POWELL. JAMES, and HURD: MATING ABILITY OF MALE KING CRAB 



Females were separated from the males which 

 held them and placed in storage ready for use 

 in the experiment. The average length of females 

 increased as the study progressed because older 

 females mate later in the season. ^ Females cap- 

 tured 15 March through 15 April averaged 128 

 mm. while those captured 20 April to 20 May 

 averaged 140. 



Twenty-four females were paired randomly 

 with the 24 males in the pens at the beginning 

 of each of the 14 separate 5-day periods begin- 

 ning 16 March and continuing through 24 May. 

 On days when a complete set of 24 females could 

 not be captured, those which had been were in- 

 troduced to males which previously had had the 

 largest number of female partners. Incomplete 

 sets of females exist for early and late spawning 

 periods — 1, 2, 3, 12, 13, and 14 — when mating 

 crabs were relatively scarce. In addition, 12 fe- 

 males escaped from six compartments on 9 May 

 when two pens were accidentally lifted; as many 

 as three females were lost from a single com- 

 partment. During the middle of the mating sea- 

 son, 31 March through 5 May, females were 

 abundant; consequently each male received a 

 female for each of the eight consecutive periods 

 within this interval. Females were introduced 

 to males as soon after capture as possible; none 

 were held more than two days and many were 

 introduced the same day. 



Females were left with males until mating 

 was completed and eggs were known to be fer- 

 tilized; consequently males commonly shared 

 their compartments with several females at a 

 time. Approximately 10 days after eggs were 

 deposited on the pleopods, a sample of approxi- 

 mately 1,000 was obtained from each female to 

 determine success of fertilization. Each sample 

 consisted of several separate groups of eggs 

 taken from scattered locations within the egg 

 mass. Samples were preserved in Bouin's solu- 

 tion. Fertilized eggs developing for 7 to 10 days 

 at approximately 37 °F showed cleavage when 

 viewed microscopically. When eggs were known 

 to be fertilized (i.e., advanced cleavage stages 

 observed), females were separated from their 

 experimental male partners and placed in stor- 

 age compartments 25-28. 



Divers made observations at daily intervals 

 recording data on underwater slates. Observa- 

 tions included collecting and measuring shed 

 exoskeletons to determine day of molting, exam- 

 ining recently-molted females to determine rela- 

 tive fullness of the brood chamber, and recording 

 activities of mating and feeding. 



RESULTS 



Relative success of male copulation was 

 measured in two ways. First, relative fullness 

 of brood chamber was determined subjectively 

 by visually deciding what proportion of the 

 brood chamber was filled with eggs and record- 

 ing same on a scale of zero to one hundred (Table 

 2). The second, percent of eggs fertilized, was 

 determined in two steps; (1) a prompt micro- 

 scopic viewing of several hundred eggs to obtain 

 quick estimates, followed later by (2) a careful 

 microscopic examination of 100 eggs (Table 3). 



Both measures of mating success provided 

 compatible results and revealed that infertile 

 eggs are scattered throughout the egg mass 

 rather than being grouped separately from fer- 

 tile eggs. 



The raw data, ij, for each of the two measure- 

 ment variables were transformed so that they 

 would be more normally distributed, using the 

 formula:'' 



z = y/n + 1/2 sin 



-1/v + 3/8* 



n + 3/4 



In this case, n is equal to 100, since both 

 methods of measurement are based on a scale of 

 100. 



A covariance analysis was performed relating 

 z to X. the number of females mated. This 

 analysis fits a least squares regression line z — 

 0/ -I- b,- X for each of the four groups, where a,-, 

 bj represent the intercept and slope respectively 

 for the ith. group. The results of this analysis are 

 presented in Table 4. Slopes for each group 

 appear to be significantly different from zero, 

 except for percent of eggs fertilized in large old- 

 shell and small new-shell males. The more nega- 

 tive the slope of the regression line, the less the 



'" Determined from 3,402 observations of grasping pairs 

 of king crabs captured over a 9-year period, 1963-1971. 

 See page 257 National Geographic Magazine. Vol. 139, 

 No. 2, Feb. 1971 for photograph of grasping pairs. 



•> Thoni, H. Transformation of variables used in the 

 analysis of experimental and observational data, a review. 

 Technical Report No. 7. Statistical Laboratory, Iowa State 

 University. Ames. Iowa, July 1967. 



173 



