f some prominent citizen,
and that, in fact, all the subjects on Dipylon vases seem to refer to
deceased persons. He points out that Virgil mentions in the _AEneid_
that games, held in honour of the deceased, commenced with a race of
ships, and that he could hardly have done this if there were no
authority for the practice. The large figures at the stern seem to
point to the bireme of Fig. 74 being about to be used for racing
purposes. The man who is going to step on board is in the act of
taking leave of a woman, who holds away from him a crown, or prize,
for which he may be about to contend. If this view be correct we have,
at once, an explanation of the very peculiar structure of this bireme,
which, with its open sides and small freeboard, could only have been
intended for use in smooth water and, possibly, for racing purposes.

There are several other representations of Greek galleys, or of
fragments of them, in existence. Nearly all have been found on
eighth-century Dipylon vases, but, hitherto, no other specimen has
been found in which all the rowers are seated on an open stage. In the
collection of Dr. Sturge there is a vase of this period, ornamented
with a painting of a bireme, which is as rakish and elegant in
appearance as Fig. 74 is clumsy. It also is propelled by 78, or
perhaps 80, rowers. Those of the lower tier are seated in the body of
the boat, while those of the upper bank on what appears to be a flying
deck connecting the forecastle and poop, and about 3 ft. to 3 ft. 6
in. above the seats of the lower tier.

In the Museum of the Acropolis there are also some fragments of
Dipylon vases, on which are clearly visible portions of biremes. The
rowers of the lower bank are here again, seated in the hull of the
galley and appear to be working their oars in large square portholes,
while the upper row are seated on a flying deck, the space between
which and the gunwale of the hull is partly closed in by what appear
to be patches of awning or light fencing. The portholes above referred
to are in fact merely open intervals between the closed-in spaces.
Similar lengths of fencing may be seen in the representation of a
Phoenician galley (Fig. 7, p. 27).

From the above description it is not difficult to see how the galley,
with two tiers of oars, came to be evolved from the more primitive
unireme. First, a flying deck was added for the accom