Cat and mouse and langskip

 A turning point in my life was when I dumped a tedious final year student project squinting down a microscope counting rye-grass chromosomes . . . in favour of coursing across the country testing the historical migration hypothesis. The idea was that

• domestic cats don't swim well
• their travel across the oceans is greatly facilitated by hitching a ride on ships
• each voyage starts somewhere . . . and finished elsewhere
• voyaging cats are a random sample of the cats in the starting port
• IF the destination has no [domestic] cats [like all the New World including the Atlantic Islands]
• THEN the gene frequencies there will resemble those in the home port
• this is an example of founder effect: it will be hard to subsequent arrivals to shift the gene frequencies of an established population

It seemed to work: I spent some time tooling around the Azores in the 1980s looking for domestic cats and noting their coat colours. We'd shown that the cats on St Pierre et Miquelon, the French department South of Newfoundland, were exactly the same profile as those in Bordeaux and the Dordogne. With some vigorous arm waving, a case could be made that NYC [Nieuw Nederland from 1614 until 1667] cats resembled those of Amsterdam and Rotterdam and were different from New England. The cats of the Azorean islands looked like each other but different from mainland Portuguese cats and missed the characteristic "Viking Profile" of higher-than-expected pure white and orange/tortie cats. Such cats are found across Shetland and Orkney and the first research grant I ever landed [£200!] was to classify the cats of Donegal for evidence of Viking settlement there. I wrote about this decade of travel before.

Why would I specify that the cats of the Azores didn't look like the cats of Trondheim or Københaven,
Q. why ever would you expect them to?
A. Because hot fresh evidence has appeared to suggest that, as with Greenland and Labrador, Nordic peoples discovered, explored and settled the mid-Atlantic islands of Madeira and the Azores hundreds of years before the Portuguese rocked up in the early 1400s.

This was flagged by a really interesting post of MetaFilter. There are links to the original papers and a really engaging 30 minute interview with genetical ecologist Jeremy Searle. The data is two-fold: a) a 2015 paper by Searle who sequenced the mitochondrial genome - ~14,000 bases - of a few hundred house mice from Madeira and several of the Azorean island b) lake bottom mud-cores which found 1) loss of tree pollen; 2) gain of rye-grass Lolium spp. pollen and 3) a layer of 5-beta-stigmastanol, a steroid molecule characteristic of sheep and cow shit . . . in cores dated 700-1200 CE at least 200 years before the advent of Diogo de Silves in 1427. Searle, who has sampled mouse DNA from all over Europe, pegs Denmark as the origin of Madeira's mice and places further North for the Azores. These data are much more extensive, robust and repeatable than our cat coat colour genetics but our data from the 1980s is a sort of proof of principal.

[who knew that the Spanish for longship is drakkar? so cool]