© Firenze University Press www.fupress.com/ah Acta Herpetologica 5(2): 151-160, 2010 Distribution of tadpoles of Ollotis occidentalis (Amphibia: Anura: Bufonidae) along the Río Salado, Puebla, México Guillermo A. Woolrich-Piña1, Geoffrey R. Smith1,2, Luis Oliver-López1, Monica Barbosa Morales1, Julio A. Lemos-Espinal1 1Laboratorio de Ecología, Unidad de Biología, Tecnología y Prototipos (FES-Iztacala, UNAM), Av. de los Barrios S/N, 54090 Los Reyes Iztacala, Tlalnepantla, Estado de México, MEXICO. Corresponding author. E.mail: smithg@denison.edu 2Department of Biology, Denison University, 43023 Granville, Ohio, USA. Submitted on: 2010, 23th January; revised on 2010, 22th September; accepted on 2010, 6th October. Abstract. We examined monthly variation in the characteristics of river sections along the Río Salado (Puebla, Mexico) and how these factors were associated with the dis- tribution of the tadpoles of the toad Ollotis occidentalis. Tadpoles were observed in the river in March 2007, and from November 2007 through February 2008 and were only found in the main river channel. Sections of the Río Salado with tadpoles were deeper, wider, and longer than sections without tadpoles. Dissolved oxygen levels were higher and salinity was lower in river sections with tadpoles compared to the sections without tadpoles. There was no difference in temperature between sections with and without tadpoles. Tadpoles were found in river sections that contained more vegeta- tion than river section without tadpoles. Our results suggest that the distribution of tadpoles of O. occidentalis is related to the permanence of water, the chemical nature of the water, and the presence of vegetation. Keywords. Dissolved oxygen, Ollotis occidentalis, pool size, salinity, tadpoles. INTRODUCTION One of the major goals of ecology is to understand what determines species distri- butions. For pond-, pool-, or stream-breeding amphibians the relevant question is often “Why are tadpoles found in some locations but not others?” In some cases, the distribu- tion of amphibians appears driven by the chemical or physical characteristics of the pool, pond, or stream (e.g., Eason and Fauth, 2001; Richards, 2002; Rocha et al., 2002; Halver- son et al., 2003; Welch and MacMahon, 2005; Girish and Krishnamurthy, 2009). In oth- er cases, it appears that the distribution can be explained by the biotic community (e.g., Blaustein and Margalit, 1995; Eason and Fauth, 2001). In addition, the interaction of the 152 G.A. Woolrich-Piña et alii chemical or physical environment with the biotic environment can potentially determine the success of amphibian larvae, and thus could affect their distribution (e.g., Sadinski and Dunson, 1992; Warner et al., 1993; Skelly, 1996; Smith et al., 2006). We investigated the distribution of the tadpoles of Ollotis [Bufo] occidentalis along the Río Salado (Puebla, Mexico) which, as its name suggests, can have relatively high salin- ity. Amphibians are typically not found in saline conditions, and are usually thought to be unable to withstand elevated salinities (Ultsch et al., 1999). However, several species of amphibians (Exerodonta xera, Hyla arenicolor, Lithobates spectabilis, and O. occidentalis) are known to occur in the Río Salado. We examined monthly variation in water chemistry (salinity, dissolved oxygen), water temperature, physical characteristics (size, distance from main channel), and vegetative cover of river sections and we evaluated how these factors differed between sections of the Río Salado with and without tadpoles of O. occidentalis. Ollotis occidentalis occurs from south-central to northern Mexico. In southern Puebla, O. occidentalis occurs in pine forest and fry tropical deciduous forests (Oliver-López et al., 2000). In the Valle de Zapotitlán Salinas, O. occidentalis breeds during the rainy season in slow moving water or in pools on the edges of rivers (Oliver-López et al., 2000; Smith and Lemos-Espinal, 2010). Very little is known about the ecology and natural history of this toad beyond reports on its reproduction (Duellman, 1961; Oliver-López et al., 2000; Oliver-López and Ramírez-Bautista, 2002; Smith and Lemos-Espinal, 2010). MATERIALS AND METHODS Study Area The Río Salado (see Fig. 1) runs through El Valle de Zapotitlán Salinas in southeastern Puebla, Mexico (see Woolrich Piña et al. 2005 for additional details). Pools along the Río Salado are seasonal, forming during the dry season as the water level in the Río Salado falls. El Valle de Zapotitlán is part of the Valle de Tehuacan-Cuicatlán that lies in central Mexico, which is considered to be an eco- logically important region due to its high levels of biodiversity and endemism (Dávila-Aranda et al., 1993). Surveys of the herpetofauna have reported between 22 and 32 species of amphibians and rep- tiles (Martín del Campo and Sánchez-Herrera, 1979; Canseco-Márquez and Gutiérrez-Mayén, 1996). More recently, there have been studies examining the effects of habitat deterioration on amphibians and reptiles (Mata-Silva, 2000) and the reproduction of the anurans in the Valle de Zapotitlán Sali- nas (Oliver-López et al., 2000; Oliver-López and Ramírez-Bautista, 2002). Woolrich Piña et al. (2005) provide a summary of the amphibians and reptiles of El Valle de Zapotitlán Salinas. Methods We conducted surveys along a 2 km segment of the Río Salado (Fig. 1) monthly from Febru- ary 2007 to June 2008 to determine the distribution of tadpoles of O. occidentalis and to characterize conditions along the Río Salado. The tadpole of this species could be actually differentiated from those of other co-occuring species in the field. We surveyed the first 10 – 19 sections of river or isolated pools we encountered on each of our surveys. We generally began the surveys in the same location and proceeding in the same direction along the river and so many of the river sections were similar among visits; however, on some visits the starting location and direction moved varied 153Distribution of Ollotis occidentalis tadpoles (e.g., when diff erent researchers conducted a monthly survey). In addition, because isolated pools dried up during some parts of the year, the selected isolated pools may have varied. Our impression is that while selection of river sections and isolated pools may have caused some of the variation observed among months, we do not believe this to be a major factor in explaining monthly varia- tion since there appeared to be much more homogeneity among sites within monthly samples than among monthly samples. Fig. 1. Photographs of the study site along the Río Salado, Puebla, Mexico. Note that the close-up photograph was taken during one of the wetter months of the year. 154 G.A. Woolrich-Piña et alii Surveys took place in the morning hours (e.g., 0800 to 1200 h). Each river section (identified as a single continuous length of river of similar water flow, either a pool with still water or a section of river with greater flow between two pools) or isolated pool (a pool without direct connection to the main channel of the Río Salado) was visually searched for the presence of larval amphibians (the river sections were small enough, and the water clear enough and shallow enough to allow for the easy detection of any tadpoles). All river sections were contiguous with, or in the case of isolated pools, parallel to, the main river channel of the river (i.e., we did not skip sections of the river dur- ing our surveys). We measured the distance between each isolated pool and the main channel of the Río Salado. The main channel consisted of the primary path of the Río Salado and which contained water throughout its length and throughout the year. For surveyed section in the main channel, this distance was recorded as 0 m. We also measured the physical dimensions (length and width) of each river section. For each sampled river section or isolated pool we measured maximum depth, salin- ity, temperature, dissolved oxygen, and pH. We measured water chemistry and temperature approxi- mately 20 – 30 cm from the river’s edge and at a depth of 10 – 20 cm. Salinity, dissolved oxygen, and temperature were measured using a YSI Model 85 Handheld DO/conductivity meter. We used DF Universal Indicator Paper pH 1-14 to measure pH. In addition, we visually estimated the percent cover of aquatic vegetation in each pool. To examine the differences in the characteristics of river sections where we found tadpoles and sections where we did not find tadpoles, we used a nested MANOVA with month and tadpole presence (nested within month) as the independent variables and the physical and chemical param- eters as the dependent variables. We limited this analysis to the November 2007 to February 2008 surveys (the months when tadpoles were present; see below). A significant MANOVA was followed by univariate nested ANOVAs to examine each dependent variable in greater detail and to deter- mine which of the measured variables differed over time and between sections with and without tadpoles. Due to the dynamic nature of the Río Salado (drying of pools and changes in water flow), we were not able to follow specific river sections from month to month, especially pools isolated from the main river channel (see above). We, therefore, could not use repeated measures ANOVAs. RESULTS Monthly variation Depth of the river sections varied among months, with the lowest values found in August 2007 and an increase from this low through June 2008, with a temporary decrease in March 2008 (Fig. 2A). Width of the river sections also differed among months (Fig. 2B), as did the length of river sections (Fig. 2C). Width and length were both higher early in the study period (e.g., March to August 2007) and lower from September 2007 through June 2008. The distance of a river section or isolated pool to the main river channel varied among months, with peaks in July and August 2007 and December 2007 (Fig. 2D). Dis- solved oxygen was fairly constant but showed distinctly low values in June 2007, and Feb- ruary to June 2008 (Fig. 2E). Salinity showed a significant trend to higher values through- out the study period (Fig. 2F). Water temperature varied significantly among months (Fig. 2G). The amount of vegetative cover in sections was low for most of the study period with distinct periods of high vegetative cover in March 2007 and from November 2007 to Feb- ruary 2008 (Fig. 2H). 155Distribution of Ollotis occidentalis tadpoles Fig. 2. Variation in characteristics of surveyed river sections throughout the study period (February 2007 to June 2008) along the Río Salado. Means are given ± 1 SE. 156 G.A. Woolrich-Piña et alii Comparison of sections with and without tadpoles Tadpoles were observed in the river in March 2007 and from November 2007 through February 2008 (Fig. 3). To further investigate the factors that might influence the distribu- tion of tadpoles in the Río Salado, we limited our analyses to the November 2007 to Feb- ruary 2008 surveys (i.e., a single contiguous breeding season). The nested MANOVA found significant differences in the river sections where we found tadpoles and sections where we did not find tadpoles (Wilks’ l = 0.005, F32,186 = 18.7, P < 0.0001). There was also a significant month effect (Wilks’ l = 0.009, F24, 146 = 24.84, P < 0.0001). Water depth varied among months (Fig. 4A; F3,57 = 3.96, P = 0.012), and tadpoles were found in deeper river sections (Fig. 4A; F4,57 = 24.3, P < 0.0001). Width of the river sections did not vary significantly from month to month (Fig. 4B; F3,57 = 1.41, P = 0.25). Sections of the river where tadpoles were found were wider than those lacking tadpoles (Fig. 4B; F4,57 = 5.98, P = 0.0004). River section length did not change from month to month (Fig. 4C; F3,57 = 1.17, P = 0.33). Sections with tadpoles were longer than sections without tadpoles (Fig. 4C; F4,57 = 5.88, P = 0.0005). Distance to the main channel varied from month to month (Fig. 4D; F3,57 = 6.39, P = 0.0008). Sections of the river that contained tadpoles were closer to the main river chan- nel than sections that did not contain tadpoles (Fig. 4D; F4,57 = 17.0, P < 0.0001). In fact, no tadpoles were observed in sections that were not along the main river channel. Dissolved oxygen levels remained relatively constant among months (Fig. 4E; F3,57 = 0.4, P = 0.75). Dissolved oxygen levels were higher in river sections where we found tad- poles compared to the sections without tadpoles (Fig. 4E; F4,57 = 42.8, P < 0.0001). Salinity varied among months (Fig. 4F; F3,57 = 7.85, P = 0.0002). Tadpoles were found in river sections that had significantly lower salinity than river sections where no tadpoles were found (Fig. 4F; F4,57 = 360.2, P < 0.0001). The difference in salinity between sec- tions with and without tadpoles was highest in November 2007 and then stabilized from December 2007 to February 2008 (Fig. 4F). Temperature showed a clear variation among months with a low in January 2008 (Fig. 4G; F3,57 = 942.3, P < 0.0001). The temperature of sections in which we found tadpoles Fig. 3. Proportion of surveyed river sections with tadpoles during monthly surveys of the Río Salado from February 2007 to June 2008. Number of surveyed river sections is provided for each survey. 157Distribution of Ollotis occidentalis tadpoles differed from the temperature of sections in which we found no tadpoles, but the differ- ence was quite small (Fig. 4G; F4,57 = 3.52, P = 0.01). The amount of vegetative cover did not vary from month to month (Fig. 4H; F3,57 = 1.56, P = 0.21). Tadpoles were found in river sections that contained more vegetation than river section without tadpoles (Fig. 4H; F4,57 = 178.0, P < 0.0001). DISCUSSION The Río Salado and the pools along it are clearly a dynamic system, both in terms of the size and depth of the river sections and the physical and chemical aspects of the water. Such a dynamic aquatic system is likely to influence the amphibians and other organisms that live in the Río Salado. Indeed, there are clear changes in the presence of tadpoles of O. occidentalis throughout the year (see Fig. 3). The dynamics of the appearance of O. occidentalis tadpoles in the Río Salado is similar to that described for a population in the nearby Río Zapotitlán (Canseco-Márquez et al., 2003). Our results also make it clear that the distribution of tadpoles of O. occidentalis is influenced by the characteristics of the river. In particular it appears that factors related to the permanence of water, the chemical nature of the water, and the presence of vegetation, are important. The importance of water-permanence is suggested by the significant dif- ference in river section length, width, and depth, with river sections with tadpoles having more water (i.e., wider, longer, deeper) than sections without tadpoles. Thus, tadpoles are located in sections of the river where water will likely persist for longer periods. In addi- tion, the fact that tadpoles were never found in pools located outside the main river chan- nel suggests the importance of water permanence. In a community of stream-breeding frogs in Sulawesi, Indonesia, Gillespie et al. (2004) found that the tadpoles of most species were found only in pools or still water along the main channel of the stream, and only one species was found to use pools not directly connected to the main channel. However, the preference for deeper sections of streams can vary depending on the species of tadpole, the stream surveyed, and on other characteristics of the stream (e.g., water flow, amount of vegetative cover) (Eterovick and Barata, 2006). In part this variation may result from variation in the overall permanence of the stream. In the case of the Río Salado, the per- manence of any section of the river is variable, and thus tadpoles may prefer to be in sec- tions of the river that might be more predictable in their duration. We also found that tadpoles of O. occidentalis were located in sections of the river with lower salinity. Indeed, sections without tadpoles were nearly twice as saline as sections with tadpoles. Thus, tadpoles use areas based in part on their salinity. Other species of anurans are also known to be less prevalent in areas with higher salinity (e.g., Davenport and Huat, 1997; Smith et al., 2007) or to avoid oviposition in water with higher salinity (Haramu- ra, 2008). However, salinity did not affect the distribution of Buergeria japonica tadpoles Fig. 4. Mean characteristics of river sections of the Río Salado with (closed circles) and without (open cir- cles) tadpoles. Means are given ± 1 SE. 158 G.A. Woolrich-Piña et alii (Haramura, 2007). In contrast, the abundance of Rhinella marina increased with salinity in introduced populations on Puerto Rico (Ríos-López, 2008). It appears that at least some species of bufonids show local adaptation to salinity, and are able to tolerate higher levels of salinity as tadpoles (e.g., Gómez-Mestre and Tejedo, 2003, but see Beebee, 1985). We also found dissolved oxygen levels were higher in areas of the river with tadpoles than in areas of the river without tadpoles. This is consistent with previous results suggest- ing that Anaxyrus terrestris tadpoles use microhabitats with higher levels of dissolved oxy- gen (Noland and Ultsch, 1981). Dissolved oxygen also appears to be important in deter- mining species richness of tadpoles in ponds along the Middle Paraná River in Argentina (Peltzer and Lajmanovich, 2004). Tadpoles also used sections of the river with higher levels of vegetative cover. For the most part, the vegetation consisted of algae, and thus this relationship of tadpole presence and vegetative cover could reflect greater food availability for the tadpoles. Alternatively, the presence of more vegetative cover could simply reflect better quality water (i.e., lower salinity, higher dissolved oxygen content, greater permanence). Given the apparent importance of water permanence and water chemistry in deter- mining the distribution of tadpoles of O. occidentalis in the Río Salado, it is worth examin- ing the potential conservation implications of the salt factories (“salineras”) that are locat- ed along the Río Salado, including just upstream of our study area. These salineras divert water from the Río Salado to harvest the salt by evaporation. Thus these salineras have the potential to greatly alter both the amount of water and the chemistry of the water of the Río Salado. Such alterations may have impacts on the amphibians of the Río Salado, either directly through changes in water levels or water chemistry, or indirectly by suble- thal effects of changes in water chemistry (e.g., Squires et al., 2008). It is not clear whether or not there is any current impact of the salineras on the Río Salado; however, any future increases in the number of extent of salineras should be carefully considered in light of the potential impacts on the native fauna, including the amphibians, living in the Río Salado. ACKNOWLEDGMENTS This research was supported by funds from Dirección General de Asuntos de Personal Académico through the project PAPIIT-IN221707 “Factores que determinan la distribución de los anfibios en las pozas asociadas al Río Salado, Puebla, México”; and for Facultad de Estudios Superi- ores Iztacala through the Programa de Apoyo a los Profesores de Carrera (PAPCA) 2007-2008 with the project “Caracterización de las pozas asociadas al Río Salado (Puebla) y su influencia en la dis- tribución de los anfibios: aspectos ecológicos y geográficos” and 2009-2010 “Efecto de la salinidad sobre las larvas de anuros que habitan el Río Salado (Puebla)”. The research was approved by the Denison University Institutional Animal Care and Use Committee (Protocol 07-004). We thank two anonymous reviewers for comments that improved the manuscript. REFERENCES Beebee, T.J.C. (1985): Salt tolerances of natterjack toad (Bufo calamita) eggs and larvae from coastal and inland populations in Britain. Herpetol. J. 1: 14-16. 159Distribution of Ollotis occidentalis tadpoles Blaustein, L., Margalit, J. (1995): Spatial distribution of Culiseta longiareolata (Culicidae: Diptera) and Bufo viridis (Amphibia: Bufonidae) among and within desert pools. J. Arid Environ. 29: 199-211. Canseco-Márquez, L., Gutiérrez-Mayén, G. (1996): Anfibios y Reptiles del Valle de Zapotitlán Salinas, Puebla. IV Reunión Nacional de Herpetología. Cuernavaca, Morelos. Canseco-Márquez, L., Gutiérrez-Mayén, G., Mendelson, J.R. III. (2003): Distribution and natural history of the hylid frog Hyla xera in the Tehuacan-Cuicatlan Valley, Mexico, with a description of the tadpole. Southwest. Nat. 48: 670-675. Davenport, J., Huat, K.K. (1997): Salinity tolerance and preference in the frog Rana rugu- losa Wiegmann. Herpetol. J. 7: 114-115. Dávila-Aranda, P., Villaseñor-Rios, J.L., Medina-Lemos, R., Ramirez-Roa, A., Salinas- Tovar, A., Sanchez-Ken, J. Tenorio-Lezama, P. (1993): Flora del Valle de Tehuacan- Cuicatlán, México. Serie: Listado florístico de México. Instituto de Biología; Univer- sidad Nacional Autónoma de México. Duellman, W.E. (1961): The amphibians and reptiles of Michoacán, México. Univ. Kansas Publ., Mus. Nat. Hist. 15: 1-148. Eason, G.W., Fauth, J.E. (2001): Ecological correlations of anuran species richness in tem- porary pools: A field study in South Carolina, USA. Israel J. Zool. 47: 347-365. Eterovick, P.C., Barata, I.M. (2006): Distribution of tadpoles within and among Brazilian streams: The influence of predators, habitat size and heterogeneity. Herpetologica 62: 365-377. Gillespie, G.R., Lockie, D., Scroggie, M.P., Iskandar, D.T. (2004): Habitat use by stream- breeding frogs in south-east Sulawesi, with some preliminary observations on com- munity organization. J. Trop. Ecol. 20: 439-448. Girish, K.G., Krishnamurthy, S.V.B. (2009): Distribution of tadpoles of large wrinkled frog Nyctibatrachus major in central Western Ghats: influence of habitat variables. Acta Herpetol. 4: 153-160. Gómez-Mestre, I., Tejedo, M. (2003): Local adaptation of an anuran amphibian to osmoti- cally stressful environments. Evolution 57: 1889-1899. Halverson, M.A., Skelly, D.K., Kiesecker, J.M., Freidenburg, L.K. (2003): Forest mediated light regime linked to amphibian distribution and performance. Oecologia 134: 360-364. Haramura, T. (2007): Microhabitat selection by tadpoles of Buergeria japonica inhabiting the coastal area. J. Ethol. 25: 3-7. Haramura, T. (2008): Experimental test of spawning site selection by Buergeria japonica (Anura: Rhacophoridae) in response to salinity level. Copeia 2008: 64-67. Martín del Campo, R., Sánchez-Herrera, O. (1979): Estudio herpetofaunístico de Zapoti- tlán de las Salinas, Puebla. Biología de Campo. Facultad de Ciencias, Universidad Nacional Autónoma de México. Mata-Silva, V. (2000): Estudio comparativo del ensamble de anfibios y reptiles en tres localidades de Zapotitlán de las Salinas, Puebla. Tesis de Licenciatura en Biología. FES-Iztacala. Universidad Nacional Autónoma de México. Noland, R, Ultsch, G.R. (1981): The roles of temperature and dissolved oxygen in micro- habitat selection by the tadpoles of a frog (Rana pipiens) and a toad (Bufo terrestris). Copeia 1981: 645-652. 160 G.A. Woolrich-Piña et alii Oliver-López, L., Ramírez-Bautista, A. (2002): Algunos aspectos de la Ecología reproduc- tiva y desarrollo larvario en un grupo de anuros del municipio de Zapotitlán de las Salinas, Puebla. Memorias de la VII Reunión Nacional de Herpetología, realizada del 25 al 28 de Noviembre en Guanajuato, Guanajuato. Oliver-López, L., Ramírez-Bautista, A., Lemos Espinal, J.A. (2000): Bufo occidentalis. Fecundity. Herpetol. Rev. 31: 39-40. Peltzer, P.M., Lajmanovich, R.C. (2004): Anuran tadpole assemblages in riparian areas of the Middle Paraná River, Argentina. Biodiv. Conserv. 13: 1833-1842. Richards, S.J. (2002): Influence of flow regime on habitat selection by tadpoles in an Aus- tralian rainforest stream. J. Zool. 257: 273-279. Ríos-López, N. (2008): Effects of increased salinity on tadpoles of two anurans from a Caribbean coastal wetland in relation to their natural abundance. Amphibia-Reptilia 29: 7-18. Rocha, C.F.D., Van Sluys, M., Begallo, H.G., Alves, M.A.S. (2002): Microhabitat use and orientation to water flow direction by tadpoles of the Leptodactylid frog, Thoropa miliaris in southeastern Brazil. J. Herpetol. 36: 98-100. Sadinski, W.J., Dunson, W.A. (1992): A multilevel study of effects of low pH on amphib- ians on temporary ponds. J. Herpetol. 26: 413-422. Skelly, D.K. (1996): Pond drying, predators, and the distribution of Pseudacris tadpoles. Copeia 1996: 599-605. Smith, G.R., Lemos-Espinal, J.A. (2010): Observations of amplexus and oviposition in Ollotis [Bufo] occidentalis in the Río Salado, Puebla, Mexico. IRCF Rept. Amphib. 17: 46-47. Smith, G.R., Temple, K.G., Dingfelder, H.A., Vaala, D.A. (2006): Effects of nitrate on the interactions of the tadpoles of two ranids (Rana clamitans and R. catesbeiana). Aquat. Ecol. 40: 125-130. Smith, M.J., Schreiber, E.S.G., Scroggie, M.P., Kohout, M., Ough, K., Potts, J., Lennie, R., Turnbull, D., Jin, C., Clancy, T. (2007): Associations between anuran tadpoles and salinity in a landscape mosaic of wetlands impacted by secondary salinisation. Freshw. Biol. 52: 75-84. Squires, Z.E., Bailey, P.C.E., Reina, R.D., Wong, B.B.M. (2008): Environmental deteriora- tion increases tadpole vulnerability to predation. Biol. Lett. 4: 392-394. Ultsch, G.R., Bradford, D.F., Freda, J. (1999): Physiology: Coping with the Environment. In: Tadpoles: The Biology of Anuran Larvae, p. 189-214. McDiarmid, R.W., Altig, R., Eds, Univ. Chicago Press, Chicago. Warner, S.C., Travis, J., Dunson, W.A. (1993): Effect of pH variation on intraspecific com- petition between two species of Hylid tadpoles. Ecology 74: 183-194. Welch, N. E., Macmahon, J.A. (2005): Identifying habitat variables important to the rare Columbia spotted frog in Utah (USA): an information-theoretic approach. Conserv. Biol. 19: 473-481. Woolrich-Piña, G.A., Oliver-López, L., Lemos-Espinal, J.A. (2005): Anfibios y reptiles del Valle de Zapotitlán Salinas, Puebla. Universidad Nacional Autónoma de México - CONABIO. México.