Spermatogenesis is an interesting topic unto itself, but I wouldn't put much stock into it. That being said, you will find certain combinations of drone mothers and queen mothers probably out performing others but whose to say it's the effect of spermatogenesis or just simply better hybridity matching. You can look at mito types, haplotypes etc... but in the end you have to build your breeding/heterotic groups and map out the best combinations. Dean hit some great points on starting with stable lines. II would capitalize on that nicely as well, great tool for trait integration and creating know hybrids/crosses. I've spent the last two years putting together potential breeding pools, still have some work to do, but I'm excited about the ride it's about to take me.