SMMK MKIIC'URY STAN! > AIM>S OF RFSlST ANCK. FTC 81 



The " End " Effect of the Standard Tube*. 



In the last column of Table IV., the calculation of the resistance of the standards 

 is based on the assumption that the termini of the resisting column are planes at 

 right angles to the axis of the tube and terminating the bore. 



When the mercury txita terminates in two vessels, the dimensions of which are 

 infinitely great compared with the diameter of the tube, and the whole is occupied 

 with a conducting liquid, the resistance between two thick leads introduced into the 

 vessels has l>een approximately calculated by Lord RAYI.KIGH. The difference between 

 this resistance and that first mentioned depends on (a) the liquid employee!, (b) the 

 terminal sections of the tube, (c) the dimensions of the end vessels and of the flange 

 of the tube, and (d) the position of the leads ; it is termed the " end correction." It 

 is conveniently expressed as the resistance of two columns of the liquid, of lengths 

 fi\ and fr.,, and of respective radii 7*j and r. 2 , where f is a constant for similar 

 dispositions of leads, vessels, &c., and r, and r 2 are the radii of the terminal sections of 

 the standard tube. 



Measurements of resistance when the standard is erected by Method I. (a description 

 of which immediately follows) are independent of this "end correction." Those taken 

 when the style of erection is that denoted by II. include it. The chief practical 

 difference in the two cases is the great care required for one method of erection, 

 and the simplicity of the other. As will afterwards be shown, Method of Erection I. 

 is difficult, and the measurements are subject to a considerable probable error unless 

 the greatest care be taken. For the purpose of checking wire standards of resistance 

 this is unfortunate, since for such observations an easily reproducible and constant 

 resistance is essential. Respecting Method II., the calculation of the theoretical 

 resistance is less rigid than with Method I., but every important detail is easily 

 reproduced. 



Hence, for the first determination of resistance in international ohms, the standards 

 have all been erected by Method I., while Method II. has been adopted for the 

 reproduction of standards. 



Erection of Standards. Method I. 



When the standard portion was cut away from the calibrated tube, the connecting 

 pieces were so marked that they and the standard could at any time be arranged in 

 line in the original manner. Under such conditions, the cross-sections facing each 

 other arc similar in value, and the general contours of the areas the same. These 

 end -connecting pieces were ground, and triple cuts made in them in a manner 

 similar to the standard lengths already described. A connector was now desired, 

 such that the three pieces of tube could be gripped together, with their axes in line, 

 and without internal irregularity occurring at the junctions. For the purpose in view, 



VOL. cciv. A. M 



