Feather sampling provides an unreliable source of DNA that may well have significant long-term impacts: a reply to Katzner et al.

Abstract

We were pleased to see that our recent point-of-view article (McDonald and Griffi th 2011) has lead to further discussion on this important issue (Katzner et al. 2012). One of our primary goals was to encourage researchers to consider both the potential impacts of their sampling methodologies upon subjects, and also the level of information that particular methods are likely to provide. Whilst we are therefore pleased that Katzner et al. (2012) have continued the discussion and helped to raise the profi le of important issues surrounding the biological sampling of avian tissues, we are disappointed that they have misrepresented our original message. Th e primary critique raised by Katzner et al. (2012) centres on our reputed suggestion that blood sampling was a ‘ one-size-fi ts-all approach ’ to ornithology. Th is main criticism of our article is somewhat weakened by the fact that nowhere in our article did we either state or imply this to be the case. In fact, the opposite is true, in several places in our article we explicitly suggested that feather sampling may well be the best practise in some situations. Th e clearest statement to this eff ect is on the very fi rst page of our article: ‘ Note that we support the obtainment of feather material, after applying an appropriate degree of ethical rigour, when no other means exists for obtaining these data ’ . At present, there is a lack of data that has specifi cally tested for impacts of feather sampling, be it feather plucking or clipping, which makes it very diffi cult to judge the longterm impacts of these techniques. Again, we implore ornithologists to publish data on any potential impacts (or lack thereof ) of feather sampling at their earliest opportunity. Th e main thrust of our article was two-fold. First, to highlight the relatively poor DNA extraction rates associated with feather versus blood sources. Second, to identify some studies that determined signifi cant negative impacts of feather sampling. Th e latter are typically swept under the carpet as ‘ insignifi cant ’ – perusal of the literature suggests that this is not the case for a range of species (Chai and Dudley 1999, Th ompson et al. 2010). Katzner et al. (2012) raise four points in their article that they claim contradict our views, we address each below. 1. Science methodology should be driven by research questions