A video arrived in my inbox this morning that absolutely made my week. Jennie Russ and Ryan Buck from Evergreen State College have adapted the LLAMA project into an animated short:

What’s going on?

Myrmecology took a technological leap in the 1990s. A protocol for standardized mass collecting of soil & leaf litter arthropods was refined from a technique called winkler sifting. A square meter of forest floor is chopped and sifted into mesh bags, and these are hung over a bag of alcohol to catch arthropods as they fall from the sample. It’s almost like magic. Strange creatures emerge from what seemed like featureless muck, and in astounding numbers.

Not only did the technique produce vast piles of specimens from a previously underexplored habitat (yielding a plethora of new species), it also introduced a singular methodology that could be replicated all over the world. With a standard sampling scheme, scientists could more directly compare the biological diversity of forests.  This new trick fueled the growth of ant macroecology, the study of global patterns of ant diversity, and scientists could really sink their teeth into the question of why some places have more species than others.

Brian Fisher’s Antweb grew out of a winkler-sifting project in Madagascar. Similar projects were launched in Paraguay, Fiji, and elsewhere. Project LLAMA– featured in the film above- surveys the litter arthropods of Central America.

What a charming video. I imagine it will become required viewing for anyone learning the ropes of tropical biodiversity.

13 thoughts on “Sifting”

  1. It’s is quite marvelous! The exact kind of thing I would expect to come out of Evergreen State too! They’ve got the most amazingly bizarre educational system there and the opportunities for student creativity are pretty awesome.

  2. Sign me up! Great video.

    The winkler sifting seems similar to using a Brelese funnel (You collect your sample and place it in a metal funnel with a light above to heat the sample and a jar of ethanol below to catch the organisms. Dr. Bill Shear, noted millipede specialist, said it was like Christmas. I learned this technique for arthropod sampling back in the dark ages). Are there important differences in these techniques?

    Also, the winkler funnels are hung up for some days(?) — does this mean that some things get eaten before you get the sample?

    Finally, would this be a good technique for me to sample especially the ants in my prairie system which essentailly has no leaf litter? I have been using pitfall traps so far.

    1. Imaging you trying to get a collection site in middle of nowhere, many hours of walking, no eletricity… The gear quantity and the weight of mini winkler are perfect to acess biodiversity in remote places – it is the great difference between both techniques, in my point of view.

      There is no way to hung up winkler extractors (see “Fisher B.L. 1999. Improving inventory efficiency: a case study of leaf-litter ant diversity in Madagascar. Ecological Applications”, and chapter 9 of the book linked by A. Wild, to know more about the technique and methods).

      For savannas in general you can also use baits. About winkler, and comparision among different techniques, in such environment you can read and “Parr, C.L. & S.L. Chown. 2001. Inventory and bioindicator sampling: Testing pitfall and Winkler methods with ants in a South African savanna. J. Insect Conserv.” and “Fisher, B. L. and Robertson, H. G. 2002. Comparison and Origin of Forest and Grassland Ant Assemblages in the High Plateau of Madagascar (Hymenoptera: Formicidae). Biotropica”

      1. Baits are not very handy because you have to retrieve them after only a short time, otherwise, all you collect are the dominant species, which have driven off the early arrivers. Besides, baits don’t collect the specialists, eg. dacetines. For my study of the ants of the Arkansas Post National Memorial, we threw out all the bait collections because they didn’t provide any information.

    1. They were real adventures with what would be considered cumbersome gear, vans packed full of you name it.

  3. The big downside to this technique is the loss of association of nestmates as a nest series and of inquilines and their hosts. This loss of information will then have to rectified by more targeted sampling of ant nests. Our sampling efforts strive to document and preserve this association by first searching for nests, collecting those separately, then sifting the left-over leaf litter.

    1. All kinds of skewed sampling effects and “lost data” come under the general category of “Observer Effects” and the like, and are pretty much unavoidable. We use a large number of techniques to try to minimize these that come under the category of “statistically valid” sampling methods. More sophisticated versions of these winkler techniques could always be used by stratified sub-sampling by ‘first searching for nests” in the larger sample quadrat.

      But there are always problems and as Clint Eastwood always said “A man’s got to know his limitations”.

Leave a Reply