LAUTH: CABEZON IN PUGET SOUND, WA 



Juan de Fuca and San Juan Islands (Fig. 1). All 

 specimens captured were weighed to the nearest 

 0.1 kg and total length measured to the nearest 



Processing of Ovaries 



From 1 to 10 females were sampled each month 

 so that the progression of ovarian development 

 could be followed throughout the study period and 

 spawning frequency could be determined. Entire 

 ovaries were excised, weighed to the nearest 0.1 g, 

 put in gauze bags, and placed in modified Gilson's 

 solution to harden the eggs and to break down 

 ovarian tissue (Simpson 1951). 



After 5 to 6 months in Gilson's solution, the eggs 

 from each ovary were separated from the ovarian 

 tissue using a mild jet of water while passing them 

 through a series of Tyler^ brass sieves with open- 

 ings of 1.651 mm, 0.295 mm, 0.180 mm, and 0.075 

 mm. Most eggs with diameters less than 0.075 mm 

 passed through the smallest screen and were dis- 

 carded. Loose eggs were retained by the sieves and 

 stored in jars with 5% formalin. 



Ova Diameter Frequencies 



Eggs and water (2.5 L) were homogeneously 

 mixed in a 4 L beaker with magnetic stirrer. A ran- 

 dom 5 mL subsample was drawn with a pipette and 

 the eggs were measured with a calibrated ocular 

 micrometer. 



At least 200 eggs were measured from each ovary. 

 Ova diameters were grouped using 0.05 mm incre- 

 ments as midpoints. Based on ova diameter frequen- 

 cy histograms, ovaries were grouped into eight 

 stages (I-VIII). Ranges, means, medians, and stan- 

 dard deviations were calculated for the apparent 

 modes within each stage. 



I calculated a wet gonadosomatic weight index 

 (WGSI) for each female using the formula of 

 Gunderson and Dygert (1988). The WGSFs for each 

 stage of ovarian development were averaged and 

 used as a measure of relative gonadal investment 

 of females. 



The Number of Eggs to be Spawned 



The subsampling procedure for estimating the 

 number of eggs to be spawned was identical to the 

 one for measuring ova diameters. Subsamples were 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



enumerated using a dissecting microscope, a gridded 

 petri dish, and a laboratory counter. When modes 

 in an ovary overlapped, the number of eggs from 

 the largest mode were counted twice and averaged. 

 At least three subsamples were taken for each ovary 

 and the mean total number of eggs to be spawned 

 was calculated, using a simple volumetric propor- 

 tion. If the coefficient of variation was greater than 

 10%, additional subsamples were made until it 

 dropped below 10%. 



Unweighted least-squares linear regression was 

 used to predict the total number of eggs to be 

 spawned using lengths and weights of females as 

 independent variables. All regression analyses were 

 done with a personal computer according to methods 

 described by Kleinbaum and Kupper (1978). 



RESULTS 



Seasonal Embryo Mass Abundance 



The beginning and end of the spawning season 

 were defined as the dates when embryo masses were 

 first and last seen. During 19 collection and tran- 

 sect dives throughout Puget Sound from mid- 

 September until the end of November, no cabezon 

 embryo masses were observed. The first embryo 

 mass observed in Puget Sound was on 6 December 

 1984. A sample of eggs from this embryo mass was 

 placed in a 5 gal bucket filled with seawater and 

 most hatched within 30 minutes. The eggs were ap- 

 parently near the end of their incubation period 

 since little of the yolk sac was remaining. Based on 

 work from this study, incubation time until hatch- 

 ing is several weeks, thus the embryo mass was most 

 likely deposited sometime in middle or late Novem- 

 ber 1984. 



Along Transects 1, 2, and 3, embryo masses were 

 first observed on 2 January 1985 and 7 and 21 

 December 1984, respectively. After the first embryo 

 masses appeared, there was a steady increase in 

 abundance with some fluctuation (Fig. 2). The peak 

 number of embryo masses at all three transects oc- 

 curred during March and April 1985, after which 

 there was a general decline. By 30 August 1985, em- 

 bryo masses were totally absent from Transects 1 

 and 2 and by mid-September 1985, none were found 

 at Transect 3 or elsewhere in Puget Sound. 



Between November 1984 and September 1985, 

 there were 35 embryo masses observed on Transect 

 2, 23 embryo masses on Transect 1, and 15 embryo 

 masses on Transect 3. It is possible that some em- 

 bryo masses may have hatched or disappeared (e.g., 

 predation, cannibalism, dislodged by physical dis- 



147 



