Social Behavior and Communication in the Neotropical Cicada Fidicina mannifera (Fabricius) (Homoptera: Cicadidae)

We investigated signalling behavior in relation to social context in the cicada Fidicina mannifera. Four types of acoustic signals were recorded from males: songs, calls, low-amplitude songs, and disturbance sounds. Signal types were similar in frequency and fine temporal structure, but differed markedly in gross temporal structure. Songs and calls were the most frequently observed signals, and these differed in the social context in which they were used. When inter-male distances were large, males produced songs, overlapping them with songs of neighboring males. As nearest-neighbor distance decreased, males switched to calls, which were typically alternated back and forth between males. Bouts of calls often were followed by a parallel walk display, which apparently was involved in settling disputes over calling sites. Singing cicadas are a prominent feature of the acoustic environment in many temperate and tropical areas. Cicada songs are diverse and often have provided a basis for taxonomic decisions and species identification (Jiang, 1985; Joermann and Schneider, 1987; Villet, 1988; Fonseca, 1991; Daniel et al., 1993). Further more, the mechanisms of sound production and sound reception have been well studied (Pringle, 1954; Moore and Sawyer, 1966; Simmons and Young, 1978; Josephson and Young, 1979, 1981; Huber, 1983; Young and Josephson, 1983; Popov and Sergeeva, 1987; Young, 1990; Popov, 1990; Bennet-Clark and Young, 1992; Fonseca and Popov, 1994; Henning et al., 1994a, b). In contrast, there are relatively few studies of the social behavior of cicadas in the field (Alexander and Moore, 1958, 1962; Claridge et al., 1979; Doolan, 1981; Gwynne, 1987; Doolan and Young, 1989; Villet, 1992), in part because many species sing from the forest canopy and are difficult to observe. As a result, the functional aspects of most cicada communication systems remain poorly known. In this study we describe the social behavior of a neotropical cicada, Fidicina mannifera (Fabricius). This species provides a rare opportunity to study cicada behavior in the field, because most of its activity takes place on tree trunks well below the canopy. In addition to describing the signals used by males, we examine the social context in which two of the most commonly-used signals are given. We also present previously unreported aspects of their social behavior. Materials and Methods The study was conducted at the BIOLAT Biological Station, located at Puesto de Vigilancia Pakitza in the Zona Reservada del Manu, about 10 km south of the 1 Present address: Section of Neurobiology and Behavior Mudd Hall, Cornell University, Ithaca, New York 14853. This content downloaded from 161.130.252.217 on Mon, 7 Jul 2014 22:40:43 PM All use subject to JSTOR Terms and Conditions 86 JOURNAL OF THE KANSAS ENTOMOLOGICAL SOCIETY Parque Nacional del Manu (11?55'48"S, 71?15'18"W; elevation 350 m). This area is in the department of Madre de Dios in southeastern Peru. The study was con ducted from 26 September to 25 October, 1990, during the transition from the dry to the wet season. Fidicina mannifera is widely distributed, ranging from Costa Rica to Argentina (Distant, 1883; Jacobi, 1907; Goding, 1925; Torres, 1953; Young, 1976). In the moist tropical forests of southeastern Peru, adults of F. mannifera have been collected from 12 September to 25 November, which includes the late dry season and early rainy season. At Pakitza, adults occur on trees in non-flooded forest (especially old alluvial terrace forest) and in the clearing surrounding the build ings. Males call on tree trunks during sunny periods in small aggregations on one or several nearby trees. Males also form a pronounced evening chorus. Sounds were recorded using either a Sony TCD-5M or a Marantz PMD 420 cassette recorder with a Sennheiser ME-80 microphone and K3U power module. Both recorder-microphone combinations provided a flat frequency response up to 14 kHz. Air temperature at or near the recorded male's calling site was measured to the nearest 0.2?C with a Miller & Weber quick-reading thermometer. Sound pressure level was measured with a Realistic Sound Pressure Level meter, using the Fast response setting and A weighting. Recordings were analyzed using a DATA 6000A Universal Waveform Analyzer for frequency and fine-temporal measurements and a Multigon Uniscan II Real Time Analyzer for gross temporal measurements. The sampling rate for waveform analysis was 40 kHz. Fourier transforms were performed on samples of 8192 points, and thus the frequency resolution (sampling rate/number of points) was about 5 Hz. In practice, factors such as small variations in tape recorder speed are likely to lessen the actual level of resolution. Five-minute observational samples of calling behavior were made on focal males and their nearest calling neighbors. Signals emitted during the sample pe riod were counted and scored as songs or calls (these signal types were differ entiated easily on the basis of temporal differences; see below). Distance between males was measured with a tape measure or, for pairs of males on the same tree but too high to measure, estimated to the nearest 0.5 m. The proportion of calls vs. songs was arcsine-transformed and regressed on the log of nearest-neighbor