X & Y cells in primates: ANSWERS

Christian Ronse ronse at dpt-info.u-strasbg.fr
Mon May 24 08:22:19 EST 1993

I received many answers to my query on the relationship between the
classification of the cat's retinal ganglion cells into X & Y, and the
magno and parvo systems in the primate. In brief: X & Y are good for
cats, but not primates!

Here are all the answers I received:

From: Peter Bell <bell-peter at YALE.EDU>
Subject: Re: X, Y, magno, parvo (in primate vision)

right you are -- the x and y cells were developed in cats, and are not 
used in primates. Most annoying decision, but that is the way it is

bell at minerva.cis.yale.edu


From: Thomas at pixel.cellbio.mcw.edu (Thomas C. Trusk, PhD)
Subject: Re: X, Y, magno, parvo (in primate vision)

Please understand that this is mostly a difference in semantics. The
original classification of retinal ganglion cells was based on
physiological response to light in cats, and were separated into X and
Y. This was shortly shown to apply to morphological differences as
well. There is somewhat of a similar dichotomy in the
morphological/physiological characteristics of primate ganglion cells
as well, and in the early going these cells were often referred to as
x-like and y-like. Many people confused the issue by referring to x
and y cells in primate retina. Today we believe that at least THREE
types of primate ganglion cells can be distinguished based upon
responsiveness, morphology, and connectivity. The labelling of primate
cells as x or y, x-like or y-like, or even A and B, has fallen out of
favor. In fact, the situation remains unsettled, and you'll see many
naming schemes. It appears that the x and y physiological response
properties of Cat ganglion cells are observed in 2 subsets of Primate
ganglion cells that both project to parvocellular layers (some refer
to them as Px and Py, others prefer P-alpha and P-beta). There is
another class of primate ganglion cell that projects to the
Magnocellular layers (M cells, and these may be divided into x-like
and y-like classes!) Finally, there is evidence that another ganglion
cell type that projects preferentially to the intercalated layers of
the LGN (stuff between parvo and magno layers). These cell have been
referred to as gamma cells.

So, to answer your question, Yes...X and Y apply ONLY to cats.

Thomas Trusk, PhD
*Thomas Trusk, PhD                   *    (DO NOT TYPE IN THIS SPACE!)    
*Dept. of Cellular Biology & Anatomy *                                    
*Medical College of Wisconsin        *  Witty saying under construction   
*Milwaukee, WI  53226                *           HARD HAT AREA            
*MaBellNet: (414) 257-8504           *                                    
*INTERNET: ttrusk at its.mcw.edu        *         (FOR OFFICE USE ONLY)      


From: James B. Hutchins <jbh at Anat.UMSMed.Edu>
Subject: Re: X, Y, magno, parvo (in primate vision)

Dr. Ronse,

There is much debate about this issue, and I'm sure you will get answers
'yes' and 'no'.  My answer is that I believe there is a pretty good
correspondence between X, Y and W and parvo (P), magno (M) and S layer/
intercalated layers/etc (respectively) in cats and primates.

The other side would argue that based on a number of properties, there
are some characteristics of (say) cat X cells which are not shared by
primate P cells. I would argue that you can always find some
characteristic to split these apart, but that the common features far
outweigh the disparate features.

Put another way, how does the variability _within_ the cat X cell
population compare to the variability _between_ X and M?

Please note that biologists are divided into two groups: "lumpers" and
"splitters".  Lumpers see the common features between cells, animals, etc.
while splitters focus on the minute differences.  I am a lumper.  Almost
all other biologists I know are splitters (it started with Linnaeus, I
think :-) ).

Hope this helps.


Jim Hutchins                    []     E-Mail: jbh at anat.umsmed.edu
Dept of Anatomy                 []
Univ Mississippi Med Ctr        []


From: julianb at cogs.susx.ac.uk (Julian Budd)
Subject: Re: X, Y, magno, parvo (in primate vision)

This topic does generate some confusion, let me see if
I can explain. 

	RETINA		dLGN	         Area 17
Monkey: "alpha"-------> PARVO ----> IVc(beta) + IVa + IIIb + VI

	 "beta"-------> MAGNO ----> IVc(alpha) + VI

Cat:	 X-type------->  X *   ----> IVb + VI

	 Y-type------->  Y **  ----> IVa + VI
*laminar A **laminar A1

Without looking these up, I'm not sure if I've got the "alpha"
and "beta" types of retinal ganglion cell the correct way round,
ie "alpha" may go to MAGNO and "beta" got to PARVO. The classes
were decide by Perry,V. et al (1983 or 1984). 

A good book on visual pathways is by Guy Orban: Orban, G. (1984) 
Neuronal Operations in Visual Cortex. Springer-Verlag. 

I hope this helps. 


From: curtis at astra.vision.mcgill.ca (Curtis L. Baker)
Subject: X/Y and M/P

	I think it is fair to say that the vast majority of the visual
neuroscience community would agree that the X/Y distinction for cat
retinal ganglion cells (and, LGN cells) is not obviously equivalent to
the M/P distinction for primate retinal and LGN cells, in spite of
some similarities. The best work on this comparison has been done by
Bob Shapley and his colleagues at NYU - they seem to believe that the
P-cells are unique to the primate, and that the cat's X and Y cells
collectively correspond to the primate's M cells. In some of their
writing, I think they have mentioned the idea of the monkey having two
kinds of M-cells, Mx and My ("X-like" and "Y-like" M-cells); however I
have not seen this idea clearly developed and supported anywhere in a
full paper.
	Sorry I don't have any ref.s, this is all from memory. You might want
to send Bob Shapley email :   shapley at cns.nyu.edu

	Curtis Baker


From: Brendan O'Brien <creator at u.washington.edu>
Subject: x,y,magno,parvo

As I understand it, these are the current conventions.

x,y are used for cats only and refer to the small and large ganglion cells
respectively.  The small ones of course getting their input from single
cones and subserving visual acuity whereas the large y cells get input
from multiple cones and are most sensitive to spatial contrast.

In the primate retina these cells are called midget (x-like) and parasol
(y-like) cells.  Presumably they serve the same function.  

alpha, beta, and gamma refer to experiments done on the optic nerve where
visual stimulation caused three seperate volleys of action potentials
(a,b,g)  due to differing conduction velocities.  These studies were done
before there the retinal morphological studies elucidating the x,y,w cells
in the cat and parvo magno cells in monkeys.


alpha cells, y-cells, and parasol cells are  teleologically "equivalent"
beta cells, x-cells, and midget cells  '	' 	'
gamma cells, w-cells, and ? (I don't remember if they have a specific name
in primates.

Brendan O'Brien

snail mail:
Depts of Psychology and
Physiology and Biophysics
Univ. of Washington
Seattle, WA 98195

creator at u.washington.edu

feel free to respond if I am not clear or you need more information.


From: charles at anatomy.ucl.ac.uk (Charles King)
Subject: Re: X, Y, magno, parvo (in primate vision)

I actually work on the hippocampus, but there is a large vision group here
at UCL and I remember talking to one of them a couple of months ago who
said the whole X/Y-cell thing should be ignored in primates (ie X and Y
cells are for cats only). I can't give any more modrn refs than the ones
you've got, though.

Charles King 
Dept of Anatomy, University College London
charles at anat.ucl.ac.uk


Subject: Re: visual pathway neurons

Funny you should ask that question, about X/Y ganglion cells vs. A/B 
ganglion cells projecting to the magno-/parvocellular pathways, because I 
was wondering the same thing. It seems to me that it all depends upon who 
you ask: clearly, a lot of people use the older, X/Y terminology (not a 
half-hour ago, my neuroanatomy instructor used the X/Y terminology when 
describing the "where" vs. the "what" information streams). In fact, I 
don't recall anybody using the A/B terminology--- and I even took a CNS 
course with Marge Livingston a couple of years ago! I'd imagine that the 
reason why she and David Hubel used the A/B terminology for the paper that 
you referenced (I'll look it up, I promise...) was probably because the X/Y
cell to magno-/parvo- cell correlation isn't precisely accurate. It makes 
for a good story, though... ;-)

Please do let me know if you hear otherwise, or if you learn the true 
reason why Livingston and Hubel used the A/B terminology. I am allegedly a
vision science researcher, you see... ;-)

Take care!

  -- yh.

Yukiharu "Yuki" Hadeishi   -   <yhadeish at biomed.med.yale.edu>
The Interdepartmental Neuroscience Program will have
absolutely nothing to do with any of my opinions...


From: "John P. Brockway" <JOBROCKWAY at apollo.davidson.edu>
Subject: Re: X, Y, magno, parvo (in primate vision)

I don't know whether or not you have seen the Chapter in 
Annual Review of Neuroscience, 1993:16:369-402
by Merigan, W. H. & Maunsell, entitled
"How parallel are the primate visual pathways?"

It may help in answering in a schematic manner some of your
question. There are approximately 140+ references

John P. Brockway, Ph.D.
Neurosciences Institute			Department of Psychology
Doctor's Building			Davidson College
Box 61 Suite 207 1012 S. Kings Drive	Davidson, NC 28036
Carolinas Medical Center
Charlotte, NC 28283-0061		704-355-5696


More information about the Neur-sci mailing list