as worked out in much detail a scheme
of chronology for this period, based upon the Egyptian dating of
Professor Eduard Meyer, places the finds from El Argar at from 2500 to
2360 B.C.[11] Allowing, therefore, some margin on the later side, we
should probably be fairly safe in placing the period when the halberds
were in use in Ireland at the end of the third and beginning of the
second millennium B.C. We must remember that the whole of the Irish
Bronze Age has to be fitted in after the copper period; and if we are
to allow sufficient room for the several periods and their approximate
correspondence with the periods of the Continental chronology, it is
not easy to see how this dating can be much reduced. It may be noted
that Montelius in his recent scheme of Bronze Age chronology for the
British Islands, treats the halberds as bronze, and places them in his
second period (first period of the true Bronze Age) dated from the
beginning of the second millennium to the seventeenth century B.C.[12]

[11] Prehistorische Zeitschrift, vol. i, 1909, p. 138.

[12] Archaeologia, vol. lxi, p. 162, and pl. xi, fig. 43.

CHAPTER III

FIRST AND LATER PERIODS OF THE BRONZE AGE

Even during the copper period an evolution can be traced in the celt.
The cutting-edge has been expanded; and the thickest part of the celt
has been moved up from just above the cutting-edge to the centre.
Until, however, we get into the Bronze Age, there has been no trace of
a stop-ridge. When we get into the true Bronze Age, we find a complete
and probably fairly rapid evolution of type from the flat celt to the
final socketed form. Analyses of Irish celts on a large scale have not
been made; but such analyses as have been done do not indicate an
experimental stage of small additions of tin, but rather show that
the bronze from the first contained a fairly large proportion of tin.
Where the tin came from is at present uncertain. The illustrations
will make the evolution of the celt clear. The first step was the
broadening of the cutting-edge, and moving the thickest part up to the
centre of the blade; the next step was hammering the sides to make
flanges to grip the handle more securely; a stop-ridge was then added
to prevent the handle slipping down over the blade; and the latter
forms are reached by increasing the flanges and broadening the
stop-ridge; in its last forms the wings are increased at the expense
of the stop-ridge; and the final sock