PropertyValue
is nif:broaderContext of
nif:broaderContext
is schema:hasPart of
schema:isPartOf
nif:isString
  • We conducted experiments in the bamboo forests at Chitou in Nantou County, Taiwan. The average monthly temperature is 18 °C, and the total annual rainfall is about 3000 mm [46]. The rainy season is from February to September. Local farmers cut bamboo (Phyllostachys edulis) for commercial purposes, and the stumps collect rain water and become suitable breeding sites for K. eiffingeri [46]. Survey and sample collection of overlapping clutches: From March—August 2007–2009, we conducted field surveys to collect adults (males and/or females) and overlapping offspring (eggs and tadpoles) for parentage analyses. We first surveyed bamboo forests and located bamboo stumps containing tadpoles. We marked the stumps and monitored them if an egg clutch was found in the inner wall of stumps above the water line. We surveyed stumps every 3–7 d., dependent upon the breeding activity of frogs. Once an egg clutch was found in a tadpole-occupied stump, we collected all eggs and tadpoles about five days later and reared them in the laboratory in order to sample tissues for parentage analyses. We captured and toe-clipped attending male frogs and preserved the tissue individually in 95% ethanol. Feeding females were more difficult to capture because they return to the nest during the night at an interval of about eight days [37]. Nevertheless, from June 27 to August 31, 2009, we set up funnel traps on the opening of stumps to trap feeding females [37]. If an egg clutch was deposited in a tadpole-occupied stump, we would remove eggs and tadpoles and transport them back to the laboratory about five days later. To continue the trapping procedure, we put ten tadpoles that had been collected from bamboo stumps at least 1–2 km away into the emptied stump. Female K. eiffingeri do not discriminate kin from non-kin[37], and we could capture them while they were feeding the substitute tadpoles. Upon capture, we toe-clipped female frogs and preserved the tissue individually in 95% ethanol. In the laboratory, we incubated egg clutches separately on moist substrates until hatching. We reared tadpoles in beakers (ca. 1 L water) and fed them with chicken egg yolk once every 4 days until they reached metamorphosis. Chicken egg yolk is a good substitute for K. eiffingeri eggs [47]. When tadpoles reached Gosner stage 40, we clipped a distal portion of tadpoles’ tail (i.e., 10% or less of total tail length) and preserved tissues in 95% ethanol for parentage analyses. From April 17 to August 15, 2013, we conducted a manipulated experiment where paired bamboo cups, with and without tadpoles, were set up to investigate nest choice of frogs. First, we collected bamboo trunks which were cut by local farmers to thin the forest and sawed a section of trunk (i.e., an internode) which contains a septum at each end. We then sawed the middle of the internode which resulted in two identical bamboo cups and tied them to a bamboo stump (Fig 1). Bamboo stumps were covered by plastic sheet to prevent frogs from depositing eggs. We randomly designated one cup as the control and the other as the experimental group. A control group contained water only but the experimental group contained water and 5 tadpoles (Gosner stage 28–35). We filled the paired-bamboo cups with 7 cm of water [46]. We drilled small holes in the bamboo wall to prevent a rise of the water level from rainfall and, where necessary, added water to prevent the drop of water level by evaporation [25]. Tadpoles were collected from bamboo stumps at least 1–2 km away. We set up 50–80 paired-bamboo cups each week during the study period, dependent upon the breeding activity of frogs and availability of tadpoles. Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0123221.g001 A set-up for a manipulated experiment on nest choice.Paired bamboo cups were attached to a stump which was covered by a plastic sheet. We surveyed the paired-bamboo cups every 4 days. Once fertilized eggs were found on the inner wall of a bamboo cup above the water line, we recorded which cup was used and the clutch size. Subsequently, these paired-bamboo cups were no longer used for the rest of our experiment, and we usually set up new paired-bamboo cups in nearby bamboo forests. In each survey, we removed the tadpoles from the bamboo cups, fed them with raw chicken egg yolk in a water bowl, and returned them to bamboo cups after feeding [47]. If tadpoles reached metamorphic stages, Gosner stage 40 or older, we replaced them with younger tadpoles. Genetic analyses on the parentage of tadpoles and adults: Detailed methods on parentage analyses were reported by Chen et al [48]. Briefly, we used five polymorphic microsatellite loci (CEd12365, CEd15688, CEd19063, CEd13854, and CEd19091) to conduct parentage analyses [48]. At first, DNA was extracted from tissues, amplified, and genotyped using standard methods [48]. Program Cervus 3.0 [49, 50] was used to calculate the number of alleles, allele frequencies, null allele frequencies and exclusion probabilities for each locus, and the combined exclusion probability. The combined exclusion probability across all loci was 0.987 when the genotype of a parent is known [48]. The likelihood-based, COLONY 2 program [51] was used to analyze genetic relationships between the attending males, feeding females and the offspring in the nests. Based on the results of the parentage analyses, we deduced the mating pattern of frogs (S1 Table). We defined (1) a synchronous polyandrous female as one which mates with multiple males at the same time [15]; (2) a sequentially polygynous male as one which mates with several different females at different periods of time [38]; (3) a sequential multi-mating event as a series of matings that occur within a short time, such as in a night or within a week. Each mating could be monogamous, polyandrous, or polygynous. Also, based on the results of parentage analyses, we further assessed whether individuals that breed earlier in a stump would reuse the same stump again. For example, if the early-laid clutch is a result of mating by a male (A) and female (X) and the late-laid clutch by two males (A and B) and a female (Y), then we conclude that the tadpole-occupied stump was reused by the same male (i.e., A) but different females (i.e., X and Y). In contrast, if the early-laid clutch is a result of mating by two males (C and D) and a female (Z) and the late-laid clutch by a male (E) and female (Z), then we conclude that the tadpole-occupied stumps were reused by the different males (i.e., C, D, and E) but same female (i.e., Z). All statistical analyses of the data were performed with SAS [52]. We used a G test to assess the occurrence of stump use by frogs and proportion of nest reuse among sexes [53]. We used the Wilcoxon rank sum test to compare the clutch size of eggs laid in empty and tadpole-occupied stumps. Unless stated otherwise, we report the mean ± SD of each variable. Kurixalus eiffingeri is not an endangered or protected species and thus no specific permission is needed for field studies. Some bamboo forests in Chitou are privately owned, and we had verbal permission from land owners (Mr. Pan Yeong-Sung and Ms Chen Mei-Chi) to conduct observational and field studies (23o41’05.4” N 130o47’45.4” E and 23o41’22.8” N 130o47’22.5” E, respectively). Institutional Animal Care and Use Committee (IACUC) of Tunghai University has specifically approved this study (Approval No. 100–19). We captured and toe-clipped attending males and feeding females and preserved the tissue individually in 95% ethanol for parentage analyses. Because only the most distal segment of the second toe was clipped, the toe-clipping is likely to have a minimal effect on the individual [48, 54]. When tadpoles reached Gosner stage 40, we clipped a distal portion of tadpoles’ tail (i.e., 10% or less of total tail length) [48]. Due to the fast healing of wounds (1–2 days) and sedentary nature of tadpoles, the tail-clipping is likely to have little effect on activity and feeding of tadpoles [38, 48].
rdf:type