Kai: the lack of access to FAN is definitely the weak link, and I would caution anyone against assuming the results are directly applicable to wort starters. I'm not sure what you mean about errors in the results though. The error in the grad cyl readings is pretty small relative to the values (0.9 mL), and while I don't want to read too much into a single dataset, the trend seems pretty compelling to me. If FAN was the (only) limiting factor I would have expected all the samples to produce the same amount of slurry.