Sunday, 28 September 2014

Genomes show why Hooded and Carrion Crows are different and stay different…and are they different species?

Over the years, as an outsider to the field, I have read about taxonomy and systematics and I have read about different species concepts and the attempts to define what species are. It has struck me that the whole discussion has been concerned with defining the species as a group as opposed to considering the individuals that make up that group. I just have the feeling that defining the group without reference to the individual came from an age when many if not most biologists argued intentionally or unintentionally for group selection with, for example, adaptations ‘important for the survival of the species’ on many lips. As the realisation grew that group selection was the result of flawed reasoning and that the individual is the unit of selection, nobody to me, or at least nobody that I could find, was writing about the species as a collection of individuals subject to natural selection (although this view does seem implicit in the Grants’ work on Galapagos finches). When I did consider the individual I realised that for many organisms, most vertebrates, for example, a species could be defined as those individuals with which an individual could breed to its selective advantage or to put it the other way round, with which an individual could breed without selective disadvantage. Therefore, to me the group—the species— is defined by the advantage it affords its individual members in terms of fitness.

My amateur definition includes those species identified as being completely reproductively isolated (i.e. those that conform to the Biological Species Concept as originally propounded) as well as those, like the Galapagos finches, in which there is gene flow between obviously different species or in which there is a hybrid zone between geographically adjacent forms that differ morphologically and/or behaviourally.  I therefore bridle when I hear only the former referred to as ‘good’ species by those wedded to the Biological Species Concept, with the others left to a ‘judgement call’ as to their status; ‘easily discernible’ should replace ‘good’.

To reiterate, my definition means that complete reproductive isolation (i.e. no gene flow) is not necessary in order to define a species. This definition provides for breeding between neighbouring closely related forms with a hybrid zone between them. If breeding with the other neighbouring form leads to the hybrid offspring having a selective disadvantage then the two forms are retained as distinct species even though introgression occurs.

It is with background that I read a recent paper in Science with great interest. Not only does it throw light on the species problem, it also occurs on my doorstep. The paper, The genomic landscape underlying phenotypic integrity in the face of gene flow in crows, is on the genetic differences between the Hooded Crow and Carrion Crow. It is also interesting not only in its own right but also for the comments it evoked from those trying to decide whether the two crows should be regarded as one or two species.

We live fairly near the hybrid zone between the Hooded and Carrion crows in the West of Scotland. A hybrid lived for several years in the surrounding gardens and school playing field even though it never looked the healthiest of birds and had, what the vets would described as, an ‘unthrifty’ appearance. As well as hybrids, Hooded Crows are reported occasionally since the hybrid zone is narrow at this point and hoodies occur on the islands of Arran and Ailsa Craig where they are said to prefer the higher ground compared with the Carrion Crow.

Carrion Crow

Hooded Crow

Discussion has centred on whether the new findings offer support to the arguments in favour of considering the two forms as two species, or as subspecies of one species. Over the years they have flitted between being regarded as two species or one. Linnaeus originally described them as two species but throughout the later part of the 20th century, the one species view prevailed with the Hooded Crow listed as a subspecies, Corvus corone cornix, alongside the Carrion Crow, C. c. corone. Then, in 2002 the British Ornithologists’ Union recommended that the two forms should be recognised as separate species (Knox et al 2002). The reasons for this (non-random mating and reduced fitness of the hybrids) and how they fit the Evolutionary Species Concept are explained by Parkin et al  (2003) and by Parkin (2003). Parkin (2003) concludes:

In the case of the crows, I believe that the Hooded and Carrion lineages maintain their separate identities through time and space, for hybrids are at a selective disadvantage. They should be regarded as separate species.

The work that has caused the flurry of interest was done on mainland Europe by Swedish, German and Spanish co-authors (Poelstra et al 2014). They compared Carrion Crows from Spain and Germany with Hooded Crows from Poland and Sweden (as in northern Britain there is a hybrid zone where the two forms meet). The differences detected in the genome and in gene expression have been described and discussed well elsewhere, for example, in a commentary in Science by Peter de Knijff. I will only state here some of the main conclusions.

Differences between the two forms were very small. Only 83 out of 8.4 million DNA positions were found to be fixed and therefore diagnostic of the two forms. As one would expect, the genes concerned were mainly associated with plumage colour. Of the 83 fixed differences 81 were found in a small region of chromosome 18.

de Knijff summed up the conclusions neatly: Poelstra et al. present a unique case of speciation whereby, despite substantial gene flow (especially from hooded crows into German carrion crows), phenotypic divergence likely caused by assortative mating and sexual selection is maintained by genetic variation in less than 1% of the genome. Invoking Darwin on species concepts he concludes: Obviously, forcing complex patterns of biological variation into a single hierarchically structured archive is a purely anthropogenic project. No matter how hard we try, there cannot be a robust, all-inclusive, objective species concept.

While the latter sentence is probably right, I am not sure that the concept of a species is entirely anthropogenic. Defined from the fitness of individuals, an assemblage does have some sort of real existence at an ecological level for example. What I am now convinced is that the Biological Species Concept has had its day in its classical mid-20th century form. It is and perhaps always was too dogmatic; its underlying hypothesis is simplistic and has been undermined by further observations and experiments. Evolution is more complex than that…and continues.


de Knijff P. 2014. Carrion and hooded crows defeat Linnaeus’s curse. Science 344 1345

Knox AG, Collinson M, Helbig AJ, Parkin DT, Sangster G. 2002. Taxonomic recommendations for British birds. Ibis 144 707–710

Poelstra JW, Vijay N, Bossu CM, Lantz H, Ryll B, Müller I, Baglione V, Unneberg V, Wikelski M, Grabherr MG, Wolf JBW. 2014. The genomic landscape underlying phenotypic integrity in the face of gene flow in crows. Science (20 June 2014) 344 1410-1414

Parkin DT, Collinson M, Helbig AJ, Knox AG, Sangster G. 2003. The taxonomic status of carrion and hooded crows. British Birds 96 274–290

Parkin DT. 2003. Birding and DNA: species for the new millennium. Bird Study 50 223-242