It seems that populations of elephants in Africa have been evolving long enough in different habitats that they look somewhat different, and have distinctly different DNA. There is a large branch of biology, taxonomy, charged with identifying and naming separate species. This is not necessarily a simple task. As one group of organisms splits and evolves into two or more, there is a period where the distinction is inherently difficult to categorize. Traditionally, the basis for separating species is whether they successfully interbreed to the point that gene flow is maintained between the populations (Many different species that could interbreed in captivity don't in the wild due to behavioral or other differences). Do these elephant populations interbreed at the edges of their separate habitats? Perhaps the DNA is informative of this, but it was not clear from the article.
One thing that struck me in the article is that it is now considered important to classify the African elephants into two species for political reasons:
It's important to classify the two as different species for the conservation aspects. You would want to develop a separate conservation plan for each one," Roca said. The African elephant is listed as endangered by the U.S. Endangered Species Act, and splitting the population into two different species places the forest elephant in much more dire straits. A little over a fifth of Africa’s 500,000 elephants are forest elephants, and their numbers are dwindling quickly as their habitats disappear and poachers kill them for their ivory tusks, Roca said.
I certainly have no objections to identifying the forest and savanna elephants as different species, and I'm completely on board with efforts to protect both. However, I object to making taxonomy a political tool.