ELLSWORTH—With the formal announcement last week that federal agencies had proposed Atlantic salmon in some Maine rivers for listing as an endangered species, debate over exactly what they are trying to protect has reached a boiling point.

As a species, Atlantic salmon are far from endangered. Wild fish exist in abundance in some parts of Europe and millions are raised commercially for aquaculture. In Maine’s rivers, however—and elsewhere in New England—wild stocks have dwindled to the point where some people on both sides of the debate question whether it will even be possible to save them.

The Endangered Species Act offers protection for "distinct population segments" as well as whole species. To qualify, however, a distinct population segment must be isolated from other populations with which it could breed and must be "evolutionarily significant." The act specifies, however, that it is not intended to protect a species which is rare in one geographic area but plentiful in another, as long as the two groups are not significantly different genetically. Those concepts are not well-defined in the law, and that is the crux of the argument over the listing proposal.

Governor Angus King said in a recent interview that he does not believe the salmon in the eight Downeast rivers proposed for listing—the Dennys, East Machias, Machias, Pleasant, Narraguagus, Ducktrap and Sheepscot Rivers and Cove Brook—are what Congress meant by a distinct population segment.

"There are now at this minute 6 million Atlantic salmon in cages off the coast of Maine," King said, "and it is hard to argue that this is an endangered species." He said he would consider suing federal authorities to establish that fact if the listing goes forward.

King’s position has the support of some geneticists.

Jeff Rodzen, who studied the question while an undergraduate at University of Maine at Machias and a volunteer at the Pleasant River Salmon hatchery, is now pursuing a Ph.D. in animal genetics at the University of California at Davis. He said the scientific literature indicates genetic mixing between the Maine rivers, as a result of both natural straying—some scientists estimate that as many as 20 percent of the fish spawned in a river could return to other nearby rivers—and 100 years of government stocking programs.

The federal stocking program, begun at Craig Brook in 1883, used fish from Canadian and New England Rivers and even Coho salmon from the West Coast during its early days. The program also introduced fish from the Penobscot, Union, Narraguagus and Machias rivers into other Maine rivers, and many of these stocks were already composites of other stocks.

Supporters of the listing do not contest the history of mixed breeding, but contend that enough genetic differences remain to make the stocks in each river worth protecting from contamination.

U.S. Geological Survey biologist Tim King—no relation to the governor—recently completed a genetic study comparing about 1,000 wild North American salmon, 380 European salmon and 250 farmed salmon. He compared the frequency of variations of certain genes in the different stocks as a way to determine how different they were, genetically.

He found major differences between European and North American populations, but concluded that the local differences, while "shallower," were also significant. He found many gene variants which were present in only one Maine river, and was able to identify fish from some rivers with considerable accuracy using only their genetic makeup as a guide.

He was struck by the "highly divergent" populations in different branches of the Penobscot and Kennebec rivers and was concerned by the discovery of European gene variations in some Maine fish, which he took as evidence of interbreeding with escaped aquaculture fish.

Rodzen, however, argued that there may be other explanations for Tim King’s findings. By the method used, he said, one cannot distinguish genetic variations which have been inherited from one group by another from those which occurred as separate mutations in each group. Since there are a finite number of alleles, as these variations are called, Rodzen said the two scenarios are equally likely. In addition, he said, the sample size was small enough that the "European" alleles might have actually existed in other Maine stocks but been so rare that they were not noticed in the study.

In small breeding pools—and the river-specific stocks are becoming very small indeed—Rodzen said the frequency with which a certain genetic trait occurs can change very rapidly. This, he said, might explain the level of genetic divergence Tim King observed. In other words, instead of being the last remnants of separate populations, the salmon in the eight targeted Maine rivers may actually being diverging because the artificially small gene pool is causing inbreeding. He said statistical analyses showed that most of the differences observed by geneticists were within rivers, rather than between one river and another.