Records of lite Western Australian il.1H5eWr! Supplement No. 69: 11-21 (2006).

The population dynamics and feeding preferences of leachii (: ) in northeast Queensland, Australia

Cathryn L. ClaTke James Cook University, Townsville, Queensland 4811, Australia Email: [email protected]

A!:-5tr~ct - Sea h",E'5 (OpisthoDranchi

Key words: Bursatella leach ii, stria/lis, sea , Anaspidea, feeding preference

INTRODUCTION preference for low~i.ntensity wave action and their The single unifying feature of all populations is preference for intertidal algae . In one year, their dynamism. Documenting natural population Plaut et aL (1998) observed a rare algal bloom in fluctuation has become increasingly important in deeper water and oculi/era was found in recent times where the need exists to distinguish greater abundance in association with this bloom natural fluctuations in systems from those caused than those populations in the exposed shallow­ by anthropogenic disturbance. The majority of water habitats. fluctuations in populations occur on a local scale The second hypothesis is that the sea settle (Smith 1996). Aggregations of large numbers of in a range of habitats and migrate to areas of highly animab in a Single area are an extreme example of dense conspecifics for the purpose of breeding. In population fluctuation. These events are referred to California surf-grass beds, Aplysia califomica [arm as population explosions or irruptions (Colgan dense breeding aggregations in summer months 1987; 8uda and l:Zibi 1998; Cote and Reynolds 1998; (Audeskirk 1979). Breeding aggregations of B. Williams et aT. 2001). These aggregations are often leachii during th~ Florida wbter \vere also observed an obvious and highly visible incident to human but in summer aggregations were mostly visitors to the site in question and can have a strong immature (Lowe and Turner 1976). effect on the local and macrophyte The third hypothesis for aggregation is that it communities. occurs where food is localized (Carefoot 1987). In A l'ange of largely anecdotal evidence suggests habitats where food is concentrated sea hares may that some sea hare species exhibit boom-and-bust aggregate incidentally as a result of larvai cycles, forming large aggregations and then settlement or feeding preferences. That is, a large disappearing. TIl.ere are a number of hypotheses as recruitment of sea hares may settle in an area where to "\.-vhy populations of sea hares exhibit this a preferred host alga is high in abundance. In Israel, dynamic. One hypothesis for the cause of A. oculi/era abundance was hlghly correlated with aggregation is a response· to hydrological the variable presence of species of Ulva and conditions. Lowe and Turner (1976) hypothesized Enteromorpha (Plaut et aL 1998). A. californica that aggregations of juvenHe 1,',rere juveniles for example, are primarily found on the caused by the hydrological conditions in subtidal Ploeamillm cartilagineum and Laurencia habitats. In addition, the intensity of water pacifica (Pennings 1991). However, Pawlik (1989) movement in Israel intertidal habitats was demonstrated that A. califomiea settles on a wide negatively related to adult sea hare abundance range of algae and crawls to the preferred species. (Plaut et ~l. 1998). The Occurrence of sea hares in Sites with greater P. cartilageum cover contained intertidal habitats may be a compromise between a higher numbers of It ca1zfam.ica recruits however 12 c.L. Clarke

this characteristic did not explain the varifltion in negatively correlated with macrophyte toughness recruihnent intensity which may be related to larval and calcification in D. auricularia. In addition, the supply (Pennings 1991). The combination of feeding preferences of A. califarnica have been found favourable larval conditions and a high abundance to expand with ontogenetic development (Penni..'lgs of a preferred algae species for competent larvae 199Gb) and this trend ~was thought to be caused by may cause high settlement of juveniles in these the soft weak mouth parts found in younger areas. individuals. The key to this last hypothesis is the feeding Biotic interactions, either competition or preference of the sea hare species under predation, have also rarely been examined as investigation. On the whole, feeding specialists influences on feeding preferences in sea hares. seem to be relatively rare in the marine Competition has been suggested by some authors environment when compared with terrestrial (Willan 1979; Achituv and Susswein 1985) although 5y~5tcms (Lubchenco attd (~aines 1981; 1-1" ay et al. never directly tested. Two studies on the influence 1989). In contrast, most sea hares are relatively of predation on feeding preferences yielded specialized feeders and there is a substantial body contradictory results (Pennings 1t,/90a; Rogers et £11. of work addressing the feeding preferences of 2000, 2002). various species (Carefoot 1967, 1987; Wu 1980; and B. leachii are often Pennings 199Gb; Pennings et al. 1993; Rogers et al. reported to be found in dense feedi..'l.g aggregations 1995; Nagle et al. 1998). These studies had mixed in association with blooms of cyanobacteria results and it can be difficult to sort out the factors (Switzer-Dunlap and Hadfield 1979; Paul and influencing feeding preference in sea hares. Factors Permings 1991). The dynamics of these aggregations such as nutritional value, abundance, secondary has never been quantitatively examined for either chemistry and biotic interactions have previously of these two species. A sole report, Lowe and been the focus of sea hare feeding preference Turner (1976) examined aggregations of B. leachii at studies. Analyses of the nutritional qualities of a beach in Florida but measured the densities of algae tested for sea hare feeding preferences have animals washed up on the shoreline, therefore largely failed to correlate speCific qualities,.. with giving no indication of the natural density of this feeding preferences (e.g., Carefoot 1967, 1970; species. The temporal dynamics of these types of Rogers et ai. 1995). aggregations have not been previously investigated Foraging theory states that feeding on the most and therefore, the underlying cause of population abundant and therefore readily obtainable food irruption in sea hares is unknown. In addition, the sources will require the least energy and thus is feeding preferences of B. leachii have not been most advantageous to the herbivore (Crawley determined for tropical populations. This report 1983). We would expect therefore that sea hares tests the third hypothesis as the cause for the would prefer abundant algae. This is true for some aggrega tions observed in tr

Cleveland Bay is tropical but experiences large aquaculture facility. Sargasslim sp. (Phaeophyta: seasonal variation in rainfall and minor Fucules) ill1d P. pinnata (Rhodophyta) were chosen temperature changes between summer and i"linter for their abundance in areas where B. leachii is months (Lanyon and Marsh 1995). found along the coast of Queensland, although they The study site, Shelly Beach, is a shallow bay "vere not abundant at the site where the anima1s covered by se.agrass beds dominated by three were collected. Red and brown algae were only species, Halodule uninervis, Halopilila (Iva lis, and present at extremely low densities in the study site Zostem caprico171is. The site is located at the northern so Sargassum sp. and P. pilllwta were collected from end of Rowes Bay and faces east towards Magnetic Kissing Point, Rowes Bay, Townsville. Island. 1'~1angrove forest lies at the northem end and For feeding preference assays, 40 individual B. a rocky poir'lt at the south end. Shelly Beach is an leac!ni were collected at low tide from t.lte study site. ideal place to closely examine the population Animals were placed in fresh seawater in lL dynamics of S. 8tril1t21s ;)nd B. lrachii because of the containers and transDorted. back to the laboratorv, rdative Simplicity of the benthic habitat. The Shelly located in the Marine Biology arl.d Aquaculture Beach seagrass beds are never completely dry Research Facility Unit (MARFU) aquarium system because some water is retained even at extremely at James Cook University (JeU). All B. leachii were low tides. Therefore sampling can be completed at placed in a large aquarium (maintenance tank) iow tide as sea hares are not forced to migrate by supplied with continuously flowing filtered dropping water levels. seawatel::,Jsalinity ~33ppt, pH - 8.0, temperature Five sampling trips were made to the Shelly 27~28acJ. Animals were held overnight to allow Beach sea grass beds on 25 June, 7 July, 21 July, 27 acclimatisation before experimentation. Both July, and 3 September 2002. On each trip three 25 x maintenance and experimental tanks ..."ere on a a.SIn belt transects were run parallel to the beach, 12h:12h light:dark diel cycle. yielding an estimate of sea hare density per 12.5m2• Transects were run at approximately 100m from the Feeding assays shore and were always oriented parallel to the Sixteen replicate L5L aquaria supplied with beach to eliminate the effect of tidSllheight and to continuously flowing filtered seawater acted as the keep the measuremen t area within the dense experimental arenas. Pre-weighed, equal amounts sea grass habitat. v\Tithin each transect the sea hare (approximately 2 g wet weight) of each of two species present and the number and length of species of macro algae or cyanobacteria were placed individual sea hares was recorded. TIle sea hare at opposite ends of each experimental arena. For species present was identified according to each two-way choice experiment, eight Coleman (2001). The length of each sea hare was experimental tanks and eight control (no-herbivore) measured by placing the individual on the transect replicates were used. Control tanks were identical tape and allowing the animal a few seconds to start in design and set-up, except for t..l-je absence of the to crawl along it. Once crawling, the distance between the base of the cephalic tentacles and the herbivore. The loss of algae in the control tanks was tip of the tail was recorded. The number of sea hare used to calculate the loss of algae due to autogenic egg masses observed within transects was also change from natural breakdown in the recorded .. although the species of egg mass could experimental tanks. Experimental tanks were not be identified in the field. randomly assigned at the start of each trial to During the five sampling trips, algae for remove the possibility of any single aquaria­ laboratory feeding preference testing were associated bias. A t the cormnencement of a trial, the collected. In total, three species of macroalgae wet weight of each alga in each tank was recorded (Enteromorplia J sp., Sargassum sp., and Pterocladia and a randomly chosen B. /eachii from the pillfwta) and one species of cyanobacterium maintenance tank ,vas placed at the centre of each (Calothrix crustacea) were tested. The latter species experimental tank. Individual sea hares were was the cyanobacterium found in the cyanobacterial allowed to move about the experimental tank and bloom where the sea hares were collected and was feed for a 24-hour period. Five times over the an observed prey item of B. leachii in the field. course of the triat the position of each individual Filamentous C. crustacea was collected directly from sea hare waS recorded and scored as to presence on seagrass beds at Shelly Beach, growing alga 1, alga 2, or on the tank walls (Attractiveness epiphytically on seagrass blades or the benthos. The trial). At the termination of the trial, aB algae was green alga, Enteromorpha sp., was chosen as a removed from the aquaria, patted dry and re­ chemically poor species (Pennings et ai. 1993) and weighed to determine percentage consumption of was collected from the algal scrubbers at the Jeu each alga (Palatability trial).

1 The genus £I/:eromarpha has recentlf been merged with the genus Ulva (rIayden et al. 2003). Throughout the present work 'EI/!t?I'~morpha' has been used to designate the filamentous growth form of the genus U/va. 14 C.L Clarke

Statistical analysis Descriptive length-frequency analysis was used to I·iOB .,eacj h'· 1/ iI examine the population dynamics of the sea hare L!dlS . striatus I population. Length-frequency analysls is a technique that examines cha..n.ges in length of the population, using a series of length histograms. rob Fh Length-frequency distribution reflects the interaction between rates of reproduction, growth, and mortality. Simple descriptive length-frequency analysis is used here to draw basic conclusions o~.~l~IL~21 Jui 27JuI 3 Sep about the characteristics of the population. 1',1ore Sampling date forrn~!l length-frequency techniques could not be uS'2d because the full life 11istory cl"1aracterist1cs of these species have not been tested. The sea hare Figure 1 The mean density (per m1) of Bursatella leachi! lengths were divided into 5mm classes as this and StyiocileiiHs ~iriallis wilhU·\ th" sc;:,grass beds of Shelly Beach in 2002. provided the best resolution of the patterns of recruitment and mortality. One-way ANOVA was transect area on 7lh July 2002. Its density peaked at performed on the length of sea hares found during 0.88 individuals per m 1 and steadily declined in the each trip. following t!':ips until no animals were recorded on 3 For the feeding preference trials, the autogenic September 2002. The mean density of S. striatus change calculated for each alga was used to c;rrect when present (7th, 21 st, and 27th July, 2002) was 0.42 the percentage consumption values (after Nicotri per m 2 (Figure 1). Formal analysis of the length of S. 1980; Peterson and Renaud 1989). Paired-samples t­ striatus was performed but is not included here tests were used to analyse the percentage because it did not illuminate any useful trends for consumption and behavioural choice data from this study. This may be a result of the relatively low each two-way choice triaL Where the data did not abundance recorded during most sampling trips. meet the assumptions of hetero,geneity and normality a non-parametric Wilcoxon-signed ranks test was used to assess significance. Pulse settlement VS. continuous recruitment Evidence against apulse settlement event is the mean size of each sampling trip. The sizes of B. RESULTS leachii found during each of the four sampling trips were significantly different (One-way ANOVA: df Aggregation characteristics "" 3, F = 35.588, P < 0.001). TIlere was a decrease in Sea hares were found in high abundance at SheUy mean length of B. leachii between 25 June and 7 Beach from 25 June to 27 July 2002. Two species of July, a small increase during 21 July and then sea hare were present B. leachii and S. striatus. another small decline during 27 July (Figure 2). The Individuals of the two species were observed in the animals found on 25 June and 21 July were same areas, often in close association. A significantly similar in size and those in 7 July and cyanobacterial bloom was also present at Shelly 27 July were significantly similar in size (Tukey's Beach during the sea hare population irruption. The cyanobacterium was identified as C. crustacea (B. Bendell, Jeu, pers. comm.) and was present both in A 8 A B large tufts on the sediment and growing epiphytically on the seagrass leaves. During the initial sampling trip, 25 th June 2002, only B. leachii was recorded in the transect area. The mean density of B. leachii at this time was 5.12 individuals per m' (Figure 1). The highest recorded density of B. leachii was observed on 7 July when it increased to 7.44 individuals per m 2 and then o decreased in the subsequent sampling trips. No 25 Jun 17 Ju! 21 J.l1 27 ju! animals were recorded on the final sampling trip, 3 Sampling date September 2002. Throughout the entire month of sampling, the predominant species remained B. Figure 2 The mean length (mm +/- standard error) of BursC/tella leachii individuals dming the four leachii (Figure 1). 111e density of S. striatus was much sampling trips to Shelly Beach in 2002. The lower than that recorded for 8. leachii for the letters denote significantly different sub­ duration of the population irruption. groups as determined by Tukey's HSD post­ Stylocheilus striatus was first recorded within the hoc tests. Population dynamics and feeding preferences of Bursatella leachii 15

60

50 A

40

30

20

10

60

50

40

30

20

10

60 (j) 50 C -:::l 0 :::::.. 40 >. l) c 30 Q) ::J 0- 20 Q} u...~ 10

60

50 D

40

30

20

10

20 25 3~. 35 40 45 50 55 60 65 70 75 Length (mm)

Figure 3 Length-frequency histograms of Blirsatella leachii present at Shelly Beach in 2002 during each of the four sampling trips where a) 25 June, b) 7 July, c) 21 July, and d) 27 July.

HSD post-hoc tests: homogeneous subsets A and B, size classes during the course of sampling in Figure Figure 2). This indicates that there was more than a 3. There was also a gradual decline in the single settlement event occurring at Shelly Beach. abundance of larger size classes and the three There was continual recruitment of smaller largest classes disappeared completely in the last individuals into the population at Shelly Beach sampling trip of. In tate JUIce 2002, the size classes during the sampling period (Figure 3). TIle small were normally distributed with a peak at the SOmm sea hares grew larger and moved into the larger length class although there were slightly more 16 c.L. Clarke

individuals in the larger size classes (Figure 3a). In early July, there was a peak of 40mm animals and £ 2.51

two smaller peaks at 30 and SOmm. There were a g 2 ~ high proportion of smaller individuals at this hme ~w i (Figure 3b). There was an increase in the number of rn if) 1.5 1 larger individuals sampled during the 21 July E i: ] If; ~ sampling trip (Figure 3c). At the end of July no one (J] E size class dominated the distribution and again E~ O'l 0.5 0> there was an increase in the number of small w individuals (Figure 3d). 25 Jun 17 Jui 21 Jul 27 -Jui 3 Sep

Egg mass variation Sampling date The pre:oence of charaderistic stl'in.g-like Sea hare Figure 4 The mean density of egg masses (per m 2, +/­ egb IT.. 2lsses l,'vithi~. the sc"!np!ing are.R indj(:atpd standard error) found dunng each sampiing spawning activity and the presence of trip within the scagras:,. bl.'ds of Shelly Beach reproductively mature sea hares. Hoy/ever, 'Nhich in 2002. sea hare species had laid each mass could not be identified from the egg masses themselves in a field setting. The egg mass density during the 2002 sea including tBe cyanobacteria, C crustacea. The mean hare population irruption peaked on July 7th (Figure percentage consumption, corrected for autogenic 4), which corresponds with the timing of the peak change, of each pair-wise trial is shown in Figure 5. in sea hare density (see Figure 1). The density of Only Enteromorpha and C crustacea were consumed egg masses at this time was 2.28 egg masses per m:? in amounts greater than 10%. The feeding preference hierarchy of B. leachii was Enteromorpha Palatability sp. > C crustacea> Sargassum sp. and p, pinnata. Bmsatelia leachii significantly preferred the green Enteromorpha was consumed significantly more than alga Ente1'On1orpha over all other species tested, C. crustacea (Wilcoxon Signed ranks test: Z = -2.240,

w (f) i: 60 * * * * * ---c: 50 o '"5. 40 § 30 (j) § 20 o :!2.o 1 0 C {lj o OJ ::2

A B c D E F Algal Pairvvise Trial

Figure 5 The mean percentage consumption (+/- standard error) of each pairvvise choice trial by Bursatella leachii. Asterisks (*) indicate significant choices as evidenced by the respective statistical test. Each colour code represents a single two-way choice experiment denoted by letters A-F where A :;;; Enteromorpha sp. vs. Calothrix cnlstacea, B =: Eilteromorpha sp. vs. Pteroc1adia pinnata, C :;;; Ertferomorpha sp. vs. Sargassum sp., D =: CalOlhrix crustacea vs. SargasSll1rt sp., E '" Calothrix crustacea vs. Pterocladia pinnata and F '" Sargasswn sp. vs. PterocIadia pinl1ata. population dynamics and feeding preferences of Bursatella leachii 17

:r '" 0.025), Sn1"gas5um (Wilcoxon signed ranks test: Z algcte presented to 8. leadeii had at least one "" -2.100, P '" 0.036) and P. pimwta (Wilcoxon signed individual found upon it at least once during the ranks test: Z == -2.521, P == 0.012). Calothrix cntstacea trials. was consumed significantly more than Sargassurn sp (Wilcoxon signed ranks test: Z '" -2.521, P = 0.012) and F pinnata (WilcoX0n signed ranks: -2.521, DISCUSSION p "" 0.012). There was no significant difference between the consumption of Sargassum sp. and P. Sea hare density pinnata (Wilcoxon signed ranks test: Z "" -1.400, The sea hare population at Shelly Beach reached p>O.05). extremely high densities during the sampling period compared to those previously reported A ttra cti yen ess (Carefoot 1987). In June 2003, another population TI1€ presence of L.'1dividual sea hares on each alga irruption event was documented at the same yielded almost the same feeding preference location (Clarke 2004) where 5. striatus reached high hierarchy as that obtained by the direct measure of denslty (3.84 per m:) while B. leacilii '''-'as ioW"\d at percentage consumption: Enteromol'pha sp. > C. lower density (0.24 per mn. Stylochcihis siriatu5 has C1'll:::t~lcea > P. pill nata and S!lrglls5um sp. The place of been reported in large numbers in association with P. pinnata and Sargasswil sp .•vere reversed in this blooms of Lyngbya r!1njl!scula. in Ha\-\'aii (Switzer­ data set but there was no significant difference Dunlap and Hadfield 1979) and Guam (Paul and between the two. The mean number of times a sea Pennings 1991; Nagle et al. 1998) but the densities hare was present on each alga during the pair-wise were never quantified. Lowe and T umer (1976) trial is shown in Figure 6. Bursatella leachii vvas reported substantially higher densities of B. leachii recorded significantly more on Enteromorpha than in Florida. However, the denSity reported for C. crustacea (Wilcoxon signed ranks test: Z = -2.058, subtida.l populations was based on the number of P '" 0.040) or P. pinnata (Wi1coxon signed ranks: Z "" animals within a single aggregation. Other sea hare -2.251, P '" 0.024). Bursatella ieachii significantly species have much lower densities than those preferred C. crustacea over P. pinnaJa (Wilcoxon recorded here. Carefoot (1987) revie;ved the signed ranks: Z "" -2.041, P '" 0.041). Each of the four reported densities of several species of Aplysia and

3 * * *

W (/) 2 i Q) () c 1 Q) V) ill..... a... 0 (\J (\J OJ .m ctl E \il E OJ .£l ..c:: Q) ...c:: .r::: OJ I:\l G \il :.J 0 :::J 8 e- e-- c:: e. U) Cfl c: a .£l 0 c:: a U) .S9 Cfl .£l c: Cfl U) V) \il r./) Q E '0. E OJ £: 2 p 2 [?l 2 2 0 2 n.: ~ C) (iJ 0 cl .i!:l .i!:l .§ (\J (f) <.) c:: U)

Figure 6 The mean number of times Bursatella leachii was recorded present on each alga within in each pairwise trial. Aster~sks (*) indicate statistically significant differences. Each colour code represents a single two-way choice expenment ?enoted by letters A-f where A = Enleromorpha sp. vs. Calothrix cnlstacea, B '" Enteromorpha sp. VS. Pterocladw pll1nata, C '" Enteromorpiw sp. vs. Sargass!£n1 sp., D", Calothrix crustacea vs. Sargassum sp., E '" Calothrix crustacea vs. Plerocladia pinnata and F '" Sargassum sp. vs. Pterocladia pinnata. 18 c.L. Clarke

the average densities never exceeded 5.0 per m" the one that is the most consumed. Stylocheihls with most reports less than 1 per m 2 (Willan 1979; striatus has only been reported to feed upon L Carefoot 1987). The low densities of Aplysla species majuscIIla overseas (Switzer-Dunlap and HadfieJd have often been used as evidence for the limited 1977; Paul and Pennings 1991; Nagle ei Id. 1998). role of all sea hare species in the marine community nlOugh S. striatus and B. leachii seem to be common (Carefoot 1987; Rogers et aL 2003). However, the together, B. lcachii is reported to feed on a number present study indicates much higher densities for B. of red and green algae. The reason for the greater leachii than the members of the more commonly dietary range of B. leachii has not been previously studied genus ApiysiCi. discussed yet recent work on the way these two sea hares digest chemicaily rich foods has been Aggregation of sea hares on cyanobacterial conducted. Capper (2003) showed that there are blooms differences in the way B. leachii and 5. striatus deal The co-occurrence of B. ieachii and S. siriatus anu 'Nith the secondary metabolites present in L. the bloorH of the cyarloba.ctEriuTI"I.,- C. crHSr2CCc. in majuscula. B. ieachii excretes secondary metabolites 2002 has not been documented previous1y. At the in the ink and faecal matter while S. striatus stores same location in a subsequent year these sea hare these metabolites in the digestive gland. The species were associated with a different strategies of these two co-occurring species thus unidentified cyanobacteria species (Clarke 2004} differ; 5. striatus may have specialized to deal with Bursatella leachii is believed to have a more catholic the speciEs metabolites of L majuscula while B. diet than S. striatus (Wu 1980; Paige 1988). In Hong leachii excretes the chemicals and thus is more likely Kong, 8. leachii consumes a range ot macroalgae, to be able to deal with a larger range of metabolites preferring Enteromorpha prolifera, although than S. striatus (Capper 2003). cyanobacteria species were not tested (Wu 1980). In Preferred choices may not offer the best long-term Florida, only species of cyanobacteria (including L. growth and fitness. consumes majuscu[a) were preferred in food choice tests and Enteromorpha sp. preferentially III the laboratory but as meta.Tllorphic substrates (paige 1988). In contrast, animals on a diet of the red alga, Plocamium sp., an previous reports have linked S. striatus with L. have lowered faecal production and better growth rnajuscula, its preferred food according to a number (Carefoot 1967). Monospecific diets of preferred of authors (Paul and Pennings 1991; Pennings et al. chemically rich species have shown increased 1996; Nagle et al. 1998). mortality for B. leachii and S. striatus (A. Capper, D1.1ring the present study, R leacMi preferred the pas. camm.) and for A. parvula (Rogers et aL 1995). green alga, Enteromorpha sp., to the cyanobacterium The next step in investigating the feeding C. crustacea, tile brown macroalga Sargassum sp., preferences of B. leachii would be to compare the and the red macroalga P. pinnata. Enteromorpha sp. long term growth and reproduction of animals fed was preferred even to the cyanobacterial species (c. Enteromorpha sp. with those fed cyanobacteria. crustacea) that the animals were seen to be feeding upon in the field. The occurrence of Enteromorpha Population irruption dynamics sp. was extremely rare at the sampling site. Several The length-frequency distribu tion of this sea hare species exhibit greater generalization of popUlation eill, be used to investigate hypotheses feeding preferences in the laboratory than are about its dynamics (see Gev et al. 1984). It is exhibited in the field (Saito and Nakamura 1961; important to note that length-frequency analysis is Winkler and Dawson 1963; Carefoot 1987; Rogers et based on the assumption that length is a reasonable al. 2003). For example, some members of the genus proxy for age. A single study found that length and Aplysiaare often considered to specialize on red age are directly related for B. leachii although only a algae in the field (Carefoot 1987) and yet laboratory small number of animals were used in experiments experiments have repeatedly shown their (Willan 1979). Other molluscs shrink under preference for green algae (Winkler and Dawson conditions of stress or low food availability 1963; Rogers et al. 2003). Winkler and Dawson (Russell-Hunter et aL 1984; Russell-Hunter 1985). (1963) found that A. cal(fomica were only inhabiting This is unlikely in the present study because the red algae in the field, but preferred Enteromorpha in concurrent cyanobacterial bloom provided an laboratory testing. Additionally, Saito and abundance of food, at least for the majority of the Nakamura (1961) found that A. juliana and A. sea hare population irruption event. The hypothesis kurodai consumed species of brown and red algae being tested was that sea hare population irruptions respectively in the field but both preferred species are the result of settlement of a single cohort on the of El1teromorpha and Ulva in the laboratory. cyanobacterial bloom. Similar studies with co­ The feeding preference hierarchy of B. leachii was occurring and in identical whether obtained by testL.'1g palatability or Israel showed a single recruitment pulse (Achituv attractiveness. Therefore the alga that individual sea and Susswein 1985). If a single pulse occurred we hares chose to inhabit (the most attractive) was also would expect that the length-frequency analysis Population dynamics and feeding preferences of BUl"sCl-tella leachii 19

would have detected a large cohort of animals of population dynamics of B. leachi; and S. striatu:: in similar size moving through the size classes. This Shelly Beach. did not seem to be the case here, because for B. The other possibility for the observed depletion of leacizil, there was a continual addition of smaii large sea hares is migration. If the animals were recruits. 1r. conjunction.' there were also a small migrating to areas unavailable to observation then number of individuals in the large size classes. we would expect to see 2L'l abrupt appeuance or Thus, the size distribution of this population disappearance of mature animals. The continual indicates that recruitment of individuals into the recruitment of small B. lmchii into the population in population is occurring throughout the duration of 2002 rules out the possibility that this high-density the cyaxlObacterial bloom. occurrence was merely a roving population of Additionally, if the sea hare population irruption adults. It would be expected that migration would was a: pulse-recruitment event we would expect to cause an increase/decrease in abundance of all size see a graou

habitats. Analysis of the population dynamics Cardoot, T. H. (1967). Growth and nutrition of Aplvsia revealed that the population irruption observed punctata feeding on a variety of marine algae. JOIlTiial was not the result of a pulse recruitment event, but of the Marine Biological Association of the United Kingdom rather a continuous supply of recruits, in contrast to 47: 565-589. previous studies. Techniques such as population Carefoot, T. H. (1970). A companson of absorption and utilization of food energy in two species of tropical genetics and larval tows are needed to shed further Aplysia. J01l1"l1al of Experimenial .Marine Biology and light on recruitment patterns in this ecosystem. Ecology 5: 47-62. The difference between the preferred alga in the Carefoot, T. H. (1987). Aplysia: its biology and ecology. laboratory and the prey species that B. leach!i was Oceanography and Marine Biology Ann1tal Review 25'. feeding upon in the field reflects the complications 167-284. associated wilh attempting to separate the various Clarke, c.L. 2004. The ecological role of sea hares factors influencing feeding preferences in sea hares. (Opisthobranchia: Ano.spidea) within tropical intertidal It seems that the most important raciors dicLatir'g haviials. iv1Sc Thesis. James Cook tJniver.~ltYJ feedilig preieTences iTt B. lcachii arc the palatability, Townsville, Australia. 141 pp.

abundance it, the natural habitatr and though not Coleman, N. (2001). 1001 Nudibranchs: CatalogHe of Indo~ directly tested here, likely the secondary chemistry Pacific Sea Sl!lgs. Neville Coleman's Underwater of the food item. The demonstrated ability of B. GeographicPty Ltd, Springwood, Qld. leachii to reach large population sizes while feeding Colgan, M. W. (1987). Coral reef recovery on Guam, on a sub-optimal food source highlights the Micronesia after catastrophic predation by Acanthaster flexibility of feeding in these sea hares. The planci. Eiology 68(6): 1592-1605. plasticity in feeding is likely to enhance the ability Cote,!. and Reynolds, J. D. (1998). Tropical fish: of sea hares to successfully exploit an abundant, but ExplOSions and extinctions. Trends in Ecology and Evoltltion 13(12): 475-476. often, unpredictable resource. The aggregations of Craw lev, M. (1983). Herbivory: The Dynamics of Animal­ B. leachii observed at Shelly Beach, Townsville are J. lvfa~rophyte Interactions. Blackwell Scientific not the result of single pulse recruitment. The Publications, Oxford. evidence presented here suggests that the Fisher, R., Bellwood. D. R. and Job, S. D. (2000). aggregations were supported by continual Development of swimming abilities in reef fish settlement in response to what may be a less than larvae. Marine Ecology Progress Series 202: 163-173. ideal food source. Gev, S., Achituv, Y. and Susswein, A. J. (1984). Seasonal determinants of the life cycle in two species of Aplysill ACKNOWLEDGEMENTS found in shallow waters along the Mediterranean This work was completed as part of the M.Sc. coast of Israel. Journal of Experimental Marine Biology and Ecology 74: 67-83. degree program at James Cook University. The Ginsburg, D. W. and Paul, V. J. (2001). Chemical defenses author would like to thank Giliarme Brodie, Rocky in the sea hare Aplysia parvula: importance of diet and de and Annette for their Nys Klussmann-Kolb sequestration of algal secondary metabolites. Ivlarine supervision. Financial support was provided by Ecology Progress Series 215: 261-274. eRC Reef, the Malacological Society of Australasia Hay, M. E., Pawlik, J. R, Duffy, J. E. and Fcnical, W. and James Cook University. Laboratory work was (1989). Seaweed-herbi vore-preda tor interactions: host~ performed at the MARFU aquarium facilities at macrophyte specialization reduces predation on small James Cook University. Two anonymous reviewers herbivores. Oecolagia 81: 418-427. provided valuable advice on an earlier version of Hayden, H.5., Blomster, L Maggs, C. A. Silva, M.J. this manuscript. Stanhope, and Waaland, J. R. (2003). Linnaeus was right all along: Ulva and Enteromorpha are not distinct genera. European Joumal of Phycology 38: 277-294. REFERENCES Johnson, P. M. and Willows, A. O. D. (1999). Defense in Achituv, Y. and Sus5wein, A. J. (1985). Habitat selection sea hares (Gastropodu, Opisthobranchia, Anaspidea); by two Mediterranean species of Aplysia~ A. fasciata Multiple layers of protection from egg to adult. Poiret and A depilans Gmelin (: Marine and Freshwater Behavioural Physiology 32: 147- Opisthobranchia). Journal of Experimental Marine 180. Biology and Ecology 85: 113-122. Lanyon, J. M. and Marsh, H. (1995). Temporal changes Audeskirk, T. E. (1979). A field study of growth and in the abundance of some tropical intertidal reproduction in Aplysia c(lii/arnica. Biological Bulletin seagrasses in North Queensland. Aquatic Botany 49: 157: 407-421. 217-237. Burla, H. and Ribi, G. (1998). Density variation of the Lowe, E. F. and Turner, R. L (1976). Aggregation and trail­ zebra mussel Dreissena polymorpha in Lake Zurich, foHawing in juvenile Bursatella leachii plei (: from 1976 to 1988. Aquatic Sciences 60(2): 145-156. Opisthabranchia). The 19(2): 153--155. Capper, A. (2003). PhD Thesis. Ecotoxicology of the Lubchenco, ]. and Gaines, S. D. (1981). A unified cyanobacterium, Lyngbya majllsclIla and five approach to marine macrophyte~herbivore herbivores in Moreton Bay, Southeast Queensland, interactions. 1. Populations and communities. Annual Australia. University of Queensland, Brisbane. Review of EcologtJ and SystematiCS 12: 405-437. Population dynamics and feeding preferences of Bursatella leachli 21

Nagle, D G., Camacho, F. T. and Paul, P J (1998). Rogers, C N, Steinberg. P. D. and De N~'s, R (1995). DHctary preferences elf the opisthobrand, mollusc Factors associated with ohgophagy in two species of Stylocl1eilus longical£da for secondary metabolites sea hare~ (M 011 usca: Anasp id ea). Journal of produced by the trcric,.11 cy2..:-:.obact~:-i~ln Lyngbya Experimen/[)/ l'i1:l'i?!~ B;~I!ogy and EU1ic1gy 192: 47-73. majuscula. MarinI? Biohgy 132: 267-273. (1984). ?"icotn, M E. (1980). Factors invo! vcd in l1erbi\"ore food Russell-Hunter, W.D. (1935). Physiological, ecological. preference. JOllrna! of Expenmental Marine Biology and and evolutionary aspects of molluscan tissue EcolDgy 42: 13-26. degrowth. American !l1alacologlcal BlI/lrfm 3(2): 213- Pa.ige, J. A. (1988). Biology, metamorphosis and 222. postlarval development of Bursatella leachii Rang Russell-Hunter, W.D., Bro,vne, R. A. and Aldridge, D. (Gastropoda: Opisthobranchia). Bu!letin of Marille W. (1984) Overwinter tissue degrowth in natural Science 42(1): 65-75. populations of freshwater pulmonate snails (Helisoma Paut V. J. and Pennings, S. C. (1991). Diet-derived trivobis and Lymnaea palustris). Ec%31J 65(1): 223-229. chemical dcfc~s::-s i:;-. t!--~c sc~ hztrc Sty!cchci!us 5.Ji~o, ~'. and ~" ..l3.k:l:T:.".J=2.i ~'J. (1961). B!o]cgy ':)f the se:!. /ollgicouda (Quo\' ct Gaimard 1824). /olnnal of hare, Aplysia itl[iana, as a predator of the brown Experimental Manni' BIOlogy and EcoioS!1 151: 227-243. seaweed, Undaria plntliltifida - 1. The feeding habit. Pawlik, J. R. (1989). Larvae of the sea hare Aplysia Blllletzn of the Japanese Society of Scientijic Fisheries cillifornica settle and metamorphose on an assortment 27(5): 395. of macroalgal :;pecie:;;. Marine Eeulogy Prl)xr~ss Series Smith, R. L. (1996). Ecology and Fidd Biology. l-iarpel 51: 195-199. Collins College Publishers, New York. Pennings, S. C. {l990a). Predator-prey interactions in Switzer-Duii!ap, M. and Hadfield, M. G. (1977). opisthobranch gastropods: effects of prey body size Observations of development, la rval growth and and habitat complexity. lvlarine Ecology Progress Series metamorphosis of four species of 62: 95-1Ol. (Gastropoda: Opisthobranchia) in laboratory culture. Pennings, S. C. (l990b). JI.·1ultiple factors promoting Journal of Experimwtal Mm·ine Biology a/hi Ecology 29: narrow host range in the sea hare, Aplysia caiifomica. 245-261. Oc,olagia 82: 192-·200. Switzer-Dunlap,!'vI and Hadfield, 1\-1 G. (1979) Pennings, S. C. (1991). Spatial and temporal variation in Reproductive patterns of Hawaiian aplysiid recruitment of Aplysia callfarnica Cooper: patterns, gastropods. In S. E- Stancyk (ed.), Reproductive Ecology mechanisms, and consequences Journal of of Marine Invertebrates 199-210. University of South Experimental Maritle Biolog1j and Ecology 146: 253-274. Carolina Press, Columbia, South Carolina. Pennings, S. c., Nadeau, M. T. and Paul, V. J. (1993). Walker, T. A. and O'Donnell, G. (1981). Observations on Selectivity and growth of the generalist herbivore nitrate, phosphate, and silicate in Cle:veland Bay. Do/abel/a auricularia feeding upon complementary Northern Queensland. Allslralian JOllrnal of Marine and resources. Ecology 74(3): 879-890. Freshwater Research 32: 877-887. Pennings, S. C. and Paul, V. J. (1992). Effect of INillan, R C. (1979). The Ecology of Two New Zealand macrophyte toughness, calcification, and chemistry Opisthobranch lvIolit/$c$. Unpublished PhD Thesis. on herbivory by Dolabeila al.lt'1cuiaTln. Ecology 73(5): University uf Auckland, Auckland, New Zeala.nd. pp. 1606-1619, 278 pp. Pennings, S. c., Weiss, A. M. am] PauL V. J. (1996). Williams, E. H, Jr., Bunckley-Williams, L., Lilyestrom, C. Secondary metabolites of the cyanobacterium G., Larson, R. L Engstrom, N. A, Ortiz-Corps, E. A. Microcoleus lyngbyacells a.nd the sea ·hare Stylachellus R, and Timber, J. H. (2001). A population explOSion !ongicauda: palatability and toxicity. Marine Biology of the rare tropical/ subtropical purple sea mane, 126: 735-743. DnJmonema dalmatil1um, around Puerto Rico in the Peterson, C. H. and Renaud, P. E. (1989). Analysis of summer and fall of 1999. Caribbean Journal of Science feeding preferencE experiments. Oecologia 80; 82-86. 37(1-2): 12.7-130. Plaut, 1., Sorut, A. and Spira, M. E. (1998). Seasonal cycle Winkler, L R. and Dawson~ E. Y (1963). Observations and population dynamics of the sea hare Aplysia and experiments on the food habits of Californian sea r]clIilfera in the northern Gulf of Eila t (Aqaba), Red hares of the genus Ap!ysia. Piic0~c Science 17: 102-105. Sea. Joun/al of MOl!lISCt1t1 Siudies 64: 239-247. Winkler, L. R. and Tilton, B. E. (1962). Predation on the Rogers, C. N., De Nys, R. and Steinberg, P. D. (2000). , Aplysia californlca Cooper, by the Preda tion on juvenile Aplysia parvllill and other small solitary great green sea anemone, Anlhopleura anaspidean, ascoglossan and gastropods xanthogrammica (Brandt) and the effect of sea hare by pycnogonids. The Vehger 43(4): 330-337. toxin and acetylcholine on anemone muscle. Pacific I{ogers, C. N., De Nys, R. and Steinberg, P. D. (2002). Science 16: 286-290. Effects of algal diet on the performance and Wu, R. S. (1980). Foraging strategy in the sea hare susceptibility to pn~da tion of the sea hare Aplysia B!lrsiliella leachii (Mollusca: Opisthobranchia. parul/la. Marine Ecology Progress Series 236: 2341-254. Proceedings of the First 111tematiorwl Marine Biological Rogers, C. N., De Nys, R. and Steinberg, P. D, (2003). Workshop: The marine flora and farma of Hong Kong and southern China, Hong Kong. Hong Kong University Ecology 0 f the sea hare Aplysia parvula (Opisthobranchia) in New South Wales, Australia. Press. Moill/scan Research 23(3): 185-198.