From erotic excrescences to pheromone shots: structure and diversity of nuptial pads in anurans

From erotic excrescences to pheromone shots: structure and diversity of nuptial pads in anurans Abstract The nuptial pad is a secondary sexual character found in anuran amphibians. It includes modified epidermal and dermal tissues and is usually located on the first digit of the hand of males. In this study, we review the structure and diversity of nuptial pads based on a large-scale sample of morphological and phylogenetic diversity. Our findings show that all nuptial pads are characterized by the presence of specialized mucous glands in the dermal component. We also report the co-occurrence of other types of glands in some species and characterize their histochemistry. We describe three primary nuptial pad morphologies: (1) nuptial pads with papillary epidermal projections (or simply, papillae), in which the epidermal projections are formed by an epidermal and dermal evagination; (2) nuptial pads with non-papillary epidermal projections, in which the epidermal projections lack a dermal core forming a papilla; and (3) smooth nuptial pads, which lack projections but have a slight thickening of the dermal area with respect to the adjacent skin. We compare nuptial pads with other secondary sexual traits and discuss several related topics, including morphological diversity, coloration, hormonal control, taxonomic usefulness, and hypotheses regarding their role in reproduction. Amphibia, histology, reproductive biology, secondary sexual characters, specialized mucous glands INTRODUCTION Sexually dimorphic characters in amphibians have caught the attention of scientists for many years. Early studies of amphibian biology and morphology called attention to sexual dimorphism in tail fin structure, nuptial coloration, vocal sacs and nuptial pads (e.g. Swammerdam, 1737; Rösel von Rosenhof, 1758; Rusconi, 1821), with sexual variation in several other traits being identified subsequently (e.g. Noble, 1931; Liu, 1936; Conaway & Metter, 1967). These traits are known as secondary sexual characters (SSCs) and include any difference between the sexes that is not connected to the gonads and their ducts (Darwin, 1871; Turner & Bagnara, 1976; Sever & Staub, 2011). Among anurans, one of the most commonly reported sexually dimorphic characters is the nuptial pad. The presence of nuptial pads and, less frequently, comments on their macroscopic/microscopic structures are commonly included as standard information in taxonomic descriptions. The common reference to the presence of nuptial pads in anuran taxonomic literature, however, is far from being equivalent to our knowledge of their structure, diversity and roles in reproductive biology. Although earlier students of amphibian biology reported the presence or absence of nuptial pads in anurans (e.g. Swammerdam, 1737; Rösel von Rosenhof, 1758), Lataste (1876) was the first to survey and describe them in detail among most species of European anurans then recognized. He provided a histological analysis that focused on the epidermal structures and defined different morphologies (see Supporting Information, Table S1). Leydig (1877) and Boulenger (1897) further described and illustrated in detail the distribution of pads along the hands of European anurans. Ecker & Wiedersheim (1882) and Gaupp (1904) provided thorough histological descriptions of nuptial pads of Pelophylax kleptonesculentus (Ranidae), and Kändler (1924) described the histology of the pads in several other European species. Noble (1931) also described the presence and diversity of nuptial pads in many anurans, as did Liu (1936) in his study of sexually dimorphic characters in Chinese anurans. Several contributions, beginning in the 1920s and continuing today, have studied nuptial pads of single species, focusing mostly on histological changes associated with androgen cycles (e.g. Aron, 1926; Horié, 1939; Greenberg, 1942; Inger & Greenberg, 1956; Lofts, 1964; Rastogi, 1976; Izzo et al., 1982; Epstein & Blackburn, 1997; Yang, Zhang & Cui, 2004). Parakkal & Ellis (1963) described the nuptial pad in Rana pipiens (Ranidae), using histology and, for the first time, transmission electron microscopy (TEM). Komnick & Stockem (1970) and Kurabuchi & Inoue (1981) were the first researchers to study the epidermal component of nuptial pads using TEM and scanning electron microscopy (SEM) in Xenopus laevis (Pipidae), respectively. This was followed by studies by Zweifel (1983) and Tyler & Lungershausen (1986) on the surface of nuptial pads of some species of pelodryadine hylids and limnodynastids using SEM, and by Kurabuchi (1993, 1994) on hylids, ranids and rhacophorids using SEM and TEM. These authors defined several morphologies, noting interspecific variation in the number of cells that formed each projection and the density of the projections. Thomas, Tsang & Licht (1993) added histological and histochemical data for 14 species with sexually dimorphic skin glands (SDSGs), glands of the dermal component of the nuptial pad and some other glandular clusters, and Brizzi, Delfino & Jantra (2003a) reviewed the available information on the dermal component of nuptial pads at that time. Brizzi, Delfino & Jantra (2003b) and Kyriakopoulou-Sklavounou, Papaevangelou & Kladisios (2012) studied nuptial pad morphology in some ranids and bombinatorids. See the Supporting Information (Table S1) for the different terminology used to describe nuptial pad morphology. In the context of several taxonomic and systematic studies, nuptial pad occurrence and morphology has also been surveyed externally with different levels of detail in various groups (e.g. Parker, 1934, 1940; Wright & Wright, 1949; Inger, 1966; Tyler, 1968; Duellman, 1970; Lynch, 1971; Tyler & Davies, 1978; Scott, 2005; Grant et al., 2006; Cisneros-Heredia & McDiarmid, 2007; Duellman & Lehr, 2009; Barrionuevo, 2017). A few studies, focusing on specific groups, made more detailed analyses. In an unpublished Master’s thesis, Wray (2000) examined the nuptial pads of the hylid genera Duellmanohyla and Ptychohyla. This author studied the external pad morphology using SEM, reporting its variation, density of projections, location, and density and types of glandular pores. Cisneros-Heredia & McDiarmid (2007) provided a revised classification of nuptial pads in Centrolenidae, taking into account previous studies (e.g. Flores, 1985; Ayarzagüena, 1992; Lynch & Ruiz-Carranza, 1996; see Supporting Information, Table S1). Luna et al. (2012) studied the diversity of nuptial pads in 26 species of phyllomedusine hylids using SEM and histological procedures. Although most of these contributions are relevant for the groups they covered, they did not study nuptial pads using histology, thus leaving the structural variation, if any, in dermal structures unknown. Other secondary sexual characters, at least superficially similar to nuptial pads, had been described on the hand and other parts of the body, such as the forearm [e.g. Bombina (Bombinatoridae); Leydig, 1877], upper arm [e.g. Hyloscirtus armatus group, (Hylidae); Duellman, De La Riva & Wild, 1997; Faivovich & De la Riva, 2006], pectoral region [e.g. Alsodes (Alsodidae); Lynch, 1978], upper lip [Leptobrachium (Megophryidae); Liu, 1950], lower jaw [e.g. Ascaphus (Ascaphidae); Noble & Putnam, 1931] and even on the toes (e.g. Bombina; Leydig, 1877). Most of these had been treated simply as nuptial pads developed in body areas other than on the hands (Duellman & Trueb, 1986) or even as states within the same transformation series as nuptial pads [e.g. the ‘thumb spines’ in Leptodactylus (Leptodactylidae); Heyer, 1998; Ponssa, 2008. Hypotheses about the reproductive role of nuptial pads in frogs are probably as old as their earliest reports (Swammerdam, 1737), with most ideas revolving around an increase in the hold on the female by the male during amplexus, even if for different reasons. For example, some authors pointed out the difficulty of the male holding the female during aquatic amplexus (Lataste, 1876; Boulenger, 1897, 1910, 1912; Noble, 1931; Parker, 1940; Duellman & Trueb, 1986; Brizzi et al., 2003a), whereas others stressed the need to avoid the gripping male being dislodged by competing males (Savage, 1932, 1934, 1961; Wells, 1977, 2007). Thomas et al. (1993) were the first authors to suggest that the SDSGs of the nuptial pad might also have a pheromone-secreting function. In this paper, we present a survey of nuptial pad morphology and diversity in anurans based on external morphology, electron microscopy and histology and compare them with other SSCs and cornified structures in anurans. On the basis of those data and a review of all other relevant available information, we present a discussion of nuptial pad structure, diversity and function in anuran reproductive biology. MATERIAL AND METHODS Specimens We examined 1145 adult males of 507 species, representing 250 genera from 52 anuran families (Supplementary Information, Lists S1 and S2). From this sample, we selected 258 specimens of 212 species for histology and SEM. This subset represented 139 genera and 46 families. We also examined other areas of the skin of species where other SSCs were obviously developed (pectoral spines, lower jaw spines and cornified modifications on finger II and the chest). To understand levels of intraspecific variation, we analysed more than one male of some species. When possible, we made an effort to select specimens in active reproductive condition, as evidenced by partly dilated vocal sacs. Institutional collection codes are those recommended by Sabaj (2016). Terminology The epidermis of amphibians consists of four to seven layers of epithelial cells (Fox, 1986). The outermost layer is the stratum corneum (one layer), followed by the stratum granulosum (one layer), the stratum spinosum (one to several layers) and an innermost stratum basale (one layer). The dermis is formed by a stratum spongiosum, where different types of glands are usually found, and a stratum compactum. Both the epidermis and the dermis might vary in thickness, depending on the region of the body that they cover. In our study, we describe the epidermal and dermal components of nuptial pads separately. Throughout this paper, we use the terminology of Thomas et al. (1993), who defined ‘breeding glands’ of Anura as sexually dimorphic skin glands (SDSGs), including specialized mucous glands (SMGs) as defined by Brizzi, Delfino & Pellegrini (2002) and specialized serous glands (SSGs) as defined by Brunetti, Hermida & Faivovich (2012). We follow the recommendations of Altig (2007) and Bragulla & Homberger (2009) and distinguish keratinization from cornification. Keratinization involves the special differentiation of the epithelia (keratinocytes) together with the synthesis of keratin filaments and proteins; it can be a product of a single cell (e.g. labial teeth of tadpoles) or many cells (e.g. jaw sheaths of tadpoles, claws). The cornification process requires previous keratinization of the cells and involves dead and cornified cells (corneocytes) connected by desmosomes in the stratum corneum, and the synthesis of keratin filament-associated proteins (KAPs) that cross-link with keratin filaments via disulphide bonds. In the particular case of amphibian epidermis, the cells of the cornified layer still retain their nucleus (parakeratosis). The term ‘pigmentation’ is used when there is evidence of pigmentary cells (chromatophores) that underlie a pattern of coloration (Bagnara & Matsumoto, 2007). Fingers are numbered II–V (Fabrezi & Alberch, 1996). Most nuptial pads include elevated structures that have received different names in the literature (see Supporting Information, Table S1). Luna et al. (2012) introduced the term ‘epidermal projection’ (EP) to refer collectively to these structures in an attempt to provide a neutral term that implies nothing about its histological structure. Throughout this paper we adopt this term, but on the basis of our results we introduce further refinements (see Results section). The following abbreviations are used throughout the text and figures: AB, Alcian blue; B, bony core of metacarpal II; D, dermis; DS, distal segment; E, epidermis; EP, epidermal projection; HRP, histidine-rich protein; KAP, keratin-associated protein; L, lumen; MEC, myoepithelial cell; N, neck; NPEP, non-papillary epidermal projection; NU, nucleus; OMG, ordinary mucous gland; OSG, ordinary serous gland; PAAB, performic acid–Alcian blue; PAS, periodic acid–Schiff; PEP, papillary epidermal projection; PN, pyknotic nuclei; PS, proximal segment; SB, stratum basale; SC, stratum corneum; SCO, stratum compactum; SDSG, sexually dimorphic skin gland; SEM, scanning electron microscopy; SG, stratum granulosum; SMG, specialized mucous gland; SP, secretory portion; SS, stratum spinosum; SSC, secondary sexual character; SSG, specialized serous gland; SSP, stratum spongiosum; and TEM, transmission electron microscopy. Optical microscopy Morphological and structural analyses of nuptial pads were carried out with a Nikon SMZ 800 or Zeiss V20 stereomicroscope. Photographs were taken using a Nikon DS-Fi1 or Sony RX-100 digital camera operated with Micrometrics1 SE Premium 4 software. Histological sections were studied with a Nikon Eclypse 200 microscope and documented using a Nikon DS-Fi1 camera with NIKON NIS-Elements D software. Measurements where made using ImageJ software (Rasband, 1997–2016; https://imagej.nih.gov/ij/). Scanning electron microscopy The specimens were fixed in a 10% solution of formaldehyde and preserved in 70% ethanol. The nuptial pads were removed, dehydrated through an ascending series of ethanol up to 100%, dried using a critical point dryer (EMS 850 and Baltec CPD 030), coated with gold:palladium (40:60; SC 7620 Mini Sputter Coater Termo VG Scientific and Cressington Sputter Coater 108A), and submitted for SEM using a Philips XL30 TMP New Look microscope. Some samples were mounted in cross-sections in order to obtain images of the secretory portion of the glands and their profile morphology, whereas others were oriented with the epidermis facing upward to facilitate microscopic visualization of the external morphology of the nuptial pad area, glandular outlets, and the density and ornamentation of the projections. In a few specimens we also removed the stratum corneum of the pad to reveal the underlying epidermal structure. Transmission electron microscopy Small samples of nuptial pad regions were fixed with 2% glutaraldehyde and 4% paraformaldehyde in 0.1 mol L−1 Sorensen’s phosphate buffer (pH 7.2) for 6 h at 48 °C. Samples were rinsed in 0.1 mol L−1 phosphate buffer and postfixed for 1 h in 1% OsO4 in the same buffer. The specimens were dehydrated and embedded in araldite resin. Semi-thin sections (1 µm) were stained with 1% Toluidine blue in 1% Na2CO3 and examined under a light microscope. Ultrathin sections were stained with uranyl acetate and lead citrate (Reynolds, 1963) and observed with a Jeol JEM 1200EX II transmission electron microscope at 60 keV. Images were obtained with an ES500W Erlanshen CCD Gatan digital camera. Histological procedures Samples were dehydrated in an ascending series of ethanol, cleared in toluene and embedded in Paraplast. All samples were sectioned at 5–6 µm and stained with modified Masson’s trichrome to describe general morphology of the pads (Goldner, 1938). Histochemical studies were conducted using standard staining techniques, including Alcian blue (AB), pH 2.5 (Bancroft, 2002) for acid mucosubstances and periodic acid–Schiff (PAS) for neutral mucosubstances, with Haematoxilin used as a counterstain (Bancroft, 2002). For some samples we also conducted performic acid–Alcian blue staining (PAAB) for protein-bound sulphur (Bancroft, 2002). We treated selected samples with hydrogen peroxide 10% for 48 h in order to determine the source of the coloration of dark nuptial pads (modified from Liu et al., 2013). The histochemical properties of glands were recorded only as positive (+), negative (−) or equivocal (±), to avoid confounding the effects of variation in the fixation histories of the specimens. RESULTS Structure of the nuptial pads Nuptial pads are a modified area of skin located at least on the medial margin of finger II of most male anurans, whose invariable characteristic is the presence of SDSGs in the dermis. The external morphology of the nuptial pads usually consists of thickened skin owing to thickened epidermis or dermis, or both, when compared with adjacent skin (Fig. 1A–D). In some instances, however, no thickening is evident; in these cases, the observation of the acinar glands through the transparent epidermis indicates that nuptial pads are present (Fig. 1E, F). Figure 1. View largeDownload slide External morphology of nuptial pads under light microscopy and scanning electron microscopy. A, Ptychohyla hypomykter (USNM 580114). B, Ptychohyla hypomykter (USNM 343566). C, D, Ischnocnema guentheri (USNM 235638). E, Dendropsophus nanus (MACN 40281). F, Dendropsophus nanus (MACN 46384). Notice the evident protrusion and thickening of the nuptial pad skin with respect to the adjacent skin in A–D. There is no thickening in E compared with the adjacent skin; however, acinar glands are visible through transparent skin and reveal the presence of a nuptial pad; no modification is evident with scanning electron microscopy (F). Arrowheads indicate the margins of the nuptial pad. Scale bars: 500 µm. Figure 1. View largeDownload slide External morphology of nuptial pads under light microscopy and scanning electron microscopy. A, Ptychohyla hypomykter (USNM 580114). B, Ptychohyla hypomykter (USNM 343566). C, D, Ischnocnema guentheri (USNM 235638). E, Dendropsophus nanus (MACN 40281). F, Dendropsophus nanus (MACN 46384). Notice the evident protrusion and thickening of the nuptial pad skin with respect to the adjacent skin in A–D. There is no thickening in E compared with the adjacent skin; however, acinar glands are visible through transparent skin and reveal the presence of a nuptial pad; no modification is evident with scanning electron microscopy (F). Arrowheads indicate the margins of the nuptial pad. Scale bars: 500 µm. Dermal component The dermal component of the pad usually shows an increase in thickness of the stratum spongiosum and stratum compactum compared with the adjacent skin (Fig. 2A). The area is characterized by the presence of SDSGs. In most species we studied, these SDSGs correspond to SMGs. In only one case (Xenopus epitropicalis) did these co-occur with SSGs (Fig. 2B; see Supporting Information, Table S2). The most common dermal arrangement includes the presence of closely packed and hypertrophied SMGs, without ordinary mucous glands (OMGs) and ordinary serous glands (OSGs). However, scattered OMGs and OSGs are found in a few species [e.g. Mixophyes fasciolatus (Myobatrachidae), Pseudophryne bibroni (Myobatrachidae), Rhinella arenarum (Bufonidae), Strongylopus faciolatus (Pyxicephalidae); Fig. 2C; see Supporting Information, Table S2). Figure 2. View largeDownload slide Dermal component of the nuptial pad. A, Aparasphenodon brunoi (CFBH 22900). Note the increase in thickness of the dermis of the nuptial pad compared with the adjacent skin; dashed line separates nuptial pad from adjacent skin. B, Xenopus epitropicalis (USNM 573445). Specialized mucous glands co-occur with SSGs. C, Rhinella arenarum (MACN 40670). Specialized mucous glands co-occur with OSGs. D, Phyllomedusa azurea (MACN 51205); semi-thin section of SMGs. The secretory portion of the gland is formed by a monolayer of columnar cells with basal nuclei. Secretory granules occur in most glandular cells. The neck is formed by a double layer of cells that lack secretory granules. E, Scinax perereca (MACN 43321). Transmission electron micrograph. Note the electron-dense secretory granules. Histological staining: periodic acid–Schiff in B; modified Masson’s trichrome in A, C. Toluidine blue in D. Abbreviations: D, dermis; E, epidermis; L, lumen; MEC, myoepithelial cell; N, neck; OSG, ordinary mucous gland; SCO, stratum compactum; SMG, specialized mucous gland; SP, secretory portion; SSG, specialized serous gland; SSP, stratum spongiousum. Scale bars: 100 µm in A–C; 15 µm in D; 4 µm in E. Figure 2. View largeDownload slide Dermal component of the nuptial pad. A, Aparasphenodon brunoi (CFBH 22900). Note the increase in thickness of the dermis of the nuptial pad compared with the adjacent skin; dashed line separates nuptial pad from adjacent skin. B, Xenopus epitropicalis (USNM 573445). Specialized mucous glands co-occur with SSGs. C, Rhinella arenarum (MACN 40670). Specialized mucous glands co-occur with OSGs. D, Phyllomedusa azurea (MACN 51205); semi-thin section of SMGs. The secretory portion of the gland is formed by a monolayer of columnar cells with basal nuclei. Secretory granules occur in most glandular cells. The neck is formed by a double layer of cells that lack secretory granules. E, Scinax perereca (MACN 43321). Transmission electron micrograph. Note the electron-dense secretory granules. Histological staining: periodic acid–Schiff in B; modified Masson’s trichrome in A, C. Toluidine blue in D. Abbreviations: D, dermis; E, epidermis; L, lumen; MEC, myoepithelial cell; N, neck; OSG, ordinary mucous gland; SCO, stratum compactum; SMG, specialized mucous gland; SP, secretory portion; SSG, specialized serous gland; SSP, stratum spongiousum. Scale bars: 100 µm in A–C; 15 µm in D; 4 µm in E. Specialized mucous glands are formed by an intraepidermal duct, a neck and a secretory portion. The secretory portion is composed of a monolayer of columnar cells and a contractile sheath of myoepithelial cells that is discontinuous (Fig. 2D). The columnar cells have basal nuclei and secretory granules (Fig. 2E). Dense vascularization can be present surrounding the secretory portion of the glands, below the epidermis, and also in the core of the papillary epidermal projections (see section on ‘Nuptial pads with papillae’ below). Ordinary mucous glands differ from specialized mucous glands in size and morphology. Ordinary mucous glands are smaller than SMGs, and their secretory portion is a monolayer of flat to cubical cells. Most frequently, SMGs are hypertrophied with respect to mucous glands in the adjacent skin (Fig. 3A). The level of aggregation of the SMGs varies from scattered to densely packed (Fig. 3B, C), and their secretory portion can be arranged in one or more layers depending on the species [e.g. one layer in Boana semilineata (Hylidae); Fig. 3D; three layers in Ascaphus montanus (Ascaphidae); Fig. 3E]. The morphology of the specialized mucous glands varies, being alveolar (e.g. the hylid tribe Cophomantini; Fig. 3F) or tubuloalveolar (e.g. the hylid tribe Lophiohylini, Fig. 3G), with a thinner or a thicker stratum spongiosum, respectively. Typically, SMGs consist of simple glands, but in two species [Ascaphus truei and Calyptocephalella gayi (Calyptocephalellidae)] they are ramified alveolar glands (Fig. 3H). Usually, the glands are elongated dorsoventrally, but in some cases we find that the longest axis of the secretory portion of the gland is parallel to the epidermis (Fig. 3I). Figure 3. View largeDownload slide Light micrographs of cross-sections of nuptial pads. A, Aglyptodactylus madagascariensis (USNM 499560). Specialized mucous glands are present only in the nuptial pad area. Notice the increment of dermal and epidermal components in the nuptial pad with respect to the adjacent skin (black arrows). B, Spea bombifrons (MACN 39156). Specialized mucous glands are scattered in the stratum spongiosum of the dermis. C, Phyllomedusa camba (CFBH 21725). Specialized mucous glands are closely packed in the stratum spongiosum of the dermis. D, Boana semilineata (CFBH 11248). Specialized mucous glands are arranged in a single layer. E, Ascaphus montanus (USNM 334425). Specialized mucous glands are arranged in two or more layers. F, Phyllomedusa azurea (MACN 40066). Specialized mucous glands have an alveolar secretory portion. G, Argenteohyla siemersi (CENAI 3082). Specialized mucous glands have a tubuloalveolar secretory portion. H, Ascaphus montanus (USNM 33425). Specialized mucous glands have a ramified alveolar structure. Note the union of two alveoli (white arrowhead). I, Rhinophrynus dorsalis (USNM 219809). The secretory portions of the SMGs are parallel to the epidermis. Histological staining: modified Masson’s trichrome in A–E, G, H; periodic acid–Schiff in F, I. Abbreviations: D, dermis; E, epidermis; OMG, ordinary mucous gland; SMG, specialized mucous gland. Scale bars: 100 µm. Figure 3. View largeDownload slide Light micrographs of cross-sections of nuptial pads. A, Aglyptodactylus madagascariensis (USNM 499560). Specialized mucous glands are present only in the nuptial pad area. Notice the increment of dermal and epidermal components in the nuptial pad with respect to the adjacent skin (black arrows). B, Spea bombifrons (MACN 39156). Specialized mucous glands are scattered in the stratum spongiosum of the dermis. C, Phyllomedusa camba (CFBH 21725). Specialized mucous glands are closely packed in the stratum spongiosum of the dermis. D, Boana semilineata (CFBH 11248). Specialized mucous glands are arranged in a single layer. E, Ascaphus montanus (USNM 334425). Specialized mucous glands are arranged in two or more layers. F, Phyllomedusa azurea (MACN 40066). Specialized mucous glands have an alveolar secretory portion. G, Argenteohyla siemersi (CENAI 3082). Specialized mucous glands have a tubuloalveolar secretory portion. H, Ascaphus montanus (USNM 33425). Specialized mucous glands have a ramified alveolar structure. Note the union of two alveoli (white arrowhead). I, Rhinophrynus dorsalis (USNM 219809). The secretory portions of the SMGs are parallel to the epidermis. Histological staining: modified Masson’s trichrome in A–E, G, H; periodic acid–Schiff in F, I. Abbreviations: D, dermis; E, epidermis; OMG, ordinary mucous gland; SMG, specialized mucous gland. Scale bars: 100 µm. Melanophores usually occur in the stratum spongiosum of the dermis (Fig. 4A–D), but occasionally they are also found in the stratum spinosum of the epidermis (e.g. Rana cascadae). The melanin is either scattered (Fig. 4A, B) or in a continuous layer (Fig. 4C, D) in the dermis. Figure 4. View largeDownload slide Light micrographs of nuptial pads showing differences in the presence and distribution of melanophores in the stratum spongiosum of the dermis. A, Telmatobius rubigo (MACN 41687), showing scattered melanophores (arrowheads). B, detail of A. C, Conraua crassipes (USNM 571261), showing melanophores in a continuous layer beneath the epidermis (arrowheads). D, detail of C. Histological staining: modified Masson’s trichrome. Scale bars: 100 µm in A, C; 50 µm in B, D. Figure 4. View largeDownload slide Light micrographs of nuptial pads showing differences in the presence and distribution of melanophores in the stratum spongiosum of the dermis. A, Telmatobius rubigo (MACN 41687), showing scattered melanophores (arrowheads). B, detail of A. C, Conraua crassipes (USNM 571261), showing melanophores in a continuous layer beneath the epidermis (arrowheads). D, detail of C. Histological staining: modified Masson’s trichrome. Scale bars: 100 µm in A, C; 50 µm in B, D. In histochemical analyses of the SMGs, all species reacted positive for neutral mucosubstances (PAS+) and some also for acidic mucosubstances (AB+) (Fig. 5A, B; see Supporting Information, Table S2). A positive result was also revealed with the PAAB reaction in the upper layer of the nuptial pads, corroborating the presence of disulphide and sulphur groups in the stratum corneum (Fig. 5C). It was not feasible to stain for lipids in our study because most of the samples came from preserved collections; hence, the differentiation of SSGs was based mainly on morphological differences evidenced by light microscopy and a negative result for PAS and AB staining. The hydrogen peroxide test revealed the loss of pigmentation in the dermal melanocytes but did not affect the epidermal coloration of the dark-coloured nuptial pads, thus showing that melanin is not involved in the coloration of the stratum corneum (Fig. 5D–G; see ‘Coloration and cornification’ below). Figure 5. View largeDownload slide Light micrographs of histochemical results in specialized mucous glands (SMGs) of nuptial pads. A, Aglyptodactylus madagascariensis (USNM 499560); positive reaction of Alcian blue (AB) in several cells of the secretory portion of the SMGs (white arrowheads). B, Rana cascadae (USNM 312045); negative AB reaction in SMGs. C, Thoropa megatympanum (USNM 218197); positive reaction of performic acid–Alcian blue staining (PAAB) in the stratum corneum and the apices of the stratum granulosum (arrowheads). D–G, Xenopus tropicalis (USNM 571230). D, cross-section of nuptial pad used as a control. E, results of the incubation of nuptial pads in peroxidase. F, higher magnification view of D. Note the occurrence of dermal melanophores (black arrows). G, higher magnification view of E. Note the bleaching of the dermal melanophores and the persistence of the coloration in the stratum corneum (white arrows), indicating that melanin is not involved. Histological staining: Alcian blue in A, B; performic acid–Alcian blue in C; Haematoxylin and Eosin in D, F (control); Haematoxylin, Eosin and treatment with peroxidase in E, G. Abbreviations: SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bars: 50 µm in A, B, F, G; 10 µm in C; 100 µm in D, E. Figure 5. View largeDownload slide Light micrographs of histochemical results in specialized mucous glands (SMGs) of nuptial pads. A, Aglyptodactylus madagascariensis (USNM 499560); positive reaction of Alcian blue (AB) in several cells of the secretory portion of the SMGs (white arrowheads). B, Rana cascadae (USNM 312045); negative AB reaction in SMGs. C, Thoropa megatympanum (USNM 218197); positive reaction of performic acid–Alcian blue staining (PAAB) in the stratum corneum and the apices of the stratum granulosum (arrowheads). D–G, Xenopus tropicalis (USNM 571230). D, cross-section of nuptial pad used as a control. E, results of the incubation of nuptial pads in peroxidase. F, higher magnification view of D. Note the occurrence of dermal melanophores (black arrows). G, higher magnification view of E. Note the bleaching of the dermal melanophores and the persistence of the coloration in the stratum corneum (white arrows), indicating that melanin is not involved. Histological staining: Alcian blue in A, B; performic acid–Alcian blue in C; Haematoxylin and Eosin in D, F (control); Haematoxylin, Eosin and treatment with peroxidase in E, G. Abbreviations: SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bars: 50 µm in A, B, F, G; 10 µm in C; 100 µm in D, E. Epidermal component The epidermis of the nuptial pad may show extreme modifications in comparison to the adjacent skin owing to an increase in the cellular layers of the stratum spinosum and/or changes in the morphology of the cells in the stratum spinosum and stratum corneum (turgid cuboidal cells and thicker cells, respectively). As a result, we recognize three main nuptial pad morphologies. Nuptial pads with papillary epidermal projections (PEPs or papillae), where epidermal projections are formed by an epidermal and dermal evagination (Fig. 6A). This is the most common morphology (74.3% of all species analysed with histology and/or SEM). Nuptial pads with non-papillary epidermal projections (NPEPs), where epidermal projections lack a dermal core. This is the least common morphology (2.9% of all species analysed with histology and/or SEM; Fig. 6B). Smooth nuptial pads, where no projections are present but there is a slight increase in the thickness of the dermal area with respect to the adjacent skin (22.8% of all species analysed with histology and/or SEM; Fig. 6C). Figure 6. View largeDownload slide Schematic representation of nuptial pads with different epidermal projections (EPs). A, nuptial pad with papillary epidermal projections (PEPs or papillae). The epidermis protrudes from the adjacent skin, and the dermis protrudes into the epidermal projection, as the core of the papilla. Several capillaries are seen in the dermal core. B, nuptial pad with non-papillary epidermal projections (NPEPs). The epidermis protrudes from the adjacent skin; there is no dermal core. The epidermal projections are produced by changes in cell morphology of the most external epidermal strata (stratum granulosum and corneum). C, smooth nuptial pad. There are no papillae or NPEPs. The protrusion of the nuptial pad, if it occurs, is attributable to the dermal component only. Abbreviations: D, dermis; E, epidermis SB, stratum basale; SC, stratum corneum; SCO, stratum compactum; SG, stratum granulosum; SS, stratum spinosum; SSP, stratum spongiosum. Figure 6. View largeDownload slide Schematic representation of nuptial pads with different epidermal projections (EPs). A, nuptial pad with papillary epidermal projections (PEPs or papillae). The epidermis protrudes from the adjacent skin, and the dermis protrudes into the epidermal projection, as the core of the papilla. Several capillaries are seen in the dermal core. B, nuptial pad with non-papillary epidermal projections (NPEPs). The epidermis protrudes from the adjacent skin; there is no dermal core. The epidermal projections are produced by changes in cell morphology of the most external epidermal strata (stratum granulosum and corneum). C, smooth nuptial pad. There are no papillae or NPEPs. The protrusion of the nuptial pad, if it occurs, is attributable to the dermal component only. Abbreviations: D, dermis; E, epidermis SB, stratum basale; SC, stratum corneum; SCO, stratum compactum; SG, stratum granulosum; SS, stratum spinosum; SSP, stratum spongiosum. Nuptial pads with papillae Papillae are characterized by an increase in the number of layers of the stratum spinosum with respect to that of adjacent skin, and by changes in the morphology of the cells of the stratum granulosum and stratum corneum (Fig. 7A–C). Morphological changes of the stratum granulosum include an increase in cell turgidity, a flat to columnar cell morphology and, in some species, the occurrence of cytoplasmic projections in the apical portion of the cells (Fig. 7B, C). The stratum basale consists of a monolayer of low columnar cells on the basal membrane. The core of the papilla is formed by a dermal evagination of the stratum spongiosum into the epidermis. This evagination is densely vascularized (Fig. 7A). Each papilla is formed by multiple epidermal cells, which may also show elaborate ornamentation (Fig. 7D, E). Scanning electron microscopy and light microscopy reveal the morphology of the stratum granulosum as a positive cast of the overlying stratum corneum (Fig. 7D–G). We frequently found some papillae that were partly fused [e.g. Phyllomedusa azurea, Tepuihyla edelcae (Hylidae)]. Along the margins of the nuptial pad, the papillae are usually less conspicuous or depressed (Fig. 7H). Figure 7. View largeDownload slide Nuptial pads with papillary epidermal projections (PEPs). A, Phyllomedusa azurea (MACN 40066). Note the evagination of both epidermal and dermal components and the presence of well-developed specalized mucous glands. Capillaries occur near or in the core of the papillae (arrows). B, Phyllomedusa azurea (MACN 51205); semi-thin section. Note the increase in the size and turgidity of the epidermal cells (white arrowheads). C, Strongylopus fasciatus (USNM 153095); scanning electron micrograph. Note the epidermal layers in profile. D, Bombina orientalis (MACN 42129); scanning electron micrograph. Stratum corneum partly removed, exposing the stratum granulosum, which is a positive cast of the morphology of the stratum corneum. E, Bombina orientalis (MACN 42129). Detail of the morphology of the stratum granulosum. F, Telmatobius rubigo (MACN 41687); scanning electron micrograph. Stratum corneum partly removed (on left), exposing the stratum granulosum. G, Agalychnis hulli (MNRJ 74779); scanning electron micrograph. Papillae are depressed and less conspicuous in the margins of the nuptial pad (black arrowheads). Histological staining: modified Masson’s trichrome in A, Toluidine blue in B. Abbreviations: D, dermis; E, epidermis; SB, stratum basale; SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bars: 50 µm in A, D, F, G; 10 µm in B, C, E. Figure 7. View largeDownload slide Nuptial pads with papillary epidermal projections (PEPs). A, Phyllomedusa azurea (MACN 40066). Note the evagination of both epidermal and dermal components and the presence of well-developed specalized mucous glands. Capillaries occur near or in the core of the papillae (arrows). B, Phyllomedusa azurea (MACN 51205); semi-thin section. Note the increase in the size and turgidity of the epidermal cells (white arrowheads). C, Strongylopus fasciatus (USNM 153095); scanning electron micrograph. Note the epidermal layers in profile. D, Bombina orientalis (MACN 42129); scanning electron micrograph. Stratum corneum partly removed, exposing the stratum granulosum, which is a positive cast of the morphology of the stratum corneum. E, Bombina orientalis (MACN 42129). Detail of the morphology of the stratum granulosum. F, Telmatobius rubigo (MACN 41687); scanning electron micrograph. Stratum corneum partly removed (on left), exposing the stratum granulosum. G, Agalychnis hulli (MNRJ 74779); scanning electron micrograph. Papillae are depressed and less conspicuous in the margins of the nuptial pad (black arrowheads). Histological staining: modified Masson’s trichrome in A, Toluidine blue in B. Abbreviations: D, dermis; E, epidermis; SB, stratum basale; SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bars: 50 µm in A, D, F, G; 10 µm in B, C, E. Nuptial pads with papillae include six different morphologies that are characterized by their shape and histological structure. Furthermore, they fall into two size classes: small papillae and spine-shaped papillae. Small papillae have a diameter of 15–170 µm and a height of 16–150 µm; they include truncate, rounded, irregular, cone-, and spicule-shaped papillae. Spine-shaped papillae have a diameter of 165–800 µm and a height of 120–700 µm. These types are described in detail below. Small papillae Truncate papillae: These papillae have a flattened apex. They have a diameter of 44–68 µm and a height of 33–68 µm. The cells of the stratum corneum are thin and flattened. The cells of the stratum granulosum are turgid, and they present a flat apical margin in contact with the stratum corneum (Fig. 8A–D). These papillae occur in 5.4% of nuptial pads with PEPs [e.g. Dendropsophus molitor (Hylidae), Edalorinha perezi (Leptodactylidae), Ischnocnema guentheri (Brachycephalidae)]. Rounded papillae: These papillae are hemispherical to elliptical, and most are proportionally uniform with a regular profile. They have a diameter of 27–75 µm and a height of 25–55 µm. Ornamentations that cover the papillae are uniformly short (Fig. 8E–H) or larger in the apex (Fig. 8I–L). These papillae occur in 59.2% of nuptial pads with PEPs [e.g. Amietia angolensis (Pyxicephalidae), Buergeria japonica (Rhacophoridae), Nannophryne variegata (Bufonidae)]. Figure 8. View largeDownload slide Scanning electron micrographs and light micrographs of nuptial pads with papillary epidermal projections (PEPs, or papillae). A–D, Ischnocnema nasuta (USNM 229856); truncate papillae. A, light-coloured nuptial pad. B, scanning electron micrograph of the smooth and flattened surface of the cells of the stratum corneum. C, cross-section showing the truncate morphology of the papillae. D, magnification of C. Note the turgidity of the cells of the stratum granulosum, and the flattened apical margin in contact with the stratum corneum (white arrowhead and black arrows, respectively). E–H, Strongylopus fasciatus (USNM 153095); rounded papillae. E, light-coloured nuptial pad. F, scanning electron micrograph of the rounded papillae with apical ornamentations. G, cross-section showing the rounded regular profile of the papillae. H, magnification of G. Note the turgidity of the cells of the stratum granulosum and the short ornamentations in the apex of each elevation (white arrowhead and black arrow, respectively). I–L, Osteopilus vastus (MACN 45309); rounded papillae. I, dark-coloured nuptial pad. J, scanning electron micrograph of the rounded papillae with apical ornamentations that are more pronounced than in F. K, cross-section showing the rounded, regular profile of the papillae; compare with J. L, magnification of K. Note the conspicuous ornamentations in the apices of the papillae, in both the stratum granulosum and the stratum corneum (white arrowheads and black arrows, respectively). M–P, Scinax nasicus (MACN 39431); irregular papillae. M, light-coloured nuptial pad. Some individual acini are seen along the dorsal extension of finger II and III, reaching the disc. N, scanning electron micrograph showing differences in morphology of papillae. O, cross-section showing the different shape and orientation of each papilla, and differences in height and diameter (black arrows). P, magnification of O. Histological staining: modified Masson’s trichrome in G, H, K, L, O, P; periodic acid–Schiff in C, D. Scale bars: 500 µm in A, E, I, M; 50 µm in B, C, F, G, J, K, N, O; 10 µm in D, H, L, P. Figure 8. View largeDownload slide Scanning electron micrographs and light micrographs of nuptial pads with papillary epidermal projections (PEPs, or papillae). A–D, Ischnocnema nasuta (USNM 229856); truncate papillae. A, light-coloured nuptial pad. B, scanning electron micrograph of the smooth and flattened surface of the cells of the stratum corneum. C, cross-section showing the truncate morphology of the papillae. D, magnification of C. Note the turgidity of the cells of the stratum granulosum, and the flattened apical margin in contact with the stratum corneum (white arrowhead and black arrows, respectively). E–H, Strongylopus fasciatus (USNM 153095); rounded papillae. E, light-coloured nuptial pad. F, scanning electron micrograph of the rounded papillae with apical ornamentations. G, cross-section showing the rounded regular profile of the papillae. H, magnification of G. Note the turgidity of the cells of the stratum granulosum and the short ornamentations in the apex of each elevation (white arrowhead and black arrow, respectively). I–L, Osteopilus vastus (MACN 45309); rounded papillae. I, dark-coloured nuptial pad. J, scanning electron micrograph of the rounded papillae with apical ornamentations that are more pronounced than in F. K, cross-section showing the rounded, regular profile of the papillae; compare with J. L, magnification of K. Note the conspicuous ornamentations in the apices of the papillae, in both the stratum granulosum and the stratum corneum (white arrowheads and black arrows, respectively). M–P, Scinax nasicus (MACN 39431); irregular papillae. M, light-coloured nuptial pad. Some individual acini are seen along the dorsal extension of finger II and III, reaching the disc. N, scanning electron micrograph showing differences in morphology of papillae. O, cross-section showing the different shape and orientation of each papilla, and differences in height and diameter (black arrows). P, magnification of O. Histological staining: modified Masson’s trichrome in G, H, K, L, O, P; periodic acid–Schiff in C, D. Scale bars: 500 µm in A, E, I, M; 50 µm in B, C, F, G, J, K, N, O; 10 µm in D, H, L, P. Irregular papillae: These papillae are neither truncate nor rounded; their size and shape vary continuously within the nuptial pad. Nuptial pads with irregular papillae occur in some species where multiple specimens were available for study, indicating that the irregular size and shape of the papillae are not the result of ontogenetic variation, teratology or sample preparation (Fig. 8M–P). These papillae occur in 5.4% of nuptial pads with PEPs [e.g. Fejervarya limnocharis (Dicroglossidae), Scinax nasicus (Hylidae)]. Cone-shaped papillae: These papillae have a pointed apex. The cells of the stratum granulosum are turgid and columnar in shape. The surface of the papillae may be smooth (Fig. 9A–D, I–L) or ornamented (Fig. 9E–H, M–P), depending on the morphology of the apical part of the cells of the stratum granulosum and stratum corneum. If the apical part of the cells of the stratum granulosum is flat and smooth, the stratum corneum is also smooth. On the contrary, if the apical part of the cells of the stratum granulosum shows cytoplasmatic projections, then these ornamentations also occur in the stratum corneum. These papillae also show differences in orientation with respect to the epidermis, being straight (Fig. 9A–H) or leaning towards the sides (Fig. 9I–P). There is also great variation in diameter (37–170 µm) and height (35–150 µm) of these papillae, but further analyses are needed to determine whether there are discrete size classes. These papillae occur in 21.5% of the nuptial pads with PEPs [e.g. Ascaphus montanus, Spea multiplicata (Scaphiopodidae), Telmatobius rubigo (Telmatobiidae)]. Figure 9. View largeDownload slide External morphology, scanning electron micrographs and light micrographs of nuptial pads with cone-shaped papillary epidermal projections (PEPs, or papillae). A–D, Telmatobius rubigo (MACN 41687); cone-shaped papillae without ornamentation and with straight axis. A, dark-coloured nuptial pad. B, scanning electron micrograph of the cone-shaped papillae. C, magnification of E. Note the surface of the papilla, lacking ornamentations. D, cross-section. E–H, Rhinella arenarum (MACN 40670); cone-shaped papillae with ornamentations and with straight axis. E, dark-coloured nuptial pad. F, scanning electron micrograph of the cone-shaped papillae. G, detail of a papilla showing ornamentations. H, cross-section. I–L, Ascaphus montanus (USNM 33425); cone-shaped papillae without ornamentation and axis leaned to the side. I, dark-coloured nuptial pad. J, scanning electron micrograph of the cone-shaped papillae. K, magnification of G. Note the smooth surface of each papilla. L, cross-section. M–P, Corythomantis greeningi (CFBH 16129); cone-shaped papillae with ornamentations and axis leaned to the side. M, dark-coloured nuptial pad. N, scanning electron micrograph of the cone-shaped papillae. O, magnification of H, showing the ornamentation of each papilla. P, cross-section. Histological staining: modified Masson’s trichrome in D, L, P; Alcian blue with Haematoxylin as counterstain in H. Scale bars: 1 mm in A, E, I, M; 50 µm in B, F, J, N; 20 µm in C, G, K, O; 100 µm in D, H, L, P. Figure 9. View largeDownload slide External morphology, scanning electron micrographs and light micrographs of nuptial pads with cone-shaped papillary epidermal projections (PEPs, or papillae). A–D, Telmatobius rubigo (MACN 41687); cone-shaped papillae without ornamentation and with straight axis. A, dark-coloured nuptial pad. B, scanning electron micrograph of the cone-shaped papillae. C, magnification of E. Note the surface of the papilla, lacking ornamentations. D, cross-section. E–H, Rhinella arenarum (MACN 40670); cone-shaped papillae with ornamentations and with straight axis. E, dark-coloured nuptial pad. F, scanning electron micrograph of the cone-shaped papillae. G, detail of a papilla showing ornamentations. H, cross-section. I–L, Ascaphus montanus (USNM 33425); cone-shaped papillae without ornamentation and axis leaned to the side. I, dark-coloured nuptial pad. J, scanning electron micrograph of the cone-shaped papillae. K, magnification of G. Note the smooth surface of each papilla. L, cross-section. M–P, Corythomantis greeningi (CFBH 16129); cone-shaped papillae with ornamentations and axis leaned to the side. M, dark-coloured nuptial pad. N, scanning electron micrograph of the cone-shaped papillae. O, magnification of H, showing the ornamentation of each papilla. P, cross-section. Histological staining: modified Masson’s trichrome in D, L, P; Alcian blue with Haematoxylin as counterstain in H. Scale bars: 1 mm in A, E, I, M; 50 µm in B, F, J, N; 20 µm in C, G, K, O; 100 µm in D, H, L, P. Spicule-shaped papillae: These are the smallest papillae and are evident only as tiny, sparsely distributed spicules at ×30 magnification (Fig. 10A, B). They have a diameter of 15–21 µm and a height of 16–19 µm. These papillae occur in only 2.3% of the nuptial pads with PEPs (e.g. Scinax fuscovarius, X. epitropicalis and Xenopus tropicalis). Scanning electron microscopy reveals that in S. fuscovarius the skin is smooth, except in areas where the papillae occur, which are slightly elevated with respect to the rest of the pad (Fig. 10C, inset). Spicule-shaped papillae in X. epitropicalis and X. tropicalis are more densely distributed than those in S. fuscovarius (Fig. 10D, inset). The stratum corneum in the three species is formed by a monolayer that invaginates into the epidermis, resulting in a folded appearance. The main difference is the dark-coloured stratum corneum in the Xenopus species (Fig. 10E vs. F), making the identification of the nuptial pad easier. The reduced dermal core is evident only in serial sections (Fig. 10G, H). Figure 10. View largeDownload slide External morphology, scanning electron micrographs and light micrographs of nuptial pads with spicule-shaped papillary epidermal projections (PEPs, or papillae). A, C, E, G, Scinax fuscovarius (MACN 34999). A, light coloured spicule-shaped papillae. C, scanning electron micrograph and inset. Note the smooth skin of the nuptial pad area, except in the areas where the EPs are projecting, which are slightly elevated from the rest of the skin. E, cross-section of spicule-shaped papillae. Note the presence of specialized mucous glands in the dermis. G, detail of a single spicule-shaped PEP. Note the invagination of the stratum corneum (black arrowhead). B, D, F, H, Xenopus epitropicalis (USNM 573445). B, dark-coloured spicule-shaped papillae in palmar surface. D, scanning electron micrograph and inset, spicule-shaped papillae. F, cross-section of dark-coloured spicule-shaped papillae. H, magnification of F. Note the increment of the stratum spinosum (black arrows) and the invagination of the stratum corneum (black arrowhead). Histological staining: modified Masson’s trichrome in E–H. Scale bars: 1 mm in A, B; 500 µm in C; 100 µm in D; 50 µm in E, F; 10 µm in G, H; 20 µm in insets in C, D. Figure 10. View largeDownload slide External morphology, scanning electron micrographs and light micrographs of nuptial pads with spicule-shaped papillary epidermal projections (PEPs, or papillae). A, C, E, G, Scinax fuscovarius (MACN 34999). A, light coloured spicule-shaped papillae. C, scanning electron micrograph and inset. Note the smooth skin of the nuptial pad area, except in the areas where the EPs are projecting, which are slightly elevated from the rest of the skin. E, cross-section of spicule-shaped papillae. Note the presence of specialized mucous glands in the dermis. G, detail of a single spicule-shaped PEP. Note the invagination of the stratum corneum (black arrowhead). B, D, F, H, Xenopus epitropicalis (USNM 573445). B, dark-coloured spicule-shaped papillae in palmar surface. D, scanning electron micrograph and inset, spicule-shaped papillae. F, cross-section of dark-coloured spicule-shaped papillae. H, magnification of F. Note the increment of the stratum spinosum (black arrows) and the invagination of the stratum corneum (black arrowhead). Histological staining: modified Masson’s trichrome in E–H. Scale bars: 1 mm in A, B; 500 µm in C; 100 µm in D; 50 µm in E, F; 10 µm in G, H; 20 µm in insets in C, D. Spine-shaped papillae Spine-shaped papillae are distinguishable with low magnification from cone-shaped papillae by the number of EPs and, in most cases, by their size. Spine-shaped papillae have a diameter of 165–800 µm and a height of 120–700 µm. They always occur in small numbers (from one to ~80, depending on the species; Fig. 11A, B), and their diameters are usually larger than those of cone-shaped papillae (Fig. 11C, D). Histological analyses show that the epidermis consists of five to ten layers, mostly attributable to an increase in cell layers of the stratum spinosum. The stratum granulosum includes a monolayer of turgid, cuboidal cells, which are notably larger than those of other papillae morphologies (Fig. 11E). The stratum corneum is still a monolayer but is thicker than that of all other nuptial pad morphologies, and lacks ornamentations. Only a few SMGs are found in the dermis, generally at the base of the EP. As a consequence, there are relatively few pores (as revealed by SEM), which occur on the lower margin of the spine-shaped papillae, near the base (Fig. 11F). The dermal core of the papilla consists primarily of abundant and loosely arranged collagen fibres. Although no quantitative data are available, there is an obvious increase in vascularization of the dermis (Fig. 11G) in proportion to the other nuptial pad morphologies. These papillae occur in only 6.2% of the nuptial pads with PEPs [e.g. Alsodes gargola, Osteopilus wilderi (Hylidae), Ptychohyla hypomykter (Hylidae), Thoropa taophora (Cycloramphidae)]. Figure 11. View largeDownload slide Dark-coloured nuptial pads with spine-shaped papillary epidermal projections (PEPs, or papillae) and comparison with cone-shaped papillae. A, Thoropa petropolitana (USNM 164137). External morphology of the spine-shaped papillae. B, Thoropa megatympanum (USNM 218197); scanning electron micrograph. Note the position and size of each spine. C, Osteopilus wilderi (USNM 251289); scanning electron micrograph. Spine-shaped papillae. D, Ascaphus truei (AMNH 169115); scanning electron micrograph. Cone-shaped papillae. Note the differences in size and density of the epidermal projections between C and D. E, Thoropa petropolitana (USNM 164137); cross-section. Notice the increment in thickness of the stratum corneum and the turgidity of the cells of the stratum granulosum and stratum spinosum. F, Thoropa megatympanum (USNM 218197); scanning electron micrograph. Detail of a spine, showing few secretory pores near the base (white arrowheads). G, Thoropa taophora (CFBH 12734); cross-section of spine-shaped papillae showing the presence of few SMGs and abundant collagen fibres and capillaries (arrows) in the dermis. Histological staining: modified Masson’s trichrome in E, G. Abbreviations: SB, stratum basale; SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bars 1 mm in A; 500 µm in B; 100 µm in C, D, G; 10 µm in E; 50 µm in F. Figure 11. View largeDownload slide Dark-coloured nuptial pads with spine-shaped papillary epidermal projections (PEPs, or papillae) and comparison with cone-shaped papillae. A, Thoropa petropolitana (USNM 164137). External morphology of the spine-shaped papillae. B, Thoropa megatympanum (USNM 218197); scanning electron micrograph. Note the position and size of each spine. C, Osteopilus wilderi (USNM 251289); scanning electron micrograph. Spine-shaped papillae. D, Ascaphus truei (AMNH 169115); scanning electron micrograph. Cone-shaped papillae. Note the differences in size and density of the epidermal projections between C and D. E, Thoropa petropolitana (USNM 164137); cross-section. Notice the increment in thickness of the stratum corneum and the turgidity of the cells of the stratum granulosum and stratum spinosum. F, Thoropa megatympanum (USNM 218197); scanning electron micrograph. Detail of a spine, showing few secretory pores near the base (white arrowheads). G, Thoropa taophora (CFBH 12734); cross-section of spine-shaped papillae showing the presence of few SMGs and abundant collagen fibres and capillaries (arrows) in the dermis. Histological staining: modified Masson’s trichrome in E, G. Abbreviations: SB, stratum basale; SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bars 1 mm in A; 500 µm in B; 100 µm in C, D, G; 10 µm in E; 50 µm in F. Nuptial pads with non-papillary epidermal projections This type of nuptial pad has EPs without a dermal evagination forming a papilla. These EPs are truncate (height, 12–15 µm; diameter, 5–10 µm) or flat shaped in profile (height, 4–6 µm; diameter, 14–20 µm). They are distinguishable from most papillary epidermal projections by their smaller diameter and by the number of cells involved in each EP. We comment below on the few species we studied with this type of nuptial pad. We observed truncate NPEPs only in Pseudis platensis (Hylidae) and flat NPEPs only in Rhinophrynus dorsalis (Rhinophrynidae). At low magnification, these structures are indiscernible from nuptial pads without EPs (smooth nuptial pads; Fig. 12A, B). Scanning electron micrographs show that each projection has a truncate apex in profile (Fig. 12C, D, insets). Each EP is formed by the modification of a single cell of the stratum corneum. The main differences between nuptial pads with truncate NPEPs and flat NPEPs are the shapes of the cells of the stratum corneum and stratum granulosum. In truncate NPEPs, the stratum granulosum has pear-shaped cells and the EP in the stratum corneum is formed by a single cell with a columnar projection that is noticeably higher than wide (Fig. 12E, G). In nuptial pads with flat NPEPs, the stratum granulosum has cuboidal cells, and the stratum corneum is formed by a single cell with a low projection that is noticeably wider than high (Fig. 12F, H). Most of the glands (SMGs) have a secretory portion formed by cylindrical cells with a large lumen, and their longest axes are usually parallel to the surface of the epidermis (Fig. 3I). These truncate or flat NPEPs cannot be confused with truncate papillae because each EP is formed by the modification of a single cell, whereas each truncate papilla is formed by multiple cells, conferring a larger size (~5–20 vs. 44–68 µm diameter, respectively). Figure 12. View largeDownload slide External morphology, scanning electron micrographs and light micrographs of nuptial pads with non-papillary epidermal projections (NPEPs). The NPEPs are distinguishable from papillary epidermal projections by the smaller size and lower number of cells involved in each epidermal projection. A, C, E, G, Pseudis platensis (MACN 40926). A, light-coloured nuptial pad. C, scanning electron micrograph and inset. Truncate NPEPs with detail in dorsal view in inset. E, cross-section of truncate NPEPs. G, higher magnification of E. Note the shape of the cells of the stratum corneum (black arrowheads) and the pear shape of the cells of the stratum granulosum (digitally delimited). B, D, F, H, Rhinophrynus dorsalis (USNM 219809). B, light-coloured nuptial pad. D, scanning electron micrograph and inset. Flat NPEPs. Detail of the NPEP in oblique view in inset. F, cross-section of flat NPEPs. H, higher magnification of F, showing flat NPEPs. Note the cuboidal cells in the stratum granulosum and the shape of the cells in the stratum corneum (black arrowheads). Histological staining: modified Masson’s trichrome in E–H. Scale bars: 1 mm in A, B; 500 µm in C; 100 µm in D; 50 µm in E, F; 10 µm in G, H; 20 µm in insets in C, D. Figure 12. View largeDownload slide External morphology, scanning electron micrographs and light micrographs of nuptial pads with non-papillary epidermal projections (NPEPs). The NPEPs are distinguishable from papillary epidermal projections by the smaller size and lower number of cells involved in each epidermal projection. A, C, E, G, Pseudis platensis (MACN 40926). A, light-coloured nuptial pad. C, scanning electron micrograph and inset. Truncate NPEPs with detail in dorsal view in inset. E, cross-section of truncate NPEPs. G, higher magnification of E. Note the shape of the cells of the stratum corneum (black arrowheads) and the pear shape of the cells of the stratum granulosum (digitally delimited). B, D, F, H, Rhinophrynus dorsalis (USNM 219809). B, light-coloured nuptial pad. D, scanning electron micrograph and inset. Flat NPEPs. Detail of the NPEP in oblique view in inset. F, cross-section of flat NPEPs. H, higher magnification of F, showing flat NPEPs. Note the cuboidal cells in the stratum granulosum and the shape of the cells in the stratum corneum (black arrowheads). Histological staining: modified Masson’s trichrome in E–H. Scale bars: 1 mm in A, B; 500 µm in C; 100 µm in D; 50 µm in E, F; 10 µm in G, H; 20 µm in insets in C, D. Smooth nuptial pads This nuptial pad is a thickened, light-coloured glandular area evident by transparency with magnification (Figs 1C, 13A). This is usually evident with SEM, when the nuptial pad is easily distinguishable from adjacent skin by its thickness (Fig. 13B) attributable to the increase of the stratum spongiosum of the dermis, associated with hypertrophy of SMGs (Fig. 13C). However, in a few cases the nuptial pad is hardly distinguishable through SEM, even though the glandular area is evident by transparency (Fig. 1F). There are no EPs, so modifications are also restricted to the dermis (Fig. 13D). In a few species, the acini are difficult to see by transparency owing to the presence of epidermal pigmentation, although they are evident by the thickening of the skin [e.g. Conraua crassipes (Conrauidae)]. In cases of doubt, a superficial dissection of the skin might reveal the underlying glandular tissue. Smooth nuptial pads occur in 22.9% of the species that we studied with histology and/or SEM [e.g. Philautus schmackeri (Rhacophoridae), Phyllodytes luteolus (Hylidae), Pseudophryne bibroni]. A schematic representation of all the nuptial pads described above is provided in Figure 14. Figure 13. View largeDownload slide Smooth nuptial pads. A–C, Phyllodytes luteolus (A, B, CFBH 890; C, CFBH 14924). A, light-coloured nuptial pad. Thickened whitish glandular area distinguishable from adjacent skin (white arrowheads). B, scanning electron micrograph. Note the limits of the nuptial pad (white arrowheads) and the presence of pores. C, cross-section showing the thickening of the stratum spongiosum of the dermis and the presence of specialized mucous glands. Note that there is no increase in layers of epidermis. D, Dendropsophus nanus (MACN 46384); cross-section of nuptial pad. Although there is a thickening of the stratum spongiosum, there is no protrusion with respect to adjacent skin (see Fig. 1C, F). Histological staining: modified Masson’s trichrome in C, D. Scale bars: 500 µm in A; 100 µm in B–D. Figure 13. View largeDownload slide Smooth nuptial pads. A–C, Phyllodytes luteolus (A, B, CFBH 890; C, CFBH 14924). A, light-coloured nuptial pad. Thickened whitish glandular area distinguishable from adjacent skin (white arrowheads). B, scanning electron micrograph. Note the limits of the nuptial pad (white arrowheads) and the presence of pores. C, cross-section showing the thickening of the stratum spongiosum of the dermis and the presence of specialized mucous glands. Note that there is no increase in layers of epidermis. D, Dendropsophus nanus (MACN 46384); cross-section of nuptial pad. Although there is a thickening of the stratum spongiosum, there is no protrusion with respect to adjacent skin (see Fig. 1C, F). Histological staining: modified Masson’s trichrome in C, D. Scale bars: 500 µm in A; 100 µm in B–D. Figure 14. View largeDownload slide Schematic representation of nuptial pad diversity. A–J, papillary epidermal projections (PEPs, or papillae). A, truncate papillae. B, C, rounded papillae. Ornamentation absent (B) or present (C). D, irregular papillae. E–H, cone-shaped papillae. Ornamentation absent (E, G) or present (F, H); main axis straight (E, F) or leaned (G, H). I, spicule-shaped papillae. J, spine-shaped papillae. K, L, non-papillary epidermal projections (NPEPs). K, truncate NPEPs. L, flat NPEPs. M, N, smooth nuptial pads, without epidermal projections (EPs). M, protruded with respect to adjacent skin. N, not protruded with respect to adjacent skin. Structures were schematized on the basis of average values. See text for ranges of measurements within each papillary morphology. Figure 14. View largeDownload slide Schematic representation of nuptial pad diversity. A–J, papillary epidermal projections (PEPs, or papillae). A, truncate papillae. B, C, rounded papillae. Ornamentation absent (B) or present (C). D, irregular papillae. E–H, cone-shaped papillae. Ornamentation absent (E, G) or present (F, H); main axis straight (E, F) or leaned (G, H). I, spicule-shaped papillae. J, spine-shaped papillae. K, L, non-papillary epidermal projections (NPEPs). K, truncate NPEPs. L, flat NPEPs. M, N, smooth nuptial pads, without epidermal projections (EPs). M, protruded with respect to adjacent skin. N, not protruded with respect to adjacent skin. Structures were schematized on the basis of average values. See text for ranges of measurements within each papillary morphology. Morphology and position of glandular pores in nuptial pads The visualization of pores in nuptial pads is possible only through SEM, although in some cases, especially when papillae are very dense, pores are hardly visible (Luna et al., 2012). When visible, it is possible to analyse the density, location and morphology of the pores. We found pores occurring in the spaces between the papillae (Fig. 15A, D), on the apex of each papilla [the most uncommon case, seen only in Dendropsophus molitor and Pristimantis cruentus (Craugastoridae); Fig. 15G], on the lower margin of each spine-shaped papilla (one to three per papilla; Fig. 15J) and between the limits of epidermal cells (in the smooth nuptial pads; Fig. 15M). Pores may be surrounded by cornified epithelial cells, which delimit a rim in some species (compare Fig. 15B, K with E, H, N). Histologically, the epidermal cells of the stratum corneum protrude from the rest of the skin and form the rim (compare Fig. 15C, L with F, I, O). Figure 15. View largeDownload slide Glandular pores in nuptial pads. A, B, Exerodonta catracha (USNM 514184). A, scanning electron micrograph, showing a pore (arrowhead) between papillae. B, detail of the pore, showing the epidermal rim. C, Agalychnis moreletii (CENAI 7821); cross-section of a nuptial pad showing a pore. Note the epithelial cells of the stratum corneum protruding and forming the rim. D–F, Duellmanohyla soralia (USNM 515414). D, scanning electron micrograph, showing pores (arrowheads) between papillae. E, detail of a pore without epidermal rim. F, cross-section of a nuptial pad, showing a pore. The epidermal cells of the stratum corneum do not form a rim. G–I, Dendropsophus molitor (ICN 33631). G, scanning electron micrograph, showing pores (arrowheads) opening on the top of each papilla. H, detail of a single pore. I, cross-section of a nuptial pad showing a pore. Note that the duct is in the middle of the papilla. J–L, Ptychohyla hypomykter (USNM 343566). J, scanning electron micrograph, showing pores (arrowheads) near the base of each papilla. K, detail of a single pore. Notice the epidermal rim, less conspicuous than in B. L, cross-section of a nuptial pad, showing a pore. Note the epithelial cells of the stratum corneum protruding and forming the rim. M–O, Hyla cinerea (MACN 42126). M, scanning electron micrograph, showing pores (arrowheads) between the limits of the epidermal cells in a smooth nuptial pad. N, detail of a pore. O, cross-section of a nuptial pad showing a pore. Histological staining: modified Masson’s trichrome in C, F, I, L, O. Scale bars: 20 µm in A, C, D, F, G, I, J, L, M, O; 5 µm in B, E, H, K, N. Figure 15. View largeDownload slide Glandular pores in nuptial pads. A, B, Exerodonta catracha (USNM 514184). A, scanning electron micrograph, showing a pore (arrowhead) between papillae. B, detail of the pore, showing the epidermal rim. C, Agalychnis moreletii (CENAI 7821); cross-section of a nuptial pad showing a pore. Note the epithelial cells of the stratum corneum protruding and forming the rim. D–F, Duellmanohyla soralia (USNM 515414). D, scanning electron micrograph, showing pores (arrowheads) between papillae. E, detail of a pore without epidermal rim. F, cross-section of a nuptial pad, showing a pore. The epidermal cells of the stratum corneum do not form a rim. G–I, Dendropsophus molitor (ICN 33631). G, scanning electron micrograph, showing pores (arrowheads) opening on the top of each papilla. H, detail of a single pore. I, cross-section of a nuptial pad showing a pore. Note that the duct is in the middle of the papilla. J–L, Ptychohyla hypomykter (USNM 343566). J, scanning electron micrograph, showing pores (arrowheads) near the base of each papilla. K, detail of a single pore. Notice the epidermal rim, less conspicuous than in B. L, cross-section of a nuptial pad, showing a pore. Note the epithelial cells of the stratum corneum protruding and forming the rim. M–O, Hyla cinerea (MACN 42126). M, scanning electron micrograph, showing pores (arrowheads) between the limits of the epidermal cells in a smooth nuptial pad. N, detail of a pore. O, cross-section of a nuptial pad showing a pore. Histological staining: modified Masson’s trichrome in C, F, I, L, O. Scale bars: 20 µm in A, C, D, F, G, I, J, L, M, O; 5 µm in B, E, H, K, N. Nuptial pad coloration The coloration of nuptial pads depends on three factors: (1) the occurrence of coloration in the stratum corneum (if present, different intensities of light brown to black); (2) the colour of the glandular acini, which is visible by transparency through the epidermis and through the non-glandular matrix of the dermis when the stratum corneum is not coloured (different intensities of white or beige); or, much more rarely, (3) melanophores in the limit between the stratum spongiosum and the epidermis, which are also visible by transparency of the epidermis when the stratum corneum is not coloured. Granules of melanin are also found in the epidermal strata (stratum spinosum and stratum granulosum) of a few species. The different intensities of colour observed among species could not only be attributable to different stages of the hormonal cycles or the age of the specimens when preserved, but also could be related to variation in the quality of preservation/conservation or be artefacts of fixation. We found it more practical to distinguish dark- and light-coloured nuptial pads, where ‘dark-coloured’ includes all tones of brown and black (54.9% of all analysed species) and ‘light-coloured’ includes beige and whitish pads (45.1% of all analysed species). Comparison of nuptial pads with other keratinized structures Lower jaw spines Lower jaw spines and nuptial pad morphology were compared in Discoglossus jeanneae, Discoglossus pictus (Discoglossidae) and Platymantis dorsalis (Ceratobatrachidae); given that Petropedetes palmipes (Petropedetidae) lacks nuptial pads, we studied the skin of finger II instead (Fig. 16A, B). The lower jaw spines in Discoglossus and Platymantis show similar proportions of OMGs and SMGs (Fig. 16C, E, G). In the cases of D. jeanneae and P. palmipes, few OSGs were detected (Fig. 16C). The SMGs are PAS+ and AB−. The lower jaw spines of both Petropedetes palmipes and Platymantis dorsalis are formed by epidermal projections without a dermal core (Fig. 16E, inset). The cells of the stratum granulosum and stratum spinosum of the spines are cuboidal and turgid (Fig. 16E inset). In Discoglossus, an evagination of the dermis and epidermis forms the papillary structure (Fig. 16C). The nuptial pad of D. pictus has cone-shaped papillae, whereas that of Platymantis dorsalis has few and scattered EPs. The former has only SMGs, whereas the latter additionally shows OMGs (Fig. 16D, H, respectively). The skin of finger II in Petropedetes palmipes shows only OMGs (Fig. 16F). Figure 16. View largeDownload slide Comparison of lower jaw spines and nuptial pads. A, B, Discoglossus pictus (USNM 165888). A, external morphology of lower jaw. Note the dark spines. B, nuptial pads with cone-shaped papillary epidermal projections. C, Discoglossus jeanneae (USNM 193588); cross-section of skin in the lower jaw. Note the presence of scattered OSGs, OMGs and SMGs. The dark spines have a papillary structure (black arrowheads). D, Discoglossus pictus (USNM 165888); cross-section of the nuptial pad. Note the presence of SMGs only. E, F, Petropedetes palmipes (USNM 571344). E, cross-section of the skin in the lower jaw. Note the occurrence of OMGs and SMGs. Inset shows that the dark spines lack a dermal core (arrows) and are remarkable in the number and turgidity of the epithelial cells (white arrowheads). F, cross-section of skin of finger II, where a nuptial pad is absent; only OMGs are present. G, H, Platymantis dorsalis (USNM 508521). G, cross-section of skin in the lower jaw. Note the presence of OMGs and SMGs. H, cross-section of nuptial pad. Note the presence of SMGs, and some OMGs. Histological staining: periodic acid–Schiff in C, E, G; modified Masson’s trichrome in D, F, H, inset in E. Abbreviations: OMG, ordinary mucous gland; OSG, ordinary serous gland; SMG, specialized mucous gland. Scale bars: 2 mm in A, B; 50 µm in C–H; 25 µm in inset in E. Figure 16. View largeDownload slide Comparison of lower jaw spines and nuptial pads. A, B, Discoglossus pictus (USNM 165888). A, external morphology of lower jaw. Note the dark spines. B, nuptial pads with cone-shaped papillary epidermal projections. C, Discoglossus jeanneae (USNM 193588); cross-section of skin in the lower jaw. Note the presence of scattered OSGs, OMGs and SMGs. The dark spines have a papillary structure (black arrowheads). D, Discoglossus pictus (USNM 165888); cross-section of the nuptial pad. Note the presence of SMGs only. E, F, Petropedetes palmipes (USNM 571344). E, cross-section of the skin in the lower jaw. Note the occurrence of OMGs and SMGs. Inset shows that the dark spines lack a dermal core (arrows) and are remarkable in the number and turgidity of the epithelial cells (white arrowheads). F, cross-section of skin of finger II, where a nuptial pad is absent; only OMGs are present. G, H, Platymantis dorsalis (USNM 508521). G, cross-section of skin in the lower jaw. Note the presence of OMGs and SMGs. H, cross-section of nuptial pad. Note the presence of SMGs, and some OMGs. Histological staining: periodic acid–Schiff in C, E, G; modified Masson’s trichrome in D, F, H, inset in E. Abbreviations: OMG, ordinary mucous gland; OSG, ordinary serous gland; SMG, specialized mucous gland. Scale bars: 2 mm in A, B; 50 µm in C–H; 25 µm in inset in E. Pectoral spines Pectoral spines and nuptial pads were compared in Alsodes gargola and Telmatobius rubigo, and studied further in Leptodactylus rhodonotus, a species without a nuptial pad (see next section). The primary difference in the arrangement of pectoral spines in these species is that T. rubigo has spines scattered in the pectoral region (Fig. 17A), whereas in Alsodes there is a discrete pectoral patch (Fig. 17D). Histological study revealed that the structure of spines and adjacent pectoral skin is different from the nuptial pad in T. rubigo. The nuptial pad has cone-shaped papillae, with particularly hypertrophied SMGs (Fig. 17B). The pectoral spines have a papillate structure, and the dermis has a few OMGs, with cubic cells and abundant collagen fibres (Fig. 17C). In A. gargola, the nuptial pad and the pectoral spines have the same structure; spine-shaped papillae with few SMGs, abundant collagen fibres (Fig. 17F, G) and extensive vascularization. The pectoral spines of L. rhodonotus are compact, cornified spine clusters (Fig. 17H); they have a papillary structure but differ from spine-shaped papillae in that the stratum corneum has multiple cellular layers. The dermis has few SMGs, is rich in collagenous fibres and is extensively vascularized (Fig. 17I, J). Figure 17. View largeDownload slide Comparison of pectoral spines and nuptial pads. A–C, Telmatobius rubigo (MACN 41687). A, external morphology of pectoral spines and nuptial pads. Note the difference in density of pectoral spines (white arrowheads) and the cone-shaped papillary epidermal projections (PEPs or papillae) in the nuptial pad (white arrows). B, cross-section of nuptial pad with cone-shaped papillae and hypertrophied SMGs. C, cross-section of pectoral spines showing ordinary mucous glands and a few SMGs. Note the density of collagen fibres. D–G, Alsodes gargola (MACN 41418). D, external morphology of pectoral spines. Note the presence of a delimited area of spines (patches). E, external morphology of nuptial pad with spine-shapes PEPs. F, cross-section of nuptial pad. Note the presence of few, well-developed SMGs and dense collagen fibres. G, cross-section of pectoral spines. Spine-shaped PEPs with few, well-developed SMGs and dense collagen fibres. Note the size of the capillaries (black arrows). H–J, Leptodactylus rhodonothus (USNM 538215). H, external morphology of cornified pectoral spines. Note their compact structure and dark, continuous coloration (white arrowheads). I, cross-section of pectoral spines. Note the few SMGs, the great density of collagen fibres and the thickness of the stratum corneum. J, detail of the stratum corneum showing multiple cornified layers (black arrowheads). Histological staining: modified Masson’s trichrome in B, C, F, G; periodic acid–Schiff in I, J. Abbreviation: SMG, specialized mucous gland. Scale bars: 2 mm in A, D, E, H; 100 µm in B, C, F, G; 500 µm in I; 25 µm in J. Figure 17. View largeDownload slide Comparison of pectoral spines and nuptial pads. A–C, Telmatobius rubigo (MACN 41687). A, external morphology of pectoral spines and nuptial pads. Note the difference in density of pectoral spines (white arrowheads) and the cone-shaped papillary epidermal projections (PEPs or papillae) in the nuptial pad (white arrows). B, cross-section of nuptial pad with cone-shaped papillae and hypertrophied SMGs. C, cross-section of pectoral spines showing ordinary mucous glands and a few SMGs. Note the density of collagen fibres. D–G, Alsodes gargola (MACN 41418). D, external morphology of pectoral spines. Note the presence of a delimited area of spines (patches). E, external morphology of nuptial pad with spine-shapes PEPs. F, cross-section of nuptial pad. Note the presence of few, well-developed SMGs and dense collagen fibres. G, cross-section of pectoral spines. Spine-shaped PEPs with few, well-developed SMGs and dense collagen fibres. Note the size of the capillaries (black arrows). H–J, Leptodactylus rhodonothus (USNM 538215). H, external morphology of cornified pectoral spines. Note their compact structure and dark, continuous coloration (white arrowheads). I, cross-section of pectoral spines. Note the few SMGs, the great density of collagen fibres and the thickness of the stratum corneum. J, detail of the stratum corneum showing multiple cornified layers (black arrowheads). Histological staining: modified Masson’s trichrome in B, C, F, G; periodic acid–Schiff in I, J. Abbreviation: SMG, specialized mucous gland. Scale bars: 2 mm in A, D, E, H; 100 µm in B, C, F, G; 500 µm in I; 25 µm in J. Thumb spines in Heleioporus and Leptodactylus The ‘thumb spines’ of the genera Heleioporus and Leptodacylus are dark-coloured, cornified caps that cover a skeletal core and project from metacarpal II, the distal element of the prepollex, or both. We studied these spines on finger II in Leptodactylus chaquensis, Leptodactylus leptodactyloides (Fig. 18A) and Heleioporus albopunctatus (Limnodynastidae). The stratum corneum of the spines has multiple, non-continuous cornified layers (Fig. 18B, C); only the most internal layer is continuous with the epithelium of the adjacent skin. In these species, the cornified cap is supported internally by a skeletal element (bone or cartilage). The stratum compactum of the dermis is highly vascularized and continuous with the perichondrium (or periosteum, when ossification is present). The dermal component of the spines lacks SDSGs; instead, multiple OMGs are found in the adjacent skin of the thumb spine (Fig. 18D). As there are no epidermal or dermal components characteristic of nuptial pads, we do not consider these structures to be homologous. Figure 18. View largeDownload slide ‘Thumb spines’. A–C, Leptodactylus leptodactyloides (USNM 568063). A, external morphology of ‘thumb spines’. B, scanning electron micrograph, showing detail of a spine. C, magnification of B. The stratum corneum is formed by multiple cornified layers (white arrowheads). D, Leptodactylus chaquensis (MACN 16244); cross-section of ‘thumb spine’. No specialized mucous glands occur in the dermis. Note the great density of collagen fibres and the dense vascularization (black arrows). Histological staining: modified Masson’s trichrome in D. Abbreviation: B, bony core from the metacarpal. Scale bar: 2 mm in A; 100 µm in B, D; 50 µm in C. Figure 18. View largeDownload slide ‘Thumb spines’. A–C, Leptodactylus leptodactyloides (USNM 568063). A, external morphology of ‘thumb spines’. B, scanning electron micrograph, showing detail of a spine. C, magnification of B. The stratum corneum is formed by multiple cornified layers (white arrowheads). D, Leptodactylus chaquensis (MACN 16244); cross-section of ‘thumb spine’. No specialized mucous glands occur in the dermis. Note the great density of collagen fibres and the dense vascularization (black arrows). Histological staining: modified Masson’s trichrome in D. Abbreviation: B, bony core from the metacarpal. Scale bar: 2 mm in A; 100 µm in B, D; 50 µm in C. Distribution of nuptial pads When nuptial pads are present, they always cover a portion of finger II (Fig. 19A, B); their extent varies interspecifically and, sometimes, intraspecifically. Nuptial pads can also occur on fingers III, IV and V (Fig. 19C–F) as a continuous structure that covers the medial margin of the fingers and coalesce dorsally on the hand, or as discontinuous segments that cover each finger separately. Figure 19. View largeDownload slide Distribution of nuptial pads. A, Espadarana audax (USNM 286621); nuptial pad on finger II. B, Rana japonica (USNM 245170); extremely thickened nuptial pad. Note the thickness of the nuptial pad area on finger II compared with adjacent skin (white arrowhead). C, Lechriodus melanopygia (USNM 521750); dark-coloured nuptial pads covering fingers II and III. D, Scinax nasicus (MACN 39438); light-coloured nuptial pads covering fingers II and III. Note the nuptial pad on finger II and some individual acini extending distally on fingers II and III (white arrowheads). E, Spea hammondii (USNM 225295); dark-coloured nuptial pad on fingers II, III and IV. F, Hemisus marmoratus (FMNH 80626); light-coloured nuptial pad as a continuous structure, covering the dorsal and medial surfaces of fingers II, III and IV and extending to the wrist (white arrowheads). G, Stefania evansi (USNM 531495); nuptial pad with a proximal segment and an adjacent distal segment. H, Stefania evansi (USNM 531498). In this specimen, the two segments coalesce in one continuous structure, but there is a notch (black arrowhead) indicating the original gap between them. I, Bombina variegata (USNM 137123); nuptial pads on fingers II, III and IV, inner metacarpal tubercle and inner margin of arm. J, Alsodes gargola (MACN 41418); nuptial pads in the pectoral area, as indicated by our results (see main text). K, Xenopus epitropicalis (USNM 573445); nuptial pad covering the palm and ventral surface of the arm and forearm. Abbreviations: DS, distal segment; PS, proximal segment. Scale bars: 1 mm. Figure 19. View largeDownload slide Distribution of nuptial pads. A, Espadarana audax (USNM 286621); nuptial pad on finger II. B, Rana japonica (USNM 245170); extremely thickened nuptial pad. Note the thickness of the nuptial pad area on finger II compared with adjacent skin (white arrowhead). C, Lechriodus melanopygia (USNM 521750); dark-coloured nuptial pads covering fingers II and III. D, Scinax nasicus (MACN 39438); light-coloured nuptial pads covering fingers II and III. Note the nuptial pad on finger II and some individual acini extending distally on fingers II and III (white arrowheads). E, Spea hammondii (USNM 225295); dark-coloured nuptial pad on fingers II, III and IV. F, Hemisus marmoratus (FMNH 80626); light-coloured nuptial pad as a continuous structure, covering the dorsal and medial surfaces of fingers II, III and IV and extending to the wrist (white arrowheads). G, Stefania evansi (USNM 531495); nuptial pad with a proximal segment and an adjacent distal segment. H, Stefania evansi (USNM 531498). In this specimen, the two segments coalesce in one continuous structure, but there is a notch (black arrowhead) indicating the original gap between them. I, Bombina variegata (USNM 137123); nuptial pads on fingers II, III and IV, inner metacarpal tubercle and inner margin of arm. J, Alsodes gargola (MACN 41418); nuptial pads in the pectoral area, as indicated by our results (see main text). K, Xenopus epitropicalis (USNM 573445); nuptial pad covering the palm and ventral surface of the arm and forearm. Abbreviations: DS, distal segment; PS, proximal segment. Scale bars: 1 mm. We recognize a proximal segment that covers the dorsal and medial surfaces and sometimes reaches the prepollical elements; ventrally, they reach or cover the inner metacarpal tubercle. This segment sometimes occurs solely on finger II, or sometimes is together with a second, distal segment. This distal segment is adjacent to and sometimes even coalesces with the proximal segment, but in almost all cases there is a distinct separation of the two (Fig. 19G, H). Both the proximal and the distal segments show extreme diversity on their extensions. In some species (Fig. 19I), the nuptial pad is restricted to the inner metacarpal tubercle or enlarged prepollex and is not continuous with the segment that covers the proximal surface of finger II. The homologies of this and the proximal segment remain to be studied. Structures that are histologically similar to the nuptial pad may also be present on the lower arm (Fig. 19I), upper arm and pectoral region (Fig. 19J). An uncommon spatial pattern of a nuptial pad structure occurs in species of Xenopus, where the nuptial pad covers the entire palm and extends onto the upper arm (Fig. 19K). DISCUSSION Dermal component of the nuptial pad Three main types of glands have been described for the skin of amphibians: mucous glands, serous glands and mixed glands (e.g. Noble & Noble, 1944; Toledo & Jared, 1995). Mucous glands are small, distributed among the entire body and characterized by the presence of neutral carbohydrates (Noble, 1931; Quay, 1972; Brizzi et al., 2003a). The secretory portion of these glands presents a single layer of cuboidal cells and a large lumen (Toledo & Jared, 1995). Mucous glands are usually involved in cutaneous respiration and homeostasis (Toledo & Jared, 1995; Brizzi et al., 2003a). Serous glands (also termed granular glands) are larger than mucous glands and characterized by their proteinaceous content. Their secretory cells form a true syncytium, with multiple secretory granules (Neuwirth et al., 1979). These glands store diverse components, including compounds involved in defense (Toledo & Jared, 1995). Mixed glands are distributed along the body of most caudates and are formed by a large granular portion and a small mucous portion (for reviews, see Dawson, 1920; Delfino, Brizzi & Calloni, 1986). Apart from these generalized types of glands, others have been described, including lipid glands, seromucous glands and different types of granular glands (e.g. Blaylock, Ruibal & Platt-Aloia, 1976; Mills & Prum, 1984; Delfino, 1998; Brizzi et al., 2002). Thomas et al. (1993) showed that among 12 anuran species some glands have shared histochemical and structural properties that make them different from the serous and mucous glands of the rest of the skin. These findings supported the recognition of a different type of gland that was independent of its location and function; these are called SDSGs. The SDSGs in anurans include abdominal glands, femoral glands, humeral glands, pectoral glands, nuptial glands, post-axillary glands and a few specialized glands without a specifically defined region on the skin (Thomas et al., 1993; Houck & Sever, 1994; Brizzi et al., 2003a). Brizzi et al. (2002) introduced the term ‘specialized mucous glands’ as a type of SDSGs. These glands are larger than the OMGs and have tall columnar cells in their secretory portion. Brunetti et al. (2012, 2015) introduced the term ‘specialized serous glands’ as a new type of SDSG (mentioned but not defined by Thomas et al., 1993; Vences et al., 2007; Gonçalves & de Brito-Gitirana, 2008). The SSGs have a syncytial internal secretory layer and a lumen filled with granules. The secretory products are embedded in a matrix of amorphous cellular components. The syncytium contains rounded nuclei and a thin myoepithelial layer that encircles the secretory portion (Brunetti et al., 2012). Ordinary serous glands and SSGs are syncytial and have similar sizes. The principal differences are attributed to their histochemical properties, size and appearance of secretory granules, and glandular outlets (Brunetti et al., 2012). Both SMGs and SSGs occur at least as mental and lateral glands of several anuran groups (Brunetti et al., 2015). The SDSGs secrete their product to the surface of the epidermal component of the nuptial pad through pores. Although known for ≥ 140 years from histological studies (e.g. Lataste, 1876), these glands were first noticed through SEM and commented on by Wray (2000) in his unpublished thesis on some species of Ptychohyla and by Luna et al. (2012) in some phyllomedusines. Luna et al. (2012) showed that histological differences exist between the outermost duct cells of the different types of pores in the nuptial pads. They also provided scanning electron micrographs that showed the same cornified structure on the ducts and in the stratum corneum of the papillae. Histochemical studies of the nuptial pad led to the characterization of the secretion produced by SDSGs during the breeding season (Thomas et al., 1993; Epstein & Blackburn, 1997). Specialized mucous glands contain neutral carbohydrates (PAS+ staining) and proteins as their major secretory product (Coomassie Blue+ and nynhidrine+ staining). Their negative responses to Sudan Black (SB) and AB stains indicate the lack of lipids and acidic or sulphated mucosubstances, respectively. However, a positive reaction for AB was reported for the SMGs of the dorsal skin of Rana dalmatina (Brizzi et al., 2002) and in the SMGs of nuptial pads of Pelophylax kl. esculentus and P. perezi (Brizzi et al., 2003b). The SSGs contain acidophilic granules (PAS−, AB− and PAAB−) and proteins (Coomassie Blue+). Thomas et al. (1993) and Brunetti et al. (2012) reported positive SB staining for lipids in SSGs. Our findings are congruent with previous results (Thomas et al., 1993; Epstein & Blackburn, 1997; Luna et al., 2012). The principal differences found were the positive results for AB staining in several species (see Supporting Information, Table S2), indicating that SMGs can also contain acidic mucosubstances as part of their secretory products (as shown in R. dalmatina by Brizzi et al., 2002). We also found that 86% of the studied species had SMGs as the only SDSGs in the dermal component of the nuptial pad. In only one case (X. epitropicalis) were SSGs found, but they co-occurred with a higher density of SMGs. Fujikura et al. (1988) reported two different serous glands in the pad of X. laevis. It is very likely that one of these (their ‘small granulated glands’) corresponds to an SSG. Our study shows the presence of OMGs and OSGs together with SMGs in the nuptial pad of some species (Figs 2C, 16H; see Supporting Information, Table S2). Given that most nuptial pads have exclusively SMGs, it will be interesting to find out whether these OMGs and OSGs play the same functional role in the nuptial pad as in other parts of the body. Our definition of nuptial pads relies on the presence of SMGs in the dermis, whereas the epidermis may or may not be modified with respect to the adjacent skin. The reason for this is that SMGs are the only structures in common among all nuptial pads that we have studied, associated with the observation that the epidermal component, if present in other areas of the body, may not be associated with SMGs, nor is it necessarily a sexually dimorphic character. In addition to distinguishing SMGs, Kaptan & Murathanoğlu (2008) were the first to address the presence of mixed glands and OSGs in the nuptial pad of Pelophylax ridibundus. They reported that these mixed glands had characteristics of both serous glands and SMGs and were similar to those described in caudates (Delfino et al., 1986). Although they did not refer to them, structurally these glands seem different from those SDSGs that have been called ‘seromucous’ glands in the dorsa of some ranids (Mills & Prum, 1984; Thomas & Licht, 1993). Kaptan & Murathanoğlu (2008) also suggested that mixed glands are possibly an intermediate stage in the transformation of one type into another type of gland, although they did not elaborate on this. To date, no author has suggested that OSGs might be ontogenetic precursors to SMGs. Regardless, these results are curious in that species of Pelophylax have been the object of several studies on nuptial pad structure, none of which ever referred to the occurrence of mixed glands or OSGs (Ecker & Wiedersheim, 1882; Gaupp, 1904; Kändler, 1924; Aron, 1926; Brizzi et al., 2002, 2003b), so perhaps they are restricted to a few species. No mixed glands were found in the nuptial pads of any of the species surveyed in the present study, of which Pelophylax nigromaculatus is phylogenetically the most closely related species to P. ridibundus. The results reported by Willaert et al. (2013) using micro-computed X-ray tomography for reconstructing the nuptial pad of Rana temporaria are also noteworthy. Using this technique, they showed two types of dermal glands (Willaert et al., 2013: fig. 1E); the smaller ones seem to correspond to SMGs, whereas the larger ones seem to represent OSGs. These observations differ from previous histological descriptions of the pad in R. temporaria (e.g. Aron, 1926). Epidermal component of the nuptial pad Coloration and cornification There are two possible explanations for the occurrence of coloration in nuptial pads. The first is the presence of melanin in the epidermis. Brizzi et al. (2003a) stated that the stratum corneum of the epidermis is filled with melanin that is dense at the peaks of the EPs and absent in the valleys separating them; no test for the presence of melanin was reported. Kyriakopoulou-Sklavounou et al. (2012) noted the presence of ‘remarkable amounts of melanin’ in the apices of the cone-shaped papillae of the nuptial pad of Bombina variegata (they included a light microscopy image but did not comment on any test for melanin). We did not observe melanin in the stratum corneum of the nuptial pads in any of the seven species from seven families that we tested (see Supporting Information, Table S3). These included nuptial pads with a dark- or light-coloured stratum corneum and varying levels of melanin among their stratum spinosum and stratum granulosum and in the dermis. The negative results of the hydrogen peroxide test on nuptial pads with dark coloration correspond to those reported by Maddin et al. (2009) for claws (see next page). Dermal melanophores and melanin within the stratum spinosum and stratum granulosum were bleached, whereas dark-coloured stratum corneum was not affected by hydrogen peroxide, indicating that melanin is not involved in the dark coloration of nuptial pads. We also found that the degree of dark coloration is independent of the number of layers of the stratum corneum, because most of the specimens studied posess a stratum corneum composed of a single layer, regardless of its coloration. The other explanation for coloration in nuptial pads is the degree of cornification of the stratum corneum. The ultrastructure of the nuptial pad of X. laevis was analysed by Komnick & Stockem (1970) and Dolder (1976). Komnick & Stockem (1970) emphasized the multicelullar formation of the spicule-shaped papillae; they used the term ‘multicellular epidermal claws’, because a barely noticeable dermal core evaginates with the epidermal cells (Komnick & Stockem, 1970: fig. 2d). Dolder (1976) discussed the progressive cornification (using the term ‘keratinization’; see next section) of the spicule-shaped PEPs of the stratum corneum. This monolayer has degenerated organelles that show a dense arrangement of interconnected keratin fibres; pyknotic nuclei (flattened) are sometimes detectable. However, no pigment cells or traces of pigmentation could be located. Dolder (1976) pointed out that the difference between the apical cells of the spicule-shaped PEPs and the cells located between them was attributable to the timing of cornification. The cells lying between the spicule-shaped PEPs cornified later than the apical cells; thus, individual differentiation of fibres formed a looser, less dense arrangement (Dolder, 1976: fig. 1). Dolder (1976) also pointed out that loosely arranged keratin fibres in the cells of the stratum granulosum contributed to the cornification process. Kurabuchi (1993) described the nuptial pads of Pelophylax porosus using TEM and revealed that the structure of the epidermis of its papillae was similar to that found in the spicule-shaped PEPs of X. laevis (Komnick & Stockem, 1970; Dolder, 1976). The stratum corneum consisted of a monolayer of electron-dense cells containing a few remnants of nuclei and cytoplasmic organelles, and most of the cytoplasm was full of filamentous dark material. The stratum granulosum showed dense granules and bundles of filaments terminating in dense plaques of desmosomes. A wide space between the stratum corneum and the stratum granulosum was observed, and these two strata were connected by long cytoplasmic processes projecting from the latter and joined by desmosomal adhesion. Rastogi et al. (1986) studied colour variation of the nuptial pad of Agalychnis dacnicolor (Hylidae) during a seasonal cycle; darker colours (brown to black) were visible when the specimens were in the breeding season, whereas a whitish coloration was visible outside the breeding season. This loss of coloration paralleled the regression of papillation of the nuptial pads. Studies of Scaphiopus couchii (Scaphiopodidae) have shown that androgens may affect the coloration of their nuptial pads (Harvey & Propper, 1997). These authors analysed groups of intact, sham-castrated and castrated males treated with exogenous testosterone or cholesterol and recorded the nuptial pad coloration using a qualitative score. In their studies, males that were treated with testosterone had darker nuptial pads than those treated with cholesterol; however, they did not differ from untreated males. If the degree of coloration of the nuptial pad is related to the concentration of testosterone, we would expect that the sensitivity of the nuptial pad to hormonal variation might differ among species. Relevant to the discussion on the origin of coloration in nuptial pads is the debate on the coloration of other structures in anurans that undergo keratinization during their development, such as larval jaw sheaths, labial teeth and claws (Altig, 2007; Maddin, Musat-Marcu & Reisz, 2007; Alibardi, 2010a, b). Luckenbill (1965) identified a few granules near the apices of the jaw sheaths of Lithobates pipiens (Ranidae) as melanin. However, this author did not present information regarding their composition. Altig (2007) argued that the reported granules did not correspond to melanin because they were of a different size and were in the wrong histological position. Maddin et al. (2009) performed a hydrogen peroxide test on histological preparations of the claw of X. laevis and reported that prolonged exposure to this compound did not affect the coloration of the claw. These results suggest that the presence of colour is not derived from melanin. Albino tadpoles are also a proof that pigmented cells are not involved in the formation of the mouthparts. It is known that this mutation eliminates the ability to produce melanocytes, but in all the albino cases reported so far, the keratinized jaw sheaths and labial teeths were still visible (e.g. Smith‐Gill, Richards & Nace, 1972; Altig, 2007; Eagleson et al., 2010). Accordingly, we conclude that the coloration of the jaw sheaths and claws of amphibians is not attributable to melanin. Albino adults of X. laevis also have dark nuptial pads, although melanin is absent (Hoperskaya, 1975). This dark coloration could be explained by differential keratinization in association with the deposition of KAPs over keratins (Altig, 2007; Maddin et al., 2007; Alibardi, 2010a). The PAAB-positive reactions obtained in our results indicate richness in sulphydryl and disulphide groups, both coincidental with the presence of keratin and KAPs in the stratum corneum of the nuptial pads. Transmission electron microscopic observations show pyknotic nuclei and electron-dense material (Fig. 20), suggesting the presence of KAPs in the upper layer of nuptial pads. The degree of darkness could be attributable to the accumulation of degenerated organelles and keratin fibres and to the concentration of KAPs in the upper layer. Although it could be argued that the different intensities of coloration of nuptial pads in different species are related to the degree of cornification of the stratum corneum (and indeed, dark-coloured nuptial pads frequently have a thicker stratum corneum than light-coloured nuptial pads), there is some evidence against it. For instance, Phyllomedusa azurea and Amolops chunganensis (Ranidae) have, respectively, a stratum corneum that is 4.0–10 and 3.3–8.6 μm thick, with the pad of the former being dark-coloured and that of the latter being light-coloured. Another example of this is found in some nuptial pads with spine-shaped papillae, such as those of Alsodes gargola. In this species, the nuptial pads show a different coloration between the EPs and the epidermis separating them; both have an equally thick stratum corneum (10.3–11.9 and 6.3–13.0 μm, respectively), but the EP is dark-coloured, whereas the epidermis between them is light-coloured (Fig. 17F, G). Although the basis of coloration of nuptial pads is still poorly understood, the above discussion provides evidence to reject chromatophores as the source of coloration in pads. Figure 20. View largeDownload slide Transmission electron micrograph of the epidermis of a papillary epidermal projection in Scinax perereca (MACN 43321). Note the pyknotic nuclei on the cell of the stratum corneum. Abbreviations: PN, pyknotic nuclei; SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bar: 2 µm. Figure 20. View largeDownload slide Transmission electron micrograph of the epidermis of a papillary epidermal projection in Scinax perereca (MACN 43321). Note the pyknotic nuclei on the cell of the stratum corneum. Abbreviations: PN, pyknotic nuclei; SC, stratum corneum; SG, stratum granulosum; SS, stratum spinosum. Scale bar: 2 µm. Although the stratum corneum is commonly described as a monolayer (see above), a few reports indicate that the stratum corneum is formed by three or four layers. This feature was seen in the cone-shaped papillae of Pelophylax nigromaculatus (Yang et al., 2004; as Rana chinensis), R. dalmatina (Kyriakopoulou-Sklavounou et al., 2012) and R. temporaria (Fox, 1986; all of these authors used different terminology, see Supporting Information, Table S1). Our study shows a multilayered stratum corneum in the nuptial pads of Argenteohyla siemersi (Hylidae), Ascaphus montanus, Spea bombifrons and X. tropicalis (Fig. 21A, B). A plausible reason to explain why we do not usually see more than a single layer in other species could be that some layers might be removed easily from the skin during the process of fixation or even during handling. We have noticed the loss of upper layers of the stratum corneum during the fixation process with Limnomedusa macroglossa (Alsodidae) in the field. These layers are easily detached, indicating that, if they occur, only the most internal layer is continuous with the stratum corneum of the adjacent skin. Figure 21. View largeDownload slide Cross-section of nuptial pads showing the stratum corneum. A, Spea bombifrons (MACN 39156). B, Xenopus tropicalis (USNM 571230). Note the stratum corneum formed by three and two cell layers, respectively (white arrowheads). Histological staining: periodic acid–Schiff in A; modified Masson’s trichrome in B. Scale bars: 50 µm. Figure 21. View largeDownload slide Cross-section of nuptial pads showing the stratum corneum. A, Spea bombifrons (MACN 39156). B, Xenopus tropicalis (USNM 571230). Note the stratum corneum formed by three and two cell layers, respectively (white arrowheads). Histological staining: periodic acid–Schiff in A; modified Masson’s trichrome in B. Scale bars: 50 µm. Other keratinized structures in anurans The study of the claw in X. laevis has shown that its cornified epidermis is composed of multiple layers of corneocytes, and only the deepest layer is continuous with the non-cornified epidermis; the rest are open-ended layers on the top of these (Maddin et al., 2007, 2009). The cytoplasm of the precorneous cells of the claw region (the stratum spinosum and stratum granulosum) is filled with mucous granules containing sulphur-rich proteins (Ceresa-Castellani, 1969) and a low concentration of histidine-rich proteins (HRPs) within an interkeratin matrix (Alibardi, 2003). The concentration of desmosomes on the periphery of the stratum spinosum is higher than that in the rest of the skin. The presence of HRPs in the epidermal strata is thought to be related to the degree of cornification (Maddin et al., 2007). These proteins are present in high concentrations in mammal epidermis, where they interact and stabilize keratin molecules, favouring their aggregation as keratohyalin (Resing & Dale, 1991). In amphibians, the normal terminal differentiation of the epidermis agrees with the pathological condition of mammals (parakeratosis), where the level of HRPs is low, keratohyalin scarce or absent, and the nuclei of keratinocytes are retained in the stratum corneum (Alibardi, 2003, 2010a). Claws and nuptial pads both have cysteine-rich proteins in their stratum granulosum and stratum corneum (Alibardi, 2010b; Luna et al., 2012; present study), and an increase of desmosomes in the epidermis, as seen in serial sections (Fig. 11E) and by TEM (data not shown). Differences between these structures lie in the number of layers of the stratum corneum and the absence of SDSGs in the dermal component of claws. The increase in density of desmosomes in the stratum spinosum in claws is thought to provide additional strength to prevent shedding during mechanical stress (Maddin et al., 2007). The same hypothesis was offered to account for the high concentration of desmosomes in lizard epidermis (Alexander & Parakkal, 1969). Thus, it seems likely that the high density of desmosomes in nuptial pads could also provide strength and prevent shedding. Histological and ultrastructural studies of mouthparts of larvae are scarce (e.g., Luckenbill, 1965; Kaung, 1975; Vera Candioti & Altig, 2010), and only a few studies address the process of keratinization. Alibardi (2010a) studied the jaw sheath of Rana dalmatina and found a thicker epithelium there than in the rest of the skin, owing to the multiple cornified layers on both the oral and the labial parts of the jaw sheath. Keratin-associated proteins are rapidly deposited as a dense and amorphous corneous material into the intermediate filaments of keratin and account for the electron-dense nature of this layer under TEM that is similar to the material seen in mammalian corneous tissues (Fraser, MacRae & Rogers, 1972; Alibardi, 2009). Alibardi (2010a) suggested that KAPs are also responsible for the hardness of the jaw sheath. Similar to the claws of adult amphibians, and different from their nuptial pads, larval jaw sheaths are composed of multiple keratinized layers of cells. Nuptial pads in females? Even though nuptial pads typically are SSCs of adult male anurans, in a few cases both sexes and/or even juveniles reportedly have nuptial pads. In all these cases, the nuptial pads have an epidermal component formed by dark-coloured spine-shaped papillae, or dark-coloured spicule-shaped papillae. We are aware of reports of nuptial pads in females of X. laevis (Kurabuchi & Inoue, 1981), Crossodactylus (Hylodidae; Lutz, 1931), Cycloramphus ohausi (Cycloramphidae; Heyer, 1983) and Insuetophrynus acarpicus (Rhinodermatidae; Diaz, Valencia & Sallaberry, 1983). In the case of Xenopus, Fujikura et al. (1988) studied the different types of glands present in X. laevis and reported that SDSGs were present only in the nuptial pads of males and not in females, whereas Van Wyk, Pool & Leslie (2003) reported the presence of only mucous and serous glands on the hand of female X. laevis. Therefore, what is found in females of Xenopus reportedly is not a nuptial pad. In Crossodactylus, spine-shaped papillae occur on finger II in both sexes (Lutz, 1931; Pimenta, Cruz & Caramaschi, 2014). Our histological study of these papillae in adults of both sexes of Crossodactylus schmidti and males of C. caramaschii, C. dispar, C. gaudichaudii, and C. werneri, indicates that no sexually dimorphic glands are associated with these structures in either sex (Fig. 22A, B). This poses an interesting problem relative to the origin of these spines and their presence in males and females. Perhaps these spines originally occurred in a nuptial pad that lost the dermal component in the common ancestor of Crossodactylus; this possibility, and the hormonal regulation of spine development in this genus, require further research. Males of I. acarpicus have spines on fingers II and III, on the inner metacarpal tubercle and in the pectoral region, whereas females have spines in the same areas but are less developed on finger III and in the pectoral region (Diaz et al., 1983). In the case of Cycloramphus ohausi, Heyer (1983) referred to ‘black cornified thumb spines’ in both sexes. We did not have access to well-preserved specimens of these species; therefore, the nature of their nuptial pads remains an open question. Figure 22. View largeDownload slide Papillae in Crossodactylus species. A, B, Crossodactylus werneri (USNM 318150). A, scanning electron micrograph of the spine-shaped papillae on finger II. B, cross-section of finger II, showing a single ordinary mucous gland (arrow), no specialized mucous glands, and dense collagen fibres. Histological staining: modified Masson’s trichrome in B. Scale bars: 500 µm in A; 100 µm in B. Figure 22. View largeDownload slide Papillae in Crossodactylus species. A, B, Crossodactylus werneri (USNM 318150). A, scanning electron micrograph of the spine-shaped papillae on finger II. B, cross-section of finger II, showing a single ordinary mucous gland (arrow), no specialized mucous glands, and dense collagen fibres. Histological staining: modified Masson’s trichrome in B. Scale bars: 500 µm in A; 100 µm in B. Comparison with nuptial pads in caudates Nuptial pads have been described for some caudates (e.g. Forbes, Dent & Singhas, 1975), but comparative studies with frogs have not been done. Furthermore, the information on structure and taxonomic distribution in salamanders, including newts, is scarce and usually restricted to taxonomic descriptions (e.g. Liu, 1950). One of the better-studied species, Notophthalmus viridescens (Salamandridae), has dark-coloured nuptial pads on the hindlimbs of males, and similar structures covering the toe pads (Forbes et al., 1975). The most remarkable difference in these structures between caudates and frogs is the composition of the epidermal area. Nuptial pads in salamanders have a thicker epidermis (up to ten layers), and conical, non-papillary epidermal projections formed by single cells of the stratum corneum, which are angled at 45° (Forbes et al., 1975). Specialized mucous glands and the so-called mixed glands occur in the dermis (Dawson, 1920; Delfino et al., 1986; M.C.L., personal observation). A thorough understanding of nuptial pad structure, diversity and evolution in caudates is a prerequisite for discussing the origin and early evolution of nuptial pads in anurans. Hormonal control It is well established that androgens, especially testosterone, induce the development of nuptial pads, in addition to some other SSCs in males (e.g. Meisenheimer, 1912; Ponse, 1924; Aron, 1926; Houssay, Giusti & Lascano Gonzalez, 1929; Berk, 1939; Horié, 1939; Cei, 1948; Lofts, 1964; Lofts, Wellen & Benraad, 1972; Izzo et al, 1982; Emerson et al., 1999; for a review, see Di Fiore et al., 2005). Iwasawa & Kobayashi (1974) analysed the different effects of testosterone and estradiol on Pelophylax nigromaculatus. They showed that administering testosterone plus estradiol resulted in a slight synergistic effect on the development of the nuptial pad, compared with administering testosterone alone. They also showed that administering estradiol alone had no effect. The application of testosterone to individuals that were sexually immature resulted in the development of a nuptial pad, whereas in adults the response differed between the sexes, suggesting that sensitivity to testosterone is lost in adult females. Although estradiol had no effect on the development of the nuptial pad in the case above, Saidapur & Nadkarni (1975a) found that the administration of exogenous estradiol to Euphlyctis cyanophlyctis and Hoplobatrachus tigerinus (Dicroglossidae) produced a regression of the nuptial pad. The same result was reported in Duttaphrynus melanostictus (Bufonidae; Kanamadi & Saidapur, 1981) and in X. laevis (Van Wyk et al., 2003). Administering testosterone and its derivatives to castrated males and to males outside the breeding season can stimulate the development of nuptial pads with the same morphological features as those that occur in pads in normal breeding conditions (e.g. Kelley & Pfaff, 1976; Izzo et al., 1982; Lynch & Blackburn, 1995). However, exogenous testosterone did not restore the breeding season condition in P. kl. esculentus (Rastogi et al., 1976) and Agalychnis dacnicolor (Rastogi et al., 1986). These studies also showed that the interaction between temperature and sexual hormones is important (Rastogi, 1976; Rastogi et al., 1978, 1986). In Pelophylax kl. esculentus, the optimal temperature range for development of nuptial pads was between 10 and 20 °C. Exposing frogs to a lower (4 °C) or a higher temperature (28 °C) was ineffective during androgen administration, although the exposure to high temperature increased the plasma concentration of testosterone (Rastogi et al., 1976; Izzo et al., 1982). d’Istria et al. (1979) suggested a possible explanation for this lack of response when they reported the presence of androgen receptors in the nuptial pads of P. kl. esculentus during only the autumn–winter period, thereby revealing the importance of environmental conditions on the physiological responses of these species. In A. dacnicolor, there was a close parallel between the seasonal changes in plasma androgen concentrations and testis weight, spermatogenic activity, nuptial pad development and reproductive behaviour. However, there was no detectable response to exogenous testosterone in the development of nuptial pads in castrated males (Rastogi et al., 1986). A comparison of plasma concentrations of different androgens (androstenedione, testosterone and 5-dihydrotestosterone) indicated that no specific candidate was responsible for triggering the seasonal development of nuptial pads, and only the combination of all of them might produce the external features observed during the breeding season. Kaptan & Murathanoğlu (2008) studied the seasonal cycle of Pelophylax ridibundus and found that its nuptial pads exhibited development opposite to spermatogenetic activity. They also reported that changes in temperature affected their development. During November and December, the seasonal period for hibernation, they reported that papillae were less evident, although the glands were hypertrophied. This suggested that apart from testosterone, other factors that affect nuptial pad structure might exist. The presence of mitochondria-rich cells (MRCs) in the epidermis of nuptial pads was also suggested to have a role in the reorganization of epidermal changes of the nuptial pad (Kaptan & Bolkent, 2014). These authors also reported the lectin-binding affinity of the nuptial pad and showed a higher binding response in the upper layers of the epidermis, which reflected functional differences of the nuptial pad glycosylated components during seasonal changes. All these findings show that although testosterone is undoubtedly involved in the development of the nuptial pads, it might not be sufficient to trigger their seasonal development in some species. These reports suggest that other hormones or testosterone derivatives, in addition to temperature and photoperiod, might be involved in the development of nuptial pads during the breeding season. The hormonal influence on the development of nuptial pads has been considered cyclic or noncyclic (Inger & Greenberg, 1963; Saidapur, 1983). The cyclic pattern was commonly attributed to temperate zone species, with marked seasonal variation throughout the year (Lofts, 1964; Lofts et al., 1972; Rastogi, 1976). Development of the nuptial pad reaches a maximum during the breeding season (hypertrophied SDSGs and maximal development of the epidermal component if differentiated), which is followed by regression (regression of SDSGs; Glass & Rugh, 1944; Inger & Greenberg, 1956; Lofts, 1964; Lofts et al., 1972). A non-cyclic pattern was reported for anurans inhabiting tropical zones, with little seasonal variation in pad development (Inger & Greenberg, 1963; Saidapur & Nadkarni, 1975b; Saidapur, 1983). Papillae and SDSGs are present during the entire year, although papillae are reported to be smaller, suggesting that the hormone production never falls to such a low level that it critically affects development of the nuptial pad (Inger & Greenberg, 1963; Saidapur, 1983). Some studies indicate that tropical and subtropical species may also show cyclic variation in nuptial pad development and in other SSCs (e.g. Cei, 1948; Saidapur, 1983; Rastogi et al., 1986). The induction of nuptial pad development in ovariectomized females after the administration of androgens has been established experimentally (Horié, 1939; Blair, 1946; Iwasawa & Kobayashi, 1974; Kelley & Pfaff, 1976), and from the perspective of pad morphology and evolution, it deserves more attention. These pads in females have been described as having differentiated dermal and (when it occurs) epidermal components, as in males of model species studied experimentally [e.g. Anaxyrus fowleri (Bufonidae), Pelophylax nigromaculatus, Pseudacris triseriata (Hylidae), X. laevis]. These studies were mostly done from an endocrinological perspective, and the resulting nuptial pad production in females has not been studied in much detail, to the point that we are unaware of any study where histochemical differences between SDSGs and mucous glands have been reported for the hormone-induced nuptial pads of ovariectomized females. This leaves a number of open questions regarding the ontogenetic and evolutionary origin of the SDSGs. For example, are the enlarged dermal glands that develop in ovariectomized females as a response to the administration of androgens histologically and histochemically identical to the SDSGs present in the nuptial pad of males, or are they simply enlarged mucous glands? Thomas & Licht (1993) showed that mucous glands, at least those on the dorsum of a few ranid species, are androgen dependent and enlarge in response to hormone administration. If this were the situation in other neobatrachians also, then perhaps the nuptial pad induced in females would not be a nuptial pad as commonly understood, because of the lack of SDSGs. Alternatively, if these enlarged glands are SDSGs, then perhaps they originate through androgen induction from ordinary mucous glands that express androgen receptors. The latter postulate, however, will require further study, as the distribution of androgen receptors in dermal glands in the body is poorly known, at best (d’Istria, Delrio & Chieffi, 1975; d’Istria et al., 1982; Yang et al., 2004), and for very few species. In Xenopus, the situation deserves further study, because Thomas & Licht (1993) demonstrated that mucous glands, at least in the dorsum, are not androgen dependent. Our results have shown a perplexing diversity in the morphology of the epidermal component of nuptial pads. When this component occurs in females in natural conditions, as in females of Xenopus (Fujikura et al., 1988) or Crossodactylus (present study), it has the same structure as that in males. However, when this component is induced in ovariectomized females, the question arises as to whether it shows the same morphology as in males. The various aspects of hormonal cycles and nuptial pad development have been studied in an extremely restricted number of species, from only a few anuran families, and with a strong bias to species of the Northern Hemisphere. All generalizations regarding this subject should be made with caution. Nuptial pads as taxonomic characters Nuptial pads have traditionally been included in taxonomic and, more recently, phylogenetic studies. The terminology that has been used is confusing, and the absence or presence of a coloured stratum corneum has heavily influenced the description of nuptial pads. Several authors (e.g. Lynch & Ruiz-Carranza, 1996; Cisneros-Heredia & McDiarmid, 2007) have explicitly treated nuptial pads as one of the possible types of excrescences (others being prepollical spines and ‘nuptial glands’; Lynch & Ruiz-Carranza, 1996). Other authors often differentiated nuptial pads from nuptial asperities or excrescences, with the latter two often treated synonymously and referred to as dark-coloured nuptial pads (Boulenger, 1890, 1897, 1912; Duellman, 1970). Others treated nuptial pad and nuptial excrescence as synonyms (Liu, 1936; Peters, 1964). Although Liu (1936) differentiated nuptial pad from nuptial asperities independent of the colour of the structure, this subtlety did not enjoy much subsequent usage. Given the frequently used and often inconsistent terminology in the literature, unless there is an explicit definition, statements like ‘nuptial excrescences absent’ provide little information regarding the presence or absence of a nuptial pad, only that, if it is present, it is not dark-coloured. Whereas smooth nuptial pads might require particular attention to assess their presence, it was evident during the present study that on several occasions authors referred to ‘nuptial pads’ or ‘nuptial asperities’ as being absent, when they indeed were present. This is the case in the Dendropsophus labialis group, as Duellman et al. (1997) included in its definition ‘nuptial excrescences and projecting prepollical spines absent in males’, although D. molitor (as Hyla labialis) in fact has a light-coloured nuptial pad with truncate papillae (Fig. 15G–I). Drewes (1984) characterized the genera Leptopelis (Arthroleptidae) and Hyperolius (Hyperoliidae) as lacking nuptial pads; however, we found some species of these genera to have smooth nuptial pads (Fig. 23A). In the same way, Brown et al. (2015) characterized all Ceratobatrachinae ceratobatrachids as lacking nuptial pads. In the course of our study, we found that at least Platymantis dorsalis has a nuptial pad with a few small and scattered EPs that have well-developed SMGs in the dermis (Fig. 23B; see also Fig. 16H). We could not characterize the nature of the EPs through histological analysis (presence or absence of a dermal core). However, SEM allowed the visualization of their conical shape (data not shown). Figure 23. View largeDownload slide External morphology of nuptial pads. A, Hyperolius adspersus (USNM 578158); light-coloured smooth nuptial pad on fingers II, III and IV. Glandular acini of the specialized mucous glands are visible by transparency (black arrows). B, C, Platymantis dorsalis (USNM 508521). B, light-coloured nuptial pad on fingers II, III and IV and forearm. C, magnification of fingers II and III. Glandular acini are visible by transparency (black arrows). Note that the pigmented skin masks the extent and limits of this nuptial pad. See Fig. 16H for cross-section of this nuptial pad. Scale bars: 1 mm in A; 2 mm in B; 500 µm in C. Figure 23. View largeDownload slide External morphology of nuptial pads. A, Hyperolius adspersus (USNM 578158); light-coloured smooth nuptial pad on fingers II, III and IV. Glandular acini of the specialized mucous glands are visible by transparency (black arrows). B, C, Platymantis dorsalis (USNM 508521). B, light-coloured nuptial pad on fingers II, III and IV and forearm. C, magnification of fingers II and III. Glandular acini are visible by transparency (black arrows). Note that the pigmented skin masks the extent and limits of this nuptial pad. See Fig. 16H for cross-section of this nuptial pad. Scale bars: 1 mm in A; 2 mm in B; 500 µm in C. In their study of centrolenid taxonomic characters, Cisneros-Heredia & McDiarmid (2007) established two different terms to refer to nuptial pads. The term ‘pad’ was used when densely packed glands were present, forming a ‘cushioned thick patch’. The term ‘cluster’ was applied when there was an aggregation of individual glands, without the formation of ‘cushioned patches’. They defined six character states based on the distribution of the glands on the body and their external morphology. This was modified by Guayasamin et al. (2009), who treated the morphologies recognized by Cisneros-Heredia & McDiarmid (2007) as nuptial pads. Cisneros-Heredia & McDiarmid (2007) also noticed the presence of individual glands in the webbing and fringes of hands and feet. Our histological analysis of a few centrolenid species (Centrolene buckleyi, their type VI; Nymphargus grandisonae, their type V) shows the presence of SMGs in both ‘pads’ and ‘clusters’. The ‘pad’ also has rounded papillae, whereas the ‘clusters’ show no epidermal modifications. In our study of the webbing of two species, Hyalinobatrachium valeroi and Nymphargus grandisonae, the former species shows the presence of OMGs, OSGs and, occasionally, SMGs, whereas the latter species shows only OMGs and OSGs. Our study of skin from the ventrolateral side of Centrolene buckleyi shows SMGs in the dermis and no epidermal modifications, similar to the findings on the ‘clusters’ of the studied species. Although we do not question the descriptive utility of the nuptial pad types used in centrolenid taxonomy, we do notice that from a histological perspective they conflate different structures of questionable homology. Nuptial pads have not been reported in most microhylids, with a few notable exceptions (some species of Kalophrynus, Metaphrynella, Ctenophryne and Stereocyclops, the two species of Hoplophryne, and the monotypic Melanobatrachus and Phrynella; Barbour & Loveridge, 1928; Parker, 1934; Lehr & Trueb, 2007; Zug, 2015). In contrast, SDSGs have been described in one species of Gastrophryne as occurring in the pectoral region of males (Conaway & Metter, 1967; Holloway & Dapson, 1971; Siegel et al., 2008), or briefly referrenced as occurring in the pectoral and/or abdominal region in other species of Gastrophryne (Fitch, 1956; Savage, 2002), Elachistocleis (Scrocchi & Lavilla, 1990) and Kaloula (Liu, 1936; Inger, 1954), and always related to the adhesion of the pair during amplexus. An adhesive-mediated amplexus has also been recorded in Dasypops, Chiasmocleis, Elachistocleis and Hypopachus (Haddad & Hödl, 1997; Savage, 2002; Pombal & Cruz, 2016), although no glandular region has been reported so far in these frogs. Our study of the skin on finger II of some microhylids that lack an evident nuptial pad reveals the presence of SMGs together with OMGs and OSGs (see Supporting Information, Table S2). This histological structure is the same as the published results for Gastrophryne and our observations of lateral and mental skin in Uperodon systoma (see Supporting Information, Table S2), suggesting that SDSGs have a broader distribution in the skin of this family than reported. A systematic study of skin in different body regions, and the study of at least some of the cases were nuptial pads do occur, will provide a better understanding of the occurrence and evolution of SDSGs in this family. A wide occurrence of SMGs is also evident in Brevicipitidae, a family that also has adhesive-mediated amplexus (Visser, Cei & Gutierrez, 1982; see Supporting Information, Table S2). Clusters of SDSGs occur in many anuran clades on parts of the body other than on the nuptial pad, and several authors have considered some of these clusters, regardless of whether they have an epidermal component or not, also as nuptial pads (e.g. Peters, 1964; Duellman & Trueb, 1986; Cisneros-Heredia & McDiarmid, 2007; Glaw & Vences, 2007). Some reports have addressed the histological composition of these glands (Le Quang Trong, 1976; Thomas et al., 1993; Brizzi et al., 2002; Vences et al., 2007; Brunetti et al., 2012, 2015), and these findings, together with our preliminary studies on lateral skin, lower jaw skin and pectoral skin in some anurans, provide a framework for comparing some of these structures with nuptial pads. The presence of sexually dimorphic, dark-coloured spines spread or aligned in rows on the lower jaw or aligned on the upper lip (e.g. Noble & Putnam, 1931; Liu, 1950; Cadle, 1995; Barej et al., 2010; Sondhi & Ohler, 2011; Zhang et al., 2016) has been reported for species in different clades of anurans. There are references in the literature to different structures that are called ‘pectoral spines’ or ‘chest spines’. These include those forming ‘patches’ in Alsodes (Lynch, 1978), Insuetophrynus (Barrio, 1970), Oreolalax (Megophryidae; Liu, 1950; Wu et al., 1993; Nguyen et al., 2013) and Scutiger (Megophryidae; Liu, 1936; Jiang et al., 2012); scattered spines in Leptodactylodon (Arthroleptidae; Amiet, 1981), Boophis (Mantellidae; Blommers-Schlösser, 1979), Hoplophryne (Barbour & Loveridge, 1928), Nanorana (Liu, 1936), Telmatobius (Barrionuevo, 2017) and Quasipaa (Liu, 1936); and compact clusters of spines reported in some species of Leptodactylus (Heyer, 1969, 2005). Our histological study of spines on the lower jaws of D. pictus, Platymantis dorsalis and Petropedetes palmipes revealed a structure that is similar histologically to that of nuptial pads, with the addition of OMGs. Our comparison of pectoral spines and nuptial pads revealed that only in the case of Alsodes, where the pectoral spines are concentrated in ‘pectoral patches’, do they represent the same structure as those found in the nuptial pads. Perhaps this is the situation in other patch-like structures, such as those reported from Scutiger (Liu, 1936; Jiang et al., 2012). It would be interesting to extend this comparison of nuptial pads with axillary glands in Scutiger to those that occur next to the pectoral patches and are also covered by black spines (e.g. Liu, 1950; Jiang et al., 2016). The cornified chest spines reported in the Leptodactylus pentadactylus group (e.g. Heyer, 1975; Ponssa, 2008; de Sá et al., 2014) are similar to the pectoral spines mentioned above, but they differ by forming compact, black patches. Our results indicated the presence of few SMGs (similar to the composition found in nuptial pads with large spines), but they showed multiple and thick upper layers in the stratum corneum. Our results indicate that centrolenid frogs are the only group that has lateral glands showing only SMGs, as they occur in their nuptial pads. Studies in other groups indicate that lateral glands have both SMGs and SSGs (Brunetti et al., 2012), or SSGs only (Thomas et al., 1993), as do femoral glands in mantellids (Vences et al., 2007) and inguinal glands in Cycloramphus (Gonçalves & de Brito-Gitirana, 2008). Evidence indicates that only histological analyses will determine whether a cluster of SDSGs or a differentiated epidermal structure is genuinely comparable to a nuptial pad; this cannot be assumed a priori. The relevance of this distinction will vary according to the context. The use of expressions such as ‘sexually dimorphic glandular cluster’ or ‘sexually dimorphic spines’ to document the presence of such structures in a specific group of anurans will be more neutral from an anatomical perspective than direct reference to ‘nuptial pads’ when describing structures occurring in parts of the body other than those where nuptial pads are known to occur (all sexually dimorphic structures considered nuptial pads and occurring on finger II that we studied have SMGs). Our surveys using optical microscopy at different magnifications, histology and SEM indicate that the diversity of nuptial pads includes differences in extent, coloration, size, occurrence and structure of EPs, pore morphology, nuptial gland structure and nuptial gland histochemistry. Taxonomic studies commonly include descriptions of nuptial pads with low magnification, and only rarely with SEM. From the perspective of our investigations and surveys on the diversity of nuptial pads, observations under low magnification provide only a superficial perspective; they can provide information about coverage, coloration, density, and the occurrence and morphology of some of the papillae (most frequently small papillae and spine-shaped papillae and, within the former, spicule-shaped and cone-shaped papillae). Of course, all this variation is only a baseline; we found obvious variation in the size of EPs among species, such as within cone-shaped and spine-shaped papillae, for example. Scanning electron microscopy allows the visualization of epidermal projections and, if present, ornamentation of the papillae, and provides data on the occurrence, position and morphology of glandular pores. When the nuptial pad is light coloured, frequently only SEM allows visualization of the EPs. Although the diameter of the EPs can be used as a tool to determine whether they are PEPs or NPEPs, the most reliable method to study the structure of the EP is through histological analysis. The occurrence of dark- or light-coloured nuptial pads is a character frequently used in taxonomic descriptions. Our study indicates that differences in coloration result from minor changes in the stratum corneum, and all other epithelial structures are similar. On the basis of the evidence discussed earlier in this paper, it is more appropriate to described nuptial pads as dark- or light-coloured, rather than pigmented or unpigmented, because there is no evidence indicating that coloration in pads is attributable to chromatophores. The nuptial pad may be thicker than the adjacent skin, owing to three histological conditions: (1) the hypertrophy of the SMGs (e.g. Conraua crassipes, Phyllodytes luteolus, Rana japonica); (2) the presence of papillary (e.g. Scinax perereca, Spea bombifrons) or non-papillary epidermal projections (e.g. Rhinophrynus dorsalis), with non-hypertrophied SMGs; and (3) both hypertrophy of SMGs and the presence of papillae or epidermal projections [e.g. Aparasphenodon brunoi (Hylidae), Ascaphus montanus]. role in reproduction: variable roles in amplexus? As has been known for a long time, and corroborated by our studies, nuptial pads are highly variable structures. This variation is likely to be a primary reason why throughout the history of amphibian biology, particularly as the morphological diversity of nuptial pads and their taxonomic distribution became somewhat better known, hypotheses regarding their role in reproductive biology were stated only in very general and sometimes vague terms that usually included a few known exceptions or alternative hypotheses. The analysis of the multiple hypotheses summarized below should start from the understanding that although nuptial pads might have been selected in the early history of anurans for one or more reasons, the multiple transformations of these structures throughout their subsequent evolutionary history do not require the same selective pressures to account for them, and therefore different explanations for their biological role might apply for different clades. The origin of the discussion on the biological role of nuptial pads is traceable at least back to Swammerdam (1737), who suggested that nuptial pads increase the hold on the female by the male during amplexus. Lataste (1876) simply referred to that explanation as ‘doubtless’, without citing earlier authors. This hypothesis was also mentioned by Boulenger (1897, 1910, 1912), Aron (1926), Noble (1931) and Rostand (1933) and persisted as one of the most often cited hypotheses (e.g. Heyer, 1975; Duellman & Trueb, 1986; Brizzi et al., 2003a). In 1940, Parker suggested that the occurrence of pads was associated with amplexus in water, whereas their absence was more common in species with terrestrial amplexus. Duellman & Trueb (1986) followed Parker’s (1940) ideas, adding that pads are better developed in species that breed in streams. They also stated that the extent and ‘spinosity’ of the pads seem to be correlated with the difficulty of maintaining amplexus, which they assumed to be greater in torrential streams. Wells (2007) stated that in certain species the pads are possibly adaptations for holding onto females in moving water. As the multiple roles proposed to account for the evolution of nuptial pads involve their action during amplexus, diversity in amplectic behaviour (e.g. Duellman & Trueb, 1986; Wells, 2007) should be considered when discussing diversity in nuptial pad morphology (and vice versa). Amplectic behaviour includes the multiple ways that a male and female interact physically and the dynamics of oviposition and fecundation. Unfortunately, the different ways in which all these factors come into play have been poorly studied. Most comments refer to the possible advantages of axillary vs. inguinal amplexus (e.g. Lynch, 1973; Duellman & Trueb, 1986; Wells, 2007) or focus on a few specific amplectic positions (e.g. Savage, 1961; Altig, 2008; Brunetti, Taboada & Faivovich, 2014; Willaert et al., 2016). As a result of the paucity of information about the interplay between behaviour and morphology in the evolutionary history of nuptial pads and breeding in frogs, we are left with offering only some general comments. This lack of knowledge precludes serious evaluation of the selective advantage(s) that nuptial pads could have played in accounting for their morphological diversity, and not only in the few generalized characterizations, such as presence/absence and ‘development’ or ‘spinosity’, as it has been called in the literature. However, even considering only these general terms, the story is complex, and multiple exceptions exist for all generalizations advanced so far regarding their possible role in reproductive biology. An association between amplexus in flowing water and the common occurrence of cone- or spine-shaped papillae in males of many stream-breeding frogs has been well documented: Telmatobius (Barrionuevo, 2017); the Hyloscirtus armatus group (Duellman et al., 1997); the Ranoidea nannotis group (Tyler & Davies, 1978; Cunningham, 2002); Insuetophrynus (Barrio, 1970); Astylosternus and Trichobatrachus (Arthroleptidae; Amiet, 1978; Perret, 1966); the dicroglossids Allopaa, Chrysopaa, Nanorana and Quasipaa (Boulenger, 1920; Pope, 1931; Ohler & Dubois, 2006; Dubois & Ohler, 2009); Thoropa (Bokermann, 1965); and Telmatobufo (Calyptocephalellidae; Formas, Núñez & Brieva, 2001). In the case of Telmatobius, Barrionuevo (2017) showed differences in the density and morphology of papillae related to habitat, indicating that cone-shaped papillae (that he called epidermal projections) occur in species that breed in high-gradient streams. However, there are exceptions, including stream-breeding species that have nuptial pads, sometimes light coloured, without cone- or spine-shaped papillae, and others that lack nuptial pads completely. Examples of the former are Conraua (Perret, 1966; see Supporting Information, Table S2), several species of Boophis (Blommers-Schlösser, 1979), and the Scinax catharinae group (Lourenço, Luna & Pombal, 2014). Examples of the latter include species of Hyloscirtus, all of which are stream breeders, but nuptial pads are absent in the Hyloscirtus bogotensis group (Brunetti et al., 2015; Guayasamin et al., 2015; M. Rivera-Correa, personal communication) and in some species of the Hyloscirtus larinopygion group (Coloma et al., 2012; Rivera-Correa & Faivovich, 2013). The same applies for most stream-breeding species of Boana (Brunetti et al., 2015) and some species of Bokermannohyla (Hylidae; e.g. Faivovich et al., 2009; Leite, Pezzuti & Drummond, 2011). Interestingly, nuptial pads are fairly enlarged in some phytotelm-breeding taxa, involving spine-shaped papillae in Crossodactylodes (Leptodactylidae; Cochran, 1955; Barata et al., 2013), Mertsensophryne micranotis (Bufonidae; Grandison, 1980), Hoplophryne (Barbour & Loveridge, 1928; Noble, 1931), some species of Kalophrynus (Matsui, 2009), and Osteopilus wilderi (see Supporting Information, Table S2), or an enlargement of the dermal component in some species of Isthmohyla (Hylidae; Isthmohyla picadoi and Isthmohyla zeteki; Köhler, 2011) and some species of Kurixalus (Rhacophoridae; Wu et al., 2016). As Luna et al. (2012) noticed, all phyllomedusines that have terrestrial amplexus also have dark-coloured nuptial pads with highly ornamented papillae (they used the more general term ‘epidermal projection’), covering an area of finger II that is larger than in most Cophomantini and Dendropsophini hylids, most of which have aquatic amplexus. There are several other exceptions involving species that have terrestrial amplexus, but with a diversity of nuptial pad morphologies, such as most centrolenids (Cisneros-Heredia & McDiarmid, 2007), several brachycephaloids (e.g. Lynch & Duellman, 1997; Duellman & Lehr, 2009), hemiphractids (Duellman, 2015), several rhacophorids (e.g. Liu, 1936; Inger, 1966; Kurabuchi, 1994; Harvey, Pemberton & Smith, 2002; Manamendra-Arachchi & Pethiyagoda, 2005), Leptodactylodon (Amiet, 1981), some species of Leptopelis (Amiet, 2012), Alcalinae ceratobatrachids (Brown et al., 2015), some species of batrachylids (Batrachyla; Formas, 1976) and the bufonid Nimbaphrynoides occidentalis (Grandison, 1980). It seems feasible that dark- and light-coloured nuptial pads are functionally different among species. A dark-coloured nuptial pad might imply a denser, more rigid stratum corneum, which would provide a stiffer, less collapsible structure when pressed against the female’s body during amplexus, and thereby strengthen the grip required to avoid slipping from the female’s dorsum, or avoid being displaced by another male, or for other relevant mechanics of amplexus, such as piercing the epidermis. However, this requires a thorough study, because in several cases the occurrence of a coloured stratum corneum is not associated with an increase in its thickness. Savage (1932, 1934, 1961) expressed scepticism that nuptial pads play a role in maintenance of a regular, undisturbed amplexus and contended that they are ineffective in retaining a female that is unwilling to engage in amplexus. Instead, he suggested that the hold on the females was to avoid the gripping male being dislodged by competing males, a perspective also mentioned by Smith (1949). Wells (1977) added that the behaviour of dislodging other males may favour the evolution of enlarged nuptial pads in explosive-breeding species, stressing that he considered it unlikely that the nuptial pads would be needed simply to hold the female, because females are usually passive and do not attempt to escape from clasping males. Wells (2007) also suggested that the association between nuptial pads and aquatic reproduction might be partly spurious, explaining its presence in the context of mate-guarding behaviour instead of mating behaviour per se. He further predicted that nuptial pads will be well developed in explosive breeders with extensive male–male competition for females, but not necessarily in prolonged-breeding, territorial species. The hypothesis of Savage (1932, 1934, 1961) and Wells (1977, 2007) might be congruent with nuptial pad evolution in certain groups of anurans (Wells specifically mentioned North American ranids) and is certainly amenable to testing in a phylogenetic context. A limitation of this hypothesis in the context of nuptial pad evolution is that it covers only a small fraction of the known diversity in nuptial pad morphology, emphasizing ‘development’ of the nuptial pad, or at best referring to a ‘large nuptial pad’ as opposed to a ‘somewhat enlarged…but keratinized nuptial pad’ (Wells, 2007: 396). The hypotheses outlined above regarding the biological role of the nuptial pads involve the morphology of the epidermal component of the pad and its correlation with a few characters related to reproductive behaviour (e.g. site of amplexus, male–male competition for females). Although sexually dimorphic skin glands were generally hypothesized to mediate some form of chemical communication (e.g. Duellman & Trueb, 1986), Thomas et al. (1993) were the first authors to suggest that the SDSGs of the nuptial pad might also have a pheromone function. This general lack of attention to the SDSGs when discussing the various functional roles of nuptial pads is paradoxical in that all available literature and our own survey indicate that the SMGs are the only structures in common among all the different morphologies associated with nuptial pads. We are unaware of any study that suggests a functional dependence between the SDSGs and the overlying epidermal component; both have been shown to have testosterone receptors (Varriale & Serino, 1994; Emerson et al., 1999; Yang et al., 2004). Instead, the fact that dark, cornified structures occur in some species in other parts of the body without any underlying SDSG (see above) supports the idea of independence between both components of the nuptial pad. How this independence affects the evolution of nuptial pad structure and its function requires discussion. The occurrence of SDSGs on the body in general has been associated with the occurrence of adhesive secretions involved in amplexus (Metter & Conaway, 1969; Holloway & Dapson, 1971; Visser et al., 1982; Duellman & Trueb, 1986; Siegel et al., 2008) or chemical communication (Duellman & Trueb, 1986; Thomas et al., 1993; Epstein & Blackburn, 1997; Brizzi et al., 2002, 2003a; Vences et al., 2007; Brunetti et al., 2012, 2014, 2015; Willaert et al., 2013). These are functional roles that have also been generalized for the SDSGs of the nuptial pads. Although this is worth considering, it has been documented that known SDSGs occurring on other parts of the body sometimes differ from those on the nuptial pad by the presence of SSGs, with or without SMGs (e.g. Thomas et al., 1993; Vences et al., 2007; Brunetti et al., 2012), and at least in some instances the SMGs in the nuptial pad lack the lipid droplets that occur in SMGs in other parts of the body (Brizzi et al., 2002). Several authors have specifically discussed the role of SDSGs of the nuptial pads, and most of these have focused on the importance of chemical communication (Thomas et al., 1993; Brizzi et al., 2002, 2003a; Willaert et al., 2013) and, occasionally, as a possible source of adhesive secretions (Aron, 1926; Brizzi et al., 2003b; Luna et al., 2012). The evidence supporting a role of nuptial pad SDSGs in chemical communication stems from their structural similarities with the so-called hedonic glands of caudate amphibians, as noticed by Thomas et al. (1993), for which there is a large amount of evidence regarding their role in pheromone secretion during courtship (e.g. Sever & Staub, 2011). More recently, Willaert et al. (2013) studied nuptial pad secretions in Rana temporaria by using transcriptomics and proteomics, and characterized a group of small proteins from the Ly-6/uPAR family that they called amplexins, which occur during the breeding season but whose expression diminishes to undetectable levels in the non-reproductive period. The amplexins found in R. temporaria showed similarities with one of the known plethodontid pheromones, the plethodontid modulating factor (PMF), suggesting the possible existence of an ancient pheromone system in amphibians. Willaert et al. (2013) hypothesized that at the moment of amplexus, amplexins are secreted and seep into the female through the wounds produced by the spines of the nuptial pad and cause a reduction in the duration of the clasping. Wounds in the pectoral regions of females that were inflicted by the male’s nuptial pads during amplexus have been known for a few ranids for > 250 years (Rösel von Rosenhof, 1758; Boulenger, 1897; Gaupp, 1904; Aron, 1926; Savage, 1961), but we are unaware of their occurrence in other groups. However, we presume that if the disruption of the epidermis is necessary (this may not be the case; perhaps a topical application might suffice), the level of injury of the skin necessary for the seeping of the secretion might not be evident macroscopically. If so, similar mechanisms of secretion seeping into a female’s tissues, as hypothesized by Willaert et al. (2013), might be more common and widely distributed among other anuran species. A first requirement to test this hypothesis would be to corroborate a role of the SDSG secretion in chemical communication. The role of the SDSGs of the nuptial pad as a source of adhesive secretions has received little attention (Brizzi et al., 2002; Luna et al., 2012) in comparison with other SDSGs (Metter & Conaway, 1969; Holloway & Dapson, 1971; Visser et al., 1982; Duellman & Trueb, 1986; Siegel et al., 2008). This is despite the fact that we have commonly observed the remains of debris stuck on the pads of both living and fixed specimens of several species, which indicates a probable adherent nature of the secretions. Also relevant to the presumed adhesive nature of the secretions are our observations that in a few species with spine-shaped papillae (Alsodes, Osteopilus wilderi and some species of Thoropa), the number of glandular pores is lower than with cone-shaped papillae, and these occur on the margins of the spines, presumably as a consequence of a lower secretory capacity per total surface area. This correlation might indicate a shift in the proportional contribution of glandular secretions and surface contacts with the epidermal component. Whether this shift implies an enhancement in the friction and holding capacity of the amplecting male on the female requires further testing. Regardless of the possible roles of SDSGs in chemical communication or as sources of adhesive secretion, when we consider the way their secretion spreads and contacts the female, the epidermal component of the pads requires further attention. Studies on synthetic adhesives and adhesive secretions from invertebrates reveal that the softness of an adhesive and the shape of the contact surface affect the intensity of its spreading (Gay, 2002; Smith, 2002). If the adhesive is soft enough, the contact will extend from the peaks of a rough surface and reach into the valleys and beyond, thereby trapping isolated air bubbles. If the adhesive is purely elastic and secreted in ample volume, the contact will be restricted to the top of the peaks, resulting in small, isolated contact regions (Gay, 2002). Although we still have no information about the physical properties of nuptial pad secretions, the morphological variability of the papillae and the degree of exposure of the secretory pores are both likely to impact the way the secretion is spread on the nuptial pad surface and, eventually, on the skin of a female (Luna et al., 2012). Extensive folding of epidermal cells is characteristic of many epithelial surfaces in vertebrates. These folds, known as microridges (Sperry & Wassersug, 1976), are thought to be involved in holding mucous secretions to cell surfaces and increasing the functional surface area (Olson & Fromm, 1973; Sperry & Wassersug, 1976). The occurrence of ornamentations on the papillae and/or the presence of microridge-like structures on the ornamentation of the nuptial pad might also facilitate the spreading of secretion. An evaluation of the properties of nuptial pad secretion, together with an analysis of the morphology of the nuptial pad, will provide a better understanding of how nuptial pad morphological diversity is related to the composition (viscosity and adhesion) of its secretion. In the context of Savage’s (1961) hypothesis that nuptial pads increase a male’s ability to hold onto an amplectant female, and thereby avoid being dislodged by competing males, he suggested that pads also serve as ‘weapons’ to avoid interference from competing males, specifically referring to the case of Bombina, in which structures that he interpreted as nuptial pads occur on the toes. Subsequently, Duellman & Trueb (1986) referred to this hypothesis as a possible general explanation for nuptial pad morphology but provided no specific data. As the discussion below indicates, there is limited evidence that nuptial pads play a role in combat; on the contrary, the structures that have been observed to play a role in male combat (‘thumb spines’ and prepollical spines) are unrelated to nuptial pads per se. The only cases of which we are aware that require further study with regard to combat are the structures occuring on the feet of Bombina variegata and the coloured spines on the upper jaw of some species of Leptobrachium (Liu, 1950; Sondhi & Ohler, 2011); in these cases, there is evidence of a role in male combat (Savage, 1961; Hudson, He & Fu, 2011). The peculiar spines in Heleioporus and some species of Leptodactylus have been variously called ‘thumb spines’ (e.g. Heyer, 1969, 1998), ‘keratinized spines’ (e.g. Heyer & de Sá, 2011), ‘nuptial spines’ (e.g. Ponssa, 2008) or simply ‘spines’ (e.g. Parker, 1940). In Leptodactylus, these are the result of modifications of both metacarpal II and the distal element of the prepollex, which are covered by a dark cap of cornified material (Lynch, 1971). The situation in Heleioporus is similar, at least for spines on finger II; we did not analyse those on other fingers, which apparently do not have a bony core (Parker, 1940). The role of ‘thumb spines’ in Leptodactylus during amplexus or territorial interactions is poorly understood; we are aware of a single observation of them being used in combat between males (Rivero & Esteves, 1969). Phylogenetic studies in Leptodactylidae have addressed the presence or absence of nuptial pads and ‘thumb spines’ as two possible states of the same character (e.g. Heyer, 1975, 1998; Ponssa, 2008). We found that in the ‘thumb spines’ the stratum corneum of the epidermis is formed by multiple cornified layers, instead of only a single layer as occurs in most pads, and no SMGs are present in the dermis. These differences call into question the homology of ‘thumb spines’ and nuptial pads and indicate that they should be considered as different characters. Other sexually dimorphic structures involve modification of the prepollex or metacarpal II into a bony spine. The latter occurs in several species of Petropedetes (Barej et al., 2010). The former occurs in Babina (Ranidae; Iwai, 2012), some species in the genera Teratohyla and Vitreorana (Centrolenidae; Lynch & Ruiz-Carranza, 1996; Señaris & Ayarzagüena, 2005), the Boophis albilabris group (Cadle, 1995; Vences et al., 2010), Boana, Bokermannohyla, Ecnomiohyla, some species of Hyloscirtus and Plectrohyla (Hylidae; Duellman & Campbell, 1992; Duellman, 2001; Faivovich et al., 2005) and is sometimes called, in the non-taxonomic literature, a ‘pseudothumb’ (e.g. Wells, 2007; Iwai, 2012). Prepollical spines are usually present in both sexes, but in females they are reduced and usually unossified (e.g. Kluge, 1981; Cadle, 1995; Garcia, Faivovich & Haddad, 2007; Garcia & Haddad, 2008). Based on direct observations or inferences based on scars on the dorsa of males (e.g. Kluge, 1981; Duellman & Campbell, 1992; Iwai, 2012), prepollical spines have been associated mostly with territorial combat among males. With the exception of Iwai’s (2012) report that the prepollical spines of male Babina subaspera are involved in clasping during amplexus and that most females had wounds on the sides of their bodies, there are no other observations of a direct role of the prepollical spine during amplexus. Although these structures might be involved in amplectic mechanics, there are also cases where both prepollical spines and nuptial pads coexist (e.g. the Boophis albilabris group, several species of the Boana semilineata group, Bokermannohyla and Plectrohyla; Duellman, 1970; Vences et al., 2010; Leite et al., 2011; Brunetti et al., 2015), which indicates that they may have different roles in reproduction. Although we do not hypothesize homologies among any of these structures, we wonder whether clusters of SDSGs in other parts of the body (e.g. mental, lateral, pectoral, inguinal and femoral glands; Vences et al., 2007; Brunetti et al., 2012) or even unrelated structures (such as the ‘thumb spines’ of Leptodactylus, and prepollical and metacarpal spines) could be functional extensions of, or replacements for, nuptial pads. Interestingly, certain associations that might suggest this include: (1) the absence of nuptial pads and the occurrence of femoral glands in Mantellinae mantellids (Glaw & Vences, 2007; Vences et al., 2007); (2) the absence of nuptial pads and the occurrence of mental glands in several hyperoliids (Le Quang Trong, 1976; Drewes, 1984); (3) the absence of nuptial pads and the occurence of prepollical spines and mental and lateral glands in several Cophomantini hylids (Brunetti et al., 2015), including the co-occurrence in a few species of both nuptial pads and clusters of SDSGs; and (4) the absence of nuptial pads and the occurrence of inguinal glands in Cycloramphus (Heyer, 1983; Gonçalves & Brito-Gitirana, 2008). These hypotheses could be tested through a number of complementary approaches. The study of the skin adjacent to the prepollical spine (in species where no nuptial pad is present) or in other areas of the body (e.g. Vences et al., 2007; Brunetti et al., 2015) would provide a test for the occurrence of SDSGs. Behavioural observations are also needed to understand the role, if any, played by ‘thumb spines’, prepollical spines and areas where SDSGs occur during amplexus (e.g. Altig, 2008; Brunetti et al., 2014, 2015). Studies on the structure (e.g. SMG or SSG), nature of the secretions, and comparative transcriptome analyses between nuptial pads and SDSGs occurring in other body regions among closely related species will help us to understand whether these structures function in similar ways during reproductive activities. A more precise interpretation of the role of the dermal and epidermal components of the nuptial pads in different groups of anurans seems to be reliant on: (1) a more thorough empirical knowledge on the role of the epidermal component in the mechanics of amplexus; (2) a better understanding of the nature of the glandular secretions of the dermal component; and (3) a careful study of nuptial pad diversity using hypotheses of homology combined with a study of the taxonomic distribution of nuptial pads in a phylogenetic context. The last of these items would allow the association of inferred transformations in pad morphology through the evolutionary history of each group with those in character systems related to reproductive biology. Information derived from items (1) and (2) would ideally facilitate interpretation of these results in a phylogenetic context. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher's web-site: List S1. List of species processed by (1) scanning electron microsopy; (2) histology; and (3) transmission electron microscopy. All specimens are males unless specified. List S2. List of species studied but not processed. All specimens are males unless specified. Table S1. Examples of nuptial pad terminology used in the literature. Table S2. Histochemical results of glands in nuptial pads and some other secondary sexual characters in the species included in this study. Table S3. Results of the hydrogen peroxide test. After 48 h of treatment, the species with dark-coloured nuptial pads showed bleaching of dermal melanophores but no difference in the coloration of the stratum corneum, indicating an absence of melanin in this stratum. ACKNOWLEDGEMENTS Fabian Tricárico (MACN) and Scott Whittaker (NMNH) provided technical assistance in SEM procedures. Helen Wimer (NIH), Lynne Parenti (NMNH) and Cristian Ituarte (MACN) kindly allowed full access to their histology laboratories. Isabel Farias prepared TEM samples. David Johnson (NMNH) kindly granted access to the Zeiss V20 stereomicroscope. Agustin Elias Costa generously executed Figures 6 and 14. Andres Brunetti, Martín Pereyra, David Sever, Erik Wild and two other anonymous reviewers read the manuscript and provided insightful comments. Ronald Altig, Karen Lips and Jose M. Padial assisted with critical literature. Marco Rada, Mauricio Rivera-Correa, Pedro Peloso and Mariane Targino provided useful information regarding their taxonomic groups of expertise. Gabriela Küppers assisted us with German translations. Darrel R. Frost, Christopher Raxworthy and David Kizirian (AMNH), Celio F. B. Haddad (CFBH), Robert F. Inger and Alan Resetar (FMNH), Jose P. Pombal Jr (MNRJ), Taran Grant (MZUSP) and Kevin de Queiroz (USNM) authorized access to collections under their care and/or provided specimen loans. M.C.L. is indebted to the staff at the National Museum of Natural History, Washington, DC, especially Kevin de Queiroz, Jeremy Jacobs, Kenneth Tighe, Robert Wilson and Addison Wynn for assistance and access to facilities and specimens during her stay. Financial support was provided by the Smithsonian Institution through a Smithsonian Predoctoral Fellowship and Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) to M.C.L., the São Paulo Research Foundation (FAPESP Procs. 2012/10000–5, 2013/50741–7), CONICET (PIP 11220110100889) and Agencia Nacional de Promoción Científica y Tecnológica (PICT 404/2013, 820/2015). REFERENCES Alexander NJ, Parakkal PF. 1969. Formation of α- and β-type keratin in lizard epidermis during the molting cycle. Zeitschrift fur Zellforschung und mikroskopische Anatomie  101: 72– 87. Google Scholar CrossRef Search ADS   Alibardi L. 2003. Ultrastructural localization of histidine-labelled interkeratin matrix during keratinization of amphibian epidermis. Acta Histochemica  105: 273– 283. Google Scholar CrossRef Search ADS   Alibardi L. 2009. Ultrastructural characteristics of the process of cornification in developing claws of the brushtail possum (Trichosurus vulpecula). Acta Zoologica  90: 285– 300. Google Scholar CrossRef Search ADS   Alibardi L. 2010a. Cornification of the beak of Rana dalmatina tadpoles suggests the presence of basic keratin-associated proteins. Zoological Studies  49: 51– 63. Alibardi L. 2010b. Cornification in the claw of the amphibian Xenopus laevis (Pipidae, Anura) and comparison with claws in amniotes. Italian Journal of Zoology  77: 399– 409. Google Scholar CrossRef Search ADS   Altig R. 2007. Comments on the descriptions and evaluations of tadpole mouthpart anomalies. Herpetological Conservation and Biology  2: 1– 4. Altig R. 2008. Notes on the breeding biology of four species of mantellid frogs from Madagascar. Tropical Zoology  21: 187– 194. Amiet JL. 1978 “1977”. Les Astylosternus du Cameroun (Amphibia, Anura, Astylosterninae). Annales de la Faculté des Sciences de Yaoundé ; 23–24: 99– 227. Amiet JL. 1981 “1980”. Revision du genre Leptodactylodon Andersson (Amphibia, Anura, Astylosterninae). Annales de la Faculté des Sciences de Yaoundé ; 27: 69– 224. Amiet JL. 2012. Les rainettes du Camerun (Amphibiens Anoures) . Saint Nazaire: J.-L. Amiet & La Nef des Livres. Aron M. 1926. Recherches morphologiques et experimenteles sur le determinisme des caracteres sexuels secondaires males chez les anoures (Rana esculenta L. et Rana temporaria L.). Archives de Biologie  36: 1– 97. Ayarzagüena J. 1992. Los centrolenidos de la Guayana Venezolana. Asociación de Amigos de Doñana  1: 1– 48. Bagnara JT, Matsumoto J. 2007. Comparative anatomy and physiology of pigment cells in nonmammalian tissues. In: Nordlund JJ, Boissy RE, Hearing VJ, King RA, Oetting WS, Ortonne JP, eds. The pigmentary system: physiology and pathophysiology , 2nd edn. Oxford: Blackwell Publishing Ltd, 11– 59. Google Scholar CrossRef Search ADS   Bancroft JD. 2002. Theory and practice of histological techniques . London: Churchill Livingstone. Barata IM, Santos MT, Leite FS, Garcia PC. 2013. A new species of Crossodactylodes (Anura: Leptodactylidae) from Minas Gerais, Brazil: first record of genus within the Espinhaço mountain range. Zootaxa  3731: 552– 560. Google Scholar CrossRef Search ADS   Barbour T, Loveridge A. 1928. A comparative study of the herpetological faunae of the Uluguru and Usambara Mountains, Tanganyika Territory, with descriptions of new species. Memoires of the Museum of Comparative Zoology  50: 85– 265. Barej M, Rödel MO, Gonwouo LN, Pauwels OSG, Böhme W, Schmitz A. 2010. Review of the genus Petropedetes Reichenow, 1874 in Central Africa with the description of three new species (Amphibia: Anura: Petropedetidae). Zootaxa  2340: 1– 49. Barrio A. 1970. Insuetophrynus acarpicus, un nuevo leptodactylido firmisternio sudamericano (Amphibia, Anura). Physis  30: 331– 341. Barrionuevo JS. 2017. Frogs at the summits: phylogeny of the Andean frogs of the genus Telmatobius (Anura, Telmatobiidae) based on phenotypic characters. Cladistics  33: 41– 68. Google Scholar CrossRef Search ADS   Berk L. 1939. Studies on the reproduction of Xenopus laevis. III. The secondary sex characters of the male Xenopus: the pads. The South African Journal of Medical Sciences  4: 53– 60. Blair AP. 1946. The effects of various hormones on primary and secondary sex characters of juvenile Bufo fowleri. The Journal of Experimental Zoology  103: 365– 400. Google Scholar CrossRef Search ADS   Blaylock LA, Ruibal R, Platt-Aloia K. 1976. Skin structure and wiping behavior of phyllomedusine frogs. Copeia  1976: 283– 295. Google Scholar CrossRef Search ADS   Blommers-Schlösser RMA. 1979. Biosystematics of the Malagasy frogs. II. The genus Boophis (Rhacophoridae). Bijdragen tot de Dierkunde  49: 261– 312. Bokermann WCA. 1965. Notas sôbre as espécies de Thoropa Fitzinger (Amphibia, Leptodactylidae). Anais da Academia Brasileira de Ciências  37: 525– 537. Boulenger GA. 1890. The fauna of British India including Ceylon and Burma. Reptilia and Batrachia . London: Taylor & Francis. Google Scholar CrossRef Search ADS   Boulenger GA. 1897. The tailless batrachians of Europe . London: Ray Society. Google Scholar CrossRef Search ADS   Boulenger GA. 1910. Les Batraciens et principalement ceux d’Europe . Paris: Octave Doine et Fils. Boulenger GA. 1912. On some treefrogs allied to Hyla caerulea, with remarks on noteworthy secondary sexual characters in the family Hylidae. Zoologische Jahrbücher  15: 211– 218. Boulenger GA. 1920. A monograph of the South Asian, Papuan, Melanesian and Australian frogs of the genus Rana. Records of the Indian Museum  20: 1– 226. Bragulla HH, Homberger DG. 2009. Structure and functions of keratin proteins in simple, stratified, keratinized and cornified epithelia. Journal of Anatomy  214: 516– 559. Google Scholar CrossRef Search ADS   Brizzi R, Delfino G, Jantra S. 2003a. An overview of breeding glands. In: Jamieson B, ed. Reproductive biology and phylogeny of anura . Enfield, NH, USA: Science Publishers, Inc., 253– 317. Brizzi R, Delfino G, Jantra S. 2003b. Structure and functional significance of the nuptial thumb pads in Rana esculenta and R. perezi. In: Ananjeva N, Tseienko O, eds. Herpetologia petropolitana , Proceedings of the 12th Ordinary General Meeting of the Societas Europaea Herpetologica, St. Petersburg: SEH, 258– 262. Brizzi R, Delfino G, Pellegrini R. 2002. Specialized mucous glands and their possible adaptive role in the males of some species of Rana (Amphibia, Anura). Journal of Morphology  254: 328– 341. Google Scholar CrossRef Search ADS   Brown RM, Siler CD, Richards SJ, Diesmos AC, Cannatella DC. 2015. Multilocus phylogeny and a new classification for Southeast Asian and Melanesian forest frogs (Family Ceratobatrachidae). Zoological Journal of the Linnean Society  174: 130– 168. Google Scholar CrossRef Search ADS   Brunetti AE, Hermida GN, Faivovich J. 2012. New insights into sexually dimorphic skin glands of anurans: the structure and ultrastructure of the mental and lateral glands in Hypsiboas punctatus (Amphibia: Anura: Hylidae). Journal of Morphology  273: 1257– 1271. Google Scholar CrossRef Search ADS   Brunetti AE, Hermida GN, Luna MC, Barsotti AMG, Jared C, Antoniazzi MM, Rivera-Correa M, Berneck BVM, Faivovich J. 2015. Diversity and evolution of sexually dimorphic mental and lateral glands in Cophomantini treefrogs (Anura: Hylidae: Hylinae). Biological Journal of the Linnean Society  114: 12– 34. Google Scholar CrossRef Search ADS   Brunetti AE, Taboada CA, Faivovich J. 2014. Mating strategy of Hypsiboas punctatus (Anura: Hylidae): resource defense, female choice, and multimodal communication. Salamandra  50: 215– 224. Cadle JE. 1995. A new species of Boophis (Anura: Rhacophoridae) with unusual skin glands from Madagascar, and a discussion of variation and sexual dimorphism in Boophis albilabris (Boulenger). Zoological Journal of the Linnean Society  115: 313– 345. Google Scholar CrossRef Search ADS   Cei JM. 1948. El ritmo estacional en los fenómenos cíclicos endocrinos sexuales de la rana criolla, Leptodactylus ocellatus (L.) del norte Argentino. Acta Zoologica Lilloana  6: 283– 331. Ceresa-Castellani L. 1969. Osservazioni ultrastrutturali e istochimiche sulla pelle del Triturus cristatus Laur. Accademia Nazionale dei Lincei  46: 279– 285. Cisneros-Heredia DF, McDiarmid RW. 2007. Revision of the characters of Centrolenidae (Amphibia: Anura: Athesphatanura), with comments on its taxonomy and the description of new taxa of glassfrogs. Zootaxa  1572: 1– 82. Cochran DM. 1955. Frogs of Southeastern Brazil. Bulletin of the United States National Museum  206: 1– 423. Google Scholar CrossRef Search ADS   Coloma LA, Carvajal-Endara S, Dueñas JF, Paredes-Recalde A, Morales-Mite M, Almeida-Reinoso D, Tapia EE, Hutter CR, Toral E, Guayasamin JM. 2012. Molecular phylogenetics of stream treefrogs of the Hyloscirtus larinopygion group (Anura: Hylidae), and description of two new species from Ecuador. Zootaxa  3364: 1– 78. Conaway C, Metter DE. 1967. Skin glands associated with breeding in Microhyla carolinensis. Copeia  1967: 672– 673. Google Scholar CrossRef Search ADS   Cunningham M. 2002. Identification and evolution of Australian torrent treefrogs (Anura: Hylidae: Litoria nannotis group). Memoirs of the Queensland Museum  48: 93– 102. d’Istria M, Citarella F, Iela L, Delrio G. 1979. Characterization of a cytoplasmic androgen receptor in the male secondary sexual character of green frog (Rana esculenta). Journal of Steroid Biochemistry  10: 53– 59. Google Scholar CrossRef Search ADS   d’Istria M, Delrio G, Chieffi G. 1975. Receptors for sex hormones in the skin of the amphibia. General and Comparative Endocrinology  26: 281– 283. Google Scholar CrossRef Search ADS   d’Istria M, Picilli A, Basile C, Delrio G, Chieffi G. 1982. Morphological and biochemical variations in the skin of Rana esculenta during the annual cycle. General and Comparative Endocrinology  48: 20– 24. Google Scholar CrossRef Search ADS   Darwin C. 1871. The descent of man . London: William Clowes and sons. Dawson AB. 1920. The integument of Necturus maculosus. Journal of Morphology  34: 486– 589. Google Scholar CrossRef Search ADS   Delfino G, Brizzi R, Alvarez BB, Kracke-Berndorff R. 1998. Serous cutaneous glands in Phyllomedusa hypochondrialis (Anura, Hylidae): secretory patterns during ontogenesis. Tissue and Cell  30: 30– 40. Google Scholar CrossRef Search ADS   Delfino G, Brizzi R, Calloni C. 1986. Mixed cutaneous glands in Amphibia: an ultrastructural study on urodele larvae. Zoologische Jahrbücher. Abteilung für Anatomie und Ontogenie der Tiere  114: 325– 344. Di Fiore MM, Baccari GC, Rastogi RK, di Meglio M, Pinelli C, Iela L. 2005. Hormonal regulation of secondary sexual characters. In: Heatwole H, ed. Amphibian biology, Vol. 6. Endocrinology . Chipping Norton, NSW, Australia: Surrey Beatty and Sons, 2228– 2249. Diaz NF, Valencia J, Sallaberry M. 1983. Life history and phylogenetic relationships of Insuetophrynus acarpicus (Anura: Leptodactylidae). Copeia  1983: 30– 37. Google Scholar CrossRef Search ADS   Dolder H. 1976. Ultrastructure et formation des crochets sexuels chez ‘Xenopus’. Revue Suisse de Zoologie  82: 716– 718. Drewes RC. 1984. A phylogenetic analysis of the Hyperoliidae (Anura): treefrogs of Africa, Madagascar, and the Seychelles islands. Occasional Papers of the California Academy of Sciences  139: 1– 70. Dubois A, Ohler A. 2009. A new species of the genus Quasipaa (Anura, Ranidae, Dicroglossinae) from northern Vietnam. Alytes  27: 49– 61. Duellman WE. 1970. Hylid frogs of Middle America. Monographs of the Museum of Natural History, University of Kansas  1–2: 1– 753. Duellman WE. 2001. Hylid frogs of Middle America , expanded edn. Ithaca, NY: Society for the Study of Amphibians and Reptiles. Duellman WE. 2015. Marsupial frogs: Gastrotheca and allied genera . Baltimore: Johns Hopkins University Press. Duellman WE, Campbell JA. 1992. Hylid frogs of the genus Plectrohyla: systematics and phylogenetic relationships. Miscellaneous Publications of the Museum of Zoology, University of Michigan  181: 1– 32. Duellman WE, De La Riva I, Wild ER. 1997. Frogs of the Hyla armata and Hyla pulchella groups in the Andes of South America, with definitions and analyses of phylogenetic relationships of Andean groups of Hyla. Scientific Papers of the Natural History Museum, The University of Kansas  3: 1– 41. Duellman WE, Lehr E. 2009. Terrestrial-breeding frogs (Strabomantidae) in Peru . Münster: Natur und Tier Verlag. Duellman WE, Trueb L. 1986. Biology of amphibians . New York, NY: McGraw-Hill. Eagleson GW, van der Heijden RA, Roubos EW, Jenks BG. 2010. A developmental analysis of periodic albinism in the amphibian Xenopus laevis. General and Comparative Endocrinology  168: 302– 306. Google Scholar CrossRef Search ADS   Ecker A, Wiedersheim R. 1882. Die Anatomie des Frosches . Braunschweig: Vieweg. Emerson SB, Greig A, Carroll L, Prins GS. 1999. Androgen receptors in two androgen-mediated, sexually dimorphic characters of frogs. General and Comparative Endocrinology  114: 173– 180. Google Scholar CrossRef Search ADS   Epstein MS, Blackburn DG. 1997. Histology and histochemistry of androgen-stimulated nuptial pads in the leopard frog, Rana pipiens, with notes on nuptial gland evolution. Canadian Journal of Zoology  75: 472– 477. Google Scholar CrossRef Search ADS   Fabrezi M, Alberch P. 1996. The carpal elements of anurans. Herpetologica  52: 188– 204. Faivovich J, De la Riva I. 2006. On “Hyla”chlorostea Reynolds and Foster, 1992, a hylid of uncertain relationships, with some comments on Hyloscirtus (Anura: Hylidae). Copeia  2006: 785– 791. Google Scholar CrossRef Search ADS   Faivovich J, Haddad CFB, Garcia PCA, Frost DR, Campbell JA, Wheeler WC. 2005. Systematic review of the frog family Hylidae, with special reference to Hylinae: phylogenetic analysis and taxonomic revision. Bulletin of the American Museum of Natural History  294: 1– 240. Google Scholar CrossRef Search ADS   Faivovich J, Lugli L, Lourenço ACC, Haddad CFB. 2009. A new species of the Bokermannohyla martinsi species group from central Bahia, Brazil (Anura: Hylidae) with comments on Bokermannohyla. Herpetologica  65: 303– 310. Google Scholar CrossRef Search ADS   Fitch HS. 1956. A field study of the Kansas ant-eating frog, Gastrophryne olivacea. University of Kansas Publications, Museum of Natural History  8: 275– 306. Flores G. 1985. A new Centrolenella (Anura) from Ecuador, with comments on nuptial pads and prepollical spines in Centrolenella. Journal of Herpetology  19: 313– 320. Google Scholar CrossRef Search ADS   Forbes MS, Dent JN, Singhas CA. 1975. The developmental cytology of the nuptial pad in the red-spotted newt. Developmental Biology  46: 56– 78. Google Scholar CrossRef Search ADS   Formas JR. 1976. Descriptions of Batrachyla (Amphibia, Anura, Leptodactylidae) tadpoles. Journal of Herpetology  10: 221– 225. Google Scholar CrossRef Search ADS   Formas JR, Núñez JJ, Brieva LM. 2001. Osteología, taxonomía y relaciones filogenéticas de las ranas del género Telmatobufo (Leptodactylidae). Revista Chilena de Historia Natural  74: 365– 387. Google Scholar CrossRef Search ADS   Fox H. 1986. III, The skin of Amphibia: epidermis. In: Bereiter-Hahn J, Matoltsy AG, Richards KS, eds. Biology of the integument. 2. Vertebrates . Heidelberg: Springer Verlag, 78– 110. Fraser RDB, MacRae TP, Rogers GE. 1972. Keratins: their composition, structure, and biosynthesis . Springfield, IL: Charles C. Thomas. Fujikura K, Kurabuchi S, Tabuchi M, Inoue S. 1988. Morphology and distribution of the skin glands in Xenopus laevis and their response to experimental stimulations. Zoological Science  5: 415– 430. Garcia PCA, Faivovich J, Haddad CFB. 2007. Redescription of Hypsiboas semiguttatus, with the description of a new species of the Hypsiboas pulchellus group. Copeia  2007: 933– 949. Google Scholar CrossRef Search ADS   Garcia PCA, Haddad CFB. 2008. Vocalizations and comments on the relationbships of Hypsiboas ericae (Amphibia, Hylidae). Iheringia, Série. Zoologia  98: 161– 166. Google Scholar CrossRef Search ADS   Gaupp E. 1904. A. Ecker’s und R. Wiedersheim’s Anatomie des Frosches. Lehre von den Eigeweiden, dem Integument und den Sinnesorganen . Braunschweig: Vieweg und Sohn. Gay C. 2002. Stickiness—some fundamentals of adhesion. Integrative and Comparative Biology  42: 1123– 1126. Google Scholar CrossRef Search ADS   Glass FM, Rugh R. 1944. Seasonal study of the normal and pituitary stimulated frog (Rana pipiens). I. Testis and thumb pad. Journal of Morphology  74: 409– 427. Google Scholar CrossRef Search ADS   Glaw F, Vences M. 2007. A field guide to the amphibians and reptiles of Madagascar , 3rd edn. Cologne: Vences & Glaw Verlag. Goldner J. 1938. A modification of the Masson trichrome technique for routine laboratory purposes. The American Journal of Pathology  14: 237– 243. Gonçalves VF, de Brito-Gitirana L. 2008. Structure of the sexually dimorphic gland of Cycloramphus fuliginosus (Amphibia, Anura, Cycloramphidae). Micron  39: 32– 39. Google Scholar CrossRef Search ADS   Grandison AGC. 1980. Aspects of breeding morphology in Mertensophryne micranotis (Anura: Bufonidae): secondary sexual characters, eggs and tadpole. Bulletin of the British Museum  39: 299– 304. Grant T, Frost DR, Caldwell JP, Gagliardo R, Haddad CFB, Kok PJR, Means DB, Noonan BP, Schargel WE, Wheeler WC. 2006. Phylogenetic systematics of dart-poison frogs and their relatives (Amphibia: Athesphatanura: Dendrobatidae). Bulletin of the American Museum of Natural History  299: 1– 262. Google Scholar CrossRef Search ADS   Greenberg B. 1942. Some effects of testosterone on the sexual pigmentation and other sex characters of the cricket frog (Acris gryllus). Journal of Experimental Zoology Part A: Ecological Genetics and Physiology  91: 435– 451. Google Scholar CrossRef Search ADS   Guayasamin JM, Castroviejo-Fisher S, Trueb L, Ayarzagüena J, Rada M, Vilá C. 2009. Phylogenetic systematics of Glassfrogs (Amphibia: Centrolenidae) and their sister taxon Allophryne ruthveni. Zootaxa  2100: 1– 97. Guayasamin JM, Rivera-Correa M, Arteaga A, Culebras J, Bustamante L, Pyron RA, Peñafiel N, Morochz C, Hutter CR. 2015. Molecular phylogeny of stream treefrogs (Hylidae: Hyloscirtus bogotensis Group), with a new species from the Andes of Ecuador. Neotropical Biodiversity  1: 2– 21. Google Scholar CrossRef Search ADS   Haddad CFB, Hödl W. 1997. New reproductive mode in anurans: bubble nest in Chiasmocleis leucosticta (Microhylidae). Copeia  1997: 585– 588. Google Scholar CrossRef Search ADS   Harvey LA, Propper CR. 1997. Effects of androgens on male sexual behavior and secondary sex characters in the explosively breeding spadefoot toad, Scaphiopus couchii. Hormones and Behavior  31: 89– 96. Google Scholar CrossRef Search ADS   Harvey MB, Pemberton AJ, Smith E. 2002. New and poorly known parachuting frogs (Rhacophoridae: Rhacophorus) from Sumatra and Java. Herpetological Monographs  16: 46– 92. Google Scholar CrossRef Search ADS   Heyer WR. 1969. Studies on frogs of the genus Leptodactylus (Amphibia, Leptodactylidae). V. Taxonomic notes on L. latinasus, rhodonotus, romani, and wuchereri. Herpetologica  25: 1– 8. Heyer WR. 1975. A preliminary analysis of the intergeneric relationships of the frog family Leptodactylidae. Smithsonian Contributions to Zoology  199: 1– 55. Google Scholar CrossRef Search ADS   Heyer WR. 1983. Variation and systematics of frogs of the genus Cycloramphus (Amphibia, Leptodactylidae). Arquivos de Zoologia  30: 235– 339. Google Scholar CrossRef Search ADS   Heyer WR. 1998. The relationships of Leptodactylus diedrus (Anura, Leptodactylidae). Alytes  16: 1– 24. Heyer WR. 2005. Variation and taxonomic clarification of the large species of the Leptodactylus pentadactylus species group (Amphibia: Leptodactylidae) from Middle America, northern South America and Amazonia. Arquivos de Zoologia  37: 269– 348. Google Scholar CrossRef Search ADS   Heyer WR, de Sá RO. 2011. Variation, systematics, and relationships of the Leptodactylus bolivianus complex (Amphibia: Anura: Leptodactylidae). Smithsonian Contributions to Zoology  635: 1– 58. Google Scholar CrossRef Search ADS   Holloway WR, Dapson RW. 1971. Histochemistry of integumentary secretions of the narrowmouth toad, Gastrophryne carolinensis. Copeia  1971: 351– 353. Google Scholar CrossRef Search ADS   Hoperskaya OA. 1975. The development of animals homozygous for a mutation causing periodic albinism (ap) in Xenopus laevis. Journal of Embryology and Experimental Morphology  34: 253– 264. Horié H. 1939. Appearance of thumb pads in the metamorphosing Rana temporaria by injecting the male hormone. Proceedings of the Imperial Academy  15: 362– 363. Houck LD, Sever DM. 1994. Role of the skin in reproduction and behaviour. In: Heatwole H, Barthalamus G, eds. Amphibian biology, Vol. 1. The integument . Chipping Norton, NSW, Australia: Surrey Beatty & Sons, 351– 381. Houssay BA, Giusti L, Lascano Gonzalez JM. 1929. Implantación de la hipófisis y estimulación sexual en el sapo. Revista de la Sociedad Argentina de Biologia  5: 397– 418. Hudson CM, He X, Fu J. 2011. Keratinized nuptial spines are used for male combat in the Emei moustache toad (Leptobrachium boringii). Asian Herpetological Research  2: 142– 148. Google Scholar CrossRef Search ADS   Inger RF. 1954. Systematics and zoogeography of Philippine Amphibia. Fieldiana, Zoology  33: 181– 531. Inger RF. 1966. The systematics and zoogeography of the Amphibia of Borneo. Fieldiana, Zoology  52: 1– 402. Inger RF, Greenberg B. 1956. Morphology and seasonal development of sexcharacters in two sympatric African toads. Journal of Morphology  99: 549– 574. Google Scholar CrossRef Search ADS   Inger RF, Greenberg B. 1963. The annual reproductive pattern of the frog Rana erythraea in Sarawak. Physiological Zoology  36: 21– 33. Google Scholar CrossRef Search ADS   Iwai N. 2012. Morphology, function and evolution of the pseudothumb in the Otton frog. Journal of Zoology  289: 127– 133. Google Scholar CrossRef Search ADS   Iwasawa H, Kobayashi M. 1974. Effects of testosterone and estradiol on the development of sexual characters in young Rana nigromaculata. Biology of Reproduction  11: 398– 405. Google Scholar CrossRef Search ADS   Izzo I, Di Matteo L, Minucci S, Iela L, Di Meglio M, Rastogi RK. 1982. The control of the frog (Rana esculenta) thumb pad. Experientia  38: 134– 135. Google Scholar CrossRef Search ADS   Jiang K, Rao D, Yuan S, Wang J, Li P, Hou M, Che M, Che J. 2012. A new species of the genus Scutiger (Anura: Megophryidae) from southeastern Tibet, China. Zootaxa  3388: 29– 40. Jiang K, Wang K, Zou DH, Yan F, Li PP, Che J. 2016. A new species of the genus Scutiger (Anura: Megophryidae) from Medog of southeastern Tibet, China. Zoological Research  37: 21– 30. Kanamadi RD, Saidapur SK. 1981. The effect of short-term treatment of methallibure [ICI compound 33, 828] on the histomorphological and enzymatic aspects of testis and thumb pad of toad, Bufo melanostictus. Current Science  50: 623– 624. Kändler R. 1924. Die sexuelle Ausgestaltung der Vorderextremität der anuren Amphibien. Jenaische Zeitschrift für Naturwissenschaft  60: 175– 240. Kaptan E, Bolkent S. 2014. Seasonal lectin binding variations of thumb pad in the frog (Pelophylax ridibundus). Journal of Morphology  275: 76– 86. Google Scholar CrossRef Search ADS   Kaptan E, Murathanoğlu O. 2008. Annual morphological cycles of testis and thumb pad of the male frog (Rana ridibunda). Anatomical Record  291: 1106– 1114. Google Scholar CrossRef Search ADS   Kaung HLC. 1975. Development of beaks of Rana pipiens larvae. The Anatomical Record  182: 401– 413. Google Scholar CrossRef Search ADS   Kelley DB, Pfaff DW. 1976. Hormone effects on male sex behavior in adult South African clawed frogs, Xenopus laevis. Hormones and Behavior  7: 159– 182. Google Scholar CrossRef Search ADS   Kluge AG. 1981. The life history, social organization, and parental behavior of Hyla rosenbergi Boulenger, a nest-building gladiator frog. Miscellaneous Publications, Museum of Zoology, University of Michigan  160: 1– 170. Köhler G. 2011. Amphibians of Central America . Offenbach: Herpeton Verlag. Komnick H, Stockem W. 1970. Die Feinstruktur der Epidermisoberfläche an den Extremitäten des Krallenfrosches. Eine Untersuchung über zelluläre Formbildungen. Archivum Histologicum Japonicum  32: 17– 40. Google Scholar CrossRef Search ADS   Kurabuchi S. 1993. Fine structure of nuptial pad surface of male ranid frogs. Tissue and Cell  25: 589– 598. Google Scholar CrossRef Search ADS   Kurabuchi S. 1994. Fine structures on the surface of nuptial pads of male Hylid and Rhacophorid frogs. Journal of Morphology  219: 173– 182. Google Scholar CrossRef Search ADS   Kurabuchi S, Inoue S. 1981. Small spiny projections in the epidermis of the mature Xenopus laevis. Annotationes Zoologicae Japonenses  54: 182– 190. Kyriakopoulou-Sklavounou P, Papaevangelou E, Kladisios N. 2012. A scanning electron microscopic study of the surface morphology of nuptial pads in male amphibians (Genus: Bombina, Pelophylax, Rana). Acta Herpetologica  7: 81– 90. Lataste F. 1876. Mémoire sur les brosses copulatrices des Batraciens Anoures. Annales des Sciences Naturelles, Zoologie  6: 1– 10. Le Quang Trong Y. 1976. Étude de le peau et des glandes cutanées de quelques Amphibiens de la famille des Rhacophoride. Bulletin de l’Institute Fondamental d’Afrique Noire  38: 166– 187. Lehr E, Trueb L. 2007. Diversity among New World microhylid frogs (Anura: Microhylidae): morphological and osteological comparisons between Nelsonophryne (Günther 1901) and a new genus from Peru. Zoological Journal of the Linnean Society  149: 583– 609. Google Scholar CrossRef Search ADS   Leite, FSF, Pezzuti TL, Drummond LO. 2011. A new species of Bokermannohyla from the Espinhaço range, State of Minas Gerais, Southeastern Brazil. Herpetologica  67: 440– 448. Google Scholar CrossRef Search ADS   Leydig F. 1877. Die Anuren Batrachier der Deutschen Fauna . Bonn: Max Cohen & Sohn. Liu C. 1936. Secondary sex characters of Chinese frogs and toads. Field Museum of Natural History. Zoological Series  22: 115– 156. Liu C. 1950. Amphibians of Western China. Fieldiana Zoology, Memoirs  2: 1– 400. Liu CH, Lin CH, Tsai MJ, Chen WT, Chai CY, Huang YC, Tsai KB. 2013. Melanin bleaching with dilute hydrogen peroxide: a simple and rapid method. Applied Immunohistochemistry & Molecular Morphology  21: 275– 279. Google Scholar CrossRef Search ADS   Lofts B. 1964. Seasonal changes in the functional activity of the interstitial and spermatogenetic tissues of the green frog, Rana esculenta. General and Comparative Endocrinology  4: 550– 562. Google Scholar CrossRef Search ADS   Lofts B, Wellen JJ, Benraad TJ. 1972. Seasonal changes in endocrine organs of the male common frog, Rana temporaria. III. The gonads and cholesterol cycles. General and Comparative Endocrinology  18: 344– 363. Google Scholar CrossRef Search ADS   Lourenço AC, Luna MC, Pombal JPJr. 2014. A new species of the Scinax catharinae Group (Anura: Hylidae) from Northeastern Brazil. Zootaxa  3889: 259– 276. Google Scholar CrossRef Search ADS   Luckenbill LM. 1965. Morphogenesis of the horny jaws of Rana pipiens larvae. Developmental Biology  11: 25– 49. Google Scholar CrossRef Search ADS   Luna MC, Taboada C, Baêta D, Faivovich J. 2012. Structural diversity of nuptial pads in Phyllomedusinae (Amphibia: Anura: Hylidae). Journal of Morphology  273: 712– 724. Google Scholar CrossRef Search ADS   Lutz A. 1931. Observações sobre batrachios Brasileiros. Taxonomia e biologia das Elosiinas. Memoria do Instituto Oswaldo Cruz  24: 195– 222. Google Scholar CrossRef Search ADS   Lynch JD. 1971. Evolutionary relationships, osteology, and zoogeography of leptodactyloid frogs. Miscellaneous Publications of the Museum of Natural History, University of Kansas  53: 1– 238. Lynch JD. 1973. The transition from archaic to advanced frogs. In: Vial JL ed. Evolutionary biology of the anurans. Contemporary research on major problems . Columbia, MO: University of Missouri Press. Lynch JD. 1978. A re-assessment of the telmatobiine leptodactylid frogs of Patagonia. Occasional Papers of the Museum of Natural History, University of Kansas  72: 1– 57. Lynch LC, Blackburn DG. 1995. Effects of testosterone administration and gonadectomy on nuptial pad morphology in overwintering male leopard frogs, Rana pipiens. Amphibia-Reptilia  16: 113– 121. Google Scholar CrossRef Search ADS   Lynch JD, Duellman WE. 1997. Frogs of the genus Eleutherodactylus in western Ecuador. University of Kansas Natural History Museum, Special Publication  23: 1– 236. Lynch JD, Ruiz-Carranza PM. 1996. A remarkable new centrolenid frog from Colombia with a review of nuptial excrescences in the family. Herpetologica  52: 525– 535. Maddin HC, Eckhart L, Jaeger K, Russell AP, Ghannadan M. 2009. The anatomy and development of the claws of Xenopus laevis (Lissamphibia: Anura) reveal alternate pathways of structural evolution in the integument of tetrapods. Journal of Anatomy  214: 607– 619. Google Scholar CrossRef Search ADS   Maddin HC, Musat-Marcu S, Reisz RR. 2007. Histological microstructure of the claws of the African clawed frog, Xenopus laevis (Anura: Pipidae): implications for the evolution of claws in tetrapods. Journal of Experimental Zoology. Part B, Molecular and Developmental Evolution  308: 259– 268. Google Scholar CrossRef Search ADS   Manamendra-Arachchi K, Pethiyagoda R. 2005. The Sri Lankan shrub-frogs of the genus Philautus Gistel, 1848 (Ranidae: Rhacophorinae), with description of 27 new species. The Raffles Bulletin of Zoology  12: 163– 303. Matsui M. 2009. A new species of Kalophrynus with a unique male humeral spine from Peninsular Malaysia (Amphibia, Anura, Microhylidae). Zoological Science  26: 579– 585. Google Scholar CrossRef Search ADS   Meisenheimer J. 1912. Über den Zusammenhang zwischen Geschlechtsdrüsen und sekundären Geschlechtsmerkmalen bei Froschën. Zoologische Jahrbücher  15: 191– 218. Metter DE, Conaway CH. 1969. The influence of hormones on the development of breeding glands in Microhyla. Copeia  1969: 621– 622. Google Scholar CrossRef Search ADS   Mills JW, Prum BE. 1984. Morphology of the exocrine glands of the frog skin. The American Journal of Anatomy  171: 91– 106. Google Scholar CrossRef Search ADS   Neuwirth M, Daly JW, Myers CW, Tice LW. 1979. Morphology of the granular secretory glands in skin of poison-dart frogs (Dendrobatidae). Tissue and Cell  11: 755– 771. Google Scholar CrossRef Search ADS   Nguyen TQ, Phung TM, Le MD, Ziegler T, Böhme W. 2013. First record of the genus Oreolalax (Anura: Megophryidae) from Vietnam with description of a new species. Copeia  2013: 213– 222. Google Scholar CrossRef Search ADS   Noble GA, Noble ER. 1944. On the histology of frog skin glands. Transactions of the American Microscopical Society  63: 254– 263. Google Scholar CrossRef Search ADS   Noble GK. 1931. The biology of the amphibia . New York, NY: McGraw-Hill. Google Scholar CrossRef Search ADS   Noble GK, Putnam PG. 1931. Observations on the life history of Ascaphus truei Stejneger. Copeia  1931: 97– 101. Google Scholar CrossRef Search ADS   Ohler A, Dubois A. 2006. Phylogenetic relationships and generic taxonomy of the tribe Paini (Amphibia, Anura, Ranidae, Dicroglossinae), with diagnoses of two new genera. Zoosystema  28: 769– 784. Olson KR, Fromm PO. 1973. A scanning electron microscopic study of secondary lamellae and chloride cells of rainbow trout (Salmo gairdneri). Zeitschrift fur Zellforschung und mikroskopische Anatomie  143: 439– 449. Google Scholar CrossRef Search ADS   Parakkal PF, Ellis RA. 1963. A cytochemical and electron microscopic study of the thumb pad in Rana pipiens. Experimental Cell Research  32: 280– 288. Google Scholar CrossRef Search ADS   Parker HW. 1934. A monograph of the frogs of the family Microhylidae . London: British Museum. Parker HW. 1940. The Australasian frogs of the family Leptodactylidae. Novitates Zoologicae  42: 1– 106. Perret JL. 1966. Les amphibiens du Cameroun. Zoolgische Jahrbücher  93: 289– 464. Peters JA. 1964. Dictionary of herpetology . New York, NY: Hafner. Pimenta BVS, Cruz CAG, Caramaschi U. 2014. Taxonomic review of the species complex of Crossodactylus dispar A. Lutz, 1925 (Anura, Hylodidae). Arquivos de Zoologia  45: 1– 33. Google Scholar CrossRef Search ADS   Pombal JPJr, Cruz CAG. 2016. Records on breeding behaviour of a rare neotropical microhylid frog, Dasypops schirchi (Gastrophryninae). Herpetology Notes  9: 317– 320. Ponse K. 1924. L’organe de Bidder et le determinisme des caracteres sexuels secondaires du Crapaud (Bufo vulgaris, L.). Revue Suisse de Zoologie  31: 177– 336. Ponssa ML. 2008. Cladistic analysis and osteological descriptions of the frog species in the Leptodactylus fuscus species group (Anura, Leptodactylidae). Journal of Zoological Systematics and Evolutionary Research  46: 249– 266. Google Scholar CrossRef Search ADS   Pope CH. 1931. Notes on amphibians from Fukien, Hainan, and other parts of China. Bulletin of the American Museum of Natural History  61: 397– 611. Quay WB. 1972. Integument and the environment glandular composition, function, and evolution. American Zoologist  12: 95– 108. Google Scholar CrossRef Search ADS   Rasband WS. 1997–2016. ImageJ . Bethesda, MD: US National Institutes of Health. Available at: https://imagej.nih.gov/ij Rastogi RK. 1976. Seasonal cycle in anuran (Amphibia) testis: the endocrine and environmental controls. Italian Journal of Zoology  43: 151– 172. Rastogi RK, Iela L, Delrio G, Bagnara JT. 1986. Reproduction in the Mexican leaf frog, Pachymedusa dacnicolor. II. The male. General and Comparative Endocrinology  62: 23– 35. Google Scholar CrossRef Search ADS   Rastogi RK, Iela L, Delrio G, Di Meglio M, Russo A, Chieffi G. 1978. Environmental influence on testicular activity in the green frog, Rana esculenta. Journal of Experimental Zoology  206: 49– 63. Google Scholar CrossRef Search ADS   Rastogi RK, Iela L, Saxena PK, Chieffi G. 1976. The control of spermatogenesis in the green frog, Rana esculenta. Journal of Experimental Zoology  196: 151– 165. Google Scholar CrossRef Search ADS   Resing KA, Dale BA. 1991. Proteins of keratohyalin. In: Goldsmith LA, ed. Biochemistry, Physiology and Molecular Biology of the Skin , 2nd edn. New York: Oxford University Press. Reynolds ES. 1963. The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. The Journal of Cell Biology  17: 208– 212. Google Scholar CrossRef Search ADS   Rivera-Correa M, Faivovich J. 2013. A new species of Hyloscirtus (Anura: Hylidae) from Colombia with comments on Hyloscirtus larinopygion (Duellman, 1973). Herpetologica  69: 298– 313. Google Scholar CrossRef Search ADS   Rivero JA, Esteves AE. 1969. Observations on the agonistic and breeding behavior of Leptodactylus pentadactylus and other amphibian species in Venezuela. Breviora  321: 1– 14. Rösel von Rosenhof AJ. 1758. Historia naturalis ranarum nostratium . Nürnberg: J.J. Fleishmann. Rostand J. 1933. La vie des crapauds . Paris: Libraire Stock. Rusconi M. 1821. Amours des salamandres aquatiques et development du tetard de ces salamandres depuis l’ouef jusqu’a l’animal parfait . Milan: P. E. Giusti. de Sá RO, Grant T, Camargo A, Heyer WR, Ponssa ML, Stanley E. 2014. Systematics of the neotropical genus Leptodactylus Fitzinger, 1826 (Anura: Leptodactylidae): phylogeny, the relevance of non-molecular evidence, and species accounts. South American Journal of Herpetology  9: S1– S100. Google Scholar CrossRef Search ADS   Sabaj MH. 2016. Standard symbolic codes for institutional resource collections in herpetology and ichthyology: an online reference , Version 6.5 ( 16 August 2016). Washington, DC: American Society of Ichthyologists and Herpetologists. Available at: http://www.asih.org/ Saidapur SK. 1983. Patterns of testicular activity in Indian amphibians. Indian Review of Life Sciences  3: 157– 184. Saidapur SK, Nadkarni VB. 1975a. The effect of 17 β-estradiol acetate on the testis and thumb pad of Rana cyanophlyctis (Schn.) and Rana tigrina (Daud.). General and Comparative Endocrinology  27: 350– 357. Google Scholar CrossRef Search ADS   Saidapur SK, Nadkarni VB. 1975b. The annual testicular cycle and its relation to thumb pad structure in the frog, Rana cyanophlyctis (Schn.). The Journal of Animal Morphology and Physiology  22: 140– 147. Savage JM. 2002. The amphibians and reptiles of Costa Rica. A herpetofauna between two continents, between two seas . Chicago, IL: The University of Chicago Press. Savage RM. 1932. The spawning, voice, and sexual behavior of Bombina variegata variegata. Proceedings of the Zoological Society of London  102: 889– 898. Google Scholar CrossRef Search ADS   Savage RM. 1934. The breeding behavior of the Common Frog, Rana temporaria temporaria Linn., and of the Common Toad, Bufo bufo bufo Linn. Proceedings of the Zoological Society of London  104: 55– 65. Google Scholar CrossRef Search ADS   Savage RM. 1961. The ecology and life history of the common frog . London: Pitman. Scrocchi GJ, Lavilla EO. 1990. Elachistocleis bicolor amplexus behavior. Herpetological Review  21: 18. Scott E. 2005. A phylogeny of ranid frogs (Anura: Ranoidea: Ranidae), based on a simultaneous analysis of morphological and molecular data. Cladistics  21: 507– 574. Google Scholar CrossRef Search ADS   Señaris JC, Ayarzagüena J. 2005. Revisión taxonómica de la familia Centrolenidae (Amphibia: Anura) de Venezuela . Sevilla: Publicaciones del Comité Español del Programa Hombre y Biosfera–Red IberoMaB de la UNESCO. Sever DM, Staub NL. 2011. Hormones, sex accessory structures, and secondary sexual characteristics in amphibians. In: Norris DO, Lopez KH, eds. Hormones and reproduction of vertebrates. Volume 2. Amphibians . London: Academic Press. Siegel DS, Sever DM, Schriever TA, Chabarria RE. 2008. Ultrastructure and histochemistry of the adhesive breeding glands in male Gastrophryne carolinensis (Amphibia: Anura: Microhylidae). Copeia  2008: 877– 881. Google Scholar CrossRef Search ADS   Smith M. 1949. British reptiles and amphibia . Harmondsworth: King Penguin Books. Smith AM. 2002. The structure and function of adhesive gels from invertebrates. Integrative and Comparative Biology  42: 1164– 1171. Google Scholar CrossRef Search ADS   Smith‐Gill SJ, Richards CM, Nace GW. 1972. Genetic and metabolic bases of two “albino” phenotypes in the leopard frog, Rana pipiens. Journal of Experimental Zoology  180: 157– 167. Google Scholar CrossRef Search ADS   Sondhi S, Ohler A. 2011. A blue-eyed Leptobrachium (Anura: Megophryidae) from Arunachal Pradesh, India. Zootaxa  2912: 28– 36. Sperry DG, Wassersug RJ. 1976. A proposed function for microridges on epithelial cells. The Anatomical Record  185: 253– 257. Google Scholar CrossRef Search ADS   Swammerdam J. 1737. Biblia naturae; sive historia insectorum, in classes certa redacta, nec non exemplis, et anatomico variorum animalculorum examine, aeneisque tabulis ilustrata , Vol. II. Leyden: I. Severinus. Thomas EO, Licht P. 1993. Testicular and androgen dependence of skin gland morphology in the anurans, Xenopus laevis and Rana pipiens. Journal of Morphology  215: 195– 200. Google Scholar CrossRef Search ADS   Thomas EO, Tsang L, Licht P. 1993. Comparative histochemistry of the sexually dimorphic skin glands of anuran amphibians. Copeia  1993: 133– 143. Google Scholar CrossRef Search ADS   Toledo RD, Jared C. 1995. Cutaneous granular glands and amphibian venoms. Comparative Biochemistry and Physiology Part A: Physiology  111: 1– 29. Google Scholar CrossRef Search ADS   Turner CD, Bagnara JT. 1976. General endocrinology . Philadelphia, PA: W. B. Saunders Company. Tyler MJ. 1968. Papuan hylid frogs of the genus Hyla. Zoologische Verhandelingen  96: 1– 203. Tyler MJ, Davies M. 1978. Species-groups within the Australopapuan hylid frog genus Litoria Tschudi. Australian Journal of Zoology Supplementary Series  63: 1– 47. Google Scholar CrossRef Search ADS   Tyler MJ, Lungershausen K. 1986. The ultrastructure of male nuptial pads in some Australopapuan frogs. Transactions of the Royal Society of South Australia  110: 37– 42. Varriale B, Serino I. 1994. The androgen receptor mRNA is up-regulated by testosterone in both the Harderian gland and thumb pad of the frog, Rana esculenta. The Journal of Steroid Biochemistry and Molecular Biology  51: 259– 265. Google Scholar CrossRef Search ADS   Vences M, Andreone F, Glos J, Glaw F. 2010. Molecular and bioacoustics differentiation of Boophis occidentalis with description of a new treefrog from north-western Madagascar. Zootaxa  2544: 54– 68. Vences M, Wahl-Boos G, Hoegg S, Glaw F, Oliveira ES, Meyer A, Perry S. 2007. Molecular systematics of mantelline frogs from Madagascar and the evolution of their femoral glands. Biological Journal of the Linnean Society  92: 529– 539. Google Scholar CrossRef Search ADS   Vera Candioti MF, Altig R. 2010. A survey of shape variation in keratinized labial teeth of anuran larvae as related to phylogeny and ecology. Biological Journal of the Linnean Society  101: 609– 625. Google Scholar CrossRef Search ADS   Visser J, Cei JM, Gutierrez LS. 1982. The histology of dermal glands of mating Breviceps with comments on their possible functional role in microhylids. South African Journal of Zoology  17: 24– 27. Google Scholar CrossRef Search ADS   Wells KD. 1977. The social behavior of anuran amphibians. Animal Behavior  25: 666– 693. Google Scholar CrossRef Search ADS   Wells KD. 2007. The ecology and behavior of amphibians . Chicago, IL: The University of Chicago Press. Google Scholar CrossRef Search ADS   Willaert B, Bossuyt F, Janssenswillen S, Adriaens D, Baggerman G, Matthijs S, Pauwels E, Proost P, Raepsaet A, Schoofs L, Stegen G, Treer D, Van Hoorebeke L, Vandebergh W, Van Bocxlaer I. 2013. Frog nuptial pads secrete mating season-specific proteins related to salamander pheromones. The Journal of Experimental Biology  216: 4139– 4143. Google Scholar CrossRef Search ADS   Willaert B, Suyesh R, Garg S, Giri VB, Bee MA, Biju SD. 2016. A unique mating strategy without physical contact during fertilization in Bombay Night Frogs (Nyctibatrachus humayuni) with the description of a new form of amplexus and female call. PeerJ  4: e2117. Google Scholar CrossRef Search ADS   Wray KP. 2000. Nuptial excrescences of Central American stream-breeding frogs: microstructure and phylogenetic significance . Unpublished MSc. Thesis, The University of Texas. Wright AH, Wright AW. 1949. Handbook of frogs and toads of the United States and Canada , 3rd edn. New York, NY: Comstock Publishings Associates. Wu SP, Huang CC, Tsai CL, Lin TE, Jhang JJ, Wu SH. 2016. Systematic revision of the Taiwanese genus Kurixalus members with a description of two new endemic species (Anura, Rhacophoridae). ZooKeys  557: 121– 153. Google Scholar CrossRef Search ADS   Wu GF, Zhao EM, Inger RF, Shaffer HB. 1993. A new frog of the genus Oreolalax (Pelobatidae) from Sichuan, China. Journal of Herpetology  27: 410– 413. Google Scholar CrossRef Search ADS   van Wyk JH, Pool EJ, Leslie AJ. 2003. The effects of anti-androgenic and estrogenic disrupting contaminants on breeding gland (nuptial pad) morphology, plasma testosterone levels, and plasma vitellogenin levels in male Xenopus laevis (African clawed frog). Archives of Environmental Contamination and Toxicology  44: 247– 256. Google Scholar CrossRef Search ADS   Yang C, Zhang Y, Cui H. 2004. Correlation between the changes of microstructure and testosterone in nuptial pad of the frog, Rana chensinensis. Zoological Research  26: 638– 644. Zhang W, Guo Y, Li J, Huang L, Kazitsa EG, Wu H. 2016. Transcriptome analysis reveals the genetic basis underlying the seasonal development of keratinized nuptial spines in Leptobrachium boringii. BMC Genomics  17: 978. Google Scholar CrossRef Search ADS   Zug GR. 2015. Morphology and systematics of Kalophrynus interlineatus-pleurostigma populations (Anura: Microhylidae: Kalophryninae) and a taxonomy of the genus Kalophrynus Tschudi, Asian Sticky Frogs. Proceedings of the California Academy of Sciences  62: 135– 190. Zweifel RG. 1983. Two new hylid frogs from Papua New Guinea and a discussion of the Nyctimystes papua species group. American Museum Novitates  2759: 1– 18. © 2018 The Linnean Society of London, Biological Journal of the Linnean Society This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Biological Journal of the Linnean Society Oxford University Press

From erotic excrescences to pheromone shots: structure and diversity of nuptial pads in anurans