Killer whales live in stable social groups, called ''pods.'' It has be
en suggested that the structure of such groups may influence the vital
rates, and hence the fitness, of their members. To test this suggesti
on we used data from a long-term study of killer whales in the Pacific
Northwest (Bigg et al. 1990). We constructed stage-classified matrix
population models for the entire population, two sub-populations, and
individuals pods. The population growth rate for the entire population
is lambda = 1.0254, with 90% bootstrap confidence interval from 1.017
8 to 1.0322. The mean female population stage distribution is not sign
ificantly different from the predicted stable stage distribution. Popu
lation growth rate is most sensitive to changes in adult and juvenile
survival, followed by fertility. Factors that cause even small changes
in survival will thus have a large impact on population growth. Pod-s
pecific growth rates range from lambda = 0.9949 to lambda = 1.0498. Mo
st of the interpod variance in growth rate is due to variance in adult
reproductive output. Randomization tests show that this variance is n
ot significantly greater than expected on the basis of variation in in
dividual life histories within the population. We conclude that there
is no evidence for an effect of social structure on pod-specific popul
ation growth rate. The restriction of population growth rates to such
a narrow range suggests, but does not prove, a possible role for densi
ty-dependent processes.