624 



Fishery Bulletin 104(4) 



30 



20 



10 



 Fall 1994 

 □ Winter 1995 



JI 



2 4 6 8 10 12 14 



Size class (cm) 



Figure 1 



Size distribution for mature female pink sea- 

 perch iZalembius rosaceus) (n=10.3) collected in 

 fall 1994 (black bars) and winter 1995 (white 

 bars). Bars represent 1-cm size classes. 



difference was significant (^test: ? = 5.15, df=55, 

 P<0.001). The incidence of premature births for winter 

 females was 18% (seven of 38 females) compared to 0% 

 (0 of 27 females) for fall females, and differed signifi- 

 cantly (Fisher's exact test: P=0.04). Female size was a 

 good predictor of brood size in fall but not in winter 

 (fall: 7-2 = 0.39; winter: r- = 0.00, Fig. 2A). 



Embryos were more developed and larger in win- 

 ter than fall, and larger females brooded larger-sized 

 embryos (Fig. 2B). Fall embryos had a mean SL of 

 21.3 mm (±4.2 mm SD) and were characterized by pale 

 pink coloration, translucent tissue, and the absence of 

 distinct scales. Winter embryos had a mean SL of 36.4 

 mm (±5.8 mm SD) and had firm and opaque tissue, 

 distinct scales, iridescent pink coloration, and appeared 

 as miniature adults. In both fall and winter, linear 

 regressions of embryo-on-female size were significant 

 (fall: r'-=0.17; winter: 7-2=0.20, Fig. 2B). 



Discussion 



Both embryo characteristics and female reproductive 

 condition provided evidence of winter parturition in 

 pink seaperch and confirmed conclusions made by Gold- 

 berg and Ticknor (1977). Winter embryos had advanced 

 morphological characteristics similar to those of adults, 

 compared to embryos examined in fall, and were similar 

 in size to those reported by Goldberg and Ticknor ( 1977 ). 

 Evidence of a relatively high incidence of abortion and 

 presence of many embryos (77 = 51) in winter collections 

 provided further evidence that parturition occurs in or 

 around January. Winter parturition is unique among 

 surfperches yet no studies, to the authors knowledge, 

 have provided an explanation for the unusual breeding 

 schedule of pink seaperch. 



Results from this study demonstrate a positive re- 

 lationship between brood size and female size in pink 



14 T 



12 



10 



N O 



u) 



•o 5 



O 



O 



m '^ 



2 

 



50 

 45 



? -"^ 

 § 35 



<^ 30 



o 



>, 



^ 25-1 



£ 



UJ 20 



15 



10 



B 



90 100 110 120 130 140 150 160 170 



Female SL (mm) 



Figure 2 



Linear regressions of I A) brood size (fall: r- = 0.39, 

 brood size = -l'2.9 + 0.14.r, ;i=26; winter: r-= 0.00, 

 ;( = 31 ), and (B) embryo size (fall: r- = 0.17, embryo 

 SL = 4.9 + O.lS.v; winter: /•■- = 0.20, embryo SL = 4.2 + 

 0.28.V) on female size for fall (triangles) and winter 

 (circles) collections of pink seaperch (Zalembius 

 rosaceus). Dashed line in A represents data from 

 Baltz (1984). 



seaperch, contrary to results of previous reports. Brood 

 size increased with female size in fall but not in winter 

 (Fig. 2A) in this study. The lack of a relationship observed 

 in winter was similar to previous conclusions (See Fig. 2 

 in Baltz, 1984). Baltz and others (Goldberg and Ticknor, 

 1977; Baltz and Knight, 1983) speculated that sample 

 effects may have explained the lack of a relationship 

 between brood size and female body size, but no clear 

 evidence was provided. In the present study, there was 

 a high incidence of females (18%) that had aborted all 

 embryos (i.e., complete abortion) among winter collections, 

 and these females were excluded from analyses. However, 

 partial abortions may occur in embiotocids (Schultz, 1993) 

 presenting the possibility of underestimating brood size. 

 Females with partially aborted broods were not identified 

 nor excluded from the present study. Therefore, if such 

 behavior occurred among large winter females, it would 

 provide a reasonable explanation for the lack of relation- 

 ship observed between brood size and female size. 



Winter females had larger brood sizes than their fall 

 counterparts and a higher frequency of abortion, pro- 

 viding evidence that winter females had partial rather 

 than complete broods. Individuals smaller than the 



