In British sparrowhawks, organo-chlorine levels and shell thickness indices varied little between eggs in the same clutch; but they varied greatly among clutches from the same (known) female in different years, and among clutches from the same territory in different years (female mostly unknown). The average tendency was for OC levels in eggs to increase between first year and older females. Within populations, OC's showed a skewed distribution, with a small proportion of clutches containing very high concentrations. (2) In contrast, shell indices were distributed 'normally' in the statistical sense. Among 315 clutches, the overall ratio of DDE:PCB was 1:0-53, and of DDE:HEOD it was 1:0-19. Comparing clutches, the levels of all these compounds were significantly correlated with one another. DDE concentration was significantly related to the extent of shell thinning among clutches, the extent of egg breakage, the extent of egg addling and the extent of hatching failure. PCB concentration was significantly related to the extent of egg addling and hatching failure. (3) No significant relationship could be shown between the levels of any OC examined and the date of first egg, the initial clutch size, the incidence of egg desertion and the extent of nestling mortality. (4) No relationship could be shown between HEOD concentration and any aspect of breeding examined, but the concentrations of HEOD were relatively low. (5) In only one of nine areas examined was there evidence for a decline in the levels of all three OCs during 1971-4, but in the overall sample there was evidence for a decline in PCB levels. (6) The regression of shell index on log DDE content in the sparrowhawk was similar to those found by other workers for Falco peregrinus, F. mexicanus and Pelecanus occidentalis.