Gail, I would suggest zooming in as much as possible on each specimen when taking the image. Also, be sure to include a scale bar in the image. Using TPSDig to digitize landmarks, you digitize the landmarks, and then set the scale on each individual image (this option is found in the Image Tool box). This will convert pixel units to some other useful size measure. By changing magnification, the object is then 'maximized' with respect to landmark placement, which minimizes digitizing error to the extent possible.
Dean > Hello all, > > Im looking at allometry of caprellid amphipods and am concerned about > relative measurement accuracy at different sizes. > > I am afraid that the error margin will be different according to the > size of the individual. For example if I place the landmarks on a small > image of a small individual, which will then be adjusted for size to > compare the shape to a large individual, the scope for error may be > larger than if I have a large image of a large individual, which would > be adjusted down for size. OR if I have the same size images of a small > and large individual, would the landmarks on the small individual be > more accurately placed? > > Can anyone offer any advice? > > Thanks, > > Gail > > > > Gail V. Ashton, > > Post-doctoral Fellow > > >·}¸¸<<< > > > > Smithsonian Environmental Research Centre > > 647 Contees Wharf Road > > Edgewater, Maryland, 21037-0028 > > Telephone: 443 482 2354 > > Fax: 443 482 2375 > > > > > -- > Replies will be sent to the list. > For more information visit http://www.morphometrics.org > Dean C. Adams Associate Professor Department of Ecology, Evolution, and Organismal Biology Department of Statistics Iowa State University Ames, IA 50011 -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
