The smell leaderboard that started in sixth place

Dark navy slide with a faint scent-wave motif, masthead “The Nose Knows · Mammalian Olfaction League Table · 2026 edition”, headlined “Humans are elite smellers”. A subhead reads “Functional olfactory receptor genes — the working hardware of smell, counted straight from whole-genome sequence. Zero-based axis, no tricks.” Three vertical columns sit on medal plinths against a zero-based y-axis running to 450: a gold bar for ★ #1 HUMAN (Homo sapiens) at 387, and blue bars for #2 MACAQUE (Macaca mulatta) at 309 and #3 PLATYPUS (Ornithorhynchus anatinus) at 265. A gold note above the human column reads “25% more than our nearest rival”. Three stat cards down the right side read 387 functional smell receptor genes in the human genome, 1.25× the macaque’s count — our nearest sequenced rival, and #1 in the primate & monotreme division. A gold kicker box adds “Whatever you were told about the dog, the genome is clear: the human nose ships with more working smell receptors than anything else in the comparison.” A footnote credits Niimura & Nei, PLOS ONE 2(8): e708 (2007), Table 1. The axis is honest, the numbers are accurate, and only three of the table’s eight species appear. — Showing the misleading chart

The same data on a cream background, kicked “The same Table 1 · all eight mammals in it” and headlined “Every mammal the league table left out beats us”. A horizontal bar chart on a zero-based axis to 1,300 ranks all eight species in the source table by functional olfactory receptor genes, amber for the five left off the first chart and blue for the three shown: rat 1,207, opossum 1,188, mouse 1,035, cow 970, dog 811, then human 387 in bold, macaque 309 and platypus 265. An amber note beside the rat reads “3.1× the human 387, from the same table”; a blue note beside the human bar reads “#6 of 8 — the league table stopped counting at #3”. A card headed “The lineup was the argument” explains that the league table shows three of the table’s eight mammals and not three at random — the macaque and the platypus are the only two species in the whole table with fewer functional OR genes than humans, so the comparison set is exactly the set humans win. A second card, “And the count isn’t the score”, notes that 415 of the 802 human OR genes are pseudogenes, 51.7%, the highest broken fraction of any species listed, but that a gene count is a parts inventory rather than a smell test — McGann (Science, 2017) points out humans out-detect dogs and rodents on some odorants — so the lesson is that a ranking is only as honest as its guest list. A footnote credits Table 1 of Niimura & Nei, PLOS ONE 2(8): e708 (2007), notes the counts come from 2004–2006 draft genomes and are called minimum estimates by the authors, and says both charts are our own demonstration. — Showing the misleading chart

Be the first to star this exhibit

A species league table crowns humans the top mammalian smeller — 387 functional olfactory receptor genes, ahead of the macaque’s 309 and the platypus’s 265. Every number is right, and all three come from one published table. That table lists eight mammals, and the two humans beat are the only two in it that score lower. Put the other five back and the rat leads with 1,207, 3.1× the human count, and the champion is sixth.

Y. Niimura & M. Nei, “Extensive gains and losses of olfactory receptor genes in mammalian evolution”, PLOS ONE 2(8): e708 (2007), Table 1 — functional olfactory receptor genes in eight mammalian species, 2007Added August 6, 2026

01The claim

The human nose is elite hardware. Counted straight from whole-genome sequence, humans carry 387 functional olfactory receptor genes — 25% more than the macaque’s 309 and nearly half again the platypus’s 265 — which makes us the top-ranked mammalian smeller in the comparison, whatever you have heard about dogs.

02The trick

Look for the axis crime and you will not find one. The bars start at zero, the scale is linear, the labels match the bars, and all three numbers are exactly right: 387, 309 and 265 functional olfactory receptor genes, straight out of Table 1 of Niimura and Nei’s 2007 genome survey. The whole argument is the guest list. That table has eight mammals in it, not three, and the two that appear alongside humans are the only two in it that score lower — the macaque at 309 and the platypus at 265. Every other species in the same column beats humans, and not narrowly: dog 811, cow 970, mouse 1,035, opossum 1,188, rat 1,207. Sorted honestly, the “#1 mammalian smeller” is sixth of eight, below the average of the five omitted species by a factor of 2.7. The tell is the qualifier on the third stat card. “#1 in the primate & monotreme division” sounds like a category until you notice it has exactly three members and was drawn around a result — humans, the one primate we beat, and the one monotreme we beat. A comparison set chosen after the numbers were in is not a finding, it is a sentence rearranged into a chart. (This exhibit is our own demonstration, drawn in the house style of a league-table slide from the paper’s published table — no such ranking was published.)

03The fix

Put the missing five back and the chart answers the question its own headline asked. Ranked against all eight mammals in the table it came from, human sits sixth at 387, the rat leads at 1,207 — 3.1× the human count, out of the same column of the same table — and the gap between the crowned champion and the field is the entire story the first chart was built to hide. The same table has one more thing to say if you read across instead of down: 415 of the 802 human OR genes are pseudogenes, 51.7%, the highest broken fraction of any species listed. And the leaderboard moved again after this table was published. In 2014 Niimura, Matsui and Touhara annotated the African elephant genome and found 1,948 intact OR genes, the largest repertoire ever identified in a single species — a competitor that was not in the 2007 lineup because its genome had not been read yet. Two honest caveats belong on the redraw rather than in the punchline. These are 2004–2006 draft assemblies and the authors call the counts minimum estimates; and a gene count is a parts inventory, not a smell test — McGann’s 2017 Science review notes humans out-detect dogs and rodents on some odorants. So the lesson is not that humans smell badly. It is the question to ask of any ranking that puts someone first: who else was in the table, and who decided which rows to draw?