[Neuroscience] Re: Rat (Long Evans) Cerebellar Cortex Size

Matthew Kirkcaldie via neur-sci%40net.bio.net (by Matthew.Kirkcaldie from removeutas.removeedu.au)
Sat Nov 1 17:42:43 EST 2008


In article 
<cef08021-a794-4dbf-871e-9367f62e0c06 from w1g2000prk.googlegroups.com>,
 Kerry <kbrownk from gmail.com> wrote:

> Sorry, I should have been more explicit w/ the given purpose. I am not
> trying to use this information for any quantitative analysis. For now
> I am just trying to get an estimation of width (e.g. Euclidean
> distance from the most extreme sagittal points end to end). However,
> if I new the volume of the cerebellar cortex I could calculate the
> convex hull volume of the section I have to get a more exact
> proportion. 

I don't know if a convex hull is really the right way to approach this 
in terms of proportion - the cerebellum is highly involuted, and its 
function is quite clearly arranged in terms of the sheet (parallel 
fibres running in one direction and the dendrite trees of Purkinje cells 
orthogonal to those fibres, contacting tens of thousands of them). 
Really the only meaningful proportion of the cerebellum is the 
proportion of the cortical sheet area, or the proportion of total 
Purkinje cell numbers. 

Your parasagittal sections are good, however, because of course the 
folding and involution is mostly at right angles to the sagittal plane. 
A good first-draft approximation of sheet proportion would be to look at 
an unfolded surface and draw lines parallel to the midline. I don't 
think the convex hull is going to help because the depth of infolding 
varies considerably.

> I would guess the total volume is more readily available,
> but if I could find a picture w/ a scale bar of the cerebellum viewed
> from the rostral or caudal direction I could get an idea of how much
> section I have. The latter may be quantitatively meaningless but
> qualitatively, for those who have knowledge of the cerebellar cortical
> size and shape, one could infer what proportion of the cortex I am
> studying. A crude approximation is good enough as I am just trying to
> provide a cartoon that gives a rough idea (e.g. it is meaningful to
> say whether I have closer to 1% or 90% of the entire cerebellar
> cortex).


If you want a dimensionally accurate view of the cerebellum in three 
planes, I'd recommend you to the atlas by Paxinos and Watson, The Rat 
Brain in Stereotaxic Co-ordinates, 6th edition 2007. It's the standard 
reference work on the subject.

If you're not sure how far from the midline your sections were taken, 
the atlas has a parasagittal series at known distances from the midline 
which may allow you to match up your sections. It also had a series 
sectioned in the horizontal plane which should show you the full lateral 
extent of the cerebellum at most levels.

Another major variable to consider is tissue processing and the 
resultant shrinkage. A fixed brain can shrink by up to 15%, and the 
degree of shrinkage varies in white matter and grey matter, making 
things even more complicated!

Sorry for such a fastidious answer, but with a complex three-dimensional 
structure like the cerebellum, back-of-the-envelope approximations are 
often startlingly wrong.

Cheers,

Matthew.
** Posted from http://www.teranews.com **


More information about the Neur-sci mailing list