Fig. 6. Phenotypic analysis of rump. Bar charts show distributions of embryonic cuticle phenotypes for the population of Drosophila embryos that complete embryonic development, binned by the number of abdominal segments produced. For each graph, embryos from different genotypes were collected in parallel under the same conditions and over the same time period. The statistical significance of differences observed between pairs of genotypes was determined using the ² test. (A) Embryos with eight complete segments (8 seg) were binned separately from those containing partial or entire segment deletions (<8, black) for each genotype. nos^BN heterozygotes (nos^-/+) do not display segmentation defects (n=210). The rump¹ distribution (n=332) differs significantly from nos^-/+ or wild-type (not shown) (P<0.001). A single copy of a genomic rump transgene, P[rump], rescues the rump defect (n=483, P<0.001 versus rump¹). Removing a single copy of nos from rump¹ embryos using a rump¹ nos^BN recombinant chromosome in trans to rump¹ (rump^- nos^-/rump^- +, n=867) increases the percentage of defective embryos beyond that of rump mutants alone (37% versus 11%, P<0.001). Similar results were obtained using the rump¹ nos^BN recombinant chromosome in combination with either of three deficiencies that delete rump (data not shown). (B) Embryos were binned with: eight complete segments (8 seg); seven complete segments or seven to eight segments with partial deletions (7 seg); six or fewer segments (0-6). tud^tux46 heterozygotes (tud^-/+) exhibit a low percentage of abdominal defects (n=367). When rump is eliminated (tud^-/+; rump^-/rump^-, n=266), the frequency and severity of abdominal defects are significantly increased (P<0.001).