Correlation of acoustic and visual signals in the cichlid fish, Tramitichromis intermedius Jennifer L. Ripley a,b & Phillip S. Lobel a a Boston University Marine Program, Marine Biological Laboratory, 7 MBL Street, Woods Hole, MA 02543, U.S.A. b Current address: Department of Biology, West Virginia University, 53 Campus Drive, P.O. Box 6057, Morgantown, WV 26506, U.S.A. (e-mail: jripley@mix.wvu.edu) Received 15 December 2003 Accepted 2 March 2004 Key words: Cichlidae, sound production, spawning behavior, behavior ontogeny, mate choice Synopsis We investigated acoustic and visual communication concurrently in wild caught adult and captive-born, first generation offspring of the East African Rift Lake cichlid fish Tramitichromis intermedius. Only males emit sound during courtship. Sound production is always accompanied by quivering, but quiver behavior is not always accompanied by sound. This separation of quivering and sound supports the hypothesis that sound production is intentional serving a communicative role. As spawning nears, both sound production and quiver behavior increase. In terms of the ontogeny of sound production, the first observation of courtship occurs just days before the first spawning event and the first sound emission accompanies the first courtship activity. The accompaniment of quivering with sound as well as the escalation of the two behaviors with the approach of spawning follows similar patterns in wild caught and captive-born males. The tight correlation between behavior and sound production in both groups indicates their simultaneous performance plays an important role in reproduction. It is probable that the ability to produce sound and perform quiver behavior at the same time may be a measure of mate quality. Introduction In teleost fishes, sound has been recorded during courtship, spawning, defense, aggression and feeding. Sound production is often accompanied by visual signals (Myrberg 2002). Observations repeatedly link sound with two motivational and/ or physiological states, being reproduction and aggression (Myrberg 2002), but few studies corre- late a specific behavior with sound production. If sound production and behavior are highly corre- lated, presentation of both may be necessary to produce the intended response. On the other hand, the ability to separate may allow a greater range of messages by communicating different meanings whether the behaviors occur alone or together. Furthermore, sound production by some species may be the indirect result of another behavior rather than intentional communication (James & Heck 1994). A behavior performed without acoustic accompaniment suggests that the fish generates sounds purposely. Sound production is widespread in the family Cichlidae with twenty species documented to produce sound during aggressive and reproductive interactions (Lobel 1998, 2001). The sound-pro- ducing mechanism in these fishes may involve sound generation by the pharyngeal mill, amplified by the swimbladder (Lobel 2001, Kaatz 2002, Rice & Lobel 2002). Most studies of cichlid bioacous- tics focus on general descriptions of the call parameters and the associated physiological states Environmental Biology of Fishes 71: 389–394, 2004. Ó 2004 Kluwer Academic Publishers. Printed in the Netherlands.