The sample cannot be identified
Updated
On this page7
How labels fail
Adhesive that let go in a cooler, ink that ran when the bottle sweated, handwriting that was fine in the field and illegible on a bench, a label applied over condensation that never stuck. All of them are wet-and-cold problems, which is exactly the environment samples live in.
Why it is unrecoverable
A sample without a confirmed identity cannot be reported against anything. Guessing from the order in the box is not identification, and for regulated work it ends the sample. Recollection is the only answer and sometimes the site or the condition is gone.
Pre-number everything
Label containers before they leave, with a durable ink on a durable label, and record in a field sheet which number went where. Then the field task is writing a number against a location rather than writing a full label in the rain.
Two independent marks beat one good one
A label and a mark directly on the container, or a label on the bottle and one on the cap, means one failure does not lose the sample. Caps get swapped, so a cap-only mark is the weaker of the two.
Check legibility at receipt, not at analysis
The moment to discover an unreadable label is when the cooler is opened, while the collector can still be asked. Finding it days later, when the holding time is half spent, loses the sample twice over.
The container shape matters more than it seems
A round bottle shows one third of a wrap label at a time and a small vial has almost no flat area. Choosing a container with room for a readable label is a real criterion alongside the material.