Wasser and his colleagues extracted DNA from elephant droppings and skin biopsy samples collected from numerous locations in 16 African nations. They used that information to build a DNA-based reference map to assign tusk origin. They noted genetic differences in populations from one location to another, and used a statistical method to extrapolate genetic signatures to fill in gaps between sampled populations.

Matthew Stephens, a UW associate professor of statistics, developed a model allowing the researchers to build genetic profiles for elephant populations from which they do not have genetic samples. The model is weighted toward genetic information obtained from populations nearest those for which information is unavailable.

The method allows a DNA sample to be assigned to a fairly specific location, with a relatively high confidence that the assignment is correct. The study indicated that 50 percent of the samples tested were accurately located within 300 miles and 80 percent were accurate to within less than 600 miles. Accuracy was much greater among forest populations, which are more clearly defined because of terrain.

The new method allows for speedy determination of where a particular ivory sample came from, Wasser said. That is important because there is mounting pressure to lift the 1989 ivory trade ban enacted under the Convention on International Trade in Endangered Species. But many experts believe any legalization of ivory trade will only increase poaching. The new sampling method can help determine quickly whether that is true in time for exemptions to be altered before elephant populations suffer catastrophic damage, he said.

Two years ago, five African nations sought, and three received, an exemption from the ban so that they could conduct one-time ivory sales. Now nu
'"/>

Contact: Vince Stricherz
vinces@u.washington.edu
206-543-2580
University of Washington
27-Sep-2004