Living in a Japanese onsen: Field observations and physiological measurements of hot spring amphibian tadpoles, Buergeria japonica

Abstract

The Japanese stream tree frog, Buergeria japonica, is widely distributed across the southern islands of Japan and Taiwan. While the species is known to inhabit hot springs, this has only been reported in Taiwan. To further understand the utilization of hot springs by B. japonica, we conducted field observations of tadpoles from a hot spring on Kuchinoshima Island, a tiny volcanic island of southwestern Japan. We found that tadpoles on Kuchinoshima Island inhabited hot spring pools with extremely high temperatures that exceeded temperatures in which any other amphibians have been found. In addition, we conducted thermal tolerance measurements and found that the thermal tolerance of B. japonica tadpoles was high. These findings suggest that high thermal tolerance of B. japonica is maintained even at the northern tip of its distribution, and this has allowed them to widen their available niche and inhabit a hot spring on the tiny island of Kuchinoshima.

Full text

1
Living in a Japanese onsen: field observations and physiological measurements of hot spring
amphibian tadpoles, Buergeria japonica
Shohei Komaki1, Quintin Lau2, Takeshi Igawa3
1Global Career Design Center, Hiroshima University, 1-7-1, Higashi-Hiroshima, Hiroshima 739-
8514, Japan
2Department of Evolutionary Studies of Biosystems, Sokendai (The Graduate University for
Advanced Studies), 1560-35, Hayama, Kanagawa 240-0193, Japan
3Division of Developmental Science, Graduate School of International Development and
Cooperation, Hiroshima University, 1-5-1, Higashi-Hiroshima, Hiroshima 739-8529, Japan
Manuscript type: short note
Word counts: 2224 (whole manuscript), 151 (abstract)
Corresponding author:
Shohei Komaki
Global Career Design Center, Hiroshima University, 1-7-1, Higashi-Hiroshima, Hiroshima 739-
8514, Japan
komaki@hiroshima-u.ac.jp
[2018/01/26]
Iwate Medical University
komaki@medicalgenome.info

2
Abstract
The Japanese stream tree frog, Buergeria japonica, is widely distributed across the southern
islands of Japan and Taiwan. While the species is known to inhabit hot springs, this has only
been reported in Taiwan. To further understand the utilization of hot springs by B. japonica, we
conducted field observations of tadpoles from a hot spring on Kuchinoshima Island, a tiny
volcanic island of southwestern Japan. We found that tadpoles on Kuchinoshima Island inhabited
hot spring pools with extremely high temperatures that exceeded temperatures in which any
other amphibians have been found. In addition, we conducted thermal tolerance measurements
and found that the thermal tolerance of B. japonica tadpoles was high. These findings suggest
that high thermal tolerance of B. japonica is maintained even at the northern tip of its distribution,
and this has allowed them to widen their available niche and inhabit a hot spring on the tiny
island of Kuchinoshima.
Keywords
Geothermal hot spring; Rhacophorus; thermal tolerance; volcanic island

3
Main text
The distribution of ectotherms is critically affected by their surrounding thermal environment.
This is important for amphibians that occupy aqueous and terrestrial environments throughout
their lifecycles (Wells, 2007). Specifically, water temperature strongly affects amphibian
survival rate during early developmental stages, whereby most tadpoles generally avoid very
cold or hot waters (Wells, 2007). However, a few studies have reported the occurrences of
amphibian species in hot springs (Brues, 1932; Mason, 1939; Chen et al., 2001; Wu and Kam,
2005).
In central and northern Taiwan, tadpoles of the Japanese stream tree frog, Buergeria japonica,
reportedly inhabit hot springs (called “onsen” in Japanese) [Jentse, Rushan, and Wulai hot
springs (Chen et al., 2001; Wu and Kam, 2005)]. These populations showed extremely high
thermal tolerance (> 40 °C) and preferred water temperatures around 37 °C (Chen et al., 2001;
Wu and Kam, 2005), which was assumed to enable them to inhabit the hot springs (Chen et al.,
2001). This species is widely distributed almost throughout the Ryukyu Archipelago of Japan
and Taiwan, and is characterized as the only naturally occurring amphibian species in the islands
of the Tokara Archipelago (Maenosono and Toda, 2007). Recently, Komaki et al. (2016)
investigated thermal and salinity tolerances of multiple populations of this species
(Amamioshima and Tokashikijima Islands of Japan and Yilan and Hualien in Taiwan) and
confirmed that the thermal tolerances have been maintained across island populations. Due to
small island sizes and volcanic origin, naturally forming freshwater resources in Tokara Islands
may be limited predominantly to hot springs. Therefore, the high thermal tolerance of B.
japonica may have allowed the species to utilize the hot springs as the site for reproduction and
survive on these islands prior to modern-human colonization which brought upon freshwater
sources with lower temperatures (Komaki et al., 2016). However, the thermal tolerance and
inhabitance in hot springs of Tokara populations were not confirmed. To better understand the
utilization of hot springs by B. japonica, we performed field observations and measurements of
thermal tolerance of B. japonica tadpoles from a hot spring located on an island of the Tokara
Archipelago.
On 9th September 2015, we performed field observations in Seramma Onsen (Seramma Hot
Spring), a geothermal hot spring on Kuchinoshima Island (29° 57.32′N 129° 55.63′E, Fig. 1),
and found Buergeria japonica tadpoles inhabiting the hot spring. The study area included
shallow and narrow streams and small pools of hot spring water (Fig. 1). The chains of small
shallow pools (< 1.5 m in diameter and < 10 cm in depth) within the streams were formed by
fallen leaves, branches, or stones that dammed up the water. In the water pools, few tadpoles
were readily visible, thus, we searched for individuals hidden under fallen leaves or stones in the

4
pools using landing nets, and inferred the existence and abundance of tadpoles. We measured
water temperatures of each pool and site tadpoles were found, using two electrical thermometers
(Japan Pet Design Co., Ltd. Tokyo, Japan) for accurate measurements. Water temperatures were
remarkably variable between pools, ranging from 34.6 to 51.8 °C. Water temperature was higher
in pools closer to the hot spring sources. The highest temperature of a pool inhabited by B.
japonica was 46.1 °C, but only a single tadpole was found. The maximum water temperature of
pools in which multiple (at least four) tadpoles were found was 41.5 °C. There were also cooler
pools that ranged from 34.6 to 37.2 °C, each inhabited by more than 20 tadpoles. No dead
tadpoles, juveniles or adults were found.
The occurrence of B. japonica tadpoles in the geothermal hot spring on Kuchinoshima Island
indicates that habitation in hot springs is not confined to the populations in Taiwan. Since
sustained high thermal tolerance of B. japonica tadpoles among northern and central Ryukyu
Archipelago and Taiwan were revealed by other studies (Chen et al., 2001; Wu and Kam, 2005;
Komaki et al., 2016), the potential use of hot springs may be maintained across their distribution,
specifically to the northern tip of its distribution, Kuchinoshima Island. As tadpoles in Seramma
tended to occur in cooler pools at temperatures below 37 °C, their thermal preference is similar
to those of the Taiwanese populations (Wu and Kam, 2005).
We observed no tadpoles swimming across to neighbouring pools, but most tadpoles were
under fallen leaves in each pool. Therefore, it is possible that tadpoles had stayed for extensive
periods at the location observed, whereby individuals did not migrate frequently between pools.
However, a single individual that was found in 46.1 °C may have accidentally or temporarily
drifted from neighbouring cooler pools. Nevertheless, the identification of multiple tadpoles at
41.5 °C indicates they can regularly inhabit such high temperatures. While more detailed surveys
are needed to track individual movement and understand the temperature range and duration of
inhabitation in hot spring pools, these are the highest recorded water temperatures in which
tadpoles were found compared with previous studies [Taiwanese populations of Buergeria
japonica at 39.7 °C (Wu and Kam, 2005); Rana sp. (likely R. pretiosa) at 106 °C (41.1 °C)
(Brues, 1927); Spea bombifrons at 38.4 °C (observed temporarily at 45 °C but soon returned to
cooler water) (Brues, 1932)].
The constant water temperature year around is advantageous to amphibians (Scott and
Jennings, 1985). In addition, high water temperature is expected to increase the growth rate of
tadpoles and decrease intra- and interspecific competition (Licht, 1971; Chen et al., 2001). It is
also suggested that temperature is linked to immune responses of amphibians, whereby warmer
water temperatures can enhance the effectiveness of immune defences (Forrest and Schlaepfer,
2011). Therefore, tadpoles may actively select hot water to maximize these benefits.
Alternatively, some individuals may simply be forced to inhabit unfavourable conditions due to

5
the limitation of water resources and temporary barriers for movement. These uncertainties
should be addressed by comparative studies of B. japonica tadpoles that focus on their
behavioural patterns, biotic factors such as food availability and predation pressures among
habitats with different thermal conditions, and experiments on survival and growth rates at
variable temperatures.
To measure the thermal tolerance of B. japonica tadpoles, we collected 40 tadpoles at Gosner
stages 26–28 (Gosner, 1960) and water from pools around 37 °C in the Seramma Hot Spring.
The hot spring water that contained tadpoles was kept inside a hotel room near the hot spring for
3 h to reduce the water temperature to that of artificial freshwater pools formed on the island
(approximately 26 °C). For the control, we prepared a main tank that contained 4000 ml of water
around 26 °C, and the water temperature was stably maintained by air conditioning so that we
could measure the survival rate of tadpoles without the thermal effect of hot spring. We then
placed a sub-tank filled with 400 ml of the hot spring water into the main tank and kept 10
tadpoles in the sub-tank. For the heat treatment, we prepared another main tank similar to the
control and placed three sub-tanks with 400 ml of hot spring water, each containing 10 tadpoles
(Table 1: experimental groups 1-3). The temperature of the main tank was then increased about
0.16 °C per min using a heater (EVERES Co., Ltd. Tokyo, Japan) with a thermo-controller
(IWAKI Co., Ltd. Tokyo, Japan). Water in the main tanks were automatically circulated, and
sub-tanks were rearranged every 10 min to eliminate the bias of water temperature resulting from
position within the main tank. We also monitored water temperature at the surface and bottom of
the main tanks to confirm that there was no temperature gradient. We stopped measuring thermal
tolerance of each individual immediately when it lost balance, ability to swim, or died, and
recorded the temperature at that time as the thermal limit. Individuals were pithed immediately
after losing such righting behaviours. Tadpoles were not fed before or during the experiment.
For the thermal tolerance measurement, all tadpoles died or lost the righting behavior around
46 °C (Table 1). The lowest and highest thermal limits were 46.0 and 46.2 °C, respectively.
Meanwhile, no mortality was observed in the control group. While Chen et al. (2001) and Wu
and Kam (2004) also measured thermal tolerance of B. japonica tadpoles, direct comparisons to
our study cannot be made due to different criteria and different heating rates applied which can
affect the thermal tolerance (Rezende et al., 2011). Nevertheless, it seems likely that B. japonica
tadpoles of Seramma population have substantially high thermal tolerance similar to that found
in Taiwanese populations (43–44 °C) (Chen et al., 2001; Wu and Kam, 2005).
In conclusion, we found that B. japonica tadpoles on Kuchinoshima Island inhabit the highest
water temperatures ever recorded for any amphibian tadpole and that the B. japonica thermal
tolerance is retained across their entire distribution (see also Komaki et al., 2016). The high
thermal tolerance is essential for tadpoles to inhabit warmer hot spring pools, contributing to the

6
widening of the species’ ecological niche on the tiny volcanic island. On the other hand, tadpoles
were distributed in hot spring pools close to the fatal temperature that were adjacent to hotter
pools exceeding the fatal temperature. Therefore, to understand the mechanisms of how B.
japonica tadpoles inhabit the geothermal hot spring and avoid fatal temperatures, further studies
on ecological and behavioural features are important.

7
Acknowledgements
Permission to perform field sampling and experiment on Kuchinoshima Island was granted by
the mayor of Toshima village. This study was supported by a Grant-in-Aid for JSPS Fellows
(number 25-5065) from Japan Society for the Promotion of Science and a grant from Hiroshima
University Education and Research Support Foundation to Komaki S.

8
References
Brues, C.T. (1927): Animal life in hot springs. Q. Rev. Biol. 2: 181–203.
Brues, C.T. (1932): Further studies on the fauna of North American hot springs. Proc. Am. Acad.
Arts Sci. 67: 185–303.
Chen, T.C., Kam, Y.C., Lin, Y.S. (2001): Thermal physiology and reproductive phenology of
Buergeria japonica (Rhacophoridae) breeding in a stream and a geothermal hotspring in
Taiwan. Zoolog. Sci. 18: 591–596.
Forrest, M.J., Schlaepfer, M. A. (2011): Nothing a hot bath won’t cure: infection rates of
amphibian chytrid fungus correlate negatively with water temperature under natural field
settings. PLos ONE. 6: e28444.
Gosner, K. L. (1960): A simplified table for staging anuran embryos and larvae with notes on
identification. Herpetologica. 16: 183–190.
Komaki, S., Igawa, T., Lin, S.-M., Sumida, M. (2016): Salinity and thermal tolerance of
Japanese stream tree frog (Buergeria japonica) tadpoles from island populations. Herpetol.
J. 26: 207–211.
Licht, L.E. (1971): Breeding habits and embryonic thermal requirements of the frogs, Rana
aurora aurora and Rana pretiosa pretiosa, in the Pacific Northwest. Ecology 52: 116–124.
Maenosono, T., Toda, M. (2007) Distributions of amphibians and terrestrial reptiles in the
Ryukyu Archipelago: A review of published records. Akamata 18: 28–46.
Mason, I.L. (1939): Studies on the fauna of an Algerian hot spring. J. Exp. Biol. 16: 487–498.
Rezende, E.L., Tejedo, M., Santos, M. (2011): Estimating the adaptive potential of critical
thermal limits: Methodological problems and evolutionary implications. Funct. Ecol. 25:
111–121.
Scott, N. J., Jennings, R. D. (1985): The tadpoles of five species of New Mexican leopard frogs.
Occas. Pap. Mus. Southw. Biol. 3: 1–21.
Wells, K.D. (2007): The Ecology and Behavior of Amphibians. The University of Chicago Press,
Chicago and London.
Wu, C.S., Kam, Y.C. (2005): Thermal tolerance and thermoregulation by Taiwanese
Rhacophorid tadpoles (Buergeria Japonica) living in geothermal hot springs and streams.
Herpetologica 61: 35–46.

9
Table
Table 1. Sample sizes (n) and the thermal limits (mean, range and SD) of five experimental
groups.
Experimental
group
n
Mean (range) (°C)
SD
1
10
46.1 (46–46.2)
0.10
2
10
46.06 (46–46.2)
0.10
3
10
46.04 (46–46.2)
0.08

10
Figure
Fig. 1. Stream of hot spring in Seramma Onsen. Water pools were continuously formed.