Glomeruli are the small knots in the olfactory bulb where nerve cells carrying the same type of receptor converge. The brain reads a smell from the combination of glomeruli switched on, as What is the sense of smell explains.
Ask how many glomeruli there are in a human olfactory bulb, and most popular sources will say about 1,100 to 1,200. The figure appears in online encyclopaedias, in lecture notes, and in the first draft of the review on odour and memory on this very site. It was corrected only when another volume of the same research traced it back to the original counts.
The real counts
Following the chain of citations back, we have not found any count on human tissue behind the figure of 1,100. The counts that do exist give quite different results:
- In 2008 a research group counted glomeruli in the olfactory bulbs of 7 people aged 39 to 89, using immunostaining and statistical correction. The mean was 5,568 glomeruli per bulb, ranging from 2,975 to 9,325.
- In 1998 another count, using Nissl staining, gave about 8,000 glomeruli in young adults, falling by about 10% per decade to under 30% by the age of 90 to 100.
The two figures agree once the age of the donors is taken into account. The conclusion you can use: each human olfactory bulb has roughly 5,000 to 8,000 glomeruli, varying a great deal between people and declining with age.
A fivefold error is not a detail
The figure is not just something to memorise. In mice, each receptor type converges on about two glomeruli per bulb, and the principle of “one receptor type, two glomeruli” is taught as a general rule of smell. If humans had 1,100 glomeruli for about 350 to 400 receptor types, the rule would roughly hold. With about 5,500 glomeruli, each receptor type in humans corresponds to about 16 glomeruli.
So the way the human olfactory system organises its signals differs from the mouse in a basic respect, and the reason is still unknown. A wrong figure hides exactly the question most worth asking.
Where the wrong figure came from
We have not been able to confirm it. There is one notable coincidence: mice have about 1,100 working olfactory receptor genes. The mouse figure may have slipped across to humans through repeated citation, but that is only a hypothesis.
Not just one number
Figures for the human sense of smell disagree in several other places:
- Olfactory bulb volume measured by MRI differs about 2.5-fold between laboratories: one group measured 67–71 mm³ per bulb, while a large study of 560 people measured a mean of 54 mm³ for both bulbs together. Measurements can only be compared within the same protocol.
- The area of the olfactory epithelium, “about 5 cm² on each side of the nose” in textbooks, traces back to a measurement made in 1892 on two cadavers. We have not found a modern measurement.
Not every figure is wrong. But every figure deserves the same questions: measured on whom, how many people, and how.
How to check a figure
- Find the original study; do not stop at the article that quotes it.
- Check the species. Mouse figures are very often copied as human figures.
- Check the sample size and the age of the sample.
- When two sources disagree, record both rather than choosing the tidier number.
That is also the rule we apply to the data on this site; see our data standard.
The underlying evidence is in our research section: Part III. The human olfactory epithelium and bulb and Part V. Synthesis.
Sources
- Maresh, Rodriguez Gil, Whitman and Greer (2008). Principles of glomerular organization in the human olfactory bulb – implications for odor processing. PLoS ONE 3(7), e2640.
- Meisami, Mikhail, Baim and Bhatnagar (1998). Human olfactory bulb: aging of glomeruli and mitral cells and a search for the accessory olfactory bulb. Annals of the New York Academy of Sciences 855, 708–715. (abstract read)