If you determined the number of cells in a sample using viable plate counts and optical density measurements, why would the results from the two methods be so different?