Nesting and Paternal Care in the Weakly Electric Fish Gymnotus (Gymnotiformes: Gymnotidae) with Descriptions of Larval and Adult Electric Organ Discharges of Two Species
We describe nesting and paternal care in two species of the weakly electric fish, Gymnotus: Gymnotus carapo from savanna swamps on the Caribbean island of Trinidad; and Gymnotus mamiraua from whitewater floodplains of the Amazon Basin. In both species, single adult males guard eggs and juveniles. Male G. carapo excavate depressions in the substrate or nest in the roots of aquatic macrophytes. Male G. mamiraua form nests exclusively in the root mass of floating meadows of macrophytes. Twelve nests of G. mamiraua were encountered, containing juveniles up to a maximum total length (TL) of 67 mm and estimated maximum age of 8–12 weeks. Two of these nests exhibit a bimodal size distribution, implying more than one spawning event. Juveniles remain close to the male during the day and disperse little more than 1 m away at night. Larval G. carapo and G. mamiraua (approximately 15– 20 mm TL) generate a characteristic Electric Organ Discharge (EOD) with a dominant positive phase followed by a very weak negative phase. The Peak Power Frequency of the larval EOD is in the range 0.015–0.04 kHz. It is possible that the EODs of larvae may be sensed by animals with ampullary electroreceptors, including the male as well as predatory catfishes. Larval G. carapo and G. mamiraua generate EODs at a low repetition rate compared to adults and with no increase at night. This low rate reduces the probability of coincident pulses and may mitigate sensory jamming between nestlings in crowded nests. The repetition rate, and the waveform and spectral features of the EODs from 40 mm TL specimens of G. mamiraua, resemble those of mature adults.Abstract
Map of Tefé area in the lowland Amazon Basin of Brazil. Localities at which nests of Gymnotus mamiraua were encountered are numbered 1–8 (see Table 1). Areas exposed to an annual flood regime are stippled. White areas are Tertiary terra firme peneplain. The terrain to the north of the Rio Solimões is Quaternary whitewater floodplain (várzea). Base map traced from 1999 1:150:000 Landsat TM5 images. Flood extent plotted from June 1995 NASDA JERS-1 images
Field recording from a Gymnotus mamiraua nest (nest 6, Table 1) in a floating meadow of the várzea floodplain. Voltage and time are displayed on the vertical and horizontal axes, respectively. The recording was made at night when the nestlings are maximally dispersed. The EODs of the single male (EOD peaks exceeds dotted lines) and many juveniles (EOD peaks do not exceed dotted lines) are in most cases coincidental. Variation in EOD amplitude is due to variation in distance from the recording electrode as well as relative voltages
Histograms of size distribution of juvenile Gymnotus mamiraua in 12 nests from floating meadows of the Amazon floodplain. Total length (mm) is displayed on the x-axis and frequency of occurrence on the y-axis. Single dots indicate mean where data fit a normal distribution. Double dots indicate modes of a bimodal distribution
Field recording of larval (A,B) and adult (C,D) EODs of Gymnotus carapo from Trinidad. Sequences of EODs (A,C) and single EODs (B,D) are plotted with head positivity upwards and the time base expanded for the single EOD. Fluctuation in the peak-to-peak amplitude of the sequences of EODs are caused by movement of the fish relative to the recording electrode. The noisy background in (A) and (B) is an artifact of magnetic tape recording. (E) Photograph of adult G. carapo from nearby Venezuela (UF 77334). Gymnotus carapo from Trinidad are indistinguishable in pigmentation, meristics and morphometrics from populations in Venezuela and the Guyana Shield (Albert and Crampton, 2003)
Developmental changes in the EOD waveform of Gymnotus mamiraua. Left row illustrates EOD plotted with head positivity upward and component phases labeled P0 through P3, where P1 is the dominant positive component. Scale bar = 1 msec. Size of recorded specimen to upper left (below letter). Size range with similar EOD shape/duration annotated in upper right corner for N recorded specimens. Right column shows Power Spectrum computed by a 2048 point Fast Fourier Transform with frequency (kHz) on x-axis, amplitude in dB on y-axis (maximum 0dB, minimum −20dB), and Peak Power Frequency (PPF) scaled to 0dB. PPF ranges are annotated in upper right corner. Scale bar for photographs = 10 mm. Photograph not available for 20 mm specimen. Waveforms in this series are from specimens from more than one locality
EOD Peak Power Frequency as a function of body length in Gymnotus mamiraua nestlings from a single nest. EODs were recorded from only 20 of the 45 juveniles in Nest 7 (see Table 1). The 10 datapoints to the left refer to specimens from the cohort in this nest composed of smaller juveniles (see Fig. 3). Datapoints to the right are from the cohort composed of larger specimens. EODs plotted with head positivity upwards. Scale bar = 1 msec
Diel patterns of EOD repetition rate in three size classes of Gymnotus mamiraua. Time on the horizontal scale begins at 1200 h and ends at 1100 h on the following day. Note that small juveniles have a lower repetition rate with no appreciable increase from day to night
Contributor Notes
Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611-7800. E-mail: (WGRC) willc@flmnh.ufl.edu Send reprint requests to WGRC.