to the
scientific world in modern times. That the Arabians were very capable
practical astronomers has long been recognised as a well-established
fact, and if it had not been for them there would have been a tremendous
blank in the history of astronomy during at least six centuries from
about the year A.D. 700 onwards. In the year 1804 there was published at
Paris a French translation of an Arabian manuscript preserved at the
University of Leyden of which little was known until near the end of the
last century. The manuscript was then sent to Paris on loan to the
French Government which caused a translation to be made by "Citizen"
Caussin, and this was published under the title of _Le Livre de la
grande Table Hakenate_.[74] Caussin was Professor of Arabic at the
College of France. Newcomb considers this to contain the earliest exact
astronomical observations of eclipses which have reached us. He remarks
that some of the data left us by Ptolemy, Theon, Albategnius and others
may be the results of actual observations, but in no case, so far as is
known, have the figures of the actual observations been handed down. For
example, we cannot regard "midnight" nor "the middle of an eclipse" as
moments capable of direct observation without instruments of precision;
but in the Arabian work under consideration we find definite statements
of the altitudes of the heavenly bodies at the moments of the beginning
and ending of eclipses--data not likely to be tampered with in order to
agree with the results of calculation. The eclipses recorded are 28 in
number and usually the beginning and end of them were observed. The
altitudes are given sometimes only in whole degrees, sometimes in coarse
fractions of a degree. The most serious source of error to be confronted
in turning these observations to account arises from the uncertainty as
to how long after the first contact the eclipse was perceived and the
altitude taken; and how long before the true end was the eclipse lost
sight of. Making the best use he could of the records available Newcomb
found that they could safely be employed in his investigations into the
theory of the Moon.

The observations were taken, some at Bagdad and the remainder at Cairo.
I do not propose to occupy space by transcribing the accounts in detail,
but one extract may be offered as a sample of the rest--"Eclipse of the
Sun observed at Bagdad, August 18, 928 A.D. The Sun rose about
one-fourth eclipsed. We look