Tarahumara Frog (Rana tarahumarae) Call Types and Characteristics
James C. Rorabaugh1 and Lang Elliot2
U.S. Fish and Wildlife Service, Phoenix, Arizona. Nature Sound Studio, Ithaca, New York.
The calls of),the Tarahumara Frog and arroyos from (Rana tarahumarae a species of barrancas
south-central Arizona to northern Sinaloa (Rorabaugh and Hale 2005), are poorly known. Hillis and Wilcox (2005) noted that all members of the Rana tarahumarae group lack vocal sacs and slits, have reduced or absent tympana, and no calls have been recorded for any of the species. Those authors believed that if calls are produced at all they are likely given underwater, owing to the highly aquatic nature of these species. In reference to the Tarahumara Frog, Zweifel (1968) noted that no mating call has been reported. However, Campbell (1934) heard Tarahumara Frogs squawk when they were poked with a stick and leapt into the water at Pea Blanca Spring, Arizona. Hale and May (1983) said Rana tarahumarae is generally considered voiceless, but both sexes have a release call, a low grunt of about one-half second duration, uttered once or twice (sometimes more) at half second intervals. In response to recent declines of Tarahumara Frogs in southern Arizona, a population was established in 2002 at Horse Tanks, Castle Dome Mountains, Kofa National Wildlife Refuge, Arizona, for the purpose of producing frogs and tadpoles for reestablishment into historical localities. While working on this project, we have noted and recorded a variety of calls by this species - more than we expected based on the previous literature. Herein we describe five Tarahumara Frog call types and elaborate on their acoustical characteristics and phenology. Methods The largest aquatic site at Horse Tanks is a deep tinaja that holds water year-round due to a downstream concrete dam and, when needed, pumping of water to the site, which routinely occurs in dry summers to maintain water for bighorn sheep and other wildlife. Occurrence and types of vocalizations of Tarahumara Frogs in the tinaja were monitored by ear 28 times from May 2002 through August 2006, typically from an hour before sunset to after dark. From 2100-2230, 4 August 2005, vocalizations were recorded at the tinaja with an Optimus CTR-114 cassette recorder and a DAK UEM-83R shotgun 134 SONORAN HERPETOLOGIST 19 (12) 2006
microphone. Recordings were made at a distance of 0.5 to a few meters from calling frogs. Water and air temperatures were 290 and 330 C, respectively. In the lab, these recordings were played back on a Nakamichi BX-300 studio cassette recorder and digitized as 16-bit, 48 kilohertz wave audio files on a PowerMac G5 computer using ProTools LE M-Box audio editing hardware and software. The best sections of the raw field recordings were identified, extracted, and sequenced to create a 1 minute 48.8 second segment presenting the clearest examples of the different call types. Analyses of the calls were then conducted using RAVEN Lite software (Cornell Lab of Ornithology, Bioacoustics Research Program). Results Calling was noted as early as 2 March and as late as 20 October, and during all visits from 20 April to 20 October. Calling was always noted when water temperatures exceeded 15.20 C, and was most common at dusk, but often continued after dark. Careful listening in the field and subsequent analyses of the recordings revealed four distinct call types, including the snore, a soft whine, a vibrant eep or eepup, and a throaty croak or squawk (given individually or preceding a snore). In addition, a fifth call type is recognized - a release or distress call - which is composed of a series of brief snores. Spectograms of all but the release or distress call are provided in Figures 1a-e and call characteristics are summarized in Table 1. Frequency range is defined by the maximum and minimum frequencies recorded for all calls of a call type. Dominant frequency is the frequency with the maximum intensity in decibels, and low and high end frequencies are the lowest and highest frequencies of a call (frequency spikes, which were common in snore calls, were eliminated from this calculation). Technically, the recordings were not very good, and deciphering some variables, such as pulse rate and number, was not possible for some calls (see Table 1 caption). The calls of this species are not very loud, and although some of the recordings were made as close as 0.5 m, they were not strongly differentiated from background noise. Extraneous noise at frequencies below 500 Hz made analysis of low frequency call characteristics difficult. The snore and whine calls were the most common calls heard at Horse Tanks, but calls of all types occurred throughout the active season from March through October. Release or distress calls were not clearly audible in the field recordings, although two instances of short snores of unmeasurable, but high pulse rate, may have been these calls. An mp3 version of the edited 1 minute 48.8 second recording can be accessed on the internet at: www.musicofnature.com/tarahumara/. Because Tarahumara Frogs lack vocal sacs, and the density of frogs was often quite high, it was usually difficult to identify which frog was calling. However, an adult male that was completely out of the water was

observed calling at close range. This frog produced typical snore calls, but it was not clear how the sounds were generated. The snores appeared to be produced during exhalations and were accompanied by a slight fluttering of the throat. Both juvenile and adult frogs were noted making both the snore and whine calls, and most calling appeared to occur out of water at the shoreline. On one occasion, a whine call was heard coming from an underwater pair in amplexus. When frogs were captured and handled, they often gave the release or distress call. Discussion Qualitatively, four of the five Tarahumara Frog call types are not unusual for ranid frogs. The snore is an advertisement call common to many ranids, including most North American leopard frogs, Red-legged Frog (Rana aurora), Mountain Yellow-legged Frog (Rana muscosa), and others (see Davidson 1995, 1996). In the Tarahumara Frog, pulse rate and intensity often increased over the duration of each snore, and four of the snore calls were preceded by introductory pulses given 0.1-0.2 s before the rest of the snore. Eeps are also made by several species, and the individual croak or squawk call is similar if not identical to the component pulses in the snore call. The whine call is unique among Arizona anurans and is somewhat similar to the groan or growl call of the California Red-legged Frog (R. a. draytonii), as well as calls of the Cascade Frog (R. cascadae, Davidson 1995). The whine call is characterized by a narrow frequency range (707-1040 Hz), as measured by the mean low and high end frequencies, and relatively high pulse rate (Table 1). It varies in duration, and appears to be given either as a short note (Figure 1.b.), similar in duration to the eep call, or in a longer form up to 1.12 s (Figure 1.c.). As noted by Hale and May (1983) and as observed by us, males, females, and juveniles utter a release or distress call that is commonly given when captured. However, rather than a grunt, as described by Hale and May, it is more precisely described as a short snore of relatively high pulse rate, given
once to several times. Hale and May (1983) referred to these short snores as release calls; alternatively, they could be distress calls. Release calls are typically given when males or non-receptive females are approached and grasped by breeding males, but may also be given when frogs are captured by hand. Distress calls are often given when a frog is captured by a predator, and may serve to distract or startle a predator into releasing the frog. It may also attract other predators; and during subsequent disputes or encounters between predators, the frog may escape (Stebbins and Cohen 1995). The purposes of the other calls are unclear. The snore call, if similar to other ranids, would be an advertisement call given by adult males. However, it is given by both adult and juvenile frogs and was heard from March through October, whereas the breeding season for this species is typically April-May and occasionally in late summer (Rorabaugh and Hale 2005). Although the whine call was heard once from a pair in amplexus, it was often heard when no amplexing frogs were found. Eeps and individual croaks or squawks were also not found to be associated with particular behaviors. Campbells (1934) observation of a squawk when Tarahumara Frogs were poked with a stick is consistent with an alarm call. American Bullfrogs (Rana catesbeiana), particularly juveniles, make a squawk alarm call, but we noted no such alarm call by Tarahumara Frogs at Horse Tanks. Horse Tanks is outside of the native range of the Tarahumara Frog, and may be hotter and drier than most or all sites where the species occurs naturally. Whether our observations at Horse Tanks are typical for the species is unknown. The recordings were made at a water temperature of 290 C. This is higher than in most natural habitats of the species (Hale et al. 1998,

Calling was noted as early as 2 March and as late as 20 October, and during all visits from 20 April to 20 October.
Figure 1a-e. Spectograms of Tarahumara Frog calls. The arrows show the location of call signatures. (Inset: Tarahumara Frog [Rana tarahumarae], Santa Rita Mountains, Arizona. Photo by Jim Rorabaugh).
SONORAN HERPETOLOGIST 19(12) 2006
CALL TYPE SAMPLE SIZE (n)
Table 1. Descriptive statistics of Tarahumara Frog call types.

SNORE (52)

WHINE (20)

EEP (13)

CROAK/ SQUAWK (8 ) 1 (1), (0), (8)
# pulses could not be counted for one snore call. 2 # pulses could only be counted for one whine call. 3 # pulses could only be counted for one eep call. 4 Dominant frequency could not be determined for two snore calls.
MEAN # PULS ES (range), variance, sample size (n) MEAN DURATION (S), (range), variance, sample size (n) MEAN PULS E RATE #/S, (range), variance, sample size (n) FREQUENCY RANGE (HZ), samp le size (n) MEAN LOW END FREQUENCY (HZ), (range), variance, sample size (n) MEAN HIGH END FREQUENCY (HZ), (range), variance, sample size (n) MEAN DOMINANT FREQUENCY (HZ), (range), variance, sample size (n)

21 (4-112), 320, (511 )

8 (NA), NA, (12 )

9 (NA), NA, (13 )

0.59 (0.09-2.19), 0.15, (52) 37.0 (14.8-70.3), 219, (511 )
0.260 (0.0341.12), 0.085, (20) 77.7 (NA), NA, (13 )
0.074 (0.027-0.17), 0.00172, (16) 101.1 (NA), NA, (14 )
0.015 (0.0070.023), 0.00003, (8) 1 (1), (0), (8)
38-9004 (52) 326 (38-766), 27345, (52)
528-1346 (20) 707 (528-943), 12567, (20)
727-7033 (16) 1113 (727-1941), 114168, (16)
113-5813 (8) 309, (113-545), 24912, (8)
2086 (1094-4818), 850714, (52)
1040 (838-1346), 20084, (20)
4474 (2000-7033), 2507340, (16)
2339 (969-5813), 2406056, (8)
740 (464-1117), 26547, (504 )
847 (657-1029), 13604, (20)
1562 (891-4136), 573412, (16)
683 (377-941), 24260, (8)
Hale 2001), although water temperatures of 27, 28, and 310 C have been recorded in some pools occupied by Tarahumara Frogs in Big Casa Blanca Canyon (from monitoring reports 2004 to the present). Call frequency and pulse rate are typically higher at elevated temperatures. Hence, if heard at lower temperatures, these calls would likely sound slower and lower pitched, and may vary in other ways as well (Davidson 1995). One of us (JCR) has heard snore and whine calls in Big Casa Blanca Canyon in June and July, which qualitatively sounded similar to the recorded calls from Horse Tanks. Literature Cited Campbell, B. 1934. Report on a collection of reptiles and amphibians made in Arizona during the summer of 1933. Occasional Papers of the Museum of Zoology, University of Michigan 289:1-10. Davidson, C. 1995. Frog and toad calls of the Pacific Coast. The Library of Natural Sounds, Cornell Laboratory of Ornithology, Ithaca, NY and USDA Forest Service. Davidson, C. 1996. Frog and toad calls of the Rocky Mountains. Library of Natural Sounds, Cornell Laboratory of Ornithology, Ithaca, NY and USDA Forest Service. Hale, S.F. 2001. The status of the Tarahumara frog in Sonora, Mexico based on a re-survey of selected 136 SONORAN HERPETOLOGIST 19 (12) 2006

localities, and search for additional populations. Report to the U.S. Fish and Wildlife Service, Phoenix, Arizona. Hale, S.F., G.M. Ferguson, P.A. Holm, and E.B. Wirt. 1998. Re-survey of selected Tarahumara frog (Rana tarahumarae) localities in northern Sonora, Mexico, in May 1998. Report to the Arizona Zoological Society and the Tarahumara Frog Conservation Team. Hale, S.F., and C.J. May. 1983. Status report for Rana tarahumarae Boulenger. Report to the Office of Endangered Species, U.S. Fish and Wildlife Service, Albuquerque, New Mexico. Hillis, D.M., and T.P. Wilcox. 2005. Phylogeny of New World true frogs (Rana). Molecular Phylogenetics and Evolution 34:299-314. Rorabaugh, J.C., and S.F. Hale. 2005. Rana tarahumarae Boulenger, 1917, Tarahumara frog. Pages 593-595 in M.J. Lannoo (ed), Amphibian Declines: The Conservation Status of United States Species. University of California Press, Berkeley. Stebbins, R.C., and N.W. Cohen. 1995. A Natural History of Amphibians. Princeton University Press, Princeton, New Jersey. Zweifel, R.G. 1968. Rana tarahumarae. Catalogue of American Amphibians and Reptiles 66:1-2.