FECUNDITY AND SPAWNING FREQUENCY OF 

 THE HAWAIIAN ANCHOVY OR NEHU, ENCRASICHOLINA PURPUREA 



Thomas A. Clarke^ 



ABSTRACT 



Female nehu can begin spawning at 35 mm standard length; almost all fish over 40 mm SL from Kaneohe 

 Bay were mature and in spawning condition. Mature females were found in all months of the year. Females 

 from summer (May-October) had higher fecundity and relative cost per batch than fish from winter 

 (November- April). In nehu and most other anchovies, fecundity appears to increase exponentially with 

 weight. Nehu appear to be distinguished from other species by a higher exponent and consequently greater 

 increase in relative fecundity over the reproductive size range. Nehu spawn during a short period 1 or 

 2 hours after sunset and begin hydrating ova only a few hours before spawning. Data on presence or 

 absence of hydrated ova or postovulatory follicles along with differences in oocyte size in fish collected 

 from throughout the diel cycle indicated that, after spawning, nehu can ripen a new batch of oocytes 

 in 2 days and that most females spawn every other day. The estimated requirements for continued spawn- 

 ing at this rate indicate that individual variation in recent feeding success or stress could be responsible 

 for observed scatter about fecundity-weight relationships and deviation from the normal spawning 

 frequency. 



The nehu, Encrasicholina 'purpurea, is a small an- 

 chovy endemic to the Hawaiian Islands. It is one of 

 the dominant planktivorous fishes in enclosed, semi- 

 estuarine areas and is the major source of bait for 

 the local skipjack tuna fishery. Nehu are short-lived; 

 growth increments on otoliths indicate a maximum 

 age of about 6 mo (Struhsaker and Uchiyama 1976). 

 Leary et al. (1975) showed that nehu can reach 

 maturity at 35 mm standard length (SL) and pre- 

 sented fecundity data for 41 females. Leary et al. 

 found very few females with hydrated ova and, on 

 that basis, suggested that nehu spawn only once per 

 lifetime. 



Reexamination of Leary et al.'s (1975) conclusions 

 was prompted both by the great variability in their 

 fecundity vs. weight relationship and by discovery 

 in recent collections that female nehu with hydrated 

 ova are not at all rare, but rather are found only 

 at restricted times of the day. This paper presents 

 results of more detailed investigations of fecundity 

 and spawning frequency in nehu in order to com- 

 pare and contrast aspects of reproductive output of 

 a tropical anchovy with those of better studied tem- 

 perate species. 



MATERIALS AND METHODS 



All nehu examined for this study were collected 



'Department of Oceanography and Hawaii Institute of Marine 

 Biology, University of Hawaii, Honolulu, HI 96822. 



from Kaneohe Bay, HI. Day samples were collected 

 by beach seine or dip net in shallow water (1-2 m 

 deep) or were taken from bait recently collected 

 from similar areas by skipjack tuna vessels. Night 

 samples were taken by blind sets with a ca. 67 m 

 long by 13 m deep purse seine over deeper (12-14 

 m ) areas of the bay. Forty-four night samples and 

 two day samples were taken in 1974-79, while 5 

 night samples and 18 day samples were taken in 

 1983-85. Samples with adult nehu were available 

 from all months of the annual cycle and, for most 

 months, from at least two different years. One or 

 more samples with adults were available from all 

 hours of the diel cycle except the period between 

 midnight and dawn, when there were few samples 

 and very few adults collected. 



In order to follow short-term oocyte development 

 in the same group of fish, on two occasions a school 

 of nehu was surrounded with a 60 m long beach 

 seine in shallow water and sampled initially and 

 twice later in the day. Samples were taken at the 

 hours of 1300, 1500, and 1700 on 13 January 1984 

 and at 1000, 1300, and 1600 on 27 January 1984. 

 Although the school was obviously disrupted by ini- 

 tial surrounding and subsequent dipnetting of sam- 

 ples, the fish held in the net appeared to resume nor- 

 mal daytime behavior shortly after each disturbance 

 and spent most of the time loosely schooled with 

 other nehu on the outside of the net. The oocyte size- 

 frequency data from these samples did not differ in 

 any obvious manner from data taken from other 



Manuscript accepted November 1986. 

 FISHERY BULLETIN: VOL. 85, NO. 1, 1987. 



127 



