Notices respecting Nno Booh. 521 



to adapt it to a different latitude ; and the movement necessary for 

 following a star in the act of measurement was communicated, as 

 previously, not by a clock-work mechanism, but by a screw move- 

 ment worked by a long handle and a Hook's joint. As respects the 

 micrometer, its readings of position are stated to have been always 

 unexceptionable, but in regard of the distances the case was far 

 otherwise. " The inherent defect of the instrument (a very costly 

 one), that of the hitching of the parallel spider's threads in the act of 

 crossing, could never be fairly overcome. It was in vain that, pre- 

 viously to quitting England, it was placed in the maker's hands for 

 the express purpose of remedying this annoying evil. For a while, 

 indeed, it performed satisfactorily, but the mischief soon recurred, 

 and grew at length so obvious a source of error, that already in 

 October 1 834, I began occasionally to substitute for the system of 

 measures by cross zeros, with the spider lines, that of an absolute 

 zero, with one-sided measures between the inner edges of two thick 

 parallel wires, which have been added to the system of wires in the 

 micrometer, and which could be brought into the centre of the field 

 by the screws, to the exclusion of the spider-threads." But this was 

 a less eligible mode of measurement, and gave rise to a great deal of 

 trouble ; nevertheless, as the evil v/ith the spider-threads continued 

 to increase, in April 1835 they were removed, and from this time till 

 June 1836 the thick wires were alone used in the measurement of 

 distances. The micrometer was then refitted by Mr. iMaclear, with 

 a pair of beautifully even and delicate threads of the Bermuda spider. 

 '• These went on very well for some time, but in April 1837 the hitch- 

 ing was again rem.arked, and from this time to the conclusion of the 

 oljservations continued a more or less frequent source of annoyance." 



'J'he measures of distance taken with the thick wires, when com- 

 j)ared with those observed with the spider-lines, exhibit a systematic 

 disagreement, in consequence of which a correction must be applied 

 before the results by the two methods can be compared. To facili- 

 tate the reduction, a table is computed, which gives the correction 

 for each second of distance from 2" to 1 10". The tabular correc- 

 tions are ajiplied to all the measurements with the thick wires, as 

 part of the reductions. 



The results set dow-n in the Catalogue are the means of the several 

 measures, and iveights are attached, depending on the number and 

 presumed goodness of the observations. " Individual measures are 

 in no case given, excepting under some remarkable circumstance, or 

 if, as sometimes hajjpens, only a single measure was procured. To 

 have done so would have entailed an immense mass of printing, 

 without any corresponding advantage." 



" Appended to the micrometrical measures registered as the results 

 of single nights' observations, is a synoptic view of tlie general mean 

 results reduced to mean epochs." 



The Catalogue of Micrometrical Measures is arranged in ten co- 

 lumns, giving the following jjarticulars : — 1st, the index number; 

 2nd, the synonyms; 3rd and 4th, the UA and NPD for 1830; oth 

 and 6th, the angle of position and weight or value attributed ; 7th 



