There are several distance algorithms that are implemented in the representation of similarities and genetic distances among species, subspecies, individuals or populations (e.g: Kimura 1980, Jukes and Cantor 1969, Nei 1978 and 1983, Jaccard 1908, Cockerham 1984, Excoffier 1992, Cavalli-Sforza...etc.). It depends also on the markers you are using: AFLP, microsatellites, mtDNA, RADP or SNPs. Mega software seems OK for comparison. You can always try Arlequin, Popgene, Genetix, Phylip, FSTAT, GDA...
OK, what is the case? You have different markers obtained from different sources, different markers from same source but with different methods, differente markers from different sources using different mehotds?
Comparing species needs a phylogenetic approach, genetic distances like Nei-D usually compares popultions of the same species, since different species not neessarily amplify the same markers. Particlarly nuclear (neutral) markers could differ substantially, i guess some mitochondrial or chloroplastidial genes (highly conservated) are the best option, sequencing this genes and clustering individual gene-sequences could solve this...