134 
Discussion 
nated structure, then I agree it is restricted 
to the mammalian grade. However, if we 
talk about the identity of cell populations, 
and we say that DVR and isocortex are 
homologous, then we are saying that we be- 
lieve that these cell populations in reptiles 
and mammals can be traced back to a popula- 
tion or populations in ancestral reptiles or 
amphibians. These populations need not have 
identical connections or functions to be hom- 
ologous. 
QUESTION: What criteria do you use: 
what boundaries are you using to distinguish 
between phylogenetic similarity and con- 
vergence? 
NORTHCUTT: If I thought DVR were a 
single modality unit — that is, a target of a 
single visual projection — then I would very 
much suspect parallelism. If what we are 
looking at were optic tectum to rotundus to 
DVR, then I could believe that the agamids, 
the varanids, and one group of teids inde- 
pendently hypertrophied the DVR, and I 
would argue parallel evolution. All I would 
have to argue is that these three groups came 
under similar selective pressues for that par- 
ticular visual pathway — whatever it might 
be — independently; I have considered that 
as a possibility. I do not believe this hap- 
pened, however, because DVR is not just a 
visual target. There is a visual target, an 
auditory target, a somesthetic target, and a 
motor area within DVR. What I am really 
saying is that we have two choices: Three 
separate radiations independently evolved 
complex brains with similarities in some 30 
neural populations encompassing several dif- 
ferent sensory modalities; or this complex 
pattern evolved only once. 
