Much ado about nothing?

So I've plotted the rounds vs. examples per round chart as total examples vs. examples per round, and I think that I can safely claim that the odd behavior of the rounds chart is an artifact of the methods consuming more total examples as the number of examples per round increases. The graphs below show that the difference in total examples between the seeding methods is nearly nonexistent, and that the graphs very closely resemble the baseline graph. So, this may indeed have been much ado about nothing.