On Tue, Jun 22, 2010 at 4:32 PM, Colin Nave <[log in to unmask]> wrote:
> Secondly, the difference in the cell dimensions (b=123.92 and c=128.89A)
> appears to be quite large and should lead to split spots which (I think)
> corresponds to non merohedral twinning. Did you observe these but integrated
> them as one?
The distinction between merohedry (incl. pseudo-merohedry) and
non-merohedral twinning is not whether the spots are split: splitting
to a greater or lesser degree is often observed in the
pseudo-merohedral case since the pseudo-twin law is never perfect.
Rather the defining feature is whether the overlap of the twin-related
lattices occurs in 3 dimensions (i.e. exact overlap for merohedry, or
approximate for pseudo-merohedry) or only 2 in the non-merohedral
case. In the latter case this means that there's no obvious
relationship between the spot positions for the components of the twin
(except possibly in the zone related to the plane of 2-D overlap).
Cases where 3-D overlap occurs only for some integer fraction of the
spots are often mistakenly termed 'non-merohedral' even though overlap
occurs in 3-D and so there's a clear relationship for the fraction of
spots that are twin-related. The correct term for this case is
'reticular merohedry' (or 'reticular pseudo-merohedry'). A nice site
where all the twinning terminology is clearly defined is:
http://www.lcm3b.uhp-nancy.fr/mathcryst/twins.htm
> (Regarding what to call the twinning I have some sympathies with Humpty
> Dumpty's view "When I use a word... it means just what I choose it to
> mean-neither more nor less" As important a philosopher as Wittgenstein.)
But it helps a lot if everyone can agree on the terminology (e.g. on
the precise definition of 'non-merohedral') - most of the arguments on
the BB seem to stem from the use of conflicting definitions!
Cheers
-- Ian
|