Predation on juvenile Aplysia parvula and other small anaspidean, ascoglossan, and nudibranch gastropods by pycnogonids

8 Pages·2000·4.6 MB·English

by  C N Rogers

THE VELIGER
© CMS, Inc., 2000
The Veliger43(4):330-337 (October2. 2000)
Predation on Juvenile Aplysia parvula and Other Small Anaspidean,
Ascoglossan, and Nudibranch Gastropods by Pycnogonids
C. N. ROGERS,* R. DE NYS and P. D. STEINBERG
School of Biological Science, University of New South Wales, Sydney 2052, NSW, Australia
Abstract. This study investigates the predation on opisthobranchs by the pycnogonid Anoplodactylus evansi. Ano-
plodactyliis evansi attacked and fed on small individuals of 13 different species ofanaspidean, ascoglossan (sacoglossan),
and nudibranch gastropods in this study. This predator was used to investigate the feeding deterrent effects of diet-
derived compounds in the sea hareAplysiaparx'ula. Anoplodactylus evansi was not deterred from feeding onA. parvula
containing diet-derived secondary metabolites from the red seaweeds Laurencia obtusa or Delisea pulchra. However,
increasing the quantity of metabolites present in A. parvula by treatment with adult extracts did deter A. evansi from
feeding, suggesting that high levels of diet-derived materials in sea hares may have deterrent effects against a predator.
A. evansi does not appear to sequester algal secondary metabolites from its prey. The abundance of A. evansi and A.
pan'ula on the alga L. obtusa was measured to determine if this pycnogonid influences the population of sea hares in
the field. A. parvula occurred in significantly lower abundance onL. obtusa thalli inhabited byA. evansi overconsecutive
years, a pattern consistent with population-level effects by this predator.
INTRODUCTION etary compounds as defenses have mainly investigated
palatability to "potential" predators. Few studies have
Opisthobranch mollusks feed on a wide range of both investigated the effects on predators ofdifferent types or
macrophytes and animals, although most species have di- quantities of compounds stored by opisthobranchs. One
ePtasulli&mitPeednntiongas,fe1w99p1r;eyTrtoywpbersid(gTeh,om19p9s1o;nReotgealr.s, e1t98a2l.;, ewxhcoeptfioounndistthahte cshteudmyicoanllsyeadehpaaruepserbaytePeAnpnliynsgisa (c1al9i9f0obr)-
1995). Opisthobranchs may deter potential predators by nica Cooper, 1863, was rejected less frequently by fishes
storing noxious materials in their tissues. Such materials than chemically rich conspecifics. The types and levels
include chemical compounds or physical structures, of defensive compounds stored by opisthobranchs may
which are obtained from dietary sources or produced by
the animals de novo (Faulkner & Ghiselin, 1983; Karuso, be important in determining a predator's response follow-
1987; Avila, 1995). A predator attempting to consume ing contact, and so influence the outcome of an attack.
Population-level effects ofpredation on opisthobranchs
different species of opisthobranchs is likely to encounter have rarely been addressed. Sarver (1979) found high
a variety of defensive materials which it must overcome rates of mortality for juvenile Aplysia Juliana Quoy &
in order to successfully feed. The trophic flexibility re-
Gaimard, 1832, and attributed this to both abiotic and
quired to feed on different opisthobranchs is thus proba-
biotic factors including predators. Trowbridge (1994)
bly high, and there are few reported generalist predators
used cages to exclude predators from mat-forming algae
of these mollusks. Exceptions are some aglajid cephal-
inhabited by ascoglossans in the field and found a signif-
aspideans which prey on other opisthobranchs (Kandel,
icant difference in the abundance of the ascoglossan Al-
1979; Pennings, 1990a; Faulkner, 1992). Other predators
deria modesta Loven, 1844, between caged and uncaged
ofopisthobranchs include anemones (Nolen et al., 1995),
areas after 12 days. These studies suggest that rates of
crabs (Carefoot, 1987; Trowbridge, 1994), fishes (Pen-
predation on some opisthobranchs may be high in the
nings, 1990a; Trowbridge, 1994), seastars (Carefoot,
1987), a pycnogonid (Piel, 1991), andbirds (Todd, 1981). field, especially for small species orjuveniles. Incontrast,
Ecological studies ofpredation on opisthobranchs have many adult opisthobranchs are generally regarded as free
found that predator success varies, depending on opistho- from predation with few reported natural enemies (Care-
branch body size (Pennings, 1990b), habitat complexity foot, 1987; Faulkner, 1992).
(Pennings, 1990b), and the presence ofchemical defenses This study investigates predation on opisthobranchs by
(Hay et al., 1989). Studies on the effects of acquired di- the pycnogonid Anoplodactylus evansi Clark, 1963. Is-
sues addressed are:
* Corresponding author: Phone: [phone], Fax: 61 02 (1) Where does A. evansi occur relative to opistho-
9385 1558, e-mail: [email protected] branchs on a local scale?
C. N. Rogers et al., 2000 Page 331
mm
(2) What is the diet breadth ofA. evansi across different other invertebrates tested were 5-15 in length. Test
sizes and species of opisthobranchs? animals were observed to determine ifthey were attacked
(3) How does the type and quantity of algal secondary by pycnogonids, and if feeding commenced. Aquaria
metabolites stored by Aplysia parvula Guilding in were checked on subsequent days to determine if prey
Morch, 1863, affect predation by this pycnogonid items were consumed. The method of attack and con-
species? sumption byA. evansi was observed in aquaria, and using
(4) What is the abundance of A. evansi relative to the a dissecting microscope for prey items isolated in a glass
abundance ofthe sea hare A. parvula on the red sea- dish. When sea hares were attacked, it was noted if ink
weed Laurencia obtusa Lamouroux, 1813? or opaline secretion was released, and the effect this se-
cretion had on pycnogonid behavior. Once pycnogonids
MATERIALS and METHODS had ceased feeding on sea hares, any remaining body
parts were identified.
Study Sites and Collection of Study Organisms To investigate more rigorously how the capture success
Collection and observation of the pycnogonidAnoplo- of pycnogonids related to prey size, individual pycnogo-
dactylus evansi, the sea hareAplysiaparvula, and the red nids and Aplysiapar\'ula were weighed (ww) and placed
alga Laurencia obtusa were done in Port Jackson, near in experimental compartments with apiece ofinsect mesh
Sydney, New South Wales (NSW), Australia, which is a (1 mm) as a substratum. Predation trials were done in 25
warm temperate region. Four sites where L. obtusa oc- L experimental aquaria which were partitioned into 20
curred were used in this study (Figure 1). All four sites separate compartments by plastic sheets. Seawater flowed
had sandstone reefs covered by mixed algal assemblages through the compartments via insect mesh-covered holes
dominated by brown macrophytes (described by Farrant before exiting into the main recirculating system. Sea
& King, 1982). Aplysia par\'ula was also collected from hares ofdifferent size classes (< 100, 100-200, 200-300,
the red alga Delisea pulchra Greville, 1830, at Bare Is- 300-400, > 400 mg ww, n = 35 sea hares) were placed
land. The other species of opisthobranch used in this into contact with pycnogonids at the start of each trial.
study (Table 1) were collected from the algal beds or Experimental compartments were checked over the fol-
sponge gardens on the reefs surrounding Bare Island (de- lowing 2 days, and the condition of each sea hare re-
scribed by Van der Velde & King, 1984), near the en- corded.
trance to Botany Bay,just south ofPortJackson. Subtidal
collection of pycnogonids and opisthobranchs for exper- Chemical Effects
iments was done by searching algae, sponges, or other
encrusting organisms in situ, removing the animals by To determine ifthe presence ofdifferent types ofalgal
hand, and placing them in a plastic container for trans- secondary metabolites (i.e., not from Laurencia obtusa)
portation to the laboratory. Animals were held in the lab- deterted Anoplodactylus evansi from feeding, pycnogo-
oratory for observation and experiments in 10 L plastic nids were offeredAplysiaparvula that had fed on the red
seaweed Deliseapulchra. D. pulchra contains halogenat-
aquaria supplied with flowing seawater, with pieces of
algae (20-40 g wet weight [ww]) provided as habitat. ed secondary metabolites, and A. parvula accumulates
Anoplodactylus evansi is reported to occur at several lo- high concentrations ofthese compounds in its tissues (de
calities in south easternAustralia (Clark, 1963). However, Nys et al., 1996), significantly more so than Laurencia
this pycnogonid broods its young and does not swim, and obtusa metabolites (Rogers et al., 2000). D. pulchra did
consequently its dispersal ability is limited to movement not occur at the site where A. evansi was found, nor in
between adjacent algae and perhaps rafting (e.g., on dis- that area of Port Jackson, so the secondary metabolites
lodged algae) to new habitats. All pycnogonids used in from this seaweed may be regarded as novel to the pyc-
experiments were non-egg-bearing adults (leg span 20- nogonids used. However, A. evansi would undoubtedly
30 mm). A voucher specimen ofthe pycnogonidA. evan- encounter D. pulchra in its wider geographic range, and
si was placed with the Australian Museum in Sydney, consequently, pycnogonids from the study area in Port
accession number AM-P49749, the specimen being an Jackson may possibly have an innate response to D. pul-
egg-bearing adult male. chra metabolites. Because we cannot rule out such a pos-
sibility, the term "novel" in reference to D. pulchra
Pycnogonid Feeding should be interpreted at the least as meaning it did not
co-occur with the pycnogonids tested. Five individual
To determine the diet breadth ofAnoplodactylus evan- pycnogonids were maintained in separate beakers and of-
si, 13 species ofanaspidean, ascoglossan, and nudibranch fered five replicatejuvenileA. parvula (100-200mg ww)
opisthobranchs, and othersmall invertebrates were placed collected from D. pulchra at Bare Island. Each sea hare
in contact with adult pycnogonids held in aquaria. Opis- was placed in contact with a pycnogonid, and the result
thobranchs in a range of sizes from 50 to 820 mg (ww) of the trial recorded after 2 days.
mm
were used (approximately 5-40 body length); the To determine if increased quantities of compounds in
Page 332 The Veliger, Vol. 43, No. 4
A
levels), or solvent-treated seahares as acontrol. further
group of solvent-treated sea hares was held separately in
aquaria cells without pycnogonids to control for "auto-
genic" changes in the weight of sea hares unrelated to
consumption by pycnogonids. Ten replicate animals were
given each treatment. The Aplysia parvula used in this
experiment werejuveniles collected fromDeliseapulchra
(mean 123 ± SE 15 mg ww, which contain on average
4.3 ± SE 0.8 mg of non-polar extract). The juvenile sea
hares were frozen, then thawed prior to use. The non-
polar extract used to treat sea hares was obtained from
adultA. parx'ula fromD. pulchra by sequential extraction
of freeze-dried animals, using 3 X 10 mL aliquots of
dichloromethane (AR). The resulting solutions werecom-
bined, syringe filtered (PTFE 0.5 ixm), then air dried.
Crude extracts from sea hares contained on average 42.2
± SE 4.5% total secondary metabolites, andjuvenile and
adult sea hares had similar levels of secondary metabo-
lites in their extracts. The relationship between crude ex-
tract levels (y in mg) and sea hare wet weight (x in g)
was determined for D. pulchra fed sea hares by linear
regression (y = 39.9 X +0.1, R^ = 0.86, p < 0.001, n =
Figure 1 42). To double the quantity of extract per sea hare in the
extract treatment group, the quantity ofcrude extract pre-
TPohretpJoasciktsioonn,oefassttuodfySsyidtneseyA,,NBe,wC,SoauntdhDWanleeasr,tAhuesternatliraa.nce to sent in each juvenile sea hare was calculated, then an
equivalent amount of adult extract was injected into the
body cavity ofthe sea hare. Ethanol was the solvent used
seahares deterred feeding byAnoplodacTyhisevansi, pyc- to apply the extract, and was also injectedinto the solvent
nogonids were held in individual cells in 25 L experi- control and autogenic control treatments at equivalent
mental aquaria and fed sea hares according to the follow- volumes to those used to administer the extract. The
ing experimental design. Pycnogonids were offered either weight of sea hares was measured at the start of the ex-
sea hares treated with adult extracts (i.e., to double the periment and after 2 days, and the proportional change in
level ofextract present), untreated sea hares (i.e., average weight calculated.
Table 1
Opisthobranch mollusks consumed by the pycnogonid Anoplodactylus evansi, grouped by order and family. Published
maximum lengths for each species are included (After Bum, 1989; Coleman, 1989; Wells & Bryce, 1993).
Maximum length
Order-Family Species (mm)
Anaspidea
Aplysiidae Aplysiaparvula Guilding in Morch, 1863 60
Aplysia dactylomela Rang, 1828 250
Aplysia sydneyensis Sowerby, 1869 150
AplysiaJuliana Quoy & Gaimard, 1825 200
Stylocheilus longicauda Quoy & Gaimard, 1824 40
Ascoglossa
Oxynoidae Oxynoe viridis Pease, 1861 10
Elysiidae Elysia australis Quoy & Gaimard, 1832 10
Nudibranchia
Polyceridae Plocamopherus imperialis Angas, 864 100
1
Chromodorididae Hypselodoris bennetti Angas, 1864 50
Glossodoris atromarginata Cuvier, 1804 50
Bornellidae Bornella stelliferAdams, 1848 40
Glaucidae Austraeolis ornata Angas, 1864 35
Aeolidiidae Spurilla australis Rudman, 1982 55
C. N. Rogers et al., 2000 Page 333
To examine whether algal secondary metabolites are tor ANOVA, and was not significantly different between
sequestered by pycnogonids after feeding, chemical anal- treatments (Fj^y = 2.14, P = 0.113).
yses were done on pycnogonids that had consumed a sea Field data for the size of Laurencia obtusa thalli, and
hare and compared to analyses of starved pycnogonids. abundance ofAplysiaparvula were analyzed using a sin-
The Anoplodactylus evansi used were maintained aerated gle factor ANOVA comparing the sites in Port Jackson,
in 2 L beakers with a piece of the red alga Laurencia followed by Tukey's HSD test. Data for abundance of
obtusa as habitat. Each treatment consisted of Aplysia Aplysia panada were Ln(l -I- x) transformed. Because
parvula or Aplysia dactylomela Rang, 1828, (both col- sites were not sampled for all dates, sampling dates were
lected from L. obtusa) or starved controls. Three repli- pooled for each site in comparisons. Data for abundance
cates were given each treatment. The pycnogonids were ofA. evansi were not homogeneous using Cochran's test
allowed to consume the sea hares over 2 days, then were because Anoplodactylus evansi only occurred at one site,
fcrruoszheendf,ortheexntrascetqiuoenn.tiTahlleypeyxctnroagcotendidussiwnegre3fXre5ezemdLrieadl,- csoomapanroen-apbaurnadmaentrciecoftesptyc(nMoagnonn-iWdhsibtentewye)enwassiteusAedvert-o
iquots of dichloromethane (AR). The resulting extract sus sites B and C.
was syringe filtered, then air dried and weighed. Samples
were prepared forgas chromatography mass spectrometry RESULTS
(GCMS) by dissolving the extract in analytical grade eth-
yl acetate (0.2 mg/mL) containing 10 |xg/mL naphthalene The pycnogonid Anoplodactylus evansi was found only
internal standard. The samples were then analyzed for at site A in SharkBay, Port Jackson, close to SharkIsland
Laurencia obtusa secondary metabolites using a Hewlett (Figure 1) where the type specimens were collected
PackardHP5980 series II gas chromatographandHP5972 (Clark 1963). This species was not observed at nearby
mass selective detector, following the procedures of de sites B, C, orD (Figure 1), or at otherlocalities in Botany
Nys et al. (1998). Bay and Port Hacking, south of Port Jackson. The pyc-
nogonids were found inhabiting several species of algae,
The Abundance ofAnoplodactylus evansi and
including the brown algae Sargassum linearifolium Turn-
Aplysia parvula on Laurencia obtusa
er, 1809, Sargassum vestitum R. Brown, 1811, Padina
The alga Laurencia obtusa was collected to measure crassa Yamada, 1931, Dictyopteris acrostichoides J.
the abundance ofAnoplodactylus evansi andAplysiapar- Agardh, 1882, as well as the red alga Laurencia obtusa
vula during consecutive years from three sites (A-C in and an arborescentbryozoan. Pycnogonids were observed
Figure 1) in Port Jackson, NSW. Laurencia obtusa was to be noctumally active, as werejuvenileAplysiaparvula
chosen to assess the abundance of both A. evansi and A. and Aplysia dactylomela. During the day, pycnogonids
parvula because both species were common on this alga. and sea hares were found sheltering in the basal sections
Algal thalli were removed from the substratum, carefully of algae.
placed into individual plastic bags with seawater, andthen Anoplodactylus evansi consumed 13 different species
taken to the laboratory for sorting. The number of pyc- of opisthobranch mollusk from eight families, encom-
nogonids and sea hares present on each thallus was re- passing three orders (Table 1). Ofthe species eaten,Aply-
corded, and the wet weight ofeach thallus was measured. sia parx'ula and the nudibranch Bornella stellifer Adams,
Sampling was commenced at site C at Parsley Bay when 1848, were collected most often at site A whereA. evansi
no L. obtusa were found at site B during 1996. occurred. Anoplodactylus evansi was also observed to
consume an unidentified errant polychaete worm. Pyc-
Statistical Analyses nogonids did not consume small (5-15 mm in length)
Results ofpredation trials on different sizes ofAplysia amphipods, prosobranch mollusks, or echinoids that were
parvula by Anoplodactylus evansi were compared in a offered to them. A. evansi consumed only small individ-
contingency table for sea hares larger and smaller than uals of each opisthobranch species. Those eaten were
200 mg WW. A G-test was used to test the hypothesis that generally less than 200 mg ww (< 10 mm in length),
consumption ofsea hares byA. evansi was equivalent for which for most species was much smallerthan adult spec-
both size groups. imens observed in the field or published maximum sizes
Pycnogonid consumption ofextract-treated versus oth- (Table 1) (Bum, 1989; Coleman, 1989; Wells & Bryce,
erAplysiaparx'ula was done using a single factoranalysis 1993). When A. evansi was offered A. parvula, ranging
of variance (ANOVA) followed by Tukey's HSD test at in size from 20 to 600 mg ww, the pycnogonids could
the a = 0.05 significance level. Data for the proportional subdue and consume only animals smaller than 300 mg
change in weight of sea hares were transformed using WW (Figure 2). The effect of prey size (< 200 versus >
Ln(l -I- x) before analysis and checked for homogeneity 200 mg ww) on consumptionbyA. evansi was significant
of variance using Cochran's test. The initial size of sea (G = 8.53, P < 0.005). At adult size this pycnogonid
mm
hares in each treatment was compared using a single fac- species has a leg span ofup to 30 across and weighs
Page 334 The Veliger, Vol. 43, No. 4
uu- u.o
b
b
5- T
80- (4) T
*«
2 0.4- a
60- (12) a. a
T T
1 o 0.3- I
(7)
40- ,
t
0.2-
*
20-
0.1-
1
0- ! r "* (4) (8)
1 1 1 1 1
<100 Auto-control 2xextract Solvent Untreated
Seahareliveweight(mg) Treatment
Figure 2 Figure 3
Consumption of different sizes of the sea hare Aplysia parvula Mean weight loss (proportion) fromAplysiaparvula + ISE, for
by Anoplodactylus evansi. Bars show percentage of individuals treatments offered to Anoplodactylus evansi (except autogenic
eaten; the number of sea hares per sample is shown in brackets controls). The result ofa one way ANOVA on Ln(l -I- x) trans-
above the bar, (n). formed data was F339 = 7.94, P < 0.01. Treatments which are
not significantly different in weight loss at a = 0.05 using Tu-
key's test share the same letters on the graph, n = 10 per treat-
ment.
50-80 mg WW. A. evansi feeding on 300 mg sea hares is
attacking prey five to six times its own body weight.
Anoplodactylus evansi subdues prey using the front dactylomela containing Laurencia obtusa secondary me-
four legs to grasp and hold, while the back four legs tabolites contained trace amounts ofthe terpene palisadin
remain secured to the substratum. Each leg ends with a A at levels below the quantitative limit of the GCMS.
long claw, which can be clamped against an opposing Such small quantities of this secondary metabolite could
clawed palm, giving the pycnogonid a secure hold on be attributed to sea hare tissue in the pycnogonids' gut
slimy opisthobranchs. Once a prey item is held,A. evansi or to low levels of accumulation. Pycnogonids which
uses its chelicerae to tear pieces of flesh from the prey, were starved as controls contained no detectable peaks
which are passed to the proboscis, or alternatively, the for L. obtusa secondary metabolites on chromatographs
proboscis is applied directly to the prey. While A. evansi of samples.
dtyipgiecsatlilvye cgolannsdusmewedreallnootfcthoenssuofmtedtis(s3ue3,%soofmecasseesa,hnar=e sitCesol(leAc-tCioinnsFoifgutrhee 1r)edduarlignagLlaauteresnucmimaerob/tauustaumant tfhrroeme
18). Sea hares often released ink when attacked, although 1995 to 1998 had significantly lower abundances of the
the crouching pycnogonid was generally well clearofany sea hare Aplysia parvula on thalli where Anoplodactylus
secretions. Sea hare secretions did sometimes contact a evansi occurred (Figure 4, Table 2). There was no sig-
front leg, and pycnogonids responded to such contact by nificant difference between sites in the size of thalli col-
vigorously waving the affected limb, as though irritated. lected (Figure 4a, Table 2). L. obtusa thalli at site A were
Where pycnogonids were held together in the same hold- observed topersist during the winter, whereas thalli at the
ing tank it was noted that they would feed in groups on other sites disappeared. A. evansi occurred only at site A.
a prey item. The abundance data forA. evansi were not homogeneous
In the trial to determine if sea hares that had fed on a using Cochran's test, so aMann-Whitneytest withnormal
"novel" algae (Delisea pulchra) would be consumed,
approximation was used to compare the abundance of
Anoplodactylus evansi ate all five juvenile Aplysia par- pycnogonids between sites (A vs. B and C). The result
vula from D. pulchra that were offered. However, A. was a significant difference in pycnogonid abundance be-
evansi was deterred from feeding on A. parvula that had tween site A versus sites B and C (Z = 6.23, P < 0.001).
artificially increased levels ofsecondary metabolites (Fig-
ure 3). Pycnogonid feeding was significantly less on sea DISCUSSION
hares that had double extract levels compared to either
solvent-treated or untreated sea hares (Figure 3). The sea Anoplodactylus evansi is a generalist predator of small
hares treated with adult extract had similar weight loss to opisthobranchs, and other soft-bodied invertebrates, as
the autogenic control animals which were kept separately, are other Anoplodactylus spp. (King, 1973; Piel, 1991).
without pycnogonids indicating that little, if any, feeding This pycnogonid hunts opisthobranchs on benthic algae,
occurred. immobilizing them with movable claws on the front legs,
Pycnogonids which had fed on Aplysia parvula or A. before consuming them. A. evansi consumed whole opis-
C. N. Rogers et al., 2000 Page 335
AO/yD A-5/96 A-6/97 A-4/98 B-5/95 B-4/98 C-5/96 C-2/97 C-4/98
"1 r
1 1 \ 1 \
A-5/95 A-5/96 A-6/97 A-4/98 B-5/95 B-4/98 C-5/96 C-2/97 C-4/98
n r
A-5/95 A-5/96 A-6/97 A-4/98 B-5/95 B-4/98 C-5/96 C-2/97 C-4/98
Sample site-date
Figure 4
(A) Mean size of Laurencia obtusa (g ww) + ISE at sampling sites in Port Jackson. Samples identified by site
letter and sample date (month/year). Number ofplants per sample is shown in brackets above bars, (n). (B) Abun-
dance ofAplysiapan'ula on Laurencia obtusa (numberofseahares/g ww) + ISEatsampling sites inPortJackson.
(C) Abundance ofAnoplodactylus evansi on Laurencia obtusa (number ofpycnogonids/g ww) + ISE at sampling
sites in Port Jackson.
Page 336 The Veliger, Vol. 43, No. 4
Table 2 derived defenses of the opisthobranch mollusks investi-
gated here had no effect in deterring Anoplodactylus
Analysis of variance comparing: (A) thalli size of Lau-
evansi from attacking and consuming them. The small
dreannccieaoofbtAupslaysaitaspiatresvuilnaPoorntLJ.acokbstouns,aa(nLdn((xB)+1th)e tarbaunns-- opisthobranchs consumed by A. evansi may lack suffi-
cient quantities of such compounds to form an effective
formed) at sites in Port Jackson. Sampling dates were deterrent, although somejuvenile sea hares can have high
pooled for each site in analyses. The results of a Tukey's
HSD test showed the abundance of Aplysia parvula at concentrations of secondary metabolites (Rogers, unpub-
site A was significantly different (a = 0.05) from sites B lished data). Artificially increasing the levels of metabo-
lites present to double their average quantity did deterA.
and C.
evansi from feeding on juvenile Aplysia parvula, dem-
A
Species onstrating a quantitative deterrent effect. possible rea-
son why A. evansi did not consume all sea hares (< 300
Source of (A) L. ohtiisa (B) A. parvula mg ww) offered to them (see Figure 2) may be because
variation df F P df F P the rejected sea hares had higher levels ofsecondary me-
Between groups 2 0.85 0.429 2 13.44 < 0.001 tabolites. Hence, concentration of metabolites can reduce
Within groups 94 94 feeding by A. evansi, but most A. parvula outgrow the
prey range of this pycnogonid before they can store the
levels of metabolites necessary for deterrence. A similar
process may occur for the other opisthobranchs investi-
thobranchs, including the nudibranchs Bomella stellifer, gated, except for the ascoglossans, which attain much
Austraeolis ornata Angas, 1864, and Spurilla australis smaller maximum sizes (Table 1).
Rudman, 1982, not just the cerata as was observed for The abundance of Aplysia parvula was significantly
Anoplodactyhis can'alhoi (Piel, 1991). The digestive lower on the red alga Laurencia obtusa at site A over
gland ofsome sea hares was not eaten byA. evansi. Aply- consecutive years compared to other nearby sites where
sia par\'ula stores the bulk of acquired algal secondary Anoplodactylus evansidid not occur. L. obtusa thalli from
metabolites in its digestive gland (de Nys et al., 1996), site A were similar in size to thalli at the other sites sam-
so pycnogonids would avoid exposure to high concentra- pled. L. obtusa thalli inhabited by A. evansi were ob-
tions by rejecting this organ. Only small opisthobranchs served to persist into the winter, whereas thalli at other
less than 300 mg were eaten, as larger animals could not sites disappeared. The disjunct distribution of the pyc-
be subdued, indicating an effect of prey body size. Pen- nogonid A. evansi and sea hare A. parvula is consistent
nings (1990a) also found an effect of prey body size on with this predator excluding A. parvula from site A, al-
the capture success of opisthobranchs by Aglaja inermis though other explanations are possible.
Cooper, although in this instance, smaller individuals Anoplodactylus evansi is able to tolerate low levels of
were consumed less often because habitatcomplexity aid- secondary metabolites and feed successfully on a wide
ed predator evasion. Group feeding by A. evansi may fa- variety of small opisthobranchs with different defensive
cilitate (by weight ofnumbers) successful attacks on opis- substances. The opisthobranchs investigated here are
thobranchs larger than the prey size range found here. known to sequester different metabolites (Gunthorpe &
The nocturnal vertical movement of A. par\>ula (Rogers Cameron, 1987; Avila, 1995; de Nys et al., 1996), and
et al., 1998) and nocturnal activity ofA. evansi may also the juvenile Aplysia parvula tested here contained sec-
increase the encounter rate between predator and prey in ondary metabolites from Laurencia obtusa or Delisea
this system. pulchra, but all were consumed nonetheless. The pres-
The local distribution ofAnoplodactyhis evansi was re- ence of "novel" secondary metabolites from D. pulchra,
stricted, with pycnogonids found only at one offournear- which did not co-occur with A. evansi in Port Jackson,
by sites that had similar benthic communities (Figure 1). suggests that this pycnogonid can tolerate a variety of
A. evansi has been collected at several sites along the secondary metabolites in its diet. Given the wide range
New South Wales coastline, and as far south as Tasmania of opisthobranchs eaten by A. evansi in this study, such
(Clark, 1963), so its limited distribution in Port Jackson compounds may also include those found in sponges, cni-
contrasts with its wide distribution along the adjacent darians, bryozoans, and other dietary items of the opis-
coastline. The limited distribution of this pycnogonid is thobranchs. One reason this pycnogonid can feed suc-
most probably due to its brooding reproduction (King, cessfully on potentially poisonous opisthobranchs could
1973), and consequent limited dispersal to nearby sea- be their digestive processes involving pinocytosis and in-
weed. At site A, whereA. evansi was found, pycnogonids tracellular digestion of food particles (Richards & Fry,
occurred on several different seaweeds and a bryozoan, 1978). Isolation of prey tissue containing secondary me-
indicating that a broadrange oforganisms is used as hab- tabolites in vacuoles by pycnogonids, with eventual ex-
itat. cretion of the residual body and presumably any toxins,
Natural levels ofsecondary metabolites and otherdiet- would render such substances harmless. Disposal of sec-
C. N. Rogers et al., 2000 Page 337
ondary metabolites by this digestive process may also Hay, M. E., J. R. Pawlik, J. E. Duffy & W. Fenical. 1989.
account for the trace amounts of compounds found inA. Seaweed-herbivore-predator interactions: host-plant special-
evansi which had consumed Aplysia. The pycnogonidA. ization reducespredationon smallherbivores. Oecologia81:
418-427.
evansi is thus a generalist predator of opisthobranchs, Kandel, E. R. 1979. Behavioural Biology of Aplysia. W. H.
which may overcome their defenses by virtue of an iso- Freeman & Company: San Francisco. 463 pp.
lating digestive process and by consuming small individ- Karuso, p. 1987. Chemical ecology of nudibranchs. Pp. 31-60
uals which may lack substantial quantities ofdiet-derived in P. J. Scheuer (ed.), Bioorganic Marine Chemistry Volume
defenses. 1. Springer-Verlag: Heidelberg.
King, P. E. 1973. Pycnogonids. Hutchinson&CompanyLimited:
Acknowledgments. We thank Alistair Poore for his encourage- London. 144 pp.
ment to investigate Anoplodactylus. Timothy Charlton for pro- NoLENT. G., P M. Johnson,C. E. Kicklighter&T. Capo. 1995.
cessing GCMS samples, Dr W. Rudman for help with identifi- Ink secretion by the marine snail Aplysia californica en-
cation of opisthobranchs, and two anonymous reviewers whose hances its ability to escape from a natural predator Journal
commentsimprovedthismanuscript. C. N. Rogerswassupported ofComparative Physiology A 176:239-254.
by an Australian Postgraduate Award during the period of this Paul, V. J. & S. C. Pennings. 1991. Diet-derived chemical de-
work. R. de Nys was supported by anAustralianResearchCoun- fences in the sea hare Stylocheiliis longicauda. Journal of
cil Research Fellowship. This study was funded by Australian Experimental Marine and Biology Ecology 151:227-243.
Research Council grant # A19530672 to R D. Steinberg. Pennings, S. C. 1990a. Multiple factors promoting narrow host
range in the sea hare, Aplysia californica. Oecologia 82:
LITERATURE CITED 192-200.
Pennings, S. C. 1990b. Predator-prey interactions in opistho-
AviLA, C. 1995. Natural products of opisthobranch molluscs: a branch gastropods: effects of prey body size and habitat
biological review. Oceanography and Marine Biology An- complexity. Marine Ecology Progress Series 62:95-101.
nual Review 33:487-559. PiEL, W. H. 1991. Pycnogonid predation on nudibranchs andcer-
Burn, R. D. 1989. Opisthobranchs. Pp. 725-788 in S. A. Shep- atal autotomy. The Veliger 34:366-367.
herd & I. M. Thomas (eds.). Marine Invertebrates ofSouth- Richards, P. R. & W. G. Fry. 1978. Digestion in pycnogonids:
em Australia. Part 2. South Australian GovernmentPrinting a study ofsome polar forms. Zoological Journal ofthe Lin-
Division: Adelaide. nean Society 63:75-97.
Carefoot, T. H. 1987. Aplysia: its biology and ecology. Ocean- Rogers, C. N., P. D. Steinberg, & R. de Nys. 1995. Factors
ography and Marine Biology Annual Review 25:167-284. associatedwith oligophagy in twospeciesofseahares. Jour-
Clark,W. C. 1963. Australianpycnogonida. RecordsoftheAus- nal of Experimental Marine Biology and Ecology 192:47-
tralian Museum 26:1-81. 73.
Coleman, N. 1989. Nudibranchs ofthe South Pacific. Sea Aus- Rogers, C. N., J. E. Williamson, D. G. Carson & P. D. Stein-
tralia Resource Centre: Springwood, Queensland. 64 pp. berg. 1998. Diel vertical movement by mesograzerson sea-
DE Nys, R., p. D. Steinberg, C. N. Rogers, T. S. Charlton & weeds. Marine Ecology Progress Series 166:301-306.
M. W. Duncan. 1996. Quantitative variation of secondary Rogers, C. N., R. de Nys, T. S. Charlton & P. D. Steinberg.
metabolitesintheseahareAplysiaparvulaanditshostplant 2000. Dynamics ofalgal secondary metabolites in two spe-
Deliseapulchra. Marine Ecology Progress Series 130:135- cies of sea hare. Journal ofChemical Ecology 26:721-744.
146. Sarver, D. J. 1979. Recruitmentandjuvenile survival in the sea
DE Nys, R., S. A. Dworjanyn & P. D. Steinberg. 1998. A new haie.AplysiaJuliana. Marine Biology 54:353-361.
method for determining surface concentrations of marine Thompson, J. E., R. P. Walker, S. J. Wratten & D. J. Faulk-
natural products on seaweeds. Marine Ecology Progress Se- ner. 1982. A chemical defence mechanism for the nudi-
ries 162:79-87. branch Cadlina luteomarginata. Tetrahedron 38:1865-1873.
Farrant, P. A. & R. J. King. 1982. The subtidal seaweed com- Todd, CD. 1981. The ecology ofnudibranch molluscs. Ocean-
munities ofthe Sydney region. Wetlands 2:51-60. ography and Marine Biology Annual Review 19:141-234.
Faulkner, D. J. 1992. Chemical defences of marine molluscs. Trowbridge, C. D. 1991. Diet specialization limits herbivorous
Pp. 119-163 in V. J. Paul (ed.). Ecological Roles ofMarine sea slugs' capacity to switch among food species. Ecology
Natural Products. Comstock: Ithaca. 72:1880-1888.
Faulkner, D. J. & M. T Ghiselin. 1983. Chemical defence and Trowbridge, C. D. 1994. Defensive responses and palatability
evolutionary ecology of dorid nudibranchs and some other ofspecialistherbivores: predationonNEPacificascoglossan
opisthobranch gastropods. Marine Ecology Progress Series gastropods. Marine Ecology Progress Series 105:61-70.
13:295-301. Van der velde, J. T. & R. J. King. 1984. The subtidal seaweed
GuNTHORPE, L. & A. M. Cameron. 1987. Bioactive propertiesof communities ofWBare Island, Botany Bay. Wetlands 4:7—22.
extracts from Australian dorid nudibranchs. Marine Biology Wells, F E. & C. Bryce. 1993. Sea Slugs ofWestern Aus-
94:39-43. tralia. Western Australian Museum: Perth. 184 pp.