Galápagos Land Iguana (Conolophus Subcristatus) As A

Home , Fernandina Island, Green iguana, Portulaca

Integrative Zoology 2016; 11: 207–213 doi: 10.1111/1749-4877.12187

1 SHORT COMMUNICATION 1 2 2 3 3 4 4 5 5 6 Galápagos land iguana (Conolophus subcristatus) as a seed 6 7 7 8 disperser 8 9 9 10 10 11 11 1 2 3 4 5 12 Anna TRAVESET, Manuel NOGALES, Pablo VARGAS, Beatriz RUMEU, Jens M. OLESEN, 12 13 Patricia JARAMILLO6 and Ruben HELENO4 13 14 14 15 1Institut Mediterrani d’Estudis Avançats (CSIC-UIB), C/ Miquel Marqués 21, Esporles, Mallorca 07190, Balearic Islands, Spain, 15 16 2Island Ecology and Evolution Research Group (CSIC-IPNA), Instituto de Productos Naturales y Agrobiología, 38206 Tenerife, 16 17 Canary Islands, Spain, 3Real Jardín Botánico (CSIC-RJB), Department of Biodiversity and Conservation, Plaza de Murillo 2, 28014 17 18 Madrid, Spain, 4Centre for Functional Ecology, Department of Life Sciences, University of Coimbra, Calçada Martim de Freitas, 18 19 5 6 19 Coimbra 3000-456, Portugal, Institute of Bioscience, Aarhus University, Aarhus, Denmark and Charles Darwin Foundation, Puerto 20 20 Ayora, Santa Cruz, Galápagos, Quito, Ecuador 21 21 22 22 23 23 24 Abstract 24 25 25 26 The role of the most common land iguana (Conolophus subcristatus) in the Galápagos Islands as an effective 26 27 seed disperser is explored in this study. A total of 5705 seeds of 32 plant species were identified from 160 scats, 27 28 4545 of which (80%) appeared visually undamaged. Germination trials of 849 seeds from 29 species revealed 28 29 that at least 10 species remained viable after passing through the iguana’s gut, although only a small propor- 29 30 tion of those seeds (4%) germinated. In any case, we argue that C. subcristatus exerts an important role on the 30 31 7 Galapagos islands where it occurs because of its abundance and capacity to ingest and disperse seeds at long 31 32 distances. Our results strongly suggest that the Galápagos C. subcristatus plays an important role as a seed dis- 32 33 perser of not only of native species but also some introduced plants in the Galápagos Islands. 33 34 Key words: Conolophus subcristatus, Fernandina Island, Galápagos Islands Saurochory, seed dispersal 34 35 35 36 36 37 37 38 38 39 INTRODUCTION lizards often undertake an important ecological role as 39 40 seed dispersers (Olesen & Valido 2003). Furthermore, 40 41 Seed dispersal and pollination are 2 key services pro- the reported niche expansion or interaction release of is- 41 42 vided by animals to plants. On oceanic islands, where land vertebrates, which tend to occupy underexplored 42 43 strong isolation limits the arrival of medium and large ecological niches and adopt super-generalized diets, 43 44 sized mammals (Gorman 1979), tortoises, iguanas or magnifies the ecological importance of insular native 44 45 fauna (MacArthur et al. 1972; Cox & Ricklefs 1977; 45 46 Traveset et al. 2015). 46 47 The capacity to disperse seeds is largely limited by 47 Correspondence: Anna Traveset, Institut Mediterrani d’Estudis 48 animal body size and, consequently, by their gape width. 48 49 Avançats (CSIC-UIB), C/ Miquel Marqués 21, 07190-Esporles, Therefore, large animals are disproportionately import- 49 50 Mallorca, Balearic Islands, Spain. ant as seed dispersers in most ecosystems (Blake et al. 50 51 Email: [email protected] 51

1 2012; Galetti et al. 2015). For instance, the Galápagos ness in this arid zone can be found in Jaramillo et al. 1 2 giant tortoise is the largest terrestrial animal in the ar- (2014). Iguanas are usually concentrated in areas with 2 3 chipelago and was found to be pivotal for vegetation vegetation (Werner 1983). The predominant plant spe- 3 4 dynamics by dispersing the seeds of many plants over cies in our study area were, in order of abundance: Bur- 4 5 long distances (Heleno et al. 2011; Blake et al. 2012). sera graveolens, Lantana peduncularis Andersson, Bo- 5 6 The second largest terrestrial animals in the Galápagos erhaavia caribaea Darwiniothamnus lancifolius Hook. 6 7 are the endemic land iguanas (Conolophus spp.), rep- f., Cordia leucophlyctis Waltheria ovata Cav., Scutia 7 8 resented by 3 endemic species currently distributed in spicata and Cryptocarpus pyriformis Fernandina has 8 9 7 islands (Fernandina, Isabela, Santa Cruz, Santa Fé, never had human settlement despite its large size, and 9 10 Plazas, Baltra and Seymour) (Jiménez-Uzcátegui et al. partially for this reason is one of the world’s most pris- 10 11 2014). However, except for a few anedoctal records (re- tine tropical islands and harbors very few introduced 11 12 viewed in Heleno et al. 2011), the dispersal potential of species (Jaramillo et al. 2014), none of them vertebrates. 12 13 the Galápagos land iguanas has never been explored. Moreover, the population of land iguana of Fernandina 13 14 The 3 Galápagos land iguanas are vegetarian and high- is the only one in Galápagos that remain mostly undis- 14 15 ly generalized (Jackson 1994): for example, consum- turbed, as all other populations have been under anthro- 15 16 ing fruits of Opuntia spp., Psidium galapageium and pogenic stressors such as direct disturbance and the in- 16 17 Scutia spicata (Carpenter 1969; McMullen 1999). They troduction of feral dogs, pigs, goats, cats and rats (Snell 17 18 feed mostly on low-growing vegetation as iguanas do et al. 1984). 18 19 not climb/creep, although they can stand on their hind In February 2010 and 2011, we collected scats of C. 19 20 legs. It is known that in Fernandina, female iguanas mi- subcristatus in the remote Cabo Douglas (northwest of 20 21 grate long distances (approximately 10 km) and ascend Fernandina). The scats were collected in an area of ap- 21 22 up to 1500 m, from the sea shore to the central crater of proximately 1 km2 of the arid zone, which holds most 22 23 this volcanic island where they lay their eggs (Werner plant diversity and endemic species and is the most 23 24 1983). The population of Conolophus subcristatus (Gray, abundant habitat type in the Galápagos (Guézou et al. 24 25 1831) reaches high densities in Fernandina, with thou- 2010). This area was scanned by 5 observers and all 25 26 sands of reproductive females (Werner 1983). iguana scats, either fresh or old, were collected as long 26 27 To disclose the potential role of Galápagos land igua- as they kept their intact structure. This allowed us to 27 28 nas as seed dispersers, we looked for intact seeds in the study its diet over a relatively long period of time, given 28 29 scats of C. subcristatus of the island of Fernandina and that scats remain intact for many months (pers. observ.) 29 30 interpreted legitimate seed dispersal based on seed ger- mainly due to the large quantity of plant fiber ingested 30 31 minability tests. and to the dry climate. Our direct observations of igua- 31 32 nas in the study area lead us to think that the scats might 32 33 33 MATERIAL AND METHODS belong to a few dozens individuals. All collected scats 34 were taken to the Charles Darwin Foundation’s labo- 34 35 Fernandina is the youngest (0.3 Myr, Ali & Aitchi- ratory in Santa Cruz, where seeds were extracted and 35 36 son 2014) and westernmost Galápagos island, consist- identified by comparison with a reference collection of 36 37 ing of a single active shield volcano (last eruption in seeds from the archipelago. All seeds were counted and 37 2 38 2009), with a 5-km wide crater. The island is 642 km visually inspected under a stereomicroscope and classi- 38 39 and the maximum elevation is approximately 1500-m fied as either damaged or undamaged. 39 40 40 a.s.l. It is very heterogeneous in ecological conditions, As a second step, to evaluate the capacity of seed ger- 41 41 bearing habitats rich in resources for iguanas as well as mination, we carried out an experiment by sowing 849 42 42 much bare soil that is used for their burrow construc- seeds from 29 plant species found in the scats. Seeds, 43 43 tion (Werner 1983). The vegetation patterns are largely previously extracted from the surrounding fecal materi- 44 44 determined by volcanic activity and deposition, and the al, were sown on 1 April 2011, on trays of 104 units pre- 45 45 slopes of the volcano are mostly barren lava (Hendrix & viously filled with a substrate composed of agricultural 46 46 Smith 1986). In the lowlands, most vegetation is found soil, volcanic lapilli and turf (2:1:1 ratio). In a shad- 47 47 in small crevices within the recent lava fields. There is ed greenhouse, soil in trays was kept moist throughout 48 48 low water availability in this zone, and the black lava the experiment. Seedling emergence was recorded for 49 49 can reach over 60 °C on sunny days during the hot sea- 2 years: every other day during the first year and once a 50 50 son (Christian et al. 1983). Information on plant rich- week during the second year. In addition, we performed 51 51

1 a seed viability analysis, by applying the bioindicator 1 2 2,3,5 triphenyl-2H-tetrazolium chloride diluted to 0.1% 2 3 for 24 h in the dark and at room temperature (Porter et 3 4 al. 1947; Marrero et al. 2007) to 4 of the fleshy-fruit- 4 5 ed plant species found in the scats and, when possible, 5 6 comparing estimated viability with that of fruits collect- 6 7 ed directly from wild plants (control treatment). 7 8 8 9 RESULTS 9 10 10 11 Of the 160 scats of C. subcristatus collected, 148 a b 11 12 scats (93%) contained at least one undamaged seed. A 12 13 total of 5705 seeds of 32 plant species were identified, 13 14 4545 of which (80%) appeared visually undamaged, al- 14 15 though the proportion of damaged seeds varied consid- 15 16 erably across species (mean = 17%, SD = 24.5%, min- 16 17 17 imum = 0%, maximum = 80%; Table 1). In the scats, 18 18 seeds were always embedded within fibers and oth- d 19 c 19 20 er plant material. Two thirds (63%) of the plant spe- 20 21 cies whose seeds were found in the scats were native, 21 22 of which a third (31%) were endemic to the Galápagos. 22 23 Two species (6%) were likely introduced (Ipomoea cf. 23 24 nil and Bidens pilosa), and the remaining 31% could not 24 25 be identified and, thus, were scored as of unknown ori- 25 26 26 gin. The vast majority (approximately 82%) of the iden- f 27 tified plant species produced dry fruits while the rest 27 e 28 were fleshy-fruited species (Table 1, Fig. 1). Seeds of 28 29 29 the endemic Scalesia affinis, Jasminocereus thouar- 30 30 sii and Euphorbia punctulata (Table 1) were present in 31 31 32 higher proportions in the scats (high frequency of occur- 32 33 rence -FO). The FO of each species in the scats ranged 33 34 from 0.6% (Bursera graveolens) to 56.3% (S. affinis, 34 35 Table 1). Figure 1 Conolophus subcristatus individuals in Fernandi- 35 36 Overall, 8 species (Fig. 1) accounted for a large frac- na Island (top image) and some of the most frequent plants, 36 37 tion (86.6%) of all intact seeds in the iguana scats; 2 of whose seeds were found in their scats: (a) Castela galapage- 37 ia, (b) Lantana peduncularis, (c) Scalesia affinis, (d) Euphor- 38 these were the fleshy-fruited endemicsJ. thouarsii and L. 38 bia punctulata, (e) Exodeconus miersii and (f) Jasminocereus 39 peduncularis and the remaining species had dry fruits: 39 40 thouarsii. Seeds of each species are shown as insets (also fruits 40 Exodeconus miersii, S. affinis, E. punctulata, Portulaca 41 in a). 41 42 sp., Spermacoce cf. remota, and an unidentified species. 42 43 Germination trials showed that of the 849 sown 43 44 seeds, only a small proportion (4%) germinated. Some trol plants (G-test: G = 0.38, df= 1, P = 0.53; Table 2), 44 45 of the species (J. thourasii, Portulaca sp., Commicar- whereas the seeds of L. peduncularis showed very low 45 46 pus tuberosus and Boerhavia coccinea) took >200 days viability. None of the seeds of B. graveolens and S. spi- 46 47 47 to germinate, whereas others such as S. affinis germinat- cata recovered from the scats were viable; however, it 48 48 ed only a few weeks after sowing (Table 1). The viabil- was not possible to estimate the effect of ingestion on 49 ity tests of fleshy-fruited species showed that the seeds 49 50 seed viability for these species as we could not evaluate 50 of Castela galapageia were as viable as those from con- 51 the viability of control (uningested) seeds. 51

1 1 2 2

19 28 19 16 103 332 320 241 266 3 139 3

4 germination 4 Mean days to 5 5 6 6 / / /

% 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.1 0.0 0.0 0.0 0.0 8.3 2.9 5.6 0.0 0.0 0.0 4.8 5.8 4.8 0.0 5.8 11.1 7 13.5 7

8 germination 8

9 Germination 9 10 10 / / / 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 2 3 1 2 0 0 0 2 3 5 0 3 36 11 14 11 n. seeds

12 germinated 12 13 13 1 1 1 1 2 4 3 4 4 0 0 0 n. 34 12 14 10 24 12 26 24 27 35 36 32 52 32 42 52 52 849 104 104 104

14 sown 14 seeds 15 15

0 0 0 0 0 0 0 0 0 0 4 0 0 0 2 0 2 0 16 0 16 % 50 80 14 17 38 31 71 28 39 42 75 22 27 seeds 17 in the island of Fernandina 2010 and 2011. 17 damaged 18 18 1 1 1 1 3 2 2 4 4 4 15 12 15 20 19 26 24 45 35 67 75 75 76 81 110 113 185 290 724 554 19 772 19 1185 4541 intact 20 20 0.0 0.0 0.0 0.1 0.0 0.1 0.1 0.0 0.1 0.1 0.2 0.4 0.3 0.3 0.6 1.2 0.6 0.6 1.2 1.3 1.3 2.3 1.3 1.4 2.0 2.0 5.2 100 13.1 12.7 13.3 17.3 21 20.8 21 % all seeds Number of seeds 22 22 1 1 2 5 2 4 3 2 4 4 14 24 18 19 35 70 36 35 70 74 75 76 81 113 112 130 747 724 758 296 985 1185 5705 n. all

23 seeds 23

24 Conolophus subcristatus 24

25 25 0.6 0.6 0.6 0.6 1.3 1.3 1.3 0.6 1.9 0.6 5.7 1.3 0.6 3.8 4.4 2.5 1.9 5.1 3.8 5.7 6.3 6.3 1.3 9.5 5.1 5.7 2.5 10.1 14.6 49.4 24.1 25.9 seeds

26 % scats 26 w/ intact 27 27

1 1 1 1 2 2 2 1 3 1 9 2 1 6 7 4 3 8 6 9 2 8 9 4 n. 23 16 10 10 15 78 38 41 w/

28 seed 28 scats intact 29 29 % w/ 0.6 0.6 1.3 2.5 1.3 1.3 1.3 0.6 1.9 0.6 5.7 1.9 0.6 5.7 7.0 2.5 1.9 5.1 3.8 8.9 7.6 6.3 1.3 9.5 8.9 5.7 30 2.5 30 10.1 22.8 56.3 24.1 30.4 scats 31 seeds 31

32 Frequency of occurrence 32 1 1 2 4 2 2 2 1 3 1 3 9 1 9 3 4 8 6 2 9 4 n. 11 16 36 14 12 10 15 14 89 38 48 w/

scats seeds

33 n=160 scats 33 2011 X X X X X X X X X X X X X X X X X X X X X X X

34 X X 34 35 2010 35 X X X X X X X X X X X X X X X X X X X X X X X X X X 36 36 37 37 type Fruit

Dry / Dry Dry Dry Dry Dry / Fleshy Dry Dry Dry Fleshy / Fleshy Dry Dry Dry Fleshy / Dry Dry Dry Dry / Fleshy / Fleshy Dry Dry 38 Dry Dry 38 39 39 40 40 41 41 Family 42 42

Malvaceae / Poaceae Poaceae Fabaceae Poaceae Cyperaceae / Boraginaceae Asteraceae Convolvulaceae Fabaceae Burseraceae / Convolvulaceae Rhamnaceae Oxalidaceae Nyctaginaceae Simaroubaceae Poaceae Convulvulaceae Asteraceae Fabaceae / Verbenaceae Rubiaceae / Cactaceae Euphorbiaceae Portulacaceae Solanaceae Asteraceae 43 43

44 (Nat) 44 (End) (End) (End) (Nat) (End) (End) (Nat) 45 (End) 45 (Nat) (Nat) (Nat) (End) (End) (Nat)

46 (Nat) 46 (Nat) remota (Int) (End)

47 punctulata 47 (Nat) (QEnd) (QInt) cf. nil anderssonii 48 cf. 48 sp. (?) cf. cf. Frequency of occurrence viable and damaged seeds retrieved from 160 scats salviifolia

49 Species (origin) 49 sp.1 (?) 50 cf. 50

Sida unknown seed 5 (?) Brachiaria multiculma Poaceae sp.2 (?) Setaria setosa Cyperus Rhynchosia minima unknown seed 4 (?) Varronia leucophlyctis Varronia Blainvillea dichotoma Tephrosia cinerea Tephrosia Ipomoea incarnata Bursera graveolens unknown seed 3 (?) Ipomoea Scutia spicata Oxalis Boerhavia coccinea Boerhavia Castela galapageia Poaceae sp.1 (?) Convulvulaceae sp.1 (?) Bidens pilosa Stylosanthes sympodiales Spermacoce unknown seed 2 (?) Lantana peduncularis unknown seed 1 (?) Jasminocereus thouarsii Jasminocereus Euphorbia Portulaca Exodeconus miersii Scalesia affinis

51 1 Table Germination rate for each seed species and mean number of days to germination are included. Plant origin indicated as: (End) Endemic, (QEnd) Questionably (Nat) Native, (Int) Introduced, (QInt) Questionably introduced, (?) Unknown. 51

1 Table 2 Seed viability (identified by the staining of active tis- Most seeds in the iguana scats were from native 1 2 sues with tetrazolium chloride) of four fleshy-fruited species plants, either endemic or not. Nevertheless, we also de- 2 3 found in the scats of Conolophus subcristatus in the island of tected seeds of likely introduced plant species, namely 3 4 Fernandina. Control seeds are directly collected from plants in Bidens pilosa and Ipomoea cf. nil, as well as seeds from 4 5 the study area. a genus comprising introduced species (e.g. Portulaca 5 6 % viability (no. viable seeds/no. 6 Plant species sp.). The seeds of the alien Ipomoea cf. nil, for instance, 7 tested seeds) were able to germinate after being ingested by the igua- 7 8 Control seeds Dispersed seeds na, which suggests that this disperser might facilitate 8 9 Bursera graveolens 0 (0/18) 0 (0/7) the invasion of introduced species in Fernandina. Our 9 10 report represents one more example of native dispers- 10 Castela galapageia 60 (18/48) 56 (5/9) 11 ers integrating alien plants in the seed dispersal interac- 11 12 Lantana peduncularis / 4 (1/24) tions networks (see Traveset & Richardson [2014] and 12 13 Scutia spicata / 0 (0/7) references therein). Other native dispersers in Galápa- 13 14 gos, such as birds, tortoises and lava lizards, have also 14 15 been shown to contribute to the dispersal of alien plants 15 16 across the islands (Heleno et al. 2013), a phenomenon 16 17 that is of major concern in the archipelago (Guézou et 17 18 al. 2010). 18 DISCUSSION 19 Many seeds escaped destruction in the iguana’s gut 19 20 Our study shows that the Galápagos C. subcristatus, and were defecated intact, as has also been reported 20 21 acts as a legitimate seed disperser of at least 10 plant for other land iguana species, such as the green igua- 21 22 species in the lowlands of Fernandina, as their seeds na (Iguana iguana) (e.g. Moura et al. 2015). More- 22 23 passed intact through the iguana’s guts. Moreover, these over, a variable fraction of the defecated seeds were via- 23 24 species, were frequently found in iguanas’ droppings, ble and capable of germinating. Some studies have even 24 25 suggesting that this land iguana is a common disperser reported an increase in the seed germination rate after 25 26 of such plants. A previous study had already suggested passing through iguanas’ guts, attributing this to their 26 27 that C. subcristatus aids in seed dispersal of some spe- long gut-passage times, often more than 5 days (Bení- 27 28 cies and that it is most important for revegetation after tez-Malvido et al. 2003; Morales-Mávil et al. 2007). In 28 29 volcanic activity (Hendrix & Smith 1986). Specifically, any case, overall germination of the seeds found in the 29 30 these authors found abundant seeds of Solanum erian- scats (some of which had been potentially deposited 30 31 thum in iguana droppings, of which approximately 10% several months ago at the time of collection) was rather 31 32 germinated. They claimed that iguanas might be impli- low. 32 33 33 cated in the distribution of this plant species into areas However, considering the local abundance of land 34 34 devoid of vegetation around the main volcanic crater. iguanas, and the large amount of seeds ingested by these 35 35 Our study confirms the role of the Galápagos land igua- animals, even if only a small proportion germinates, 36 36 na as a seed disperser and expands its likely contribu- they can be considered important for plant dissemina- 37 37 tion to the dispersal of many plants in the dry lowland tion to new areas on this young island. Keystone plant 38 38 areas. species in the community, such as L. peduncularis, J. 39 39 This land iguana includes both fleshy and dry fruits thouarsi or S. affinis, may to a large extent benefit from 40 40 in its diet, as found in other studies on C. subcristatus in the dispersal by this land iguana. Seed dispersal studies 41 41 the mainland (e.g. Traveset 1990; van Marken Lichten- in Galápagos have received considerable attention, with 42 42 belt 1993; Blázquez & Rodríguez-Estrella 2007; Mou- approximately 50 studies having reported the disper- 43 43 ra et al. 2015) or in continental islands (Govender et al. sal of at least 178 plant species by 42 disperser species 44 44 2012; Burgos-Rodríguez 2014). Fleshy fruits, togeth- (see Guerrero & Tye 2011; Heleno et al. 2011 and refer- 45 45 er with leaves, are probably a good source of water for ences therein; Blake et al. 2012; Heleno et al. 2013; and 46 46 these animals in arid environments, and in particular in Heleno et al. unpublished data, including the screening 47 47 the dry season. Some studies in the mainland have also of over 4000 animal droppings collected by our team 48 48 reported land iguanas as effective dispersers of seeds in over 5 years across 12 Galápagos islands). Although 49 49 sites that are suitable for germination and establishment sampling in Galápagos across sites and seasons is still 50 50 (e.g. Benítez-Malvido et al. 2003). scarce, it is remarkable that entire seeds of at least 4 51 51

1 species found in this study, J. thouarsii, S. affinis, Sty- Benítez-Malvido J, Tapia E, Suazo I, Villaseñor E, Al- 1 2 losanthes sympodiales and Tephrosia cinerea, all native, varado J (2003). Germination and seed damage in 2 3 have not been found yet to be dispersed by any other an- tropical dry forest plants ingested by iguanas. Jour- 3 4 imal (Table 1). These data, together with the still rela- nal of Herpetology 37, 301–3. 4 5 tively small effort dedicated to evaluate seed dispersal Blake S, Wikelski M, Cabrera F et al (2012). Seed dis- 5 6 by land iguanas, clearly show the potential role of this persal by Galápagos tortoises. Journal of Biogeogra- 6 7 species as a seed disperser in the archipelago and high- phy 39, 1961–72. 7 8 8 light the need for further research on the topic. Blázquez MC, Rodríguez-Estrella R (2007). Microhab- 9 9 The species-poor plant community of Fernandina, itat selection in diet and trophic ecology of a spiny- 10 10 mostly formed by scattered Bursera trees, Castela and tailed iguana Ctenosaura hemilopha. Biotropica 39, 11 11 Lantana shrubs and Jaminocereus cacti, is served by a 496–501. 12 scarce community of seed dispersers, namely mocking- 12 13 Burgos-Rodríguez JA (2014). Effects of introduced green 13 birds (Mimus parvulus) and some finch species (Thraup- iguanas (Iguana iguana) on tropical plant communi- 14 idae), land iguanas (C. subcristatus) and lava lizards 14 15 ties through seed dispersal and germination (Masters 15 (Microlophus albemarlensis) (pers. observ.). In this pris- Thesis). University of Rhode Island, Kingston. 16 tine and simplified community, the land iguanas might 16 Carpenter C (1969). Behavioral and ecological notes on 17 well exert an important role because of their abundance 17 the Galápagos land iguana. Herpetologica 25, 155– 18 and capacity to ingest and disperse seeds at long dis- 18 64. 19 tances, as inferred by their long gut-retention times; that 19 20 is, they might be acting as a keystone species. Nonethe- Cox PA, Ricklefs RE (1977). Species diversity and eco- 20 21 less, iguanas might also have an important antagonis- logical release in Caribbean land bird faunas. Oikos 21 22 tic effect, at least on some of the species, by consum- 28, 113–22. 22 23 ing their leaves and flowers. Therefore, in those cases, Christian K, Tracy CR, Porter WP (1983). Seasonal 23 24 the positive effect of the iguana on the plants could be shifts in body temperature and use of microhabitats 24 25 counteracted by its negative effect. Whether such ef- by Galápagos land iguanas (Conolophus pallidus). 25 26 fects are overall positive or negative for each plant spe- Ecology 64, 463–8. 26 27 cies would be worth examining in future studies. Our re- Galetti M, Bovendorp RS, Guevara R (2015). Defau- 27 28 sults strongly suggest that the Galápagos C. subcristatus nation of large mammals leads to an increase in seed 28 29 plays an important, and mostly underappreciated, role as predation in the Atlantic forests. Global Ecology and 29 30 a seed disperser in Fernandina, and that the role of this Conservation 3, 824–30. 30 31 31 and other species of land iguanas in the archipelago de- Gorman M (1979). Island Ecology. Chapman and Hall, 32 32 serves further attention. London. 33 33 34 Govender Y, Muñoz MC, Ramírez-Camejo AR, Pu- 34 35 ACKNOWLEDGEMENTS ente-Rolón AR, Cuevas E, Sternberg L (2012). An 35 36 This study is framed within two projects partially fi- isotopic study of diet and muscles of the green igua- 36 37 nanced by Fundación BBVA (Spain) and Ministerio na (Iguana iguana) in Puerto Rico. Journal of Herpe- 37 38 de Economía y Competitividad (CGL2013-44386-P). tology 46, 167–70. 38 39 We are also grateful to the Charles Darwin Station and Guerrero AM, Tye A (2011). Native and introduced 39 40 Galápagos National Park (research permits nos. PC-026- birds of Galápagos as dispersers of native and intro- 40 41 09 and PC-04-11) for logistic support in this archipel- duced plants. Ornitologia Neotropical 22, 207–17. 41 42 ago. R.H. was funded by the FCT grant IF/00441/2013 Guézou A, Trueman M, Buddenhagen CE et al (2010). 42 43 (Portugal) and the Marie Curie Action CIG-321794 (Eu- An Extensive Alien Plant Inventory from the Inhabit- 43 44 ropean Union). ed Areas of Galapagos. PLoS ONE 5, e10276. 44 45 Heleno R, Blake S, Jaramillo P, Traveset A, Vargas P, 45 46 REFERENCES Nogales M (2011). Frugivory and seed dispersal in 46 47 the Galápagos: what is the state of the art? Integra- 47 Ali JR, Aitchison JC (2014). Exploring the combined 48 tive Zoology 6, 110–29. 48 49 role of eustasy and oceanic island thermal subsidence 49 50 in shaping biodiversity on the Galápagos. Journal of Heleno R, Olesen JM, Nogales M, Vargas P, Traveset A 50 51 Biogeography 41, 1227–41. (2013). Seed dispersal networks in the Galápagos and 51

1 the consequences of alien plant invasions. Proceed- McMullen CK (1999). Flowering Plants of the Galapa- 1 2 ings of the Royal Society of London B 280, 20122112. gos. Comstock Publishing Associates, Cornell. 2 3 Hendrix LB, Smith SD (1986). Post-eruption revegeta- Morales-Mávil JE, Vogt RC, Gadsden-Esparza H (2007). 3 4 tion of Isla Fernandina, Galapagos (Ecuador). II. Na- Desplazamientos de la iguana verde, Iguana iguana 4 5 tional Geographic Research 2, 6–16. (Squamata, Iguanidae) durante la estación seca en La 5 6 6 Jackson MH (1994). Galapagos: A Natural History, Re- Palma, Veracruz, México. Revista de Biología Tropi- 7 7 vised and Expanded. University of Calgary Press, cal 55, 709–15. 8 8 Calgary. Moura AC, Cavalcanti L, Leite-filho E, Mesquita DO, 9 9 McConkey KR (2015). Can green iguanas compen- 10 Jaramillo P, Guézou A, Mauchamp A, Tye A (2014). 10 sate for vanishing seed dispersers in the Atlantic for- 11 CDF checklist of Galapagos flowering plants. In: 11 est fragments of north-east Brazil? Journal of Zoolo- 12 Bungartz F, Herrera H, Jaramillo P et al., eds. Charles 12 gy 295, 189–96. 13 Darwin Foundation Galapagos Species Checklist; 13 14 http://checklists.datazone.darwinfoundation.org/. Olesen J, Valido A (2003). Lizards as pollinators and 14 15 Charles Darwin Foundation, Puerto Ayora, Galapa- seed dispersers: An island phenomenon. Trends in 15 16 gos. Ecology and Evolution 18, 177–81. 16 17 Jiménez-Uzcátegui G, Ruiz D, Snell HL (2014). CDF Porter R, Durrell M, Romm H (1947). The use of 2, 3, 17 18 Checklist of Galapagos Terrestrial & Marine Verte- 5-triphenyl-tetrazoliumchloride as a measure of seed 18 19 brates - FCD Lista de especies de Vertebrados terres- germinability. Plant Physiology 22, 149. 19 20 tres y marinos de Galápagos. In: Bungartz F, Herrera Snell HL, Snell HM, Tracy CR (1984). Variation among 20 21 H, Jaramillo P et al., eds. Charles Darwin Founda- populations of Galapagos land iguanas (Conolophus): 21 22 tion. Galapagos Species Checklist – Lista de Espe- Contrasts of phylogeny and ecology. Biological Jour- 22 23 cies de Galápagos de la Fundación Charles Darwin. nal of the Linnean Society 21, 185–207. 23 24 Charles Darwin Foundation/Fundación Charles Dar- Traveset A (1990). Ctenosaura humilis Gray (Iguanidae) 24 25 win, Puerto Ayora, Galapagos: http://www.darwin- as a seed disperser in a Central American deciduous 25 26 foundation.org/datazone/checklists/vertebrates/. Last forest. American Midland Naturalist 123, 402–4. 26 updated 18 Mar 2011. 27 Traveset A, Richardson DM (2014). Mutualistic inter- 27 28 MacArthur RH, Diamond JM, Karr JR (1972). Density actions and biological invasions. Annual Review of 28 29 compensation in island faunas. Ecology 53, 330–42. Ecology, Evolution and Systematics 45, 89–113. 29 30 30 Marrero P, Padilla DP, Valdés F, Nogales M (2007). Traveset A, Olesen JM, Nogales M et al. (2015). 31 31 Comparison of three chemical tests to assess seed vi- Bird-flower visitation networks in the Galápagos un- 32 32 ability: The seed dispersal system of the Macarone- veil a widespread interaction release. Nature Commu- 33 33 sian endemic plant Rubia fruticosa (Rubiaceae) as an nication 6, 6376. 34 example. Chemoecology 17, 47–50. 34 Werner DI (1983). Reproduction in the iguana Conol- 35 van Marken Lichtenbelt WD (1993). Optimal foraging 35 36 ophus subcristatus on Fernandina Island, Galapagos: 36 of a herbivorous lizard, green iguana, in a seasonal Clutch size and migration costs. American Naturalist 37 environment. Oecologia 95, 246–56. 37 38 121, 757–75. 38 39 39 40 40 41 Cite this article as: 41 42 42 43 Traveset A, Nogales M, Vargas P et al. (2016). Galápagos land iguana (Conolophus subcristatus) as a seed dis- 43 44 perser. Integrative Zoology 11, 207–13. 44 45 45 46 46 47 47 48 48 49 49 50 50 51 51