SEMILUNAR REPRODUCTIVE CYCLES IN 



FUNDULVS HETEROCLITUS (PISCES: 



CYPRINODONTIDAE) IN AN AREA 



WITHOUT LUNAR TIDAL CYCLES 



Although lunar spawning rhythms are relatively 

 common in species of shallow-water fish, semilunar 

 reproductive cycles have been reported in a small but 

 growing number of species (Korringa 1947; 

 Johannes 1978). Species with the best documented 

 semilunar gonad and spawning cycles include the 

 California grunion, Leuresthes tenuis, (Clark 1925); 

 Atlantic silversides, Menidia menidia, (Middaugh 

 1981); the tropical coral reef saddleback wrasse, 

 Thallassoma duperrey, (Ross 1983); two tropical 

 damselfishes, Pom,acentrus flairicauda and P. wardi, 

 (Doherty 1983), and gulf killifish, Funduhus grandis, 

 (Greeley and MacGregor 1983); and the salt marsh 

 mummichog, Fundulus heteroclitus, (Taylor et al. 

 1979; Taylor and DiMichele 1980). On the east coast 

 of North America, F. heteroditics move up onto 

 marsh surfaces during high spring tides to spawn 

 either in empty mussel (Geukensia demissa) shells or 

 in the outer leaves of salt marsh plants (Spartina 

 altemiflora) where oxygen levels are relatively high 

 and sedimentation rates are relatively low (Able and 

 Castagna 1975; Taylor et al. 1977; Kneib and Stiven 

 1978; Taylor and DiMichele 1983). Its eggs are ex- 

 tremely tolerant of desiccation and hatch on the next 

 series of high spring tides. In addition, semilunar 

 rhythms of larval hatching are well known for inter- 

 tidal chironomid insects (Newmann 1978) and for 

 several species of intertidal and estuarine crabs (see 

 Christy 1982; Forward et al. 1982). 



In most cases, the proximal factors (sensu Giese 

 and Pearse 1974) responsible for initiating and syn- 

 chronizing these semilunar cycles are unknown. 

 However, among the factors postulated as cues are 

 tidal rhythms in water turbulence and hydrostatic 

 pressures (Korringa 1947; Newmann 1978; Weld 

 and Meier 1982; Ross 1983); tidal regimes in the 

 habitat (Forward et al. 1982); photoperiod in combi- 

 nation with tidal fluctuations in water temperature 

 (Miller et al. 1981), in combination with time of high 

 tide (Middaugh 1981; Middaugh and Takita 1983), or 

 in combination with interrupted current velocity 

 (Middaugh and Hemmer 1984); and moonlight 

 (Saigusa 1980). In addition to its wide distribution in 

 coastal marshes with distinct lunar cycles of spring 

 and neap tides, F. heteroclitus is abundant in exten- 

 sive shoreline habitats and marshes of Chesapeake 

 Bay, where tidal ranges are small and changes in 

 water level caused by barometric pressure and wind 

 frequently and unpredictably override lunar tidal 



levels and obliterate semilunar tidal cycles. In the 

 present study, we measured the reproductive activi- 

 ty of F. heteroclitus in an area without lunar tidal 

 cycles in order to determine if semilunar reproduc- 

 tive rhythms occur. The occurrence of semilunar 

 reproductive rhythms would suggest that the prox- 

 imal cues regulating the reproductive cycles are not 

 factors associated with changes in tidal levels, such 

 as pressure, or turbulence of currents. 



Methods 



This study was conducted from May through 

 August 1982 at a small tidal creek (Muddy Creek) 

 which flows into the Rhode River (lat. 38°5rN, long. 

 76°32'W), a subestuary on the western shore of cen- 

 tral Chesapeake Bay, located about 11.3 km south of 

 Annapolis, MD. The creek bottom consists of fine 

 clays and silts, and its banks are fringed by a cattail 

 {Typha angvstifolia) marsh. Water level was 

 measured with a Honeywell^ diffused silicon differ- 

 ential pressure transmitter (accurate to ± 1.238 cm) 

 at a station operated by the U.S. Geological Survey 

 on the Rhode River near the mouth of Muddy Creek 

 about 1,000 m from the site where fish were sam- 

 pled. Changes in water level at the monitoring sta- 

 tion accurately reflect water levels at the sampling 

 site (D. L. Correll^ and R. L. Cory^). Measured daily 

 high water levels were compared with predicted high 

 tidal levels published for the Rhode River (National 

 Oceanic and Atmospheric Administration 1981). 



Fundulus heteroclitus is the most abundant fish in 

 the creek (Mines unpubl. data). Fish were sampled 

 every 2-3 d during the hours of 0730-1230, using un- 

 baited minnow traps set just above low water level. 

 Each sample consisted of 8-12 males and 8-12 

 females which were tested for readiness to spawn, 

 and another 8-12 of each sex were taken to deter- 

 mine gonad indices. Readiness to spawn was deter- 

 mined in the field by applying gentle pressure from 

 anterior to posterior along the ventral surface of the 

 fish. Release of sperm or eggs was interpreted as the 

 fish being ripe. The gonad index for each fish in the 

 sample was determined by dissecting out the testes 

 or ovaries and by drying the gonad and body to con- 

 stant weight at 60°C and weighing them to the near- 



'Reference to trade names does not imply endorsement by the Na- 

 tional Marine Fisheries Service, NOAA. 



^D. L. Correll, Assistant Director, Smithsonian Environmental 

 Research Center, P.O. Box 28, Edgewater, MD 21037, pers. com- 

 mun. September 1982. 



3R. L. Cory, Oceanographer, U.S. Geological Survey, Smithsonian 

 Environmental Research Center, P.O. Box 28, Edgewater, MD 

 21037, pers. commun. September 1982. 



FISHERY BULLETIN: VOL. 83, NO. 3, 1985. 



467 



