Thanks to all who replied.  It seems that most agree that Goodes is a
problem to deal with.
There is some C code to deal with the projection at:
ftp://edcftp.cr.usgs.gov/pub/software/misc/gihll2ls.c
which is said to be:
"sample code in C that converts Goodes homolosine map projection to
nice, manageable
rectangular lat / lon coordinates"
according to
http://www.geocities.com/TimesSquare/1658/earth.html

There is a veru good and explanatory article about Goodes in Arcinfo at:
http://www.esri.com/library/userconf/proc98/PROCEED/TO850/PAP844/P844.HTM


Finally, there is actually a projection tool that converts raster data from
Homolosine into a few others (have not used it yet).  It can be found at
http://ag.arizona.edu/~epfirman/windisp3.html


Other info can be found at:

 There was a post on the ESRI support site about this projection, maybe you
can contact that person directly. Here's the link to the post:
http://support.esri.com/search/ViewListPost.asp?dbid=2960

Also, there is an ESRI Conference listing about the projection, it's from
1997, but may help you. Check out:
http://www.esri.com/library/userconf/proc97/proc97/to600/pap578/p578.htm

Thanks again.



^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Matt Reeves
Ph.D. Candidate/Research Assistant
University of Montana
School of Forestry/NTSG
Missoula, MT
59812
phone:  406 243 6228
Fax    : 406 243 4510
reeves@ntsg.umt.edu
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^