'Unethical' flipper tags are damaging to penguins
The standard way of tagging penguins for science - putting bands around their flippers - affects their survival and reproduction, a study has found.
French researchers, reporting their work in the journal Nature, found king penguins had 40% fewer chicks if they were banded, and lived shorter lives.
They say continuing to use the tags would in most situations be unethical.
Flipper bands have been used for decades to identify individual penguins so they can be tracked on land and sea.
They allow for easy visual identification of individual birds from a distance.
Some studies down the years had suggested they harmed the birds - for example, by creating extra drag when they swam, or by reflecting sunlight in a way that could attract predators.
But others had suggested there was no problem.
"There was a debate about whether bands have an effect or not - and you could find studies and some would say 'yes' and some would say 'no'," said Claire Saraux from the University of Strasbourg and the French Centre National de la Recherche Scientifique (CNRS).
"So our idea was to try to make sure - instead of doing one-year studies, to try to find out what's going on over 10 years," she told BBC News.
The French researchers followed a group of 100 king penguins in a colony on Possession Island off the Antarctic coast.
Half were tagged with flipper bands, while the other half had tiny transponders implanted under their skin.
After a decade of observation, 18 of the transponder group were alive - but only 10 of the banded birds.
They arrived at the island for breeding significantly later in the Antarctic spring, needed to make longer foraging trips for food, and overall reared far fewer chicks.
"This study is far and away the longest and most systematic that's been done - it eclipses everything else," said Rory Wilson from the UK's Swansea University.
"It's conclusive - it's going to be very hard for anyone to argue against it."
Professor Wilson, who was not involved in the current research project, tagged his first penguin more than 30 years ago.
Changing climate for research
In recent years, studying penguins around the Antarctic shores has been given a new impetus by concerns over impacts of climate change on the marine ecosytem.
In some parts of the region, krill - the tiny animals near the base of the food chain - are in decline, with potential effects on everything from birds to fish to marine mammals.
But in the light of this study, use of the principal penguin research tool may become unethical.
"I would say no [it is not ethical]," said Ms Saraux.
"The exception would be using them only on land, and that probably won't be a problem so long as you take them off the birds before they go to sea - and that could still be useful, because I can tell you that when you go into a colony of 50,000 penguins to find yours, it's not easy.
"But there are a lot of groups that are still banding [in sea-based research], and I'm pretty sure it's going to be controversial - some may want to continue with other species of penguin, but I'm pretty sure the effect is going to be the same for other species too."
The logic behind this conclusion is that if the bands increase drag in the water - which has been seen in captive Adelie penguins - that is something that should affect all species.
The Strasbourg group is one that has adopted implantable transponders as an alternative tool.
But unlike flipper bands, they cannot be read from a distance - the birds must come into close proximity with the antenna that "reads" the bird's identity.
In the meantime, scientists may have to go back to research performed using flipper bands and ask whether the results still stand, or whether they were distorted by the very tools used in the research.
"If you compared population trends in chinstrap penguins in different areas, say, and both groups were banded, and one does better under certain ice conditions, that wouldn't necessarily be invalidated," said Rory Wilson.
"But if you've tried to measure the mortality rate and say 'it's down to over-fishing' or something, and you haven't considered bands as contributing to mortality, then you'd need to re-assess your data."