Annals of Botany 101: , 2008 doi: /aob/mcm323, available online at

Annals of Botany 101: 501 507, 2008 doi:10.1093/aob/mcm323, available online at www.aob.oxfordjournals.org Comparison between the Anatomical and Morphological Structure of Leaf Blades and Foliar Domatia in the Ant-plant Hirtella physophora (Chrysobalanaceae) CÉLINE LEROY 1,2, ALAIN JAUNEAU 3,ANGÉLIQUE QUILICHINI 2, ALAIN DEJEAN 2,4 and JÉRÔME ORIVEL 2, * 1 Centre Universitaire de Formation et de Recherche Jean-François Champollion, Place de Verdun, 81012 Albi, France, 2 Laboratoire Evolution et Diversité Biologique, UMR-CNRS 5174, Université Toulouse III, 118 route de Narbonne, 31062 Toulouse cedex 9, France, 3 Laboratoire Signaux et Messages Cellulaires chez les Végétaux, IFR 40 Pôle de Biotechnologie Végétale, 24 Chemin de Borde Rouge, B.P. 17 Auzeville, 31326 Castanet-Tolosan, France and 4 CNRS-Guyane, UPS 2561, Résidence Le Relais, 16 avenue André Aron, 97300 Cayenne, France Received: 9 October 2007 Returned for revision: 6 November 2007 Accepted: 26 November 2007 Published electronically: 25 January 2008 Background and Aims Myrmecophytes, or ant-plants, are characterized by their ability to shelter colonies of some ant species in hollow structures, or ant-domatia, that are often formed by hypertrophy of the internal tissue at specific locations (i.e. trunk, branches, thorns and leaf pouches). In Hirtella physophora (Chrysobalanaceae), the focal species of this study, the ant-domatia consist of leaf pouches formed when the leaf rolls over onto itself to create two spheres at the base of the blade. Methods The morphological and anatomical changes through which foliar ant-domatia developed from the laminas are studied for the first time by using fresh and fixed mature leaves from the same H. physophora individuals. Key results Ant-domatia were characterized by larger extra-floral nectaries, longer stomatal apertures and lower stomatal density. The anatomical structure of the domatia differed in the parenchymatous tissue where palisade and spongy parenchyma were indistinct; chloroplast density was lower and lignified sclerenchymal fibres were more numerous compared with the lamina. In addition, the domatia were thicker than the lamina, largely because the parenchymatous and epidermal cells were enlarged. Conclusions Herein, the morphological and anatomical changes that permit foliar ant-domatia to be defined as a specialized leaf structure are highlighted. Similarities as well as structural modifications in the foliar ant-domatia compared with the lamina are discussed from botanical, functional and mutualistic points of view. These results are also important to understanding the reciprocal evolutionary changes in traits and, thus, the coevolutionary processes occurring in insect plant mutualisms. Key words: Anatomy, ant plant mutualism, Chrysobalanaceae, extra-floral nectaries, French Guiana, Hirtella physophora, secondary domatia. INTRODUCTION Interactions between ants and myrmecophytic plants are widespread and provide one of the best known example of mutualism (Bronstein et al., 2006). Resident ants protect their host plant from herbivorous insects, encroaching vines and competing plants (Janzen, 1966; Fonseca, 1994; Gaume and McKey, 1998; Heil and McKey, 2003). In exchange, the myrmecophytes provide ants with a nesting place in natural, hollow structures (known as ant-domatia), and often also food through food bodies and/or extra-floral nectar (Beattie, 1985; Benson, 1985; Beattie and Hughes, 2002). Ant-domatia have been defined by Beattie and Hughes (2002) as plant structures that appear to be specific adaptations for ant occupation, often formed by the hypertrophy of internal tissue at particular locations in the plant, creating internal cavities attractive to ants. These hollow structures have been recorded in a variety of phylogenetically distant families of plants. They may be located in the trunk, the petiole, the stipule or the leaf blade depending on the plant species and they differ greatly in their morphology and * For correspondence. E-mail orivel@cict.fr anatomy (Jolivet, 1996). Based on the extent to which pre-existing structures become modified, Benson (1985) distinguished primary from secondary domatia. Primary domatia derive from natural cavities, such as hollow stems, petioles or thorns; secondary domatia are described as qualitatively distinct organs of ant-plants that derive from a modified structure (Benson, 1985). Few studies have discussed the morphological and anatomical characteristics of these structures and they have mainly been concerned with the anatomy of primary domatia (Bailey, 1922, 1923; Brouat et al., 2001; Federle et al., 2001). Ideally, to be able to assess Benson s principle, the structure of domatia should be compared with the structure of the same organ on a conspecific individual that did not develop domatia. Such individuals are, however, extremely rare (Blüthgen and Wesenberg, 2001; Gaume et al., 2005), so that at best one can make comparisons at the genus level between myrmecophytic and non-myrmecophytic species (Tepe et al., 2007a). This approach allowed Tepe et al. (2007a) to describe new qualitative modifications in existing structures in cauline ant-domatia. With regard to leaf pouches, their shapes have played an important role in the taxonomy of myrmecophylous Melastomataceae and comparative # The Author 2008. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

502 Leroy et al. Morphology and Anatomy of Leaf Domatia in Hirtella studies at the genus level of this family have revealed morphological and anatomical variations (Michelangeli, 2000; Michelangeli and Stevenson, 2004). However, leaf pouches, which result from the curling under of the leaf margin, have not yet been studied with regards to Benson s principle, although they can be directly compared with the lamina of the same leaf. In this way, Sampson and McLean (1965) and Nishida et al. (2006) have revealed clear qualitative anatomical differences for mite-domatia. In the present study, Benson s principle (Benson, 1985) was tested to highlight the nature and degree to which leaf pouches (as secondary ant-domatia) in the ant-plant Hirtella physophora (Chrysobalanaceae) derived from structural changes in the leaf lamina. The morphology and anatomy of the domatia and lamina on mature leaves were investigated with three main objectives in mind: (1) to describe qualitatively the morphological and anatomical structures of the domatia and lamina in such a way as to make their differences and similarities clear; (2) to compare quantitatively the different characters described; and (3) to determine whether, from both morphological and anatomical points of views, ant-domatia not only fulfil a novel function, but might actually be considered as a specialized leaf structure. MATERIALS AND METHODS Model system and study site The genus Hirtella (Chrysobalanaceae) comprises 98 species, of which only seven are mymecophytic (Prance, 1972). Hirtella physophora Martius & Zuccharini is a small tree of the Amazonian rainforest understorey that can reach 6 m in height. This myrmecophyte is mainly associated with the ant Allomerus decemarticulatus Wheeler (Myrmicinae) (Dejean et al., 2005). Ant-domatia are present on all of the leaves of an adult tree and they are used as permanent nest-sites by the resident ants (one colony per tree). The domatia consist of two spheres on either side of the petiole at the base of the leaf blade on the abaxial surface of the leaf (Fig. 1A). The natural entrances to the two adjacent leaf pouches are on the abaxial surface of the leaf, on either side of the main vein, near the lamina. Hirtella physophora bears extra-floral nectaries (EFNs) on its leaves (Prance and White, 1988). Plant samples were collected in a pristine forest in the area around Petit Saut, Sinnamary, French Guiana (5803 0 39 00 N, 53802 0 36 00 W) in February, 2007. One mature leaf from each of seven adult trees was fixed in FAA (5 % formalin, 5 % acetic acid, 50 % ethanol) for 1 2 weeks, and then transferred to 70 % ethanol for longterm storage. Additional observations were also made on freshly collected samples. Morphological measurements For each leaf, the surface areas (mm 2 ) of all of the EFNs were recorded in each domatia and for the same number in the corresponding lamina that were calculated by measuring the diameters (length and width) under a stereomicroscope equipped with an ocular micrometer. Stomatal density and guard cell lengths were determined for the abaxial surface of each lamina and the inner part of each domatia from imprints made using transparent nail varnish. The number of stomata per mm 2 was then recorded under an optical microscope for each preparation in five randomly selected areas. Stomatal aperture lengths, defined as the distance between the junctions or ends of the guard cells, were measured using an ocular micrometer. Within each area selected, the lengths of four stomatal apertures were measured and 20 stomata were measured for each lamina and ant-domatia sample. Anatomical structure and measurements The anatomical structure of all of the leaves (lamina and ant-domatia) was investigated to describe the structure and to quantify tissue thickness and cell area. For the lamina, a median portion excluding the main vein was selected on each leaf studied to avoid any anatomical variations; for the domatia the part opposite from the petiole was sampled (see Fig. 1B). Both lamina and domatia samples were dehydrated in 80, 90 and 100 % ethanol. The samples were then progressively infiltrated with medium-grade LR White resin (London Resin White, Agar Scientific, Redding,CA, USA). After polymerization of the resin, 1-mm-thick crosssections were obtained with an ultra microtome (UltraCutE, Reichert-Leica, Germany); they were then stained with 0.05 % toluidine blue O. The different cross-sections were observed by using an inverted microscope (Leica DMIRBE, Rueil-Malmaison, France). Images were acquired with a CCD camera (Color Coolview, Photonic Science, Robertsbridge, UK) and processed using image analysis software (Image-ProPlus, Media Cybernetics, Silver Spring, MD, USA). For each section, a high-resolution picture was made of its most visually representative part. The following were then measured: (1) the thickness (mm) of the lamina and domatia (20 replicates on regions of the section away from protruding veins); (2) the thicknesses of the different subsequent tissues (20 replicates per tissue and per image); and (3) the thickness of the outer surface of the upper and lower epidermal cell wall including the cuticle (20 replicates for both the abaxial and the adaxial surfaces per image). The areas (mm 2 ) of the epidermal (upper and lower epidermis) and the parenchymatous cells were determined from two to four images of different sections of the domatia and the lamina, respectively. Lignin visualization (Wiesner reagent) was made possible by staining 50-mm-thick sections of freshly collected material with phloroglucinol in such a way as to characterize the composition of the sclerenchymal fibres around the vascular bundles both in the lamina and in the domatia. From these freshly collected samples, other semi-thin sections were observed via confocal microscopy (LSM-SP2 Leica, Germany) with a 40 water immersion lens (NA 0.75). Auto-fluorescence was collected (1) in the range between 410 460 nm and 500 560 nm to visualize cell walls (using a 405-nm diode laser and the 488-nm excitation line of an Ar laser, respectively), and (2) in the

Leroy et al. Morphology and Anatomy of Leaf Domatia in Hirtella 503 FIG. 1. (A) Young Hirtella physophora branch bearing mature leaves with ant-domatia (leaf pouches located at the base of the lamina where the associated ants nest). (B, C) Whole and open ant-domatia with one chamber serving as a shelter for the ants. The arrows indicate the natural entrance of the domatia on the adaxial surface. The circles indicate the positions of the extrafloral nectaries that form a spiral. The rectangle indicates where the anatomical study was conducted. (D) Location of the extrafloral nectaries on the abaxial surface of the lamina (black circles). (E, F) Close-up of the extrafloral nectaries of ant-domatia and lamina, respectively. Scale bars: (A D) ¼ 0.5 cm; (E, F) ¼ 0.3 mm. range 645 720 nm for chloroplasts (633-nm excitation line of an HeNe laser). Images were computed using a projection of 20 30 plan-confocal images acquired in the z dimension. Measurement values were compared for individuals using the paired t-test (SYSTAT v.11.0 software). RESULTS The domatia result from the curling under of the leaf margin on either side of the petiolar insertion. The EFNs are distributed in a spiral inside the domatia in continuity with the leaf blade (Fig. 1C, D). No anatomical differences were noted in the EFNs located inside the domatia or on the leaf blade (data not shown). The average number of EFNs inside the domatia was quite constant at 8 + 1.04 (mean + s.e.), while it was more variable on the leaf blade (13 + 3.65). Moreover, the EFNs located inside the domatia were three times as large as those on the abaxial surface of the laminas (Fig. 1C F and Table 1). As expected, the lamina exhibited a typical leaf structure with well-developed palisade and spongy parenchyma on the adaxial and abaxial surfaces, respectively (Fig. 2A). These parenchyma were delimited by a layer of epidermal cells covered by a thin cuticle. Stomata were present in the abaxial epidermis. The vascular bundles were composed of phloem towards the abaxial surface and xylem towards the adaxial surface. Sclerenchyma fibres were always associated with these two tissues, but were more numerous on the abaxial surface. Phloroglucinol-stained sections revealed that these sclerenchyma cells were lignified (in red, Fig. 2C, D).

504 Leroy et al. Morphology and Anatomy of Leaf Domatia in Hirtella TABLE 1. Comparisons of morphological and anatomical characteristics of the lamina and ant-domatia n Lamina Ant-domatia t-test P Morphology EFN area (mm 2 ) 62 0.029 + 0.001 0.091 + 0.003 19.387,0.0001 Stomatal size (mm) 140 19.18 + 0.18 26.35 + 0.24 21.593,0.0001 Stomatal density (mm 2 ) 35 115.61 + 2.96 23.30 + 2.26 27.657,0.0001 Thickness Total (mm) 140 82.81 + 0.94 208.33 + 3.78 33.580,0.0001 Upper outer ECW/cuticle (mm) 140 1.65 + 0.03 2.51 + 0.05 14.240,0.0001 Upper epidermis (mm) 140 19.72 + 0.30 20.56 + 0.40 2.036 0.044 Mesophyll (mm) 140 54.25 + 1.01 154.02 + 3.93 29.525,0.0001 Lower epidermis (mm) 140 10.41 + 0.21 19.12 + 0.38 18.228,0.0001 Lower outer ECW/cuticle (mm) 140 1.35 + 0.02 4.98 + 0.07 50.310,0.0001 Cell area Upper epidermis (mm 2 ) 345 303.09 + 8.71 320.29 + 7.92 2.614 0.009 Mesophyll (mm 2 ) 1787 104.41 + 1.03 322.48 + 4.85 53.811,0.0001 Lower epidermis (mm 2 ) 266 166.88 + 9.66 362.87 + 5.05 28.360,0.0001 Values are shown as mean + s.e. and statistical comparisons were performed using paired t-tests. ECW, epidermal cell wall. Conversely, the domatia did not exhibit this typical type of tissue organization. In none of the transverse sections of the samples were either palisade or spongy parenchyma observed (Fig. 2B). Moreover, the walls of the domatia were 2 3 times thicker than the leaf laminas (Fig. 2A, B and Table 1). The greater thickness of the ant-domatia was mainly related to enlargement of all the parenchymatous cells, including the epidermal cells on the abaxial surface (they were 2 3 times larger in domatial structures than in the leaf laminas). Moreover, changes in the abaxial epidermal layer, which delimits the nesting space of the ants, should be underlined. First, the thicknesses of the surface of the lower and upper epidermal outer cell walls, including the cuticle, were clearly greater in ant-domatia when laminas were compared. Second, the density of the stomata was dramatically lower. It was obvious from imprints of the abaxial surface that there were only a few stomata on the lower epidermis in the domatia and that the apertures of these stomata were, in contrast, larger (Table 1). Third, within the parenchyma the enlarged cells (as indicated previously) showed some cell-wall thickening, especially on the abaxial surface (Fig. 2B). Fourth, there were relatively few intercellular spaces between cells. Fifth, although the overall organization of the vascular bundles was not dramatically different (Fig. 2C, D), lignified sclerenchymal fibres surrounded the vascular tissues and were clearly more numerous than the leaf lamina. Finally, chloroplasts were observed in both the lamina and the domatia and their density was higher in the adaxial epidermis than in the abaxial epidermis (Fig. 2E, F). However, the density of the chloroplasts was lower in the parenchymatous tissue of the domatia than in the lamina tissue. DISCUSSION The results demonstrate that the foliar ant-domatia of H. physophora differ both morphologically and anatomically from the leaf lamina. Observations here show that leaf pouches, which are located at the base of each leaf, are created not only when the lower part of the lamina rolls under itself, but are inherent plant structures that also undergo important structural and functional modifications. When compared with the lamina, palisade and spongy parenchyma are entirely absent from H. physophora ant-domatia. The palisade parenchyma is commonly thought to be a specialized type of photosynthetic tissue, bringing the chloroplast into a more favourable position with reference to light (Esau, 1965). The lack of differentiation of the palisade tissue and the low chloroplast density among parenchymatous cells suggest that the photosynthetic capacity of the domatia is lower than for the lamina. In addition, the persistence of stomata in the domatial cavity suggests that there may be an exchange of gases between the plant and its occupants. Treseder et al. (1995) demonstrated that ant respiration is an additional source of CO 2 in ant-occupied leaves of the myrmecophytic epiphyte Dischidia major (Asclepiadaceae). Indeed, in D. major as in H. physophora, the stomata are located on the internal surfaces of leaf cavities and absorb the carbon dioxide produced through ant respiration. The cavity of the H. physophora ant-domatia could thus be characterized as a particular micro-environment with a higher CO 2 content and lower incident light compared with the surface of the lamina. Higher CO 2 levels or a reduction in the amount of incident light are the two major parameters that could explain a reduction in stomatal density (Oberbauer and Stain, 1985; Woodward and Kelly, 1995; Gutschick, 1999; Aranda et al., 2001; Brownlee, 2001; Woodward et al., 2002). In the present study, comparisons between the number and the size of the stomata on ant-domatia and on the lamina reinforced the general premise that the larger the length of the guard cells, the lower the stomatal density (Hetherington and Woodward, 2003). Ant-domatia also have a very compact structure characterized by a thick layer of parenchymatous tissue, a thick outer surface to the lower epidermal cell wall, including the cuticle, a highly compact arrangement of the larger parenchymatous cells and a thick secondary parenchymatous wall.

Leroy et al. Morphology and Anatomy of Leaf Domatia in Hirtella 505 FIG. 2. (A, B) Transverse section (1 mm) of (A) lamina and (B) ant-domatia of Hirtella physophora. (C, D) Transverse sections (50 mm) of (C) fresh lamina and (D) fresh domatia stained with phloroglucinol showing the distribution of lignin in the sclerenchymal fibres around the vascular bundle. (E, F) Transverse sections (50 mm) of (E) fresh lamina and (F) fresh domatia observed with a confocal microscope to highlight the distribution of chloroplasts (in red). For all images the adaxial surface of lamina and domatia is to the top. Scale bars ¼ 100 mm. These abaxial parenchymatous cells could be considered as collenchyma-like cells. Niklas (1999) pointed out by using a biomechanical approach that parenchymatous and collenchymal tissues are both highly viscoelastic tissues, thus enabling the plant to use the parenchyma as load-bearing tissue (Spatz et al., 1997; Niklas, 1999). Furthermore, cellulosic deposits thicken the primary cell walls of the abaxial parenchyma, thus reinforcing the rigidity of the ant-domatia. In addition, ant-domatia are clearly highly vascularized, something that has also been observed in Hirtella myrmecophila (Izzo and Vasconcelos, 2002), and each vascular bundle is heavily sclerenchymatized. Roth-Nebelsick et al. (2001) pointed out that venation density and geometry contribute to the mechanical stability of the leaf based on the lignified xylem and sclerified elements. In contrast to what is generally observed in leaves, the thickness of the outer surface of the epidermal cell walls, including the cuticle, was higher on the abaxial surface than on the adaxial surface. The cuticle serves different functions, such as acting as an impermeable barrier against water vapour loss from tissues (Schreiber and Riederer, 1996), attenuating short-wave radiation (Krauss et al., 1997), and/or deterring herbivorous insects (Peeters, 2002) and certain pathogens (Carver and Gurr, 2006). The cuticle may also play an important role in preventing cell damage when ants traverse the cavity of the domatia. Besides housing, H. physophora also provides food rewards to its guest ants in the form of nectar from EFNs situated on the abaxial part of the lamina and of the domatia, while the congeneric myrmecophyte H. myrmecophila does not bear any EFNs at all (Romero and Izzo, 2004).

506 Leroy et al. Morphology and Anatomy of Leaf Domatia in Hirtella Food-bodies can be produced on various part of the plants (Webber et al., 2006), including inside the primary ant-domatia of various myrmecophytes (Risch and Rickson, 1981; Linsenmair et al., 2001; Fischer et al., 2002) and even the leaf pouches of myrmecophytic Melastomataceae (Roth, 1976; Cabrera and Jaffe, 1994; Alvarez et al., 2001; Nery and Vasconcelos, 2003; Tepe et al., 2007b). Nevertheless, to the best of our knowledge this is the first time that EFNs have been observed inside ant-domatia. The presence of these EFNs inside the domatia will contribute to the success of the claustrally founded A. decemarticulatus colonies because all the individuals are fed until some workers become foragers. In other myrmecophytes, EFNs are most often located on the leaf blades (Linsenmair et al., 2001; Gaume et al., 2005) or the petiole/rachis between pairs of leaflets, sometimes along the midvein (Elias, 1980) and exceptionally on the trunk at the base of the branches, along the branches and on the leaves (Breteler, 1999). The present study has made it possible to show that the coevolutionary processes occurring in insect plant mutualisms brought about changes in the leaf laminas, which, in turn, led to the formation of novel, specialized leaf structures. Furthermore, a thorough examination of the anatomical characters of these leaf pouches has shown that they have reduced their photosynthetic ability, while at the same time they have become more specialized in housing and feeding ants. Given their own structural and functional characteristics, the leaf pouches of H. physophora may be defined as a new leaf-derived plant structure. From a developmental point of view, Michelangeli and Stevenson (2004) found differences in the ability of leaves to expand based on the position of the leaf pouches in the Melastomataceae. Maieta guianensis pouches, which are immersed in the leaf blade, develop during leaf expansion; by contrast, those of Tococa guianensis, situated at the apex of the petiole, develop prior to leaf expansion. In contrast to H. physophora, pouches of which persist throughout the entire life of the leaf, old H. myrmecophila pouches dry out and fall off, while the leaf blades remain active for more than 2 years (Izzo and Vasconcelos, 2002). Thus, leaf pouches and laminas are characterized by different developmental growth rates and overall life spans that could be attributed to their own meristematic activity. The present study focused on mature leaves to highlight the major morphological and anatomical differences between the lamina and the ant-domatia. However, further studies are necessary to understand the mechanisms behind the formation and development of H. physophora ant-domatia during leaf expansion throughout the ontogeny of the plant. Finally, other myrmecophytic species bearing leaf pouches need to be studied in such a way as to define the morphological and anatomical changes common to these specialized leaf structures. ACKNOWLEDGEMENTS We are grateful to M. Gibernau and D. Barabé for useful comments on the manuscript and to Andrea Dejean for checking the text. This study was supported by a research programme of the French Agence Nationale de la Recherche (research agreement no. ANR-06-JCJC- 0109-01) to J.O. LITERATURE CITED Alvarez G, Armbrecht I, Jimenez E, Armbrecht H, Ulloa-Chacon P. 2001. Ant plant association in two Tococa species from a primary rain forest of Colombian Choco (Hymenoptera: Formicidae). Sociobiology 38: 585 602. Aranda I, Bergasa LP, Gil L, Pardos JA. 2001. Effects of relative irradiance on the leaf structure of Fagus sylvatica L. seedlings planted in the understory of a Pinus sylvestris L. stand after thinning. Annals of Forest Science 58: 673 680. Bailey IW. 1922. Notes on neotropical ant-plants. I. Cecropia angulata, sp. nov. Botanical Gazette 74: 369 391. Bailey IW. 1923. Notes on neotropical ant-plants. II. Tachigalia paniculata sp. nov. Botanical Gazette 75: 27 41. Beattie AJ. 1985. The evolutionary ecology of ant plant mutualisms. Cambridge: Cambridge University Press. Beattie AJ, Hughes L. 2002. Ant plant interactions. In: Herrera C, Pellmyr O, eds. Plant animal interactions: an evolutionary approach. Oxford: Blackwell, 211 235. Benson WW. 1985. Amazon ant-plants. In: Prance GT, Lovejoy TE, eds. Amazonia. Oxford: Pergamon Press, 239 266. Blüthgen N, Wesenberg J. 2001. Ants induce domatia in a rain forest tree (Vochysia vismiaefolia). Biotropica 33: 637 642. Breteler F. 1999. Barteria Hook. F. (Passifloraceae) revised. Adansonia 21: 307 318. Bronstein JL, Alarcón R, Geber M. 2006. The evolution of plant insect mutualisms. New Phytologist 172: 412 428. Brouat C, Garcia N, Andary C, McKey D. 2001. Plant lock and ant key: pairwise coevolution of an exclusion filter in an ant plant mutualism. Proceedings of the Royal Society of London 268: 2131 2141. Brownlee C. 2001. The long and the short of stomatal density signals. Trends in Plant Science 6: 441 442. Cabrera M, Jaffe K. 1994. A trophic mutualism between the myrmecophytic Melastomataceae Tococa guianensis Aublet and an Azteca ant species. Ecotropicos 7: 1 10. Carver TLW, Gurr SJ. 2006. Filamentous fungi on plant surfaces. Annual Plant Reviews 23: 368 392. Dejean A, Solano PJ, Ayroles J, Corbara B, Orivel J. 2005. Arboreal ants build traps to capture prey. Nature 434: 973. Elias TS. 1980. Foliar nectaries of unusual structure in Leonardoxa africana (leguminosae), an African obligate myrmecophyte. American Journal of Botany 67: 423 425. Esau K. 1965. Plant anatomy, 2nd edn. New York: John Wiley and Sons. Federle W, Fiala B, Zizka G, Maschwitz U. 2001. Incident daylight as orientation cue for hole-boring ants: prostomata in Macaranga ant-plants. Insectes Sociaux 48: 165 177. Fischer RC, Richter A, Wanek W, Mayer V. 2002. Plants feed ants: food bodies of myrmecophytic Piper and their significance for the interaction with Pheidole bicornis ants. Oecologia 133: 186 192. Fonseca CR. 1994. Herbivory and the long-lived leaves of an Amazonian ant-tree. Journal of Ecology 82: 833 842. Gaume L, McKey D. 1998. Protection against herbivores of the myrmecophyte Leonardoxa africana (Baill.) Aubrèv. T3 by its principal ant inhabitant Aphomomyrmex afer Emery. Comptes Rendus de l Académie des Sciences 321: 593 601. Gaume L, Zacharias M, Grosbois V, Borges RM. 2005. The fitness consequences of bearing domatia and having the right ant partner: experiments with protective and non-protective ants in a semimyrmecophyte. Oecologia 145: 76 86. Gutschick VP. 1999. Biotic and abiotic consequences of differences in leaf structure. New Phytologist 143: 3 18. Heil M, McKey D. 2003. Protective ant plant interactions as model systems in ecological and evolutionary research. Annual Review of Ecology, Evolution and Systematics 34: 425 453. Hetherington AM, Woodward FI. 2003. The role of stomata in sensing and driving environmental change. Nature 424: 901 908.

Leroy et al. Morphology and Anatomy of Leaf Domatia in Hirtella 507 Izzo TJ, Vasconcelos HL. 2002. Cheating the cheater: domatia loss minimizes the effects of ant castration in an Amazonian ant-plant. Oecologia 133: 200 205. Janzen DH. 1966. Coevolution of mutualism between ants and acacias in Central America. Evolution 20: 249 275. Jolivet P. 1996. Ants and plants: an example of coevolution. Leiden: Backhuys Publishers. Krauss P, Markstadter C, Riederer M. 1997. Attenuation of UV radiation by plant cuticles from woody species. Plant, Cell & Environment 20: 1079 1085. Linsenmair KE, Heil M, Kaiser WM, Fiala B, Koch T, Boland W. 2001. Adaptations to biotic and abiotic stress: Macaranga-ant plants optimize investment in biotic defence. Journal of Experimental Biology 52: 2057 2065. Michelangeli FA. 2000. A cladistic analysis of the genus Tococa (Melastomataceae) based on morphological data. Systematic Botany 25: 211 234. Michelangeli FA, Stevenson DW. 2004. Comparative morphology, anatomy and development of ant-domatia in Neotropical Melastomataceae. International Congress of Botany, August 2004, Snowbird, Utah. Nery A, Vasconcelos H. 2003. Growth and survival of incipient ant colonies in two Amazonian ant-plants: effects of habitats, host-plant, and mode of colony founding (Hymenoptera: Formicidae). Sociobiology 42: 151 162. Niklas KJ. 1999. A mechanical perspective on foliage leaf form and function. New Phytologist 143: 19 31. Nishida S, Tsukaya H, Nagamasu H, Nozaki M. 2006. A comparative study on the anatomy and development of different shapes of domatia in Cinnamomum camphora (Lauraceae). Annals of Botany 97: 601 610. Oberbauer SF, Stain BR. 1985. Effects of light regime on the growth and physiology of Pentaclethra macroloba (Mimosaceae) in Costa Rica. Journal of Tropical Ecology 1: 303 320. Peeters PJ. 2002. Ecophysiology of cuticular transpiration: comparative investigation of cuticular water permeability of plant species from different habitats. Biological Journal of the Linnean Society 77: 43 65. Prance GT. 1972. Monograph of the Chrysobalanaceae. Flora Neotropica 9: 1 410. Prance GT, White F. 1988. The genera of Chrysobalanaceae: a study in practical and theoretical taxonomy and its relevance to evolutionary biology. Philosophical Transactions of the Royal Society 320: 1 184. Risch SJ, Rickson FR. 1981. Mutualism in which ants must be present before plants produce food bodies. Nature 291: 149 150. Romero GQ, Izzo TJ. 2004. Leaf damage induces ant recruitment in the Amazonian ant-plant Hirtella myrmecophila. Journal of Tropical Ecology 20: 675 682. Roth I. 1976. Estructura interna de los domacios foliares en Tococa (Melastomataceae). Acta Biologica Venezuela 9: 227 258. Roth-Nebelsick A, Uhl D, Mosbrugger V, Kerp H. 2001. Evolution and function of leaf venation architecture: a review. Annals of Botany 87: 553 566. Sampson FB, McLean J. 1965. A note on the occurrence of domatia on the under side of leaves in New Zealand plants. New Zealand Journal of Botany 3: 104 112. Schreiber L, Riederer M. 1996. Ecophysiology of cuticular transpiration: comparative investigation of cuticular water permeability of plant species from different habitats. Oecologia 107: 426 432. Spatz HC, Beismann H, Bruchert F, Emanns A, Speck T. 1997. Biomechanics of the giant reed Arundo donax. Philosophical Transactions: Biological Sciences 352: 1 10. Tepe EJ, Vincent MA, Watson LE. 2007a. Stem diversity, cauline domatia, and the evolution of ant plant associations in Piper sect. Macrostachys (Piperaceae). American Journal of Botany 94: 1 11. Tepe EJ, Vincent MA, Watson LE. 2007b. The importance of petiole structure on inhabitability by ants in Piper sect. Macrostachys (Piperaceae). Botanical Journal of the Linnean Society 153: 181 191. Treseder KK, Davidson DW, Ehleringer JR. 1995. Absorption of ant-provided carbon dioxide and nitrogen by a tropical epiphyte. Nature 375: 137 139. Webber BL, Abaloz BA, Woodrow IE. 2006. Myrmecophilic food body production in the understorey tree, Ryparosa kurrangii (Achariaceae), a rare Australian rainforest taxon. New Phytologist 173: 250 263. Woodward FI, Kelly CK. 1995. The influence of CO 2 concentration on stomatal density. New Phytologist 131: 311 327. Woodward FI, Lake JA, Quick WP. 2002. Stomatal development and CO 2 : ecological consequences. New Phytologist 153: 477 484.