164 



Fishery Bulletin 89(1), 1991 



Males 



100 110 120 130 140 150 160 170 ISO 190 200 210 



Length (cm) 



Figure 4 



Serum testosterone concentration as a function of length for 

 31 male Dall's porpoises collected in 1982. Two-phase regres- 

 sion lines are shown. 



testosterone increased from undetectable levels to 

 relatively high levels. The best fitting two-phase regres- 

 sion model (Fig. 4) indicated that the transition point 

 occurred at 183.0cm. Hence, the LSM for males was 

 estimated as 183.0cm. The mean (± SE) testosterone 

 concentration was 1.00 ± 0.52ng/mL for males shorter 

 than the LSM, and 10.46 ± 1.20 ng/mL for males longer 

 than the LSM. 



Discussion 



Newby (1982), using the presence of ovarian scars and 

 the 50th percentile method (see DeMaster 1984). esti- 

 mated the LSM to be 170.5cm for females in this 

 western North Pacific population. The use of hormonal 

 data alone provided a similar, but slightly lower, LSM 

 estimate of 169.0 cm. However, based on the growth 

 curves for females of this population (fig. 31 in Newby 

 1982), this 1.5-cm difference translates into small dif- 

 ferences in ages. 



The estimate for LSM in males of 183.0cm is in very 

 close agreement with the previously estimated LSM 

 of 182.6cm based on the 50th percentile method using 

 testis-epididymal weight (Newby 1982). Direct com- 

 parison with previous studies are not possible due to 

 the lack of variance estimates for LSM by the methods 

 used. Nevertheless, the hormonal estimates of LSM 

 provided in this study agree quite well with previous, 

 non-hormonal methods. 



Radioimmunoassay is a quick, economical, and accu- 

 rate measure of reproductive condition. Small samples 

 of blood (< 1 mL) provide adequate serum for replicate 

 assays. Previous studies have shown the high predic- 

 tive value of serum progesterone level to corpus luteum 

 size and reproductive state (Temte and Spiel vogel 1985, 

 Kirby and Ridgway 1984, Sawyer-Steffan etal. 1983). 

 Moreover, such methods could allow the rapid collec- 

 tion of samples for maturity status assessment from 

 large numbers of animals in the field. 



Hormonal assessment of reproductive status may 

 prove to be a non-lethal technique to estimate popula- 

 tion parameters such as LSM. It is, at present, more 

 applicable to captive animal studies. However, situa- 

 tions in which animals are killed incidentally may pro- 

 vide opportunities for obtaining concordant blood and 

 reproductive tissue samples. Such sampling could allow 

 the direct comparison of hormonal and histological 

 methods of estimating maturity and establish appro- 

 priate hormonal criteria for future studies. 



Seasonality in breeding may well limit the usefulness 

 of this method in males (see Perrin and Donovan 1984) 

 because testosterone undergoes seasonal fluctuation. 

 In addition, the presence of environmental toxins may 

 interfere with steroid hormone production. For exam- 

 ple, Subramanian et al. (1987) have shown a significant 

 negative relationship between testosterone concentra- 

 tion and DDE residue level. 



There is potential for using progesterone, testos- 

 terone, and other hormonal parameters, such as chori- 

 onic gonadotropin, FSH, and LH, in the estimation of 

 sexual maturity and reproductive status in marine 

 mammal populations. As we expand our reproductive 

 database in the cetacea, we also need to correlate 

 histological and morphological states with hormonal 

 parameters. Research protocols which include 10-mL 

 samples of fresh blood and assessment of hormonal 

 state should be encouraged. 



Acknowledgments 



I thank L.L. Jones, L.M. Tsunoda, and personnel of 

 the National Marine Mammal Laboratory, NMFS, 

 Seattle, Washington for collection of blood samples and 

 exchange of information. F.L. Moore of Oregon State 

 University Department of Zoology provided laboratory 

 space for RIA. L.L. Jones and two anonymous re- 

 viewers made very helpful comments that greatly im- 

 proved this manuscript. This study was made possible, 

 in part, by support to the author from the Lutheran 

 Brotherhood Medical Research Fund. 



