Today I ran the remaining three trials of the 5mM series. These results confirmed that there is something wrong with the second and fourth samples, but other than that they were very consistent. The results were also strangely linear, which we saw some of in our results last summer.
