Same as Fig. 3 except that the expansion factor plotted for each of the four curves has been corrected to account for the time delay expected in the spherical collapse model and given by equations (13) and (14). The density inside each radius exceeds 10 times the critical density to the right of the black dots. The correction is expected to work well for a spherical collapse above an overdensity of around 10 (see the main text). Even though the collapse is not perfectly spherical applying the correction does greatly improve the match between the four curves particularly to the right of the black dots. The spacing of the outputs is insufficient to resolve features around a= 0.018 where shell crossing becomes important in the lower two curves.