genetic identity of chiffchaffs trapped in The Netherlands 2009-2011 (2013)

Chiffchaff (Phylloscopus collybita; left) and Siberian Chiffchaff (Ph. (collybita) tristis), Meijendel, Wassenaar, 31 October 2009 (Vincent van der Spek/ Vrs Meijendel)

DISCLAIMER: On this page, I made a summary of the article Peter de Knijff and I wrote for Dutch Birding about the genetic identity of "grey" chiffchaffs trapped in The Netherlands between 2009-2011. Peter's student Johannes Fischer did the mtDNA analyses. 

The full article can be downloaded here (pdf).  

Much has been said and written about the subject, also after our publication (which caused quite some discussion, especially in the UK). This was our first contribution to the ongoing discussion about the occurrence of tristis (and after our publication: the occurrence of abietinus) in NW Europe. I've added a few editorial notes in the summary.

Editorial note 2017: a follow-up survey with a (mostly) random sample method was conducted between 2013 and 2016, in order to get a better understanding of the occurrence of abietinus. Over 400 birds were sampled. The intriguing results (teaser; but it's true!) will be published elsewhere (scheduled for 2018), but a summary will be posted on this site.   


There are three chiffchaff taxa on the Dutch list: the Common (collybita), Scandinavian (abietinus) and Siberian (tristis). Collybita is a very common breeding bird and migrant that winters in small numbers. Abietinus is usually regarded a migrant and wintering bird, though there are no detailed publications supporting this assumption. Tristis is (editorial note 2017: was!) regarded a vagrant, with 39 accepted records up till and including 2011, mostly of singing or calling birds in late autumn and winter. But how to treat silent birds, like most birds trapped on ringing sites?

Nominate collybita has the largest amount of yellow and green in the plumage, tristis the least. Abietinus is often regarded intermediate, but in reality they hardly differ from collybita. On average this should be the largest taxon. A classic tristis is brownish on the upperparts, but many birds give a more grey impression. They can show more yellow or greenish parts on the head, upper- and/or underparts than previously thought (editorial note: though the final word about introgression hasn't been said).  The Dutch rarity committee CDNA however only accepted records of silent birds that show no traces of yellow and/ or green (editorial note: after the publication of this paper, tristis no longer has to be submitted). By emphasizing the importance of yellow and green tones, many ringers started to believe - in error - that birds showing these colours diagnostically ruled out tristis and therefore were abietinus.  

Some ringers disagreed with this widely accepted point of view and wondered if DNA analyses could clarify the situation. Therefore we developed a pragmatic protocol.  Between 2009-11 feathers were collected on five ringing sites. 41 birds were sampled, mainly of birds that were suspected not to be collybita (though some were sampled as a reference).  The ringers identified these birds in the field, based on their current knowledge and perception. DNA was isolated from the sampled feathers and a fragment of 939 basepairs of the mt cytochrome B-gen was sequenced. The sequences were compared with 17 sequences from the breeding areas of all three taxa available in GenBank (1 collybita, 8 abietinus and 8 tristis). 


30 out of the 41 Dutch samples matched the sequences of tristis, 11 of collybita and not a single one matched abietinus. When the genetics were compared with the field identification of the ringers, a few things stand out. All sampled birds identified as collybita (9), also showed the mtDNA of collybita. Five birds identified at tristis, were tristis based on mtDNA. However, all birds (n=23) identified as abietinus or abietnius/tristis, appeared to be tristis based on mtDNA.  Two birds with large wings (67 mm), according to available literature (editorial note: at the time) out of range for both collybita and tristis, appeared to be collybita based on mtDNA. It raises the question how useful measurements are in this matter.

This small scale survey on a limited number of ringing sites shows that between 2009 and 2011 no less than 31 tristis were trapped. Even taken into account that 2010 was an extraordinary year for "a-typical" chiffchaffs (according to all ringers involved), these results show that tristis might be scarce, but not a vagrant. The participating ringers were very capable of recognizing "suspicious" chiffchaffs but, despite their experience, they were unable to distinguish abietinus and tristis in many cases. Useful criteria to identify abietinus are not yet available and its status in The Netherlands is now unclear (editorial note: see note above about the follow-up survey).