What does my OD600 reading actually mean in cells per millilitre?
Measure how dense a culture is
Reading optical density honestly, and converting it to cell density if you must.
OD600 is not absorbance. Cells scatter light rather than absorbing it at 600 nm, which has two consequences. The first is that the reading depends on your instrument’s geometry, so a benchtop spectrophotometer and a plate reader genuinely disagree about the same culture. The second is that any conversion to cells per millilitre is a calibration, not a constant: the widely quoted 8 × 10⁸ cells/mL per OD for E. coli is a nominal figure that moves with strain, growth phase and instrument.
The reading is only linear at low density. Above about 0.4 on a 1 cm path, multiple scattering sets in and the instrument under-reports — badly, and without complaining. Dilute into the linear range, read that, and multiply back. The check belongs on the number the machine saw, not on the corrected culture value: a culture at OD 2.4 read as 0.24 on a 1 in 10 dilution is perfectly well measured.
A plate reader looks down through the liquid, so its path length is the fill depth, not 1 cm. Two hundred microlitres in a 96-well well is roughly 0.6 cm, which means a plate reading of 0.3 is about 0.5 per centimetre — already at the edge.
If you need cells per millilitre to better than a factor of two, calibrate against plate counts on your own instrument with your own strain. Quoting someone else’s conversion factor to three significant figures is precision the method does not have.