Lambir butterfly project (proposal)

From BiodivBorneo09

Jump to: navigation, search

An Analysis of Patterns of Coloration and Size Differences in Forest and Forest Edge Populations of Lepidoptera

Abigail Schoenberg and Kristina Prus



Our project will focus on butterflies and their patterns of location. We believe that the system of butterfly habitation is interesting because, while much attention is paid to the locations and strategies of butterfly larvae, it seems that little attention has been paid to the strategies of the organisms once they reach adulthood because they don't live long afterward. In our opinions, this system is just as interesting as the system of the larvae. We imagine that the organisms still have the instinct to survive despite their short lifespan, and so patterns will be evident.


Our question is: "Is there a difference in morphological traits of butterflies between the outskirts of the forest and the inside of the forest in Lambir, and if so, are there significant patterns of differences in morphologies between the two populations?" We hypothesize that yes, we will find different species compositions of butterflies and moth between the outskirts of the forest and the inside of the forest in Lambir. Those found inside the forest will be smaller and neutrally colored, and those found outside the forest will be larger and brightly colored. Our rationale is that smaller, darker, non-poisonous butterflies would habitate inside the forest because they need to be nimble and blend into the dark shades; they are also relatively safe from birds, and so they do not require a poisonous defense system or bright warning colors. Butterflies on the outskirts, however, would be larger and brighter because they need not be nimble, yet they need to have a warning signal so that predatory birds think they are poisonous; they also have to blend into a more colorful environment. We predict that color, above all, will have the strongest significance in the butterfly's location. A large, clumsy butterfly could possibly get around in the forest, but a bright one would be too exposed against the neutral backgrounds to survive predatory attacks. This question is interesting because there is a great deal of diversity among butterflies in Lambir, but to the naked eye, there does not seem to be any order to it; this question could answer the mystery of why certain butterflies are found in certain places. This topic could also aid future butterfly research projects, as experimenters will know where to look for butterflies with certain morphology.


When we are finished collecting our specimens, we will have placed 3 pairs of traps in 3 different areas, for a 24-hour time period each (Saturday --> Sunday, Sunday --> Monday, Tuesday --> Wednesday; we lose access to the traps on Wednesday when Rod leaves). We will collect the specimens in the trap, bring them back, keep records of species abundance and richness in each habitat, wing-span, colors, poison levels, etc. We will try to pin at least one individual from each species we catch. We will then walk a transect of the habitat nearby each trap by measuring a 200m walk. We will then take 10 minutes to walk that length, and another 10 minutes to walk it in the reverse. We will count all butterflies that we see on either side simultaneously and record the number, predominant color, approximate size (big or small), and the flight style (slow or fast).


We will analyze our data in a number of ways:

  • To determine whether there is a difference in species composition, we will use the T.Test and/or Wilcox Test.
  • To determine the individual significance of each factor on the chosen habitat, we will do a chi-square test and/or a correlation test.
  • To determine the interaction effect between all the characters in the morphology, we will do a general linear model.

Therefore, we have 3 different statistical hypotheses:

  • We hypothesize that the T. Test and/or Wilcox Test WILL reveal a significant P.value, suggesting that there is a difference in species composition between the two habitats.
  • We hypothesize that each character (color, wingspan, and poison level) WILL be shown to have a significant effect, according to the chi-square tests.
  • We hypothesize that, as a result of the general linear model, we will learn that color has the greatest impact on the choice of habitat for the specimen.

The null hypothesis is that there is no difference in species composition, and that none of these characters have a significant effect. If we see significant P.values in the tests for which we have hypothesized, we will know that there is a difference in species composition, and that each character has a significant effect. If there are no significant P.values, we will know that our null hypothesis is true.

Back to Projects