CAMP ANA: DAILY GROWTH INCREMENTS IN OTOLITHS OF PLAINFIN MIDSHIPMAN 



jum to 0.3 jixm) until the growth increments in the 

 region of maximal growth were most visible. I defined 

 a growth increment as a bipartite structure, consist- 

 ing of a narrow opaque band and an adjacent broad 

 translucent region. Growth increments between the 

 otolith periphery and the hatch check were counted 

 at least twice through a compound microscope at a 

 magnification of 400X. Duplicate counts of an otolith 

 never differed by more than 10 f A. The use of a hand 

 counter eliminated the possibility of a count converg- 

 ing on an expected value. There was little doubt con- 

 cerning the nature of the hatch check; its radius 

 matched that of radii of otoliths removed at hatch. 

 Growth increments in 14L:10D/CT fish sampled 10 

 September were counted as above. However, a 

 second series of counts was made from the hatch 

 check to the prominent 10 August check; the second 

 data set served as a substitute for the actual sampling 

 of 14L:10D/CT fish on 10 August. 



Increment counts were made from both the left- and 

 right-hand side sagittae. Since the two sides did not 

 differ systematically under any of the environments 

 (paired /-test, P > 0.05), the means were used in all 

 data analyses. 



Increment widths were measured from photo- 

 graphs with a micrometer. Expected increment 

 widths were calculated from radial measurements 

 (central nucleus to rostral tip) of otoliths from all 

 environments and a variety of sampling dates (N = 

 10 per date). Values for mean increase in radial 

 otolith growth per day were then compared to ob- 

 served values. 



Since individual otoliths often displayed erratic but 

 discernable width trends through time, a measure of 

 the similarity of the widths of two adjacent daily 

 increments was calculated: 



IR, 



w,- w;._, 



(W, + W^/2 



where IR, is the index of increment width regularity 

 for day/, and W, is the increment width for day (.Such 

 an index gives low values when adjacent increments 

 are similar in width, despite any trends in the data. 

 Index values were calculated for a range of ages in 

 otoliths from a given environment. 



RESULTS 



Porichthys larvae and juveniles survived and grew 

 under all laboratory environments. Survival ex- 

 ceeded 95 '/< after hatch. Fish sampled about 1 mo 

 after hatch (10 August) did not differ significantly in 

 standard length (ANOVA, P > 0.05). By the end of 



the study, only those fish maintained in the 24L/ 

 1 4T, : 1 0T 2 environment were significantly smaller in 

 length (Scheffe's test P < 0.01); the difference was 

 apparently due to unintentional overcrowding from 

 the date of transfer. 



Hatching was initiated simultaneously in two of the 

 three initial environments, but started 4 d later in the 

 24L/CT aquaria. The delay did not appear to be due 

 to the artificial environment, since embryo develop- 

 ment among the 24L/CT egg masses lagged behind 

 that of the others at the time of collection. In the 

 aquarium, about 959? of the viable ova hatched 

 within 4 d of hatch initiation. Intratank hatch-date 

 variance would be expected to affect the variance of 

 increment counts. However, the 17-d range of larval 

 release dates (from the rock) was not reflected in the 

 otolith microstructure. 



Unground sagittae derived from both pre- and 

 posthatch fish were extremely lobulated in structure. 

 The origin of the numerous lobes was 5-10 

 "peripheral" nuclei, from which the majority of the 

 growth increments emanated. A central nucleus also 

 had growth increments associated with it, although 

 these were incorporated into the peripheral incre- 

 ments within 10-20 d/increments. A prominent hatch 

 check occurred within 5-10 major increments of the 

 central nucleus. The most prominent check of the 

 older otoliths was that associated with the sub- 

 division/transfer date of 10 August. 



Many growth increments were visible in the 

 polished otoliths sampled after hatch. When plotted 

 as a function of time, total increment counts were 

 significantly greater than those expected of daily pro- 

 duction (P < 0.05) (Fig. 2). Diel light and tempera- 

 ture cycles both produced an increment: age slope of 

 about 3.0, suggesting that numerous subdaily 

 increments were being counted with any daily 

 increments present. Increment clarity, prominence, 

 and width varied substantially within an otolith. 

 However, most increments could be assigned to one 

 of two "levels" — visually prominent/relatively wide 

 and visually faint/relatively narrow. To determine if 

 the first level consisted primarily of daily increments, 

 the expected width of a daily increment was 

 calculated. 



23 July 30 July 9 Aug. 10 Sept. 

 Mean otolith 



radius (jam): 270 430 620 875 



Daily increments on the order of 12-23 and 5-8 fim 

 wide would be expected in the first and second month 

 posthatch, respectively. These expected increment 

 widths were similar to those observed in the first 

 "level" of growth increments. 



167 



