Fertility and viability of hybrids

I occasionally hear people claim that different species can't produce hybrids.  That's clearly absurd but it probably comes from a misrepresentation of another false claim that is much more widely held to be fact.  That is that different species can't produce fertile hybrids, and if they do then that indicates they aren't really different species at all and have been classified incorrectly.  That's also wrong, but I've heard it so often that I suspect it's sometimes taught in schools and textbooks.


Fertility in bird hybrids

It is demonstrably the case that some hybrids can be fertile.  There are hundreds of examples I could give you - let's say Western Gull x Glaucous-winged Gull, two species with an extensive overlap in their ranges and where in the centre of the overlap zone most birds are hybrids and freely and successfully breeding with one another.

a fertile type of hybrid - Western Gull x Glaucous-winged Gull hybrid, Mukilteo, Snohomish County (Washington, USA), 31st January 2011 - copyright Steve Mlodinow
(photo ID: 1453)

It is less easy to prove that some hybrids are necessarily infertile.  However if a hybrid is very common but backcrossed hybrids are unknown, then it seems reasonable to conclude that they are.  One example of a hybrid that is apparently always infertile is Muscovy Duck x Mallard.  Domestic examples of these have been bred together, deliberately and accidentally, and such a hybrid has never (I believe - correct me if I'm wrong) produced any offspring of its own.

an apparently infertile type of hybrid - Muscovy Duck x Mallard hybrid, Rheinauenpark, Bonn (Germany), summer 2004 or 2005 - copyright Joern Lehmhus
(photo ID: 0848)

Another good example is Greylag Goose x Canada Goose.  Such hybrids are very common but I have never seen or heard of any evidence to suggest that they can produce offspring.

another apparently infertile type of hybrid - Greylag Goose x Canada Goose hybrid, Swanton Morley (Norfolk, UK), 8th March 2008 - copyright Dave Appleton
(photo ID: 0052)

The fact that most Muscovy Duck x Mallard hybrids or Greylag Goose x Canada Goose hybrids don't produce offspring doesn't prove that they are necessarily infertile.  It doesn't mean that one day we might find one that does produce offspring.  But with common hybrids like this there is sufficient evidence at least to say that these hybrids are either much less likely to be fertile, or much less fertile.

And here's a point.  It may not be a black and white question of fertile or infertile, but a shades of grey question of how fertile.  Western Gull x Glaucous-winged Gull hybrids are very fertile and Muscovy Duck x Mallard hybrid are very infertile. Are there other hybrids that are a bit fertile, but not very fertile?  Do you know of any examples of hybrids that have shown they can be fertile but appear to have reduced fertility compared to their parent species

In an earlier draft of this article I speculated that Mallard x Northern Pintail hybrids might provide such an example.  They turn up fairly often, especially in NW USA/SW Canada it seems, and they're pretty consistent in appearance most of the time.

Mallard x Northern Pintail hybrid, Sweden, 10th February 2012 - copyright Carl Gunnar Gustavsson
(photo ID: 0925)

If Mallard x Northern Pintail hybrids were as fertile as their parents, I figured we should expect to see a few backcrossed birds now and then, looking much more like one or other parent than the typical F1 (first-generation) hybrid?  I was aware of one or two records that we've wondered about and that do seem quite likely to have been backcrossed hybrids, but not often and not many.  I concluded tentatively that perhaps Mallard x Northern Pintail hybrids are fertile, but not very fertile, or perhaps not all of them are fertile.  Well, Joern Lehmhus has brought to our attention studies of captive birds that have shown that Mallard x Northern Pintail hybrids are in fact very fertile.  This seems to be true of many other ducks in the genus Anas (and also in Aythya).

Most hybrids of very closely related taxa - like where the two taxa are treated as conspecific by some authorities and distinct species by others - seem to be very fertile.  Several records of vagrant Black Brants in Norfolk (UK)'s Dark-bellied Brent Goose flocks have remained with their carrier species to breed and now Dark-bellied Brent Goose x Black Brant hybrids are rather frequent.  Many of these have bred with Dark-bellied Brent Geese and produced young, although once the family ties are broken they are nigh-on impossible to identify.  They are evidently very fertile.

Dark-bellied Brent Goose x Black Brant hybrid with 3 presumed backcrossed young (Dark-bellied Brent Goose x Black Brant) x Dark-bellied Brent Goose hybrids, Burnham Overy (Norfolk, UK), 1st January 2015 - copyright Dave Appleton
(photo ID: 1738)

So it seems a rule might be:

• The more closely related two taxa are, the more likely their hybrid offspring are to be fertile (and the more distantly related they are the more likely their hybrid offspring are to be infertile)

and/or

• The more closely realted two are, the more fertile their hybrid offspring are likely to be (and the more distantly related they are the less fertile their hybrid offspring are likely to be)

A rule of thumb seems to be that hybrids between species that belong to different genera are infertile while hybrids between different species that belong to the same genus may be fertile.  I'm sure that cannot be more than a guide, as a hybrid's fertility cannot depend on how a taxonomist classifies its parents, but can anyone think of any exceptions?  Let us know.

Of course in nature things aren't always simple, so it's likely that hybrids of some closely related species won't be fertile and it might be that some hybrids of more distantly related species are fertile.  It would be good to hear from you if you know of cases that fall outside of expectations.


Finding evidence of hybrid fertility

In captivity it might be possible to do more, but with observations of birds in the wild we can only rely on collecting data from hybrid pairings where the hybrids are clearly different from either parent species.  If you see a lot of hybrids that look completely different from their parent species but consistent with one another then they're probably all first-generation hybrids.  If you see most hybrids are consistent but a minority look much closer to one or other parent species, it may be reasonable to conclude these are second-generation backcrossed hybrids.

For more variable hybrids, or hybrids that are not very dissimilar from one or both parent species it can be much harder to detect backcrosses and thereby prove that the hybrid is fertile.  Sometimes, as with the Dark-bellied Brent Goose x Black Brant hybrid's family shown above, their behaviour might give it away even when the birds' appearance does not.

Joern Lehmhus has also pointed out another factor that may influence our perception of fertillity.  He points out that the display of most Anas species has a series of several elements which are varied between the species.  Several species may even lack some elements.  These behavioral traits are inherited (this is shown in a study of Mallard x Northern Pintail hybrids in Behaviour 27: 259-272 (Sharpe & Johnsgard, 1966).  Therefore a hybrid does often not show the same combination as either of the parent species.  If  a male F1 hybrid is displaying to a female of one of the parents (or a male of one parent species displays to an F1 hybrid female) some elements will be 'wrong' and may reduce the success of reproduction in a hybrid, even if the bird is as healthy and viable as the parent species.


Viability in hybrids

Viability in hybrids (the ability for a hybrid to maintain life) is much harder to gather information about.  At least we can say that those hybrids that have been recorded are, or can be, viable.  But we cannot say that those hybrids we do not observe are not viable - it may simply be that they are rare.

There are some examples that might give us clues:

• In his "Handbook of Avian Hybrids of the World" (Oxford University Press 2006), Eugene M McCarthy's small section on viability/inviability includes a reference to a brood of European Greenfinch x Yellowhammer eggs.  Three eggs were sterile, one died in its shell and the fifth survived to maturity.  McCarthy admits that such data is of limited value for predicting viability but its inclusion presumably infers some relevance.  I would take this as anecdotal evidence that supports the idea that Greenfinch x Yellowhammer hybrids may have low viability, but of course much more data is required before conclusions can be drawn.  There's also a report of a European Goldfinch x Yellowhammer pairing - hybrids reportedly started developing in the eggs but died before hatching.  Both of these examples were published in Cage Birds journal in the early 50s - we're not quite sure how reliable they are.
• Also in McCarthy's book there is a comment about Mandarin hybrids.  Many breeders and researchers have tried and failed to breed Mandarins with other species without any success.  It has been suggested that because Mandarins have a different chromosome count to other ducks they cannot hybridise (the implication being that those few Mandarin hybrids that have been reported are erroneously identified) but McCarthy points to frequent hybrids between different species of Muntjac deer which have different chromosome counts and suggests that a better explanation might be that Mandarin does hybridise but only rarely.  I wonder if it might even be that they do hybridise but have poor viability.  McCarthy cites a case of two captive Mandarin x Laysan Duck hybrids - both lacked eyes on hatching with one surviving at first but dying in juvenile plumage.
• When a Swoose (swan sp. x goose sp. hybrid) was discovered by the Radipole Ringing Group their research found that such hybrids do not normally survive past fledging (Radipole Ringing Group Report 2007).  They didn't cite any references so I'm not sure of the source of that information, but if it was based on good evidence then that would suggest that swan x goose hybrids are not normally very viable.  Exceptions exist though, as the bird they were reporting on was discovered as a cygnet/gosling in 2002 and was still surviving when I saw it in 2010 and thereafter until at least December 2011.

8-year old Mute Swan x domestic goose hybrid, Wool (Dorset, UK), 24th October 2010 - copyright Dave Appleton
 (photo ID 0001)

I would be willing to speculate that hybrids of distantly related species are less likely to be viable than hybrids of closely related species.

Let us know if you have any information or experience that might help us build a clearer picture.


Species mentioned:
Mute Swan Cygnus olor
domestic goose - in this case Anser cygnoides x Anser anser
Greylag Goose Anser anser
Canada Goose Branta canadensis
Dark-bellied Brent Goose Branta (bernicla) bernicla
Black Brant Branta (bernicla) orientalis (formerly nigricans) or Branta (nigricans) orientalis
Muscovy Duck Cairina moschata
Mandarin Aix galericulata
Mallard Anas platyrhynchos
Laysan Duck Anas laysanensis
Northern Pintail Anas acuta
Western Gull Larus occidentalis
Glaucous-winged Gull Larus glaucescens
Yellowhammer Emberiza citrinella
European Greenfinch Chloris chloris (formerly Carduelis chloris)
European Goldfinch Carduelis carduelis