Glomeruli

Overview

From: https://en.wikipedia.org/wiki/Glomerulus_(cerebellum):

The cerebellar glomerulus is a small, intertwined mass of nerve fiber terminals in the granular layer of the cerebellar cortex. It consists of post-synaptic granule cell dendrites and pre-synaptic Golgi cell axon terminals surrounding the pre-synaptic terminals of mossy fibers. (from: http://neurolex.org/wiki/Category:Cerebellar_glomerulus)

Density

In cat:

In 1 cu.mm of the granular layer 98,800 glomeruli are found on average. citation?

“With direct counting of the glomeruli the corrected density was found to be 98,879 +/- 4,065/cu.mm; …” [PalkovitsM+2-1972], p. 23.

Quantity

In cat:

The granule cell-glomerulus ratio is 27-28:1. [PalkovitsM+2-1972] The density of granule cells was found earlier 1° to be 2,800,000/cu.mm (for the living), hence the glomerulus-granule cell ratio is 1:28.32. [PalkovitsM+2-1972], p. 23.

“According to earlier data [PalkovitsM+2-1971b] the granule cell-glomerulus ratio is 27-28:1, the mossy fiber-granule cell ratio is therefore 1:460.” [PalkovitsM+2-1972], p. 28.

The granular layer contributed to 29.09% of the total cerebellar volume, their absolute number being 2.2 x 10^9. [PalkovitsM+2-1971b] p. 29.

Connection to grannule cells

Divergence

Cat:

From [PalkovitsM+2-1972], p. 26: “Four mossy fibers entering a folium give rise to 16 rosettes each, hence a total of 64 glomeruli. Since one glomerulus has synaptic contacts with an average of 28 granule cells, the total number of granule cells reached by the 4 mossy fibers will be 1,792. Each granule cell is presumed to pick up excitatory impulses from 4 glomeruli belonging to different mossy fibers by as many dendrites.”

but also, from the same paper, page 28: “The granule cells have 4.17 dendrites, on average; the average mossy rosette is contacted by 112 granule dendrites. The number of postsynaptic units (dendrite digits) is 10.2/dendrite and 1,142/glomerulus.”

This seems contradictory, (28 granule cells vs 112 granule dendrites). I think the “dendrite digits” refers to a dendrite having multiple protrusions as shown in [EcclesJC+2-1967] Fig. 75 (see note for this paper), so those can be ignored for the purpose of calculating the divergence.

So, the question is whether the divergence from glomeruli is 28 or 112? I think it is 112, and that the value of 28 is calculated by taking into account the convergence of 4 (112 / 4) = 28. This seems compatable with the divergence:convergence from mossy fibers to granule cells given in the [LoebnerEE-1989] fig 2 which is 1.7x10^3:4 because 1,792 is given as the number of granule cells reached by 4 mossy fibers on average (quote above from [PalkovitsM+2-1972], p. 26) and Loebner is using that as the number reached by one mossy fiber, which means that mossy fiber has 4x the divergence than that given by one mossy fiber.

Another potential contradiction is the total number of mossy fibers. In the Loebner paper, the divergence:convergence from mossy fibers to granule cells is given as 1.7x10^3:4 and the number of granule cells is given as 2.2 x 10^9. From this, the number of mossy fibers would be about (2.2x10^9 * 4 / 1.7x10^3) = 5.2x10^6 mossy fibers. This is more than the estimated number of 2.4 x 10^6 mossy fibers, which might already be an overestimate.

Rat:

From [JakabRL+HamoriJ-1988], abstract The results demonstrate that, in the rat cerebellum, there is a high degree of convergence of granule cells at a glomerulus (53 to 1); and that there is a rich inhibitory input to about 60% of all granule cell dendrites.

Convergence

In our Golgi-Kopsch stained material the number of dendrites ranged from 2 to 7. 61.6% of cells had 4 dendrites, 22.4% had 5 dendrites and 15.2% had 3 dendrites. The average dendrite number was calculated as 4.17. [PalkovitsM+2-1972] p. 24.