trials, and specimens yielding milt in the catheter 

 were selected. The selected males and females 

 were transferred to a spawning tank identical in 

 dimensions to the holding tanks but with water 

 depth reduced to 0.61 m. Two females and two 

 males were used in the first trial, and one female 

 and one male in the second. Their sizes ranged 

 from 47.0 to 52.9 cm fork length and 2.25 to 3.04 



kg. 



Hormone treatments administered were mod- 

 ified after those used by Leong (1977) to spawn 

 Pacific mackerel. Scomber japonicus. Combina- 

 tions of triturated, acetone-dried salmon pituitary 

 glands (SP), human chorionic gonadotropin 

 (HCG), and pregnant mare serum gonadotropin 

 (PMS) were injected into the dorsal musculature 

 with tuberculin syringes and 24-gage needles. The 

 hormone preparations were suspended or dis- 

 solved in physiological saline and administered in 

 injection volumes of 0.10-0.50 ml (Table 1). Each 

 treatment consisted of two series of injections 24 h 

 apart, starting at 2 p.m. in the first trial and 6 p.m. 

 in the second. Responses of females were moni- 

 tored by ovarian biopsies taken immediately pre- 

 ceding each series of injections, and spawning was 

 detected by the appearance of the buoyant, pelagic 

 ova in plankton nets (egg strainers) placed in the 

 outflow from the treatment tank. Responses of 

 males were based on subjective evaluation of hy- 

 dration of milt, on the basis of whether milt could 

 be expressed by moderate stripping pressure, and 

 the fluid consistency of the milt obtained. 



In the first trial, neither female showed any 

 significant increases in sizes of largest ova within 

 24 h after the first injections (Table 1). However, 

 ova were found in the strainers the next morning 

 at about 8:15, about 19 h after the second injections. 

 Biopsy of both females indicated that ovulation 

 had been induced only in the one treated with 

 higher dosages; similarly, examination of both 

 males indicated that milt could be expressed only 

 from the individual treated with the higher dos- 

 ages. Eggs and milt were stripped from the re- 

 sponsive pair to effect fertilization. Most of the ova 

 from this spawning were not viable and sank when 

 placed in seawater, and of those that remained 

 buoyant none progressed beyond early cleavage 

 stages. This poor viability was probably caused by 

 our failure to detect or respond to this induced 

 ovulation early enough, with a consequent de- 

 terioration of the ovulated eggs within the ovarian 

 lumen (Stevens 1966). 



In the second trial, biopsy of the female indi- 



cated that its largest ova had increased in mean 

 diameter from the original 0.56 to 0.70 mm within 

 24 h after the first injections. Ova were found in 

 the strainer at 5:30 the next morning, 12.5 h fol- 

 lowing the second injections. Milt could be easily 

 expressed from the male at this time, and the pair 

 were stripped to effect fertilization. At incubation 

 temperatures which ranged between 21° and 26° 

 C, the first cleavage divisions were observed about 

 1 h after fertilization and the first hatching at 

 about 31 h. With limited facilities for maintaining 

 the developing embryos, we were able to sustain 

 only a small fraction of the fertilized eggs ob- 

 tained. About 300 larvae were hatched and they 

 developed and absorbed their yolk sacs over 2 d. It 

 was not our intent or purpose at this stage of these 

 investigations to attempt to feed or rear these lar- 

 vae, and they all died by the end of the third day. 

 Previous attempts to produce young tuna in cap- 

 tivity have been concentrated in Japan, and along 

 two lines of effort. One is the fertilization at sea of 

 ova obtained by stripping freshly captured, 

 running-ripe fish (Yasutake et al. 1973; Harada 

 1978; Ueyanagi 1978). Since running-ripe females 

 are only infrequently taken in current commercial 

 fishing operations, these efforts have produced 

 only occasional successes. The other approach is to 

 maintain adult-sized specimens in large, netted 

 enclosvu-es in coastal waters with the hope that 

 they will eventually start spavming naturally. (By 

 coincidence, the first known spawning of a species 

 of tuna in such a situation was reported to have 

 occurred with bluefin tuna, Thunnus thynnus, on 

 20 June 1979 (Sogo 1979) at the same time that our 

 first induced spawning of E. affinis was taking 

 place.) Whether either of these two approaches can 

 be developed into reliable spawning operations for 

 tunas remains to be determined. The two consecu- 

 tive hormone-induced spawnings that we achieved 

 with E. affinis suggest that this technique can be 

 developed into a routine procedure, at least for 

 this species. Work is expected to continue at the 

 Honolulu Laboratory on refinement of effective 

 treatments and to be initiated on rearing of the 

 resultant larvae. 



Acknowledgments 



We thank Roderick Leong of the Southwest 

 Fisheries Center La Jolla Laboratory, National 

 Marine Fisheries Service, NOAA, for providing 

 salmon pituitaries and for his words of encour- 

 agement. 



186 



