First record of growth rings for 11 native subtropical anuran species of South America

: Skeletochronology is the most accurate method to estimate a population age structure. The methodology is based on the analysis of secondary bone tissue in order to count growth rings. This study aimed to provide initial data, allowing researchers to further work out in the age of individuals and populations, sampling evidence of the presence of growth rings in 11 native species (representing nine families) of a subtropical region of southern Brazil. Four bone samples of each specimen were used to perform the skeletochronological analysis: the penultimate phalanges of the 3rd and 4th fingers, the humerus, and the femur. The presence of growth rings was confirmed in the periosteal layer of the bones of all analyzed species. In comparison with phalanges, growth rings of humeri and femora are more irregular and less distinguishable. This is the first record of growth rings to the native species herein analised. The skeletochronology was proved to be an effective tool in determining the age of anuran amphibians from a subtropical region, since this environment presents well defined climatic seasonality.


INTRODUCTION
The study of amphibians' population age structure is crucial to understanding population dynamics and species natural history, since it provides information regarding growth rate, sexual maturity, longevity, and reproductive lifespan (li et al. 2010). The analysis of bone tissue to estimate age (known as skeletochronology) has been used in both fossil (Botha and Chinsamy 2000, Sander 2000, Tütken et al. 2004, Botha-Brink et al. 2018 and extant tetrapods, such as 'reptiles' (Avens et al. 2009(Avens et al. , goshe et al. 2010, birds (Ricqlès et al. 2003), and amphibians (Socha and Ogielska 2010, Cajade et al. 2013, Bionda et al. 2015, Sun et al. 2016, Kumbar and lad 2017, Tessa et al. 2017. Skeletochronology is based on the presence of growth rings (lines of arrested growth -lAgs) in transverse sections of the bones, in which broad lines represent the growth period and narrow lines represent a growth pause. Thus, each narrow line characterizes one growth year, allowing the estimation of the age of individuals (Castanet and Smirina 1990).
The formation of annual growth marks in anurans from temperate zones occurs due to seasonal climatic variation, since growth is delayed along the winter (Castanet et al. 2003). According to Castanet et al. (1993), annual formation of lAgs has a genetic basis and under natural conditions their formation is synchronized with the climatic seasonality. It is noteworthy that age estimation by LAGs counting is effective only if the species' growth pattern is annual (gibbons and McCarthy 1983). However, although some studies point that anurans from environments with low seasonal climatic variation usually present an irregular pattern of lAgs formation (Kusrini and Alford 2006), other studies indicate that species from tropical regions may also have evident lAgs (Rebouças et al. 2018).
Southern Brazil has a subtropical climate and resembles temperate regions concerning seasonal temperature variation (Overbeck et al. 2007). The heterogeneous environment in southern Brazil hosts an anuran fauna distributed in nine recognized families: Alsodidae, Bufonidae, Centronelidae, H y l i d a e , H y l o d i d a e , l e p t o d a c t y l i d a e , Microhylidae, Odontophrynidae, Phyllomedusidae . However, studies of subtropical herpetofauna are mainly focused on investigating the communities' structure (eterovick et al. 2005), and describing the species distribution patterns (Rosset et al. 2006). Therefore, studies involving skeletochronology of native species from South American subtropical regions are scarce (echeverria and Filipello 1990, Marangoni et al. 2009, Iturra-Cid et al. 2010, Cajade et al. 2013, Caldart et al. 2019). Thus, we investigated the presence of lAgs in anurans representing all families (11 native species) of southern Brazil.

MATERIALS AND METHODS
Skeletochronology was applied in species of wide distribution in southern Brazil (Paraná, Santa Catarina, and Rio grande do Sul states). According to Koppen's climate classification, the region's climate is considered humid subtropical (Cfa-Cfb), with well-defined temperature seasonality (Alvares et al. 2013). All sampled species have seasonal activity pattern linked to the warm season (see Sá and gerhau 1983, Kwet and Miranda 2001, Kaefer et al. 2007, Both et al. 2008, Narvaes and Rodrigues 2009, Machado et al. 2014, Caldart et al. 2016 Three specimens of each species were examined. The following measurements were made for each specimen: body mass (BM), using a balance (0.01-g precision) and snout-vent length (SVl), using a digital caliper to the nearest 0.01 mm. The specimens were dissected to extract bone samples from penultimate phalanges of the 3 rd and 4 th fingers of the foot, and long bones (humerus and femur). As a standardization procedure, all bone samples were collected from the limbs on the right side of the specimens.
After extraction, the bone samples were dehydrated on an increasingly series of alcohol (70-100%) for one hour for each step. Then, they were decalcified in 10% ethylenediaminetetraacetic acid (eDTA) for one week. Thereafter, the bones were processed in historesin, transversely sectioned in a leica RM 2245 rotary microtome, and stained using toluidine blue for ten minutes (Caputo et al. 2010). After stained, the samples were dried in kiln for one day. For each specimen, four sections from the middle region of the bone diaphysis were taken. Slices were independently analyzed using a Zeiss Axio Scope A1 microscope with an Axiocam MRc 5 digital camera by at least two different authors.
To identify the occurrence of bone resorption in long bones, the lAgs counting was comparatively analyzed considering phalanges as references. The bone resorption was also determined based on the absence of remaining cartilage tissue from the larval stage (line of metamorphosis) (Rozenblut and Ogielska 2005).

RESULTS, DISCUSSION AND CONCLUSIONS
Transverse sections of phalanges and long bones of all species exhibited lAgs. Although the rings were not uniformly clear in all bones, lAgs were observed in the periosteal layer of the bones of all analyzed species. Phalanges and long bones often had the same number of lAgs. However, the phalanges allowed a better identification of lAgs than long bones and considering long bones, identification of lAgs was better in the femur than in the humerus. likewise, regarding the phalanges, the identification of LAGs was better in the phalanges of the 4 th finger than in phalanges of the 3 rd finger.
The number of LAGs identified in the studied species was (see Table I   5; Fig. 1h), V. uranoscopa (1-5; Fig. 1d), and E. bicolor (2-7; Fig. 1j). Skeletochronology was effective for both phalanges and long bones, since all matched the corresponding number of lAgs for each specimen studied. Furthermore, skeletochronology can be a non-lethal procedure if performed using the specimens' phalanges solely (Sinsch et al. 2007, ginnan et al. 2014, Sinsch 2015, Hudson et al. 2017, since using long bones demands the euthanized of the specimens. Besides, lAgs are more evident in phalanges than in long bones, thus, phalanges are the most suitable bones for determining age in anurans (Kumbar and Pancharatna 2001). Another advantage of the method is the possibility of analyzing specimens from museum collections, without having to capture new specimens (see Sinsch 2015).
Because of the low sample of specimens per species herein analyzed, we consider this study as an exploratory work, providing evidence that this method is eff ective in estimating the age of AMANDA J.C. BRUM et al. subtropical climate anurans of southern Brazil. Additionally, this contribution presents the first record of lAgs to 11 native species of southern Brazil. Thereby, we encourage the development of new studies regarding anurans species from tropical and subtropical regions, to better understand the evolution of ecological patterns associated to this environment, and their significance in a broader view concerning anuran amphibians, one of the most ecologically endangered group of vertebrates worldwide.

ACKNOWLEDGMENTS
We thank friends of Herpetology laboratory of Federal University of Santa Maria (UFSM), Brazil, for helpful comments, suggestion and discussions, and specially thank Maurício garcia (UFSM) for comments improving the quality of this manuscript.

AUTHOR CONTRIBUTIONS
AJCB, ll, MBS, and SZC designed the study; AJCB and MBS extracted the samples; AJCB and ll performed the histological protocol; ACJB and ll performed the analyses; AJCB, ll, MBS, and SZC discussed the results; AJCB, ll, MBS, and SZC wrote the article.