Bamboo Ga-38 Koa

[Jomega Gibson]

Overlappingoffspring occurs when eggs are laid in a nest containing offspring from earlier reproduction. Earlier studies showed that the parentage is not always obvious due to difficulties in field observation and/or alternative breeding tactics. To unveil the parentage between overlapping offspring and parents is critical in understanding oviposition site selection and the reproductive strategies of parents. Amplectant pairs of an arboreal-breeding frog, Kurixalus eiffingeri, lay eggs in tadpole-occupied nests where offspring of different life stages (embryos and tadpoles) coexist. We used five microsatellite DNA markers to assess the parentage between parents and overlapping offspring. We also tested the hypothesis that the male or female frog would breed in the same breeding site because of the scarcity of nest sites. Results showed varied parentage patterns, which may differ from the phenomenon of overlapping egg clutches reported earlier. Parentage analyses showed that only 58 and 25% of the tadpole-occupied stumps were reused by the same male and female respectively, partially confirming our prediction. Re-nesting by the same individual was more common in males than females, which is most likely related to the cost of tadpole feeding and/or feeding schemes of females. On the other hand, results of parentage analyses showed that about 42 and 75% of male and female respectively bred in tadpole-occupied stumps where tadpoles were genetically unrelated. Results of a nest-choice experiment revealed that 40% of frogs chose tadpole-occupied bamboo cups when we presented identical stumps, without or with tadpoles, suggesting that the habitat saturation hypothesis does not fully explain why frogs used the tadpole-occupied stumps. Several possible benefits of overlapping offspring with different life stages were proposed. Our study highlights the importance of integrating molecular data with field observations to better understand the reproductive biology and nest site selection of anuran amphibians.

Copyright: 2015 Tung et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

In this study, we used a Taiwanese frog (Kurixalus eiffingeri (Anura: Rhacophoridae)) that breeds in water-filled bamboo stumps as a model animal to study the parentage between overlapping offspring and its ecological consequence on reproductive strategy and nest site selection. Specifically, we used (1) five microsatellite DNA markers to analyze the parentage of adults and tadpoles and (2) paired bamboo cups with and without tadpoles to study the nest choice of frogs and to reveal the possible causes of nest reuse.

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].

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.

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.

We selected 24 out of 64 stumps to conduct parentage analyses. The selected stumps consisted of at least one sampled adult (attending male and/or feeding females) and 6 tadpoles each from early- and late-laid clutches (12 tadpoles total). We analyzed a total of 301 and 451 samples from early- and late-laid clutches, respectively.

COLONY deduced eight different types of parentage among overlapping offspring (Table 1). The most common parentage among overlapping offspring was partially the same father and different mothers (N = 8), followed by different parents (N = 7). The former is a case when a male frog was involved in matings that sired offspring in the early- and late-laid clutches, and at least one of the clutches was either synchronous polyandry or sequential multi-mating events that included other male frogs, which resulted in multiple paternity. In contrast, the latter is a rather simple case where two different pairs of frogs produced the two cohorts of offspring. The remaining types of parentage among overlapping offspring occurred at much lower frequency (Table 1 and S1 Table).

In this study, K. eiffingeri oviposited in tadpole-occupied stumps, resulting in overlapping offspring with different life stages; molecular and field evidence strongly suggest that the overlapping offspring in this study is not a continuum of overlapping egg clutches reported earlier [38] and represents a different breeding phenomenon. Two lines of evidence support our contention: first, only 50% of overlapping offspring in this study were sired (completely or partially) by the same male frog compared to 93% of overlapping egg clutches in Cheng et al. [38]; second, the earlier field evidence indicates that the prevalence of overlapping offspring with different life stages was about 36% [40] compared to only 5% of overlapping egg clutches in Cheng et al. [38]. In fact, overlapping offspring of different life stages in this study is more similar to the breeding strategy employed by pigeons [9, 10, 12] and fishes [3].

On the other hand, results of parentage analyses showed that about 42 and 75% of males and females respectively bred in tadpole-occupied stumps where tadpoles were genetically unrelated, which leads to the following question: why did frogs breed in the tadpole-occupied stumps of others? Lin, Lehtinen and Kam [40] reported that reuse of nests by K. eiffingeri occurred more commonly in the second half of breeding season. One explanation is that amplectant pairs may be forced to lay eggs in tadpole-occupied nests when most suitable breeding habitats are already used later in the breeding season (i.e., the saturated habitat hypothesis) [32]. These tadpole-occupied stumps are less favorable nests because larger tadpoles from early clutches outcompete smaller tadpoles from late clutches for trophic eggs that may result in tadpole starvation, retarded growth, and mortality of the latter [39, 45]. In theory, amplectant pairs would avoid tadpole-occupied stumps to minimize the detrimental impact of residing tadpoles to their offspring. However, in the nest choice experiment, when we presented identical stumps, without or with tadpoles, 40% of frogs still chose tadpole-occupied stumps, suggesting that the habitat saturation hypothesis does not fully explain why frogs used the tadpole-occupied stumps. One explanation is that females are unable to detect the tadpoles in the stumps, and the chance of stump use is random. This possibility, however, is unlikely because adults are selective in oviposition sites [46, 85], and females purposely deposited egg clutches in stumps with fewer tadpoles [40]. The alternative explanation for the re-use of stumps is that even though the quality of tadpole-occupied stumps is discounted due to inter-clutch tadpole competition, they are still as good as, if not better than, the remaining unoccupied stumps [40].

3a8082e126