Fig. 2. Numbers of cell corpses in pag-3 mutants. The numbers of cell deaths in pag-3 mutants was assayed by counting cell corpses in the ventral cords of 60-65 early L3 ced-1; pag-3 hermaphrodites using Nomarski microscopy. The difference between the observed number of corpses in ced-1 mutants and the number of deaths that occur in the corresponding lineages of developing animals can be accounted for by an incomplete block in engulfment of cell corpses in the ced-1 mutants and by a role for engulfment in cell killing (Ellis et al., 1991; Reddien et al., 2001). In pag-3 mutants, no cell corpses were observed in regions of the ventral cords that lack cell deaths in wild-type animals. For the analysis of pag-3 alleles, we constructed strains heterozygous for the indicated pag-3 alleles and for a small chromosomal deficiency (Df) which deletes the pag-3 locus (see Materials and Methods). Using an unpaired Student’s t-test, we found the differences in mean corpse number±s.e.m. between pag-3(ls20) and pag-3(ls20)/Df to be significant (P<0.02 for both anterior and posterior ventral cord data), while those between pag-3(n3098) and pag-3(n3098)/Df were not (P<0.40 and P<0.77, anterior and posterior, respectively). Anterior ventral cord, cell corpses generated by W, P1 and P2; posterior ventral cord, corpses generated by P9-P12.
