Having finally gotten round to reading the Birding World Article that John mentions (I was waiting to do so before replying to John’s post) I have to say that I’m not that impressed by the article! I think it is a bit misleading in that the paper does not describe what I would call ‘typical’ middendorffi and hugely overstates the bill structure differences between middendorffi and serrirostris.
I have also re-read Ruokonen et al. 2008 (Taxonomy of the bean goose – pink-footed goose) which is an excellent and helpful paper. That paper concludes that middendorffi is a valid species (Middendorff’s Goose) but fabalis, rossicus and serrirostris are subspecies of a second species – Bean Goose (note that they propose a treatment which is not biome related); they also fail to recognise johanseni. If anyone would like a PDF of this paper, please send me a PM.
On the whole however, the two papers are in broad agreement - that ‘bean’ geese are usually separable as follows based on a combination of size, structure, and bill structure:
• middendorrfi
• serriostris/rossicus
• fabalis
Both conclude that on present knowledge not all individuals can be identified based on external morphology alone.
Applying this to the Mai Po birds I think three individuals are easy to identify, the two large birds with long slender necks and long sloping bills and gently sloping head profile (recalling Whooper Swan) – both these are to my eyes typical middendorffi – and the obviously smaller bird with the short thick neck, shorter looking bill, obvious forecrown angle and rather angular head which is a typical serrirostris.
The fourth bird is far more problematic having the size of middendorffi but the proportions and neck length/thickness of serrirostris. In the absence of any evidence that there is any marked size variation within serrirostris, I can only think this is an extreme middendorrfi. Curiously, in the Birding World paper there are photos of a similar bird, and it seems that the paper uses this rather odd bird for the basis of the discussion on what middendorffi looks like.
I see no reason why both cannot occur together, other than mathematical likelihood! I have seen both taxa in mixed flocks on migration in Xinjiang (in arable farmland for what it’s worth), and it has been known for a long time that fabalis and rossicus have overlapping wintering ranges. Add to the fact that ‘only’ three bean geese were originally located, it seems likely that the serrirostris joined later. I also noticed last week that one of the middendorffi was regularly driving the serrirostris away when foraging.
