FISHERY BULLETIN: VOL. 74, NO. 1 



of the first holding experiment were taken in 

 order to examine the effects of captivity on sagit- 

 tal growth. Four measurements for the 5 April 

 sample were arbitrarily deleted because their 

 values were well below the distribution of the 

 majority of the sample. All 24 measurements from 

 the 9 May sample were utilized. There are signifi- 

 cant relationships between sagitta length and fish 

 length for the two samples (P <0.001, r^ values: 5 

 April, 0.82; 9 May, 0.70) (Figure 5). The first 

 experiment demonstrated that there was a signifi- 

 cant increase in the mean number of increments 

 between the two samples. Analysis of covariance 

 of sagittae lengths indicated that there were no 

 significant differences between the means of the 

 independent variables, regression coefficients, or 

 elevations of the regression curves for the two 

 samples (respective F ratios: 2.5, 1.0, 1.2) presum- 

 ably because of intrinsic variation, limited preci- 

 sion of measurements, and the relatively short 

 time period between samples. Although there 

 were no statistically significant differences found 

 in the comparison of the two curves, the two 

 regression coefficients exhibit perhaps expectable 

 trends. The lesser regression coefficient and r^ 

 value for the 9 May sample may be indicative of a 

 decreased growth rate and more variable re- 

 sponses of individuals in the population to the 

 highly variable, and probably less than optimal, 

 conditions of the holding facility. In addition, the 

 differences between the unadjusted and adjusted 

 means of sagittal lengths between the 5 April 

 (1.094 mm; 1.070 mm, respectively) and 9 May 

 (1.176 mm; 1.201 mm, respectively) samples of 

 0.082 mm and 0.131 mm are to be expected with 

 daily growth increments of about 3-4 /u.m. 



We have noted one apparent example of pro- 

 visioning rates affecting the growth rates of sagit- 

 tae of captive nehu. Sagittae from the 19 January 

 sample of the second holding experiment usually 

 exhibited 23-24 distinctive, more widely spaced 

 increments on the edge of the otolith. The num- 

 bers of distinctive increments approximately cor- 

 respond to the number of days during which the 

 daily amount of food provided the sample popula- 

 tion was double the initial ration. As might be ex- 

 pected otoliths collected 7 days later in the 26 

 January sample exhibited 30-31 distinctive incre- 

 ments. Indeed, the wider increments observed 

 after provisioning rates were doubled were much 

 more effective in "labeling" the sagitta than our 

 attempts to accomplish the same objective with 

 Tetracyclene. Possibly, controlled experiments 



MAY 9,1972 



APRIL 5,1972 



32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 

 STANDARD LENGTH (mm) 



Figure 5. — Stolephorus purpureas: Growth of sagittae during 

 first holding experiment. 



with rapidly growing fish species incorporating 

 this treatment would be a much more expeditious 

 test of the daily growth increment hypothesis. 



Age and Growth in Wild Populations 



We examined larval, juvenile, and adult nehu 

 collected in Kaneohe Bay to obtain an estimate of 

 age and growth of a wild population based on the 

 assumption that the smallest observable growth 

 layers in the sagittae represent daily growth in- 

 crements. We examined 213 specimens from 13 

 collections made during most seasons between 

 spring 1972 and summer 1973 (no collections were 

 made in the months November through January). 

 The growth curves obtained from the individual 

 collections are given in Figure 6. Because all 

 individuals in a sample have been exposed to the 

 vagaries of the environment during their ob- 

 served lifespan, a composite growth curve for all 

 collections is presented in Figure 6F. Although 

 some variation between samples is apparent, the 

 composite scattergram serves as a first estimate of 

 the growth pattern of nehu in Kaneohe Bay. 



There are two well-defined segments to the 

 composite growth curve (Figure 6F). Young lar- 

 vae exhibit exponential growth to a length of 

 about 15-17 mm. At about 20 mm the population 

 enters an almost linear growth phase to about 60 

 mm. The composite scattergram obscures another, 

 lesser inflection at about 20-30 mm exhibited by 

 the spring 1972 collections (Figure 6A). Yama- 

 shita (1951) has demonstrated that nehu have 

 completed larval metamorphosis at about 30 mm. 

 The major inflection at a length of about 17 mm 

 appears to reflect the fact that nehu begin to 

 exhibit exponential growth in body depth at this 



14 



