A hunter says DNA tests have confirmed the existence of a Canadian moose in a remote part of Fiordland.

The results are the second since June 2001 to suggest at least one of the animals - introduced from North America in 1910 and last seen more than 30 years ago - is still very much alive.

At his home at Bull Creek, Otago, scientist and renowned moose-hunter Ken Tustin called the "very exciting" findings independent proof of a "subjective" hunch he had pursued since the 1970s.

"Significantly, this means that we have real, scientific proof that an animal people said died out decades ago was actually still alive close to when we collected this sample," Mr Tustin said.

"We know that instead of being dead, that a moose stood on a beach on the northern side of the Wet Jacket Arm, that it fed there, and that it moved on to feed somewhere else. This puts the hunt outside the realm of a hoax, which is what people said when the first hairs were found."

Mr Tustin and his wife, Margie, collected the sample during a regular visit to the thickly wooded and steep area opposite Oke Island in October 2002.

The find was a considerable distance from Shark Cove on the south side of Dusky Sound where, one year earlier, Kelvin and Charlie Harper found hair that DNA tests said came from a moose.

Hamilton-based researchers Deer Improvement sent the Tustins' sample, the already tested Harper hair, and about 40 other samples to a forensic laboratory at Trent University in Ontario, Canada, this year.

The laboratory returned the results late last month and both samples were confirmed as having come from a moose, said Deer Improvement director Peter Gatley.

He had "complete confidence" in the laboratory and its methods, and said he was especially excited that the hair could only have been exposed to the elements for "weeks, rather than years" to give a viable DNA reading.

The DNA tests did not reveal whether the two samples came from different animals, or their sex, Mr Gatley said.

That a couple of dozen samples of suspected moose faeces, collected by Mr Tustin since the early 1990s, were likely to be DNA-tested before the end of the year was also "tantalising and exciting".

"This is not quite the photograph that people who do not believe there are moose there might want, but boy, this really is the next best thing," said Mr Gatley.

Ten Canadian moose were introduced to Fiordland's Supper Cove in 1910, but their population declined steadily under pressure from red deer. The last rumoured sighting was in 1971.

Spurred by finding a moose antler in 1972, and now backed by the NZ Wildlife Trust, Mr Tustin spends about two months a year in the Dusky Sound area documenting evidence of moose occupation.

Moose had eluded cameras in the area but bedding spots and browsing and antler marks suggested up to 20 animals might live there, Mr Tustin said.

A Natural History NZ camera took a blurred image of a suspected moose in 1995.