than the biceps (5:6), and if long syllable was used to fill the
biceps it had to be dragged a little. When verse was sung rather
than recited, the long positions seem usually to have been given
a more precise mathematical relationship to the shorts, nor-
mally 2:1 but occasionally in some metres 3:1. The 3:1
('triseme') longs occur in iambic or trochaic song, where we
may find, for example, - ̮ - or ̮ - - or - - counting as a
metron side by side with, and evidently equivalent to, the
normal × - ̮ -. Using the symbol ⌞ or ⌟ employed by ancient
rhythmicians, we could interpret these as ⌞ ̮ -, ̮ - ⌞, ⌞ ⌞.
In a few exceptional instances such 'syncopated' metra, as they
are conventionally called, are found in responsion with unsyn-
copated metra, but normally the same syncopations occur in
strophe and antistrophe.

The recurrence of princeps-longs at every second or third
position in most metres gives a clear sense of rhythm, a rhythm
which often continues smoothly from one period into the next.
But we should guard against our natural inclination in reading
Greek verse to place the principes at equal intervals when the
number of positions separating them is in fact unequal. We
have this inclination partly because our own poetry is stressed,
and stressed at equal intervals (and even in our speech we tend
towards equal spacing of stresses, hurrying over sequences of
unstressed syllables and slowing down when there are fewer);
partly also because we are used to music with rather banal
rhythms and regular bar-lengths. Greek metre, like the tradi-
tional folk music of eastern Europe and Asia, sometimes
presents more intricate rhythms and changing bar-lengths. It is
precisely the asymmetrical distribution of the longs and shorts
that gives many metres their characteristic quality--the fact
that in - ̮ ̮ - ̮ -, for instance, the principes are not equidis-
tant. We ought not to read that sequence as if it were the same
as 'under the greenwood tree' but, counting two units (morae)
for the longs and one for the shorts, as 'tata-ta-ta-tata-ta-tata'.
When we see - ̮ ̮ - - ̮ ̮ - we should not give it a 'buckle my
shoe, buckle my shoe' rhythm, but something more like 'gin a
body meet a body'; and similarly 'coming through the rye' is a
more accurate realization of the dochmiac rhythm ̮ - - ̮ -
than 'the wise kangaroos' or 'the mome raths outgrabe',
provided that 'rye' is not prolonged for more than two morae.

-7-