Those distances are not that difficult to compute. Simply read
the tps file into tpsSmall, click the 'compute' button, and then
go the File|Save menu and you will see that saving a matrix of
Procrustes distances is one of the choices.

----------------------------------------------
F. James Rohlf, Distinguished Professor & Grad. Prog. Dir.
Dept. Ecology & Evolution, Stony Brook University
Stony Brook, NY 11794-5245


-----Original Message-----
From: morphmet [mailto:[EMAIL PROTECTED]
Sent: Friday, February 17, 2006 2:01 PM
To: [email protected]
Subject: Procrustes Distance between individuals

Hello everybody,



I'm working with a group of fish called hamlets and I have a
total
sample size of 226 individuals.  I have used 14 landmarks,
Procrustes
superimposition and various statistical techniques to compare 6
groups,
which these fish are divided into (based on sampling location
and
morphotype).  One useful indicator of morphological similarity
between
groups is Procrustes distance.



One thing which I am interested in doing is obtaining a data
set of the
morphological distances between individuals.  However, as this
equates
to 1000's of separate comparisons, I have calculated that if I
were to
start aligning each pair of individuals, and find the
procrustes
distance between them, then I wouldn't finish working until
somewhere
near 2007.  I doubt it would be worth the effort to be honest.



This leaves me with 3 alternatives as I see it:



    1. I can use a spreadsheet and the co-ordinates of all the
aligned
       landmarks to obtain the distance between pairs of
individuals.
        This would be quite easy to do but, as the pairs are
not aligned
       with each other (but with the entire dataset), the
distances are
       not Procrustes distances.
    2. I could alternatively just analyse a small subsample,
with
       individuals taken from each group.
    3. Or I could just give the whole thing up as a bad idea.
I've not
       found an example of somebody else using this kind of
data, maybe
       there is a reason why it shouldn't be done.



I am new to morphometrics and if anybody has any ideas on this
I would
be very grateful for your help.



If you wish to reply to me personally then my email address is:
[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>.



Cheers,



Ben Holt, University of East Anglia


--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org




--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org

Reply via email to