166 SOME OBSERVATIONS ON THE 
One has often noticed in old wood cuts, and in most pictures drawn 
by children, an attempt to exhibit two opposite sides of an object, 
without regard for the perspective. Now one way of doing this— 
one sometimes sees it done intentionally in drawings of machinery— 
is to raise the outer side above the other. _ As I take it, in the few 
instances in which we find a second tier of oars, the artist, knowing 
that a spectator would see the oars only of the rowers nearest to 
him, the rowers themselves being partly hidden by the bulwarks, 
while the rowers on the other side, being further from the inter- 
vening bulwarks, would be more conspicuous, wished to bring their 
oars also into view. No doubt this error in the perspective, once 
introduced by the original artist, would be carried still further by 
the copyist, who possibly never saw such a vessel in his life; and 
this too would explain some of the strange comments which are to 
be found in later writers. With regard to the supraposition of the 
rowers, I cannot but think that, especially in very large vessels, 
where each oar was manned by ten or sixteen rowers, it would be 
necessary for the men at the upper extremity of the oar to be placed 
higher than those nearer to the thole pin; otherwise they would 
hardly have been able to reach the end of the oar when it was dipped 
in the water. As the upper part of the oar would necessarily 
describe a greater curve than the lower, it would be natural that 
the pay of the Zranite should be higher than that of the Thalamite. 
In the case of Ptolemy’s ship, it is probable that the rowers relieved 
one another, and did not all row at the same time. When I had 
arrived at the above conclusion, it occurred to me that the term 
Oahapizrns admitted of a very significant derivation (it is ordinarily 
supposed to be connected with @ddapos, i.e. the man who sits in 
the hold”). The aperture through which the oar projected was 
called 7 @adapuia scil. éx7; and, as I take it, both these words are 
derived from oxaiyds, “the thole pin” to which the oar was fastened ; 
oxadpes naturally passes into oxadayos. On calculating the probabili- 
ties in favour of this derivation, I came across one or two other words, — 
for which it seemed to me more natural to assume a parallel phonetic 
change, than to assign them to the roots to which they are ordinarily 
referred: ¢.g., 0dztw is suggestive of oxiztw, Odxtw of oxdztw (cp. 
tdppos). Accordingly 6 §adayirns would be the rower who sat nearest 
- tothe thole pin. As I thought that the probabilities were in favour 
of this view, I ventured to communicate it to the Admiral, who had 
