Good news from the Galapagos! After nearly 60 years, the Española giant tortoise, once feared extinct, has a viable population again.

A long-running captive breeding program has restored the population of Española tortoises from 14 to 2,000, enough to sustain itself on its isolated island home. It’s a remarkable achievement for an iconic animal, but is captive breeding really a viable approach to conservation? How well does it work?

Captive breeding has been controversial ever since conservationists first started talking about it. A 1996 article by Noel Snyder and colleagues in Conservation Biology sums up many of the issues surrounding the practice. One obvious challenge Snyder raises is that not all organisms readily breed in captivity, and others can’t even be kept in captivity at all (whales come to mind). Assuming that a viable captive population can be established, successfully reintroducing a species to the wild is not guaranteed. In fact, Snyder describes only 16 successful reintroductions out of 145 attempts. The reasons for failure are variable, but at a minimum there needs to be adequate protected habitat where a reintroduced species can survive, which is hardly a given. Additionally, it is very difficult to simulate the natural environment in captivity and many reintroducees have a hard time coping after release. This is particularly true for species that have a lot of parental care—the nurturing of a parent is especially difficult to create artificially.

So what about situations, like the tortoise, where conditions are ripe for a successful reintroduction? Their isolated island can be protected with proper resources, but even then there are difficulties. Rebuilding a population with just 14 individuals results in a far smaller gene pool than is healthy. To make matters worse, post-release studies indicate that tortoises do not mate equally—a select few enjoy most of the mating—further depressing the genetic viability of the population and threatening its long-term prospects. Careful monitoring and manipulation of the ongoing breeding program can help, but it is difficult to declare absolutely that the program is a “success” if it requires ongoing maintenance.

If a species is going extinct, do genetics or habitat concerns even matter? After all, its choice is either less-than perfect circumstances or oblivion. That’s why the real issue with captive breeding is the opportunity cost, as the resources used for these programs might be better spent preserving entire ecosystems. It’s also why Snyder and others argue that captive breeding is the last resort, the “Hail Mary pass” of conservation. Just as there is no guarantee that captive breeding will work, there is no guarantee that other approaches will be successful either. If there were a sure-fire method for species conservation, there would be no need for it. So, when all else is lost, why not give captive breeding a try? For now, it’s working for the tortoise.