142 
Psyche 
[Vol. 90 
Size and Mating 
Individual weevils were compatible with mating partners of many 
sizes. In 52 different pairings, females mated with males as much as 
10 mm shorter than themselves, and males with females as much as 
16 mm shorter. Despite this, mating was size-assortative overall. 
The Pearson product moment correlation for male and female 
length was r = .323 (p = .021) for the 52 different pairings, and r = 
.398 (p = .002) if multiple matings of a pair were included. 
Females, however, tended to reject males smaller than themselves 
when such males attempted to mate. In 57% of the cases of rejection 
(4 out of 7) the female was larger, whereas in only 37% of the cases 
of mating (22 out of 60), was the female larger. When lengths of 
males rejected and accepted for mating were examined, it was found 
that rejected males were shorter (p < .05, 1 -tailed, Mann-Whitney U 
test). 
Given the more frequent rejection of small males, and the greater 
success of larger males in aggressive encounters over females, it was 
expected that males found mating would tend to be larger than 
males simply present in the aggregation. Whereas females that 
mated were larger than unattended drilling females (p < .02, 2- 
tailed, Mann-Whitney U test), males that mated were not signifi- 
cantly larger than guarding males, males in a random sample, or 
males that were alone (Table 1). Instead, a frequency histogram of 
mating males showed a bimodality in the size of males that mated 
compared to an unimodal distribution of males in the breeding 
aggregation (Fig. 4). There appeared to be a dearth of medium-sized 
mating males. Indeed, a chi-square test on the 52 different pairings 
found that mating males were significantly more likely to be large 
(^ 31 mm) or small (^ 22 mm) than would be expected if they 
mated in proportion to their abundance in the random sample (x^ = 
4.87, 1 df, p = .027). 
Extra opportunities for small males to mate could arise if guard- 
ing males drove away small rivals less frequently than they did rivals 
more their size. With this in mind, I compared the 7 cases in which 
two males co-occurred at a drilling female for 3 min or more with 
the 19 cases in which one male drove off the other within the first 
minute. In 6 of the 7 cases of co-occurrence, one male was small 
(^ 22 mm) and the other large (^ 31 mm). In the remaining case 
both males were medium-sized. In the 7 cases of co-occurrence the 
