Segregation distorters are selfish genetic elements that unfairly achieve biased transmission through the germline. Found across a wide variety of taxa including fungi, plants, insects and mammals, segregation distorters can rapidly increase in frequency in natural populations and trigger the evolution of suppressors across the genome. One of the best-studied male segregation distorters is the autosomal Segregation Distorter (SD) gene complex of Drosophila melanogaster. Males heterozygous for SD and a wild type chromosome transmit SD to over >95% of their progeny, whereas heterozygous females transmit SD fairly to 50% of their progeny. SD is found at frequencies of ~1-5% in natural populations across the globe. Cosmopolitan SD chromosomes involve a main driving locus on chromosome 2L—Sd-RanGAP—and several upward enhancers that strengthen drive on both 2L and 2R. SD chromosomes often tighten linkage among enhancers and Sd-RanGAP via the recruitment of chromosomal inversions that suppress recombination. Suppressed recombination prevents the distorter from recombining onto a sensitive target background and generating self-distorting “suicide” genotypes. African and European SD chromosomes appear molecularly different—they do not share inversions and we find that suppressors of European SD chromosomes do not suppress African SD chromosomes. While inversions provide short-term benefits to SD, the reduced recombination entails long-term costs due to the associated reduced efficacy of natural selection. To study the consequences of suppressed recombination, we performed whole-chromosome population genomics analyses of SD chromosomes sampled from African and European populations. We Illumina-sequenced 10 haploid embryos from Zambia and 10 adults from France. We find a dearth of nucleotide variation on French SD chromosomes that begins at Sd-RanGAP on chromosome 2L, spans the centromere and extends 4 Mb into chromosome 2R. In contrast, we find a more dramatic dearth of nucleotide variation on Zambian SD chromosomes that begins at Sd-RanGAP on chromosome 2L, spans the centromere and extends for ~22.5 Mb across chromosome 2R— a massive sweep signal that suggests very recent and strong selection. We therefore detect a signature of parallel sweeps of independent multi-locus selfish SD complexes in Africa and France. Taken together, differences in (i) the structure of the selective sweeps, (ii) the chromosomal inversions involved, and (iii) responses to genetic suppressors of SD, imply that Zambian and French SD gene complexes have functionally diverged from one another.