the same manner as whole otoliths. If two read- 

 ings by a single observer did not agree, otoliths 

 were deleted from analyses. Additional verifica- 

 tion of counts was then obtained from a second 

 observer who read 200 otoliths (359r ) selected 

 randomly. 



Because monthly samples did not contain simi- 

 lar porportions of large and small fish, calcula- 

 tions of monthly mean marginal increments of 

 sagittae were biased. For instance, if primarily 

 small fish are sampled in one month followed by 

 mostly large fish in the next month, the mean 

 marginal increment will decrease regardless of 

 the time of year. To alleviate this bias, marginal 

 increments were standardized by converting each 

 measurement to a proportion of the maximum 

 recorded for that age group. Thus, a measurement 

 of 2.5 ocular units in an age group for which the 

 maximum is 10.0 becomes equivalent to a mea- 

 surement of 0.5 ocular units in an older group for 

 which the maximum is 2.0. 



Sex and reproductive conditions were deter- 

 mined from histological sections of gonads, which 

 were preserved in lO'yf formalin, and later pro- 

 cessed through an Autotechnicon Duo Model 2A 

 automatic tissue processor, then embedded in 

 paraffin, and sectioned with a rotary microtome 

 at approximately 7 |xm (Humason 1972). Tissues 

 were then stained with Harris' hematoxylin and 

 counterstained with eosin-Y. Sexes were identi- 

 fied as male, female, and hermaphroditic female 

 or "transitional" (gonad primarily ovarian, with 

 some traces of active testicular tissue present). 



Maturity was described following the synopses 

 listed in Waltz et al. (1982). Terminology used in 

 histological descriptions follow Hyder (1969), 

 Combs (1969), and Wallace and Selman (1981). 



Results 



A total of 1,039 gag ranging in total length (TL) 

 from 153 to 1,150 mm was examined for life his- 

 tory information. Of the 652 otoliths on which age 

 determinations were attempted, 87'/ showed dis- 

 cernible rings verified by two readings. No 

 otoliths were deleted from analyses because of 

 disagreement between primary and secondary 

 readers. Marginal increment measurements from 

 the outer edge of the last opaque band to the dor- 

 sal margin of whole sagittae indicate that these 

 bands are laid down in late spring to midsummer. 

 Bands are apparently laid down earlier and over 

 a longer time period (May-August) in ages <VIII 

 than in older gag. Although sample sizes of fish 

 >age IX on which marginal increments were 

 measured are small in relation to those of 

 younger gag, it appears that ring formation is 

 concentrated in August (Fig. 1). Twenty-two age 

 groups were identified (Table 1). 



The gag ovary is a hollow, bilobed organ sus- 

 pended in the posterior region of the body cavity 

 from the swimbladder by mesenteries. Blood ves- 

 sels and nerves enter the gonad at the anterior 

 point of each lobe's suspension and course medi- 

 ally to the mesenteries along the dorsomedial sur- 

 face of each lob. The lobes fuse posteriorly, their 



N^ 



D LU 



CC DC 



< O 

 Q Z 



z — 



< _J 

 I- < 



CO z 



z o 



< DC 

 m < 



.70 -I 



.50- 



.30- 



.10- 



o < age VIII 

 • > age IX 



6 



3 



>\27 66 



5 7 



MONTH 



~l 1 



11 



Figure l. — Mean standardized marginal increments by month for gag sage VIII and 

 ^age IX, and sizes of monthly samples. 



649 



