Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Marine benthic algal communities in the Flat Top Islands area of Georgia Strait Lindstrom, Sandra Christine 1973

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

Item Metadata

Download

Media
831-UBC_1973_A6_7 L55_5.pdf [ 5.84MB ]
Metadata
JSON: 831-1.0101315.json
JSON-LD: 831-1.0101315-ld.json
RDF/XML (Pretty): 831-1.0101315-rdf.xml
RDF/JSON: 831-1.0101315-rdf.json
Turtle: 831-1.0101315-turtle.txt
N-Triples: 831-1.0101315-rdf-ntriples.txt
Original Record: 831-1.0101315-source.json
Full Text
831-1.0101315-fulltext.txt
Citation
831-1.0101315.ris

Full Text

MARINE BENTHIC ALGAL COMMUNITIES IN THE FLAT TOP ISLANDS AREA OF GEORGIA STRAIT by Sandra C h r i s t i n e Lindstrom B. A. Reed Coll e g e , 1971 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n the Department of Botany We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1973 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood t h a t c o p y i n g or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department o f The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada Date ABSTRACT Data on 75 species from 124 quadrats c o l l e c t e d from the i n t e r t i d a l and s u b t i d a l regions i n the F l a t Top Islands area of Georgia S t r a i t i n l a t e spring to early autumn, 1972, were subjected to a number of com-munity analysis techniques including a Williams and Lambert a s s o c i a t i o n a n a l y s i s , a species c o n s t e l l a t i o n diagram, a c l u s t e r a n a l y s i s of quadrats, an inverse analysis of sp e c i e s , a Zurich-Montpellier a n a l y s i s , and a Bray and Curtis o r d i n a t i o n . A s s o c i a t i o n analysis and c l u s t e r analysis c l a s s i f i c a t i o n of quadrats delimited four communities: an i n t e r t i d a l , an upper s u b t i d a l , a lower s u b t i d a l , and a sandy bottom community. Species c o n s t e l l a t i o n diagram and inverse analysis c l a s s i -f i c a t i o n of species delimited six communities: an i n t e r t i d a l , a shallow red a l g a l , a laminarian, a t u r f , a deep water, and a sandy-bottom community. The Zurich-Montpellier analysis and the Bray and Cu r t i s ordination did not produce c l e a r l y definable groups of quadrats. The r e s u l t s of a l l the analyses i n d i c a t e that marine benthic algae form broadly overlapping d i s t r i b u t i o n s from which communities of varying a f f i n i t i e s can be extracted. i i i TABLE OF CONTENTS Introduction 1 Methods The study area 8 C o l l e c t i o n methods 8 A n a l y t i c a l methods 13 Associ a t i o n a n a l y s i s .15 Species c o n s t e l l a t i o n diagram 18 Cluster and inverse a n a l y s i s . . 19 Zurich-Montpellier analysis 22 Bray and Curti s o r d i n a t i o n 23 Results Asso c i a t i o n analysis 26 Species c o n s t e l l a t i o n diagram... 40 Cluster and inverse analysis 49 Zurich-Montpellier analysis 73 Bray and C u r t i s o r d i n a t i o n 76 Discussion C o n t r o l l i n g environmental factors 82 Comparison and evaluation of a n a l y t i c a l methods 86 Comparison x^ith other s u b t i d a l communities 92 Bibliography 95 Appendices 100 i v LIST OF TABLES Page Table 1. Species used i n the computer a n a l y s i s . 25 Table 2. Standardized species-group/quadrat-group coincidence table f o r weighted c l u s t e r and inverse analyses r e s u l t s . 63 Table 3. Standardized species-group/quadrat-group coincidence table f or unweighted c l u s t e r and inverse analyses r e s u l t s . 63 Table 4. Constant and c h a r a c t e r i s t i c species of the s i x benthic a l g a l communities defined by selected quadrat groups for the weighted c l u s t e r and inverse analyses r e s u l t s . 68 Table 5. Constant and c h a r a c t e r i s t i c species of the s i x benthic a l g a l communities defined by selected quadrat groups for the unweighted c l u s t e r and inverse analyses r e s u l t s . 70 Table 6. Species-quadrat table of the Zurich-Montpellier analysis using the 50-20 diagnostic species formula. 74 V LIST OF FIGURES Page Figure 1. Ass o c i a t i o n analysis of 75 species and 124 quadrats at a p r o b a b i l i t y l e v e l of .001. 28 Figure 2. Ass o c i a t i o n analysis of 75 species and 124 quadrats at a p r o b a b i l i t y l e v e l of .05. 30 Figure 3. The four major groupings of the ass o c i a t i o n analysis quadrats at BI01, BI02, and SI02. 32 Figure 4. Arrangement of the important subdivisions of the p = .05 a s s o c i a t i o n analysis according to the depths of the quadrats contained i n the sub d i v i s i o n s . 37 Figure 5. The as s o c i a t i o n matrix showing p o s i t i v e and negative associations between a l l p a i r s of species. 42 Figure 6. The species c o n s t e l l a t i o n diagram showing highly p o s i t i v e associations (p<.001). 44 Figure 7. The species c o n s t e l l a t i o n diagram showing the p o s i t i v e associations not shown i n Figure 6 (.001< p <.01). 46 Figure 8. Dendrogram of the c l u s t e r analysis using the weighted pair-group average method. 51 Figure 9. Dendrogram of the c l u s t e r analysis using the unweighted pair-group average method. 53 Figure 10. Dendrogram of the inverse analysis using the weighted pair-group average method. 59 Figure 11. Dendrogram of the inverse analysis using the unweighted pair-group average method. 59 Figure 12. Quadrats common to both the weighted and unweighted analysis c o i n c i d e n t a l quadrat/species groups at BI01, BI02, and SI02. 66 Figure 13. The Bray and C u r t i s . o r d i n a t i o n of a l l 124 quadrats. 78 Figure 14. The Bray and Cu r t i s ordination of a selected group of 83 quadrats. 81 v i ACKNOWLEDGEMENTS The author wishes to thank the following people: J u l i e Celestino of the U.B.C. Phycology Herbarium f o r i d e n t i f i c a t i o n of a l g a l specimens, Steve Borden of the U.B.C. Biology Computing Centre for w r i t i n g the computer programs, and Brent P a t r i q u i n , Dave Turpin, and Ke i t h Hutchinson for c o l l e c t i o n of the quadrat data. Without t h e i r a s s i s t a n c e , t h i s thesis could not have been w r i t t e n . Dr. R. F. Scagel also aided i n the species i d e n t i f i c a t i o n . INTRODUCTION The study of benthic marine ecology began i n the 1830's with Milne Edwards i n France and Edward Forbes i n England. E a r l i e r studies were not w e l l p u b l i c i z e d and were generally overlooked. Both men proposed widely-recognized schemes of shore and shallow water zonations of organisms. Forbes' four-part zonation consisted of an i n t e r t i d a l l i t t o r a l zone, a shallow s u b t i d a l laminarian zone, a deeper c o r a l l i n e zone, and a deep-sea c o r a l zone at the lower l i m i t s of sea-inhabiting l i f e .1 One of the most d e t a i l e d early works on marine ecology was published i n German by J . R. Lorenz i n 1863. The author presented graphs of seasonal changes i n temperature at d i f f e r e n t depths to 30 fathoms, tables of the t i d a l l e v e l s and currents, and figures of the d i s t r i b u t i o n of plants and animals i n the Gulf of Quarnero, Yugoslavia. Lorenz recognized various communities by the presence of c h a r a c t e r i s t i c species. He also introduced the terms supra- and s u b l i t t o r a l , even now i n general use. Unfortunately, h i s early death precluded the wide-scale d i s t r i b u t i o n of t h i s work, and Lorenz was generally ignored by l a t e r e c o l o g i s t s . Although A. E. V e r r i l l , an American, grasped the concept of community i n h i s 1873 report on the invertebrates and the p h y s i c a l c h a r a c t e r i s t i c s of the Vineyard Sound area, K. Moebius f i r s t c l e a r l y defined the term and gave i t the name "biocoenosis" i n h i s monumental work on the oyster and oyster c u l t u r e published i n 1877. . ^The early h i s t o r y of benthic marine ecology has been summarized by Hedgpeth (1957), whose discussion i s followed here from Forbes through G i s l e n . References are found i n Hedgpeth. 2 In an e f f o r t to q u a n t i t a t i v e l y estimate the secondary p r o d u c t i v i t y of the sea i n the 1910's, C. G. J . Petersen of the Danish B i o l o g i c a l S t a t i o n designated bottom communities on the basis of the most dominant and c h a r a c t e r i s t i c (constant) organisms for a p a r t i c u l a r type of bottom. Petersen was the f i r s t marine e c o l o g i s t to obtain q u a n t i t a t i v e samples and (ignoring Lorenz) to apply the concept of dominant organism(s) to naming communities. His work was influenced by the Danish plant e c o l o g i s t Warming, and much of the subsequent h i s t o r y of q u a n t i t a t i v e marine ecology follows c l o s e l y the paths blazed by t e r r e s t r i a l plant e c o l o g i s t s . Most early studies of marine ecology and a continuing number today are l i t t l e more than species l i s t s with some marine natural h i s t o r y . An exception to t h i s g e n e r a l i z a t i o n i s the work by Gislen i n Gullmar F j o r d , Sweden, i n the 1920's. By using d i v i n g equipment and counting frames, Gislen catalogued the various hard-bottom associations and computed t h e i r production. G i s l e n also gave d e t a i l e d a t t e n t i o n to hydrographic and other e c o l o g i c a l f a c t o r s . A deluge of papers on i n t e r t i d a l zonation began around 1930. The most important contributors to t h i s l i t e r a t u r e were T. A. and Anne Stephenson, who summarized t h e i r observations by d e l i m i t i n g three i n t e r t i d a l zones of almost u n i v e r s a l occurrence. These zones were the l i t t o r i n e zone, or s u p r a l i t t o r a l f r i n g e ; the balanoid, or m i d l i t t o r a l , zone; and the sub- or i n f r a - l i t t o r a l f r i n g e (Stephenson & Stephenson, 1949). Although the Stephensons acknowledged that p h y s i c a l factors are probably the dominant c o n t r o l l i n g agents of d i s t r i b u t i o n s i n these zones, they saw no clear c o r r e l a t i o n with any single c r i t i c a l f a c t o r , and therefore preferred to define t h e i r 3 zones i n terms of the organisms which occurred there. The occurrence of subzones wi t h i n these zones i s not u n i v e r s a l and i s therefore only b r i e f l y mentioned. Their u n i v e r s a l zones seem to merely represent the i n t e r t i d a l zone and the ecotones between i t and the land at the upper extreme and between i t and the sea at the lower extreme. However, the nearly u n i v e r s a l occurrence of l i t t o r i n e s n a i l s i n the s u p r a l i t t o r a l f r i n g e and barnacles and limpets i n the m i d l i t t o r a l zone foreshadowed the observations by Thorson (1957) of " p a r a l l e l " marine animal communities in h a b i t i n g s i m i l a r types of bottoms i n widely d i f f e r e n t areas. Recent work on i n t e r t i d a l zonation has followed several l i n e s — one seeking p h y s i c a l causes, one b i o l o g i c a l causes, and one continuing the d e s c r i p t i v e work with the a i d of more sophisticated mathematical analyses. Connell (1972) has recently reviewed patterns of species d i s t r i b u t i o n s and population i n t e r a c t i o n s on rocky i n t e r t i d a l shores. Paine and Connell have shown that the i n t e r t i d a l d i s t r i b u t i o n and abundance of c e r t a i n animal species are subjected to control by predation and i n t e r s p e c i f i c competition. For example, Paine (1972) has demonstrated that the preying of a s t a r f i s h on the mussels of the open coast keeps the mussels from dominating the i n t e r t i d a l zone to the exclusion of other sedentary species. Connell (1961a and b) has shown that i n t e r s p e c i f i c competition between two species of barnalces can be modified by the s e l e c t i v e predation of a s n a i l . In 1946, Doty suggested that tide l e v e l s are the major deter-minants of i n t e r t i d a l plant d i s t r i b u t i o n s because there can be as much as a three-fold increase i n the maximum time of submergence or emer-gence of organisms found at a p a r t i c u l a r t i d a l l e v e l . A number of obvious c r i t i c i s m s have been leveled at t h i s proposal: the occurrence 4 at d i f f e r e n t l e v e l s of the same organisms on north and south-facing shores, i n d i r e c t sunlight and i n the shade of a c r e v i c e , on exposed and semi-sheltered shores, or i n the presence and absence of a grazer. Nevertheless, authors continued to provide evidence for and against the hypothesis of " c r i t i c a l t i d e l e v e l s . " Widdowson (1965) examined the upper and lower l i m i t s of i n t e r t i d a l organisms at 34 stat i o n s i n an area t r a n s i t i o n a l between oceanic and estuarine s a l i n i t i e s and between semidiurnal mixed tides and d i u r n a l t i d e s . He found only 25% of the species upper l i m i t s coincided with a c r i t i c a l t i d e f a c t o r , and interpreted h i s r e s u l t s as c o n t r a d i c t i n g Doty's hypothesis. Saito and Atobe (1970) supported the idea of c r i t i c a l t i d e l e v e l s from the i n t e r -t i d a l d i s t r i b u t i o n s of three groups of benthic algae. Yet, the c r i t i c a l t i d e l e v e l hypothesis remains an unproved explanation of i n t e r t i d a l zonation, those workers seeking to deny i t exacting a s t r i c t i n t e r p r e -t a t i o n of t h e i r data; those seeking to sustain i t apparently doing so by the subjective i n t e r p r e t a t i o n of t h e i r observations. However, one cannot deny that t i d e i s a c o n t r o l l i n g factor of i n t e r t i d a l zonation. Although the q u a n t i t a t i v e d e s c r i p t i o n of i n t e r t i d a l seaweed communities may be considered an end i n i t s own r i g h t , i n t e r t i d a l p hytosociologists have tended to use t h e i r r e s u l t s to support one or another viewpoint of i n t e r t i d a l zonation. Working on the B r i t i s h coast, R u s s e l l (1972) subjected h i s q u a n t i t a t i v e samples to an asso-c i a t i o n analysis and a determination of s i m i l a r i t y c o e f f i c i e n t s . He found a two-zone shore—one group of species representing the supra-l i t t o r a l f r i n g e and the other, the e u l i t t o r a l zone, thereby supporting the Stephensons' and Lewis' (1964) concept of a three-zoned rocky shore (the s u b l i t t o r a l . f r i n g e species being absent because of the 5 proximity of estuarine sediments at the lower l i m i t s of the e u l i t t o r a l zone.) Saito and Atobe (1970) also found three groups of i n t e r t i d a l plants based on s i m i l a r i t y measurements of q u a n t i t a t i v e data, but they preferred to view these groups as supporting Doty's c r i t i c a l t i d e l e v e l hypothesis. Most of the recent q u a n t i t a t i v e work i n s u b t i d a l marine ecology has been with animals occurring i n sediments. There are several reasons for t h i s preference—the ease of sampling (which may be done i n d i r e c t l y by dredging or grab samples or d i r e c t l y using SCUBA d i v i n g techniques and a suction apparatus), the neatness of the numerical data (numbers of i n d i v i d u a l s or biomass), and the e x c e l l e n t pioneer work of Petersen and others i n the f i e l d . A p p l i c a t i o n of numerical methods to c l a s s i f i c a t i o n and/or o r d i -nation of benthic marine infa u n a l communities has now been c a r r i e d out i n both the i n t e r t i d a l and the s u b t i d a l regions. Studies i n the i n t e r t i d a l zone have occurred i n Old Tampa Bay, F l o r i d a (Bloom, Simon & Hunter, 1972) and near Auckland, New Zealand (Cassie & Michael, 1968.) Subtidal studies have been done i n Moreton Bay, A u s t r a l i a (Stephenson, Williams & Lance, 1970), Kingston Harbour, Jamaica (Wade, 1972), Moresby Island near the southern tip. of Vancouver Island (Popham & E l l i s , 1971), Puget Sound and the outer Washington coast ( Lie & K e l l e y , 1970), Port Madison, Washington (N i c h o l s , 1970), the continental shelf of North Carolin a (Day, F i e l d & Montgomery, 1971), False Bay, South A f r i c a ( F i e l d , 1971), and i n several bays along the eastern seaboard of Canada (Hughes & Thomas, 1971a and b; Hughes, Peer & Mann, 1972). Stephenson, Williams and Cook (1972) have compared Petersen's o r i g i n a l c l a s s i f i c a t i o n of soft-bottom communities to one derived by modern numerical methods of c l a s s i f i c a t i o n . Subtidal a l g a l communities are best developed on rocky substrates. Therefore, they cannot be sampled q u a n t i t a t i v e l y by dredging or grabbing but require d i v i n g and scraping. Furthermore, benthic algae may not occur as e a s i l y d i s t i n g u i s h a b l e i n d i v i d u a l s , and synthetic q u a n t i t a t i v e data, such as cover values, may need to be employed for the a n a l y s i s . Neushul (1967) did some pioneer work on s u b t i d a l plant ecology near Friday Harbor, Washington, by noting the number and i d e n t i t y of plants crossed by a transect l i n e running from the i n t e r t i d a l to the lower l i m i t s of plant occurrences. From h i s data, Neushul derived a c l u s t e r analysis using the unweighted pair-group method with a r i t h -metic average, a method derived from numerical taxonomy (Sokal & Sneath, 1963.) Because none of the species showed a high l e v e l of a s s o c i a t i o n and no obvious groups appeared, Neushul concluded that h i s r e s u l t s supported the continuum hypothesis of vegetation d i s t r i b u t i o n better than h i s previously proposed three-zone system of s u b t i d a l vegetation (Neushul, 1965). However, his r e s u l t s a r e generally incon-c l u s i v e , probably due to the method by which h i s data were obtained. Using both qu a n t i t a t i v e and q u a l i t a t i v e data, Boudouresque and coworkers have applied the methods of the Zurich-Montpellier school of phytosociology to the study of s u b t i d a l a l g a l communities i n the Mediterranean. Blanc and Boudouresque (1970) have also performed p r i n c i p a l components analysis on quadrat samples from the c o r a l l i n e zone i n the lower l i t t o r a l . However, most of Boudouresque's work has been on communities of shade-loving species. Boudouresque (1971) 7 has recently summarized his methods for studying marine benthic plant communities. The present study i s the f i r s t known attempt to apply a v a r i e t y of. numerical methods to systematic samples of benthic marine vege-t a t i o n which were neither s u b j e c t i v e l y chosen (Boudouresque) nor only q u a l i t a t i v e l y sampled (Neushul). The objectives of t h i s study are (1) to describe the benthic marine vegetation i n the study area and provide baseline data on community composition and structure for future studies r e q u i r i n g such information, (2) to apply and compare several mathematical analyses i n order to determine which method provides the best d e s c r i p t i o n of the communities or continua for the amount of time and e f f o r t expended, and (3) to suggest the major c o n t r o l l i n g f a ctor(s) of community composition i n the study area. The samples analyzed i n t h i s thesis are part of a long-term study on s p a t i a l and temporal v a r i a t i o n s i n community structure and function being conducted by Dr. R. E. Foreman and supported by grants from the National Research Countil of Canada (Operating Grant A6241) and the U n i v e r s i t y of B r i t i s h Columbia. METHODS The study area The F l a t Top Islands are located along the southeastern t i p of Gabriola I s l a n d , which i s the northernmost of a s e r i e s of islands known as the San Juan Islands i n northern Washington and the Gulf Islands i n southern B r i t i s h Columbia. To the east of the F l a t Top Islands i s Georgia S t r a i t , extending approximately 20 miles to the mainland of B r i t i s h Columbia. Bath Island i s the most ea s t e r l y i s l a n d of the F l a t Top Islands group and i s eight acres i n s i z e . Sear Island i s 26 acres and i s located between Gabriola Island on the west, Tugboat Island on the northeast (to which i t i s joined during extreme low water,) and Breakwater Island to the south. A l l of the F l a t Top Islands were formed i n the upper Cretaceous or T e r t i a r y as part of the Gabriola formation. They are composed c h i e f l y of sand-stone, with shale and conglomerate as minor constituents (Muller, 1971.) C o l l e c t i o n methods (aft e r Foreman, manuscript i n preparation) Samples were c o l l e c t e d along transect l i n e s running perpendicular to the shore o f f Bath and Sear Islands."*" Each transect l i n e was anchored on shore near mean ti d e l e v e l and weighted with lead at 25 m i n t e r v a l s to hold i t i n p l a c e . Ropes with f l o a t s were secured to the transect l i n e at 50 m and 100 m so the l i n e could be found e a s i l y from the s u r f a c e . The transect l i n e s were marked at meter i n t e r v a l s , with a coded marking at each 5 and 10 m i n t e r v a l to i n d i c a t e the distance from shore. ''"The l o c a t i o n of the transect l i n e s and the dates sampled appear i n Appendix 1. 9 The transect l i n e s were sampled at 5 m i n t e r v a l s (with the excep-tions of BI02 and SI02 x = 025, which were both sampled at 10 m i n t e r v a l s . ) A 0.5 x 0.5 m frame was placed along the left-hand side of the l i n e such that the 0.25 m mark on the frame coincided with the meter mark on the transect l i n e . A f t e r p l a c i n g the quadrat, the diver transmitted to the surface a p h y s i c a l d e s c r i p t i o n of the sample area v i a a SUBC0M underwater wireless communications unit model 142. The diver then proceeded to i d e n t i f y the species of algae found within the quadrat and describe t h e i r general d i s p o s i t i o n and abundance. The • i d e n t i t y of obvious invertebrates was also noted. A l l information was coded d i r e c t l y on IBM computer sheets by the surface operator. F r e -quently, l a r g e r algae such as Laminaria or Agarum would have to be c o l l e c t e d before the r e s t of the quadrat could be "enumerated." Only those portions of the kelp d i r e c t l y over or attached within the quadrat were cut out and placed i n c o l l e c t i n g bags. Occasionally, large animals found wi t h i n the quadrat would also be c o l l e c t e d . Following complete enumeration of the quadrat, a second diver proceeded to " c o l l e c t the quadrat" using a putty k n i f e to free the algae from the substrate and an underwater a i r l i f t to c o l l e c t the algae. The organisms sucked up by the a i r l i f t were c o l l e c t e d i n a 2 bag with a mesh s i z e of approximately 2.5 mm . The area enclosed by the quadrat frame was scraped, leaving only a few filamentous i n d i v i d u a l s and some encrusting forms. In quadrats where the rock was fragmented, the l a r g e r rocks were scraped underwater and the smaller rocks were put into c o l l e c t i n g bags to be scraped on the surface. A f t e r the quadrat was c o l l e c t e d , the mesh bag on the a i r l i f t was removed and placed i n a c o l l e c t i n g bag. Because of the force of 10 the a i r l i f t , f r e e - f l o a t i n g algae were sometimes also sucked i n t o the mesh bag. I d e n t i f i c a t i o n of these "contaminants" l a t e r presented some d i f f i c u l t y . In general, three s p a t i a l l y consecutive quadrats were c o l l e c t e d per d i v e . Several quadrat c o l l e c t i o n s were made s p e c i f i c a l l y to determine the optimal quadrat s i z e based on the species-area curve method of a 10% increase i n the number of species f o r a doubling of quadrat s i z e . A l l c o l l e c t i o n s were made i n the shallow s u b l i t t o r a l region and i n d i -2 cated an optimum quadrat s i z e of 0.125 to 0.25 m . This s i z e agrees with that found by other inv e s t i g a t o r s i n the area ( F r a l i c k , 1971.) No c o l l e c t i o n s were made to determine the number of quadrats needed to adequately sample ;a community since i t was assumed that with the systematic sampling method being used a more than adequate number of quadrats would be c o l l e c t e d . For most sampling, the 0.5 x 0.5 m quadrat was used, but occ a s i o n a l l y a 0.25 x 0.25 m quadrat was used to sampled areas of the i n t e r t i d a l region which appeared p a r t i c u l a r l y homogeneous. Further work on s u b t i d a l seaweed communities requires a more adequate determination of optimum quadrat s i z e needed to sample plant communities from the i n t e r t i d a l to the l i m i t s of t h e i r d i s t r i b u t i o n s . Depth was measured during slack water using a c a l i b r a t e d l i n e with a f l o a t on the end. A di v e r swimming along the transect l i n e released the f l o a t to the surface, p u l l e d i n the slack on the l i n e u n t i l he could f e e l the tension of p u l l i n g the f l o a t under, and then read the depth from the l i n e . Depths were measured every f i v e meters along the transect l i n e s BI01 x = 050 and x = 085j SI02 x = 000 and x = 025, and BI02. Depths from mean tide l e v e l were cal c u l a t e d according to the method i n 11 the Canadian Tide and Current Tables (1972) f o r tides at S i l v a Bay. Depths at stations BI01 x = 000 and SI02 x = 050 were estimated from depths along the other transects at those s i t e s . Each quadrat c o l l e c t i o n was sorted as soon as p o s s i b l e . Bags containing quadrat c o l l e c t i o n s which would probably not be sorted that day were suspended from a dock i n n a t u r a l l y flowing seawater. For s o r t i n g , the organisms from the bags were emptied into p l a s t i c buckets. Several handfuls at a time were put into trays containing seawater. A l l smaller animals were put into a j a r of unneutralized 4% formalin i n seawater. In cases where a s i n g l e large animal or a large number of moderately-sized animals occurred i n a quadrat, the organisms were counted, weighed, and returned to the water as soon as p o s s i b l e . Usually several of the smaller i n d i v i d u a l s would be preserved. The formalin i n the animal j a r s was l a t e r n e u t r a l i z e d with borate, and the j a r s were stored f or future i d e n t i f i c a t i o n and enumeration of the organisms. The algae were sorted by species as completely and as accurately as p o s s i b l e . A f t e r s o r t i n g , each species was damp-dried and weighed. In most cases, a subsample of the alga was taken for l a t e r determination of species i d e n t i t y and reproductive condition at U.B.C. When the algae were small or where there was only a s i n g l e plant or the i d e n t i f i c a t i o n was i n doubt, the e n t i r e sample was preserved. The subsamples were put into i n d i v i d u a l p l a s t i c bags with tags i n d i c a t i n g the c o l l e c t i o n number, the d i s h number and the tent a t i v e species i d e n t i f i c a t i o n and preserved i n 4% formalin i n seawater. For l a t e r reference, the specimens were preserved on herbarium sheets with the c o l l e c t i o n number, dish number, species name and reproductive 12 condition i f f e r t i l e . A l l c o l l e c t i o n s are f i l e d with the U.B.C. Phycology Herbarium. The dishes were stored i n a drying c l o s e t (open near the top and bottom to allow flow of a i r through the closet) before being packed i n cardboard boxes and taken to U.B.C. There they were dried to a constant weight at 105° C i n f o r c e d - a i r drying ovens. A f t e r being reweighed, the samples were stored again i n cardboard boxes u n t i l they could be ashed at 430° C for four hours i n a muffle furnace and reweighed again to determine ash content. A l l weights f o r each sample were recorded on IBM cards. The residuum of rock, s h e l l , d e b r i s , fragments of algae and a few i n d i v i d u a l s of the smaller species remaining a f t e r s o r t i n g was also damp d r i e d , weighed, and ashed as miscellaneous to estimate the biomass of the unsorted algae. When the remainder contained only s h e l l and rock i t was discarded. Data coded f o r keypunching thus consisted of three kinds of cards: one card f o r each quadrat containing the p h y s i c a l data, one card containing the d e s c r i p t i v e data for each species i n that quadrat, and one card containing the weight data for that species. A computer pro-gram converted the weight data i n t o a l g a l biomass (wet, dry, and ash-2 free dry weights/m ) , and these values were punched onto the d e s c r i p t i v e data cards for each species. Missing dry weights and ash-free dry weights from samples which were completely preserved were estimated from a regression analysis based on those samples of the same species for which complete weight data were a v a i l a b l e . Because of the imprecision of the balance used to obtain the wet weights, the dry and ash-free dry weights for very small specimens were inaccurate. These samples were given minimum weights (0.01 g wet, 0.005 g dry, and 0.001 g ash-free dry 13 weight) or provided with weight extrapolated from the regression a n a l y s i s . A n a l y t i c a l methods Not a l l of the species c o l l e c t e d or i d e n t i f i e d were used i n the computer analysis of the data. I n i t i a l l y , a l l cards which were not species (such as MLSC, TIDBITS, etc.) were removed. Weights from these cards were frequently used to estimate the biomass of the species which occurred i n that sample on a percentage b a s i s . Other samples, which were discovered to be mixtures of species a f t e r the i n i t i a l s o r t i n g and weighing, had th e i r weights adjusted to the proper percentages. The i d e n t i t y of many of the algae had to be v e r i f i e d . Bossiella and Gelidium proved to be p a r t i c u l a r l y r e f r a c t i v e to species deter-minations, and since a l l specimens appeared to be of the same species, a l l cards were given only the generic name. Because of the method of c o l l e c t i o n , many of the quadrat samples contained algae which were l y i n g loose on the bottom and which did not normally grow there. These algae were considered contaminants and an e f f o r t was made to remove these cards on the basis of the quadrats i n which they had not been reported underwater and personal observations. The most common contaminants were Enteromorpha intestinalis, Fucus distichus, Gigartina papillata, Microcladia borealis, and Porphyra perforata—all i n t e r t i d a l species probably detached by the force of waves i n this zone. Some s u b t i d a l species were probably also contaminants, but these were not so e a s i l y detected because of the greater uncertainty of t h e i r usual d i s t r i b u t i o n s . The computer analyses used i n t h i s study also pointed to discrepancies i n species occurrences, and obvious mistakes 14 were corrected before further analyses were performed. In general, i removal of contaminating species was probably s l i g h t l y conservative. More quadrat c o l l e c t i o n s w i l l point out what a d d i t i o n a l species were contaminants. Also not considered were cards i n which only the genus was given and no preserved specimen was a v a i l a b l e to determine the species ( i n a l l cases these specimens had an extremely small biomass), j u v e n i l e specimens for which i d e n t i f i c a t i o n was questionable, species which occurred i n three or fewer quadrats, and species which occurred i n more than three quadrats, but, because of their, small s i z e and the high p r o b a b i l i t y of having been overlooked i n some of the quadrats, were ignored pending more complete data on t h e i r d i s t r i b u t i o n s and abundances. Names o f , and notes on, these species are found i n Appendix 2. Five methods of analyzing the species data were used. These were (1) an a s s o c i a t i o n a n a l y s i s , (2) a species c o n s t e l l a t i o n diagram, (3) a c l u s t e r analysis of quadrats and an inverse analysis of species, (4) a Zurich-Montpellier a n a l y s i s , and (5) a Bray and Curti s o r d i n a t i o n . Detailed descriptions of these methods foll o w . 1. Asso c i a t i o n analysis Although D. W. Goodall (1953) pioneerred the use of i n t e r s p e c i f i c associations i n the c l a s s i f i c a t i o n of vegetation, h i s methods were s i g n i f i c a n t l y altered by Williams and Lambert (1959, 1960), the founders of a s s o c i a t i o n a n a l y s i s . A s s o c i a t i o n analysis i s a hi e r a r c h i c a l - d i v i s i v e - m o n o t h e t i c method, of c l a s s i f i c a t i o n ( P i e l o u , 1969a). Quadrats are subdivided into groups such that a l l associations between pai r s of species disappear and the 15 ultimate groups may be considered homogeneous i n composition. Since a large number of a l t e r n a t i v e subdivisions would f u l f i l l t h is r e q u i r e -ment, a su b d i v i s i o n i s needed such that the r e s u l t i n g two subgroups of each d i v i s i o n produce the smallest t o t a l number of r e s i d u a l asso-c i a t i o n s . As an approximation of fa c t o r a n a l y s i s , the most e f f i c i e n t a s s o c i a t i o n analysis would di v i d e the population on the s i n g l e species with the l a r g e s t sum of a selected a s s o c i a t i o n index. Of the several 2 as s o c i a t i o n indices a v a i l a b l e , uncorrected X was chosen for this study. Williams and Lambert (1959) have stated that i t i s the "theo-r e t i c a l l y p referable" index of a s s o c i a t i o n , and P i e l o u (1969b) has found that i t can give the same d i v i s i o n s of quadrats as homogeneity t e s t s . 2 C a l c u l a t i o n of the X a s s o c i a t i o n index requires only species presence l i s t s for each quadrat. With a s u f f i c i e n t number of samples the l e v e l of a s s o c i a t i o n between every p a i r of species can be deter-mined. The number of observed quadrat combinations of the two species can be compared with the expected number of quadrat combinations, assuming 2 the species are not associated, by a X test using a 2 x 2 contingency 2 table (see below.) The c a l c u l a t i o n of X follows the formula: (ad-bc) n with df = 1 (a+b)(c+d)(a+c)(b+d) where the values of these symbols are found i n the following table: Species A present present a absent b a+b Species B absent c d c+d a+c b+d n 16 2 These c a l c u l a t i o n s r e s u l t i n an n x n matrix of X values for a l l pai r s of species. 2 In a s s o c i a t i o n a n a l y s i s , only s i g n i f i c a n t X values are considered. Values that are not s i g n i f i c a n t or are indeterminate (such as the values along the major diagonal of the matrix) are considered to be zero. A l l 2 s i g n i f i c a n t X values for each species are summed, and the species with 2 the highest t o t a l X i s chosen as the species on which to begin the a n a l y s i s . The quadrats are then divided into two groups, c o n s i s t i n g of a l l quadrats containing that species and a l l quadrats la c k i n g that species. The e n t i r e procedure i s repeated separately on these two groups of quadrats, and each group i s further subdivided on the basis of the species 2 with the highest t o t a l X for that group. The d i v i s i o n of quadrats into 2 subgroups i s continued u n t i l the highest s i n g l e X for the species on which the d i v i s i o n i s being made i s no longer s i g n i f i c a n t , the person performing the analysis s e l e c t i n g the l e v e l of s i g n i f i c a n c e d e s i r e d . The end products of a s s o c i a t i o n analysis are groups of quadrats characterized by the presence and absence of p a r t i c u l a r s p e c i e s . The r e s u l t s are usually presented i n the form of a dendrogram showing the successive d i v i s i o n of the groups of quadrats on the basis of p a r t i c u l a r species at d i f f e r e n t l e v e l s of s i g n i f i c a n c e . The f i n a l groupings of quadrats may be suggestive of c o n t r o l l i n g environmental f a c t o r s . Besides d e l i m i t i n g groups of quadrats of s i m i l a r species composition, a s s o c i a t i o n analysis produces " c h a r a c t e r i s t i c " species which define the groups of quadrats. Since each d i v i s i o n of quadrats i s based on the presence or absence of a s i n g l e s pecies, the f i n a l subgroups are charac-t e r i z e d by the presence and absence of the p a r t i c u l a r species on which that portion of the dendrogram was d i v i d e d . This r e s u l t of as s o c i a t i o n 17 analysis i s of value i n i t s own r i g h t beyond the grouping together of p a r t i c u l a r quadrats. Association analysis tends to pick out charac-t e r i s t i c species of high f i d e l i t y and constancy as w e l l . These species may serve as i n d i c a t o r s of the p a r t i c u l a r environmental factors respon-s i b l e for the existence of the communities or may even be used as measured of community p r o d u c t i v i t y i f t h e i r c o r r e l a t i o n s with the abun-dance of other species are p a r t i c u l a r l y strong. The groups of charac-t e r i s t i c species r e s u l t i n g from a s s o c i a t i o n analysis may d i f f e r from the choice of c h a r a c t e r i s t i c species by other workers using more subjective methods because the species need not be the most dominant or abundant but only those which are c o n s i s t e n t l y present i n or absent from a p a r t i c u l a r group of quadrats. There are a number of assumptions i n a s s o c i a t i o n analysis which make thi s method an approximation at best. However, i n comparing i t to the s u b d i v i s i o n of quadrats on the basis of homogeneity, t e s t s , P i e l o u (196%) obtained the same r e s u l t s as did Williams and Lambert (1959) using an 2 uncorrected X to measure associations between species p a i r s . P i e l o u admits that the defects i n the mathematical arguments are s l i g h t and are not l i k e l y to be overly misleading; the r e s u l t s often according highly with the i n t u i t i o n of the experienced observer. 2. Species c o n s t e l l a t i o n diagram The r e l a t i o n s h i p s among species can be represented i n the form of a two-dimensional s p a t i a l arrangement of species c a l l e d a species c o n s t e l -l a t i o n diagram (Kershaw, 1964). The species c o n s t e l l a t i o n diagram i s a r e t i c u l a t e as opposed to a h i e r a r c h i c a l method of c l a s s i f i c a t i o n . The c o r r e l a t i o n s between pai r s of species are f i r s t represented i n a matrix 18 i n d i c a t i n g the degree and q u a l i t y ( p o s i t i v e or negative) of a s s o c i a t i o n . 2 In t h i s study, the X values obtained from the as s o c i a t i o n analysis were used as the measure of a s s o c i a t i o n . To draw a species c o n s t e l l a t i o n diagram, the species are arranged such that those that are highly p o s i -t i v e l y associated are positioned close to one another. The arrangement i s checked by seeing that species negatively c o r r e l a t e d are further removed from each other. Because of the d i f f i c u l t y of reducing the multi-dimensional r e l a t i o n s h i p among the species to a two-dimensional diagram, l i n e s representing the degree of p o s i t i v e a s s o c i a t i o n are drawn between species. I t i s the i n t e n s i t y of these l i n e s rather than t h e i r s p a t i a l separation which more accurately r e f l e c t s the r e l a t i o n s h i p s . 3. Cluster analysis Cluster analysis i s a hierarchical-agglomerative-polythetic method of c l a s s i f i c a t i o n . The method originated i n psychometrics and was subsequently applied to numerical taxonomy f o r s o r t i n g out taxa from a v a r i e t y of i n d i v i d u a l measurements of d i f f e r e n t c h a r a c t e r i s t i c s (Sokal & Sneath, 1963.) B a s i c a l l y , a measure of s i m i l a r i t y f o r each p a i r of e n t i t i e s (taxa i n the case of numerical taxonomy, quadrats i n plant ecology) i s derived from the measurements describing the e n t i t i e s . The e n t i t i e s are then grouped into c l u s t e r s on the basis of t h e i r s i m i l a r i t i e s to each other. The c r i t e r i a for e s t a b l i s h i n g the c l u s t e r s may be s t r i c t or l a x , r e s u l t i n g i n many or few c l u s t e r s , r e s p e c t i v e l y . A double standardization of the data was performed before s i m i l a r i t y values were computed. The biomass of each species i n each quadrat was made a percentage of the maximum value f o r that species among a l l the 19 quadrats. Then, the biomass values for a l l of the species i n any one quadrat were adjusted so that the t o t a l equalled 1.00 for each quadrat. A s i m i l a r i t y value between each pair of quadrats was then obtained by summing the minimum of the two wet weights for a l l species common to each p a i r of quadrats. Differences i n c l u s t e r i n g methods are based mainly on whether c l u s t e r s are defined by the lowest or highest or average resemblance. In t h i s study, c l u s t e r i n g was done using the weighted and unweighted pair-group average methods. The admission of any i n d i v i d u a l into a c l u s t e r i s based on the simple average of the s i m i l a r i t i e s of that i n d i v i d u a l with the other members of the c l u s t e r . This method permits only the two most highly correlated groups to j o i n at each c l u s t e r i n g c y c l e . Following each c y c l e , a new s i m i l a r i t y matrix of a l l the c l u s t e r s with each other and with unclustered i n d i v i d u a l s i s recalculated using the arithmetic averages of a l l the c o e f f i c i e n t s involved i n the c o r r e l a t i o n s of any two c l u s t e r s . In the weighted method, each i n d i v i d u a l c a r r i e s as much weight as the e n t i r e c l u s t e r to which i t i s being joined i n determining i t s r e l a t i o n to that c l u s t e r . In the unweighted method, a l l i n d i v i d u a l s are weighted equally throughout the c l u s t e r i n g procedure. Cl u s t e r i n g i s continued u n t i l every c l u s t e r has been joined together. Cluster analysis of species i s c a l l e d inverse a n a l y s i s . Instead of c a l c u l a t i n g s i m i l a r i t y values for a l l p a i r s of quadrats by the species which occur i n them, one can i n v e r t the data matrix and c a l c u l a t e s i m i -l a r i t y values for a l l p a i r s of species by the quadrats i n which they occur. C l u s t e r i n g of species also uses both the weighted and the unweighted pair-group methods with arithmetic average. 20 A method of comparing species groups and quadrat groups has been elaborated by Stephenson et a l . (1970). Their method has been followed here for comparison of the quadrat groups of c l u s t e r analysis with the species groups of inverse a n a l y s i s . The t o t a l number of occurrences of a l l species i n each species group was c a l c u l a t e d for each quadrat group. The occurrences were presented as a matrix with the columns representing quadrat groups and the rows, species groups. The matrix was standardized f i r s t by adjusting each entry to make the t o t a l of each row equal to one hundred, then adjusting the values i n the columns to make each column t o t a l equal to one hundred. Standardization of rows f i r s t and then columns was found to be more s u i t a b l e than the reverse procedure which was used by Stephenson et a l . (1970) because i t provided a greater range i n the values of the standardized matrix and was therefore considered to be more s e n s i t i v e to d i f f e r e n c e s i n quadrat group and species group composition. Only i n cases where the values were equivocal anyway did the two standardization procedures y i e l d d i f f e r e n t r e s u l t s . No o v e r r i d i n g argument can be given for p r e f e r r i n g one method to another. 4. Zurich-Montpellier analysis Because no mathematical c l a s s i f i c a t i o n procedure can simultaneously c l a s s i f y quadrats and s p e c i e s , a Zurich-Montpellier analysis was applied to the species presence-absence data for a l l quadrats. This a n a l y t i c a l method was developed by a European plant e c o l o g i s t , Braun-Blanquet. B a s i c a l l y , i t involves the rearrangement of the o r i g i n a l data matrix so that quadrats t y p i f i e d by p a r t i c u l a r groups of species occur together, and species t y p i f i e d by p a r t i c u l a r groups of quadrats occur together. 21 T r a d i t i o n a l l y , t h i s "tablework" ( i . e . , the rearrangement of the rows and columns of the matrix) was done by hand. However, i t has recently been programmed for computer by Ce £ k a and Roemer (1971). Their program was used-in the present study with a diagnostic species formula of 50-20 ( i . e . , the species must occur i n at l e a s t 50% of the quadrats belonging to a p a r t i c u l a r quadrat group and i n not more than 20% of the quadrats outside t h i s group). This formula was a r r i v e d at e m p i r i c a l l y as giving the best groupings of species and quadrats; other values ei t h e r did not d i f f e r e n t i a t e groupings or casued too great a f r a c t i o n i z a t i o n . A s i m i l a r formula was used for d e f i n i n g diagnostic quadrats: a quadrat belongs to a quadrat group i f i t contains at l e a s t 50% of the diagnostic species. The actual computer procedure begins by a r b i t r a r i l y s e l e c t i n g a species and using a l l the quadrats i n which i t occurs as the f i r s t working quadrat group. A l l species which f i t the diagnostic species formula are considered as preliminary members of the diagnostic species group. A new quadrat group i s found by applying the diagnostic quadrat r u l e . The pro-cedure i s repeated, a l t e r n a t e l y applying the diagnostic species formula and the diagnostic quadrat formula to obtain improved sets of diagnostic species and quadrats u n t i l no further changes occur i n e i t h e r the species or quadrat groups. An i n i t i a t i n g species which i s eliminated from the species group i s also omitted from further c o n sideration, thereby forming a one-species group. The e n t i r e process i s repeated by s e l e c t i n g another i n i t i a t i n g species from among the remaining species u n t i l a l l diagnostic species groups of the table are extracted. 5. Bray and Curt i s ordination Ordination was f i r s t used i n the 1930's when i t was applied to psychological t e s t i n g i n an attempt to extract from a group of tests 22 the p a r t i c u l a r mental processes being examined ( H o t e l l i n g , 1933). Goodall (1954) was the f i r s t plant e c o l o g i s t to apply the method to the study of vegetation. In 1957, Bray and Cu r t i s published a less rigorous method of ordin a t i o n than the fac t o r analysis used by t h e i r predecessors. I t i s the method of Bray and Curtis which i s used i n th i s study to order quadrats along axes of v a r i a t i o n . The Bray and Curtis method uses the same s i m i l a r i t y c o e f f i c i e n t c alculated i n c l u s t e r analysis a f t e r transforming i t to a distance measurement. Because the maximum s i m i l a r i t y between any two quadrats i s always l e s s than 1.00, even i f the quadrats are from i d e n t i c a l s i t e s , each s i m i l a r i t y c o e f f i c i e n t i s usually subtracted from the maximum s i m i -l a r i t y c o e f f i c i e n t i n the c l u s t e r analysis matrix to obtain interquadrat distances. However, Bannister (1968) has shown that the maximum s i m i l a r i t y value makes l i t t l e d i f f e r e n c e , and Gauch (1973) has found that overestimating the maximum s i m i l a r i t y value causes l e s s d i s t o r t i o n than underestimating i t . Gauch therefore concludes that there i s l i t t l e p r o f i t from using anything other than 1.00. For these reasons and because i t was easier to program, each s i m i l a r i t y c o e f f i c i e n t was subtracted from a maximum s i m i l a r i t y value of 1.00. From the new matrix containing the interquadrat distances, one sel e c t s as one end point of the f i r s t axis the quadrat with the greatest sum of interquadrat distances. The other end point has the maximum distance from the f i r s t . Each of the other quadrats i s located i n r e l a t i o n to these end points by the rearrangement of the Pythagorean theorem to give the equation 2 2 2 Xn = L + D - D7 1 a b 2L 23 where X^ i s the distance along the a x i s , L i s the distance between the end p o i n t s , and and are the distances between the quadrat under consideration and each of the two end p o i n t s . This procedure i s repeated for a l l the quadrats. The end points f o r second and subsequent axes are chosen from among the quadrats with the poorest f i t to the previous axis ( i . e . , among those quadrats located near the middle of that axis.) Two quadrats are chosen which are s p a t i a l l y close together on the f i r s t axis but are nevertheless separated by a great interquadrat distance so that they w i l l form an axis perpendicular to the f i r s t when t h e i r points are j o i n e d . Such a p a i r of quadrats can be found by subtracting the distance between each p a i r of 1 quadrats on the f i r s t axis from t h e i r actual interquadrat distance and choosing the p a i r with the maximum value. Each of the other quadrats i s located along t h i s axis using the same equation f o r the f i r s t a x i s . Usually no more than three axes are constructed because most of the v a r i a t i o n i s accounted for by the f i r s t two or three axes. r RESULTS Associ a t i o n analysis A s s o c i a t i o n analyses using the 75 species l i s t e d i n Table 1 and 2 the 124 quadrats c o l l e c t e d i n 1972 were performed using s i g n i f i c a n t X values equal to 10.871 (p = .001) and to 3.841 (p = .05). The two as s o c i a t i o n analyses appear i n Figures 1 and 2, r e s p e c t i v e l y , and the quadrats are l i s t e d i n Appendices 3 and 4. Diagrams of the l o c a t i o n of the quadrats along the transect l i n e s at the three s i t e s are given i n Figure 3. These maps include depth, occurrence of sea urchins and grazing, substrate type, and quadrat attenuation. This f i g u r e also shows the four major groupings of quadrats from the two a s s o c i a t i o n analyses. The f i r s t d i v i s i o n i n both analyses i s based on the presence or absence of Gigartina papillata, a d i s t i n c t i v e l y i n t e r t i d a l species. This d i v i s i o n thus separates the majority of the i n t e r t i d a l quadrats from the lowest i n t e r t i d a l and the s u b t i d a l quadrats. The 20 quadrats containing Gigartina are next subdivided on Sargassum mnticum for p = .001 and on Herposiphonia plumula for p = .05 to y i e l d i d e n t i c a l subgroups of quadrats. The s i x quadrats with both Sargassum and Herposiphonia also a l l contain Ceramium washingtoniensis, Microcladia borealis, Odonthalia floccosa, and Plocamium coccineum v a r . pacificum as w e l l as Gigartina. In species composition, these quadrats seem to represent a t r a n s i t i o n from the i n t e r t i d a l to the s u b t i d a l , as would be expected from t h e i r positions along the transect l i n e s . Further notable s u b d i v i s i o n of the groups of quadrats lacking Herposiphonia and Sargassum does not occur for p = .001. For p = .05, a su b d i v i s i o n on Ceramium washingtoniensis 25 Table 1. Species used i n the computer analyses. Acronym Species Number of occurrences A g a r f i Agarum fimbrj.atum Harvey 23 Agarte Agardhiella tenera ( J . Ag.) Schmitz 11 Ahnfpl Ahnfeltia plicata (Hudson) Fr i e s 4 Amplpa Amplisiphonia pacifica Hollenberg 90 Analja Analipus japonicus (Harvey) Wynne 5 Antide Antithamnion defectum K y l i n 55 Bonnno Bonnemaisonia nootkana (Esper) S i l v a 23 Boss Bossiella S i l v a 66 Botrps Botryocladia pseudodichotoma (Farlow) K y l i n 10 Branwo Branchioglossum woodii ( J . Ag.) K y l i n 16 C a l l f l Callophyllis flabellulata Harvey 70 Callha Callophyllis haenophylla S e t c h e l l 12 C a l l l i Callophyllis linearis (Kylin) Abbott & Norris 29 C a l l t u Calliarthron tuberculosum (Manza) 36 Ceraca Ceramium californicum J . Agardh 20 Cerawa Ceramium washingtoniensis K y l i n 57 Conssu Constantinea subulifera S e t c h e l l 65 Coraofc Corallina officinalis v a r.chilensis (Harvey) Kutzing 66 Costco Costaria costata (Turner) Saunders 24 Crypru Cryptopleura ruprechtiana ( J . Ag.) K y l i n 53 Crypwo Cryptosiphonia woodii J . Agardh 5 Desmvi Desmarestia viridis (Muller) Lamouroux 38 Entein Enteromorpha intestinalis (Linnaeus) Link 10 Fucudie Fucus distichus (DeLa Pylaie) Powell 19 G e l i Gelidium Lamouroux 78 Gigaex Gigartina exasperata Harvey & Bai l e y 12 Gigapa Gigartina papillata (C. Ag.) J . Agardh 20 Gracve Gracilaria verrucosa (Hudson) Papenfuss 12 Gratca Grateloupia doryphora (Montagne) Howe 8 Grifpa Griffithsia pacifica K y l i n 13 Gymnle Gymnogongrus leptophyllus J . Agardh 27 Herpri Herposiphonia plumula ( J . Ag.) Hollenberg 103 Hetede Heterosiphonia densiuscula K y l i n 15 Hollsu Hollenbergia subulata (Harvey) Wollaston 27 Irid c o c Iridaea cordata v a r . cordata (Turner) Bory 42 Kallob Kallymenia oblongifructa (Setchell) S e t c h e l l 19 Lami Laminaria Lamouroux 46 Laursp Laurencia spectabilis Postels & Ruprecht 86 Leatdi Leathesxa difformis (Linneaus) Areschoug 20 L i t h Lithothamnion F o s l i e 69 Lomesp Lomentaria hakodatensis Yendo 13 Micrbo Microcladia borealis Ruprecht 35 Micrco Microcladia coulteri Harvey 15 Monofu Monostroma fuscum (P. & R.) Wittrock 73 Nerelu Nereocystis luetkeana (Mertens) Postels & Ruprecht 36 Nien' Nienburgia K y l i n 23 Nitomi Nitophyllum mirabile K y l i n 12 Odonfl Odonthalia floccosa (Esper) Falkenberg 57 Peyspa Peyssonelia pacifica K y l i n 80 Phyc Phycodrys isabellae Norris & Wynne 7 26 Table 1. Continued. Acronym Species Number of occurrences P l a t c l Platysiphonia clevelandii (Farlow) Papenfuss 4 Platpe Platythamnion pectinatum K y l i n 41 P l a t v i Platythamnion villosum K y l i n 35 Pleova Pleonosporium vancouverianum J . Agardh 18 Ploccop Plocamium cocdneum v a r . pacificum (Kylin) Dawson 75 Polyheg Polysiphonia hendryi v a r . gardneri (Kylin) Hollenberg 14 P o l y l a Polyneura latissima (Harvey) K y l i n 39 Polypaz Polysiphonia pacifica Hollenberg 23 Polyuru Polysiphonia urceolata (Dillwyn) G r e v i l l e 9 P r i o l a Prionitis lanceolata Harvey 58 P t e r b i Pterosiphonia bipinnata (P. & R.) Falkenberg 16 Pterde Pterosiphonia dendroidea (Montagne) Falkenberg 100 Ptergr Pterosiphonia gracilis K y l i n 72 R a l f f u Ralfsia fungiformis (Gunner) S e t c h e l l & Gardner 48 Rhodla Rhodomela larix (Turner) C. Agardh 11 Rhodper Rhodymenia pertusa (P. & R.) J . Agardh 31 Rhodpl Rhodoptilum plumosum (Harvey & Bailey) K y l i n 23 Rhodro Rhodoglossum roseum (Kylin) Smith 27 Sargmu Sargassum muticum (Yendo) Fensholt 34 Schipa Schizymenia pacifica K y l i n 12 Sphaca Spahcelaria norrisii Hollenberg 4 Stenin Stenogramme interrupta (C. Ag.) Montagne 5 Ulva • Ulva Thuret 27 Weekfr Weeksia fryeana S e t c h e l l 8 Zostmal Zostera marina v a r . l a t i f o l i a (Linnaeus) Morong 5 27 Figure 1. A s s o c i a t i o n analysis of the 75 species and 124 quadrats c o l l e c t e d i n 1972 at a p r o b a b i l i t y l e v e l of .001. Each box includes the species on which the. d i v i s i o n occurred and the number of quadrats e i t h e r containing (+) or lacking (-) that species. The boxes are located i n the f i g u r e according to the X^ value at which that s u b d i v i s i o n of quadrats occurred. 100 . 90 . 80 . 70 .-60.. 23 I j g -1- ; N 1241 •»|Odonfl SO prli fsa nhodo _L 1 .ftUrt Kollcb 1 "ff>" p a J Rho*l 19— JET -|fctoni. lUJva 3L. JBo«» J55_ 2_ i i 29 Figure 2. A s s o c i a t i o n analysis of the 75 species and 124 quadrats c o l l e c t e d In 1972 at a p r o b a b i l i t y l e v e l of .05. Each box includes the species on which the d i v i s i o n occurred and the number of quadrats e i t h e r containing (+) or l a c k i n g (-) that species. The boxes are located i n the figure-according to the X value at which that s u b d i v i s i o n of quadrats occurred. Maximum X' T3 o 31 Figure 3. The four major groupings of the a s s o c i a t i o n analysis quadrats at BI01, BI02, and SI02: +Gigapa, quadrats containing Gigartina papillata, +Odonfl, quadrats cut o f f the from group lac k i n g Gigartina papillata but containing Odonthalia floccosa, -Odonfl, quadrats lacking both Gigartina papillata and Odonthalia floccosa, and +Zostmal, quadrats lacking Gigartina papillata but containing both Odonthalia floccosa and Zostera marina. SI02 Y-005 010 015' 020 025 030 035 040 045 050 055 060 065 070 075 080 085 090 095 100 X-000 X-025 Depth (In feet) _P moon tide level 4-io 9 3 ^ ^ ^ ^ 9 1 I N -odon^-l |50 A-Odftv* X-050 20 30 .40 25 —T~—^OdiW ., 36 ,"?^vOdMW 60 J.I0 20 430 .40 .50 ||0 160 20 .30 3 ^ N : -OdonVI ko X-050 86 | , ^^ WwA +0<io<\W OdonW 6 6 6 7 6 8 69 ' 7 0 71 75 76 Odowil Odotift 10 20 30 40 |50 60 70 80 .90 llOO 84 Bid Y.005 0I0 015 020 025 030 035 040 045 050 055 060 065 070 075 080 085 090 095 100 X-OS5 Depth (In feet) 0 mean tide level 33 produces four quadrats containing t h i s species and i n a d d i t i o n Leathesia difformiSf Microcladia borealis, and Polysiphonia hendryi v a r . gardneri and nine quadrats la c k i n g Ceramium but a l l containing Fucus distichus. A f t e r removing one quadrat on the occurrence of Bossiella, f i v e of the remaining eight quadrats contain Viva. These quadrats occur i n the mid i n t e r t i d a l from 0 to 5 feet below mean t i d e l e v e l . In summary, then, the i n t e r t i d a l region i s delimited by the presence of Gigartina papillata and can be divided into three subregions which show a general r e l a t i o n s h i p with increasing depth: (1) those containing Gigartina and Fucus but no Ceramium, Sargassum, or Herposiphonia, (2) those containing Gigartina and Ceramium but no Sargassum or Herposiphonia, and (3) those containing a l l four species. The 104 quadrats lacking Gigartina are f i r s t subdivided on the occurrence of Odonthalia floccosa i n both analyses. In the p = .001 a s s o c i a t i o n a n a l y s i s , the f i f t y quadrats containing Odonthalia f i r s t cut o f f a s i n g l e deep water quadrat containing Sphacelaria norrisii, then three upper s u b t i d a l quadrats containing Rhodomela larix. The f i r s t s u b d i v i s i o n of the 54 quadrats lacking Odonthalia cuts o f f two upper s u b t i d a l quadrats which contain Rhodomela. The f i v e quadrats with Rhodomela would probably be recombined i n an a s s o c i a t i o n analysis following Goodall's method (1954) because of t h e i r s i m i l a r i t i e s , but i n the Williams and Lambert analysis these quadrats must remain separate. Four of the f i v e quadrats containing Zostera marina are next removed from the main group of Odonthalia quadrats i n the p = .001 a s s o c i a t i o n a n a l y s i s . (The f i f t h quadrat containing Zostera lacks Odonthalia.) This d i v i s i o n i s followed by two more subdivisions which remove f i v e deeper water quadrats on the occurrence of Griffithsia pacifica in two 34 of them and Bonnemaisonia nootkana i n the other three. The next d i v i -s i on of the quadrats divides them nearly i n h a l f on the basis of the occurrence of Polyneura latissima,.the 15 quadrats containing the species occurring at intermediate depths between both shallower and deeper quadrats l a c k i n g that species. The remainder of the s i g n i f i c a n t sub-d i v i s i o n s cut o f f only one species at a time, except for the f i n a l sub-d i v i s i o n of the quadrats l a c k i n g Polyneura. This s u b d i v i s i o n produces 9 quadrats containing Rhodymenia pertusa and 11 quadrats l a c k i n g t h i s species. Once again there are no d i s t i n c t differences i n depth between the quadrats containing and those l a c k i n g t h i s species. In the p = .05 a s s o c i a t i o n a n a l y s i s , the 50 quadrats containing Odonthalia are subdivided f i r s t on Zostera marina, removing the four quadrats containing that species from further consideration. A l l four quadrats also contain Agardhiella tenera, Gelidium, Laminaria spp., Laurencia spectabilis, Monostroma fuscum, Plocamium coccineum var. pacificum, and Pterosiphonia dendroidea. (The f i f t h quadrat containing Zostera but l a c k i n g Odonthalia also contains a l l thase species.) The 46 quadrats l a c k i n g Zostera are subdivided on Constantinea subulifera. Only f i v e quadrats lack t h i s species. A l l occur at depths of 27-35 feet, and a l l contain Bossiella, Herposiphonia plumula, Lithothamnion, Peyssonelia pacifica, Plocamium coccineum var. pacificum, and Pterosiphonia dendroidea. The 41 quadrats containing Constantinea occur at depths of 10-29 feet. They are next subdivided on Antithamnion defectum. The 24 quadrats with Antithamnion also contain Callophyllis flahellulata, Herposiphonia plumula, and Pterosiphonia dendroidea. Those la c k i n g Antithamnion contain Laurencia spectabilis. No g e n e r a l i z a t i o n as to depths of occurrences of these two groups of quadrats can be made. 35 Quadrats containing Antithamnion i n the p = .05 a s s o c i a t i o n analysis next cut o f f two lower i n t e r t i d a l quadrats containing Rhodomela larix. The 22 quadrats la c k i n g Rhodomela lose two quadrats containing Rhodoptilum plumosum before being subdivided on Corallina officinalis v a r . chilensis. A l l but four of the 20 quadrats l a c k i n g Rhodoptilum contain Corallina. These 16 quadrats also contain Cryptopleura ruprechtiana and Laurencia spectabilis i n addition to the species previously l i s t e d for the quadrats with Antithamnion. These quadrats occur i n 12 to 23 feet of water. Although nine more s i g n i f i c a n t subdivisions occur on t h i s group of quadrats, a l l but one cut o f f only one or two quadrats at a time. The s i n g l e excep-t i o n divides a group of eight quadrats into 3 and 5, also groups generally too small to be of great e c o l o g i c a l s i g n i f i c a n c e . The 17 quadrats l a c k i n g Antithamnion i n the p = .05 a s s o c i a t i o n analysis are subdivided on the occurrence of Nereocystis luetkeana, the eight quadrats with that species also containing Amplisiphonia pacifica, Gelidium, Herposiphonia plumula, Laurencia spectabilis, Monostroma fuscum, Prionitis lanceolata, and Pterosiphonia dendroidea. These eight quadrats occur at an average depth of 23 + 3.0 f e e t . The nine quadrats la c k i n g Nereocystis occur at an average depth of 1 3 + 2 . 5 f e e t . They contain Bossiella, Corallina officinalis v a r . chilensis, Cryptopleura ruprechtiana, and Laurencia spectabilis. Once again, the subsequent subdivisions of the quadrats remove only one to three quadrats at a time, and the resultant groups are of l i t t l e s i g n i f i c a n c e . In the p = .001 ass o c i a t i o n a n a l y s i s , the large group of quadrats l a c k i n g both Odonthalia and Rhodomela i s next.subdivided on the occurrence of Bossiella i n 19 of the 52 quadrats. The majority of these 19 quadrats occur early i n the summer, but there i s no apparent a s s o c i a t i o n of the 36 species with a p a r t i c u l a r depth. These quadrats are subdivided.on the occurrence of Platythamnion pectinatum, which e f f e c t i v e l y cuts o f f most of the deeper quadrats containing Bossiella. The l a s t s i g n i f i c a n t sub-d i v i s i o n of t h i s group cuts o f f a s i n g l e quadrat containing Rhodoptilum plumosum. The 33 quadrats l a c k i n g Bossiella i n the p = .001 a s s o c i a t i o n analysis undergo ten more s i g n i f i c a n t subdivisions. A l l but three cut o f f only one or two quadrats at a time and are disregarded. A s u b d i v i s i o n on Branchioglossum woodii divides the remaining quadrats i n t o f i v e containing that species and 26 l a c k i n g i t . Four of the f i v e occur at the deeper s i t e s along the BI01 085 transect l i n e . Another s i g n i f i c a n t d i v i s i o n occurs on Nienburgia, producing four quadrats with that species and 17 without i t . These four quadrats occur at moderate depths at both BI01 and SI02 i n mid summer. The t h i r d s u b d i v i s i o n of note occurs on Nereocystis luetkeana. The four quadrats containing i t occur at the l i m i t of i t s d i s t r i b u t i o n ; the 12 quadrats lacking i t generally occur below i t s d i s t r i b u t i o n a l l i m i t s . In the p = .05 a s s o c i a t i o n a n a l y s i s , the 54 quadrats l a c k i n g Odonthalia are also subdivided f i r s t on Rhodomela larix. This s u b d i v i s i o n removes two lower i n t e r t i d a l - u p p e r s u b t i d a l quadrats containing this species. Of the 52 quadrats lacking both species, 39 contain Pterosiphonia gracilis, the species on which the next s u b d i v i s i o n i s made. The quadrats containing Pterosiphonia gracilis occur at an average depth of 37.6 + 12.2 feet with a range from 23 to 81 f e e t , and those l a c k i n g i t occur from +0.5 to 37 feet with an average of 23.6 + 9.4 f e e t . A l l subsequent subdivisions of the quadrats lacking Pterosiphonia gracilis cut o f f only one or two quadrats at a time, i n d i c a t i n g that t h i s group of quadrats i s quite heterogeneous. 3.8+2.8 6.6+1.2 8.5+2.1 13.0+2.5 15.9+3.3 23.0+3.0 23.6+9.4 31.3+4.1 32.5+7.1 35.8+5.8 52.9+24.4 +Gigapa +Gigapa +Gigapa +0donfl +Odonfl +0donfl -Odonfl +Odonfl (+Fucud±e) +Cerawa +Herpri +Conssu +Conssu +Conssu -Ptergr -Conssu (+Cerawa) (+Coraofc) +Antide +Nerelu (9) (4) +Coraofc (13) (5) (6) (1£ -Odonfl -Odonfl -Odonfl +Ptergr +Ptergr +Ptergr -Coraofc +Coraofc +Botrps 7) INTERTIDAL UPPER SUBTIDAL LOWER SUBTIDAL Figure 4. Arrangement of the important subdivisions of the p = .05 a s s o c i a t i o n a n a l y s i s according to the depths ( i n feet below mean tide l e v e l ) of the quadrats contained i n the s u b d i v i s i o n s . Species presences i n parentheses denote species found i n a l l quadrats of the s u b d i v i s i o n but which were not species on which the subdivisions were made. Numbers i n parentheses i n d i c a t e the number of quadrats i n that s u b d i v i s i o n . Species absences are denoted only i n those cases where necessary for d i s t i n g u i s h i n g p a r t i c u l a r s u b d i v i s i o n s . 38 The 39 quadrats la c k i n g Pterosiphonia gracilis eliminate a s i n g l e quadrat containing Iridaea cordata v a r . cordata and two quadrats l a c k i n g Herposiphonia plumula before being subdivided on Botryocladia pseudo-dichotoma,a deep water a l g a . The seven quadrats containing Botryocladia occur at an average depth of 52.9 + 24.4 feet with a range from 31 to 81 f e e t . A l l further s i g n i f i c a n t subdivisions of this group remove only one quadrat at a time. The 29 quadrats l a c k i n g Botryocladia occur at an average depth of 33.2 + 7.0 feet with a range from 23 to 46 f e e t . The next s u b d i v i s i o n of th i s group cuts o f f the remaining sandy bottom quadrat containing Zostera marina. Then a su b d i v i s i o n on Corallina officinalis v a r . chilensis divides the remaining group into nine quadrats containing Corallina and 19 lacking i t . The average depths of the two groups of quadrats are 35.8 + 5.8 and 32.5 + 7.1 f e e t , r e s p e c t i v e l y , not a s i g n i f i c a n t d i f f e r e n c e . Of greater importance, however, i s the occur-rence of Corallina i n quadrats only i n the l a t e spring and early summer. The next subd i v i s i o n on the quadrats with Corallina cuts o f f four quadrats with Branchioglossum woodii, a l l of which occur at the deep end of the BI01 085 transect. The only other s u b d i v i s i o n of the as s o c i a t i o n analysis with each subgroup containing at l e a s t four quadrats occurs on Desmarestia viridis. The four quadrats with Desmerestia occur at the BI01 s i t e i n early summer at the deeper end of the transect l i n e s . I n general, the 54 quadrats la c k i n g both Gigartina and Odonthalia occur at the deeper ends of the transects and are more heterogeneous i n composition than the shallower quadrats containing Odonthalia. The two l e v e l s of as s o c i a t i o n analysis agree completely on the four basic community subd i v i s i o n s—a Gigartina papillata community occurring i n the i n t e r t i d a l , an Odonthalia floccosa community occurring i n the 39 upper s u b t i d a l , a Zostera marina community occurring on the sandy bottom, and a community lac k i n g a l l three diagnostic species occurring i n the lower s u b t i d a l . The main d i v i s i o n of the quadrats occurs p r i m a r i l y on depth and secondarily on substrate. The i n t e r t i d a l quadrats (+Gigapa) occur at a depth of 5.7 + 3.1 f e e t , the upper s u b t i d a l quadrats (+Odonfl) at 19.2 + 6.3 f e e t , and the lower s u b t i d a l quadrats (-Odonfl) at 33.2 + 13.9 f e e t . Both analyses produce v a r i a t i o n s i n further subdivisions of the quadrats into d i s t i n c t communities. The important subdivisions of the p = .05 a s s o c i a t i o n analysis are arranged for comparison according to depth i n Figure 4. Any generalizations regarding these subdivisions need support from other methods of a n a l y s i s . The subdivisions of the quadrats into one group containing only one or a few and the other group containing the remainder of the quadrats occurs quite frequently i n the a s s o c i a t i o n a n a l y s i s . Such a s u b d i v i s i o n r e s u l t s from two or more infrequently occurring species showing c o i n c i -dental occurrences and i s generally useless i n suggesting e c o l o g i c a l r e l a t i o n s h i p s . However, i t can be extremely us e f u l i n p o i n t i n g out p o s s i b l e discrepancies i n species i d e n t i f i c a t i o n s , i n occurrences due to previously undetected contamination, and i n missing data due to incomplete c o l l e c t i o n s . Many of the early subdivisions of some of the l a r g e r groups are p u r i f i c a t i o n procedures that eliminate quadrats which do not f i t together w e l l with the majority of the quadrats i n that group for one reason or another. Even more s i g n i f i c a n t l y , subdivision of only one or two quadrats at a time r e f l e c t s the fact that most of the species i n the analysis show d i s t r i b u t i o n s independent of one another: even the "communities" of species composed of f i v e or more quadrats are probably the r e s u l t of overlapping AO species d i s t r i b u t i o n s and are not e n t i t i e s which can be completely separated from every other community s t r i c t l y on the basis of species composition. Furthermore, such subdivisions show that species d i s t r i -butions are not e n t i r e l y p r e d i c t a b l e and that a complex of factors may be responsible for the absence of a s i n g l e species i n a given quadrat i n which i t s occurrence i s expected. F i n a l l y , i t i s worth noting that the ultimate subdivisions of the quadrats i n the a s s o c i a t i o n analysis generally produce groups of ad-jacent quadrats from the same transect l i n e , groups of quadrats from s i m i l a r depths at the same s i t e , or groups from s i m i l a r depths at d i f f e r e n t s i t e s . This r e s u l t supports the continuity of species d i s t r i -butions i n both time and space. Species c o n s t e l l a t i o n diagram The a s s o c i a t i o n matrix from which the species c o n s t e l l a t i o n diagram was derived appears i n Figure 5. The species have been rearranged, by hand, so that the p o s i t i v e associations tend to c l u s t e r around the main diagonal. The species c o n s t e l l a t i o n diagram for species associations with p *.001 i s shown i n Figure 6 and for .001 p .01 i n Figure 7. (Seventeen of the .001 <-p <.01 associations were not included i n Figure 7 because of geometric problems.) Six groups of species showing varying degrees of association can be determined from t h i s diagram. Two species {Ralfsia fungiformis and Polysiphonia pacifica) were e n t i r e l y ignored because the former showed no s i g n i f i c a n t p o s i t i v e associations at p = .01, and the s i g n i f i c a n t associations of the l a t t e r species did not reveal c o r r e l a t i o n s with any p a r t i c u l a r group of species. In p a r t i c u l a r , i t i s thought that Polysiphonia pacifica may include two subspecies or that i t occurred as a contaminant i n a s i g n i f i c a n t number of quadrats. 41 Figure 5. The a s s o c i a t i o n matrix showing p o s i t i v e (-H-, p < .001 and +, p ^.01) and negative (=, p <.001 and p < .01) associations between a l l pa i r s of species. Giflopa P^+ty Fucudie It £ _ I 5 ? ±5 +• + Entein Polyheg Crypwo Pterbi Ulvo Leatdi Analja Gratca Rhodla Lomesp Sargmu Cerowa Rhodro Ahnfpl !ttt± Odonfl Iridcoc Ploccop C i i Crypru Priolo Boss Conssu Colli i Coraofc i i i Micrco Lours p Calltu Gymnle Rhodper Herpri Geli x5 Conn Nion Desmvi Lt. Pterde Amp I pa Nerelu Lami Costco Nltomi Platvi Polyuru Hollsu Schipo "nfi Ii A ride Polyla Lith Peysoo Bronwo Monofu Ptergr Pleova Platpe Cera c a Bonnno Agarfi Phyc H< otede Rhodpl Callha Grifpa Botrps Plofel Weekfr Sphoca Stenln Zostmal Asar'e Grocve Kallob Glgaex 43 Figure 6. The species c o n s t e l l a t i o n diagram with l i n e s connecting species which are highly p o s i t i v e l y associated (p <.001). 45 Figure 7. The species c o n s t e l l a t i o n diagram with l i n e s connecting species which are p o s i t i v e l y associated (.OOKp <.01) and not included i n Figure 6. 47 An i n t e r t i d a l a s s o c i a t i o n showing strong i n t e r n a l a f f i n i t i e s occurs i n the upper left-hand corner of the f i g u r e s . This group of species i s composed of s t r i c t l y i n t e r t i d a l species and a few species with d i s t r i -butions that overlap the s u b t i d a l species but which nevertheless have t h e i r strongest a f f i n i t i e s with the i n t e r t i d a l species. Microcladia borealis, a mid to lower i n t e r t i d a l species, i s intermediate between t h i s group and the group of lower i n t e r t i d a l - u p p e r s u b t i d a l s p e c i e s , showing strong a f f i n i t i e s to species i n both groups. A group of lower i n t e r t i d a l - u p p e r s u b t i d a l species occurs on the upper left-hand side of the c l u s t e r i n the center of the page. These species include the f o l i o s e red algae which occur abundantly from near the lower l i m i t s of the i n t e r t i d a l into the upper s u b t i d a l region. They have the most i n t e r n a l p o s i t i v e associations of any group. The lower right-hand side of the c e n t r a l c i r c l e i s composed of another group with strong i n t e r n a l a f f i n i t i e s but with a large number of external a f f i n i t i e s as w e l l . This group may be considered a shade-l o v i n g group because the species frequently occur as understory elements i n shallow water but among the canopy species i n deep water. Besides a f f i n i t i e s with the f o l i o s e red group, these species show a f f i n i t i e s with the other three groups to be considered. In general, t h i s group i s com-posed of turf-forming filamentous or encrusting species and includes the most ubiquitous species i n t h i s study. A deep-water a s s o c i a t i o n of species occurring at the lower l i m i t s of s u b t i d a l plant d i s t r i b u t i o n s can be seen i n the upper right-hand corner of the diagram. As previously noted, several of these species show strong associations with the shade-loving group of the large c e n t r a l c i r c l e , and indeed some of the species i n t h i s group occur i n shallower waters i n the 48 understory of the Zostera community, i n d i c a t i n g that the d i s t i n c t i o n between these two deep-water groups i s probably not quite as sharp as pictured i n the diagram. The group of species on the lower left-hand side of the diagram i s found on sand. Although Zostera marina i s the only species i n t h i s group req u i r i n g a s o f t bottom, the other species seem to p r e f e r rocks surrounded by sand. In general, t h i s group contains few s p e c i e s , prob-ably because of the lack of a permanent and hard substrate for attachment. Because a l l the s i t e s . c o n t a i n i n g Zostera were contiguous and occurred at approximately the same depths, one cannot generalize on the species composition of t h i s community u n t i l further c o l l e c t i o n s have been made. Personal observation i n the lower i n t e r t i d a l and upper s u b t i d a l regions seems to confirm the preference of Agardhiella tenera, Gracilaria verrucosa, Kallymenia oblongifructa, and Gigartina exasperata for rocks surrounded by sand (although large Gigartina also seem to favor v e r t i c a l rocks i n channels with s w i f t currents.) The group of species occurring on the lower right-hand side of the diagram i s dominated by the large brown kelp Nereocystis luetkeana. Also included i n t h i s group are two other annual brown algae {Costaria costata and Desmarestia viridis), several species of f o l i o s e red algae (Nitophyllum mirabile and Schizymenia pacifica), and a number of filamentous or poly-siphonous understory or epiphytic red algae. Laminaria spp. also occur i n many of the same quadrats. The element of time i s p a r t i a l l y represented by the occurrence of Desmarestia viridis, a widely d i s t r i b u t e d spring annual which i s usually absent by mid summer. I t shows strong p o s i t i v e associations with Nereocystis and with Calliarthron tuberculosum and weaker a f f i n i t i e s to 49 Constantinea subulifera, Laurencia spectabilis, Corallina officinalis v a r . chilensis, Bossiella, Herposiphonia plumula, Bonnemaisonia nootkana, Pterosiphonia dendroidea, and Costaria costata. In the cases of Herpo-siphonia, Pterosiphonia, Laurencia, and Calliarthron, the a s s o c i a t i o n appears to be due to overlapping s p a t i a l d i s t r i b u t i o n s . However, for Nereocystis, Costaria, Constantinea, and Bonnemaisonia, the overlap appears to be temporal. A l l four species are numerically most abundant i n spring and disappear through the summer months, the three former species being p a r t i c u l a r l y susceptible to bleaching i n the lower i n t e r -tidal-upper s u b t i d a l region during the daytime spring ebb t i d e s . The overlap with Corallina and Bossiella i s equivocal whether i t i s s p a t i a l or temporal. Cluster analysis The dendrogram of the quadrat c l u s t e r analysis using the weighted and unweighted pair-group methods appear i n Figures 8 and 9, r e s p e c t i v e l y . Three major groups of quadrats are delimited by both c l u s t e r analyses. These groups correspond to an i n t e r t i d a l , an upper s u b t i d a l , and a lower s u b t i d a l assemblage. A sandy bottom species association can also be distinguished i n both analyses. Both s u b t i d a l assemblages are further subdivided into four and three groups, r e s p e c t i v e l y . Differences i n depth may account f o r the three major groups of quadrats but to a l e s s e r degree for the subgroups: the upper s u b t i d a l quadrats occurring at 19.4 +6.1 feet and 18.8 + 6.1 feet for the weighted and unweighted methods, r e s p e c t i v e l y , and the lower s u b t i d a l quadrats at 37.7 + 11.9 feet and 34.8 + 12.4 f e e t , r e s p e c t i v e l y . 50 Figure 8. Dendrogram showing the c l u s t e r i n g of the 124 quadrats c o l l e c t e d i n 1972 using the weighted p a i r group average method of c l u s t e r a n a l y s i s . Lettered groupings of quadrats are discussed i n the t e x t . I-saxr -83XT" Ezocr-9 3 X T ~ 0 9 0 0 -T 9 0 0 -9S0Cr" 6 9 3 0 — TSOCT" asto— 3 3 0 0 -UUJ • s x r E a x r ono-ssro-3*30-aaxr-saxr czxr ocrrcr ZSCXT" vmr-xaxr 3 B 0 0 -a s t o -c w x r 2 0 1 0 — BOO"" * o c r E O T O -osoo-B T O O -b i o r racer" •SOO" 3E00-E3XT EBOJ-aaocr b l W E 3 0 0 -FKOQ-g e c c r > racxr ESOO" saxr 90ta ( € 0 0 " oaxr S900" CTKT 93X0" 3B0O" ei gaxr 2X00" •stcr EflOO" soar cew fvnrr 6 3 X 0 " osocr xoxcr saxr EETO" FFfrr £socr £ 2 X 0 " owxr »exr 3 6 0 0 " TBOCT saocr zoxcr wocr B O K T aaccr 0TO0 — • tJUUU ™ 52 Figure 9, Dendrogram showing the c l u s t e r i n g of the 124 quadrats c o l l e c t e d i n 1972 using the unweighted pair-group average method of c l u s t e r a n a l y s i s . Lettered groupings of quadrats are discussed i n the t e x t . wccr ancr osuu~ »ccr aoxr-aco-EOTO-TEOO-taxr aooo-2BD0-i"3no~ SBXT /pcrr COTCT taxr OBOcr aaxr S20C-•axr- o e s x r sxr »E00-svsu SBxr tlUJ BOKT i/rcr EBXT aaxr isocr 3&XT •=sxr aaxr rrKT-rsxr rexr E s x r -retxr EBXT SBQT" OOCT" Boxr-srocr 54 Both, the weighted and unweighted pair-group methods cut o f f a group of i n t e r t i d a l quadrats—25 i n the case of the former and 21 i n the l a t t e r . In both analyses, these quadrats are the l a s t to j o i n the c l u s t e r , i n d i -c a t i n g that they are the most d i s t i n c t from the other quadrats. In the weighted method, a l l but f i v e quadrats occur above zero t i d e l e v e l , and four of these f i v e quadrats are separated from the next most s i m i l a r quadrat at a s i m i l a r i t y l e v e l of .11 i n the weighted method and are combined with the upper s u b t i d a l quadrats rather than with the i n t e r t i d a l quadrats i n the unweighted procedure. The depth of the f i f t h quadrat was estimated at one foot below zero t i d e l e v e l . The f a c t that i t occurs with the other i n t e r t i d a l quadrats may i n d i c a t e that i t s r e a l depth was shallower than estimated. The l e v e l of s i m i l a r i t y among the l a s t two groups of i n t e r -t i d a l quadrats to unite i s quite low, .06, and many of the other quadrats are joined at l e v e l s considerably lower than among the upper s u b t i d a l quadrats, thereby supporting the idea that the i n t e r t i d a l zone represents a more i n t e r n a l l y heterogeneous assemblage of species. The f i v e quadrats occurring on sand j o i n the other s u b t i d a l quadrats at a s i m i l a r i t y l e v e l of .06 to .07 i n both methods. The two most d i s s i m i -l a r groups of quadrats within t h i s group j o i n at a s i m i l a r i t y l e v e l of .32, i n d i c a t i n g a high.degree of i n t e r n a l homogeneity. Both the weighted and unweighted methods delimit four groups of upper s u b t i d a l quadrats, but the composition of the four groups i s s l i g h t l y d i f f e r e n t . I f the four groups are designated R, S, T, and U i n that order from l e f t to r i g h t on the unweighted p a i r group dendro-gram (R', U', S', and T' on the weighted dendrogram), quadrats i n group R occur at 11.9 + 1.6 feet (16.3 + 4.5 feet for R'), i n group S at 22.2 + 5.6 55 feet (24.6 + 4.9 feet f o r S*), i n group T at 21.9 + 5.3 feet (22.0 + 5.9 f e e t f or T ' ) , and i n group U at 15.9 + 4.0 feet (14.8 + 3.0 feet for U'). The differences i n depths of the two groups are not s i g n i -f i c a n t f o r S and S', T and T', and U and U' at p = .05. The differences i n depths for R and R' cannot be tested because t h e i r variances are s i g -n i f i c a n t l y d i f f e r e n t at p = .05. In the unweighted method, group R i s d i s t i n c t . I t contains the shallowest quadrats. However, i t s d i f f e r e n c e from the other groups cannot be tested s t a t i s t i c a l l y because the variances are s i g n i f i c a n t l y d i f f e r e n t . In the dendrogram, i t i s the l a s t of the four upper s u b t i d a l groups to j o i n the remainder at a s i m i l a r i t y l e v e l of .16. Groups S and T contain quadrats whose depths are not s t a t i s t i c a l l y d i f f e r e n t at p = .05. These two groups j o i n at the highest s i m i l a r i t y l e v e l (.21) of the four upper s u b t i d a l groups. The depths of quadrats i n group U are s i g n i f i c a n t l y d i f f e r e n t from those i n both S and T. Group U j o i n s the S-T supergroup at a s i m i l a r i t y l e v e l of .20. In the weighted method, the depths of quadrats R' and U' and those of S' and T' are not s i g n i f i c a n t l y d i f f e r e n t . However, a l l other combi-nations are s i g n i f i c a n t l y d i f f e r e n t . These s i m i l a r i t i e s and differences are suggested by the dendrogram—with R' and U' j o i n i n g at a s i m i l a r i t y l e v e l of .22 and S' and T1 j o i n i n g at a s i m i l a r i t y l e v e l of .21; groups R'-U' and S'-T' not j o i n i n g u n t i l a s i m i l a r i t y l e v e l of .18. From these comparisons, i t seems obvious that differences i n depth can account for the major dif f e r e n c e s i n species composition of the quadrats under consideration. In the unweighted pair-group method, the three groups of lower s u b t i d a l quadrats are designated V, W, and X. The 12 quadrats i n group 56 V occur at depths from 28 to 53 feet with an average depth of 39.6 + 9.5 f e e t . A l l quadrats contain Agarum fimbriatum. Group W contains 15 quadrats ranging i n depth from 27 to 81 feet with an average depth of- 38.3 + 13.1 f e e t . The only quadrats with Agarum occur o f f Sear Island at depths of 32 to 44 f e e t . The ten quadrats i n group X range i n depth from 10 to 29.5 feet with an average depth of 23.7 + 7.3 f e e t . This group contains no quadrats with Agarum and no quadrats from Sear I s l a n d . There appears to be no obvious explanation i n terms of p h y s i c a l factors or species composition for d e l i m i t a t i o n of these three groups of deep water quadrats. Groups V and W are joined at a s i m i l a r i t y l e v e l of .18. There i s no s i g n i f i c a n t d i f f e r e n c e i n the depths at which t h e i r quadrats occur. Group X i s added to this supergroup at .10. The depths of i t s quadrats are s i g n i f i c a n t l y d i f f e r e n t from those of group V but not of group W at p = .05. In the weighted c l u s t e r i n g , the quadrats are s l i g h t l y rearranged so that group W i s enlarged to include three a d d i t i o n a l quadrats from group V, and group V i s subdivided into the four quadrats from the Bath Island s i t e which contain Agarum (group V^) and into the remaining f i v e quadrats i n the group and an a d d i t i o n a l deep water quadrat from SI02 which was only poorly clustered i n the unweighted method (group V^). Group joi n s group W at a s i m i l a r i t y l e v e l of .19, and group j o i n s t h i s supergroup at .16. Group X' i s no longer joined to the deeper water quadrats. I t has been s p l i t i n t o two, one group of eight quadrats (group X|) showing closer a f f i n i t y to the shallow s u b t i d a l quadrats (.08) and the other group of four quadrats (plus one from group T—thus forming group X^) having i t s c l o s e s t a s s o c i a t i o n (.12) with the quadrats occurring on sand. A l l but one quadrat i n group X~ contains Laminaria as do a l l 57 the quadrats on sand. Therefore, i t would be best to consider group X' as t r a n s i t i o n a l between the deeper and shallower s u b t i d a l communities. None of the groups show s i g n i f i c a n t differences i n depths at p = .05. Four of the deeper quadrats i n both methods c l u s t e r poorly, not j o i n i n g the other deep water quadrats u n t i l j u s t before or even a f t e r they have clustered with the other s u b t i d a l groups. The inverse analysis of species using the weighted and unweighted pair-group methods are diagrammed i n Figures 10 and 11. In general, both methods produce very s i m i l a r r e s u l t s to those of the species c o n s t e l l a t i o n diagram. Once again, s i x associations of species are evident. The deep water a s s o c i a t i o n generally contains the same species as i n the species c o n s t e l l a t i o n diagram but now includes Peyssonelia pacifica, previously put i n t o the turf-forming group, and lacks Ceramium californicum and Rhodoptilum plumosum, which show closer a f f i n i t i e s to the sandy bottom and turf-forming a s s o c i a t i o n s , r e s p e c t i v e l y . The sandy bottom a s s o c i a t i o n contains i n a d d i t i o n to Zostera marina, Agardhiella tenera, Gracilaria verrucosa and Ceramium californicum, two filamentous s p e c i e s , Platythamnion villosum and Polysiphonia urceolata, i n both analyses and Gigartina exasperata and Hollenbergia subulata i n only the unweighted a n a l y s i s . The l a s t two species are joined to the laminarian community i n the weighted a n a l y s i s . In a d d i t i o n to the large browns (Laminaria, Nereocystis, and Costaria), t h i s a s s o c i a t i o n contains the turf-forming species Amplisiphonia pacifica, Gelidium, Pterosiphonia dendroidea, and Monostroma fuscum. Calliarthron tuberculosum, Corallina officinalis v a r . chilensis, and Laurencia spectabilis also occur i n t h i s community i n the unweighted method. 58 Figure 10. Dendrogram showing the c l u s t e r i n g of the 75 species using the weighted pair-group average method of inverse a n a l y s i s . Figure 11. Dendrogram showing the c l u s t e r i n g of the 75 species using the unweighted pair-group average method of inverse a n a l y s i s . otaxtr NI31KT atairc-VtMJIDT-3iorr jru-M31K3-rrxTin— V » V S 3 J -aaxnar-in l-1 £ Z V d A T M -3 T * U £ T -rubrwr-V31WO -W W ?! b aajmr-iTnvj-n 9 9 C J ~ ru»c*r-r n i u i I o E i f V T " soaaur-r w v<n*rv— V d l K E -ZVdATttT" lAlV-kT" vjwccr •MUSQZ-TUV-kf-tuioar-3 D 3 L X -O H O B -VcSA3kT" 3UIV -U -3mix~ 60 However, i n the weighted method, these l a s t three species are components of the shallow s u b t i d a l f o l i o s e red a s s o c i a t i o n which also contains Callophyllis flabellulata, Constantinea subulifera, Callophyllis linearis, Cryptopleura ruprechtiana, Iridaea cordata v a r . cordata, Herposiphonia plumula, Ceramium washingtoniensis, Prionitis lanceolata, and Plocamium coccineum v a r . pacificum. Species with uncertain a f f i n i t i e s to t h i s group include Desmarestia viridis, Nitophyllum mirabile, and Microcladia coultcri. The composition of the turf community has been reduced to Ptero-siphonia gracilis, Antithamnion defectum, Polyneura latissima, Nienburgia, Rhodymenia pertusa, and Ulva, with Grateloupia doryphora, Ralfsia fungi-form! s , Branchiogldssum woodii, Rhodoptilum plumosum, Gymnogongrus leptophyllus, Ahnfeltia plicata, Polysiphonia pacifica, and Schizymenia pacifica showing uncertain a f f i n i t i e s to t h i s group. The i n t e r t i d a l a s s o c i a t i o n has been expanded from the species i n the c o n s t e l l a t i o n diagram to include Bossiella, Lomentaria, Lithothamnion, Odonthalia floccosa, Kallymenia oblongifructa, Sargassum muticum, and Rhodoglossum roseum i n both analyses. As with the quadrats i n the c l u s t e r a n a l y s i s , the species i n the inverse analysis j o i n at the lowest s i m i l a r i t y l e v e l s for the i n t e r t i d a l and deep su b t i d a l groups and at the highest l e v e l s for the sandy bottom and shallow s u b t i d a l a s s o c i a t i o n s . In general, the weighted and unweighted methods of c l u s t e r and inverse analysis produce the same r e s u l t s . Results of the weighted species procedure tend to confirm the r e s u l t s of the species c o n s t e l l a t i o n diagram more than those of the unweighted procedure, whereas the unweighted 61 quadrat procedure follows more c l o s e l y the a s s o c i a t i o n analysis quadrat d i v i s i o n than does the weighted method. Sokal and Rohlf (1962) found that among four c l u s t e r i n g methods tested-—including the two used i n t h i s study—the unweighted pair-group method with arithmetic average gave the highest c o r r e l a t i o n with the o r i g i n a l data. The l e v e l of s i m i l a r i t y at which the d i f f e r e n t groups are delimited i n the c l u s t e r and inverse analyses varies g r e a t l y . Tn general, the l e v e l of s i m i l a r i t y i s lowest f o r both i n t e r t i d a l quadrats and species and nearly as low f o r the deep water species and quadrats. The highest a f f i n i t i e s occur for the group of shallow water red algae i n the inverse analysis and for a s i m i l a r group of uppper s u b t i d a l quadrats i n the c l u s t e r a n a l y s i s . A s i m i l a r i n t e n s i t y of a s s o c i a t i o n among the species i s seen i n the species c o n s t e l l a t i o n diagram. The most probable expla-nation for this phenomenon i s not that associations become weaker at the extremes of marine a l g a l d i s t r i b u t i o n s but that as such p h y s i c a l l i m i t s are reached, not only do the plants become scarcer but also the l i m i t s of p a r t i c u l a r species become more evident by the r e l a t i v e paucity of the f l o r a . Although o r i g i n a l l y i t was thought that these low s i m i l a r i t y l e v e l s were the r e s u l t of fewer samples and inadequate quadrat s i z e , i t i s now believed that the data accurately represent the r e a l nature of the com-munities . Quadrat group/species group coincidence tables were calculated separately for the weighted and unweighted c l u s t e r i n g methods using ten quadrat groups (R through Z as given i n Figures 8 and 9 and a group containing a l l quadrats not belonging to one of the other nine groups) and the. s i x species groups (as they appear i n Figures 10 and 11.) The 62 species which d id not c l u s t e r w e l l i n the unweighted method were not used. A l l species were included i n the weighted method. The standardized r e s u l t s f o r both the weighted and the unweighted analyses appear i n Tables 2 and 3, r e s p e c t i v e l y . For both analyses, the i n t e r t i d a l quadrat group shows the highest coincidence with the i n t e r t i d a l species group. The r e l a t i v e l y high coincidence of the i n t e r t i d a l species group with quadrat group R i n the unweighted analysis r e s u l t s from the occurrence of four of the i n t e r t i d a l (according to the weighted analysis but not according to t i d a l data) quadrats i n that group. The quadrats occurring on sand show the highest coincidence with the species p r e f e r r i n g a sandy h a b i t a t . The deep water species show the highest l e v e l of coincidence with quadrat group V and secondarily with quadrat group W i n both the weighted and unweighted analyses. As previously noted, these two groups contain the deepest quadrats with mean depths of 39.6 + 9.5 feet and 38.3 + 13.1 f e e t , r e s p e c t i v e l y , f o r the unweighted groups. The coincidence of the three remaining species groups with the shallower s u b t i d a l quadrat groups i s not so c l e a r . The f o l i o s e red group shows the highest coincidence with quadrat group R for both the weighted and unweighted methods and shows nearly as high a coincidence f o r quadrat group S i n the unweighted method. This high coincidence occurs because eight quadrats" occurring i n group R1 i n the weighted method occur i n group S i n the unweighted method. Because many of the species i n the f o l i o s e red group have wide d i s t r i b u t i o n s , t h i s group shows moderate coincidence values with a l l quadrat groups, incl u d i n g the i n t e r t i d a l , 63 Table 2. Standardized species-group/quadrat-group coincidence table for weighted c l u s t e r and inverse analyses r e s u l t s . Groups considered c o i n c i -dental are i t a l i c i z e d . Quadrat Groups R' S' T* U' V* W« X' Y' Z' 0 I n t e r t i d a l 21 10 12 19 5 10 11 8 56 11 Fo l i o s e red 33 23 20 20 10 12 10 10 17 14 Laminarian 19 28 20 12 16 19 16 15 8 16 Turf 16 11 18 20 14 18 22 8 15 20 Deep water 5 11 8 5 47 31 19 0 5 23 Sand 14 17 22 24 9 11 21 59 0 16 Table 3. Standardized species-group/quadrat-group coincidence table f or unweighted c l u s t e r and inverse analyses r e s u l t s . Groups considered c o i n c i -dental are i t a l i c i z e d . Quadrat Groups CO o. g I n t e r t i d a l F o l i o s e red m Laminarian £ Turf Deep water & Sand R S T U V W X Y Z 0 39 13 12 16 6 10 11 10 59 9 28 26 19 22 9 10 11 12 15 13 17 30 17 14 13 20 14 15 7 13 6 8 19 25 16 19 24 10 15 17 6 9 7 5 43 29 22 0 4 30 6 15 26 18 12 12 17 55 0 17 64 sandy bottom, and deep water regions as w e l l as the r e s t of the upper s u b t i d a l . The laminarian group shows the highest coincidence with quadrat group S i n both analyses, but once again i t shows moderate to high l e v e l s of coincidence with a l l other quadrat groups, the lowest being with the i n t e r t i d a l groups i n both analyses. The group of turf-forming species shows highest coincidence with quadrat groups X and U i n both analyses. According to t h e i r l o c a tions on the c l u s t e r analysis dendrograms, groups U and X are t r a n s i t i o n a l between the shallower s u b t i d a l quadrats and the deep water quadrats. This species group shows moderate coincidences with a l l other quadrat groups except groups R and S i n the unweighted analysis and group Y' i n the weighted a n a l y s i s . In general, the l a s t three species groups discussed show c o i n c i -dences with quadrat groups of increasing depth, the f o l i o s e red community occurring i n the shallowest water and the turf community i n the deepest water, with the laminarian community i n between. The l o c a t i o n of the quadrats common to both the weighted and unweighted analysis c o i n c i -dental quadrat/species groups i s shown i n Figure 12 for a l l three s i t e s . Constancy i s the percentage occurrence of a species within a group of quadrats representing a s i n g l e a s s o c i a t i o n . Plants occurring i n at l e a s t 80% of the quadrats are commonly c a l l e d constants (Daubenmire, 1968). In phytosociology, f i d e l i t y i s a measure of the degree to which a species i s r e s t r i c t e d to a p a r t i c u l a r community. In general, a subjective 5-degree s c a l e has been used to express the degree of r e s t r i c t i o n . Species which are preferents ( i . e . , present i n several communities but predominant 65 Figure 12. Quadrats common to both, the weighted and unweighted analysis c o i n c i d e n t a l quadrat/species groups at BI01, BI02, and SI02. SI02Y-005 010 015 020 025 030 035 040 045 050 055 060 065 070 075 080 065 090 095 100 Depth (In feat) p meon tide level 66 67 68 69' 70 7, 75 76 30 40 50 160 70 80 .90 .100 Bl CH Y-005 010 015 020 025 030 035 040 045 050 055 060 065 070 075 080 085 090 095 100 X-063 Depth (In feet) 0 mean tide level Substrate -— Continuous rocky bottom 49 Discontinuous rocky bottom • Shell tt Sand Sea urchins Number •toy Grazed Attenuation 10 20 30 em 6 "6 15 I Q I .0 JO .20 .30 10 Contnunlbiv L«»«ii\wia/i Q D-e«.p wjwtcr 67 i n only one), s e l e c t i v e s ( i . e . , r a r e l y found outside t h e i r preferred community), and exclusives are termed the c h a r a c t e r i s t i c species of a 2 community (Oosting, 1958). Except for the use of a X test to deter-mine the preference of two species for the same or d i f f e r e n t communities (Greig-Smith, 1964), the concept of f i d e l i t y has not been q u a n t i f i e d . I n . t h i s study, a species i s considered c h a r a c t e r i s t i c of a p a r t i c u l a r community i f 33% of i t s occurrences are r e s t r i c t e d to quadrats contained i n that community. The constancy and f i d e l i t y of the species occurring i n t h e i r respective quadrat groups were examined, and the r e s u l t s are given i n Tables 4 and 5 f o r the weighted and unweighted analyses, r e s p e c t i v e l y . Only the r e s u l t s of the weighted analysis w i l l be discussed, however, since a l l the species are included only i n t h i s analysis and not i n the unweighted a n a l y s i s . In general, the i n t e r t i d a l and deep water species have high f i d e l -i t i e s but low constancies f o r t h e i r respective communities. This l a t t e r observation i s accounted for by the f a c t that the two communities are at the p h y s i c a l l i m i t s of marine a l g a l d i s t r i b u t i o n s and r e l a t i v e l y few species occur i n any s i n g l e quadrat. C h a r a c t e r i s t i c species of the i n t e r t i d a l zone are Gigartina papillata, Fucus distichus, Analipus japonicus, Enteromorpha intestinalis, Rhodomela larix, Polysiphonia hendryi v a r . gardneri, Leathesia difformis, Cryptosiphonia woodii, Pterosiphonia bipinnata, and Microcladia borealis. Species c h a r a c t e r i z i n g deep waters by showing high f i d e l i t i e s to group V are Platysiphonia clevelandii, Stenogramme interrupta, Heterosiphonia densiuscula, Callo-phyllis haenophylla, Botryocladia pseudodichotoma, and Agarum fimbriatum. By incl u d i n g group W i n the deep water a s s o c i a t i o n , Bonnemaisonia 68 Table. 4. Constant and c h a r a c t e r i s t i c species (enclosed i n boxes) of the s i x benthic a l g a l communities defined by selected quadrat groups for the weighted c l u s t e r and inverse analyses r e s u l t s . F i d e l i t y Constancy I n t e r t i d a l species Z' z' Analipus japonicus 100 20 Gigartina papillata 95 76 Fucus distichus 95 72 Enteromorpha intestinalis 90 36 Rhodomela larix 90 36 Polysiphonia hendryi v a r . gardneri 71 40 Leathesia difformis 65 52 Cryptosiphonia woodii 60 12 Pterosiphonia bipinnata 56 36 Microcladia boreal is 46 64 Rhodoglossum roseum 30 32 Sargassum muticum 26 36 Lomentaria 23 12 Bossiella 18 48 Odonthalia floccosa 16 36 Kallymenia oblong ifructa 11 4 Liththamnion 9 24 Deep water species V W V* W V + W Platysiphonia clevelandii 75 25 30 6 14 Stenogramme interrupta 60 0 30 0 11 Heterosiphonia densiuscula 57 29 80 22 43 Callophyllis haenophylla 50 17 60 11 29 Botryocladia pseudodichotoma 50 10 50 6 21 Agarum fimbriatum 43 39 100 50 68 Bonnemaisonia nootkana 26 22 60 28 39 Weeksia fryeana 25 25 20 11 14 Sphacelaria norrisii 25 25 10 6 7 Platythamnion pectinatum 24 29 100 66 79 Griffithsia pacifica 23 39 30 28 29 Pleonosporium Vancouverianum 22 33 40 33 36 Phycodrys 14 14 10 6 7 Peyssonelia pacifica 10 22 80 100 93 Sand Y' Y' Zostera marina v a r . l a t i f o l i a 100 100 Agardhiella tenera 45 100 Gracilaria verrucosa 33 80 Ceramium californicum 15 60 Platythamnion villosum 11 80 Polysiphonia urceolata 0 0 69 Table 4. Continued F i d e l i t y Constancy F o l i o s e red R' R' Laurencia spectabilis 17 100 Prionitis lanceolata 24 93 Constantinea subulifera 22 93 Corallina officinalis var. chilensis 21 93 Herposiphonia plumula 14 93 Cryptopleura ruprechtiana 25 87 Ceramium washingtoniensis 19 73 Callophyllis flabellulata 16 73 Plocamium coccineum var. pacificum 15 73 Calliarthron tuberculosum 25 60 Iridaea cordata var. cordata 19 53 Callophyllis linearis 24 47 Desmarestia viridis 18 47 Microcladia coulteri 45 45 Nitophyllum mirabile 25 20 Laminarian S' Sf Gelidium 15 100 Amplisiphonia pacifica 13 100 Pterosiphonia dendroidea 12 100 Nereocystis luetkeana 30 92 Monostroma fuscum 15 92 Costaria costata 29 58 Laminaria 15 58 Gigartina exasperata 42 42 Hollenbergia subulata 19 42 Turf U' X' U' X' U' + J Antithamnion defectum 15 19 100 77 86 Pterosiphonia gracilis 10 17 87 92 90 Polvneura latissima 18 26 87 92 81 Rhodymenia pertusa 23 10 87 23 48 Nienburgia 22 26 63 46 52 Gymnogongrus leptophyllus 19 15 63 31 43 Ralfsia fungiformis 8 15 50 54 52 Ulva 15 4 50 8 24 Schizymenia pacifica 17 42 25 38 33 Polysiphonia pacifica 0 33 0 69 43 Grateloupia doryphora 0 20 0 15 10 Branchioglossum woodii 0 12 0 15 10 Ahnfeltia plicata 0 25 0 8 5 Rhodoptilum plumosum 0 4 0 8 5 70 Table 5. Constant and c h a r a c t e r i s t i c species (enclosed i n boxes) of the s i x benthic a l g a l communities defined by selected quadrat groups for the unweighted c l u s t e r and inverse analyses r e s u l t s . F i d e l i t y Constancy I n t e r t i d a l species Z Z Gigartina papillata 90 86 Fucus distichus 89 81 Analipus japonicus 60 14 Rhodomela larix 60 29 Polysiphonia hendryi var. gardneri 57 38 Leathesia difformis 55 52 Cryptosiphonia woodii 40 10 Microcladia borealis 34 57 Pterosiphonia bipinnata 31 24 Rhodoglossum roseum 19 24 Lomentaria 15 10 Sargassum muticum 15 24 Bossiella 14 43 Odonthalia floccosa 9 24 Lithothamnion 4 14 Kallymenia oblongifructa 0 0 Deep water species V W V W V + w Platysiphonia clevelandii 75 25 25 7 15 Heterosiphonia densiuscula 64 14 75 13 41 Stenogramme interrupta 60 0 25 0 11 Agarum fimbriatum 52 26 100 40 67 Sphacelaria norrisii 50 0 17 0 7 Callophyllis haenophylla 50 10 50 7 26 Botryocladia pseudodichotoma 40 10 33 7 19 Weeksia frueana 38 12 25 7 15 Platythamnion pectinatum 27 24 92 67 78 Bonnemaisonia nootkana 26 22 50 33 41 PIeonosporiurn vancouverianum 22 28 33 33 33 Griffithsia pacifica 15 38 17 33 26 Phycodrys 14 14 8 7 7 Peyssonelia pacifica 12 19 83 100 93 Sand Y Y Zostera marina v a r . l a t i f o l i a 100 100 Agardhiella tenera 45 100 Gracilaria verrucosa 33 80 Gigartina exasperata 17 40 Ceramium californicum 15 60 Platythamnion villosum 11 80 Hollenbergia subulata 4 20 Polysiphonia urceolata 0 0 71 Table 5. Continued. F i d e l i t y Constancy F o l i o s e red R R Constantinea subulifera 12 100 Prionitis lanceolata 14 100 Plocamium coccineum v a r . pacificum 11 100 Ceramium washingtoniensis 14 100 Herposiphonia plumula 7 88 Cryptopleura ruprechtiana 13 88 Iridaea cordata v a r . cordata 14 75 Callophyllis linearis 14 50 Callophyllis flabellulata 3 25 Laminarian S S Pterosiphonia dendroidea 19 100 Laurencia spectabilis 22 100 Amplisiphonia pacifica 20 95 Gelidium 22 89 Calliarthron tuberculosum 44 84 Corallina officinalis v a r . chilensis 24 84 Monostroma fuseurn 19 74 Nereocystis luetkeana 32 63 Laminaria 24 58 Costaria costata 33 42 Turf U X U X U + X Antithamnion defectum 20 17 85 82 83 Pterosiphonia gracilis 15 15 85 100 92 Rhodymenia pertusa 35 6 85 18 54 Polyneura latissima 23 23 69 82 75 Nienburgia 35 22 62 45 54 Viva 22 4 46 9 29 72 nootkana, Weeksia fryeana, Sphacelaria norrisii, Platythamnion pectinatum, Griffithsia pacifica, and Pleonosporium Vancouverianum are added to the l i s t of c h a r a c t e r i s t i c species of deep water. The quadrats occurring on sand contain species of both high con-stancy and f i d e l i t y . These species are Zostera marina, Agardhiella tenera, and Gracilaria verrucosa. The upper to mid s u b t i d a l quadrats contain species of rather low f i d e l i t i e s to t h e i r respective quadrat groups because of the wide d i s t r i b u t i o n s of most of the species. However, these species generally show high constancies. Constant species occurring i n the f o l i o s e red a l g a l a s s o c i a t i o n are Laurencia spectabilis, Prionitis lanceolata, Constantinea subulifera, Corallina officinalis v a r . chilensis, Herpo-siphonia plumula, and Cryptopleura ruprechtiana. Microcladia coulteri i s the only species considered f a i t h f u l to t h i s community by the present c r i t e r i o n . I t occurs only as an epiphyte on several of the species i n t h i s community. Constant species occurring i n the laminarian a s s o c i a t i o n are Gelidium, Amplisiphonia pacifica, Pterosiphonia dendroidea, Nereocystis luetkeana, and Monostroma fuscum. Only Gigartina exasperata has a f i d e l i t y greater than 33%. Constant species occurring i n the t u r f a s s o c i a t i o n include only Antithamnion defectum, Pterosiphonia gracilis, and Polyneura latissima when both quadrat groups U' and X; are considered. The f i d e l i t i e s of only Schizymenia pacifica and Polysiphonia pacifica to group X' equal or exceed 33%. Because there are more quadrat groups than species groups, a one-to-one correspondence between the two groups would leave a surplus of three 73 quadrat groups. I t has already been noted that group W contains quadrats with many deep water species and that groups U and X are both composed of turf-containing quadrats. The s i n g l e remaining quadrat group (T) contains quadrats with species from a l l s i t e s but with the l a r g e s t percentage of species from the sandy bottom a s s o c i a t i o n . However, the overlapping species are almost e x c l u s i v e l y those which were poorly associated with the c h a r a c t e r i s t i c sandy bottom species of the inverse a n a l y s i s . Hence group T i s best considered without important a f f i n i t i e s to a p a r t i c u l a r species group. The quadrat group containing quadrats not associated with another s i t e groups has a preponderance of deep water quadrats. Because of the paucity of vegetation at the lower l i m i t s of a l g a l d i s t r i b u t i o n s , quadrats from these depths contain fewer species and lower biomassses than any other quadrats. This dearth of species of necessity r e s u l t s i n a poor f i t with other quadrats i n the c l u s t e r a n a l y s i s . Zurich-Montpellier analysis The r e s u l t s of the Zurich-Montpellier analysis using the 50-20 d i a g -n o s t i c species formula appear i n Table 6. Thirteen diagnostic species groups were extracted from the presence-absence data. Although several of the species groups have equivalents i n the other analyses,''" none of ^Antithamnion defectum, Lithothamnion, Peyssonelia pacifica, and Pterosiphonia gracilis i n the t u r f community of inverse a n a l y s i s ; Callo-phyllis linearis, Cryptopleura ruprechtiana, Iridaea cordata v a r . cordata, Odonthalia floccosa, and Prionitis lanceolata i n the f o l i o s e red community of inverse a n a l y s i s ; Calliarthron tuberculosum, Desmarestia viridis, and Nereocystis luetkeana i n the implied time-element community of the species c o n s t e l l a t i o n diagram; Microcladia borealis and Sargassum muticum i n the lowest i n t e r t i d a l community of a s s o c i a t i o n a n a l y s i s ; Agarum fimbriatum, Bonnemaisonia nootkana, and Rhodoptilum plumosum and Botryocladia pseudo-dichtomoa, Phycodrys, and Stenogramme interrupta i n the deep water community 74 T a b l e 6. S p e c l e B - q u a d t a t t a b l e of the Z u r i c h - M o n t p e l l i e r a n a l y s i s u s i n g the 50-20 d i a g n o s t i c s p e c i e s f o r m u l a . A n t i d e L i t h Peyspa P t e r g r Cal111 C r y p r u I r i d c o c O d o n f l P r i o l a C a l l f l C o r a o f c G e l i P l o c c o p Ccrawa Monofu Boss Conssu P l a t p e P o l y l a C a L l t u Dcsravi M i ctbo Sargrou Gyranle HOIIBU A g a r f i Bonnno Rhodpl E n t e i n F u c u d i e Gigapa L e a t d i P o l y h e g N i t o m i S c h i p a B o t r p s Phyc S t e n i n A g a r t e A n n f p i Amnio* Ana1J a Branwo C a l l h a C e r a c a C o s t c o Crypwo Cigaax Grncve G r a t c a G r t f p a H e r p r i Hetode K u l l o b Loral Loursp Lomesp M i c r c o N i e n P l a t c l P l a t v i P l e o v a P o l y p a z P o l y u r u P t e r b i P t e r d a R a l f f u Rhodla Rhodper Rhodro Sphaca U l v a Ueek.fr Z o s t n a l CO s n 3 5 •3- 5 o cr. vO 3 -a vD o O co DO to oo co CO co ON o> o o o o rl co rl o 2 \0 ri co rl CO S P; s s eo CO co w o 4- + + + + + + + + _ + __ + + + 4- + 4- + + + + + + + 4- + 4- 4- 4 + 4- + + + + + + + + + + + + + + + + + + + + 4- 4- 4- + 4- 4- + + + 4- + + + 4 + 4- 4- + + 4 4- 4- 4- 1-4-+ + 4-+ + + + + + + + + + + + + + + + + + 4- 4- + 4- + 4- + 4 + 4- + 4- 4- + + + 4- 4- + 4- 4- 4- + 4- 4- 4- 4-4-+ 4- 4- 4- 4-+ + + + + + + + + + ± + ±_ + + + + + + + t + + 4 + t + 4 . 4- + 4 '+ + + + + ,. + 4- + 4 4- + t i t + + "T" + + +- + + + + + + + 4- +-"+ 4- + 4- 4 4-f + + + + + + + + + + + + + + + + + + + + + + + + 4- + + 4- 4- + + + + + 4- 4- + + + + + + + 4-f + + + + + + + + t + + + + + + + + + + + + + + + + + 4- + + + 4- + + + + 4- + + 4 f + + . + + + + + + + + + + + + +• + + + + + + + + + t + + + + + + 4- 4- 4- + 4- + 4- + 4- + . + + + *• + + + + + + + + + + + + + + + + + + t + + + + + 4 4 f + t + + 4 4 + T + 4- •t- 4* 4-f + -=f— + + + 4- + + •F-4- 4- 4 4 4- + 4- 4 4- 4- 4- 4- 4-4- 4- + + + + + 4 4- 4-+ + + + + + + + + + + + + + + + + + + + + + 4 + + + + + + + + + + 4- + + + + + 4- + + + + + + + 4- + 4- 4- + + + + + + + + + + + + + + + + + + + + + + + + + + + 4 4- + 4- + + + 4- + + + + + + + + + + 4 + + + 4- + 4- 4- 4* + 4 4" 4-+ + + + .+ + + + + + + + + + + + t + + + + t t + + + + + + 4- 4 + + + + t t +  4 + •1- t t + + 4- 4 + 4 t + 4-f — T + t + -r+ + + 4- + + t + 4- + + + + + + + + + + + + 4 4-+ + 4- + 4-+ + 4 4- + t 4-+ + 4-4-t 4 + 4-4-+ + + 4- t 4- 4 f t + 4 t... + + t + 4 t 4- + 4 + t t + +• + + + + T + + + + + + + • + 4- + + + + + + 4- 4 4- 4- 4- 4- + + 4- 4- 4- + 4- + + + + + 4- + + + T T 4 4- 4 + + 4- + t t + t t t + + + t + + + + + t + + + + + + + + + t + + + + + 4- + 4-. +" + + 4 + 4- 4-± — - -re— -1 t t O — O CM »0 r- O O 4- + 4- 4 + + + + 4 + + 4-4- + 4 - 4 - 4 - + 4 + + + 4 - 4 - 4 - + 4-4- + + + 4- + 4 -+ + + + + + -f + + + + ± ± _ + + + + + + + 4- -4-4- + + +| + + 4 + 4- L + + + + + + + + + + + _t + + + + + if + ± + + + + + + + + + + + t + "7 + T + + + ±. T + + + -±. + + + T T T ±. ±_±. * + * + + + + + + + + + + _ t _ t _ •t ± . ± - i . T T T T T T ^ + + + + +• + ± ± ± ± t t t T T i T T 1 + + + ±. ±—± + + + + + + + ±_ T + + + + + +| H 3 + +.—+ + + + + + + + + + + + + + + * + + + + + + ± ±. i. T + + + + T T" ± _ _ T T T T T =T + + + ± + + + • * * * + + + + + + + + + + + + + + + + + ±-A + + + + + + + * * * + + + + + + + + + T T T + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + t + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + t t + + + + + + + + + + + + + + + + + + + + + t + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + * + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + t- + + + + + + + + + + + + + + + + + + + + + + 75 these groups appears to be represented by a unique quadrat group i n t h i s a n a l y s i s . In general, a few species form large quadrat groups when the 50-20 l i m i t s are used. With more stringent group admittance, fewer species would form smaller quadrat groups. With more lax group admittance, more species would form larger quadrat groups. Because of the d i f f e r e n t l e v e l s of species constancy and f i d e l i t y i n the d i f f e r e n t communities, no s i n g l e diagnostic species formula can s a t i s f a c t o r i l y d e l i m i t the same communities determined by the other analyses. The Zurich-Montpellier analysis was designed to extract species-quadrat groupings, but the present data were not c o l l e c t e d such that the analysis was capable of producing the same r e s u l t s as the other methods of a n a l y s i s . Because t h i s method requires c o l l e c t i o n of t y p i c a l quadrats, the sand and/or number of quadrats c o l l e c t e d i n the i n t e r t i d a l and deep s u b t i d a l probably need to be increased and those i n the upper and mid s u b t i d a l decreased. This problem was compensated f o r i n the c l u s t e r a n a l y s i s , f or example, by choosing lower s i m i l a r i t y l e v e l s f o r d e l i m i t i n g the communities at environmental extremes and higher l e v e l s for determining the communities i n between. A l s o , c o l l e c t i o n of more representative quadrats would enhance the c l a r i t y of the analysis r e s u l t s . This method does not appear to be generally suited f o r analysis of s u b t i d a l seaweed communities. of inverse analysis and species c o n s t e l l a t i o n diagram; and Enteromorpha intestinalis, Fucus distichus, Gigartina papillata, Leathesia difformis, and Polysiphonia hendryi v a r . gardneri i n the i n t e r t i d a l communities of the various analyses. 76 Bray and C u r t i s o r d i n a t i o n The Bray and C u r t i s o r d i n a t i o n of a l l 124 quadrats appears i n Figure 13. Quadrat 23 had the highest sum of interquadrat distances, and i t was used as one end point of the f i r s t axis X. Several quadrats had a maxi-mum distance of 1.00 from t h i s quadrat. Quadrat 48 was chosen as the second end point among these quadrats because i t showed higher s i m i l a r i t i e s to some of the other quadrats occurring at near maximum distances from quadrat 23 than d i d other quadrats i n that group. The second axis Y was chosen from among those quadrats which occurred close together on the f i r s t axis but which were nevertheless separated by large interquadrat d i s t a n c e s . Quadrats 5 and 83 were chosen as end points of the second a x i s . They occurred at .5101 and .5100, r e s p e c t i v e l y , along the X axis with an a c t u a l interquadrat distance of .9999. A t h i r d axis Z was extracted following the same procedure as f o r the previous axes. However, when the quadrats were plotted along this a x i s , most quadrats occurred along the X and Y axes, the end points of the Z axis (12 and 131) apparently representing quadrat extremes with very few quadrats showing any s i m i l a r i t y to e i t h e r end p o i n t . Therefore, only the ord i n a t i o n along the X and Y axes appears i n Figure 13. Both the X and Y axes of the ordi n a t i o n tend to separate the quadrats on the basis of depth. The end points of the X axis represent the lower i n t e r t i d a l and the moderate to deeper s u b t i d a l—the mid i n t e r t i d a l , shallow to mid s u b l i t t o r a l , and the deepest s u b t i d a l quadrats tending to c l u s t e r near the center of the graph. The end points of the Y axis represent the upper s u b t i d a l and the deepest s u b t i d a l—the mid to lower s u b t i d a l quadrats tending toward the center of the axes. 77 Figure 13. The l o c a t i o n of the 124 quadrats along the X and Y axes the Bray and C u r t i s o r d i n a t i o n . 78 Y io a 0 iue .34 -is-0 3 6 O « 3 0° ° I 0 9 Ol.5 7 5 ^ 1 3 9Sf" ~ « 47 O OI 0 9 0 ,101 0 9 0 2 9 09265 0 " 2 0 5 8 0 6 2 C * 6 n 0 8 9 „ O l 2 2 „ O 0120 OlW O SO 46 73 0 , s 0 «OO44 B 0 Q53 „ ° ° 7 47 O ° « Oin 0 3 3 O 4 0 0 4 3 0 0 9 8 O " „ „ OO 96 0 4 2 ° O l 2 6 TC0» 0 8 0 Q 3 3 71 o 0 6 4 0 'X>'28 0 9 7 O 4 9 0119 16 -118 O ° OS2 OI7 0 , 9 O T O 0=0 £ • « > • 0 " 6 Qr7 O122 OT2 Ck ° 8 6 0 44 0 4 5 0 5 4 O " 0 9 4 OO O 48 100 79 Because of the great v a r i a t i o n i n species composition along the transect, no s i n g l e axis can be used to represent depth. Because many of the i n t e r t i d a l and deeper s u b t i d a l quadrats represent extremes i n species composition, a l l quadrats but those occurring on sandy bottoms with a sum of interquadrat distances greater than 111 ( a r b i t r a r i l y chosen because most species-poor i n t e r t i d a l and deeper s u b t i d a l quadrats had greater t o t a l interquadrat distances) were eliminated from consider-a t i o n . The new Bray and C u r t i s o r d i n a t i o n of these 83 quadrats appears i n Figure 14. In this o r d i n a t i o n the end points of the X and Y axes are 54 and 121, and 43 and 85, r e s p e c t i v e l y . The end points of the f i r s t axis represent the mid to deeper s u b t i d a l zone and the shallow s u b t i d a l r e g i o n , r e s p e c t i v e l y , and the end points of the second a x i s , the mid s u b t i d a l and the deepest s u b t i d a l areas, respec-t i v e l y . No obvious quadrat groups are delimited by e i t h e r o r d i n a t i o n , thereby supporting the concept of continuous and overlapping d i s t r i b u t i o n s of the d i f f e r e n t species. 80 Figure 14. The l o c a t i o n of a selected group of 83 quadrats along the X and Y axes of the Bray and C u r t i s o r d i n a t i o n . 81 OI07 075 091 074 092 083 037 OCX ° ° 0 76 090 041 OI23 094 0129 O40 039 0 1 7 C 8 6 o87 069 0 6 S Q o 0 0 5 8 OK) ° 2 7 OI9 OI8 Ol28 OM8 02' 09 y J l i a CHO On OS 032 <5T3 °> CB9 050 CUB OI22 06 O'buo 0 080 0 7 ! 0 5 7 044 OI02 Q43 °»< 098 „ ° < 3 OSS 0 8 2 049 43 O I DISCUSSION C o n t r o l l i n g environmental f a c t o r s As argued by Lewis (1961) and others, the use of benthic marine organisms to characterize shore communities i s a necessity because of our i n a b i l i t y to measure a l l of the p h y s i c a l parameters which may be responsible f o r the d i s t r i b u t i o n of the sp e c i e s . This f a c t was e a r l i e r recognized by t e r r e s t r i a l plant e c o l o g i s t s (Goodall, 1954) who have since developed elaborate mathematical methods to describe communities by t h e i r species composition i n such a way that the major environmental factors determining community composition become obvious. Most previous attempts to apply community analysis techniques to benthic marine organisms have dealt p r i m a r i l y , i f not e x c l u s i v e l y , with animals. Cassie and Michael (1968) were among the f i r s t marine ecologists to apply numerical methods to benthic organisms. They subjected twelve species of i n t e r t i d a l animals from s o f t bottoms to a p r i n c i p a l com-ponents a n a l y s i s . The two communities separated by the f i r s t axis were found to be p o s i t i v e l y correlated with sediment type. F i e l d (1971), using a c l u s t e r a n a l y s i s , found species assemblages associated with p a r t i c u l a r depths and bottom types, as did L i e and Kell e y (1970), using a method of factor a n a l y s i s , and Hughes, Peer and Mann (1972), employing both c l u s t e r analysis and p r i n c i p a l components a n a l y s i s . Nichols (1970) also found bottom type and depth to be the important determinants of benthic polychaete assemblages but s p e c i f i e d clay content as the dominant factor i n bottom type and subjugated depth to secondary importance, as i t i s r e l a t e d to deposition of s i l t s and clays i n deeper, quieter waters. 83 Other workers have r e l i e d only on the species composition and general f i e l d observations to suggest the determining environmental v a r i a b l e s . Hughes and Thomas (1971a) suggested that d i f f e r e n c e s i n s a l i n i t y , the proportion of s i l t and organic matter i n the sediment, the mean p a r t i c l e s i z e , and p o s s i b l y the e f f e c t s of scouring by i c e i n winter accounted f o r community differences along four transects i n the Bideford River estuary, northwest of Prince Edward I s l a n d . In Bedeque Bay, P. E. I . , they (1971b) found s a l i n i t y and bottom type to be the probable determinants of community d i f f e r e n c e s . In the present study, only selected environmental factors were measured. Factors thought to be most important were depth and substrate, and these were recorded f o r each quadrat sampled, along with substrate attenuation and o r i e n t a t i o n . Presence of or recent grazing by sea urc h i n s , the dominant herbivore, was also noted. No consistent temperature or s a l i n i t y measurements were taken at the study s i t e s , but data for the area are a v i l a b l e from surface stations at P o r l i e r Pass, s i x miles southeast of S i l v a Bay, and Entrance I s l a n d , s i x miles to the northwest. D a i l y recordings for 1971 are found i n H o l l i s t e r (1972). Mean monthly temperatures and s a l i n i t i e s from 1914 to 1970 are reported i n H o l l i s t e r and Sandnes (1972) . Diving observations and monthly temperature readings at d i f f e r e n t depths with a Yellow Springs telethermometer confirmed, with rare exception, that the temper-ature and s a l i n i t y of the bottom water i n the study area were quite close to that of the surface. Temperature maxima occur i n July and August, reaching about 69° F at Entrance Island and 63° F at P o r l i e r Pass, i n 1971; minima of about 43° F occur at both s t a t i o n s from December to March. Monthly mean temperatures range from 44.1° F i n January and 84 February "at P o r l i e r Pass and Entrance I s l a n d , r e s p e c t i v e l y , to 59.0° F i n J u l y and 62.6° F i n J u l y and August at the two s t a t i o n s . Surface s a l i n i t y attained a minimum value of about 17.5%« at Entrance Island i n June 1971 (data for that month are missing for P o r l i e r Pass) and a maxi-mum of 29%. i n January to A p r i l at both s t a t i o n s . Monthly means at Entrance Island range from 22.75%, i n J u l y to 28.3% i n March. Monthly means are not a v a i l a b l e for P o r l i e r Pass. No temperature or s a l i n i t y extremes over the period of observation are given. Although many workers suggest that i t i s these extremes, rather than monthly or yearly averages, which are the d e c i s i v e c o n t r o l l i n g factors of plant d i s t r i b u t i o n s (Gessner, 1970), i t i s u n l i k e l y that the l o c a l d i f f e r e n c e s i n d i s t r i b u t i o n s could be a t t r i b u t e d to these extremes. When control of s u b t i d a l plant d i s t r i b u t i o n s i s a t t r i b u t e d to depth, i t i s assumed that l i g h t i s the depth-related factor of importance to the algae. Preliminary measurements of incident l i g h t of d i f f e r e n t wavelengths were taken at d i f f e r e n t depths during the summer of 1972. These measure-ments were continued on a more regular basis i n the summer of 1973 (Foreman, personal communication), but as yet no data have been published on the r o l e of d i f f e r e n t wavelengths and quantities of l i g h t i n determining s u b t i d a l d i s t r i b u t i o n s of p a r t i c u l a r species or communities of algae. No current measurements were made. The major c o n t r o l l i n g factor of community d i s t r i b u t i o n s i n t h i s study appears to be due to the emergence-submergence phenomenon associated with the i n t e r t i d a l zone. The f i r s t major break i n the a s s o c i a t i o n analysis occurred on a d i s t i n c t l y i n t e r t i d a l species as did both the c l u s t e r analysis of quadrats and the inverse analysis of species. 85 No completely d e f i n i t i v e studies have yet been c a r r i e d out to show what main factors underlie i n t e r t i d a l species d i s t r i b u t i o n s . Doty (1946) and others have supported the emergence hypothesis of zonation, but this " c a t c h - a l l " phrase f a i l s to d i s t i n g u i s h the p a r t i c u l a r p h y s i c a l phenomenon responsible f o r the e f f e c t s of emergence. A d i f f e r e n c e i n d e s i c c a t i o n tolerances i s considered the p r i n c i p a l reason for i n t e r t i d a l zonation, and the osmotic gradient, r a d i a t i o n and temperature are the p h y s i c a l factors a f f e c t i n g d e s i c c a t i o n tolerance. In general, tolerance to d e s i c c a t i o n i s d i r e c t l y r e l a t e d to height of occurrence on the shore. I n t e r t i d a l species generally show greater tolerances than s u b t d i a l algae to hyperosmotic conditions (Gessner & Schramm, 1971), to d i r e c t sunlight (Hellebust, 1970), and to temperature extremes, some even being able to withstand freezing (Gessner, 1970). However, tolerance to d e s i c c a t i o n and other factors may r e l a t e l e s s to i n d i v i d u a l tolerance than to a t o t a l "population tolerance." A whole population may so modify the microclimate of i t s own habitat that i t can grow and reproduce under conditions more severe than i t could normally t o l e r a t e . L i g h t a v a i l a b i l i t y as represented by depth of occurrence of the algae i s a second major c o n t r o l l i n g factor of the composition of benthic marine a l g a l communities. A l l analyses but the Bray and C u r t i s o r d i n a t i o n separated s u b t i d a l communities on the general basis of depth. Although there i s l i t t l e doubt that l i g h t i s the dominant environ-mental factor determining the v e r t i c a l d i s t r i b u t i o n s of s u b t i d a l benthic marine algae, there i s s t i l l an incomplete understanding of j u s t how l i g h t determines d i s t r i b u t i o n s . The early controversy between Englemann and Gaidukov, s t r e s s i n g q u a l i t y of l i g h t , and Berthold and Oltmanns, emphasizing quantity, has not been resolved (Hellebust, 1970). Most 86 modern marine e c o l o g i s t s conceed that both quantity and q u a l i t y of l i g h t are important. D i f f e r e n t seaweeds show d i f f e r e n t compensation and l i g h t saturation responses which generally r e f l e c t t h e i r v e r t i c a l p o s i -t i o n on the shore (Mathiesen & Burrs ,1971). Growth rates and repro-duction periods also respond v a r i o u s l y to l i g h t . However, the d i f f e r e n t p h y s i o l o g i c a l responses of the algae to l i g h t can only be resolved by autoecological studies of selected species. Substrate type represents a t h i r d major f a c t o r c o n t r o l l i n g community composition. The e e l grass bed occurring on sand was separated out by a l l of the methods of analysis but the Zurich-Montpellier analysis and the Bray and C u r t i s o r d i n a t i o n . Although most marine plants are i n d i f f e r e n t to the chemical pro-p e r t i e s of the substrate, the p h y s i c a l properties determine whether a plant can s e t t l e , grow and reproduce. Degree of d i s p e r s i o n , texture, hardness, color ( i n the case of deep-water i n h a b i t i n g s p e c i e s ) , and water-r e t a i n i n g capacity (important to i n t e r t i d a l plants) are s i g n i f i c a n t p h y s i c a l f e a t u r e s . S t a b i l i t y of the substrate and topographical factors are also important: a plant on a v e r t i c a l w a l l must absorb the f u l l impact of the waveshock, while a h o r i z o n t a l plant i s merely washed (den Hartog, 1972). Comparison and evaluation of a n a l y t i c a l methods Both the a s s o c i a t i o n analysis and the c l u s t e r analysis roughly di v i d e the quadrats into four groups. In both cases, the i n t e r t i d a l quadrats are the f i r s t to be distinguished from the r e s t . The second order of d i v i s i o n separates the quadrats into an upper s u b t i d a l and a lower s u b t i d a l group. Quadrats occurring on sandy bottoms are also 87 delimited i n both, analyses, but t h e i r a f f i n i t i e s to the shallower or deeper subgroups v a r i e s between as we l l as w i t h i n the analyses. The early subdivisions of the two s u b t i d a l groups of the a s s o c i a t i o n analysis serve to p u r i f y these groups of the quadrats which more properly belong i n the other group. A f t e r these quadrats have been removed, a comparison of a s s o c i a t i o n analysis and c l u s t e r analysis shows that 14 lower s u b t i d a l and one i n t e r t i d a l quadrat from the a s s o c i a t i o n analysis have been placed i n the upper s u b t i d a l group of the weighted c l u s t e r a n a l y s i s , and three upper s u b t i d a l and one lower s u b t i d a l quadrat (wrongly placed because i t contained no Gigartina papillata or Odonthalia floccosa) occur i n the i n t e r t i d a l group of the weighted c l u s t e r a n a l y s i s . Only four lower s u b t i d a l quadrats from the a s s o c i a t i o n analysis were placed i n the unweighted c l u s t e r analysis upper s u b t i d a l group, and two upper s u b t i d a l and one i n t e r t i d a l quadrat were placed in-the lower s u b t i d a l group. The i n t e r t i d a l group from the unweighted c l u s t e r analysis con-tained only one lower s u b t i d a l quadrat (the one lac k i n g both Gigartina and Odonthalia) from the a s s o c i a t i o n a n a l y s i s . Both the a s s o c i a t i o n analysis and the species/quadrat coincidence tables of c l u s t e r analysis d e l i m i t c h a r a c t e r i s t i c species. However, few of the species groups are comparable. The a s s o c i a t i o n analysis tends to produce small groupings of species not unique to those quadrats, whereas the c l u s t e r analysis coincidence tables determine l a r g e r groups of species that seem to predominate i n p a r t i c u l a r quadrats. I n comparing the two h i e r a r c h i c a l methods of c l a s s i f i c a t i o n , Stephenson et a l . (1970) argued that a s s o c i a t i o n analysis tends to pro-duce more cle a r - c u t community d i s t i n c t i o n s than c l u s t e r a n a l y s i s . This occurs because the l a t t e r method aims at reproducing r e l a t i o n s h i p s as 88 accurately as p o s s i b l e , seeking i n t e r n a l l y homogeneous groups i f these e x i s t . In a s s o c i a t i o n a n a l y s i s , c l u s t e r i n g i s imposed on the data and boundaries are a r t i f i c i a l l y sharpened. However, i n t h i s study, the community d i s t i n c t i o n s produced by the c l u s t e r analysis method were more obvious and generally more meaningful than those produced by the a s s o c i a t i o n a n a l y s i s . Comparison of the species c o n s t e l l a t i o n diagram, derived from presence-absence data, and inverse a n a l y s i s , based on q u a n t i t a t i v e data, produces s i x communities of very s i m i l a r species composition. Differences brought out by the inverse a n a l y s i s which do not contradict the data on which the. species c o n s t e l l a t i o n diagram were based could be accomodated i n the species c o n s t e l l a t i o n diagram by (1) adding Peyssonelia pacifica to the deep water community, (2) adding Gymnogongrus leptophyllus, Rhodymenia pertusa, and Rhodoptilum plumosum to the turf-forming commun-i t y , (3) moving Monostroma fuscum and Hollenbergia subulata closer to the other turf-forming species showing a f f i n i t i e s to the laminarians, and (4) moving Bossiella c l o s e r to the i n t e r t i d a l group but s t i l l keeping i t i n the f o l i o s e red community. In comparing c l a s s i f i c a t i o n and ordination techniques, Hughes et a l . (1972) found that p r i n c i p a l components analysis was more s e n s i t i v e to heterogeneity than was c l u s t e r a n a l y s i s . They concluded that i t would be a more e f f e c t i v e method f o r analyzing r e l a t i v e l y homogeneous data f o r e c o l o g i c a l l y s i g n i f i c a n t gradients. From a comparative study of d i f f e r e n t o rdination techniques, Gauch and Whittaker (1972) also concluded that, p r i n c i p a l components ord i n a t i o n i s most use f u l when applied to samples with r e l a t i v e l y small species differences and l i m i t e d the usefulness of any of the methods considered by them (i n c l u d i n g the Bray and C u r t i s 89 ordination) to communities with l e s s than f i v e half-changes i n the beta d i v e r s i t y . The f a i l u r e of a s i n g l e axis of the Bray and C u r t i s ordination i n t h i s study to account for the changes i n depth may be a t t r i b u t e d to the several complete turnovers i n species composition that occur from the mid i n t e r t i d a l to the depth l i m i t s of plant d i s t r i b u t i o n s . Because s i m i -l a r i t y values based on q u a n t i t a t i v e data tend to deemphasize sharp d i f f e r -ences, no d i s t i n c t communities are delimited by the o r d i n a t i o n . The main weakness of the Bray and C u r t i s method has tended to be i n the s e l e c t i o n of the axes (Erman & Helm, 1971). The same weakness was discovered i n t h i s study, the s e l e c t i o n of the end points of the f i r s t axis being p a r t i c u l a r l y d i f f i c u l t because a number of quadrats shared the same maximum distance from the most d i s s i m i l a r quadrat. Erman and Helm (1971) have also c r i t i c i z e d the Bray and C u r t i s method of ordination because axes subsequent to the f i r s t are frequently r e l a t e d to i t ( i . e . , the axes are not orthogonal) and provide l i t t l e new information (as was the case i n t h i s study). However, i t remains uncertain whether other more objective methods produce any more r e a l i s t i c axes. In f a c t , recent authors have also c r i t i c i z e d p r i n c i p a l components analysis and other mathematically sop h i s t i c a t e d o r d i n a t i o n techniques on t h i s b a s i s . The arguments against the use of p r i n c i p a l components analysis a r i s e from the v i o l a t i o n of the two basic assumptions inherent i n f a c t o r a n a l y s i s : (1) that the factors are independent of one another, and (2) that the occurrence of a species i s l i n e a r l y r e l a t e d to the f a c t o r responsible for i t s abundance and d i s t r i b u t i o n (Greig-Smith, 1964). In a recent paper, Beals (1973) summarized the case against p r i n c i p a l components analysis by s t a t i n g 90 . . . I t does not take i n t o account the normal-curve r e l a t i o n s h i p between species success and environ-ment, nor the e c o l o g i c a l ambiguity of species ab-sence i n a stand...Each plant species i n a p a i r of stands responds to the t o t a l environmental d i f f e r e n c e of those two stands; not to factors independent of those to which other species are responding. Both Beals (1973) and Gauch and Whittaker (1972) pref e r a Bray and C u r t i s o r d i n a t i o n because i t represents a more r e a l i s t i c model by defining species-dimensional space i n terms of vegetation changes from point to p o i n t , the reference points representing r e a l d i f f e r e n c e s i n vegetation space. I f the Bray and C u r t i s technique i s the preferred method of o r d i n a t i o n , one can only conclude that i t s a p p l i c a t i o n to the data i n hand was inappropriate because of the several complete changes i n species composition along the t r a n s e c t s . Several workers have obtained s i m i l a r r e s u l t s from employing d i f -ferent methods of a n a l y s i s . Hughes and Thomas (1971a and b) and Hughes e_t aJL. (1972) found that the highest l e v e l of agglomeration i n the c l u s t e r analysis and the f i r s t axis i n the p r i n c i p a l components analysis y i e l d e d s i m i l a r groups of st a t i o n s or s p e c i e s . L i e and K e l l e y (1970) found that the matrix of a f f i n i t i e s between a l l possible p a i r s of s t a t i o n s and Q f a c t o r a n a l y s i s produced s i m i l a r groups of stations although the f i r s t method uses only presence-absence data, and the second method, c o r r e l a t i o n c o e f f i c i e n t s based on the numbers of animals present. Similar groups of species were also generated by two d i f f e r e n t methods, one employing only q u a l i t a t i v e data and the other q u a n t i t a t i v e data, but the agreement was not as close as between the groups of s t a t i o n s . 91 Erman and Helm (1971) found l i t t l e d i f f e r e n c e i n two ordinations using d i f f e r e n t importance values, one emphasizing large but r e l a t i v e l y low density organisms, the other, small but r e l a t i v e l y high density species. The use of biomass obviates the predicament of what measure to use to properly weight each species. However, a standard such as dry weight should be employed when using biomass data i n order to e l i -minate differences i n the a b i l i t i e s of plants to r e t a i n moisture. I t i s e s p e c i a l l y noteworthy that few of the communities examined have a unique species composition which sets them apart from every other community. F i e l d (1971) found that comparison among the species of inverse analysis revealed that many of the species were not c l o s e l y associated i n communities. In a comparison of site-groups and species-groups , Stephenson et: a l . (1970) found no site-groups uniquely defined by the presence of a species-group, d i f f e r e n c e s i n composition being mainly q u a n t i t a t i v e and not q u a l i t a t i v e . The r e s u l t s of these studies along with those of other authors (Neushul, 1967; N i c h o l s , 1970) and the present study support the continuum hypothesis of independent species d i s t r i b u t i o n s . The evaluation of d i f f e r e n t methods of community analysis must consider not only i t s representativeness and i t s effectiveness but also the amount of time which must be put i n t o the a n a l y s i s . The a s s o c i a t i o n analysis i s a p a r t i c u l a r l y time-consuming method on the IBM 1130 computer. The c l u s t e r analysis i s r e l a t i v e l y rapid but, because of the s i z e of the matrix, requires the use of the IBM 360/67-2 computer under MTS operating system, which i s more expensive. The Bray and C u r t i s method also requires the use of t h i s computer but takes considerably l e s s time. However, i f the o r d i n a t i o n i s constructed by hand, a large number of man-hours i s required. 92 Construction of the species c o n s t e l l a t i o n diagram i s e s s e n t i a l l y a man-hour consuming method, the 1130 computer being required only to generate the species a s s o c i a t i o n matrix. In summary, methods based on d i s t i n c t community differences (such .as a s s o c i a t i o n a n a l y s i s , Zurich-Montpellier a n a l y s i s , and Bray and C u r t i s ordination) did not produce d i s t i n c t communities. Methods which could accomodate gradual t r a n s i t i o n s between communities (such as c l u s t e r and inverse analysis) provided recognizable quadrat and species units which could be i d e n t i f i e d as communities. For the v i s u a l representation of communities, the species c o n s t e l l a -t i o n diagram i s probably the most l u c i d of a l l the methods. However, i t s usefulness i n suggesting c o n t r o l l i n g environmental factors i s probably more l i m i t e d than any of the other methods. Since the inverse analysis gave e s s e n t i a l l y the same r e s u l t s and represents a more accurate method of community a n a l y s i s , i t i s recommended that the two procedures be used i n conjunction, the r e s u l t s of the inverse analysis being used to sub-s t a n t i a t e the p o s i t i o n i n g of the species i n the c o n s t e l l a t i o n diagram. Comparison with other s u b t i d a l communities The observation that the s u b t i d a l region represents an area of con-tinuous changes i n species composition rather than sharply defined commun-i t i e s has been noted by other authors taking a more subjective approach to the d e l i n e a t i o n of s u b t i d a l seaweed communities. Although the only known previous attempt to i d e n t i f y s u b t i d a l seaweed communities with the use of mathematical analysis was i n c o n -c l u s i v e , the l i t e r a t u r e contains a number of verbal and schematic d e s c r i p -tions of s u b t i d a l communities, p a r t i c u l a r l y from Europe and North America. 93 A comparison of some of these descriptions with the present r e s u l t s i n d i -cate s t r i k i n g , d i f f e r e n c e s with the communities of Europe and C a l i f o r n i a and s t r i k i n g s i m i l a r i t i e s with those of S t . Margaret's Bay, Nova S c o t i a , and Friday Harbor, Washington. B a s i c a l l y , the s u b t i d a l f l o r a of Europe i s dominated by Laminaria communities from the lower i n t e r t i d a l to the depth l i m i t s of the k e l p , with no mention of a deep water a l g a l a s s o c i a t i o n . Other species are cf secondary importance i n the kelp beds and can be divided into two assoc i a t i o n s : shade-loving species occurring i n the lower canopy and sun-lov i n g species occurring on the upper s t i p e s or on rock outcrops ( K i t c h i n g , 1941; Smith, 1967.) Although Mann (1972) found a s i m i l a r s u b t i d a l dominance pattern of the Laminariales i n S t . Margaret's Bay, Nova S c o t i a , four of h i s eight communities appear to be analogous to communities i n the present study area i n terms of community s t r u c t u r e , species composition, depth, and substrate. His Fucus-Ascophyllum, Chondrus crispus, Zostera marina, Agarum-Ptilota zones resemble the i n t e r t i d a l , f o l i o s e r e d , sandy bottom and deep water communities of the present study. Since three of the other zones are defined on the occurrence of two species of Laminaria and therefore resemble the laminarian community of the present study, only his Chorda filum community lacks an analog i n the Georgia S t r a i t area. The s u b t i d a l f l o r a of c e n t r a l C a l i f o r n i a has been divided into two communities—an upper zone dominated by Calliarthron cheilosporioides but containing a large number of s u b t i d a l . brown and lower i n t e r t i d a l red algae and a lower zone dominated by Pterygophora californica and other large brown algae, with a gentle t r a n s i t i o n between the two communities (McLean, 94 1962). Similar zones have been observed on the outer coast of Vancouver Island (personal observation). In 1915, Muenscher defined f i v e a l g a l associations from San Juan I s l a n d , Washington, by t h e i r species composition. His upper i n t e r t i d a l Endocladia muricata zone has no counterpart i n the present study because a l l but one quadrat was c o l l e c t e d below mean tide l e v e l . His four other associations are r e a d i l y recognizable by t h e i r species composition as the i n t e r t i d a l , f o l i o s e r e d , laminarian and sandy bottom communities of the present study. Common species to the two analyses include Rhodomela larix, Microcladia borealis, Polysiphonia sp., Ulva lactuca, and Colpomenia sinuosa i n h i s Fucus-Gigartina a s s o c i a t i o n ; Odonthalia floccosa, Corallina officinalis, Gigartina exasperata, Plocamium coccineum v a r . pacificum, Ahnfeltia plicata, Monostroma fuscum, Iridaea cordata, i n h i s Ulva asso-c i a t i o n ; Nereocystis luetkeana, Costaria costata, Agarum fimbriatum, Laminaria spp., Gigartina exasperata, Rhodymenia pertusa, Cryptopleura ruprechtiana and Iridaea cordata i n h i s Laminariaceae a s s o c i a t i o n , and Agardhiella tenera i n h i s Zostera marina a s s o c i a t i o n . The observation that the communities i n t h i s study represent the normal s i t u a t i o n i n protected inner coastal waters needs to be substan-t i a t e d by more thorough documentation of s u b t i d a l seaweed communities around the world. 1? BIBLIOGRAPHY Bannister, P. 1968. An evaluation of some procedures used i n simple o r d i n a t i o n s . Journal of Ecology 56: 27-34. Beals, E. W. 1973. Ordination: mathematical elegance and e c o l o g i c a l n aivete. Journal of Ecology 61-; 23-35. Blanc, p., & Ch.-F. Boudouresque. 1970. S i g n i f i c a t i o n des peuplements Pr^ c o r a l l i g e n e s de Mediterranee par 1'analyse f a c t o r i e l l e en facteurs principaux. C. R. Acad. Sc. Paris 271:493-496. Bloom, S. A., J . L. Simon, & V. D. Hunter. 1972. Animal-sediment r e l a t i o n s and community analysis of a F l o r i d a estuary. Marine Biology ]_3_: 43-56. Boudouresque, Ch.-F. 1971. Methodes d'etude q u a l i t a t i v e et q u a n t i t a t i v e du benthos (en p a r t i c u l a r du phytobenthos). Tethys _3: 79-104. Bray, J . R., & J . T. C u r t i s . 1957. An ordination of the upland forest communities of northern Wisconsin. E c o l o g i c a l Monographs 2J7: 325-349. Canadian Hydrographic S e r v i c e . 1972. Canadian Tide and Current Tables. Volume 5. Juan de Fuca and Georgia S t r a i t s . 87 pp. C a s s i e , R.M., & A. D. Michael. 1968. Fauna and sediments of an i n t e r -t i d a l mud f l a t : a m u l t i v a r i a t e a n a l y s i s . Journal of experimental marine Biology and Ecology 2_: 1-23. Ceska, A., & H. Roemer. 1971. A computer program for i d e n t i f y i n g species-releve groups i n vegetation s t u d i e s . Vegetatio 23_: 255-277. Connell, J . H. 1961a. E f f e c t of competition, predation by Thais lapillus, and other factors on n a t u r a l populations of the barnacle Balanus balanoides. E c o l o g i c a l Monographs _31: 61-104. Connell, J . H. 1961b. The influence of i n t e r s p e c i f i c competition and other factors on the d i s t r i b u t i o n of the barnacle Chthamalus stellatus. Ecology 42: 710-723. Connell, J . H. 1972. Community i n t e r a c t i o n s on marine rocky i n t e r t i d a l shores. Annual Review of Ecology and Systematics J3: 169-192. Daubenmire, R. 1968. Plant communities: a textbook of plant synecology. New York: Harper & Row. 300 pp. Day, J . H., J . G. F i e l d , & M. P. Montgomery. 1971. The use of numerical methods to determine the d i s t r i b u t i o n of the benthic fauna across the continental shelf of North C a r o l i n a . Journal of Animal Ecology 40: 93-125. Doty, M. 1946. C r i t i c a l t i d e f actors that are corr e l a t e d with the v e r t i c a l d i s t r i b u t i o n of marine algae and other organisms along the P a c i f i c coast. Ecology 27: 315-328. 96 Erman, D. C , & W. T. Helm. 1971 Comparison of some species importance values and ordi n a t i o n techniques used to analyze benthic invertebrate communities. Oikos 22: 240-247. F i e l d , J . G. 1971 A numerical analysis of changes i n the soft-bottom fauna along a transect across False Bay, South A f r i c a . Journal of experimental marine Biology and Ecology _7: 215-253. F r a l i c k , J . E. 1971. The e f f e c t of harvesting Iridaea on a s u b l i t t o r a l marine plant community i n northern Washington. M. S. t h e s i s . Western Washington State U n i v e r s i t y , Bellingham. 60 pp. Gauch, H. G., J r . 1973. A qu a n t i t a t i v e evaluation of the Bray-Curtis o r d i n a t i o n . Ecology 54: 829-836. Gauch, H. G., J r . , & R. H. Whittaker. 1972. Comparison of ordi n a t i o n techniques. Ecology 53_: 868-875. Gessner, F . 1970. Temperature: p l a n t s . In Marine Ecology: A compre-hensive, integrated t r e a t i s e on l i f e i n oceans and coastal waters. Otto Kinne, ed. London: Wiley-Interscience. Pp. 363-406. Gessner, F., & W. Schramm. 1971. S a l i n i t y : p l a n t s . In Marine Ecology: A comprehensive, integrated t r e a t i s e on l i f e i n oceans and coastal waters. Otto Kinne, ed. London: Wiley-Interscience. Pp. 705-820. Goodall, D. W. 1953. Objective methods of the c l a s s i f i c a t i o n of vege-t a t i o n . I . The use of p o s i t i v e i n t e r s p e c i f i c c o r r e l a t i o n . A u s t r a l i a n Journal of Botany _1: 39-63. Goodall, D. W. 1954. Objective methods for the c l a s s i f i c a t i o n of vege-t a t i o n . I I I . An essay i n the use of f a c t o r a n a l y s i s . A u s t r a l i a n Journal of Botany 2_: 304-325. Greig-Smith, P. 1964. Quantitative plant ecology. London: Butterworth's. Second e d i t i o n . 256 pp. Hartog, C. den. 1972. Substratum: m u l t i c e l l u l a r p l a n t s . In Marine Ecology: A comprehensive, integrated t r e a t i s e on l i f e i n oceans and co a s t a l waters. Otto Kinne, ed. London: Wiley-Interscience. Pp. 1277-1289. Hedgpeth, J . W., ed. 1957. Tr e a t i s e on marine ecology and paleoecology. Volume 1. New York: The Geological Society of America. Pp. 1-16. Hellebust, J . A. 1970. L i g h t : p l a n t s . In Marine Ecology: A comprehensive, integrated t r e a t i s e on l i f e i n oceans and coastal waters. Otto Kinne, ed. London: Wiley-Interscience. Pp. 125-158. H o l l i s t e r , H. J . 1972. Observations of seawater temperature and s a l i n i t y at B r i t i s h Columbia shore s t a t i o n s 1971. Marine Sciences D i r e c t o r a t e , P a c i f i c Region, V i c t o r i a , B. C. Report No. 72-14. 97 H o l l i s t e r , H. J . , & A. M. Sandnes. 1972. Sea surface temperatures and s a l i n i t i e s at shore s t a t i o n s on the B r i t i s h Columbia coast 1914-1970. Marine Sciences D i r e c t o r a t e , P a c i f i c Region, V i c t o r i a , B. C. Report No. 72-13. H o t e l l i n g , H. 1933. Analysis of complex v a r i a b l e s into p r i n c i p a l com-ponents. Journal of Educational Psychology 24_: 417-441, 498-520. Hughes, R. N., D. L. Peer, & K. H. Mann. 1972. Use of m u l t i v a r i a t e a n a l y s i s to i d e n t i f y f u n c t i o n a l components of the benthos i n St. Margaret's Bay, Nova S c o t i a . Limnology and Oceanography 17_: 111-121. Hughes, R. N. , & M.L.H. Thomas. 1971a. C l a s s i f i c a t i o n and ordination of benthic samples from Bedeque Bay, an estuary i n Prince Edward I s l a n d , Canada. Marine Biology 10: 227-235. Hughes, R. N., & M.L.H. Thomas. 1971b. The c l a s s i f i c a t i o n and o r d i -nation of shallow-water benthic samples from Prince Edward I s l a n d , Canada. Journal of experimental marine Biology and Ecology 7_: 1-39. Kershaw, K. A. 1964. Quantitative and dynamic ecology. London: Edward . Arnold. 183 pp. K i t c h i n g , J, A. 1941. Studies i n s u b l i t t o r a l ecology. I I I . Laminaria f o r e s t on the west coast of Scotland: a study of zonation i n r e l a t i o n to wave ac t i o n and i l l u m i n a t i o n . B i o l o g i c a l B u l l e t i n 80: 324-337. Lewis, J . R. 1961 The l i t t o r a l zone on rocky shores—a b i o l o g i c a l or p h y s i c a l entity? Oikos .12: 280-301. Lewis, J . R. 1964. The ecology of rocky shores. London: E. U. P. 323 pp. L i e , U., & J . C. K e l l e y . 1970. Benthic infauna communities o f f the coast of Washington and i n Puget Sound: i d e n t i f i c a t i o n and d i s t r i b u t i o n of the communities. Journal of the Fi s h e r i e s Research Board of Canada 27: 621-651. Mann, K. H. 1972. E c o l o g i c a l energetics of the seaweed zone i n a marine bay on the A t l a n t i c coast of Canada. I . Zonation and biomass of sea-weeds. Marine Biology 12: 1-10. Mathieson, A. C , & R. L . Burns. 1971. E c o l o g i c a l studies of economic red algae. I . Photosynthesis and r e s p i r a t i o n of Chondrus crispus Stackhouse and Gigartina stellata (Stackhouse) B a t t e r s . Journal of experimental marine Biology and Ecology 7_: 197-206. McLean, J . H. 1962. S u b l i t t o r a l ecology of kelp beds of the open coast area near Carmel, C a l i f o r n i a . B i o l o g i c a l B u l l e t i n 122: 95-114. Muenscher, W. L. C. 1915. A study of the a l g a l associations of San Juan I s l a n d . Puget Sound B i o l o g i c a l Station P u b l i c a t i o n 1: 59-84. 98 M u l l e r , J . E. 1971. Geological reconnaissance map of Vancouver Island and Gulf I s l a n d s . Revised. Geological Survey of Canada open f i l e map. Neushul, M. 1965. SCUBA diving studies of the v e r t i c a l d i s t r i b u t i o n of benthic marine p l a n t s . I n The v e r t i c a l and h o r i z o n t a l d i s t r i b u t i o n s of marine organisms. T. L e v r i n g , ed. Proceedings of the F i f t h Marine B i o l o g i c a l Symposium, Goteborg^ Pp. 161-176. Neushul, M. 1967. Studies of s u b t i d a l marine vegetation i n Western Washington. Ecology 48_: 83-94. N i c h o l s , F. H. 1970. Benthic polychaete assemblages and t h e i r r e l a t i o n -ships to the sediment i n Port Madison, Washington. Marine Biology 6.: 48-57. Oosting, H. J . 1958. The study of plant communities: an i n t r o d u c t i o n to plant ecology. San Francisco: W. H. Freeman & Company. 440 pp. Paine, R. T. 1972. I n t e r t i d a l communities. Seminar. U n i v e r s i t y of B r i t i s h Columbia. October 18. P i e l o u , E. C. 1969a. An i n t r o d u c t i o n to mathematical ecology. New York: Wiley-Interscience. 286 pp. P i e l o u , E. C. 1969b. As s o c i a t i o n tests versus homogeneity t e s t s : t h e i r use i n subdividing quadrats into groups. Vegetatio JJ8: 4-18. Popham, J . D., & D. V. E l l i s . 1971. A comparison of t r a d i t i o n a l c l u s t e r and Zurich-Montpellier analyses of infaunal pelecypod associations from two adjacent sediment beds. Marine Biology 8_: 260-266. R u s s e l l , G. 1972. P h y t o s o c i o l o g i c a l studies of a two-zone shore. I . Basic p a t t e r n . Journal of Ecology 60_: 539-545. S a i t o , Y., & S. Atobe. 1970. P h y t o s o c i o l o g i c a l study of i n t e r t i d a l marine algae. I . U s u j i r i benten-jima, Hokkaido. B u l l e t i n of the Faculty of F i s h e r i e s , Hokkaido U n i v e r s i t y 2^: 37-69. Smith, R. M. 1967. S u b l i t t o r a l ecology of marine algae on the North Wales coast. Helgol'ander wissenschaftliche Meeresuntersuchungen 15: 467-479. Sokal, R. R., & F. J . Rohlf. 1962. The comparison of dendrograms by objective methods. Taxon 11: 33-40. Sokal, R. R., & P. H. Sneath. 1963. P r i n c i p l e s of numerical taxonomy. San Francisco: W. H. Freeman. 359 pp. Stephenson, T. A., & A. Stephenson. 1949. The u n i v e r s a l features of zonation between tidemarks on rocky, coasts. Journal of Ecology 37: 289-305. 99 Stephenson, W., W. T. Williams, & S. D. Cook. 1972. Computer analyses of Petersen's o r i g i n a l data on bottom communities. E c o l o g i c a l Monographs 42: 387-415. Stephenson, W., W. T. Williams, & G. N. Lance. 1970. The macrobenthos of Moreton Bay. E c o l o g i c a l Monographs 40: 459-494. Thorson, G. 1957. Bottom communities ( s u b l i t t o r a l or shallow s h e l f ) . In T r e a t i s e on marine ecology and paleoecolrgy. Volume 1. J . W. Hedgpeth, ed. New York: The Geological Society of America. Pp. 461-534. Wade, B. A. 1972. A d e s c r i p t i o n of a highly diverse soft-bottom com-munity i n Kingston Harbour, Jamaica. Marine Biology 13_: 57-69. Widdowson, T. B. 1965. A survey of the d i s t r i b u t i o n of i n t e r t i d a l algae along a coast t r a n s i t i o n a l i n respect to s a l i n i t y and t i d a l f a c t o r s . Journal of the F i s h e r i e s Research Board of Canada 22; 1425-1454. Williams, W. T., & J . M. Lambert. 1959. M u l t i v a r i a t e methods i n plant ecology. I . Association analysis i n plant communities. Journal of Ecology 47: 83-101. Williams, W. T., & J . M. Lambert. 1960. M u l t i v a r i a t e methods i n plant ecology. I I . The use of an e l e c t r o n i c d i g i t a l computer for a s s o c i a t i o n a n a l y s i s . Journal of Ecology 4_8: 689-710. 100 Appendix 1. Sites of transects and c o l l e c t i o n dates of quadrats. A t o t a l of seven s u b t i d a l transect l i n e s were sampled i n 1972. Three 100-m transect l i n e s were l a i d o f f the SE end of Bath Island (Site BI01), running approximately 110° from magnetic north and ranging i n depth from near mean t i d e l e v e l to 29 feet below mean ti d e l e v e l . The l i n e s were run from an x-axis p a r a l l e l to the shore at x = 000 m, x = 050 m, and x = 085 m. These transect l i n e s were sampled May 18 - 30, June 20 - July 4, and August 2 - 1 5 , r e s p e c t i v e l y . A 200-m l i n e was run o f f the north shore of Bath Island ( S i t e BI02) at 20° from magnetic north and sampled from a depth of 10 feet to 98 f e e t . C o l l e c t i o n s were made along t h i s transect l i n e from July 7 - 1 9 . The three transect l i n e s o f f the SE end of Sear Island ( S i t e SI02) were run at an angle of about 110° from magnetic north, from near mean t i d e l e v e l to a depth of approximately 53 f e e t . The x-coordinate of the transect l i n e s and the dates sampled were x = 050, May 31 - June 13; x = 000, July 18 - 31; and x = 025, October 6 - 7 . 101 Appendix 2. Species not used i n the computer analyses. Species not used because they occurred i n fewer than three quadrats and whose i d e n t i t i e s and/or d i s t r i b u t i o n s may be questionable. Species Acrochaetium K y l i n Quadrats Remarks 10 31 119 An epiphyte—occurrence probably more frequent than noted Antithamnion floccosum 108 (Muller) Kleen 109 Blidingia minima v a r . minima 12 (Naegeli) K y l i n Callithamnion Lyngbye 117 Callophyllis firma (Kylin) 8 Norris 128 Species taxonomy uncertain Choreocolax Reinsch Codiolum petrocelidis Kuckuck Colpomenia bullosus (Saunders) Yamada Colpomenia sinuosa (Roth) Derbes & S o l i e r Cryptonemia obovata J. Agardh Derbesia marina (Lyngbye) S o l i e r Ectocarpus Lyngbye Elachista fucicola (Velley) Areschoug Enteromorpha linza (Linnaeus) J . Agardh 12 49 34 23 23 31 29 100 123 33 118 10 119 87 12 102 119 On Pterosiphonia bipinnata On Herposiphonia plumula An endophyte—may have been over-looked i n other quadrats Erythrocladia subintegra Rosenvinge 28 29 102 Species Quadrats Remarks Goniotrichopsis sublittoralis Smith Goniotrichum elegans (Chauvin) Zanardini Hildenbrandia prototypus Nardo Membranoptera Stackhouse Petalonia debilis (C. Agardh) Derbes & S o l i e r Petrocelis franciscana S. & G. Pilayella tenella S. & G. Porphyra gardneri Smith & Hollenberg Punctaria G r e v i l l e Rhizoclonium implexum (Dillwyn) Kutzing Rhodymenia californica K y l i n Vrospora Areschoug 29 120 63 77 78 130 25 34 115 11 39 45 9 95 96 105 131 53 54 74 This alga occurs as the encrusting base of Gigartina papillata an i s probably more common than i n d i c a t e d . None attached to Laminaria, the usual host. Small specimen. Species not used because they occurred i n whose i d e n t i t i e s were unmistakable. fewer than three quadrats but Alaria tenuifolia S e t c h e l l Codium setchellii S. & G. Delesseria decipiens J . Agardh Desmarestia ligulata var. ligulata (Lightfoot) Lamouroux 39 78 47 126 27 51 67 Last year's plant—badly grazed This year's j u v e n i l e . Dictyota binghamiae J . Agardh 11 103 Species Farlowia mollis (Harvey & Bailey) Farlow & S e t c h e l l Fauchea fryeana S e t c h e l l Quadrats Remarks Perhaps Pikea californica Harvey 46 64 125 128 This alga makes i t s f i r s t appear-ance i n l a t e summer. Halosaccion glandiforme (Gmelin) Ruprecht Opuntiella californica (Farlow) K y l i n Porphyra perforata J . Agardh 105 29 12 34 35 Frequently a contaminant i n other quadrats, these were the only ones i n which i t was attached. Sarcodiotheca furcata (S. & G.) 29 K y l i n 109 123 Scytosiphon lomentaria (Lyngbye) J . Agardh Syringoderma abyssicola (S. & G.) Levring 25 78 83 Species not used because of uncertain i d e n t i f i c a t i o n s but which occurred i n more than three quadrats. Fryeella garnderi (Setchell) K y l i n 7 29 53 73 101 A l l specimens s m a l l . Halicystis ovalis (Lyngbye) Areschoug 14 40 45 55 112 Hymenena G r e v i l l e 72 102 119 126 127 Tentatively i d e n t i f i e d as j u v e n i l e Cryptopleura ruprechtiana 104 Species Quadrats Remarks Pugetia fragillisima K y l i n 74 A l l specimens s m a l l . 98 100 107 108 110 111 112 118 125 Rhodymenia palmata (Linnaeus) 96 A l l specimens small and none G r e v i l l e 97 reproductive 127 128 Species not used because t h e i r small sizes suggested that they may have been overlooked i n a number of quadrats. Choreocolax polysiphoniae 51, 54, 58, 61, 97 Reinsch Cladophora Kutzing 20, 24, 26, 59, 120, 126 c o l o n i a l diatoms 7, 21, 25, 27, 28, 31, 33, 39, 40, 46, 58, 77, 85, 96, 107, 108, 109, 110, 116, 123, 128, 130 E r y t h r o t r i c M a caraea 10, 20, 31, 65, 96, 119, 120, 121 (Dillwyn) J . Agardh Gonimophyllum skottsbergii 8, 13, 21, 31, 36, 60, 72, 79, 97 S e t c h e l l Janczewskia gardneri 13, 14, 18, 19, 49, 79, 86 S e t c h e l l & Guernsey Pilayella l i t t o r a l i s 25, 64, 86, 95 (Lyngbye) Kjellman Sphacelaria subfusca S. & G. 29, 39, 44, 48, 52, 61, 66, 72, 86, 95, 93, 98, 100, 105, 107, 120, 126 Trailiella intricata Batters 10, 29, 37, 76, 77, 83, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 129, 130 105 Appendix 3. Summary of as s o c i a t i o n ana Species on which d i v i s i d n ( s ) occurred +Gigapa " ,+Pterde " ',+Rhodla " ,+Sargmu +0donfl " ,+Bonnno " ,+Grifpa " ,+Kallob ' " ,+Polyheg " ,+Polyla " ,+Rhodla " ,+Rhodper " ,+Sphaca " ,+Zostmal " ,+Polyla,+Iridcoc " , " ,+Polypaz " , " ,+Rhodpl +Agarte +Boss +Branwo +Kallob +Micrco +Nerelu +Nien + P l a t c l +Polyuru +Rhodla +Rhodper +Ulva +Boss,+Platpe " ,+Rhodpl —0— s i s r e s u l t s for p = .001. Quadrats 12, 20, 24, 25, 26, 34, 35, 65, 95, 113, 114, 115 96 23 22, 59, 105, 119, 120, 126 5, 18, 19, 21, 33, 71, 79, 80, 82, 118, 122 45, 61, 110 88, 125 102 121 57, 74 63, 64, 86 43, 44, 70, 81, 97, 98, 99, 100, 103 76 66, 67, 69, 73 32, 42, 46, 47, 49, 50, 60, 72, 104, 127, 128 31 75 55, 68 6, 8, 13, 14, 15, 16, 17, 27, 28, 116, 117 41, 107, 108, 111, 112 29, 129 124 7, 39, 48, 53 54, 58, 90, 101 11 51 36, 62 93 106 9, 10, 37, 87, 92, 94, 130 40 38, 52, 77, 78, 83, 85, 89, 91, 109, 123, 131 106 Appendix 4. Summary of a s s o c i a t i o n analysis r e s u l t s f o r p = .05. Species on which d i v i s i o n ( s ) occurred Quadrats +Gigapa 12, 20, 115 " ,+Boss 25 " ,+Cerawa 24, 95, 114 " ,+Herpri 126 " ,+Pterde 96 " ,+Ulva 26, 34, 35, 65, 113 " ,+Cerawa,+Rhodla 23 " ,+Herpri,+Ralffu 22 , " , " ,+Priola 59, 105, 120 " , " , " ,+Pterbi 119 +0donfl 74, 75, 76 " ,+Conssu 63, 81, 82 " ,+Nien 88 " ,+Ralffu 110 " ,+Zostmal 69 " ,+Conssu,+Antide 47, 57, 60 " , " ,+Leatdi 5, 21, 121 " , " ,+Micrco 18, 19 " , " ,+Nerelu 79, 80, 99 " ,+Zostmal,+Sargmu 66, 67, 73 " ,+Conssu,+Antide,+Coraofc 127 " , " , " ,+Rhodla 64, 86 " , " , " ,+Rhodpl 45, 125 " , " , " ,+Ulva 61 " , " ,+Leatdi,+Sargmu 33 " , " ,+Nerelu,+Platpe 97, 100 " , " ,+Ralffu 71 " , " , " ,+Rhodro 44, 70 " , " ,+Antide,+Coraofc,+Amplpa 43 " , " , " , " ,+Costco 46, 72 " ,+Amplpa,+Ploccop 98 " , " ,+Polyheg 49 +Ploccop,+Kallob 102 " ,+Micrco 42, 50, 104 " ,+Priola 122 ,+Pterbi 32, 118 ." ,+Priola,+Rhodper 31, 103, 128 —0— 106 +Laursp 14, 16 +Ptergr 29, 129, 92 +Rhodla . 36, 62 +Weekfr 94 +Laursp,+Branwo 116, 117 11 ,+Crypru 6 " ' ,+Ploccop 15, 17 " ,+Ralffu 27, 28 " ,+Ulva 8, 13 107 Species +Ptergr Quadrats +Botrps 90 +Branwo 111 +Coraofc 9, 10, 11 +Iridcoc 124 +Pterde 131 +Zostmal 68 +Botrps,+Pterde 83, 85, 89 " ,+Rhodper 130 +Branwo,+Rhodpl 107, 108, 112 +Coraofc,+Rhodpl 37, 40, 87 " ,+Schipa 39 +Pterde,+Desmvi 48, 51, 58 " ,+Monofu 38, 77, 101 " ,+Polypaz 52, 53, 54 " ,+Pterbi 123 " ,+Sargmu 55 +Botrps,+Pterde,+Rhodpl 78 " , " ,+Sargmu 109 +Coraofc,+Rhodpl,+Rhodper 93 " , " ,+Sargmu 41 +Pterde,+Desmvi,+Weekfr 7 " ,+Monofu,+Polyla 91 

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.831.1-0101315/manifest

Comment

Related Items