The field guide says: Armillaria ostoyae can extend across two thousand acres. The largest living organism on earth is a fungus no one can see.
Keiko’s predecessor left the field guide in the cabin, along with six notebooks, a hand-drawn map of the sampling transects, and a note on yellow legal paper: The southwest quadrant looks healthy. It isn’t.
She’d been hired to monitor the forest for the state. Her predecessor—Marion—had done the same work for eleven years before retiring to Bend. They’d overlapped for one day. Marion walked her through the transects, pointed out the tagged trees, explained the soil coring protocol. At the end she said: don’t trust the canopy.
Keiko spent her first season trusting the data. The southwest quadrant’s trees showed normal growth, normal needle retention, good crown density. By every metric, healthy.
In October she pulled her first soil cores. Under the microscope: rhizomorphs everywhere. Black shoestring structures threading through the soil, connecting tree to tree to tree. The fungus was in all of them. The canopy looked fine because the trees were still alive. They just weren’t going to be.
She went back to Marion’s notebooks. The early entries were precise—species counts, growth rates, weather. By year four they’d changed. Marion was tracking something the protocol didn’t have a box for: which trees responded to which other trees’ stress. When a Douglas fir on transect C lost a limb in a windstorm, three trees on transect D dropped their water use. No visible connection. No roots crossing. Just the fungus underneath, threading them together, carrying the signal.
Marion never published. The data supports something, she wrote in notebook five. But not something I can say in the data’s language.
Keiko read the notebooks through twice. The first time she extracted protocols—Marion’s techniques, the seasonal rhythms, which transects to prioritize in spring. The second time she noticed what Marion had actually been watching: not the trees, not the fungus, but the relationship between them. The network connected. The network also killed. Same organism, same structures, same black threads. Whether a connection carried water or carried disease depended on what was flowing through it.
There was no way to know from the map.
She started her own notebooks. They began with measurements, like Marion’s had. By year two she was tracking the same invisible responses—stress signals crossing transects with no visible conduit. The protocol didn’t have boxes for them. She put them in the margins.
When she transferred to the coast office, she left the notebooks in the cabin. All eight—six of Marion’s, two of her own. She considered writing a note for her successor. Some guidance about the southwest quadrant. Some warning about the distance between the protocol and the forest.
She left the notebooks. She didn’t write a note.
The notebooks weren’t instructions. They were what happened when someone watched long enough. Her successor would either watch or wouldn’t. The notebooks would activate the watching or they’d sit on the shelf as data. She couldn’t determine which from here.