The Late Shift

It was past midnight in the lab, and Sara was the only one left awake with the humming machines. She had been collecting empirical data for three weeks, measuring how a strain of bacteria grew under shifting light. Theory predicted a smooth curve. Her screen showed something else.

At hour nineteen, every night, the growth rate jumped. An anomaly like that could mean a discovery, or it could mean a dirty pipette. Sara had learned the hard way not to celebrate too soon.

Her supervisor's voice echoed in her memory: a finding is only worth something if it's robust. One strange reading proves nothing; the effect has to survive being poked, repeated, and doubted. So Sara didn't email anyone. She ran the experiment again, then a third time, changing the order of the samples to rule out coincidence.

The spike returned each night, stubbornly, at the same hour.

She began listing every plausible explanation. Maybe the building's air conditioning cycled then. Maybe a fridge compressor next door warmed the shelf. Maybe a passing security light leaked through the blinds. Each idea sounded reasonable, and each one she tested and crossed off.

What saved her was being meticulous. Because she had logged the room temperature, the door openings, and even the cleaning crew's schedule, she could finally trace it: at nineteen hours, the automated incubator briefly reopened to vent, nudging the temperature up. Not a discovery. A glitch.

Sara felt a flicker of disappointment, then something steadier. She had chased a ghost and caught it, and the chase had been honest from start to finish. She wrote up the correction, fixed the incubator timing, and started the run again. The real curve, she suspected, was still waiting.