University of New Hampshire University of New Hampshire Scholars' Repository

Doctoral Dissertations Student Scholarship

Fall 1990

Population ecology and feeding of nudibranchs in colonies of the hydroid geniculata

Walter J. Lambert University of New Hampshire, Durham

Follow this and additional works at: https://scholars.unh.edu/dissertation

Recommended Citation Lambert, Walter J., "Population ecology and feeding biology of nudibranchs in colonies of the hydroid Obelia geniculata" (1990). Doctoral Dissertations. 1623. https://scholars.unh.edu/dissertation/1623

This Dissertation is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected]. INFORMATION TO USERS

The most advanced technology has been used to photograph and reproduce this manuscript from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Bell & Howell information C o m p an y 300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600 Order Number 0108313

Population ecology and feeding biology of nudibranchs in colonies of the hydroidObelia geniculata

Lambert, Walter J., Ph.D.

University of New Hampshire, 1990

UMI 300 N. Zeeb R& Ann Arbor, MI 48106 NOTE TO USERS

THE ORIGINAL DOCUMENT RECEIVED BY U.M.I. CONTAINED PAGES WITH PHOTOGRAPHS WHICH MAY NOT REPRODUCE PROPERLY.

THIS REPRODUCTION IS THE BEST AVAILABLE COPY. Population ecology and feeding biology of nudibranchs

in colonies of the hydroid Obelia geniculata

by

W alter J . Lam bert B.S.. University of Rhode Island, 1982 M.S., University of New Hampshire, 1985

Dissertation

Doctor of Philosophy

in

Zoology

September. 1990 This dissertation has been examined and approved by:

" ■ *4 . ; . ..) ^ 1 1-0 Dissertation Director, Larrv G. Harris Professor of Zoology

-— \ J^nes T. Taylor J Associate Professor of Zoology

• — ^ S £T* James F. Haney Professor of Zoology

O'/' / S Lr( I '-&f Yun-Tzu Kiang / Professor of Plant Biology y

Ronald H. Karlson Associate Professor of Ecology University of Delaware

m * k ^ = L Alan M. Kuzirian Assistant Scientist Marine Biological Lab. Woods Hole. MA Acknowledgements

The completion of this work benefitted from the support and encouragement of many. My wife. Debbie deserves my deepest gratitude and appreciation for her love, support and faith that this project would be completed. Jody Berman and Maiy Sue Potts provided encouragement, advice and friendship. Also, thanks to Paul and Cindy Martin. Mary-Jane James. Marianne deS and Kelly Gestring for their confidence and support during this project. Many divers braved the cold waters of Nubble Light and Newcastle. NH assisting in the collection of samples. Particularly, thanks to Paul Martin. Karen Vemy, Gwen Weisgarber. Steve Truchon. Phil Levin, Nadine Folino and Larry Harris. Their help and faithfulness is greatly appreciated. A number of individuals generously offered advice and assistance during various stages of this work. Marian Litvaitis and Chuck Walker provided technical assistance and space for EM and histology. Drafts of this dissertation were labored over by Adam Marsh. Jody Berman and Marian Litvaitis. Their willingness to "cut and paste" improved the dissertation immensely. Jim Taylor offered time, a computer and assistance in experimental design and statistical analyses as well as conversations on the biology of populations. Hunt Howell provided space and equipment at the Coastal Marine Lab. I also thank my committee: Larry Harris, Jim Taylor, Jim Haney. Yun-Tzu Kiang. Ron Karlson and Alan Kuzirian for advice and encouragment throughout this project. Special thanks are due to Dr. "T" for always keeping the cookie jar filled and to Steve Frank for providing time and help in making slides. This work was supported in part by a Central University Research Fund grant, a summer teaching assistant fellowship and a dissertation fellowship from the University of NH Graduate School.

lii Table of Contents

Acknowledgements ...... lii

List of Figures ...... vii

List of Tables ...... x

A b stra c t ...... xii

Dissertation objectives ...... 1 B ac k g ro u n d ...... 2 descriptions ...... 8 S tu d y s ite ...... 10

CHAPTER 1. The epifauna associated with colonies of the hydroid Obelia geniculata. Introduction ...... 13

Materials and M ethods ...... 15 Collection of anim als ...... 15 Statistical analysis ...... 16

R e su lts...... 17 S p ecies p attern s o f a b u n d a n c e ...... 17 Hydroid abundance ...... 17 Abundances of epifauna ...... 18 Factors affecting species abundances ...... 21 Motile epifauna ...... 21 Sessile epifauna ...... 22

D isc u ssio n ...... 23 Motile epifauna ...... 23 Sessile epifauna ...... 25

F ig u res...... 28

T ab les ...... 44

CHAPTER 2. Distribution and abundance of nudibranchs within colonies of the hydroid Obelia geniculata.

Introduction ...... 49

Materials and Methods ...... 51 Collection of anim als ...... 51 Statistical analysis ...... 52

iv R e s u lts ...... 53 Hydroid abundance ...... 53 abundance ...... 53 Species ...... 54

D isc u ssio n ...... 58 Nudibranch population biology ...... 58 Seasonality ...... 58 S iz e ...... 58 Habitat selection ...... 60

F ig u re s ...... 63

T a b le s ...... 87

CHAPTER 3. Resource partitioning by nudibranchs in colonies of the hydroid Obelia geniculata.

Introduction ...... 90

Materials and M ethods ...... 92 Collection of anim als ...... 92 Analysis of radulae ...... 92 Feeding behavior and feeding damage r . . ' ...... 93 Nudibranch habitats ...... 94 Perisarc analysis ...... 95

R e s u lts ...... 96 Structure of the radulae ...... 96 Feeding behavior ...... 97 Habitats of nudibranchs ...... 101 Perisarc analysis ...... 103

D isc u ssio n ...... 105 Feeding biology ...... 105 S iz e ...... 108 Habitats and feeding ...... 109 Resource partitioning ...... I l l

F ig u re s ...... 114

T a b le s ...... 130

v CHAPTER 4. Behavioral interactions among nudibranchs in colonies of the hydroid Obelia geniculata.

Introduction ...... 138

Materials and Methods ...... 141 Collection of anim als ...... 141 Behavioral interactions ...... 141 Displacement and nearest neighbor ...... 142

R e s u lts ...... 144 Behavioral interactions ...... 144 frondosus ...... 144 coronata ...... 145 Tergioes ...... 146 exiguus ...... 147 Intraspecific interactions ...... 147 Displacement experiment ...... 147

D isc u ssio n ...... 150

F ig u re s ...... 154

T a b le s ...... 172

Summary and Conclusions ...... 174

Literature Cited ...... 177

vi List of Figures

Figure 1. Map of the study site ...... 12

Figure 1-1. Annual variation in size, density and height of Obelia geniculata colonies on collected on Laminaria b la d e s ...... 29

Figure 1-2. Seasonal variation in size, density and height of Obelia geniculata colonies on collected on Laminaria b la d e s ...... 31

Figure 1-3. Percent occurrence and mean abundance of motile epifauna on kelp blades ...... 33

Figure 1-4. Abundance of crustacean epifauna on kelp blades covered with Obelia geniculata ...... 35

Figure 1-5. Abundance of molluscan epifauna on kelp blades covered with Obelia geniculata ...... 37

Figure 1-6. Abundance of mites, ostracods and barnacles on kelp blades covered with Obelia geniculata ...... 39

Figure 1-7. Abundance of turbellarian flatworms and syllid polychaetes on kelp blades covered with Obelia geniculata ...... 41

Figure 1-8. Abundance of ectoprocts on kelp blades covered with Obelia geniculata ...... 43

Figure 2-1. Proportion of kelp blades covered with Obelia geniculata containing nudibranchs ...... 64

Figure 2-2. Proportion of juveniles in the nudibranch population population sampled on kelp blades covered with Obelia geniculata ...... 66

Figure 2-3. Proportion of Obelia-covered kelp blades with Dendronotus frondosus ...... 68

Figure 2-4. Mean abundance of nudibranchs collected on Obella-covered kelp blades from June. 1987 - May. 1989 ...... 70

Figure 2-5. Mean abundance by size class of Dendronotus frondosus in colonies of Obelia geniculata from June. 1987 - May. 1989 ...... 72

Figure 2-6. Size frequency histograms of the four nudibranch species in colonies of Obelia geniculata on kelp blades ...... 74

vii Figure 2-7. Proportion of Obelia-covered kelp blades with and ...... 76

Figure 2-8. Mean abundance by size class of Doto coronata in colonies of Obelia geniculata from June. 1987 - May. 1989 ...... 78

Figure 2-9. Mean abundance of collected on Obelia-covered kelp blades from June, 1987 - May. 1989 ...... 80

Figure 2-10. Mean abundance by size class of Eubranchus exiguus in colonies of Obelia geniculata from June. 1987 - May. 1989 ...... 82

Figure 2-11. Proportion of Obelia-covered kelp blades with Tergines tergipes ...... 84

Figure 2-12. Mean abundance by size class of Tergipes tergipes in colonies of Obelia geniculata from June. 1987 - May. 1989 ...... 86

Figure 3-1. Schematic drawings of a radular tooth row showing the measurements made to quantify radular structure .... 115

Figure 3-2. Scanning electron micrographs of radulae from Dendronotus frondosus and Doto coronata ...... 117

Figure 3-3. Scanning electron micrographs of radulae from Eubranchus exiguus and Tergipes tergipes ...... 119

Figure 3-4. Scanning electron micrographs of a from Obelia geniculata and damage produced by Dendronotus frondosus after feeding ...... 121

Figure 3-5. Scanning electron micrographs of stolons from Obelia geniculata showing drill holes produced by Doto c o ro n a ta ...... 123

Figure 3-6. Scanning electron micrographs of hydrothecae from Obelia geniculata showing penetration holes produced by Eubranchus exiguus ...... 125

Figure 3-7. Field observations of the vertical habitat occupied by nudibranchs in Obelia geniculata colonies at Cape Neddick. M E ...... 127

Figure 3-8. Histogram of perisarc thicknesses of structures from colonies of Obelia geniculata ...... 129

viii Figure 4-1. Ethograms of behaviors involving Dendronotus frondosus before and after an encounter with another n u d ib ra n c h ...... 155

Figure 4-2. Ethogram of Initial behaviors of Dendronotus frondosus with another nudibranch and the reaction of the other nudibranch ...... 157

Figure 4-3. Ethogram of behaviors involving Doto coronata before and after an encounter with another nudibranch ...... 159

Figure 4-4. Ethogram of initial behaviors of Doto coronata with another nudibranch and the reaction of the other n u d ib ra n c h ...... 161

Figure 4-5. Ethogram of behaviors involving Tergipes tergipes before and after an encounter with another n u d ib ra n c h ...... 163

Figure 4-6. Ethogram of initial behaviors of Tergipes tergipes with another nudibranch and the reaction of the other nudibranch ...... 165

Figure 4-7. Ethogram of behaviors of Eubranchus exiguus before and after an encounter with other nudibranchs ...... 167

Figure 4-8. Ethogram of initial behaviors of Eubranchus exiguus with another nudibranch and the reaction of the other nudibranch ...... 169

Figure 4-9. Mean height occupied by nudibranchs on hydrocauli in pair-wise intra- and interspecific density manipulations ...... 171

ix List of Tables

Table 1-1. Common taxa found on kelp blades covered with Obelia geniculata ...... 44

Table 1-2. Multiple regression analyses of parameters of the kelp/hydroid community with abundances of crustacean epifauna ...... 45

Table 1-3. Multiple regression analyses of parameters of the kelp/hydroid community with abundances of Mvtilus edulis. Lacuna vincta and flatworms ...... 46

Table 1-4. Multiple regression analyses of parameters of the kelp/hydroid community with abundances of mites. ostracods. and syllids ...... 47

Table 1-5. Multiple regression analyses of parameters of the kelp/hydroid community with abundances of barnacles and Membranipora membranacea ...... 48

Table 2-1. Indices of dispersion for each nudibranch species in colonies of Obelia geniculata on kelp b la d e s ...... 87

Table 2-2. Analysis of variance tables of parameters of the kelp/hydroid community with abundances of Dendronotus frondosus and Doto coronata ...... 88

Table 2-3. Analysis of variance tables of parameters of the kelp/hydroid community with abundance of Tergipes te rg ip e s ...... 89

Table 3-1. Multiple analyses of variance of radular structures for the four species of nudibranchs feeding on Obelia geniculata ...... 130

Table 3-2. Parameters of the radulae measured to quantify radular structure of nudibranchs feeding on Obelia g e n ic u la ta ...... 131

Table 3-3. Parameters of the penetration hole produced by Eubranchus exiguus in hydrothecae and width measurements of radulae from £. exiguus ...... 132

Table 3-4. Multiple analysis of variance of habitat parameters of nudibranchs in colonies of Obelia geniculata and size of the nudibranchs ...... 133

x PLEASE NOTE

Page xi is missing print. New page not available from school or au­ thor.

University Microfilms International ABSTRACT

POPULATION ECOLOGY AND FEEDING BIOLOGY OF NUDIBRANCHS IN COLONIES OF THE HYDROID OBELIA GENICULATA

by

Walter J. Lambert University of New Hampshire, September, 1990

Kelp blades fl^amtnaria spp.) with colonies of Obelia geniculata were collected over a 28 month period from Cape Neddick, ME to document the epifaunal community and to determine the possible mechanisms allowing four nudibranch species to coexist. Regression analyses correlated the abundances of epifauna with habitat parameters. The feeding biology of nudibranchs was described. Density manipulations of nudibranchs in pair-wise behavioral interactions tested the effects of interference on the location of nudibranchs in the colony.

Obelia geniculata provides habitat and food for many invertebrate epifauna. The hydroid colony acts as the island for most species. Hydrocauli provide structure to motile epifauna. Denser and taller colonies have higher abundances of motile epifauna.

Q. geniculata is a nursery habitat for nudibranchs. Adults of

Dendronotus frondosus. Doto coronata and Eubranchus exiguus were infrequent inhabitants of the hydroid colony. In contrast. Tergipes tergipes consistently occupied the colony for its entire life cycle suggesting that Q. geniculata is the primary habitat of Tergipes.

Each nudibranch species utilized a separate portion of the hydroid colony as food and habitat. Feeding behavior of Dendronotus was size dependent; small nudibranchs (<5 mm) penetrated perisarc and large nudibranchs (>5 mm) bit polyps. Doto drilled through stolons and Eubranrhus penetrated

hydrothecae. Tergipes rasped naked tissue from polyps. Dendronotus was

found throughout the colony on hydrocauli. Doto was on the kelp surface and

Eubranchus on hydrocauli at the edge of colonies. Tergipes occupied Lhe

central area of colonies and atop hydrocauli.

Behavioral interactions among the nudibranchs occurred frequently but

did not influence the habitat or feeding locations of nudibranchs. The vast majority (71%) of reactions of nudibranchs in pair-wise encounters between nudibranchs were non-aggressive. This suggests that nudibranchs are unaware of each other, similar to many insect communities.

Although some separation of the hydroid food resource is present, resource partitioning is unlikely to be a major factor allowing coexistence of the nudibranchs. Interference competition is too rare or weak to cause the observed patterns. Differential recruitment by all four nudibranch species may overwhelm the hydroid but equilibrium conditions necessary for exclusion are unlikely to occur or persist for long periods.

xiii Dissertation Objectives

This study examined the factors regulating the distribution and abundance of nudibranch molluscs within colonies of the coelenterale hydroid

Obelia geniculata on Iaminarian seaweeds. The objectives of this research were:

1. To describe the temporal abundance patterns of epifauna

associated with colonies of O. geniculata on blades of Iaminarian

kelps.

2. To document the temporal changes in recruitment and

population structure of the nudibranchs with respect to physical

parameters (area, height, density) of the hydroid colony.

3. To document the nudibranch feeding biology and relate this to

the structure of the radula. feeding behavior, location within the

hydroid colony and skeletal structure of Q. geniculata.

4. To determine if behavioral interactions among the nudibranchs

affect the distribution of nudibranchs within the hydroid colony.

The results are presented in chapters related to each of the above objectives. The following background section presents an overview of species coexistence and niche partitioning and an introduction to the hydroid- nudibranch assemblage.

1 BACKGROUND

The concept of "limited membership" (Ellon, 1927) in a community states

that all species in one community must interact to some degree. Coexistence of species within a community is the result of evolved interactions that have allowed a species to be present at a particular time and place (Roughgarden and

Diamond, 1986). Interactions between species that utilize similar resources within a community are expected to be strong, where competition for food and shelter can lead to displacement and death (Diamond and Case, 1986).

Resource partitioning within a habitat results in niche differentiation via morphological (genetic) changes or behavioral modifications in species.

Interspecific interactions will determine which species coexist in a habitat and which resources are utilized by these species.

Many researchers have modeled species coexistence (Hutchinson. 1959;

MacArthur and Levins. 1967: Schoener, 1974: Birch, 1979: Vance, 1984;

Chesson, 1986). From these works and empirical research utilizing a wide variety of organisms (see Diamond and Case, 1986), a num ber of mechanisms promoting coexistence of species in a community have been proposed.

Resource partitioning can occur between two species via habitat selection

(Cody. 1968: Arnold, 1972: Chew, 1981; Tokeshi and Townsend. 1987), recruitment patterns (Denslow, 1984: Naeem. 1988) and dietary specialization

(Grant. 1966, 1986; Fraser. 1976b; Cates, 1980; Brown. 1989). In the above studies coexistence was demonstrated by a separation in species niche requirements. Niche differentiation is required for coexistence when resources are limiting (Yodzis. 1986) and is often accomplished with evolutionary trade­ offs in body size and locomotory ability that promote diversity of resource use within a community (Kotler and Brown, 1988). Interference behaviors (Rathcke.

2 1976: Morse, 1980: Karieva, 1982) and non-limiting resources (Rathcke. 1976:

Schoener. 1982) can also allow coexistence.

When utilize similar food resources within a particular habitaL. differences in their temporal occurrence or habitat selection of specific sites to live within the larger area allow their coexistence (Arnold, 1972; Birch, 1979).

Temporal differences in habitat use promote coexistence because the foraging efficiencies of species differ and may generate variability in resource abundance

(Cody, 1968. Kotler and Brown, 1988. Brown, 1989). For example, in pierid butterflies, the distribution of species is more dependent upon the composition and abundance of the crucifer flora than upon interspecific competition (Chew,

1981). Here, the structural characteristics of the habitat provide refugia for coexisting species, allowing a spatial release from competitive interactions

(Birch, 1979).

Recruitment can promote coexistence (Chesson, 1986; Chesson and Case.

1986). Most species exhibit some degree of density-dependent recruitment with high densities new recruits at low adult densities. The timing of a larval dispersal phase in a species' life history will determine the degree to which that species can colonize a habitat. The relationship between the timing of species' dispersal phase and the duration of the open patch may explain why many species assemblages are composed of groups of species with different life histories (Naeem, 1988). Chesson and Warner (1981) propose that in addition to temporal differences in recruitment, overlapping generations are necessary for species coexistence. They suggest that differences In competitive abilities between species and mortality due to environmental variability allow one species to be favored at one time and other species at another. These non­ equilibrium theories postulate coexistence resulting from the possession of certain life history traits (Chesson and Case. 1986: Tokeshi and Townsend.

3 1987) where no species has a large net advantage over another.

Coexistence of many terrestrial species has occurred through dietary specialization. Morphological variation in bill or jaw size (Van Valen. 1965:

Grant. 1969) and body size (Fraser. 1976a) between coexisting species has been used to indicate whether organisms use different food types. This method has commonly suggested the occurrence of competition because it quantifies resource exploitation patterns and interactions between individuals.

Specialization on particular food types may be dictated by prey chemistry.

Monophagous and oligophagous insects preferred young leaves which were more easily digested rather than mature leaves, while polyphagous insects preferred mature leaves (Cates. 1980).

Behavioral aggression and interference promote coexistence of species by limiting access to resources by competitors (Morse. 1980: Strong, et al., 1984).

In wood rats (Dial. 1988) and stem-boring insects (Rathcke. 1976) aggressive encounters between individuals Inhibited resource use by other competitors and provided the necessary niche separation for coexistence. In some instances agonistic behaviors between species may result in severe injury and death of a subordinate. Coexistence can be maintained if inferior species have a larger fundamental niche than the dominant species (Morse, 1980).

Lastly, coexistence of species can occur when resources are not limiting and intraspeciflc competition is low. The potential for competition was high within a guild of stem-boring insects, but species interactions were infrequent because resources, particularly food and space were seldom limiting (Rathcke.

1976). Schoener (1968) described a similar scenario for Anolis lizards in the

Bimini Islands. He suggested that for similarly sized individuals to inhabit the same area, food m ust be abundant to allow increased overlap in resource use.

4 The coexistence of sessile species in aquatic environments has received

much attention; most studies are examples of preemptive or overgrowth

competition of clonal invertebrates (Schoener, 1983). The co-occurrence of

motile invertebrates in epifaunal and algal communities is well documented bui

the mechanisms allowing their coexistence is generally unknown. Freshwater snails show differences in diets (Fernandez, et al.. 1989). structure of the radulae (Blinn. et al., 1989: Hawkins, et al. 1989), feeding behaviors (Hawkins, et al., 1989) and habitat preferences (Fernandez, et al., 1989). These differences have been suggested as consequences of resource and habitat partitioning that allow coexistence. In the marine environment, snails and opisthobranchs also exhibit partitioning of resources. Here, variability in food preference (Radwin and Weils, 1968: Nybakken and Eastman. 1977; Shonman and Nybakken, 1978), structure of the radula (Bloom, 1976: Nybakken and

Eastman. 1977), habitat preference (Bloom, 1981) or temporal occurrence on the food resource (Yoshioka. 1986) provide explanations for the coexistence of these predators.

For other groups of marine, opisthobranch mollusks, very little work has been done on interspecific interactions. Nudibranchs are important inhabitants of hydroid colonies because they may control hydroid colony structure (Gaulin, et al. 1986). In particular, assemblages of aeolid and dendronatacean nudibranchs coexist in hydroid colonies, but mechanisms for this coexistence are unknown.

Hydroids colonies are conspicuous members of marine subtidal communities. They commonly grow on any hard substratum, alga or other organisms (L'Hardy, 1962; Pequegnat. 1964; Hagerman. 1966; Weis, 1968;

Hughes, 1975; Hayward. 1980; Boero, 1981; Seed. 1986; Boero and Fresi,

1986: Lambert and Laur, 1987). Hydroids approximate prairie grasses in the

5 structure they provide to motile epifauna and the attraction to consumers.

However, the relationship between co-occurring nudibranchs on colonies of

hydroids is not well documented (Todd. 1981). Previous studies (Clark. 1975:

Lambert. 1985: Harris. 1987: Todd and Havenhand, 1989) suggest that

nudibranchs may destroy fouling communities by their grazing activities.

Increases in nudibranch abundance and predation probably cause the rapid

turnover of local populations of hydroids by decreasing the hydroid colony's

abilities to filter settling larvae (Standing. 1976).

In the southern Gulf of Maine the hydroid Obelia geniculata (Linnaeus.

1758) often dominates the surfaces of Iaminarian kelps. Laminaria saccharina

(L.) Lamour and L. digltata (Huds) Lamour. Q. geniculata is a thecate hydroid

with occurring in the free-swimming medusoid generation.

The short-lived medusa (3-4 weeks in the plankton) is considered the dispersal

stage of the life cycle, but the hydroid may also be transported over long

distances as a fouling organism on ships, drift algae and swimming vertebrates

(Cornelius. 1982). The major recruitment period of Q. geniculata to kelp blades

occurs in late spring. The increase in abundance appears to be a combination

oflarval settlement and vegetative growth. O. geniculata is most abundant

during the summer months (June-August) while blooms also occur in autumn

(November) and late winter (March).

Six nudibranch species appear to have a strong association with Q.

geniculata as a prey source: Corvphella verrucosa (M. Sars. 1829).

Dendronotus frondosus (Ascanius. 1774). Doto coronata (Gmelin. 1791).

Eubranchus exiguus (Alder and Hancock. 1848). E. pallldus (Alder and

Hancock. 1842). and Tergipes tergipes tForskal. 1775) (Meyer, 1971; Clark.

1975: Franz, 1975: Todd. 1981: Lambert. 1985). This study focuses on four

. nudibranchs (Dendronotus. Doto. E. exiguus. and Tergipes) that exhibit

6 seasonal abundance maxima on Q. geniculata during late spring and summer.

Doto is most frequently found among the stolons. Eubranchus and Tergipes on the hydrocauli and Dendronotus Is often found clinging to the tops of hydrocauli and branches. The spatial distribution of nudibranchs within the colony appears to suggest a separation in their use of the hydroid colony as food and habitat. Thus, the location of a nudibranch within the colony results from its feeding behavior.

7 Species Descript ions

Obelia geniculata (Linnaeus. 1758) is a thecate, coelenterate hydroid with a cosmopolitan distribution in shallow waters (Calder, 1970; Cornelius. 1975:

1982). Q. geniculata is found almost exclusively growing on the blade surfaces oflaminarian kelps. Vertical hydrocauli (uprights) arise from stolons that are attached to the kelp surface. Hydrocauli are seldom branched and reach a height of 25 mm (pers. obs.). Cornelius (1975) reports heights to 40 mm.

Hydranths have a single whorl of 15-20 tentacles and are borne within smooth- rimmed, bell-shaped hydrothecae. Medusae are produced within gonothecae and are individually released at maturity. In this study, gonangia were found in all months except January and February although medusae were seldom seen in winter months (Lambert, unpub. data).

Dendronotus frondosus (Ascanius, 1774) is a dendronatacean nudibranch with a cosmopolitan distribution in the northern hemisphere. It is common in bays, fouling communities and shallow, subtidal habitats in association with hydroids. Juveniles have been found almost exclusively on colonies of thecate hydroids in the spring (Clark. 1975) whereas adults are associated with athecate hydroids. primarily Tubularia spp. (Swennen, 1961: Miller. 1961:

Todd. 1981). Dendronotus has a laterally compressed body with 4-8 pairs of branched and a narrow foot. The body is commonly whitish, whereas large individuals (>12 mm) may be marbled with red and brown. European researcher have reported the life span to be two years (Swennen. 1961: Miller.

1961) but Robillard (1970, Pacific Ocean) and Kuzirian (pers. comm.) feel that most individuals live about 1 year and die after spawning. Maximum body size is reported to be 100 mm (Thompson and Brown, 1984) with the majority of adults in the Gulf of Maine being closer to 50-60 mm.

8 Doto coronata (Gmelin, 1791) is a small dendronotacean nudibranch with an amphiatlantic distribution. In the spring and summer, it is commonly associated with thecate hydroids (Obelia spp., Dvnamena pumila. and

Sertularia spp.) (Clark, 1975). Doto has a short, squat, pale white-yellow body with 5-8 pairs of cerata. The cerata have circlets of red tubercles on their sides.

Maximum body size is usually 12 mm. Miller (1961) suggests that D. coronata has at least 2 generations per year whereas Clark (1975) states that individuals live 3-4 months and populations are not self-sustaining. Spawn has been found throughout the year in the North Sea (Swennen, 1961).

Eubranchus exiguus (Alder and Hancock, 1848) is a small, aeolid nudibranch with a boreal, amphiatlantic distribution. It commonly occurs on colonies of thecate hydroids (Obelia geniculata) growing epiphytically on algae

(Clark. 1975), in the shallow subtidal (Edmunds and Kress, 1969) and it is capable of living in brackish water (Swennen, 1961). E. exiguus has a small

(maximum body size = 8 mm), pale white-yellow body with up to 5 pairs of inflated, white-tipped, um-shaped cerata. The cerata and usually have 2-3 olive green-brown bands along their lengths. E. exiguus is an opportunist, specializing on thecate hydroids (Todd, 1981). It has multiple generations each year (Clark, 1975) with the majority of individuals spawning in the summer (Swennen, 1961),

Tergipes tergipes (Forskal, 1775) is a small, aeolid nudibranch with a boreal, amphiatlantic distribution. It is very common on colonies of Obelia geniculata on Laminaria spp. and Zostera sp. (Clark. 1975), and it is most abundant in early summer fTodd, 1981). Tergipes has a pale body with 7-8 cerata, a narrow foot and a long, thin tail. Usually, a red-brown streak is present on the dorsal surface between the cerata. Maximum body length is

9 about 8 mm. Tergipes has continuous recruitment throughout the year (Clark.

1975). Its generation time can be as short as 5 weeks (Swennen. 1961).

S tudy site

Animals were collected from a shallow (4-10m). subtidal kelp bed at Cape

Neddick, York. ME (43° 10 N. 70° 36 W). Water temperature fluctuated

seasonally with a minimum of 1°C in February and a maximum of 18°C in

August (Lambert, unpub. data). A permanent kelp bed has been present at the

tip of Nubble Island at Cape Neddick for at least 20 years (Harris, pers.comm.)

(Figure 1). The bottom is granitic rock to a depth of 12m and below that sand.

Laminarian kelps (Laminaria saccharina and L. digitata) dominate the canopy between depths of 3-12m. The understory is an assemblage of filamentous and foliose red and green algae. The dominant invertebrates present on primary substrata are mytillid bivalves and colonial suspension feeders (bryozoans. cnidarians and tunicates). The majority of kelp blade surfaces are covered with the hydroid Obelia geniculata and the ectoproct Membranipora membranacea

(Berman, et al.. in prep).

10 Figure 1. Location of the study site at Cape Neddick, York, ME (43° 10' N, 70° 36' W). Inset shows greater detail of the area at Cape Neddick and Nubble Island.

11 NB

100 m

to

0 50100 1 i i i Km

71 72 CHAPTER 1

THE EPIFAUNA ASSOCIATED WITH COLONIES OF THE HYDROID OBELIA

GENICtJLATA

INTRODUCTION

EpibloLic species relationships have been studied on a variety of marine

algal substrates: Fucus senratus (Stebbing. 1973: Seed, et al.. 1981),

Sargassum (Fine. 1970: Ryland, 1974). Posidonia (Boero. 1981: Casola. et al..

1987), Thalassia (Heck and Orth. 1980; Orth, et al., 1984), and laminarian

kelps (Sloane. etal.. 1957; L'Hardy. 1962: Foster. 1975). Macroalgae and seagrasses provide four primary resources exploitable by marine invertebrates:

1. surface area for sessile organisms for attachment: 2. shelter as either a permanent habitat for sedentary and sessile organisms or a temporary habitat for motile species; 3. sediment traps as detrital material accumulates within holdfasts and fronds: 4. food for algal herbivores and for grazers on microbial films and sessile epifauna (Hayward. 1980). On kelp blades, the primary resource provided is two-dimensional space. Space is best colonized by opportunistic species with short life spans and high growth rates (Seed and

O’Connor. 1981) because the physical parameters in these habitats produce a very dynamic community.

Hydroids are common epiphytes of laminarian kelps (L'Hardy, 1962;

Foster. 1975; Hayward, 1980). Once established, hydroids influence the settlement of many other organisms. Tunicate larvae (Standing. 1976: Schmidt.

1983), mussel larvae (Bayne. 1964, 1965) and caprellid amphipods (Keith.

1971) are attracted to hydroid colonies. These investigators suggest that the attraction of larvae to hydroid assemblages is attributed to a physical refuge provided by the colony understory. Certain hydroid colonies have also been

13 shown to inhibit sessile organisms from attaching to the substrate. Stolons and

hydrocauli of Obelia dlchotoma interfere with the cementing action of Balanns cyprids (Standing. 1976) and Hydractlnia echinala prevents settlement and encroachment by other organisms to space it occupies (Sutherland and

Karlson. 1977).

In the Gulf of Maine stands of the kelps Laminaria saecharina and L. digitata are relatively abundant in shallow, coastal subtidal habitats. The hydroid Obelia geniculata often dominates the surfaces of kelp blades and provides a habitat for many motile epifauna. This study examines the relationship between Q. geniculata and its non-nudibranch epifauna. The physical structure provided by the hydroid community on kelp blades, the seasonal variation of epifauna and the association between the hydroid and the epifauna are described for a study site in the southern Gulf of Maine.

14 MATERIALS AND METHODS

Collection of Animals

Blades of Laminaria spp. were collected from the kelp bed at Cape

Neddick, ME. Five kelp blades with high percent cover (50-100%) of Q. geniculata were selected twice monthly from June. 1987 - May. 1989. Bare kelp blades (n = 25) were collected during the summer of 1989 (May - August).

Each kelp blade was individually placed in a plastic bag underwater and then brought to the lab. In the lab, each kelp blade was placed In a 0.5 mm mesh bag and stored in a recirculating seawater table until processed, within 1-3 days. Processing consisted of placing each kelp blade into 8% MgCl 2 for 20 minutes to narcotize all motile epifauna. All organisms were sorted in sea water and identified to species or taxonomic group (copepods. gammaridds, caprellids, flatworms. ostracods. polychaetes) while alive with a dissecting microscope.

Observations on the general life stage (Juvenile, adult) of the organisms were also recorded.

A permanent record of the kelp blade and all attached epifauna was obtained by tracing the outline of the kelp blade onto paper. Colony sizes of Q. geniculata and all ectoprocts were obtained by poking holes into the kelp blade with a pencil onto the tracing. Afterwards, all areas (cm^) were digitized with an Apple II Plus computer and graphics tablet. Hydrocauli were counted to quantify the density of the hydroid community (#/cm^) and the height (mm) of a random subsample (15%) of all hydrocauli were measured with a dissecting microscope to identify a vertical component of the community.

15 Statistical Analysis

Multiple regression techniques were used to analyze the community characteristics and define the most important components of the O. geniculata community correlated to the abundance of epifauna. Abundance data were transformed, square root (x+0.5) or natural log (x+1), to adjust for nonnormality and heterogeneous variances (Zar, 1984). Bonferoni adjustments were made to accomodate multiple statistical tests. Abundances of each species were standardized ((abundance/hydroid colony area) * 100) to the size of the hydroid colony to adjust for different sized sampling units (Seed, et al.. 1981).

16 RESULTS

Species Patterns of Abundance

Hvdroid Abundance

The size (cm^) of the kelp blades between the two years of the census were similar (F = 0.111. df = 1, 228. p = 0.739). Generally, colonies of Q. geniculata recruited heavily in the spring, showed high abundances in the spring and summer due to recruitment and vegetative growth before dying back in the fall.

A minor recruitment event also occurred during late fall (early November).

Colony size of Obelia geniculata on Laminaria blades collected during the entire census period averaged 161.97 cm2 (SE = 8.68). They altered the physical characteristics of the smooth, flat blades by providing discontinuities

(stolon network) and a vertical component (hydrocauli) to the kelp surface.

Overall, these colonies averaged 8.2 mm (SE = 0.2) in height with a density of

10.6 (SE = 0.3) hydrocauli per cm2, but there were differences in these parameters between years (Figure 1-1). To determine if components of the hydroid colony differed within a year, the year was divided into 4 seasons: summer (June-August), fall (September-November), winter (December-February) and spring (March-May). Seasonal patterns of Q. geniculata colonies revealed a spring peak and a fall/winter low in size, density and average height of the hydroid colony (Figure 1-2). This pattern was seen for all parameters except colony size during year 1 (1987-1988). In the first year colony size was largest in the winter (157.53 cm2) and smallest in the summer (83.00 cm2). This may be attributed to the absence of an August sample in 1987 when hydroid colonies in the kelp bed are normally large and luxuriant.

17 Abundances of Epifauna

Seventeen taxa representing 6 phyla were found frequently (in >10°/o of samples) in the hydroid colony: 10 were motile microfauna. 4 were predatory nudibranchs (Chapter 2) and 3 were sessile species (Table 1-1).

Motile epifauna were usually present on all blades collected during the census (Figure 1-3A); representatives of each taxon were collected in most months. Overall, abundances of each taxon were less than 20 individuals per

100 cm^ of kelp surface (Figure 1-3B). The high overall abundance of Mvtilus edulis is attributed to heavy recruitment during the summer. The motile epifauna on kelp blades without Q. geniculata was more sporadic and abundances were much lower (Figure 1-3B), suggesting that the hydroid colony has a positive influence on the recruitment of these organisms. The seasonal abundance patterns of each common taxon are described below.

Gammarid amphipods were present on 91.7% of all kelp blades collected during the census period. Abundances peaked in the spring of each year (May), declined through the summer and were low in the fall and winter (Figure 1-4A).

The small peaks of abundance in the early winter may represent reproduction of the year's first cohort (Donn. 1983).

Both calanoid and harpacticoid copepods were found in samples, however, harpacticoids were a very minor component and therefore all copepod abundances were pooled. Copepods were absent from only 2.2% of kelp blades collected. Highest abundances of copepods were found in the winter and spring with lows in the summer (Figure 1-4B). Peaks in abundance occurred at regular intervals of 6-8 weeks except through the summer.

Caprellid amphipods were abundant in the spring and sum m er and decreased in July of each year of the census (Figure 1-4C). They were present

18 on 59% of kelp blades collected. In the spring of 1988 a large number of juvenile caprellids recruited to the hydroid colonies on kelp blades. 90% of all

caprellids were newly settled. This peak was followed in May, 1988 by a large

number of adults carrying ova. One-to-three peaks in abundance occurred in

the spring and summer with a decline in overall abundance from late summer

through winter.

Mvtilus edulis was the most abundant organism on kelp blades covered

with Q. geniculata. Mussels were present on 95.7% of all kelp blades. A large

pulse of recruitment of Mvtilus occurred in the summer of each year of the

census and a smaller peak occurred in the fall of 1988 (Figure 1-5A). All

mussels collected were less than 1 mm in valve length suggesting that pediveligers use the hydroid colony as an area for primaiy settlement (Bayne.

1964) before secondarily settling into adult habitats.

Lacuna vincta showed highest abundances in the winter and spring while being essentially absent in each summer (Figure 1-5B). Lacuna was found on

87.4% of kelp blades collected and was found in each month of the census except September. 1987. Egg masses were found attached to the surface of kelps in winter (December-Februaiy) only. Winter samples were dominated by snails with shells less than 1 mm in length. All snails found at other times were larger (>2 mm).

Mites were common epifauna on kelp blades covered with Q. geniculata: they were present on 93% of kelp blades. Although numerically higher abundances of mites were found in the spring and summer of each year, abundance peaks occurred at regular intervals of 10-12 weeks throughout the year (Figure 1-6A).

Ostracods were collected from 87.8% of all kelp blades. Although the highest abundances were found in the late fall and winter of each year, they

19 had regular peaks in abundance at 8-10 weeks intervals throughout the census (Figure 1-6B).

Semibalanus balanoides settled on kelp blades in the spring of each year

and abundantly in 1989 (Figure 1-6C). Barnacles were present only during this

period of high cyprid settlement. There was no distinct distribution pattern to

the barnacles on the kelp blade surface (sensu Standing. 1976). Barnacles

were present on 9.6% of kelp blades collected during the two years.

Turbellarian flatworms were rare in terms of abundance and were present

on only 29.6% of kelp blades, being most abundant in the spring and summer

(Figure 1-7A). Peaks in abundance occurred regularly (every 10-12 weeks)

during the first 14 months of the study, but flatworms were essentially absent

from September. 1988 - February. 1989.

Syllid polychaetes were common during the summer of each year (Figure

1-7B). Although other polychaetes (cirratulids. terebellids and polynoids) also occurred, syllids dominated all samples when polychaetes were present.

Juvenile syllids (<5 segments) were found abundantly in the summer with a smaller pulse in early winter (December). Syllid polychaetes were present on

47% of kelp blades collected during the census.

Two species of ectoproct were present on kelp blades: Electra pilosa and

Membranipora membranacea. The abundance of ectoprocts was low in the first year of the study (June, 1987 - May. 1988). when only Electra occupied space on kelp blades (Figure 1-8A). In the second year (June, 1988 - May. 1989) abundances of ectoprocts on kelp blades increased by an order of magnitude and was solely attributable to the presence of Membranipora membranacea

(Figure 1-8B). The abundance of Electra on kelp blades between the two years of the census was similar (year 1 = 0.61 cm^ (SE=0.14), year 2 = 1.00 cm^

(SE=0.34); F=2.459, df=l. 228, p=0.118). Electra was found on 46.1% of all

2 0 kelp blades and occupied 0.81 cm2 (SE=0.19) per 100 cm2 kelp surface.

Membranipora was first collected in July. 1988 and was present on 67.5% of

kelp blades during the second year of the census while being absent in year 1.

Membranipora occupied 7.25 cm2 (SE=1.03) per 100 cm2 of kelp surface.

Factors Affecting Species Abundances

Motile Epifauna

Abundances of all motile epifauna except flatworms were positively associated with the size (area) of the hydroid colony (Table 1-2 to 1-4). The relationship between epifaunal abundance and hydroid colony size was not dependent on kelp blade size although the size of a hydroid colony is related to the size of the kelp blade (t=8.403, p<0.001). The size of the kelp blade was negatively associated with the abundance of syllid polychaetes (Table 1-4C).

These data suggest that hydroid colony size in the island influencing recruitment of motile epifauna.

Colonies of Q. geniculata altered the surface of the kelp by providing surface discontinuities (stolons and hydrocauli) and a vertical component.

Denser colonies of Q. geniculata had higher abundances of gammarid amphipods (Table 1-2A), copepods (Table 1-2B). Lacuna (Table 1-3B), mites

(Table 1-3C) and flatworms (Table 1-4B) while syllid polychaetes were found on colonies with lower densities of hydrocauli (Table 1-4C). Three taxa

(gammarids, caprellids and mites) showed a positive relationship with colony height (Table 1-2A, 1-2C, 1-3C respectively) and Mvtilus (Table 1-3A) had a negative relationship. This suggested that the physical structure provided by colonies of Q. geniculata affects recruitment of motile epifauna.

21 Sessile Organisms

Barnacle abundance was positively associated with only one parameter of the community, average height of the hydroid colony (Table 1-5A). This suggested that cyprids selected substrates which were clear but provided some structure or an understory. The abundances of Electra and Membranipora showed very different patterns. Electra was not associated with any physical parameter of the community, while Membranipora was positively related to the size of the kelp blade and negatively associated with the colony size of Q. geniculata and density of hydrocauli (Table 1-5B). This suggested that larvae of

Membranipora settle on the largest substrate available that is clear of other sessile epifauna.

22 DISCUSSION

Abundance peaks of most species in the epifaunal community on

Laminaria spp. blades covered with Obelia geniculata occurred in the spring

and summer months (Figure 1-4 to 1-7). Although seasonality in reproductive

patterns of organisms has been suggested to be the primary cause of variation

observed in many systems (Sutherland and Karlson, 1977; Osman. 1978;

Edgar. 1983b; Bros. 1987a), community structure (complexity) also plays an

important role (Coyer. 1984; Bros. 1987b; Hall and Bell. 1988). The paucity of

organisms on bare kelp blades collected during the summer of 1989 (Figure 1-

1B) suggested that seasonality in reproductive patterns was a minor force

structuring the kelp/hydroid community when compared to kelp blades covered

with (2. geniculata.

Motile Epifauna

Habitat complexity increased the abundance of many organisms on algal

and artificial substrates (Edgar, 1983c; Bros. 1987b; Hall and Bell, 1988;

Lambert, unpub. data). Although Obelia is a predator and can act as a larval

filter (Standing, 1976), invasion by epifauna can occur by larvae overwhelming

the hydroid colony (differential survival during settlement) or by active

behavioral mechanisms (Meadows and Campbell. 1972). Once organisms have

successfully recruited to a kelp blade with Q. geniculata. they benefit by refuge

from predation by fish, food resources or physical structure for attachment

(Vimstein and Howard, 1987).

Structural complexity interferes with prey capture rates of fish (Heck and

Orth. 1980; Martin, 1988). The wrasse Tautogolabrus adsoersus (cunner) is a

common predatory fish on epifaunal communities in the southern Gulf of Maine

(Martin. 1988; pers. obs.) and recruits heavily to shallow, subtidal habitats in

July and August (Levin, unpub. data: pers. obs.). Stomach contents of Juvenile

23 cunner from Cape Neddick were dominated by crustaceans and juvenile mussels (Clark, unpub. data). The decrease in abundances of crustacean epifauna on Laminaria blades during late summer months may be due to high predation by cunner. Martin (1988) showed that caprellid amphipods are afforded greater protection from cunner by filamentous rather than flat-bladed algae. Hall and Bell (1988) manipulated the sessile community of blades of

Thallassia and showed that abundances of crustacean epifauna were positively correlated with the amount of epiphytic algae. They suggested these habitats served as a predator refuge from fish and large invertebrates.

The herbivorous snail Lacuna vincta is often found on laminarian kelps

(Fralick, et al.. 1974). While bare kelp blades provide an abundant food resource for Lacuna, they do not afford any protection from predation. Very few snails were found on bare kelp blades (Figure 1-1B) compared to kelp blades covered with Q. geniculata. Snails can find protection from fish predation within a hydroid colony and still retain access to the rich food resource.

Active selection for food or physical structure for attachm ent may be an alternative mechanism structuring the epifaunal community within colonies of

Q. geniculata on Laminaria. Caprellid amphipods select substrates that can be easily grasped (Bynum, 1978; Caine. 1978) and increase their feeding effectiveness (Keith. 1971; Caine. 1978). As suspension and detrital feeders, caprellids may capture food more easily and find more food in filamentous habitats because food particles may be slowed and suspended longer in the colony understory. The majority of caprellids found on kelp blades covered with

Q. geniculata in the spring and summer were juveniles (<3 mm) whereas Martin

(1988) at the same site found lower abundances of caprellids in phytal communities in the summer than the winter. Juvenile caprellids are translucent-white and possibly more cryptic in Obelia colonies. Also they may

24 find hydroid hydrocauli easier to grasp than the relatively thicker thalli of

filamentous algae, thus a hydroid community provides a "better" habitat for

these small animals.

Mvtilus edulis is a suspension feeding bivalve whose primary settlement

occurs into filamentous communities (hydroids, bryozoans and algae) (Bayne,

1964; 1965). Much greater abundances of mussels settled onto kelp blades covered with £>. geniculata than onto bare kelp blades (Figure 1-1). In addition, mytilids settled more heavily on plexiglas panels with artificial vertical structure

(hydroid mimics) than bare panels (Lambert, unpub. data). Bayne (1964) suggested that settlement of pediveligers may include a tactile recognition of suitable substrates. It is likely that this mechanism occurred in the hydroid community. Also, hydroids alter the physical conditions and hydrodynamics at the surface level (Williams. 1964; Dean, 1981) of the substrate they colonize.

The modified habitat may facilitate food particle capture by Juvenile mussels by maintaining particles In suspension, slowing particles through the colony understory and by producing eddies that extend the time in which food is available for capture.

Sessile Epifauna

Barnacles (Semibalanus balanoides) settle heavily to natural and artificial substrates in the southern Gulf of Maine during the spring (March-April) (Harris and Irons. 1982; Lambert. 1985). Higher abundances of barnacles were present on lower sides of opaque surfaces (Verny, unpub. data). In this study, barnacles were found only on blades of Laminaria spp. from April-June (Figure

1-4C). Kelp blades provide a flat, opaque surface for settlement but the flexibility of the frond and movement by surge and waves must detach barnacles. The settlement of barnacles onto kelp blades appears to be a

25 consequence of heavy recruitment from the plankton; cyprids that are competent to metamorphose attach to almost any rigid, opaque surface.

The distribution of barnacles (3- balanoidesl on kelp blades covered with

O. geniculata showed no discemable pattern. Attached barnacles were found on portions of the kelp lacking Q. geniculata and In the center of colonies where hydrocauli are tallest. At times barnacle tests were also atop stolons. Standing

(1976) working with Balanus crenatus showed that barnacles settle more abundantly on surfaces without . He suggested that hydrocauli and stolons block contact between the cyprid and the settling surface. B. crenatus has a calcareous base, while 3 . balanoides has a membranous base. The membranous base may provide flexibility to 3- balanoides that reduces the interference effects of hydrocauli and stolons and allows settlement within the hydroid colony.

Prior to 1987. Electra pilosa was the only encrusting ectoproct on blades of Laminaria spp.. Membranipora membranacea was first observed in the summer of 1987 (Berman, et al., in prep.). Since 1987 the abundance of ectoprocts on Laminaria has increased by an order of magnitude (Figure 1 -6) and is solely due to the presence of Membranipora.

The presence of Electra on kelp blades was not related to any parameters i of the kelp/hydroid community. Electra is found much more abundantly on the red alga Chondrus crlspus (Berman, et al., in prep.; pers. obs.). Kelp blades are a secondary habitat for Electra. The area covered by Membranipora on kelps is positively associated with the size of the kelp blade but negatively related to hydroid colony size and density of hydrocauli in the colony (Table 1-5B). It is not surprising that colony size of Membranipora is related to kelp blade size, but larger colonies of Membranipora are on kelp blades with less Q. geniculata.

This phenomenon suggested that either the cyphonautes larvae of

26 Membranipora at settlement actively select kelp surfaces without O. geniculata or if no selection by larvae occurs, that Q. geniculata prevents settlement of

Membranipora. possibly by feeding on larvae as they swim into the colony fsensu Standing. 1976). Pre-settlement community structure seemed to create the observed patterns because once settled. Membranipora appears to overgrow all other organisms (Berman, et al., in prep.; pers. obs.). Q. geniculata does have a temporary vertical refuge from overgrowth by Membranipora but eventually even hydrocauli are covered by Membranipora (pers. obs.). Although

Membranipora appears to be a dominant competitor (sensu Jackson. 1977). the impact of this new species on benthic communities in the Gulf of Maine is still unknown and requires further attention.

27 Figure 1-1 Analysis of parameters of hydroid colonies on laminaria spp. blades between the two years of the census (year 1: June, 1987 - May, 1988, year 2: June, 1988 - May. 1989).

28 F = 5.270 p = 0.040 Obelia area df = 1. 228

100 -

F = I 1.390 Obelia density p = 0.001 df = 1, 220

F = 0.329 Obelia height p = 0.004 df = 1,220

Year 1 Year 2

2 9 Figure 1-2. Mean (+SE) seasonal variation in size, density and mean height of Obelia geniculata colonies on Laminaria spp. blades collected at Cape Neddick. ME. Horizontal lines indicate significant differences (p<0.006) between seasons for each parameter as determined by Tukey-Kramer HSD multiple comparison test.

30 400 Obelia area

300

200

100

0 Vinter Spring Fall Summer Summer Spring Fall Vinter Obelia density 15

10

5

0 Spring Summer Winter Fall Spring Vinter Fall Summer Obelia height 10

•

6

4 •

2

0 Spring Vinter Summer Fall Spring Summer Fall Vinter 1987-88 1988-89

31 Figure 1-3 (A) Percent occurrence and (B) mean abundances (+SE) of motile epifauna on kelp blades fLaminaria spp.) covered with Qbcl.ia geniculata (n=230) and bare kelp blades (n=25).

32 In x spp / 100 cm^ Kelp Percent Occurrence Figure 1- Mean abundances of crustacean epifauna (+SE) at Cape Neddick, ME on blades of Laminaria spp. covered with Obelia geniculata. No samples collected in August, 1987.

34 Mean # / 100 cm^ Hydroid Area 100 200 120 140 160 180 20 40 60 100 120 80 100 125 200 150 175 GO 80 25 50 75 ■ - - - J J A S O N D JFUAHJJASON D JF M A M 97 98 1989 1988 1987 m ? ? T t » t i y? i ? i 35 Caprel Caprel lids Copepods GammoridAmphipods Figure 1-5 Mean abundances (+SE) of molluscan epifauna at Cape Neddick, ME on blades of T-aminaria spp. covered with Obelia geniculata. No samples collected in August, 1987.

36 Mean * / 100 cm^ Hydroid Area 2000 1000 - 2500 3000 50 - 1500 100 125 200 150 - 500 175 25 - 25 50 75 - - JJASON D J F M A M J J A SONDJF M A M 97 98 1989 1988 1987 A

Figure 1-6. Mean abundances (+SE) of mites, ostracods and barnacles at Cape Neddick. ME on blades of Laminaria spp. covered with Qbelia geniculata. No samples collected in August, 1987.

38 Mean * / 100 cm^ Hydroid Area 20 40 80 - 80 80 H 80 * i t i i i i i i i i i T r i t i i i i i r i i i r r i i i i i i i i i i i t i i i r T i i i i i < i i i i i i i i i t t i i i itt i *i 0 - * * ON O N O O "*0** JJAS0N0J FJJAS0N0J n AH J J AS • N DJF H A M 9 7 98 1989 1988 1987 I TT 1 ft^t tft t f Tf tt t t f f ^ T t I T Tf T tT T1 » t f "IT t i t f f TTTTf 39 BjQrnocles Mites ' I i ‘ 1-1 t

Figure 1-7 Mean abundances (+SE) of turbellarian flatworms and syllid polychaetes at Cape Neddick, ME on blades of Laminaria spp. covered with Qbelia geniculata. No samples collected in August. I OCT

40 Mean * / 100 cm2 Hydroid Area 0 2 10 - 30 40 - 40 0 * 50 10 ■ • - JJASONDJ FtlAMJJASON 97 98 1989 1988 1987 41 Flotworms SyTlids M A M Figure 1-8. Mean abundance (+SE) of ectoprocts at Cape Neddick. ME on blades of Laminaria spp. covered with Obelia geniculata from A) June. 1987 - May. 1988 and B) June. 1988 - May. 1989. Abundances (cm^j standardized to area of the kelp blade. No samples collected in August. 1987.

42 CM Area (cm 20 30 30 40 40 50 50 10 0 o 2 3 4 5

0 o — o — o i e in ex in — JJASONDJFMAM J J AO S ND JF MA M 7 8 9 1 9 8 9 1 8 8 9 1 0 N 0 N 0 M I

O N O M - I N — '■-NO-"N — n 43 **N-N0 II ll 0 N - N 8 8 9 1 Electra □ Membra ■ 1 1 . Liilii Electra □ - - Membranlpora

n

0 « 0 (M O

Table I -1. Taxa commonly found on blades of Laminaria spp. covered wilh Qbelia geniculata.

A nnelida Polychaeta Syllids

A rthopoda A rachnida M ites C ru stacea CaprelUd amphipods Copepods Gammarid amphipods O stracods Semibalanus balanoides

E ctoprocta Electra pllosa Membranlpora membranacea

M ollusca Bivalvia Mvtlius edulis G astropoda Dendronotus frondosus Doto coronata Eubranchus exiguus Lacuna vincta Tergipes terglpes

N em atoda

Platyhelmlnthes Turbellarian flatworms

44 Table 1-2. Multiple regression analyses of parameters of the kelp/hydroid community. Model: In species abundance = constant + sqr kelp area (cm2) + sqr hydroid area (cm^) + sqr density of colony (#/cm2) + height of colony (mm).

Variable C oefficient (SE) t p

A. Gammarid amphipods

C o n stan t -2.033 (0.611) -3.328 0.001

Kelp area -0.016 (0.014) -1.166 NS

Otaelia area 0.106 (0.024) 4.504 <0.0001

Obella densitv 0.618 (0.152) 4.076 0.0001

Obelia height 0.168 (0.039) 4.348 <0.0001

B. C ooepods

C onstant -1.212 (0.574) -2.110 0.036

Kelp area 0.014 (0.017) 0.813 NS

Obelia area 0.074 (0.020) 3.668 0.0003

Obelia densitv 0.812 (0.131) 6.178 <0.0001

Obelia height 0.017 (0.032) 0.520 NS

C. Caprellid amphipods

C o n stan t -46.045 (38.819) 1.363 NS

Kelp area -1.390 (0.755) -1.841 NS

Obelia area 3.285 (1.309) 2.509 0.013

Obelia densitv 0.737 (8.390) 0.088 NS

Obelia heieht 6.872 (2.133) 3.220 0.002

45 Table 1-3. Multiple regression analyses of parameters of the kelp/hydroid community. Model: In species abundance = constant + sqr kelp area (cm^) + sqr hydroid area (cm^) + sqr density of colony (#/cm^) + height of colony (mm).

Variable Coefficient (SE) t p

A. Mvtilus edulis

C o n stan t -1.844 (9.306) -0 .198 NS

Kelp area 0.290 (0.208) 1.394 NS

Obelia area 2.036 (0.360) 5.6 4 9 <0.0001

Obelia densitv -1.993 (2.309) -0.863 NS

Obelia height -1.288 (0.587) -2.193 0.029

B. Lacuna vincta

Constant -1.083 (0.605) -1.790 NS

Kelp area -0.006 (0.013) -0.414 NS

Obelia area 0.086 (0.023) 3 .6 6 8 0 .0 0 0 3

Obelia densitv 0.682 (0.150) 4.548 <0.0001

Obelia height -0.002 (0.038) -0.039 NS

C. Mites

C onstant -2.551 (0.784) -3.253 0.001

Kelp area 0.033 (0.017) 1.864 NS

Obelia area 0.079 (0.030) 2.601 0.010

Obelia densitv 0.710(0.194) 3.653 0 .0003

Obelia height 0.207 (0.050) 4.146 <0.0001

46 Table 1-4. Multiple regression analyses of paramaters of the kelp/hydroid community. Model: In species abundance = constant + sqr kelp area (cm2) + sqr hydroid area (cm2) + sqr density of colony (#/cm2) + height of colony (mm).

Variable Coefficient (SE) t p

A. O stracods

C o n stan t 0.524 (0.581) 0.901 NS

Kelp area 0.007 (0.013) 0.504 NS

Obelia area 0.149 (0.023) 6.628 <0 .0 0 0 1

Obelia densitv 0.068 (0.144) 0.473 NS

Obelia heieht -0.028 (0.037) -0.759 NS

B. Flatw orm s

C o n stan t -1.335 (0.874) -1.528 NS

Kelp area -0.009 (0.020) -0.482 NS

Obelia area 0.017 (0.034) 0.490 NS

Obelia densitv 0.436 (0.217) 2.022 0.044

Obelia heieht 0.078 (0.055) 1.421 NS

C. Svlllds

C onstant 0.866 (0.466) 1.859 NS

Kelp area -0.030(0.010) -2.852 0.005

Obelia area 0.139 (0.018) 7.727 <0.0001

Obelia densitv -0.485 (0.116) -4.199 <0.0001

Obelia heieht 0.045 (0.029) 1.545 NS

47 Table 1-5. Multiple regression analysis of paramaters of the kelp/hvdroid community. Model: In species abundance = constant + sqr kelp area (cm2) + sqr hydroid area (cm2) + sqr density of colony (#/cm2) + height of colony (mm).

Variable Coefficient (SE) t p

A. B arnacles

C onstant -1.074 (0.715) -1.502 NS

Kelp area 0.017 (0.016) 1.090 NS

Obelia area -0.029 (0.028) -1 .055 NS

Obelia densitv 0.234 (0.177) 1.319 NS

Obelia height 0.162 (0.045) 3.591 0.0 0 0 4

B. Membranioora

C onstant 7.692 (2.009) 3 .8 2 8 0.0002

Kelp area 0.090 (0.045) 2 .0 1 7 0 .0 4 5

Obelia area -0.237 (0.078) -3 .0 4 7 0 .0 0 3

Obelia densitv -0.866 (0.499) -1.738 NS

Obelia height -0.086 (0.127) -0 .678 NS

48 CHAPTER 2 .

DISTRIBUTION AND ABUNDANCE OF NUDIBRANCHS WITHIN COLONIES

OF THE HYDROID OBELIA GENICULATA

INTRODUCTION

Colonization processes of epibiotic organisms have been studied on many

algal substrates (Fine, 1970; Stebbing. 1973; Heck and Orth. 1980; Boero,

1981; Casola, et al.. 1987). Macroalgae provide four primary resources for exploitation by marine organisms: surface area, shelter, sediment traps and food (Hayward. 1980). On kelp blades the major limiting resource is two-

dimensional living space. Space can be colonized best by opportunistic species with relatively short life spans and high growth rates (Seed and O'Connor, 1981) because the physical parameters of these habitats produce a veiy dynamic community. Some predators show a very close affinity to particular prey inhabiting the algal fronds, but little work has been done describing these relationships (Seed. 1986).

Hydroids are conspicuous epifauna on laminarian kelps. Once established, a hydroid colony influences the settlement of many organisms, particularly predators (Harris, 1973; Todd. 1981). The association between nudibranchs and their food resource is unique because the nudibranchs are generally specialized consumers (Braams and Geelen, 1953; McBeth. 1971).

They are also partial predators (Harvell, 1984) with the abundance and availability of their prey often determining their reproductive potential and survivorship (Thompson, 1964; Harris. 1973). Metamorphosis of nudibranch larvae is stimulated by cues associated with adult food and habitat (Todd. 1981:

Burke. 1983; Hadfield. 1984). Nudibranchs accumulate within a hydroid colony, and potentially overwhelm it (Lambert. 1985; Todd and Havenhand.

49 1989). although this phenomenon is not verified. Often, many nudibranch species live within a single hydroid colony (Kuzirian. 1979; pers. obs.).

In the Gulf of Maine stands of the kelps Laminaria saccharina and L. digitata are relatively abundant in shallow subtidal habitats. The hydroid

Obelia geniculata often dominates the surfaces of kelp blades and harbors 4-5 nudibranch species. This study examined the association between Q. geniculata and its nudibranchs. The seasonal variation of nudibranchs and the association between the hydroid and the nudibranchs are described for a study site in the Gulf of Maine. The physical structure provided by the hydroid on kelp blades was described previously (Chapter 1).

50 MATERIALS AND METHODS

Collection of Animals

Colonies of Obelia geniculata were collected from the kelp bed at Cape

Neddick, ME twice monthly from June, 1987 to May. 1989. Five kelp blades

fLaminarla saccharina and L. digitatal with high coverage of O. geniculata (50-

100%) were selected during each sampling period. Bare kelp blades (n = 25) were collected during the summer of 1989 (May - August. 1989). Each kelp blade was placed individually in a plastic bag underwater and then brought to the lab. In the lab. each kelp blade was placed in a 0.5 mm mesh bag and stored in a recirculating seawater table until processed, within 1-3 days.

Processing consisted of placing each kelp blade into 8% MgCl2 for 20 minutes to narcotize all motile epifauna. Nudibranchs were sorted alive, in sea water, identified to species with a dissecting microscope and sized to the nearest mm.

Presence of juvenile nudibranchs was used to determine major recruitment periods of each nudibranch species.

A permanent record of the kelp blade and all attached epifauna was obtained by tracing the outline of the kelp blade onto paper. Colony sizes of Q. geniculata were obtained by poking holes into the kelp blade with a pencil onto the tracing. Afterwards, all areas (cm^) were digitized using an Apple II Plus computer and graphics tablet. Hydrocauli were counted to quantity the density

(#/cm^) of the hydroid community and the height (mm) of 15% of all hydrocauli was measured with a dissecting microscope to identify the vertical component of the community.

51 Statistical Analysis

Multiple regression techniques were used to analyze the community characteristics and deline the most important components of the Q. geniculata community. F-tests were used to determine if abundances of nudibranchs were related to components of the hydroid community (Model: nudibranch abundance = constant + kelp area + hydroid area + density of hydrocauli in colony + height of colony). Nudibranch abundances were square root (x+0.5) transformed to adjust for non-normality (Zar. 1984). A negative binomial test was performed and an index of dispersion (variance-to-mean ratio) was calculated to test whether species of nudibranchs aggregate with respect to conspeciflcs (Krebs. 1989).

52 RESULTS

Seasonal Patterns in Abundance of Organisms

Hvdroid Abundance

Colonies of Obelia geniculata recruited heavily to kelp blades in the

spring, showed high abundance in the spring and summer due to recruitment

and vegetative growth and died back in the fall and winter. Monthly collections

of colonies of Q. geniculata revealed a spring peak and a fall/winter low in

colony size, density of hydrocauli and average colony height (Chapter 1).

Nudibranch Abundance

Six species of nudibranchs were found on colonies of Q. geniculata during

the two year census: Corvphella verrucosa. Dendronotus frondosus. Doto

coronata. Eubranchus exiguus. E. pallidus and Tergipes tergipes. Both Q. verrucosa and E. pallidus were very infrequent; they .were present on only 7.8% and 3.0% of colonies sampled, respectively and thus were excluded from further analysis. Overall, nudibranch abundance was highest between March and

September. This abundance peak coincided with peaks in abundance of Q.

geniculata on blades of Laminaria spp. suggesting that nudibranch abundance patterns track their prey populations. On bare kelp blades a total of only four nudibranchs were found on three of the 25 bare kelp blades sampled. All nudibranchs were immature Tergipes (body size <2 mm).

Generally, nudibranchs were found on kelp blades covered with O. geniculata: 96.5% of all kelp blades collected had at least 1 species of nudibranch present (Figure 2-1). However, it was also uncommon to Find all four species of nudibranchs on a single kelp blade (7.4%). Tergipes was present on most kelp blades with Q. geniculata. Kelp blades with a single species of nudibranch (93.4%). all two species groupings and most three species

53 groupings (98.4%) contained Tergipes. suggesting that Q. geniculata on blades

of Laminaria spp. is a primary habitat for Tergipes.

Population growth of nudibranchs in colonies of O. geniculata on

Laminaria spp. occurs via recruitment processes only. Recruitment of juveniles was continuous throughout the census period (Figure 2-2). Very few adults were present in the total population sampled and cohorts of any nudibranch species were impossible to follow.

Species

Dendronotus frondosus was present on 57% of all kelp blades collected

(Figure 2-1, 2-3) and in all months except October, 1987 (Figure 2-4A). A large index of dispersion supported by a negative binomial test (Krebs, 1989) (Table

2-1) suggested that Dendronotus aggregated in colonies of Q. geniculata.

Dendronotus had continuous recruitment of juveniles throughout the year and major peaks In abundance occurred twice, June. 1987 and May, 1989

(Figure 2-4A). Very small nudibranchs (<2 mm) were present throughout the census (Figure 2-5A,B) and comprised 87.0% of all Dendronotus collected

(Figure 2-6). Mature Dendronotus (>12 mm) were found in the winter of both years (Figure 2-5E), but only 0.6% of the Dendronotus were adults (Figure 2-6).

Egg m asses were found infrequently on kelp blades. A total of six egg masses were found during the census: one in June. 1987. 4 on one kelp blade in

December, 1987 and one in December, 1988. The presence of spawn either coincided with or preceded peaks of recruitment. The pattern of habitat use by

Dendronotus suggested that colonies of Q. geniculata on Laminaria spp. are a nursery habitat for Dendronotus.

The pattern of recruitment exhibited by Dendronotus was significantly related to the density of hydrocauli and the height of the colony (F= 10.498, df=4, 225. pcO.OOl) (Table 2-2A).

54 Doto coronata was present on 25% of kelp blades (Figure 2-1. 2-7A) and in

all months except September and October. 1988 (Figure 2-4B). Individuals of

Doto aggregated in colonies of Q. geniculata (Table 2-1).

Doto was common in colonies of Q. geniculata on Laminaria spp. blades

during the summer and winter of both years (Figure 2-4B). Two primary peaks

in abundance of Doto were evident in the summer and winter of each census year; these pulses were dominated by small, immature nudibranchs (<4 mm)

(Figure 2-8 A-C). The majority (85.0%) of Doto present on kelp blades was immature (<4 mm) (Figure 2-6B). Large, mature Doto were collected throughout the year (Figure 2-8D) but were a minor component of the population (15.0%). Ribbons of spawn were found in the spring and summer of

1988 and in December. 1987 and 1988. suggesting that Doto reproduced twice each year.

Recruitment of Doto to colonies of Q. geniculata was related to the hydroid community (F=5.297, df=4. 225. p<0.001). Doto recruited most heavily to Q. geniculata when colony size was largest (winter, 1987 and summer. 1988).

Colony size was the only significant parameter correlated with recruitment of

Doto (Table 2-4B).

Eubranchus exiguus was found on 27% of kelp blades collected (Figure 2-

7B) and was never the only species present (Figure 2-1). Eubranchus was absent in 4 of 23 months (January. October, 1988 and January. February.

1989). A large index of dispersion and non-significant Chi-square test suggested that Eubranchus aggregated on colonies of Q. geniculata (Table 2-1).

During each sampling period, it was common to find all individuals of

Eubranchus on 1 or 2 of the five kelp blades sampled. Recruitment of

Eubranchus was not related to any component of the kelp/hydroid community examined (F=1.518. df=4. 225. p=0.198) (Table 2-2C).

55 Eubranchus was common during the first year of the study but rare in the

second year (Figure 2-4C). Peaks of recruitment occurred in the summer of

1987 and in early winter of both years. Abundances of Eubranchus were

highest in December of each year and coincided with populations lows of other

nudibranchs (Figure 2-5. 2-9). This suggested that Eubranchus utilized the

habitat when others used it less.

Although small Eubranchus (<3 mm) were collected throughout the first year of the census (Figure 2-10 A-C), small pulses of recruitment occurred in

December. 1987 and April. 1988. During the first year, large Eubranchus also were collected during these pulses of new recruits (Figure 2-10D). The first six months of the second year were similar to year 1. a small pulse of small

Eubranchus occurred in December. 1988, but essentially no Eubranchus were collected from January-May. 1989 (Figure 2-10).

Tergipes was the most abundant nudibranch on colonies of Q. geniculata on Laminaria spp.. Tergipes was present in all samples (Figure 2-9) and on

94.7% of all kelp blades collected. Four to five peaks of abundance could be identified during each census year, suggesting that Tergipes reproduced at least

4 times each year. Peaks in abundance of new recruits (< 1 mm) (Figure 2-9A) were followed within 2-4 weeks with a peak in the population of larger Tergipes

(Figure 2-9B). Spawn was found in every month of the census.

The population of Tergipes was separated into two sub-populations to discern new recruits from newly settled nudibranchs. New recruits were considered any nudibranch of < 1 mm in body size. New recruits and larger

Tergipes were present on 97% and 91% of kelp blades collected, respectively

(Figure 2-1, 2-11). Immature Tergipes (<4 mm) were abundant throughout the census period (Figure 2-12 A-C) and comprised 95.1% of the total population of

Tergipes collected (Figure 2-6D). Distributions of each sub-population have

56 large indices of dispersion and Chi-square lesis are non-significant (Table 2-1) suggesting that each tended to clump in colonies where conspecifics were present.

Abundances of both sub-populations of Tergipes were related to the physical components of the hydroid community (new recruits: F=22.050. df=4.

225. p<0.001: large Tergipes: F=30.755. df=4, 225. p<0.001). Each sub­ population was most abundant when the hydroid colony was largest and most dense. New recruits of Tergipes were positively associated with the size of the hydroid colony and density of hydrocauli but negatively associated with the size of the kelp blade (Table 2-3A). Because the size of the hydroid colony was limited by the size of the kelp blade, larvae of Tergipes settled into the largest hydroid colony on the smallest kelp blades. The abundance of larger Tergipes was associated with the size of the hydroid colony, density of hydrocauli and average height of the colony (Table 2-3B).

57 DISCUSSION

Nudibranch Population Biology

Seasonality

The analysis of the nudibranch community in Obelia geniculata colonies on blades of Laminaria spp. showed large summer and small winter abundance peaks. Previous studies on the seasonality of nudibranchs suggested that patterns of abundance were related to environmental factors and the appearance of species-specific food resources. Temperature has been suggested to be a primary factor limiting the distribution of many species of nudibranchs

(Clark. 1975). The nudibranchs in this study are boreal-subarctic species. The general disappearance of nudibranchs in late summer may be due to higher water temperatures. Todd and Doyle (1981) suggested the settlement-timing hypothesis to explain seasonal patterns of invertebrate predators and their prey. Predictable blooms of Q. geniculata occur in February. May and October.

Peaks in abundances of the nudibranchs generally followed these periods of hydroid abundance (Figure 2-4. 2-9). The concentration of reproductive effort to times when food is available for post-metamorphic animals has obvious selective advantages for any species. This scenario appears possible for the nudibranchs at Cape Neddick. ME.

Size

Size distributions of species revealed differences in reproductive cycles, mortality and habitat use. Each of the populations of nudibranchs in this study was dominated by small individuals (Figure 2-6). Other studies documenting abundances of nudibranchs have not reported the presence of very small nudibranchs (<2 mm) in the population (Swennen, 1961; Miller.

1962; Clark, 1975: Nybakken, 1978). Describing the presence and abundance

58 of only large, mature nudibranchs. may obscure reproductive patterns, and fail to reveal differential mortality in some habitats over others.

Small Dendronotus utilize thecate hydroids as food while larger, mature nudibranchs are associated with the hydroid Tubularia spp. (Swennen, 1961:

Thompson, 1964; Todd, 1981). Q. geniculata functioned as a nursery habitat for Dendronotus: 87% of all nudibranchs were <3 mm. The size at which

Dendronotus switches prey (10-12 mm) is larger than the maximum size of any of the other nudibranchs in the hydroid colony. This suggests that the use of

Q. geniculata as a food resource may not provide the necessary caloric requirements to sustain larger nudibranchs. In general, it is uncommon to find any nudibranchs larger than 10 mm in colonies of Q. geniculata (pers. obs.;

Kuzirian, unpub. data). Also, large Dendronotus may leave Q. geniculata for hydroid colonies which provide a better refuge or they may be eaten by the wrasse Tautogolabrus adspersus. Small Dendronotus are cryptic in colonies of

O. geniculata and T. adspersus readily eats Dendronotus when made available

(pers. obs.: Harris, pers. comm.). Large Dendronotus feed on polyps atop hydrocauli (Chapter 3) and thus are more exposed to fish. Poor caloric content and fish predation may provide the selective pressures necessary for

Dendronotus to benefit from switching prey with size.

Doto coronata also utilized Q. geniculata as a nursery habitat. Eighty-five percent of the Doto collected from colonies of Q. geniculata on Laminaria spp. were immature (<3 mm). Ribbons of spawn were found infrequently and usually only when abundances of other thecate hydroids were low. Kuzirian

(unpub. data) found high abundances of Doto (>3 mm) on colonies of the hydroids Thuiarla argentea. Obelia commlssuralis and Dvnamena pumila and generally at times of the year when low abundances of Doto were found on Q. geniculata (Figure 2-4. 2-8). It is likely that small Doto are cryptic in the short

59 colonies of Q. geniculata but require taller hydroid colonies for protection from

fish predation when adult pigmentation is acquired.

Eubranchus exiguus and Tergipes tergipes are both small, cryptic aeolids;

maximum body size is <10 mm (Thompson and Brown. 1984). Both nudibranchs have similar reproductive periods (Swennen. 1961: Clark. 1975;

Thompson and Brown. 1984; Lambert. 1985; Kuzirian. unpub. data) but recruit to different hydroid species. In early summer, high abundances of E. exiguus were found on colonies of Obelia commissuralis (Lambert. 1985; Kuzirian. unpub. data), while Tergipes recruited to £>. geniculata (Figure 2-9). Egg capsules of Tergipes were found on Q. geniculata throughout the year and only one Eubranchus egg mass was discovered (December. 1988). This was the only sample date between June, 1988 - May, 1989 when mature Eubranchus were found on Q. geniculata and it corresponded to a winter recruitment peak

(Figure 2-10). Kuzirian (unpub. data) found adults and egg capsules of E. exiguus in Q. commissuralis from April-July in four consecutive years (1972-

1975) at Gerrish Island. ME. This suggests that Q. geniculata is an alternative food resource and fringe habitat of E. exiguus. whereas it is the primary food and habitat for Tergipes tergipes.

Habitat Selection

A good living place provides breeding sites, foraging areas and safety from enemies. Marine larvae recognize habitats via structural complexity, color and chemistry (Thorson, 1950: Meadows and Campbell. 1972). When certain cues are reliably associated with superior sites, innate preferences for particular habitats evolve. The evolution of sensory structures and behaviors in larvae are important to their evaluating settlement sites.

60 Chemical cues, especially from food, induce the settlement and metamorphosis of nudibranch veligers (Todd. 1981: Hadfield. 1984). A majority of nudibranchs have very specialized diets, among them especially the hydroid eaters (Thompson, 1976; Todd. 1981). Since many nudibranchs mature, spawn and die within a few weeks, a genetic basis in the selection process is necessary for the survival of adults.

Many nudibranch veligers select large, dense hydroid colonies as habitats

(Harris, 1973: Todd. 1981) because they provide food and habitat heterogeneity.

The adult food induces metamorphosis in a number of nudibranch larvae (Todd.

1981: Hadfield, 1986, reviews). The benefit to survival is obvious as posi- metamorphic nudibranchs do not need to search further for food. Large, dense colonies have a more diverse stolon network and more polyps than small, sparse colonies (Crowell, 1957; Bravermann. 1971). As wide, flat-bladed algae are colonized by epiphytes, the structural complexity of the kelp surface is increased providing more microhabitats for settlement (Hall and Bell. 1988) and refuge from fish predation (Hayward, 1980: Edgar, 1983b: Orth, et al.. 1984).

Hydroid colonization of kelps essentially creates a 'lawn' on the surface of the blades. Nudibranchs settling into the colony can select habitats with respect to height above the surface and proximity to the edge of the colony. This mechanism is similar to selection of grass microhabitats by insects (Strong, et al.. 1984).

Hydroid colonies can also provide protection from fish predation. Small nudibranchs are generally cryptic: both Eubranchus and Tergipes are hidden within colonies of Q. geniculata. however Dendronotus and Doto become more conspicuous with size and acquisition of adult pigmentation. The red pigmentation often found on large Dendronotus (>10 mm) may partially explain the prey switching behavior to colonies of Tubularia by this because

61 these nudibranchs are usually visible atop hydrocauli (pers. obs.). Although it is common to find Doto at the bottom of colonies (Chapter 3) presumably well protected from fish, better refuge may be alforded by the taller, bushier colonies of the hydroids Thuiaria argentea and Obelia commissuralis where Kuzirian

(unpub. data) found many mature Doto.

62 Figure 2-1. Proportion of kelp blades, Laminaria spp. (n=230) covered with Obelia geniculata containing nudibranchs. Data within each bar represent proportion of kelp blades having Individual or groups of nudibranchs. (Df = Dendronotus frondosus. Dc = Doto coronata. Ee = Eubranchus exiguus. Tt = Tergipes tergipesl

63 Proportion 0.0 - .5 0 0.2 - 0.3 - 0 .4 0.1 - - -

uirnh pce o a Kelp Blade a on Nudibranch Species 64 TtDc ZTtDfE* TtDc Ee Df DcEe

Figure 2-2. Mean proportion (+SE) of the nudibranch population sampled which were Juveniles in colonies of Obelia geniculata on Tjuninaria spp at Cape Neddick. ME from June. 1987 - May. 1989.

65 X Proportion o.o 0.2 - .4 0 0.8 0.6 1.0- - a Fb a Ar a u Jl u Sp c Nv Dec Nov Oct Sep Aug Jul AprMay MarJun Feb Jan 66 X Month X Figure 2-3. Proportion of kelp blades. Laminaria spp. (n=230) with Dendronotus.

67 89

Proportion

o o O O o M M CD

■ CT ~5 Q) 3 O =X CO

25

28 29 30

44 Figure 2-4. Mean abundances l+SE) of nudibranchs collected at Cape Neddick. ME from June. 1987 - May, 1989. Data were standardized to the size of the hydroid colony (no samples collected in August. 1987.).

69 Mean * Nudibranchs / 100 cm2 Hydroid Area 0 2 4 « 20 - 30 10 0 - 40 0 - 50 0 2 I 3 4 5 - - • N T * DO-*Oe« lTftafi* i f a t f Ml T « e O * - O D O M T v* N O N O V f O N O t f O N H ^ M * < O « f t r M M | i ( M N O N 0 V4 O M O W IN ••K J JAS 0 N F DJ fl FA ft ft J O A B O J S J AII 97 98 1989 1988 1987 70 TT Dendronotus Eubranchus « e t f « N o T t?r? T IT r ? Hi1t o t o D ‘NOMO^ O M O N •‘

Figure 2-5. Mean abundance (+SE) by size class of Dendronotus frondosus collected at Cape Neddick, ME from June. 1987 - May. 1989. Data were standardized to size of the hydroid colony (no samples collected in August. 1987.

71 Mean * Nudibranchs / 100 cm2 Hydroid Area 0 2 23 - 23 15 - 15 - t t t t i t f t t t t T t *40 • o *• 0 4 0* n 11 2-5 mm 2-5 < 1 mm 1 mm

Figure 2-6. Size frequency histograms for the four species of nudibranchs in colonies of Obelia geniculata on blades of Laminaria spp. at Cape Neddick. ME, June, 1987 - May. 1989.

73 Proportion 0.3 0.4 OS 02 02 0 0 0 I0 9 10 0 7 6 5 11 4 3 2 1 12 1 < tS 16 20 23 24 D A

75 Proportion 0.0 0.2 0.4 - 0.6 0.8 00 0.2 0.4 - 0.6 0.8 - - - - 0 0 1 Nudibranchs 2 76 2 3 3 4 Eubranchus 4 5 Doto 6 5 Figure 2-8. Mean abundance (+SE) by size class of Doto coronata collected at Cape Neddick, ME from June. 1987 - May. 1989. Data were standardized to size of the hydroid colony (no samples collected in August, 1987).

77 Mean # Nudibranchs / 100 cm2 Hydroid Area

-4 oo

- 1

55

t i Figure 2-9. Mean abundances (+SE) of Tergipes tergipes collected at Cape Neddick, ME. June. 1987 - May, 1989. Data were standardized to the size of the hydroid colony (no samples collected in August. 1987.).

79 Mean * Nudibranchs / 100 cm2 Hydroid Area 100 200 ISO 50- 100 120 20 40 60 - - - J J A S O N D JFflAMJJA S ‘ O ND JF M A f l 97 98 1989 1988 1987 80 Tergipes New Recruits (< New 1 Recruits mm) Tergipes Tergipes (>1 mm) Tergipes

Figure 2-10. Mean abundance (+SE) by size class of Eubranchus exiguus collected at Cape Neddick. ME from June. 1987 - May. 1989. Data were standardized to size of the hydroid colony (no samples collected in August. 1987).

81 < 1 mm

A * t"t-TA r & r - t^ T ! I 1 I I t ! 1 m m

2 mm

1 0 kj.fyfrAfv v f ^ tA? tt i-frTfrj f i 7 6 >2 mm 5

4

3 2' 1 0 A rrrr T f t f T T lAt^1! PTttTftftf T f T t » r?T t t N B oo * •>NBB_iD^-4in'«rv —»e»mr^e»xinnmo-«iBTn«»

NaNo87 1968 1969

82 Figure 2-11. Proportion of kelp blades Lamlnaria spp. (n=230) containing A. new recruits of Tergipes tergipes (< 1 mm) and B. T. tergipes (> 1 mm).

83 Proportion o.o o 0 0 O 0 0 0 O 0 0 0 p 0 0 o 0 0 0 o o - i i i • i i i i i 1 cn

— — — i i t i i — — — —k — — > > *e - t e * r> ♦ * Nudibranchs •* Tergipes New TergipesRecruits New (<1mm) 4 8 — M V) « M N — O Tergipes (>I mm) ...... Figure 2-12. Mean abundance (+SE) by size class of Tergipes tergipes collected at Cape Neddick. ME from June. 1987 - May. 1989. Data were standardized to size of the hydroid colony (no samples collected in A ugust. 1987).

85 Mean * Nudibranchs / 100 cm2 Hydroid Area

8 8 8 - i - —L--- .___ 1_

?* co°*\14*b o 01- as-1 x »»-ao- ° 10- m C.i i -1 !£rfj H **■01- > £ i ' 00 0# a 11- 0) > 1 ] ^ 3 S:f

c . 20- 0 9 - > I > w at-1 * 1 5 :?

— c. ~ _ M - - *0£ — IS «■ ■* ai Table 2-1. Indices of dispersion (I.D.) (variance/mean abundance) for each species of nudibranch in colonies of Obelia genieulata on Laminaria spp.. A Chi-square test was performed: H0: The negative binomial is an adequaie distribution to describe the pattern (Krebs, 1989). Outliers were identified by large Studentized residuals and removed from the analysis.

Species I.D. Chi-square d f P

Dendronotus 12.35 40.144 42 NS

Doto 2.00 4.201 3 NS

Eubranchus 2.37 3.818 4 NS

TereiDes 55.31 69.350 58 NS new recruits (< 1 m m)

Tergipes 88.28 55.680 58 NS (>1 m m )

87 Table 2-2. Analysis of variance tables of paramaters of the kelp/hydroid community. Model: Sqr nucjibranch abundance = constant + sqr kelp area |c m i + sqr hydroid area (cm*) + sqr density of colony (#/cm^) + height of colonyr»r-k1r\n\r (mm). fmml

Variable Coefficient (SE)

A. Dendronotus

Constant -1.313(0.687) -1.912 NS

Kelp area -0.014(0.015) -0.907 NS

Obelia area 0.039(0.027) 1.464 NS

Obella density 0.498(0.170) 2.922 0.004

Obelia height 0.150(0.043) 3.473 0.001

B. Doto

Constant 0.760(0.239) 3.175 0.002

Kelp area -0.008(0.005) -1.464 NS

Obelia area 0.040(0.009) -4.308 <0.001

Obelia density -0.019(0.059) -0.322 NS

Obelia height -0.012(0.015) -0.763 NS

C. Eubranchus

Constant 0.475(0.266) 1.789 NS

Kelp area 0.006(0.006) 1.007 NS

Obelia area -0.010(0.010) -0.932 NS

Obelia density 0.090(0.066) 1.371 NS

Obelia height 0.024(0.017) 1.446 NS

8 8 Table 2-3. Analysis of variance tables of paramaters of the kelp/hydroid community. Model: Sqr nuchbranch abundance = constant +■ sqr kelp area (cm2) + sqr hydroid area (cm2) + sqr density of colony (#/cm2) + height of colony (mm).

V ariab le C o efficien t (SE)

A. Tergines - New recruits (< 1 mm)

C o n stan t -3.002 (1.480) -2.029 0.044

Kelp a re a -0.076 (0.033) -2.305 0.022

Obelia area 0.424 (0.057) 7.404 <0.001

Obelia density 1.136 (0.367) 3.095 <0.001

Obelia height 0.111 (0.093) 1.192 NS

B. Tergjpes - (>1 mm)

C o n stan t -7.513 (1.864) -4.030 <0.001

Kelp a re a -0.053 (0.042) -1.264 NS

Obelia area 0.363 (0.072) 5.036 <0.001

Obelia density 1.302 (0.462) 2.816 0.005

Obelia height 0.687 (0.118) 5.846 <0.001

8 9 CHAPTER 3

RESOURCE PARTITIONING IN NUDIBRANCHS IN COLONIES OF

THE HYDRO ID OBELIA GENICULATA

INTRODUCTION

Information on the feeding biology of predators is needed to understand the relationship between predators and their prey. In terrestrial bird communities. MacArthur (1958) showed that spatial segretation and variation in bill morphologies were related to the types of food consumed by finches. In aquatic communities, grazing by herbivorous gastropods can modify algal associations on the marine rocky, intertidal zone (Lubchenco, 1978; Creese,

1988) and in freshwater, macrophyte communities (Bronmark, 1989; Osenberg,

1989) by eliminating opportunistic algae. In epifaunal and in fouling assemblages, nudibranch molluscs can alter community structure in a similar manner (Clark, 1975: Lambert, 1985; Gaulin, et al.. 1986; Todd and

Havenhand, 1989).

Nudibranchs are gastropod predators that often specialize on particular prey types (Harris. 1973; Todd. 1981). They possess a rasping organ, the radula. that allows them to penetrate the skeleton of their prey. The radula is a membranous belt covered with renewable chitinous teeth. Hughes (1986) proposed that radular shape and tooth morphology dictate the diets of gastropods and correlations between tooth form and diet are most common for specialized feeders (Nybakken, 1970; Bloom. 1976). Nudibranchs that feed upon hydroids tend to either penetrate the perisarc and suck coenosarc or attack exposed polyps directly (pers. obs.). According to Nybakken and

McDonald (1981). nudibranchs with a uniseriate radula should feed by piercing through stolons (perisarc) and nudibranchs with a triseriate radula should feed

90 upon athecate polyps and gonophores.

The majority of nudibranchs in the Gulf of Maine prey on hydroids (Meyer.

1971; Clark. 1975: Lambert. 1985). Often nudibranchs demonstrate considerable overlap in their distribution with many nudibranch species feeding in the same hydroid colony (Kuzirian, 1979). Coexistence in these hydroid colonies may be mediated by differences in habitat or food, but the underlying mechanisms are unknown.

The hydroid Obelia geniculata grows epiphytically on laminarian kelps; often covering the entire blade surface. Four species of nudibranchs inhabit colonies of Obelia simultaneously: Dendronotus frondosus. Doto coronata.

Eubranchus exiguus and Terglnes tergipes (Chapter 2). This study focuses on the feeding biology of these nudibranchs inhabiting colonies of Q. geniculata.

For each nudibranch species the following parameters are described: morphology of the radula, feeding behavior and location of the nudibranchs within the hydroid colony. The skeletal morphology of the hydroid prey is also described. Differences in trophic morphology and feeding mechanisms between predators may reflect a predator s diet within the hydroid colony. W hether these differences in feeding biology and habitat help partition resources among these nudibranchs is considered.

91 MATERIALS AND METHODS

Collection of Animals

Nudibranchs and colonies of Q. geniculata were collected from the kelp bed from September. 1988 to September, 1989. Kelp blades (Lamlnaria

saccharlna and L, dlgltatal were examined underwater for nudibranchs. individually placed in plastic bags and brought to the lab. In the lab. four species of nudibranchs. Dendronotus frondosus. Doto coronata. Eubranchus exiguus and Tergipes terglpes were isolated from the hydroid colonies, sized to the nearest mm and separated for different parts of the study with a dissecting microscope.

Analysis of Radulae

Radulae from 4 individuals each of Dendronotus. Doto and Eubranchus and 7 individuals of Tergipes were removed from the nudibranchs by dissolving their bodies in 10% NaOH overnight (10-12 hours). Radulae were teased free from the disintegrated muscle tissue and placed into 70% ethanol until mounting for scanning electron microscopy (SEM). Clean radulae were mounted onto glass coverslips in a drop of distilled water and allowed to air-dry before attaching to stainless steel stubs. Radulae were sputter coated with a

200-300 A° coating of Au/Pd (Hummer V Sputter Coater) before viewing with an

AMR 1000 Scanning Electron Microscope at 20 kV.

Six parameters from 3 tooth rows of each radula were quantified to identify the morphology of each nudibranch species' radula. From a dorsal perspective the width (um| of the radula. the length (um) and width (um) of the rachidian tooth were measured and the number of denticles on one side of the rachidian tooth was counted (Figure 3* 1A). The rake angle of the radula and

9 2 the curvature of the radula were quantified from a lateral view (Figure 3- IB).

The rake angle was defined as the angle of the radula to the feeding surface when protruded from the mouth. Rake angle was measured as the angle made from the tip of the rachidian tooth to the base of the tooth row. Curvature is the degree of hook or amount of concavity along the inner margin of the radula

(Bloom. 1976). Curvature was measured as the ratio between the height of the concavity along the inner margin of the rachidian tooth to the length of the rachidian tooth. All length measurements and angles were digitized using

Sigma Scan on a Jandel Scientific digitizing tablet (Model 2210.3.C)

A MANOVA was performed on In (x+1) and square root (x+0.5) transformed data to elucidate morphological differences of radulae from each species of nudibranch (Model: Radula width, length of rachidian tooth, width of rachidian tooth, number of denticles, rake angle = constant + species of nudibranch) (Zar, 1984: Harris. 1985). Ratios of morphological measurements to size of the nudibranch were calculated to account for differences in size of individual nudibranchs.

Feeding Behavior and Feeding Damage

Feeding mechanisms were identified for each species of nudibranch.

Nudibranchs were starved overnight before being placed on kelp blades covered with Obelia geniculata in 10 cm diameter stacking dishes. Nudibranch feeding behaviors were observed with an Olympus SZ-III/SZ-Tr zoom stereo microscope. Individual feeding bouts were photographed with an Olympus OM-

2 camera attached to the microscope. A minimum of 6 feeding bouts was observed for each nudibranch species. A feeding bout consisted of approach, attack, manipulation and consumption of the hydroid prey. Each feeding bout was considered complete when the nudibranch retreated from the polyp or from

93 the stolon it was eating.

Portions of predator-manipulated hydroid sections were obtained directly

from feeding nudibranchs. Pieces of stolon were cut 0.5 cm from a site of

feeding and lifted from the kelp surface. If the stolon could not be easily

removed, the kelp was cut to obtain the feeding area. Uprights were also

removed after feeding bouts by nudibranchs. Hydroid skeletal pieces were

obtained from each nudibranch species except Tergipes because this species attacks naked tissue when feeding. All hydroid pieces were immediately

preserved in 2.5% phosphate buffered glutaraldehyde for 1-2 hours and then stored in 70% ethanol. Tissues were prepared for SEM by dehydration through a graded ethanol series and then placed in hexamethyldisilizane (HMDS) to finish drying hydroid skeletons and tissue (Nation. 1983; Polysciences. Inc..

1985). Skeletons were mounted on stainless steel stubs with double-stick tape and prepared for SEM as previously described for the radulae.

Nudibranch Habitats

The habitat of each species of nudibranch was identified by noting the location of individuals within the colony. After collection, hydroid samples were kept in a flowing sea water system at the University of New Hampshire Coastal

Marine Lab and processed within 24 hours. Observations were made with an

Olympus SZ-III/SZ-Tr zoom stereo-microscope by placing kelp blades in large metal trays. Four habitat parameters were quantified: (a) height of the nudibranch above the kelp surface (mm), (b) density of hydrocauli around the nudibranch, (c) mean height of hydrocauli (n=5) around the nudibranch and (d) the distance (cm) the nudibranch was from the edge of the colony.

Measurements of density of hydrocauli and mean height of hydrocauli were made within a 2.25 cm^ area around each nudibranch.

94 All habitat parameters were analyzed with a multiple ANOVA on In (x+ 1) and reciprocal (x+1) transformed data (Zar. 1984: Harris, 1985).

Perisarc Analysis

Samples of stolons and hydrocauli were isolated from the central and peripheral (outer 2 cm) portions of 5 colonies of Q. geniculata and fixed overnight in Bouin’s solution. After washing the hydroid pieces in 50% ethanol, they were dehydrated, stained with acid fuchsin. embedded in paraffin and sectioned. Lugol's solution was used to stain the perisarc of hydroid sections

(Lillie. 1954) following procedures adapted from Drury and Wallington (1980).

Measurements of perisarc thickness (um) of stolons, hydrocauli and theca from each area of the colony were made using a Nikon Fluophot. Microphot series V compound microscope equipped with an ocular micrometer.

A two-way analysis of variance was performed to compare the thickness of perisarcs from the center and from the periphery of the hydroid colony. A

Tukey-Kramer HSD multiple comparison test was used to test comparisons of interest among structures within and between areas of the colony (Zar, 1984;

Day and Quinn. 1989).

95 RESULTS

Structure of the Radulae

The radulae of the 4 species of nudibranchs differed markedly in the six

parameters measured (Table 3-1). Only the radulae of Tergpies were curved,

others showed no degree of hook. The radulae showed very little variation

among individuals of a species (Table 3-2). The similarity among conspecifics

and the large differences among species is not surprising because radulae are

used as a taxonomic indicator for opisthobranchs (Thompson and Brown.

1984). These results are similar to previously published figures and

observations.

Dendronotus frondosus had a wide, multiseriate radula with 5-9 lateral

teeth flanking the rachidian tooth (Figure 3-2A). The rachidian tooth was

triangularly shaped with many small, lateral denticles. The lateral teeth had a

single cusp and 3-5 denticles on each tooth and the radula had a wide rake

angle (58°) (Figure 3-2B).

The uniseriate radula of Doto coronata was very narrow (Figure 3-2C).

The rachidian tooth was short and stubby with 3-4 lateral denticles on each side. The central denticle of the rachidian tooth was also depressed, making the dorsal surface of the radula concave. The rake angle was very narrow (21°)

(Figure 3-2D).

Eubranchus exiguus had a wide, triseriate radula (Figure 3-3A). A single, lateral tooth flanked the rachidian tooth. The rachidian tooth was relatively long and thin with few (3-7) large, lateral denticles on each side: it appeared rake-like. The lateral teeth were wide at their base and tapered sharply to a thin, narrow, single cusp (Figure 3-3B). The rake angle was moderately wide

(36°).

96 Tergipes tergipes had a narrow, uniseriate radula (Figure 3-3A) with a

short, wide rachidian tooth that bears a prominent, central denticle and many

(5-10) large, lateral denticles. The central denticle was approximately 2 times

longer than the lateral denticles and it had a narrow rake angle (22°) (Figure 3-

3D.E). The radula of Tergipes was the only one of the four to show any degree

of hook. The curvature index for all 21 tooth rows was 0.293 (SE=0.04) but

there was considerable variation (Figure 3-3D.EJ. Some of this variability could

be due to the drying process used to prepare the radulae for viewing with the

SEM.

Feeding Behavior

Direct observations of nudibranch feeding showed species specific

behaviors for handling and consuming hydroid prey. The feeding behaviors

am ong conspeciflcs showed very little variation.

Two feeding mechanisms were observed for Dendronotus: these were

dependent on the animal's size. Small individuals of Dendronotus were

suctorial feeders while larger individuals were polyp biters. Individuals smaller

than 5 mm in length fed by penetrating the perisarc of thecae and stolons and

suctioning tissue from the hydroid (n=7). On thecae. nudibranchs positioned

their mouthes at the base of the cup by grasping it with the anterior portion of

the foot, thus securing the mouth against the perisarc. Muscular protrusions

of the buccal apparatus everted the radula from the mouth which scraped the

theca. The average time necessary to penetrate a theca was 4.2 min (SE=2.9).

When a hole was made, suction was created by the buccal apparatus that withdrew either the hydranth from a hydro theca or Juvenile medusae from a gonotheca. When feeding on gonangla, juvenile medusae were plucked

individually from the bottom of the gonotheca. On stolons 2-3 violent pulses of

97 the buccal apparatus created a hole through the perisarc that allowed withdrawal of the coenosarc. Tissue was extracted from the stolon by a few, quick pulses of the buccal apparatus.

Large Dendronotus (>5 mm) were polyp biters or grabbers. Nudibranchs crawled up hydrocauli and positioned themselves over an exposed hydranth.

The mouth was positioned directly above a polyp by light contact of the anterior portion of the foot and mouth with polyp tentacles. It appeared that both a suctorial and grasping mechanism was used to consume polyps. The radula was everted from the mouth creating a disturbance around the polyp that pulled the polyp towards the nudibranch's mouth. The radula never contacted the polyp (n=14), so polyp tissue was not hooked or pulled toward the mouth.

Immediately following, the jaws were protruded and the polyp was clipped off at its base. After Dendrontous bit a polyp, only the annulated base of the hydrotheca remained (Figure 3-4). A nudibranch would continue this behavior along a hydrocaulus following the alternating pattern of polyp branching.

Feeding on whole polyps was observed at a rate of 1-2 polyps per minute

(SE=0.2) over a 5 min period (n=6).

On one occasion during a feeding bout by a large Dendronotus (8-10 mm) a juvenile Tergipes (0.5 mm) was consumed along with a hydranth. The juvenile Tergipes had been atop the polyp feeding when Dendronotus arrived.

Apparently Dendronotus was unaware of the presence of the other nudibranch.

After swallowing the polyp and the juvenile Tergipes. Dendronotus paused, stopped feeding and crawled from the hydrocaulus. It is possible that some epidermal discharge from the eaten nudibranch was unpalatable to

Dendronotus.

Doto coronata was a suctorial feeder preying predominantly on the stolons of Q. geniculata. Nudibranchs crawled atop a stolon and clutched the stolon

9 8 with the anterior portion of the foot. The clutching behavior secured the stolon

against the animal's mouth. During pulsations of the buccal apparatus, the anterior portion of the foot flattened suggesting that the nudibranch was exerting additional pressure to assist in the penetration process. When a hole was produced through the perisarc. coenosarc was withdrawn from the stolon into the animal’s mouth. Tissue moved unldirectionally toward the nudibranch and into its mouth. The movement of tissue from a stolon coincided with pulses of the buccal apparatus. This process appeared similar to the action of drawing liquid through a straw.

The mechanism of perisarc penetration by Doto is by rasping with the radula. Drill holes produced by Doto (body size = 4-5 mm) were round (35.7 um

(SE=4.8) diam. Figure 3-5). Holes had bevelled sides with grooves corresponding to the denticles of the rachidian tooth. There was a raised lip around the perimeter of the holes. This raised lip was distinct from the perisarc and is either coenosarc of the hydroid or mucus produced by glands in the foot of the nudibranch. Also, these holes often showed a jagged or deteriorated perisarc at their center. The deteriorated perisarc may be the remains of an incomplete bore hole or suggests that the nudibranch secreted a substance to assist the penetration process.

The time necessary to penetrate a stolon by an individual of Doto was difficult to determine. A hole was complete when coenosarc was drawn toward the nudibranch. It was not uncommon for an individual nudibranch to position itself atop a portion of stolon for many hours (4-6). Over any single hour several periods of buccal apparatus pulsations alternating with longer periods of apparent Inactivity were observed. The periods of pulsations of the buccal apparatus were short, usually 1-3 minutes long with up to 10-15 minutes between sets of buccal pulses. A minimum of 1 hr was necessary for a

99 nudibranch to penetrate the perisarc but penetration averaged 3.3 hrs (SE= 1.2: n=6). This type of behavior suggested a chemical component to the penetration process because the nudibranch paused between muscular or mechanical manipuations.

Eubranchus exiguus was also a suctorial feeder, but fed only through the perisarc of hydrothecae. Nudibranchs (n=6) crawled up hydrocauli and positioned themselves at the base of a hydrotheca. The anterior portion of the foot grasped the base of the theca securing the nudibranch's mouth against the perisarc. Contractions of the odontophore everted the radula which scraped the thecal surface. When a hole was created, the tentacles and hypostome of the hydranth disappeared into the theca and tissue was withdrawn through the base. The nudibranch left when all tissue was removed. The entire feeding process occurred in 1-2 min (SE= 10.6 sec). Each successive polyp eaten by a nudibranch took longer to consume.

The penetration hole produced by E. exiguus was elliptical (Figure 3-6).

The sides of the holes were jagged and the perisarc of the theca showed a ripped, rather than worn or scraped perisarc as in holes in stolons produced by

Doto. The width of the hole at its center and widest point was similar to the width of the top 3 denticles on the rachidian tooth (Table 3-3). The lateral teeth on radulae of Eubranchus were not observed contacting the thecal perisarc during the feeding process.

The feeding mechanism used by Tergipes tergipes differed from those of the previous species: Tergipes attacked naked tissue. Small individuals (<4 mm) placed their mouth atop a polyp and rasped tissue directly from the area around the hypostome. If the hydranth was retracted into the theca, the nudibranch would curl its body over the thecal rim to reach the polyp tissue.

Smaller individuals (<1 mm) would crawl into hydrothecae to feed upon polyps.

100 While direct observations clarified the use of the radula in raking tissue from the polyp, I could not determine whether the jaws were also utilized. Small individuals (n=5) of Tergipes required up to 2 hours to consume an enLire polyp

(mean= 75 min; SE=45 min).

Larger individuals (n=8) of Tergipes (>3 mm) were able to either rip the upper half of the polyp or the entire polyp from a hydrotheca. This process was very rapid compared to the slow rasping of smaller Tergipes. The approach to a polyp, consumption of the polyp and departure from the branch took as little as

60 sec (mean=150 sec; SE=75 sec).

In addition to eating whole polyps, individuals of Tergipes >2 mm cropped tentacles. The nudibranch positioned itself at the tip of a tentacle and consumed the whole tentacle. The tentacle was then snipped off at its base. A

3-4 mm Tergipes could eat 60-80% of a whorl of tentacles on a polyp of Q. geniculata in 25-30 sec.

Habitats of Nudibranchs

Each nudibranch species occupied a particular area within the hydroid colony. Dendronotus was conspicuous on hydrocauli and was generally towards the center of the colony among tall hydrocauli at high densities. Doto was mostly found along the edge of the colony on the kelp surface or on stolons.

Eubranchus was usually on hydrocauli along the edge of the colony. Tergipes was commonly found in the center of the hydroid colony where tall hydrocauli at high densities are present. Tergipes was seldomly found off hydrocauli but when it was on the kelp surface it was actively crawling to another hydrocaulus.

A multiple analysis of covariance found no interaction between habitat parameters, and nudibranch species and size (Table 3-4).

A multiple analysis of variance tested habitat effects on nudibranch size

101 within a species. Sizes of Dendronotus. Doto and Eubranchus were not significantly related to any habitat parameter. However, two habitat parameters

(height and density of hydrocauli around a nudibranch) were related to size of

Tergipes (Table 3-5). Large nudibranchs were in areas with taller, more numerous hydrocauli than smaller nudibranchs.

The habitat occupied by each nudibranch species in colonies of O. geniculata differed. Each habitat parameter measured was important for the distinction of where a particular species of nudibranch could be found (Table 3-

6). A Tukey-Kramer HSD test was performed to determine differences between species within each habitat parameter (Table 3-7).

Individuals of each nudibranch species occupied a different height on a hydrocaulus (Table 3-6). Individuals of Tergipes were consistently found on hydrocauli well above the bottom of the hydroid colony. Tergipes was found at the bottom of the colony in only 6% of all observations (n=193). At these times they were crawling on the kelp surface. Tergipes also positioned itself higher on hydrocauli (6.0 mm. SE=0.3) than both Dendronotus and Doto (pcO.OOl). but not significantly different from Eubranchus (5.6 mm. SE=0.1). Eubranchus was found on hydrocauli 69% (n=16) of the time and its position also differed from both Dendronotus and Doto (p<0.001). Dendronotus maintained an intermediate position within the vertical component of the habitat. Individuals of Dendronotus were found to be 3.0 mm (SE=0.5) above the kelp surface, but in 56% (n=77) of the observations Dendronotus was on the stolons or crawling along on the kelp surface. Doto was consistently found at the bottom of the colony (0.6 mm. SE=0.3): in 90% (n=76) of all observations Doto w as on stolons or on the kelp surface. These data were similar to field observations (Figure 3-

7).

Two parameters were quantified to describe the immediate area (2.25 cm^)

102 around each species of nudibranch: mean height and density of hydrocauli.

The hydrocauli around Dendronotus {15.7 mm, SE=0.6) were significantly taller than those around all other nudibranchs (p<0.001). The height of hydrocauli around Tergipes {13.3 mm, SE=0.3), Eubranchus (12.3 mm. SE=0.3) and Doto

(11.5 nun, SE=0.5) were not significantly different from each other {Table 3-7).

The density of hydrocauli around individuals of each species of nudibranch showed the same general pattern (Table 3-7). The areas around Dendronotus

(14.6 hydrocauli/2.25 cm2, SE=0.6) and Tergipes (13.4 hydrocauli/2.25 cm2,

SE=0.5) contained a similar and higher, density of hydrocauli and both were significantly different (pcO.OOl) from the areas around Eubranchus (10.7 hydrocauli/2.25 cm2, SE=0.4) and Doto (10.3 hydrocauli/2.25 cm2. SE=0.5).

Eubranchus and Doto occupied areas that were not significantly different from each other (Table 3-7).

Individuals of Tergipes tended to occupy a more central area of the colony.

They were found 4.05 cm (SE=0.2) from the edge of the colony. Tergipes was further from the edge of the colony than Doto and Eubranchus (p<0.001) but not significantly different from Dendronotus. Dendronotus was found 2.65 cm

(SE=0.2) from the edge of the colony. Doto and Eubranchus occupied similar areas of the colony as Dendronotus (Table 3-7). Doto was 2.62 cm (SE=0.2) from the edge of the colony while Eubranchus was 2.28 cm (SE=0.3) from the edge of the colony.

Perisarc Analysis

The width of the perisarc of colonies of Q. geniculata differed among structures (stolons, hydrocauli. theca) within an area and between areas of the hydroid colony (Figure 3-8). The perisarc surrounding stolons and hydrocauli was thicker in the center than the edge of the colony. The perisarc on

103 hydrothecae was similar between both areas of the colony. Along the edge of the colony, the thickness of the perisarc differed (Stolons > Hydrocauli. 1=3.105. p<0.002: Stolons > Hydrotheca. t=l 1.661. p<0.001: Hydrocauli > Hydrothecae. t=9.367. p<0.001). In the center of the colony the perisarc surrounding stolons and hydrocauli was similar, but both were significantly thicker than the perisarc covering the hydrothecae (Stolons > Hydrothecae. t=19.781. p<0.001:

Hydrocauli > Hydrothecae, t=20.152, p<0.001).

104 DISCUSSION

Radular structures and feeding mechanisms vaiy in nudibranchs co­ occurring within colonies of Q. geniculata. Each nudibranch species used a distinct method to ingest hydroid tissue; little overlap existed between species.

Individuals within a species showed some variation with respect to the hydroid structure they fed upon, but small nudibranchs (<6 mm) appeared to be restricted mostly by the structure of their radula which determines how they feed (Hinde. 1958; Nybakken. 1970; Mills, 1977).

Feeding Biology

Nybakken and McDonald (1981) predicted that nudibranchs with wide, triseriate radulae should feed upon naked hydranths and gonophores.

Dendronotus frondosus has a wide, multiseriate radula (Figure 3-2) and

Robilliard (1970) described the general feeding behavior of Dendronotus as biting whole polyps and suctioning coenosarc. My findings support Robilliard. but it appears that the mechanisms used by Dendronotus are size dependent.

Small individuals (<5 mm) feed by piercing the perisarc and suctioning the coenosarc. while larger nudibranchs (>5 mm) bite whole polyps. The jaws are the primary structures used to penetrate the perisarc of stolons and thecae

(Waters. 1966; Robilliard. 1970). A large tear was found in the stolon after a feeding bout by a 4 mm individual (Lambert, unpub. SEM observation). In this instance, the radula was only used as a rasp to pull coenosarc from the open stolon (Waters. 1966; pers. obs.). When feeding on polyps, the oral papillae are used to orient the nudibranch to the hydranth by lightly touching the polyp's tentacles. The radula creates a suction that pulls the polyp towards the mouth while the Jaws bite the polyp at its base. Christiansen (1977) describes a

105 similar mechanism of coordinating Jaws and radula for feeding by Preeuthona

peachi on Hvdractlnia. Because P. peachi has a uniseriate radula (Thompson and Brown, 1984) in contrast to Dendronotus. it suggests that nudibranch size relative to hydroid prey is a better indicator of feeding mode than is radular structure for large individuals.

Although Eubranchus exiguus has a triseriate radula, individuals only feed by penetrating the base of hydrothecae. contrary to the predictions of

Nybakken and McDonald (1981). and withdrawing the hydranth from the cup by suction. The jaws may assist penetration of a hydrotheca by pinching the thin perisarc: however, the width of the narrow, longitudinal penetration hole

(Figure 3-6) corresponds to the top 3 denticles on the rachidian tooth (Table 3-

3). The lateral teeth are very thin, delicate structures that appear to have little functional utility. These data suggest that the lateral teeth are vestigial structures and are unimportant in the feeding process by E. exiguus. Schmekel

(1985) suggested that a reduction from broad radulae to narrow radulae has occurred in the evolution of opisthobranch radulae. Also, the break off early in the phylogeny within the superfamily Acleioprocta and are one of only two families within this superfamily that have retained a triseriate radula (Schmekel and Portmann, 1982). Two mechanisms for removing polyp tissue from the hydrotheca are feasible: (1) the rachidian tooth rasps tissue and may pull the polyp out from the bottom of the cup or (2) the buccal apparatus assists by creating a vacuum to suction (Kohn, 1983). The second mechanism is more probable because rasping would be a slower process than suctorial feeding and Eubranchus rapidly penetrates a hydrotheca and removes the polyp (15-30 seconds; pers. obs.).

According to Nybakken and McDonald (1981), nudibranchs with uniseriate radulae should feed upon stolons that are covered with perisarc.

106 They suggested that a uniseriate radula is better adapted for drilling holes than

radulae with three teeth per row. Doto and Tergipes both have narrow,

uniseriate radulae, but veiy different feeding mechanisms. Doto feeds by

rasping a hole through stolons while Tergipes rakes naked tissue of hydranths.

The radula of Doto is flat with the central denticle of the rachidian tooth

depressed below the lateral denticles (Figure 3-2C. 3-2D). The penetration

holes are circular (Figure 3-5) and similar to those produced by muricid snails

(Carriker, 1969), octopods (Nixon and Macommachie. 1988) and by the dorid

nudibranch Okadaia elegans (Young, 1969). Each of these predators prey upon organisms that have a calcareous exoskeleton (molluscs) or live in calcareous

tubes (serpulld polychaetes). These authors show evidence of chemical assistance during the penetration process. Muricid and naticacid snails utilize secretions from the accessory boring organ (Carriker. 1969: 1981), octopods utilize saliva from the posterior salivary glands (Nixon, 1979a, b: Nixon and

Maconnachie, 1988) and Young (1969) suggests that Q. elegans uses secretions from glandular cells around the lumen of the mouth which he feels are part of the stomodeal gland. The penetration hole produced by Doto shows an etching pattern along the sides (Figure 3-5B) as well as a deteriorated top (Figure 3-5D) that suggests the presence of a caustic agent. The behavior of Doto sitting atop stolons for long periods of time with short pulses of the buccal apparatus alternating with longer periods of inactivity is similar to the drilling-related behavior of the snail Urosalplnx described by Carriker (1969). This suggests

Doto may use chemicals to assist in the penetration of chilinous perisarc of Q. geniculata. No reports exist that show dotoids to possess glands capable of secreting substances for dissolution of prey skeleton. The salivary glands are unlikely candidates because they do not produce secretions containing lytic enzymes (Hyman, 1967; Welsch and Storch. 1973). The drilling process may be

107 purely mechanical as the nudibranch uses the denticles of the radula to rasp

through the chitin of the perisarc but histochemical work is needed to further

elucidate this mechanism.

Tergipes tergipes has a narrow, hooked radula that Is well adapted to rake

naked tissue. Individuals of all sizes (< 1 mm to 6 mm) feed by rasping tissue directly from polyps. Nybakken and Eastman (1977) showed that the curved radula of Triopha carpenteri is better for ripping flesh from erect biyozoa than the flat radula of Triopha maculata which is used to scoop polypides from colonies of Membranipora membranacea. an encrusting bryozoan. The feeding mechanism observed for Tergipes is similar to the mechanism used by T. carpenteri. Feeding on polyp tentacles by slurping seems to be restricted to larger individuals (>3 mm). This may solely be a function of the mouth size and the ability to engulf whole strands of tissue.

Size

The nudibranchs inhabiting colonies of Obelia geniculata are all veiy small; nudibranchs greater than 5 mm are seldom found (Chapter 2). The mechanism each species uses to feed may be in part dictated by its size, in addition to the structure of its radula. The dorids Triopha carpenteri and T. maculata reach sizes of several centimeters (Behrens. 1980). They are able to either bite off or scoop whole sections of ectoproct colonies (Harris, pers. comm.). Heavy Jaws are used to break through the relatively thin exoskeletons encasing zooecia of the ectoproct colonies. Dendronotus may grow to 100 mm

(Thompson and Brown. 1984). In colonies of Q. geniculata. Individuals of

Dendronotus greater than 8 mm are very rare (Chapter 2). but animals larger than 5 mm use Jaws to bite off whole polyps. Large Dendronotus feed similarly to Triopha spp. (Nybakken and Eastman. 1977). No reports exist describing the

108 feeding behaviors of small Triopha spp. and whether they use their radula to

penetrate ectoproct skeletons. This suggested that the structure of the radula

is a more accurate predictor of how a nudibranch feeds when the animal is

small, and at larger sizes the Jaws play an important role in feeding.

Habitats and Feeding

The analysis of nudibranch habitats suggested that each species of

nudibranch is most likely to be found in an area of the colony in which it can

feed. Dendronotus is essentially a generalist predator within colonies of Q.

geniculata. Although a pattern between feeding behavior and size of a nudibranch is discernable. size is not a factor in determining the nudibranch's position within a hydroid colony. Large nudibranchs (>5 mm) are infrequent occupants of colonies of O. geniculata; only 2.5% of all nudibranchs on colonies were larger than 5 mm during the two years of censusing (Chapter 2). Larger individuals (>14 mm; Miller, 1961)) of Dendronotus feed upon athecate hydroids. primarily Tubularia sp p . (Swennen, 1961; Thompson. 1964;

Robilliard. 1970; Clark. 1975; Todd, 1981). Large Dendronotus are very obvious in colonies of Q. geniculata (pers. obs.). Switching from Q. geniculata may in part be a function of Tubularia colonies providing a better refuge from fish predation as well as necessary caloric requirements to sustain large nudibranchs.

Doto coronata is found along the edge of colonies, at the bases of hydrocauli or on the kelp surface and among few, short hydrocauli. Doto feeds predominantly on stolons. The analysis of perisarc thickness showed that structures along the perimeter of colonies have thinner skeletal coverings than structures at the center. The relationship between habitat and feeding biology is consistent with the idea that the predator should select food items that are

109 easier to handle (Pyke, et al.. 1977). This preference to feed at the growing edge of colonies Is similar to findings by Harvell (1984) and Todd and Havenhand

(1989). showing dorid nudibranchs to feed at the edges of bryozoan colonies.

They interpret this behavior to be related to the lower calcification and strength of edge zooids (Best and Winston. 1984) and to differences in palatibility of structures between zooids in a colony. Harvell (1984) suggests that colonies of

Dendrobeania lichenoides have an ontogenetic gradient marked by morphological and physiological variation in tissue content of zooids. The gradient affects the grazing patterns of nudibranchs: nudibranchs prefer to feed upon zooids at the perimeter that are free of brown bodies and reproductive structures. In the case of Doto. the preference shown for stolons along the perimeter of colonies seems to be a result of a greater mechanical ease to penetrate these stolons. Composition of the coenosarc differs between the growing edge and the center of colonies of Q. geniculata with respect to cellular activity (Crowell, 1957: Braverman, 1971), but no reports discuss differences in nutritional value between areas in hydroid colonies as were found in bryozoan colonies (Harvell. 1984).

Eubranchus exiguus is generally found high on hydrocauli that are relatively numerous and tall. This position is consistent with its mechanism of feeding. E. exiguus feeds by penetrating hydrothecae and suctioning tissue.

The hydrothecae have the thinnest perisarc coverings (Figure 3-8) and taller hydrocauli have more polyps (Crowell, 1957). Eubranchus is also found along the perimeter of the colony. This Inconsistency in the pattern that a nudibranch s habitat is related to its feeding biology could be an artifact of the small sample size (n=16). A second possibility is that Eubranchus is less selective in where it feeds in colonies of Q. geniculata because this hydroid is an alternative food resource.

110 The habitat of Terglpes tergipes corresponds to its feeding behavior.

Individuals of Terglpes are found among many, relatively tall hydrocauli in the center of the colony. Crowell (1957) and Braverman (1963) showed that the centers of hydroid colonies are taller, denser and have more polyps than the periphery. Since Tergipes feeds on exposed polyps one might expect to find them where food is most abundant.

Resource Partitioning

The main differences among the four species of nudibranch in colonies of

Obelia geniculata are in habitat/feeding position and radular morphology (Table

3-8). Areas of the colony and different structures within an area show variation in density and height of hydrocauli (Crowell. 1957) and in skeletal thickness

(Figure 3-7). Differences in radular structure among nudibranch species are expected (Thompson and Brown. 1984), thus feeding mechanisms are also likely to differ (Hinde. 1958; Hughes. 1980). This study suggested that some separation of the hydroid food resource as a consequence of habitat selection within the colony and radular structure/feeding biology is present.

Circumstantial evidence suggests that the hydroid food resource is limiting in epiphytic and fouling communities (Clark. 1975; Harris, 1987; Todd and Havenhand, 1989). If food is limiting, then switching prey or prey colonies may relieve competitive pressure. Dendronotus is capable of swimming (Harris.

1973: pers. obs.) and adults greater than 12 mm feed on Tubularia (Miller.

1961; Clark. 1975; Todd, 1981). Many studies show that Tergipes. Doto and

Eubranchus are specialists on Obelia (Clark. 1975: Todd. 1981), but often do not report the species of Obelia these nudibranch species feed upon. Kuzirian

(unpub. data) and Lambert (1985) document high abundances of adults and egg masses of E. exiguus on colonies of Obelia commissuralis (= £). dlchotoma

111 (Cornelius. 1975)). Only one egg mass of E. exiguus was found on colonies of O.

geniculata on Laminaria spp. blades during the two years of the census

(Chapter 2). Doto coronata was never very abundant on O. geniculata: highest

abundances occurred in late summer and reached three individuals per 100

cm2 of hydroid colony. Egg ribbons of D. coronata were also infrequently

found. Kuzirian (unpubl. data) found Doto on colonies of Thuiarla argentea.

Obelia commissuralis. Q. geniculata on Laminaria spp. blades and

Campanularia flexuosa at Gerrish Island. Kittery, ME. Egg ribbons were found

associated with each prey species but predominantly with T. argentea. Kuzirian

only found individuals of Doto larger than 3 mm. while 71% of all Doto on O.

geniculata censused during 1987-1989 were less than 3 mm (Chapter 2). This

suggested that Doto leaves colonies of Q. geniculata and switches thecate

hydroid prey. An opposite scenario is found for Tergipes tergipes. Tergipes was

by far the most abundant nudibranch on colonies of Q. geniculata on Laminaria

blades: 94% of all nudibranchs censused were Tergipes (Chapter 2). T. tergipes

occurs on Q. commissuralis (Kuzirian. unpub. data: Lambert. 1985) but is

much less abundant than on Q. geniculata.

Resource partitioning of the hydroid colony as a consequence of present

competition is unlikely to be a major factor allowing coexistence of this suite of

nudibranch predators. Interference competition is seemingly too rare or weak

to cause the observed patterns of predation. One could expect large separations

in resource use in extant communties due to the ghost of competition past

(Connell. 1980). but I would still expect aggressive behaviors to occur more

frequently when nudibranchs encountered each other (Chapter 4). It appears

that Q. geniculata is a fringe food resource and habitat for Eubranchus and

Doto. a nursery habitat (Chapter 2) for Dendronotus and possibly for Doto. and the primary resource for Tergipes. Differential settlement and recruitment by

112 all four species of nudibranchs can at times overwhelm the hydroid, but equilibrium conditions necessary for exclusion are unlikely to occur or persist for long periods. Therefore resource partitioning need not be invoked to explain the continued coexistence of these predators.

113 Figure 3-1. Drawings of a radula tooth from a A.) dorsal and B.) lateral perspective. Schematics show the measurements made to quanitfy radular structure. (WR = Width of Radula. Wj. - Width of rachidian tooth. Lj- = Length of rachidian tooth)

114 distance EF Curvature Index = distance CD

(Bloom, 1976)

115 Figure 3-2. Scanning electron micrographs of radulae from Dendronotus frondosus (a.b) and Doto coronata (c.d) from a dorsal (a.c) and lateral view (b.d). Scale bars: a = 20um, b = 15 um, c.d = 5 um .

116

Figure 3-3. Scanning electron micrographs of radulae from Eubranchus exiguus (a.c) and Tergipes tergipes (c.d.e) from a dorsal (a,c) and lateral (b.d.e) view. Scale bars: a.b.c.d = 10 um. e = 15 um .

118

Figure 3-4. Scanning electron micrographs of A.) a healthy feeding polyp of Obelia geniculata and B.) the damage produced by Dendronotus frondosus (size 8 mm) after feeding. Scale bars 200 um .

120

Figure 3-5. Scanning electron micrographs of stolons of Obelia geniculata showing drill holes produced bv Doto coronata during feeding. Scale bars: a.c = 20 um. b.d = 50 um.

122

Figure 3-6. Scanning electron micrographs of hydrothecae of Qbelia geniculata showing penetration holes produced by Eubranchus exiguus during feeding. In a and c the arrow points to the area of damage; b,d are magnifications of these areas. Scale bars: a.c = 50 um, b,d = 10 um,

124

Figure 3*7. Field observations of proportions of nudibranchs occupying different heights on hydrocauli in Q. geniculata colonies on Laminaria spp. blades at Cape Neddick, ME.

126 Eubranchus Eubranchus Tergipes Dendronotus Dendronotus Doto

127 Figure 3-8. Histogram of the thickness of the perisarc of Obelia geniculata from stolons, hydrocauli and thecae from central and peripheral areas of the hydroid colony. (* p < 0.001)

128 Edge of colony Edge ■ Center of colony Hydrocauli Hydrocauli Thecae Structure Stolons - - 10 20 30 -

129 ! Table 3-1. Multiple analyses of variance of parameters of radular structure for the 4 species of nudibranchs IDendronotus frondosus. Doto corona ta. Eubranchus exiguus and Tergipes tergipes). Measurements of radula width fWR). and length (Wi) and width (Wr) of the rachidian tooth were adjusted to the size of the individual nudibranch and In (x+1) transformed to stabilize variance. The number of denticles were square root (x=0.5) tranformed. Model: In WR, In VVj, In Wr, sqr denticles, angle = constant + species of nudibranch.

Variable SS d f MSF P

In WR 40.716 3 13.572 397.356 <0.001 E rror 1.810 53 0.034

In Wr 6.772 3 2.257 56.743 <0.001 E rror 2.101 53 0.040

In 10.741 3 3.580 83.263 <0.001 E rror 2.279 53 0.043

Sqr(denticles) 2.946 3 0.952 94.119 <0.001 E rror 0.553 53 0.010

Angle 11646.59 3 3882.19 65.73 <0.001 E rro r 3130.28 53 59.06

130 Table 3-2. Parameters of the radulae of nudibranchs feeding on Obelia genlculata. Length measurements are standardized to accouont for size differences between individuals within a species. Three tooth rows from 4 individuals of Dendronotus frondosus. Doto corona ta and Eubranchus exiguus and 7 individuals of Tergipes tergipes were used. Values reported are means (standard errors) of all tooth rows measured.

Dendronotus Doto Eubranchus Tergipes n u m b e r of to o th 12 12 12 21 rows examined

Width of Radula 39.43 3 .56 31.31 8.02 (1.13) (0.05) (1.40) (0.51)

Width of Rachidian 9.84 3 .5 4 7.43 7.15 Tooth (0.25) (0.05) (0.40) (0.46)

Length of Rachidian 7.48 1.56 6.87 4.44 Tooth (0.33) (0.07) (0.37) (0.28)

Number of Lateral 10.58 3.33 4.92 7.52 Denticles (0.31) (0.14) (0.31) (0.36)

A ngle of Radula(°) 58.20 21 .5 7 36.45 22.45 (2.83) (0.95) (1.94) (1.86)

C urv atu re -- -- — 0.293 (0.04)

131 Table 3-3. A) Parameters of the penetration hole (Figure 3-6) produced in hydrotheca of Obelia ffeniculata by Eubranchus exiguns (size=4 mm). B) Width measurements (um) from radulae of 4 mm individuals of E. exiguus. Values are means (standard errors).

A. W id t h o f H o l e . Shape of Penetration Hole Top Middle Bottom Length

Ellipitical 5.70 8.00 4.08 32.86 (0.26) (1.07) (1.21) (3.36)

B.

Radula R achidian Top Top 3 Top 5 T ooth D en ticle D en ticles D en ticles

103.70 28.62 4.24 6.46 11.49 (3.86) (2.80) (0.26) (0.48) (0.33)

132 Table 3-4. Multiple analysis of covariance table of habitat parameters of nudibranchs within colonies of Qbelia ffeniculata and size of nudibranchs. Model: inverse (height on hydrocauli+1), mean height of hydrocauli around a nudibranch. In (density hydrocauli around a nudibranch+1). In (distance from the edge of the colony+1) = constant + species + size + (species*size).

Variable df F p

Effect = Species

Inverse height 3. 354 14.404 <0.001 Mean height 3. 354 4.968 0.002 In density 3. 354 1.217 NS In distance 3. 354 1.346 NS from edge

Effect = Size

Inverse height 3. 354 0.001 NS Mean height 3. 354 0.170 NS In density 3. 354 0.045 NS In distance 3. 354 0.010 NS from edge

Effect = Species’Size

Inverse height 3. 354 0.597 NS Mean height 3. 354 2.282 NS In density 3. 354 1.200 NS In distance 3. 354 1.119 NS from edge

133 Table 3-5. Analysis of variance table testing the relationship between habitat parameters of nudibranchs in colonies of Obelia genieulata and size of the nudibranch. Model: Inverse (height on an hydrocaulus+1). Mean height of hydrocauli. In (density of hydrocauli around a nudibranch^ 1), In (distance from the edge of the colony+1) = constant + size.

V ariable df F P

Dendronotus

Inverse height 1. 75 0.055 NS Mean height 1. 75 0.118 NS In density I. 75 0.982 NS In distance 1. 75 2.550 NS from edge

Doto

Inverse height 1. 74 0.040 NS Mean height 1. 74 0.838 NS In density 1. 74 0.013 NS In distance I. 74 1.123 NS from edge

E u b ran ch u s

Inverse height 1. 15 0.122 NS Mean height 1. 15 0.554 NS In density 1. 15 0.183 NS In distance 1. 15 0.813 NS from edge

Terglpes

Inverse height 1. 191 2.847 NS Mean height 1. 191 13.948 <0.001 In density 1. 191 6.689 0.010 In distance 1. 191 0.437 NS from edge

134 Table 3-6. Multiple analysis of variance table testing the relationship between habitat parameters of nudibranchs in colonies of Qhelia geniculata and species of nudibranch. Model: Model: Inverse (height on an hydrocaulus+1). Mean height of hydrocauli. In (density of hydrocauli around a nudibranch+1). In (distance from the edge of the colony* 1) = constant + species.

Variable df F P

Inverse height 3.358 51.983 <0.001 Mean height 3.358 11.115 <0.001 In density 3 .3 5 8 8.827 <0.001 In distance 3 .3 5 8 13.981 <0.001 from edge

135 Tabic 3-7. Summary of habitats of nudibranchs inhabiting colonics of Obclia gcniculata. Vertical lines within a habitat parameter indicate significant differences (Tukcy’s, p<0.05) between species of nudibranchs for each parameter.

Height on Height of Uprights Density of Uprights Distance from Edge Upright around Nudibranch around Nudibranch of Colony

High 1 Tcrgjpcs Dsodmaatus CfiDlllQDPlUS TereiDes I Eubranchus

Mid 1 Deodronoius TereiDcs Eubranchus Dendronotus Eubranchus

Low | Doto Doto V m Doto Eubranchus Table 3-8. Resource partitioning among nudibranchs inhabiting colonies of Ohelia peniculaia on blades of Laminaria spp..

Species Habitat Food Radula Feeding Feeding Position Mechanism

DcmlranQius tall hydrocauli at polyps/ wide, flat edge, high pierce & (small) high density stolons suck l^cmlraaaius tall hydrocauli at polyps wide, flat edge, high bile (large) high density

Eubranchus moderately tall polyps wide, flat edge, high pierce & hydrocauli at suck moderate density

Tenures moderately tali polyps narrow, center, high rasp hydrocauli at high curved density

Dfllu short hydrocauli stolons narrow, edge, low pierce & at low density flai suck CHAPTER 4

BEHAVIORAL INTERACTIONS AMONG NUDIBRANCHS IN COLONIES

OF THE HYDROID QBELIA GENICULATA

INTRODUCTION

Behavioral interactions between species in a community often affect the

use of resources by those species. Aggression by stem boring insects (Rathcke.

1976) and dung beetles (Bartholomew and Heinrich. 1978) limit access to food resources of congeners and other species. In these studies, when individuals of different species encountered each other, attacks by the superior resulted in injury or death of the subordinate. Also, interactions between species may be infrequent and occur only at high population densities. Populations of the white butterfly fPleris rapae) depressed the abundance of flea beetles on collards by inhibiting access to the food (Chew. 1981). Each of these cases describes interference competition where the actions of individuals directly affect how species use resources. Although studies show interference competition in insect communities, these situations are rare (Strong, et al.. 1984).

Studies on interference competition are concerned with the symmetry of interactions between species (Lawton and Hassel. 1981: Persson, 1985). Since the effects of one species on another are not equal, reciprocal effects of inferior species on the superior species are usually very small or undetectable. To determine the relative importance of interference competition between different populations or the proportion of possible species interactions which are significant, one needs to examine single species pairs for interspecific competition compared with intraspeciflc interactions (Keddy, 1989).

In marine, intertidal mudflats, habitat use by native snails Is restricted by interference from introduced snails. The Pacific mud snail. Cerithidea

138 califomica is confined to marsh pans in San Francisco Bay by behaviorallv avoiding Ilvnassa obsoleta (Race. 1982). In Barnstable Harbor. MA the periwinkle Llttorina littorea limits the microhabitat distribution of I. obsoleta in the mid-intertidal zone (Brenchley and Carlton. 1983). L. littorea stimulates an avoidance response in I. obsoleta by grazing on shell epifauna. This behavior interferes with the foraging, locomotory and reproductive activities of the native snail. Although behavioral interactions between marine gastropods affect distributions in some systems, the phenomenon is not universal (Berman.

1989)

In marine epifaunal communities, structure that mimics plants is provided by sessile invertebrates such as hydroids. Nudibranchs either graze and crop polyps or penetrate outer skeletons and suction tissue (Nybakken and

McDonald. 1981; Chapter 3). These mechanisms are analagous to insect feeding mechanisms on grasses and shrubs (Strong, et al.. 1984; Howe and

Westley. 1988). Often, the abundance and presence of particular species of nudibranchs are unpredictable in epifaunal communities but when multiple species are present competition for food and habitat are likely.

In the southern Gulf of Maine blades and stipes of the kelps Laminaria saccharina and L. digltata are often dominated by the campanularid hydroid

Obelia geniculata. At least four nudibranch species frequently inhabit colonies of Obelia spp.: Dendronotus frondosus. Doto coronata. Eubranchus exiguus and Tergipes terglpes (Swennen. 1961; Clark. 1975; Todd, 1981; Chapter 2).

Encounters between these nudibranchs occur frequently, but the extent to which interactions dictate how each species uses the hydroid colony is unknow n.

This study utilized two techniques to document the degree of interaction between nudibranchs within Q. geniculata colonies in the Gulf of Maine. The

139 behaviors of each nudibranch were observed in pair-wise interactions and experiments manipulated densities of individuals of each nudibranch species.

The potential for whether interference between nudibranchs plays a role in how nudibranchs are distributed within a hydroid colony is discussed.

140 MATERIALS AND METHODS

Collection of Animals

Nudibranchs and hydroids were collected from a shallow (4-10m), subtidal

kelp bed at Cape Neddick, York, ME from May - September, 1989. Kelp blades

were tom from stipes and placed in plastic bags while underwater. In the tab

the four nudibranch species were isolated and kept in mesh containers in

flowing sea water at University of New Hampshire Coastal Marine Lab in

Newcastle, NH. Hydroid colonies were also kept in tanks supplied with flowing

sea water until needed.

Behavioral Interactions

Nudibranchs were placed on portions of kelp blades covered with Q.

geniculata in 10 cm diameter stacking dishes. Interactions between

nudibranchs were observed with an Olympus SZ-III/Sz-Tr zoom stereo

microscope. All encounters between nudibranchs were recorded. Behaviors

were catergorized into the following patterns (after Allmon and Sebens. 1988):

touch (contact of oral tentacles or rhinophores with the other nudibranch), taste

(contact of mouth with another nudibranch), climb (movement of a nudibranch

over or onto the back of another nudibranch), cringe (quick, muscular contraction of a nudibranch's body), aversion or avoidance (movement away from another nudibranch), bristle (the erection or movement of cerata toward another nudibranch), no reaction (NR) (either retraction of rhinophores or no apparent movement).

141 Displacement and Nearest Neighbor

Pair-wise, manipulative experiments were conducted to test whether the location of a nudibranch differs when among conspecifics or heterospecifics.

Nudibranch densities in each treatment were consistent to field densities. Six pair-wise combinations are possible with 4 nudibranch species. Two combinations of species were not tested due to a shortage of nudibranchs

(Tergipes:Eubranchus. Dendronotus:Eubranchus). The densities used for each species pair were: Tergipes:Dendronotus (7:1). Tergjpes.Doto (8:1),

Dendronotus:Doto (6:1). Doto:Eubranchus (2:3).

Interspecific treatments were performed by separately placing individuals of two species of nudibranchs (A.B) onto a series of three microscope slides with pieces of Obelia-covered kelp attached. Individuals of species A were introduced to the slides and allowed to establish themselves. Slides were suspended in open slide trays in an aquarium at ambient water temperature. After 24 hours the location of each nudibranch was documented. Four parameters were measured to identify the location of each nudibranch: height on an upright, density of hydrocauli in a 1 cm^ area around the nudibranch. the distance between any two nearest nudibranchs (nearest neighbor) and the identity of the nearest neighbor. After the location of individuals of species A were recorded, species B was introduced to the hydroid colony and the slides were resuspended in aquaria for 24 hours. The location of each nudibranch was then again documented. A reciprocal pair-wise treatment was run simultaneously allowing species B to establish first.

To control intraspecific interactions against a density effect, monospecific treatments were conducted. Protocols were identical to heterospecific treatments and were run simultaneously.

142 Analysis of variance techniques were utilized in a randomized block design

(Zar. 1984) to determine if the location of a nudibranch differed when among conspeciflcs and heterospecifics. The pattern of spatial dispersion of individuals was determined using nearest neighbor methods (Clark and Evans.

1954).

143 RESULTS

Behavioral Interactions

The initial behaviors displayed by individuals of each species of nudibranch when it approached another nudibranch were similar (Figures 4-1 to 4-8). Encounters occurred while one nudibranch was crawling across the kelp surface; meetings between any two nudibranchs on a hydrocaulus were infrequent. On almost all occasions a nudibranch made contact with a heterospecific with its rhinophores or oral tentacles (Touch) or mouth (Taste).

These encounters were brief and the response of any nudibranch to contact varied but both the initiator and the recipient nudibranch generally reacted non-aggressively. The individual behavioral patterns for each species of nudibranch are described below,

Dendronotus frondosus

Dendronotus was very active; individuals were sedentary very briefly and only when feeding. When a Dendronotus approached another nudibranch it usually made contact with oral tentacles or rhinophores regardless of the identity of the other nudibranch (Figure 4-1). After making contact with another nudibranch. Dendronotus usually retracted its rhinophores resulting in a "no reaction" response or climbed over the other nudibranch. However, these responses to initial contact were inconsistent. In 33% of the encounters with

Tergjpes. Dendronotus turned away and retreated while in 3% of the encounters

Dendronotus contracted its body quickly after contact which was quantified as a "cringe" response. This was a reaction to nematocysts after touching the cerata of Tergipes. Dendronotus retreated from 24% of encounters with Doto.

A num ber of non-contact (Aversion) encounters were also observed. In

21% of the interactions with Tergioes and 18% with Doto. Dendronotus turned

144 away when within 2-3 mm and avoided any contact. This behavior suggested

an ability by Dendronotus to recognize a heterospecific without any tactile

contact (chemoreception).

The responses of Tergloes and Doto to advances by Dendronotus were

generally non-aggressive (Figure 4-2). Most reactions were simply a retraction

of oral tentacles or rhinophores by both species (NR). After contact by

Dendronotus. Tergipes retreated in 15% of encounters and Doto retreated in

11%. Avoidance behaviors were the second most common responses to contact

by Dendronotus and resulted when large (8-10 mm) individuals of Dendronotus

approached much smaller (<5 mm) heterospecifics. In a few encounters (8%) with Dendronotus. Tergipes turned and bristled its cerata at the head of

Dendronotus. This behavior caused Dendronotus to cringe or leave the immediate area.

Doto coronata

Individuals of Doto were relatively sedentary. It was not uncommon for any individual to remain atop a stolon for 2-3 hours during any observation period. Encounters initiated by Doto were very similar: Doto approached and touched Dendronotus in 87% of encounters and in 73% of encounters with

Tergipes (Figure 4-3). Most often Doto followed contact behavior by retracting its rhinophores (NR) or by climbing over the other nudibranch (Figure 4-3).

Doto occasionally retreated from encounters. An avoidance behavior as previously described for Dendronotus occurred in 13% and 14% on encounters with Dendronotus and Tergipes. respectively.

Reactions of heterospecifics to an encounter initiated by Doto were similar

(Figure 4-4). Dendronotus either did not respond (NR) (57%) or retreated from the encounter (28%). A cringe response was elicited by Dendronotus in 14% of the encounters. This behavior is curious because Doto does not possess stored

145 nematocysts or the ability to secrete noxious chemicals from its epidermis

(Thompson and Brown, 1984). Tergipes responded less often than

Dendronotus. Encounters with Doto either elicited a reaction, in this case

Tergipes retreated (21%) or individuals of Tergipes were unaroused (NR) (78%)

(Figure 4-4).

Tergipes tergipes

Individuals of Tergipes were generally active; they would crawl across the kelp surface and up and down hydrocauli stopping only briefly to feed on an exposed polyp- Tergipes initiated almost all encounters by either touching another nudibranch with its oral tentacles and rhinophores or tasting (Figure 4-

5). An aggressive behavior (bristling cerata at a heterospecific) was elicited to both Dendronotus (4%) and Doto (3%) but was infrequent. The behavior of

Tergipes after it initiated an encounter was variable. A non-aggressive response

(NR or climb) occurred in 52% of encounters with Dendronotus and 47% with

Doto. Also. Tergipes avoided contact with heterospecifics by turning and crawling away when within 2-3 mm in 16% and 23% of all encounters with

Dendronotus and Doto. respectively (Figure 4-5).

The majority of reactions of heterospecifics to Tergipes were retraction of rhinophores (NR) (Figure 4-6). Dendronotus retreated or cringed in 43% of encounters with Tergipes while Doto retreated from only 3% of encounters.

Doto was usually approached while it was atop a stolon, presumably feeding.

The contact (Touch or Taste) made by Tergipes seldom provoked Doto to respond with any reaction beyond retracting its rhinophores. Never was Doto displaced from a stolon. Tergipes would either turn away from or crawl around Dots-

146 Eubranchus exiguus

Interspecific encounters with Eubranchus were infrequent due mostly to a

scarcity of nudibranchs, therefore all encounters with heterospecifics were

pooled (Figure 4-7, 8). When Eubranchus approached another nudibranch it

initiated contact by touching or tasting and generally did not show any further

response (Figure 4-7). Eubranchus avoided contact in 15% of all possible encounters. Avoidance by Eubranchus was not evoked in response to any particular heterospecific. Retreats from encounters and a cringe response occurred in 6% and 3% of encounters with Doto. respectively.

When other nudibranchs approached Eubranchus. their behaviors were similar to those previously described. Touching and tasting occurred in 82% of all encounters and avoidance of an interaction occurred in the remaining 18% of possible encounters (Figure 4-8). The combination of a nudibranch avoiding contact (18%) with Eubranchus and retreating from an encounter after contact

(27%) is similar to the patterns seen for Tergipes. Both Eubranchus and

Tergipes store nematocysts in their cerata which provide a defensive ability.

Intraspeciflc Interactions

When any nudibranch approached a conspecific their behaviors were very similar (Figure 4-1 to 4-8). Retraction of rhinophores (NR), tasting and climbing were the most frequent responses by an initiator and recipient of all species. In all species except Tergipes. individuals did retreat from an encounter on occasion (Figure 4-2, 4-3, 4-8).

Displacement Experiment

The overall area utilized by a nudibranch within a hydroid colony is not generally affected by additions of heterospecifics or conspecifics. Height on a hydrocaulus did not vary significantly in any of the four treatments (Figure 4-9).

147 The density of hydrocauli (tt/cm^) around a nudibranch was different after

an addition of nudibranchs in some intraspecific trials. This difference in

hydrocauli density indicates that nudibranchs moved and not that they

modified the hydroid colony. When Doto were added to treatments with

conspecifics the areas occupied by nudibranchs differed, but this change was

not consistent between treatments. The hydrocauli density around Doto after

nudibranchs were added was greater in one treatmeant (Before=4.8. After=7.2.

F=5.48, p=0.03), less in another (Before=6.3, After=4.4, F=7.92. p=0.01) and

similar in a third treatment (Before=5.6. After=6.9). The area around

individuals of Tergipes had fewer hydrocauli in both treatments after

conspecifics were added to the hydroid colonies. For Dendronotus. the pattern

was also inconsistent: in one treatment nudibranchs were found among fewer

hydrocauli after the addition of more individuals (Before=9.7. After=6.2.

F= 13.29, p=0.002), while in the other treatment densities of hydrocauli around

the nudibranchs showed no significant change (Before=6.5. After=6.3).

Only the interspecific treatment between Tergipes:Dendronotus showed

differences in the density of hydrocauli around nudibranchs. After the

introduction of Tergipes to the hydroid colonies, individuals of Dendronotus were among more hydrocauli than individuals of Tergipes (Df=9.0. H=5.4.

F= 10.03, p=0.006). This change in hydrocauli density around a nudibranch is only an indication of the behavior of the nudibranchs in the hydroid colony.

Individual spacing between conspecifics did not vary between treatments

(Table 4-1). Individuals of Tergipes. Dendronotus and Doto maintained a similar distance from conspecifics regardless of the identity of the other nudibranch species present. Within each treatment involving these three nudibranchs. interspecific distances were similar to intraspecific distances for each of the species of nudibranchs (Table 4-1, Treatment 1,2,3). In each

148 treatment Intraspecillc distances for Dendronotus and Doto in the treatments with Tergipes and for Doto with Dendronotus could not be measured because only one individual of these species was used so that field densities could be maintained. In Table 4-1 intraspecific distances for these species from monospecific trials are reported to show similarities with other treatments.

Individuals of Doto are three times closer to each other than are individuals of

Eubranchus (Table 4-1. Treatment 4). Intraspecific distances between Doto are

2.5 times closer than interspecific distances to Eubranchus. but distances between individuals of Eubranchus did not differ from interspecific distances w ith Doto.

Nearest neighbor analysis (Clark and Evans. 1954) was used to determine the pattern of dispersion among conspecifics for the four species of nudibranchs. The pattern of dispersion was determined at field densities and then after an increase above field densities in monospecific associations. At field densities, each species of nudibranch is randomly distributed in the hydroid colony (Table 4-2). There was a tendency for the pattern of dispersion to change when densities of Dendronotus and Doto were increased. In two of the three replicates Dendronotus exhibited a regular pattern after densities were increased while Doto tended to aggregate (clump) at higher densities. In both Eubranchus and Tergipes no changes in the dispersion pattern occurred when densities were increased.

149 DISCUSSION

Behavioral interactions between nudibranchs in colonies of Obelia

geniculata are frequent, but do not influence the distribution or feeding

ecologies of other nudibranchs. Though nudibranchs were sufficiently

abundant in both observational and manipulative experiments to contact one

another, and densities were similar to those found naturally, nudibranchs were

not displaced from their preferred portion of the colony by competitors. Similar

conclusions were shown for insects on the same trophic level on plants (Root.

1973: Simberloff, 1978; Lawton. 1982) and in a salt marsh community for snails (Berman. 1989). The nudibranch community in Q. geniculata is probably similar to insect communities where resource-based competition is relatively unimportant in structuring this community (see Lawton and Strong. 1981:

Strong, et al., 1984).

Many ecological patterns are attributed to competition but alternative hypotheses need to be explored (Lawton and Strong. 1981). Interspecific competition has been designated the cause of niche division, character displacement and density compensation (Schoener. 1982: Cody and Diamond.

1975). however, despite the apparent lack of competition for food, the feeding ecologies of these four Gulf of Maine nudibranchs are different (Chapter 3).

Lawton and Strong (1981) stress that the important question to ask is not whether differences between species exist, but are the differences greater than other factors dictate?

A primary reason to reject interspecific competition as a major force structuring communities of insects and other organisms is the lack of intraspectfic competition (Miller. 1967; Strong, et al.. 1984). Behavioral interactions between conspeciflc nudibranchs were primarily non-aggressive

(Figure 4-1 to 4-8). On many occasions, the meeting of two conspecifics

150 resulted in the animals mating (pers. obs.). This suggested that nudibranchs

were generally unaware of others in the hydroid colony unless they are in

physical contact.

Displacement from a feeding position has not been observed and although

the location and dispersion {Table 4-2) of nudibranchs within the hydroid

colony was altered by increased densities of conspecifics, the pattern was

inconsistent. Two hypotheses are suggested to account for these obsevations.

Firstly, food is not at a shortage. Although many have inferred that

nudibranchs can destroy fouling communities by consuming particular hydroid

prey (Clark, 1975: Harris, 1987; Todd and Havenhand, 1989), no work has yet

demonstrated that food is limiting to these nudibranchs. Most hydroid-eating

nudibranchs are opportunistic, fugitive species with short life spans and a

single reproductive period just prior to death (Clark. 1975; Todd, 1981). By the

time a hydroid colony is decimated, nudibranchs within the colony have concluded reproducing and are beginning senescence. Secondly, nudibranchs within the hydroid colony are generally unaware of each other as are many insects on plants (Root, 1973: SimberlofT. 1978). Nudibranchs contacted each other only while crawling within hydroid colonies (Figure 4-1 to 4-8). The majority of these interactions were non-aggressive and recognition of other nudibranchs by a mechanism other than tactile contact was rare.

Differences in colonization, morphology and physiology, susceptibility to predation and food patchiness all contribute to patterns of species abundances and distributions and may not be influenced by interspecific competition

(Lawton and Strong. 1981: Diamond and Case. 1986). In the guild of nudibranchs in colonies Q. geniculata. biotic factors in the form of variable recruitment and predation may operate to regulate the community and reduce competition. Although peaks in colonization of all nudibranchs occurred during

151 the summer months (Chapter 2), food abundance was also high and alternative

foods were available (pers. obs.: Kuzirian. unpub. data). Q. geniculata also

grows epiphytically on Agarum cribosum (pers. obs.; Berman, et al.. in prep.)

and there are other common thecate hydroids on rock substrata. These

represent potential alternative habitats and food resources for settling veligers

that would reduce competitive interactions and promote coexistence.

The wrasse Tautogolabrus adspersus (cunner) feeds upon nudibranchs

which could facilitate coexistence of the nudibranchs. The vast majority of

nudibranchs inhabiting Q. geniculata colonies are less than 3 mm (82.6%.

Chapter 2). These animals are immature (Swennen. 1961; Miller. 1962;

Robillard, 1970. Clark. 1975) and cryptic. Large nudibranchs are likely to be

more susceptible to fish predation by the wrasse T. adspersus. Cunner readily

eat nudibranchs in the laboratory (Harris, unpub. data) and in the field (pers.

obs.). If large nudibranchs are picked from hydroid colonies, nudibranch populations will be dominated by small individuals.

Competition may be limited by prey switching as the nudibranchs get larger. Prey switching can reduce competition by decreasing feeding pressure in the habitat where veligers are most likely to settle. Dendronotus switches to the hydroid Tubularia spp. w hen large (> 10 mm) (Swennen, 1961; Thompson and

Brown, 1984). Kuzirian (unpub. data) showed that Doto feeds on other hydroids as adults (>3 mm). The majority of Doto found on the hydroid

Thularla a r gen tea were greater than 3 mm over 4 years (1972-1975) (Kuzirian, unpub. data) whereas 71.3% of Doto in Q. geniculata were less than 3 mm

(Chapter 2).

A general assumption in ecology has been that similar species of animals will compete most intensely (MacArthur. 1958; Miller. 1967). This is not supported by this study: hydroid eating nudibranchs do not interact at a level

152 where competition could be a major force structuring the community. In contrast, the aeolid Hermissenda crassicornis on the Pacific coast is a predator of hydroids and nudibranchs in colonies of thecate hydroids. Both small aeolid and dendronatacean nudibranchs actively avoid H. crassicornis by occupying peripheral areas of hydroid colonies (Harris, pers. comm.). The association between nudibranchs and hydroids appears very similar to that of insects and plants (Strong, et al.. 1984; Howe and Westley, 1988), Mechanisms regulating insects communities, such as predation, colonization and patchily distributed food, are likely to be operating in the nudibranch community in colonies of O. geniculata.

153 Figure 4-1. Ethogram of behaviors involving Dendronotus frondosus before and after an encounter with other nudibranchs on colonies of Obelia geniculata. (T = touch, Ta = taste. Cl = climb, Bri = bristle cerata, Av = avoid/aversion, NR = no reaction)

154 % Total Interactions - 0 4 20 - 0 3 10 0 - - -

-R -T Av -r aC TaAv iNr vi T- r Ta-N Avoid r ri-N B v a-A T Ta-Cl T-Cr v -A T T-CT T-NR Behaviors 155 Dendronotus (#*11) Doto(n=34) Tergipes (n=33)

Figure 4-2. Ethogram of behaviors of Dendronotus frondosus initiating an encounter with another nudibranch and reaction of other nudibranchs. fT = touch, Ta = taste. Bri = bristle cerata, Cr = cringe. Av = avoid/aversion. NR = no reaction)

156 % Total Interactions 100 -R Av Br T-R T Ta- iAv ri-A B v -A a T -T T Ta-NR ri -B T v -A T T-NR 157 Behaviors Dendronotus n= ( 11) Doto(n=28) Tergipes (n-26)

Figure 4-3. Ethogram of behaviors of Doto coronata before and after an encounter with other nudibranchs on colonies of Obelia geniculata. (T = touch. Ta = taste. Cl = climb. Av = avoid/aversion. NR = no reaction)

158 * 0 U) 04 -i -i u — Total Interactions o o o o o o o %

T-NR T-Cl T-Av Ta-NR T a -A v

Behaviors 159 i I

Figure 4-4. Ethogram of behaviors of Doto coronata initiating an encounter with another nudibranch and reaction of other nudibranchs. (T = touch. Ta = taste, Cr = cringe, Av = avoid/aversion, NR = no reaction)

160 H u 9 IS) e e 9 «* A, 1 a H** || n t o II mmt (* H 9 1 9 9 w V tO II a * o o 9 ■ n % Total% Interactions

Behaviors 161

I

i | Figure 4-5. Ethogram of behaviors of Tergipes Tergipes before and after an encounter with other nudibranchs on colonies of Obelia geniculata. (T = touch, Ta = taste. Cl = climb, Bri = bristle cerata. Cr = cringe. Av = avoid /aversion, NR = no reaction)

162 tn ro % 7 u i a * __ *■ i ? c © ~ n ■H ■H O tt O ft O 2 ii ~ £ ■ ■ I <0 © Gk ___ i— i ___ i ____ I— i — r o w __ Total Interactions 4— * o o o o o o %

T-NR T-Bri T-Cl T-Av T- TV NR T*-Av Bri-NR Avoid

Behaviors 163

I I I Figure 4-6. Ethogram of behaviors of Tergipes tergipes initiating an encounter with another nudibranch and reaction of other nudibranchs. (T = touch. Ta = taste. Bri = bristle cerata. Av = avoid/aversion. NR = no reaction)

164 % Total Interactions Behaviors

0 5 Ol Figure 4-7. Ethogram of behaviors of Eubranchus exiguus before and after an encounter with other nudibranchs on colonies of Obelia geniculata. (T = touch. Ta = taste. Av = avoid/aversion. NR = no reaction)

166 % Total Interactions 100 T-NR Av aC Avoid A Ta-Cl v -A T Behaviors 167 Reactions of others to of others Reactions (n«=33) Eubranchus urnhs n- ) 2 -3 (n Eubranchus EZ - r Ta-NR Cr T- z B t Q C ----

Figure 4-8. Ethogram of behaviors of Eubranchus exiguus initiating an encounter with another nudibranch and reaction of other nudibranchs. (T = touch. Ta = taste. Cl = climb. Cr = cringe. Av = avoid/aversion. NR = No reaction)

1 6 8 % Total Interactions T-NR Av aAv Ta-A v -A T Behaviors 169 Other Nudibranchs Other Reactions of EubranchusReactionsof to others (n= 11) (n= others to mm (n-11) Avoid

Figure 4-9. Mean height (+SE) occupied by nudibranchs on hydrocauli in pair-w ise intra- [AA.BB) an d inter-specific (AB.BA) treatm en ts. Designations of before and after refer to an addition of individuals to manipulate densities of nudibranchs. No significant differences were found for any species pair.

170 Height (mm) 10 12H 10 12 0 10 12 2 4- 6 8 0 2H 4 6 -8 - - ■ - -

- - m fter A ■ Before ^ 171 A Tergipes/Dendronotus T " T BA Dendronotus/Doto Doto/Eubronchus Tergipes/Doto BB Table 4-1. Summary of patterns of spacing in interspecific associations of nudibranchs on colonies of Obelia. Values are mean distances (mm) between any two nearest neighbors. (* Distances between individuals from monospecific trials.)

Species A Species B Distance (mm) between p individuals (SE)

1. Tergipes Tergipes 14.2 (2.2) NS Tergipes Dendronotus 16.2 (4.0) * Dendronotus Dendronotus 10.2 (1.8)

2 . Tergipes Tergipes 10.5(1.1) NS Tergipes Doto 11.8(4.0) * Doto Doto 6.7 (1.9)

3. Dendronotus Dendronotus 11.7(1.6) NS Dendronotus Doto 11.2(1.2) * Doto Doto 7.8 (1.7)

4. a. Doto Doto 6.4 (2.9) a:b 0.027 b. Doto Eubranchus 15.1 (2.1) c. Eubranchus Eubranchus 21.0(3.7) b:c NS

172 Table 4-2. Patterns of interspecific dispersion of nudibranch species on colonies of Obelia. Designations of before (B) and after (A) refer to an addition of individuals to manipulate densities of nudibranchs in pair-wise experiments. Patterns of dispersion were determined by methods described in Clark and Evans (1954).

Species Mean Distance (mm) Pattern of between Individuals (SE) Dispersion

Before After Before After

Dendronotus 13.3 13.0 Random Random - (1.8) (2.1) Regular Doto 11.2 2.5 Random Random- (1.5) (0.9) Clum ped

E u b ran ch u s 12.2 10.4 R andom No C hange (4.9) (0.4)

Tergipes 14.1 13.3 Random No C hange (1.9) (1.8)

1 7 3 SUMMARY AND CONCLUSIONS

Kelp blades (Laminaria spp.) with colonies of Obelia geniculata were

collected over a 28 month period to document the epifaunal community

inhabiting the hydroid colony and to determine the possible mechanisms

allowing the four nudibranch species present to coexist. Regression analyses

correlated the abundances of epifauna with four habitat parameters: kelp blade size (cm^). hydroid colony size (cm^), density of hydrocauli (per cm^l and height of hydrocauli (mm). The feeding biology of the four nudibranch species inhabiting O. geniculata was described by the following criteria: morphology of the radula, feeding behavior and location of the nudibranchs within the hydroid colony. Density manipulations of nudibranchs in pair-wise behavioral interactions tested the effects of interference on the location of nudibranchs in the colony.

Colonies of Obelia geniculata provide habitat and food for many invertebrate epifauna (Chapter 1,2). The hydroid colony is the effective island for most species: higher abundances of all motile species except flatworms and

Dendronotus frondosus were found on kelp blades with larger hydroid colonies. The size of colonies of the ectoproct Membranipora membranacea was negatively associated with hydroid colony size but positively correlated to the size of the kelp blade. This pattern showed that larvae of M- membranacea settle on the largest substrate available that is clear of other sessile epifauna.

Hydrocauli in colonies of Q. geniculata provide structure to motile epifauna.

Denser and taller colonies generally have higher abundances of motile epifauna.

Colonies of Q. geniculata are a nursery habitat for nudibranchs (Chapter

2). Adults of Dendronotus frondosus and Doto coronata were infrequent

174 inhabitants of the hydroid colony. Eubranchus exiguus was found less

frequently, but juveniles were more abundant than adults. Greater

abundances of Eubranchus are found in other hydroid colonies (Lambert.

1985; Kuzirian, unpub. data), thus Q. geniculata seems to be a fringe habitat

for Eubranchus. Q. geniculata is the primary habitat of Tergipes tergipes.

Adults and egg masses of Tergipes were found in most months of the census.

Tergipes was the only nudibranch that consistently occupied the hydroid colony for its entire life cycle.

Each nudibranch species utilizes a separate portion of the hydroid colony as food and habitat (Chapter 3). Dendronotus feeds by penetrating the perisarc of thecae or stolons or by biting off whole polyps. This difference in feeding preference is size dependent: small nudibranchs (<5 mm) drill through perisarc and large nudibranchs (>5 mm) bite polyps. Both Doto and

Eubranchus feed by penetrating the hydroid perisarc and suctioning out tissue, but they feed upon different structures. Doto drills through stolons, while Euhranchus penetrates hydrothecae. Lastly, Tergipes attacks and rasps naked tissue of polyps. The feeding preferences of the nudibranchs correlated with the structure of their radulae suggests that the radula is an accurate predictor of how a nudibranch feeds when the animal is small, but at larger sizes (>5 mm) the jaws play a more important role in feeding.

Encounters among the nudibranchs occur frequently but do not appear to influence the location of nudibranchs within the colony or where nudibranchs feed (Chapter 4). The vast majority of all pair-wise intra- and interspecific encounters between nudibranchs were non-aggressive.

Displacement from an area by a heterospeciflc was infrequent. This suggests that nudibranchs are unaware of each other, similar to many insect communities (Strong, et al.. 1984; Howe and Westley, 1988).

175 Although some separation of the hydroid food resource is present within the hydroid colony, resource partitioning is unlikely to be a major factor allowing coexistence of the nudibranchs. Interference and competition are Loo rare or too weak to cause the observed patterns, although these patterns could be a consequence of the ghost of competition past (Connell. 1980). Only one nudibranch. Tergipes uses colonies of Q. geniculata as a primary habitat.

Both Dendronotus and Doto use the habitat primarily as a nursery and prey switching by these nudibranchs when larger facilitates coexistence.

Eubranchus is found more abundantly on other hydroids (Lambert. 1985;

Kuzirian, unpub. data) and Q. geniculata seems to be a alternative food resource. Differential settlement and recruitment by all four nudibranch species can overwhelm the hydroid at times but equilibrium conditions necessary for exclusion are unlikely to occur or persist for long periods.

Therefore, resource partitioning need not be invoked to explain the continued coexistence of these predators.

176 LITERATURE CITED Allmon. RA. and K.P. Sebens. 1988. Feeding biology and ecological impact of an introduced nudibranch. plebia. New England. USA. Mar. Biol. 99:375-385.

Arnold, S.J. 1972. Species densities of predators and their prey. Am. Nat. 106:220-236.

Bartholomew. G.A. and B. Heinrich. 1978. Endothermy in African dung beetles during flight, ball making and ball rolling. J. Exp. Biol. 73:65- 83.

Bayne. B.L. 1964. Primary and secondary settlement in Mvtilus edulis L. (). J. Anim. Ecol. 33:513-523.

Bayne, B.L. 1965. Growth and delay of metamorphosis of the larvae of Mvtilus edulis. Ophelia. 2:1-47.

Behrens. D.W. 1980. Pacific Coast Nudibranchs. Sea Challengers. Los Osos. CA. 112 pp.

Berman. J. 1989. The ecology of biological invasions: Interactions between native and introduced salt marsh gastropods. MS Thesis. University of Oregon.

Best, B.A. and J.E. Winston. 1984. Skeletal strength of encrusting cheilostome bryozoans. Biol. Bull. 167:390-409.

Birch, L.C. 1979. The effect of species of animals which share common resources on one another's distribution and abundance. Fortschur. Zool. 25:197-221.

Blinn. D.W.. RE. Truitt and A. Pickart. 1989. Feeding ecology and radular morphology of the freshwater limpet Ferrissia fragilis. J. N. Am. Benthol. Soc. 8:237-242

Bloom, SA 1976. Morphological correlations between dorid nudibranch predators and sponge prey. Veliger 18:289-301.

Bloom. S A 1981. Specialization and noncompetitive resource partitioning among sponge-eating dorid nudibranchs. Oecologia. 49:305-315.

Boero. F. 1981. Systematics and ecology of the hydroid population of two Posidonla oceanica meadows. Mar. Ecol. 2:181-197.

Boero, F. and E. Fresi. 1986. Zonation and evolution of a rocky hydroid community. Mar. Ecol. 7:123-150.

Braams, W.G. and H.F.M. Geelen. 1953. The preference of some nudibranchs for certain coelenterates. Arch. Neerland. de Zool. 10:241-264.

Braverman, M.H. 1963. Studies on hydroid differentiation. II. Colony growth and the initiation of sexuality. J. Embryol. exp. Morph. 11:239-253.

178 Braverman, M.H. 1971. Studies on hydroid differentia lion. VI. Regulation of hydroid formation in Podocorvne cam ea. J. Exp. Zool. 176:361-382.

Brenchley, G.A. and J.T. Carlton. 1983. Competitive displacement of native mud snails by introduced periwinkles in the New England intertidal zone. Biol. Bull. 175:543-558.

Bronmark, C. 1989. Interactions between epiphytes, macrophytes and freshwater snails: A review. J. Moll. Stud. 55:299-311.

Bros. W.E. 1987a. Temporal variation in recruitment to a fouling community in Tampa Bay. FL. J. Coast. Res. 3:499-504.

Bros. W.E, 1987b. Effects of removing or adding structure (barnacle shells) on recruitment to a fouling community in Tampa Bay. Florida. J. Exp. Mar. Biol. Ecol. 105:275-296.

Brown, J.S. 1989. Coexistence on a seasonal resource. Am. Nat. 133:168- 182.

Burke. R.D. 1983. The induction of metamorphosis of marine invertebrate larvae: Stimulus and response. Can. J. Zool. 61:1701-1719.

Bynum, K.H. 1978. Reproductive biology of Caprella penantis Leach. 1814 (Amphipoda:Caprellidae) in North Carolina. USA Est. Coast. Mar. Sci. 7:473-485.

Caine, E_A 1978. Habitat adaptations of North American caprellid amphipods (Crustacea). Biol. Bull. 155:288-296.

Calder. D.R. 1970. Thecate hydroids from the shelf waters of northern Canada. J. Fish. Res. Bd., Canada. 27:1501-1547.

Carriker, M. R. 1969. Excavation of boreholes by the gastropod. Urosalpinx: An analysis by light and scanning electron microscopes. Am. Zool. 9:917-933.

Carriker, M.R. 1981. Shell penetration and feeding by Naticacean and Muricacean predatory gastropods: A Review. Malacologla. 20:403-422.

Casola, E., M. Scardi, L. Mazzella and E. Fresi. 1987. Structure of the epiphyte community of Posidonia oceanica leaves in a shallow meadow. Mar. Ecol. 8:285-296.

Cates. R.G. 1980. Feeding patterns of monophagous. oligophagous and polyphagous insect herbivores: The effect of resource abundance and plant chemistry. Oecologia. 46:22-31.

Chesson. P.L. 1986. Environmental variation and the coexistence of species. In: Diamond. J. and T.J. Case (eds.). Community Ecology. Harper and Row Publ.. Inc.. New York, pp 240-256.

179 Chesson, P.L. and T.J. Case. 1986. Overview: Nonequilibrium community theories: Chance, variability, history and coexistence. In: Diamond. J. and T.J. Case (eds.). Community Ecology. Harper and Row Publ.. Inc.. New York, pp 229-239.

Chesson. P.L. and R.R. Warner. 1981. Environmental variability promotes coexistence in lottery competitive systems. Am. Nat. 117:923-943.

Chew. F.S. 1981. Coexistence and local extinctions in two pierid butterflies. Am. Nat. 118:655-672.

Christensen, H. 1977. Feeding and reproduction in Precuthona peachi (Mollusca, Nudibranchia). Ophelia. 16:131-142.

Clark, K.B. 1975. Nudibranch life cycles in the Northwest Atlantic and their relationship to the ecology of fouling communities. Helgol. wiss Meeres. 27:28-69.

Clark, P.J. and F.C. Evans. 1954. Distance to nearest-neighbor as a measure of spatial relationships in a population. Ecology. 35:445-453.

Cody, M.L. 1968. On the methods of resource division in grassland bird communities. Am. Nat. 102:107-147.

Cody. M.L. and J.M. Diamond. 1975. Ecology and Evolution of Communities. Harvard Univ. Press. Cambridge, MA. 545 pp.

Connell, J. 1980. Diversity and the coevolution of competitors, or the ghost of competition past. Oikos. 35:131-138.

Cornelius, P.F.S. 1975. The hydroid species of Obelia (Coelenterata, :). with some notes on the medusa stage. Bull. Br. Mus. nat. Hist. (Zool.). 28:251-293.

Cornelius. P.F.S. 1982. Hydroids and medusae of the family Campanulariidae recorded from the eastern North Atlantic, with a world synopsis of genera. Bull. Br. Mus. nat. Hist. (Zool.). 42:37-148.

Coyer, J.A, 1984. The invertebrate assemblage associated with the giant kelp. Macrocvstls pvrtfera. at Santa Catalina Island. California: A general description with emphasis on amphipods, copepods. mysids and shrimps. Fish. Bull., US. 82:55-66.

Crowell. S. 1957. Differential responses of growth to nutritive level, age and temperature in the colonial hydroid Campanularia. J. Exp. Zool. 134:63-90.

Creese. R.G. 1988. Ecology of molluscan grazers and their interactions with marine algae in north-eastern New Zealand: A review. New Zealand J. Mar. Freshwater Res. 22:427-444.

Day. R.W. and G.P. Quinn. 1989. Comparisons of treatments after an analysis of variance in ecology. Ecol. Monogr. 59:433-463.

180 Dean, T.A. 1981. Structural aspects of sessile invertebrates as organizing forces in an estuatine fouling community. J. Exp. Mar. Biol. Ecol. 53:163-186.

Denslow, J.S. 1984. Disturbance-mediated coexistence of species. In: Pickett. S. and P. White, (eds.). The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Inc., New York, pp 307-323.

Dial. K.P. 1988. Three sympatric species of Neotoma: Dietary specialization and coexistence. Oecologia. 76:531-537.

Diamond, J. and T.J. Case (eds.). 1986. Community Ecology. Harper and Row Publ., Inc., New York. 665 pp.

Donn. T.E. 1983. Population ecology and production of the sandy beach amphipod Haustorius canadensis (Crustacea:Haustoriidae). Ph.D. Dissertation, University of New Hampshire. 103 pp.

Drury. R.A.B. and E.A. Wallington. 1980. Carleton's Histological Techniques. Oxford Univ. Press. Oxford. 520 pp.

Edgar, G.J. 1983a. The ecology of south-east Tasmanian phytal animal communities. II. Seasonal change in plant and animal populations. J. Exp. Mar. Biol. Ecol. 70:159-179.

Edgar, G.J. 1983b. The ecology of south-east Tasmanian phytal animal communities. III. Patterns of species diversity. J. Exp. Mar. Biol. Ecol. 70:181-203.

Edmunds. M. and A. Kress. 1969. On the European species of Eubranchus (Mollusca. Opisthobranchia). J. mar. biol. Assoc., U.K.. 49:879-912.

Elton, C.S. 1927. Animal Ecology. Sidgwick and Jackson, London.

Fernandez, E., R. Anadon and C. Fernandez. 1988. Life histories and growth of the gastropods Blttium reticula turn and Barleeia unifascia ta inhabiting the seaweed Gelidium latlfollum. J. Moll. Stud. 54:119-129.

Fine, M.L. 1970. Faunal variation on pelagic Sargassum. Mar. Biol. 7:112- 122.

Foster, M.S. 1975. Regulation of algal community development in a Macrocvstis pyrtfera forest. Mar. Biol. 32:331-342.

Fralick, R.A., K.W. Turgeon and A.C. Mathieson. 1974. Destruction of kelp populations bv Lacuna vincta (Montagu). Nautilus. 88:112-114.

Franz, D.R. 1975. An ecological interpretation of nudibranch distribution on the northwest Atlantic. Veliger. 18:79-83.

Fraser. D.F. 1976. Empirical evaluation of the hypothesis of food competition on salamanders of the Plethodon. Ecology. 57:459-471.

181 Gaulin. G.. L. Dill, J. Beaulieu and L.G. Harris. 1986. Predator induced changes in growth form in a nudibranch-hydroid association. Veliger 28:389-393.

Grant. P.R 1966. Ecological compatibility of bird species on islands. Am. Nat. 100:451-462.

Grant, P.R 1969. Colonization of islands by ecologically dissimilar species of birds. Can J. Zool. 47:41-43.

Grant. P.R 1986. Interspecific competition in fluctuating environments. In: Diamond, J. and T.J. Case (eds.). Community Ecology. Harper and RowPubl., Inc., New York, pp 173-191.

Hadfield, M.G. 1984. Settlement requirements of molluscan larvae: New data on chemical and genetic roles. Aquaculture. 39:283-298.

Hagerman, L. 1966. The macro- and microfauna associated with Fucus serratus L., with some ecological remarks. Ophelia. 3:1-43.

Hall, M.O. and S.S. Bell. 1988. Response of small motile epifauna to complexity of epiphytic algae on seagrass blades. J. Mar. Res. 46:613-630.

Harris, L.G. 1973. Nudibranch associations. In: Cheng, T.C. (ed.). Current Topics in Comparative Pathobiology, Vol. 2. Academic Press, NY. pp 213-315.

Harris, L.G. 1986. Size-selective predation in a sea anemone, nudibranch and fish food chain. Veliger. 29:38-47.

Harris, L.G. 1987. Aeolid nudibranchs as predators and prey. Amer. Malacolog. Bull. 5:287-292.

Harris. L.G. and K.P. Irons. 1982. Substrate angle and predation as determinants in fouling community succession. In: J. Cairns (ed). Artificial Substrates. Ann Arbor Publ., Inc., Ann Arbor. MI. pp 1331- 174.

Harris, RJ. 1985. A Primer of Multivariate Statistics. 2nd ed.. Academic Press, O rlando, FL. 5 7 6 pp.

Harvell, D. 1984. Why nudibranchs are partial predators: Intracolonial variation in biyozoan palatabiiity. Ecology. 65:716-724.

Hawkins, S.J., D.L. Watson, A.S. Hill, S.P. Harding, M.A. Kyriakides. S. Hutchinson and T.A. Norton. 1989. A comparison of feeding mechanisms in microphagous, herbivorous, intertidal prosobranchs In relation to resource partitioning. J. Moll. Stud. 55:151-165.

Hayward, P.S. 1980. Invertebrate epiphytes of coastal marine algae. In: Price, J.H., D.E.G. Irvine and W.F. Farnham (eds.). The Shore Environment, 2: Ecosystems. Academic Press, London, pp 761-787.

182 Heck, K.L., Jr. and R.J. Orth. 1980. Seagrass habitats: The roles of habitat complexity, competition and predation in structuring associated fish and motile macroinvertebrate assemblages. In: V.S. Kennedy (ed.). Estuarine Perspectives. Academic Press, pp. 449-464.

Hinde,. 1958. Food and habitat selection in birds and lower vertebrates. 15th Intemat. Congress Zool., Sect. 10. 18:1-2.

Howe, H.F. and L.C. Westley. 1988. Ecological Relationships of Plants and Animals. Oxford Univ. Press. 276 pp.

Hughes. R.G. 1975. The distribution of epizoites on the hydroid Nemertesia antennina (L.). J. mar. biol. Assoc.. U.K. 55:275-294.

Hughes. RN. 1980. Predation and community structure. In: J.H. Price, D.E.G. Irvine and W.F. Famham (eds.). The Shore Environment. Vol. 2: Ecosystems. Academic Press, London, pp. 699-728.

Hughes. RN. 1986. A Functional Biology of Marine Gastropods. Croom Heelm, Australia. 245 pp.

Hutchinson. G.E. 1958. Homage to Santa Rosalia or why are there so many kinds of animals? Am. Nat. 93:145-159.

Hyman. L. 1967. The Invertebrates, vol. VI. Mollusca I. McGraw-Hill Book Co.. New York. 792 pp.

Jackson. J.B.C. 1977. Habitat area, colonization and development of epibenthic community structure. In: B.F. Keegan, P.O. Ceidigh and P.J.S. Boaden (eds.). Biology of Benthic Organisms. Pergamon Press. London, pp. 349-358.

Karieva, P. 1982. Exclusion experiments and the competitive release of insects feeding on collards. Ecology. 63:696-704.

Keddy, P.A. 1989. Competition. Chapman and Hall, New York. 202 pp.

Keith, D.E. 1971. Substrate selection in Caprellid amphipods of southern California, with emphasis on Caprella califomica and Caprella equilibra Say (Amphipoda). Pacific Science. 25:387-394.

Kohn. A.J. 1983. Feeding biology of gastropods. In: Wilbur, K.M. (ed.). The Mollusca, Vol.5, Physiology, Part 2. Academic Press. New York, pp 1- 63.

Kotler, B.P. and J.S. Brown. 1988. Environmental heterogeneity and the coexistence of desert rodents. Ann. Rev. Ecol. Syst. 19:281-307.

Krebs. C.J. 1989. Ecological Methodology. Harper and Row Publ.. New York. 654 pp.

Kuzirian. A.M. 1979. and biology of four New England coryphellid nudibranchs (: Oplsthobranchia). J. Moll. Stud. 45:239- 261.

183 L'Hardy. J.P. 1962. Observations sur le peuplement epiphyte des lames Lamlnaria saccharina (Linne) Lamouroux. en Baie de Morlai (Finistere). Cah. Biol. Mar. 3:115-127.

Lambert. W.J. 1985. The Influence of predators on early colonists in a fouling community. MS Thesis. Univ. of New Hampshire. 72 pp.

Lambert. W.J. and D.R Laur. 1987. Observations on the association of a hydroid. its specialist predator and associated epibionts. East Coast Benthic Ecology Meetings, Raleigh, NC.

Lawton, J.H. 1982. Vacant niches and unsaturated communities: A comparison of bracken herbivores at sites on two continents. J Anim. Ecol. 51:573-595.

Lawton. J.H. and M.P. Hassel. 1981. Asymmetrical competition in insects. Nature. 289:793-795.

Lawton, J.H. and D.R. Strong, Jr. 1981. Community patterns and competition in folivorous insects. Am Nat. 118:317-338.

Lillie, R.D. 1954. Histopathologic Technic and Practical Histochemistry, 2nd edition. Blakiston. New York.

Lubchenco. J. 1978. Plant species diversity in a marine intertidal community: Importance of herbivore food preference and algal competitive abilities. Am. Nat. 112:23-39.

MacArthur. R.H. 1958. Population ecology of some warblers of northeastern coniferous forests. Ecology. 39:599-619.

MacArthur. R and R Levins. 1967. The limiting similarity, convergence and divergence of coexisting species. Am. Nat. 101:377-385.

Martin, P.D. 1988. The ecology of caprellid amphipods: Population patterns and the role of algal complexity in mediating predation by wrasse. MS Thesis, Univ. of New Hampshire. 63 pp.

McBeth, J.W. 1971. Studies on the food of nudibranchs. Veliger. 14:158-161.

Meadows. P.S. and J.I. Campbell. 1972. Habitat selection by aquatic invertebrates. Adv. Mar. Biol. 10:271-382.

Meyer. K.B. 1971. Distribution and zoogeography of fourteen species of nudibranchs of northern New England and Nova Scotia. Veliger. 14:137-152.

Miller, M.C. 1961. Distribution and food of the nudibranchiate Mollusca of the south of the Isle of Man. J. Anim. Ecol. 31:95-116.

Miller, M.C. 1962. Annual cycles of some Manx nudibranchs. with a discussion of the problem of migration. J. Anim. Ecol. 31:545-569.

Miller, RS. 1967. Pattern and process in competition. Adv. Ecol. Res. 4:1-74.

184 Mills. P.M. 1977. Radular tooth structure in three species of Terebridae (Mollusca:Toxoglossa). Veliger. 19:259-265.

Morse. D.H. 1980. Behavioral Mechanisms in Ecology. Harvard Univ. Press. C am bridge. MA. 383 pp.

Naeem. S. 1988. Resource heterogeneity fosters coexistence of a mite and a midge in pitcher plants. Ecol. Monogr. 58:215-227

Nation, J.L. 1983. Specialization in the alimentary canal of some mole crickets (Orthoptera:Giyllotalpidae). Int. J. Insect. Morph. Embryol. 12:201- 210.

Nixon, M. 1979a. Hole boring in shells by Octopus vulgaris Cuvier in the Mediterranean. Malacologia. 18:431-433.

Nixon. M. 1979b. Has Octopus vulgaris a second radula? J. Zool., London. 187:291-296.

Nixon, M. and E. Maconnachie. 1988. Drilling by Octopus vulgaris (Mollusca:Cephalopoda) in the Mediterranean. J. Zool.. London. 216:687-716.

Nybakken. J. 1970. Correlation of radula tooth structure and food habits of three vermivorous species of Conus. Veliger. 12:316-318.

Nybakken. J. 1978. Abundance, diversity and temporal variability in a California intertidal nudibranch assemblage. Mar. Biol. 45:129-146.

Nybakken, J.W. and J. Eastman. 1977. Food preference, food availability and resource partitioning in Triopha maculata and Triopha carpenter! (Opisthobranchia, Nudibranchia). Veliger. 19:279-289.

Nybakken. J. and G. McDonald. 1981. Feeding mechanisms of west American nudibranchs feeding on Bryozoa, and Ascidiacea. with special respect to the radula. Malacologia. 20:439-449.

Orth. R.J., K.L. Heck. Jr. and J. van Montfrans. 1984. Faunal communities n seagrass beds: A review of the influence of plant structure and prey characteristics on predator-prey relationships. Estuaries. 7:339-350.

Osenberg, C.W. 1989. Resource limitation, competition and the influence of Sfe history in a freshwater snail community. Oecologia. 79:512-519.

Osman. R.W. 1978. The influence of seasonality and stability on the species equilibrium. Ecology. 59:383-399.

Pequegnat, W.E. 1964. The epifauna of a California siltstone reef. Ecology. 45:272-283.

Persson, L. 1985. Asymmetrical competition: Are larger animals superior? Am. Nat. 126:261-266.

Polyscience. Inc. 1985. Data sheet # 382.

185 Pyke, G., H.R. Pulliam and E.L. Chamov. 1977. Optimal foraging: A selective review of theory and tests. Quart. Rev. Biol. 52:137-154.

Race, M.S. 1982. Competitive displacement and predation between introduced and native mud snails. Oecologia. 54:337-347.

Radwin. G.E. and H.W. Wells. 1968. Comparative radular morphology and feeding habits of muricid gastropods from the Gulf of Mexico. Bull. Mar. Sci. 18:72-85.

Rathcke. B.J. 1976. Competition and coexistence within a guild of herbivorous insects. Ecology. 57:76-87.

Robilliard. G.A. 1970. The systematics and some aspects of the ecology of the genus Dendronotus (Gastropoda:Nudibranchia). Veliger. 12:433-479.

Root, R.B. 1973. Organization of a plant-arthropod association in simple and diverse habitats: The fauna of collards (Brassica oleraceal. Ecol. Monogr. 43:95-124.

Roughgarden. J. and J. Diamond. 1986. Overview: The role of species interactions in community ecology. In: Diamond, J. and T.J. Case (eds.). Community Ecology. Harper and Row Publ.. Inc.. New York, pp 333-344.

Ryland, J.S. 1974. Observations on some epibionts of gulfweed Sargassum natans (L.) Meyen. J. Exp. Mar. Biol. Ecol. 14:17-25.

Schmekel, L. 1985. Aspects of evolution within the Opisthobranchia. In: Trueman. E.R. and M.R. Clarke (eds.). The Mollusca. vol. 10. Evolution. Academic Press, NY. pp. 221-267.

Schmekel, L. and A. Portmann. 1982. Opisthobranchia des Mittelmeeres. Springer-Verlag, Berlin and New York.

Schmidt, G.H. 1982. Random and aggressive settlement in some sessile marine invertebrates. Mar. Ecol. Prog. Ser. 9:97-100.

Schoener. T.W. 1968. The Anolis lizards of Bimini: Resource partitioning on complex fauna. Ecology. 49:704-726.

Schoener, T.W. 1974. Resource partitioning in ecological communities. Science. 185:27-39.

Schoener, T.W. 1982. The controversy over interspecific competition. American Scientist. 70:586-595.

Schoener, T.W. 1983. Field experiments on interspecific competition. Am. Nat. 122:240-285.

Seed. R. 1986. Ecological pattern in the epifaunal communities of coastal macroalgae. In: Morre, P.G. and R. Seed (eds.). The Ecology of Rocky Coasts. Columbia Univ. Press. NY. pp 22-35.

186 Seed. R.. M.N. Elliott. P.J.S. Boaden and R.J. O'Connor. 1981. The composition and seasonal changes amongst the epifauna associated with Fucus serratus L. in Strangford Lough, northern Ireland. Cah. Biol. Mar. 22:243-266.

Seed, R. and R.J. O’Connor. 1981. Community organization in marine algal epifauna. Ann. Rev. Ecol. Syst. 12:49-74.

Shonman, D. and J.W. Nybakken. 1978. Food preferences, food availability and food resource partitioning in two sympatric species of cephalispidean opisthobranchs. Veliger. 21:120-126.

SimberlofT, D.S. 1978. Using island biogeographic distributions to determine if colonization is stochastic. Am. Nat. 112:713-726.

Sloane, J.F.. F.J. Ebling, J.A. Kitching and S.J. Lilly. 1957. The ecology of Lough Ine rapids with special reference to water currents. V The sedentary fauna of the Laminarian algae in the Lough Ine area. J. Anim. Ecol. 26:197-211.

Standing. J. 1976. Fouling community structure: Effects of the hydroid, Obelia dlchotoma. on larval recruitment. In: G.O. Mackie (ed.). Coelenterate Ecology and Behavior. Plenum Press. New York. pp. 155- 164.

Stebbing, A.R.D. 1973. Competition for space between the epiphytes of Fucus serratus L.. J. mar biol. Assoc., U.K. 53:247-261.

Strong, D.R., Jr. 1982. Harmonious coexistence of hispine beetles on Heliconia in experimental and natural communities. Ecology. 63:1039-1049.

Strong, D.R., J.H. Lawton and T.RE. Southwood. 1984. Insects on Plants: Community Patterns and Mechanisms. Blackwell Publ., Oxford. 313 pp.

Sutherland. J.P. and R.H. Karlson. 1977. Development and stability of the fouling community at Beaufort, North Carolina. Ecol. Monogr. 47:425- 448.

Swennen. C. 1961. Data on distribution, reproduction and ecology of the nudibranchiate molluscs occurring in the Netherlands. Neth. J. Sea. Res. 1:191-240.

Thompson, T.E. 1964. Grazing and the life cycles of British nudibranchs. In: Crisp, D.J. (ed.). Grazing in Terrestrial and Marine Environments. Blackwell, Oxford. England, pp 275-297.

Thompson. T.E. 1976. Biology of Opisthobranch Molluscs, Vol. I. The Ray Society. London. 207. pp.

Thompson, T.E. and G.H. Brown. 1984. Biology of Opisthobranch Molluscs, Vol. II. The Ray Society, London. 229 pp.

187 Thorson. G. 1950. Reproduction and larval ecology of marine benthic invertebrataes. Biol. Rev. 25:1-45.

Todd. C.D. 1981. The ecology of nudibranch molluscs. Oceanog. Mar. Biol. Ann. Rev. 19:141-234.

Todd. C.D. and R.W. Doyle. 1981. Reproductive strategies of marine benthic invertebrates: A settlement-timing hypothesis. Mar. Ecol. Prog. Ser. 4:75-83.

Todd. C.D. and J.N. Havenhand. 1989. Nudlbranch-biyozoan associations: The quantification of ingestion and some observations on partial predation among Doridoidea. J. Moll. Stud. 55:245-259.

Tokeshi. M. amd C.R. Townsend. 1987. Random patch formation and weak competition: Coexistence in an epiphytic chironomid community. J. Anim. Ecol. 56:833-845.

Vance. R.R. 1984. Interference competition and the coexistence of two competitors on a single limiting resource. Ecology. 65:1349-1357.

Van Valen. L. 1965. Morphological variation and width of ecological niche. Am. Nat. 99:377-390.

Vimstein, R.W. and R.K. Howard. 1987. Motile epifauna of marine epiphytes in the Indian River Lagoon. Florida. II. Comparisons between drift algae and three species of seagrasses. Bull. Mar. Sci. 41:13-26.

Waters, V. 1966. Feeding ecology and other aspects of the natural history of the nudibranch Eubranchus olivaceus. M.S. Thesis. Univ. of Washington.

Weis, J.S. 1968. Fauna associated with pelagic Sargassum in the Gulf stream. Am. Mid. Nat. 80:554-558.

Welsch, U. and V. Storch. 1973. Comparative Animal Cytology and Histology. Univ. Washington Press, Seattle, WA.

Williams. G.B. 1964. The effects of extracts of Fucus serratus in promoting settlement of larvae of Splrorbis borealis Daudin (Polychaeta). J. mar. biol. Assoc., U.K.. 44:397-414.

Yodzis. P. 1986. Competition, mortality and community structure. In: Diamond. J. and T.J. Case (eds.). Community Ecology. Harper and Row Publ.. Inc.. New York, pp 480-491.

Yoshioka, P.M. 1986. Competitive coexistence of the dorid nudibranchs Doridella steinbergae and Corambe paciftca. Mar. Ecol. Prog. Ser. 33:81-88.

Young, D.K. 1969. Okadala elegans. a tube-boring nudibranch mollusc from the central and west Pacific. Am. Zool. 9:903-907.

188 Zar. J.H. 1984. Biostatical Analysis. 2nd edition. Prentice-Hall. Inc.. New Jersey. 718 pp.

189