On Friday I created a new calibration set to run more cobalt-only samples. In those and the samples I made today to run, I included an internal standard (Sr). The standard worked well in all the calibration solutions on Friday, but I must have made the samples less precisely this morning because the concentration of Sr was no where near the value it should have been, so the internal standard was dropped. The samples I ran today were a factor of 10 lower than the cobalt-only samples I ran more than a month ago. I am still unsure if this is a calculation mistake or a trend that actually exists in the data.