[R] Box's M/likelihood ratio test for equal v-c matrices

[Thomas Lotze]

Hopefully a quick question:

is there a package/routine to perform Box's M (like proc discrim in SAS) 
or some other test for whether two multivariate samples (presumed 
multivariate normal) have the same v-c matrix?  I've looked, but can't 
find it.

If not, I can certainly code it up myself (and perhaps submit it for 
inclusion in the mvtnormtest package), but I wanted to check if it exists.

Many thanks,
Thomas Lotze

"Isn't sanity just a one-trick pony anyway?  I mean, all you get
is that one trick--rational thinking--but when you're good and crazy,
well, the sky's the limit!"
 				-The Tick