M. 8. Woodward — Iced Bar Base Apparatus. 51 



requirements of the iced bar apparatus. This comparator was 

 constructed in a lot adjacent to the Survey Office. Briefly 

 described it consists of six brick piers resting at intervals of a 

 meter on a foundation of six cubic meters of well rammed 

 concrete. The foundation and the piers are set in Portland 

 cement and weigh about twelve tons. The foundation, which 

 rises to near the ground surface, is covered to the depth of a 

 decimeter with sawdust. Each pier carries a micrometer 

 microscope. The latter and their connections with the piers 

 are wrapped heavily with cotton batting. The meter and 5 m 

 bars were moved under the microscopes on portable tracks sup- 

 ported on posts isolated from the piers. The track for the 5 m 

 bar was that used in the field, and the mode of handling this 

 bar was precisely the same in all respects as that followed in 

 measuring a line. 



The microscopes as mounted on this comparator proved to 

 be very stable notwithstanding the fact that they were subject 

 to the daily range in air temperature except so far as they 

 were protected by cotton batting wrappings. The effect of 

 the observers' presence near the comparator was undoubtedly 

 less than in the office comparing room, and the use of artificial 

 light was avoided entirely. 



The program followed in the use of the new comparator 

 comprised three measures of the distance between the end 

 microscopes with the meter and two measures with the 5 m bar 

 in each set of observations. The measures with the meter 

 were made in opposite directions alternately along the com- 

 parator. Twenty sets of observations were made with the 5 m 

 bar in each of its two orientations relatively to the Y-trough 

 and microscopes. The consistency of the results attained 

 leaves little to be desired. The probable error of a single 

 determination of the length of the 5 m bar in terms of the 

 Prototype is ±1'"*8 from one group of measures and dbl^'7 

 from the other. The range in the one case is 12' a, 3 and in the 

 other 8^5. 



The following table gives an abstract of the results obtained 

 in the manner described above' for the length of the 5 m bar. 

 The values given are subject to corrections for flexure of the" 

 5 m bar, which, however, cannot exceed a few tenths of a micron. 

 The probable errors are those which come from the discrepan- 

 cies between the individual and mean results of a group of 

 determinations. Each result is derived from an equal number 

 of measures with the meter and 5 m bars in their two different 

 orientations. The first, second, and third results were obtained 

 from the Office comparator, the third one depending on the 

 method wherein the auxiliary bar, No. 18, was used. The 

 fourth result was obtained from the new comparator. 



