666 



Lieutenant Rennt on the Constants of 



Table XIII. 



Comparing the above Table XIII. with similar Table IV. of page 639, 1 re- 

 joice to be able to modify my opinion very much indeed, concerning the merits 

 of the large Tables of Horary Corrections of pages 650 and 657. The column 

 of errors, p.m., of the above Table, furnishes facts very satisfactory indeed, par- 

 ticularly for the seven months from March to September p. m., both months 

 inclusive. The mean error for these seven months p.m. is only 1-85 metres. 

 Now, it is not to be forgotten, that the vanishing moments of Table III., page 634, 

 were obtained as the mean often years' observation, from 1841 to 1850, both 

 years inclusive, so that the deviations of one particular year (viz., 1855) from 

 the mean of ten previous years, is small indeed. Moreover, M. Plantamour 

 informs us, at page 184 of the "Resume Meteorologique" for the year 1855, 

 that the mean temperature for this year (1855) was one-tenth of a degree colder 

 than the mean temperature of the twenty preceding years, so that, making 

 allowance for this circumstance, the mean error of calculated heights, p.m., for 

 the seven months from March to September, being only 1*85 metres, will be 



