Daniel Manrique-Castano, Neuroscientist, Germany

16 April 2020

This article was originally published in Spanish at https://hablemosdeneurociencia.com/cuantas-celulas-hay-en-el-cerebro/

Since the school years, the idea that neurons are the cells of the brain has entered our heads. While doing social research for my next popular science book, I learned that no one outside of the field of neuroscience knows that there are up to a dozen other cells in the brain besides neurons. Well, I think it is absolutely necessary that we start to change this very limited perspective on the central nervous system (CNS). In reality, the brain is composed of a diverse multicellular population, and although neurons add up to a significant number, the rest of the cells are by no means negligible.

Last year I published an article called “The brain is much more than neurons”, where I gave some estimates about the percentages or numbers of various cell groups in the brain. This time I want to correct the mistakes I made, mistakes that are not only mine, but of the whole neuroscientific community. Fortunately, a couple of years ago I found an article by the Brazilian neuroanatomist Suzana Herculano-Houzel [5], which opened my eyes to a huge mistake, which, believe it or not, has been transmitted only by tradition from generation to generation.

The perpetuation of an error

In the aforementioned article of my authorship I wrote “Astrocytes are the main cell population in the CNS, doubling the neurons in the human and a third in the mouse”. This statement is derived from the widely held view that the cell group known as glia or neuroglia, which includes astrocytes, oligodendrocytes, microglia, Müller’s glia, radial glia, NG2+ cells, tansycites, Schwann cells and Bergmann’s glia, outnumber neurons by a factor of ten. This can be found in eminent neuroscience books such as Erik Kandel’s Principles of Neuroscience [3] and Mark Bear’s Neuroscience: Exploring the Brain [1], as well as multiple scientific articles in different research contexts [4].

It is also stated that astrocytes are the most abundant cells within the group of neuroglial cells, so they would greatly outnumber neurons. However, Herculano-Houzel and his colleagues let us know that all these estimates are due to an error of interpretation that arose in the 1960s and 1970s, and that by inertia was transcribed from book to book and from article to article without any reference or experimental basis to confirm it. It is interesting how an error can be perpetuated in science. In his book Ignorance [2], neuroscientist Stuart Firestein of Columbia University, expresses his surprise when he learns of this circumstance. For him, it was so normal to think that glia were ten times more numerous than neurons just because everyone said so.

“It was so normal to think that glia were ten times more numerous than neurons just because everyone said so”

The number of cells in the CNS

Cellular quantification in the CNS is a difficult task that has been carried out for at least a hundred years. With the methods we have available it is impossible to count all the cells in the brain, so representative portions must be taken and generalized, which brings with it various biases and errors in the estimates. The crux of the problem uncovered by the Brazilian researcher was that the erroneous estimates of the ratio of neurons to neuroglias were not the result of technical limitations, but rather the indiscriminate adoption of contradictory and unproven notions that morbidly passed from the oldest to the youngest neuroscientists. I suggest that you read the Herculano-Houzel article carefully; I think it represents one of the biggest blunders in our discipline.

Let’s move on to the numbers. Based on the analysis of different estimates and own experiments, the researcher gives us some figures about the amount of cells that make up the CNS. It is estimated that the cortex, which occupies 80% of the total volume of the brain, has between 10 and 20 billion neurons. For its part, the cerebellum, which occupies 10% of the brain’s volume, contains some 50-70 billion neurons. Yes, you read that correctly and it is not a mistake on my part, there are three times more neurons packed into this small structure in the back of our heads, than in all the rest of the brain. Finally, what can be called the rest of the brain, which includes the diencephalon, the striatum and the brain stem, has about 700 million neurons. In summary, the data we have at present indicate that the brain contains some 90 billion neurons.

The CNS is conformed at least by similar proportions of glia and neurons. 

Version 8.25 from the Textbook OpenStax Anatomy and Physiology


Let us now turn to the glia, the cell group that poses the problem here. The glia are not ten times more numerous than the neurons; on the contrary, their number is similar at best. In my previous article I mistakenly referred to the fact that with evolution the ratio between glia and neurons had tilted in favor of the former, but Herculano-Houzel let us know that such a statement is not true in light of the available data. The proportions between glia and neurons are maintained along the animal evolutionary branches without significant differences. This is an error that is still quite widespread among experts and must be corrected. Quantifications indicate that there are 20-40 billion glial cells in the cerebral cortex; their number is 3 billion in the cerebellum and about 7 billion in what I called the rest of the brain. This means that in the cortex and the rest of the brain the glial cells are more numerous than the neurons, particularly, in the rest of the brain they could be, indeed, ten times more numerous than the neurons.

Brain astrocytes labeled with an antibody against the Glial Fibrillary Acid Protein (GFAP).

However, this is only a small percentage of the cells that inhabit our heads. If we take all of these values and put them in the context of the entire brain, the glia (including endothelial cells, which are not glia, but are within this value) total is no more than 50 billion cells, about ¾ of the number of neurons. In addition, it has been estimated that in the glial cell group 45-75% are oligodendrocytes, 20-40% are astrocytes and the microglia are 10%. I know that the ranges are quite wide, but the task of quantification is difficult, and look at it from any perspective, in no way does it support the claim of “ten times more glia than neurons”. Neither glia are the largest cell population, nor astrocytes the largest population of glia, oligodendrocytes are, cells that are located in the white matter and myelinize the neuronal axons. As far as I know, there are no similar data on the estimation of the rest of the glial cells that I mentioned at the beginning. Perhaps it’s time we got down to business.

The article written by Herculano-Houzel  is not difficult to understand and is quite revealing, yet largely unknown. A few months ago I referred to this aspect of quantification in a comment I made to an article on the PubPeer website that started “astrocytes are the largest cell population in the CNS”. The author of the text replied that he was not familiar with the work of Herculano-Houzel, but that I was getting into an “interesting” discussion. I do not understand what discussion he was referring to, when the data provided by different methods of cellular quantification are unequivocal in the main concussion: there are no more glia than neurons in the CNS.


[1] Bear, M., Connors, B., y Paradiso, M. (2007). Neuroscience: Exploring the Brain. Philadalphia, Estados Unidos: Lippincott Williams & Wilkins.

[2] Firestein, S. (2012). Ignorance: How It Drives Science. Londom, UK. Oxford University Press.

[3] Kandel, E. Schwartz, J., y Jessell, T. (2000). Principles of Neural Science. Estados Unidos, McGraw-Hill Medical.

[4] Ronaldson, P. y Bendayan, R. (2008). HIV-1 viral envelope glycoprotein gp120 produces oxidative stress and regulates the functional expression of multidrug resistance protein-1 (Mrp1) in glial cells. Journal of Neurochemistry, 106(3):1298-313.

[5] von Bartheld, C. S., Bahney, J., & Herculano-Houzel, S. (2016). The search for true numbers of neurons and glial cells in the human brain: A review of 150 years of cell counting. Journal of Comparative Neurology, 524(18), 3865–3895.

Want to know what’s new in ScienceLogs?


Pin It on Pinterest

Share This