- Reproductive Ecology of the Gobiid Fish Eviota abax at Nobeoka, Japan, with Notes on Geographic Variation
The reproductive behavior and spawning cycle of the gobiid fish Eviota abax were observed in a rocky tide pool at Nobeoka, Miyazaki, Japan. Both sexes maintained nonterritorial, overlapping home ranges. The spawnings took place at the low tide of neap to spring tidal periods. Matings varied in each spawning cycle, but males did not simultaneously mate with multiple females. Males were larger than females in the spawning pairs. After spawning, only the male guarded the egg mass. Although separated by 900 km, the basic patterns of reproductive ecology were similar at Nobeoka to those reported earlier for this species from Kominato, Chiba, Japan; nest entrances were smaller at Nobeoka than at Kominato, and larger males kept their home ranges longer at Nobeoka.
Intraspecific Geographical variation in behavior has been the focus of recent research on the mechanism of evolution of behavior brought about by environmental variation (Foster and Endler 1999). Such behavioral variation could result in reproductive isolation of individuals from different habitats. Therefore, it is important to discuss conspecific behavioral variation among populations from different habitats to elucidate possible evolutional sequences of specialization (Verrell 1999).
However, comparative studies on geographic variation of behavior among reef fishes whose larvae disperse by water currents and settle out in different habitats following metamorphosis are rare. Yanagisawa (1982) investigated the reproductive ecology of the gobiid fish Amblyeleotris japonica from two different populations; Akagawa and Okiyama (1997) examined the reproductive and feeding behavior of the file fish Rudarius ercodes from four populations.
In this paper we describe the spawning cycle, mating system, and egg-guarding behavior of the gobiid fish Eviota abax at Nobeoka, Miyazaki, Japan, and compare it with the reproductive ecology of another population 900 km distant at Kominato, Chiba, Japan, previously studied by Taru and Sunobe (2000).
Eviota abax is a small species (<50 mm total length), inhabiting shallow subtidal and lower intertidal rocky shores and coral reefs in southern Japan (Lachner and Karnella 1980). The spawning season is from May to August in Kyushu (Dotsu et al. 1965). Larvae have been collected from June to October, but are particularly abundant from July to August at Kominato (Okabe 1998). Sunobe (1998) described courtship behavior of E. abax in an aquarium. The female lays an egg mass on the wall of the nest, and the male guards the embryos for approximately 5 days at 22-28°C until hatching (Dotsu et al. 1965, Sunobe and Nakazono 1987).
Materials and Methods
This study was carried out on the rocky shore in front of the Fisheries Research Laboratory of Miyazaki University, located on the east coast of Kyushu Island (32° 31´ N, 131° 42´ E). We observed the behavior of Eviota abax by snorkeling one to three times a day between 0530 and 1900 hours from 7 June to 31 August 1997, except for stormy days. Observations [End Page 35] were made on 67 days for a total of approximately 169 hr. The water temperature varied from 21.3 to 28.7°C during the observation period.
A quadrat (1.8 by 5.8 m) divided into 213 20 by 20 cm grid squares was set on a rock face and cobble field at 0-1.5 m depth inside a tide pool. Before the beginning of observations, from 23 May to 6 June, we captured all individuals of E. abax in the observation area with a hand net. They were individually discriminated by the color pattern on the nape and by clipping of the upper or lower part of the caudal fin. Standard length (SL) and total length (TL) were measured and sex was determined by the degree of elongation of the first two dorsal spines (which are elongated by 150-200% in males) and the shape of the urogenital papilla (Lachner and Karnella 1980). Previously unencountered individuals appearing after the start of observations on 7 June were cataloged in the same manner. The total number of identified males and females was 34 and 50, respectively.
We recorded the swimming tracks for each individual for 5 min on each...