upon British Ethnology. 193
that, though the five groups into which he divides the Finno-
Tatar family differ a good deal in their physical characteristics,
they are all alike brachycephalic. There is thus a strong
probability that our invaders of the Bronze Age were men
from what is now Denmark, but of Finnish and not Teutonic
affinities.
Before treating of the Celtic peoples who succeeded the
broad-headed man of the Bronze Age, it seems to me that a
few words about the makers of the great stone circles and the
builders of the lake dwellings should not be omitted. Of the
great stone circles the most advanced, as regards construction,
is Stonehenge, and for this reason we may infer that it is the
latest in date. Sir John Lubbock notes that it is the centre
of a very large number of barrows, these barrows dating
evidently from the Bronze Age. We may therefore conclude
that Stonehenge itself probably dates from the Bronze Age ;
also that the ruder stone circles, such as that at Abury, near
Marlborough, that near Keswick, and "Long Meg and her
daughters," in Cumberland, are most likely of still greater
antiquity.
As regards lake-dwellings in the British Isles, it has been
found that while many of our crannoges were originally formed
in Neolithic times, there is also abundant evidence, in many
cases, to show that they were still inhabited in Post-Roman
times. Both stone circles and lake-dwellings are found in
countries so remote from each other, and so diversely peopled,
that there can be little doubt that they originated in-
dependently in many lands. In Britain both stone circles
and lake-dwellings appear to have existed long before we
have any evidence of a Celtic immigration.
Leaving the pre-Celtic peoples of the long and round
barrows, we come to the Celtic races, who must have been
settled here in considerable numbers long before the Roman
occupation, since it is not known that any pre-Celtic languages
were spoken in Roman Britain. To the Romans the Celts
were a tall fair race. As regards the typical shape of the
Celtic skull, it may be said to have been dolicho-cephalic.
The late Prof. Rolleston remarks ihat "many authorities