The bulb map
Each receptor type converges on its own few glomeruli (observed: about 2 per bulb in mouse, about 16 in human), so the capacity 2N = 128 is carried into a spatial odour map. Sorting the organisers by spinodal(γ) gives EBF1 → EMX2 → LHX2, a code output read as interpretation. Glomerular targeting is a named [O].
Sorted by spinodal(γ), the 3 OSN organisers read EBF1 → EMX2 → LHX2 (code output; reading it as developmental order is interpretation — building is [O], dna §RB); receptor-type convergence onto glomeruli (observed: ≈2 per bulb in mouse, ≈16 in human) preserves capacity 2N = 128 as a spatial map. The targeting coordinate is a named [O].
The organisers sorted by measured γ (interpretation, not developmental order)
Smell's substitute for a spatial image is the bulb map: axons of the same receptor converge onto specific glomeruli, turning the receptor-space code into a spatial odour map. The organisers that lay it down (the OSN-identity transcription factors) are sorted here by spinodal(γ).
Reading note. The order below is a code output — the organisers sorted by their measured γ — and reading it as developmental order, or dwell ∝ γ1.5 as relative size, is interpretation. In the corpus, developmental order comes from regulatory-cascade depth in the DNA atlas, and building (order, size, timing) is open [O] (dna §RB).
| organiser | order | γ | h* = spinodal | dwell ∝ γ1.5 | role |
|---|---|---|---|---|---|
| EBF1 | 1 | 1.4097 | 0.64423 | 1.52159 | OSN differentiation (early B-cell factor) |
| EMX2 | 2 | 1.4574 | 0.67720 | 1.59947 | OR expression regulation |
| LHX2 | 3 | 1.5172 | 0.71930 | 1.69891 | OSN differentiation / OR gene choice |
The γ-sorted order is EBF1 → EMX2 → LHX2 — monotone in measured γ by construction of the sort (code output). In the model each Organ is absent below 0.9·h* and present above 1.1·h*: an intact downstream pathway with the master switch off still yields absence, a real organ rather than parts.
Convergence preserves the capacity 2N
Observation: the axons of sensory neurons expressing one receptor type converge on a small set of glomeruli — about 2 per bulb in mouse (one medial, one lateral; Mombaerts et al. 1996, Cell 87:675) and about 16 in human (~5,500 glomeruli for ~350–400 intact OR genes; Maresh et al. 2008, PLoS ONE 3:e2640). The map is one receptor type → several glomeruli, not a bijection. The capacity conclusion is unchanged: copies of one channel carry the same ON/OFF bit, so distinct subsets of the 7 receptor channels still light distinct sets of glomeruli, and 2N = 128 distinct spatial images remain.
Because this map preserves subset inclusion, the receptor-space nested thermometer maps to a nested spatial thermometer — 4 patterns at the 8 drive values the script samples (code output; a dense sweep of the same code gives N+1 = 8), glomeruli lighting in the same order. The map inherits the code's order and capacity from the frozen switch and measured γ.
The honest [O] — the targeting coordinate is not in γ
The substrate fixes the map's order and capacity, but not which physical glomerulus a given receptor targets. That coordinate is axon-guidance chemistry — the receptor's own signalling shaping a Neuropilin-1/Sema3A gradient — in its coding sequence, not its promoter γ. This is the same kind of molecular [O] as the odorant key: structure is substrate-derived, specificity is not.