FISHERY BULLETIN: VOL. 73. NO. 4 



tiveness of sex pheromone has also been demon- 

 strated by Rossi (1969) in Colisa lalia and C. 

 lahiusa. In this case a female pheromone can in- 

 duce nest building in heterospecific males. 



As to the chemical nature of piscine sex 

 pheromones, Amouriq (1965) identified an es- 

 trogen as the pheromone inducing hyperactivity 

 in male Lebistes reticulata, and Tavolga (1956) 

 identified the internal fluid of the ovary as the 

 source of the chemical stimulus eliciting courtship 

 behavior of male in Bafhijgobius soporator. In the 

 present study, of six trials consisting of placing 

 test females into fresh tap water previously oc- 

 cupied for 24 h by a male, only four positive re- 

 sponses were recorded. Since the metabolites 

 produced by a male in 24 h are far below the 

 threshold level for inhibiting ovulation, as we have 

 experienced and as Greene (1966) has reported, the 

 two failures out of six trials in the above 

 experiment suggest either that the male 

 pheromone is short lived or that less male 

 pheromone was released in the chemical absence 

 of the female. 



The selective advantage of an ovulating 

 pheromone in the zebrafish is not clear. Although 

 Hart and Messina (1972) claimed to be able to ob- 

 tain sperm from male zebrafish at all times under 

 laboratory conditions, we often encountered un- 

 successful milking of males. If both sexes are not 

 sexually ready at all times, it would be advan- 

 tageous to synchronize sexual readiness between 

 sexes. If release of the ovulating pheromone 

 corresponds with male readiness, synchronization 

 would be guaranteed. However, males appeared to 

 release the pheromone quite regularly, even while 

 in the presence of metabolites as suggested by the 

 50% ovulation in experiment 4. Yet, it is possible 

 that while the metabolites may repress testicular 

 development or spermiation and the release of the 

 pheromone by the males, the test male introduced 

 at the same time as the test female into the two 

 small chambers of experiment 4 was often already 

 sexually ready at that time and would thus release 

 the pheromone regardless of the presence of me- 

 tabolites. 



Such a pheromone, if functioning in 

 synchronization, would be advantageous for a 

 species with a long spawning interperiod. 

 Although female zebrafish can spawn every 1 or 2 

 days under laboratory conditions (Eaton and 

 Farley 1974), the spawning interperiod in the na- 

 tive habitat is not known. 



In some fishes, it is possible that the active 



chasing of the female by the male prior to spawn- 

 ing takes part in stimulating ovulation. An 

 ovulating pheromone would conserve such chasing 

 energy and therefore, be selectively advan- 

 tageous. 



The inhibitory effect of metabolic wastes on fish 

 reproduction has been reported by Greene (1966) 

 who found that an increase in metabolite concen- 

 tration resulted in a decrease in the number of 

 successful natural spawnings in the zebrafish. Lin 

 (1935) observed that grass carp, Ctenopharyn- 

 godon ideUu^, would spawn only after a rise in the 

 river water due to rain. Similar observations 

 confirming the coincidence of heavy rain and 

 spawning have been made by von Ihering and 

 Wright (1935) and Lake (1967). Lake suggested 

 that the stimulatory effect of rain on fish spawn- 

 ing was through addition of soil elements through 

 runoffs. However, according to Swingle (1956), 

 draining a pond crowded with goldfish or 

 largemouth bass and refilling it subsequently with 

 new water could induce spawning in the pond fish. 

 One of us (Chen) had observed on numerous oc- 

 casions that goldfish spawned during or after rain 

 in outdoor concrete tanks. In these cases, spawn- 

 ing occurred without input of soil elements. 

 Swingle (1956) suggested that the effect of rain 

 was to dilute a spawning repressive factor. It is 

 obvious from the present experiment that this 

 repressive factor is metabolites. 



Tang (1963) noted that the testes of silver carp, 

 Hypophthalmichthij!^ moUtn'x, would develop only 

 after the volume of the reservoir had been 

 increased by rain, thus suggesting that matura- 

 tion of testes may be retarded by waste products 

 from fish and that new water, or dilution of these 

 wastes, is necessary for sexual development. It is 

 possible that removal of the metabolites can also 

 induce the release of the pheromone by the male 

 zebrafish, indirectly stimulating ovulation in the 

 females. In the present study, however, removal of 

 the metabolites apparently had a direct effect on 

 the females, as mere exposure of the females to 

 fresh tap water resulted in ovulation in nearly half 

 of the trials (experiment 7). Tang (1957) reported 

 that female common carp inhibited from spawn- 

 ing by metabolic wastes would release eggs in the 

 absence of males upon introduction of new water. 



The chemical nature of the inhibiting me- 

 tabolites is not known. Greene (1966) believed that 

 they were ammonia. FVom the observation made 

 by Swingle (1965) that crowding of bluegill 

 inhibited spawning in the largemouth bass in the 



892 



