FISHERY BULLETIN: VOL. 74, NO. 3 



megalops and Scopeloberi/x opisthopterus, the 

 maximum size of females was only slightly greater 

 than that of males, but there were relatively few 

 males larger than the smallest mature female. 

 Thus there appears to be a slight but real differ- 

 ence in size of the two sexes. The two smallest 

 melamphaids, Mehmphaet^ danae and S. rohustus 

 showed no sexual differences in size. In all cases, 

 except P. crassiceps and P. osciiam^, there were 

 sufficient small females to indicate that the size 

 differences were not due to protandrous 

 hermaphroditism. 



Sexual differences in size have been reported for 

 many species of dioecious mesopelagic fishes. 

 Large differences comparable to that observed in 

 Bafhyleptus occur in the ceratioid anglerfishes 

 (Bertelsen 1951) and the stomiatoid Idiacauthus 

 (Gibbs 1964). Differences of the order observed for 

 some of the melamphaids are known for several 

 stomiatoids: Stomias (Gibbs 1969), Echiosfoma 

 (Krueger and Gibbs 1966), and Cyclothone 

 (Kobayashi 1973). The usual explanation of the 

 adaptive significance of smaller males (Marshall 

 1971) is that in a food limited environment-such 

 as the deep-sea probably is-the energy required 

 by the population is lowered without diminished 

 fecundity. 



Sexual dimorphism (as opposed to differences 

 only in size) is quite common among meso- and 

 bathypelagic fishes. Males of several groups ex- 

 hibit better developed swimming muscles or sen- 

 sory apparatus than the females (Marshall 1971). 

 In many myctophids and stomiatoids, there are 

 sexual differences in light organs. In most cases, 

 sexual dimorphism seems related to increasing 

 reproductive success by increasing the probability 

 of heterosexual encounter. 



No obvious external sexual dimorphism was 

 observed in any of the species considered here 

 (with the exception of Isisfius), but at least two 

 species appear to have uneven sex ratios (Table 2), 

 an adaptation which, like the dimorphisms noted 

 above, serves to increase the probability of a 

 female meeting a conspecific male. Actually, the 

 sex ratios of five species were significantly differ- 

 ent from 1:1, and Anoplogaster cornnia showed a 

 nearly significant trend. However, it seems wise to 

 view the estimates for Poromitra crassiceps, P. 

 oscita7is, and Diplospinus nudtistriatits with 

 suspicion since numbers were rather low and 

 biases due to inadequate sampling and avoidance 

 may be involved. 



For both Melamphaes danae and Scopelagadus 

 mizolepis, the numbers involved are relatively 

 high and there is no indication that the popula- 

 tions were not adequately sampled. The estimated 

 sex ratios for these two species indicate that the 

 probability of an individual female encountering a 

 male is about twice that expected for a population 

 with the same density of females and 1:1 sex ratio. 

 (The probability of an individual male encounter- 

 ing a female is lowered, but this has no con- 

 sequences to population reproductive success.) In 

 the case of M. da nae, where the sexes are the same 

 size, population fecundity would be less than that 

 of a population of equal total biomass and 1:1 sex 

 ratio because about two-thirds of the biomass are 

 males. The males of S. mizolepis are, however, 

 smaller than the females. Consequently, the effect 

 of uneven sex ratio on population fecundity is to 

 some extent balanced by the more nearly even 

 division of population biomass between males and 

 females. Better data on stages of matu- 

 rity-particularly for males-and size distri- 

 bution of mature fish of each sex would be 

 needed to quantitatively describe the "trade off" 

 between uneven sex ratio and sexual size 

 difference. 



The difference between M. danae and S. 

 mizolepis may simply be due to the fact that M. 

 danae is already a "dwarf" species-the smallest at 

 maturity of all mesopelagic fishes in our collec- 

 tions. There may be other factors which select 

 against the males being smaller than the already 

 tiny females. On the other hand, M. danae may in 

 some sense be less "food-limited" than S. m izolepis 

 and thus able as a population to afford having 

 two-thirds of the biomass as males. Further study 

 of the interplay between sexual dimorphism, 

 differences in size, and departure of sex ratio from 

 1:1 might prove to be a fruitful approach toward 

 understanding the diverse life history features 

 shown by mesopelagic fishes. 



ACKNOWLEDGMENTS 



We are indebted to the captain and crew of the 

 RV Teritu and to the many people who assisted in 

 collecting and rough-sorting the material. We also 

 thank A. W. Ebeling for confirming our iden- 

 tifications of representatives of each of the 

 melamphaid species considered here. This re- 

 search was supported by NSF GB23931, NSF 

 GA38423, and funds from the University of 

 Hawaii, Hawaii Institute of Marine Biology. 



644 



