In article ,
Alastair McDonald writes:

"John Hall" wrote in message
.. .


So I think, even if you're very cautious, you can rely on the values
from at worst 1772 to be high accurate.

Yes, but how good was the data the the series is based on? It was
not from Met Office approved sites, like that since 1910. Manley,
and later Parker, have done a lot of good work, but you can't
make a silk purse out of a sow's ear.

There were a lot of very careful observers much earlier than 1910, but
of course the questions of the quality of the thermometers and the
exposure of the instrument are important ones. One of the main problems
is that the use of Stevenson screens only became standard towards the
end of the 19th century, before then the less good Glaisher stand being
used. Even earlier, there would have been no stand at all, of course.
But Manley was well aware of the problems, and spent many decades
overcoming them. To see how he did this I recommend reading the papers
to which I have provided the links.

There's also the point that the CET series has been adopted as a
standard by modern climatologists, and they wouldn't have done that were
they not convinced of its validity. Some of them will have investigated
the quality of the series much more thoroughly than it's practical for
you or I to do.
--
John Hall "He crams with cans of poisoned meat
The subjects of the King,
And when they die by thousands G.K.Chesterton:
Why, he laughs like anything." from "Song Against Grocers"