R. S. Woodward — Iced Bar Base Apparatus. 



47 



may differ for the two ends of the bar. Each section of the 

 kilometer was measured in both directions. The bar way kept 

 in the same position, relatively to the Y-trough, it had when 

 used on the comparator. A measure from west to east on the 

 comparator corresponds then to a measure from north to south 

 on the kilometer. 



Two series of measures of the kilometer were made ; one 

 immediately after the intermediate section stones were set and 

 while the ground was very unstable, and one after the ground 

 was dry and very stable. In addition to errors arising from 

 unstable ground the first measures were subject to some errors 

 arising from delay and temporary stops, which were avoided 

 in the second measures. Accordingly, the first series of 

 measures must be regarded as much less precise than the 

 second. 



The following table gives the results of these two series of 

 measures. The numerical quantities given are the excesses 

 measured with the cut-off scale over a round number of bar 

 lengths, the number of bar lengths being fifty for each sec- 

 tion and two hundred for the whole kilometer. The first 

 measures were made on Sept. 10 to 15 and the second on Sept. 

 26 to 30. 



1-2 



3 

 4 



3-4 



Results of measures of kilometer. 

 1st series, Sept. 10 to 15, 1891. 



No. of 



Direction 



Section 



Section 



Section 



Section 





meas- 



of 



m to 



250 m to 



500 m to 



T50 m to 





ures. 



measure. 



250 m . 



5(>0 m . 



750 m . 



1000 m . 1 



a -1000 n 







mm 



mm 



mm 



mm 



mm 



1 



N to S 



+ 6-35 



— 5-43 



+ 19-08 



— 20-56 



— 0-56 



2 



StoN 



+ 6-52 



-4-55 



+ 19-37 



— 19-28 



+ 2-06 



-0*17 —0-88 - 0-29 — 1-28 — 2'62 



N to S 

 S to N 



2d series, Sept. 26 to 30, 1891. 



+ 6-97 -5-50 +19-67 —21-28 

 + 7-07 -5-60 +19-76 —21-19 



—0-10 +0-10 



0-09 



0-09 



-0-14 



+ 0-04 



-0-18 



Without attempting to discuss these results here, it may be 

 pointed out that they indicate systematic differences depend- 

 ing on the direction of measurement. These differences agree 

 in sign with those developed in the forward and backward 

 measures of the 100 m comparator as explained above. It may 

 be remarked also that the largest differences in the first series 

 are found in the measures of the second and fourth sections 

 where the ground was least stable. 



