t this passage from Varro is a new and hitherto unnoticed proof,
and certainly it ought to be a most convincing one, because it seems
impossible to give Varro's words a rational meaning without the admission
of this hypothesis, while with it everything is clear and consistent.

The commentators, driven by the necessity I have just pointed out, either
to admit the afternoon position of _sexta hora_, or to abstain from reading
it as a _space_ of time, have attempted to force a meaning by reading
_sexta hora_ in its other sense, an absolute mathematical point, the
_punctus ipse_ of noon.

In so doing they have not scrupled to libel Varro's common sense; they
represent his idea of the absurd to consist in the embarrassment that would
be caused by the birth occurring at the critical moment of change,--split
as it were _upon the knife-edge of noon_; so that, in the doubt that would
arise as to which day it should belong, it must be attributed partly to
both!

This interpretation is so monstrous, and so evidently wide of the meaning
of the words, that its serious imputation would scarcely be believed, if it
were not embalmed in the Delphin edition of Aulus Gellius, where we read
the following footnote referring to the _argumentum ad absurdum_ of Varro:

"Infirmum omnino argumentum, et quod perinde potest in ipsum Varronem
retorqueri. Quid enim? Si quis apud Romanos Calendis hora vi. noctis
fuerit natus, nonne pariter dies ejus natalis videri debebit, et partim
Calendarum, et partim ejus dici qui sequetur?"

It is not worth while to inquire what may have been the precise dilemma
contemplated by the writer of this note, since most certainly it is not a
reflex of Varro's meaning. The word _dimidiatus_ is completely cushioned,
although Gellius himself has a chapter upon it a little farther on in the
same volume.

The anomaly that amused Varro was the necessity of piecing together two
halves not belonging to the same individual day and with the hiatus of a
night between them; a necessity that would assuredly appear most absurd to
one who had no other idea of birthday than the twelve consecutive hours of
artificial day, which he would call "the natural day."

This proneness of the Romans to look upon the _dies solis_ as the only
effective part of the twenty-four hours, is again apparent in their
commencement of horary notation at sunrise, six hours later than the actual
commencement of the day. And in our own a