View metadata, citation and similar papers at core.ac.uk brought to you by CORE
provided by Aquatic Commons THE DEMONSTRATION OF CHEMOSENSORY FOOD DETECTION IN HYMENOCERA PICTA DANA (DECAPODA, CARIDEA), A PROPOSED PREDATOR OF THE CROWN-OF THORNS STARFISH ACANTHASTER PLANCI (L)
PHILIP S. RAINBOW Cambridge Coral Starfish Research Group P.O. Box 99, Port Sudan, Democratic Republic of lhe Sudan
ABSTRACT HymerlOcera pieta, the painted shrimp, is a possible predator of A canthaster planci. the Crown~of-Thorns slarfish. H. piela detects food by chemical cues alone and visual cues play no part in the initial location of prey. The presence of food in the water causes the shrimp to become more active, and distance chemoreceptors are probably present in the antennules of the shrimp. Extract of A. pfand has statistically similar attractive powers to an extract of Linckia multi/ora, the starfish supplied as food to the .<;hrimps. The painted shrimp was not attracted to fish extract (Chaetodon sp.) and may respond only to starfish. It is suggested that although H. pieta is able to kill and feed on small juvenile A. plan~i. it is probably an ineJTeclive predator against larger adult Crown·of-Thoms starfish.
Present address: N.E.R.C. Unit. Department of Marine Science. U.C.N.W.• Menai Bridge, Anglesey. Wales.
INTRODUCTION an Aean/haster individual, probabty as a Part of the research programme of the result of attack by the shrimps. From our Cambridge Coral Starfish Research Group own observations it seems probable that (ORMOND and CAMPBELL 1971; OR under natural conditions this shrimp would MOND et al. 1973) has been concerned cause little serious injury to adult A. planei, with an evaluation of potential predators of due to the discrepancy in she between pre Aeanthaster pland, the starfish responsible dator and prey, as the shrimps do not for widescale destruction of living coral usually exceed 5 cm in length whilst the (VINE 1972; CHESHER 1969). smallest starfish found on the reefs near Hymenoeera pieta Dana (Decapoda, Cad Port Sudan measured 15 Cln in diameter. dea, Gnathophyllidae), the painted shrimp, The ,hrimp, however, might well offer a has been suggested as one possible controlling serious threat to juvenile Aeanthaster planei predator (WICKLER and SEIBT 1970). as pairs of Hymenoeera pieta are able to These authors record the eventual death of kill other starfish species up to 10 cm such 184 PHILlP S. RAINBOW as Linekia multifora (Ll, Linekia laevigata fringing reef, to the laboratory in Port Sudan. lids (sealed witl and Fromia ghardaqana (personal observa The shrimps were kept in their natural tightly hy elaslic tions). As yet it has not been possible to pairs, one pair to an aquarium, and main· fora were placed test the effect of predation by H. piela on tained at an approximate temperature of tainers which '" A. planci of small diameter as specimens 28 0 C. The high ambient air temperature (l) Transparent of this size penetrate deeply into the corals defeated attempts to lower the temperature a clear view of n of the reef and are so well concealed that of the water to the 'lower average vC'iues otf cues alone) (2) " they are difficult to collect. Larger starfish the Kenyan coast but the shallow iagoons allowing the Line (greater than 15 em diameter) were found of the natural habitat of the shrimp vary chemical Clles to relatively easily and were thus available for considerably in temperature, higher values mental tank (vi experiment. often rising above 30" C. The salinity was (3) Painted white
the passage of cI WICKLFR and SEIBT (1970) suggest that maintained at 35S'~. To avoid the necessity (chemical cues al H. pieta detect their prey .either by chance for acclimation to experimental conditions, not perforated th or more normally by chemical cues. The all experiments were conducted in the tank the enclosed stal
present experiments were designed to in in which each pair of shrimps permanently ~hemosensitivity clues). A pair 0 vestigate food of the shrimp resided. The shrirn!Js were fed wi th whole one of these four and to determine whether the Crown-of .,taffish (Linekia multifora) every evening, one hour, the con Thorns starfish provides any attraction as and any food not consumed was removed a food source. 10 em from the
the next day prior to any experiment close together a~ MATERlALS AND METIlODS Small transparent perspex petridishes (5 em distance between Hvmetlocera piela were transported from diameter, I cm deep) were u:,cd in the first tainer were take the .Kenyan coast where' they are found series of experiments. These dishes were the distance mm living in pmrs in the lagoons within the weigh led down by small pebbles and the preceding two m side of eaeh tanl units to facilitat, Table 1. Behavioural characteristics uf H piela evoked during 60 minute experiments. 4 pairs of shrimps were
presented sequentiaHy with four different situations. of sensory cues for deteetion of food material. The number Any movements in parentheses is. the number of expcrirnemal re~lIlts averaged (totalled for antennules) to obtain the final shrimps were a1 value given. The experimen offer the shrimps Behavioural Chemical --I ------Yisua( Chemical and Concrol food 0'o.iects. CI
characteristic cues alone I cues alone Visual cues I No sensory cues (2emx2cmx -'-~
I~-~ .--1----- extracts of whole LATENCY by macerating a time until 45 58 I 46 I 60 shrimp alights (max 60 25 ml of seawa min) mins. (42) (38) (30) I (33) specimens were ( Percentage time 20 3 24 0 ditcH resin 10 gi (43) (40)
spent on food object (42) (33) the starfish, wi ~~ Activity chemical cues. of shrimp 1.5 0.8 1.2 0.4 behaviour of H) cm/min (41) (38) I (40) (33) apparent repelle Antennules total no of coat. The spon recordings and per were p centage Linckia moving 847 i.e. 75 313 i.e. 18 767 i.e. 74 289 i.e. 29 with the shrimps still 283 25 793 72 272 26 7DO 71 the apex of a 10 (42) (17) (41) (33) with the spong FOOD DETECTION IN HYMENOCERA PICTA 185 ilory in Port Sudan. lids (sealed with vaseline) were held on at either end of the base. Experiments again t in their natural tightly by clastic bands. Small Liuckia multi lasted one hour and at two minute intervals uarium, and rnain fora were placed in seawater in these con measurements were taken of the distance of lte temperature of taincr;, which were modified as follows: each shrimp from both sponge and starfish. nt air temperature (I) Transparent and unperforated, allowing A similar arrangement wa<; lIsed in a con· er the temperature a clear view of the starfish within (Le. visual trol experiment comparing the reactions of , average v<:!lues ofI cues alone). (2) Transparent but perforated, the shrimps to sponge soaked in starfish Ie shallow Jagoons aHowing the Unekia to be seen and allowing extract with those to sponge soaked in sea f the shrimp vary chemical cues to be released lnto the experi water, in order to discount the IX>ssibiIity ure, higher values mental tank (visual and chemical cues). of attraction to the visual s,timulus of the '. The sa linity was (3) Painted white but perforated. pennitting blue synthetic sponge. A third experiment lvoid the necessity the passage of chemical but not visual cues compared extracts of Linekia /nullilara to mental conditions, (chemical cues alone). (4) Painted white and extracts of Acunrhaster planet (the same • • lucted in the tank not perforated thus giving no indication of weight in seawater of starfish was crushed the enclosed starfish (control situation, no in 25 ml of seawater prior to soaking of the ~imps permanently -e fed with whole clues). A pair of shrimps was exposed to sponge). Final experiments compared starfish '0) eyery evening, one of these four containers for a periud of extract to fish extract (Cheatodan sp.). one hour, the container being placed initially ned was removed 10 cm from the shrimps which remained RESULTS ly experimen t. close together as a pair. Readings of the The first series of experim~nts were an x petridishes (5 em distance betw~en each shrimp and the con attempt to discover whether H. piela uses ;: used in the firs t tainer were taken every two minutes and chemical or visual cues in di'scovering its ~ese dishes were the distance moved by each shrimp in the prey. From the recordings taken it was pebbles and the preceding two minutes was estimated. The possible to estimate the average percentage side of each tank was marked off in 5 em of the experime~tal hour, spent by a shrimp airs of shrimps were units to facilitate these distance estimates. on any particular container, and -the average latt:ria1. The number Any movements of the antennules of the latency period of each situation~the time (0 obtain the final shrimps were also recorded. taken until a shrimp alighled on a container, The experiments were then redesigned to remaining there at least five minutes (this •• offer the shrimps a dir~t choice of potential latter condition overcomes errors caused by Control food ""iceIs. Cuboids of synthetic sponge chance ~ncounters). The results are shown No sensory cues (2 em x 2. em x 1 cm deep) were soaked in in Table 1. , extracts of whole Unekia /nllitilara. prepared It was noted" in preHminary observations by macerating a 5 I.:m diameter starfish in that the shrimps became more active when 25 ml of seawater. Other similar Linekia in the vicinity of food. agreeing with (33) spo.:imens were dried and covered in "AraI o observations made on other marine decapod dite" resin to give a transparent layer over crustacea (HAZLETf 1'171a and 1971b). <33) the starfish. which prevented leaking of Similarly. it was noted that the antennules chemical cues. Close observation of the were repeatedly active when food was intro 04 behaviour of Hymenoeera pieta revealed no duced to the tank and so Table 1 includes a (33) apparent repellent etrects of the "Araldite" measurement of the average distance moved coat. The sponge and the resin covered by the shrimps (excluding time actually spent !""'inckia were placed together in the tank situated 011 the containers) and a ratio of the /j~ i.e. 29 with the shrimps. the pair of H. pieta fonning number of recordings of antennules actively 00 7t the apex of a 10 cm sided equilateral triangle, moving to the number of recordings of the (33) with the sponge and resin-coated starfish antennuJes being stationary. The number of •
186 PHil lP S. R\INBOW Table 2. Statistical compari~olls of results in four situations shown in Table I. actu. in tI Situations -----Behayjoural Student's expel
compared j characteristic T x' p the t ~~I~_~_ I , , fresh \I. Latency 1.52 p>O.1 Visual Control disturl Percentage time 1.22 p>O.1 Activity 2.10 O.05>p>O.O:! most I Anrennule movt I 0.72 p>O.5 fully a Visual v. Chemical LatlOncy I 3.S5 p
expedments carriej out for each situation two minutes of the distance between each DISCUS is also given in the tahle. the figures given shrimp and each choice objocl and thc~;e It wou being an average lJ"f all experimental results. distances have been totalled to give the to starfis\ Table 2 shows statistical comparisons values (D) shaWl!. A percentage difference by chern11
bety..'cen the different situations using Stu· (Do-D,) of these values is also presented. and 1971t denf~ T and Chi s4uared analyses. There The first choice expedme.llt5 confirm thc SON (196
is a .'iignificant ditIcrence between results attra..:tive powers of the starfish extracts for demonstra obtained for chemical cue.';; alone and those the control experirnent renlOves the pos~ihility attraction fOf visual cues alone, whilst visual cues of v.isual attraction lo synthetic sponge. In Visual stil produce statistically similar results to the the next experiment, a direct comparison was in the ini control situation where no stimuli arc appar made between the reactions caused by Linckia pMsiblc II ent. Chemical cues simihuly do not seem 10 multi/ora and those evoked by Acamhaster ing activil have an increased effect when associated planci. The totalled values of the distance reported with visual cues. between shrimp and Linekia extract, and V;ffU{US (f From this experiment it is apparent that shrimp and Acanthasler extrac,t are seen 10 likely f"r H. picta uses chemical cues alone. The be similar when the percentage difference observed results are, however, open to criticism, for (Do-D,) is compared with those of other to a dark there exists the possibility that a Lillckiu choice experiments. From consideration of Moreover contained in a. transparent perspex container this value and of the other L.'Omparisons made starfish, may not provide an adequate visual stimulus. in 'fable 3, it can be seen that the t\'10 SYMON, A second series of experiments provides a starfish have a slmilar attraction. cues have more natural s,timulus (a coated starfish) Experiments were also carried out to dis "f the cr \,:oupled with an extremely unnaliJral chemical cover the degree of specificity of the chemo In prel
source (synthetic sponge). Table 3 shows sensory response of H. pictl1. The effect of that the results obtained when the shrimps were fIsh extract was compared with that of star~ move ani presented with various choices. A shorler fish extract and results suggesl that fish' was plact lateHcy period probably indicates a stronger ex.tract does not attract Hymenocera piela. that the chemical source than in previous experj In the experiments described, it may be chemical mcnts. As stated, rocordings were made every noted Ihat in many cases no choice was (Table I FOOD DETECTION IN HYMENOCERA PICTA 187
actually made. This IS probably due -to faults together but were able to move independ in the experimental design, for although ently, and in some cases shrimps were seen experiments were carried out at intervals, moving their chelipects over their antenllules p the tanks were only flushed through with as though cleaning them. When nearing the fresh seawater at fortnightly intervals to avoid food source, the shrimps would often make p>O.1 p>O.! disturbance of the shrimps. The shrimps repeated deep bows, bringing their antennules 0.05> p>O.O:!. most used for experiments moulted success closer to the food source, but once lhe p> 0.5 fully at approximately eighteen days' intervals shrimp was actually feeding, all antennular p
In the list of hehaviour patterns exhibited -----chemical cues alone. vi.mal cues alone,
by the hermit crab Pctrochirus diogenes chemical and visual cues together, and a during I..~hemotaetic stimulation of feeding, control situation with no cues apparent. HAZLETT (197Ib) includes increased ]oco Chemical cues attract the shrimp 1O a poten motion, A similar increase in locomotor tial food source, a process involving an activity is also shown by H. picla. Other increas.e jn the locomotor acHvity of the behaviour p (haster "plagues", where it is the adult situated in the antennules of the shdmp. starfish that are present in large numbers A second series of experiments confirm~d (CHESHER 1969; DANA, NEWMA:-l and the attractive power of starfish extract as a FAGER 1972; VINE 1972), Against these stimulant for food-seeking behaviour in H. larger adults H, piela probably has little pichL Ex[ract of A. planci attracts the shrimp significant predatory capacity. A high con to a statistically similar extent as does lhal centration of adult A. fllanci wm possibly of Linckia multifora, a starfish to which have heen precedej by a high concentration the shrimps were accustomed as a food of juvenile starl1sh, susceptible to I.:ontrol source. H. piela was not attracted to fish hy predation by H, pieta. However a further extract (Chaetodon sp,) and may be allmcted explanation exists in that A. planci release only to starfish under natural conditions. an altractant during feeding (ORMOND It is suggested that H. picta is possibly (!f uf. lQ73) leading to the migration and an effective predator aglJ.inst juvenile A. aggregation of adult starfish. planei, but is ineffc:ctive against the larger adult starfish making up an Acanthasler SUMMARY "p,lague". Hymenocera pieta, the painted shrimp, has been put forward as one potenlial pre RESUME dator of Acanrha.\ter plunci, the CrO'Nn·of· Ce travail -rente d'evaluer l'lmportance de lhorns ::,tarfish, which i" known to feed on 1a Crevette Hymenocera pieta ~n tant que coral and may destroy coral reefs, Experi· predateur de l'Echinoderme Aeanthas/a ments were conducted to discover tht: method piand, ce dernier etant bien connu camme used by the shrimp to detect food and to destructeur de coraux. Des experiences en detenn'ine whether H. pieta js attracted to vue de determiner les methodes de detection A. pfanci as a food source. de naurriture par H. piChi et son attirance A first series of experiments compared the vers A. pranei comme source de nourriture reactions of the shrimps to food objects, sont decrites. offering four different forms of sensory cues Dans une pn:miere serie d'experiences les FOOD DETECTION IN HYMENOCF.RA PleTA 189 Jal cues alone, ogether, and a I cue.... apparent. imp (0 a poten 1_"_1 5 i involving an 8 o 8 activity of the :i "ently play no "o o o 6' o 6' of the shrimp > pattern evoked " offering visual imilar to that lon, where no imilarly. visual }fi of chemiL:al ical and visual similar to the o o o o ,. The distance are probablv " I the shrimp. ~"§ II :nts confirmed '§ ~ 1 extrac( as a o 'D lavjour III H, ~ I ~ cts the shrimp --I I as does that ~ t ish to which g ~ -1 8 , I as a food o o o , o v V f\ v acted to fish " " " Vbe attrac ted conditjans. 002l __I 'D •.. " JS possibly juvenile A. ~ 51 the larger of _ Uof 'D o .::! o "~ c c ..... '0 Acanthaster .c , u -"u -"u ,. o o c oS c c [partance de en tant que A canthaster I nnu camme ~riences en ie detection -L )0 aWrance , naurriture eriences les o - 190 PHILIP S. RAINBOW mg behaviour oj reactions de la crevette soot analysees lorsque nourriture--chez 1a crevette. Statistiquement, vittatus (Decapc celle-ci est mise en preseIh,.:e de diffe-rents il n'y a aucune difference significative entre 15: 305-311. types de "nourriture~objets" offrant les la reponse de H. picla aux extraits d'A. planci Hazlett, B. A. (1971. differentes formes d'indices de detection et 1a reponse aux extraits de l'Echinoderme in marine decapl Phys;o[. 18 (1): suivantes: des indices chimiques seuls. de.'" Linckia multifora. une etoile de mer sur Hazlett, B. A. (1971 indices visuels seuls, des indices chimiques laquelle se nourdt habituellemenl Ja crevette. stimulation of fe et yisuels combines et un temain n'offrant H. picla n'est pas attiree par des extraits crab Petrochiru aueuo ind ice apparen1. Les experiences de poisson (Chaetodon sp.). Par ailleurs, dans Physiol. 39 (41 montrent que les indices chimiques attirent des conditions naturelles H. picla est prob Levandowsky, M., Amino acid an< aHin~e la crevette vers une source potentielle de ablemenl uniquement des etoiles de Compo Biochem nourriture; ce processus s'accompagne d'uo mer. Mackie, A. M. (197: accroissement de I'activite locomo(rice de H. picla est peut-etre une predatrice effi detection in the 1a crevette. Les indices visuels ne jouent, cace de juveniles d'A. planci mais 1a crevelte Mar Bioi, Berti apparamment, aucun role dans la phase est inefficace contre l'etoile de mer adulte, Ormond, R. F. G., Observations o~ constit~ initiale de I'attraction de 1a crevette vers sa celie qui Ie "Reau" Acanlhasler. coral reef echiI nourriture. II n'y a aucune difference signi ficative enlre 1a reponse provoquee par un stimulus "nourriture-objet" otfrant des in dices visuds seuls et 1a reponse provoquee par Ie temoin. De ]a meme falYon, les indices visuels n"accroissent pas'la reponse aux in dices chimiques lorsque les deux sont com ACKNOWLEDGEMENTS: [ wish to express my bines: il n'y a aucune difference significative thanks to members of the Cambridge Coral Star entre 1a 'reponse aux indices chimiques et fish Research Group, especially Dr. C. H. Roads, for assistance and co-operation during this work visue1s conlbines et celie aux indices chimi I am especially indebted to Dr. R. F. G. Ormond gues seul:;. Les chemorecepteurs de H. picla and ]\'1r. J. K. G. Dart for advice and helpful som probablement siwes dans les antennu1es discussion in the course of the experiments. My de 1a crevette. thanks go to Dr. D. A. Jones and Dr. Graham Une seconde serie d'experiences a confirme Walker for their critical review of the manuscript and valuable comments. I also wish to thank que des extraits d'etoile de mer declenchaient Edwin Bourget for his kind assistance in trans etfectivemcnt lc comportement-recherche de lating the summary. REFERENCES by Ihe sea star Acan/haster planci. Science N.Y. 165, 280·283. Ache, B. W. (1969). The sensory basis of host Dana, T. F., W. A. Newman, and E. W. Fager recognition in symbiotic shrimps of the genus (l972). Acarllhastt'r aggregations: Interpreted Betaeus. Am. Zool. 9: 1096. as primarily responses to natural phenomena. Ache, B. W., and D. Davenport (1972). The Padf. Sci. 26 (4): 355-372. sensory basis of host recognition by symbiotic Gravier, C. (1921). Sur un Decapode macrome shrimps, genus Betaeus. Bioi. Bull. mar. bioI. des mers indo-pacifiques (Hymenocera eIegan.~ Lab. Woods Hole 143: 94·111. Heller). Bull. Mus. nalll. Hisi. naI., Pari.\'. Balss, H. (J 940). Decapoda. Bronn, Kl. Ord. 1921, 424·427. Tierreichs. V, J Abt., 7. Buch, 1. Lfrg: l-I60. Gray, 1. E .. L. R. McCloskey and S. C. Wiehe Case, J. (1964). Properlies of the dactyl chemo (1968). The commensal crab Dissodaclylus receptors of Cancer antennarius Stimpson and mellfrae and its reaction to sand dollar host Published c. produclus Randall. Bioi. BulL. mar. bioI. factor. J. Elisha Milchell scient. Soc. 84: Headqua Lab. Woods Hole 127: 428-446 , 472-481. Chesher, R. H. (J 969). Destruction of pacific corals Hazlett. B. A. (1968). Stimuli involved in the feed. FOOD DETECTiON IN HYMENOCERA PleTA 191 Slalistiquemenl, mg behaviour of the hermit crab Clibanarius Sea. Symp. zool. Soc. Land., No. 28: 413-454. vittatus (Decapoda, Paguridae). Crustoceana Ormond, R. F. G.• A. C. Campbell, S. H. Head, ignificative entre 15, 305-311. R. J. Moore, P. :So Rainbow and A. P. traits d'A. planci Hazlett, B. A. (197Ia). Antennule chemosensitivity Saunders (1973). Formation and breakdown e [' Echinoderme in marine decapod crustacea. J. Anim. Morph. of aggregations of the Crown~of-Thoms star i1e de mer sur Physiol. 18 (1): 1-10. fish, Acanthaster planci (L). Nature, Lond., nent la crevelle. Hazlett, B. A. (1971b). Chemical and chemotactic 246: 167-169. stimulation of feeding behaviour in the hermit Symons, P. E. K. (1964). Rehavioural responses. of mr des extraits crab Petrachirus diogenes. Camp. Biochem. the crab Hemigrapsus oregonensis to tempera ·ar aH1eurs, dans Physiol. 39 (4A): 665-670. ture, diurnal light variation and food stimuli. piela est prob Levandowsky, M., and E. S. Hodgson (1965). Ecology 45: 580-591. des etoiles de Amino acid and amine receptors of lobsters. Vine, P. J. (972). Recent research on the Crown Compo Biochem. Physiol. 16: 159-161. of-Thorns starfish. Underwater Journal 4 (2): Mackie, A. M. (1973). The chemical hasi.o> of food 64-73. predatrice cffi detection in the lobster Homarus grammarus. Wickler, von W., and U. Seibt (1970). Das Ver mais la creYet!c Mar Bio'. Berlin 21: 103-108. halten von Hymenocera pieta Dana, einer de mer adulte, Onnond. R. F. G., and A. C. Campbell (1971). Seesrerne fressenden Gamete (Decapoda, Observations on Acanthaster planci and other Natantia, GnathophyUidae). Z. TierpsychoI " A canthaster. ·I coral reef echinoderms in the Sudanese Red 27: 352-368. :h to express my idge Coral Star k C. II. Roads, luring this work . F. G. Ormond 'ice and helpful I :xperiments. My nd Dr. Graham the manuscript wish to thank :tance in trans. •• • pland. Science i E. W. Fager ns; Interpreted al phenomena. 'Ode macrourc nocera rlegans I t. nal., Paris, I. S, C. Wiehe DissodQcfylus d dollar host - ~nt. SaL 84: Published by East African Literature Bureau. East African Community Regional Headquarters. Ngong Road, P.O. Box 30022, Nalmbi. and printed by E.n~1ish Press. P.O. Box 30127, Accra Road, Nairobi, Kenya. d in the feed.