UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Population ecology of the clonal red alga Mazzaella Cornucopiae from Barkley Sound, Canada Scrosati, Ricardo Augusto 1996

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

Item Metadata

Download

Media
831-ubc_1997-196534.pdf [ 9.91MB ]
Metadata
JSON: 831-1.0088141.json
JSON-LD: 831-1.0088141-ld.json
RDF/XML (Pretty): 831-1.0088141-rdf.xml
RDF/JSON: 831-1.0088141-rdf.json
Turtle: 831-1.0088141-turtle.txt
N-Triples: 831-1.0088141-rdf-ntriples.txt
Original Record: 831-1.0088141-source.json
Full Text
831-1.0088141-fulltext.txt
Citation
831-1.0088141.ris

Full Text

POPULATION ECOLOGY OF THE CLONAL RED A L G A MAZZAELLA CORNUCOPIAE F R O M BARKLEY SOUND, C A N A D A by RICARDO AUGUSTO SCROSATI Lie, Universidad de Buenos Aires, 1987 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department of Botany) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA A p r i l 1997 © Ricardo Augusto Scrosati, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract This Thesis examines aspects of the population ecology of clonal algae using as a model Mazzaella cornucopiae (Rhodophyta, Gigartinaceae), an abundant species i n the high intertidal zone of Prasiola Point, Barkley Sound, British Columbia, Canada. Research focused on investigating the kin d of interactions that occur among fronds during growth, discussing its significance for M. cornucopiae i n light of what is known for the better studied clonal and non-clonal terrestrial plants. M. cornucopiae develops from holdfasts as a turf, which results i n both costs and benefits for thalli, and they are analyzed i n relation to the characteristics of its habitat. Additionally, the role of holdfasts and spores for population maintenance, recolonization after disturbance, and frond regeneration after harvesting was also studied. During the growth season for Mazzaella cornucopiae, between late winter and summer, frond density and stand biomass increases simultaneously, hence no self-thinning occurred among fronds as it occurs among individual thalli of actively growing, non-clonal algae. I introduced the use of randomization tests to solve the lack of randomness i n data collection. The decrease of frond size hierarchy observed during the growth season indicates that the growth increment of small fronds per time unit was higher than that of large fronds (symmetric competition model). A n increase of frond crowding, as it occurs during the growth season, involves a cost expressed as reduced net photosynthetic rates; however, frond crowding is beneficial overall. Laboratory and field experiments indicated that high frond densities protect fronds from high desiccation and irradiance and the associated bleaching and tissue loss that spring and summer daytime low-tides bring. Additionally, frond crowding confers protection against the negative effects of wave action on frond survival. Holdfasts are fundamental for population maintenance. They are mostly perennial and constitute the main source of the annual production of fronds; the annual contribution of spores is very low. Finally, experimental pruning of fronds showed that, if commercial harvesting of this species is ever done, a spring collection of all frond biomass, without damaging holdfasts, gives the highest possible biomass yield while allowing frond biomass to be annually renewed. iv Table of Contents Page Abstract i i Table of Contents iv List of Tables v i i i List of Figures ix Acknowledgements x i v Chapter 1: General introduction 1 Chapter 2: Structure and dynamics of the population from Prasiola Point 14 Introduction 14 Materials and Methods 16 Study site and sampling program 16 Statistical analysis 18 Gametophyte:tetrasporophyte ratio 19 Relationship between population parameters and environmental variables 21 Results 22 Structure and dynamics of the population 22 Gametophyte:tetrasporophyte ratio 23 Relationship between population parameters and environmental variables 24 Discussion and Conclusions 25 Structure and dynamics of the population 25 Production of reproductive fronds 28 Gametophyte:tetrasporophyte ratio 29 Relationship between population parameters and environmental variables 36 Summary of conclusions 39 Chapter 3: The relationship between stand biomass and frond density 51 Introduction 51 Materials and Methods 54 Study site and sampling program 54 V Statistical analysis 55 Choice of stand biomass over mean frond biomass 55 Functional relationship between frond density and stand biomass 57 Random samples and randomization tests 58 Comparison with the "ultimate biomass-density line" for terrestrial plants 60 Results 61 Test for self-thinning among fronds of Mazzaella cornucopiae 61 The annual biomass-density relationship and its site-dependence 61 Comparison with the "ultimate biomass-density line" for terrestrial plants 62 Discussion and Conclusions 63 Randomization tests and parametric tests 63 Lack of self-thinning among fronds of Mazzaella cornucopiae 63 Comparison with the "ultimate biomass-density line" for terrestrial plants 67 The annual biomass-density relationship and its site-dependence 69 Is frond crowding beneficial for Mazzaella cornucopiae? 72 Interactions among growing genets 72 Chapter 4: Factors affecting the production, growth, and mortality of fronds 83 Introduction 83 Materials and Methods 86 Effects of density on the rate of production of fronds 86 Temporal variation of frond size hierarchy 88 Effects of frond density on desiccation and understory irradiance 89 Effects of desiccation and irradiance on net photosynthesis 90 Effects of frond density on bleaching and subsequent loss vi of frond tissues 93 Effects of wave action on frond mortality mediated by frond density 94 Results 94 Effects of density on the rate of production of fronds 94 Temporal variation of frond size hierarchy 95 Effects of frond density on desiccation and understory irradiance 96 Effects of desiccation and irradiance on net photosynthesis 97 Effects of frond density on bleaching and subsequent loss of frond tissues 98 Effects of wave action on frond mortality mediated by frond density 99 Discussion and Conclusions 99 Effects of density on the rate of production of fronds 99 Temporal variation of frond size hierarchy 100 Interactions between frond density, desiccation, irradiance, and net photosynthesis 104 Effects of wave action on frond mortality mediated by frond density 108 Chapter 5: Recruitment, holdfast perennation, and regrowth after harvesting 121 Introduction 121 Materials and Methods 124 Relative contribution of recruitment from spores and of perennating holdfasts to the maintenance of populations 124 Relative contribution of recruitment from spores and of perennating, bordering holdfasts to the recolonization of disturbed areas 125 Time for full recovery after disturbances occurring at different seasons 126 Frond regeneration after harvesting 127 Results 129 Relative contribution of recruitment from spores and of perennating holdfasts to the maintenance of populations 129 Relative contribution of recruitment from spores and of perennating, bordering holdfasts to the recolonization of disturbed areas 130 Time for full recovery after disturbances occurring at different seasons 131 Frond regeneration after harvesting 131 Discussion and Conclusions 134 Importance of recruitment from spores and of perennating holdfasts for population maintenance and recolonization after disturbance 134 Frond regeneration after harvesting 139 Production of reproductive fronds during frond regeneration 144 Chapter 6: General conclusions and suggested areas for future research 156 Structure and dynamics of the population from Prasiola Point 156 Frond interactions during growth 159 The estimation of stand biomass from frond density 162 Mode of competition among fronds 162 The adaptive significance of frond crowding 163 Recruitment from spores and holdfast perennation 165 Frond regeneration after harvesting 166 Final remarks 168 References Authors of scientific names cited i n the text 171 188 viii List of Tables Page Table 2.1: Functional relationships between frond biomass (B, i n g) and frond length (L, i n cm) for Mazzaella cornucopiae from Prasiola Point. 46 Table 2.2: Dates of frond collection for the calculation of biomass-length relationships and dates to which these relationships were applied to estimate stand biomass. 47 Table 2.3: Summary table for one-way repeated-measures A N O V A for percent cover of Mazzaella cornucopiae from Prasiola Point. 48 Table 2.4: Summary table for one-way repeated-measures A N O V A for frond density (fronds cm - 2) of Mazzaella cornucopiae from Prasiola Point. 48 Table 2.5: Summary table for one-way repeated-measures A N O V A for mean frond length (cm) of Mazzaella cornucopiae from Prasiola Point. 48 Table 2.6: Summary table for one-way repeated-measures A N O V A for stand biomass (g cm - 2) of Mazzaella cornucopiae from Prasiola Point. 49 Table 2.7: Relative abundance of gametophytic (G) and tetrasporophytic (T) fronds of Mazzaella cornucopiae from Prasiola Point, expressed as percentage of G and T and as the G:T ratio. 49 Table 2.8: Pearson correlation coefficients between population parameters (PC = percent cover, FD = frond density, MFL = mean frond length, and SB = stand biomass) for Mazzaella cornucopiae from Prasiola Point and five environmental variables measured near Prasiola Point (see text for details). 50 ix List of Figures Page Fig. 1.1: Fronds of a stand of Mazzaella cornucopiae from the intertidal zone at Prasiola Point. The largest fronds may reach up to 5 cm in length. 11 Fig. 1.2: Location of Prasiola Point. 12 Fig. 1.3: General view of the moderate wave-exposure side (East) of Prasiola Point. 13 Fig. 2.1: Seasonal variation of percent cover of Mazzaella cornucopiae from Prasiola Point (mean + SEM). 41 Fig. 2.2: Seasonal variation of frond density (fronds cm - 2) of Mazzaella cornucopiae from Prasiola Point (mean + SEM). 41 Fig. 2.3: Seasonal variation of mean frond length (cm) of Mazzaella cornucopiae from Prasiola Point (mean + SEM). 42 Fig. 2.4: Seasonal variation of stand biomass (g cm - 2) of Mazzaella cornucopiae from Prasiola Point (mean + SEM). 42 Fig. 2.5: Seasonal variation of the density of cystocarpic fronds and tetrasporic fronds of Mazzaella cornucopiae from Prasiola Point (mean + SEM). 43 Fig. 2.6: Seasonal variation of mean monthly air and sea surface temperature (°C) at La Perouse Bank buoy. 43 Fig. 2.7: Seasonal variation of mean monthly wave height (m), from crest to trough, at La Perouse Bank buoy. 44 Fig. 2.8: Seasonal variation of mean monthly salinity (%o) at Amphitrite Point. 44 Fig. 2.9: Seasonal variation of mean monthly daylength (h) at Tofino . 45 Fig. 3.1: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae from Prasiola Point during the 1994 growth period. 76 X Fig. 3.2: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae from Prasiola Point during the 1995 growth period. 77 Fig. 3.3: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #1 at Prasiola Point throughout 2 years. 78 Fig. 3.4: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #2 at Prasiola Point throughout 2 years. 78 Fig. 3.5: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #3 at Prasiola Point throughout 2 years. 79 Fig. 3.6: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #4 at Prasiola Point throughout 2 years. 79 Fig. 3.7: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #5 at Prasiola Point throughout 2 years. 80 Fig. 3.8: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #6 at Prasiola Point throughout 2 years. 80 Fig. 3.9: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #7 at Prasiola Point throughout 2 years. 81 Fig. 3.10: Combinations between logio rnean frond biomass and logio frond density for stands of Mazzaella cornucopiae from Prasiola Point (n=96). The plotted line is the "ultimate biomass-density line" proposed for terrestrial plants (see text for an explanation). 81 Fig. 3.11: Combinations between logio mean frond biomass and logio frond density for clumps of Mazzaella cornucopiae from Prasiola Point and Nudibranch Point (n=56). The plotted line is the "ultimate biomass-density line" proposed for terrestrial plants (see text for explanation). 82 Fig. 4.1: Temporal change of percent desiccation of fronds of Mazzaella cornucopiae from Prasiola Point within a photosynthetic chamber of an infra-red gas analyzer (IRGA) under 20 uE n r 2 s _ 1 of xi irradiance. I l l Fig. 4.2: Temporal change of percent desiccation of fronds of Mazzaella cornucopiae from Prasiola Point within a photosynthetic chamber of an IRGA under 515 uE m~2 s _ 1 of irradiance. I l l Fig. 4.3: Temporal variation of the coefficient of variation (C.V.) for frond length (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. 112 Fig. 4.4: Temporal variation of the coefficient of variation (C.V.) for frond length (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. 112 Fig. 4.5: Temporal variation of frond length (cm; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. 113 Fig. 4.6: Temporal variation of frond length (cm; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. 113 Fig. 4.7: Temporal variation of the coefficient of variation (C.V.) for frond biomass (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. 114 Fig. 4.8: Temporal variation of the coefficient of variation (C.V.) for frond biomass (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. 114 Fig. 4.9: Temporal variation of frond biomass (mg; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. 115 Fig. 4.10: Temporal variation of frond biomass (mg; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. 115 Fig. 4.11: Net photosynthesis (mean ± SEM) of fronds of Mazzaella cornucopiae from Prasiola Point, expressed as uptake rates of CO2, exposed to air at 20 uE m~2 s _ 1 as a function of percent desiccation. 116 Fig. 4.12: Net photosynthesis (mean ± SEM) of fronds of Mazzaella cornucopiae from Prasiola Point, expressed as uptake rates of CO2, exposed to air at 515 uE m~2 s _ 1 as a function of percent desiccation. 117 xii Fig. 4.13: Net photosynthesis (mean + SEM) of fronds of Mazzaella cornucopiae from Prasiola Point, expressed as uptake rates of CO2, exposed to air at 20 uE m~2 s"1 (low irradiance) and at 515 LIE n r 2 s"1 (high irradiance) as a function of percent desiccation. 118 Fig. 4.14: Number of cystocarps per frond of Mazzaella cornucopiae from Prasiola Point versus frond biomass (mg); fronds were collected on 27-28 October 1996. 119 Fig. 4.15: Number of tetrasporic sori per frond of Mazzaella cornucopiae from Prasiola Point versus frond biomass (mg); fronds were collected on 27-28 October 1996. 120 Fig. 5.1: Density of Mazzaella cornucopiae thalli (mean + SEM) that were recruited from spores i n quadrats cleared i n June 1993 at Prasiola Point. 146 Fig. 5.2: Percent cover (mean + SEM) of perennating thalli of Mazzaella cornucopiae in undisturbed quadrats and of thalli that were recruited from spores in quadrats cleared in June 1993 at Prasiola Point. 146 Fig. 5.3: Frond length (cm, mean + SEM) of perennating thalli of Mazzaella cornucopiae in undisturbed quadrats and of thalli that were recruited from spores i n quadrats cleared i n June 1993 at Prasiola Point. 147 Fig. 5.4: Percent cover (mean + SEM) of Mazzaella cornucopiae thalli that were recruited from spores in quadrats cleared in June 1993 at Prasiola Point and of new holdfast areas that grew from perennating holdfasts that bordered the same cleared quadrats. 148 Fig. 5.5: Percent cover (mean + SEM) of Mazzaella cornucopiae in natural quadrats (controls) and i n quadrats cleared at different seasons between 1992 and 1993 (see text for a complete explanation) at Prasiola Point. 149 Fig. 5.6: Seasonal variation of percent cover (mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). 150 Fig. 5.7: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes in percent cover (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). 150 Fig. 5.8: Seasonal variation of frond density (fronds cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). Fig. 5.9: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes i n frond density (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). Fig. 5.10: Seasonal variation of frond length (cm; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). Fig. 5.11: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes i n frond length (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). Fig. 5.12: Seasonal variation of total stand biomass (g cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). 154 Fig. 5.13: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes i n stand biomass (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). 154 Fig. 5.14: Seasonal variation of the density of cystocarpic fronds (fronds cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point in quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and in undisturbed, control quadrats (C). 155 Fig. 5.15: Seasonal variation of the density of tetrasporic fronds (fronds cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point in quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). 155 151 151 152 153 Acknowledgements I wish to especially thank my supervisor, Robert E. De Wreede, for his guidance and support throughout this Thesis, and for all that I learned from him during the past few years. I also thank Gary Bradfield and Paul G. Harrison, for their useful suggestions during my research. A special mention goes for Frank Shaughnessy and Jeong Ha Kim, from whom I benefited from their experience i n life and with who I spent great moments, including fieldtrips to Bamfield and the International Seaweed Symposium i n Chile. M y fieldwork at Barkley Sound received the invaluable and fun assistance mainly of Tarda Thenu, Jean-Paul Danko, Laura Wong, and Eduardo Jovel. In statistics, I was lucky to count with the advice from Alfred Briilisauer, Robert Schutz, and Donald Jackson. I am also indebted to Kanti Patel, for his complete guidance for the use of the infra-red gas analyzer to measure photosynthetic rates, and to Mary Berbee, who kindly provided me with additional laboratory equipment. Rob De Wreede's lab was a very enjoyable workplace, and I'm also grateful to its crew: Brent Phillips, Kristen Drewes Milligan, Russell Markel, Andrea Sussmann, Nick Grabovac, Len Dyck, and Fatima Mouhajir. I couldn't forget to especially thank Carol Ann Borden and Shona Ellis, for the excellent experience of being their teaching assistant and for their patience when I had to combine fiel d trips w i t h teaching responsibilities. I also extend my deep gratitude to the personnel of Bamfield Marine Station for their logistic support for field work. This Thesis was mainly funded by an operating grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) to Rob De Wreede. During my first year at UBC, additional funding was generously provided by Ron Foreman. I am deeply indebted to UBC for granting me a University Graduate Fellowship (UGF), an Edith Ashton Memorial Scholarship, and a Kit Malkin Scholarship. The Organization of American States (OAS) also granted me a PRA Fellowship. It was also very helpful to count with financial aid from the X l l t h International Seaweed Symposium organizing committee and from the Faculty of Graduate Studies of UBC to attend the XVth International Seaweed Symposium i n Valdivia, Chile. Another important friend for me was Gustavo Zuleta, who significantly helped me during my first few months at UBC. M y best lucky shot during my stay i n Vancouver was to meet who finally became my wife, Oliva Ortiz Alvarez. I am enormously indebted for the infinite support, patience, and love that I encountered i n her (and this gratitude includes our little 'sporeling', Pablo Mauricio). Finally, I would like to express that nothing that what I accomplished here i n Canada would have been possible without the never-ending, loving support from my parents, Amalia Ines Abad de Scrosati and Angel Mauricio Scrosati (Haedo, Buenos Aires, Argentina). It is to them both and to my wife that I dedicate this Thesis. Chapter 1 General introduction The ecology of populations of marine macroalgae is a fascinating field of research. Traditionally, it has been less studied than the ecology of populations of terrestrial plants, both because of a generally higher interest i n terrestrial plants, from scientific and commercial viewpoints, and because of the easier accesibility of terrestrial compared with marine habitats. Thus, progress i n the ecological understanding of seaweed populations was, and still is, greatly influenced by the ecological knowledge gained for terrestrial plant populations. For example, an important book that summarized the ecology of terrestrial plant populations was published 20 years ago by Harper (1977). Relevant publications that reviewed and discussed different aspects of the ecology of seaweed populations had to wait for data to be specifically gathered for seaweeds and such publications appeared from the following decade onwards [i.e., Dayton 1985, Chapman 1986, Russell 1986, De Wreede and Klinger 1988, Kain and Norton 1990, Santelices 1990, John et al. 1992, Lobban and Harrison 1994). The above body of evidence pointed out that seaweeds have many ecological traits that are similar to those of terrestrial plants, but seaweeds also have some unique ecological properties that derive from their inherent biology or from characteristics of the marine habitat. Some of these unique properties are the existence of complex life-histories for most red algae, which include an alternation between three generations, two of them with independent life (van den Hoek et al. 1995), the coalescence that occurs among holdfasts of actively growing, neighboring thalli (Maggs and Cheney 1990, Santelices et al 1996), and the particular adaptations of algae to the intertidal zone, where they have to withstand the cyclic and dramatic changes of abiotic factors such as desiccation, irradiance, and hydrodynamic forces that result from tidal movements. The fact that the ecology of seaweed populations is less well known than the ecology of terrestrial plant populations makes the former an attractive field for research. The added importance of this kind of research comes when we consider that macroalgae are primary components of coastal marine communities worldwide, and a knowledge of their ecology is necessary to predict responses to increased human-caused perturbations on these habitats or to better manage algae as economic resources subjected to regular exploitation. The objective of this Thesis is to provide new insights on the ecology of populations of seaweeds, i n particular, of clonal red seaweeds (phylum Rhodophyta). Clonal seaweeds are those that produce several upright fronds that have the potential capacity for independent life if they become separated from the parent thallus together with an associated portion of holdfast. Common examples of clonal seaweeds are Chondrus crispus Stackhouse (Irish moss, a red seaweed) and Ascophyllum nodosum (Linnaeus) Le Jolis (a brown seaweed), both abundant i n eastern Canada. The entire thallus of a clonal alga that is derived from a single spore constitutes a genet, whereas fronds constitute ramets [sensu Harper 1977, Jackson et al. 1985, van Groenendael and de Kroon 1990). The species selected as a model for my research was Mazzaella cornucopiae (Postels et Ruprecht) Hommersand [=Iridaea cornucopiae), which belongs to the family Gigartinaceae, order Gigartinales (Fig. 1.1). This species occurs in coasts of the North Pacific ocean, from Japan to northern California, including Alaska, British Columbia, Washington, and Oregon (Mikami 1965, Abbott 1971, Hommersand et al. 1993, Selivanova and Zhigadlova 1997). The reason for choosing M. cornucopiae for my study is that it is a highly abundant species i n the high intertidal zone of Prasiola Point (48° 49' N, 125° 10' W, located in southern Barkley Sound, in the West coast of Vancouver Island, British Columbia, Canada, Figs. 1.2 and 1.3), which was selected as the study site because of its adequate accesibility and safety for fieldwork. A t this site, M. cornucopiae can cover extensive areas of the rocky substratum. The size of individual genets is as yet unknown due to possible holdfast coalescence and high frond densities, which prevent from an accurate identification of genets. Fronds of M. cornucopiae can reach up to 5 cm in length at Prasiola Point and have been reported to be as long as 15 cm for other localities (Mikami 1965, Abbott 1971). Prasiola Point is a conglomerate of fractured/layered volcanic material and it has two regimes of wave exposure. The West side usually receives the direct impact of large waves that come from the open ocean, whereas the East side receives smaller waves i n comparison, since it is protected by the tip of Prasiola Point. Biogeographically, this coast belongs to the cold temperate northeastern Pacific region (Luning 1990) and, within this region, to the Oregonian province, which comprises the western coast of North America between Dixon Entrance (North of Queen Charlotte Islands) and Point Conception, California (Briggs 1974, Foster et al. 1991). Close to Prasiola Point, mean monthly sea surface temperature oscillated between 8.6 °C (winter) and 16.1 °C (summer) between the summer seasons of 1993 and 1995, whereas seawater salinity oscillated between 26.7 %o (winter) and 31.5 %o (summer) during the same period (see Chapter 2). At Prasiola Point, Mazzaella cornucopiae is the dominant algal species in the high intertidal zone of the wave-exposed side, but it shares the space with other co-dominant algae i n the wave-protected side. These additional algae are Fucus gardneri and Pelvetiopsis limitata, which are brown algae that may constitute dense, almost-monospecific patches as well (throughout this Thesis, for ease of reading, authors of all scientific names cited i n this Thesis are listed at the end of the Bibliography). Other algae also occur i n this zone, but they are less abundant or occur only during certain seasons. These algae are Cladophora columbiana (green), Mastocarpus papillatus (red), Endocladia muricata (red), Callithamnion pikeanum (red), Odonthalia floccosa (red), Analipus japonicus (brown), Scytosiphon dotyi (brown), and a crustose, unidentified species (different from the Petrocelis phase of Mastocarpus). Between late winter and early spring, the red alga Pterosiphonia gracilis appears i n small quantities. The rocky surface is also inhabited by sessile or low-mobility invertebrates. Barnacles of the species Balanus glandula, Balanus cariosus, and Chthamalus dalli occur here. Limpets, e.g. Lottia, and snails, e.g. Littorina, graze in the same area; carnivorous snails, e.g. Nucella, are also present i n the area. Occasionally, small specimens of the mussel Mytilus trossulus grow attached to the rocky substratum. Isopods and amphipods are frequent organisms that occur in this area as well. Fish ecology has been studied by N.J. Wilimovsky and coworkers for this area. The high intertidal zone is also regularly visited by seagulls, crows, and oyster catchers during low tides. Mostly bare rock is above the M. cornucopiae zone, which is inhabited by marine lichens. Below the M. cornucopiae zone, the rocky substratum is mainly dominated by mussels [Mytilus trossulus and Mytilus californianus), barnacles, and goose-neck barnacles [Pollicipes polymerus, i n the wave-exposed sites); the alga Mazzaella heterocarpa appears mainly between late winter and late spring i n this area. Sea stars and sea urchins, which occur at even lower tidal levels, seem not to reach the high intertidal zone. Some of the key biotic interactions that occur between the main species of the high intertidal zone from Prasiola Point are discussed in Kim (1995) and Kim and De Wreede (1996a, b). At the beginning of the fieldwork that was done as part of this Thesis, little was known about the basic structure and dynamics of the population of Mazzaella cornucopiae from Prasiola Point (see K i m 1995). Chapter 2 discusses the general structure and dynamics of the population of M. cornucopiae from Prasiola Point. The seasonal variation of total thallus cover, frond density, mean frond length, stand biomass, and density of reproductive fronds was examined. The population was also characterized i n terms of the spatial and temporal variation of the relative abundance of gametophytes and tetrasporophytes, which are the two independent-life generations of this species. The possible influence that abiotic factors such as daylength, air and sea surface temperature, seawater salinity, and wave action have on the dynamics of the population of M. cornucopiae is also discussed. Observations were compared with the structure and dynamics of populations of other red algae of the family Gigartinaceae to place M. cornucopiae within the context of what is presently known about the general population ecology of these species. Chapter 3 discusses the interactions that occur among fronds of Mazzaella cornucopiae during the period of active growth. For monospecific stands of even-aged, non-clonal plants that are actively growing i n crowded conditions, the smallest plants start to die at a certain point mainly as a consequence of competition for light with large plants (Hara 1988, Weiner 1988, 1990). This process is known as self-thinning, during which plant density and total stand biomass are inversely related (Weller 1987). The few reports available on this matter for non-clonal algae indicate that they also undergo self-thinning during active growth (Dean et al. 1989, Reed 1990, Ang and De Wreede 1992, Creed 1995). It is interesting to see how ramets of clonal plants interact during growth, since they are physically related to one another, unlike non-clonal plants. For stands of terrestrial clonal plants from seasonal habitats, shoots (ramets) do not undergo self-thinning during the growth season (Hutchings 1979, Pitelka 1984, de Kroon 1993, de Kroon and Kalliola 1995). However, for stands of clonal algae, the situation for fronds (ramets) is not clear yet. Pybus (1977) found a positive relationship between stand biomass and frond density for Chondrus crispus and Mastocarpus stellatus from Ireland, whereas Martinez and Santelices (1992) found no relationship between mean frond biomass and frond density for Mazzaella laminarioides from Chile, which was interpreted as evidence of a lack of self-thlnrung among fronds (these three species belong to the Gigartinales). However, Pybus (1977) and Martinez and Santelices (1992) analyzed "static" relationships between biomass and density (data were collected from different algal stands, which were not followed through time), and a static biomass-density relationship may not allow us to infer exactly what the dynamic biomass-density relationship (biomass-density being measured through time for a given stand) might be (Weller 1989). The only report presently available on this matter for clonal seaweeds indicates that fronds from a subtidal population of the red species Gelidium sesquipedale (Gelidiales) from Portugal do not undergo self-thinning during growth (Santos 1995). The main objective of Chapter 3 was to analyze the dynamic relationship between stand biomass and frond density for Mazzaella cornucopiae, which provides evidence whether self-thinning among fronds occurs during growth. Chapter 3 also discusses the rationale for using stand biomass and frond density as the correct way to test for self-thinning. This was necessary because some authors presently analyze the correlation between mean plant (or frond) biomass and plant (or frond) density (see, for example, Cousens and Mortimer 1995, Petraitis 1995, Flores-Moya et al. 1996) as means of testing for self-thinning, but this method has problems of interpretation that may lead to incorrect conclusions. The possible causes for the observed dynamic relationship between stand biomass and frond density for Mazzaella cornucopiae, and the similarities and differences with the dynamics of non-clonal plants/algae and of ramets of clonal terrestrial plants are also discussed in Chapter three. A n additional objective of Chapter 3 was to find a quick and reliable way of estimating stand biomass of Mazzaella cornucopiae. The estimation of stand biomass for seaweeds may be achieved through destructive sampling (harvest of parcels, De Wreede 1985), which may not be desirable because of conservation concerns, through plotless techniques, which may not be accurate (Littler and Littler 1985), or through the measurement of the size structure of a stand, which is usually time-consuming. The relationship between stand biomass and frond density for M. cornucopiae was calculated for several stands, in order to search for a quick and reliable way of estimating stand biomass from a count of fronds i n a given plot. It was also of interest during my research to make inferences about the kind of competition that fronds of Mazzaella cornucopiae undergo while fronds are growing and stands become more dense. The temporal variation of the frond size hierarchy (=variability or inequality) for a given stand offers information about the mode of competition among fronds. Again, most of the knowledge on competition models is available for non-clonal terrestrial plants. Before self-thinning operates during the growth of non-clonal plants, plant size hierarchy increases, since the growth rate of the small plants progressively decreases as the large plants grow i n size and shade the small plants. When self-thinning commences, however, plant size hierarchy starts to decrease because of the progressive death of the smallest, growth-suppressed plants (Schmitt et al. 1986, Weiner and Thomas 1986, Hara 1988, Weiner 1988,1990, Knox et al. 1989). Stands of non-clonal plants then follow the asymmetric competition model (Hara 1988, Weiner 1990, de Kroon 1993). For ramets of clonal terrestrial plants from seasonal habitats, there is no asymmetric competition among ramets and ramet size hierarchy is not expected to increase during growth (de Kroon 1993). For clonal seaweeds, the only available information on the dynamics of frond size hierarchy was obtained for the subtidal red species Gelidium sesquipedale (Santos 1995), for which frond length hierarchy decreased during growth. The mode of competition among fronds of M. cornucopiae is discussed i n Chapter 4, as inferred from the temporal variation of frond size hierarchy during the growth period. Preliminary observations had shown that the density of fronds of Mazzaella cornucopiae increases during its growth season. A n increase i n frond density and stand biomass results in a progressive decrease of irradiance levels at the understory. This generally results in a reduced photosynthesis of fronds (Hay 1981, Taylor and Hay 1984). Why would an alga increase its frond density continuously during the growth season if that implies a cost expressed as reduced net photosynthesis? Existing evidence for other intertidal algae, including red, brown, and green species, indicates that high frond densities provide protection against strong desiccation and subsequent bleacMng of fronds (Hay 1981, Padilla 1984, Taylor and Hay 1984). This suggests that there may be more benefits for M. cornucopiae derived from frond crowding that would outweigh the above mentioned cost. I explored the adaptive significance that frond crowding represents for M. cornucopiae, especially i n terms of the benefits that a high frond density might confer for growth and survival at the high intertidal zone. Specifically, I studied the interactions between frond density, desiccation, irradiance, frond bleaching, mortality through wave action, and reproduction. This is discussed i n Chapter 4. For clonal algae within the Gigartinaceae, the relative importance of spores to the maintenance of populations is highly variable, and it depends on the species and/or the site considered (Hansen and Doyle 1976, Hansen 1977, May 1986, Santelices and Norambuena 1987, Lazo et al. 1989, A ng et al. 1990, Gomez and Westermeier '1991). Chapter 5 discusses the role that carpospores and tetraspores together have on the annual production of fronds for the population of Mazzaella cornucopiae from Prasiola Point and compares results with what is known for other species of the Gigartinaceae. Disturbances at Prasiola Point may occur during unusually hot, sunny, and dry summer seasons, as a result of the loss of abundant algal biomass (R. A. Scrosati, pers. obs.), or through strong wave action, which is highest mainly during fall and winter (see Chapter 2). The relative importance that spores have on recolonization after disturbances, compared with recolonization from perennating holdfasts that border the disturbed area, was also assessed and discussed i n Chapter 5. The time for a full recovery of M. cornucopiae (reaching control levels of total thallus cover) after disturbances that occur at different seasons was determined by monitoring experimental quadrats that had been artificially cleared before the beginning of the fieldwork done for this Thesis (see K i m and De Wreede 1996a). Additional kinds of disturbance for Mazzaella cornucopiae are those that result from harvesting fronds. M. cornucopiae is not presently exploited for commercial purposes, but its high abundance i n the high intertidal zone of outer Barkley Sound could some day make this species economically interesting. The potential interest for M. cornucopiae may result from the presence of carrageenans in cell walls, which are sulphated polysaccharides used i n a variety of fields (McLachlan 1985), or from the antiviral activity found i n extracts from this species (J. H. Kim, pers. comm.). The response of M. cornucopiae populations to different intensities of frond harvesting, in terms of recovery rates and the effects on the production of reproductive structures, was also analyzed i n Chapter 5. The dynamics of the recovery process was also compared i n Chapter 5 with the recovery from harvesting experienced by other algal species within the Gigartinaceae and other related red algae (MacFarlane 1952, Taylor 1959-60, Burns and Mathieson 1972, Mathieson and Burns 1975, Hansen 1977, Carter and Anderson 1985, Carter and Simons 1987, Santelices and Norambuena 1987, Westermeier et al. 1987, Pringle and Semple 1988, Santelices et al. 1989, Gomez and Westermeier 1991), i n search for general relationships between harvesting impact and regeneration capacity for these species. It is my hope that this Thesis w i l l contribute to improve our understanding of the causes that determine the structure and dynamics of populations of clonal red algae. Fig . 1.1: Fronds of a stand of Mazzaella cornucopiae from the intertidal zone at Prasiola Point. The largest fronds may reach up to 5 cm i n length. 1 2 Fig. 1.2: Location of Prasiola Point. 13 Fig. 1.3: General view of the moderate wave-exposure side (East) of Prasiola Point. Chapter 2 Structure and dynamics of the population from Prasiola Point Introduction The red alga Mazzaella cornucopiae occurs frequently i n the high intertidal zone of rocky areas that are located close to the open ocean i n Barkley Sound, British Columbia. Some of the key biological interactions that exist between this species and other frequent members of the intertidal community at Prasiola Point and surrounding areas, including interspecific competition with fucoid algae and herbivory by snails, are discussed i n K i m (1995) and K i m and De Wreede (1996a, b). At the beginning of the fieldwork for this Thesis, the structure and dynamics of the population of M. cornucopiae from Prasiola Point (and from its entire biogeographical range i n general; see Chapter 1) was poorly known. This knowledge was necessary to allow me to formulate and answer the ecological questions that are posed throughout the rest of this Thesis. The present Chapter w i l l discuss the structure and the dynamics of the population of Mazzaella cornucopiae from Prasiola Point. A possible coalescence of adjacent holdfasts (Maggs and Cheney 1990, Santelices et al. 1996) and the high frond densities observed at Prasiola Point makes the accurate identification of neighboring t h a l l i of M. cornucopiae almost impossible, which is particularly true for large stands. This is an unresolved problem for the study of the dynamics of thalli of some members of the Gigartinaceae (Bhattacharya 1985, Lazo et al. 1989). In addition, it is also possible that different parts of the same thallus of M. cornucopiae survive independently if they are physically separated from one another by some kind of disturbance (R. A. Scrosati, pers. obs.), so counts of separate holdfasts do not guarantee that one is dealing with distinct thalli even if coalescence has not occurred. The population structure and dynamics of M. cornucopiae w i l l be discussed i n this Chapter i n terms of the seasonal variation of total thallus cover, frond density, mean frond length, and stand biomass. The period of reproduction by spores was estimated for this population by periodically quantifying the density of cystocarpic and tetrasporic fronds. The presence of fronds with either cystocarps or tetrasporic sori at Prasiola Point indicates that M. cornucopiae alternates between gametophytes and tetrasporophytes at this site, which are the two independent-life, reproductive generations found i n most red algae (van den Hoek et al 1995). The proportion between both generations is an important descriptor of the structure of a red algal population, and it varies with seasons or site conditions for other species of the Gigartinaceae (Mathieson and Burns 1975, Hansen and Doyle 1976, Craigie and Pringle 1978, Bhattacharya 1985, Dyck et al. 1985, Hannach and Santelices 1985, May 1986, Santelices and Norambuena 1987, Westermeier et al. 1987, Lazo et al. 1989, Luxoro and Santelices 1989, De Wreede and Green 1990, Olson 1990, Bolton and Joska 1993, Dyck and De Wreede 1995, Lindgren and Aberg 1996, Piriz 1996, Shaughnessy et al. 1996). The seasonal and spatial variation of the ratio between gametophytes and tetrasporophytes was also determined for the population of M. cornucopiae from Prasiola Point. Abiotic factors exert a major influence on the population dynamics of seaweeds that have seasonal behaviour (Kain and Norton 1990, Luning 1993, Luning and Kadel 1993, Lobban and Harrison 1994). Little is known about the possible effects of the environment on the seasonality of Mazzaella cornucopiae. In this Chapter, correlations between total thallus cover, frond density, mean frond length, and stand biomass and environmental factors such as air and sea surface temperature, wave height, salinity, and daylength are examined to determine the possible influence of the environment on M. cornucopiae population dynamics. A i r temperature was considered i n this study because M. cornucopiae is exposed to atmospheric conditions for long periods, given its location i n the high intertidal zone. Wave height was the only available descriptor of wave energy for the study area, and wave energy is known to have important biomechanical effects on algal survivorship (Denny et al. 1989, Carrington 1990, Gaylord et al. 1994, Denny 1995, Shaughnessy et al. 1996). Materials and Methods Study site and sampling program The field study was carried out i n the high intertidal zone of the East side of Prasiola Point (Figs. 1.3 and 1.4), which is protected against strong wave action by a vertical rocky wall. On 3 June 1993, a preliminary sampling was done to determine an appropriate number of replicates to be monitored during the entire study period. The percent cover of Mazzaella cornucopiae was measured i n an area where thalli were abundant, using a 10 cm x 10 cm quadrat (with 100 subdivisions) that was repeatedly placed i n a random fashion along a transect line. Throughout this Thesis, randomness was achieved using a random number table and by following Krebs' (1989) recommendations. This particular size of quadrat was selected based on preliminary measurements suggesting that for the variables of interest the error would be acceptable. For each quadrat, percent cover was determined as the number of subdivisions were M. cornucopiae occupied half or more than half of the subdivision. With the mean percent cover (m=43.2, n=40) and the sample standard deviation (S=25.3), a number of 9 replicates was determined as the minimum to keep the index of precision at a generally accepted level of 20%, following n = 25 S 2 n r 2 (c/. Elliott 1977). The position of 9 permanent transects was delimited by cementing a concrete nail to the rock at both extremes of each transect. The position of one 10 cm x 10 cm quadrat, where data were repeatedly collected on successive sampling dates, was established in each transect using a random number table. This design was selected over randomly placing 9 quadrats along just one transect because data from these 9 quadrats had to be used for additional questions (harvesting experiment; see Chapter 5), due to time constraints during each field trip. Five population variables, namely total percent cover, frond density, mean frond length, stand biomass, and density of reproductive fronds, were measured approximately at bimonthly intervals, which was the best possible option given time and funding available. There were 14 sampling dates: 4-6 June 93, 16-20 August 93, 15-18 October 93, 10-12 December 93, 24-27 February 94, 25-29 A p r i l 94, 21-25 June 94, 18-22 August 94, 5-9 October 94, 2-6 December 94, 28-31 January 95, 30 March-3 A p r i l 95, 13-17 May 95, and 10-14 July 95. The time available for fieldwork was limited during sampling dates, as measurements for this and other purposes could only be taken during low tides. Because of that, percent cover was measured for each entire quadrat, but frond density and frond length were estimated from six 2 cm x 2 cm, randomly chosen subquadrats in each quadrat (total area of 6 subquadrats = 24 cm 2). The same subquadrats were examined on successive sampling dates. A l l of the fronds present in the subquadrats were counted, and their frond length was measured to the nearest 5 mm. Stand biomass was estimated by applying power relationships between frond length and frond biomass, determined for different dates (Table 2.1), to the frond density and mean frond length found for the quadrats. Frond biomass-length relationships were not examined for most of the first year of this study, so the relationships measured mainly during the second year were applied to comparable dates that lacked biomass-length measurements (Table 2.2). Statistical analysis The number of replicates available for the statistical analysis of the population dynamics of Mazzaella cornucopiae was six. This was decided because two disturbances removed a significant amount of biomass from two quadrats during the study (one disturbance was probably caused by the indirect effect of a nearby concrete nail on the stability of the substratum), and because a third quadrat was difficult to relocate exactly on successive sampling dates. Differences among monthly values of percent cover, frond density, mean frond length, and stand biomass were tested using one-way, repeated-measures analyses of variance (ANOVAR, Howell 1992), since values were repeatedly estimated from the same quadrats. The Huynh-Feldt adjustment for probability values (which corrects for the lack of compound symmetry of the covariance matrix, Howell 1992) was preferred over the Greenhouse-Geisser adjustment because it is considered to be a more powerful and reliable method (Myers and Well 1991, Howell 1992). The assumption of normality of scores within each sampling date was tested using the "Probability Plot-Normal" option i n SYSTAT 5.2.1 for Macintosh (Wilkinson et al. 1992), and it was satisfactory for all cases, so no data transformations were done. With respect to the assumption of homogeneity of variance, the criterion of Howell (1992) was followed: if the data distributions are symmetric, and if the largest variance is no more than four times the smallest, the analysis of variance is most likely to be valid. This was applicable to my data set. To graphically describe the temporal trend for the four population variables analyzed, values were smoothed following the LOWESS algorithm (Cleveland 1979 i n Wilkinson et al. 1992), which involves a locally weighted robust regression. This was done using SYSTAT 5.2.1, with a tension factor of 0.5. Gametophyte:tetrasporophyte ratio Fronds of Mazzaella cornucopiae were sampled from different sites within Prasiola Point and on different dates to determine the spatial and temporal variation of the gametophyte:tetrasporophyte (G:T) ratio. To characterize the area where the population dynamics was studied, 15 fronds were haphazardly collected from each of the 9 permanent transects on 29 A p r i l 1994 and on 20-21 May 1995 (n=135 for each date), i.e., from the moderate wave exposure side (E) of Prasiola Point. To test for possible seasonal differences on the G:T ratio, frond sampling was repeated for the same 9 transects on 22 December 1995. To test for a possible difference on the G:T ratio under high wave exposure, 100 fronds were haphazardly collected from the high wave exposure side (W) of Prasiola Point on 20 May 1995 and on 28 October 1995. Additionally, to detect if the G:T ratio changes with tidal height, the two frond samplings done i n 1995 at the high wave exposure site were carried out at the high and at the low zone of the M. cornucopiae belt (approximately 1 m of difference in height). Fronds taken from the high zone came from thalli that covered almost all of the available substratum, and they were collected far enough apart to ensure as much as possible that they represented different thalli. Fronds collected from the low zone belonged to different clumps clearly separated from one another. Once collected, fronds were air dried until laboratory analyses were done at the University of British Columbia, i n Vancouver. The reproductive phase for each frond was determined according to the resorcinbl method (Garbary and De Wreede 1988), which is designed to detect the presence of kappa-carrageenan, found only i n gametophytes for members of the Gigartinaceae (Waaland 1975, McCandless and Craigie 1979, McCandless et al. 1983). The accuracy of the resorcinol method for determining phases of Mazzaella cornucopiae depends on the experimental protocol used (Shaughnessy and De Wreede 1991). For the samples analyzed here, frond fragments around 0.5-1 mg i n weight were placed i n test tubes with 2 ml of the resorcinol reagent, and then submerged for 2 min i n a water bath at 70-75 °C, to be finally transferred to a cold bath immediately after the 2 minutes. Solutions containing gametophytic material showed dark red, while the ones containing tetrasporophytic material were light orange, pink or transparent. To determine the correct phase for the few doubtful cases, the color of the solutions were compared with the color obtained when using reproductive fronds, so that the phase could be known with certainty (4 gametophytic fronds and 4-7 tetrasporophytic fronds were used for each run of the resorcinol test). The percentage of doubtful identifications using this protocol oscillated between 0 and 5% for the different groups of fronds analyzed. Relationship between population parameters and environmental variables Environmental data collected i n the vicinity of Prasiola Point were gathered from the Institute of Ocean Sciences (Sidney, BC) and Environment Canada (Vancouver, BC). A i r temperature, sea surface temperature, and wave height, from crest to trough, were recorded daily at an offshore buoy located in La Perouse Bank (48° 50' N, 126° W), salinity was recorded at Amphitrite Point (48° 33' N, 125° 19' W), on the north side of the mouth of Barkley Sound, and daylength was recorded at Tofino (48° 50' N, 125° 8' W). These are the closest sites to Prasiola Point that regularly and reliably measure environmental data. The particular characteristics of Prasiola Point may determine some environmental variation w i t h respect to the above mentioned sites. The pattern of seasonal changes of air temperature may be similar at Prasiola Point and La Perouse Bank, although the temperature amplitude may be higher i n Prasiola Point, because it is a coastal area. Sea surface temperature and salinity should be similar between Prasiola Point, La Perouse Bank, and Amphitrite Point, since Prasiola Point is just a few km inwards i n Barkley Sound. With respect to wave height, although absolute values are probably lower at the protected side of Prasiola Point than at La Perouse Bank, my personal observations suggest that the seasonal trend of changes is approximately the same for both sites. Daylength differs only a few minutes between Tofino and Prasiola Point, because of the small difference i n latitude between these two localities. The relationship between population parameters and environmental variables was analyzed by doing separate linear correlation analyses (Howell 1992) for each possible combination. Population data were correlated separately with environmental data collected on the same month, on the previous month, and two months previous. These two lag periods were considered to detect possible delays i n the influence that the environment may have on the population dynamics of Mazzaella cornucopiae. Data analyses were conducted using SYSTAT 5.2.1, and the significance of correlation coefficients was assessed using Fisher and Yates' (1963) table. Results Structure and dynamics of the population The discontinuous belt formed by Mazzaella cornucopiae where my quadrats were located at the wave-protected side (E) of Prasiola Point occurs approximately between 3.07 m and 4.03 m above lowest normal tide (LNT, Canadian Chart Datum). Percent cover, frond density, mean frond length, and stand biomass of Mazzaella cornucopiae varied significantly among months (Tables 2.3, 2.4, 2.5, and 2.6). Monthly mean percent cover varied between 48.8 ±- 11.7 % and 80.2 + 5.8 % (mean ± SEM, n=6), with a minimum absolute value of 10% and a maximum absolute value of 9 9 % for the different quadrats. Monthly mean frond density oscillated between 5.2 ±0.8 fronds cm - 2 and 10.4 ±- 1.5 fronds cm - 2, with a minimum value of 3.4 fronds cm - 2 and a maximum value of 20.6 fronds cm - 2 . Monthly mean frond length oscillated between 0.7 ± 0.1 cm and 1.1 ± 0.1 cm, with a minimum mean value of 0.4 cm and a maximum mean value of 1.6 cm (the largest fronds were approximately 5 cm long). Monthly mean stand biomass ranged between 15 ± 5 mg cm - 2 and 115 ±- 15 mg cm - 2, with a minimum value of 3 mg cm"2 and a maximum value of 205 mg cm"2. These population variables generally followed a seasonal pattern, with maximum values between mid-spring and summer and rrtinimum values i n winter (Figs. 2.1, 2.2, 2.3, and 2.4). Two exceptions to this general behaviour were spring and summer of 1995 for percent cover and for mean frond length, which followed a decreasing trend instead of an increasing one as expected from previous years. • Density of reproductive fronds The density of reproductive fronds followed a seasonal pattern, being present during fall and winter and completely or almost completely absent during summer (Fig. 2.5). The few cystocarps observed during the summer seemed to be empty or in poor condition. Cystocarpic fronds were more abundant than tetrasporic fronds at a l l sampling dates, and cystocarps appeared before tetrasporangia in both reproductive seasons. Gametophyte:tetrasporophyte ratio Gametophytic fronds of Mazzaella cornucopiae were more abundant than tetrasporophytic fronds at all studied sites and seasons at Prasiola Point (Table 2.7). A t the wave-protected side (E) of Prasiola Point, where the population dynamics was investigated, the G:T ratio was similar i n spring 1994 (3.3), spring 1995 (2.1), and winter 1995/6 (2.6). The G:T ratio had a higher seasonal variation at the high wave exposure side (W) of Prasiola Point. A t this site, the G:T ratio was 1.5 in spring 1995 and 4.3 in fall 1995. There was an apparent dependence of the G:T ratio on tidal height at the wave-exposed site. The proportion of gametophytic fronds was higher at the high zone of the M. cornucopiae belt (the G:T ratio was 1.6 in spring 1995 and 9.0 i n fall 1995) than at the low zone (1.4 i n spring 1995 and 2.6 i n fall 1995). In the fall of 1995 at the wave-exposed site, sori-bearing tetrasporophytic fronds were mostly found at the low zone of the M. cornucopiae belt, and cystocarp-bearing fronds were rare i n that area. At the high zone, the reverse was true. Relationship between population parameters and environmental variables A i r and sea surface temperature, wave height, salinity, and daylength followed a seasonal trend of change i n the vicinity of Prasiola Point (Figs. 2.6 to 2.9). Pearson correlation coefficients between Mazzaella cornucopiae population parameters and the above environmental variables are shown i n Table 2.9. In general, frond density and stand biomass had the highest correlation with environmental variables; percent cover and mean frond length were poorly or not associated with the environment. Among the environmental variables, temperature, wave height, and daylength had the highest levels of significance with the population parameters (positive correlations for temperature and daylength and negative correlations for wave height). Significance levels decreased when population values were correlated with environmental values from the previous month, and only daylength presented some degree of significant correlation when population values were correlated with environmental values that corresponded to a two-month backwards shift. Discussion and Conclusions Structure and dynamics of the population The population dynamics of Mazzaella cornucopiae from Prasiola Point generally followed a seasonal pattern between the late spring of 1993 and the summer of 1995. Holdfasts of M. cornucopiae are mostly perennial and bear fronds all year-round at Prasiola Point. The presence of small fronds (0-0.5 cm long) at all seasons indicates that fronds originated from holdfasts continuously, but production rates varied strongly with seasons. The period of strongest frond production started i n winter, and high production of new fronds continued throughout the spring until the summer. The largest fronds (3-5 cm long) became most abundant during mid-spring for both studied years. Frond density and stand biomass peaked i n late spring and summer. During mid-spring and summer, frond areas that were directly exposed to sunlight and the open air became heavily bleached, making them easier to be removed than healthy frond areas, presumably by wave action and herbivory by littorinid snails (Kim and De Wreede 1996b, R. A. Scrosati, pers. obs.). After the summer, a strong reduction i n the production rate of fronds, the loss of bleached frond areas, and an increasing frond mortality (the density of all size-classes progressively declined) determined minimum values of stand biomass for the winter. The annual cycle of strongest frond production and growth resumed when the initiation of new fronds started to increase during mid-winter. Since the summer of 1994, percent cover and mean frond length showed a decreasing trend until the last sampling date (July 1995). One year later, on 30 July 1996, percent cover was 23 ± 10 % (mean ± SEM) on the same six quadrats, which represented a record low value for any summer season studied for this area. The causes for this decrease i n cover of M. cornucopiae are not clear. Apparently, May 1995 and July 1996 were sunnier and warmer compared with similar months of previous years. At the end of July 1996, bleaching was unusually strong, in a way that even many holdfast areas were bleached, which was rare i n previous years. A simultaneous increase i n desiccation, irradiance, and temperature (see Chapter 4) may partially account for the high bleaching observed between 1995 and 1996. Seaweeds of the family Gigartinaceae mostly occur i n cold and warm temperate coasts of the world (Kim and Norris 1981, Hannach and Waaland 1986, Hommersand et al. 1993, 1994). The abiotic environment has varying degrees of seasonality at these sites, and population parameters of these algae are expected to vary generally following these seasonal patterns. Evidence of this was found for populations of Mazzaella cornucopiae from Hokkaido, Japan (Hasegawa and Fukuhara 1952, 1955, as Iridophycus cornucopiae), for Mazzaella capensis from South Africa (Bolton and Joska 1993, as Iridaea capensis), for Mazzaella flaccida from California (Foster 1982, as Iridaea flaccida), for Mazzaella laminarioides from Chile (Hannach and Santelices 1985, Santelices and Norambuena 1987, Westermeier et al. 1987, as Iridaea laminarioides), for Mazzaella splendens from western North America (Hansen 1977, 1981, Hansen and Doyle 1976, May 1986, all as Iridaea cordata, De Wreede and Green 1990, as Iridaea splendens, Dyck and De Wreede 1995), for Chondrus crispus from northeastern North America (Prince and Kingsbury 1973, Mathieson and Burns 1975, Tveter-Gallagher et al. 1980, Bhattacharya 1985) and from western Europe (Pybus 1977, Fernandez and Menendez 1991a, b), for Chondrus nipponicus from Japan (Masuda and Hashimoto 1993), for Chondracanthus pectinatus from the gulf of California (Pacheco-Ruiz et al. 1992, as Gigartina pectinata), for Gigartina polycarpa and Sarcothalia stiriata from South Africa (McQuaid 1985, as Gigartina radula and Gigartina stiriata, respectively), for Gigartina skottsbergii from Argentina (Piriz 1996), for Sarcothalia crispata from Chile (Hannach and Santelices 1985, Poblete et al. 1985, both as Iridaea ciliata), and for the related species Mastocarpus stellatus (Petrocelidaceae) from New Hampshire, United States (Burns and Mathieson 1972, as Gigartina stellata), and from Ireland (Pybus 1977, as G. stellata). For many of the above species, the crustose holdfast is perennial and gives rise to foliose fronds either annually or continuously throughout the year. Fronds of some species are mostly annually deciduous, with large fronds being rare i n winter, such as for Mazzaella flaccida (Foster 1982) and Mazzaella splendens (Norris and K i m 1972, Hansen 1977, Dyck et al. 1985, May 1986). The frond dynamics of Mazzaella cornucopiae is similar to that for other species such as Mazzaella laminarioides and Chondrus crispus, all with abundant year-round fronds (Santelices and Norambuena 1987, Westermeier et al. 1987, McLachlan et al. 1988). The timing of frond initiation and growth may vary both within and among species of these algae. For M. cornucopiae from Prasiola Point, the main period of production of new fronds started i n winter, but for M. cornucopiae from Hokkaido, Japan, the highest production of new fronds occurred during the fall (Hasegawa and Fukuhara 1955). For other morphologically similar algae, the main period of frond production occurs at varying times: i n the fall, for Mastocarpus stellatus from Ireland (Pybus 1977); i n the winter, for M. stellatus from New Hampshire (Burns and Mathieson 1972), for Mazzaella heterocarpa from Prasiola Point (R. A. Scrosati, pers. obs.), for Mazzaella splendens from California (Hansen and Doyle 1976) and from Barkley Sound (Dyck et al. 1985), and for Chondracanthus pectinatus from the gulf of California (Pacheco-Ruiz et al. 1992); i n winter-spring, for Chondrus crispus from New Hampshire (Mathieson and Burns 1975) and from Ireland (Pybus 1977); i n the spring, for M. stellatus from Britain (Marshall et al. 1949 i n Hansen and Doyle 1976), and for Mazzaella laminarioides from southern Chile (Gomez and Westermeier 1991). No seaweeds with this life form are reported to have the main period of production of new fronds i n the summer for temperate coasts. Production of reproductive fronds Reproduction by spores, as indicated by the presence of mature cystocarps and tetrasporic sori, followed a seasonal pattern, occurring mainly during fall and secondarily during winter. Spore production during fall and winter may be advantageous for an enhanced recruitment, compared with spore production during spring and summer. The removal of algal and animal biomass by strong wave action and the natural decrease of cover of these seasonal species during fall and winter open up space that can be colonized by spores (see also Kain and Norton 1990, Molenaar and Breeman 1994). In addition, low tides occur mainly at night time during fall and winter at Prasiola Point, so desiccation and high irradiance do not constitute stressful factors for sporelings that are recruited during that time of the year. Sporeling recruitment during fall and winter thus could be higher than a hypothetical recruitment during spring and summer. Temperature and daylength could be the main factors that promote reproduction during the cold season (Kain 1987, Kain and Bates 1993, Molenaar and Breeman 1994). For members of the Gigartinaceae, the production of reproductive fronds depends on the species and the site. Some species produce both cystocarpic and tetrasporic fronds all year-round, with varying degrees of seasonality, such as Mazzaella splendens from California (Hansen and Doyle 1976, Hansen 1977), Mazzaella laminarioides from southern C h i l e (Westermeier et al. 1987), Chondrus crispus from Massachusetts (Prince and Kingsbury 1973), and Gigartina skottsbergii from Argentina (Piriz 1996). Other species (or the above species occurring i n different habitats) show a seasonal reproduction, as it did Mazzaella cornucopiae from Prasiola Point. For these species, one (or both) reproductive phases is absent during certain seasons, such as C. crispus from New Hampshire (Mathieson and Burns 1975) and from Ireland (Pybus 1977), and M. laminarioides and Sarcothalia crispata from central Chile (Hannach and Santelices 1985, Santelices and Norambuena 1987). Gametophytertetrasporophyte ratio Gametophytic fronds of Mazzaella cornucopiae were more numerous than tetrasporophytic fronds at Prasiola Point at all seasons analyzed and under different degrees of wave exposure, as shown by resorcinol tests done on haphazard samples. This coincides with the predominance of cystocarpic fronds over tetrasporic fronds observed during the reproductive season (Fig. 2.5). Studies based on the presence of reproductive structures and/or resorcinol tests of field samples of fronds showed that the overall annual predominance of gametophytic fronds i n populations of members of the Gigartinaceae is common, as found for Mazzaella capensis, Mazzaella laminarioides, Mazzaella splendens, Chondrus crispus, Gigartina skottsbergii, and Sarcothalia crispata (Mathieson and Burns 1975, Bhattacharya 1985, Hannach and Santelices 1985, Poblete et al. 1985, Santelices and Norambuena 1987, Luxoro and Santelices 1989, Bolton and Joska 1993, Dyck and De Wreede 1995, Piriz 1996). Studies based on resorcinol tests done on frond samples of M. splendens and C. crispus that were collected only between spring and fall also showed a predominance of gametophytic fronds (Craigie and Pringle 1978, May 1986, Lazo et al. 1989, Scrosati et al. 1994, Lindgren and Aberg 1996). Add i t i o n a l information on the G:T ratio for M. cornucopiae exists for populations from Hokkaido, Japan (Hasegawa and Fukuhara 1952), but no clear pattern was evident, since counts of reproductive fronds, which may or may not accurately reflect the G:T ratio, showed an overall annual dominance of cystocarpic fronds over tetrasporic fronds at two sites, but no clear dominance by any phase occurred at two different sites. The field identification of different genets of Mazzaella cornucopiae was not possible, because of the high frond densities encountered, a possible coalescence of holdfasts, and the potential rupture of individual genets into more than one independent-life clonal fragment. To infer the G:T ratio at the genet level from the G:T ratio at the frond level, it is necessary to assume that fronds have been taken from separate genets. I tried to achieve this by collecting fronds that were separated from one another by, at least, 10 cm, when they were located i n the same large, continuous stand, or that belonged to different clumps, spatially separated from one another. This is a problem to consider for other species of the Gigartinaceae that may form extensive turfs as well, such as Mazzaella laminarioides and Chondrus crispus (R. A. Scrosati, pers. obs.). Based on a generally accepted idea that the expected ratio between gametophytes and tetrasporophytes should be 1 for populations of members of the Gigartinaceae, field and laboratory experiments have attempted to explain the frequent overall dominance of gametophytes i n these populations by the differential effects that abiotic and biotic factors have on both reproductive phases. Some factors that have been suggested as being partially responsible for a gametophyte predominance i n the Gigartinaceae are a higher resistance of gametophytes to desiccation (Luxoro and Santelices 1989, Olson 1990), a lower feeding preference for gametophytes by invertebrate herbivores (Hannach and Santelices 1985, Luxoro and Santelices 1989), different growth rates between phases (May 1986), a lower susceptibility of gametophytes to infection by endophytes (Correa and McLachlan 1991), and apomixis (Hannach and Santelices 1985). The differential effects that those factors have on the two reproductive phases have been suggested to be related to the distinct types of carrageenan that each phase produces i n this family (Luxoro and Santelices 1989): kappa-carrageenan by gametophytes and lambda-carrageenan by tetrasporophytes (Waaland 1975, McCandless and Craigie 1979, McCandless et al. 1983). However, the expected ratio between gametophytes and tetrasporophytes is not necessarily one. In the simplest possible scenario, each male and female gametophytic thallus (assuming dioecious species; no monoecious species have been reported for the Gigartinaceae) only produce one spermatium and one carpogonium, respectively, whereas a single tetrasporophytic thallus produces four tetraspores, as a result of meiosis from only one tetrasporangium. If, following the above model, we assume that reproduction only occurs at specific intervals that are equal for both phases and that both phases respond equally to the environment, the abundance of gametophytes at a time t (Gf) is given by: Gf = Gt-1 + ATt-l , whereas the abundance of tetrasporophytes at time t (Tj) is given by: Tt = Tt-1 + 0.5 Gt-1 Regardless of the initial relative abundance of both phases, successive iterations of this dynamic model lead to an equilibrium G:T ratio of about 2.8, which should be regarded as the expected ratio between gametophytes and tetrasporophytes instead of one. This value of 2.8 remains the same when spore output increases by the same multiple factor for both phases. Factors that would alter this expected value of 2.8 i n natural populations are: (1) differential spore output (fecundity) between both life-history phases, (2) differential spore viability (fertility) between both phases, (3) differential recruitment between both phases, and (4) differential mortality rates of adults of both phases (by either natural senescence, herbivory, and competition). It is then possible that the predominance of gametophytes reported for natural populations of some of the above species may not be a deviation from an expected G:T ratio previously assumed to be 1, but a result to be expected instead. For the population of Mazzaella cornucopiae from Prasiola Point, the G:T ratio found for the moderate wave exposure site on three sampling dates (2.1, 2.6, and 3.3) was similar to the expected value of 2.8, but the G:T ratio found for the wave-exposed site on two different sampling dates deviated more from 2.8 (1.5 and 4.3; additional variation also resulted from differences i n tidal height). Therefore, gametophytes and tetrasporophytes seem to behave differently at the wave-exposed site i n some of the four categories detailed i n the previous paragraph. Knowledge about the effects that certain biological and physical factors have on the spatial and temporal variation of the G:T ratio for M. cornucopiae and for related species allow us to speculate about possible reasons for the variation observed at the wave-exposed site. The degree of wave action may affect the G:T ratio of populations of Mazzaella splendens (a low-intertidal to shallow-subtidal species) from Barkley Sound. For this species, tetrasporophytes, which are more resistant to detachment by hydrodynamic forces than gametophytes, dominate i n wave-exposed sites, whereas gametophytes dominate i n more sheltered sites (Dyck et al. 1985, Shaughnessy et al. 1996). The G:T ratio for Mazzaella cornucopiae from the wave-exposed side of Prasiola Point was estimated only for two different dates and it was either higher (4.3) or lower (1.5), depending on the season, than the G:T ratio generally found for the moderate wave exposure site (between 2.1 and 3.3). A relationship between the degree of wave exposure and the G:T ratio is therefore as yet unclear for M. cornucopiae. However, a shift of phase predominance according to the degree of wave exposure has not been observed as for M. splendens. The largest fronds of M. cornucopiae are much smaller compared with those of M. splendens and are densely packed in extensive turfs, unlike those of M. splendens, which may confer on them a high resistance to detachment by waves (see also Chapter 4). Therefore, even if fronds of both reproductive phases of M. cornucopiae differ i n biomechanical properties, the degree of wave action would not be as important i n shaping its G:T ratio as it is for M. splendens. Tidal height, with its associated vertical variation i n physical and biological factors, affects the G:T ratio of Mazzaella cornucopiae at the wave-exposed site of Prasiola Point, with gametophytes being relatively more abundant than tetrasporophytes at higher levels. A similar pattern of distribution of life-history phases was also reported for other species of the Gigartinaceae such as Chondrus crispus and Mazzaella laminarioides. Differential resistance to desiccation or herbivory between both phases have been suggested as possible causes, based on laboratory experiments (Mathieson and Burns 1975, Craigie and Pringle 1978, Hannach and Santelices 1985, Luxoro and Santelices 1989). For C. crispus however, additional surveys by Bhattacharya (1985) and Lazo et al. (1989) d i d not show unequivocal differences i n distribution of phases at different tidal levels or depths. For M. cornucopiae from the coast of Oregon, gametophytes and tetrasporophytes showed a similar pattern of vertical distribution, and field experiments indicated that differential resistance to desiccation and grazing by limpets between phases may explain this pattern for that site (Olson 1990). Desiccation and herbivory could therefore play an important role i n the differential vertical distribution of phases of M. cornucopiae at Prasiola Point. For some species of the Gigartinaceae, a seasonal shift i n phase predominance occurs. Tetrasporophytes predominate over gametophytes during the winter for Mazzaella splendens from British Columbia (De Wreede and Green 1990, Dyck and De Wreede 1995), during the fall for Mazzaella capensis from South Africa (Bolton and Joska 1993), and also possibly during the fall for Sarcothalia crispata from central Chile (Hannach and Santelices 1985). This seasonal reversal of phase dominance has not been observed for Mazzaella laminarioides (Santelices and Norambuena 1987, Westermeier et al. 1987) or for Chondrus crispus (Mathieson and Burns 1975). Tetrasporophytes of M. splendens from California were reported to be the predominant phase between 1972-73 (Hansen and Doyle 1976), but resampling i n 1982-83 showed a high gametophytic predominance (Dyck et al. 1985). Based on these results, a seasonal alternation of phase predominance for species of "Iridaea" (except C. crispus, the above species belonged to the genus Iridaea in the past, but are currently recognized within the genera Mazzaella and Sarcothalia, Hommersand et al. 1993, 1994) was suggested to occur at latitudes higher than 48°, but phases would alternate annually or at longer periods i n populations from lower latitudes (De Wreede and Green 1990). The constant predominance of gametophytes of Mazzaella cornucopiae throughout the year at Prasiola Point, just above 48° N, indicates that factors other than latitude, such as thallus longevity, may determine the occurrence or not of a seasonal alternation of phase predominance for this and possibly other species. M. cornucopiae has a perennial crustose holdfast that continuously produces fronds. In Prasiola Point, the rocky substratum of the high intertidal zone is generally stable, so thalli stay i n place for long periods. Because of this, a change towards a tetrasporophyte predominance would take place only after a severe disturbance removes most of the existing thalli and a subsequent recruitment occurs mostly from carpospores, which seems an unlikely event on an annual basis at this site. Future research on the factors that affect the G:T ratio of different populations of Mazzaella cornucopiae and other species of the Gigartinaceae should consider a possible relationship between latitude and changes i n the life-history. For example, for species of the morphologically similar (only the gametophyte) genus Mastocarpus (Petrocelidaceae), direct development of gametophytes from carpospores predominates over heteromorphic alternation of generations at the northern part of their geographical distribution i n the northern hemisphere. In the southern limit of their distribution, direct development of gametophytes does not occur, and life cycles are only heteromorphic and alternate between foliose gametophytes and crustose tetrasporophytes (Polanshek and West 1977, Guiry and West 1983, Zupan and West 1988). For Mastocarpus, three processes have been proposed to explain the direct development of gametophytes. One is apomixis, which involves the production of dipl o i d gametophytes from parthenogenetic carpospores, the original d i p l o i d gametophytes being produced by diploid "tetraspores" after meiotic failure. Another process is somatic meiosis, which may occur within diploid crustose thalli, producing haploid gametophytic fronds afterwards (Chen et al. 1974, Polanshek and West 1977, Guiry and West 1983, Zupan and West 1988). A third process would occur through fertilization of carpogonia with self-produced spermatia (which thus bear a "female" genome), producing "female" d i p l o i d carpospores that originate diploid gametophytes i n turn (Maggs 1988). Relationship between population parameters and environmental variables The seasonality of seaweeds may be detenruned by (1) a direct response to environmental factors (such as temperature, irradiance, and nutrients), (2) environmental signals (daylength, temperature) that trigger a response i n the seaweed through a signal receptor and a circadian oscillator, and (3) an endogenous growth rhythm with circannual periodicity (Luning and torn Dieck 1989). Biological factors such as competition and herbivory, which may also be seasonal, can also affect the seasonal behaviour of seaweeds. The annual variation of growth rates of the sporophytes of some perennial kelps follow circannual rhythms, and they are of endogenous nature, i.e., produced by thalli themselves. As such, growth rates i n those kelps are not driven by environmental factors directly, but they are synchronized to a period of 12 months mainly by the annual cycle of daylength, and, to a lesser extent, possibly by temperature (torn Dieck 1991, Liming 1993, Luning and Kadel 1993). Circannual rhythms i n kelp sporophytes have been demonstrated i n laboratory cultures with constant conditions of daylength, temperature, and nutrient supply. They have not been observed yet i n red seaweeds, but this is possible, since they are thought to occur i n all eukaryotic organisms (Luning 1993). Environmental factors are thought to directly control the dynamics of organisms that occur mainly i n relatively unpredictable environments (Gwinner 1986 in Luning and torn Dieck 1989), which is not the case for Prasiola Point, so it is unlikely that Mazzaella cornucopiae dynamics is mainly a direct response to environmental variation. A circannual growth rhythm for M. cornucopiae might be indicated by the fact that daylength was the factor that showed the highest correlation with frond density and stand biomass, and daylength is the main synchronizing factor for circannual rhythms to a 12-month period (torn Dieck 1991, Luning 1993, Luning and Kadel 1993). However, the initiation of fronds during winter being a photoperiodic response, as occurs for other red algae (Kain and Norton 1990), remains possible until further research proves the contrary. Both air and sea surface temperature were significantly positively correlated to all population parameters separately (except air temperature with percent cover) when same-month data were considered. Confidence levels for correlation coefficients were similar to those obtained for daylength, but correlations using temperature data for previous months was less or non-significant. This suggests that the possible response of Mazzaella cornucopiae to temperature is shorter-term compared wi t h that to daylength. A s mentioned above, temperature could be a less important synchronizing factor than daylength if circannual rhythms are driving the seasonal variation of population parameters (Luning 1993). Up to a maximum level, increasing temperatures should benefit the growth of M. cornucopiae fronds through increased photosynthesis and physiological activity i n general, but seasonal changes i n temperature may not affect photosynthetic and respiration rates proportionally, because thermal acclimation of enzymes may occur (Lobban and Harrison 1994). M. cornucopiae is eurythermic, since it is subjected to strong annual oscillations of temperature. In winter, fronds must survive below-zero air temperatures, which occur during low tides at night time, and, i n summer, fronds have to withstand high temperatures, especially during sunny days at low tides. For other seaweeds, temperature and/or daylength are commonly associated with the production of biomass (Burns and Mathieson 1972, Mathieson and Burns 1975, Hansen 1977, McQuaid 1985, Poblete et al. 1985, Sheath et al 1985). Wave height was negatively correlated with all population parameters when same-month data were utilized, but population parameters and environmental variables showed low levels of significance, or were non-significant, when environmental data from previous months were considered. Wave action thus seems to have an immediate effect on the population dynamics of Mazzaella cornucopiae. Strong waves occur mainly during fall and winter at Prasiola Point. Increased hydrodynamic forces exerted on the fronds during that time may determine higher removal rates of fronds (Denny et al. 1989, Carrington 1990, Gaylord et al. 1994, Denny 1995, Shaughnessy et al. 1996) than during the summer, therefore being an important environmental factor that contributes to shape the seasonality of M. cornucopiae. Salinity was positively correlated to frond density and stand biomass when using same-month data. Higher salinity i n the summer may result from less rainfall than i n the winter, but the association with higher frond density and stand biomass might be coincidental and not causal. Intertidal seaweeds are euryhaline (Lobban and Harrison 1994), because they are adapted to the frequent osmotic stress that the high variation of the salinity of frond surface water poses. Salinity is low when there is heavy rainfall during low tides and it is high when evaporation of surface water occurs during daytime at summer low tides. Changes i n seawater salinity probably do not affect population parameters strongly, because of their small magnitude relative to changes i n the salinity of frond surface water during low tides. Summary of conclusions Thalli of Mazzaella cornucopiae are mostly perennial at Prasiola Point, and their long-lived, crustose holdfasts produce fronds throughout the year. Gametophytes predominate over tetrasporophytes, regardless of season or degree of wave exposure, although tetrasporophytes seem to be relatively more abundant i n the low zone than i n the high zone of the vertical range of distribution. The population of M. cornucopiae from Prasiola Point generally followed a seasonal pattern between the summer of 1993 and the summer of 1995. Major descriptors of the population such as percent thallus cover, frond density, mean frond length, and stand biomass commonly showed maxima i n spring-summer and minima i n winter. The period of active population growth occurred between winter and summer. Reproductive structures, cystocarps and tetrasporic sori, appeared mainly during the fall and, secondarily, during the winter. Daylength, temperature, and wave action seem to be important environmental factors that shape the seasonality of M. cornucopiae. Frond density and stand biomass were positively associated with both daylength and temperature, and negatively related to wave action, all with high significance levels. 90 80 > O O 70 60 50 40 i 1 1 1 r "i i r J A O D F A J A O D J M M J 1993 | 1994 | 1995 M o n t h Fig. 2.1: Seasonal variation of percent cover of Mazzaella cornucopiae from Prasiola Point (mean + SEM). The functional relationship was calculated by LOWESS. ^ 13 | 1 1 1 1 1 1 1 i 1 1 1 1 i r E 12 " t o T 2 10 V J A O D F A J A O D J M M J 1993 | 1994 | 1995 M o n t h Fig. 2.2: Seasonal variation of frond density (fronds cm - 2) of Mazzaella cornucopiae from Prasiola Point (mean + SEM). The functional relationship was calculated by LOWESS. 1 . 2 5 I i i i i i i i i i i i i i r J A O D F A J A O D J M M J 1993 | 1994 | 1995 M o n t h F i g . 2 . 3 : Seasonal variation of mean frond length (cm) of Mazzaella cornucopiae from Prasiola Point (mean + SEM). The functional relationship was calculated by LOWESS. F i g . 2 .4 : Seasonal variation of stand biomass (g cm"2) of Mazzaella cornucopiae from Prasiola Point (mean + SEM). The functional relationship was calculated by LOWESS. CM CD Q TJ C o 0.6 0.5 -\ E 0 co 1 0 - 4 H 0.3 H to 0.2 0.1 H Cys toca rp i c F ronds Tetrasporic Fronds 11 i—r i—r ~\—i—r J A O D F A J A O D J M M J 1993 | 1994 | 1995 M o n t h Fig. 2.5: Seasonal variation of the density of cystocarpic fronds and tetrasporic fronds of Mazzaella cornucopiae from Prasiola Point (mean + SEM). O CD L. 17.0 14.8 12.6 CD Q. | 10.4 8.2 6.0 I I I I I I I I I I I I I I 1 1 1 V rj O _ 1 I I I I O S E A H AIR _ - • ^ fo * ' © /*•' ii'' •' >> 1 1 1 1 i i i i i A J A O D F A J A O D F A J 1993 1994 1995 M o n t h Fig. 2.6: Seasonal variation of mean monthly air and sea surface temperature (°C) at La Perouse Bank buoy. Fig. 2.7: Seasonal variation of mean monthly wave height (m), from crest to trough, at La Perouse Bank buoy. 32 i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i i i i r 8 31 & 30 § 29 28 27 26 i i i i i i i i i i i i i i i i i i i i i i i i i i i A J A O D F A J A O D F A J 1993 I 1994 I 1995 M o n t h Fig. 2.8: Seasonal variation of mean monthly salinity (%o) at Amphitrite Point. Fig. 2.9: Seasonal variation of mean monthly daylength (h) at Tofino 46 Date Function 2 r n 29 A p r i l 1994 B = 0.011 X L " 0.86 167 5 December 1994 „ n n n M 2.813 B = 0.004 x L 0.92 162 30 January 1995 2 728 B = 0.004 x L ' 0.92 168 3 A p r i l 1995 1.977 B = 0.008 x L 0.93 172 21 May 1995 „ 1.948 B = 0.009 x L 0.89 164 14 July 1995 B = 0.012 x L ™ 0.94 180 28 October 1995 ^ ™ ~ T 2.569 B = 0.007 x L 0.91 118 Table 2.1: Functional relationships between frond biomass (B, i n g) and frond length (L, in cm) for Mazzaella cornucopiae from Prasiola Point. Date of frond collection for determining B-L relationships Sampling dates to which B-L relationships were applied 29 A p r i l 1994 A p r i l 1994 5 December 1994 December 1993 and 1994 30 January 1995 February 1994 and January 1995 3 A p r i l 1995 March/April 1995 21 May 1995 June 1993 and May 1995 14 July 1995 August 1993 and 1994, June 1994, July 1995 28 October 1995 October 1993 and 1994 Table 2.2: Dates of frond collection for the calculation of biomass-length relationships and dates to which these relationships were applied to estimate stand biomass. 1-way A N O V A R for Percent Cover Source SS DF MS F p H-Fp Time 5427.57 13 417.51 3.16 0.001 0.004 Error 8583.00 65 132.05 Huynh-Feldt epsilon = 0.747 Table 2.3: Summary table for one-way repeated-measures A N O V A for percent cover of Mazzaella cornucopiae from Prasiola Point. 1-way A N O V A R for Frond Density Source SS DF MS F p H-Fp Time 205.09 13 15.78 3.40 0.001 0.007 Error 301.60 65 4.64 Huynh-Feldt epsilon = 0.548 Table 2.4: Summary table for one-way repeated-measures A N O V A for frond density (fronds cm - 2) of Mazzaella cornucopiae from Prasiola Point. 1-way A N O V A R for Mean Frond Length Source SS DF MS F p H-Fp Time 0.601 13 0.046 2.04 0.031 0.031 Error 1.475 65 0.023 Huynh-Feldt epsilon = 1.000 Table 2.5: Summary table for one-way repeated-measures A N O V A for mean frond length (cm) of Mazzaella cornucopiae from Prasiola Point. 1-way A N O V A R for Stand Biomass Source SS DF MS F p H-Fp Time 0.109 13 0.008 12.09 <0.001 <0.001 Error 0.045 65 0.001 Huynh-Feldt epsilon = = 0.986 Table 2.6: Summary table for one-way repeated-measures A N O V A for stand biomass (g cm - 2) of Mazzaella cornucopiae from Prasiola Point. Collection Date Site %G-%T G:T ratio n 29 A p r i l 1994 E 77-23 3.3 135 20-21 May 1995 E 68-32 2.1 135 20 May 1995 W (all) 60-40 1.5 100 20 May 1995 W (high) 62-38 1.6 50 20 May 1995 W (low) 58-42 1.4 50 28 October 1995 W (all) 81-19 4.3 100 28 October 1995 W (high) 90-10 9.0 50 28 October 1995 W (low) 72-28 2.6 50 22 December 1995 E 72-28 2.6 105 Table 2.7: Relative abundance of gametophytic (G) and tetrasporophytic (T) fronds of Mazzaella cornucopiae from Prasiola Point, expressed as percentage of G and T and as the G:T ratio. The E side of Prasiola Point is generally protected from large waves, whereas the W side is usually exposed to strong wave action. 'Low' and 'high' refer to the low and the high zone of the M. cornucopiae belt, respectively, and 'all' represents both data sets combined. Air Temperature Sea Surface Temperature Wave Height Salinity Daylength PC 0.54 0.60 * -0.57 * -0.04 0.38 FD 0.73 ** 0.78 ** -0.78 ** 0.71 ** 0.80 *** MFL 0.68 * 0.62 * -0.25 0.07 0.35 SB 0.83 *** 0.86 *** -0.74 *** 0.54 * 0.91 *** PC-1 0.24 0.25 -0.25 -0.26 0.31 FD-1 0.57 * 0.59 * -0.63 * 0.39 0.86 *** MFL-1 0.40 0.45 0.01 0.03 0.41 SB-1 0.58 * 0.62 ** -0.49 * 0.23 0.91 *** PC-2 -0.06 -0.03 0.01 -0.40 0.17 FD-2 0.09 0.26 -0.39 0.01 0.70 ** MFL-2 0.27 0.26 -0.02 -0.21 0.39 SB-2 0.06 0.18 -0.21 -0.24 0.68 ** Table 2.8: Pearson correlation coefficients between population parameters (PC = percent cover, FD = frond density, MFL = mean frond length, and SB = stand biomass) for Mazzaella cornucopiae from Prasiola Point and five environmental variables measured near Prasiola Point (see text for details). -1 indicates that population data were correlated to environmental data for the previous month; -2 indicates that the lag period was two months. * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001. Chapter 3 The relationship between stand biomass and frond density Introduction The general structure and dynamics of the population of Mazzaella cornucopiae from Prasiola Point have been discussed i n the preceding Chapter. This Chapter w i l l focus on the interactions that occur among fronds of M. cornucopiae. The analysis of the relationship between the density of fronds and the total biomass of a given stand during the growth season offers a first approach to understanding such interactions. Below is a general account of what is presently known about the relationship between biomass and density for terrestrial plants and for the few seaweeds where it has been investigated, which is given as the basis for establishing the objectives of this Chapter. For terrestrial plants that are actively growing i n monospecific, even-aged, and crowded stands, density-dependent mortality starts to operate on the smallest plants at a certain point i n time, as they are negatively affected by increasing plant sizes. During this stage, which is known as self-thinning, the total biomass of a stand continues to increase, because the surviving plants' growth outweighs biomass losses due to the death of the smallest plants. The temporal variation of the relationship between log stand biomass (expressed per unit area) and log plant density is described by a straight line with negative slope. This line is called the self-thinning line, and its slope and intercept depend on both the population and the site considered (Zeide 1985, Weller 1987, 1990). Self-thinning lines may be considered "dynamic" relationships between biomass and density (Weller 1989, 1990), since data points are repeatedly measured from the same plots. Clonal plants are those which grow and propagate by self-replication of genetically identical units, termed ramets, which can potentially or actually function and survive on their own if they become separated from one another by natural processes or injuries (Jackson et al. 1985, van Groenendael and de Kroon 1990, Herben et al. 1994). For vascular clonal plants from seasonal habitats, shoots (ramets) generally do not undergo self-thinning during growth (Hutchings 1979, Pitelka 1984, Dickerman and Wetzel 1985, Lapham and Drennan 1987, Room and Julien 1994, de Kroon and Kalliola 1995). Red seaweeds such as those of the family Gigartinaceae are clonal. The genet (sensu Harper 1977) is the entire thallus derived from a single spore, and it is composed of several foliose fronds produced by a crustose holdfast. Fronds may be considered as ramets, because of their potential capacity of independent life, together with an associated portion of holdfast. The only clonal seaweed where the temporal relationship between frond density and stand biomass was studied is the red species Gelidium sesquipedale (Gelidiales) from Portugal (Santos 1995). During the growth season for this species, fronds do not undergo self-thinning, but their density is positively correlated to stand biomass. Its habitat is subtidal, and it is not known what occurs for intertidal species. A positive relationship between stand biomass and frond density was found for two intertidal clonal algae: Chondrus crispus (Gigartinaceae) and Mastocarpus stellatus (Petrocelidaceae) from Ireland (Pybus 1977, M. stellatus as Gigartina stellata). This is indirect evidence that self-thinning does not occur among fronds of these two algae, because biomass-density pairs were measured at single instants for different clumps, and a "static" biomass-density relationship such as this one may not reflect correctly the nature of the dynamic relationship (Weller 1989). The first objective of this Chapter is to test for the occurrence of self-thinning among fronds of Mazzaella cornucopiae during its growth season (winter to summer at Prasiola Point, Chapter 2). Results w i l l be used to discuss the interactions that occur among fronds during their growth. The so-called "ultimate thinning line" (White 1985) describes the maximum average biomass of individual plants possible for any plant density, and it is thought to constrain all plant populations, even those that do not self-thin (Weller 1989). Since it is a static constraint, the name "thinning" applied to this line i n the past does not describe accurately its meaning. This line is the upper boundary of the interspecific biomass-density "band", which includes all of the biomass-density pairs used to calculate self-thinning lines for different terrestrial plants, and its slope is negative (Weller 1989). It has been suggested that ramets of clonal plants with a seasonal behaviour attain their maximum average biomass when stands approach this "ultimate biomass-density line" for a given density (usually at the end of the growing season). Ramets are thought to almost stop their growth at this stage, simultaneously with the occurrence of sexual reproduction and senescence. This may contribute to prevent self-thinning among ramets because the ultimate biomass-density line is not surpassed (Westoby 1984, de Kroon 1993, de Kroon and Kalliola 1995). If it is found that self-thinning does not occur among fronds of M. cornucopiae, combinations between average frond biomass and frond density w i l l be compared with the proposed ultimate biomass-density line to determine the possible validity of the above hypothesis for the frond dynamics of M. cornucopiae. The second objective of this Chapter is to determine a possible annual dynamic relationship between stand biomass and frond density for Mazzaella cornucopiae, i.e., considering both the growth and decline periods for this species. This is relevant to the estimation of stand biomass for different purposes, whether they are of scientific, conservation, or commercial nature. Usually, the estimation of algal biomass is time-consuming because it may require a measurement of the size structure of the stand, it may not be very accurate when plotless techniques are used (Littler and Littler 1985), or it may involve destructive methods, such as harvesting of parcels (De Wreede 1985). If found, a biomass-density relationship could be a quicker, non-destructive tool to estimate the stand biomass of M. cornucopiae just from a count of fronds i n a given area. Materials and Methods Study site and sampling program The field study was carried out i n the moderate wave exposure side (E) of Prasiola Point. Seven permanent transects (each one delimited by two concrete nails at both extremes) were placed over stands with high cover of Mazzaella cornucopiae, and the location of one permanent, 10 cm x 10 cm quadrat was randomly established for each transect. Frond density and stand biomass were repeatedly estimated for these quadrats from June 1993 until July 1995. There were 14 sampling dates: 2-6 June 93, 16-20 August 93, 15-18 October 93,10-12 December 93, 24-27 February 94, 25-29 A p r i l 94, 21-25 June 94, 18-22 August 94, 5-9 October 94, 2-6 December 94, 28-31 January 95, 30 March 95, 13-17 May 95, and 10-14 July 95. Given the limited time available for fieldwork on each date, data were estimated from six 2 cm x 2 cm, randomly chosen subquadrats on each quadrat (total area of 6 subquadrats = 24 cm 2). The same subquadrats were examined on successive sampling dates. A l l of the fronds present i n subquadrats were counted, and their length was measured to the nearest 5 mm. Stand biomass was estimated by applying power relationships between frond length and frond biomass, determined for different dates (Table 2.1), to the frond density and mean frond length found for the quadrats. Frond biomass-length relationships were not examined for many of the earlier sampling dates, so the relationships measured mainly during the second year of the study were applied to comparable dates for which biomass-length measurements were not calculated (Table 2.2). Statistical analysis • Choice of stand biomass over mean frond biomass Self-thinning among terrestrial plants has been studied initially by determining the correlation between mean plant biomass and plant density during growth, i n an attempt to detect if surviving plants grow as some plants die and also to study their growth rates (Yoda et al. 1963 and papers cited i n Weller 1987). Weller (1987) showed that there are problems of interpretation of data associated with this method, suggesting the use of stand biomass instead of mean plant biomass as a valid alternative to test for self-thinning. However, a recent paper (Petraitis 1995) questioned Weller's (1987) conclusion, and claimed that the choice of either mean or total biomass depends solely oh the research question of interest and is not affected by other considerations. The discussion of whether to choose mean or total plant biomass for the study of dynamic biomass-density relationships is relevant for this Chapter, with the only difference being that 'frond' and not 'plant' w i l l be considered for Mazzaella cornucopiae. This section w i l l show that Petraitis' (1995) conclusion presents problems, which have to be clarified again to prevent doing an inadequate test of self-thinning for M. cornucopiae or for other seaweeds (for example, a recent analysis of seaweed self-thinning done by Flores-Moya et al. -1996- also considered mean plant biomass). A simultaneous increase i n mean plant biomass and decrease i n plant density could be a result of two different processes that are not resolved by the functional relationship between both variables (Weller 1987): (1) mean biomass may increase when small plants die even if the surviving, larger individuals do not grow, or (2) mean biomass may increase with an increase i n stand biomass (through the growth of surviving plants) simultaneously with the decrease of the number of plants. Only the second case corresponds to what is understood as self-thinning. However, even for this second case there is an associated problem, since the variable "mean plant biomass" increases faster than the real biomass of surviving individuals. Thus, the functional relationship between mean plant biomass and plant density may not tell exactly if surviving plants are growing or not, therefore failing to detect the active growth period that defines self-thinning, and may not describe accurately plant growth rates when they are growing, regardless of the occurrence of self-thinning. The study of the relationship between total •9 biomass and plant density enable us to detect more appropriately if self-thinning is occurring or not (Weller 1987). For the reasons discussed above, the dynamic biomass-density relationship for M. cornucopiae w i l l be analyzed here by studying the relationship between total stand biomass and frond density, discarding the use of mean frond biomass. • Functional relationship between frond density and stand biomass The linear relationship between the logio of stand biomass and the logio of frond density was calculated using principal components analysis (PCA). This is one of the appropriate techniques when a Model I regression can not be applied such as in this case, because the measurement of both variables is subject to error, which is not experimentally controlled (Weller 1987, Sokal and Rohlf 1995). The temporal variation of the relationship between biomass and density was plotted for each quadrat separately. The Pearson correlation coefficient was calculated for every case, and its significance was assessed using the appropriate table given by Fisher and Yates (1963). The percentage of variance explained by the first principal component was discarded as an appropriate measure of bivariate linear association, since it is never less than 50%, even if the two variables are uncorrelated (Weller 1987). The slopes of the relationships were compared to one another by calculating confidence limits on the slopes (Weller 1987, Sokal and Rohlf 1995). Statistical analyses were conducted using SYSTAT 5.2.1 for Macintosh (Wilkinson et al 1992). • Random samples and randomization tests If the Pearson product-moment correlation coefficient, r, is computed only for describing the association between two samples, no assumptions regarding the sampling procedure or the population from which data were sampled are necessary. However, if r is used to test the hypothesis that there is no linear relationship between the two variables (population rho=0), then it has to be assumed that data pairs have been sampled randomly from a bivariate-normal distribution (Howell 1992). One of the objectives of this Chapter is to investigate the possible existence of an annual dynamic relationship between stand biomass and frond density for Mazzaella cornucopiae. For this purpose, one possibility is that measurements of biomass and density are taken randomly from different nearby plots through the year. Then, the assumption of random sampling necessary for doing parametric significance tests for the correlation between biomass and density is met. Preliminary observations indicated that even nearby stands with an area of 100 cm 2 would give different biomass-density relationships. Thus, to investigate the existence of a dynamic biomass-density relationship and its possible variation among sites, data were repeatedly collected from the same seven experimental quadrats mentioned above. With this procedure, the random sample condition no longer holds. However, significance tests for correlation may be carried out by performing randomization tests (Edgington 1987, Manly 1991). Randomization tests are distribution-free statistical tests which can be applied to test for the significance of any statistic (Edgington 1987, Manly 1991). For randomization tests, assumptions regarding random sampling, normality, and homogeneity of variance are not required (Edgington 1987, Manly 1991). The significance of the desired statistic is calculated as follows: after the statistic is computed for the data set, data are randomly rearranged repeatedly, and the test statistic is calculated for each data permutation. No tables are used for determining significance, because all of the data permutations constitute the reference set. The significance or p value for a one-sided test is the proportion of test statistics which are greater than or equal to (or less than or equal to, depending on the statistic) the statistic obtained from the original sample (Edgington 1987, Manly 1991). When random sampling is not done and randomization tests are used to calculate the significance of a statistic, statistical inferences can only be applied to the subjects from which data have been taken. Generalization of results to other subjects, such as the entire population or a different population for the same species, has to be done through nonstatistical inference, i.e., without a basis i n probability (Edgington 1987, Manly 1991). However, this is a standard procedure i n scientific research. Ideally, if a certain characteristic is studied for a given species, random samples should be taken from its entire biogeographical range to allow for a generalization of results through statistical inference. However, ideal random sampling is usually not possible to do because of restrictions of funds, time, or other resources. In spite of this, most of the hypotheses concerning entire populations or the species itself are commonly tested using parametric statistical tests, which require random sampling. Generalization of results to a broader range than that which was sampled is done also by nonstatistical inference, based on the similarity between the studied subjects and others that were not studied (Edgington 1987, Manly 1991). Because of the lack of random sampling i n studying the dynamic biomass-density relationship for Mazzaella cornucopiae, randomization tests were carried out to test for significant linear correlations between frond density and stand biomass for the seven quadrats that were monitored. Analyses were done with the program "Randomization tests", developed by Eugene S. Edgington (University of Calgary, AB) for DOS. Given the number of data pairs for each replicate quadrat (n=13-14), the number of posible data permutations (2 n) is high, so 1000 random permutations were done for each quadrat. Probability values for correlation coefficients using randomization tests were compared with those obtained by using Fisher and Yates' (1963) table to see how non-randomness affects parametric tests of significance for this case. Edgington (1987, p. 11) quoted Cotton and Fisher when saying that, i n the absence of random sampling, "determining significance by the use of conventional published statistical tables is of questionable validity until it has been shown that the significance so obtained would agree closely with that given by randomization tests". Comparison with the "ultimate biomass-density line" for terrestrial plants The ultimate biomass-density line for terrestrial plants is thought to be approximately described by the relationship Y = -1.5 X + 5.0, where Y = logio average plant biomass and X = logio plant density (White 1985, Weller 1989). Combinations between average frond biomass and frond density for Mazzaella cornucopiae w i l l be plotted together with this hypothetical line to determine if this constraint in biomass for a given density applies also to this clonal seaweed. Results Test for self-thinning among fronds of Mazzaella cornucopiae During the two years of study of the population of Mazzaella cornucopiae from Prasiola Point, frond density oscillated between 3 and 17 fronds cm - 2 (7.9 ± 0.3 fronds cm - 2, mean ± SEM, n=98), whereas total biomass ranged between 3 and 205 mg cm - 2 (57 ± 5 mg cm - 2, n=98). Maximum values were recorded for spring-summer, and minimum values, for winter. The time trajectory of combinations of monthly means (n=7) of logio stand biomass and logio frond density during the growth seasons of 1994 and 1995 are shown i n Figs. 3.1 and 3.2. Log stand biomass presented a significant positive linear correlation with log frond density for the growth period of 1995 (r=0.98, p<0.05, df=2), but, although the correlation coefficient was high and positive for the growth period of 1994, it was not significant (r=0.68, p>0.05, df=2). Stand biomass never increased significantly together with a significant decrease i n frond density for any of the 2-month intervals analyzed (p<0.05, paired f-tests), which indicates that self-thinning did not occur i n the population of M. cornucopiae during the growth periods of 1994 and 1995. The annual biomass-density relationship and its site-dependence Log stand biomass presented a significant positive linear correlation with log frond density for the seven experimental quadrats analyzed separately for the entire study period (Figs. 3.3 to 3.9). Spring-summer values were usually found at the upper-right area of the data cloud, whereas winter values were commonly found at the lower-left side. The significance of Pearson correlation coefficients obtained by randomization tests was the same as those obtained using parametric tables. Stand biomass is expressed i n g m~2 and frond density is expressed as fronds m - 2 to allow a comparison with results reported by previous researchers mainly for land plants. The slopes determined by P C A for the seven experimental quadrats are statistically similar to one another (p<0.05). Comparison wi t h the "ultimate biomass-density line" for terrestrial plants Approximately 6 0 % of the combinations between average frond biomass and frond density for Mazzaella cornucopiae from the moderate wave exposure side (E) of Prasiola Point lays above the hypothetical ultimate biomass-density line proposed for terrestrial plants (Fig. 3.10). In addition, the average biomass of fronds of M. cornucopiae measured from 52 small clumps collected between October 1993 and August 1994 from the wave-exposed side (W) of Prasiola Point and from Nudibranch Point (about 400 m from Prasiola Point) was always higher than the value predicted by the ultimate biomass-density line for any density (Fig. 3.11). Thus, the constraint i n average biomass for any density predicted by the ultimate biomass-density line developed from work with land plants does not apply for crowded stands of fronds of M. cornucopiae. Discussion and Conclusions Randomization tests and parametric tests Parametric tests and randomization tests of the significance of a given test statistic usually show good agreement when they use equivalent statistics and when observations seem to come from a normal distribution. This agreement gives some justification for using parametric tests on non-random samples (Manly 1991). The similar significance level obtained through both methods for the linear correlation between biomass and density supports, under this approach, the valid use of parametric tables for the significance of r for this data set, even considering that sampling was not done at random. It is possible that this conclusion can be also extended to the analysis of self-thinning for land plants, where estimation of biomass and density is also usually done repeatedly on the same experimental plots. Lack of self-thinning among fronds of Mazzaella cornucopiae Stand biomass of Mazzaella cornucopiae never increased w i t h a concomitant decrease of frond density for any of the 2-month intervals studied. This shows that fronds did not undergo self-thinning at the natural densities observed. Fronds of M. cornucopiae are produced continuously throughout the year (Chapter 2), which is common for other species of the Gigartinaceae (Gomez and Westermeier 1991, Scrosati et al. 1994), and also for ramets of clonal vascular plants (Cook 1985, Dai and Wiegert 1996). The continuous production of fronds would appear to guarantee the lack of self-thinning for M. cornucopiae, because of constant addition of new fronds to the stand. However, another possible outcome could have been that large fronds had suppressed the production and growth of the smallest fronds (through shading, for example) once large fronds attained a certain size, causing mortality of the smallest fronds. Then, overall frond density could have decreased (with a simultaneous increase i n stand biomass due to the continuous growth of large fronds) if the mortality of small fronds had been higher than the production of new fronds. A lack of self-thinning among fronds was suggested for Mazzaella laminarioides from Chile (Martinez and Santelices 1992, as Iridaea laminarioides). The evidence was indirect however, since biomass and density were measured for different clumps at single points i n time, and the temporal variation of the biomass-density relationship was not followed for indi v i d u a l clumps. The static biomass-density relationship may not tell whether stands are undergoing self-thinning or not (Weller 1989). The analysis of the dynamic relationship between stand biomass and frond density for M. laminarioides may provide results similar to those found for Mazzaella cornucopiae, since both species are morphologically similar (R. A. Scrosati, pers. obs.). Until recently, self-thinning was believed not to occur among ramets of clonal terrestrial plants (Hutchings 1979, Pitelka 1984, Dickerman and Wetzel 1985, Lapham and Drennan 1987) or even clonal aquatic ferns (Room and Julien 1994). Physiological integration of ramets occurs at varying degrees within clonal plants (Pitelka and Ashmun 1985, Marshall 1990, Hester et al. 1994, Stuefer et al. 1994, Alpert 1996). For most of the clonal plants studied to date, interconnections among ramets may help to prevent mortality of the smallest ramets, up to a certain point (de Kroon 1993), by allowing translocation of assimilates from large ramets when the small ramets become shaded at high densities, and it may also allow density-dependent regulation of ramet production within the clone to prevent an overproduction of ramets (Hutchings 1979, Pitelka 1984, Pitelka and Ashmun 1985, Marshall 1990). Density-dependent regulation of ramet production may be also achieved through changes i n light quality that occur below the canopy at high ramet densities (de Kroon 1993, Murphy and Briske 1994). Fronds of clonal red seaweeds that arise from the same holdfast may have a certain degree of physiological integration (Maggs and Cheney 1990), although long-distance transport of large molecules may not occur as it does in large brown algae (cf. Pueschel 1990, Schmitz 1990, Penot 1992). Recent evidence (Gonen et al. 1996) indicates that translocation of photoassimilates occurs within thalli of the red alga Gracilaria cornea (Gracilariales). For Mazzaella cornucopiae, an indirect indication that physiological integration may occur is the existence of frond-like short (<5 mm) projections that occasionally grow downwards from the lower surface of holdfasts and penetrate crevices. They are almost colorless, and the very low to n i l irradiance that occurs at these microsites suggests that they should be nourished from the upper regions of thalli, involving a certain degree of translocation. The possible physiological integration among fronds of M. cornucopiae may contribute to prevent self-thinning by enhancing both survival and growth rates of the smallest fronds i n crowded stands. Density-dependent production of fronds occurs during the growth period of Mazzaella cornucopiae (Chapter 4), so this factor may also be partially responsible for the lack of self-thinning among fronds. A third reason that may partially explain the lack of frond self-thinning for Mazzaella cornucopiae is a possible acclimation of the smallest fronds to the low irradiance observed close to the holdfasts of crowded stands (3 to 30 umol m - 2 s _ 1, depending on the closeness to the holdfast and the irradiance outside the algal turf, Chapter 4). If the light compensation point for the smallest fronds is lower than such irradiance value, then their growth would be guaranteed, thus preventing self-thinning, regardless of the intensity of physiological integration or the degree of density-dependent frond production. On the contrary, if the light compensation point for the smallest fronds is higher than the irradiance observed close to the holdfasts, then the causes of the lack of frond self-thinning might be related to a possible physiological integration among fronds or to the density-dependent production of fronds that occurs during the growth period. The light compensation point and understory irradiance for high intertidal seaweeds have been so far only reported for the tropical red species Ahnfeltiopsis concinna from Hawaii (Beach and Smith 1996a, 1996b). For this alga, the compensation point for understory tissue (26.3 umol m~2 s"1, measured underwater) is higher than the irradiance observed at the understory (usually less than 10 umol n r 2 s _ 1). Only one high intertidal species has been investigated on this respect, but it nonetheless suggests that crowding could limit the growth of the smallest fronds of high intertidal red algae through shading. If we consider that the reported compensation point for A. concinna could be higher as desiccation progresses during low tides (net photosynthesis generally decreases as desiccation increases; see Chapter 4), the idea of growth limited by low irradiance is reinforced. Comparison w i t h the "ultimate biomass-density line" for terrestrial plants It has been suggested that ramets of seasonal clonal plants would attain their maximum average biomass when stands approach the "ultimate biomass-density line" for a given density (Westoby 1984, de Kroon 1993, de Kroon and Kalliola 1995), which would usually occur at the end of the growth period for seasonal plants. Sexual reproduction and senescence of the older ramets mainly occur at this time, and ramets would almost stop growing. This would contribute to prevent self-thinning because average ramet biomass does not increase beyond the value predicted by the ultimate biomass-density line for a given density. This hypothesis has not been sufficiently tested, so it has not been fully accepted yet (de Kroon 1993). Combinations between average frond biomass and frond density for Mazzaella cornucopiae are not constrained by the ultimate biomass-density line proposed as universal for the plant kingdom (White 1985, Weller 1989). Additional measurements of biomass and density done i n 561 clumps of Chondrus crispus from Prince Edward Island and 32 clumps of Mastocarpus papillatus (Petrocelidaceae) from Vancouver harbour (R. A. Scrosati, unpublished data) lead me to the same conclusion. Whether there is a different constraint for biomass-density combinations that is specific for clonal red algae remains to be determined. Various terrestrial plant species had to be studied to notice the existence of the ultimate biomass-density line, although the functional relationship is not accurately known yet (Weller 1989). This suggests that several algal species have to be studied to determine possible constraints for the maximum average biomass of fronds for any given density. For the moment, the lack of self-thinning among fronds of Mazzaella cornucopiae cannot be explained by the "ultimate biomass-density" hypothesis as it is presently known. Recently, self-thinning was detected among shoots (ramets) of the perennial clonal grass Gynerium sagittatum (Poaceae) from tropical floodplains from Amazonian Peru (de Kroon and Kall i o l a 1995). These authors suggested that the seasonality of shoot growth is a major influence on the expression of self-thinning i n clonal plants: periodic density-independent mortality would not allow shoots to reach a self-thinning stage. A low seasonality of the environment (such as i n the Amazon) would allow ramets of clonal plants to reach self-thinning after years of continuous growth (de Kroon 1993, de Kroon and Kalliola 1995). To fully accept this hypothesis, I believe that it is necessary to consider the degree of physiological integration among ramets that exists after prolonged growth. If physiological integration is reduced to a minimum or nonexistent level after prolonged growth, ramets would relate to one another as non-clonal plants do, thus leading to mortality of small "disconnected" ramets and, therefore, self-thinning. If physiological integration remains strong, probably self-thinning among ramets may not occur at all even after prolonged growth. The frond dynamics of Mazzaella cornucopiae is markedly seasonal at Prasiola Point, and it is generally coupled with environmental variables (Chapter 2). Frond density and stand biomass both reach their highest levels i n spring-summer, after which they decrease (Chapter 2). Under hypothetical constant conditions that favour growth, would fronds undergo self-thinning after prolonged growth as predicted above? To answer this question through field experimentation, conditions which favour growth should be kept constant throughout time. However, this is very difficult, if not impossible, to achieve i n the field, because of the strong seasonality of the environment i n the study area (Chapter 2). Cultivation i n the laboratory under controlled conditions, considering periods of emergence and submergence characteristic of the intertidal zone, may provide a more feasible way to answer this question i n the future. The annual biomass-density relationship and its site-dependence The dynamic relationship between stand biomass and frond density for Mazzaella cornucopiae throughout two years is described by a straight line with positive slope i n a bilogarithmic scale. Combinations of stand biomass and frond density "moved" back and forth along fitted lines according to seasons, reaching the highest values i n spring-summer and the lowest ones i n winter. Cyclical seasonal movements of biomass-density combinations were also observed for stands of shoots (ramets) of clonal perennial herbs (Hutchings 1979), for which there was no significant correlation between mean shoot biomass and shoot density. The slopes of the biomass-density relationships determined for Mazzaella cornucopiae from the seven experimental quadrats are not significantly different from one another. This suggests that there is only one slope for the dynamic relationship between stand biomass and frond density for this population. In that way, stand biomass could be predicted from a count of fronds i n a given plot. However, the seven slopes obtained range from 2.9 (quadrat #7) to 7.4 (quadrat #2). This means that stand biomass is expected to increase about 3 times per unit change of log frond density i n quadrat #7, but the increase i n stand biomass is expected to be about 15 times for quadrat #2. The statistical similarity among the seven seemingly different slopes results from the variation observed for the data (correlation coefficients for biomass-density relationships ranged between 0.48 and 0.80). Confidence limits for slopes are inversely related to correlation coefficients. Moderate correlation coefficients result i n the overlapping of confidence intervals for slopes, and therefore i n a failure to detect significant differences between slopes. The differential increase i n biomass per unit change of log frond density between the quadrats with the most different slopes suggests that the dynamic biomass-density relationship might be site-dependent, i n spite of statistical results obtained from 14 sampling dates. This would agree with the fact that a given species of terrestrial plant presents different self-thinning lines depending on site conditions (Weller 1987). The several biotic and abiotic factors that affect M. cornucopiae stands are continuously varying in their relative intensity, which may be responsible for the high variability i n biomass-density combinations. As a result, the dynamic biomass-density relationship would be different each year and would also depend on the site considered. Therefore, it seems unlikely that even adding data for more sampling dates would lead to an accurate estimation of stand biomass for a given quadrat at any time from biomass-density relationships obtained for specific quadrats and for specific years. The positive dynamic relationship between stand biomass and frond density was also reported for a subtidal population of the clonal red seaweed Gelidium sesquipedale (Gelidiales) from Portugal (Santos 1995). For the clonal algae Chondrus crispus and Mastocarpus stellatus from Ireland, stand biomass is also positively correlated with frond density (Pybus 1977, M. stellatus as Gigartina stellata], but data points were collected from different stands at single instants. This constitutes an indirect evidence that a positive dynamic relationship might occur i n these intertidal species as well. The dynamic biomass-density relationship was investigated for an additional number of clonal algae (Cousens and Hutchings 1983), but mean frond biomass was used instead of stand biomass, which presents limitations (see Materials and Methods in this Chapter). Thus, it is not possible to draw clear conclusions from such analysis. For non-clonal seaweeds, evidence that self-thinning does occur among actively growing individuals has been found for the brown algae Egregia menziesii (Black 1974), Carpophyllum maschalocarpum and Sargassum sinclairii (Schiel 1985), Macrocystis pyrifera (Dean et al. 1989), Pterygophora californica (Reed 1990), Fucus gardneri (Ang and De Wreede 1992, as Fucus distichus), and Himanthalia elongata (Creed 1995). For the subtidal species C. maschalocarpum and S. sinclairii, thalli progressively grew in size while density-dependent mortality occurred, indicating that self-thinning occurred for these algae. However, thalli grew faster and became larger i n high-density stands compared with stands at low density after the same study period (Schiel 1985), which is not common during self-thinning among terrestrial plants; reasons for this difference remain unclear (Schiel 1985). For P. californica, growth rates were negatively related to thallus density but mortality seemed unaffected by density (Reed 1990). For F. gardneri, there was inverse (negative) density-dependent mortality at the germling stage, but mortality became positively related to density when thalli grew older (Ang and De Wreede 1992). Growth rates were unrelated or either positively or negatively associated with thallus density depending on the experimental design employed or the season considered (Ang and De Wreede 1992). Is frond crowding beneficial for Mazzaella cornucopiae? Holdfasts of Mazzaella cornucopiae produce fronds continuously during the growth period at Prasiola Point (Chapter 2), and stand biomass increases simultaneously, as the dynamic biomass-density relationships found for each experimental quadrat indicate. The increase i n frond density theoretically would imply reduced productivity because the irradiance decreases below the canopy, but the continuous increase i n density suggests that the benefits of crowding outweigh this apparent disadvantage. Some important environmental factors that affect the growth and survival of M. cornucopiae are desiccation, irradiance, temperature, and wave action. High frond densities may be a way to cope with a stressful intensity of those physical factors as observed i n the intertidal zone. A high frond density may prevent strong desiccation and an excessive increase i n temperature during spring and summer low tides, may protect understory tissues against high irradiance during spring and summer, and may decrease the detachment rate of large fronds due to wave action during high tides. Chapter 4 w i l l analyze the effects of these environmental factors on the growth and survival of M. cornucopiae as they interact with the density of fronds. Interactions among growing genets It is important to note again that the dynamic biomass-density relationship has been analyzed for Mazzaella cornucopiae at the ramet (frond) level, not at the genet (whole thallus) level. How might the biomass-density relationship be among growing genets of this species or other clonal red algae? Does self-thinning occur among genets that are growing i n crowded conditions? Do they present density-dependent patterns of growth and mortality? A n important and simplifying characteristic for the study of these algae is that they grow attached to rocks or other hard surfaces and that they lack roots. Thus, competition is restricted to the level above the substratum and should be easier to study than for land plants. However, a complicating factor is that thalli of some species (such as M. cornucopiae and Chondrus crispus) frequently form dense aggregations of fronds that cover a high percentage of the substratum, making the field identification of neighboring genets difficult. In addition, holdfasts of different genets of clonal red algae may coalesce when they get in contact during growth (Tveter and Mathieson 1976, Rueness 1978, Tveter-Gallagher and Mathieson 1980, Maggs and Cheney 1990, Santelices et al. 1996). When high abundance of fronds and coalescence of holdfasts occur, it is virtually impossible to distinguish individual genets in the field. In spite of this, it is reasonable to expect that the negative relationship between stand biomass and genet density observed for non-clonal terrestrial plants (Weller 1987) or even for non-clonal brown algae (Creed 1995) during self-thinning probably does not occur for clonal red algae, because neighboring genets possibly coalesce instead of one competitively excluding the other. Thus, patterns of genet growth and mortality may differ between clonal red seaweeds and non-clonal plants and algae, and even terrestrial clonal plants, which do not coalesce. For any demographic study of genets of clonal red algae, mapping the position of different genets from early stages should be done regularly i n order to describe correctly their patterns of growth, coalescence, and mortality. This is especially important when a given genet becomes partitioned into two or more fragments (clonal fragments, sensu Eriksson and Jerling 1990) due to some kin d of disturbance and the surviving portions behave like separate individuals. However, an important problem is that we do not know what the outcome is when two growing genets meet through holdfast contact. W o u ld they coalesce on a l l occasions? W o uld one of them slowly competitively exclude the other? Would they react differently depending whether meeting genets belong to the same reproductive phase (gametophytes and tetrasporophytes) or not? Does genetic diversity affect the outcome even between genets of the same phase? These questions might be difficult to answer by doing just a visual inspection at regular intervals i n the field, because of the difficulty of detecting a possible mobile contact line between holdfasts, which is i n turn complicated by high frond densities and the isomorphism between reproductive phases. Techniques of molecular biology might be useful i n solving these problems. Isozyme electrophoresis, D N A fingerprinting, and the RAPD (random amplified polymorphic DNA) technique have been applied to sampled ramets to distinguish among different neighboring genets ranging from mosses to flowering plants (Ellstrand and Roose 1987, Schaal et al. 1991, Widen et al. 1994, Jonsson et al. 1996), for which identification was not possible or was difficult by regular visual or morphological methods. For clonal red algae such as Mazzaella cornucopiae, these methods could be useful to differentiate genets when growing thalli are followed over time and their holdfasts make contact. Thus, the dynamic biomass-density relationship could theoretically be studied at the genet level, and patterns of growth, coalescence, and mortality be determined. Mapping w i l l be essential though, because molecular methods might not distinguish between different genets that bear the same genotype. This can occur because some tetrasporophytes are formed from genetically identical carpospores (released from carposporophytes within gametophytes, as part of the triphasic life-cycle, van Oppen et al. 1995), and potentially when gametophytes originate from parthenogenetic spores from other gametophytes (Zupan and West 1988). Thus, a combination of molecular methods and field mapping techniques may allow us i n the future to document and understand the structure and dynamics of populations of clonal red algae at the genet level, as opposed to the ramet level studied here. CM 'E CO CO E g co. c CO CO CT) O r 4.85 4.9 4.95 5 5.05 Log (Frond Density) (fronds r r f 2 ) Fig. 3.1: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae from Prasiola Point during the 1994 growth period. Data points are monthly means based on data from seven experimental quadrats. 3 CM 'E C D — 2.75 CO CO CO E g in c CO CO o 2.5 H 2.25 H 4.6 Ju l M a y Mar J a n r = 0.98 4.8 I 4.9 5 4.7    5.1 Log (Frond Density) (fronds m" 2 ) Fig. 3.2: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae from Prasiola Point during the 1995 growth period. Data points are monthly means based on data from seven experimental quadrats. CVJ E O ) CO CO CO E o in T3 c CO 3.4 2.9 2.4 1.9 CO o 1.4 4.50 1 1 Y = 6.33 X - 29.40 1 I ,-' r • / • / * • r = 0.64(n = 13) • r -• • j I 1 i i 4.66 4.82 4.98 5.14 5.30 Log (Frond Density) (fronds r r f 2 ) Fig. 3.3: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #1 at Prasiola Point throughout 2 years. Each data point corresponds to one month. The functional relationship was calculated by PCA. CAJ E 3.4 2.9 co CO CO 1 2.4 DO c <o - ~ CO 1 9 o 1.4 4.50 Y = 7.42 X - 33.20 r = 0.55 (n = 14) 4.66 4.82 4.98 5.14 5.30 Log (Frond Density) (fronds m" 2 ) Fig. 3.4: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #2 at Prasiola Point throughout 2 years. Each data point corresponds to one month. The functional relationship was calculated by PCA. CM I E cn to to CO E g bo T3 C CO 3.4 2.9 2.4 £2 1.9 CO Y = 3.83 X - 1 6 . 4 8 r = 0.80 (n = 14) 1.4 4.50 4.66 4.82 4.98 5.14 5.30 Log (Frond Density) (fronds m" 2 ) F i g . 3 .5: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #3 at Prasiola Point throughout 2 years. Each data point corresponds to one month. The functional relationship was calculated by PCA. CM E CD CO CO CO E o bo c co CO o 3.4 2.9 2.4 CO i a 1, ,4 4.50 Y = 5.07 X - 22.60 r = 0.48 (n = 13) 4.66 4.82 4.98 5.14 5.30 Log (Frond Density) (fronds m" 2 ) F i g . 3.6: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #4 at Prasiola Point throughout 2 years. Each data point corresponds to one month. The functional relationship was calculated by PCA. CM I E CD CO CO CO E o in T3 C CO CD O 3.4 2.9 2.4 1 Q CO 1 , a 1.4 I • 1 t i 1 • • • Y = 3.62 X - 1 4 . 3 7 1 r = 0 . 6 7 (n = 14) I I I . . 4.50 4.66 4.82 4.98 5.14 5.30 Log (Frond Density) (fronds m" 2 ) F i g . 3.7: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #5 at Prasiola Point throughout 2 years. Each data point corresponds to one month. The functional relationship was calculated by PCA. 3.4 CM E CD 2.9 CO CO CO E 2.4 o in T J C CO CO 1.9 CD O _1 1.4 i 1 7 1 1 ,-' • • • / m • Y = 5.40 X - 23.21 . i #• = 0.52 (n = 14) i i i 4.50 4.66 4.82 4.98 5.14 5.30 Log (Frond Density) (fronds m" 2 ) F i g . 3.8: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #6 at Prasiola Point throughout 2 years. Each data point corresponds to one month. The functional relationship was calculated by PCA. 81 ^ 3.4 CM I E D ) 2.9 co co CO E o in CO CO O ) o 2.4 & 1.9 1.4 I 1 1 1 • -• • Y = 2.85 X - 1 0 . 9 9 I r = 0.72 (n = 14) i i i 4.50 4.66 4.82 4.98 5.14 5.30 Log (Frond Density) (fronds m" 2 ) Fig. 3.9: Relationship between logio stand biomass and logio frond density for Mazzaella cornucopiae for experimental quadrat #7 at Prasiola Point throughout 2 years. Each data point corresponds to one month. The functional relationship was calculated by PCA. 0.0 i i i Io -0.5 Y = - 1 . 5 X + 5 co CO E -1.0 _ \ o in T3 -1.5 \ C \ y, • o LL -2.0 _ « • ' c CO CD -2.5 - • • : j \p D ) • \ • • \ O _ l -3.0 - • \ "."" • \ •Vi • \ -3.5 1 1 \ 1 3 4 5 6 7 Log (Frond Density) (fronds m" 2 ) Fig. 3.10: Combinations between logio mean frond biomass and logio frond density for stands of Mazzaella cornucopiae from Prasiola Point (n=96). The plotted line is the "ultimate biomass-density line" proposed for terrestrial plants (see text for an explanation). Fig. 3.11: Combinations between logio mean frond biomass and logio frond density for clumps of Mazzaella cornucopiae from Prasiola Point and Nudibranch Point (n=56). The plotted line is the "ultimate biomass-density line" proposed for terrestrial plants (see text for explanation). Chapter 4 Factors affecting the production, growth, and mortality of fronds Introduction The basic population dynamics of Mazzaella cornucopiae from Prasiola Point and the influence that abiotic factors may have on its seasonality have been discussed i n Chapter two. Holdfasts of this species are perennial, and they produce upright fronds i n a continuous fashion throughout the year. The period of maximum production and growth of fronds occurs between mid-winter and summer. Frond density and stand biomass increase between mid-winter and summer, leading to a high degree of frond crowding and a biomass peak by late spring/early summer. Fronds do not undergo self-thinning during the growth season, even at the highest densities observed i n the field (Chapter 3). The present Chapter w i l l examine important factors that affect the production, growth, and mortality of fronds of the population of M. cornucopiae from Prasiola Point. In particular, the effects that frond density and key environmental factors such as irradiance, desiccation, and wave action have on these processes w i l l be analyzed, with implications for the adaptive significance that frond crowding confers to M. cornucopiae. This Chapter w i l l firstly explore one of the possible causes of the lack of frond self-thinning during growth. For clonal land plants, a reduction of the rate of production of ramets at high ramet densities is thought to be one of the possible reasons for the observed lack of self-thinning among growing ramets (de Kroon and Kwant 1991, de Kroon 1993). A density-dependent rate of production of ramets would help to prevent the overproduction of ramets which otherwise would lead to self-thinning, but this has not been fully demonstrated yet (de Kroon 1993). The first objective of this Chapter is to test if the production of fronds (ramets) of M. cornucopiae during the growth season is density-dependent, and to compare this with the production of ramets of clonal terrestrial plants, discussing its ecological significance. The temporal variation of combinations between frond density and stand biomass (Chapter 3) gives information on whether self-thinning occurs or not. The temporal change of frond size hierarchy (---inequality or variability) is another important descriptor of the growing process. The dynamics of the inequality of a given size parameter offers information about the effects of density on the performance of growing individuals, i.e., the kind of competition that they undergo (Weiner and Solbrig 1984, Hara 1988, Weiner 1988, Bendel et al. 1989). Two measures of size inequality are the coefficient of variation and the Gi n i coefficient. Both coefficients have advantages and disadvantages, but they are highly correlated (Weiner 1988, Knox et al. 1989). The second objective of this Chapter is to analyze the dynamics of the inequality of frond length and frond weight for Mazzaella cornucopiae and its implications for the effects that frond density has on the competitive interaction among fronds during their growth. The increasing shade that occurs below the canopy of Mazzaella cornucopiae because of the growth i n size of existing fronds and the continuous addition of new fronds between winter and spring or summer does not cause frond mortality as it does for non-clonal plants growing i n crowded stands (Chapter 3). As irradiance decreases beneath seaweed canopies at increasing frond densities, photosynthetic rates also decrease, as observed for Dictyota bartayresii, Halimeda opuntia, Laurencia papillosa (Hay 1981), Corallina vancouveriensis, Gelidium coulteri, and Mazzaella affinis (Taylor and Hay 1984; M. affinis as Rhodoglossum affine). Net photosynthetic rates also decrease at high frond densities for M. cornucopiae (see Results). Then, why does crowding of fronds increase as the growing season proceeds? Are there advantages i n crowding which outweigh the disadvantage of a lower net photosynthesis? In the intertidal zone, desiccation during low tides constitutes a stress which may severely limit photosynthesis and growth rates of seaweeds (Johnson et al. 1974, Brinkhuis et al. 1976, Quadir et al. 1979, Hay 1981, Hodgson 1981, Oates and Murray 1983, Oates 1985, 1986, Brown 1987, Madsen and Maberly 1990, Levitt and Bolton 1991, Bell 1993, Britting and Chapman 1993, Dudgeon et al. 1995). Additionally, the combination of high desiccation and high irradiance, which may be frequent during spring and summer low tides, may cause bleaching of fronds (Kain and Norton 1990). A n increased frond density may be beneficial for an intertidal alga by preventing high desiccation, through increased water retention both among and wi t h i n fronds, and by decreasing the high irradiance that is observed at the canopy level. The third objective of this Chapter is to assess the effects that frond density have on desiccation and understory irradiance and the response of net photosynthesis to varying levels of desiccation and irradiance. The interactions between frond density, desiccation, irradiance, and net photosynthesis w i l l provide an understanding of the significance of crowding for the growth and survival of M. cornucopiae i n the high intertidal zone. The hydrodynamic forces imposed by waves constitute a limit to the maximum size that organisms can reach i n the intertidal zone (Denny et al. 1989, Carrington 1990, Gaylord et al. 1994, Denny 1995, Shaughnessy et al. 1996). Mazzaella cornucopiae occurs i n areas that are moderately to highly exposed to waves at Prasiola Point, and this becomes an important mortality factor during the fall and winter (Chapter 2). A n increase i n frond density may reduce the rate of detachment of large fronds induced by wave action because of protective effects given by the surrounding fronds. The fourth objective of this Chapter is to examine the effects of wave action on frond mortality mediated by frond density. Materials and Methods Effects of density on the rate of production of fronds The study site was the moderate wave exposure side (E) of Prasiola Point. In an area with a high cover of Mazzaella cornucopiae, ten 2 cm x 2 cm quadrats were randomly located along a permanent transect line on 1-2 A p r i l 1995. A l l of the fronds longer than 1 cm were completely removed. Ten additional 2 cm x 2 cm quadrats (randomly selected along the same transect line) were used as controls. Although measurements of frond size were not done for this experiment, the manipulation likely reduced mean frond size for treatment quadrats compared with control quadrats. To keep the same frond size structure for both set of quadrats, I would have had to remove a similar proportion of fronds for different size classes i n treatment quadrats. This proved to be difficult i n the field without damaging neighboring fronds, especially when trying to remove the smallest fronds. The removal of fronds longer than 1 cm made pruning easier and enabled me to leave only intact fronds i n place for the experiment. The effects of a reduction of mean frond size are similar to a reduction i n frond density, i n the sense that they both allow a higher irradiance to reach lower areas of the M. cornucopiae turf and for a higher desiccation to occur, for instance. Therefore, results of this experiment may have also been influenced by this possible change i n size structure, i n addition to the observed reduction i n frond density. On 1-2 A p r i l and on 14-16 May 1995, total frond density was estimated i n both groups of quadrats. The rate of production of fronds was expressed as the difference i n total frond density between May and A p r i l 1995, relative to the ini t i a l frond density, i.e., a proportional increase. Tagging the small fronds to follow their individual survival was impossible, due to their small size and location i n the high density stands. For the estimation of the rate of production of fronds, I assumed that no further losses of fronds occurred during the study period. This assumption is supported by the following observations: frond density increased i n both groups of quadrats, fronds looked healthy, and holdfast scars that indicate frond losses were not seen during this experiment. To test for a differential rate of production of fronds between the low-density and control quadrats, an independent f-test was performed through SYSTAT 5.2.1 for Macintosh (Wilkinson et al. 1992). Temporal variation of frond size hierarchy The parameters considered for this section were frond length and frond biomass. Data came from the six experimental quadrats used for the study of natural frond dynamics (Chapter 2), but only data collected during the period of active growth of Mazzaella cornucopiae were considered, i.e., between February and August 1994 and between January and July 1995 (these months correspond to the available sampling dates). Frond biomass was estimated for each quadrat from the frond length structure (Chapter 2) and different frond biomass-length relationships (Table 2.1). The coefficient of variation (C.V.) was preferred for this study over the Gird coefficient simply because of its availability i n the employed statistical program (SYSTAT 5.2.1 for Macintosh); both coefficients are highly correlated (Weiner 1988, Knox et al. 1989). To test for differences between monthly values of frond length, frond biomass, and the coefficient of variation for both frond length and frond biomass, one-way repeated-measures ANOVA's (ANOVAR's) were performed, followed by trend analyses (Howell 1992). For a few occasions, it was necessary to logarithmically transform the data to meet the homoscedasticity assumption. A l l of the analyses were done with SYSTAT 5.2.1. A one-way A N O V A R through randomization tests was done for the coefficient of variation for frond length for 1995 because of heteroscedasticity even after data transformations. For this particular case, 1000 random permutations were done using the program Randomization Tests, written for DOS by Eugene S. Edgington (University of Calgary, AB; see also Edgington 1987). The particular analysis done for each variable is detailed i n the Results. Effects of frond density on desiccation and understory irradiance To test for the effects of frond density on frond desiccation during low tides, eight 2 cm x 2 cm quadrats were randomly located along a transect line, and the small fronds were completely removed on 19 May 1995. Only large fronds were left attached to holdfasts. Control areas consisted of eight 2 cm x 2 cm quadrats randomly located along the same transect line, where all of the fronds were left intact. On 20 May, fronds were collected from both control and pruned quadrats (n=49 and n=44, respectively) and carried i n closed plastic bags to the laboratory at Bamfield Marine Station, where their biomass ("desiccated" biomass) was calculated to the nearest mg. Fronds were then rehydrated i n seawater until maximum biomass was attained, which was recorded as "fully hydrated" biomass. Percent desiccation of fronds was estimated as: %D = 100 (Bh - B d) Bh"1 where Bh is "fully hydrated" biomass and Bd is "desiccated" biomass. On 20 May, the high tide previous to the collection time occurred approximately at 5:30 hs. Given the height of the experimental site (about 3.3 m above LNT) and considering that waves were small on 20 May, the time of emersion between the last previous wave and the time of collection of fronds (about 15:30 hs) was approximately 9.5 hours. The experimental plots were fully exposed to sunlight and moderate winds during that time. The difference between percent desiccation of fronds i n low-density and control quadrats was tested with an independent f-test through SYSTAT 5.2.1. To assess the effects of frond density on the irradiance that reaches different strata within stands of Mazzaella cornucopiae, the irradiance was measured at Prasiola Point with a probe (connected to a light meter, model LI-COR LI-189) that was placed both above and below the frond canopy, under both full cloud cover and direct sunlight on 2 June 1996. A i r temperature was measured with a thermometer on the same date, both above and below the canopy of M. cornucopiae and under cloudy and sunny conditions. Effects of desiccation and irradiance on net photosynthesis Fronds of Mazzaella cornucopiae were collected during low tide from Prasiola Point, at approximately 20:00 hs on 3 June 1996, shortly before sunset. Fronds were clumped together and placed i n atmospheric conditions within a closed cooler, to be later transported to the laboratory at the University of British Columbia, i n Vancouver. On 4-5 June, fronds were placed i n a glass containing seawater at 12-15 °C under room light for 30-60 min before photosynthesis was measured. The aerial net photosynthesis for fronds was measured by monitoring changes i n CO2 concentration within a closed chamber using an infrared gas analyzer (IRGA; Nortech Control Equipment Inc., Delta, BC). Two irradiances were used for this experiment: a "low" irradiance of 20 umol m - 2 s _ 1, which simulated the average irradiance found below the canopy of fronds (see Results), and a "high" irradiance of 515 umol m~2 s _ 1, which may occur above the canopy of fronds or among fronds from experimentally thinned stands (see Results). The value used here as high irradiance was chosen because it is within the range of saturation irradiances for intertidal seaweeds (400-600 umol m - 2 s - 1, Lobban and Harrison 1994). To measure net photosynthesis, fronds were blotted dry after being taken from the container with seawater, and placed inside the photosynthetic chamber. To avoid self-shading, fronds were attached along a string that was hung from the top of the chamber. The air temperature was 21.5 and 25.0 °C for the low and high irradiance experiments, respectively. To prevent shortage of CO2 within the photosynthetic chamber during each experimental run, as a consequence of frond photosynthesis, additional CO2 was added at the beginning of each ru n i n the form of exhaled air. The i n i t i a l concentration of CO2 was 522 ±- 5 ppm (mean ±- SEM, n=5) for the low-irradiance experiments, whereas the final concentration was 479 ± 5 ppm. For the high-irradiance experiments, the initial concentration of CO2 was 512 ± 12 ppm (n=5), whereas the final concentration was 466 ± 6 ppm. Only for the high-irradiance experiments, the concentration of CO2 reached its minimum value during the experiment (435 ± 4 ppm), to increase up to the final value mentioned above. Therefore, the concentration of CO2 was never lower than that of atmospheric air at ground level (350-400 ppm). The air that flows within the photosynthetic chamber of the IRGA caused fronds to desiccate more with time. To estimate the desiccation rates for the two experimental irradiances, two groups of medium-to-large fronds were placed within the photosynthetic chamber (1.60 g of total fresh biomass, for low irradiance, and 0.91 g, for high irradiance). With lights on and the air pump of the IRGA working, total frond biomass was estimated to the nearest 0.01 g at 1 min intervals during 70 min (for low irradiance) and 30 min (for high irradiance). The string that had the fronds was removed from the chamber for about 5 seconds every 1 min to estimate total frond biomass. After the 30 and 70 min periods, all of the fronds were dried to constant biomass. Total water content was 7 1 % and 7 3 % of the total biomass for the groups of fronds under low and high irradiance, respectively. Percent desiccation of fronds (referred to total water content) had a high positive linear correlation with time for both irradiances (r=0.997; Figs. 4.1 and 4.2). By both variables being highly linearly correlated, the linear relationship % Desiccation = A . Time (A = constant) allowed me to accurately estimate the temporal variation of percent desiccation for the photosynthesis experiments. For these photosynthesis experiments, which are explained below in detail, total initial (fully hydrated) biomass, total final (partially desiccated) biomass (at the end of each run), and dry (fully desiccated) biomass were determined for the different groups of fronds. The temporal variation of percent desiccation was determined by using linear regressions between percent desiccation and time that were calculated using the initial total biomass and final total biomass for each group of fronds. In other words, the constant A was calculated for each run. Aerial net photosynthesis was estimated for five replicate groups of fronds for each of the two irradiances tested. Total initial weight was 2.97 ± 0.04 g (mean ± SEM) for the low irradiance experiment and 2.04 ± 0.12 g for the high irradiance experiment. The concentration of CO2 was determined every 1 minute for 60-90 min. Net photosynthesis was estimated as the changes i n CO2 concentration that occurred during 2-min intervals around specific desiccation levels (0%, 10%, 20%, 30%, etc), and values were finally expressed i n umol CO2 g"1 (dry biomass) min - 1. The desiccation reached by each group of fronds at the end of each run was averaged separately for the two irradiances; net photosynthesis for those two final desiccation levels was estimated as the change i n CO2 that occurred during the last two minutes for each run. The final desiccation levels for both treatments (41% for low irradiance and 80% for high irradiance) were purposely reached to match the two desiccation levels observed at Prasiola Point after 9.5 hours of being exposed to the air at natural and low-density quadrats, respectively. To test for the effects of frond desiccation on net photosynthesis, a one-way repeated-measures analysis of variance (ANOVAR, Howell 1992) was performed separately for each irradiance through randomization tests (using the program "Randomization tests"), because of the violation of the homoscedasticity assumption (tested as i n Chapter 2). After the two ANOVAR's were done, two-tailed paired f-tests were used to compare mean photosynthetic rates between consecutive desiccation levels. To test the hypothesis that net photosynthesis is lower at low irradiance than at high irradiance (at high hydration levels), one-tailed independent f-tests were used to compare photosynthetic rates at 0%, 10%, and 20% desiccation separately. Parametric analyses were done using SYSTAT 5.2.1. Effects of frond density on bleaching and subsequent loss of frond tissues The effects that the density of fronds of Mazzaella cornucopiae have on bleaching and subsequent loss of frond tissue were assessed at Prasiola Point between late spring and mid-summer of 1996. On 3 June 1996, ten 2 cm x 2 cm quadrats were randomly selected along a permanent transect line on the wave-protected side (E) of Prasiola Point, and stands were thinned, leaving only fronds longer than 1 cm. Forty healthy-looking fronds were tagged with a string tied around their stipe, and their length was measured to the nearest mm. Eighteen additional, undisturbed quadrats served as controls, where additional forty healthy-looking fronds were tagged and their length also measured to the nearest mm. On 30 July 1996, frond length was measured again for the tagged fronds to estimate their growth or reduction i n size, and the degree of bleaching was recorded. Changes i n frond length were compared through independent f-tests performed i n SYSTAT 5.2.1. Effects of wave action on frond mortality mediated by frond density To test for the effect of waves on frond mortality through changes i n frond density, five 2 cm x 2 cm quadrats were randomly located on a transect line at the wave-exposed side (W) of Prasiola Point, and fronds were selectively pruned on 26 October 1995. Only fronds longer than 1.5 cm were left attached to holdfasts, while all of the fronds were left intact i n five additional, randomly located 2 cm x 2 cm control quadrats. The density of fronds was estimated i n all of the quadrats, and the density of fronds longer than 1.5 cm was also determined i n the control quadrats. Two days later, on 28 October 1995, the density of fronds longer than 1.5 cm was measured for both treatment and control quadrats. The percentage of large (>1.5 cm long) fronds lost during the two days was calculated for both sets of quadrats and then compared using a f-test. This f-test was carried out through a randomization test with systematic permutations (Edgington 1987, Manly 1991), using the program "Randomization tests", because of the violation of the assumption of normality of the data. Results Effects of density on the rate of production of fronds Frond density was 5.5 ±- 0.5 fronds cm - 2 (mean ±- SEM; n=10) i n low-density quadrats (after pruning) and 8.3 ± 0.6 fronds cm - 2 i n control quadrats on 2 A p r i l 1995, which is a significant difference (independent f-test, p=0.003). The density of fronds increased i n both groups of quadrats between A p r i l and May 1995, but the rate of production of fronds was higher i n low-density quadrats than i n control quadrats. Low-density quadrats had a proportional increase of 2.24 ± 0.49 (mean ± SEM), whereas that for control quadrats was 0.42 ± 0.11, which is a significant difference (independent f-test, p=0.005). This shows a density-dependent production of fronds for Mazzaella cornucopiae. Temporal variation of frond size hierarchy The coefficient of variation for frond length (CVFL) varied among months during the growth season of 1994 (one-way A N O V A R on log-transformed data, F=6.20, Huynh-Feldt adjusted p=0.017), and it decreased linearly during this growth period (trend analysis, F=7.38, p=0.035; Fig. 4.3). For the growth season of 1995, CVFL also differed between months (one-way A N O V A R through randomization tests, F=3.09, p =0.030). The statistical program employed, "Randomization tests", can not perform trend analysis, but a linear decrease is apparent i n the data (Fig. 4.4), as it occurred for 1994. Mean frond length differed among months for the 1994 growth period (one-way ANOVAR, F=8.46, Huynh-Feldt adjusted p=0.002), and it increased linearly during this time (trend analysis, F=12.87, p=0.016; Fig. 4.5). For the 1995 growth period, mean frond length apparently d i d not change significantly between winter and spring, and it decreased only between May and July. A one-way A N O V A R indicated significant differences between months (F=3.61, Huynh-Feldt adjusted p=0.038), and trend analysis suggested a linear decrease between January and July (F=7.79, p=0.038; Fig. 4.6), but separate f-tests done between monthly means for 1995 showed that the only significant difference (decrease) occurred between January and July (p<0.05). The coefficient of variation for frond biomass (CVFB) varied significantly among months for the 1994 growth period (one-way ANOVAR, F=19.69, Huynh-Feldt adjusted p<0.001), and it followed a decreasing trend, with both significant linear (F=21.90, p=0.003; Fig. 4.7) and quadratic (F=15.38, p=0.008) components. For 1995, CVFB also varied between months (one-way A N O V A R on log-transformed data, F=16.19, Huynh-Feldt adjusted p<0.001), showing a decreasing linear trend (F=30.56, p=0.001; Fig. 4.8). The mean biomass of fronds significantly varied during 1994 (one-way ANOVAR, F=10.68, Huynh-Feldt adjusted p=0.002), and it followed an increasing trend with both significant linear (F=15.36, p=0.008; Fig. 4.9) and quadratic (F=8.30, p=0.028) components. Mean frond biomass did not change significantly during the 1995 growth period (ANOVAR, F=0.16, Huynh-Feldt adjusted p=0.86; Fig. 4.10). Effects of frond density on desiccation and understory irradiance • Desiccation The density of fronds i n the low-density quadrats was 1.5 ± 0.2 fronds cm - 2 (mean ±- SEM; n=8), and the mean biomass of fully hydrated fronds was 50 ± 2 mg (n=44). The density of fronds i n the control quadrats was 11.1 ± 0.7 fronds cm"2 (n=8), and the fully hydrated biomass of the large fronds selected was 81 ± 6 mg (n=49). Desiccation was 58.6 ± 1.3 % (n=44) i n the low-density quadrats and 31.2 ± 1.2 % (n=49) i n control quadrats approximately 9.5 hours after the last wave of the previous high tide had reached the site. Percent desiccation was significantly higher in low-density quadrats compared with control quadrats (two-tailed independent f-test; p<0.001). • Irradiance On 2 June 1996, between 9:30 hs and 10:00 hs, the irradiance recorded just above the canopy of Mazzaella cornucopiae was approximately 2000 umol n r 2 s"1 during exposure to direct sunlight and around 800 umol n r 2 s _ 1 during periods of full cloud cover. Below the canopy, the irradiance was 20-30 umol m~2 s"1 during sunny periods and about 10 umol m~2 s _ 1 under total cloud cover. During cloudy periods, the irradiance was as low as 3 umol n r 2 s _ 1 near the holdfast. The temperature also differed above or below the canopy. During sunny periods on the same day and time, the temperature was 23-24°C above the canopy and 21-22 °C within the algal turf. Under total cloud cover, the temperature decreased to 21-22 °C on the surface of the turf and to 20-21 °C within the turf. Effects of desiccation and irradiance on net photosynthesis Fronds of Mazzaella cornucopiae were capable of doing photosynthesis when experimentally exposed to the air, which simulated a low tide. Maximum net photosynthetic rates were observed at 10-30% desiccation under low irradiance and at 0-20% under high irradiance (Figs. 4.11 and 4.12). A t higher desiccation levels, net photosynthetic rates declined ( ANOVAR through randomization tests, p=0.034 for low irradiance, and p=0.001 for high irradiance). Net photosynthesis decreased to zero at 4 1 % desiccation under low irradiance (two-tailed f-test, p<0.05) and at 30% desiccation (two-tailed f-test, p<0.05) under high irradiance. Between 5 0% and 80% desiccation, net photosynthetic rates were negative at high irradiance (two-tailed f-tests, p<0.05). When fronds were highly hydrated (0%, 10%, and 20% desiccation), net photosynthetic rates were significantly lower at low irradiance than at high irradiance (one-tailed independent f-tests, p<0.05; Fig. 4.13). Effects of frond density on bleaching and subsequent loss of frond tissues On 3 June 1996, the initial frond density was 4.4 ± 0.3 fronds cm" 2 (mean ±- SEM, n=10) and the initial length of tagged fronds was 1.5 ± 0.1 cm (n=40) i n treatment quadrats. In control quadrats, the initial frond density was 8.5 ± 1.4 fronds cm - 2 (n=18) and the initial length of tagged fronds was 1.9 ± 0.1 cm (n=40). On 30 July 1996, 30 tags were found out of the original 40 (75%) in treatment quadrats, whereas 31 tags (77.5%) were found i n the control quadrats. Mean length of fronds significantly decreased i n both treatments after two months: mean frond length was 1.0 ± 0.1 cm (n=30) i n treatment quadrats, which was significantly lower than for June (p<0.001, two-tailed paired r-test), and it was 1.5 ± 0.1 cm (n=31) i n control quadrats (p=0.032, two-tailed paired f-test). The percent loss of tissue, expressed as: 100 - (100 l e n g t h y / l e n g t h J U N E ) , was significantly greater i n treatment quadrats (31.6 ± 5.4 %, n=30) compared with control quadrats (14.0 ± 6.5 %, n=31) (p=0.042, two-tailed independent t-test). A collection of fronds from treatment quadrats (n=108) showed that 59.3% of the fronds presented evidence of bleaching i n their tips, whereas only 22.4% of fronds showed signs of bleaching i n their tips i n control quadrats (n=219). Effects of wave action on frond mortality mediated by frond density The initial density of fronds i n the low-density quadrats was 2.4 ± 0.2 fronds cm - 2 (mean ± SEM; n=5), while the initial density i n control quadrats was 11.8 ± 1.6 fronds cm - 2 (n=5). After two days of being subjected to strong wave action during high tides, the percent loss of fronds longer than 1.5 cm was 22.8 ± 8.0 % (n=5) i n low-density quadrats and 5.0 ± 3.3 % (n=5) i n control quadrats. A one-tailed f-test done through a randomization test w i t h systematic permutations showed that the loss of large fronds was significantly greater in low-density quadrats compared with control quadrats (p=0.048). Discussion and Conclusions Effects of density on the rate of production of fronds During the period of active growth for Mazzaella cornucopiae at Prasiola Point, i.e., between mid-winter and summer (Chapter 2), the continuous production of new fronds is density-dependent. This is also what generally occurs during the growth period for shoots (ramets) of terrestrial clonal plants (Hutchings 1979, Hartnett and Bazzaz 1985, Briske and Butler 1989, de Kroon and Kwant 1991, de Kroon 1993, Hara et al. 1993). The density-dependent production of ramets during growth of clonal land plants from seasonal habitats is thought to be one of the possible causes of the lack of ramet self-thinning (de Kroon and Kwant 1991, de Kroon 1993). The same is potentially applicable then to the dynamics of fronds of M. cornucopiae. The cause of the density-dependent production of ramets of clonal terrestrial plants is not clear. Some authors think that the reduced production of ramets at high densities may result from the inhibiting effects of the low irradiance and the low red:far-red ratio occurring near the bottom of ramets on the development of basal axillary buds (Casal et al. 1985, Solangaarachchi and Harper 1987, de Kroon 1993). However, recent experiments with the bunchgrass Schizachyriun scoparium var. frequens d i d not support the hypothesized role of the red:far-red ratio as a density-dependent signal regulating bud development, but the red:far-red ratio may still play a yet unknown (direct or indirect) role on ramet production, mediated by phytochrome (Murphy and Briske 1994). Among clonal seaweeds, nothing is known about the causes for a density-dependent regulation of frond production. As terrestrial plants, seaweeds also present a variety of physiological responses to the light regime. Marine algae produce phytochrome, although its presence i n red algae is i n doubt, and other pigments possibly related to photomorphogenesis (Dring 1988, Lobban and Harrison 1994). This suggests that light could affect the density-dependent production of fronds of Mazzaella cornucopiae and possibly other clonal red algae, pointing out the need for future research i n this area. Temporal variation of frond size hierarchy The dynamics of frond size hierarchy for Mazzaella cornucopiae during its growth period differs i n fundamental ways from the dynamics of plant size hierarchy for non-clonal plants that are actively growing. For even-aged stands of non-clonal plants during early growth, plant size hierarchy increases through time, because plant differences i n growth and size that result from genotypic variation are increased by competition for light (Hara 1988, Weiner 1988, 1990). As a result of this one-sided or asymmetric competition, where large plants progressively supress the growth of the smallest plants, self-minning commences at some point i n time and plant size hierarchy begins to decrease because of the mortality of the smallest plants (Schmitt et al. 1986, Weiner and Thomas 1986, Weiner 1988, 1990, Knox et al. 1989). For stands of Mazzaella cornucopiae, frond size hierarchy, as described by the coefficient of variation for frond length and frond biomass, decreased consistently during the two growth periods analyzed. The mean size of fronds, expressed as both length and biomass, increased during the 1994 growth period and remained statistically similar during the 1995 growth period. The increase of frond density that occurred during both growth periods (Chapters 2 and 3), together with an increase of mean frond size (for 1994) or even if mean frond size remains constant (for 1995), likely determine a progressive reduction of the irradiance that reaches the understory. The progressive decrease of the variability of frond size during the growth period, while stands were getting crowded and the understory irradiance was likely decreasing, suggests that the increasing shade within the algal turf does not determine a decrease or suppression of growth of the smallest fronds. Rather, this indirectly indicates that the growth rate of small fronds were likely higher than for large fronds. This is contrary to what happens during the growth of stands of non-clonal terrestrial plants prior to self-thinning, during which the growth of the smallest plants progressively decreases until it is completely suppressed, mainly due to competition for light. For clonal terrestrial plants, ramet size inequality has been observed to either decrease or remain constant as ramets grow i n size and their density increases (de Kroon 1993). Under low irradiance such as below the canopy of fronds, net photosynthetic rates of fronds of Mazzaella cornucopiae are significantly lower compared with under high irradiance such as at the canopy level (see below). However, the competition for light among fronds does not appear to be asymmetric (one-sided), as it is for actively growing non-clonal plants (Weiner 1988, 1990). Rather, frond competition for M. cornucopiae responds to the symmetric (two-sided) competition model, for which the growth increment of ramets is proportional to, independent of, or negatively related to their size (de Kroon et al. 1992, de Kroon 1993, Hara et al. 1993). The progressive reduction of frond size hierarchy during the growth period for M. cornucopiae suggests that the growth increment of fronds is negatively related to frond size. Genetic differences among non-clonal plants constitute one of the causes that determine the increase of plant size inequality during early stages of growth, before self-thinning operates (Weiner 1988, 1990). For Mazzaella cornucopiae, genetic differences among fronds from the same genet would be low or non-existent, so the decrease of frond size hierarchy and the lack of asymmetric competition among fronds during the growth period would apparently be facilitated by this low or ni l genetic variability among fronds. A factor that may contribute to the progressive decrease i n frond size hierarchy and the lack of asymmetric competition for M. cornucopiae is the possible physiological integration among fronds that belong to the same holdfast (Maggs and Cheney 1990), which would allow photosynthates to be translocated from large fronds to small, shaded fronds (see Gonen et al. 1996). The causes that may determine higher growth rates for small fronds of M. cornucopiae compared with large fronds are possibly related with the effects that desiccation and high irradiance have on the tips of large fronds, which are generally directly exposed to sunlight and wind during low tides. This is discussed i n the following section of this Chapter. The decrease of frond size hierarchy for Mazzaella cornucopiae may also result from the combination of wave action and herbivory. Snails of genus Littorina are common inhabitants of M. cornucopiae stands (Kim 1995, K i m and De Wreede 1996b). Between mid-spring and summer, the tips of the largest fronds become bleached, possibly as a result of strong desiccation and high irradiance (see later i n this Chapter). During low tides, snails frequently graze on those tips (Kim 1995, K i m and De Wreede 1996b, R. A. Scrosati, pers. obs.), producing holes and other forms of damage on them. The combined effect of grazing by littorinid snails and removal of damaged tips by waves may contribute to reduce the frond size hierarchy by selectively reducing the size of the largest fronds. Very little is known about the dynamics of frond size hierarchy and its meaning for clonal seaweeds i n general. For the clonal red species Gelidium sesquipedale (Gelidiales) from Portugal, frond length hierarchy also decreased between winter and summer (Santos 1995). Neither frond bleaching nor herbivory on frond tips has been reported for this subtidal species, so it not possible to speculate about the causes for the observed decrease of frond length hierarchy. Frond size hierarchy was also assessed for Mazzaella laminarioides from the intertidal zone of central Chile (Martinez and Santelices 1992); however, the existence of only two sampling dates (spring and fall) prevented from analyzing its temporal variation during the growth season for this species. Interactions between frond density, desiccation, irradiance, and net photosynthesis The irradiance that reaches fronds at the understory level of stands of Mazzaella cornucopiae is greatly reduced compared with the irradiance observed above the canopy. During the growth period, frond density increases progressively as fronds grow i n biomass (Chapter 3), which determines that the understory-level irradiance progressively decreases. For a number of red, brown, and green seaweeds, increasing densities result i n lower photosynthetic rates (Hay 1981, Taylor and Hay 1984). For M. cornucopiae, net photosythetic rates are also lower at irradiance levels observed at the understory compared with at the canopy, at least when desiccation is not strong. In spite of this cost, an increasing crowding of fronds has been selected. Why is this? Are there more benefits due to frond crowding which outweigh the cost of reduced net photosynthetic rate? Desiccation affects the rate of net photosynthesis of fronds of Mazzaella cornucopiae. Maximum photosynthetic rates occur when desiccation is between 10-30% (at low irradiance) and 0-20% (at high irradiance), whereas increased levels of desiccation significantly reduce net photosynthesis. After an i n i t i a l increase i n photosynthesis from 0% to 10-20% desiccation, photosynthesis is also inversely correlated to desiccation for an additional number of intertidal seaweeds, including red, brown, and green species (Johnson et al. 1974, Brinkhuis et al. 1976, Quadir et al. 1979, Hay 1981, Hodgson 1981, Oates and Murray 1983, Oates 1985, 1986, Brown 1987, Madsen and Maberly 1990, Levitt and Bolton 1991, Bell 1993, Britting and Chapman 1993, Dudgeon et al. 1995). The average desiccation of fronds i n natural, high-density stands of Mazzaella cornucopiae was 4 3 % after 9.5 hours of exposure to sunny and windy conditions i n the spring. Net photosynthesis decreases to zero at 4 1 % desiccation under an experimental irradiance of 20 umol m~2 s _ 1, which is normally observed within high-density stands. The average desiccation of fronds that were i n experimental low-density stands was much higher after the same period of 9.5 hours, about 81%. The irradiance that reaches fronds within low-density stands is comparable to the irradiance observed only at the canopy level i n high-density stands. Under an experimental irradiance of 515 umol m"2 s _ 1, net photosynthesis is negative for desiccation levels between 50-80%. This suggests that fronds i n low-density stands may consume their energetic reserves during a relatively long period during low tides. This could seriously compromise the growth of fronds, especially when desiccation is high during spring and summer. By preventing strong desiccation and reducing the irradiance that reaches the fronds below the canopy, crowding of fronds would thus protect against strong decreases of net photosynthesis as time progresses during low tides. The negative values of net photosynthesis observed under high irradiance and high desiccation (Fig. 4.12) may be a result of photoinhibition (cf. Herbert and Waaland 1988, Herbert 1990, Lobban and Harrison 1994). Photoinhibition may involve bleaching if it occurs frequently (Kain and Norton 1990), which constitutes a disruptive stress, because the affected tissue is unable to fully recover (Davison and Pearson 1996). Extensive bleaching of fronds and frond tips of Mazzaella cornucopiae that are directly exposed to sunlight (up to 2000 umol m~2 s _ 1 at Prasiola Point) during low tides occurs during mid-spring and summer (Chapter 2). The total loss of pigments is accompanied by intense grazing by littorinid snails, frequently producing holes and other forms of physical damage on bleached areas (Chapter 2). Bleached areas of the fronds are softer than healthy areas, which, together with the damage produced by the herbivores, seems to make bleached portions more susceptible to be removed by wave action. Fronds or parts of fronds that are located below the canopy do not undergo bleaching and look healthy during spring and summer. Crowding of fronds of M. cornucopiae thus confers protection against bleaching and the subsequent loss of bleached areas. In fact, this was proven i n the field by means of experimental thinning of stands, which led to a higher degree of frond bleaching and a higher loss of frond tissue compared with natural, high-density stands. The importance of an increased density for the prevention of high desiccation and subsequent bleaching was also observed for tropical algae such as Dictyota bartayresii (brown species), Halimeda opuntia (green), and Laurencia papillosa (red) (Hay 1981), and for additional temperate algae such as Corallina vancouveriensis (red) (Padilla 1984, Taylor and Hay 1984) and Gelidium coulteri (red) (Taylor and Hay 1984). H i g h irradiance may also cause a strong increase of the thallus temperature during low tides, which may constitute a serious stress for intertidal algae (Henley et al. 1992, Britting and Chapman 1993, Henley 1993, Kubler and Davison 1993, Carrington Bell 1995, Dudgeon et al. 1995). This is particularly true when frond desiccation is strong during calm and sunny days, as observed for the high intertidal species Mastocarpus papillatus (Carrington Bell 1995). When both desiccation and irradiance are low, thallus temperature may stay lower than air temperature, because of the energy transfer from the thallus to the air through evaporation (Carrington Bell 1995). If this situation holds true also for Mazzaella cornucopiae, with its similar morphology, then high frond density could offer protection against a stressful increase i n thallus temperature by decreasing both irradiance and desiccation below the canopy. As mentioned i n Results, air temperature is lower within natural turfs of M. cornucopiae compared with air temperature just above the canopy, suggesting that high frond density could contribute to the protection against strong increases i n thallus temperature. The effects of temperature on growth and survival of M. cornucopiae were not specifically tested i n this study, but a temperature stress, as a result of high irradiance and desiccation, may be partially responsible for the bleaching observed i n fronds at Prasiola Point. The above discussion points to the general conclusion that high density of fronds of Mazzaella cornucopiae involves a cost expressed as reduced net photosynthesis when frond desiccation is low, but frond crowding seems to be an adaptation to cope with the harsh environmental conditions especially characteristic of spring and summer low tides, which otherwise would limit the growth and survival of this alga. This may be a common strategy adopted by many intertidal algae, as a similar trade-off was also observed for Dictyota bartayresii, Halimeda opuntia, Laurencia papillosa (Hay 1981), Corallina Vancouveriensis and Gelidium coulteri (Taylor and Hay 1984). For the latter two algae, if desiccation stress is absent [i.e., for thalli that inhabit tide pools), frond density decreases and total biomass per unit area increases compared with dense stands, which shows the cost i n terms of productivity that crowding involves (Taylor and Hay 1984). Crowding of fronds has also been found to be a defensive mechanism against herbivores for some seaweeds. High density of fronds of the tropical algae Dictyota bartayresii, Halimeda opuntia, and Laurencia papillosa reduced their consumption by the parrot fish Sparisoma rubripinne and the sea urchin Diadema antillarum (Hay 1981). The main herbivores for fronds of Mazzaella cornucopiae from Prasiola Point are snails of genus Littorina, which graze mainly during low tides (Kim 1995, K i m and De Wreede 1996b). Most of the snails are small (< 5 mm long) and apparently they find refuge from desiccation and heat during low tides i n dense stands of M. cornucopiae. Further experiments are required to determine the extent to which high frond densities offer protection against herbivory i n M. cornucopiae. Reproductive structures of Mazzaella cornucopiae (cystocarps and tetrasporic sori) occur mainly during fall and, secondarily, during winter at Prasiola Point (Chapter 2). A n analysis of a collection of reproductive fronds done on 27-28 October 1996 at the moderate wave-exposure side of Prasiola Point showed a positive relationship between frond biomass and number of reproductive structures per frond for both phases (Figs. 4.14 and 4.15). High frond densities during spring and summer may, through their protective action against strong bleaching and tissue loss, ensure an increased abundance of large fronds by the end of the growing season, therefore favouring a higher production of reproductive structures during fall and winter. The beneficial effects of frond crowding for Mazzaella cornucopiae have been discussed based on observations made during low tides. Is crowding important for frond growth when tides are high and thalli are fully hydrated? Higher seaweed densities are associated with a decrease i n the velocity of water passing among individuals (Gerard 1982, Jackson 1984, Mork 1996), which i n turn increases the thickness of the surface boundary layer. This may lead to a reduction i n rates of nutrient uptake and photosynthesis (Wheeler 1980, 1988, Gerard 1982, Norton et al. 1982, Norton 1991, Neushul et al. 1992, Koch 1993), meaning that high algal density would limit growth rates. However, M. cornucopiae occurs i n the high intertidal zone of rocky shores where waves are commonly breaking during high tides. The high energy imposed by waves may make water movement and subsequent distribution of nutrients independent of frond density. During high tides, frond density is a more important factor for the survival of fronds, as discussed below. Effects of wave action on frond mortality mediated by frond density The effect of wave action on frond mortality, as mediated by frond density, has been assessed during the fall. In the fall, low tides occur mainly during night time at Prasiola Point. For this reason, desiccation of fronds and irradiance do not constitute stressful factors for the survival of fronds as they are during the spring and summer. Wave action is generally stronger during fall than during spring and summer (Chapter 2). Therefore, the experimental thinning of stands done during the fall is assumed to have tested for the effects of wave action on frond mortality rather than for the effects of desiccation and irradiance. Mortality of large fronds of Mazzaella cornucopiae through detachment by wave action is inversely density-dependent: crowding of fronds, characteristic of natural stands, resulted i n a decreased rate of frond detachment compared with thinned stands. One of the possible advantages of this protection against waves relates to reproduction. As mentioned above, the largest fronds are the ones that bear the highest number of cystocarps or tetrasporic sori during the reproductive season. It seems advantageous that t h a l l i keep the large fertile fronds attached to holdfasts u n t i l their reproductive structures are fully mature and the spores are released, and that would be enhanced under high frond densities. High frond densities might reduce the detachment rate of large fronds in two ways. Large fronds may be "cushioned" among a large number of surrounding fronds, therefore the angle formed by the substratum and the frond that is being bended by a passing wave would not be as low than if fronds were solitarily arranged; this would result i n a decreased probability of breakage of the stipe-holdfast junction. In addition, the surface area of large fronds that are directly exposed to passing waves may be reduced when more fronds surround them. This may be translated into less drag, which is one of the main hydrodynamic forces and is proportional to the exposed frond surface area (Denny et al. 1989, Carrington 1990, Gaylord et al. 1994, Denny 1995, Shaughnessy et al. 1996). Desiccation of fronds and high irradiance are important factors that affect the growth and survival of fronds (through bleaching) mainly during spring and summer i n the high intertidal zone of Prasiola Point, whereas wave action would be important for frond survival (through detachment) mainly during fall and winter at this site. Therefore, crowding of fronds of M. cornucopiae appears to be an adaptation to cope with high desiccation and high irradiance, which are typical of spring and summer, and with strong wave action, which is characteristic of fall and winter. 10 20 30 40 50 60 Time (min) Fig. 4.1: Temporal change of percent desiccation of fronds of Mazzaella cornucopiae from Prasiola Point within a photosynthetic chamber of an infra-red gas analyzer (IRGA) under 20 umol n r 2 s - 1 of irradiance. 70 60 c o g 50 o CD 40 Q c£ 30 20 10 0 —i 1 1 1 1 r r =0.997 (n = 31) 10 15 20 25 30 Time (min) 35 Fig. 4.2: Temporal change of percent desiccation of fronds of Mazzaella cornucopiae from Prasiola Point within a photosynthetic chamber of an IRGA under 515 umol n r 2 s"1 of irradiance. 1.00 0.60 Feb Apr Jun Month Aug Fig. 4.3: Temporal variation of the coefficient of variation (C.V.) for frond length (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. 0.85 0.60 Jan Mar May Month Jul Fig. 4.4: Temporal variation of the coefficient of variation (C.V.) for frond length (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. 1.30 r-1.16 -1.02 h 0.88 - r-"—j t T 0.74 - p . - . . -f? " 0.60 I u»iw H i i mm Feb Apr Jun Aug M o n t h Fig. 4.5: Temporal variation of frond length (cm; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. 1.1 1 3 1.0 cp % 0.9 T3 c o LL 0.8 0.7 Jan Mar May Jul M o n t h E J C -4—• c CD T 3 C o Fig. 4.6: Temporal variation of frond length (cm; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. Fig. 4.7: Temporal variation of the coefficient of variation (C.V.) for frond biomass (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. CO CO £ o ho. "O c p > 6 Jan Mar May Jul M o n t h Fig. 4.8: Temporal variation of the coefficient of variation (C.V.) for frond biomass (mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. Feb Apr Jun Aug M o n t h Fig. 4.9: Temporal variation of frond biomass (mg; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1994 growth season. Jan Mar May M o n t h Fig. 4.10: Temporal variation of frond biomass (mg; mean + SEM) of Mazzaella cornucopiae from Prasiola Point during the 1995 growth season. 116 c 'E 0 04 CM 0 03 o o "o 0 02 E 3 . CO 0 01 CO CD • c 0 .00 >» CO o o .c -0 01 D_ •§—1 CD z 0 1 0 20 30 41 % Desiccation Fig. 4.11: Net photosynthesis (mean ±- SEM) of fronds of Mazzaella cornucopiae from Prasiola Point, expressed as uptake rates of CO2, exposed to air at 20 umol m - 2 s*1 as a function of percent desiccation (relative to total water content). Desiccation levels that are grouped by a bar have statistically similar values of net photosynthesis (two-tailed paired f-tests, p<0.05). 0.2 n 1 1 1 1 1 1 r O) CM o o "o E 3. CO 'co CD •#—' c >« CO o •I—» o CL CD 0.1 0 .0 - 0 . 1 J I L 0 1 0 2 0 3 0 4 0 5 0 6 0 8 0 % Desiccat ion Fig. 4.12: Net photosynthesis (mean ±- SEM) of fronds of Mazzaella cornucopiae from Prasiola Point, expressed as uptake rates of CO2, exposed to air at 515 umol n r 2 s _ 1 as a function of percent desiccation (relative to total water content). Desiccation levels that are grouped by a bar presented statistically similar values of net photosynthesis (two-tailed paired f-tests, p<0.05). 2 0 10 20 % Desiccation F i g . 4.13: Net photosynthesis (mean + SEM) of fronds of Mazzaella cornucopiae from Prasiola Point, expressed as uptake rates of CO2, exposed to air at 20 umol n r 2 s - 1 ( low irradiance) and at 515 umol n r 2 s - 1 (high irradiance) as a function of percent desiccation (relative to total water content). 100 "O c I CO O .Q E =3 100 150 200 250 F r o n d B i o m a s s (mg) Fig. 4.14: Number of cystocarps per frond of Mazzaella cornucopiae from Prasiola Point versus frond biomass (mg); fronds were collected on 27-28 October 1996. -O 60 £ 50 h O o co .<1> 40 30 20 10 t r r = 0.85 (n = 50) 50 100 150 200 250 F r o n d B i o m a s s (mg) Fig. 4.15: Number of tetrasporic sori per frond of Mazzaella cornucopiae from Prasiola Point versus frond biomass (mg); fronds were collected on 27-28 October 1996. Chapter 5 Recruitment, holdfast perennation, and regrowth after harvesting Introduction Holdfasts of Mazzaella cornucopiae from Prasiola Point are long-lived, and they constitute an important source for the seasonal renewal of fronds, which have high turnover rates (Chapter 2). Thalli produce carpospores and tetraspores mainly during the fall and, secondarily, during the winter (Chapter 2). On an annual basis, recruitment of new thalli from spores appears not to be important for the maintenance of populations, but empirical evidence is not available yet. Among members of the Gigartinaceae, the relative importance of recruitment of new thalli from spores to the annual population maintenance is highly variable, depending on the species and on the site (Hansen and Doyle 1976, Hansen 1977, May 1986, Santelices and Norambuena 1987, Lazo et al. 1989, A ng et al. 1990, Gomez and Westermeier 1991). The first objective of this Chapter is to assess the contribution of recruitment from spores of M. cornucopiae to the maintenance of populations, relative to the contribution from existing holdfasts that perennate from previous years. Disturbances occur occasionally i n the high intertidal zone of Prasiola Point. They may result from the dislodgment of surface rock fragments, mainly during fall and winter, or from unusually hot, sunny, and dry spring or summer seasons, which is associated with strong bleaching of algal tissues (Chapter 4). To potentially maintain its high abundance at the high intertidal zone, Mazzaella cornucopiae needs an effective way of recolonizing disturbed areas. Recolonization of M. cornucopiae is achieved by two mechanisms: recruitment of new thalli from spores and vegetative horizontal growth of existing holdfasts that border the disturbed area. The second objective of this Chapter is to determine the relative contribution of recruitment from spores and of horizontal growth of perennating holdfasts to the colonization of disturbed microhabitats. The combination of recruitment from spores and horizontal growth of perennating holdfasts finally determine the recovery rate of Mazzaella cornucopiae i n a disturbed area. Recovery rates may be affected by the season of disturbance. Percent cover of M. cornucopiae from the moderate wave-exposure side (E) of Prasiola Point apparently increased faster i n summer- and fall-cleared 25 cm 2 quadrats than i n winter- and spring-cleared 25 cm 2 quadrats, but seasonal differences were not evident for 100 cm 2 quadrats that were cleared on the same dates (Kim and De Wreede 1996a). The recovery up to undisturbed, natural levels apparently did not occur for the experimental quadrats during the monitored period for that study (20-24 months, depending on the season of clearing). Given the dates of clearings and the date of my first visit to the same study site, K i m and De Wreede's (1996a) quadrats constitute a unique opportunity to determine the rates of f u l l recovery of populations of M. cornucopiae after disturbances occur at different seasons. The third objective of this Chapter is to determine the time involved for full recovery of M. cornucopiae, through recruitment from spores and horizontal growth of bordering holdfasts, by monitoring the quadrats cleared by K i m and De Wreede (1996a) for an additional period of approximately two years. Additional types of disturbance for Mazzaella cornucopiae are the complete or partial loss of fronds from holdfasts. If holdfasts remain i n place after frond biomass is lost, this type of disturbance does not fit exactly into the ecological definition of disturbance, which involves the provision of new space that becomes available for colonization after biomass is removed. A complete or partial loss of fronds may result from moderate or severe bleaching during spring-summer (aided by grazing by littorinid snails and wave action, Chapters 2 and 4), from the destructive action of logs that are brought to the coast by tides and waves, or from intentional human harvesting. Reasons for an interest i n harvesting M. cornucopiae relate to its potential economic value. This species has carrageenans i n the cell walls, sulphated polysaccharides that are widely used as stabilizers and thickeners i n a variety of foods, cosmetics, and pharmaceutical products (McLachlan 1985). In addition, sulphated polysaccharides and certain lipids from red seaweeds may have antiviral and/or antitumor activity (Neushul 1990, Noda et al. 1990). Although fronds of M. cornucopiae may grow only up to 5 cm i n length, this species is abundant and is present all year-round at Prasiola Point (Chapter 2). If future research on M. cornucopiae finds bioactive compounds that prove useful for antiviral or antitumor treatments (preliminary results indicate so, J. H. Kim, pers. com.), or if this species is targeted as a source of carrageenan for the traditional uses, M. cornucopiae may become of economic interest. Natural populations could be harvested regularly, or collections may be also done from cultivated beds, taking advantage of the alga's capacity for extensive growth on the rocky substratum. The fourth objective of this Chapter is to determine the intrinsic capacity of holdfasts for survival and later regeneration of new fronds after a moderate to complete loss of fronds occurs. Questions related to this objective are: (1) do holdfasts survive a partial and a complete loss of frond biomass?, (2) do thalli recover completely after one year?, (3) do recovery rates vary depending on the intensity of harvesting?, and (4) does harvesting intensity affect further production of reproductive structures? The morphology of Mazzaella cornucopiae is similar to that of other species of the Gigartinaceae that are presently exploited for the extraction of carrageenan such as Mazzaella laminarioides and Chondrus crispus. Results for this experimental harvesting on M. cornucopiae w i l l be compared with those obtained for M. laminarioides, C. crispus, and other related seaweeds, i n a search for patterns of regrowth after harvesting for these species. Materials and Methods Relative contribution of recruitment from spores and of perennating holdfasts to the maintenance of populations The study site was the moderate wave-exposure side (E) of Prasiola Point. To determine the contribution of perennating holdfasts to the annual production of fronds, six 10 cm x 10 cm quadrats were randomly located where Mazzaella cornucopiae was abundant. The reason for using six quadrats was given i n Chapter 2, since quadrats used there for the description of the natural population dynamics were also used for the present section due to time constraints during field trips. The perimeter of holdfasts present i n those quadrats was mapped every two months, from June 1993 to July 1995, with a water-proof marker on quadrats made with transparent plastic. Percent cover and frond length, to the nearest 5 mm, were determined as explained i n Chapter two. The periodic examination of these quadrats allowed me to detect the growth or reduction of the area of perennial holdfasts through time. To determine the contribution of recruitment of new thalli from spores to the annual production of fronds, additional 10 cm x 10 cm, randomly located quadrats were completely cleared by removing the surface layer of substratum with a hammer and a chisel on June 1993. The position of newly recruited thalli of M. cornucopiae that appeared on successive sampling dates, until July 1995, was mapped using a quadrat with 100 subdivisions. Percent cover and frond length, to the nearest 5 mm, for recruited thalli were measured. Therefore, patterns of natality, growth, and mortality of thalli that were recruited from spores were monitored through time. There were 14 sampling dates: 4-6 June 93, 16-20 August 93, 15-18 October 93, 10-12 December 93, 24-27 February 94, 25-29 A p r i l 94, 21-25 June 94, 18-22 August 94, 5-9 October 94, 2-6 December 94, 28-31 January 95, 30 March-3 A p r i l 95, 13-17 May 95, and 10-14 July 95. Relative contribution of recruitment from spores and of perennating, bordering holdfasts to the recolonization of disturbed areas The study site was the moderate wave-exposure side of Prasiola Point, and the same 6 quadrats referred to above were monitored for this section. By means of a 10 cm x 10 cm quadrat that was subdivided i n 100 subquadrats, both the percent cover of thalli that recruited from spores and the horizontal expansion of perennating thalli that bordered the cleared quadrats was measured every two months from June 1993 to July 1995. The sampling dates were the same as for the experiment described above. Time for full recovery after disturbances occurring at different seasons The same 10 cm x 10 cm experimental quadrats established by K i m and De Wreede (1996a) i n the moderate wave-exposure side of Prasiola Point were monitored for this study. The quadrats had been cleared i n summer (August 1992), fall (October 1992), winter (February 1993), and spring (April 1993) (corrected from K i m and De Wreede 1996a, J. H. Kim, pers. com.); 15 replicate quadrats had been assigned for each clearing season and they had been last examined i n August 1994 (Kim and De Wreede 1996a). For the present study, the quadrats were monitored on 25-27 October 1995 and on 20-21 October 1996. Total percent cover of Mazzaella cornucopiae was measured with a 10 cm x 10 cm quadrat subdivided i n 100 subquadrats. To compare percent cover i n disturbed quadrats with the percent cover of natural, undisturbed quadrats, 21 additional 10 cm x 10 cm plots were randomly located along the original transect. The community at one of the extremes of the original transect (about 2/9 of its full length) was different compared with the rest of the transect, apparently because of differences i n elevation. Therefore, I did not consider the quadrats from those 2/9 of the original transect for the present analysis. Percent cover for the four clearing treatments (seasons) and for the controls were compared separately for October 1995 and October 1996 through one-way ANOVA's (Howell 1992). Because of lack of normality and homoscedasticity (checked as explained i n Chapter 2), ANOVA's were done through randomization tests with 1000 random permutations (Edgington 1987, Manly 1991), using the program "Randomization tests" for DOS, written by Eugene S. Edgington (University of Calgary, AB). Only the A N O V A for October 1995 was significant (see Results), so percent cover for each clearing treatment was compared separately with control percent cover by doing independent t-tests through randomization tests. Frond regeneration after harvesting On 4-6 June 1993, 7 permanent transects were established i n the high intertidal zone of the moderate wave exposure side of Prasiola Point by placing concrete nails at each extreme of transects. One "partial pruning" (PP) quadrat, one "total pruning" (TP) quadrat, and one control (C) quadrat were randomly located on each of the 7 transects. A l l frond biomass higher than 1 cm was pruned with scissors i n PP quadrats, all of the fronds were pruned from the base of their stipes i n TP quadrats, and no manipulation was effected in C quadrats. Holdfasts were not disturbed for any of the treatments. Special care was taken not to disturb any other sessile organisms that were found on the experimental quadrats. This constituted a randomized complete block design (Krebs 1989), because each treatment (PP, TP, and C) appeared once i n each block (transect), so each block consisted of 3 experimental units (quadrats). The same quadrats were monitored on 6 bimonthly sampling dates: 4-6 June 1993, 16-20 August 1993, 15-18 October 1993, 10-12 December 1993, 24-27 February 1994, 25-29 A p r i l 1994. Five population parameters were measured each time: total percent cover of Mazzaella cornucopiae, frond density, mean frond length, stand biomass (expressed per unit area), and density of reproductive fronds. The methodology for estimating the above variables was described in Chapter 2. To examine the recovery status of thalli after approximately 1 year (April 1994), one-way ANOVA's (Howell 1992) were performed for each population parameter separately. Since data came from a randomized complete block design, transect (block) was incorporated as a source of variation i n ANOVA's. The advantage of using blocks was that variation among blocks was removed from the error term i n ANOVA's, thus increasing the precision of the experiment (Krebs 1989). The assumptions of normality and homogeneity of variance were checked as explained i n Chapter 2, and they were satisfactory for these ANOVA's. To examine the effects of harvesting intensity on recovery rates, the bimonthly difference between values for each quadrat and for each population parameter was calculated. The mean growth rates thus obtained for each pruning treatment were compared by performing separate one-way ANOVA's (Howell 1992) for each month. Transects were considered as blocks, as explained i n the preceding paragraph. When treatment effects were significantly different according to ANOVA's, f-tests were done between monthly means for each growth rate. The use of f-tests after ANOVA's followed suggestions made by Carmer and Walker (1982), Hurlbert (1990), Mead (1991), and Soto and Hurlbert (1991). The least significant difference (LSD) method has also been suggested to do pairwise comparisons after ANOVA's (Carmer and Walker 1982), but f-tests are preferred when within-treatment variances are heterogeneous, this being a subjective decision (Soto and Hurlbert 1991, S. H. Hurlbert, pers. com.). Statistical analyses were performed using SYSTAT 5.2.1 for Macintosh (Wilkinson et al. 1992). Results Relative contribution of recruitment from spores and of perennating holdfasts to the maintenance of populations Regular mapping between June 1993 and July 1995 indicated that the survival rate of perennating thalli was high and that recruitment of new thalli from spores was very low i n control quadrats. The detection of potential thalli that might have recruited from spores close to perennating holdfasts was not possible. I assumed that the undetected thalli that might have settled close to perennating holdfasts d i d not compromise the accurate determination of percent cover of perennating thalli, given the small size and the slow growth rate of recruited thalli observed i n cleared quadrats. Thalli that were recruited from spores were easily identified i n cleared quadrats. During the first year after the clearings were done, the number of thalli that were recruited from spores progressively increased until the spring, but some thalli died during the summer (Fig. 5.1). Thalli that were recruited from spores only during the second year were fewer than those that were recruited during the first year. The highest monthly average density of recruited thalli was 4 ± 3 thalli / 100 cm 2 (mean ±. SEM; n=6) and it occurred i n May 1995, after the second reproductive season. Percent cover of recruited thalli was much lower than percent cover of perennating thalli during the two-year study period (Fig. 5.2). At the end of the first reproductive season (April 1994), percent cover of recruited thalli was 2.0 ± 0.7 %, which was significantly lower (p<0.01; independent f-test) than percent cover of perennating thalli for the same date, 71.0 ±- 12.2 %. As mentioned above, fewer thalli were recruited from spores during the second year; the percent cover of thalli that were recruited only during the second year was 0.1 ± 0.2 % in July 1995, which was significantly lower (p<0.01; independent f-test) than percent cover of perennating thalli for the same date, 46.7 ± 12.8 %. The relative contribution of annually-recruited thalli to the total percent cover for the population of Mazzaella cornucopiae, with respect to perennating thalli, was 3% i n A p r i l 1994 and 0.4% i n July 1995. Mean frond length was generally lower for recruited thalli than for perennating thalli during the study period (Fig. 5.3). Relative contribution of recruitment from spores and of perennating, bordering holdfasts to the recolonization of disturbed areas The recolonization of disturbed quadrats by Mazzaella cornucopiae was slow. Mean percent cover for the combination of thalli that were recruited from spores and horizontal growth of bordering holdfasts never reached 10% during the two-year study period (Fig. 5.4). Those values were always much lower than the percent cover of thalli i n undisturbed, natural quadrats (Fig. 5.2). Percent cover of horizontal growth or perennating, bordering holdfasts was not measured i n December 1993, February 1994, and A p r i l 1994, but both sources of increase i n percent cover were measured regularly for the rest of the study period. There were no significant differences i n percent cover between recruited thalli and horizontal growth of perennating holdfasts between June 1994 and July 1995 (p>0.05, independent f-tests done separately for each month). Therefore, the recolonization of M. cornucopiae i n disturbed areas was slow and recruitment of thalli equaled horizontal growth of bordering holdfasts i n their relative importance as mechanisms of recolonization. Time for f u l l recovery after disturbances occurring at different seasons Recolonization rates of Mazzaella cornucopiae led to a complete recovery i n October 1995 i n quadrats cleared i n August 1992 (summer), October 1992 (fall), and February 1993 (winter), but quadrats cleared i n A p r i l 1993 (spring) still presented a lower mean percent cover than control areas (Fig. 5.3). The A N O V A done for October 1995 (through a randomization test) gave an F=2.85 (p=0.023), while a f-test comparing the A p r i l 1993 clearing treatment with controls led to p=0.009; the other clearing treatments were not significantly different from controls (f-tests, p>0.05). By October 1996, all clearing treatments reached control levels of percent cover (Fig. 5.5, A N O V A through randomization test, F=1.27, p>0.05). However, percent cover i n quadrats cleared i n A p r i l 1993 did not change between October 1995 and October 1996 (Fig. 5.5). The statistical similarity of A p r i l 1993-cleared quadrats with control, undisturbed quadrats was a result of a significant decrease i n percent cover of control quadrats between October 1995 and October 1996 (paired f-test, p=0.001). Recolonization of disturbed areas was thus complete i n about 3 years after a disturbance occurred, depending on the season of disturbance: 3 years and 2 months for the summer clearing, 3 years for the fall clearing, 2 years and 8 months for the winter clearing, and 3 years and 6 months for the spring clearing. Frond regeneration after harvesting • Percent cover Mean initial percent cover was similar among the three treatments just before frond pruning on June 1993 (one-way A N O V A on a randomized block design, F=1.68, p=0.230). The effects of partial pruning of fronds d i d not significantly reduce percent cover immediately after pruning (f-test, p=0.248), but percent cover did significantly decrease immediately after total pruning was done (f-test, /?<0.001). For total pruning quadrats, percent cover gradually increased up to control levels during the year that followed pruning (Fig. 5.6). In A p r i l , percent cover was similar among the three treatments (one-way ANOVA, F=3.31, p=0.072), thus showing full recovery for harvested quadrats. Percent cover decreased for partial pruning and control quadrats between June and August, but remained similar for total pruning quadrats during the same period (Fig. 5.6 and 5.7). Between August and A p r i l , although mean growth rate (expressed as changes i n percent cover) was generally higher for total pruning quadrats than for the two other treatments (Fig. 5.5), one-way ANOVA's did not detect significant differences. Growth rates were highest, i n general, between February and April. • Frond Density Between August and A p r i l , frond density was always similar for the three treatments (Fig. 5.8). For A p r i l , one-way A N O V A resulted i n F=0.75 (p=0.492), so recovery from harvesting was complete also when considering frond density. Bimonthly changes i n frond density were not affected by pruning treatment (Fig. 5.9). Growth rates were generally negative between August and December and generally positive between December and April. • Mean frond length For partial pruning quadrats, mean frond length reached control levels two months after pruning (f-test, p=0.066), but, for total pruning quadrats, values similar to those for controls were reached in February (f-test, /?=0.745). On Ap r i l , the three treatments presented a similar mean frond length (one-way ANOVA, F=0.37, p=0.669; Fig. 5.10). Growth rates, expressed as bimonthly changes i n mean frond length (Fig. 5.11), were positive for both pruning treatments between August and October; mean frond length decreased during the same period i n control quadrats. Later on, only total pruning quadrats presented a consistent pattern of positive growth rates up to A p r i l ; comparisons with the two other treatments showed no consistent pattern during this period. • Stand biomass Partial pruning quadrats took 4 months to reach control levels of stand biomass (f-test, p=0.863), while total pruning quadrats needed 8 months (f-test, p=0.827). In A p r i l , there were no significant differences among the three treatments (one-way ANOVA, F=0.17, p=0.850; Fig. 5.12). Growth rates, expressed as changes i n stand biomass, were negative between August and December, but the decrease was more pronounced i n control quadrats than i n treatment quadrats, with total pruning quadrats presenting the highest values (Fig. 5.13). Growth rates were highest between February and A p r i l for the three treatments. • Production of reproductive fronds during frond regeneration Cystocarpic fronds were first detected i n October and at similar densities for the three treatments (Fig. 5.14). The density of cystocarpic fronds decreased gradually until A p r i l , presenting similar values on each sampling date for the three treatments. The production of tetrasporic fronds differed among treatments. Tetrasporic fronds appeared at least two months earlier (October) and were present for a longer period (up to April) i n partial pruning quadrats compared with control quadrats (Fig. 5.15). Stands subjected to total pruning did not produce tetrasporic fronds during the study period. Discussion and Conclusions Importance of recruitment from spores and of perennating holdfasts for population maintenance and recolonization after disturbance Holdfasts of Mazzaella cornucopiae from Prasiola Point are mostly perennial and constitute the principal source of new fronds that are generated each year i n the population. Regular mapping done at the study site showed that the shape of the area of perennating holdfasts slowly changes i n a continuous manner; some areas of holdfasts grow horizontally while others die. As mentioned i n Chapters 2 and 3, it is not possible at present to recognize different genets of M. cornucopiae i n mature stands because of the high frond densities encountered and, possibly, holdfast coalescence (see Maggs and Cheney 1990). However, thalli that were recruited from spores were easily identified i n cleared quadrats. Recruitment occurred during fall and winter, which are the reproductive seasons for Mazzaella cornucopiae at Prasiola Point (Chapter 2), but some recruits died during the summer, possibly because they were directly exposed to strong desiccation, irradiance, and temperature during low tides at that time (see Chapter 4). The number of thalli that were recruited from spores was low and their growth rates were slow. This determined that the relative contribution of recruited thalli to the annual production of fronds, compared with mature thalli, was very low for the population at Prasiola Point. The extensive horizontal growth of holdfasts and their perennial nature appears to be a mechanism that allows M. cornucopiae to become competitively dominant or frequent i n its habitat (see also K i m 1995). The high intertidal zone at Prasiola Point presents a relatively predictable environmental seasonality (Chapter 2). Disturbances that remove the surface layer of the rocky substratum are uncommon at this site, and the biological community changes i n a relatively predictable way through seasons (Kim and De Wreede 1996a). In that general context, species whose main life-history traits fit relatively better to a K-strategy than to a r-strategy (Mac Arthur and Wilson 1967, Pianka 1982) are expected to be common. Mazzaella cornucopiae falls into this category w i t h respect to some important demographic characteristics. Thalli are dominant i n the high intertidal community and they are mostly perennial, so population size does not experience high interannual variation. Reproduction is delayed after recruitment (at least, it did not occur on recruited thalli during the first two years after experimental clearings) and thalli are iteroparous, although individual fronds are usually detached after reproduction. Recruitment, as discussed above, is a minor source of annual frond production. High levels of competition are expected for K-selected species (Pianka 1982); for M. cornucopiae, holdfast coalescence may potentially lessen the intensity of at least intraspecific competition among genets (Chapter 3). The strategy of persistence shown by Mazzaella cornucopiae is also observed for similar algae that occupy a similar ecological niche. The mid- to high intertidal species Mazzaella laminarioides from central Chile also produces several fronds from extensive holdfasts and occurs i n a similar habitat. For populations of this species, spores also play a secondary role i n the maintenance of populations, compared w i t h perennating holdfasts (Santelices and Norambuena 1987, Gomez and Westermeier 1991, as Iridaea laminarioides). Other species within the Gigartinaceae may show different strategies. For Mazzaella splendens (a low intertidal to shallow subtidal species) from San Juan Island, northern Washington, perennation played the major role i n the production of fronds between winter and spring of 1982 (May 1986, as Iridaea cordata). Although the sampling design carried out by May (1986) did not allow for distinguishing thalli that were recruited from spores from those that were potentially recruited from a possible bank of microscopic forms (Chapman 1986, Santelices 1990), it was evident that thalli that were recruited from spores accounted for 2 0 % (or less, if recruitment from the bank of microscopic forms occurred) of the thalli observed i n the immediate spring that followed the winter of 1982. In accordance w i t h May's (1986) observations, matrix models that simulated the demography of M. splendens from Vancouver harbour, southern British Columbia (Ang et al. 1990, as Iridaea splendens) suggested that established thalli were more important than recruited thalli (from either spores or from another microscopic form) for the production of fronds between summer and the following winter and between winter and the following summer. However, recruitment of thalli (from either spores or from another microscopic form) was predicted to be the major contributor to the production of fronds on an annual basis (Ang et al. 1990). Why does the relative importance of spores differ between Mazzaella cornucopiae and M. splendens? Holdfasts of M. splendens are much smaller in area than those of M. cornucopiae (pers. obs.). Therefore, a given absolute decrease i n holdfast area due to partial death is more likely to increase the probability of death for the entire holdfast of M. splendens than for those of M. cornucopiae. Then, spores would be expected to be more important for the population maintenance of M. splendens than for M. cornucopiae. For Mazzaella splendens from central California, though, the great majority of annually produced fronds were reported to derive from perennating thalli (Hansen and Doyle 1976, Hansen 1977, as Iridaea cordata). Causes for the observed differences between populations from southern British Columbia and northern Washington and from central California are unknown. The North Atlantic alga Chondrus crispus is another member of the Gigartinaceae that presents intraspecific variation i n the importance of perennation vs. recruitment for the maintenance of populations. For this species, this variation depends on substratum characteristics and the degree of disturbance. For subtidal populations from Prince Edward Island, the importance of recruitment from spores is high, because the friable sandstone substratum determines frequent dislodgment of mature thalli, opening up new space for colonization (Lazo et al. 1989). On the contrary, the relatively more stable substratum of intertidal sites i n the Atlantic coast of Nova Scotia and the characteristic longevity of thalli of C. crispus makes holdfast perennation the major process that is responsible for frond production each year (Lazo et al. 1989). The population of Mazzaella cornucopiae from Prasiola Point behaves then like the population of C. crispus from Atlantic Nova Scotia. The strategy of population maintenance of Mazzaella cornucopiae from Prasiola Point is essentially similar to the one followed by vascular clonal plants. Many species that occur at high latitudes and altitudes, aquatic habitats, the temperate or boreal forest understory, i n grazed grasslands, and fire-disturbed vegetation are clonal. Their population dynamics are mainly determined by birth and death rates of ramets, which are continuously produced by some form of vegetative propagation. Recruitment of new genets through seed germination is infrequent i n those populations (Cook 1985, Hara 1994). Given that vegetative propagation of ramets is the main source of demographic change i n these species, including M. cornucopiae, it has been suggested that the evolutionary potential would be limited for them (Cook 1985). Having a low relative importance for the annual production of fronds, compared with adult holdfasts, spores of Mazzaella cornucopiae would then be important for the recolonization of disturbed areas where all biomass has been lost. The relative importance of spores for recolonization depends on two factors: size of disturbance and presence of mature tha l l i of M. cornucopiae surrounding the disturbed area. The role of M. cornucopiae spores as recolonizing agents decreases with increasing disturbance size. For example, recruitment from spores did not occur i n 400 cm 2 cleared quadrats for, at least, two years after experimental clearings were done i n stands where M. cornucopiae was present (close to Prasiola Point), but recruits did appear i n 100 cm 2 and 25 cm 2 cleared quadrats (in Prasiola Point), presumably because a larger disturbance implies increased desiccation and heat stress (Kim and De Wreede 1996a). Higher recruitment i n smaller disturbed areas could also result from reduced water flow owing to the closer presence of mature algal stands, which might enhance spore settlement (Foster 1975). With respect to mature thalli that may border a disturbed area, they w i l l likely grow to slowly recolonize the bare rocky surface and this process may be more important than recolonization from spores. If mature thalli are mostly absent i n the surroundings of the disturbed area, then spores w i l l play the major role i n recolonizing the area. Recruitment from spores was less variable among the cleared experimental quadrats than recolonization by horizontal growth of bordering holdfasts (Fig. 5.6). This is possibly a result of spore dispersal covering greater distances than the slow horizontal growth of perennating holdfasts, making recruitment less dependant on the distance of mature thalli to a disturbed area. With both recruitment from spores and horizontal growth of bordering holdfasts combined, the recolonization of 100 cm 2 cleared quadrats up to undisturbed levels of percent cover took place after 3 reproductive seasons (comprising fall and winter each, Chapter 2). The spring clearing took more time (3 years and 6 months) than the other clearing treatments (around 3 years) to fully recover possibly because of the longer period between the time of clearing (April 1993) and the reproductive season that followed that clearing (fall and winter 1993-94). Frond regeneration after harvesting The recovery of stands of Mazzaella cornucopiae subjected to partial and total frond pruning done i n the late spring of 1993 was complete by the spring of 1994. This points out the importance of holdfasts for the maintenance of populations, acting as space-savers and quick sources of new fronds i n case of frond losses. Holdfasts showed a high capacity for vegetatively regenerating fronds even after being directly exposed to sunlight and wind i n late spring after total frond pruning was done. During the summer, holdfasts survived a largely increased irradiance and probably an increased desiccation and temperature as well (see Chapter 4). This means that, on a first approach, if a harvesting program is ever carried out for M. cornucopiae, one annual complete removal of upright biomass i n the spring would give the highest yield without compromising the f u l l recovery of stands one year later, thus allowing for economically viable annual harvests in the following years. Average growth rates, expressed as changes i n percent cover, frond density, mean frond length, and stand biomass between consecutive sampling dates, were negative for undisturbed, natural quadrats during late summer and fall. For percent cover, mean frond length, and stand biomass, growth rates were less negative or even positive i n pruned quadrats compared with control quadrats during the same period. Total pruning of fronds generally brought about a faster regrowth rate than partial pruruhg of fronds. There was then an inverse relationship between the degree of frond biomass removal and growth rates, even during supposedly worsening environmental conditions. The stimulation of frond production and growth that follows a removal of frond biomass also occurs for other species of the Gigartinaceae such as Mazzaella laminarioides (Gomez and Westermeier 1991) and Chondrus crispus (Chopin et al. 1988, McLachlan et al. 1988); see also Chapter four. The effects that different harvesting programs have on algal recovery have been investigated for other species of red algae (MacFarlane 1952, Taylor 1959-60, Burns and Mathieson 1972, Mathieson and Burns 1975, Hansen 1977, Carter and Anderson 1985, Carter and Simons 1987, Santelices and Norambuena 1987, Westermeier et al. 1987, Pringle and Semple 1988, Santelices et al. 1989, Gomez and Westermeier 1991). Based on results obtained here for Mazzaella cornucopiae and results from the above studies, the following paragraphs w i l l discuss what presently is known about the dynamics of red algal recovery after harvesting. The different species of seaweeds investigated i n those studies inhabit different geographical regions and/or different tidal levels, and harvesting techniques employed were not always similar; however, some general conclusions may still be drawn. For example, when frond harvesting is done and care is taken not to damage holdfasts, recovery through production of new fronds from undamaged holdfasts is usually guaranteed. The removal of algal biomass that includes holdfasts generally leads to a fairly slow re-establishment of populations, if any (see, for example, the first part of this Chapter's conclusions). The time required for a f u l l recovery (matching natural levels) depends, among other factors, on the season of frond harvest. If collection of fronds is done at the beginning of spring or summer, regrowth rates are likely to be higher than if harvesting is done i n autumn or winter. In addition to this advantage, a summer harvest may also allow for a higher biomass yield (Pringle and Semple 1988). The intensity of harvesting also affects algal recovery rates. Mazzaella cornucopiae had higher regrowth rates when fronds were most severely pruned. As mentioned above, the production of fronds i n members of the Gigartinaceae is usually stimulated by frond removal (Chopin et al. 1988, McLachlan et al. 1988, Gomez and Westermeier 1991). The opposite was found for the related species Mastocarpus stellatus (Petrocelidaceae) (Burns and Mathieson 1972, as Gigartina stellata), although those observations were done with no replication. The methodology selected for harvesting algal biomass may also affect regrowth rates. For Gymnogongrus furcellatus ( G i g a r t i n a l e s , Phyllophoraceae), plucking the largest fronds is recommended over harvesting with scissors, because plucking leaves several small fronds with intact apices, which allows for a higher regrowth i n biomass than if all of the fronds were cut (Santelices et al. 1989). Similar results were observed for Mastocarpus stellatus (Marshall et al. 1949 i n Carter and Anderson 1985) and Chondrus crispus, where recovery after raking the largest fronds was three times faster than after shearing close to the holdfasts (MacFarlane 1952). A comparison of recovery rates from plucking and shearing is not possible for Mazzaella cornucopiae for now, because plucking was not done for this study, but the similarity of its growth pattern with that of M. stellatus and C. crispus suggests that similar results might be obtained. The particular mode of regrowth of branches or fronds that are partially cut when they are harvested also affects recovery rates. For example, Gelidium pristoides (Gelidiales) reacts to branch pruning by producing multiple apices i n the tips of cut branches, which is thought to enhance later productivity (Carter and Simons 1987). I observed the same pattern of regrowth after I experimentally cut apices of Gracilaria chilensis (Gracilariales) from Chile. This pattern of regrowth does not occur among members of the Gigartinaceae. A n increase of harvesting frequency may result i n a higher annual yield, but only up to a certain point. If harvesting frequency becomes too high, the alga may not recover before the following harvest. For example, subtidal areas of Prince Edward Island, Canada, that are subjected to frequent raking for collection of fronds of Chondrus crispus have a lower biomass than areas without harvest pressure (Pringle and Semple 1988). Biomass yields may decrease with frequent collections i n the same area possibly because carbohydrate reserves within holdfasts are consumed during frond regrowth but are not produced and stored at the same rate (Gomez and Westermeier 1991). The above observations on frond regeneration after harvesting are va l i d for a number of red algal species that have been studied so far. However, specific studies have to be done for predicting the quantitative responses for a given target species that occurs i n a specific habitat; this is important for the design of an adequate harvesting strategy that aims for a sustainable biomass yield. For example, for Mazzaella cornucopiae from Prasiola Point, the eleven months after frond pruning i n the late spring of 1993 were long enough for stands to match control levels i n terms of percent cover, frond density, mean frond length, and stand biomass. This suggests that recovery rates should allow for one annual harvest on stands of this species. For other algae that occur i n different habitats, one year might not be enough for a complete recovery. For example, for Chondrus crispus from eastern Canada, both plucking and clipping left a population of only 24.5-29% of what it was in the preceding year (Taylor 1959-60). Production of reproductive fronds during frond regeneration Frond pruning affected i n different ways, or did not affect at all, the production of reproductive fronds of Mazzaella cornucopiae during the following reproductive season. Neither the appearance nor the abundance of cystocarpic fronds were affected by frond pruning, but pruning did affect the appearance and the abundance of tetrasporic fronds. Partial pruning of fronds induced an earlier production of tetrasporic fronds and for a longer period (six months at least) than i n control stands. However, total pruning of fronds completely inhibited the production of tetrasporic fronds during the entire study period. A t present, little is known about the effects of harvesting on the reproduction of red seaweeds, and no clear pattern is apparent yet. A contrasting case is given by Gymnogongrus furcellatus: plucking accelerated the increase i n abundance of cystocarpic fronds with respect to controls, but pruning had opposite effects (Santelices et al. 1989). For subtidal populations of Chondrus crispus from Prince Edward Island, annual harvests done by dragraking reduced the reproductive capacity of cystocarpic fronds, but no effects of harvesting were observed on the reproductive capacity of tetrasporic fronds (Chopin et al. 1988). For Mastocarpus stellatus, different degrees of harvesting reduced the reproductive potential for the first and second years f o l l o w i n g harvesting (Burns and Mathieson 1972). A depletion of carbohydrate reserves that may occur during frond regrowth may partially explain the negative impact of harvesting on reproduction, although the cost of reproduction among seaweeds is thought to be low (De Wreede and Klinger 1988). A complete understanding of the effects that different harvesting programs have on algal reproduction requires extensive research on different algal morphological groups that occur i n different habitats. This represents a fertile field of research for the future, which is essential for designing any strategy with the objective of achieving the highest possible sustainable yield of biomass. A O F A J A O D J M M J 1993 I 1994 I 1995 M o n t h Fig. 5.1: Density of Mazzaella cornucopiae thalli (mean + SEM) that were recruited from spores i n quadrats cleared i n June 1993 at Prasiola Point. Fig. 5.2: Percent cover (mean + SEM) of perennating thalli of Mazzaella cornucopiae i n undisturbed quadrats and of thalli that were recruited from spores in quadrats cleared in June 1993 at Prasiola Point. • Perennating Thalli Recruited Thalli j 1993 1 1994 I 1995 M o n t h Fig. 5.3: Frond length (cm, mean + SEM) of perennating thalli of Mazzaella cornucopiae i n undisturbed quadrats and of thalli that were recruited from spores i n quadrats cleared i n June 1993 at Prasiola Point. Data from cleared quadrats for December 1993 are not available. > O 0 1 > c <1> CL 5-4-3-2-1-0-m Recruited Thalli • Horizontal Growth of Holdfasts i i. i A O D . F A J A O D J M M J 1993 | 1994 1995 M o n t h Fig. 5.4: Percent cover (mean + SEM) of Mazzaella cornucopiae thalli that were recruited from spores i n quadrats cleared i n June 1993 at Prasiola Point and of new holdfast areas that grew from perennating holdfasts that bordered the same cleared quadrats. Data for recruited t h a l l i corresponding to December 1993 and data for holdfast areas that grew from bordering, perennating holdfasts corresponding to December 1993, February 1994, and A p r i l 1994 are not available. 40 O O • C O N T R O L m S U M M E R FALL DD WINTER m SPRING <1> Oct 95 Oct 96 Month Fig. 5.5: Percent cover (mean + SEM) of Mazzaella cornucopiae i n natural quadrats (controls) and i n quadrats cleared at different seasons between 1992 and 1993 (see text for a complete explanation) at Prasiola Point. J u n Aug O c t Dec F e b Apr J u n 1993 1994 M o n t h Fig. 5.6: Seasonal variation of percent cover (mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond piwring (PP) and in undisturbed, control quadrats (C). M o n t h Dec Feb 1993 I 1994 Fig. 5.7: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes i n percent cover (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). For each sampling date considered separately, bars at the top of the graph join means that were statistically similar (independent f-test, p>0.05). 1993 1994 M o n t h Fig. 5.8: Seasonal variation of frond density (fronds cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). o O Dec Feb 1993 1994 M o n t h Fig. 5.9: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes i n frond density (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). For each sampling date, bars join means that were statistically similar (independent f-test, p>0.05). 1993 1994 M o n t h Fig. 5.10: Seasonal variation of frond length (cm; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and in undisturbed, control quadrats (C). 153 Fig. 5.11: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes i n frond length (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). For each sampling date, bars join means that were statistically similar (independent f-test, p>0.05). For A p r i l , the curvilinear bar indicates that the mean growth rate for C was statistically similar to the mean growth rate for TP but not to the mean growth rate for PP. 1 50 Oct Dec Feb 1993 1994 Month Fig. 5.12: Seasonal variation of total stand biomass (g cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). CO CO co E o co T3 C CO -*—» CO c co CD O) c CO CD CO rx JZ 100 50 <NJ E 0 h -50 h -100 o O Oct Apr Dec Feb 1993 I 1994 Month Fig. 5.13: Growth rates of Mazzaella cornucopiae from Prasiola Point, expressed as bimonthly changes i n stand biomass (mean ± SEM), for quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and for undisturbed, control quadrats (C). For each sampling date, bars join means that were statistically similar (independent f-test, p>0.05). Fig. 5.14: Seasonal variation of the density of cystocarpic fronds (fronds cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). co p o 'v_ O Q_ co & O CO c d> Q E o CO "O c 0.3 0.2 \-0.1 k 0.0 M o n t h Fig. 5.15: Seasonal variation of the density of tetrasporic fronds (fronds cm - 2; mean + SEM) of Mazzaella cornucopiae from Prasiola Point i n quadrats subjected to total frond pruning (TP) and partial frond pruning (PP) and i n undisturbed, control quadrats (C). Chapter 6 General conclusions and suggested areas for future research The preceding four Chapters discussed diverse aspects of the population ecology of the clonal red alga Mazzaella cornucopiae from Prasiola Point, Barkley Sound. This constitutes the final Chapter of this Thesis and its two main objectives are: (1) to summarize the topics that were investigated, including problems that were encountered during my research and a discussion of the conclusions obtained, and (2) to suggest new research pathways that appeared as a consequence of the observations and analyses done throughout this Thesis. The overall objective of this Thesis was to provide new insights on the population ecology of clonal red algae, choosing the intertidal species Mazzaella cornucopiae as a model because of its high abundance and amenability for observations and manipulations both at the study site and i n the laboratory. To understand the structure and the dynamics of populations of this species, and the factors that affect them, different areas of research were targeted, as discussed in the following paragraphs. Structure and dynamics of the population from Prasiola Point The role of Mazzaella cornucopiae w i t h i n the high intertidal community at Prasiola Point and at Nudibranch Point has been investigated by Kim (1995). However, at the beginning of the fieldwork for my Thesis, little was known about the population ecology of M. cornucopiae. Thus, a detailed characterization of the structure and dynamics of the population from Prasiola Point was done to provide basic knowledge that would allow me to investigate the ecological questions posed throughout the rest of this Thesis. This baseline study was presented i n Chapter 2. Holdfasts of Mazzaella cornucopiae from Prasiola Point are mostly perennial and fronds are short-lived, possibly being annually renewed. Thallus percent cover, frond density, mean frond length, and stand biomass generally showed seasonal dynamics between June 1993 and July 1995. The period of active growth for the population started i n winter, when frond production resulted i n a progressive increase i n frond density. Both the density of fronds and stand biomass increased until late spring or mid-summer. The color of the fronds changed from dark red i n winter to yellowish-green i n late spring and summer. Frond tips or other frond areas that were directly exposed to sunlight and wind during low tides became bleached during late spring and summer. Bleached areas were weaker than healthy areas, and the former suffered from pronounced grazing by littorinid snails (see also K i m and De Wreede 1996b), resulting i n the loss of frond tissues, apparently aided i n that effect by wave action. The period of active growth ended i n summer and frond mortality (complete detachment of fronds) determined a progressive decrease of frond density and stand biomass until both parameters reached their lowest values i n winter. The growth cycle resumed approximately at the beginning of each year. Reproduction started after the period of active growth. The abundance of cystocarpic fronds and tetrasporic fronds was highest during the fall; fewer reproductive fronds were also present until early to mid-winter. Based on resorcinol tests (Garbary and De Wreede 1988, Shaughnessy and De Wreede 1991), I found that gametophytes predominated over tetrasporophytes at Prasiola Point consistently through the seasons and regardless of the degree of wave exposure. I also proposed that an expected (equilibrium) value for the ratio between gametophytes and tetrasporophytes would be about 2.8 if both phases were ecologically similar; the theoretical value of 2.8 is derived from the simplest possible case, where each phase produces the minimum possible number of reproductive cells. The G:T ratio for Mazzaella cornucopiae from the moderate wave exposure side of Prasiola Point was close to that value, but the two samples from the wave-exposed site showed more variation, including variation due to tidal elevation, which suggests that gametophytes and tetrasporophytes behave ecologically differently at the wave-exposed site. The elucidation of causes for this difference needs further research. A t present, the only evidence of ecological differences between the reproductive phases of M. cornucopiae comes from populations from Oregon, where gametophytes were reported to be more resistant to desiccation and tetrasporophytes more resistant to limpet grazing (Olson 1990). The seasonality of the population of Mazzaella cornucopiae from Prasiola Point was expected, since the physical environment is markedly seasonal at this site. Frond density and stand biomass were positively correlated to daylength and to both air and sea surface temperature and negatively correlated to wave height. It is thought that a l l eukaryotic organisms have endogenous, circannual growth rhythms, which would be synchronized to a 12-month period mainly by daylength and, secondarily, by temperature (Luning 1993). Significant correlations do not necessarily imply a direct cause-effect relationship, but the high correlation coefficients (Table 2.8) obtained for daylength and, secondarily, for temperature may indicate the existence of such circannual growth rhythms, although the possibility that the initiation of strong frond production is a photoperiodic response (Kain and Norton 1990) also remains viable. Wave action appears to be another important factor that contributes to shape the seasonality of M. cornucopiae through hydrodynamic effects on frond survival. Frond interactions during growth Chapter 3 reported on the interactions that occur among fronds of Mazzaella cornucopiae during the period of active growth. Non-clonal terrestrial plants (Weller 1987) and non-clonal seaweeds (Black 1974, Dean et al. 1989, Reed 1990, A n g and De Wreede 1992, Creed 1995) undergo self-thinning once stands become crowded, because the smallest individuals die due to the asymmetric competition from the largest individuals, mainly by shading (Hara 1988, Weiner 1988, 1990). For clonal terrestrial plants that occur in seasonal habitats, ramets do not undergo self-thinning (de Kroon 1993, de Kroon and Kalliola 1995). Prior to my research, little was known about the dynamics of the relationship between stand biomass and frond density for clonal algae. There was indirect evidence that self-thinning would not occur among fronds of clonal red algae such as Chondrus crispus and Mastocarpus stellatus (Pybus 1977), as a "static" biomass-density relationship (each data pair came from a different algal clump) suggested, but a static relationship may or may not reflect the "dynamic" biomass-density relationship that describes the growth period (Weller 1989). A n attempt to determine whether fronds of clonal red algae undergo self-thinning or not (for Mazzaella laminarioides, Martinez and Santelices 1992) focused inadequately on the static biomass-density relationship. A better analysis of the dynamic biomass-density relationship for fronds of clonal red algae was first offered for Gelidium sesquipedale (Gelidiales), concluding that fronds do not undergo self-thinning during growth under natural frond densities (Santos 1995). The analysis of the dynamic biomass-density relationship for fronds of Mazzaella cornucopiae showed that fronds do not undergo self-thinning at natural densities. Rather, frond density and stand biomass increase simultaneously un t i l late spring or mid-summer, declining afterwards (similar to what occurred for Gelidium sesquipedale, Santos 1995). Physiological integration among ramets is thought to be one of the reasons for the lack of self-tWnning among ramets of terrestrial clonal plants, because it would prevent mortality of the smallest, shaded ramets by allowing translocation of photosynthates from the largest ramets (Pitelka 1984, Pitelka and Ashmun 1985, Marshall 1990, de Kroon 1993). Evidence of translocation of assimilates within thalli of red algae has been recently provided (for Gracilaria cornea, Gonen et al. 1996), so the above hypothesis could be valid also for clonal red algae such as Mazzaella cornucopiae or Gelidium sesquipedale. A second hypothesis proposed to explain the lack of ramet self-thinning is that of density-dependent production of ramets during growth. Density-dependent production of fronds has been shown to occur for M. cornucopiae during the growth period (Chapter 4), and this may potentially contribute to prevent an overproduction of fronds, thus preventing the occurrence of frond self-thinning. A third hypothesis also proposed for clonal terrestrial plants is that ramets would stop growing when they reach a maximum average biomass, a biomass determined for any ramet density by the "ultimate biomass-density line" (renamed from de Kroon and Kalliola 1995), thus avoiding self-thinning because such a line, thought to constrain all biomass-density combinations (at either plant or ramet level) for the plant kingdom, would not be transgressed (de Kroon and K a l l i o l a 1995). Combinations of average frond biomass and frond density for M. cornucopiae do transgress the "ultimate biomass-density line". Thus, this third hypothesis does not seem to explain the lack of frond self-thinning for M. cornucopiae. A fourth hypothesis that may explain the lack of frond self-thinning for Mazzaella cornucopiae is that the light compensation point for the smallest fronds may be lower than the irradiance observed close to the holdfast, so the growth of the smallest fronds would be guaranteed even at high frond densities. The only evidence available on this point for high intertidal red algae points to the contrary. For the tropical red alga Ahnfeltiopsis concinna, the compensation point for understory tissue is higher than the irradiance observed at the understory (Beach and Smith 1996a, 1996b). This is the only high intertidal species where the above two parameteres have been measured so far, and results suggest that the growth of the smallest fronds would be shade-limited. A definitive explanation for the lack of self-thinning among fronds of Mazzaella cornucopiae could be achieved through the measurement of size-dependent growth rates of fronds, combined with a study of the possible translocation of photosynthates from large to small fronds and a determination of light compensation points (including the effects of desiccation) for the smallest fronds. For this purpose, special techniques have to be developed for tagging small fronds, which is currently not possible, i n order to determine their growth rates with accuracy. The estimation of stand biomass from frond density A side objective of Chapter 3 was to search for a relationship between stand biomass and frond density that would be site-dependent and would provide a reliable estimate of stand biomass from an enumeration of fronds in a given plot. This would be important when a quick, non-destructive estimation of stand biomass needs to be done. Combinations of stand biomass and frond density obtained for seven permanent plots over two years (between June 1993 and July 1995) fit a linear model with positive slope when plotted on a bilogarithmic scale. However, a quick and non-destructive estimation of stand biomass from a calculation of just frond density remains unreliable, given the site- and year-specific variation encountered for the dynamic biomass-density relationship for the seven experimental quadrats. Mode of competition among fronds The temporal variation of the size hierarchy of individuals or ramets offers information about the k i n d of competition among individuals or ramets. The dynamics of the coefficient of variation for frond length and frond biomass of Mazzaella cornucopiae decreased during the two growth periods (1994 and 1995) analyzed i n Chapter 4. The decrease of frond size hierarchy occurred with a lack of frond self-thinning, while fronds were continuously being produced by holdfasts and were actively growing. This suggests that competition among fronds of M. cornucopiae is symmetric (two-sided), because the growth of the smallest fronds was not reduced or suppressed by the growth of large fronds. For M. cornucopiae, the growth increment during a given time interval for small fronds appears to be higher than for large fronds. Physiological integration among fronds could play an important role on the avoidance of growth suppression for small fronds, by allowing translocation of photoassimilates from large to small fronds. Between mid-spring and the end of the growth period (late spring or summer) a combination of bleaching, grazing by littorinid snails, and wave action selectively removed the tips of large fronds, which may partially explain the progressive decrease of frond size hierarchy. Therefore, the reduction of frond size hierarchy during the growth period for M. cornucopiae appears to be caused by both intrinsic and extrinsic factors. The adaptive significance of frond crowding Frond density increases continuously during the period of active growth for Mazzaella cornucopiae (Chapter 3), which, together with the growth of existing fronds, results i n a decreasing irradiance below the canopy. Given that net photosynthesis of fronds decreases at decreasing irradiances (see also Hay 1981, Taylor and Hay 1984), why has a continuous production of fronds during growth been selected, if that implies a cost i n terms of reduced net photosynthesis? The study of the adaptive significance of frond crowding for M. cornucopiae was one of the objectives of Chapter 4. Desiccation and high irradiance constitute stressful factors i n the high intertidal zone, especially during spring and summer low tides, which occur during daytime at Prasiola Point. In addition to reducing the irradiance below the canopy, high frond densities prevent strong desiccation during low tides. Laboratory experiments demonstrated (Chapter 4) that a combination of high desiccation and high irradiance, which occur i n experimentally thinned stands, results i n negative net photosynthetic rates, which may last for as much as half of a low tide. This could be a result of photoinhibition [cf. Herbert and Waaland 1988, Herbert 1990, Lobban and Harrison 1994), which likely results i n bleaching of frond areas if it occurs frequently (Kain and Norton 1990). The experimental thinning of stands of M. cornucopiae done i n the spring of 1996 at Prasiola Point resulted i n a greater percentage of bleached fronds and a greater loss of frond tissue by mid-summer. Therefore, high frond densities prevent widespread frond bleaching and strong tissue loss during spring and summer, at the cost of reduced net photosynthesis when fronds are highly hydrated. This protection from strong desiccation and high irradiance could enhance the reproductive capacity of populations during fall and winter. The largest reproductive fronds bear the highest number of cystocarps and tetrasporic sori, and more large fronds are expected to survive the summer under high frond densities than under low densities. Frond crowding also confers protection against wave action. Experimental thinning of stands of Mazzaella cornucopiae done at the wave-exposed side of Prasiola Point i n the fall of 1995 resulted i n an increased mortality rate of large fronds, compared with natural stands. A high frond density may reduce the actual area of large fronds that are i n direct contact with incoming waves, and it could also determine that fronds "cushion" one another when waves impact on the shore. These could be the mechanisms by which high frond densities protect thalli of M. cornucopiae against a strong removal of fronds by waves. Hydrodynamic models of frond survival should test for these hypotheses. The protection against strong wave action would also be advantageous for M. cornucopiae because it would allow large, reproductive fronds to stay attached to holdfasts until reproductive structures mature and spores are released. The strongest wave action occurs during fall and winter at Prasiola Point, coincidentally with the reproductive season for M. cornucopiae. Since spring and summer low tides occur mainly during daytime and the strongest wave action occurs mainly during fall and winter at Prasiola Point, frond crowding seems to be more important i n terms of protection against strong desiccation and high irradiance during spring and summer and more important i n terms of protection against wave action during fall and winter. Recruitment from spores and holdfast perennation The importance of reproduction through spores for different ecological processes was discussed i n Chapter 5. Holdfasts of Mazzaella cornucopiae are mostly perennial and the contribution of carpospores and tetraspores to the annual production of fronds was less than 3% for 1993/94 and 1994/95, compared with the contribution of perennating holdfasts. In case of disturbance (the experimental size of disturbance was 100 cm 2), M. cornucopiae is capable of recolonizing the area. The relative contribution of spores to the recolonization of a 100 cm 2 disturbed area depends on the proximity of mature thalli of M. cornucopiae. If mature thalli are bordering the disturbed patch, horizontal growth of their holdfasts may be relatively more important than recruitment from spores for recolonization. In the absence of mature thalli bordering the disturbed area, recruitment from spores w i l l be the only source of recolonization, so recolonization w i l l therefore proceed at a slower rate. Full recovery to undisturbed, natural levels of percent cover (including recruitment from spores and horizontal growth of bordering, perennating holdfasts) was reached i n approximately 3 years for experimental quadrats cleared at different seasons between 1992 and 1993. Recovery from a spring (1993) disturbance took more time than disturbances imposed i n summer (1992), fall (1992), and winter (1992/3) presumably because of the lag period between the spring disturbance and the reproductive season that immediately followed it, i.e., the fall of 1993. The above observations show that Mazzaella cornucopiae presents some demographic traits that are characteristic of a K-selected species. These traits are: (1) thalli are mostly perennial (although fronds seem to have a high turnover rate), (2) reproduction is delayed after recruitment, (3) reproduction is iteroparous (although fronds are lost after cystocarps and tetrasporic sori mature), (4) recruitment from spores has a low importance for the annual renewal of fronds compared with that of perennating holdfasts. Frond regeneration after harvesting The capacity of recovery of holdfasts of Mazzaella cornucopiae from different intensities of frond harvesting was also assessed i n Chapter 5. Both partial (all frond biomass higher than 1 cm) and total (all of the fronds) pruning of fronds were experimentally done i n natural stands i n the late spring of 1993, taking special care not to damage holdfasts while pruning fronds. Compared with undisturbed stands, and considering total thallus cover, frond density, mean frond length, and stand biomass, both treatment quadrats showed complete recovery after one year. For these population parameters, growth rates were generally directly related to the degree of damage, i.e., the highest growth rates were observed i n totally pruned quadrats. Results of these pruning experiments stressed the importance of holdfasts of M. cornucopiae for the maintenance of populations by acting as space-savers and as a source of new fronds. The effects of frond pruning on the production of reproductive structures (cystocarps and tetrasporic sori) after frond regeneration were not consistent for M. cornucopiae. The appearance and abundance of cystocarpic fronds was not affected by frond pruning, but pruning did affect the dynamics of tetrasporic fronds. Tetrasporic fronds appeared earlier and lasted for a longer period i n partially pruned quadrats compared with control quadrats, but tetrasporic fronds were not present i n totally pruned quadrats during the reproductive season that followed the experimental harvest. From what is known from this and other studies on the impact of harvesting on red algal reproduction (MacFarlane 1952, Taylor 1959-60, Burns and Mathieson 1972, Mathieson and Burns 1975, Hansen 1977, Carter and Anderson 1985, Carter and Simons 1987, Santelices and Norambuena 1987, Westermeier et al. 1987, Pringle and Semple 1988, Santelices et al. 1989, Gomez and Westermeier 1991), no trend is sufficiently clear to predict responses for unknown species. Thus, the effects that different harvesting programs have on the reproduction of red algae can itself constitute another extensive topic for research. Mazzaella cornucopiae could be targeted as a source of carrageenan or as a source of antiviral compounds (antiviral activity has been recently detected in extracts from this species, J. H. Kim, pers. comm.) in the future, given its high abundance i n the high intertidal zone at Prasiola Point and adjacent areas. If so, the recovery rates shown by this species after a complete disturbance and the results of the experimental pruning of fronds suggest that holdfasts should not be damaged during any harvesting program. They also suggest that pruning fronds at the base of their stipes would guarantee the maximum harvestable biomass while still allowing for a complete recovery of stands after one year, assuming that repeated annual harvests do not decrease the capacity for recovery. Final remarks Through the study of the population of Mazzaella cornucopiae from Prasiola Point, this Thesis has documented some important ecological patterns that apply to clonal red seaweeds. It has also shown that some ecological traits depend on the species or the habitat, and it has made evident areas where little or almost nothing is known, therefore opening the possibility for exploring new research areas. Main topics that have been only partially examined during the past 20 years and others that have never been approached yet may be summarized as follows: • The relative abundance of gametophytes and tetrasporophytes What abiotic and biotic factors determine the relative abundance of these two life-history phases for different species and for a given species from different habitats? To answer such a question, the main population parameters (recruitment, growth, mortality, and reproduction) should be quantified for each reproductive phase, and the effects that key abiotic and biotic factors have on them should be determined. • The significance of physiological integration for frond dynamics The role of possible physiological integration among fronds on the frond dynamics (including the lack of frond self-thinning) of clonal red algae is another fertile field for future research. This topic is still under discussion even for terrestrial clonal plants, on which all research has been done to date. For clonal algae, age- and size-specific growth rates and translocation patterns with in the thallus should be studied during the growth period to determine the possible l ink between physiological integration among fronds and frond dynamics. • The field identification of genets of clonal red algae The accurate field identification of genets constitutes an unresolved problem that currently prevents studying the demography of thalli of several clonal red algae. The possible frequent occurrence of coalescence between neighboring holdfasts suggests that the dynamics of clonal algae might differ not only from the dynamics of non-clonal plants i n general, but even from the dynamics of clonal terrestrial plants, which do not coalesce as they grow. A combination of field mapping techniques and molecular techniques such as isozyme analysis, D N A fingerprinting, and R A P D analysis could make the demography of clonal thalli amenable for study. • The effects of harvesting on reproduction of clonal red algae A s mentioned above, frond harvesting affects differently reproduction during frond regeneration for the few seaweeds investigated, but no clear relationship between harvesting program, seaweed morphology or growth strategy, and reproductive response is apparent yet. The effects that the harvesting mechanism, intensity and frequency have on the reproduction of clonal red algae wi th different morphologies or growth strategies [i.e., foliose vs. terete, apical growth vs. diffuse growth, annuals vs. perennials) may let us identify such trends, therefore al lowing us to predict responses to harvesting for unexploited species or known species from unexplored habitats. These new areas of research should allow us to improve even more the understanding of the population structure and dynamics of clonal red algae and the role that they play i n marine communities. Challenges and possible results are exciting. References Abbott, LA. 1971. On the species of Iridaea (Rhodophyta) from the Pacific coast of North America. Syesis 4: 51-72. Alpert, P. 1996. Nutrient sharing i n clonal fragments of Fragaria chiloensis. Journal of Ecology 84: 395-406. Ang, P.O. and R.E. De Wreede 1992. Density-dependence i n a population of Fucus distichus. Marine Ecology Progress Series 90: 169-181. Ang, P., R.E. De Wreede, F. Shaughnessy and L. Dyck 1990. A simulation model for an Iridaea splendens (Gigartinales, Rhodophyta) population i n Vancouver, Canada. Hydrobiologia 204/205: 191-196. Beach, K.S. and C M . Smith 1996a. Ecophysiology of tropical rhodophytes. I. Microscale acclimation i n pigmentation. Journal of Phycology 32: 701-710. Beach, K.S. and CM. Smith 1996b. Ecophysiology of tropical rhodophytes. II. Microscale acclimation i n photosynthesis. Journal of Phycology 32: 710-718. Bell, E.C 1993. Photosynthetic response to temperature and desiccation of the intertidal alga Mastocarpus papillatus. Marine Biology 117: 337-346. Bendel, R.B., S.S. Higgins, J.E. Teberg and D.A. Pyke 1989. Comparison of skewness coefficient, coefficient of variation, and G i n i coefficient as inequality measures within populations. Oecologia 78: 394-400. Bhattacharya, D. 1985. The demography of fronds of Chondrus crispus Stackhouse. Journal of Experimental Marine Biology and Ecology 91: 217-231. Black, R. 1974. Some biological interactions affecting intertidal populations of the kelp Egregia laevigata. Marine Biology 28: 189-198. Bolton, J.J. and M.A.P. Joska 1993. Population studies on a South African carrageenophyte: Iridaea capensis (Gigartinaceae, Rhodophyta). Hydrobiologia 260/261: 191-195. Briggs, J.C 1974. Marine zoogeography. McGraw-Hill, New York, 475 p. Brinkhuis, B.H., N.R. Tempel and R.F. Jones 1976. Photosynthesis and respiration of exposed salt-marsh fucoids. Marine Biology 34: 349-359. Briske, D.D. and J.L. Butler 1989. Density-dependent regulation of ramet populations w i t h i n the bunchgrass Schizachyrium scoparium: interclonal versus intraclonal interference. Journal of Ecology 77: 963-97'4. Britting, S.A. and D.J. Chapman 1993. Physiological comparison of the isomorphic life history phases of the high intertidal alga Endocladia muricata (Rhodophyta). Journal of Phycology 29: 739-745. Brown, M.T. 1987. Effects of desiccation on photosynthesis of intertidal algae from a southern New Zealand shore. Botanica Marina 30: 121-7. Burns, R.L. and A.C. Mathieson 1972. Ecological studies of economic red algae. III. Growth and reproduction of natural and harvested populations of Gigartina stellata (Stackhouse) Batters i n New Hampshire. Journal of Experimental Marine Biology and Ecology 9: 77-95. Carmer, S.G. and U.M. Walker 1982. Baby Bear's dilemma: a statistical tale. Agronomy Journal 74: 122-124. Carrington, E. 1990. Drag and dislodgment of an intertidal macroalga: consequences of morphological variation i n Mastocarpus papillatus Kutzing. Journal of Experimental Marine Biology and Ecology 139: 185-200. Carrington Bell, E. 1995. Environmental and morphological influences on thallus temperature and desiccation of the intertidal alga Mastocarpus papillatus Kutzing. Journal of Experimental Marine Biology and Ecology 191: 29-55. Carter, A.R. and R.J. Anderson 1985. Regrowth after experimental harvesting of the agarophyte Gelidium pristoides (Gelidiales: Rhodophyta) i n the eastern Cape province. South African Journal of Marine Science 3: 111-118. Carter, A.R. and R.H. Simons 1987. Regrowth and production capacity of Gelidium pristoides (Gelidiales, Rhodophyta) under various harvesting regimes at Port Alfred, South Africa. Botanica Marina 30: 227-231. Casal, J.J., V.A. Deregibus and R.A. Sanchez 1985. Variations i n tiller dynamics and morphology i n Lolium multiflorum Lam. vegetative and reproductive plants as affected by differences i n red/far-red irradiation. Annals of Botany 56: 59-65. Chapman, A.R.O. 1986. Population and community ecology of seaweeds. Advances in Marine Biology 23: 1-161. Chen, L.C.-M., T. Edelstein and J. McLachlan 1974. The life history of Gigartina stellata (Stackh.) Batt. (Rhodophyceae, Gigartinales) i n culture. Phycologia 13: 287-294. Chopin, T., J.D. Pringle and R.E. Semple 1988. Reproductive capacity of dragraked and non-dragraked Irish moss [Chondrus crispus Stackhouse) beds i n the southern gulf of St. Lawrence. Canadian Journal of Fisheries and Aquatic Sciences 45: 758-66. Cook, R.E. 1985. Growth and development i n clonal plant populations. In: (Jackson, J.B.C., L.W. Buss and R.E. Cook, eds.) Population biology and evolution of clonal organisms. Yale University Press, New Haven and London, pp. 259-296. Correa, J.A. and J.L. McLachlan 1991. Endophytic algae of Chondrus crispus (Rhodophyta). III. Host specificity. Journal of Phycology 27: 448-459. Cousens, R. and M.J. Hutchings 1983. The relationship between density and mean frond weight in monospecific seaweed stands. Nature 301: 240-241. Cousens, R. and M. Mortimer 1995. Dynamics of weed populations. Cambridge University Press, Cambridge, 332 p. Craigie, J.S. and J.D. Pringle 1978. Spatial distribution of tetrasporophytes and gametophytes i n four Maritime populations of Chondrus crispus. Canadian Journal of Botany 56: 2910-2914. Creed, J.C. 1995. Spatial dynamics of a Himanthalia elongata (Fucales, Phaeophyta) population. Journal of Phycology 31: 851-859. Dai, T. and R.G. Wiegert 1996. Ramet population dynamics and net aerial primary productivity of Spartina alterniflora. Ecology 77: 276-288. Davison, I.R. 1991. Environmental effects on algal photosynthesis: temperature. Journal of Phycology 27: 2-8. Davison, I.R. and G.A. Pearson 1996. Stress tolerance i n intertidal seaweeds. Journal of Phycology 32: 197-211. Dayton, P.K. 1985. Ecology of kelp communities. Annual Review of Ecology and Systematics 16: 215-245. Dean, T.A., K. Thies and S.L. Lagos 1989. Survival of juvenile giant kelp: the effects of demographic factors, competitors, and grazers. Ecology 70: 483-495. de Kroon, H. 1993. Competition between shoots i n stands of clonal plants. Plant Species Biology 8: 85-94. de Kroon, H. and R. Kal l i o l a 1995. Shoot dynamics of the giant grass Gynerium sagittatum i n Peruvian Amazon floodplains, a clonal plant that does show self-thinning. Oecologia 101: 124-131. de Kroon, H., T. Hara and R. Kwant 1992. Size hierarchies of shoots and clons in clonal herbs monocultures: do clonal and non-clonal plants compete differently? Oikos 63: 410-419. de Kroon, H. and R. Kwant 1991. Density-dependent growth responses in two clonal herbs: regulation of shoot density. Oecologia 86: 298-304. De Wreede, R.E. 1985. Destructive (harvest) sampling. In: (M.M. Littler and D.S. Littler, eds.), Handbook of phycological methods. Ecological field methods: macroalgae. Cambridge University Press, London, pp. 147-160. De Wreede, R. and T. Klinger 1988. Reproductive strategies i n algae. In: (Lovett-Doust, J. and L. Lovett-Doust, eds.) Plant reproductive ecology: patterns and strategies. Oxford University Press, Oxford, pp. 267-284. De Wreede, R.E. and L.G. Green 1990. Patterns of gametophyte dominance of Iridaea splendens (Rhodophyta) i n Vancouver Harbour, Vancouver, British Columbia, Canada. Journal of Applied Phycology 2: 27-34. Denny, M. 1995. Predicting physical disturbance: mechanistic approaches to the study of survivorship on wave-swept shores. Ecological Monographs 65: 371-418. Denny, M.W., V. Brown, E. Carrington, G. Kraemer and A. Mi l l e r 1989. Fracture mechanics and the survival of wave-swept macroalgae. Journal of Experimental Marine Biology and Ecology 127: 211-228. Dickerman, J.A. and R.G. Wetzel 1985. Clonal growth i n Typha latifolia: population dynamics and demography of the ramets. Journal of Ecology 73: 535-552. Dring, M.J. 1988. Photocontrol of development i n algae. Annual Review of Plant Physiology and Plant Molecular Biology 39: 157-174. Dudgeon, S.R., J.E. Kiibler, R.L. Vadas and I.R. Davison 1995. Physiological responses to environmental variation i n intertidal red algae: does thallus morphology matter? Marine Ecology Progress Series 117: 193-206. Dyck, L.J. and R.E. De Wreede 1995. Patterns of seasonal demographic change i n the alternate isomorphic stages of Mazzaella splendens (Gigartinales, Rhodophyta). Phycologia 34: 390-395. Dyck, L., R.E. De Wreede and D. Garbary 1985. Life history phases i n Iridaea cordata (Gigartinaceae): relative abundance and distribution from British Columbia to California. Japanese Journal of Phycology 33: 225-232. Edgington, E.S. 1987. Randomization tests. Second Edition. Marcel Dekker, New York and Basel, 341 p. Elliott, J.M. 1977. Some methods for the statistical analysis of samples of benthic invertebrates. Freshwater Scientific Association, Sci. Publ. 25. Titus Wilson & Son, Kendal, 160 p. Ellstrand, N.C. and M.L. Roose 1987. Patterns of genotypic diversity i n clonal plant species. American Journal of Botany 74: 123-131. Eriksson, O. and L. Jerling 1990. Hierarchical selection and risk spreading i n clonal plants. In: (van Groenendael, J. and H. de Kroon, eds.) Clonal growth in plants: regulation and function. SPB Academic Publishers, The Hague, pp. 79-94. Fernandez, C. and M.P. Menendez 1991a. Ecology of Chondrus crispus Stackhouse (Rhodophyta) i n the Northern coast of Spain. I Seasonal patterns. Scientia Marina 55: 475-481. Fernandez, C. and M.P. Menendez 1991b. Ecology of Chondrus crispus Stackhouse (Rhodophyta) on the Northern coast of Spain II. Reproduction. Botanica Marina 34: 303-310. Fisher, R.A. and F. Yates 1963. Statistical tables for biological, agricultural and medical research. 6th Edition. Oliver & Boyd, Edinburgh, 138 p. Flores-Moya, A., J.A. Fernandez and F.X. N i e l l 1996. Growth pattern, reproduction, and self-thinning i n seaweeds, fournal of Phycology 32: 767-769. Foster, M.S. 1975. Regulation of algal community development i n a Macrocystis pyrifera forest. Marine Biology 12: 331-342. Foster, M.S. 1982. Factors controlling the intertidal zonation of Iridaea flaccida (Rhodophyta). Journal of Phycology 18: 285-294. Foster, M.S., A.P. de Vogelaere, J.S. Oliver, J.S. Pearse and C. Harrold 1991. Open coast intertidal and shallow subtidal ecosystems of the northeast Pacific. In: (Mathieson, A.C. and P.H. Nienhuis, eds.) Ecosystems of the world 24. Intertidal and littoral ecosystems. Elsevier, Amsterdam, London, New York, and Tokyo, pp. 235-272. Garbary, D.J. and R.E. De Wreede 1988. Life history phases i n natural populations of Gigartinaceae (Rhodophyta): quantification using resorcinol. In: (Lobban, C.S., D.J. Chapman and B.P. Kremer, eds.) Experimental phycology. A laboratory manual. Cambridge University-Press, Cambridge, pp. 174-178. Gaylord, B., C.A. Blanchette and M.W. Denny 1994. Mechanical consequences of size i n wave-swept algae. Ecological Monographs 64: 287-313. Gerard, V.A. 1982. In situ water motion and nutrient uptake by the giant kelp Macrocystis pyrifera. Marine Biology 69: 51-54. Gomez, I.M. and R.C. Westermeier 1991. Frond regrowth from basal disc i n Iridaea laminarioides (Rhodophyta, Gigartinales) at Mehuin, southern Chile. Marine Ecology Progress Series 73: 83-91. Gonen, Y., E. Kimmel, E. Tel-Or and M. Friedlander 1996. Intercellular assimilate translocation i n Gracilaria cornea (Gracilariaceae, Rhodophyta). Hydrobiologia 326/327: 421-428. Guiry, M.D. and J.A. West 1983. Life history and hibridization studies on Gigartina stellata and Petrocelis cruenta (Rhodophyta) i n the North Atlantic. Journal of Phycology 19: 474-494. Hannach, G. and B. Santelices 1985. Ecological differences between the isomorphic reproductive phases of two species if Iridaea (Rhodophyta: Gigartinales). Marine Ecology Progress Series 22: 291-303. Hannach, G. and J.R. Waaland 1986. Environment, distribution and production of Iridaea. Aquatic Botany 26: 51-78. Hansen, J.E. 1977. Ecology and natural history of Iridaea cordata (Gigartinales, Rhodophyta) growth. Journal of Phycology 13: 395-402. Hansen, J.E. 1981. Studies on the population dynamics of Iridaea cordata (Gigartinaceae, Rhodophyta). Procedings of the 8th International Seaweed Symposium (Bangor, Wales, 18-23 August 1974): 336-341. Hansen, J.E. and W.T. Doyle 1976. Ecology and natural history of Iridaea cordata (Rhodophyta; Gigartinaceae): population structure. Journal of Phycology 12: 273-278. Hara, T. 1994. Growth and competition i n clonal plants - Persistence of shoot populations and species diversity. Folia Geobotanica & Phytotaxonomica, Praha, 29: 181-201. Hara, T. 1988. Dynamics of size structure i n plant populations. Trends in Ecology and Evolution 3: 129-133. Hara, T., J. van der Toorn and J. Mook 1993. Growth dynamics and size structure of shoots of Phragmites australis, a clonal plant. Journal of Ecology 81: 47-60. Harper, J.L. 1977. Population biology of plants. Academic Press, London, New York & San Francisco, 892 p. Hartnett, D.C. and F.A. Bazzaz 1985. The regulation of leaf, ramet and genet densities i n experimental populations of the rhizomatous perennial Solidago canadensis. Journal of Ecology 73: 429-443. Hasegawa, Y. and E. Fukuhara 1952. Ecological studies on Iridophycus cornucopiae (P. et R.) Setchell et Gardner. 1. On the seasonal change of the number of gametophytes and sporophytes. Bulletin of the Hokkaido Reg. Fish. Res. Lab. 3: 20-30. Hasegawa, Y. and E. Fukuhara 1955. Ecological studies on Iridophycus cornucopiae (P. et R.) Setch. et Gardn. 5. On the seasonal and local variations i n the weight of fronds. Bulletin of the Hokkaido Reg. Fish. Res. Lab. 12: 29-39. Hay, M.E. 1981. The functional morphology of turf-forming seaweeds: persistence i n stressful marine habitats. Ecology 62: 739-750. Henley, WJ. 1993. Measurement and interpretation of photosynthetic light-response curves i n algae i n the context of photoinhibition and diel changes. Journal of Phycology 29: 729-739. Henley, W.J., S.T. Lindley, G. Levavasseur, C.B. Osmond and J. Ramus 1992. Photosynthetic response of Viva rotundata to light and temperature during emersion on an intertidal sand flat. Oecologia 89: 516-523. Herben, T., T. Hara, C. Marshall and L. Soukupova 1994. Plant clonality: biology and diversity. Folia Geobotanica and Phytotaxonomica, Praha 29: 113-122. Herbert, S.K. 1990. Photoinhibition resistance i n the red alga Porphyra perforata. The role of photoinhibition repair. Plant Physiology 92: 514-519. Herbert, S.K. and J.R. Waaland 1988. Photoinhibition of photosynthesis i n a sun and a shade species of the red algal genus Porphyra. Marine Biology 97:1-7. Hester, M.W., K.L. McKee, D.M. Burdick, M.S. Koch, K.M. Flynn, S. Patterson and I.A. Mendelssohn 1994. Clonal integration i n Spartina pattens across a nitrogen and salinity gradient. Canadian Journal of Botany 72: 767-770. Hodgson, L.M. 1981. Photosynthesis of the red alga Gastroclonium coulteri (Rhodophyta) in response to changes i n temperature, light intensity, and desiccation. Journal of Phycology 17: 37-42. Hommersand, M.H., M.D. Guiry, S. Fredericq and G.L. Leister 1993. New perspectives i n the taxonomy of the Gigartinaceae (Gigartinales, Rhodophyta). Hydrobiologia 260/261: 105-120. Hommersand, M.H., S. Fredericq and D.W. Freshwater 1994. Phylogenetic systematics and biogeography of the Gigartinaceae (Gigartinales, Rhodophyta) based on sequence analysis of rbcL. Botanica Marina 37: 193-203. Howell, D.C. 1992. Statistical methods for psychology. Third Edition. Duxbury Press, Belmont, 693 p. Hurlbert, S.H. 1990. Pastor binocularis: now we have no excuse. Ecology 71: 1222-1223. Hutchings, M.J. 1979. Weight-density relationships i n ramet populations of clonal perennial herbs, with special reference to the -3/2 power law. Journal of Ecology 67: 21-33. Jackson, G.A. 1984. Internal wave attenuation by coastal kelp stands. Journal of Physical Oceanography 14: 1300-1306. Jackson, J.B.C., L.W. Buss and R.E. Cook 1985. Population biology and evolution of clonal organisms. Yale University Press, New Haven and London, 530 p. John, D.M., S.J. Hawkins and J.H. Price (eds.) 1992. Plant-animal interactions in the marine benthos. Clarendon Press, Oxford, 570 p. Johnson, W.S., A. Gigon, S.L. Gulmon and H.A. Mooney 1974. Comparative photosynthetic capacities of intertidal algae under exposed and submerged conditions. Ecology 55: 450-453. Jonsson, B.O., I.S. Jonsdottir and N. Cronberg 1996. Clonal diversity and allozyme variation i n populations of the arctic sedge Car ex bigelowii (Cyperaceae). Journal of Ecology 84: 449-459. Kain, J.M. 1987. Photoperiod and temperature as triggers i n the seasonality of Delesseria sanguinea. Helgolander Meeresuntersuchungen 41: 355-370. Kain (Jones), J.M. and M.J. Bates 1993. The reproductive phenology of Delesseria sanguinea and Odonthalia dentata off the Isle of Man. European Journal of Phycology 28: 173-182. Kain, J.M. and T.A. Norton 1990. Marine Ecology. In: (Cole, K.M. and R.G. Sheath, eds.) Biology of the red algae. Cambridge University Press, Cambridge, pp. 377-422. Kim, D.H. and R.E. Norris 1981. Possible pathways i n world distribution of the Gigartinaceae. Procedings of the 8th International Seaweed Symposium (Bangor, Wales, 18-23 August 1974): 377-382. Kim, J.H. 1995. Intertidal community structure, dynamics and models: mechanisms and the role of biotic and abiotic interactions. Ph.D. Thesis, The University of British Columbia, Vancouver, Canada, 147 p. Kim, J.H. and R.E. De Wreede 1996a. Effects of size and season of disturbance on algal patch recovery i n a rocky intertidal community. Marine Ecology Progress Series 133: 217-228. Kim, J.H. and R.E. De Wreede 1996b. Distribution and feeding preference of a high intertidal littorinid. Botanica Marina 39: 561-569. Knox, R.G., R.K. Peet and N.L. Christensen 1989. Population dynamics i n loblolly pine stands: changes i n skewness and size inequality. Ecology 70: 1153-1166. Koch, E.W. 1993. The effect of water flow on photosynthetic processes of the alga Ulva lactuca L. Hydrobiologia 260/261: 457-462. Krebs, C.J. 1989. Ecological methodology. Harper & Row, New York, 654 p. Kfibler, J.E. and I.R. Davison 1993. High-temperature tolerance of photosynthesis i n the red alga Chondrus crispus. Marine Biology 117: 327-335. Lapham, J. and D.S.H. Drennan 1987. Intraspecific regulation of populations of the clonal herb, Cyperus esculentus. Journal of Applied Ecology 24: 1011-1024. Lazo, M.L., M. Greenwell and J. McLachlan 1989. Population structure of Chondrus crispus Stackhouse (Gigartinaceae, Rhodophyta) along the coast of Prince Edward Island, Canada: distribution of gametophytic and sporophytic fronds. Journal of Experimental Marine Biology and Ecology 126: 45-58. Levitt, G.J. and J.J. Bolton 1991. Seasonal patterns of photosynthesis and physiological parameters and the effects of emersion i n littoral seaweeds. Botanica Marina 34: 403-410. o Lindgren, A. and P. Aberg 1996. Proportion of life cycle stages of Chondrus crispus and its population structure: a comparison between a marine and an estuarine environment. Botanica Marina 39: 263-268. Littler, M.M and D.S. Littler 1985. Nondestructive sampling. In: (M.M. Littler and D.S. Littler, eds.), Handbook of phycological methods. Ecological field methods: macroalgae. Cambridge University Press, London, pp. 161-175. Lobban, CS. and P.J. Harrison 1994. Seaweed ecology and physiology. Cambridge University Press, Cambridge, New York, and Melbourne, 366 P-Luning, K. 1990. Seaweeds. Their environment, biogeography, and ecophysiology. John Wiley & Sons, New York, Chichester, Brisbane, Toronto, and Singapore, 527 p. Luning, K. 1993. Environmental and internal control of seasonal growth i n seaweeds. Hydrobiologia 260/261: 1-14. Luning, K. and I. torn Dieck 1989. Environmental triggers i n algal seasonality. Botanica Marina 32: 389-397. Luning, K. and P. Kadel 1993. Daylength range for circannual rhythmicity i n Pterygophora californica (Alariaceae, Phaeophyta) and synchronization of seasonal growth by daylength cycles i n several other brown algae. Phycologia 32: 379-387. Luxoro, C. and B. Santelices 1989. A d d i t i o n a l evidence for ecological differences among isomorphic reproductive phases of Iridaea laminarioides (Rhodophyta: Gigartinales). Journal of Phycology 25: 206-212. Mac Arthur, R.H. and E.O. Wilson 1967. The theory of island biogeography. Princeton University Press, Princeton, 203 p. MacFarlane, C. 1952. A survey of certain seaweeds of commercial importance in southwest Nova Scotia. Canadian Journal of Botany 30: 78-97. Madsen, T.V. and S.C. Maberly 1990. A comparison of air and water as environments for photosynthesis by the intertidal alga Fucus spiralis (Phaeophyta). Journal of Phycology 26: 24-30. Maggs, C.A. 1988. Intraspecific life history variability i n the Florideophycidae (Rhodophyta). Botanica Marina 31: 465-490. Maggs, C.A. and D.P. Cheney 1990. Competition studies of marine macroalgae i n laboratory culture. Journal of Phycology 26: 18-24. Manly, B.F.J. 1991. Randomization and Monte Carlo methods in biology. Chapman and Hall, London, New York, Tokyo, Melbourne, and Madras, 281 p. Marshall, C. 1990. Source-sink relations of interconnected ramets. In: (van Groenendael, J. and H. de Kroon, eds.) Clonal growth in plants: regulation and function. SPB Academic Publishing, The Hague, pp. 23-41. Marshall, S.M., L. Newton and A.P. Orr 1949. A study of certain British seaweeds and their utilization in the preparation of agar. Ministry of Health and Supply, London, 101 p. Martinez, E. and B. Santelices 1992. Size hierarchy and the -3/2 "power law" relationship i n a coalescent seaweed. Journal of Phycology 28: 259-264. Masuda, M. and Y. Hashimoto 1993. Autecological studies on Chondrus nipponicus Yendo (Gigartinales, Rhodophyta). Japanese Journal of Phycology 41: 99-111. Mathieson, A.C. 1989. Phenological patterns of northern New England seaweeds. Botanica Marina 32: 419-438. Mathieson, A.C. and R.L. Burns 1975. Ecological studies of economic red algae. V. Growth and reproduction of natural and harvested populations of Chondrus crispus Stackhouse i n New Hampshire. Journal of Experimental Marine Biology and Ecology 17: 137-156. May, G. 1986. Life history variations i n a predominantly gametophytic population of Iridaea cor data (Gigartinaceae, Rhodophyta). Journal of Phycology 22: 448-455. McCandless, E.L. and J.S. Craigie 1979. Sulfated polysaccharides i n red and brown algae. Annual Review of Plant Physiology 30: 41-53. McCandless, EX., J.A. West and M.D. Guiry 1983. Carrageenan patterns i n the Gigartinaceae. Biochem. Syst. Ecology 11: 175-182. McLachlan, J. 1985. Macroalgae (seaweeds): industrial resources and their utilization. Plant and Soil 89: 137-157. McLachlan, J.L., N.I. Lewis and MX. Lazo 1988. Biological considerations of Chondrus crispus Stackhouse (Rhodophyta, Gigartinaceae) i n the southern gulf of St. Lawrence, Canada. Gayana, Botanica 45: 29-54. McQuaid, CD. 1985. Seasonal variation i n biomass and zonation of nine intertidal algae i n relation to changes i n radiation, sea temperature and tidal regime. Botanica Marina 28: 539-544. Mead, R. 1991. The design of experiments. Statistical principles for practical applications. Cambridge University Press, Cambridge, New York, Port Chester, Melbourne, Sydney, 620 p. Mikami, H. 1965. A systematic study of the Phyllophoraceae and Gigartinaceae from Japan and its vicinity. Sci. Papers Institute Algol. Res. Fac. Sci. Hokkaido University 5 (2): 181-285. Molenaar, F.J. and A.M. Breeman 1994. Ecotypic variation i n Phyllophora pseudoceranoides (Rhodophyta) ensures winter r e p r o d u c t i o n throughout its entire geographic range. Journal of Phycology 30: 392-402. Mork, M. 1996. The effect of kelp in wave dumping. Sarsia 80: 323-327. Murphy, J.S. and D.D. Briske 1994. Density-dependent regulation of ramet recruitment by the red:far-red ratio of solar radiation: a field evaluation with the bunchgrass Schizachyrium scoparium. Oecologia 97: 462-469. Myers, J.L. and A.D. Well 1991. Research design and statistical analysis. Harper Collins, New York, 713 p. Neushul, M. 1990. A n t i v i r a l carbohydrates from marine red algae. Hydrobiologia 204/205: 99-104. Noda, H., H. Amano, K. Arashima and K. Nisizawa 1990. Antitumor activity of marine algae. Hydrobiologia 204/205: 577-584. Norris, R.E. and D.H. K i m 1972. Development of thalli i n some Gigartinaceae. In: (Abbott, LA. and M. Kurogi, eds.) Contributions to the systematics of benthic marine algae of the North Pacific. Japanese Society of Phycology, Kobe, pp. 265-275. Norton, T.A. 1991. Conflicting constraints on the form of intertidal algae. British Phycological Journal 26: 203-218. Norton, T.A., A.C. Mathieson and M. Neushul 1982. A review of some aspects of form and function in seaweeds. Botanica Marina 25: 501-510. Oates, B.R. 1985. Photosynthesis and amelioration of desiccation i n the intertidal saccate alga Colpomenia peregrina. Marine Biology 89: 109-119. Oates, B.R. 1986. Components of photosynthesis i n the intertidal saccate alga Halosaccion americanum (Rhodophyta, Palmariales). Journal of Phycology 22: 217-223. Oates, B.R. and S.N. Murray 1983. Photosynthesis, dark respiration and desiccation resistance of the intertidal seaweeds Hesperophycus harveyanus and Pelvetia fastigiata f. gracilis. Journal of Phycology 19: 371-380. Olson, A.M. 1990. Algal life history stages respond differently to desiccation and herbivory. (Abstract) Bulletin of the Ecological Society oj America (Suppl.) 71: 274. Pacheco-Ruiz, I., J.A. Zertuche-Gonzalez, A. Cabello-Pasini and B.H. Brinkhuis 1992. Growth responses and seasonal biomass variation of Gigartina pectinata Dawson (Rhodophyta) i n the Gulf of California. Journal of Experimental Marine Biology and Ecology 157: 263-274. Padilla, D.K. 1984. The importance of form: differences i n competitive ability, resistance to consumers and environmental stress i n an assemblage of coralline algae. Journal of Experimental Marine Biology and Ecology 79: 105-127. Penot, M. 1992. Le transport a longue distance chez les grandes algues -Acquis et perspectives-. Cryptogamie, Algologie 13: 197-224. Petraitis, P.S. 1995. Use of average vs. total biomass i n self-thinning relationships. Ecology 76: 656-658. Pianka, E.R. 1982. Ecologia evolutiva. Omega, Barcelona, 365 p. Piriz, M.L. 1996. Phenology of a Gigartina skottsbergii Setchell et Gardner population i n Chubut province (Argentina). Botanica Marina 39: 311-316. Pitelka, L.F. 1984. Application of the -3/2 power law to clonal herbs. The American Naturalist 123: 442-449. Pitelka, L.F. and J.W. Ashmun 1985. Physiology and integration of ramets i n clonal plants. In: (Jackson, J.B.C., L.W. Buss and R.E. Cook, eds.) Population biology and evolution of clonal organisms. Yale University Press, New Haven and London, pp. 399-435. Poblete, A., A. Candia, I. Inostroza and R. Ugarte 1985. Crecimiento y fenologia reproductiva de Iridaea ciliata K t i t z i n g (Rhodophyta, Gigartinales) en una pradera submareal. Biologia Pesquera 14: 23-31. Polanshek, A.R. and J.A. West 1977. Culture and hybridization studies on Gigartina papillata. Journal of Phycology 11: 434-439. Prince, J.S. and J.M. Kingsbury 1973. The ecology of Chondrus crispus at Plymouth, Massachusetts. II. Field studies. American Journal of Botany 60: 964-975. Pringle, J.D. and R.E. Semple 1988. Impact of harvesting on Irish moss [Chondrus crispus Stackhouse) frond size-class structure. Canadian Journal of Fisheries and Aquatic Sciences 45: 767-773. Pueschel, C M . 1990. Cell structure. In: (Cole, K. M. and R.G. Sheath, eds.) Biology of the red algae. Cambridge University Press, Cambridge, pp. 7-42. Pybus, C. 1977. The ecology of Chondrus crispus and Gigartina stellata (Rhodophyta) i n Galway bay. Journal of the Marine Biological Association of the United Kingdom 57: 609-628. Quadir, A., P.J. Harrison, and R.E. De Wreede 1979. The effects of emergence and submergence on the photosynthesis and respiration of marine macrophytes. Phycologia 18: 83-88. Reed, D.C. 1990. A n experimental evaluation of density dependence i n a subtidal algal population. Ecology 71: 2286-2296. Room, P.M. and M.H. Julien 1994. Population-biomass dynamics and the absence of -3/2 self-thinning i n the clonal weed Salvinia molesta. Australian Journal of Ecology 19: 26-34. Rueness, J. 1978. A note on development and reproduction i n Gigartina stellata (Rhodophyta, Gigartinales) from Norway. British Phycological Journal 13: 87-90. Russell, G. 1986. Variation and natural selection i n marine macroalgae. Oceanography and Marine Biology Annual Review 24: 309-377. Santelices, B. 1990. Patterns of reproduction, dispersal and recruitment i n seaweeds. Oceanography and Marine Biology Annual Review 28: 177-276. Santelices, B. and R. Norambuena 1987. A harvesting strategy for Iridaea laminarioides i n central Chile. Hydrobiologia 151/152: 329-333. Santelices, B., P. Camus and A.J. Hoffmann 1989. Ecological studies for harvesting and culturing Gymnogongrus furcellatus (Rhodophyta, Gigartinales) i n central Chile. Journal of Applied Phycology 1: 171-181. Santelices, B., J.A. Correa, I. Meneses, D. Aedo and D. Varela 1996. Sporeling coalescence and intraclon a l v a r i a t i o n i n Gracilaria chilensis (Gracilariales, Rhodophyta). Journal of Phycology 32: 313-322. Santos, R. 1995. Size structure and inequality i n a commercial stand of the seaweed Gelidium sesquipedale. Marine Ecology Progress Series 119: 253-263. Schaal, B.A., S.L. O'Kane, and S.H. Rogstad 1991. D N A variation i n plant populations. Trends in Ecology and Evolution 6: 329-333. Schiel, D.R. 1985. Growth, survival and reproduction of two species of marine algae at different densities i n natural stands. Journal of Ecology 73: 199-217. Schmitt, J., D.W. Ehrhardt and M. Cheo 1986. Light-dependent dominance and supression i n experimental radish populations. Ecology 67: 1502-1507. Schmitz, K. 1990. Algae. In: (Behnke, H.-D. and R.D. Sjolund, eds.) Sieve elements. Springer-Verlag, Berlin, pp. 1-18. Scrosati, R., D.J. Garbary and J. McLachlan 1994. Reproductive ecology of Chondrus crispus (Rhodophyta, Gigartinales) from Nova Scotia, Canada. Botanica Marina 37: 293-300. Selivanova, O.N. and G.G. Zhigadlova 1997. Marine algae of the Commander Islands. Preliminary remarks on the revision of the flora III. Rhodophyta. Botanica Marina 40: 15-24. Shaughnessy, F.J. and R.E. De Wreede 1991. Reliability of the resorcinol method for identifying isomorphic phases i n the Gigartinaceae (Rhodophyta). Journal of Applied Phycology 3: 121-127. Shaughnessy, F.J., R.E. De Wreede and E.C. Bell 1996. Consequences of morphology and tissue strength to blade survivorship of two closely related Rhodophyta species. Marine Ecology Progress Series 136: 257-266. Sheath, R.G., K.L. Van Alstyne and K.M. Cole 1985. Distribution, seasonality and reproductive phenology of Bangia atropurpurea (Rhodophyta) i n Rhode Island, U.S.A. Journal of Phycology 21: 297-303. Sokal, R.R. and F.J. Rohlf 1995. Biometry. The principles and practice of statistics in biological research. T h i r d Edition. W.H. Freeman and Company, New York, 887 p. Solangaarachchi, S.M. and J.L. Harper 1987. The effect of canopy filtered light on the growth of white clover Trifolium repens. Oecologia 72: 372-376. Soto, D. and S.H. Hurlbert 1991. Long-term experiments on calanoid-cyclopoid interactions. Ecological Monographs 61: 245-265. Stuefer, J.F., H.J. During and H. de Kroon 1994. High benefits of clonal integration i n two stoloniferous species, i n response to heterogeneous light environments. Journal of Ecology 82: 511-518. Taylor, A.R.A. 1959-60. Growth and productivity of Irish moss. Annual Report and Investigators' Summaries, Fisheries Research Board of Canada, Biological Station St. Andrews, N.B. Taylor, P.R. and M.E. Hay 1984. Functional morphology of intertidal seaweeds: adaptive significance of aggregate vs. solitary forms. Marine Ecology Progress Series 18: 295-302. torn Dieck (Bartsch), I. 1991. Circannual growth rhythm and photoperiodic sorus induction i n the kelp Laminaria setchellii (Phaeophyta). Journal of Phycology 27: 341-350. Tveter, E. and A.C. Mathieson 1976. Sporeling coalescence i n Chondrus crispus (Rhodophyceae). Journal of Phycology 12: 110-118. Tveter-Gallagher, E. and A.C. Mathieson 1980. A n electron microscopy study of sporeling coalescence i n the red alga Chondrus crispus. Scanning Electron Microscopy 3: 571-579. Tveter-Gallagher, E., A.C. Mathieson and D.P. Cheney 1980. Ecology and developmental morphology of male plants of Chondrus crispus (Gigartinales, Rhodophyta). Journal of Phycology 16: 257-264. van den Hoek, C, D.G. Mann and H.M. Jahns 1995. Algae. An introduction to phycology. Cambridge University Press, Cambridge, 623 p. van Groenendael, J. and H. de Kroon (eds.) 1990. Clonal growth in plants: regulation and function. SPB Academic Publishing, The Hague. van Oppen, M.J.H., J.L. Olsen and W.T. Stam 1995. Genetic variation within and among North Atlantic and Baltic populations of the benthic alga Phycodrys rubens (Rhodophyta). European Journal of Phycology 30: 251-260. Waaland, J.R. 1975. Differences i n carrageenan i n gametophytes and tetrasporophytes of red algae. Phytochemistry 14: 1359-1362. Weiner, J. 1988. Variation i n the performance of individuals i n plant populations. In: (Davy, A.J., M.J. Hutchings and A.R. Watkinson, eds.) Plant population ecology. Blackwell Scientific Publications, Oxford, London, Edinburgh, Boston, and Melbourne, pp. 59-81. Weiner, J. 1990. Asymmetric competition i n plant populations. Trends in Ecology and Evolution 5: 360-364. Weiner, J. and O.T. Solbrig 1984. The meaning and measurement of size hierarchies i n plant populations. Oecologia 61: 334-336. Weiner, J. and S.C. Thomas 1986. Size variability and competition i n plant monocultures. Oikos 47: 211-222. Weller, D.E. 1987. A reevaluation of the -3/2 power rule of plant self-thinning. Ecological Monographs 57: 23-43. Weller, D.E. 1989. The interspecific size-density relationship among crowded plant stands and its implications for the -3/2 power rule of self-thinning. The American Naturalist 133: 20-41. Weller, D.E. 1990. W i l l the real self-thinning rule please stand up? - A reply to Osawa and Sugita. Ecology 71: 2004-2007. Westermeier, R., P.J. Rivera, M. Chacana and I. Gomez 1987. Biological bases for management of Iridaea laminarioides Bory i n southern Chile. Hydrobiologia 151/152: 313-328. Westoby, M. 1984. The self-thinning rule. Advances in Ecological Research 14: 167-225. Wheeler, W.N. 1980. Effect of boundary layer transport on the fixation of carbon by the giant kelp Macrocystis pyrifera. Marine Biology 56: 103-110. Wheeler, W.N. 1988. A l g a l productivity and hydrodynamics - a synthesis. Progress in Phycological Research 6: 23-58. White, J. 1985. The thinning rule and its applications to mixtures of plant populations. In: (White, J., ed.) Studies on plant demography: a Festchrift John L. Harper. Academic Press, London, pp. 291-309. Widen, B., N. Cronberg and M. Widen 1994. Genotypic diversity, molecular markers and spatial distribution of genets i n clonal plants, a literature survey. Folia Geobotanica & Phytotaxonomica, Praha 29: 245-263. Wilkinson, L., M.A. H i l l and E. Vang 1992. SYSTAT: Statistics, Version 5.2 Edition. SYSTAT, Inc., Evanston, 724 p. Yoda, K., T. Kira, H. Ogawa and K. Hozumi 1963. Self-thinning i n overcrowded pure stands under cultivated and natural conditions (Intraspecific competition among higher plants XI). Journal of Biology, Osaka City University 14: 107-129. Zeide, B. 1985. Tolerance and self-tolerance of trees. Forest Ecology Management 13: 149-166. Zupan, J.R. and J.A. West 1988. Geographic variation i n the life history of Mastocarpus papillatus (Rhodophyta). Journal of Phycology 24: 223-229. 188 Authors of scientific names cited in the text Ahnfeltiopsis concinna (J. Agardh) Silva et De Cew Analipus japonicus (Harvey) Wynne Ascophyllum nodosum (Linnaeus) Le Jolis Balanus cariosus Pallas Balanus glandula D a r w i n Callithamnion pikeanum H a r v e y Carpophyllum maschalocarpum (Turner) Greville Chondracanthus pectinatus (Dawson) L. et R. Aguilar Chondrus crispus Stackhouse Chondrus nipponicus Yendo Chthamalus dalli P i l s b u r y Cladophora columbiana C o l l i n s Corallina vancouveriensis Yendo Diadema antillarum P h i l i p p i Dictyota bartayresii Lamouroux Ecklonia radiata (C. Agardh) J. Agardh Egregia menziesii (Turner) Areschoug Endocladia muricata (Postels et Ruprecht) J. Agardh Fucus gardneri S i l v a Gelidium coulteri H a r v e y Gelidium sesquipedale (Clemente) Bornet et Thuret Gigartina polycarpa (Kutzing) Setchell et Gardner Gigartina skottsbergii Setchell et Gardner Gracilaria chilensis Bird, McLachlan et Oliveira Gracilaria cornea J. Agardh Gynerium sagittatum (Aublet) Palisot de Beauvois Halimeda opuntia (Linnaeus) Lamouroux Himanthalia elongata (Linnaeus) S. F. Gray Laurencia mcdermidiae (J. Agardh) Abbott Laurencia papillosa (Forsskal) Greville Macrocystis pyrifera (Linnaeus) C. Agardh Mastocarpus papillatus (C. Agardh) Kutzirig Mastocarpus stellatus (Stackhouse) Guiry Mazzaella affinis (Harvey) Fredericq Mazzaella cornucopiae (Postels et Ruprecht) Hommersand Mazzaella flaccida (Setchell et Gardner) Fredericq Mazzaella heterocarpa (Postels et Ruprecht) Fredericq Mazzaella laminarioides (Bory de Saint-Vincent) Fredericq Mazzaella splendens (Setchell et Gardner) Fredericq Mytilus californianus Conrad Mytilus trossulus Gould Odonthalia floccosa (Esper) Falkenberg Pelvetiopsis limitata Gardner Pollicipes polymerus Sowerby Pterosiphonia gracilis K y l i n Pterygophora californica Ruprecht Sarcothalia crispata (Bory de Saint-Vincent) Leister Sarcothalia stiriata (Turner) Leister Sargassum sinclairii Hooker et Harvey Schizachyriun scoparium var. frequens (Hubb) Gould Scytosiphon dotyi Wynne Sparisoma rubripinne Cuvier et Valenciennes 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088141/manifest

Comment

Related Items