vations collected with infinite pains from an endless variety of
authors ancient and modern.

In 1757 the well-known James Ferguson reprinted in his
_Astronomy_,[153] but in a very condensed form, all Struyck's eclipses
from 721 B.C. to A.D. 1485. Then he carried on his catalogue to 1800 by
means of the materials furnished by Ricciolus and _L'Art de verifier les
Dates_. Ferguson also invented a machine for illustrating mechanically
the circumstances of an eclipse. He called it the "Eclipsareon." A full
description is given in his book, mentioned above, but I do not know
whether any such instrument is still in existence, or, if so, where it
is to be found.

Ferguson apologises[154] for the incompleteness of his eclipse
information in the following words:--"I have not cited one half of
Ricciolus's list of portentous eclipses, and for the same reason that he
declines giving any more of them than what that list contains, namely,
that 'tis most disagreeable to dwell any longer on such nonsense, and as
much as possible to avoid tiring the reader. The superstition of the
ancients may be seen by the few here copied. My author further says that
there were treatises written to show against what regions the malevolent
effects of any particular eclipse was aimed, and the writers affirmed
that the effects of an eclipse of the Sun continued as many years as the
eclipse lasted hours, and that of the Moon as many months."

The most comprehensive (indeed almost the only) modern English book on
eclipses is the Rev. S. J. Johnson's,[155] of which frequent use has
already been made in these pages. It contains a vast amount of matter
put together in a condensed form but the references to authorities are
rather defective and deficient. Less comprehensive in one sense yet
exceedingly valuable and interesting as a succinct summary of solar
eclipse knowledge up to the date of 1896 is Mrs. D. P. Todd's excellent
little volume[156] which has been several times quoted on previous
pages. On various occasions in 1890 and following years Professor J. N.
Stockwell contributed to the American _Astronomical Journal_ a number of
papers[157] discussing in a very interesting and exhaustive manner many
of the eclipses recorded by the ancient classical authors. These papers
should be consulted by all who desire to realise the value of eclipse
records in connection with mundane chronology.

The calculation of eclipses is a matter of some interest. It is be