Attributions
Every rule in the simulation traces to one of the papers below. Where the model simplifies or scales a published result, the entry says so. Citations are given in full. None of this work is mine, and none of the authors are affiliated with this project.
On the model
AGAR is an illustration, not an instrument. It reproduces the qualitative behaviour these papers describe — saturating growth, biased random walks, constant-increment division, replicative limits, diffusion-limited branching — at interactive speed and at a scale of a few thousand agents. Absolute numbers are not calibrated to any organism; the shapes of the curves are the claim.