A WAVE-LENGTH COMPARATOR FOR STANDARDS OF LENGTH. 29 



"Tresca" section, shown in fig. 11. The defining lines on this bar are similar to 

 those of the Imperial Standard Yard, except that only one transverse line is engraved 

 instead of three, and the engraving was performed directly on the platinum-iridium, 

 on that one of the surfaces of the horizontal transverse portion of the bar which 

 contains the centre of the section. The appearance of the defining line under the 

 ordinary comparator microscope is represented in fig. 12. 



The defining lines on the Imperial Standard Yard were engraved by Messrs. 

 Troughton and Simms, its constructors, and were the finest lines it was considered 

 advisable for the sake of permanency should be engraved. The defining lines on the 

 platinum-iridium yard were engraved at the Bureau International des I'oids et 

 Mesures at Sevres by M. BENOIT, the Director. They are only one-third as thick as 

 the Troughton and Simms lines, and are similar to those on the French metre employed 

 by MroHELSON. 



The lines on the platinum-iridium yard, under a magnifying power of 200 diameters, 

 are observed to have very irregular edges, instead of appearing under this only mode- 

 rately high power as a broad band of equal thickness with sharp edges. The centre of 

 this line has to be estimated as the actual defining limit of the yard, and it can be well 

 understood how difficult the operation is with any but the lowest-power objectives. 

 Its average thickness is such as includes no less than 15 half wave-lengths of red 

 hydrogen or cadmium light. That is to say, on traversing the microscope of the 

 apparatus described in the preceding communication through the distance covered by 

 the average width of. either of the defining lines on the platinum-iridium standard 

 yard, no less than 15 interference bands pass the reference spot in the centre of the 

 field of view. 



When the defining lines on the Imperial Standard Yard are similarly investigated, 

 they are found to be much more regular, although three times as thick as the lines 

 just referred to. No less, however, than 45 interference bands pass the reference 

 spot on traversing the microscope through the width of either of these lines. 



It will be clearly obvious that defining lines which are considered sufficiently 

 accurate for the degree of refinement of measurement possible to the micrometer- 

 microscope comparator, are very coarse and inadequate for wave-length measurement 

 by the delicate instrument described in the preceding communication. Like the 

 constant struggle between guns and armour-plate, there is a similar struggle going on 

 between fineness of measurement and defining lines, and for the moment refinement 

 of measurement has triumphed. The only weak point in MIUHELSON'S wonderfully 

 accurate determination of the total number of wave-lengths of red cadmium light in 

 the French metre is, if the author may be permitted to say so, and as MICHELSON 

 himself states in his book on ' Light Waves,' in the final comparisons between his 

 glass-plate e"talons (or the Fabry and Perot etalons) affording the interference fringes, 

 and the standard bar itself, on which the defining lines were similar to those on 

 the British platinum-iridium bar. For the centre of each line has to be estimated, 



