Would edit but HC doesn't allow after a certain period....

Would edit but HC doesn't allow after a certain period. Essentially every reading in a lab has an uncertainty associated with it. If the duplicate (which should always be freshly sampled and not just one sample split into two) is significantly different from the original then it's possible their was an error made in the lab, an error in sampling all the sample itself wasn't reproducible,
O'm sure Blythefan can give a better answer, remembering I am speaking from a forensic/pharmaceutical lab background but the principles would be similar I'm assuming.