tract 2G.

MEAN DENSITY OP THE EARTH.

1 »

long connected chain of triangles, dependent on one another,the effects of such small errors at length become too great tohe tolerated, in a computation requiring much accuracy.—Another method is, first to compute, from both bases, thelength of the line kn, extended along the ridge of the hillfrom east to west, and from it, as a secondary base, to com-pute all the other lines in the plan. This method admits ofmuch more, accuracy than the former, supposing this second-ary base to be truly assigned ; because that, from the elevatedand central situation of this line, all or most of the other pointsin the survey are visible, from one or both of its extremities;by which it happens, that the other lines are mostly deter-minable from it alone, without so close a connection with oneanother as in the other method of computation. By both ofthese methods then, and by a.11 the triangles furnished by eachof them, I computed all the principal lines in the plan, andeither took a mean among the several values of each, or elseselected out of them such as from various circumstances Ijudged it safest to rely on, as nearest the truth. The trigo-nometrical computations were always accurately made, bythe common numbers, and generally repeated by logarithms,and the result of every proportion determined to two or threeplaces of decimals. I shall here abstract the mean or cor-rected values of some of the principal lines, or horizontaldistances, so computed, as well as the secondary base kn,from the Eastern to the Western cairn.

The mean among a great number of ways of computationfrom the South base, gives the horizontal distance of the se-condary base, from k to n, = 4052'2, and the mean of ail theresults from the North base a/3y gives kn = 4058-9, and themean between these two gives 4055-5 for the mean distancebetween k and n. And this value of kn was used in comput-ing most of the other lines, whose mean results are as herefollows.