84 AN HISTORICAL DISQUISITION

SECT. principles with the moderns , yet it was by"XI. means of instruments very inferior in their con-struction to those now used , and without thesame minute attention to every circumstance thatmay affect the accuracy of an observation , anattention of which long experience only candemonstrate the neceffity. In order to ascertainthe latitude of any place , the ancients observedthe meridian altitude of the sun , either by meansof the shadow of a perpendicular gnomon , orby means of an astrolabe , from which it waseasy to compute how many degrees and minutesthe place of observation was distant from theEquator. When neither of these methods couldbe employed , they inferred the latitude of anyplace from the best accounts which they couldprocure of the Jength of its longest day.

With respect to determining the longitude ofany place, they were much more at a loss, asthere was only one set of celestial phenomenato which they could have recourse. These werethe eclipses of the moon ( for those of the sunwere not so well understood as to be subser-vient to the purposes of geography): the differ-ence between the time at which an eclipse wasi observed to begin or to end at two different^ places, gave immediately the difference between'the meridians of those places. But the difficultyof making those observations with accuracy, andthe impossibility of repeating them often, ren-dered them of so little use in geography , thatthe ancients in determining longitudes were