114 Frugivory by birds in Myrsine coriacea (Myrsinaceae) inhabiting fragments of mixedAraucaria Forest in the Aparados da Serra National Park, RS, BrazilAparecida Brusamarello Basler; Eliara Solange Müller and; Maria Virginia PetryPizo (2004) recor<strong>de</strong>d the consumption of fruit from Myrsineumbellata by several bird species.Avian ecology studies have recor<strong>de</strong>d four cotingaspecies (Cotingidae) consuming the fruit of Myrsine coriaceaand Myrsine lancifolia (Pizo et al. 2002); Bailoniusbailloni (Ramphastidae) consuming Rapanea ferrugineafruit (Galetti et al. 2000); and Pipile jacutinga (Cracidae)consuming Rapanea umbellata and R. ferrugineafruit (Galetti et al. 1997, Galetti et al. 2000). Fransciscoand Galetti (2001) studied frugivory and seed dispersalin Myrsine lancifolia and found 11 bird species as potentialdispersers. Pineschi (1990) studied seed-dispersingbirds for seven species of Rapanea and recor<strong>de</strong>d 104 speciesconsuming fruits, of which 60 species were potentialdispersers.Myrsine sp. fruit is consumed by birds of all sizesand according to Carvalho (1994) and Backes and Irgang(2002), Myrsine seeds exhibit dormancy caused by the endocarp,but can easily germinate in any kind of soil afterpassing through the digestive trait of an animal. Therefore,fauna feeding on the fruit is also important to thelife cycle of Myrsinaceae species.The aim of the present study was to i<strong>de</strong>ntify birdsassociated to Myrsine coriacea and investigate their behaviourin or<strong>de</strong>r to answer the following questions: (i) Dobirds use M. coriacea preferentially for fruit consumption?(ii) Does the size of a fragment of forest influencethe number of bird species using M. coriacea and the frequencyof use categories? (iii) Does bird behaviour on thetrees influence the number of bird species using M. coriacea?(iv) Does bird behaviour on the trees influence thenumber of visit events? The potentiality of birds as seeddispersers was also evaluated.MethodsData collection was performed in the Aparados daSerra National Park (ASNP). Three fragments of MixedAraucaria Forest approximately 20 ha in size, classifiedas small fragments (P1, P2 and P3), and three fragmentsapproximately 200 ha in size, classified as largefragments (G1, G2 and G3), were selected. A total ofsix individuals of M. coriacea were selected, correspondingto one individual per fragment. The individualswere located at bor<strong>de</strong>rs or clearings.Three individualswere approximately 13 m high and the other threewere approximately 4 m high, distributed randomly byfragments.The i<strong>de</strong>ntification of the species from the genusMyrsine was performed by collecting samples and performingexsiccates for comparisons to material from theAnchietan Herbarium of the Vale do Rio dos Sinos University.Authors were also consulted as Barroso 1999 andBackes and Irgang 2002, for example.In the month of July 2004 were the first search fieldto individuals of M. coriacea, but they did not have fruityet. Some individuals exhibited initial indications of fructificationin August, but only in November 2004 when allsix individuals were with fruit, began the remarks, whichexten<strong>de</strong>d to the month of February 2005.Varying quantities of fruit were found in both immatureand ripe stages within the same tree. Two randomobservations per month were ma<strong>de</strong> for each individual,totalling 96 hours of observation. The sampling effortwas the same per day periods (6 am to 7 pm). Each observationspen<strong>de</strong>d two hours a day, comprising four hoursper month for each individual of M. coriacea.Observations were performed with the use of10 x 40 mm binoculars. For bird i<strong>de</strong>ntification in locus,field gui<strong>de</strong>s by Narosky and Yzurieta (1987) and De LaPeña and Rumboll (1998) were employed. The followingdata were sampled: species of each bird observed on theplant species; bird behaviour (fruit consumption, insectconsumption and perching); number of visiting birds;time and visiting duration (measured by chronometer);visiting pattern (the manner in which the bird reachedthe tree: individually, by pairs, in mono-specific or mixedflocks), agonistic encounters (intraspecific and interspecific);and whether visits were complete (when the birdwas seen flying in coming, if not of food or fruit and / orinsects and when he left) or incomplete (when just partof the visit could be followed) according to Krügel et al.(2006), when we were unable to view all the activity ofthe bird (due to factors such as the limitation of binocularsfocus or sunlight, for example).Visiting birds were classified into different feedingguilds, as <strong>de</strong>scribed in Sick (1997) and Azpiroz (2001), asfrugivorous (FR, diet predominantly of fruit, vegetablesand occasionally invertebrates), granivorous (GR, dietbased on grains), nectarivorous (NC, diet based on nectarand occasionally small invertebrates), insectivorous(IN, diet exclusively based on invertebrates) and onivorous(ON, diet including fruit, invertebrates and smallvertebrates).The observation of frugivorous species inclu<strong>de</strong>dthe recording of fruit consumed, collection behaviourand fruit ingestion treatment. Collection behaviour wasclassified according to <strong>de</strong>scriptions by Moermond andDenslow (1985): Stalling (S): the bird flies toward thefruit and plucks it without stopping; Hovering (Hv): thebird stops in the air in front of the fruit; Picking (P): thebird collects the fruit next to its perch without stretchingor assume a special position; Reaching (R): the birdstretches out its body to pluck the fruit; Hanging (Hg):entire body and legs are un<strong>de</strong>r the perch, with the ventralsi<strong>de</strong> facing upward. Fruit ingestion treatments wereclassified into Swallowing (when the fruit is swallowedwhole); Mashing (when the bird macerates or grinds thefruit with its jaws before ingesting); Pecking (when theRevista <strong>Brasileira</strong> <strong>de</strong> <strong>Ornitologia</strong>, 17(2), 2009
Frugivory by birds in Myrsine coriacea (Myrsinaceae) inhabiting fragments of mixedAraucaria Forest in the Aparados da Serra National Park, RS, BrazilAparecida Brusamarello Basler; Eliara Solange Müller and; Maria Virginia Petry115bird pecks the fruit with its beak, pulling out portionsof it).Factorial ANOVA was used to evaluate bird behaviour,taking into account the size of the forest area, speciesrichness and number of events. Kruskal-Wallis wasapplied to compare the richness of birds visiting M. coriaceain different forest fragments. Repeated MeasuresANOVA was applied to assess the variation in richnessbetween months according to the size of the area.RESultsThirty-one bird species were recor<strong>de</strong>d visiting theM. coriacea individuals, among which 23 species consumedfruits; 18 also used the tree for perching and fivealso foraged, capturing invertebrates. Consi<strong>de</strong>ring speciesrichness (F = 7.536; df = 2.12; P = 0.008) and the numberof events for each behaviour (F = 6.809; df = 2.12;P = 0.011), the birds preferentially used the tree for fruitconsumption. The interaction between bird behaviourand area size did not affect species richness or the numberof events recor<strong>de</strong>d for each behaviour (F = 1.488;df = 2.12; P = 0.265 and F = 1.291; df = 2.12; P = 0.310,respectively) (Figure 1 and 2).Among the birds visiting M. coriacea, nine specieswere only observed in large areas, eight were only seenin small areas and another 14 were recor<strong>de</strong>d in both areasizes. Area size (F = 0.062; df = 3.12; P = 0.979), variationbetween months from November to February (F = 3.5;df = 1; P = 0.658) and the interaction of these factors(F = 0.639; df = 3.12; P = 0.604) did not affect speciesrichness of birds visiting M. coriacea (Figure 3 and 4).74.2% of the birds recor<strong>de</strong>d (23 species) were classifiedas insectivorous and, <strong>de</strong>spite feeding basically oninvertebrates, 82.6% of these (19 species) also consumedfruits, whereas 17.4% (four species), namely Colaptescampestris, Xiphorhyncus fuscus, Leptasthenura setaria andXenops rutilans, consumed only insects. The frugivorousguild represented 12.9% of the total. 75% of this guild(four species) consumed fruit; Tangara preciosa is a specieswithin this guild, but did not consume fruit. The omnivorousguild represented 6.5% of the total (two species):Cyclarhis gujanensis, which was only observed perching;and Cyanocorax caeruleus, which consumed fruit. Thegranivorous and nectarivorous guilds each represented3.2% of the total, each with one species, respectively,Zonotrichia capensis and Leucochloris albicollis (Table 1).A total of 348 visits were recor<strong>de</strong>d, 31, 03% (108visits) of which were complete, 60.34% (210) were incompleteand 8.63% (30) were of pairs or flocks reachingthe tree, which were not accompanied. 284 visits werema<strong>de</strong> by birds arriving to the tree individually, 26 by birdsin pairs, four by mono-specific flocks and four by mixedflocks. For complete visits, fruit consumption totalled64 visits (60%); perching totalled 36 visits (34%); andFIGuRE 2: Frequency of fruit consumption, insect consumption andperching in different sizes of Araucaria Forest fragments.FIGuRE 1: Bird species richness regarding fruit consumption, insectconsumption and perching in different sizes of Araucaria Forestfragments.FIGuRE 3: Species richness of birds visiting Myrsine ferruginea indifferent sizes of Araucaria Forest fragments.Revista <strong>Brasileira</strong> <strong>de</strong> <strong>Ornitologia</strong>, 17(2), 2009