Fitzhugh et al.: Reproductive biology of Pogonias cromis in Louisiana 



245 



tion changes and potential egg production to maintain 

 future fishing harvests. We characterize ovarian de- 

 velopment, seasonal spawning duration, and frequency 

 as determined by ovarian histology and batch fecun- 

 dity in Louisiana coastal waters. 



Materials and methods 



We sampled black drum monthly from commercial land- 

 ings during March, June, July, October, November, and 

 December 1986 and July 1987 to obtain reproductive 

 information. We increased sampling effort during the 

 period of reported peak seasonal reproductive activity 

 and sampled 25 commercial landings during February, 

 March, April, May, and June 1987. We also sampled 

 recreational hook-and-line landings during March and 

 July 1986 and April 1987. Landings sampled from in- 

 shore waters (bays and sounds) were primarily taken 

 by gillnet, haul-seine, and hook-and-line. Landings 

 sampled from offshore waters were taken by trawl and 

 purse-seine. 



In order to contrast size-at-maturity with other stud- 

 ies, we made gross visual classifications of gonads dur- 

 ing sampling. Macroscopic characteristics for classify- 

 ing gonads as mature correspond to Bagenal (1968) 

 and Nielson & Johnson (1983). Female characteristics 

 included the presence of eggs visible to the naked eye 

 and light-yellow to reddish appearance from increased 

 vascularization of the ovary. Characteristics for ma- 

 ture males included white appearance and relative 

 enlargement of testes within the body cavity. Measure- 

 ments included fork length (FL), sex, gutted (viscera 

 removed) body weight (BW), and gonad weight (GW; 

 wet weight blotted dry to nearest 0.1 g). We documented 

 reproductive development by expressing gonad weight 

 as a function of body size using the gonosomatic index 

 (GSI) (Htun-Han 1978, Nielson & Johnson 1983). 



We held gonads in ice up to 24 h after sampling and 

 then fixed gonads in 10% formalin. One tissue sample 

 was randomly selected from the preserved ovary and 

 placed in an OmniSette tissue cassette. For histologi- 

 cal observation, tissue samples were dehydrated, em- 

 bedded in paraffin, sectioned, stained with Gill's he- 

 matoxylin, and counter-stained with eosin*. We 

 classified oocytes from the prepared histology slides 

 following Wallace & Selman (1981), DeVlaming ( 1983), 

 and Selman & Wallace (1986). These stages include 

 primary growth (PG), cortical alveoli (CA), vitellogen- 

 esis (V), and hydration (H). 



* Preparation of histological slides, including washing, embedding, 

 sectioning and staining, were completed by the Louisiana State Uni- 

 versity School for Veterinary Medicine, Department of Pathology. 



To determine relative frequency of oocyte stages, we 

 located a random starting point on a histological sec- 

 tion and counted and staged all identifiable oocytes 

 within a field before moving to a new microscope field 

 using manual stage drive. Field movement was in- 

 ward along the ovigerous lamellae, from the outer tu- 

 nica albuginea toward the center of the ovary, with 

 realignment along a vertical axis. To be counted, >50% 

 of an oocyte must have been within a field of view. We 

 counted and staged a minimum of 200 oocytes from 

 each female and expressed tallies of the four oocyte 

 stages as a percentage of the total count ( Htun-Han 

 1978, Holdway & Beamish 1985). The Bioquant IV 

 image analysis system software, IBM PC, and Hous- 

 ton Instrument digitizing pad (Hipad model DT-11) 

 were used in conjunction with an Olympus microscope 

 (with video attachment) to facilitate counts and 

 measurements. 



In addition to relative frequency of developmental 

 stages, we classified each histological section for the 

 presence of postovulatory follicles (POF) and atretic 

 oocytes to aid in determination of spawning frequency. 

 Our atresia classification was modified from that for 

 northern anchovy (Hunter & Macewicz 1985). If no 

 atresia of yolked oocytes was observed, we denoted the 

 ovary as atretic state 0. Tissue sections exhibiting 

 yolked oocytes undergoing atresia at <50%, >50%, 

 and 100% were classed as atretic states 1, 2, and 3, 

 respectively. 



To estimate oocyte development rate, we related our 

 histological observations of actively-spawning black 

 drum to published accounts of spawning time of black 

 drum (Mok & Gilmore 1983, Holt et al. 1985) and 

 rates of hydration and postovulatory follicle degenera- 

 tion in sciaenids (DeMartini & Fountain 1981, Brown- 

 Peterson et al. 1988). With this time-calibrated histo- 

 logical information, we estimated the hours from 

 spawning for females displaying yolk coalescence, hy- 

 dration, postovulatory follicles, and atresia, and clas- 

 sified day-0, day-1, and nonspawning females. Our es- 

 timate of seasonal spawning frequency was determined 

 by taking the average of the fractions of day-0 and 

 day-1 spawning females relative to total females ob- 

 served histologically with vitellogenic oocytes. Batch 

 fecundity, the number of hydrated oocytes which com- 

 prise the leading "batch" of eggs immediately prior to 

 spawning, was determined from formalin-fixed tissue 

 samples taken from each visibly-hydrated ovary. We 

 took replicate ovarian tissue samples (l-2g) from an- 

 terior, mid-, and posterior regions of left and right 

 lobes. In order to obtain 100-300 hydrated oocytes, 

 tissue subsamples of 90-100 mg (weighed to the near- 

 est 0.05 mg) were placed on a slide, glycerin added, 

 and hydrated oocytes counted (Hunter et al. 1985). 

 After observation of histological sections, any ovaries 



