must remain tentative and inconclusive until 

 growth has been monitored directly in one or more 

 captive or free-ranging, tagged individuals. 

 Some alternatives that can be considered are: 



1. Two layers per year until the teeth are 

 occluded. 



2. Two layers in the first year and one per year 

 thereafter until the teeth are occluded. 



3. Two layers per year until puberty (about 

 nine layers in males and seven in females; 

 see section below on age at puberty), and one 

 per year thereafter. 



This list of alternatives can be extended to great 

 length by making assumptions such as that layers 

 are laid down at irregular intervals, males and 

 females lay down layers at different rates, layers 

 disappear with age, etc., but the above are proba- 

 bly the main possibilities that should be con- 

 sidered. All references below to age are in terms of 

 layers, with the above alternative possibilities 

 considered or implied. None of the alternatives 

 can be eliminated with certainty. One tooth layer 

 deposited per year has been inferred for the west- 

 ern Pacific population of S . attenuata by Kasuya et 

 al. (1974). One layer per year has also been sug- 

 gested for other closely related delphinids, includ- 

 ing S. coeruleoalba (Kasuya 1972) and Tursiops 

 truncatus (Sergeant et al. 1973). Two tooth layers 

 per year have been found in Delphinapterus leucas 

 (Sergeant 1973), but this form is less closely re- 

 lated to Stenella. Thus, there is more support in 

 the literature for the one-layer-per-year model 

 (number 2 above) than for the others. 



Length-Weight Relationships 



Length-weight relationships were determined 

 for 218 fetuses, 66 postnatal males, and 33 

 nonpregnant, postnatal females by using linear 

 regressions of log weight on log length. 



Fetuses 



The fetuses ranged from 20 to 897 mm long and 

 weighed from 2 to 7,588 g. Ten fetuses less than 20 

 mm long were not included. The regression equa- 

 tion is 



log W = 3.5532 + 2.501 logL, 



where W = weight in grams 



L = length in millimeters. 



242 



FISHERY BULLETIN: VOL. 74, NO. 2 



In exponential form, the relationship is 



W = 2.79 X 10-4 L2-501. 



Females 



The females ranged from 100 to 200 cm and 

 weighed from 12.0 to 69.1 kg. The regression equa- 

 tion is 



log W = -4.1576 + 2.6120 log L, 



where W = weight in kilograms 



L = length in centimeters, or in exponen- 

 tial form, W = 6.95 x 10"^ L^^'^. 



Males 



The males ranged from 86 to 218 cm and 

 weighed from 6.8 to 90.0 kg. The regression equa- 

 tion is 



log W = -4.7135 + 2.873 logL, 



where W = weight in kilograms 



L = length in centimeters, or in exponen- 

 tial form, W = 1.93 x lO'^ L^-^^s 



The slopes of the regression equations are 

 statistically different (^-test ata = 0.05) for males 

 and females. Males are lighter for their length at 

 birth, and heavier for their length after about 135 

 cm has been attained. 



Color Pattern 



Perrin (1970b) has previously described the de- 

 velopment of the color pattern of S. attenuata in 

 the offshore eastern Pacific. The animal begins 

 life unspotted, develops dark spots ventrally that 

 later coalesce, as light spots develop dorsally. The 

 ontogenetic continuum can be divided into five 

 stages as defined below and as shown in Figures 13 

 and 14: 



1. Newborn stage. Dark purplish-gray dorsal 

 surfaces and lateral blazes, with white ven- 

 tral surfaces and no spots; about 80 to 160 cm. 



2. Two-tone stage. General two-tone pattern 

 with dark-gray surfaces above, lighter gray 

 lower surfaces, and a well-defined pattern in 

 varying shades of gray about the head and 

 flippers; no spots; about 95 to 175 cm. 



