterad to fleshy flap which makes up i t s posterior margin. 33. One to ten simple c i r r i in cluster anterad to uppermost preopercular spine. 32. One to three simple c i r r i in cluster midway between top of pectoral base and l a t e r a l l i n e . Specimens Examined UMMZ 141866 1 Bird Rock, San Diego (holotype); BC 63-979 3 Mexico:Baja; CAS 19800 2 CArBird Rock, San Diego; CAS 19671 1 Mexico:Baja; CAS 50119 1 Mexico:Baja; CAS 19516 2 CArSan Diego; CAS 19797 2 CArSan Diego; CAS 25382 1 CArSanta Catalina Island; CAS 19626 3 MexicorBaja; LACM 6587-5 39 CArPalos Verdes; LACM 4917 3 CArSan Luis Obispo; LACM 32082-1 3 MexicorBaja; LACM 32045-3 4 MexicorCedros Island; LACM 32053-11 3 MexicorBaja; LACM 32041-9 1 MexicorCedros Island. 44 Ruscarius meanyi Ruscarius meanyi Jordan and Starks 1895 (Fig. 7) Range: Alaska to Arena Cove ( C a l i f o r n i a ) . Habitat: I n t e r t i d a l to 269 feet. Synonymy Ruscarius meanyi Jordan and Starks, 1895, p.805, p i . 80 (Port Orchard Puget Sound, Washington); Jordan and Evermann, 1898, p. 1908 (in key, and d i s t r i b u t i o n ) ; Halkett, 1913, p.99 (name only, in c h e c k l i s t ) ; Kincaid, 1919, p. 29, f i g . 66 (checklist - Puget Sound); Hubbs, 1928 (name only, in c h e c k l i s t ) ; Schultz and De Lacy, 1936, p.127 (in c h e c k l i s t , and d i s t r i b u t i o n ) ; Schultz, 1936, p. 177 (key). Artedius meanyi Rosenblatt and Wilkie, 1963 (redescription, d i s t r i b u t i o n : f i r s t B r i t i s h Columbia record); Peden, 1972, p. 168 ( d i s t r i b u t i o n - B r i t i s h Columbia); Quast and H a l l , 1972, p. 20 (name only - Alaska c h e c k l i s t ) ; Blackburn, 1973, (larvae: Cottid 3); Hart, 1973, p. 483 (figure, l i f e history, d i s t r i b u t i o n ) ; Lea, 1974 ( f i r s t C a l i f o r n i a record); F i t c h and Lavenberg, 1975, p. 128 (name only); Peden and Wilson, 1976, p. 235, ( f i r s t Alaskan record); 45 Moulton, 1977 (Puget Sound: habitat, depth d i s t r i b u t i o n ) ; Richardson, 1977, (larvae: Icelus sp. 1); Richardson and Pearcy, 1977, (larvae: Icelus sp. 1);. Howe and Richardson, 1978, p.14 (in key, plus diagnosis, synonymy, meristic v a r i a t i o n ) ; Hubbs, F o l l e t t , and Dempster, 1979, p. 19 (name only, in C a l i f o r n i a c h e c k l i s t ) ; Richardson and Washington, 1980, (larvae: Icelus spp.); Washington, 1982, (larvae); Eschmeyer, Herald and Hamman, 1983, p. 163, p i . 18. Diagnos i s 1. In adult males, urogential opening abuts o r i g i n of anal f i n . 8. One spine and two soft rays in pelvic f i n s . 37. Maximum adult size 50 mm. 38. In adult males, f i r s t three anal rays thickened and short. 44. No c i r r i on nasal spine. 47. Scales cover upper o n e - f i f t h of surface of eye, never covering p u p i l . Spec imens Examined SU 3127 2 WA:Cotypes; CAS 37281 1 BC:Saanich Inlet; CAS 27695 2 CA:Mendocino; BC 62-495 1 BC:Sooke; LACM 38248-1 3 BC:Blind Bay; LACM 38247-1 3 BC:Raymond Island; NMC 77-0148 27 BC:Klaquack 46 Channel; NMC 68-0373 1 BC:Queen Charlotte Islands; SIO 63-599 2 BC:Howe Sound; SIO 73-227-55 1 CArArena Cove; UW 20721 1 WA:Shaw Island; UW 20720 1 WA:Seattle; UW uncat. 2 cleared and stained. 47 PHYLOGENETIC ANALYSIS L i s t of Adult Characters 1 . Position of anus 2. Shape of snout 3. Form of scale ridge 4. Body color pattern 5. Scales above a x i l l a 6. Scales on snout 7. Shape of upper preopercular spine 8. Number pelvic rays 9. C i r r i on opercle 10. Spinelets on main preopercular spine 11. Male color 12. Chin coloration 13. Mandibular pore pattern 14. Pores on l a t e r a l l i n e scales 15. Form of head scales 16. C i r r i on transverse head tubercles 17. C i r r i on suborbital stay 18. Form of preopercular spine 19. Nasal pores 20. Form of teeth 21 . Branchiostegal number 22. Number of scale rows above l a t e r a l l i n e 48 23. Anal f i n membrane 24. Anal f i n pigmentation 25. Penis 26. C i r r i on upper l i p 27. Po s t c l e i t h r a 28. Scales under anterior of orbi t 29. Scales behind a x i l l a 30. Scales on caudal peduncle 31. Preorbital c i r r i 32. C i r r i above a x i l l a 33. C i r r i anterad to upper preopercular spine 34. C i r r i on preopercular margin 35. Scale ridge shape 36. Scale ridge placement 37. Adult size 38. Anal ray form (males) 39. Form of head tubercles 40. Serrations on posttemporal and supracleithrum 41. Size of c i r c l e s on body 42. White throat pigmentation 43. White spots on body 44. C i r r i on nasal spine 45. Pterotic flange 46. O s s i f i c a t i o n of opercle 47. Scales on eye 49 L i s t of Larval Characters 1. Number of preopercular spines 2. Relative size of preopercular spines 3 . Basal preopercular spines 4 . Inner shelf preopercular spines 5. Skin bubble on nape 6. Dorsal gut d i v e r t i c u l a 7. P a r i e t a l spines 8. Nape melanophores 9 . Snout shape 10. Hindgut length 11. Number of pelvic rays Adult Character Analysis Character 1. Position of anus. State 0: urogenital opening located s l i g h t l y anterad to or i g i n of anal f i n base. State 1: in adult males, urogenital opening abuts o r i g i n of anal f i n . An anterior anus is found in adults of Clinocottus, Triglops, Blepsias and Orthonopias. Ruscarius creaseri and 50 Artedius sensu s t r i c t o have the piesiomorphic condition (state 0). The condition in Ruscarius meanyi (state 1) i s not found in any of the outgroups and i s therefore hypothesized to be autapomorphic for Ruscarius meanyi. Character 2. Snout shape. State 0: snout long; distance from anterior edge of orbit to upper l i p about equal to o r b i t a l width. State 1: snout short and steep in p r o f i l e ; distance from anterior edge of orbi t to upper l i p about one-half width of o r b i t . ( f i g . 8) Snout shape in the co t t i d s i s highly variable. A majority have a long, f l a t snout, but a very short snout i s found in only a few genera, including Orthonopias, Gymnocanthus and Ocynectes (state 1). Artedius sensu str i c t u and Ruscar ius have longer snouts (state 0). Character 3. Form of scale ridge. ( f i g . 9) State 0: scale ridge almost perpendicular to basal plate, with c t e n i i curving to an angle p a r a l l e l with plate ( f i g . 9a). State 1: scale ridge almost perpendicular to basal plate, c t e n i i curving very l i t t l e ( f i g . 9b). State 2: no scale ridge; c t e n i i o r iginating d i r e c t l y from basal plate ( f i g . 9c). State 3: scale ridge curving to become p a r a l l e l with plate, 51 c t e n i i radiating from dorsal, ventral and caudal margins of ridge ( f i g . 9d). State 4: scale ridge curving to become p a r a l l e l with plate, c t e n i i radiating from caudal margin of ridge only. Antero-l a t e r a l edge of ridge without c t e n i i , forming a d i s t i n c t shelf ( f i g . 9d). State 5: no scale ridge, no c t e n i i . The scales in co t t i d s are not the usual teleostean c y c l o i d or ctenoid bony ridge scale, rather they are bony structures consisting of a basal plate from which arise various projections. These projections ( c t e n i i ) may a r i s e d i r e c t l y from the plate, or from an intervening ridge (spinous ridge). State 0 i s found in Hemilepidotus hemilepidotus and Orthonopias, and is hypothesized to be plesiomorphic. The scale ridge in Orthonopias i s not complete and decreases in height as i t approaches the midline of the basal plate. As a result, the three or four c t e n i i which l i e on or near the midline of the plate arise d i r e c t l y from the base, with no intervening ridge. State 1 is found in Ruscar ius meanyi and Ruscarius crea s e r i , and i s hypothesized to be synapomorphic for Ruscar ius. By comparison with the outgroups, state 3 i s hypothesized to be synapomorphic for Artedius sensu s t r i c t o . State 4 is hypothesized to be synapomorphic for the subgenus Artedius, which includes Artedius harrinqtoni, Artedius l a t e r a l i s , and Artedius c o r a l l i n u s . State 2 i s found only in Chitonotus  puqetensi s. Although t h i s character serves to d i s t i n g u i s h several 52 monophyletic lineages, i t has a non-additive d i s t r i b u t i o n at the higher l e v e l of analysis and is therefore of l i t t l e value in determining inter-generic relationships, p a r t i c u l a r l y with regards to the lineage including Ruscarius, Chitonotus, and Icelinus. There are nine e q u a l l y - l i k e l y transformation series, none of which can be favored at present. Character 4. Body color pattern. (Figs. 1-8) State 0: in preserved specimens, the area bounded an t e r i o r l y by pectoral f i n 'base, p o s t e r i o r l y by caudal f i n , dorsally by rows of body scales, ventrally by a l i n e dorsal to anal f i n covered with uninterrupted darker pigment. This may be arranged in alternating stripes of r e l a t i v e l y l i g h t e r and darker pigment ( f i g . 8). State 1: l a t e r a l body surface covered with l i g h t s t i p p l i n g of darker pigmentation, interrupted by well-defined c i r c u l a r areas without any darker pigmentation ( f i g s . 6 and 7). State 2: generally darker background pigmentation of l a t e r a l body surface interrupted by c i r c u l a r areas lacking darker pigmentation (the l a t t e r the same as ventral body surface). Lighter c i r c l e s become larger closer to ventral edge of dark pigmentation. The largest c i r c l e s , at the ventral margin, are incomplete, producing a scalloped appearance above anal f i n base. Margins of each c i r c l e c r i s p and well-defined ( f i g s . 1-5). The d i s t i n c t i v e "Artedius" pattern of pigmentation (state 53 2) i s found only in Artedius sensu s t r i c t u and not in any of the outgroups. It i s therefore postulated to be synapomorphic for the genus Artedius. A similar type of coloration pattern occurs in the c o t t i d genus Myoxocephalus, but on close examination t h i s i s seen to be a pattern of irregular yellow blotches on a darker background. These blotches continue to the anal f i n base and do not extend above the l a t e r a l l i n e : hence t h i s is hypothesized to be a d i f f e r e n t state from that observed in Artedius. Irregular blotches of l i g h t e r pigment, similar to the Artedius c i r c l e s , are also found in Clinocottus. However, these do not have the c h a r a c t e r i s t i c a l l y c r i s p outline of Artedius c i r c l e s and they decrease in diameter as they continue down to the anal f i n . In 01igocottus, the pattern is much the same as in Clinocottus: the margins of the spots are i n d i s t i n c t and they decrease in size v e n t r a l l y . In 01igocottus, though, the middle portion of each spot is covered with darker black speckles. In Chitonotus, Icelinus, and Ruscarius (state 1 ) , the l a t e r a l body surface is covered with a s t i p p l i n g of black pigment, with widely-spaced melanophores. Below the l a t e r a l l i n e , this s t i p p l i n g is broken up by areas without pigment. These areas are c i r c u l a r in Icelinus and Chi tonotus, and form v e r t i c a l bars in Ruscarius meanyi and Ruscarius c r e a s e r i . The pattern varies within Icelinus, with Icelinus boreali s, I_. f ilamentosus and possibly J_. burchami having roughly c i r c u l a r areas of l i g h t pigment, though they are more irregular than those in Artedius. In Icelinus cavifrons these seem to be roughly aligned in v e r t i c a l bars, much as in Ruscar ius. 54 This character has a d i s j o i n t d i s t r i b u t i o n consisting of two transformation subseries: 0-1-2 and 3-4. Its main use in this hypothesis seems to l i e in supporting the separate monophyly of Ruscarius and Artedius sensu s t r i c t u . Character 5. Scales above a x i l l a . State 0: no scales located above a x i l l a . State 1: a small patch of ten to twenty scales located just caudad to dorsal edge of pectoral f i n base, not arranged in rows. This patch of scales (in Ruscarius creaseri and R. meanyi) is in a d i f f e r e n t location from the scale patch found in Hemilepidotus hemilepidotus, which is behind the f i n and not above the a x i l l a . The other pattern of scales behind the f i n i s one of p a r a l l e l rows, not a patch (see character 39). This patch i s not found in Artedius, Orthonopias, Icelinus, Stelgistrum, Ricuzenius or Chitonotus pugetensi s, and i s postulated to be synapomorphic for Ruscarius. Character 6. Scales on snout. State 0: no scales on snout. State 1: scattered scales on snout above upper l i p , at t i p of ethmoid hump and above ascending processes of premaxillae. Scales are found on the snout in Ruscar ius meanyi and R. creaseri as well as two other genera (Chitonotus and 55 Ricuzenius). F a r r i s optimization leads to a hypothesis of two independent origins for th i s feature: once in Ricuzenius and once in the lineage leading to Ruscarius and Chitonotus. These scales are not found in Artedius .sensu s t r i c t u . Character 7. Shape of upper preopercular spine. State 0: upper (fourth) preopercular spine not extended caudally, having two to three prongs. State 1: upper preopercular spine narrow and extended caudally. This spine b i f i d , forking only in posterior quarter of i t s length ( f i g s . 6 and 7 ) . A narrow, caudally elongated preopercular spine is found in several c o t t i d genera. A long, unbranched spine is t y p i c a l of Porocottus and Myoxocephalus. Variations on the " a n t l e r - l i k e " preopercular spine are found in Chi tonotus, Gymnocanthus, Leptocottus, Enophrys, Icelinus and the Japanese genera Stlenqis and Daruma. The number of times t h i s " a n t l e r - l i k e " spine has arisen in the family cannot be estimated u n t i l a phylogenetic hypothesis including a l l cotto i d genera i s generated. For the time being, then, only the state found in Ruscar ius meanyi and Ruscar ius creaseri (state 1) can be hypothesized to be a unique, derived t r a i t . A l l of the species in Artedius sensu s t r i c t u have state 0. Character 8 . Number of pelvic rays. 56 State 0: one spine, three soft rays. State 1: one spine, two soft rays. There has been some contention as to whether Ruscarius  meanyi actually has two or three pelvic rays, primarily caused by i t s small size and the d i f f i c u l t y of distinguishing individual rays without clearing and staining. Lea (1974) found only two rays in his specimens. Howe and Richardson (1981) re-examined Lea's (1974) specimens and found only one with two: the rest had three. Washington (1982) also re-examined Lea's specimens, th i s time clearing and staining. She found two rays in a l l of the f i s h she examined. I examined cleared and stained Ruscar ius meanyi in t h i s study and found them to have two rays, never three. Without clearing and staining, i t i s easy to mistake the branched end of the one thickened ray for two normal rays. Two pelvic rays are also c h a r a c t e r i s t i c of the genera Icelinus and Stlenqis. Washington (1982) suggested that Ruscarius meanyi be placed in a taxon with Icelinus based on th i s character. However, the state 1,2 i s hypothesized to have arisen independently in Ruscar ius meanyi and Icelinus, since the present analysis indicates that Ruscar ius c reaser i (1,3) is the s i s t e r taxon to Ruscar ius meanyi, and, further, that Chitonotus (also 1,3) i s the s i s t e r taxon to these two. An indication of the independent o r i g i n of this state comes from Washington (1982). Referring to Ruscarius meanyi, she stated: "cleared and stained specimens have 1,2 pelvic f i n rays. The outermost ray is greatly thickened and branched at the t i p in a l l specimens 57 examined" (1982, p. 148). By contrast, she observed that, in Icelinus, "both f i n rays are r e l a t i v e l y short and fine" (p. 172, i t a l i c s mine). These two observations suggest that the 1,2 states in Ruscarius meanyi and Icelinus are not s t r u c t u r a l l y homologous. A l l of the species in Artedius sensu s t r i c t u have the state 1,3 (state 0). Character 9. C i r r i on opercle. State 0: one to three c i r r i clustered at dorsalmost posterior edge of opercle, just anterior to fleshy f l a p which makes up i t s posterior margin. State 1: as in state 0; but three to ten c i r r i per cl u s t e r . State 2: no c i r r i on opercle. In Orthonopias, Clinocottus, 01igocottus, Icelinus, Ruscarius meanyi and Artedius sensu s t r i c t o , there are one to three c i r r i on the opercle, which is hypothesized to be the plesiomorphic state. The possession of up to ten c i r r i here by Ruscar ius creaseri (state 1) is hypothesized to be autapomorphic. If Chi tonotus is hypothesized to have secondarily l o s t these c i r r i , t his character has the additive transformation series 0-1-2. Character 10. Spinelets on main preopercular spines. State 0: absent. State 1: additional short, rounded spinelets encrusting two or 58 three main upper preopercular spines ( f i g . 5). These additional spinelets (state 1), found in adult Artedius notospilotus, are absent from the rest of .the cott i d s examined (which a l l have state 0) and are therefore hypothesized to be autapomorphic. Character 11. Male color. State 0: in preserved specimens, ventral surface of male the same as female: branchiostegal membranes, ventral body, and fin s l i g h t colored. State 1: males with a dusky black pigmentation covering the lower jaw, branchiostegal membrane, and ventral body surface. In addition, a l l of the fin s are dusky, with the f i r s t dorsal having a darker, almost black band along i t s dorsal margin. In a l l of the outgroups and in Artedius, the body coloration i s either v i r t u a l l y the same for males as i t i s for females, or at most only the ventral f i n s and throat are dusky (state 0). Therefore, the overall dusky coloration of males of Ruscar ius meanyi and Ruscarius creaseri i s hypothesized to be an apomorphic state (state 1). Character 12. Chin c o l o r a t i o n . State 0: no pattern of c i r c l e s on ventral surface of throat. State 1: in preserved specimens, a pattern of large c i r c u l a r areas of l i g h t e r background pigmentation interrupting dark 59 s t i p p l i n g , sometimes extending onto anteriormost portion of branchiostegal membrane ( f i g . 10a,b). State 2: in preserved specimens, a pattern of c i r c u l a r blotches of l i g h t background pigmentation interrupting darker pigmentation on ventral surface of head, extending caudally to branchiostegal membrane ( f i g . 10c). State 3: as in state 2, but c i r c l e s very small ( f i g . I0d). The only outgroup with a d i s t i n c t l y pigmented throat i s Orthonopias; however, i t s arrangement of white vermiculations on a dark background is not the same as the condition in Artedius  sensu s t r i c t u , which consists of an absence of darker pigment, not the presence of white pigment superimposed on a uniformly dark background. Ricuzenius, CIinocottus, 01igocottus, Chitonotus, and Stelgistrum lack any throat pigmentation at a l l (state 0). Ruscarius meanyi and R. creaseri have scattered s t i p p l i n g on the throat. The presence of li g h t e r c i r c u l a r areas of throat pigment i s hypothesized to .be synapomorphic for Artedius sensu s t r i c t o . In state 1, found in Artedius  notospilotus and Artedius f e n e s t r a l i s , the pigment does not extend onto the branchiostegal membrane to any appreciable extent, while in states 2 and 3 i t covers v i r t u a l l y a l l of the membrane. Artedius harringtoni has state 2, while the very small c i r c l e s of l i g h t e r pigmentation (state 3) are found in A. c o r a l l i n u s and A. l a t e r a l i s . This character has a non-additive d i s t r i b u t i o n within the genus Artedius, and i s therefore of l i t t l e value in elucidating the relationships therein. However, i t serves to distinguish 60 two subgroups within Artedius; the subgenus Astrolytes, including Artedius f e n e s t r a l i s and Artedius notospilotus, and the lineage including Artedius c o r a l l i n u s and Artedius  l a t e r a l i s . Three transformation series are equally l i k e l y : 0-1-2-3; 0-1-3-2; and 0-1-2 character 13. Mandibular pore pattern. State 0: beginning at posterior end of maxilla, three single pores on each side. Additionally, two (rarely one) pores on ventral midline, just caudad to lower l i p ( f i g s . 10c,d). State 1: ten to twenty minute pores on each side, occurring singly or in clusters, plus four to ten pores clustered around ventral midline. One or two midline pores located caudally to anterior midline cluster ( f i g s . 10a,b). McAl l i s t e r (1968) i l l u s t r a t e d the pattern of mandibular pores in many c o t t i d genera. Based on his study and examination here of Orthonopias, Ruscarius, Clinocottus and 01igocottus, which a l l share state 0, the pattern found in Artedius  f e n e s t r a l i s and Artedius notospilotus (state 1) is postulated to be apomorphic. Character 14. Pores on l a t e r a l l i n e scales. State 0: one pore located near posterior margin of each l a t e r a l l i n e plate. 61 State 1: one to three extra pores located on ten to twenty-five of last t h i r t y l a t e r a l l i n e scales. Extra pores minute, e a s i l y distinguishable from main pore located at posterior margin of scale. These extra pores are not found in Ruscarius, Orthonopias, Oligocottus, Clinocottus, Chitonotus, Icelinus or Stelqistrum. In a l l of these genera, there i s only one pore per l a t e r a l l i n e scale (state 0). Extra, minute pores (state 1) are therefore hypothesized to be apomorphic and are found only in Artedius  f e n e s t r a l i s and Artedius notospilotus. Character 15. Form of head scales. State 0: scales on head not distinguishable in form from scales covering body. State 1: scales on occiput deeply embedded, c t e n i i radiating in a l l d i r ections d i r e c t l y from margin of basal scale plate. Ctenii much shorter and blunter than c t e n i i of body scales. State 2: no scales on head. In Artedius harringtoni, Ruscar ius meany i , R. creaseri and other genera such as Ricuzenius and Stelgistrum with scales extending onto the head, the scales on the occiput are not markedly d i f f e r e n t in form from the scales on the body (state 0). Therefore the embedded, s t e l l a t e scales (state 1) on the heads of Artedius f e n e s t r a l i s and A. notospilotus are, by out-group comparison, apomorphic. Artedius c o r a l l i n u s and A. l a t e r a l i s have no scales on the head. Some Gymnocanthus 62 species also have bony "scales" on the head, but these consist of conical points a r i s i n g perpendicular to a basal plate, not a r i s i n g at an angle from the margin of the plate, as in Artedius  sensu s t r i c t u . This character has the additive transformation series 0-1-2. Character 16. C i r r i on transverse head tubercles. State 0: one to three simple c i r r i just anterad to four bony tubercles on head, tubercles in transverse l i n e posterior to o r b i t a l margins. State 1: three to ten c i r r i . l o c a t e d on each tubercle, in a l i n e . The plesiomorphic condition for tubercle c i r r i i s one to three c i r r i on each tubercle. Artedius f e n e s t r a l i s is the only species with state 1. The remainder of Artedius sensu s t r i c t u and Ruscarius have state 0. Clinocottus embryum, C. analis, and C. qlobiceps also have a high number of c i r r i on these tubercles, but in these species they are not arranged in a linear fashion as in Artedius f e n e s t r a l i s and are in a c l o s e l y -packed bunch. The presence of up to ten c i r r i (state 1) on each tubercle in Artedius f e n e s t r a l i s i s interpreted to be autapomorphic. Character 17. C i r r i on suborbital stay. State 0: one to two simple c i r r i on suborbital stay, midway between posterior end of preopercle and posterior end of o r b i t . 63 State 1: as in state 0, but one to two c i r r i at anterior end of suborbital stay at a point posterior to end of o r b i t , plus one to two simple c i r r i at posterior end of suborbital stay, at a point just anterad to uppermost preopercular spine. State 2: no c i r r i on suborbital stay. In Orthonopias there i s one c i r r u s located midway along the suborbital stay (state 0), while in Clinocottus and Oligocottus there are none (state 2). Artedius f e n e s t r a l i s also has none, and t h i s i s here interpreted to be a secondary loss. Artedius  l a t e r a l i s and A. c o r a l l i n u s have, in addition, c i r r i at both the anterior and posterior ends of the bony stay (state 1). This s i t u a t i o n i s , by comparison with the outgroups, apomorphic. The two Ruscar ius species both have state 0. This character has the additive transformation series 0-1-2. Character 18. Form of preopercular spines. State 0: in adults, rounded to sharp, always simple. State 1: in adults, serrated at the posterior edge, extended dorso-ventrally along preopercular margin ( f i g . 7). The form of the lower three preopercular spines varies greatly in c o t t i d genera. However, the range of variation runs from rounded to sharp, never branched or elaborated in any way. The serrated form of the preopercular spines in Artedius  notospilotus (state 1) is therefore postulated to be autapomorphic. Ruscarius and the remainder of Artedius sensu s t r i c t u have state 0. 64 Character 19. Nasal pores. State 0: two to four evenly spaced pores located in groove bounded a n t e r i o r l y by ethmoid hump and p o s t e r i o r l y by anterior o r b i t a l margin, one to two on either side of midline. State 1: six to twelve i r r e g u l a r l y spaced pores, not evenly divided by midline. In Orthonopias and CIinocottus there are four evenly spaced nasal pores, two on either side of the midline. In 01iqocottus, Chitonotus and Icelinus, there are usually two. Possession of numerous i r r e g u l a r l y arranged pores (state 1) i s interpreted to be autapomorphic for Artedius notospilotus. Ruscarius and the remainder of Artedius sensu s t r i c t u have state 0. Character 20. Form of teeth. State 0: teeth on dentaries, premaxillae, vomer and palatines in v i l l i f o r m bands. State 1: cardiform teeth on dentaries, premaxillae, vomer and palatines. State 2: in adult males, teeth on vomer and palatines cardiform, those on premaxillae and dentaries i r r e g u l a r l y canine, with numerous d i s t i n c t i v e l y longer canine teeth on both upper and lower jaws. In Ruscarius, Orthonopias, CIinocottus, 01igocottus, Chitonotus and Stelqistrum the teeth are uniformly v i l l i f o r m , never elongate or canine (state 0). The cardiform bands of 65 teeth in Artedius c o r a l l i n u s (state 1) are therefore postulated to be autapomorphic. Further, the "fangs" (state 2) found in adult male Artedius harringtoni are also interpreted to be autapomorphic for that species. This character has the additive transformation series 0-1-2. Character 21. Branchiostegal number. State 0: six. State 1: seven, one additional on ceratohyal. In a l l cottoids,' there are six branchiostegals on each side, save for the psychrolutids, which have seven (two on the epihyal, five on the ceratohyal) and are placed by some authors in a separate family. At the l e v e l of analysis including Artedius and i t s immediate s i s t e r taxa, possession of seven branchiostegals by A. harrinqtoni (state 1) i s interpreted to have been derived separately from the seven in psychrolutids. Seven branchiostegals are also infrequently found in Artedius  f e n e s t r a l i s and Artedius l a t e r a l i s : however, the frequency of occurrence is apparently very low and the true frequency in nature i s unknown. Character 22. Number of scale rows above the l a t e r a l l i n e . State 0: 29 to 38 rows. State 1: 24 to 26 rows. State 2: 43 to 46 rows. 66 In Artedius £enestralis, Chitonotus, Orthonopias and Stelgistrum, there are from twenty-nine to th i r t y - e i g h t rows of scales on the body above the l a t e r a l l i n e , counted from the o r i g i n of the f i r s t dorsal f i n to the caudal end of the second dorsal base (state 0). The high number of rows found in Artedius c o r a l l i n u s and A. harringtoni (state 2) is therefore hypothesized to be apomorphic. The lower counts found in Artedius l a t e r a l i s and Artedius notospilotus (state 1) are hypothesized to have arisen twice. This character has the additive transformation series 0-1-2. Character 23. Membrane of anal f i n . State 0: incised membrane of anal f i n concave in males and females. State 1: incised membrane of anal f i n convex in males, concave in females ( f i g . 3 ) . In males of a l l of the outgroups examined, as well as in Ruscar ius and Artedius sensu s t r i c t u save A. harr ington i , the incised membrane of the anal f i n i s concave in both males and females (state 0). The condition in Artedius harrinqtoni, where the membrane i s convex in males (state 1), i s hypothesized to be autapomorphic. Character 24. Male anal f i n pigmentation. State 0: anal f i n dusky, sometimes with alternating l i g h t and 67 dark areas. State 1: anal f i n covered with hexagonal latticework of areas lacking pigmentation on darker background. ( f i g . 3 ) . In Orthonopias and Ruscar ius, the anal f i n is uniformly dusky, while in Chitonotus the dusky area i s confined to a narrow band set in s l i g h t l y from the d i s t a l margin of the f i n . A- pattern of l i g h t and dark str i p e s i s also found in several other c o t t i d genera such as Hemilepidotus, 01igocottus and Myoxocephalus ( a l l state 0). Therefore the pattern of hexagonal latticework on the anal f i n of Artedius harrinqtoni (state 1) i s hypothesized to be autapomorphic. Character 25. Penis. State 0: absent. State 1: present, in form of small cone (fig.3) state 2: present, much longer, with or without elaborate d i s t a l processes. Modifications of the urogenital p a p i l l a are widespread in the Cottidae, and are usually lumped under the generic label of "penis". However, differences in penis structure in various c o t t i d genera ranges from the massive curved cylinders in Clinocottus, Icelinus, Pseudoblennius, and •Psychrolutes  paradoxus to delicate filaments in Eurymen, Bero elegans, Nautichthys oculof asc iatus and 01 igocottus Between these extremes are the long, slender conical penes in Radulinus and the complex structures in Chi tonotus, Ocynectes, and 68 Gymnocanthus. In the immediate outgroups under consideration here, Chitonotus has a unique and elaborate penis (state 2), but neither Hemilepidotus nor Orthonopias have any penis (though Bolin in 1941 reports that the female of Orthonopias has an extensible oviduct). The presence of a short, caudally-directed conical penis in Artedius harr ingtoni (state 1) i s tentatively hypothesized to be an autapomorphic t r a i t at the current l e v e l of analysis, as no obvious s t r u c t u r a l homologue can be found within the family. Ruscar ius and the remainder of Artedius  sensu s t r i c t u a l l lack penes (state 0). Character 26. C i r r i on upper l i p . State 0: none. State 1: two to sixteen simple paddle-shaped c i r r i dorsad to upper l i p , located anywhere from end of maxilla anterad. Not d i s t r i b u t e d evenly on each side. Usually two smaller c i r r i at premaxillary symphysis. The presence of paddle-shaped c i r r i above the upper l i p in Artedius c o r a l l i n u s (state 1) appears to be autapomorphic, since these c i r r i are not found in Ruscar ius, Chitonotus, Orthonopias, CIinocottus, Oligocottus or Stelgistrum or other members of Artedius (state 0). The only other c o t t i d with c i r r i above the upper l i p is Dasycottus setiger. These c i r r i , though are only found above the posterior portion of the l i p , and are long and slender, as opposed to the paddle-shaped c i r r i of A. c o r a l l i n u s . 69 Character 27. Postcleithra. State 0: p o s t c l e i t h r a present. State 1: p o s t c l e i t h r a absent. The p o s t c l e i t h r a are present in the outgroups, but are absent in Artedius l a t e r a l i s , A. f e n e s t r a l i s , A. harrinqtoni, A. c o r a l l i n u s and A. notospilotus. Apparently the loss of pos t c l e i t h r a has occurred several times within the family. A survey of the available c o t t i d s revealed that Myoxocephalus  niger, three species of Clinocottus (C. acuticeps, C. embryum, and C. qlobiceps), G i l b e r t i d i a sigalutes, Ascelichthys rhodorus (which also lacks pelvic fins) and in Psychrolutes paradoxus a l l lack p o s t c l e i t h r a . Character 28. Scales under anterior of o r b i t . State 0: none. State 1: a single row of scales extending forward to a point under anterior margin of o r b i t . In Orthonopias and Chitonotus pugetensis, the scales on the head do not continue forward under the anterior portion of the orbit (state 0). Stelgistrum stejnegeri, Ricuzenius pinetorum , Artedius f e n e s t r a l i s , Ruscar ius meanyi and R. creaseri do have scales under the anterior orb i t (state 1). The remainder of Artedius sensu s t r i c t u have no scales in t h i s region. F a r r i s optimization based on the most parsimonious cladogram suggests four origins for this feature: once in Stelgistrum, once in 70 Ricuzenius, once in Artedius f e n e s t r a l i s , and once in the lineage leading to Ruscar ius. t Character 29. Scales behind a x i l l a . State 0: none. State 1: two to three p a r a l l e l rows of five to twenty scales extending from top to ventral end of pectoral f i n , closely confined to f i n base. Scattered scales are found in the area behind the a x i l l a in Gymnocanthus, Ricuzen ius, Stelgistrum, and Hemilepidotus. Scattered scales or p r i c k l e s are also found in the western P a c i f i c genera Bero, Ale ichthys, Furc ina, and Pseudoblennius. In a l l of the preceding genera, however, the scales are not clos e l y confined to the pectoral f i n base as they are in Artedius, and extend caudally along the body. F a r r i s optimization suggests that in Artedius sensu s t r i c t u , a row of scales confined to the pectoral base (state 1) arose once and was subsequently lost twice, once in A. notospilotus and once in A. l a t e r a l i s . Character 30. Scales on caudal peduncle. State 0: body scales continuing onto dorsal surface of caudal peduncle, covering area extending from end of second dorsal base to o r i g i n of f i r s t dorsal caudal r a y l e t . State 1: scales absent from dorsal surface of caudal peduncle. 71 Scales on the caudal peduncle (state 0) are found in Stelqistrum, Ricuzenius, Orthonopias, and Chitonotus pugetensis. They are absent in Artedius c o r a l l i n u s , A. l a t e r a l i s and A. notospilotus (state 1). The most parsimonious interpretation i s that they were lost twice in Artedius - once in the lineage including A. c o r a l l i n u s and A. l a t e r a l i s , and once in A. notospilotus. Character 31. Preorbital c i r r i . State 0: none. State 1: one to four c i r r i located at upper anterior margin of orbit ( f i g s 6,7). State 2: one very large c i r r u s anterior to o r b i t , plumose in mature males ( f i g . 3). Preorbital c i r r i are not found in I eelinus, Orthonopias or Chitonotus pugetensis (state 0). However they are present in Artedius harrinqtoni (state 2), Ruscarius, Jordania zonope and a l l species of 01iqocottus ( a l l state 1). At present the most parsimonious interpretation is that they arose four times: once in Jordania, once in Artedius harr ingtoni, once in Ruscar ius and once in 01iqocottus. A. harrinqtoni is the only Artedius species with preorbital c i r r i . This character has the additive transformation series 0-1-2. Character 32. C i r r i above a x i l l a . 72 State 0: one to three simple c i r r i in cluster at point midway between pectoral f i n base and l a t e r a l l i n e scale row. State 1: none. One to two c i r r i appear above the a x i l l a (state 0) in Artedius harrinqtoni, Ruscarius meanyi, Orthonopias, a l l Clinocottus except C. acuticeps, Icelinus fimbriatus and Icelinus oculatus. F a r r i s optimization postulates four origins for this state: once in Clinocottus (with a loss in C. acuticeps), once in A. harrinqtoni, once in R. meanyi, and once in Icelinus, i f Bolin's (1947) phylogenetic diagram is correct. The remainder of Artedius sensu s t r i c t u lack these c i r r i ( state 1). Character 33. C i r r i anterad to upper preopercular spine. State 0: none. State 1: one to ten c i r r i in cluster anterad to uppermost preopercular spine, at margin of opercle and preopercle. By outgroup comparison, the cluster of up to ten c i r r i just in front of the upper preopercular spine in Ruscar ius creaseri (state 1) i s autapomorphic, as these c i r r i do not occur in R. meanyi, Artedius sensu s t r i c t u , Orthonopias, Clinocottus, 01igocottus, Chitonotus, Icelinus, Ricuzen ius or S t e l g i strum ( a l l state 0). Character 34. C i r r i on preopercular margin. 73 State 0 : one to five simple c i r r i confined to each of preopercular spines. State 1: seven to twelve simple c i r r i along preopercular margin, not confined to preopercular spines. In Artedius sensu s t r i c t u except A. harringtoni, in Ruscar ius and in the other outgroups, there are from one to five simple c i r r i on the preopercular margin, with one or more on each preopercular spine (state 0 ) . The derived condition (state 1) found in Artedius harringtoni, with seven to twelve c i r r i not confined to the spines, is shared with Clinocottus embryum, and is hypothesized to have arisen twice within the c o t t i d s . Character 3 5 . Scale ridge shape. State 0 : scale ridge a broad curve ( f i g . 11a). State 1: scale ridge a pointed arch ( f i g s . 11b,11c,11d). State 2: scale ridge a semicircle ( f i g s . 11e,11f,11g,11h). State 3 : no scale ridge. The plesiomorphic state is that found in Hemilepidotus, Icelinus, and Orthonopias. State 1 i s found in Ruscarius. Chitonotus, though i t lacks a ridge (state 3 ) , has i t s c t e n i i aligned in the same pattern. The semicircular arrangement of Artedius sensu s t r i c t o (state 2) is not found in any of the other scaled outgroups, and is hypothesized to be synapomorphic. This character has a d i s j o i n t d i s t r i b u t i o n and i t s u t i l i t y in t h i s study is severely l i m i t e d . Although a transformation series cannot be established, i t nonetheless provides support 74 for the monophyly of two lineages: Artedius sensu s t r i c t o and the lineage including Ruscar ius and Chitonotus. Character 36. Scale ridge placement. State 0: scale ridge not reaching margins of basal plate ( f i g . 11a). State 1: scale ridge reaching anterolateral edge of basal plate ( f i g . 11b). State 2: scale ridge reaching anterior edge of plate only, not extending to l a t e r a l edge ( f i g s . 11c-11h). In Hemilepidotus, the scale ridge does not extend to the edges of the plate (state 0). This condition is hypothesized to be plesiomorphic. State 1, the ridge reaching the antero-l a t e r a l edge of the plate, i s found only in Orthonopias. State 2, the ridge reaching only the anterior edge, i s found in Artedius sensu s t r i c t o , Ruscar ius, and Icelinus. Chitonotus  pugetensis, though i t lacks a ridge, also has c t e n i i extending a l l the way to the anterior margin. Having the scale ridge reach the margins of the plate is hypothesized to be apomorphic r e l a t i v e to Hemilepidotus, and having the c t e n i i or the ridge extending only to the anterior edge is hypothesized to be apomorphic r e l a t i v e to the condition in Orthonopias. This character has an additive transformation series, 0-1-2. Its main use l i e s in separating Hemilepidotus and Orthonopias from the rest of the genera under consideration. Unfortunately no other character shares t h i s d i s t r i b u t i o n . 75 Further corroboration is needed. Character 37. Adult size. State 0: maximum greater than 130 mm standard length. State 1: maximum less than 50 mm standard length. Adult size in the outgroups, and in the cottidae as a whole, i s generally well over 100 mm, ranging up to about 750 mm in Myoxocephalus (state 0). The small (less than 50 mm) adult size of Artedius meanyi (state 1) is not found in any other c o t t i d , and i s therefore t e n t a t i v e l y hypothesized to be autapomorphic. Character 38. Form of anal rays. State 0: f i r s t three anal rays unmodified in adult males. State 1: f i r s t three anal rays thickened and short (50% of the length of the f i f t h ray) ( f i g . 7). Many co t t i d s have modified anal rays in mature males, notably 01igocottus maculosus and Oligocottus snyderi. In these two species the f i r s t two anal rays are substantially thickened, and are at least as long as, i f not longer than, the t h i r d , unmodified, ray. In 0. snyderi, they are also set off by a notch in the f i n membrane i t s e l f . The rest of the outgroups under consideration do not have modified anal rays in the males. In Ruscarius meanyi (state 1), the f i r s t two anal rays are thickened, their length about half of the length of the f i f t h 76 (the f i r s t normal) ray. The t h i r d ray, which i s not thickened, is the same length as the f i r s t two, while the fourth ray is approximately 75% the length of the f i f t h . This is interpreted to represent a d i s t i n c t state from that found in 01igocottus, and to be autapomorphic for R. meanyi at the current level of analysis. Character 39. Form of head tubercles. State 0: head tubercles unmodified. State 1: tubercles raised dorsally, adorned with irregular groups of knoblets. Porocottus, Ereunias, Dasycottus, Icelus and Myoxocephalus also have elaborate bony structures on the head, none of which occur in the same regions as those found in Artedius  notospilotus (state 1). In Myoxocephalus, these are either in the form of longitudinal ridges of various forms (along the 'fronto-parietal ridge', Bolin 1947), such as in M. scorpius, M. scorpioides, M. octodecimspinosus, and M. s t e l l e r i , or they are short sections of ridge with erose d i s t a l surfaces, found in M. quadricornis and M. groenlandicus. Although four of the tubercles in Artedius notospilotus are aligned with the fronto-p a r i e t a l ridge (which is not raised in Artedius), they arise from rounded bases and are globose d i s t a l l y . The bony head structures in other c o t t i d genera and A. notospilotus are therefore not considered to be homologous, and the state in A. notospilotus is hypothesized to be autapomorphic. 77 Character 40. Serrations on the posttemporal and supraclei thrum. State 0: absent. State 1 : serrations present on posterior of posttemporal and upper border of supracleithrum. In Orthonopias, Clinocottus, 01igocottus, Chitonotus, Hemilepidotus, Icelinus and Ruscar ius the posttemporal and supracleithrum are unmodified (state 0). The presence in Artedius notospilotus of serrations (state 1) i s therefore postulated to be autapomorphic. Character 41. Size of body c i r c l e s . State 0: no tiny c i r c l e s of l i g h t e r pigment scattered over l a t e r a l body surface ( f i g s . 6,7). State 1: tiny c i r c l e s of l i g h t e r background pigmentation scattered over l a t e r a l body surface ( f i g s . 1,2). State 2: as in state 1, but c i r c l e s dense ( f i g s . 3,4,5). These c i r c l e s are not found in the outgroups (state 0). Within Artedius, they are found only in A. harr ingtoni, A. c o r a l l i n u s and A. l a t e r a l i s (state 1). The condition in A. l a t e r a l i s and A. c o r a l l i n u s , in which the small c i r c l e s are p a r t i c u l a r l y dense (state 2), is hypothesized to be a further derived condition. This character, because of i t s non-additive d i s t r i b u t i o n 78 within Artedius, cannot serve as strong evidence for i n t r a -generic relationships. It can, however, contribute support for the s i s t e r group status of Artedius c o r a l l i n u s and Artedius  l a t e r a l i s , as well as the i n t e g r i t y of the clade containing A. l a t e r a l i s , A. c o r a l l i n u s , and A. harrinqtoni. Three transformation series are equally l i k e l y : 0-1-2, 0-2-1 and 0-1 character 42. White throat c o l o r a t i o n . State 0: none. State 1: white spots present on throat, in regular pattern ( f i g . I0e). The presence of white vermiculations on the underside of the throat in Orthonopias (statel) is hypothesized to be autapomorphic, as i t is not found in any of the other outgroups, in Ruscarius or in Artedius sensu s t r i c t u ( a l l state 0). The spots appear in a regular pattern, with a "Y" just behind the median mandibular pore, and a series of regularly grouped spots proceeding caudally along the branchiostegal membrane. Character 43. White spots on body. State 0: absent. State 1: present ( f i g . 8). In the area which in Artedius is covered by c i r c l e s , Orthonopias has a series of white spots superimposed over the 79 underlying darker pigmentation (state 1). This white pigment is d i s t i n c t l y l i g h t e r that the l i g h t background pigmentation of the ventral body surface, so that, beyond the ventral margin of darker pigment, the white c i r c l e s are c l e a r l y v i s i b l e against the l i g h t e r body color. The margins of the c i r c l e s are i n d i s t i n c t , in contrast to the sharply-defined c i r c l e s of Artedius sensu s t r i c t u (state 0). Character 44. C i r r i on nasal spine. State 0: a simple slender c i r r u s a r i s i n g from base of each nasal spine. State 1: no c i r r u s on each spine. In Orthonopias and Chitonotus pugetensis, there i s a long slender c i r r u s on each nasal spine (state 0). This is also found in Ruscar ius c r e a s e r i , Artedius harringtoni, A. c o r a l l i n u s and A. l a t e r a l i s . Possession of a nasal c i r r u s i s hypothesized to be the plesiomorphic condition for Artedius. F a r r i s optimization suggests losses (state 1) for this feature in Ruscar ius meanyi, in the stem leading to A. f e n e s t r a l i s and A. notospilotus, once each in 01igocottus (0. maculosus) and Clinocottus (C. globiceps), and at least three times for the four Icelinus species which lack i t (l_. burchami, I_. cavif rons, I_. quadriseriatus, I_. tenuis) , i f Bolin's (1947) diagram is correct. Character 45. Pterotic flange. 80 State 0: triangular flange at posterior edge of pterotic short, not reaching posterior edge of cranium ( f i g . 12a). State 1: flange long, extending to posterior margin of cranium ( f i g . 12b). State 2: as in state 1, but d i s t a l edge of lower flange raised to form a ridge perpendicular to plane of main flange ( f i g . 12c). The posterior flange of the pterotic consists of two p a r a l l e l triangular elements which extend in a ventro-caudal d i r e c t i o n . The lower of the two i s always the longest. In Icelinus, Hemilepidotus, Clinocottus, Oligocottus, Ruscar ius  meanyi, R. creaseri and Orthonopias, this flange i s always short, never extending to the posterior edge of the cranium (state 0). In Artedius sensu s t r i c t o this flange i s very much longer, continuing to the posterior end of the neurocranium. This state (state 1) is hypothesized to be synapomorphic for Artedius sensu str i c t o . State 2, found in A. c o r a l l i n u s and A. l a t e r a l i s , is hypothesized to represent a further derived condition and be apomorphic at that l e v e l . This character i s considered to provide strong evidence for these groups as the transformation series i s additive (0-1-2). Character 46. O s s i f i c a t i o n of opercle. State 0: opercle completely o s s i f i e d . State 1: ventro-caudal t h i r d of opercle not o s s i f i e d . In CIinocottus, 01iqocottus and Artedius sensu str i c t o , the 81 ventro-caudal section of the opercle i s never completely o s s i f i e d and does not take up a l i z a r i n red (state 1). In Chitonotus, Orthonopias, Hemilepidotus, Icelinus, Ruscarius  meanyi and R. c r e a s e r i , the opercle i s always completely o s s i f i e d (state 0). State 1 i s hypothesized to be a derived t r a i t . Character 47. Scales on eye. State 0: no scales on eye. State 1: scales on surface of eye, arranged in crescent covering upper f i f t h of eye, never covering p u p i l . Scales on the eye are found in Chitonotus pugetensis and Ruscar ius meanyi (state 1). F a r r i s optimization postulates that the most parsimonious explanation i s one of a single o r i g i n , in the stem leading to Ruscarius and Chitonotus, with a loss in Ruscar ius c r e a s e r i . These scales are not found in Artedius sensu s t r i c t u (state 0). 82 RESULTS Adult Trees Monophyly and Relationships of Artedius The Wagner program produced three trees of 86 steps ( f i g s . 13, 14, 15), each with a consistency index of 76.74. Each tree recognizes a monophyletic clade composed of fiv e of the seven nominal Artedius species (A. notospilotus, A. f e n e s t r a l i s , A. harringtoni, A. l a t e r a l i s , and A. c o r a l l i n u s ) . R. meanyi and R. creaseri are placed in a separate clade, not c l o s e l y related to the remainder of Artedius. The monophyly of Artedius  sensu s t r i c t o i s supported by six synapomorphic characters: 3 (form of scale ridge), 4 (body color pattern), 12 (chin col o r a t i o n ) , 27 ( p o s t c l e i t h r a ) , 35 (scale ridge shape), and 45 (pterotic flange) (see f i g . 16). Within Artedius, two separate clades are i d e n t i f i a b l e : A. notospilotus plus A. f e n e s t r a l i s supported by synapomorphies in characters 3 (body color pattern), 12 (chin coloration), 11 (male c o l o r ) , 14 (extra l a t e r a l l i n e pores), 15 (form of head scales), 32 (cirri.above a x i l l a ) , and 44 ( c i r r i on nasal spine); and a clade including A. harringtoni, A. c o r a l l i n u s and A. l a t e r a l i s , with synapomorphies in characters 4 (body color pattern), 12 (chin coloration) and 41 (size of c i r c l e s on body). Within the l a t t e r clade, A. c o r a l l i n u s and A. l a t e r a l i s are 83 s i s t e r species, sharing apomorphic states in characters 12 (chin co l o r a t i o n ) , 17 ( c i r r i on suborbital stay), 32 ( c i r r i above a x i l l a ) , 41 (size of c i r c l e s on body) and 45 (pterotic flange) (see f i g . 16). Each of the three trees placed Artedius, Oliqocottus and Clinocottus in one monophyletic lineage. The evidence supporting t h i s p a r t i c u l a r clade i s sl i g h t and consists only of character 49 ( o s s i f i c a t i o n of opercle). Oliqocottus and Clinocottus are hypothesized to be si s t e r taxa in thi s study mainly on the basis of their lack of scales. Since many c o t t i d genera lack scales t h i s i s obviously not strong evidence for genealogical relationship. In the absence of further character evidence no refutation i s possible but alternatives are apparent. For present purposes i t should be s u f f i c i e n t to note that these two aspects of the cladogram (the Oliqocottus-Clinocottus-Artedius clade, and the 01igocottus-Clinocottus clade) are provisional and require further t e s t i n g . Monophyly of Ruscar i us Ruscar ius meanyi and Ruscar ius creaseri do not belong to the monophyletic lineage made up of the other five Artedius species, although they do constitute a separate clade of their own as evidenced by characters 3 (form of scale ridge), 4 (body color pattern), 7 (shape of upper preopercular spine) and 11 (male color) (see f i g . 17). Thus as Washington (1982) 84 suggested the genus Artedius sensu Bolin (1947) is demonstrably d i p h y l e t i c . Alternative trees - the relationships of Ruscar ius The only ambiguity present in the trees for the adult data concerns the relationships of Chitonotus, Ruscarius meanyi, and Ruscarius c r e a s e r i , e s p e c i a l l y in r e l a t i o n to the lineage of Artedius, Clinocottus, and 0 1igocottus. Three equally parsimonious alternatives emerge from the Wagner analysis, each d i f f e r i n g only in the placement of Chitonotus: 1 . Ruscarius is the s i s t e r lineage to the 0 1igocottus-Clinocottus-Artedius lineage, while Chitonotus and Icelinus are placed sequentially below Ruscar ius on the tree ( f i g . 13). 2. Chitonotus plus Ruscarius form a monophyletic lineage, while Icelinus i s placed between the basal genera (Orthonopias and Hemilepidotus) and the Chitonotus-Ruscarius 1ineage ( f i g . 14). 3. Chitonotus i s placed as the s i s t e r taxon to the 0 1igocottus-Clinocottus-Artedius lineage, followed by the increasingly plesiomorphic Ruscarius and Icelinus ( f i g . 15). Unfortunately, the only characters which are derived at the l e v e l that includes R. meanyi, R. c r e a s e r i , Chitonotus, and Icelinus are non-additive multistate characters, which, are not very useful in elucidating relationships. In the absence of binary characters at this l e v e l , the branching sequence remains 85 ambiguous. This ambiguity appears in the Adams consensus tree as a trichotomy, from which stem Chitonotus, Ruscarius and the Artedius-01igocottus-Clinocottus clade ( f i g . 18). Although there is ambiguity in the Adams tree in the placement of Chitonotus, the genus Ruscar ius appears to be closer to Chitonotus and the Artedius-Qligocottus-Clinocottus clade than i t is to Icelinus. Washington's (1982) diagram suggests the reverse. At present there is no evidence on which to base a preference for one of the three trees ( f i g s . 15, 16, 17) over another, since the position of Chitonotus cannot be resolved given the characters at hand. In binary characters, Chitonotus has losses, in two (characters 9 and 17) and is hypothesized to be homoplasious in another (character 47), leaving only character 6 as a useful binary character. Chitonotus also shares with Icelinus apomorphic states in two multistate characters which have d i s j o i n t d i s t r i b u t i o n s over the entire tree (characters 4 and 35) and has a state in another character which i s non-additive (character 3). This leaves the additive character 36, for which i t shares the same state as Ruscar ius, Icelinus and Artedius. F i n a l l y , Hemilepidotus and Orthonopias remain unresolved at the base of the phylogenetic hypothesis, since they are plesiomorphic for a l l of the characters with respect to the rest of the taxa at hand. Orthonopias i s highly autapomorphic, es p e c i a l l y in body coloration (characters 45 and 46). The Wagner program produced only one tree ( f i g . 24) (for the data set excluding R. meanyi to make i t comparable to 86 Bolin's 1947 tree). Artedius, 01igocottus and Clinocottus are f u l l y resolved, with the same branching sequence as in the trees in which R. meanyi was included. R. creaseri and Chitonotus are placed together as a separate lineage sharing a common ancestor with the Artedius-Qligocottus-Clinocottus clade. Icelinus i s plesiomorphic r e l a t i v e to a l l of the above taxa. F i n a l l y , Hemilepidotus and Orthonopias are again relegated to a basal polytomy. Larval Trees Three trees of seventeen steps ( f i g s . 21, 22, and 23) emerged from the Wagner analysis of the l a r v a l data extracted from Washington's (1982) study. A l l three trees have a consistency index of 82.35. None of these trees i s id e n t i c a l to Washington's collapsed tree. One tree ( f i g . 19) duplicates exactly Washington's (1982) branching sequences within her genera, although Artedius, 01igocottus, and Clinocottus emerge from an unresolved trichotomy. The part of the tree including R. meanyi, R. creaseri and Icelinus is the same as in the tree resulting from collapsing unsupported branches in Washington's dichotomous tree ( f i g . 25). In the second tree ( f i g . 20), Clinocottus is resolved as in the collapsed tree, as i s the R. meanyi-R. creaseri-Icelinus clade. 01igocottus i s now the s i s t e r taxon to Artedius, (less A. harringtoni, which emerges from a trichotomy with Clinocottus 87 and the 01igocottus-Artedius lineage). The t h i r d and f i n a l l a r v a l cladogram ( f i g . 21) also has 01igocottus and Artedius (less harringtoni) in one lineage. This branch arises from a trichotomy involving two other lineages: Clinocottus (less C. acuticeps) and a clade made up of Artedius harringtoni and CIinocottus acut iceps. The R. meanyi-R. creaseri-Icelinus portion of the tree i s the same as in the f i r s t two trees. Figure 24 i s the Adams consensus tree derived from these three l a r v a l Wagner trees. With respect to Artedius, the l a r v a l tree is of l i t t l e use, since i t does not include a l l of the species and the relationships of those i t does include are not resolved. The l a r v a l analysis does demonstrate, however, that R. meanyi and R. creaseri do not belong to the lineage including the rest of the Artedius species under consideration. 01igocottus and CIinocottus are in fact more clos e l y related to Artedius than are either R. meanyi or R. c r e a s e r i . This can be seen in the Adams consensus tree for the three l a r v a l trees ( f i g . 22). Icelinus plus R. meanyi plus R. creaseri constitute one resolved clade, while the remainder of the tree i s composed of f i v e branches emerging in a polytomy: 01igocottus, Artedius  f e n e s t r a l i s plus Artedius harringtoni plus Artedius type 3, Artedius harringtoni, Clinocottus acut iceps, and. the remainder of Clinocottus. 88 COMPARISON OF CLASSIFICATIONS Adults The tree published by Bolin (1947) ( f i g . 23) requires fiv e steps more than the Wagner tree calculated in the absence of R. meanyi ( f i g . 24) and raises the number of homoplasious characters from six to seventeen, or from twelve percent to thi r t y - f o u r percent. In addition, when the character data from this study are mapped onto Bolin's (1947) tree, six branches remain unsupported. Although he can hardly be faulted for f a i l i n g to represent my data as well as my tree did, based on my analysis I cannot accept his hypothesis of relationships. Compared to the Manhattan distance matrix calculated from the character data set, the Wagner tree has an R value of 0.97, whereas Bolin's tree has an R value of 0.45. Clearly the Wagner tree better represents the data in this study. Interestingly, as far as could be determined from his text, my study includes most of Bolin's (1944) character data. The main difference between his tree and my wagner tree l i e s in the relationship between Artedius harrinqtoni and Ruscar ius c r e a s e r i . Both studies recognize two eas i l y i d e n t i f i a b l e pairs of s i s t e r species within Artedius: Artedius  fenestralis-Artedius notospilotus, and Artedius l a t e r a l i s - A r t e d i u s c o r a l l i n u s . Bolin placed Artedius harrinqtoni as the s i s t e r species to A. c r e a s e r i , in the absence 89 of Ruscarius meanyi. The j u s t i f i c a t i o n for placing A. harrinqtoni and A. creaseri together lay mainly in the similar d i s t r i b u t i o n of scales. Since no other corroborative characters have been discovered, t h i s character i s not a r e l i a b l e indicator of phylogenetic relationship. The remaining Artedius species with reduced squamation and no p r e o r b i t a l c i r r i are separated from R. creaseri and A. harringtoni. F i n a l l y , he placed Orthonopias as the s i s t e r taxon of Artedius. The present analysis offers quite a d i f f e r e n t hypothesis of relationships. The two pairs of Artedius s i s t e r species are again recognized, but Artedius harr ingtoni i s placed closest to Artedius c o r a l l i n u s plus Artedius l a t e r a l i s . In the absence of R. meanyi, A. creaseri i s placed with Chitonotus and i s removed from the genus Artedius e n t i r e l y . The lineage including 01iqocottus and Clinocottus is hypothesized to be the s i s t e r group to Artedius sensu str i c t o . The formerly hypothesized s i s t e r species of Artedius sensu Bolin, Orthonopias t r i a c i s , i s removed to a basal trichotomy with Hemilepidotus, separate from Chitonotus, Icelinus, Ruscar ius, Oligocottus, Clinocottus, and Artedius• Contrary to Bolin's (1947) claim that "Orthonopias i s c l e a r l y derived from the Artedius l i n e " (p. 162). This study shows that the characters which Bolin (1947) used to unite Orthonopias with Artedius are plesiomorphic at that l e v e l and therefore inadmissable as evidence for s i s t e r group status. In addition, Orthonopias i s highly autapomorphic, especially in the form of the pelvic f i n s in the male, the form of the body scales 90 and in the pattern of body coloration. Such c h a r a c t e r i s t i c s y i e l d l i t t l e evidence for placement of Orthonopias with any of the other taxa under consideration in t h i s study, although Howe and Richardson's (1978) suggestion that i t be included in Artedius appears to be t o t a l l y unsupported. The shortcomings of Bolin's (1947) analysis seem to stem from methodology, since his character analysis i s c e r t a i n l y very detailed. The root of the problem l i e s in placing forms with similar scalation patterns together, on the a p r i o r i assumption that, within the Cottidae, there is a trend towards reduction in squamation. This assumption i s then used to order the species within the Cottidae. The point that i s missed is that such trends cannot be i d e n t i f i e d in the absence of a l o g i c a l l y and empirically constructed phylogenetic hypothesis. In this p a r t i c u l a r instance, such a generalization has l i t t l e , i f any, u t i l i t y in reconstructing genealogy. Witness the placement of the "scaly" Artedius harringtoni with the the two species which show the highest degree of reduction of scalation within Artedius, A. l a t e r a l i s and A. c o r a l l i n u s , and the placement, though admittedly s t i l l p r ovisional, of two t o t a l l y non-scaled forms as the s i s t e r lineage of the scaled genus Artedius. In addition, not a l l things that look l i k e an "Artedius" are Artedius, as evidenced by the removal of R. creaseri and R. meanyi to a separate genus outside the 01iqocottus-Clinocottus-Artedius group and the placement of Orthonopias at the unresolved base of the tree diagram far removed from Artedius. These two results provide an example of 91 the f u t i l i t y of grouping by o v e r a l l s i m i l a r i t y when one i s trying to establish genealogical relationships. Larvae None of the mininum-length trees ( f i g s . 19, 20, and 21) from the Wagner analysis are the same as those presented by Washington (1982) ( f i g s . 25 and 26), one of which ( f i g . 25) i s supposedly derived from an "unrooted Wagner analysis". In the following comparisons, the two trees derived from the l a r v a l study w i l l be referred to as the collapsed tree and the f u l l y resolved or dichotomous tree. Inspection of her data reveals that t h i s dichotomous tree has no character j u s t i f i c a t i o n for the complete resolution of relationships within Artedius and Clinocottus. The j u s t i f i c a t i o n for her f u l l y resolved tree remains mysterious. It is possible that the relationships depicted in her cladogram are in fact those proposed by Bolin in 1947, who placed A. l a t e r a l i s and A. f e n e s t r a l i s in a lineage separate from that including A. creaseri and A. harringtoni. Since Washington removed A. creaseri to a lineage with Icelinus and A. meanyi, this would lead to the placement of A. l a t e r a l i s and A. f e n e s t r a l i s in a monophyletic group apart from A. harringtoni. However, the problematical Artedius type 3 clouds the issue. This i s either A. c o r a l l i n u s , A. notospilotus or a combination of both, and i t s inclusion in the analysis 92 results in Artedius type 3, A. f e n e s t r a l i s and A. l a t e r a l i s a r i s i n g from a trichotomy. The j u s t i f i c a t i o n for the resolution of relationships within Clinocottus in the dichotomous tree cannot be found with Bolin (1947), since the branching sequence depicted by Washington (1982) does not correspond to his tree, the only other published tree for t h i s genus. In the absence of character data in the body of her text, Artedius collapses to A. harrinqtoni plus a trichotomy including A. f e n e s t r a l i s , A. l a t e r a l i s and A. type 3_. Cl inocottus collapses to C. acut iceps plus a trichotomy involving C. anali s, C. embryum and C. recalvus. As with the adult tree of Bolin, the goodness of f i t of the l a r v a l tree presented by Washington is considerably less than that of the Wagner trees (see table 1). Table 1. Comparison of f i t of l a r v a l trees. R-value Wagner tree 1 96.96 Wagner tree 2 97. 1 0 Wagner tree 3 96.57 Washington (1982) dichotomous 76.43 Washington (1982) polytomous 68.99 One of the Wagner trees ( f i g . 19) reproduces exactly the 93 branching sequence within genera shown in Washington's collapsed tree ( f i g . 26). However, in my tree, Artedius, 01igocottus and Clinocottus emerge from a trichotomy and are not resolved as in Washington's tree. Three explanations may be offered to explain my i n a b i l i t y to reproduce her trees: I seriously erred in the preparation of a data matrix from Washington's text; the characters were poorly analyzed in the text, but I coded them as she intended; or, i f I coded the data cor r e c t l y and the o r i g i n a l character analysis was sound, her a n a l y t i c a l technique is not, as claimed, a Wagner (parsimony) method. F i r s t , Washington's analysis of the characters. With one notable exception, Washington (1982) polarized the characters using outgroup analysis. Given the information in the text these are above suspicion. One character, number 11, she apparently polarized by use of the common-equals-primitive c r i t e r i o n . "This condition (rounded snout) i s probably the primitive condition r e l a t i v e to larvae of Artedius, Clinocottus, and 01igocottus, because i t is widespread in (larvae of) several divergent genera of Cottids and Scorpaeniformes. (therefore) the pointed snout appears to be a derived condition." (p. 172). It seems, though, that the alternative character state also exhibits a widespread d i s t r i b u t i o n , which caused her to comment: "snout length is variable in the outgroup taxa." (p. 171). Unfortunately, t h i s character (snout length) is one of two that i s used to j u s t i f y the monophyly of a group including Icelinus, A. c r e a s e r i , and A. meanyi. It must be noted here 94 that the shape of the snout in l a r v a l forms does not correspond to snout shape in adult forms (character 2 in the adult a n a l y s i s ) . The other l a r v a l character used to j u s t i f y this group's monophyly (basal preopercular spine) was polarized by outgroup comparison, but not a l l Icelinus species were examined. A single character, pelvic f i n ray number, was used to j u s t i f y the group A. meanyi plus Icelinus. Next l e t us consider the p o s s i b i l i t y that I did not code the data as Washington (1982) intended. Washington (1982) offered 11 characters to be used in the phylogenetic analysis (although for reasons not stated, two of these were not included on the cladogram). Of these, seven are binary characters and four are multistate. Again assuming that Washington was able to determine the plesiomorphic state for the binary characters, there i s l i t t l e doubt that these were coded as she intended. The analysis of the multistate characters was also straightforward. In the interest of fairness, alternative coding schemes were t r i e d , in an attempt to reproduce the branching sequence in her collapsed tree. These invariably resulted in a tree with one additional step r e l a t i v e to the o r i g i n a l coding scheme. None of these trees had the same topology as the trees in Washington's text. I can only conclude that no coding scheme w i l l y i e l d her tree when subjected to a Wagner analysis. I am s a t i s f i e d , though, that my o r i g i n a l coding of her data results in the most parsimonious cladogram. Although I can reproduce ( f i g . 19) the same topology within genera as in her collapsed tree ( f i g . 26) there does not seem 95 to be support in her character data for the placement of Oligocottus and Clinocottus as s i s t e r taxa. Although Richardson's (1981) discussion of intergeneric relationships within the cott i d s was preliminary and presented no branching diagram, Washington (1982) pointed out that Ruscarius meanyi was included in Richardson's (1981) study under "Icelus spp.", which makes possible a comparison of Richardson's (1981) phenetic groupings with the results from th i s study. Two of her groups are of p a r t i c u l a r interest: (1) the Artedius-01igocottus-Clinocottus-Orthonopias group and (2) the group including Chitonotus, Icelus and Icelinus. Two major points of agreement emerge: the separation of the Artedius~Clinocottus-01igocottus group from Icelinus and the notion that R. meanyi does not belong with Artedius. A l l three of the l a r v a l analyses (one phenetic, one phylogenetic in name only and one Wagner) agree that R. meanyi (or R. meanyi and R. creaseri) does not share an immediate common ancestor with the rest of Artedius sensu s t r i c t o . 96 CONGRUENCE OF LARVAL AND ADULT CLASSIFICATIONS Figure 27 shows the consensus tree derived from the consensus of the adult Adams tree and the l a r v a l Adams tree, (note that only those taxa common to both analyses are present in the f i n a l consensus). The large amount of amgibuity stems from the i n a b i l i t y of the l a r v a l tree to resolve relationships. This ambiguity i s also a function of the size of the data set; more characters are needed. Despite the shortcomings, there are two areas of agreement which merit discussion. Both adult and l a r v a l Adams trees recognize that the closest r e l a t i v e of Artedius is either 01igocottus or Oligocottus plus Clinocottus. This is important in that neither of the l a t t e r genera is scaled. A widespread assumption in c o t t i d systematics is that there is a general trend toward reduction of scales within the family (see Bolin, 1947). Unsealed forms are therefore grouped with other unsealed forms. In the context of my analysis, the use of this trend has limited u t i l i t y in sorting out relationships, as there are no corroborative characters with the same d i s t r i b u t i o n . As a general rule of thumb the reduction of scales within the c o t t i d s may be a useful concept, but i t s use in p o l a r i z i n g characters and taxa is l i m i t e d . This study hypothesizes that two completely unsealed genera, 01igocottus and Clinocottus, are the closest r e l a t i v e s of Artedius (although Oligocottus rimensis does have dermal bony p r i c k l e s ) . Orthonopias, which has a scalation pattern generally similar to Artedius, i s not hypothesized to be c l o s e l y related to the l a t t e r . 97 The second major area of agreement between the adult and l a r v a l Adams trees is the hypothesis that Ruscar ius meanyi and Ruscarius c r e a s e r i , formerly in Artedius, do not in fact belong with the rest of the nominal Artedius, although the l a r v a l tree is unable to completely resolve the placement of Ruscar ius. Although Washington's (1982) analysis c o n f l i c t e d with the adult analysis in i t s placement of R. meanyi, R. cre a s e r i , and Icelinus, i t did serve to focus attention on a central issue in systematics: monophyly. With respect to methodology, however, her study i s i n s t r u c t i v e . The l a r v a l study served to focus attention on the lack of j u s t i f i c a t i o n for the monophyly of Artedius• At the le v e l at which there was character support, Washington's study did recognize that R. meanyi and R. creaseri did not belong with the rest of Artedius sensu str i c t u , which is something that the evolutionary taxonomic study of Bolin (1947) f a i l e d to uncover. A s t r i c t congruence study using the results of Richardson (1981) was not possible, due to the uncertainties in i d e n t i f i c a t i o n and the di f f e r e n t taxa examined. One point is worth noting: even th i s admittedly phenetic grouping suggested that an Artedius including R. meanyi is not monophyletic, a result supported by the analysis based on adult characters. Both adult and l a r v a l cladograms remove R. creaseri and R. meanyi from Artedius sensu str i c t u . Due to the unresolved nature of much of the l a r v a l cladogram, i t i s not possible to test for further congruence. S t i l l , in the area that jjs resolved in both l a r v a l and adult analyses, there i s congruence. 98 In the presence of a great deal of noise in the l a r v a l data, phylogenetic analysis provided the same hypothesis for the placement of Ruscarius r e l a t i v e to Artedius as did the adult analysis. The only other systematic treatment of Artedius and i t s r e l a t i v e s (Bolin, 1947) did not recognize the d i p h y l e t i c condition of Artedius. Congruence of c l a s s i f i c a t i o n s i s also important in comparing theories of evolutionary change when e x p l i c i t predictions regarding congruence can be derived from those theories. Within current evolutionary theory there seems to be a range of statements regarding congruence. De Beer (1958) found that c l a s s i f i c a t i o n s based on adult and l a r v a l characters were most often congruent, even in the face of the obvious adaptation of the l a r v a l forms to their habitats. Charlesworth et a l . (1982), l i k e De Beer (1958), do not make any s p e c i f i c predictions. They state that phylogenetic patterns are due predominately to the action of natural selection (1982, p. 490). I_f natural selection alone shapes the organism, morphological change should be better correlated with environment than with phylogeny. If we then compare c l a s s i f i c a t i o n s derived from d i f f e r e n t l i f e history stages such as larvae and adults, which are presumably subjected to vastly d i f f e r e n t s e l e c t i v e regimes, we would not predict congruence. A s t r i c t "selective regime" explanation grows more convoluted each time congruence is found. The non-equilibrium framework of Wiley and Brooks (1983, also Brooks and Wiley, 1983) does predict congruence, since i t e x p l i c i t l y recognizes the 99 constraining role of history. Congruence i s then a consequence of d i f f e r e n t data sets r e f l e c t i n g the same phylogenetic history. Eventually we w i l l reach a point where our understanding of the pattern of order in nature forces us to re-examine our notions about causal processes. Either we restructure our ideas of process to accomodate features of the natural world such as congruence, or we abandon our former ideas e n t i r e l y in favor of something that better accounts for the data. 1 0 0 SUGGESTIONS FOR FUTURE STUDIES Now that a phylogeny i s available for Artedius sensu  s t r i c t u , there are other important questions that can be examined. F i r s t l y , we can map morphometric data onto th i s cladogram in order to i d e n t i f y h i s t o r i c a l trends in changes in body shape and s i z e . This mapping technique can in fact be generalized to test for trends in any continuous character. Other types of analyses are also possible with respect to what might be c a l l e d " h i s t o r i c a l ecology". The a v a i l a b i l i t y of a cladogram makes possible analysis of host-parasite relationships, leading to studies of coevolution in general. With th i s cladogram, f i e l d ecologists now have the necessary background for studies of speciation. Community ecologists have a v i t a l piece of information necessary in estimating the h i s t o r i c a l component of ecological associations. Biogeographers can study the relationship between the history of these f i s h and the history of the areas in which they l i v e . In the l i g h t of the si t u a t i o n in Artedius, the obvious higher-level question a r i s e s : which other c o t t i d genera are in need of phylogenetic analysis? In how many of the other genera has a robust estimate of phylogeny remained obscure, even in the face of evolutionary-taxonomic study? Of course t h i s question can reasonably be asked of any taxon, not just c o t t i d s . Within the problem at hand i t seems c r i t i c a l that the in t e r r e l a t i o n s h i p s of Icelinus, Chitonotus, Ruscar ius, 01igocottus, Clinocottus and Artedius be examined, especially in 101 the l i g h t of the meagre evidence for the relationships of Artedius, Clinocottus and 0 1igocottus. These three genera have commonly been grouped together, but l i t t l e character support has been forthcoming. Icelinus i s probably next in l i n e for thorough study, followed by Clinocottus and 0 1igocottus. F i n a l l y , other scaled genera such as Icelus merit examination, since i t is becoming apparent that "scaliness" i s not a very r e l i a b l e indicator of intergeneric relationships. Our knowledge of v i r t u a l l y a l l of the c o t t i d s in the northern and western P a c i f i c , p a r t i c u l a r l y those in Soviet waters, would benefit from phylogenetic analysis. It is possible that the Cottidae as i t stands today i s not a monophyletic assemblage, but suggestions to that effect have as yet been made without a n a l y t i c a l support. 102 LITERATURE CITED Adams, E.N. 1972. Consensus techniques and the comparison of taxonomic trees. Syst. Zool. 21:390-397. Andersen, N.M. 1979. Phylogenetic inference as applied to the study of evolutionary d i v e r s i f i c a t i o n of semiaquatic bugs (Hemiptera: Gerromorpha). Syst. Zool. 28:554-578. Ayres, W.O. 1855. A communication to attempt a correct exposition of the two species of the genus Hemilepidotus Proc. Cal. Acad. Nat. S c i . 1:75-77. Baird, R.C, and M.J. Eckardt. 1972. Divergence and relat i o n s h i p in deep-sea hatchetfishes (Sternoptychidae). Syst. Zool. 21:80-89. Baverstock, P.R., S.R. Cole, B.J. Richardson, and C.H.S. Watts. 1979. Electrophoresis and c l a d i s t i c s . Syst. Zool. 28:214-219. Bean, B.A., and A.C. Weed. 1920. Notes on a c o l l e c t i o n of fishes from Vancouver Island, B.C. Trans. Roy. Soc. Canada, ser. 3 K3 -.69-83 . 4 p i s . Bean, T.H. 1882a. A preliminary catalogue of the fishes of Alaskan and adjacent waters. Proc. U.S. Nat. Mus. 4:239-272. Bean, T.H. 1882b. Notes on a c o l l e c t i o n of fishes made by Captain Henry E. Nichols, U.S.N., in B r i t i s h Columbia and southern Alaska, with descriptions of new species and a new genus (Delolepis). Proc. U.S. Nat. Mus. 4:463-474. Berg, L.S. 1940. c l a s s i f i c a t i o n of fishes, both recent and f o s s i l . Akad. Nauk SSSR Zool. Inst. T.5, 2. Blackburn, J.E. 1973. A survey of the abundance, d i s t r i b u t i o n , and factors a f f e c t i n g d i s t r i b u t i o n of ichthyoplankton in Skagit Bay. M.Sc. Thesis, Univ. Washington, Seattle. 136 pp. 103 Bolin, R.L. 1937. Notes on C a l i f o r n i a fishes. Copeia 1937:63-64, Bolin, R.L. 1941. Embryonic and early stages of the c o t t i d f i s h Orthonopias Starks and Mann. Stanford Ichthyol. B u l l . 3_: 73-82. Bolin, R.L. 1944. A review of the marine Cottid fishes of C a l i f o r n i a . Stanford Ichthyol. B u l l . 3_:1-135. Bolin, R.L. 1947. The evolution of the marine Cottidae of C a l i f o r n i a , with a discussion of the genus as a systematic category. Stanford Ichthyol. B u l l . 3_:153-1 68. Brooks, D.R., and E.O. Wiley. 1983. Systematics, ontogeny and evolution, pp. 00-00. in N. Platnick and C. Humphries (eds.), Advances in C l a d i s t i c s I I I . Columbia University Press, New York. Budd, P.L. 1940. Development of the eggs and larvae of six C a l i f o r n i a fishes. C a l i f . Div. Fish and Game, Fish B u l l . 56, 53 pp. Charlesworth, B., R. Lande, and M. S l a t k i n . 1982. A Neo-Darwinian commentary on macroevolution. Evolution 36:474-498. Churchland, P.M. 1982. Is thinker a natural kind? Dialogue 21: 223-238. Clemens, W.A., and G.V. Wilby. 1946. Fishes of the P a c i f i c coast of Canada. 1st ed. Canada B u l l . Fish. Res. Bd. 68:1-368. Clemens, W.A., and G.V. Wilby. 1961. Fishes of the P a c i f i c Coast of Canada (2nd ed.) . Fish. Res. Board. Canada B u l l . 68:1-368. Colless, D.H. 1981. P r e d i c t i v i t y and s t a b i l i t y in c l a s s i f i c a t i o n s : some comments on recent studies. Syst. Zool. 30:325-331. 1 04 De Beer, G. 1958. Embryos and Ancestors. (3rd ed.) Oxford Unversity Press, Oxford. 197 pp. DeLacy, A.C, B.S. M i l l e r , and S.E. Borton. 1972. Checklist of Puget Sound Fishes. Washington Sea Grant Prog. Publ. WSG 72-3, 43 pp. Dingerkus, G., and L.D. Uhler. 1977. Enzyme clearing of alcian blue stained whole small vertebrates for demonstration of ca r t i l a g e . Stain Tech. 52:229-232. Eigenmann, C.H., and R.S. Eigenmann. 1892. A catalogue of the fishes of the p a c i f i c coast of America north of Cerros Island. Ann. N.Y. Acad. S c i . 6:349-358. Eldredge, N., and J. Cracraft. 1981. Phylogenetic Patterns and  the Evolutionary Process. Columbia Univ. Press, New York. 349 pp. Eschmeyer, W.N., E.S. Herald, and H. Hammnan. 1983. A F i e l d Guide to P a c i f i c Coast Fishes of North America. Houghton M i f f l i n Co. Boston. 335 pp. Evermann, B.W., and E.L. Goldsborough. 1906. The fishes of Alaska. B u l l . U.S. Bur. Fish. 26:219-360. 28 p i s . , 114 f i g s . F a r r i s , J.S. 1970. Methods for computing Wagner trees. Syst. Zool. 19:83-92. F a r r i s , J.S. 1971. The hypothesis of nonspecificity and taxonomic congruence. Ann. Rev. Ecol. Syst. 2_:227-302. F a r r i s , J.S. 1974. Formal d e f i n i t i o n s of paraphyly and polyphyly. Syst. Zool. 23:548-554. F a r r i s , J.S. 1977. On the phenetic approach to vertebrate c l a s s i f i c a t i o n , pp. 823-850 i n : Major Patterns in  Vertebrate Evolution. (M.K. Hecht, P.C Goody, and B.M. Hecht, eds.). NATO Adv. Stud. Inst. Symp. no. 14. Plenum Press, New York. 105 F a r r i s , J.S. 1979a. On the naturalness of phylogenetic c l a s s i f i c a t i o n . Syst. Zool. 28:200-214. F a r r i s , J.S. 1979b. The information content of the phylogenetic system. Syst. Zool. 28:483-519. F a r r i s , J.S. 1982. Simplicity and informativeness in systematics and phylogeny. Syst. Zool. 31:413-444. F a r r i s , J.S., A.G. Kluge, and M.J. Eckhardt. 1970. A numerical approach to phylogenetic systematics. Syst. Zool. 19:172-189. Fink, W.L. 1979. Optimal c l a s s i f i c a t i o n s . Syst. Zool. 28:371 -374. Fink, W.L. 1982. The conceptual relationship between ontogeny and phylogeny. Paleobiology 8 (3_): 254-264 . F i t c h , J.E., and R.J. Lavenberg. 1975. Tidepool and nearshore fishes of C a l i f o r n i a . C a l i f o r n i a Nat. His t . Guide, University of C a l i f . Press, 156 pp. G i l b e r t , C.H., and CV. Burke. 1912. Fishes from the Bering Sea and Kamchatka. B u l l . U.S. Bur. Fish. 30:33-96. 38 f i g s . G i l b e r t , C.H., and J.C. Thompson. 1905. Notes on the fishes of Puget Sound. Proc. U.S. Nat. Mus. 28:973-987. G i l l , T.N. 1862. Notice of a c o l l e c t i o n of the fishes of C a l i f o r n i a , presented to the Smithsonian I n s t i t u t i o n by Mr. Samuel Hubbard. Proc. Acad. Nat. S c i . Philad. 1862(1863):274-282. Girard, C F . 1856. Contributions to the ichthyology of the western coast of the United States, from specimens in the Smithsonian I n s t i t u t i o n . Proc. Acad. Nat. S c i . Philad. 8: 131-137. 106 Girard, C F . 1857a. Report upon fishes c o l l e c t e d on the survey (In: Report of Lieut. Henry L. Abbot, upon explorations for a r a i l r o a d route from the Sacramento valley to the Columbia r i v e r . U.S. Senate Misc. Doc. no. 78, 1857. 33rd Congress, 2nd sess. Girard, C F . 1857b. A l i s t of the fishes c o l l e c t e d in C a l i f o r n i a by Mr. E. Samuels, with descriptions of the new species. Boston Jour. Nat. S c i . 6:533-544. p i s . 24-26. Girard, C F . 1859. Fishes. (In: general report of the zoology of the several P a c i f i c r a i l r o a d routes, 1857). U.S. Senate Misc. Doc. no. 78. 1859. 33rd Congress, 2nd Sess. 400 p. 27 p i s . Greeley, A.W. 1897. Notes on the tide-pool fishes of C a l i f o r n i a , with a description of four new species. B u l l . U.S. Fish. Comm. 1897:7-20. Greenwood, P.H., D.E. Rosen, S.H. Weitzman, and G.S. Myers. 1966. Phyletic studies of teleostean fishes, with a provisional c l a s s i f i c a t i o n of l i v i n g forms. B u l l . Amer. Mus. Nat. Hist. 131:339-456. Gunther, A.C. 1860. Catalogue of the fishes in the B r i t i s h Museum. London. Printed by order of the Trustees. 1859-1870. Halkett, A. 1913. Check l i s t of the fishes of the Dominion of Canada and Newfoundland. King's Printer, Ottawa. 138 pp. Hart, J.L. 1973. P a c i f i c Fishes of Canada. Fish. Res. Bd. Canada B u l l . 180:1-739. Hempel, CG. 1965. Aspects of Sc ient i f ic Explanat ion and other  essays in the philosophy of science. The Free Press, New York. Hennig, W. 1965. Phylogenetic systematics. Ann. Rev. Entomol. 10:97-116. 107 Hennig, W. 1 9 6 6 . Phylogenetic Systematies. University of I l l i n o i s Press, Urbana. 2 6 3 pp. Hood, C.S., and'J.D. Smith. 1 9 8 2 . C l a d i s t i c a l analysis of female reproductive histomorphology in Phyllostomatoid Bats. Syst. Zool. 3J_( 3 ) : 2 4 1 - 2 5 1 . Howe, K.M., and S.L. Richardson. 1 9 7 8 . Taxonomic review and meristic variation in marine sculpins (Osteichthys:Cottidae) of the Northeast P a c i f i c Ocean. MS., NOAA-NMFS Rept. 1 4 2 pp. Hubbard, J.D., and W.G. Reeder. 1 9 6 5 . New l o c a l i t y records for Alaskan fishes. Copeia 1 9 6 5 ( 4 ) : 5 0 6 - 5 0 8 . Hubbs, C L . 1 9 2 6 . Descriptions of new genera of Cottoid fishes related to Artedius. Occ. Pap. Mus. Zool. Univ. Mich. 1 7 0 : 1 - 1 6 . Hubbs, C L . 1 9 2 8 . A check l i s t of the marine fishes of Oregon and Washington. Jour. Pan-Pacific Res. Inst. 3 ( 3 ) : 9 - 1 6 . Hubbs, C.L., and K.F. Lagler. 1 9 5 8 . fishes of the Great Lakes Region. Ann Arbor. Univ. Michigan Press. 2 1 3 pp. Hubbs, C.L., and L.P. Schultz. 1 9 4 1 . Contributions to the Ichthyology of Alaska, with descriptions of two new fishes. Occ. Pap. Mus. Zool. Univ. Michigan. 4 3 1 : 1 - 3 1 . Hubbs, C.L., W.I. F o l l e t t , and L.J. Dempster. 1 9 7 9 . L i s t of the fishes of C a l i f o r n i a . Occ. Pap. Cal. Acad. S c i . 1 3 3 : 1 - 5 5 . Humphries, J.M., F.L. Bookstein, B. Chernoff, G.R. Smith, R.L. Elder, and S.G. Poss. 1 9 8 1 . Multivariate discrimination by shape in r e l a t i o n to size. Syst. Zool. 3 0 : 2 9 1 - 3 0 8 . Jensen, R.J., and CD. Barbour. A phylogenetic reconstruction of the Mexican Cyprinid f i s h genus Algansea. Syst. Zool. 3 0 : 4 1 - 5 7 . 108 Jordan, D.S. 1887. A catalogue of the fishes known to inhabit the waters of North America, north of the tropic of Cancer, with notes on the species discovered in 1883 and 1884. Rept. U.S. Fish. Comm. 1885(1887):789~973. Jordan, D.S. 1905. A Guide to the Study of Fishes. Henry Holt and Co. New York. 2 vols. 624 + 599 pp. Jordan, D.S. 1923. A c l a s s i f i c a t i o n of fishes including families and genera as far as known. Stanford Univ. Publ. Univ. Ser. 3(2):77-243+i-x. Jordan, D.S. 1925. Fishes. D. Appleton and Co. New York. 773 pp. Jordan, D.S., and B.W. Evermann. 1896. A check-l i s t of the fishes and f i s h - l i k e vertebrates on North and Middle America. Rept. U.S. Fish. Comm. 1895 (1896):207-584. Jordan, D.S., and B.W. Evermann. 1898a. The fishes of North and Middle America. Part I I . B u l l . U.S. Nat. Mus. 47 (2):xxx + 1241-2183. Jordan, D.S., and B.W. Evermann. 1898b. The fishes of North and Middle America. Part I I I . B u l l . U.S. Nat. Mus. 47 (3):xxv + 2!83a-3l36. Jordan, D.S., and B.W. Evermann. 1900. The fishes of North and Middle America. Part IV. B u l l . U.S. Nat. Mus. 47 (4):ci + 3137-3317. 958 f i g s . Jordan, D.S., and CH. G i l b e r t . 1881. L i s t of the fishes of the Pa c i f i c coast of the United States, with a table showing the d i s t r i b u t i o n of the species. Proc. U.S. Nat. Mus. 3: 452-458. Jordan, D.S., and CH. G i l b e r t . 1882a. Notes on the fishes of the P a c i f i c coast of the United States. Proc. U.S. Nat. Mus. 4:29-70. Jordan, D.S., and CH. G i l b e r t . 1882b. Synopsis of the fishes of North America. B u l l . U.S. Nat. Mus. 16:lvi + 1018 pp. 109 Jordan, D.S., and CH. G i l b e r t . 1883. Description of a new species of Artedius (Artedius f e n e s t r a l i s ) from Puget Sound. Proc. U.S. Nat. Mus. 5:577-579. Jordan, D.S., and CH. G i l b e r t . 1899. The fishes of the Bering Sea. i n : Jordan, D.S., ed., The Fur Seals and Fur-seal Islands of the North P a c i f i c Ocean, part 3:433-492. 43 p i s . Jordan, D.S., and P.L. Jouy. 1882. Check-list of the duplicates of fishes from the P a c i f i c coast of North America, d i s t r i b u t e d by the Smithsonian I n s t i t u t i o n in behalf of the United States National Museum, 1881. Proc. U.S. Nat. Mus. 4:1-18. Jordan, D.S., and V.L. Kellogg. 1907. Evolution and Animal L i f e . D. Appleton and Co., New York. 489 pp. Jordan, D.S., and E.C. Starks. 1895. The fishes of Puget Sound. Proc. Cal. Acad. S c i . ser. 2 5:785-855. p i s . 76-104. Kincaid, T. 1919. An annotated l i s t of Puget Sound fishes. State of Wash. Fish. Dept. Olympia 51 pp. Kluge, A. 1983. Ontogeny and phylogenetic systematics. pp. OO-OO. in N. Platnick and C. Humphries (eds.), Advances in  C l a d i s t i c s . I I I . Columbia University Press, New York. Lea, R.N. 1974. F i r s t record of the Puget Sound sculpin Artedius  meanyi from C a l i f o r n i a . Canada Jour. Fi s h . Res. Bd. 31: 1242-1243. MacPhee, C , and W.A. Clemens. 1962. Fishes of the San Juan Island Archipelago, Washington. Northwest S c i . 36(2):27-38. Markle, D.F. 1982. I d e n t i f i c a t i o n of l a r v a l and juvenile Canadian Atlantic gadoids with comments on the systematics of gadid subfamilies. Can. Jour. Zool. 60:3420-3438. 1 1 0 Marliave, J.B. 1975. The behavioral transformation from the planktonic l a r v a l stage of some marine fishes reared in the laboratory. Ph.D. Thesis, Univ. B r i t i s h Columbia, Vancouver. 231 pp. Mayr, E. 1942. Systematics and the or i g i n of species. Columbia Univ. Press, New York. 334 pp. Mayr, E. 1969. Princip l e s of Systematic Zoology. Mcgraw-Hill, New York. 428 pp. McAl l i s t e r , D.E. 1960. L i s t of the Marine Fishes of Canada. B u l l . Nat. Mus. Canada 168:1-76. Mc A l l i s t e r , D.E. 1968. Mandibular pore pattern in the sculpin family Cottidae. B u l l . Nat. Mus. Canada 223:58-69. Mickevich, M.F. 1978. Taxonomic congruence. Syst. Zool. 27:143-1 58. Mickevich, M.F. 1980. Taxonomic congruence: Rohlf and Sokal's misunderstanding. Syst. Zool. 29:162-176. Mickevich, M.F. 1982. Transformation series analysis. Syst. Zool. 31:461-478. Mickevich, M.F., and J.S. F a r r i s . 1981. The implications of congruence in Menidia. Syst. Zool. 30:351-370. Mickevich, M.F., and M.S. Johnson. 1976. Congruence between morphological and allozyme data in evolutionary inference and character evolution. Syst. Zool. 25:260-270. M i l l e r , D.J., and R.N. Lea. 1972. Guide to the coastal mar ine f i shes of C a l i f o r n i a . C a l i f . Dept. Fish and Game Fish B u l l . 157:1-249. Miyamoto, M.M. 1981. Congruence among character sets in phylogenetic studies of the frog genus Leptodactylus. Syst. Zool. 30:281-290. 111 Moulton, L.L. 1977. An ecological analysis of fishes inhabiting the rocky nearshore regions of northern Puget Sound. Ph.D. Diss. Univ. Washington. 181 pp Mundinger, P. 1979. C a l l learning in the Carduelinae: ethological and systematic considerations. Syst. Zool. 28: 270-283. Nelson, G.J. 1970. Outline of a theory of comparative biology. Syst. Zool. 19:373-384. Nelson, G.J. 1971. "Cladism" as a philosopy of c l a s s i f i c a t i o n . Syst. Zool. 20:373-376. Nelson, G.J. 1972. Phylogenetic relationship and c l a s s i f i c a t i o n . Syst. Zool. 2J_:227-230. Nelson, G.J. 1973. C l a s s i f i c a t i o n as an expression of phylogenetic relationship. Syst. Zool. 22:344-359. Nelson, G.J. 1978. Ontogeny, phylogeny, paleontology, and the biogenetic law. Syst. Zool. 27:324-345. Nelson, G.J. 1979. C l a d i s t i c analysis and synthesis: p r i n c i p l e s and d e f i n i t i o n s , with a h i s t o r i c a l note on Adanson's Families des Plantes (1763-1764). Syst. Zool. 28:1-21. Nelson, J.S. 1976. Fishes of the World. Wiley-Interscience, New York. 416 pp. Osgood, W.H. 1910. Natural History of the Queen Charlotte Islands, B r i t i s h Columbia, and Natural History of the Cook Inlet Region, Alaska U.S. Dept. of Agric. Div. of B i o l . Surv. North American Fauna 21. 87 pp. Pearcy, W.G., and S.S. Myers. 1979. Larval fishes of Yaquina Bay, Oregon: a nursery ground for marine fishes? Fish. B u l l . 72(1):201-213 . . 1 12 Peden, A.E. 1972. New records of sculpins (Cottidae) from the coasts of B r i t i s h Columbia and Washington. Can. F i e l d Nat. 86(2):169-170. Peden, A.E., and D.E. Wilson. 1976. Di s t r i b u t i o n of i n t e r t i d a l and subtidal fishes of Northern B r i t i s h Columbia and southeastern Alaska. Syesis 9:221-248. Quast, J.C. 1968. New records of thirteen c o t t o i d and blennioid fishes for southeastern Alaska. Pac. S c i . 22(4):482-487. Quast, J . C , and E.L. H a l l . 1972. L i s t of the fishes of Alaska and adjacent waters with a guide to some of their l i t e r a t u r e . NOAA Tech. Rept. NMFS SSRF-658. 47 pp. Regan, C T . 1913. The osteology and c l a s s i f i c a t i o n of the Teleostean fishes of the order Scleroparei. Ann. Mag. Nat. Hist. 11:169-184. Richardson, S.L. 1 9 8 1 . Current knowledge of larvae of sculpins (Pisces:Cottidae and a l l i e s ) in northeast P a c i f i c genera with notes on intergeneric relationships. Fish. B u l l . 7_9(J_) : 1 0 3 - 1 22. Richardson, S.L., J.L. Laroche, and M.D. Richardson. 1980. Larval f i s h assemblages and associations in the North-east P a c i f i c Ocean along the Oregon coast, winter-spring 1972-1975. Estuar. Coast. Mar. S c i . 11:971-699. Richardson, S.L., and W.C Pearcy. 1977. Coastal and oceanic f i s h larvae in an area of upwelling of Yaquina Bay, Oregon. Fish. B u l l . 75(1):125-145. Richardson, S.L., and B.B. Washington. 1980. Guide to i d e n t i f i c a t i o n of some sculpin (Cottidae) larvae from marine and brackish waters off Oregon and adjacent areas of the northeast P a c i f i c . U.S. Dept. Comm., NOAA Tech. Rept. NMFS C i r c . 430, 56 pp. Rosen, D.E. 1974. Cladism or gradism?: a reply to Ernst Mayr. Syst. Zool. 23:446-451. 1 1 3 Rosenblatt, R.H., and D. Wilkie. 1963. A redescription of the rare c o t t i d f i s h , Artedius meanyi, new to the fauna of B r i t i s h Columbia. Canada Jour. Fish. Res. Bd. 20:1505-1511. Schuh, R.T., and J.S. F a r r i s . 1981. Methods for investigating taxonomic congruence and their application to the Leptopodomorpha. Syst. Zool. 30:331-351 . Schuh, R.T., and J.T. Polhemus. 1980. Analysis of taxonomic congruence among morphological, e c o l o g i c a l , and biogeographic data sets for the leptopodomorpha (Hemiptera). Syst. Zool. 29:1-26. Schultz, L.P. 1936. Keys to the fishes of Washington, Oregon, and clo s e l y adjoining regions. Univ. Wash. Publ. Zool. 2(4):103-228. Schultz, L.P., and A.C. DeLacy. 1936. Fishes of the American Northwest. A catalogue of the fishes of Washington and Oregon, with d i s t r i b u t i o n a l notes and a bibliography. Jour. Pan-Pacific Res. Inst. (1935) 10:365-380 and (1936) 11:63-78, 127-142, 211-226, 275-290. Simpson, G.G. 1944. Tempo and Mode in Evolution. Columbia Univ. Press, New York. 237 pp. Simpson, G.G. 1961. P r i n c i p l e s of Animal Taxonomy. Columbia Univ. Press, New York. 247 pp. Somerton, D., and C. Murray. 1976. F i e l d Guide to the f i sh of Puget sound and the northwest coast. Wash. Sea Grant Publ. Univ. of Wash. Press, 70 pp. Starks, E.C. 1896. L i s t of fishes c o l l e c t e d at Port Ludlow, Washington. Proc. Cal. Acad. S c i . ser 2 6^:549-562. p i s . 74-75. Starks, E.C. 1911. Results of an Ichthyological survey about the San Juan Islands, Washington. Ann. Carnegie Mus. 7:162-213. 1 14 Starks, E.C., and E.L. Morris. 1907. The marine fishes of southern C a l i f o r n i a . Univ. C a l i f . Publ. Zool. 3:159-252. Taranets, A.Y. 1941. On the c l a s s i f i c a t i o n and o r i g i n of the family Cottidae. Inst. Fi s h . Univ. B.C. Mus. Cont. 5:1-28. Washington, B.B. 1982. I d e n t i f i c a t i o n and systematics of larvae of Artedius, Clinocottus, and 01igocottus (Scorpaeniformes:Cottidae). M.Sc. Thesis, Oregon State University. 192 pp. Wiley, E.O. 1975. Karl R. Popper, systematics, and c l a s i f i c a t i o n : a reply to Walter Bock and other evolutionary taxonomists. Syst. Zool. 24:233-242. Wiley, E.O. 1978. the evolutionary species concept reconsidered. Syst. Zool. 27:17-26. Wiley, E.O. 1979. An annotated Linnean hierarchy, with comments on natural taxa and competing systems. Syst. Zool. 28:308-337. Wiley, E.O. 1981. Phylogenet i c s . The Theory and Pract ice of Phylogenetic Systematics. John Wiley and Sons, New York. 439 pp. Wiley, E.O. 1982. Convex groups and consistent c l a s s i f i c a t i o n s . Syst. Botany 6:346-358. Wiley, E.O., and D.R. Brooks. 1982. Victims of history - a nonequilibrium approach to evolution. Syst. Zool. 31:1-24. Wilimovsky, N.J. 1954. L i s t of the fishes of Alaska. Stanford Ichth. B u l l . 4(5):279-294. Wilimovsky, N.J. 1958. Provisional keys to the fishes of Alaska. U.S. Fish and W i l d l i f e Serv. MS. 113 pp. Wilimovsky, N.J. 1963. Inshore f i s h fauna of the Aleutian archipelago. Proc. fourteenth Alask. S c i . Conf.: 172-190. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Hemilepidotus 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 o 1 1 0 0 0 0 Orthonopias 1 0 1 2 1 1 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 2 1 1 0 0 0 0 0 1 0 0 1 R.meanyi 0 0 1 2 1 1 1 0 2 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 1 0 1 2 0 0 0 0 0 0 0 0 0 0 0 R.ereaseM 0 0 2 1 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 0 0 0 0 0 0 0 0 0 0 1 Chitonotus 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 IeelInus 0 0 3 3 0 0 0 0 0 0 0 1 1 1 1 1 2 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 0 0 2 2 0 0 0 0 0 0 0 1 t 1 0 A.fenestra!4s 0 0 3 3 0 0 0 0 0 1 0 1 1 1 1 0 0 1 1 0 0 2 0 0 0 0 1 0 0 0 0 1 0 0 2 2 0 0 1 1 0 0 0 1 1 1 0 A.notospilotus 0 0 4 3 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 1 1 1 1 1 1 0 1 0 1 0 2 0 0 1 2 2 0 0 0 0 1 0 0 0 1 1 0 A.herrtngtont 0 0 4 3 0 0 0 0 0 0 0 3 0 0 0 0 1 0 0 2 0 1 0 0 0 1 1 0 1 0 0 1 0 0 2 2 0 0 0 0 2 0 0 0 2 1 0 A.coral 1inus 0 0 4 3 0 0 0 0 0 0 0 3 0 0 0 0 1 0 0 0 0 2 0 0 0 0 1 0 0 0 0 1 0 0 2 2 0 0 0 0 2 0 0 0 2 1 0 A. lateral is 0 0 S 4 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 3 0 0 0 0 0 0 0 0 0 1 0 01igocottus 0 0 3 4 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 3 3 0 0 0 0 0 0 0 0 0 1 0 Clinocottus > W Z O •-< M > a a r > > W APPENDIX 2. LARVAL DATA MATRIX 0 0 0 0 0 0 0 0 0 0 Ancestor 0 0 1 0 0 0 O 1 0 1 Icelinus 0 0 1 0 0 0 0 1 0 1 Meanyi 0 0 1 0 0 0 0 1 0 0 Creaser1 1 2 0 0 0 0 2 0 1 0 A.harr1ngtoni 1 2 0 0 0 2 1 0 1 0 A.lateral 1s 1 2 0 0 0 2 1 0 1 0 A.fenestra 1 1 s 1 2 0 0 0 2 1 0 1 0 A.type 3 1 1 0 2 0 0 1 0 1 0 C.globiceps 1 1 0 2 o 0 0 0 1 0 C.embryum 1 1 0 2 0 0 0 0 1 0 C.anali s 1 1 0 2 0 0 0 0 1 0 C.recalvus 1 1 0 1 0 0 2 0 1 0 C.acut iceps 1 1 0 0 1 1 1 0 1 0 0.snyder i ! o 0 1 1 1 o ! 0 0.maculosus 1 1 7 HE STUDENT WANTED TO LEAVE A BLANK PAGE BETWEEN THE END OF THE TEXT AND THE BEGINNING F THE APPENDIX MATERIAL, 118 Figure 1. Artedius c o r a l l i n u s after Bolin (1944). 119 1 20 Figure 2. Artedius f e n e s t r a l i s from Bolin (1944). Fig. 2 122 Figure 3. Artedius harringtoni after Bolin (1944). 123 Fig. 3 1 24 Figure 4. Artedius l a t e r a l i s from Bolin (1944). 125 Fig. 4 1 26 Figure 5. Artedius notospilotus from Bolin (1944). 1 28 Figure 6. Ruscarius creaseri from Bolin (1944). 129 Fig. 6 130 Figure 7. Ruscar ius meanyi from Jordan and Evermann (1895). 131 Fig. 7 1 32 Figure 8. Orthonopias t r i a c i s from Bolin (1944). Fig. 8 134 Figure 9. Side view of scale ridge (schematic). Figure 9a, state 0. Figure 9b, state 1. Figure 9c, state 2. Figure 9d, states 3 and 4. Fig. 9c Fig. 9d 135a Ctenii r Fig. 9b 1 36 Figure 10. Chin pigmentation patterns.. Figure 10a, Artedius f e n e s t r a l i s (CAS 40354, 55.1 mm). Figure 10b, Artedius notospilotus (SIO 63-1054, 56.0 mm). Figure 10c, Artedius harringtoni (CAS 29521, 62.9 mm). Figure I0d, Artedius l a t e r a l i s (UMMZ 42958, 62.9 mm). Figure I0e, Orthonopias t r i a c i s (SU 18132, 54.8 mm). Areas outlined in f i g s . I0a-I0d indicate l i g h t background pigmentation. Areas outlined in f i g . 1Oe indicate white pigment. Fig. 10a Fig. 10b 137b Fig. 10c 1 3 7 c Fig. 10d 137d Fig. 10e 1 38 Figure 11. Top view of scale ridge (schematic). Figure 11a, Hemilepidotus hemilepidotus. Figure 11b, Chitonotus pugetensis. Figure 11c, Ruscarius c r e a s e r i . Figure 11d, Ruscarius meanyi. Figure 11e, Artedius  f e n e s t r a l i s . Figure 11f, Artedius notospilotus. Figure 11g, Artedius harringtoni. Figure 11h, Artedius l a t e r a l i s . 139a Fig. 11d 139b Fig. 11f 1 40 Figure 12. Pterotic flange. Figure 12a, Chitonotus pugetensis (UW uncat., 46.1 mm). Figure 12b, Artedius  f e n e s t r a l i s (BC57-210, 55.5 mm). Figure 12c, Artedius  c o r a l l i n u s (CAS 48970, 62.3 mm). Abbreviations: PT -pt e r o t i c , PA - p a r i e t a l , EP - e p i o t i c , SO -supraoccipital, EX - e x o c c i p i t a l , TB - tabulars, SP -sphenotic, EC - ethmoid c a r t i l a g e , VO - vomer, NA -nasal, ME - mesethmoid, LE - l a t e r a l ethmoid, FR -f r o n t a l . Fig. 12a 141a Fig. 12b Fig. 12c 1 42 Figure 13. Adult Wagner tree 1. 1 44 Figure 14. Adult Wagner tree 2. ''<3 •''6 *4 / . \ V % "a. o 146 Figure 15. Adult Wagner tree 3. 147 Figure 16. Cladogram of Artedius sensu s t r i c t u . Characters: 3 - form of scale ridge. 4 - body color pattern. 10 - spinelets on main preopercular spine. 12 - chin coloration. 13 - mandibular pore pattern. 14 - pores on l a t e r a l l i n e scales. 15 - form of head scales. 16 - c i r r i on transverse head tubercles. 17 - c i r r i on suborbital stay. 18 - form of preopercular spines. 19 - nasal pores. 20 - form of teeth. 21 -branchiostegal number. 22 - number of scale rows above l a t e r a l l i n e . 23 - anal f i n membrane. 24 - anal f i n pigmentation. 25 - penis. 26 - c i r r i on upper l i p . 27 - p o s t c l e i t h r a . 28 - scales under anterior o r b i t . 29 - scales behind a x i l l a . 30 - scales on caudal peduncle. 31 - p r e o r b i t a l c i r r i . 32 - c i r r i above a x i l l a . 34 - c i r r i on preopercular margin. 35 -scale ridge shape. 39 - form of head tubercles. 40 -serrations on posttemporal/supracleithrum. 41 - size of c i r c l e s on body. 45 - pterotic flange. 149 1 50 Figure 17. Cladogram of Ruscarius. Characters: 1 -position of anus. 3 - form of scale ridge. 4 - body color pattern. 5 - scales above a x i l l a . 6 - scales on snout. 7 - shape of upper preopercular spine. 8 -number of pelvic rays. 9 - c i r r i on opercle. 28 -scales under anterior o r b i t . 31 - p r e o r b i t a l c i r r i . 32 - c i r r i above a x i l l a . 33 - c i r r i anterad to upper preopercular spine. 35 - scale ridge shape. 37 -adult s i z e . 38 - anal ray form (males). 44 - c i r r i on nasal spine. 47 - scales on eye. 1 52 Figure 18. Adams consensus tree for adult trees. 153 1 54 Figure 19. Larval Wagner tree 1. 155 1 56 F i g u r e 20 . L a r v a l Wagner t r e e 2. 157 1 58 Figure 21. Larval. Wagner tree 3. 159 160 F i g u r e 22. Adams consensus t r e e f o r l a r v a l t r e e s . 161 Figure 23. Phylogenetic tree of Bolin (1947) 163 1 64 Figure 24. Wagner tree calculated in the absence of Ruscarius meanyi. 1 66 Figure 25. Dichotomous l a r v a l tree of Washington (1982). 1 6 7 168 Figure 26. Polytomous l a r v a l tree from Washington (1982). 169 1 70 Figure 27. Adams consensus tree of adult and l a r v a l Adams trees. 1 7 1 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0095983/manifest

Comment

Related Items