[Wien] question about sgroup

[Fred Nastos]

Hi,

In some cases it seems like sgroup does not provide all the space
group symmetry operations.  It seems to provide all the rotations
(proper and improper), but does not seem to provide all 
rotations+translations.  I think my question/misunderstanding is
most easliy conveyed through an example.

Consider the two systems: zincblende and wurtzite.  Let's assume
we have a struct file for each case: zincblende.struct and wurtzite.struct.

If I run 'x sgroup' for the wurtzite case, sgroup finds that the structure
belongs to spacegroup 186, and the file wurtzite.struct_sgroup
has 12 symmetry matrices appended at the end of the file.  Those
12 matrices are the same as what one finds at the website:
	http://www.cryst.ehu.es/cgi-bin/cryst/programs/nph-getgen
under spacegroup 186.

Of these 12 symmetry operations, six of them are the point group
operations, and the other six are rotations+translations.  It seems
like sgroup provides the user with all the space group symmetry
matrices.

But, if I run 'x sgroup' for the zincblende case, sgroup finds that the
structure belongs to space group 216, and it lists 24 symmetry matrices
at the end of zincblende.struct_sgroup.  But space group 216, according
to the above site, has 96 space group symmetry operations (the product
of 24 point group operations and four different trasnlation vectors).  So, 
there appears to be a contradiction.

Thank you for your time.