380 



On a close examination of all tbe observations embraced in the 

 series, we find one hundred and eighty-nine (189) which we think 

 ought to be rejected, because influenced in an extraordinary degree 

 by unfavorable winds. This would reduce the number of observa- 

 tions in the series to 8995 ; and each co-ordinate of altitude would 

 depend on a mean of three hundred and thirty-three observations, 

 and stand as follows, viz. : 



TABLE 2. 



Shotving the Jialf-Jwurli/ (and in two places the quarter-hourly) co- 

 ordinates of altitude of the average semi-diurnal lunar tidal 

 wave, at Chicago, on. Lake 3Iichigan, as derived from 8995 

 observations, made between January 1, and July 1, 1859. 



