The Distribution and Abundance of the Intertidal Prosobranchs Littorina Scutulata (Gould 1849) and L

THE DISTRIBUTION AND ABUNDANCE OF THE INTERTIDAL PROSOBRANCHS LITTORINA SCUTULATA (GOULD 1849) AND L. SITKANA (PHILIPPI 1845)

by Sylvia Behrens B.Sc,, Honors, University of British Columbia, 1968

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OP THE REQUIREMENTS FOR THE DEGREE OP MASTER OP SCIENCE

in the Department

of

Zoology

We accept this thesis as conforming to the

required standard

THE UNIVERSITY OP BRITISH COLUMBIA

July, 1971 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 representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.

Depa rtment

The University of British Columbia Vancouver 8, Canada i

ABSTRACT

An attempt was made to explain the distribution and abundance of

the intertidal prosobranch snails Littorina scutulata (Gould, 1649), and £. sitkana (Philippi, 1845) on beaches near the city of Vancouver,

in the Gulf Islands and on the west coast of Vancouver Island, British

Columbia. L. scutulata has a planktonic dispersal stage and is widely distributed. L. sitkana develops directly from benthic egg masses and

tends to be more restricted in its distribution. The egg masses of L.

sitkana are susceptible to desiccation at low tide, and consequently

this species thrives best in damp places such as mud flats, tide pools and crevices. L. sitkana appears to be selected against in wave-swept places, because it offers more resistance to wave action than does the comparatively streamlined L. scutulata.

Experimental manipulation of densities and species composition indicated that food limitation may take place in the summer (decreased growth rates at higher densities of snails and low food abundance) but not in the winter. Density dependent survivorship and natality were found for L. sitkana, indicating that a regulatory mechanism may be oper• ating. L. scutulata showed no such density dependent response at the densities examined.

The presence of scutulata reduced the survival of JL. sitkana and vice versa in comparison to control populations consisting of each

species alone.

Possible evolutionary survival strategies and competitive relation•

ships of these two species are discussed. ii

TABLE OF CONTENTS

Page ABSTRACT i LIST OF FIGURES iT LIST OF TABLES vi ACKNOWLEDGMENTS viii GENERAL INTRODUCTION 1 EXTERNAL MORPHOLOGY OF L. SITKANA AND L, SCUTULATA 5 GEOGRAPHIC DISTRIBUTION 8 SOME ASPECTS OF THE REPRODUCTIVE BIOLOGY OF L. SITKANA AND £. SCUTULATA 10 Introduction 10 Description of juvenile stages 10 Breeding Season 16 Desiccation of juvenile stages as a possible factor restricting L. sitkana's distribution 19 Materials and Methods 19 Results 21 Discussion 23 General Discussion 23

LOCAL DISTRIBUTION OF LITTORINES 25 A. HORIZONTAL DISTRIBUTION 25 Introduction 25 Survey of Beaches 27 Materials and Methods 29 Results 30 Responses of L. sitkana and L. scutulata to Physical factors 34 Materials and Methods 34 Results 37 Submergence of L. scutulata 57 Wave exposure 38 Desiccation 38 ^ Temperature 39 Salinity 39 Crevices 39 Discussion 42 B. VERTICAL DISTRIBUTION 43 Introduction 43 I. Upper Distribution 44 Materials and Methods 44 Results 45 iii

II. Limits to the Lower Distribution 45 Materials and Methods 48 Results 51 Acmaea scutum. L. sitkana interaction 51 Effect of Leptasterias hexactis predation on Acmaea paradigitalis and L. scutulata 51 Summary 51

THE ABUNDANCE OP LITTORINES 53 A. BEHAVIORAL RESPONSES 53 Introduction 53 Materials and Methods 54 Results 55 Densities 55 Pood Levels 57 Pood Detection 57 Shelter (crevices) 57 Discussion 58 B. DENSITY-SPECIES INTERACTION EXPERIMENT 58 Introduction 58 Materials and Methods 59 Food Abundance 67 Growth Rates 73 Methods 73 Results and Discussion 77 L. sitkana Natality 84 Survivorship and Mortality 90 Materials and Methods 90 Results 92 Discussion 103

DISCUSSION AND CONCLUSIONS 105

LITERATURE CITED 111

APPENDIX, Tables 1 to 56 114 iv

LIST OP FIGURES Page 1 Shell morphology of L. scutulata and L. sitkana 6 2 Map of study areas between Tofino and Coos Bay 9 3 IJ. scutulata egg capsule containing 5 eggs 11 4 L. scutulata eggs showing one-cell to eight-cell stages 12 5 L. scutulata veliger just before hatching 13 6 L. sitkana eggs inside an egg mass 14 7 Newly hatched L. sitkana 15 8a Number of L. sitkana egg masses produced inside experimental cages as a function of time 17 8b High splash pool on Edward King Island 26 9 Map of study areas between Sunset Marina and Victoria 27 10 Map of study areas on San Juan Island 28 11 Size frequency distributions of L. sitkana and L. scutulata 31 12 Size frequency distributions of L. sitkana and L. scutulata 32 13a Number of animals of both species found on rocks of different roughness 40 13b Tolerance of littorines to distilled water 41 14a Transect from Bowman's Bay 46 14b Length of Acmaea scutum versus length of L. sitkana of equal dry weight 49 15 Cages used for Acmaea-littorine interaction experiment 50 16 Modification of water table for food detection experiment 56 17 Cement stepping stones removed from vexar cages 60 18 Arrangement of cages at the Cantilever Pier Beach 61 19 Optical density versus wet weight and dry weight versus wet weight of Fragellaria 64 20 The relationship of subjective and objective estimate of standing crop of algae 66 21 Numerical estimate of standing crop of algae at three snail densities as a function of time 69 V

22 Numerical estimate of algae abundance on control slabs as a function of time 71

23 Diagram illustrating the relationship between lip increment and length increment and between original length and final length of L. scutulata 74 24 Relationship between length and lip increment of L. sitkana 75 25 Relationship between length and lip increment of L. scutulata 76 26 to Growth indices under different density and species 78 to 29 treatments 82 30 Number of egg masses produced from September 17 to December 9» 1969 under the three density treatments (uncorrected for equal density of L. sitkana) 86 31 Number of egg masses produced from September 17 to December 9, 1969 under the three density treatments (corrected for equal density of L. sitkana) 87 32 Spring natality for L. sitkana as a function of density 88 33 Mortality index of L. sitkana and L. scutulata as a function of time 93 34 Original number of littorines surviving one year as a function of density 94 35 L. sitkana mortality data in summer and winter 95 36 L. scutulata mortality data in summer and winter 96 37 Variability of summer mortality in right and left cages 97 38 Variability of winter mortality in right and left cages 98 39 Crushed shell mortality as a function of density 99 40 Number of non-sheltered animals as a function of density 101 vi

LIST OP TABLES Page 1 Time sequences for the developmental stages of L_. scutulata 115 2 Hatching success of L. sitkana egg masses suspended 116 from a boat moored in Vancouver Harbor 116 3 Survival of two size classes of juvenile L. sitkana caged at the 9 foot and 13 foot tidal levels at Lilly- Point, from May 17 to May 18, 1969 117 4 Hatching success of L. sitkana egg masses at Lilly Point 118 5 Position of L. sitkana egg masses found in 16 cages at Cantilever Pier Beach 119 6 Local distribution of littorines 120 7 Number of animals of both species found on the surface and in crevices of a quadrat taken at Marvista Resort 122 8 Ability of the two species of littorines to resist wave exposure in the field 124 9 Ability of the two species of littorines to resist being washed off the substrate with a jet of sea water 125 10 Tolerance of the two species of littorines to 3 and 4 days of desiccation at room temperature 126 11 Tolerance of littorines to high water temperature 127 12 Tolerance of littorines to salinity extremes 128 13 Electivity coefficients for Searlesia dira 129 14 Electivity coefficients for Leutasterias hexactis 130 15 Pood choice by Leptasterias hexactis as determined by contact with prey 131 16 Survival of L. scutulata caged in high intertidal and splash zone at Lilly Point 132 17 Mean growth increments of Littorina sitkana and Acmaea scutum from single and mixed species cages 133 18 Starfish predation on littorines and limpets at two tide levels 134 19 Dispersal behavior of L. sitkana under two densities 135 20 Number of littorines of both species found on cement slabs with and without algae 136 vii

21 Total number of littorines of both species leaving diatom covered and clean slabs 137 22 Position of L. sitkana 14 hours after introduction into modified water table 138 23 Behavioral response of littorines to crevices 139 24 Comparison between abundance of food in L. sitkana and L. scutulata cages throughout the year 140 25 Biomass and length increment relationships for L. sitkana and L. scutulata in August from single species low density treatments 141 26 The effects of density on L. sitkana natality 142 27 Tests on the effect of species composition on the natality rate of L. sitkana 143 28 Comparisons between the natality rate in single and mixed species cages and between left and right cages 144 29 Comparison of survivorship curves in 24 cages for species and density effects 145 30 L. sitkana survivorship and mortality data 147 31 Lf scutulata survivorship and mortality data 148 32 Incidence of cercaria shedding experimental animals 149 33-56 Friday Harbor density-species interaction experiment 150 33-40 Demography and food abundance from July 1969 to June 1970 150-159 41-48 L. sitkana growth data from July 1969 to June 1970 160-167 49-56 L. scutulata growth data from July 1969 to June 1970 168-175 viii

ACKNOWLEDGMENTS

I thank my advisor, Dr. Robin Harger for allowing me to formulate and attack my own research problems. His advice and criticisms were extremely helpful during the writing phase of this study.

I am grateful to Dr. John Stimpson for his interest in my work and his many valuable suggestions.

I would like to thank the faculty, staff and graduate students of the University of Washington Marine Laboratories and the Zoology

Department of the University of British Columbia for their cooperation.

Dr. Pu-shing Chia, Dr. N. Gilbert, Dolores Lautiente, Bruce Menge,

Dr. N.J. Wilimovsky, Dr. A.G. Lewis and Dr. D. McPhail were especially helpful.

This research was financed by the National Research Council of

Canada Grant # 67660 to Dr. J.R.E. Harger. 1

GENERAL INTRODUCTION

Littorines, or periwinkles are cosmopolitan in their distribution,

being found on rocky shores throughout the intertidal regions of the

world (Stephenson and Stephenson, 1949). At low tide they can be found

between barnacles, under rocks and seaweed and in crevices. Since they

are ubiquitous and easy to collect, littorines have been the object of

many studies.

As early as 1911 Baseman attempted to describe the physical

factors responsible for the oscillatory movements of Littorina, littorea.

located on vertical surfaces, which corresponded to tidal cycles. In

1916 Kanda looked at the negative geotrophic response of littorines to

a combination of factors such as light, angle of inclination, sub•

mergence and emergence, texture and moisture of substrate. He noted,

that animals appeared to be sensitive to desiccation; moving upward if

the substrate was dry.

Hertling and Ankel (1927) described the mode of development of

various Atlantic Lacuna and Littorina species. Littorina littorea and

Jj. neritold.es, both have planktonic egg capsules and veliger larvae. L.

littoralis fasten their gelatinous egg masses to the fronds of fucoids.

The veliger stage is passed inside the egg and the young snails hatch

as miniature adults. L. saxatilis on the other hand is viviparous.

Struhsaker and Costlow (1968) reared the Hawaiian Littorina picta from

egg capsules to miniature adults by feeding the veligers on the phyto-

plankton Phaeodactylum tricornutum. Just prior to metamorphosis the 2 veligers were observed to prefer substrates with algal cover to sub•

strates without algae. Although much attention has been devoted to the development of Atlantic and Hawaiian littorines, no one has studied the mode of development of the north-eastern Pacific littorines. In the present study I describe the development of L. sitkana and L. scutulata for the first time.

In California, L. scutulata is found in the intertidal zone and

]J, p3,anaxis lives in the spray zone, above the high water mark. Bock and Johnson (1968) feel that this zonation can be attributed to L,. planaxis', greater tolerance to desiccation and an inability of this species to obtain the right kind of food (e.g. lichens) in the inter• tidal. On be°:hes '•-^•'ind Vancouver and on the Gulf Islands L. sitkana and L. gcutoilata occur together without differentiation in pattern of vertical zor- ; -\. On exposed coasts of Vancouver Island however, one finds L. scutulata inhabiting the intertidal and L. sitkana living in

splash pools which are isolated from the ocean for most of the year.

The present study attempts to explain this and other distribution patterns of L. sitkana and L. scutulata,.

Littorines are grazers on microscopic and macroscopic marine algae

(Dahl, 1964; Poster, 1964). The effect of their grazing activity in reducing the standing crop of marine diatoms has been reported by

Castenholz (l96l). In the present study littorines were caged at three densities and their effect on the standing crop of algae for one year was measured.

The investigation of factors limiting the number of animals in a 3 population forms a topic of central interest in ecology. Animals need shelter from environmental extremes and from predators, as well as food for body maintainance, growth and reproduction. As the density of animals increases, both shelter and food generally become restricted in availabi• lity and an increased mortality from exposure, predation, starvation and susceptibility to diseases usually follows as well as a decrease in natality. Such density dependent responses by animal populations to the favorability of their physical and biological environment has the effect of allowing populations to increase when densities are low and resources abundant and decrease when densities are high and resources scarce.

According to Murdoch (1970) long-lived species, such as elephants, generally are more independent of environmental changes and thus tend to be numerically constant over time. Shortlived species such as bacteria react faster to fluctuations in environmental favorability by increasing or decreasing their numbers. In the present study a comparison of the numerical responses of two species of littorines to three density levels is made.

The competitive exclusion principle, or Gause's principle states that two ecologically similar species using the same resource, be it food or shelter, cannot coexist indefinitely for one species would be more efficient at utilizing that resource and thus would increase in numbers and displace the other species (Hardin, I960). This was the case with Paramecium caudatum and P. aurelia grown in culture vials (Gause,

1934). In single species cultures both species survived indefinitely but when grown together, the smaller species P. caudatum, with a 4 greater rate of increase, could acquire food more efficiently than P. aurelia and thus _P. caudatum increased in numbers and displaced P. aurelia. In nature, however environmental conditions are more variable in time and space and coexisting species utilising the same resource are not uncommon (Low, 1970; Harger, 1967).

L. sitkana and L_. scutulata eat the same food and overlap in their distribution in most habitats. In the present study littorines were caged in the field in single and mixed species treatments for a year in an attempt to determine whether competitive relations exist between the two species. 5

EXTERNAL MORPHOLOGY

Unweathered specimens of the two littorine species may be dis• tinguished by their external shell morphology (see Pig.l). The shell of L. scutulata is smooth and has a pointed apex, that of L. sitkana is roundish and has transverse grooves. L. sitkana is the faster growing of the two species and as a rule is larger than ,L. scutulata. Both species may reach up to 2 cm in length. Size however varies with loca• tion and is not particularly useful in littorine taxonomy. Shells of both species usually appear black when wet and grey when dry. L, sitkana is often more variable in colour, sculpture and shape than L. scutulata.

Shells of L. sitkana can be white, grey, orange, yellowish, brown or black. Complex banding patterns of one, two, three or more grey, brown or black stripes on a white background are found in L. sitkana.

The characteristic grooves of L. sitkana are sometimes eroded away in older specimen. The ratio of the height of spire to maximum width is approximately 1.2/l.

L. scutulata does not appear to be as variable in shell colour and morphology. This littorine as its Latin name indicates, is often speckled with many white dots. The occasional white L. scutulata with one black or brown band has been noted, but the elaborate banding and colour patterns of L. sitkana is absent in this species. Strong surf erodes the apex of the spire, thus giving L. scutulata on exposed beaches the squat shape of L. sitkana. Height of spire to maximum width ratio is approximately I.4/1. Littorina sitkana

Dorsal view Ventral view

Littorina scutulata

Figure 1. Shell morphology of Littorina sitkana

and Littorina scutulata. (Magnif«c*Hor\ 7

Erosion and attack by boring algae often imbue littorine shells with a perforated appearance. Lightly pigmented specimens may also be coloured by the epiphytic growth of green and brown algae. 8

GEOGRAPHICAL DISTRIBUTION

Only three species of littorines occur on the west coast of North

America (Urban,1962). Littorina scutulata has the widest distribution, occurring from Alaska to Baja, California (58°N to 19°N latitude).

Orldroy (1929) and Keen (1937) state that L. sitkana is present from the

Bering Sea to Puget Sound (48°N) and that L. planaxis occurs from Puget

Sound southward.

Thomas (1966) found L. sitkana in bays just north and south of

Lincoln City, Oregon. I confirmed the presence of Littorina sitkana in

Siletz Bay just south of Lincoln City where this species was found on scattered basaltic rocks on the island near Schooner Creek Bridge.

Thomas also claimed to have found L. sitkana at Sunset Bay and Winchester

Bay near Coos Bay and North Bend, Oregon. I however did not find L. sitkana at Sunset Bay. The photographs he took of these littorines looked like striped L. scutulata. IJ. sitkana does however occur much further south than Puget Sound.

I have never found L. planaxis in Puget Sound. The specimens of

L. planaxis in the Friday Harbor museum are those of L. sitkana with the grooves eroded. Neither Thomas in 1966 nor I have found L. planaxis north of Coos Bay.

10

SOME ASPECTS OP THE REPRODUCTIVE BIOLOGY OF L. SITKANA AND L. SCUTULATA

Introduction

Hertling, 1927 reports on the diverse modes of development of the

Atlantic littorine species. Littorina littorea and L. neritoides both have planktonic egg capsules and veliger larvae. L. littoralis fasten their gelatinous egg masses to the fronds of fucoids. The veliger stage is passed inside the egg and the young snails hatch as miniature adults.

L. saxatilis on the other hand is viviparous.

There appears to be no information in the literature concerning the development of IJ. sitkana and L_. scutulata.

L. scutulata released floating egg capsules (Fig. 3) in the laboratory during the week of June 22, 1970 and juvenile L. sitkana hatched from gelatinous egg masses (Fig. 6) collected from Brockton

Point (Fig. 9) in March of 1969. These observations confirmed that

L. scutulata (like L. littorea and L. neritoides) has planktotrophic development and L. sitkana (like L. littoralis) has le cithotrophic development.

Description of Juvenile Stages of L. scutulata and L. sitkana

Materials_and Methods

0 0

Three female L. scutulata were kept at 13 to 15 C in separate

stacking dishes (5 cm in diameter) with a little sea water. The water was changed at least once a day and the number of newly laid egg 11 a. capsules noted. Egg capsules were followed through the developmental stages to the hatched veligers at two temperatures (l3° to 15° and at

22 C).

Five IJ. sitkana egg masses in different stages of development were collected from Brockton Point on March 12, 1969. Each egg mass was divided into four more or less equal sections. One quarter of each of the five egg masses was placed in the bottom of a petri dish (5 cm in diameter) with a section of fine plankton netting wrapped around it.

Pour such dishes were prepared. Two nylon lines (lOO lb tested)' were suspended from a boat moored in Vancouver harbour. Two petri dishes- were attached with rubber bands to each line at points 50 cm and 100 cm below the surface of the water. After 13 days these dishes were examined.

Results

In a period of one week three female L_. scutulata produced from

749 to 1034 egg capsules each.

L_. scutulata egg capsules are 840/(in diameter, resemble trans• parent saucers and may contain from 1 to 6 eggs measuring 100 /\,±n diameter (Fig. 3).

The veligers hatched from the egg capsules in 7 to 8 days after laying in 13° to 15°C water and after 3 days in 22° water. Newly hatched larvae measured l60/{_in length.

L. sitkana egg masses (Fig. 6) vary in length from 5 mm to 15 mm and contain about 50 to 150 eggs. Eggs range from 0.9 mm to 1.0 mm in lib

Fig.3 L. scutulata egg capsule containing 5 eggs. Fig.4 L. scutulata eggs showing one-cell, two-cell, four-cell and eight-cell stages. Note the micromeres and macromeres in the top right hand egg. Fig.5 L. scutulata veliger just before hatching. Note the dense shell and the velum. 14 15

Fig.7 Newly hatched L. sitkana. Note the foot and tentacles of the uppermost snail. 16 diameter and embryo sizes range from 0.33 to 0.40 mm in width. Newly hatched snails are approximately 0.45 mm wide.

The jelly encasing the embryos is transparent and relatively hard.

As the shell of the embryo develops, the colour and consequently the colour of the entire egg mass changes from white to yellow to pink and finally to red. Sometimes the whole egg mass is covered with diatoms and will appear opaque brown in colour.

Less than half of the eggs in the petri dishes hatched into

•roundish snails having a dark brown and smooth shell (Table 2). The older snails showed the characteristic grooves of L. sitkana (Fig. 7).

Breeding Season

L. scutulata may breed for most of the year. Copulation was ob• served in the fall and early winter of 1969 and 1970 and in the summer of 1969. Newly settled individuals were found in February 1968 and in spring 1969. Egg capsules were produced in June 1970 at Friday Harbour

Marine Laboratories and in November 1970 at Sunset Bay near Coos Bay,

Oregon (Fig. 2).

L. sitkana egg masses can be found at various beaches throughout the year except during the summer months. Four egg masses were found in the cages by Cantilever pier (to be discussed in the last section) on June 22, 1970, but these had dried out, (Table 40). Peaks of egg mass abundance occur in the fall and early spring (Fig.8a). An extreme• ly mild spring in 1970 may have been responsible for a peak of abun• dance occurring earlier than in the previous year. (E.g. egg masses Fig. ^a Number of L. sitkana egg masses produced inside experimental cages as a function of time. 18 were collected from the Montague Harbour mud flat, Galiano Island, on

May 16, 1969 whereas on May 19, 1970 none were found, but young snails were already 2 mm in length.)

Egg laying is not synchronous from beach to beach; for example the first egg masses on San Juan Island were found at Pile Point in

August of 1969, whereas at Cantilever pier they did not appear until

September. '

Observations made on L. sitkana caged on a cement slab suspended at the 1 foot tide mark (U.S. tables) at the Friday Harbour Labora• tories pier from September 20 to October 12, 1969 indicated, that a fe• male L.sitkana can lay at least three egg masses in a three week period. 19

Desiccation of Juvenile Stages as a possible Factor Restricting L. sitkana's Distribution

Survival of juvenile L. sitkana in a location inhabited naturally by L. scutulata only was investigated at Lilly Point (Fig. 9). The rocky foreshore in this area is characterised by barnacle covered cobble and rocks not larger than 15 cm in diameter resting on sandy bottom. The low intertidal area is mostly sand interspersed with four barnacle covered concrete blocks (50 cm by 50 cm by 50 cm). I worked on the site of an abandoned fish cannery where an artificial substrate consisting of compressed tin can scraps and cobble is completely covered with barnacles and extends from the mid to the high intertidal region.

Numerous barnacle covered pilings (the remains of the cannery's pier) run in rows from the mid to high intertidal. Absence of shade, as well as good drainage tends to make the Lilly Point site a dry beach at low tides in sunny weather. Animals cannot find shelter under the cobble and rocks for these are resting on sand.

To determine the critical life history stage preventing _L. sitkana from living at Lilly Point, adults, newly hatched snails and egg masses were transplanted to the area. Egg masses recovered from experimental cages containing L. sitkana yielded information on the preferred sites for egg laying.

Materials and Methods a) Transplant Experiments

To determine whether adult _L. sitkana could live at Lilly Point, 20 five hundred young L. sitkana (ca. 5 mm in length) were released on the "compressed tin can rock" and on one piling stump in May of 1969.

On May 16, 1970 newly hatched L. sitkana were collected from the

Montague Harbor mud flat on Galiano Island (Fig. 9 ). The oyster shells and little neck clam shells on which the egg masses were attached were kept moist with sea water. The following day eight "cages" were pre• pared by pulling a square of fine plankton netting over the concave half of the little neck clam shells. Four of the cages contained 10 newly hatched snails each and 4 cages contained five older snails (l mm or longer). One of each type of cage was set up in the following location: on pilings at the 13 foot tidal level, in artificial tide pools (32 oz orange juice jars) at the 13 foot tidal level, on pilings at the 9 foot tidal level and in artificial tide pools at the

9 foot tidal level. The "tide pools" and cages were attached to the piling stumps using rubber bands cut from an inner tube. The number of surviving animals, salinity and temperature of the tide pools and air and water were recorded the next day.

L. sitkana egg masses were collected from False Bay, San Juan

Island (Fig. 10) on August 21, 1969 and taken to Lilly Point on

August 23. All egg masses were sectioned into two parts. Each half was placed into a plastic petri dish lid and fine plankton netting was wrapped around the dishes. The dishes were attached to the pilings at

the high tide levels and to the concrete blocks at the low tide levels

so that half of each egg mass was represented at each tidal level. The 21 number of hours the 5 foot and 12 foob tide level was exposed to direct mid-day sunshine was estimated from weather data and a tide table. The condition of the egg masses and the number of hatched snails were recorded on September 1 and September 7 (Table 4). b) Site preference for egg laying

,L. sitkana were retained in 16 cages located at the 2 foot tide level (U.S. tables) at Cantilevel pier beach. The cages consisted of cement stepping stones (l9 cm by 4 cm by 39 cm) sewn into plastic mesh

(vexar) bags (Fig. 18). The location of all the egg masses laid by the snails in the cages was noted and grouped into 7 position categories

(Table 5).

Results

After one year six L. sitkana were recovered on the tin can rock.

All the animals had grown to roughly 10 cm in length. The rest of the animals had either died or dispersed from the immediate area. One yellow egg mass located on the wave and sun sheltered side of a piling stump was found in May 1970. The egg mass, however, dried up before hatching. The fact that the transplanted L_. sitkana survived for a year, grew and even demonstrated reproductive potential would indicate that no major selective factor was operating during this period of time preventing adult L_. sitkana from living at Lilly Point. Lack of shelter coupled with the abrasive action of shifting sand in this area may prevent L. sitkana egg masses and newly hatched snails from devel• oping. 22

All the young L. sitkana caged for 26 hours at the mid tidal

level and those retained inside the high tide pool survived

(Table J>). However all the small snails (less than 1.0 mm) caged to the

high piling stump were dead. The cages and pool at the high tide level

(l3 ft) were calculated to be exposed to roughly 11 hours of direct

sunshine, those at the mid tide level (9 foot) for roughly 6 hours

during the duration of the experiment. It would seem that desiccation

and not the high temperatures per se killed the small snails at the

high tide level since the temperature in the high tide pool was 7°

higher than air temperature at the time of measurement (Table 3).

From weather and tide data for the period August 23 to September

1, the 5 foot tidal level was estimated to be exposed to 'a total of 6

hours to direct mid-day sun and the 12 foot level to at least 30

hours. A total of 161 young L_. sitkana hatched at the 5 foot level as

opposed to only 2 at the 11 foot level. All the egg masses at the high

tide level had dried out by September 7, (Table 4). This correlation

suggests that desiccation was responsible for egg mass mortality at

the high tide level but not at the lower.

Wo eggs were laid on the wave exposed side of the experimental

slabs and only one on top of the slab, whereas 29 egg masses were re•

covered from the wave sheltered side of the slab and 26 from the bottom

of the slabs (Table 5). Fourty seven egg masses were laid in cievices

of the cage seams which were on the wave sheltered side of the slabs.

It therefore appears as if L. sitkana prefer to lay their eggs in

wave sheltered and dark places. 23

Discussion

- Desiccation acting on egg masses and newly hatched individuals of

L. sitkana may be a critical factor preventing this species from living at Lilly Point and other dry beaches. It is conceivable,, that a perma• nent population of L. sitkana could be established at Lilly Point if tide pools or damp crevices were added. Egg masses hatched at low tide levels, but the abrasive action of shifting sand and silt, especially during storms seems to prevent any grazers from living there permanent• ly. The preference L_. sitkana shows in laying their egg masses in sheltered places appears to be an adaptive trait.

General Discussion

Thorson (l950) showed that a greater proportion of northern prosobranch species pass the veliger stage inside the egg capsules than do their southern counterparts. He suggests that this form of develop• ment may be an adaptation to the unpredictability of the phytoplankton food for the veligers in northern waters. This trend 'appears to hold for littorines. Both JJ. sitkana and the European L./littoralis have benthic egg masses and are considered northern species. These two species do not extend into warmer climates as do L_. scutulata. L. planaxis (imai, 1964) and the European L. neritoides with plankto- trophic development. Susceptibility of egg masses of L. sitkana to desiccation at low tide may be one reason why this species does not extend further south. Such susceptibility combined with a limited dis• persal potential restricts L. sitkana to beaches which offer protection 24 from desiccation. L. scutulata with its planktonic dispersal stage appears to be able to invade more habitats (Table 6). 25

LOCAL DISTRIBUTION PATTERNS OF LITTORINES

A. HORIZONTAL DISTRIBUTION

Introduction

Littorina scutulata and L. sitkana coexist on beaches near

Vancouver, B.C., on the islands in the Gulf of Georgia and on Vancouver

Island. Vertical distribution of these snails, unlike that of the

California littorines (Bock and Johnson, 1968) and the Atlantic littorines (Gowanloch and Hayes, 1926) does not fall into discrete bands along the intertidal exposure gradient. This absence of any obvious pattern in the vertical distribution of these two species was also noted by Urban (l962). In the San Juan Archepel^go and at Port

Renfrew Urban (op. cit.) noted, that L. scutulata was the more numer• ous and more widespread species. He attributed this to a greater ability of L. scutulata to tolerate environmental extremes such as low salinity and surf action, but did not test the idea experimentally.

The two species of littorines were found separated into two horizontal zones only on extremely exposed surf-swept beaches of

Vancouver Island such as Chesterman's Island near Tofino. There L. scutulata occupied the intertidal zone and L. sitkana was found only in extremely high splash pools. These pools receive ocean water only during the winter storm tides (Andreas Schroeder, personal communi• cation). In the summer the pools are isolated from the ocean and both water level and salt concentration probably fluctuate between dry and rainy periods. In the spring of 1970 the pools dropped one inch in Fig. 8b High splash pool on Edward King Island near Bamfield, containing numerous newly hatched L. sitkana. 9 Taken from Chart # 3001, Vancover Island Published by the Canadian Hydrographic Service Marine Science Branch, Department of Mines and Technical Surveys, Ottawa 6-vtN\ Reef- 28

Cove

N

X

GiUii't SAN DUAN Cove. UniVersi °f U^shWo, La bom furies o >nHlever Pier

ISLAND FRIMY

JDegcitnaii's Bay

PILE POINT BAY' Harsjista O Resort

SCALE MILES

Figure 10 Cattle Point 29 height from May 5 to May 6. Many snails were stranded in the salt crust at the edges of the pools, however they were not dead for they opened

their operculum and moved their foot when immersed in the pool. During

the week of July 1, 1970, Schroeder noted,that the water level was extremely low and that many snails were stranded.

Twenty-eight beaches were surveyed and presence of the two species

of littorines was correlated with type of substrate, degree of wave

exposure and salinity. A series of experiments was performed to deter• mine whether the distribution patterns could be explained on the basis

of differential adaptation by the two species to aspects of wave action, submergence,desiccation, water temperatures and salinity.

Survey of Beaches

Materials and Methods

Tidal levels were estimated by reference to biological zones or

else measured from the low water mark at the time of the survey. Popu•

lation densities were estimated using standard quadrats of 1 meter by

1 meter or 5 cm by 5 cm. Size frequency distributions of animals were

noted either by measuring animals with calipers to the nearest mm or

else by sieving animals through a series of soil test sieves (mesh

sizes 0.5 mm, 1 mm, 2 mm, 3 mm, 4 mm, 5 mm, 6 mm). Animals retained by

the 6 mm mesh diameter were called size 8 animals, those by the 5 mm

mesh size 7, those by the 4 mm mesh size 6, those by the 3 mm mesh size

5, those by the 2 mm mesh size 4, those by the 1 mm mesh size 3, those

by the 0.5 mm mesh size 2 and those passed through the 0.5 mm mesh

size 1. Salinity measurements were taken with a hydrometer and wave 30 exposure ranked subjectively according to the degree of water movement at the time of the survey.

Results

A number of generalizations regarding population densities, size frequency distributions as well as distribution patterns of the two species of littorines were determined from the survey.

1. L, sitkana in any one place are larger than L. scutulata.

Frequency distributions based on shell length do not always show differences in size since at any given length L. sitkana is bulkier and heavier than L. scutulata. For example dry body weight of a 5 mm

L_. scutulata is .0018 grams and a 5 mm L. sitkana .0030 grams (data courtesy of Menge). Greater growth rate of L. sitkana may in part explain this size difference (Fig. 26).

2. Populations of L_. scutulata at any one place are character• ised by a unimodal size frequency distribution and populations of L_. sitkana by a bimodal distribution (Figs. 11, 12 ). The presence of reproductive peaks in the spring and fall combined with more rapid growth may make season classes detectable. Evidence indicates that

L. scutulata larvae settle from at least late winter to early fall.

Such continual settlement combined with slow growth may obscure season classes.

3- L. scutulata live at higher densities than L. sitkana.

Evidence from a density-species interaction experiment to be discussed later indicates that JL. scutulata survive better at higher than at 31 I«*>

70

L. scutulata

5»

Ho-

01

e •H cs io - o n j 1 1_ 0) io i» •4i r

to

L. sitkana Xo

IO

l J- 5 H >5* 6 1 1 10 II 12. 13 J* If /fte l I?

Fig.ii size Frequency Distributions for L. sitkana and L. scutulata. • L. scutulata was collected at Lilly Point of January 27, 1968; L. sitkana was collected from Jekel's Lagoon October 12, 1969. 32

i6 r L. scutulata xo >

10 •

3 H r "> 1 <\ lo

r IOO CO H e •H 8o *

U_l o loh L. sitkana

*6 •

50

Ho

30

10

10

8 ^ io »

Fig. 12 Size Frequency Distributions of littorines from Montagaue Harbour Mud Flat July 30, 1969. 33 lower densities.

4. L. scutulata is more widespread than L. sitkana. occurring in

36 of the 40 sites, whereas L. sitkana occured in only 28 of them

(Table 6).

5. The ratio of L. scutulata to L. sitkana increases as exposure increases. For example only L_. scutulata were present on the heavily exposed transect on Chesterman's Island and on the Victoria Break• water (Table 6).

6. The size of littorines varies inversely with wave exposure.

At Chesterman's Beach exposed transect, the largest L. scutulata was

4 mm in length, at Deception Pass State Park relatively sheltered transect, they were 20 mm in length.

7. The highest densities of L. sitkana occur in sheltered damp beaches with many crevices. For example at Brockton Point there were more L. sitkana under boulders and seaweed than on the dry seawall

(Tables 6,7).

8. Vertical shores tend to have higher ratios of L. scutulata to L_. sitkana. than horizontal ones. (See Saxe Point horizontal versus vertical site, and pilings versus cobbles at the Anacortes Ferry landing, Table 6).

9. The lower the salinity ( surveying beaches from the Maritime

Museum to Jericho to Spanish Banks) the lower the ratio of L. sitkana to L. scutulata (Table 6). 34

Responses of L. sitkana and L. scutulata to Physical Factors

Materials and Methods

a) Continual Submergence

Jekel's Lagoon on San Juan Island (Fig. 11) is similar to the splash pools of the west coast of Vancouver Island in that it contains

L. sitkana but no L. scutulata and is relatively isolated from ocean water which flows into the lagopn only during high high tide. To determine whether L. scutulata could live submerged for an extended period, 50 marked animals were released into the lagoon in February

1970 and three hundred marked L. scutulata were added to the very high splash pools on Chesterman's Island on May 6 1970.

b) Wave Exposure

Extremely small L_. scutulata only are found on wave swept beaches such as Chesterman's Island near Tofino. To investigate the action of intense surf as possible factor acting selectively against L. sitkana and larger L_. scutulata the following series of laboratory and field tests were performed.

An equal number of similar sized animals (matched for biomass) were collected and painted with cellulose base paint. The animals were then wetted and allowed to attach to the rock or barnacle substrate of the beach, and then subjected to the wave action of the incoming tide.

After a trial period (from 6 hours to two days) the test site and ad• jacent areas were carefully searched and all the missing animals were assumed to have been dislodged by waves. Laboratory experiments using 35 cement slabs as a substrate and a running sea water jet to simulate wave force were performed to check field results.

c) Temperature

Since both species of littorines occur in Alaska, it was assumed, that differential adaptation to cold temperatures would not be a key- factor in explaining their local distribution patterns. Since L. sitkana is found in higher proportions than L. scutulata in high tide- pools and mud flats, this species might be expected to be more tolerant than L. scutulata to summer high temperatures. To test this, four lots of 20 animals (size 5) of each species were placed into four pyrex finger bowls (lO cm in diameter), containing sea water at

30°C, 30°C, 44°0, and 45°C. Animals were left in the water for 3 minutes after which they were tested for viability by putting them into room temperature sea water. The number of animals showing signs of life after 2 minutes and after 24 hours recovery time was noted.

Since all the animals exposed to the 30°C water were active after 2 minutes they were subjected to higher temperatures. The water in the finger bowls was heated gradually (from 30°C to 40°C in three hours) and continually stirred. Animals were removed from the warm water periodically and tested for viability. One dish was raised to 42°C, another to 45°C. Animals were then put into room temperature water and the number showing signs of life after 2 minutes and after 19 hours recovery time was noted.

d) Desiccation

To determine which species of littorines was more tolerant to 36 desiccation size 4 animals of both species were collected from Lonesome

Cove, San Juan Island (Pig.10 ) in January 1970. These animals were allowed to dry on paper at room temperature and after 3 days desicca•

tion 20 animals of each species were put into sea water and the number of animals showing signs of life after 24 hours recovery time noted.

Animals ranging in size from 4 mm to 11 mm in length were also collected from Brockton Point in April 1971 and exposed to 4 days desiccation at room temperature,

e) Salinity Tolerance

A survey of the beaches down a salinity gradient from Second

Beach to Maritime Museum to Jericho Beach to Spanish Banks (Fig. 9) indicated that L. sitkana disappeared between Maritime Museum and

Jericho Beach. A few L. scutulata occur as far as Spanish Banks. This distribution pattern may indicate that L. sitkana is less tolerant to

low salinities than L. scutulata. The following tests were performed

to determine the tolerance of littorines to salinity extremes.

Specimens of both species from Brockton Point were matched for

equal biomass. Ten animals of each species were placed into each of

14 finger bowls. Animals in-fen of the bowls were exposed to distilled water at room temperature and animals in the other 4 finger bowls to

sea water at room temperature. At intervals the water in all the bowls was drained and replaced with fresh sea water. All animals which

showed activity after 5 minutes recovery time were placed into clean

distilled water. Those animals which did not show signs of life were given another hour to recover in the sea water after which they were considered dead. 37

f) Crevices

In the Marvista Resort quadrat (Table 7), significantly more L. sitkana than L. scutulata were found in crevices. To determine whether

L. sitkana has a superior ability to find these crevices fifty animals of both species were introduced into a dishpan containing sea water and 15 rocks (not larger than 5 cm in diameter) of different roughness*.

Five of the rocks were extremely smooth, 5 were extremely jagged with many crevices and 5 were of medium roughness. After 24 hours the number of snails of each species on each type of rock was noted.

Results

a) Submergence

Most of the fifty marked L. scutulata which were released into

Jekel's Lagoon in February were recovered two months later. All of these animals had grown, indicating that continual submergence did not harm them. The marked L_. scutulata released in the high splash pools on Chesterman's Island on May 6, 1970 were still alive during the first week in July (Schroeder, personal communication).

It would seem that the physical factors measured do not prevent adult L. scutulata from living in high splash pools and lagoons. The absence of this species from the splash pools and lagoons could perhaps be explained by the inability of the planktonic larvae to reach these pools. The larvae are in the plankton from at least

February to early fall (see p. 16 ) and if these pools are flushed only during winter storm tides, then the larvae may not have a change 38 to enter.

Struhsaker and Costlow (1968),working with Littorina picta larvae in the laboratory,state:

"Removal of water from the covered rearing bowls after approximately 3 weeks of development appears to be one of the major stimulii to settlement (of L. pieta larvae). Preliminary results show that larvae settle earlier when they are damp, but not submerged at this period of development."

If such drainage action is also necessary for the settlement of L. scutulata larvae, in the field, then one would not expect to find newly settled individuals in tide pools or lagoons at any level.

b) Wave Exposure

Both field and laboratory data indicate that L. scutulata are less likely to be dislodged by waves than L. sitkana (Tables 8 and 9).

L. sitkana with its round shape and many grooves perhaps offers more resistance to wave action than the more streamlined JL. scutulata.

Large L,. sitkana were more easily dislodged by wave impact than the smaller ones (Table 9b). However large L_. scutulata appeared as resistant to such forces as smaller ones (Table 8, see Port Renfrew .

Nov. 21 to 22).

The effect of wave force on littorines would depend on the slope of the beach and the number of crevices. L_. sitkana can live on moderately exposed horizontal beaches with crevices, but not on vertical rock walls experiencing the same type of wave action.

c) Desiccation

Both species of littorines were equally resistant to desiccation 39

(Table 10). After having been exposed to air for 4 days most animals of both species were extremely active when returned to sea water. It is unlikely that adult animals are ever exposed to such desiccation stress in the field. As mentioned earlier the susceptibility of the egg masses and newly hatched snails to desiccation tends to restrict

L. sitkana to beaches which offer some protection.

d) Temperature

li' scutulata was significantly more resistant to high water temperatures than L. sitkana (Table 11). Temperature sensitivity therefore does not appear to be the factor preventing L. scutulata from living in high tide pools and lagoons.

e) Salinity Tolerance

Littorina sitkana survived exposure to low and high salinities significantly better than L. scutulata (Table 12, Fig. 13b). The time at which 50 percent of the Brockton Point animals died due to exposure to distilled water was 63 hours for L. scutulata and 89 hours for L. sitkana. The presence of L. scutulata closer to the mouth of the /

Fraser River is not related to this species' greater tolerance to low salinity. Planktonic dispersal would allow a few L. scutulata to periodically invade marginal habitats such as Spanish Banks. However these individuals may not have the ability to reproduce and may be killed during periods of low salinity.

f) Crevices

Significantly more L. sitkana were found on the extremely jagged rocks than L. scutulata (Fig. 13a). This may suggest that L. sitkana 40

smooth medium rough

Fig. 13a Number of animals of both species found on rocks of different roughness. Control Animals

K ^t«^^. — v Jfc • >^ ^\ N=40

0N ^ \ \ \ \ \ \

L. scutulata \ L. sitkana

N=100 \ ^ N=10°

%o 30 kO sro fco TO 8© <*o \°»

Exposure Time to Distilled Water at Room Temperature (Hours).

Figure 13 b Tolerance of Littorines to Distilled Water 42 is better able to find shelter than L_, scutulata.

Discussion

Littorina scutulata is the more widespread species, presumably because it has planktonic dispersal. Even marginal or unstable habitats probably receive recruitment every year. This would not be the case for

L. sitkana with its le ci tH orophic development. One bad season in a year (for example low salinity due to river run off such as at Spanish

Banks) could have the effect of eliminating a species with such devel• opment.

It would seem that L. sitkana needs crevices for two reasons:

1) in providing shelter from direct wave force (especially on

exposed beaches);

2) in providing a damp place for egg mass development. L_.

scutulata displays no such dependence on crevices.

L. sitkana may attempt to compensate for this disadvantage by having a greater ability to find crevices.

High temperatures, continual submergence, and desiccation are not factors preventing L. scutulata from living in the high splash pools and lagoons since this species is equally or more tolerant than L. sitkana. A lower tolerance to salinity extremes could prevent L. scutulata from living in the high splash pools at some time of the year. L_. scutulata survived in these pools from May 6 to at least July

1, 1970. High salinities encountered in the late summer due to desic• cation or low salinities encountered during the winter rains could 43

select against L. scutulata. Planktonic larvae probably prevent L.

scutulata from living in these pools since these may be absent during

the winter when the pools have communication with the ocean.

Low salinities and a lack of planktonic recruits cannot explain

the absence of L. scutulata in Jeckel's Lagoon, for it receives fresh ocean water every high high tide. In this case it appears as if the L.

scutulata veliger larvae do not settle when submerged.

B. VERTICAL DISTRIBUTION

Introduction

The most striking observation one makes when visiting a rocky

sea shore is that the various species of organisms are distributed in

horizontal bands between the low and the high tide mark. Thus an eel

grass zone, a kelp zone, a barnacle zone, a littorine zone and a black

lichen zone can easily be distinguished. The higher an organism is on

the shore, the greater will be the time it. is exposed to air. Environ•

mental stresses such as desiccation, extreme temperatures and extreme

salinities in the case of tide pools become increasingly important as

one moves up the intertidal gradient. As a rule, the upper distribu•

tion of intertidal organisms is limited by their tolerance to physical

factors such as desiccation and high temperatures and their lower

distribution is limited by biological factors such as predation or

interspecific competition (Connell, 1961). 44

I. The Upper Distribution of Littorines

Littorines become extremely rare in the splash zone. Desiccation must ultimately limit their upper distribution however limitation of available food also seems to become increasingly important up the shore. Littorines migrate up the shore in early fall probably in response to an easing of the desiccation stress encountered in the summer as well as in response to an increased abundance of diatoms

(which in turn also is a result of decrease in desiccation) at high tidal levels.

Two pilot experiments were undertaken in an attempt to evaluate the relative importance of food limitation and desiccation in setting the upper limit -to,littorine distribution.

Materials and| Methods

a) Survival of L. scutulata in the splash zone

The first test was to determine whether littorines can survive higher on the shore than they are normally found. On February 26, 1970, three cages containing 10 L. scutulata each were attached to the pilings at Lilly Point, Point Roberts (Fig. 9) from to l-j feet above the highest observed littorine. After 38 days the cages were examined for live snails.

b) Response of Littorines to Food in the Upper Intertidal Region

To determine the response of littorines to the presence of food at high tide levels, two slabs were set up on a shelf just below the black lichen zone at Cantilever pier San Juan Island (Fig. 10). One 45 of the slabs was clean and the other was surfaced by a mat of diatoms.

Ten littorines of each species were allowed to attach on each of the

slabs. On the next low tide (the high tide just washed the slabs), the number of littorines on each slab was noted.

Results

a) Survival of L. scutulata in the splash zone

After being caged for 38 days in the black and yellow lichen zone,

18 out of 23 L. scutulata were still alive. This indicates that desic•

cation at these levels does not kill the animals. However limited algal

abundance combined with a limited time available for feeding would make

the high intertidal and splash zone an unfavorable habitat.

b) Response of Littorines to Food in the Upper Intertidal Region

Two L. scutulata and 3 out of 10 L. sitkana remained on the diatom

covered slab in the high intertidal zone at Cantilever pier after one

high tide. No animals remained on the claan slab. Unfortunately, this*

experiment was of short duration and desiccation was not a factor in

that the atmosphere was. noist. This test indicated only that littorines

may tend to stay in places where there is food. (Responses of

littorines to food will be discussed in the next section).

II. Limits to the Lower Distribution of Littorines

When taking transects at Saxe Point, Victoria and at Deception

Pass State Park it 'was observed that limpets of the genus Acmaea

(especially Acmaea scutum) become more abundant and littorines become

less abundant at the bottom of the shore (Fig. 14s). Similarly, at 46

Figure 14a 47

Cantilever pier around the 0 ft. tidal level (U.S. tables) smooth bolders resting in the gravel, appeared to be without algae but bore many Acmaea scutum and no littorines. Fencing off an area and removing the limpets resulted in abundant algal growth within a month. It was conceivable that the grazing action of Acmaea scutum could make these bolders unattractive to littorines. To test this hypothesis a littorine-

Acmaea interaction experiment was set up.

The drop in littorine abundance at lower tidal levels is also correlated with an increase in intertidal predators such as the star• fish Leptasterias hexactis and Pisater ochraceus and the snail Searlesia dira (M. Lloyd, B. Menge, personal communication). All these species have been observed to feed on littorines in the field. The calculated electivity coefficients of the prey species for the predators

Leptasterias hexactis (Table 14, courtesy of Menge) and Searlesia dira

(Table 13, courtesy of M. Lloyd), indicate that littorines are always taken at greater proportions than they are found in the environment.

Menge (personal communication) is of the opinion that littorines are a convenient food source for Leptasterias in that this starfish can be digesting a littorine', Cand at the same time be foraging for more food.

Perhaps the increased abundance of Acmaea and decreased abundance of littorines at lower tidal levels can in part be explained by dif-- . ferential predation pressure. Menge (1970) suggests, that Leptasterias can catch littorines more easily than Acmaea spp. since littorines cannot clamp down and have relatively weak escape responses when com• pared to Acmaea spp. (Table 15). 48

To determine whether predators feed selectively on littorines as opposed to limpets, experimental cages containing Acamea paradigitalis,

L. scutulata and Leptasterias hexactis were set up in the mid and low intertidal.

Materials and Methods

a) Acmaea scutum, L. sitkana Interaction

A cement slab (50 cm by 50 cm by 4 cm) was suspended from the

Friday Harbor Laboratories pier on August 16, 1969 between the 0 and 2 foot tide mark (U.S. tables). Six stainless steel cages (Fig. 15)

(10 cm by 20 cm) were screwed to the slab using plastic washers and stainless steel screws. Limpets (Acmaea scutum) between 9-5 and 16 mm in length and L. sitkana between 7 and 12 mm in length were matched for biomass by using information contained in Fig. 14b. The limpets were marked by tagging their shell with self-sticking wire markers (Brady

Micro Markers, W.H. Brady Co. 727 W.Glendale Avenue, Milwaukee 9,

Wis.) and adding a drop of decophane technical cement (Rona Pearl

Corporation, Bayonne, New Jersey). Twenty animals, either all A.scutum, all L. sitkana or 10 of each species were then put into the cages. On

September 20, 1969 the growth increments of the animals were measured.

b) Effect of Leptasterias hexactis predation on A.- paradigitalis

and L_. scutulata

On July 16, 1969 eight mesh bags, sewn from plastic screening, with a medium sized rock placed in each, together with 10 small L.- scutulata. 10 small A. paradigitalis and two Leptasterias hexactis Length ( mm ) of Littorina sitkana Fig. 15 Cages used for Acmaea-Littorine interaction experiment. Cement slab and stainless steel cages were suspended from the Friday Harbour Marine Laboratories pier. 51 were closed with rubber bands. Four of the bags were placed at the

-0.5 foot tidal level and four at the +2 foot tidal level (U.S. tables), at Cantilever Pier. After three days the number of live animals and empty shells were noted.

Results

a) Acmaea scutum. L. sitkana interaction

There was no significant difference in the growth rate of A. scutum grown alone and grown with L. sitkana, nor was there a signifi• cant difference in the growth rate of L. sitkana grown by themselves and with A. scutum (Table 17). However the direction of the growth rates may indicate that in mixed species cages A. scutum was growing at the expense of L_. sitkana. More data is needed to clarify this point.

b) Effect of Leptasterias hexactis predation on Acmaea

paradigitalis and L. scutulata.

Under the experimental conditions L_. hexactis appeared to feed on

L_. scutulata and A. paradigitalis at the same rates (Table 18). How• ever predation pressure was significantly more intense at the low tide level than the mid tide level (Table 18J.

Summary

It would seem that the upper distribution of littorines is set by desiccation acting on the animals directly and/or indirectly in decreasing their food abundance. In either case the intertidal habitat becomes increasingly less favorable up the shore. 52

If it is true that the drainage action encountered in the inter•

tidal is a major stimulus to the settlement of littorine larvae then

one would not expect L. scutulata to settle subtidally. The maximum

density of this species then should be found in the intertidal.

The lower distribution of littorines is correlated with an in•

creasing abundance of predators such as Pisaster ochraceus,

.,L;e_ptasterias hexactis and Searlesia dira. All these species have been

observed to feed on littorines in the field, and evidence is presented

that the intensity of predation of L. hexactis on littorines and

limpets increases down the shore.

The abundance of limpets such as Acmaea scutum increases down

the shore as the abundance of littorines decreases. Positive evidence

was not found that competitors limit the lower distribution of

littorines. 53

THE ABUNDANCE OF LITTORINES

A. Behavioural Responses

Introduction

The abundance of sessile organisms such as barnacles and dandelions is a function of the number of juveniles (barnacle larvae or dandelion seed) reaching a particular habitat and their subsequent survival success. In the case of mobile animals abundance is also affected by their ability to migrate from unfavorable to favorable habitats.

The highest densities of adult and recently settled L. scutulata were found on barnacle covered beaches such as those on Chesterman's

Island and at Lilly Point. Newly settled L. scutulata are strongly associated with barnacle interstices which probably act to trap the larvae. None were ever found on smooth rock surfaces.

The number of new recruits entering a L_. sitkana population is directly dependent on the number of egg masses the females in that area lay, their hatching success and subsequent survival.

Such mortality factors as food scarcity, desiccation, predation, surf action etc. act to thin out littorine populations. Particular responses by the animals may mitigate the effect of such factors.

Migration from an unfavorable habitat (e.g. one without food or shelter) to a more favorable habitat would be such a response. In order to study such behaviour a number of experiments was performed to determine the response of littorines to food abundance, shelter and crowding. 54

Materials and Methods

a) Behavioural Response to Crowding

To investigate the possibility of dispersal behaviour in response

to overcrowding, a number of L. sitkana (l2 or 36) were allowed to attach to the center of four concrete paving slabs (l9.5 cm by 19.5 cm).

These slabs were submerged in a water table in the laboratory and the number of animals reaching the edge of the slabs after 5 minute trial periods was noted.

b) Behavioural Response to Food Levels

1. Field Observation

On August 11, 1969 an observation was made while diving at high tide over the cement slabs used in a density-species interaction

experiment (to be discussed later). Many littorines had moved on to cement slabs covered with diatoms and very few had moved on to clean slabs although equal access to both types was possible. The number of littorines of each species was counted on 6 clean slabs and

6 diatom covered slabs.

2. Laboratory Study

During September an experiment was set up to investigate the

responses of littorines to a clean and a diatom covered slab. The

slabs were put into a tank with running sea water. Ten L. scutulata

and 10 L. sitkana were placed into the center of each slab and the

times at which the animals reached the edges were noted. Those animals

reaching the edge were replaced to the center of the slab as soon as

their tentacles protruded over the edge. 55"

c) Response of L. sitkana to the presence of food

To investigate the capacity of L. sitkana to use sensory stimulii in finding food, a glass plate was used to divide a water table into two sections (Fig. 16). A cement slab (l9.5 cm by 39 cm by 4 cm), covered with diatoms was introduced into one section and a clean slab into the other section. The water level was 8 cm deep (enough to cover the slabs and the snails). Fresh sea water was run over each slab to' the drain at the other end. Fifty L. sitkana were distributed evenly between the edges of the glass plate and the drain so that 25 snails were in each section. The whole water table was covered with black plastic so that light gradients would not play a role in the orienta• tion.

d) Behavioural Response to Crevices

To test the idea that crevices are more attractive to littorines than smooth surfaces four oyster shells (length 12 cm) were placed in a finger bowl (diameter 10 cm). Two of the oyster shells were com• pletely covered with dead barnacle shells (therefore possessing many crevices) and two were smooth and devoid of barnacles. Water was poured into the bowls and 5 littorines of each species were placed on each of the shells. After 10 minutes, the number of littorines remain• ing on each shell was noted. The same tests was run 3 more times changing the order of the oyster shells and using new animals.

Results

a) Behavioural Response to Density

The results of Table 19 indicate that dispersal of L. sitkana Direction of

running sea water I \\ No

Food Food

—Glass plate

Position of 50 Littorina sitkana

0"~ 'Drain

Fig.JJ-6 Modification of a water table to test the hypothesis that littorines can detect the presence of food at a distance. Scale: 1mm = 1cm 57 was significantly greater at low densities than at high density. At high density the animals tended to attach and move over each other more so than at low density.

b) Behavioural Response to Food Levels

1. Field Observation

Littorina sitkana appears to be sensitive to food levels in that they move onto diatom covered slabs at a significantly greater rate than onto clean slabs. L_. scutulata did not show this trend

(Table 20).

2. Laboratory Experiment

After one hour significantly more animals had reached the edges of the clean slab than the diatom covered one. Both species spend more time on the diatom covered slab than the clean ones

(Table 21).

c) Response of L. sitkana to the presence of Food

After 14 hours the positions of the snails in the food choice tank were noted (Table 22). There was no difference between the number of snails located in each section, therefore it would seem, that L. sitkana does not use water born chemical stimulii to detect food.

d) Behavioural Response to Crevices

The results in Table 23 indicate that oyster shells with barnacles are more attractive to both littorine species than are smooth oyster shells without barnacles. 58

Discussion

When L. sitkana were crowded they did not disperse because crowded animals exhibited a tendency to attach to one another.

Animals of both species reacted to the presence of food in that they dispersed less on diatom covered slabs than on clean slabs..

L. sitkana does not appear to use chemical stimulii in finding food. Animals may move randomly with respect to food and when finding it stay there.

Both species of littorines react to crevices in that they dis• perse less when put on barnacle covered oyster shells than on smooth ones. Crevices would protect littorines from predators, direct wave force and provide a damper subhabitat thus decreasing desiccation and possibly increasing algal productivity.

B. Density-Species Interaction Experiment

Introduction

If density dependent factors operate to regulate littorine densities at a certain level, greater mortality rates and perhaps lower growth and reproductive rates would be expected at densities higher than normally found whereas the reverse might be true at lower densities. To test this^an experimental investigation into the effects of density on survival, growth and reproduction of littorines was ini• tiated. This investigation also helped to determine if food limitation affected littorines in the course of a year and if competitive 59 relationships existed between the two species.

Materials and Methods

To manipulate both densities and specific composition of littorine populations over a period of time, it was necessary to retain the animals in cages. The cages used were plastic mesh cylinders constructed of vexar (Crown Zellerbach Co.) height 45 cm., circumference 100 cm and bottom diameter 33 cm sewn together with 20 lb tested nylon monofilament. The cages were weighted down by concrete stepping stones (4 cm by 19.5 cm by 39 cm) which also provided grazing surfaces for the littorines (Fig. 17). Twenty-six stepping stones were placed at the 2 to 3 food tidal level (U.S. tables) near Canti• lever Pier close to the Friday Harbor Laboratories.. This site was chosen because it offered medium wave exposure and because both species of littorines were abundant. The slabs were left on the shore for 5 days prior to the animals being placed into the cages so that they would weather naturally and so that algae could settle on them.

Animals of both species were collected at Cantilever Pier and on Gull Reef (north of San Juan Island) on June 15, 1969. A series of screens was used to match animals for size in all the treatments.

Measurements indicated that sieving was a convenient method of sepa•

rating animals of similar biomass.

The lips of all experimental animals were marked with tech pen

paint (Mark-Tex Corp. 161 Coolide Ave., Englewood, N.J.) so that new 6o

Fig.17 Cement stepping stones removed from vexar cages used in the density-species interation experiment. Note the diatom matts on the right bottom slabs. Fig. 18. Arrangement of cages at Cantilever pier beach. 62

growth could be measured by lip increment (Fig. 23). Either 20, 40,

or 80 animals (all L. sitkana, all L_. scutulata or half of each

species) were put into the 24 cages on June 18, 1969. Twice as many mixed species cages as single species cages were used so that an

equal number of animals would be subjected to mixed and single species

treatments. All mixed species treatments were replicated four times

and single species twice.

All experimental animals were moistened so that they could

attach to the slabs before the cages were sewn up. One cage was used

as a control giving an indication of the quantity of algae produced

in the absence of grazers . The associated slab was turned over at

each checking .time giving a rough estimate of algal productivity

during the period of littorine growth. Another slab which had broken

in two served as an uncaged control giving an indication of the

standing crop of algae under a natural density of grazers, which

corresponded roughly to the medium density treatments.

At each inspection measurements of lip increment and final

length of all snails were obtained. The total number of individuals

per cage was recorded together with the number of empty and broken

shells. Surviving snails were repainted with a contrasting colour

at each inspection and dead or missing animals replaced with new

ones every two months to maintain original densities.

A chlorophyll extraction method was used to estimate the

standing crop of diatoms (Castenholz, 196l). Two standard samples

were taken for each slab. Each sample consisted of scraping an area 63

(5 cm by 1.3 cm) five cm from the edge of the slab five times. Algal scrapings and rock chips were placed in screw-cap vials (18 mm by

100 mm) containing 20 cc of absolute ethanol. The samples were allowed to extract at room temperature for 48 hours. The relative abundance of algae was estimated by determining the absorption of the chlorophyll extract in a Bechman D.U. spectrophotometer at slit width 0.3 and wave length of 665 m,A. A visual estimate of standing crop of algae was made concurrently with the chlorophyll estimate of standing crop by scoring each slab on a numerical scale.

Heavy mat of algae (diatoms mostly) 5

Medium thick mat of diatoms 4

Thin mat of diatoms 3

Brown film 2

Faint Brown film 1

Clear Surface 0

In July 1969. all the diatoms were found to belong to the genus

Fragellaria. The relationships between optical density at 665 m-/t6 versus wet weight of Fr.agellaria and between wet and dry weight of

Fragellaria were obtained by use of regression analysis (Fig. 19). '

From December 1969 to June 1970, visual estimates of the standing crop of algae were used since a high correlation (R=0.88 p <.00l) was found to exist between subjective and objective measurements

(Fig. 20).

An additional piece of information was obtained from this experiment when L. sitkana commenced laying egg masses in the cages. Figure 19.

OPTICAL DENSITY vs. WET WEIGHT aw* J?RV WEIGHT vs. WET WEIGHT of Frajellaria

<0 . ,

a. so O-iO o.xo O.HO

0.10

Wet Wei^Kb o£ FrageUaria sp.. in grams

* Opl-ical3)ensiof Standard algal Scraping CScm v I.Scm) exhraclfed in 20 m\ of absotare e-ftanol. 65

Pig. 20. The relationship of subjective estimate of standing crop of algae ( Numerical estimate squared ) and objective estimate ( optical density of chlorophyll extract at 665 m/t).

The correlation coefficient was 0.880 NUMERICAL ESTIMATE OF STANDING CROP SQUARED 67

The position of the egg masses and their colour were noted. Colour of the egg masses is indicative of the stage of development of the embryos.- Egg masses were not disturbed, but unfortunately, not one young snail was recovered, presumably because they were washed off the slabs and through the coarse mesh.

FOOD ABUNDANCE

The standing crop of algae in an area depends on the settling rate and growth of the algae minus the grazing intensity minus physi• cal removal by such factors as wave wash. The algal cover in the cages was monitored for a year using the subjective numerical system described previously.

Castenholz (l96l) found that in the colder seasons carpets of dark brown diatoms cover the rocky intertidal of the southern Oregon coast. In the summer these rocky surfaces were almost devoid of dia• toms except for patches correlated with low grazer abundance. The same phenomenon seemed to occur at Cantilever Pier.

In the winter (September, October and December) food was abun• dant in all the density treatments (Fig. 2l). This was probably due to a high productivity of the diatoms combined with low grazing rates.

In February and April it appears as if the grazing activity of the littorines in the high density cages was beginning to reduce the standing crop of algae. In the summer (July, August and June) three distinct food levels were maintained by the three snail densities. A great amount of food was present in the low density cages and the 68

Figure 21 Numerical estimate of standing crop of agae at three snail densities as a function of time. One standard error is plotted on each side of the mean.

solid line low density dashed line medium density dotted line high density 69

A 70 70

Fig. 22 Numerical estimate of diatom abundance on control slabs. Uncaged controls ( mean and range ) give an indication of standing crop of diatoms under natural conditions; the caged control slab was turned over each checking time thus giving a rough estimate of productivity.

solid line Uncaged controls (standing crop of diatoms)

dashed line Caged control ( productivity of diatoms )

73

• high density cages were devoid of food with the middle density cages supporting a greater amount of food than the high density cages (Fig.2l).

High grazing intensities may have prevented macroscopic algae from establishing themselves in the high density cages (twice natural density). Ulva appeared in the medium and low density cages by Septem• ber and Enteromorpha and Coilodesme were found in October (Tables 35,

36).

The abundance of diatoms on the two uncaged control slabs was within the range of that of the experimental cages with a low standing crop in the summer and high standing crop in the winter (Fig. 22). The rough estimates of diatom productivity ranged from scale 2 to 4 with no seasonal trend (Fig. 22). This observation may suggest that the low observed standing crop of diatoms in the summer is due to grazing pressures.

GROWTH RATES

Methods

For the purposes of analysis it was necessary to convert growth data representing lip increment and final length to original length and length increment. Lip and length increment are directly proportional since the angle subtended by the whorl at the axis of the animals remains constant regardless of snail size (Fig. 23). Regression of lip increment on length increment was obtained for both L. sitkana and L. scutulata (Figs. 24, 25). These relationships were then used to convert all lip increment data to length increment. The length increments were 73

Fig. 23 Diagram illustrating the relationship between lip increment and length increment and between original length and final length of L. scutulata.

75

Fig. 24. Relationship between length and lip increment of L. sitkana.

GRAPH NO- 1 L-siVKana Y = -0-5011E 00 + 0-4055E 01X N = 42G

THE PROBABILITY OF THE SLOPE BEING ZERO IS 0.0000

20 4-

XX XX X>K )K X X-)SKX >K. XX .16 X XX XXM X XX X/ XX XX3KXX/ X XX X X X/ XSK>iK>B80p6o< X X X w X o X >B8K X,>&B3KX X H h3

S

0 0 3 4 LENGTH INCREMENT (mm) 76

Fig. 25. Relationship between.length and lip increment of L. scutulata.

GRAPH NO" 2 -L.scu>ulal& Y = 0-2791E.00 + 0°3137E OIX N = 395

THE PROBABILITY OF THE SLOPE BEING ZERO . IS 0.000

. 20 +

LENGTH INCREMENT (am) 77 then subtracted from the final length of animals to yield the original length.

Growth indices were calculated for all species and density treatments by dividing the adjusted mean growth increment by the number of days growth. These indices were then plotted against time (Fig. 26 to 29).

Results and Discussion

Growth rates in L. sitkana are much greater than for In scutulata at all times of the year and at all densities. This trend holds when length increment is converted to increment in biomass. For example, in low single species cages from July to August L_. sitkana (biomass .01225 grams) increased 82$ in biomass whereas the same dry weight L. scutulata increased jfo in biomass (Table 25). Since as mentioned previously the measured effect of the two species on the diatom standing crop was similar it would indicate that Li. sitkana is more efficient than L. scutulata in converting food to biomass under the conditions of this experiment.

Maximum growth for both species occured in the summer months

(April to September) and a minimum growth occured in the winter (October through February). Low growth was associated with low temperatures? however food was actually more abundant during the winter than the summer months. Activity was much less in the winter than in the summer.

In the summer most of the animals were found crawling on the slabs and on the vexar, whereas in the winter most of them were found aggregated under or on the sides of the slabs. This behaviour is doubtless 78

Figures 26,27, 28, 29.

Growth indices ( length increment x 1000 mm/ number of days ) under density and species treatments as a function of time.

Legend

Density comparisons

Low density

Medium density

High density

Species Comparisons

Single species

Mixed species

Stars between two adjacent data points indicate the

level of significance at wich the null hypothesis of

no difference between the values has been rejected.

Absence of a star indicates no significant difference.

.05

.01

.001 79

•j^H Aug. SepV- Oc-V- 2>ec. Feb. April 3u.ne-

82 83 adaptive serving to offer the snails protection from storms and winds.

In the summer (July, August, September and June) L. sitkana grew at different rates under the three density treatments. The high density, low food availability treatments produced the lowest growth rate and the low density treatment having high food availability showed the highest growth rate (Fig. 26). L. scutulata responded similarly, however, in this case, there was no difference between medium and low densities (Fig. 26). This would indicate that L. scutulata grew at its physiological maximum at medium densities and that increasing food- abundance did not result in a growth response.

Although algae could not be detected in the high density cages in July, August and June, the animals contained therein continued to grow over this period indicating some nutrient input. This suggests, that the animals were removing the algae at ;he rate of production.

Since L. scutulata shrank in size during July (Fig. 26) it is apparent that erosion must have been responsible. This could conceivably be associated with the fact that this species just recovered from re• production.

The differences in growth rates among the three densities for

Jj. sitkana in October, December, and February are apparently not direct-; ly related to food abundance, as during this time food was equally abundant under all densities. An explanation of these differences could either be due to individual interaction effects, whereby increased competition for sheltered places as density increased could result in 84 lower observed growth rates, or perhaps as a consequence of a delayed response to differences in food availability over the previous summer..

There was no difference in growth rates between single and mixed species treatments for either species at low densities (Fig. 27)*-''j«-.. sitkana survived better with animals of their own species than with L_. scutulata at medium densities in the fall (September, October, December) and poorer in the spring (April and June) (Table 28). There was no difference in the food availability between mixed and single species treatments. However L_. sitkana experienced a reproductive peak in the fall and the presence of L. scutulata may in someway have interfered • with the reproductive behaviour and food foraging ability of L. sitkana at this time. In the spring the situation was reversed, indicating that intraspecific effects were more important than interspecific effects- at this time. A similar effect was observed in September for high density L_. scutulata when animals in mixed species treatments grew better than in single species treatments (Fig. 29).

L. SITKANA NATALITY "\ ^.

In both the fall and in the spring proportionately fewer egg masses were laid by L. sitkana in single species treatments at high densities than at lower densities (Figs. 30, 32a and Table 26). This densitiy dependent effect was more pronounced in spring than in the fall in that numerically fewer egg masses were laid in high density cages than in low density cages (Fig. 32a).

In the fall such a density dependent response could have been 85 mediated through food limitation of the adult snails in the high

density cages during the summer, since at low densities animals' had more food and also grew more (Figs. 21, 26).

It is unlikely that direct availability of food resulted in the

density dependent natality in the spring because even in the high

density cages there was abundant food in the winter and early spring

(Fig. 2l). Since, as already mentioned, littorines tend to aggregate

in sheltered sites which are also prefered for egg deposition, it is

possible, that increased interference with egg laying occured as the

densities increased.

The number of egg masses produced at 20 and 40 L. sitkana per

cage (uncorrected for equal density of animals per cage) were signifi•

cantly greater for single species than for mixed species cages in the

fall and spring (Figs. 30, 32a, Table 27). This would indicate that the

presence of L. scutulata has an effect on the natality of L. sitkana.

However such a comparison does not take account of total snail densi•

ties per cage. In Figures 31 and 32b natality rates for pure L. sitkana

populations have been halved to estimate production as influenced by

an equal number of animals of the same species. Thus in a cage con•

taining a total of 20 L_. sitkana producing 50 egg masses, yield the

estimated natality of 25 egg masses for 10 L. sitkana in competition

with an equal number of their own kind.

A greater number of egg masses were laid per individual L.

sitkana in single species cages than in mixed species cages in the fall 86

Fall Natality of L. sitkana

40 r

O Single species

X Mixed species

ON

10 20 40 80

Number of L. sitkana per cage (uncorrected for equal density)

Fig. 30 The number of egg masses produced from September 17 to December 9 1969 under the three density treatments. Note that the L. sitkana densities for mixed species cages are 10. 20, and 40 L. sitkana per cage. 87

Fall Natality Of L. sitkana

A

Number of L, sitkana per Cage (corrected for equal density)

Fig. 31. The Number of Egg Masses Produced from September 17 to December 9 1969 as a Function of Density. Note that the values in the single species cages were divided in half to correct for an equal number of animals and an equal number of L. sitkana per cage. The number of egg masses produced was significantly greater in the two lower density treatments in single than in mixed species cages.

Single Species

Mixed Species 88

Spring Natality of L. sitkana

0 N 0»

0.10

3

2 -lew-

v> Nunber of L. sitkana per Cage (uncorrected for 10 equal density)

in

L _Q 10 i •i 2

|2

Number of L. sitkana per Cage (corrected for equal density)

Fig. 32 The Number of Egg Masses Produced by L. sitkana by April 17, 1970 as a Function of Density.

Single species Mixed species 89

(Tables27 ,28, Fig. 3l). This discrepancy was greatest at low and medium densities (lO and 20 animals per cage). This could be related either to the fact that there were only half the number of L. sitkana in each mixed species cage and individual animals may have had a greater difficulty in finding a mate at lower densities or to direct inter• specific interference. At 40 animals per cage there was no significant difference between production of egg masses in mixed and single species cages. The reason for this seems to be associated with generally reduced egg mass production in the high density pure _L. sitkana cages.

If interspecific interference rather than mate finding ability is responsible for depressed natality at lower densities, then at high densities intraspecific interference would appear to equal interspecific interference.

In the spring the same trend was observed. Both at 20 and at 40

L. sitkana per cage (corrected for equal density of animals) signifi• cantly fewer egg masses were produced per individual .L. sitkana in mixed species cages than in single species cages (Fig. 32b, Table 27).

This indicates that L. scutulata in some way interfered with egg mass production or deposition at these higher densities.

Significantly fewer egg masses were found in the spring in the more wave exposed cages (Table 28) than in the more sheltered cages.

This phenomenon was not oberved in the fall indicating that exposure to winter storms may have influenced natality. 90

SURVIVORSHIP AND MORTALITY

Methods and Materials

Mortality and survivorship data should complement each other.

However a few animals were always lost, possibly by emigration through tears in the cage fabric. Losses by this means were most pronounced in

December when storm action resulted in large rips in cages 6 and 21

(Table 38).

Survivorship data was analysed (courtesy of Dr. Niel Gilbert) by plotting survivorship curves Log Original Number of Animals versus

Time) for all 24 cages. Analysis of co-variance was used to compare the slopes of these curves (representing rate of survival) between single and mixed species treatments and among density treatments.

Since snails have hard shells, mortality can be measured by counting the number of empty and crushed shells directly. It was assumed that the causes of mortality for crushed shells were due to physical factors (e.g. being hit by water born logs and rocks) and these data were considered separately from empty shell mortality.

Since food availability differed from summer to winter, the empty shell mortality data was partitioned into two parts. As mentioned previously, winter food was not limiting at any density but in the summer distinct food levels were maintained by the three snail densi• ties. To test the idea that shelter may have been limiting at high' densities, the number of animals in exposed places (the top of the slabs and on the exposed cage walls) were counted on March 3 and March 6,1970 . 91

All the low numbered cages (l to 12) were situated on one side of the beach, whereas the higher numbered cages (l3 to 24) were on the other side (Fig. 18). The right cages were resting on a shelf, whereas the.left cages were resting on a cobble gradient which sloped into deeper water. The left cages were thus exposed to direct wave action, whereas the shelf on the right side of the beach tended to break the wave force, and thus offered these cages more protection. Data was stratified into two groups to test for position effects.

Chi-squared tests were performed to determine whether mortality- data and survivorship data and the number of non-sheltered animals conformed to the assumptions that no difference between the single and mixed species treatments existed and further that there was no differ• ence among the proportional mortality rates of the three density treatments (Figs. 35, 36; Tables 30, 3l). The total number of animals was split equally between single and mixed species cages and parti• tioned among the density treatments in the ration of 1 to 2 to 4 for the overall goodness of fit to the two assumptions. Further chi- squared tests were performed to determine whether species effects and or density effects could explain the discrepancy from the expected values (Tables 30, 31).

Animals of both species often released fluke larvae. Upon dissection it was revealed that these parasites had taken over the entire digestive gland and gonad. It was conceivable that such heavy parasitism could lead to death of the snails. To determine whether species composition and density of snails had an effect on the 92 proportion of animals parasitised, original animals of both species and from all densities were isolated in individual stacking dishes in

June 1970 in an attempt to observe the release of fluke larvae.

Results

The comparison of the slopes of survivorship curves showed that neither species nor density effect were detectable (Table 29). However the number of original experimental animals surviving one year is greater for L. sitkana under single species treatments than under mixed species treatments and also greater under lower than under higher densities (Fig. 34; Table 30). L. scutulata showed no difference between survival in mixed and single species treatments at all densi• ties. However survival in the high density single species treatments was proportionately higher than in the single species lower density treatments. This indicates that survival improved as the density was increased (Table 31; Fig. 34).

In this location JJ. scutulata appeared to live longer than L_. sitkana for 49$ of the original L. scutulata survived one year, where• as only 9$ of the L. sitkana did.

No species or density effects were detected for the crushed shell mortality of either species. The overall chi-squared tests for goodness of fit of observed values to the two assumptions of even mortality with regard to species composition and density treatment was poor for both species (Tables 30,' 3l). None of this discrepancy however could be attributed to species or density effects.

The summer empty shell mortality for both species was smaller 93

luly Aug. Sept. Oc*. Secamfeer Feb. April 3une-

33 Fig. Mortality Index of L. sitkana and L. scutulata as a function of Time. ZO 40 80 Number of Animals per Cage

Fig.34 Original Number of Littorines surviving one Year as a function of Density.

Single Species Cages Mixed species Cages 95

i_ 1 • 20 **-0 80

Number of Animals per Cage

Fig. 35 Number of Empty Shells of L. sitkana found in Winter and Summer as a Function of Density. In mixed species treatments significantly more animals died in the winter. 96

Mortality Data L. scutulata .A io Mixed sp. bo Winter

SO

30 Single species

"S io

£a in io to 90

0 i- J) 5b

Summer

io- Mixed species

2o Single species

•o

xo HO 80 Number of animals per cage

pig«36. Number of Empty Shells of L. scutulata found in Winter and Summer as a Function of Density. In the winter L. scutulata at high densities survived significantly better in single species cages than in mixed species cages. Proportionately fewer animals died under high densities single species treatment than under lower densities single species treatments in the winter. Summer Mortality

L. sitkana

Ho —i 2o 80

L. scutulata

.. *

t TB >^ ^' ' •"—"

r 1 lt> HO SO

Number of Animals per Cage

Mortality Patterns in right and left cages. Fig.37 Number of Empty Shells found in the summer as a function of Density. Note, that there is no difference between left and right cages.

Right Cages Left Cages ^ Single species cages X Mixed species cages 98 Winter Mortality Mixed species Left cages

Single species L. sitkana Left Cages

Mixed species Right Cages

Single species Right cages

80 HO

L. scutulata .X Mixed sp. Left

Mixed sp. Right

Single sp, Left

Single sp, Right

Niunber of Animals per Cage

Pig. 38 Number of Empty Shells found in the Winter as a Function of Density. Note, that the animals in the right cages survived better than those in the left cages. 99

Crushed Shell Mortality

..•X Mixed sp. 2zo L, sitkana

0) A S. • Single sp. n lis

0*

«> 10 n1

2$

V 3 c. 3LO MO to

lOr L. scutulata Mixed Species 2 _ ST A

Single species

20 MO ZO Number of Animals per Cage

Fig.39. Total Number of crushed shells from June 1969 to June 1970 as a Function of Density. 100

Fig. 40. Total Number of animals found in exposed places in the experimental cages on two occasions ( March 3 and March 6 1970 ) as a function of density. Proportionately more animals were found in non-sheltered places as the density increased ( Tables 30, 31 ). The top of the slabs and the roof of the mesh cages were considered exposed sites. 101

Number of Animals per Cage 102 than the winter mortality (Pigs. 33> 35, 36), this data conformed to the assumptions of no density or species effects, whereas the winter mortality did not, since in this case both species survived better under single species treatments than under mixed species treatments

(Figs. 30, 35; Tables 30, 3l). Mortality proportional to density was observed for L. sitkana at single and mixed species treatments and for

L.. scutulata for mixed species treatments. L_. scutulata single species treatments had proportionately more animals dying at medium density than at high density.

In the summer there was no difference in the mortality between left and right cages for either species. In the winter however, animals of both species at all the densities did not survive as well in the exposed (left) cages as in the sheltered (right) cages (Fig.38).

On March 3 and March 6, 1970 there were more animals of both species in exposed sites in higher than in lower densities (Tables 30,

31; Fig. 40). This may indicate that shelter became less and less available as densities increased.

Two types of fluke larvae were released by the snails,a large echinostome cercaria and a smaller microphallid cercaria (Table 32).

Ten percent of both species of littorines released microphallids cercariae and 7 percent of the L_. scutulata released echinostome cercaria. Hilda Ching (personal communication) reports absence of echinostome larvae in L. sitkana. There was no difference in the pro• portion of microphallids affecting snails at high density and at medium and low density. 103

Discussion

At any density L. sitkana mortality was greater than L. scutulata mortality regardless of the season. Survivorship data showed the reverse.

A shorter life span combined with a faster growth rate would indicate: that L. sitkana had a greater turnover rate than JL. scutulata.

Since all the cages contained abundant food in the winter, it does not appear that food limitation could account for most of the mortality. Starvation appeared not to play a direct role in the summer mortality because animals in the high density treatments (which were devoid of food) did not survive less well than those at lower densities.

Species interaction effects likewise were not directly related to food limitation, for these effects were only measurable in the high density mixed species cages in the winter.

There is some evidence that shelter from storms may be a clue in explaining the mortality patterns. The highest mortality rates were observed in the storm season. Also it can be noted, that during this time animals of both species did not survive as well in the more wave exposed cages than animals in the more sheltered cages (Fig. 38). In the summer, x^hen storms did not occur, no such difference between left and right cages could be detected (Fig. 37).

Animals of both species survived less well in high density mixed species treatments than in high density single species treatments 35 36; 30, 3l). (Figs. t Tables Neither food nor shelter appeared to be less abundant in mixed species cages. Perhaps the mutual inhibitory effects these two species have on each other is related to their 104

social behaviour.

An equal proportion of L. sitkana died at all densities but pro•

portionately fewer original animals survived one year as the density

increased (Fig. 34). This would indicate that living conditions remained

as favorable or else became less favorable as densities increased. The

opposite trend held for _L. scutulata especially in single species treat• ments (Fig. 38b; Table 3l). Littorines have been observed aggregated

in sheltered places during the winter and perhaps this behaviour is

beneficial in protecting animals from cold winds, surf and ice formation

as their densities increased.

More experimental data is needed to clarify the observed density

and species interaction effects. 105

DISCUSSION AND CONCLUSIONS

Hairston, Smith and Slobodkin (i960), suggest, that as a rule, grazing populations are not food limited and are prevented from over- exploting their food supply by the action of predators and parasites.

Benthic marine predators seem to thin out littorines at low tidal level where algal food is most abundant, but these predators were never found in the mid and high tidal zones. Stomach samples of crows and ducks revealed littorine shells (Low, personal communication), however no estimate of their effect on littorine populations was made.

Two species of fluke larvae were found in the digestive gland and gonad of littorines, but these parasites did not appear to exert a density dependent mortality.

Both L_. sitkana and L. scutulata appear to be food limited in the high density cages (twice the natural density) in the summer, in that they grew significantly less in these cages than in the low density cages with a high standing crop of algae. Decreased growth due to a scarcity of food however did not affect numbers directly. It is possible that a nutritional limitation in the summer and possibly competition for sheltered places in the fall was responsible for the decreased natality rate shown by L. sitkana in the high density cages.

Eisenberg (l966, 1970) demonstrated that nutritional limitation in high density cages of the pond snail Lymnaea elodes had the effect of reduc• ing the number of eggs produced but had no effect on adult mortality.

Natality rates for L_». scutulata could not be measured, but mortality 106 rates decreased with density.

The lower the density of L. sitkana and the higher the availabi• lity of food, the greater were growth rates, natality and survivorship'.

These trends indicate that density responsive mechanisms may be prevent• ing this species from overexploiting its food supply. Such mechanisms would be important to a species such as L. sitkana with direct develop• ment and limited dispersal potential, for if it became extinct in one place, the chances of it reinvading that place might be very small.

L. scutulata grew less but survived better at higher than at lower densities. L. scutulata can live at much higher densities and"can utilise much lower food levels than L. sitkana. Extremely high densities of L. scutulata (l68 animals per 25 cm ) were observed on the exposed transect at Chesterman's Beach. These high densities were maintained possibly because these sites were very favorable for settlement of the larvae. These animals however were extremely stunted and the oldest animals (at least one year old) were not more than 4 mm long. High densities of stunted adult L. sitkana were never found. Prom the density species interaction experiment and from growth data gathered at other sites it was found that L. scutulata would often shrink in size (Fig.

29) whereas L. -qi tkana would either maintain its size or else die.

Higher growth rates together with a greater locomotory activity indi• cates that _L. sitkana may have a greater metabolic rate and may require more food per individual than L_. scutulata and thus cannot maintain as dense a population. L. scutulata, at high densities, may contribute little if any energy to reproduction. This is perhaps unimportant for a 107 species with planktotrophic development since recruitment can take place from a more favorable area.

The competitive exclusion principle leads to the expectation that two ecologically similar species using the same resource cannot co• exist indefinitely, for one species would be more efficient at utilising that resource and thus would increase in numbers and displace the other species (Hardin, I960). When two such species overlap in their distri• bution it is often assumed that competition keeps them apart and that coexistence implies lack of competition. Resource partitioning whereby both specie specialise on different aspects of the same resource is one way whereby such competition can be minimised. For example, five species of northeastern warblers in mature, homogeneous, coniferous forests each feed in one of five locations in the same tree (MacArthur,

1958)* Since both species of littorines feed on benthic diatoms and shelter in the same places, resource partitioning does not appear to play a role.

No two species are exactly alike in their ecological requirements and thus habitats exist where only one of a species pair can live. This is the case with L. sitkana and L. scutulata. L. scutulata does not live in high splash pools or lagoons, presumably because its planktonic larvae cannot settle there. L. sitkana cannot live on exposed coasts without shelter from waves and sun, as adults tend to be dislodged by wave action and juvenile stages tend to die from desiccation. In intermediate habitats both species can co-exist. Lack of competitive interactions is not responsible for the co-existence of 108

L. sitkana and L. scutulata in the intermediate habitat at Cantilever

Pier, for at high densities both species died at faster rates in mixed species cages than in single species cages and L. scutulata inter• fered with the natality of L. sitkana .The degree to which two species overlap in their distribution would depend on the extent of the inter• mediate habitats and the degree of balance in their respective adaptive advantages.

Hutchinson (1957) suggests that if the advantage of one species over the other is constantly reversed by environmental fluctuations, co-existence might be possible. This was shown by Harger (1968, 1970a,

1970b) for mixed population of mussels consisting of Mytilus califomianus and M. edulis. M. edulis had the ability to crawl to the outside of mussel clumps thus preventing themselves from getting smothered and crushed in the matrix of the clump. M. califomianus were attached to the substrate by stronger b3gsal threads. During storms

M. edulis would get washed off, thus freeing the formerly inprisoned

M. califomianus. Calm weather would allow M. edulis to re-establish themselves. This phenomenon did not appear to apply to L. sitkana and

L. scutulata grown at Cantilever Pier, a site equally favorable to L. sitkana and L. scutulata, for growth and mortality patterns showed the same trends (Figs. 26, 33). L. sitkana. however, suffered a greater mortality during the winter and produced fewer egg masses in the spring in the more exposed (left) cages than in the more sheltered (right) cages (Table 28). This together with susceptibility to wave action, would select against L. sitkana during storms. In the summer the 109 advantage of L. sitkana over L. scutulata may be associated with its • greater growth rates.

Williamson (l957)> postulates that two ecologically similar species can co-exist if they do not have the same controlling factor.

Predators such as Leptasterias hexactis and Searlesia dira seem to prefere L_. sitkana to L. scutulata (Tables 13, 14). The preference of these predators for L. sitkana. counterbalanced by the apparent preference of the echinostome fluke larvae for L. scutulata, may very well contribute to the maintenance of both species of littorines in an area. Unfortunately, there is no information bearing on the degree to which parasites and predators control littorine populations.

Nicholson (1954) points out that continual recolonization from

species specific refuges would allow two similar species to live in

the same habitat. The co-existence of L_. sitkana and L. scutulata at

Cantilever Pier beach however does not depend on species specific

refuges for both species demonstrated reproductive potential.

Co-existence is possible if each species inhibits itself more

than the other species when it becomes numerically more abundant.

Drosophila willistoni and D. pseudoobscura exhibit such a balance

situation (Ayala, 197l). At high D. wiT)istoni and low D. pseudoobscura

densities I), willistoni was at a disadvantage and D_. pseudoobscura at

an advantage as far as natality was concerned. At high D. pseudoobscura

and low D. willistoni densities D. pseudoobscura was at a disadvantage

and D. willistoni at an advantage as far as survivorship was concerned.

As densities increased the number of egg masses produced per individual 110

L. sitkana decreased in single species cages but not in mixed species cages (Fig. 30). This may inidcate that intraspecific competition but not interspecific competition became more important as densities in• creased.

L_. scutulata appears to be the generalist of the two species of littorines in that they are found almost everywhere. The highest densities of L. scutulata adults and juveniles were associated with barnacle cover whose interstices may act to trap the larvae. L. scutulata larvae do not appear to settle in pools and mud flats to any great extent and thus this species would not inhibit L_. sitkana in these areas. In more exposed locations L,. i:\ear\p showed greater mortality and lower natality. This together with their susceptibility of being dislodged by waves would select against L. sitkana in exposed places. L. scutulata had the advantage of planktonic recruitment, longevity, resistance to wave action and ability to graze on a lower standing crop of algae. L. sitkana apeears to be a specialised species in that it does best in areas such as mud-flats, high tide pools,

(where L_. scutulata does not appear to settle in large numbers) and rough shale rocky beaches. All these habitats offer protection to adults from being dislodged by waves and to egg masses and juvenile snails from desiccation. In such habitats L. sitkana appears to be more efficient than L_. scutulata in finding food and crevices, in converting food to biomass and in having a less wasteful mode of development

(leicithotrophic as opposed to planktotrophic). Ill

LITERATURE CITED

Ayala, F.J., 1971. Competition between Species: Frequency Dependence. Science, February 1971, 820-824.

Bock, C.E., R.E. Johnson, 1968. The Role of Behavior in Determining the Intertidal Zonation of Littorina planaxis and Littorina scutulata. Veliger, 10; No. 1:42-54.

Castenholz, R.W., 1961. The Effect of Grazing on Littoral Diatom Populations. Ecology, 42:783-794.

Connell, J.H., 1961. Effects of Competition, Predation by Thais lapillus and other Factors on Natural Populations of the Barnacle Balanus balanoides. Ecol. Mon. 31:61-104.

Dahl, A.L., 1964. Macroscopic Algal Food of Littorina planaxis and Littorina scutulata. Veliger, 7; No. 2:139-143.

Eisenberg, R.M., 1966. The Regulation of Density in a Natural Population of the Pond Snail, Lymnaea elodes. Ecology 47:889-906.

— 1970. The Role of Food in the Regulation of the Pond Snail, Lymnaea elodes. Ecology 51:680-684.

Foster, M.S., 1964. Microscopic Algal Food of Littorina planaxis and Littorina scutulata. Veliger, 7; No. 2:149-152.

Gause, G.F., 1934. The Struggle for Existence. Wilkins Co., Baltimore.

Gowanloch, J.N., F.R.Hayes, 1926. Contributions to the Study of Marine Gastropods. The Physical Factors, Behaviour and Intertidal Life of Littorina. Contr. Canad. Biol.Fish. N.S. 133-165.

Hairston, N.G., F.E.Smith, and L.B. Slobodkin, I960. Community Structure Population Control, and Competition. American Naturalist 94:421-425.

Hardin, G., I960. The Competitive Exclusion Principle. Science, 131:1292-1297.

Harger,J.R.E., 1967. Population Studies on Mytilus Communities. Ph.D. thesis, Biology Department, University of California at Santa Barbara. 112

Harger, J.R.E., 1968. The Role of Behavioral Traits in Influencing the Distribution of Two Species of Sea Mussels, Mytilus californianus and Mytilus edulis. Veliger 11:45-49•

, 1970a. The Effect of Wave Impact on some Aspects of the Biology of Sea Mussels. Veliger 12:401-414.

, 1970b. The Effect of Species Composition on the Survival of Mixed Populations of Sea Mussels, Mytilus californianus and Mytilus edulis. Veliger 13:147-152.

Baseman, J.D., 1911. The Rhythmic Movements of Littorina littorea Synchronous with Ocean Tides. Biol. Bull., Woods Hole, 21: 113-121.

Hertling, H., W.E.Ankel, 1927. Bemerkungen uber Laich und Jugendformen von Littorina und Lacuna. Wissenschaftliche Meeresuntersuchungen- Kommission zur Untersuchung der Deutschen Meere in Kiel und der Biologischen Anstalt auf Helgoland. Neue Polge. Abteilung Helgo• land, XVI. Band, Abhandlung Nr. 7.

Hutchinson, G.E., 1957. Concluding Remarks, Cold Spring Harbor Symposia on Quantitative Biology 22:416-427.

Imai, E., 1964. Some aspects of Spawning Behavior and Development in Littorina planaxis and L. scutulata. Student Spring Report, Hopkins Marine Station, Monterey, California.

Kanda, S., 1916. Studies on the Ceotropism of the Marine Snail Littorina littorea. Biol.Bull., Woods Hole, 39:57-84.

Keen, M.A., 1937. An Abridged Check-List and Bibliography of West North American Marine Mollusca. Standford University Press, Stanford, California.

Low, C., 1970. Factors Affecting the Distribution and Abundance of Two Species of Beach Crabs Hemigragsus oregonensis and H. nudis M. Sc. Thesis, Zoology Department, University of British Columbia, Vancouver.

Mac Arthur, R.H., 1958. Population Ecology of some Warblers of North Eastern Coniferous Forests. Ecology, 39:599-619.

Menge, B., 1970. Ph.D. Thesis, Zoology Department, University of Washington.

Murdoch, W.W., 1970. Population Regulation and Population Inertia. Ecology 51:497-502. 113

Nicholson, A.J., 1954. An Outline of the Dynamics of Animal Populations. Australian J. Zool. 2:9- -5.

Orldroy, I.S., 1929. The Marine Shells of the West Coast of North America. Stanford University Press, Stanford, California.

Siegel, S., 1956. Nonparametric Statistics for the Behavioural Sciences. McGraw-Hill Book Company Inc., New York.

Sokal, R.R., F.J. Rohlf, 1969- The Principles and Practice of Statistics in Biological Research. W.H. Freeman Co., San Francisco.

Stephenson, T.A., A. Stephenson, 1949. The Universal Features of Zonation between Tide Marks on the Rocky Shores. J.Ecology, 37:289-305.

Struhsaker, J.W., J.D. Costlow Jr., 1968. Larval Development of Littorina picta reared in the Laboratory. Proc. Malac. Soc. Lond. 38:153-160.

Thomas, R.I., 1966. The Distribution and Zonation of Prosobranch Molluscs of the Genus Littorina on the Central Oregon Coast. Student Summer Report, Newport Marine Station.

Thorson, G., 1950. Reproductive and Larval Ecology of Marine Bottom Invertebrates. Biol.Rev., 25, pp.1-45.

Urban, E.K., 1962. Remarks on the Taxonomy and Intertidal Distri• bution of Littorina in the San Juan Archipelago, Washington. Ecology, 32:320-323.

Williamson, M.H., 1957. An Elementary Theory of Interspecific Competition. Nature 180:422-425. APPENDIX

TABLES 1 to 56

Significance Levels

Stars were used to indicate the level of significance at which the null hypothesis of no difference between values has been rejected.

Absence of a star or N.S. indicates no significant difference.

.05

.01

.001 115

TABLE 1.

Time sequences for the developmental stages of Littorina scutulata.

Egg Laying 0 hours

Polar body formation 2 hours

Two cell stage 3 hours

Four cell stage

Late cleavage 12 to 24 hours

Blastula

Gastrula 21 hours

Young veliger 3 days

Pre-hatched veliger 4 days

Hatching 7 to 8 days at 13° to 15°C 3 days at room temperature (22°C) 116

TABLE 2.

Hatching success of Littorina sitkana egg masses suspended from a boat moored in Vancouver Harbor.

Number of Eggs

Hatched not hatched

Higher 'cages' 50 cm below surface of the water line 1 40 40

line 2 47 40

Lower 'cages1 100 cm below surface of the water line 1 15 50

line 2 8 50

Total 110 180 117

TABLE 3.

Survival of two size classes of juvenile L. sitkana caged at the 9 foot and 13 foot tidal levels at Lilly Point from May .17 to May 18, 1969.

POSITION SALINITY TEMPERATURE RECOVERY OF L.SITKANA OF CAGES OF POOLS

medium 27°C 7 small (newly hatched) tidal water snails all with their 3 evel foot moving. ( 9 ft.) cage with larger (1.0 mm pool or longer)snails was lost

medium 20°C 9 small snails, all alive tidal air 5 large snails, all but level one opened operculum ( 9 ft.) when moistened. dry

high 25£ 27°C 8 small snails, all alive tidal water 5 large snails, all alive level (13 ft.) pool

high 20°C 9 small snails, all dead tidal air k large snails, 3 alive level one with broken shell (13 ft.) dry 118

TABLE 4.

Hatching success of L. sitkana egg masses at Lilly Point.

Five L. sitkana egg masses were sectioned in two. Half of each egg mass was attached to pilings at the high tide level and half were attached to concrete blocks at the low tide level at Lilly Point.

INITIAL TIDAL NUMBER OF HATCHED LITTORINA SITKANA COLOR OF HIGHT OF EGG MASS CAGE September 1 September 7

pink 5 ft. not sampled 50 alive * { 11 ft. red egg mass G alive light 5 ft. not sampled 7 alive * pink ( 13 ft. yellow and dry 0 alive dark 6 ft. 25 hatched 32 alive pink 13 ft. red and dry 0 alive light f 6 ft. red egg mass 47 alive pink V13 ft. 2 hatched 0 alive light (9 ft. lost pink 113 ft. covered with covered with sediment silt

* indicates puncture in cage 119 Table 5.

Position of L. sitkana egg masses found in 16 cages at Cantilever pier beach.

Surface Area Number of egg Position ( cnr) masses on Oct.22

1. wave exposed side 76 0 of slab

2. wave sheltered side 76 29 of slab

3. top of slab 760 1

4. bottom of slab 760 26

5. right or left side 320 19 of slab

6. in crevice of cage seam ( wave sheltered) small 47

7. on side or bottom of cage large 8

y Wave impact Diagram showing cage removed from cement slab with positions marked. 120

Table 6. Local Distribution of Littorines.

Lowest Record. Wave Sa• Salinity Presence Expos• lin• Sept.* 68 of Place Substrate ure ity to Jun. '70 sit. scut.' Lonesome Cove, rough shale M H yes yes SJ Marvista Resort ,rough shale H ' H yes yes SJ Cantilever Pier,rough shale M H yes yes SJ boulders & M H yes yes gravel Pile Pt., SJ rough shale H H yes yes boulders & H H yes yes gravel Deception Pass rough shale M H yes yes State Park, boulders M H rare yes near Anacor- tes Anacortes Fer• gravel M H yes rare ry landing pilings M H rare yes Saxe Pt. Park rough shale M H yes yes near Victoria shelf smooth wall M H no yes Albert Head rough shale M? H yes yes Light shelf

False Bay,SJ mud flat w/rocks L H yes yes & gravel Montague Har• mud flat w/grav• L H yes yes bor, on Gal- el & oyster eano Is. shells shore-gravel & M H rare yes

oyster shells • Jekel's La• mud flat w/ grav• L H yes no goon, SJ el & rocks 1 Fanny Basin mud flat L yes rare 1 near Nanaimo pilings no yes Chesterman's wave exposed VH H no yes Is. near shore Tofino high splash L ? yes no pool Victoria granite cubes w/ H H no yes Breakwater barnacles Port Renfrew sand.shore flat H H yes yes June, 1969 Nov., 1969 VH H rare yes high tide pools H H yes rare 121

Table6.? continued Lowest Record. Wave Sa• Salin. Presence Expos• lin• 9/68 - of Place Substrate ure ity 6/70 sit. scut. Edward King shore-rough H H yes yes Is. near' shale Bamfield high tide pools L yes no Lilly Pt., compressed tin M M 21.2#. No yes Pt. Roberts cans, rock & boulders Brockton Pt. seawall L M 16.9%« rare yes boulders & gravel L M yes yes Second Beach boulders L M yes yes Eng. Bay by boulders L L no yes Sylvia Hotel Under Burrard boulders L L no yes St. Bridge Maritime boulders & L L yes yes Museum seawall Jericho barnacled pier L L no yes Beach Spanish Banks by: 1. oceano• Barnacled boulders L VL no yes graphy plat• form it 2. guntowers L VL 3.0/O0 no no

Gales Park boulders L L? no yes Light House sandstone M M? no yes Park shelves Sunset Marina boulders L M? yes yes

Wave Exposure Estimates VH = very exposed H = exposed M = medium exposure L = sheltered

Salinity Estimates (for values around Vancouver, courtesy of Ora Johannsson) H = high salinity, around 31ppt. most of the year. M = salinity may drop as low as 15 ppt. in the summer L = salinity may drop as low as 7 ppt. in the summer VL = salinity may drop below 7 PPt. in the summer

SJ = San Juan Island TABLE 7

Number of animals of both species found on the surface and in crevices of a quadrat ( 50 cm by 50 cm ) taken at Marvista Resort ( Fig. 11 ). Animals in surface sample were brushed off and animals in crevice sample were picked out with forcepts.

Small Animals ( size 1 to 4 )

L. sitkana L. scutulata

Obs. Exp. Obs. Exp.

crevice sample 59 48.33 118 128.66

surface sample 12 22.66 71 60.33

X2 = 10. **

Large Animals ( size 5 and. 6 )

L. sitkana L. scutulata

Obs. Exp. Obs. Exp.

Crevice sample 15 4.25 12 22.75

Surface 10 20.75 122 111.25 sample

X = 38.95 *** TABLE 8.

Ability of the two species of littorines

to resist wave exposure in the field. 124

WAVE EXPOSURE Proportion of animals Difference remaining betw«n»r«*s Place Date Substrate L. sitkana L. scutulata X2

Lilly point Oct. 20-22 barnacled 7/10 7/10 1968 tin can rock, horizontal

barnacled 2/10 8/10 ' 1.979 tin can rock, vertical

barnacled 4/20 6/20 piling, vertical

April vertical tin 11/20 16/20 15-16 can rock 2.990 1969 horizontal 14/20 16/20 tin can rock

Brockton Oct. 2-4 Point 1968 sea wall 11/20 13/20 7.466 March sea wall 21/50 35/50 13-14

Saxe Feb. 8-11 vertical 4/38 15/38 7.017 Point 1969 smooth wall

Chesterman1s May 1-2 exposed 37/49 47/49 6.740 Island 1969 shore, barnacles

Port Renfrew June 5-6 low level, flat 26/50 46/50 8.734 1969 sandstone shelf

high level, 33/50 47/50 10.562 on Fucus sp.

High level, 31/50 47/50 13.119 sandstone shelf

Nov. Low level, 34/50 35/50 0.047 21-22 sandstone shelf 1969 small animals

low level, 19/50 34/50 9.0325 sandstone shelf, large animals

POOLED DATA 254/487 366/487 55.657

(Y)* Yates correction for small cell frequency was used. 125

TABLE 9.

WAVE EXPOSURE

Laboratory Experiments

Number of animals remaining on substrate after squirting them with a jet of sea water

Number L. sitkana L. scutulata used 18 1 5 18 1 10 18 1 4

Total 54 19 12.842 (Y)*

Proporion of animals remaining on substrate after squirting them with a jet of sea water.

Large Small L. sitkana L, sitkana

4/27 10/24 4/24 10/22 2/23 5/24 0/12 1/12 3/20 5/19 2/20 6/20

Total 15/126 37/121 12.947

Yates correction for small cell frequencies was used. 126

TABLE 10.

Tolerance of two species of littorines to 3 and 4 days desiccation at room temperature.

Time of Exposure L. scutulata L. sitkana (days)

Number

alive 20 1?

Number Dead 0 3

Number

alive 44 35

Number dead 16 25

3.00 N.S. 127 TABLE 11

HIGH WATER TEMPERATURES

Proportion of animals recovering after Three minute 2 minutes 24 hours exposure to L. sitkana L. scutulata L. sitkana L.scutultata

44° water 1/25 2/25 1/25 11/25

45° water 0/25 0/25 0/25 5/25

30° water 25/25 25/25

30° water 25/25 25/25

Exposure to Proportion of animals recoverin after radual heating 2 minutes 219 Hours f15° C in 3 hours) L. sitkana L. scut. L, sitkana L. scut,

Dish b , heated up to 45° C o/25 0/25 17/25 25/25

Dish a, heated up to 42° C 1/25 4/25 11/25 17/25

Total L. sitkana L. scutulata Number of Alive 29 IB animals Dead 71 42

x2 = 17.109 *** D.f.= 1 PC .001 128

TABLE 12.

Tolerance of littorines to salinity extremes.

Groups of 50 ( size 4 ) animals of each species were subjected to 16°C sea water at various salinities. After 28 hours the number of dead animals was noted.

SALINITY NUMBER OF ANIMALS COUNTED DEAD AFTER 28 HOURS. (parts per thou thousand) L. sitkana L, scutulata

7.5 12 22

15 2 3

30 0 1

50 2 13

Differences between species

7.5 parts per thousand X2 = 4.455 *

50 parts per thousand X2= 7.843 **

X2 1 D.f. (K .05 = 3.84

^ .005= 7.79 129

Table 13 . Electivity Coefficients for Searlesia dira (proportion of prey organisms of one species in diet divided by the proportion of that prey species in the environment). An electivity value greater than 1 indicates prey selection, one less than 1 indicates prey avoidance.

Total Feeding Number in Electivity Prey Species Observations Environment Coefficient

Littorina sitkana 57 357 3.3^

L. scutulata 30 1+95 1.28

Acmea digitalis & A. paradigitalis 21 1399 0.315

A. scutum 13 37b 0.73

A. pelta 7 211 0.70

Total prey organisms observed being eaten = 138

Total prey organisms in environment = 2920

Data courtesy of Margaret Lloyd Colin's Cove study area October, 1968 to July, 1969 130

Table 14 . Electivity Coefficients for Leptasterias hexactis (pro• portion of prey organisms of one species in diet divided by the proportion of that prey species in the environment). An electivity value greater than 1 indicates prey selection, one less than 1 indicates prey avoidance.

Lonesome Cove Lonesome Cove Dead Man's Cattle Species Resort Area Far Point Bay Point

Balanus glan- dula 0.61 2.14 0.88 0.23 B. cariosus 1.04 0.24 2.78 0.35 Acmaea scutum 5.25 0.51 0.84 1.82 Littorina scutulata 2.91 1.54 _ 6.53 Lacuna spp. 5.3^ 0.15 0.71 1.54

Acmaea pelta 2.69 1.95 4.47 4.99 A. paradigitalis 1.72 3-59 0.43 2.38 L. sitkana 14.53 - - 5.86 Tonicella lineata 8.45 _ Margarites spp. - 1.42 0.23 0.28 Ishnochiton sp. - 5.89 - - Chthamalus dalli 0.22 - O.69 0.16 Katherina tun- icata 0.25 0.09 0.4i

Data courtesy of Bruce Menge 131

Table 15 . Food choice by Leptasterias hexatis as determined by contact c prey. Relative Escape # Obs. Feedings Preference Coef. Responses of Prey Species # Contacts (%" success, encounters) Prey Species

Lacuna sp. 6/25 24.00 + Littorina scutulata 9/40 22.50 + L. sitkana 9/hh 20.40 + Acmaea digitalis 2/15 13.30 + A. paradigitalis V79 5.10 - Mytilus edulis 1/42 2.38 sessile A. pelta 2/74 2.70 . ++ Calliostoma ligatum 1/58 1.72 +++ Balanus glandula 2/121 1.65 sessile

A. scutum 0/124 0.00 +++

Katherina tunicata 0/35 0.00 - Tonicella lineata 0/25 0.00 - Balanus cariosus 0/12 0.00 sessile Thais canaliculata 0/17 0.00 - Thais lamellosa 0/17 0.00 - Thais emarginata 0/16 0.00 +

Data courtesy of Bruce Menge Table 16. Survival of L. scutulata caged in high intertidal and splash zone at Lilly Point, Point Roberts.

Height of cage Proportion of animals alive after 38 days

1 ft. above highest observed L. scutulata 3/7

l£ ft. above highest observed L. scutulata, sun and wave sheltered side of piling, in yellow lichen zone. 5/6

i ft above highest observed L. scutulata, in black lichen zone 10/10 133

TABLE 17.

Mean Growth Increments of Littorina sitkana and Acmaea scutum from single and mixed species cages.

Species Treatment Mean Length N standard T Increment deviation value

Acmaea scutum

single species 0.1263 19 0.2795

mixed species 0.2222 9 0.3589

0.773 N.S.

Littorina sitkana

single species 2.644 17 1.889

mixed species 1.977 9 1.104

0.794 N.S. 134 TABLE 18.

Starfish predation on littorines and limpets at two tide levels. L. scutulata A. paradigitalis Tidal Level ( feet) # empty, # attached, # empty, # attacked

-0.5 4 0 4 0

112 0

50 5 0

4 0 0 0

Total "~14 1 11 o

+ 2 1 0 0 0 t 0 0 0 0 0 0 0 2 1 0 0 1 Total 0

The number of Littorina scutulata and

Acmaea paradigitalis eaten by Leptasterias

hexactis at two tidal levels.

Difference between number of Acmaea paradigitalis and number of L. scutulata eaten X2= 0.164 N.S.

Difference between number of animals eaten 2 at low and mid tidal level X =16.397 *** 135

TABLE 19,

Dispersal behaviour of L. sitkana under two densities, Number of animals reaching the edge of the slab after 5 minute trials.

Number of 12 12 36 an j. max o per slab /slab number tA 2 3 Total

4 1 3 8 6 16 13 7 36 24 9 8 11 28 20 8 6 14 28 26 12 17 10 39 21 7 14 10 31 20

Total 56 59 55 170 117 Observed Expected 143 143

X2 calculated =11.0 *** 136

TABLE 20.

Number of littorines of both species found on cement slabs with and without algae.

Diatom covered slabs Clean slabs

Number of Number of Number of Number of L. sitkana L. scutulata L. sitkana L. scutulata

6 0 0 0

10 2 0 2

10 1 2 0

11 1 1 0

6 2 3 2

4 3 2 1

47 9 8 5 Total Observed

28 7 28 7 Expected

Difference between diatom and clean slabs.

L. sitkana X2 = 13. 0 ***

L. scutulata X2 = 1.2 N.S. 137

TABLE 21

Total number of littorines of both species leaving diatom covered and clean slabs.

diatom covered slab clean slab

L. sitkana L. scutulata L. sitkana L. scutulata

Total

Observed 24 9 51 20

Expected 37.5 14.5 37.5 1^.5

Difference between diatom covered and clean slab.

L. sitkana X2 = 9.720 ** o L. scutulata X = 4.172 * TABLE 22.

Position of L. sitkana 14 hours after introduction into modified sea water table ( Fig. 16 ).

Number of Animals

No Food Slab 4

Food Slab 4

Bottom of water Table 5

Around Drain 14

Side of water Table 23 139 TABLE 23.

Behavioral response of littorines to crevices.

Five littorines of each species were allowed to attach to oyster shells with barnacles ( creviced substrate ) and to oyster shells without barnacles ( smooth substrate). After 10 minute intervals the number of littorines remaining on each .type of shell was noted.

Littorina sitkana

Smooth shells barnacled shells

0 0 5 8 3 3 5 4 1 0 4 7

Total 9 47

2 X 1 d.f. 27.557 ***

Littorina scutulata

Smooth shells barnacled shells

0 2 3 11 4 2 7 6 0 1 4 7 1 0 6 7

Total 10 51

X2 1 D.F. = 25.786 *** 140

TABLE 24.

Comparison between abundance of food in L. sitkana and h* scutulata cages through out the year. ( Data was taken from tables 32 to 40 ). The Wilcoxan Matched-Pairs Signed- T Ranks tests were used ( Siegel,(1956) pp. 75-83 ). Low numbered or high numbered cages of L. sitkana and L. scutu• lata comprised a matched pair..

Density Number of Number of T paired ranked Rank with tabled values differences less fre• <* =.05 quent sign

Low 16 13 35.5 N.S. 14 or less

Medium 16 10 21 N.S. 8 or less

High 16 6 6.5 N.S. 0 141

TABLE 25

Biomass and length increment relationships for L. sitkana and L. scutulata in August from single species low density treatments.

Species Original Original Weight Percent adjusted dry weight increment increase mean length ( grams ) ( grams ) in weight ( mm )

7 L. scutulata 9.750 0.01275 0.00095

L. sitkana 8.244 0o01225 0.01005 82

Species Original Length Percent adjusted increment increase length (mm) ( mm ) in length

.2462 L. scutulata 8.997 o 3

L. sitkana 8.997 2.14075 24 142

TABLE 26

The effect of density on natality.

Chi-squared tests were used to test the hypothesis that an equal proportion of egg masses were laid under the three density treatments.

Number of egg masses produced

Low Medium High density density density XX2 Fall Single species 31 5^ 48 Observed

Expected 19 38 76

27.03 ***

Mixed species Observed 10 26 52

Expected 12.6 25.1 50.3

0.61 N.S,

Spring Single species

Observed 15 H 8

Expected 4.9 9.7 19.43 35.81 ***

Mixed species

Observed 14 1 1

Expected 2.3 4.6 9.1 sample size too small 143

TABLE 27.

Tests on the Effect of Species Composition on the Natality Bate of L. sitkana.

Range tests ( Sokal, Rohlf, 1°69 ) were used to test the hypothesis that there was no difference between the number of egg masses laid in single and mixed species cages.

DENSITY MEAN NUMBER OF EGG MASSES RANGE RANGE OF TEST DENSITY Single sp. Mixed sp. VALUES VALUE

Natality as a Function of L. sitkana density un• corrected for equal density 20 15.5 6.5 23 .41 * FALL 40 27 13 28 .50 *

Natality, single species corrected for equal number of animals per cage. 10 7.25 2.25 12.5 .40 * FALL 20 13.5 6.5 13 .50 * 40 12 13 23 .04 N.S.

Natality as a Function of L. sitkana density un• corrected for equal density 20 7.5 .25 15 .48 * SPRING 40 5.5 .25 8 .38 *

Natality, single species corrected for equal number of animals per cage 10 3.75 3.50 7.5 .03 N.S, SPRING 20 2.75 .25 4. .50 * 40 2.00 .25 3 .58 *

Range Test = Difference of 2 Means/ Range

cx=.05, N= 6, Range Test Value = .312 144

TABLE 28.

Comparisons between the natality of L. sitkana in single and mixed species cages and between left and right cages ( pooled data ).

Number of Egg masses produced

Single species Mixed species Difference X2 Pall Observed 133 88 Expected 110.5 110.5 4.58

Spring Observed 34 16 Expected 25 25 6.48 *

Right cages Left cages Difference X2

Pall Observed 116 105 Expected 110.5 110.5 0.54 N.S,

Spring Observed 44 6 Expected 25 25 28.88 *** 145

TABLE 29.

Comparison of survivorship curves of littorines in 24 cages for species and density effects.

L. sitkana

Analysis of Variance D.F. S.S. M.S. F Densities 2 7.3390 3.6695 0.2415 N.S, Species 1 6.1479 6.1479 0.4046 N.S,

Interaction 2 4.6888 2.3444 Remainder 12 182.3126 15.1927

L. scutulata

Analysis of Variance D.f. S.S. M.S. F Densities 2 2.3632 1.1916 3.0956 N.S, Species 1 0.0313 0.0313 0.0820 N.S,

Interaction 2 0.9741 0.4871 Remainder 12 4.5807 0.3817

F, d.f. 2,12 (^=.05 )= 3.8853

F, d.f. 1.12 (©\=.o5 )= 4.7472 TABLES 30, 31.

Survivorship, and Mortality Data for L. sitkana and L. scutulata.

Chi-squared tests were performed to test the following hypotheses.

1) There is no difference in the survivorship, mortality and number of non-sheltered animals in single and in mixed species cages.

2) There is no difference in the proportion of animals surviving, dying, and being non- sheltered in the three density treatments.

3) There is no difference in the mortality, survi vorship and number of non-sheltered animals in left ( more sheltered ) and right ( less sheltered ) cages.

Data for these tests were taken from figures 3^ t CHI-SQUARED D.F. SURVIVORSHIP EMPTY SHELL MORTALITY CRUSHED-SHELL • NON-SHELTERED June 69-June 70 SUMMER WINTER MORTALITY ANIMALS March 3&6,1970 Fig. 34 Fig.35 ,37 Fig.35,38 Fig. 39 Fig. 40 Total Overal good• ness of fit 2 23.396*** 5.703* 17.333*** 7.781* 13.7173*

Species Low Density 1 .642 1.195 0.000 2.400 ' 3.200 Med.Density; 1 1.066 1.067 1.841 .016 1.636 High Density 1 14.087*** .203 15.138*** .116 .006 Total 3 15.795*** 2.464 16.979 2.531 4.842 Species pooled density 2 15.076*** .446 13.586** .953 .008 Heterogeneity 1 .719 2.018 * 3.392 1.578 4.834*

Density .994 1.250 3.447 9.553** Single Species 2 2.2189 2 .952* Mixed Species 2 sample size too 4.128 .606 4.197 3 small Total 4 5.117 3.856 7.644 13.505** Density pooled- species 2 7.328* 2.976 .398 - 4.397 9.929** Heterogeneity 2 2.140 3.458 3.247 3.576

Left vs right cagew 1 27.769*** .018 40.533*** 1.613 19.854*

TABLE 30 L. sitkana CHI-SQUARED D.F. SURVIVORSHIP EMPTY SHELL MORTALITY CRUSHED SHELL NON SHELTERED • June 69-June 70 SUMMER WINTER MORTALITY ANIMALS March 3&6,1970 Fi?. 34 Fig. 36 Fig. 36,37 Fiej. 39 Fig. 40

Total Overal good• [ ness of fit 2 9.500* 2.07 18.304* 8.327* 19.555*

Species Low Density 1 .421 .643 .044 sample too 0.000 small Med^ Density . 1 .071 .037 .056 1.066 .190 High Density 1 3.613 .011 21.631*** .642 2.919 Total 3 4.095 .691 21.730*** 3.109 St>ecies pooled density 2 1.008 .287 8.563 1.778 .478 Heterogeneity 1 3.087 .404 '13.167*** 2.631

Density Single Species 2 7.530** .659 10.459* 1.229 3.855 Mixed Species 2 ,580 .938 2.248 sample too 12.725** small Total .4 8.110 1.597 12.707* 16.580** Density pooled snecies > 2 4.390 .518 4.725 4.772 15.755*** Heterogeneity 2 3.720 1.079 7.982* .825

Left vs right cages 1 5.27* 1,391 15.223* 1.111 32.507***

TABLE 31. L. scutulata CO.. 149

TABLE 32.

Incidence of cercariae shedding in experimental animals.

Number of Echinostome Microphallid animals Number % Number % examined

L. sitkana 56 0 0 6 10.7

L. scutulata 60 4 6.7 6 10.o

Number of animals ( lumped species ) shedding Microphallid cercariae at two desnities.

Low and Medium High Density Number shedding .' Densities Microphallids 3 9 Number not shedding 51 51 Microphallids

Chi-squared 1.946 N.S. Friday Harbor Density-Species Interaction Experiment, June 17 to July 15, I969

# Empty OD at 665 Estimated # Survivors Shells millimicrons Amount Density Species Cage #f sit, scut, sit, scut, sam. a sam. b of Algae Control Flyscreen con- 0 Littor• trol .236 .019 ++ ines uncaged control 25 .001 .001 0 • control next to #5 .004 0 control next to wall .004 0 Low L. sitkana 1 19 2 .452 .507 +++ 20 L. sitkana . 13 19 1 .026 .019 ++ L. scutulata 2 21 0 .066 .066 + L. scutulata Ik 17 2 .085 .032 Mixed spp. 3 7 8 k 1 .025 .037 ++ Mixed spp. 10 9 . 11 1 . 0 1.250 .630 Mixed spp. 15 8 8 1 0 .014 .090 + Mixed spp. 22 12 10 0 2 .062 .028 ++ Medium L. sitkana 6 39 2 .002 .0035 0 4o L. sitkana 17 36 k .003 .002 0 L. scutulata k ko 2 .082 .0125 ++ L. scutulata 16 39 1 .005 .005 0 Mixed spp. 5 22 20 1 0 .001 .0025 0 Mixed spp. 11 19 19 1 0 .005 .008 0 Mixed spp. 18 17 19 k 2 .003 .004 0 Mixed spp. 23 15 19 0 0 .002 .003 0 High L. sitkana 7 70 7 .002 .003 0 80 L. sitkana 19 75 7 .0025 .0025 0 L. scutulata 8 77 1 .002 .005 0 L. scutulata 20 78 0 .006 .0025 0 Mixed spp. 9 29 38 7 0 .002 .002 0 Mixed spp. 12 37 38 1 . 0 .004 .002 0 Mixed spp. 21 3k 38 6 0 .002 .004 0 Mixed spp. 2k 39 38 1 1 .003 .003 0 Friday Harbor Density-Species Interaction Experiment, August 11

# Surviv• # Empty OD at 665 ors Shells millimicrons # Unaccounted For Density Species Cage # sit. scut. sit. scut. sam.a sam.b sit, scut. Controls Caged control .113 .277 uncaged control? 25 .018 .012 Control next Vcage effect .010 .016 to #5 no cage ;ef. Control next "7 cage effect .012 .008 to rock wallj no cage ef. .OOU .004

Low L. sitkana 02 19 0 .090 .165 N = 20 11 13 5 .obQ .040 L. scutulata 01 21 0 .353 .!+59 n 14 16 1 .Obk .031 Mixed spp. 03 7 8 . 212 .171 11 10 9 8 .221+ .357 11 i-3 15 6 7 2 1 .025 .027 CD 11 22 8 11 1 0 .025 .O69

Medium • L. sitkana 6 .36 3 .005 .029 N = bO 11 17 3b l 0 .006 2 L. scutulata 04 36 3 .015 .015 1 11 16 39 .007 .005 Mixed spp. 5 21 20 0 0 .005 .005 11 11 17 17 1 1 .012 .007 1 1 ti 18 16 18 1 1 tube.broken.Oil 11 23 13 19 2 0 .008 .007

High L. sitkana 7 57 5 .0025 .003 N = 80 11 19 69, 5 .002 .002 1 L. scutulata 8 76 1 .005 .003 11 20 77 1 . 002 .005 Mixed spp. 9 29 39 0 1 .001+ .003 II 12 32 3b 3 0 .001+ .003 1 1 11 21 3b 38 •h 2 .001+ .001+ +1+ 11 24 36 39 b 0 0 .002 Friday Harbor Density-Species Interaction Experiment, Sept. 17

# Empty # Original # Survivors Shells Animals Left # Unaccounted For Density Species Cage # sit. scut. sit. :scut. sit. scut. sit. scut.

Controls Caged control uncaged 25 cage effect 25 uncaged Next to cage effect ' #5 Low L. sitkana 01 18 2 17 0 L. sitkana 13 15 2 10 3 L. scutulata 02 20 0 20 0 L. scutulata Ik 20 1 16 +1 Mixed spp. 03 9 9 1 0 6 7 1 k 15 6 11 k 2 8 +3 t-3 Mixed spp. a* Mixed spp. 10 8 10 2 0 6 8 M Mixed spp. 22 6 3 0 5 9 CD VJ1

Medium L. sitkana 06 32 8 29 0 L. sitkana 17 32 2 27 k L. scutulata Ok 38 2 33 0 L. scutulata 16 39 2 38 +1 Mixed spp. 05 18 19 3 1 18 +1 0 19 -2 • Mixed spp. 11 19 17 1 1 0 Ik 19 6 1 11 0 0 Mixed spp. 18 17 Mixed spp. ".. 23 15 18 5 1 8 0 -1

High L. sitkana 07 70 15 ~W +5 L. sitkana 19 50 17 1+9 -3 L. scutulata 08 81 o 81 +1 L. scutulata 20 78 2 75 0 38 6 23. 37 +1 +1 Mixed spp. 09 35 3 -k Mixed spp. 21 26 35 9 3 21 33 -2 Mixed spp. 12 35 39 5 3 29 36 o +2 Mixed spp. 2k 32 36 0 2 -2 Friday Harbor Density-Species Interaction Experiment, Sept. 17 (con'td)

Number of OD at 665 Estimated egg masses millimicrons Abundance Microscopic in cages Density Species Cage # sam. a sam. b of Algae Algae red pink yellow Total

Controls caged control 1 .041 .090 +++ uncaged 25 i .020 .020 ++ cage effect .025 .013 ++ uncaged Next to .030 .010 ++ cage effect #5 .010 .042

Low L. sitkana 01 .040 .060 ++ b Ulva L. sitkana 13 .100 .140 ++ L. scutulata 02 .240 .270 +++ L. scutulata 14 .070 .192 .++ 3 Ulva co Mixed spp. 03 .080 .040 ++ 1 aM Mixed spp. 15 .020 .045 ++ ro Mixed spp. 10 .180 .202 +++ heavy Ulva cover/diatoms Mixed spp. 22 .070 .095 +++ 9 Ulva

Medium L. sitkana 06 .100 .130 +++ L. sitkana 17 .090 .060 ++ 12 12 L. scutulata Ok .060 .110 +++ 2 Ulva • L. scutulata 16 .090 .145 +++ Mixed spp. 05 .040 • 305 +++ ' 9 9 Mixed spp. 11 • 060 .110 ++ Mixed spp. 18 .220 .360 ++++ Ulva: Mixed spp. 23 .080 .065 ++ 1 2 3

High L. sitkana 07 .030 .032 ++ greenish L. sitkana 19 .060 .010 ++ L. scutulata 08 .030 .042 ++ greenish L. scutulata 20 .110 .062 ++ Mixed spp. 09 .030 .040 ++ greenish 1 1 Mixed spp. 21 .070 .050 ++ Mixed spp. 12 .040 .045 ++ greenish Mixed spp. 2k • 055 .075 ++ k k Friday Harbor Density-Species Interaction Experiment, October 22, 1969

# Empty # Original j //Survivors Shells Animals Left # Unaccounted For Density Species Cage # sit, scut. sit. scut. sit. scut. sit. scut.. Controls caged uncaged 25 cage effect Next to uncaged #5 cage effect Low L. sitkana 01 Ik 1 13 0 L. sitkana 13 5 10 6 0 L. scutulata 02 16 2 16 0 H3 p> L. scutulata : 11+ 18 1 17 0 C I-1 Mixed spp. • 03 5 7 2 1+ - 6 0 0 CD Mixed spp. 10 8 9 7 6 0 -1 CTl Mixed spp. 15 6 11 ' 1+ 8 0 0 Mixed spp. 22 1+ 8 1 1 k 8 0 0 Medium L. sitkana Ok 19 13 16 0 L. sitkana • 17 2l+ 12 21 0 .L. scutulata 06 36 1 33 -1 L. scutulata 16 35 1+ 3k 0 Mixed spp. 05 11 16 5 11 16 -1 -3 Mixed spp. 11 13 15 12 13 0 -2 Mixed spp. 18 9 17 k 1 7 13 -1 -1 Mixed spp. 23 ll+ 17 2 0 7 16 +1 -1 High L. sitkana 07 59 7 k5 -4 L. sitkana 19 1+7 1+ 1+6 +1 L. scutulata 08 75 0 73 -5 - L. scutulata 20 69 8 67 -1 Mixed spp. 12 30 35 k 3 25 32 -*8 +3 Mixed spp. 09 19 35 5 0 17 33 -10 -2 Mixed spp. 21 31 36 3 l 27 36 -1 -2 Mixed spp. 2k 30 36 1 1 28 36 -1 +1 Friday Harbor Density-Species Interaction Experiment, October 22, I969

O.D. at 665 Estimated macrosc0pic Number of Egg masses millimicrons abundance ^ color of Egg masses j k Density Species Cage if | samp, a samp, b of algae _ red p n yellow Controls caged i .102 .093 ++ uncaged *) . 25 .104 .086) cage effect J" next to .092 .119J uncaged "I #5 .205 .177? ++++ cage effect \ .064 .095J Low L. sitkana 01 .187 • .159 +++ 3 L. sitkana 13 .147 .321 ++H-+ 6 L. scutulata; 02 .103 .020 ++ L. scutulata 14 .084 .202 ! 4 Ulva ++++ Pt3-3 Mixed spp. 03 , .320 .194 ++++ ,. Enteromorp* a" Mixed spp. 10 .244 .470 all Ulva Mixed spp.• 15 .500 .272 ++++ V>1 Mixed spp. 22 .058 .069 +-H-+ Medium L. sitkana 04 .570 .232 +++++ 3 brown 4 9 11 L. sitkana 17 .280 .108 ++++ 3 1 L. scutulata 06 .280 .106 ++++ L. scutulata ' 16 ++++•)- 2 brown . Mixed spp. 05 .114 .087 ++++ Mixed spp. 11 .110 .068 ++ Mixed spp. 18 .157 .195 +++ Mixed spp. 23 .382 .500 +++++ High L. sitkana 07 .175 .210 ++++ 1 15 L..sitkana 19 .800 .214 +++++ 3 12 L. scutulata 08 .073 .086 ++ L. scutulata 20 .598 .398 +++++ 2 brown Mixed spp. 12 .105 .035 +++ ui Mixed spp. 09 .275 .195 ++++ 7 8 VJl Mixed spp. [ 21 .580 .213 +++++ 8 Mixed spp. ! 24 .200 .330 +++++ 1 20 Friday Harbor Density-Species Interaction Experiment, December 9? 19^9

# Empty # Original Est. Abun. #Egg # Alive Shells Animals # Missing of Masses Density Species Cagesit. scut. sit, scut, sit, scut. sit..scat. Diatoms red pink yel. Controls Caged 25 i uncaged cage effect next to #5 Low L. sitkana 02 15 5 12 0 ++++ 1 3 10 L. sitkana • 13 5 3" k -12 ++ L. scutulata 01 17 3 14 0 +++ L. scutulata 14 18 0 15 0 +++ k Ut,rd i-3 Mixed 03 8 10 1 0 8 • 0 0 0 - 0 1 0 P a1 Mixed 10 6 9 1 0 5 0 -1 +++ Kttf* 1 3 0 M Mixed 15 0 10 7 0 0 . io 0 0 CD 1 Mixed 22 4 7 k 0 2 -1 0 +++

Medium L. sitkana 04 19 5 15 -6 bit c- 0 k L. s itkana 17 31 6 19 -3 ++++ o 2 6 L. scutulata 06 20 2 20 -18 ++++ L. scutulata 16 35 2 31 0 4-f 1 brow* Mixed 05 14 17 5 3 8 14 ' 0 0 ++++ o 0 1 . Mixed 11 5 12 3 3 2 11 -7 -1 ++ . Mixed 18 10 18 . 9 1 k 17 -1 -1 ' 0 0 Mixed 23 6 15 6 4 6 13 -7 0 ++ High L. sitkana 07 57 13 39 -5 + k 0 3 L. sitkana 19 56 15 32 -1 +++ 3 1 3 L. scutulata 08 73 2 69 -3 +++ L. scutulata 20 79 0 70 -1 ++ Mixed 09 o). 36 3 23 32 0 0 + 0 0 1 Mixed 12 17 36 . 21 4 13 32 +3 0 + 21 10 10 11 -8 Mixed 2k 4 21 -13 ' ++ . 0 0 1 h Ik Mixed 24 3'+ 24 5 13 29 0 0 ++++ c Friday Harbor Density-Species Interaction Experiment, February 2, 1970

# Empty # Original Est. Abun- I # Alive Shells Animals # Missing dance of Density Species Cage # sit. scut. sit. scut. sit. scut. sit. scut. Diatoms

Caged ++++ Control Cage effect +++ uncaged ++ Low L. sitkana 02 18 1 10 1 +++++ L. sitkana 13 10 9 2 I ++++ L. scutulata 01 20 0 Ik 0 +++++ ^3 L. scutulata 14 17 2 12 1 +++ P> Mixed 03 6 9 3 0 3 0 1 ++++ n CD Mixed 10 O 9 2 0 k - 0 1 Small Mixed 15 7 10 3 0 0 8 0 0 +++-<- 00 Mixed 22 9 8 1 0 +1 -2 +++ Medium L. sitkana Ok 28 k 2 +++ L. sitkana 17 26 9 12 5 ++++ L. scutulata 06 31 7 0 ++++ L. scutulata 16 32 7 23 1 +++ • Mixed 05 Ik 19 k 1 2 0 +++ Mixed 11 8 18 8 2 0 k 0 • +++ Mixed 18 12 12 6 7 2 9 0 0 ++++ Mixed 23 12 Ik 9 6 5 9 +1 0 ++++ High L. sitkana 07 . 57 12 7 ? L. sitkana 19 35 31 13 14 +++ L. scutulata 08 Ik 3 1 ++ L. scutulata 20 71 10 62 +1 ++ Mixed 09 27 3k 10 6 2 0 + Mixed 12 19 36 16 0 5 4 + k Mixed 21 19 32 14 7 14 7 1 +++ Mixed 2k 6 28 28 8 0 2k 6 4 ++ Friday Harbor Density-Species Interaction Experiment,-April 17, 1970

# Empty # Unac• Est. Abun• # Alive Shells counted For # Egg Masses dance of Density Species Cage |# sit. scut. sit. scut. sit. scut. red pink yellow Diatoms Caged control ++ cage effect 0 uncaged control 0

Low L. sitkana 2 17 2 1 12 3 ++ uw* L. sitkana 13 11 9 0 ++++ br«,-*Uw",v,Vo^4 L. scutulata 1 18 0 -2 L. scutulata Ik 16 3 -1 Mixed 3 9 9 1 1 0 0 2 +++++ Mixed 10 6- 5 2 5 2 . k 3 3 6 +++ kuw«. Mixed 15 2 7 7 4 1 +1 +++ Mixed 22 k 10 5 1 • 1 +1 ++++

Medium L. sitkana k 32 7 -i 2 6 ++++ L. sitkana 17 23 14 -3 3 ++++ b«W4ul^ L. scutulata 6 36 4 0 +++ 16 8 -1 +-H-++ £3 L. scutulata 31 CO Mixed 5 17 19 3 1 0 0 1 ++ cr-j j Mixed 11 19 19 3 1 +2 0 +++ CD Mixed 18 8 17 10 3 -2 0 +++ VJJ Mixed 23 9 15 10 k -1 -1 ++++ High L. sitkana 7 68 12 - 0 6 0 L. sitkana 19 30 37 -13 1 1 ++++ L. scutulata 8 73 7 0 + L. scutulata 20 74 6 0 + Mixed 9 22 33 18 8 0 +1 + ...... - 0 Mixed 12 2k 36 5 3 -11 -1 « Mixed 21 13 31 25 10 -2 -8 1 +++ CO Mixed 2k 26 27 13 12 -1 -1 +++ Friday Harbor Density-Species Interaction Experiment, June 22, 1970 Est. # Empty # Orig. Anim- # Unaccount, # Egg Jbun. i # Survivors Shells als Alive for Masses 8ia- Density Species Cage //; sit, scut. sit. scut. sit. scut. sit. scut. red pink yel. toms. Control caged control +++ cage effect uncaged control ++

L. s itkana 2 19 2 9 +1 L. s itkana 13 17 3 0 0 l ++ L. scutulata l 17 1 10 2 ++++++ UlUA. L. scutulata l4 20 7 0 +++ U\l>*. Mixed spp. 3 7 10 2 0 0 6 •1 0 2 ++++ Mixed spp. 10 9 7 1 3 5 •k 0 0 dry +++++ Mixed spp. 15 7 10 2 0 0 5 1 ++ Mixed spp. 22 10 10 0 . 0 6 ++ Ii. s itkana 4 36 4 k 0 + L. sitkana. 17 39 2 6 +1 + fD L. scutulata 6 36 1 11 3 L. scutulata 16 37 3 18 + <° .Mixed spp. 5 18 19 2 2 5 10 0 +1 Mixed spp. 11 19 19 2 1 0 10 +1 0 + Mixed spp. 18 18 18 2 2 0 8 + Mixed spp. 23 16 20 3 0 0 6 1 ++ L. sitkana, 7 72 6 20 2 0 L. sitkana 19 67 12 1 -1 0 L. scutulata 8 75 5 58 0 L. scutulata 20 66 12 41 2 0 Mixed spp. 9 38 37 1 2 2 24 1 1 0 Mixed spp. 12 38 39 6 1 0 23 1 0 0 H Mixed spp. 21 37 35 2 k 0 17 1 1 0 ^ Mixed spp. 2k 36 32 k 2 0 10 0 6 0 TABLES 41 to 56

Growth Data for L. sitkana and L. scutulata for the period July 1969 to June 1970. Tables 41 to 55 follow the following pattern:

Treatment Density Species Number Composition

1 low single

2 low mixed

3 medium single

4 medium mixed

5 high -single

6 high mixed

Table 56 follows•the reverse pattern. ( REGRESSION RESIDUAL N SSX SPXY SSY B SS DF SS DF 38 12 2.51 8. 56 9. 31 0.070 0.60 1. 8.71 36 34 131 .06. 2.70 6. 41 0.021 0.06 1 6.35 3? 75 319.62 -8. SO 9. 8 6 -0.023 0.24 1 9.61 73 73 39li9 2 -21.79 10. 49 -0. 05 6 1.21 1 9.28 71 136 3 74.98 -12.32 9. 0 8 -0.033 0.40 1 8.67 134 .133 .45.8, .06 -12. 45 10. 9 6....-0.0 2 7 0.34 1 10.62 131

POOLED REGRESSION CX EFFICIENT I S -0 .02 5

SOURCE SS OF MS F REGRESSION (BB AR ) 1.08 1 1 .0 3 .,- 9.69 AMONG SAMPLE RS 1.77 5 0.35- 3. 17 POOL ED RES I DUAL 53.25 477 0.11 WITHIN SAMPLE 56. 10 433 1,180 LovJi 'single £ • Co I—' cr1 0. 752 Hect si^e <

pi THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 8.218 CO c+ CO

o REGRESSION RES IOUAL N SSX SPXY SSY 3 SS D SS DF 1 31 105.87 ' 4.7C 36.17 0.044 0.21 35. 96 29 2 29 132.00 -12.26 23.55 -0.093 1.14 22 41 27 3 67 263. 37 -39.23 27.62 -0. 146 5. 73 21.89 65 62 2 85.£2 -50 .65 34 .0 7 -0.177 8.99 2 5.03 60 125 353. J55 -41. 73 21. 56 -0.118 4.92 16.64 123 118 349.30 •36.39 20. 2 4 •0.104 3. 79 16.45 116

POOLED REGRESSION COEFFICIENT IS -0.117

SOURCE SS DF MS F REGRESSION (BBAR) 20.62 1 20.62 i 62.57 AMONG SAMPLE BS 4.16 5 0.8 3 2. 53 POOLED RFSI DUAL 138 .43 420 0 .3 3 WITHIN SAMPLE 163. 21 426 2.107 1.8 74 1.281 K.,S. 1.354 M..H, rr 0.883 U.S. 0.841 H-'H, fD Y ADJUSTED 54. 21 10.8 4 32.89

X X-= a- THE VALUE TO WHICH YS ARE ADJUSTED IS 8.244 fO •U1U 5>tfk C. 5,*/" REGRESSION RESIDUAL N _ ssx SPXY SSY B SS DF SS DF 1 31 63. 15 -9. 57 17. 1 6 -0.152 1.45 1 .15.71 29 2 28 131.03 -8 .09 9 .12 -0.06 2 0. 50 1 8.62 26 -36.69 < 3 64 194. 85- 24 .24 -C. 18 8 6.91 1 17.33 62 ? 4 66 , . 234.J95 -30.81 ?0 . 56 -0. 131 4.04 1 16.52 64 5 119 Z69.0H -38.84 17 . 79 -0.144 5.61 1 12.18 117 . 6_ 119 385.29 -28. 34 _ 13 • 6 9_._-Q.._07 4 2_.08 1 16.61 117

POOLEO' REGRE SSIQN COEFFICIENT IS -0. 119.

SO URCE SS DF MS F REGRESSION (BRAR) 18.16 1 18 . 16 > 86. 65 AMONG SAMPLE BS 2.44 5 J.49 - 2.33 POOLED RESIDUAL 86.97 4 15 0.21 WITHIN SAMPLE _ 107.56 421 •

1.431 k,*., wit-" H9 1 .2.03 U, M.. _ :. <§ • g. 1.02 0 s. W M N•) CD H- CD 0. 837 K. n- ci- ( s. & 0.484' CD s ^ H C.477 H. Y ADJUSTED ~~ "40 .0 2" ' 5 8.00 38.19 O —. d- V A< THE VALUE TO WHICH YS ARE ADJUSTED IS 9.334 P> X X= c+ Co

ON ro REGRESS ION RES I DUAL N SSX SP XY SSY B SS DF SS DF 1 19 48 .66 1. 39 6.80 0.029 0.04 1 c.76 17 2 21 8 8. 70 0. 36 1. 79 0.004 0.00 1 1.79 19 3 42 59.80 -7.72 7.76 •0. 129 1.00 1 6.76 40 y 4 45 8 7> 22 -A . 53 6.22 -0.052 0. 24 1 5.99 43 5 105 197!. 33 -20. 10 14.56 -0. 102 2.05 1 12.52 103 .-17 ft ... 107 256.24 .M4 11.13 •0.070 1. 24 1 9.89 105

POOLED REGRESSION COEFFICIENT IS -0 .06 6

SOURCE SS DF MS F REGRESSION ( B B A R ) 3.18 1 3 .1 8 . 23. 78 AMONG SAMPLE BS 1.38, 5 0.2 3- 2.07 POOLED RESIDUAL 43.70 3 27 0 .' 1 3 WITHIN S AMPL E _ 48.27 333 0.729 Oltr^ 1-3 C. 544 o • to c+ 0.654 0 to M H- CD 0.432

c+ dat a THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 9.914

ON REGRESSION RES I DUAL N _ ssx SPXY SSY B SS D SS DF 18 142.13 •2.33 0.34 -0.016 0.04 0.30 16 i 18 79. 32 •3. 46 0.38 -0.044 0.15 0.23 16 2 39 166 . 18 -4. 16 0.61 -0.025 0. 10 0. 50 37 3 4^ 27 72.j76 -1 .12 0.05 -0.015 0.02 0.03 25 5 75 196. ;i6 •4. 5 8 0.55 •0.023 0. 11 0.44 73 .6. 54 172.32 •4.85 1.06 -0.028 0. 14 0. 92 52.

POOLED REGRESSION COEFFICIENT. IS -0 .0 2 5

SOURCE SS DF MS F REGRESSION (BBAR) 0.51 1 0.51 \ 45.66 AMONG SAMPLE BS 0.05 5 0 .01 0.85 POOLED RESIDUAL 2.43 219 0 .01 WITHIN SAMPLE. _ 2.98 225 0.27 5 Wit* H3 0. 219 CD • fD OCD & 0. 209 B H- CD

0. 153 CD 0. 145 H _ 0.182 Y ADJUSTED 0.31 0. 06 5. 66 CD VJ1 o—«

13' p- THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 10.198 f» c+ P

ON iTable 46. ' !L. si kana growth data February II >0 (C H vl- CO O f\J t\i ITi-4 " (X) ZD

rH in CM vT ,00 l/) f\J CM CM CM O: 00 oo CM CM rn UJ O o O O O Oj o LU CO a: cr o m r- a OS (M ro. re CM M o o O O o Oj X CM o• o • o• o • o • o i CM o ro CM 00 1 1 1 1 1 v ? o 1 ! • • • t—I o C >- co s t\j »C CO «Cf. LLJ oo so cc r- f- O t— 00 . . . OO 1 ZD O o — rH o rH ~> .—i IT. o G o r-i- <

UJ >• in L' o UJ 00 o I o o X x CJ m >-H oo o r- r-r*~ i—* CM < oo • • . x- r- 4- oo CD < : c isTn vC C- CM m m -4" cr G 00 in « n o UJ LU o z ifi ^ m _J cc Cj) UJ r-i CM CM f\J CM rH CM ""J O O z: -i i • • • • • • G LU ^ IA >0 o o o t- O O o a LU s: a ~ o o o < X < O 2 -A.

auaxns mi SNMI sswsna KOIUH I Table 47. i L. si kana growth data i •April IX. CO r-l CM sJ- in o (XI rH in in O- oo < c 00 r~ <- in vO >!• —'o o cn O r—1 o r- » oo • • a « o m o CM a ct: r-l CM m rn

u_ r—1 r-l —i r-l r-l r-l o 1 o i » oo r-l m co m CO a> CM UJ on cc a- -o cn CT' I— 00 OO in o cn in r- «—l —> 1—1 in CV CM o cn o t—i' r- r-l CV CM X r-l m CO' o CM m m < a. • • • • oo cn m (V m •m • •i n• CM oO cn r\! m m ii—n' m CM m 1 1 l 1 UJ 00 > 1 1 j r-l r-l: I o i X X m m r- CM: o tx 00 cn in CC r—; -i a: IU o 00 z. Q 1 UJ UJ < UJ < cc oo < CO o cn CM CM IT in c- CO oo o -z r- CM r~ CM CO CM i o a O UJ • cn CM cv in ov- o- 1 UJ _J X • . •

•WKOhM Oil 157

Table 48. L. s_ i Itkana growth data June

u_ m r- N- CO CM >0 O —I <

Q i/l >T O m co r- >— 1/1 & >? •r in 5> 1/1 • » t » • UJ o c •T m N «a- ,H m

2 O a to o m sT co to to IT. f\i to • t • • • t CO UJ o CO in r» (NJ •c —1 m cn ; in cco 1—1 CO o 1° at 1 « 00 j m i CO CO CO o 3- O • • O • o• o• O NO SO o to 1 l 1 1 l 1 m rn r» CO to • < • o m o cn > o —I CT o> m i £ CM to in co CM m CM (/> • • t • t • CM to in CM in cc IT) «o m r» CM o o -5 i—i U- m o in o r-i. O «~i CM < UJ UJ >• ffj CO m CO CM ON CO CM cc so CO o 0> co m en o < a. • • t t fl • * • • » to —1 o U. CNi *G O <0 to r~ j- or- o o CM LL tO O CM «—< O >- 1 1 1 •4-CM —< UJ tO CM 1 1 1 D X O o ^4 X I—< CM oo r- vO o or X 00 _J to o « « r- o m < to a to CO o CM• • • co < cn in to CO UJ (NJ CO u o -o oa- UJ UJ o CM cn m •d- •i" o UJ o oc o. to _J UJ ID 71 jr UJ < o < or < in 0> o> cr o cc «i" CM m to > to ato to m m m ^ —t ;D r-1 UJ UJ o uj •—i o co m I-ION") -J cc Z -J X • * •

6 140v 361.79 -6 .06 0.72 -0.017 0. 10 1 0.62 138

POOLED REGRESSION COEFFICIENT IS -0 .017

SOURCE SS DF MS F REGRESSION (BBAR) 0. 46 1 0.46 40.92 AMONG SAMPLE 3S • 0. 04 5 0 .01 0.78 POOLED RESIDUAL 5. 18 4 61 0 .01 WITHIN SAMPLE 5. 69 467

-0.006 H3 -0.009 • CO c ca Ma" -0.00 3 o 0 -0.046 PH' • -0.055 c-t- Y ADJUSTED 0. 20 5 0 .04 3.65 ,' CO U°.H o t3* THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 9.881 REGRESSIGN RF SI DUAL N SSX SPXY SSY B SS DF SS DF 1 37- 208.09 -10 .64 3.34 -0.051 0. 54 2. 79 35 2 30 172.19 -14.27 3.94 -0.083 1.18 2.76 28 3 50 194.06 -18.86 7.41 -0.097 1.83 5.58 48 y 4 5 8 244.29 •15 .79 7.91 •0.06 5 1.02 6.88 56 5 87 239.55 -4. 41 1. 77 •0.018 0.08 1.68 85 6 10 2 257.32 13. 77 5.0 4 •0. 054 0. 74 4.30 100

POOLED REGRESSION COEFFICIENT IS -0.059

SOURCE SS DF • MS F REGRESSION (BBAR) 4.59 1 4.59 67.36 AMONG SAMPLE BS 0.80 5 0.. 16 2. 36 POOLED RESIDUAL 24.0 1 0 .07 WITHIN SAMPLE 3 52 29. 40 358 0.20 8 0.231 0. 239 0. 180 £ O 0 .00 4 M » 6 Co C.03

THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 9.750 a. CO CO

U3 L70

Table 51. L. s ciitulata grow;h data u_ in cr CM r- 4- oc September o ro (\j «o in m oj

c oo m ro o CT1 «fr ro 1/1 s m n f\J CO LO

<\J LU O -H IA rH fx ro

U_ rH r-i -H 2 C O r- vD ro i-t oo •-«(*_ ro H N >0 yi y> H in o> O1 00 • • • O. in LU O O O in >t o in u. . UJ . on in r-H ! O o CO O >4- (NJ OJ. t-H ro r- ro o II O O o o o O X • • • • ro o o• o o• o o on' l-H in ro 00 l 1 1 1 1 1 o o o . 00 . . o s: ro o o o > o v0 o o 1 LU 00 vl- rH 00 co o 00 r—1 ro LC. ro ST OO _3 O •—. LL —J in in rH h- o o < z X?" co CO in

z X X O O CO r\i f < l/> 00 • • • oo co CQ < o CO r-- H O CC oO CO LU rvi ro r- _ : •—< ro co LU LU LU O C- rH r-l o; _J »-H rH-H-O <\1 o a: a. 00 c '.JJ r> o LU < (X < en 00 LU < 1 0o0 OO Z r- r-( O -c o oo o o in I— > in in rf, Q oo ~Z CSJ ro 0=^ ro ro vo M rH rH rH O- ro O oo X U—J1 LCCU UJ (M O D s: _i •I— rH O LU o o o O o • • . Q f\j ro .4- in vo c LU »—> X c. a: < a. 3 o a o

-A 300

REGRESSION RESIDUAL N SSX SPXY SSY B SS OF SS DF 1 28 87.05 -2.94 0.38 -0.034 0.10 0.28 26 2 28 113.31 -3.52 0.51 -0.031 0.11 0.40 26 3 69 161.69 -5.02 1.39 -0.031 0.16 1.23 67 4 55 15 3. 13 -2.99 0. 76 •0. 019 0. 06 0.70 53 5 133 26 5.82 •13.02 2.68 •0.049 0.64 2.04 131 6 133 224. 14 -6. 22 2. 54 -0.02 8 0.17 2.36 131

POOLED REGRESSION COEFFICIENT IS -0.034

SOURCE SS DF MS F REGRESSION (BBAR) 1.13 1 1.13 69.81 AMONG SAMPLE BS 0.10 5 0.02 1.27 POOLED RESIDUAL 7.02 434 0.02 WITHIN SAMPLE 8 .26 4 40 0.095 C .049 O • CD M 0.044 CD 0.-021 T= 0.045 £ ro G .08 1 Y ADJUSTED 0.2 2 0 .04 2.72 fo 0*3 O

THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 9.967 to

POOLED REGRESSION COEFFICIENT IS -0.024

SOURCE SS OF MS F REGRESSION < BBAR ) 0.47 1 0. 47 50.07 AMONG SAMPLE BS O.C 2 5 0 .00 0.47 POOLED RESIDUAL 3.22 342 0.01 WITHIN SAMPLE 3.71 348 -0.011 wltr» i-3 0. 02 5 1 a -0.041 o (BOH -0.039 -J3- " m -0.018 M • -0.037 SO Y ADJUSTED 0 . 10 0.02 2.21 CO

o

THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 10.081 p. c+ .

REGRESSION RESIDUAL N SSX SPXY SSY B SS OF SS DF 1 35 76.45 -1.26 0.27 -0.016 0.02 1 0.25 33 2 35 133.14 -3.93 0.54 -0.03G 0.12 1 0.43 33 3 61 168.98 -5.92 0.48 -0.035 0.21 1 0. 27 59 4 61 171.01 -4. 95 0.53 -0.C29 0. 14 1 0. 39 59 5 143 295. 23 -6. 18 0.68 -0.021 0. 13 1 0.55 141 6 130 267.43 -3.91 0 .42 ..-0.01 5 0.06 .1 0. 37 128

POOLED REGRESSION COEFFICIENT IS -0.024

SOURCE SS DF MS F REGRESSION (BBAR) 0.61 1 0.61 123.27 AMONG SAMPLE BS 0.06 5 0.0 1 2.37 POOLED RESIDUAL' 2.26 453 0.00 WITHIN SAMPLE 2.93 459 -0.02 6 hrjhr. ^ ,-. .-, „ CO • CO 0. Ou 3 a- a- -0.022 a : : 2 -0.029 ; -0.041 -0.055 ...... p5- * Y ADJUSTED 0.12 5 0 .0 2 4.84 3

THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 9.619 p. co

POOLED REGRESSION COEFFICIENT IS -0.115

SOURCE SS DF MS F REGRESSION ( BB AR ) 21.53 1 21 .5 3 159.50

AMONG SAMPLE BS 3.37 5 0.67 5. 00 POOLED RESIDUAL 60.0 7 4 45 0.13 WITHIN SAMPLE 84. 98 451 0.554 'iglt. tr" Ue<*t, si • i» 0.53 5 Low a" i T'*e<* co M o CD 0. 636 MecXiiA,^, SI »v>) I-t. 0.54 6 iM eciiV^v, ^i\e

1 0 .426 i-W^W, S,*5IC«M

POOLED REGRESS ION COEFFIC IENT IS -0.237

SOURCE SS DF MS F REGRESSION (BBAR) 48.86 1 48.86 182.92 AMONG SAMPLE BS 1.95 5 0.39 1. 46 POOLED RESIDUAL 102.31 383 0.27 WITHIN SAMPLE 153.12 389 0.782 'r^Mifd • Co 0. 762 W S C CD ca 1.158 M M 1.301 ~TTJ~ 0.978 M

1.124 Lowe 1 r- • co Y ADJUSTED 16. 94 3.39 12.68 <3<-)- coo -rr- THE X VALUE TO WHICH YS ARE ADJUSTED IS X= 9.834 P- co <+ co

VJ1