Alexander,
The image you shared is a curious mix of homologous and non-homologous points.
I would chose points that can be found in each individual of the species in
both fore and hind wings. That is the only way a comparison can be made with
traditional landmark analysis. Your forewing landmark# 52 corresonds to a
hindwing landmark #58 and hindwing #59 does not have a homologous forewing
landmark.
In addition forewing landmarks 1 and 7 seemingly correspond to hindwing 1 and
15. In my experience, these cross vein landmarks are quite variable from
individual to individual within a species … a larger specimen will have more
crossveins between homologous landmarks such as forewing #31 and 53 and 60 and
10, … which appear homologously in all Anisoptera and perhaps all Odonata. You
have named these ‘true’ order-wide landmarks in the hind wing 37, 60, 29 and 10
… a mix of similar and different designations. Lesson, do not use minor cross
veins as landmarks and learn the basic vein naming and branching of Odonates to
guide you in chosing landmarks to use.
Foregive me if I sound derisive but I can not see how your naming of landmarks
would work or how you could use them to compare the shape of front and hind
wings.
Look at this poster for a set of hindwing landmarks but know that the for and
hindwings have some differences in the availability of homologous landmarks,
URL:
http://www.bio.umass.edu/biology/kunkel/pub/reprints/Warrell_SympetrumPoster-S2014.pdf
Joe
-·. .· ·. .><º>·. .· ·. .><º>·. .· ·. .><º> .··.· >=-
=º}><
Joseph G. Kunkel, Research Professor
122C/125 Pickus Center for Biomedical Research
Marine Science
University of New England
Biddeford ME 04005
http://www.bio.umass.edu/biology/kunkel/
> On Jul 7, 2017, at 6:18 AM, Alexander Blanke
> wrote:
>
> Dear all,
>
> the world of morphometrics is relatively new to me, so forgive me if this
> question is an old hat:
>
> I would like to know whether two structures, which are not connected to each
> other, show concerted shape variation (potentially in response to functional
> demands which act on both structures in the same way).
>
> The example here are dragonfly fore- and hindwings. To test whether these two
> structures show correlated shape variation I created a set of landmarks like
> in the attached file. However, I am a bit unsure how to proceed. Obviously it
> is not advisable to use tests which require one set of landmarks which have
> been Procrustes superimposed since we are talking here about two independent
> structures which are not attached to each other (Fore- and hindwings can vary
> in their position to each other which has nothing to do with a real “shape
> signal” but is simply due to the placement of the wings when capturing the
> images).
>
> I think I need to perform two independent Procrustes superimpositions and go
> from there. If I understood correctly, it is then not possible to use
> functions such as the integration.test in geomorph since all partitions used
> in this test need to stem from one set of Procrustes aligned landmarks. Maybe
> I am too focused on tests concerned with integration and modularity…
>
> My search in the literature was so far not successful since most of the tests
> of modularity and integration concern connected structures. My expectation
> for the wings is that fore- and hindwings show correlated shape variation due
> to the functional demands of flight. Certain parts of the forewing will
> “influence” the hindwing shape due to alterations of the flight performance.
> I have gathered a few functional parameters to do further testing of this
> hypothesis (the total dataset contains 192 species each with ~122 landmarks)
> but first I want to get this shape analysis right.
>
> If I missed literature on that topic I would be grateful for any hint to
> further reading.
>
> Thanks in advance for your thoughts,
>
> Alex
>
>
> --
> MORPHMET may be accessed via its webpage at http://www.morphometrics.org
> ---
> You received this message because you are subscribed to the Google Groups
> "MORPHMET" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to morphmet+unsubscr...@morphometrics.org.
>
--
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
---
You received this message because you are subscribed to the Google Groups
"MORPHMET" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to morphmet+unsubscr...@morphometrics.org.
