Acoustitic Characteristics of Three Species of the Genus Amolops : Amphibia, Anura, Ranidae

5 Pages·1993·1.8 MB·English

by  Masafumi Matsui

ZOOLOGICAL SCIENCE 10: 691-695 (1993) 1993Zoological SocietyofJapan
Acoustitic Characteristics of Three Species of the Genus
Amolops (Amphibia, Anura, Ranidae)
Masafumi Matsui1 Gan-fu Wu2 and Hoi-sen Yong3
,
^Graduate School ofHuman and Environmental Studies, Kyoto University,
Sakyo-ku, Kyoto 606, Japan, 'Chengdu Institute ofBiology, Academia
Sinica, Chengdu, Sichuan, People's Republic ofChina, and
^Department ofZoology, Faculty ofScience, University
ofMalaya, 59100 Kuala Lumpur, Malaysia
—
ABSTRACT Advertisement call characteristics of Amolops chunganensis from western China, A.
larutensis from Peninsular Malaysia, and A. jerboa from Borneo are described. The last species is
sometimesgroupedinto adifferentgenus, butallthethreespeciesareconsideredtoconstituteasingle
genus Amolops (sensu lato) from their unique larval morphology. Calls of the three species differ
considerably from each other both in temporal and frequency patterns, and from these acoustic
characteristics, the idea ofsubdividing this genus into discrete subgenera is supported, and the three
speciesareclassifiedasA. (Amolops)chunganensis, A. (Amo)larutensis,andA. {Meristogenys)jerboa.
Calls of the three species differ in temporal patterns, but are similar in having rather high dominant
frequencies andmore orlessclearfrequencymodulations. These commonacousticpropertiesseemto
be well adapted for the call to emerge above the heavy environmental noises.
characteristicsoffrogs ofthe genusAmolops [1,5,
INTRODUCTION
9]. On the otherhand, insome anurangenerathat
Asian ranid genus Amolops includes more than were studied acoustically as well as morphologi-
32 species [10], and is characterized by peculiar cally, call structure has been known to parallel the
larval ecology and morphology; they inhabit swift morphological groups, and, therefore, has some
torrents and bear a large abdominal sucker [4]. phylogeneticsignificance [3, 6, 8]. Inthispaperwe
Recently, Yang [10] and Dubois [2] divided this describe call characteristics of three species, i.e.,
genus into distinct genera and subgenera, respec- A. chunganensis from China, A. larutensis from
tively. However, their classifications are based Peninsular Malaysia, and A. jerboa from Borneo,
chiefly on adult and larval morphology, and other andassessvalidityofnewtaxonomictreatmentson
taxonomically important characteristics such as this genus.
acoustic and biochemical ones are not included.
Thus, it seems worth to reassess taxonomic treat- MATERIALS AND METHODS
mentsofthese authorsfrom approachesotherthan
morphology. Recordingsofcallswere made inthe field bythe
The acousticsignalplaysaveryimportantrolein senior author using a cassette tape recorder (Sony
the breeding behavior of anuran species, and is TC-D5) with an external microphone (Sony ECM-
thus regarded asone ofthe keycharacteristics that 23F). Temperature measurements were made at
are responsible for the speciation events in this the time of recording using a quick recording
animal group. Nevertheless, there are only a few thermistor thermometer (Takara A 600). The
anecdotal or incomplete reports on the acoustic recorded callswere analyzed usingcomputerprog-
rams, SoundEdit Vers. 2 or SoundEdit Pro (Mac-
Accepted June 28. 1993 roMind-Paracomp, Inc.) by a Macintosh com-
Received Mav 27. 1993 puter.
692 M. MAtsui, G. F. Wu and H. S. Yong
number of pulses included (0.133sec in four
RESULTS
pulsed note to 0.259sec in seven pulsed note).
Thelengthofinter-noteinterval, however,wasnot
Amolops chunganensis always proportional to the note length. The pulse
Calls were recorded on Mt. Omei-shan at the repetitionratewasnearlysimilaramongnoteswith
altitude of700 m, Sichuan, China, on27July 1992. different number of pulses (30.0per sec in four
Air temperature at the time of recording was pulsed note and27.0persecinsevenpulsednote).
27.4°C. Males were observed to form a loose Weak frequency modulations are present within a
aggregation and call even in the daytime (1400hr) note, and the dominant frequency was initially
either in the low bushes along streams (width<8 about 3000-3400Hz, but rose towards the end of
m) or on rocks in the streams with a moderate the pulse series to about 3500-3800Hz. Harmo-
current. nics were absent.
The call (Fig. 1A) consisted of a continuous
seriesofnoteslastingmore than30sec. Each note Amolops larutensis
consisted of four to seven clear pulses (Table 1). Calls were recorded at Frasers Hill, Larut Hill,
The note length increased with the increment of and Gombak, Peninsular Malaysia, on 10-12 De-
8
N
X
6
>-
o
z
4
LU tiH*i ,j#HH ,H«*4 tiiWi H«**< ,Hl*M
o
LU
2
A —
n , • i i.. _ j i
0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4
TIME IN SEC
11
10
N
I 8
>-
LLi
o=> *""""> *
/ «f
LU 4
LL
2
b c
;
0.3 0.6 0.9 1.2 0.3 0.05 0.1
TIME IN SEC
Fig. 1. Sonagrams of advertisement calls ofAmolops chunganensis (A), A. larutensis (B), andA. jerboa (C, D).
Acoustitic Characteristics of Amolops 693
Table 1. Call characteristics of three Amolops species (Mean+ISD, followed by sample size)
Species Note length Note interval Initial frequency Final frequency
(sec) (sec) (Hz) (Hz)
Amolops chunganensis
4 pulsed note 0.133±0.002 0.217+0.009 3440.0+142.9 3798.0+ 97.9
10 9 10 10
5 pulsed note 0.170+0.005 0.186+0.021 3163.6+192.4 3895.4+169.4
11 10 11 11
6 pulsed note 0.201+0.004 0.219+0.027 3250.0+ 70.7 3487.5+ 64.0
10 8 7 8
7 pulsed note 0.259+0.018 0.188+0.023 3004.5+230.7 3604.5+ 96.0
11 11 11 11
Amolops larutensis
1st note 0.175+0.006* 0.077+0.003 5440.0+114.0 4625.0+137.6
6 6 5 6
2nd note 0.029+0.002 0.090+0.002 4804.1+ 84.6 4387.5+213.2
6 —6 6 5
3rd note 0.032+0.001 4776.6+ 39.4 4362.5+175.4
—
6 6 6
Amolops jerboa
—
0.042+0.014 7152.9+403.1 4266.6+159.8
—
17 17 6
initial weak phase not included.
cember 1992, 2-3 January 1993, and 24 January the first note in frequency pattern. No harmonics
1993. respectively. Airtemperatures at the time of are evident in either of the three notes.
recordingvariedfrom 19.3 to21.7°C. Malescalled
at night on rocks in or at the edge of streams Amolopsjerboa
(width<5 m) with swift currents. Calls were recorded on Mt. Gunung Serapi, at
Calls were emitted with long intervals (22-37 the altitude of 150m, near Matang, Kuching,
sec). A call lasted about 0.49sec and included Sarawak, on 12 December 1990. Body tempera-
three discrete notes (Fig. IB). The first note was ture of males measured immediately after record-
long and unpulsed. and included two continuous ing averaged 25.2°C. Males were calling at night
phases. The first phase wasshort (about0.085 sec) on low trees along the edge of a swift stream
and had marked frequency modulation; it rose (width<5 m) with waterfalls.
gradually in frequency from 4300 Hzupto 5400 Hz The call (Fig. 1C, D) was emitted sporadically
and descended quickly to 4900 Hz. This phase, (intervals about 65-104sec), and consisted of a
especially from the start to the point with the very short, unpulsed note, lasting about 0.042sec
highest frequency, was weak and seldom traced in (Table 1). A call included two continuous phases,
the sonagram. and therefore, its length is omitted ofwhich the first onewasveryshort (about0.004-
in Table 1. The second phase was long (0.175 sec) 0.014sec) and had a marked frequency modula-
and almostconstant in frequency (about4900 Hz), tion; it descended quickly in frequency from over
but with a slight frequency modulation in the final 11000 Hz down to about 10000 Hz and again rose
portion (to 4600 Hz). Each ofthe subsequent two toover 11000 Hz. Thisphasewasonlyveryweakly
notes were produced with an interval of 0.077- traced in the sonagram (Fig. ID). The second
0.091 sec. They were short pulses (about 0.03 sec phase was longer (0.020-0.074sec), but with an
in length), but were similarto the second phase of even conspicuous frequency modulation, and the
694 M. MAtsui, G. F. Wu and H. S. Yong
frequency decreased rapidly from about 11000Hz reported a call of a Nepalese frog which might be
to 3000 Hz. Dominant frequencies were traced at that ofA. afghanus (A. marmoratus, according to
about 7100 and 4200 Hz ranges, and harmonic Dubois [2]), the type species of the genus Amo-
bands are often found between 4200-7800Hz lops. No sonagram was given, but the call was
ranges. reported to be a single note (length=0.2sec) with
a dominant frequency between 2000-2500Hz.
This frequency range is significantly lower than
DISCUSSION
that of the three species described above, but this
In describing three new species of the genus might be attributable to a difference in the body
AmolopsfromBorneo, the seniorauthor[7] noted size. In the note length, thecallseemstoresemble
remarkable differentiation in eggs and larval mor- the first noteofA. larutensis, but itisimpossibleto
phology between Chinese and Bornean members make further comparisons because of the lack of
of this genus and suggested that they represent additional data. No call data are available at
different evolutionary lineages. Later, Yang [10] present for members of Huia [10], either.
dividedthisgenus into three distinct genera: Amo- From our acoustic analyses, Dubois' treatment
lopsCope, 1865frommainlandSoutheasternAsia, [2] seems better than Yang's one [10] in that A.
Huia Yang, 1991 from Java, Sumatra, Borneo, chunganensis and A. larutensis are quite dissimilar
Thailand, and southern China, and Meristogenys in call characteristics, and are therefore to be split
Yang, 1991 endemictoBorneo. Dubois [2], on the taxonomically. Thus, the three species are clas-
otherhand, retained only one genusAmolops, but sified as A. (Amolops) chunganensis, A. (Amo)
divided it intofoursubgenera; he relegatedYang's larutensis, and^4. (Meristogenys)jerboa. Itshould
[10] three genera {Amolops, Huia, and Meris- be noted, however, Dubois' taxonomy [2] is far
togenys) to subgenera, and added a new subgenus from complete, including important mistakes such
Amo Dubois, 1992. According to their classifica- as inclusion of Rana taiwaniana in his subgenus
tions, A. chunganensis and A. jerboa belongs to Amolops [2: p. 321]. The species actually is not a
the genus orsubgenusAmolops and Meristogenys, memberofthe genusAmolops, but issynonymous
respectively, whereas A. larutensis is assigned to with Ranaswinhoana (Matsui, Unpublisheddata).
the genusAmolopsby Yang [10] orsubgenusAmo Future studies on additional species will demons-
byDubois [2]. AlthoughYang [10] adoptedcladis- trate the relationships of acoustic characteristics
tic methods in his revision, his interpretation of a and phylogeny of this genus.
monophyletic group is ambiguous, and we think it Although fairly differentiated in characteristics,
more reasonable to recognize only one genus calls ofthe three species described herein were all
Amolopsforfrogsfrom SoutheasternAsiathat are emitted at the edge ofrapidcurrents, where heavy
united by a distinct synapomorphy (larval abdo- noises, with additional songs ofother frog species
minal sucker). The problem is the number of and insects, seem to prevent frogs from transmit-
lineages present within this genus. ting acousticsignals. The three speciesincommon
As shown above, calls ofthe three species differ have calls with rather high dominant frequencies
considerably both in temporal and frequency pat- and more orlessclearfrequencymodulations. On
terns, and from these acousticcharacteristics, they the other hand, they are different in the degree of
seem to represent different lineages. Pope [9] these traits, and they even more differin temporal
noted that Rana (—Amolops) chunganensis from patterns of calls. Amolops chunganensis gives
Fujian produced several kinds of calls, one of nearlycontinuouscallswith distinct pulses, but the
which seems to conform to the call described frequency and degree of frequency modulation
above. Kiew [5] noted that Staurois (—Amolops) were lower than those of A. larutensis and A.
larutensis had a shrill call hardly audible above the jerboa. The call of A. larutensis is shorter and
roar of the water. No exact data for calls are emittedonlysporadically, but dominant frequency
available forother membersofthe genusAmolops is higher and frequency modulation is more
(sensu lato) except for Dubois [1], who briefly marked than in A. chunganensis. Finally, calls of
Acoustitic Characteristics of Amolops 695
A. jerboa are shortest and emitted with longest Ranidae (Amphibiens Anoures). Bull Mens Soc
intervals, but dominant frequency and degree of Linn Lyon 61: 350-352
frequency modulation are most conspicuous 3 Duellman WE (1970) The hylid frogs of Middle
America. Monog Mus Nat Hist Univ Kansas 1: 1-
amongthe three species. All ofthese differentcall
753
characteristics seem to be well adapted for the call 4 IngerRF(1966) The systematicsandzoogeography
to emerge above the heavy environmental noises. ofthe Amphibia ofBorneo. Fieldiana: Zool 52: 1-
402
5 Kiew BH (1990) Amphibian fauna of Taman
ACKNOWLEDGMENTS
Negara. J Wildlife and Parks 10: 96-108
WF
T. Hikida. K. Araya. D. Labang, and A. A. Hamid 6 Martin (1972) Evolution ofvocalization in the
hheellppeedd MlaMboriantotrhye fwioelrdktraipndtocrMiatliacaylsliya.reTa.dHtihkeidmaaanlus-o tEodadbygeWnuFs BBluafior,InUn"iEvvoolfutTieoxnasinPtrheessG,eAnuusstiBnu,fop"p
sSctraitpet.GoSvoecrino-mEecnotnoomfiScarReasweaakrckhindUlnyitpeorfmiMtatleadysMiaMantdo 7 2M7at9s-u3i09M (1986) Three new species of Amolops
work in Malaysia. This study was partly supported by from Borneo. Copeia 1986: 623-630
Grants under The Monbusho International Scientific 8 Nelson CE (1973) Matingcallsofthe Microhylinae:
Research Program (Field Research, Nos. 02041051, descriptionsandphylogeneticandecologicalconsid-
03041044. and04041068) andNational GeographicSoci- erations. Herpetologica 29: 6-17
ety (No.4505-91) to MM. 9 Pope CH (1931) NotesonamphibiansfromFukien,
Hainan, and other parts of China. Bull Amer Mus
Nat Hist 61: 397-611
REFERENCES 10 Yang DT (1991) Phylogenetic systematics of the
Amolops group ofranid frogs ofSoutheastern Asia
1 DuboisA(1976) Chantsetecologiechezlesamphi- and the Greater Sunda Islands. Fieldiana: Zool
biens du Nepal. Coll Int CNRS 268: 109-118 (New Ser) 63: 1-42
2 Dubois A (1992) Notes sur la classification des