----- Forwarded message from "Smith,
Vann E" <vsmi...@tulane.edu> -----
Date: Wed, 3 Apr 2013
21:34:31 -0400
From: "Smith, Vann E" <vsmi...@tulane.edu>
Reply-To: "Smith, Vann E" <vsmi...@tulane.edu>
Subject: extended
eigenshape analysis for shark teeth
To: "morphmet@morphometrics.org"
<morphmet@morphometrics.org>
Hello everyone,
I'm a master's student in geology at Tulane University. I'm planning on doing my
master's thesis on morphometric
techniques for identification of modern and fossil shark teeth. I have been
using Iwata's SHAPE program for elliptic fourier analysis. After reading Dr.
MacLeod's 1999 paper on eigenshape analysis ("Generalizing and Extending the
Eigenshape Method of Shape
Space Visualization and Analysis"), it seemed like his method of extended
eigenshape analysis might be ideal for shark teeth, with three landmarks, one at
the tip of the tooth, and two at the crown/root boundary on the outline. I
would like to see how well
this method differentiates species and higher taxonomic levels.
My problem now is finding a software
program to take my pictures, add the landmarks, and run the analysis so that it
can be plotted with principal component analysis (or any x-y graph to visualize
variation). On the website http://www.morpho-tools.net/ there
is a link to download a program that Dr. Jonathan Krieger developed for his Ph.D
for extended eigenshape analysis,
but the link is broken. If anyone could help me with a link to this program, or
with information about a similar program that performs the same function, I
would be very grateful.
I am not very familiar with morphometric
techniques or statistics, so this is a slow learning process. I have read
criticisms of elliptic fourier analysis that argue the lack of landmarks is a
serious drawback, but landmark analysis that ignores curvature seems crude to
me. I have become acquainted with Iwata's SHAPE program, and may continue to use
it for my thesis, but Dr. MacLeod's method as described in his 1999 paper seemed
ideal for shark
teeth.
My email address is vsmi...@tulane.edu if
anyone has any information they think might be helpful. Thanks in advance for
helping out a novice in the field-
----- End forwarded message -----
