PLEASE SCROLL DOWN FOR ARTICLE

Transcription

1 This article was downloaded by: On: October 00 Access details: Access Details: ree Access Publisher Taylor & rancis Informa Ltd Registered in England and Wales Registered Number: 0 Registered office: ortimer House, - ortimer Street, London WT JH, UK Italian Journal of Zoology Publication details, including instructions for authors and subscription information: eeding habits of Triturus vulgaris, cristatus and (Amphibia, Urodela) in the northern apennines (Italy) auro asola a ; Luca Canova a a Dipartimento di Biologia Animale, Università di Pavia, Pavia, Italy Online Publication Date: 0 January To cite this Article asola, auro and Canova, Luca()'eeding habits of Triturus vulgaris, cristatus and (Amphibia, Urodela) in the northern apennines (Italy)',Italian Journal of Zoology,:, 0 To link to this Article: DOI: 0.00/0000 URL: PLEASE SCROLL DOWN OR ARTICLE ull terms and conditions of use: This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

2 i Boll. Zool. : -0 () eeding habits of Triturus vulgaris, cristatus and (Amphibia, Urodela) in the northern Apennines (Italy) AURO ASOLA LUCA CANOVA Dipartimento di Biologia Animale, Università di Pavia, p.za Botta, I-00 Pavia (Italy) ABSTRACT Although the aquatic cycle of some newts spans the entire year with various life stages, newt diet had only been studied during single phases of the cycle. We analyzed the diet of the warty newt, alpine newt and smooth newt, living syntopically in the same pond, during all the life stages (, juveniles,, ) and throughout one entire annual cycle. ood samples were obtained by stomach-flushing newts. The newts were found to exploit a large variety of prey, mainly aquatic insects (imagoes and ), and other aquatic invertebrates, together with small numbers of terrestrial Arthropoda that had fallen into the pond. The dietary similarity among species and life stages was high. Although newts were able to exploit most of the range of available prey sizes, on average they selected prey size in relation to their own weight. The frequency of capture of some prey differed significantly between day and night. The six most frequent prey, equivalent to % of prey items, were captured with marked seasonal variations. Gastropoda, Collembola, Homoptera and Diptera were captured in relation to their cycles of availability in the pond, while the other variable prey were continuously available, and were probably exploited only when the fluctuating prey were absent. The patterns of dietary similarity among species and of seasonal and diel plasticity, indicate that newts are opportunistic and do not select specific prey taxa. Prey are selected on the basis of size, although all newts may catch a wide range of prey, with some limitations for large items. In newts, as in the other Urodela, food size is more important than type in the partitioning of feeding resources. KEY WORDS: Diet - Prey selection - Triturus vulgaris - Triturus cristatus -. ACKNOWLEDGEENTS We wish to thank Sergio ezzadri, Raffaella Alien, rancesca ogavero and Laura Pettiti for having greatly helped our work, and Cristina Giacoma for her friendly advice. (Received June - Accepted September ) INTRODUCTION Information on the feeding habits of newts for southern Europe is very scanty. Prey types and other aspects of feeding activities have been studied for the warty newt Triturus cristatus, the marbled newt marmoratus, the alpine newt, the smooth newt vulgaris, the palmate newt helveticus and the Italian newt italicus (Braña et al., ; Dolce & Stoch, ; Joly, ; Joly & Giacoma, in press). These studies generally describe the diet of single life stages of the newts, during a single phase of their aquatic cycle. However, some species of newts span their aquatic cycle over the entire year, with various life stages, and their feeding activity exhibits large seasonal differences (Chacornac & Joly, ; Verrell, ). In this paper we analyze the diet of warty newts, alpine newts and smooth newts, living syntopically in one pond. We aimed at comparing the newt diets by species and by all their aquatic life stages (, juveniles,, ), to study the variations in the prey exploited throughout the annual cycle, to check whether newts are selective or opportunistic in the choice of their prey, and to test whether they select their prey by prey type or by size. Our community of newts, with three species, is one of the richest to be found in Europe. This is the first complete description of the feeding relations of newt species, covering all the aquatic life stages and the entire annual cycle. ATERIALS AND ETHODS The newts were captured in a permanent pond, at 0 m a.s.l. in the Apennine mountains (province of Piacenza, northern Italy, UT T-NQ-0). The pond reached its maximum size (x0 m, 0 cm deep) during April, and shrank to a minimum size (x m, 0 cm deep) during August. We conducted sessions of capture, once or twice per month, from April to arch 0. Occasional captures were conducted in two other ponds, within km of the main pond. The number of newts of each species and the frequency of capture of the 0 most frequent classes of prey did not differ significantly between ponds Ct test), and we therefore pooled the data in order to obtain sufficient samples. At each trapping session we placed 0-0 funnel traps, regularly spaced over the pond, suspended on the surface and on the bottom; other traps were placed at depths of 0 cm and 00 cm when available. The traps, made from plastic bottles (Griffiths, ), were left in position during h, and were checked in the evening and morning (within h of sunrise). rom November to ebruary the traps were placed through holes in the ice covering the pond. The traps were effective in catching only when they had grown over 0. g. Supplementary newts were captured by hand netting, and this was effective in catching over 0. g. Each captured newt was weighted, and its stomach was flushed by injecting 0 cc of water, using a syringe fitted with a capillary covered by plastic tubing. The newts were weighed with their full stomachs before flushing, because repeated measurements had shown that after flushing the weight of the newts with empty stomachs was biased by retention of water in their digestive tract. The prey were classified as accurately as possible, and the three dimensions of all intact prey were measured to the nearest 0. mm, under a binocular microscope. The volume of each prey item was calculated by the appropriate formula: cylinder (worms, insect lar-

3 . ASOLA, L. CANOVA vae, newt ), hemi-spheroid (Aracnida, Copepoda, metamorphosed insects), spheroid (Bivalvia, Ostracoda, Cladocera), cone (Gastropoda), following a similar method used by Griffiths (). The efficiency of stomach flushing was tested by dissecting 0 recently flushed alpine newts and warty newts: flushing had removed % by volume and % by number, of the prey present in their stomachs. Chacornac & Joly () found that the teguments of medium sized arthropods are broken in the stomachs of alpine newts in - h at C. Therefore the prey we found in the morning and in evening could not have been captured before the preceding night or day, and this allowed us to compare the diurnal variations in diet. Breadth in diet was calculated as: B = l/r'sp? (Levins, ), where p is the proportion of the prey i taken by the newt, out of all the R prey types used; this index may range from /R (lowest breadth, only one resource used) to (all resources used in equal proportions). The differences between diurnal and nocturnal frequency of the prey were compared using the Bonferroni confidence intervals (Byers & Steinhorst, ), that tests the frequency of each prey type during a given period with the frequency expected under the null hypothesis that the prey were captured evenly during day and night. Diet diversity was estimated by the Shannon index: -(p,) (log p,). RESULTS ood samples were obtained from newts: and of warty newts, alpine newts, and smooth newts, branchiate juveniles and of alpine newts, and a few juveniles and of warty newts (Table I). We categorized as juveniles those newts that had hatched in a previous year and were not sexually active, and as those branchiate newts that were sexually active, as shown by their external characters. The diets of the juveniles and of the of warty newts are not described in this paper, because they were very scarce and our sample size was insufficient. The species and the life stages differed in the length of their' residency in water throughout the year (Table I). The residency in water of the newts in our study pond is described in detail in asola & Canova (in press). Diet The newts exploited a large variety of prey (Appendix A). Aquatic insects (imagoes and ), and other aquatic invertebrates were the most frequent prey of all the newt species and life stages. Certain terrestrial Arthropoda that had fallen in the pond were captured in small numbers, except for winged instars of Aphididae (Homoptera) of which many were captured in large numbers (Appendix A). Eggs and of amphibians were captured infrequently: these, a relatively large prey, were only taken by warty newts, the largest of the three species. Although the variety of the prey taxa was high, the bulk of the diet consisted in a few prey only, for all newt species and life stages. When all the newts are taken together, six taxa make up % of the prey: aquatic Crustacea %, terrestrial Homoptera %, Diptera %, Collembola 0%, Gastropoda %, Trichoptera % (Appendix A). Each newt species and life stage relied on the same main types of prey, and their dietary similarity was high (ig. ). The diet of the of the three species was mainly Homoptera, with large numbers of Gastropoda, Crustacea, Collembola, and of Trichoptera and Diptera. The branchiate life stages (the of the three TABLE I - requency and seasonal presence of the individuals of the three species of newts, from which food samples were obtained. Species Sex Stage Total number. Seasonal presence from to Triturus cristatus juveniles (non-branchiate) 0 0 mid July - mid November early ay - early July mid June mid ebruary juveniles (branchiate) 0 0 mid July - mid January Triturus vulgaris 0 mid July - mid November mid April - mid ay early ay - mid July

4 EEDING HABITS O THREE NEWT SPECIES aquatic benthos aquatic neuston îivalviï _.-usucei undetermined Insecta Collemboli Heteroptera r Jibia eggs of Amphibia Tiphenieroptera Bl, Odonata Plecoptera Neuroptera Trii Co, cristatus B=. B=. I vulgaris B=. terrestrial aquatic benthos aquatic neuston Nemi JJivilvia lmdeterminedlsseaa Collembola Heteroptera Cole of Amr eggs of Amphibia larvu Tricfco larvátüóleol juveniles cristatus B=. vulgaris B=.O terrestrial proportion of prey items ig. - Prey taxa in the diet of the three newt species and of their life stages throughout the entire year. The proportions are calculated from the total number of items of each species. Diet breadth is expressed by the index B. species, and the juveniles and neotenic alpine newts) mainly captured aquatic Crustacea,'and also large numbers of Collembola, Homoptera, and of Ephemeroptera, Diptera, and Trichoptera. The dietary breadth was slightly larger for than their respective (ig. ). Size relations The weights of the newts ranged only from 0. g for smooth newt to 0 g for adult warty newts, and some species or life stages had little or no overlap with others: e.g., adult smooth newts and the did not overlap in size with warty newts (Table II). On the other hand the prey ranged widely, from 0.00 mm for Crustacea and Ostracoda to 0 mm for a preyed newt larva, and they largely overlapped (five orders of magnitude out of six, Table II). All newt species and life stages were able to exploit the smallest prey (Crustacea, Ostracoda), whereas the largest prey (above 00 mm, newt, and Annelida) were only captured by the adult warty newts, and the relatively large prey (from 0 to 00 mm, mainly insect ) only by the of the three species and by the neotenic alpine newts. Therefore the exploitable prey were limited by their maximum size. On average, the prey size was selected in relation to newt weight (ig..). This relation holds both when the newts are distinguished on the basis of species and life stage (left of ig. ), and when they are grouped on the basis of size classes (right of ig., limits of the weight classes: 0.,,.,,,,, 0, 0 g). The maximum and minimum prey sizes were also correlated to the weight of the newts, except for the minimum prey when classed by newt species, which was not significant. The correlations were generally closer when prey volume was classed by newt weight than when classed by newt species (ig. ). The selection of given prey volumes is therefore due to the size of the individual newts, and not to any speciesspecific preferences by the newts. Although the newts captured their prey over most of

5 . ASOLA, L. CANOVA TABLE II - Weight of the newt species and of their life stages, and volume of their prey items. Newts weight (g) Prey volume (mm ) median lower-upper range median lower-upper range quartile quartile Triturus cristatus Triturus vulgaris Triturus cristatus Triturus vulgaris juveniles the range of available sizes, on average they were selective and adjusted the average size of their prey in relation to their own weight. No significant relationship was found between newt weight and the taxonomic or dimensional diversity of the diet, measured by the Shannon index. ta o cu I BY SPECIES P<0.0 V=-.+.W P<0.00 V=-+W N.S. BY WEIGHT CLASS.0 r P<0.00 V=-.+W newt weight (g) P<0.00 V=-+W P<0.0 V=O.00O+O.0O0W ig. - Correlation and regression of the average, the maximum and the minimum prey volume on the weight of the newts, grouped by the eigth species and life stage (left) or by nine size weight (right). The axes show the ranges of the values. The scale for maximum prey volume is logaritmic. The P values show the probability associated with the Pearson correlation coefficient. The equations show the linear regression between prey volume (V) and newt weight (W). Z Day-night differences In all the newt species and life stages the total volume of prey did not differ significantly between the newts captured within h of sunrise and those captured in the afternoon (ann-whitney test); they fed both during the day and the night. However the frequency of capture of some prey differed significantly (P<0.0) between day and night (ig. ). The prey whose capture did not differ (not shown in ig. ) were the least abundant, and it is likely that a larger sample would have revealed differences in these prey, too. Seasonal variations The six most frequent prey, which together made up % of prey items, were captured with marked seasonal variations (ig. ). These variations occurred in parallel among the four life stages and species of igure, as shown by the significant correlations among the prey frequencies of all four classes of newts, when tested in pairs (Spearman rank correlation, P<0.0). This pattern Annelida \^ Gastropoda Crustacea Collembola Ephemeroptera Trichoptera Díptera Homoptera (terrestrial) f Heteroptera (terrestrial) fc Thysanoptera (terrestrial), night day proportion of prey items ig. - Differential capture of some prey taxa by day and night. Only the taxa that differed significantly (Bonferroni statistics based on prey, P<0.0) are shown.

6 EEDING HABITS O THREE NEWT SPECIES Vi e I 0 o 0. o ti 0. ocl, Gastropoda Crustacea Collembola Trichoptera Díptera Homoptera l/l. /u aquatic aquatic VA/ terrestrial A 0 j AJ'J'A'S'OÑD jïaj'j'a's'owd J ÎA'J"A'S'OVD J"AJ'J'A'S'OWD J "AJ'J'A'S'OD J'AJ'J'A'S'OWD! cris ta tus juveniles ig. - Annual variations of the six main prey types in the diet of the adult, neotenic, and juvenile newts. The prey of smooth newts and their are not shown, because these newts were present for short periods. The proportions are calculated from total prey of the species through each month. suggests that all the newts responded opportunistically to the same variations in the abundance of their prey. DISCUSSION The most frequent prey described in other regions for adult warty newts, alpine newts and smooth newts differ from those found in our pond. Elsewhere, adult warty newts were found to prey mainly on Crustacea and Hirudinea (Griffiths & ylotte, ), while in our study Homoptera and other insects were prevalent; adult alpine newts mainly on Crustacea, Díptera and Ephemeroptera (Braña et al, ; Joly, ), while in our study Crustacea and Collembola; adult smooth newts mainly on Crustacea and aquatic insect (Dolce & Stoch, ; Griffiths, ) while in our study mainly on Collembola, terrestrial Homoptera and Crustacea. Therefore, the prey taxa taken by adult newts vary from one site to another. Adult newts presumably exploit differing prey taxa opportunistically in relation to their local abundance, and the inclusion of certain prey types in the diet depends on the availability of the more profitable prey. On the other hand the diet of the of warty newts and smooth newts seems to be consistently based on Crustacea (Cladocera, Copepoda and Ostracoda) as shown by Avery (), Dolce & Stoch (), and by our study. These Crustacea are probably the only abundant and small prey available to newt in all the areas. Cannibalism and oophagy, observed in warty newts, is widespread among amphibians, as a by-product of normal predatory behaviour (Polis & yers, ). Joly () found the same differences in the rate of capture of Diptera and of Crustacea in alpine newts between night and day as were found in our study. Newts adopt a plastic predatory behaviour, switching prey types between night and day. Size relations are important in aquatic prey-predator systems (Osenberg & ittelbach, ), and correlations between the size of the predator and that of the preferred prey have been observed in some species of Urodela (Braña et al., ) and Anuran (Toft, 0). Prey size is however highly variable within a species of predators (itchell & Taylor, ). In our study, the average and the maximum prey size were correlated to the size of the newt, whereas the minimum prey size was less, or not at all, correlated with the size of the predator, a pattern found in some newts and salamanders (Lynch, ; Griffiths & ylotte, ). The newts are therefore able to exploit a wide range of prey sizes with some limitations for the large ones, even though on average the newts select prey of the preferred size. The increase of prey diversity with newt size, observed by Braña et al. (), was not confirmed in our study. arked seasonal variations occurred in the prey exploited by each newt species and life stage, and the variations were similar among all the species. The Gastropoda peaked from October to November; of these Lymnaea (Polmönata), reproducing during the summer, and with young growing during the autumn (Girod et al., 0), was the most abundant Gastropod in the diet. It is likely that the few Gastropoda captured during spring and summer were those born in the preceding year, and that they were preyed upon abundantly only during the autumn when those born in the preceding summer attained a suitable size. This is supported by the size of the captured Gastropoda: their average length was. mm

7 . ASOLA, L. CANOVA from January to September, and was smaller (. mm, probably snails born in the same year) when their abundant capture began in October; in late October their size grew to. mm and to. mm in November. The Collembola peaked in the diet from August to November; this peak may be related to the general abundance of the Collembola during autumn (Houlbert, ). The Díptera (mainly Chironomidae) emerge from their pupae in large swarms and in differing periods from April and September, depending on the species and thermal conditions (Bertrand, ); they were probably captured when available in large numbers during emergence outbreaks. The two peaks of the Homoptera (Aphididae) capture, in ay and in October, may be related to the development of the winged instars, which disperse from the host plants biannually, in spring and in late summer or autumn (Grandi, ; Johnson, ). Only the winged instars are prone to fall into water during the dispersal flights. These four prey taxa were therefore captured in relation to their cycles of availability in the pond. The other seasonal prey generally have complex life cycles, with a succession of generations through the year (Crustacea), or with overlapping pluriannual cycles (Trichoptera: Grandi, ): these prey were captured throughout the year. Their variations may be due to the switching by the newts to these prey, which are continuously available, when the other fluctuating prey were absent. The newts therefore exhibit a seasonal plasticity (also observed by Chacornac & Joly,.;, in alpine newts), allowing them to exploit temporary resources.. In conclusion, the prey taxa are highly similar among the newt species and life stages, while: wide differences exist in prey of the same species between areas. These patterns, together with the prey variations between night and day and between seasons, indicate opportunism and no selection of prey taxa by the newts!.prey are selected on the basis of their size, although all the newts may catch a wide range of prey, with some limitations for large ones. In newts, as in the other Urodela (Toft, ), food size is more important than type in the partitioning of feeding resources. REERENCES Avery R. A., - ood and feeding relations of three species of Triturus (Amphibia Urodela) during the aquatic phases. Oikos, : 0-. Bertrand H., - Les insectes aquatiques d'europe. Le Chevalier, Paris, pp. Braña., de la Hoz., Lastra C., - Alimentaciòn y relaciones tròficas entre las larvas de Triturus marmoratus, y helveticus (Amphibia: Caudata). Doñana Acta Venebrata, : -. Byers C. R., Steinhorst R. K., - Clarification of a technique for analysis of utilization-availability data. J. Wildl. anage, : Chacornac J.., Joly P., - Activité prédatrice du triton alpestre () dans un lac alpine ( m, Alpes françaises) Acta Oecologica, : -0. Dolce S., Stoch., - Osservazioni sull'alimentazione degli Anfibi: I «Triturus vulgaris meridionalis» (Boul.) degli stagni del Carso triestino (Italia nordorientale). Atti us. civ. St. nat., Trieste, : -. asola., Canova L. - Residency in water by the newts Triturus vulgaris, cristatus and in a pond in Northern Italy. Amphibia-Reptilia, (in press). Girod A., Bianchi I., ariani., 0 - Gasteropodi (Gastropoda: Palmonata. Prosobranchia: Neritidae, Viviparidae, Bithyniidae, Valvatidae). Guide per il riconoscimento delle specie animali delle ácque interne italiane,. CNR, AQ//. Consiglio Nazionale delle Ricerche, Roma, pp. Grandi G., - Introduzione alio studio dell'entomologia. Edagricole, Bologna, pp. Griffiths R. A., - A simple funnel trap for studying newt populations and an evaluation of trap behaviour in smooth and palmate newts, Triturus vulgaris and belveticus. Herpetol. J., : -0. Griffiths R. A., - eeding niche overlap and food selection in smooth and palmate newts, Triturus vulgaris and belveticus, at a pond in mid-wales. J. anim. Ecol., : 0-. Griffits R. A., ylotte V. J., - icrohabitat selection and feeding relations of smooth and warty newts, Triturus vulgaris and cristatus, at an upland pond in id-wales. Holarct. Ecol., 0: -. Houlbert C., - Thysanoures, Dermaptères et Orthoptères de rance et de la faune europèenne. Doin, Paris, pp. Johnson C. G., - igration and dispersal of insects by flight. ethuen, London, pp. Joly P., Giacoma C. - Limitation of similarity and feeding habits in three synthopic species of newts (Triturus, Amphibia). Holarct Ecol., (in press). Joly P., Giacoma C., - Limitation of similarity and feeding habits in three synthopic species of newts (Triturus, Amphibia). Holarctic Ecol., (in press). Levins R., - Evolution in changing environments. Princeton Univ. Press, Princeton, 0 pp. Lynch J.., - The feeding ecology of Aneides flavipunctatus and sympatric Plethodontid salamanders in Northwestern California. J. Herpetol., : -. itchell J. C., Taylor J. A., - Predator - prey size relationships in a North Carolina population of Pletbodon jordani. J. Herpetol., 0: -. Osenberg C. W., ittelbach G. G., - Effects of body size on the predator-prey interaction between pumpkinseed and gastropods. Ecol. onogr., : 0-. Polis G. A., yers C. A., - A survey of intraspecific predation among reptiles and amphibians. J. Herpetol., : -0. Toft C., 0 - eeding ecology of thirteen syntopic species of anurans in a seasonal tropical environment. Oecologia, : -. Toft C., - Resource partitioning in Amphibians and Reptiles. Copeia, : -. Verrell P. A., - eeding in adult smooth newts (Triturus vulgaris), with particular emphasis on prey changes in the aquatic phase. Amphibia-Reptilia, : -.

8 EEDING HABITS O THREE NEWT SPECIES APPENDIX A - Prey obtained from the newts over the entire year. The numbers are relative frequencies, per thousand prey in each column. ND, not determined. cristatus alpestris vulgaris r alpestris alpestris juveniles cristatus alpestris vulgaris Total no. prey items Aquatic invertebrates Turbellaria Triclada Nematoda Hermitidae Nematomorpha Gordiidae Annelida ND Hirudinca Lumbricidae Gastropoda Polmonata Prosobranchia Bivalvia Pisidae Crustacea ND Amphipoda Cladocera Copepoda Ostracoda Insecta imagines ND Collembola ND Poduridae Isotomidae Ephemeroptera ND Ephemeridae Caenidae Baetidae Leptophlebiidae Odonata ND Coenagridae Platycnemididae Lestidae. Aeshnidae ' Corduliidae Libellulidae Plecoptera Leuctridae Heteroptera ND Nepidae Corixidae Notonectidae Neuroptera Osmylidae Sialidae Tricoptera ND Philopotamidae Limnephilidae Sericostomatidae Tricoptera Hydropsychidae Psycomiidac Leptoceridae Glossostomatidae Beraeidae Lepidostomatidae Díptera ND Psycodidae Tipulidae Culicidae Chironomidae Ceratopogonidae Tabanidae Ephydridae Athereidae Anthomidae ^ s 0.

9 0. ASOLA, L. CANOVA (continued) cristatus alpestris vulgaris alpestris alpestris cristatus alpestris vulgaris juveniles Coleóptera Dytiscidae Gyrinidac Haliplidae Hydrophilidae Dryopidae Elmidae Helodidae Coleóptera imagines ND Dytiscidae Gyrinidae Hydrophilidae Hydraeniidae Elmidae Terrestrial invertebrates Aracnida ND Acari Arancidae Pseudoscorpionidae Crustacea Isopoda Diplopoda Iulidae Thysanoptera Ortóptera ND Grullidae Tettigonidae Acrididae Tetrigidae Dictyoptera Blattidae Heteroptera Acanthosomidae Lygaeidac icrophysidae Cicadellidae Delphacidae Psyllidae Afididae Lepidoptera ND Díptera imagines ND Simulidae Cecidomyiidae Bibionidae Phoridae Drosophilidae uscidae Hymenoptera imagines ND ormicidae Hymenoptera imagines Proctotrupoidea Tingidae Coleóptera imagines Carabidae Staphilinidae Coccinellidae Curculionidae Amphibians Amphibia newts ND eggs NS