A new and reliable field mark to distinguish African and Eurasian Collared Dove?

Last autumn I went for a vacation to the beautiful island of Tenerife, together with my non-birding girlfriend. Birding was not a primary activity, still I managed to see all my lifer targets. Or at least, so I thought: Upon returning home, I found out that I had strung African Collared Dove Streptopelia roseogrisea (->ACD).

Prior to my vacation I hadn't given much thought to the topic. The Canary Islands are supposedly a stronghold for the domesticated form of ACD (Barbary Dove S. risoria) and in many trip reports birders seemingly found the species with ease. 

Arriving at Puerto de la Cruz, Tenerife, I soon noticed that things might be more complicated than I had thought. All Collared Doves I could find looked just like the Eurasian species, none would vocalize and while trying to further read up on the topic I only found rather vague identification criteria.

HBW Alive describes the species like this:

26 cm; 130–166 g. Upperparts pale greyish fawn, outermost secondary-coverts and primary-coverts blue-grey, primaries black; two central rectrices greyish drab; head, neck and breast mauve-pink merging into white on chin and belly; black half-collar edged with white; underwing-coverts white; outermost rectrices grey above with white tips but black below and white-tipped; iris red; orbital skin pale grey; bill black; legs purplish red. Sexes alike. Juveniles paler and duller, contour feathers with paler edges and legs dull greyish; iris colour changes from yellow through orange to red of adult. Race arabica usually darker with pale grey underwing-coverts.

This is the description of the wild form in the Collins Guide:

Very like Collared Dove, but a bit smaller and has slightly shorter tail. Plumage is a shade paler than Collared's, with whiter belly, but flight-feathers and uppertail on the other hand are darker, giving slightly greater contrast. The multisyllabic song is recognised by the first syllable being higher and drawn out, followed by hint of a pause and then a falling section with rolling r-sound 'kaaw, kurroo-ooh'.

For the domesticated form S. risoria (or S. roseogrisea forma domestica), which is the one that should occur on Tenerife, the description reads:

Very like Collared Dove but smaller, has slightly shorter tail, paler buffy grey-white plumage, and is almost white on belly (Collared Dove light grey). Black neck-side marking is often broader, more rounded, not narrow cross-bar. Is also a touch smaller and paler than the wild ancestral form. Call a pleasing 'ko k'rooh', repeated a few times.

On feederwatch.org the field marks are summarized like this:

Pale tan or grayish primaries that do not contrast obviously with the upperparts
11”, slightly smaller but heavier than Mourning Dove
Relatively pale undertail coverts
Inner portion of tail shows only very limited black, giving the bird a pale appearance underneath
Generally quite pale overall, sometimes almost ghostly
Song is a hollow, rolling, two-part cooing, which contrasts with the Eurasian’s coarse, rapidly delivered, three-part cooing

Basically, what all descriptions state, is that (domestic) ACD is smaller and paler than its Eurasian counterpart. Most of the field marks are very weak. Apart from the song, the only 'hard' field mark I extract from these descriptions was the white colouration of the undertail coverts.

This is what I then started looking for in the birds on Tenerife, and after several days I finally found two birds in the village of Garachico, displaying the feature.

This bird has pure white undertail coverts and limited black in its tail. The primaries have a bluish-grey hue to them. Any other feature seems difficult to judge. Garachico, Tenerife, 17.09.17

A Eurasian Collared Dove for comparison. Note the grey undertail coverts and darker primaries. Valencia, Spain, 21.02.16

The second contender for African Collared Dove I found. Again the undertail coverts are very pale. Garachico, Tenerife, 17.09.17

Compare the undertail covert coloration of an obvious (?) Eurasian Collared Dove (below) with the candidate for African Collared Dove (top). Seems striking, doesn't it? Garachico, Tenerife, 17.09.17

I could've just left it like this. ACD would now be on my life list and I wouldn't have given the topic much further thought. Unfortunately, I stumbled upon an ID request in the german Club300 forum: A very pale, almost ghostly looking Dove was being debated. Consensus was, this is a Barbary Dove until hybrids expert Jörn Lehmhus chimed in. He was not convinced, and he introduced a field mark that I had never heard or read of before: 

In ACD the black pigmentation in the outer web of the (outmost ?) tail feathers is absent or strongly reduced. It doesn't project past the black pigmentation on the inner web.

He used these two pictures to illustrate his point: 

Eurasian Collared Dove

Barbary Dove (Wild type)

Here's the whole french article about this feature, including a hybrid!

Looking back at my own pictures was disappointing. Both birds, which I had believed to be textbook examples of ACD, just weren't. Most likely, they are hybrids, as both Eurasian and African Collared Dove colonized the Canary Islands in the late 1980s and are known to hybridize; one species was certainly human introduced (Barbary Dove), while the other one possibly reached the islands on its own wings. Interestingly, the tail pattern in the 'obvious' Eurasian Collared Dove above is closer to that of ACD, than the pattern of my candidate bird above it. Could it be a hybrid, too? It would be very interesting to get some feedback concerning sightings of pure ACD and hybrids!

To check, if this field mark really applies to all Collared Doves, I have since checked dozens of Eurasian Collared Doves here in Germany. They all show the projection of black pigmentation on the outer web. Digging through pictures of ACD on the internet, there are only few that show the underside of the tail, but many of these fit the hypothesis. However, it seems there are also a lot of misidentifications (which isn't surprising at all) and/or hybrid pictures.
Finally, I visited the local natural museum's collection. Here is one of my pictures; you can probably guess by now, how the specimens were labelled.

The tail patterns of three specimens of Collared Doves found in the 'Landesmuseum für Natur und Mensch Oldenburg'. In the top bird there is some dark pigmentation on the outer web, which very slightly projects past the pigmentation on the inner web. Still the difference to the Eurasian CD is obvious. From top to bottom: Barbary, Barbary and Eurasian Collared Dove. Oldenburg, 09.01.18

Here are some pictures that show wild African Doves (ssp. arabica) from southern Arabia.
Here is another bird from Tenerife, which is probably a hybrid too

So far, I have not seen any mention of this useful and unambiguous field mark anywhere else (which is why I have decided to write this blog post). Please let me know, if there's anything I have missed, be it publications concerning this topic or further useful field marks. Feel free to share your own observations in the comments.

Finally, I would like to especially thank 

• Jörn Lehmhus, for sharing this great field mark and commenting on the first draft of this post.
• The 'Landesmuseum für Natur und Mensch Oldenburg', where I was allowed to take pictures of the collection
• My girlfriend, for always being patient with me, even when I'm looking at the arses of boring pigeons during a romantic couple holiday