﻿980 Dr. Thomas on Discharge of Air through Small 



ordinates and abscissae respectively. The relation between 

 the logarithms in the case of the larger orifices is clearly 

 linear, the points corresponding to orifices nos. 1-7 lying- 

 very accurately on the best-fitting line given by 



/ 



509 d 1 ' 892 . 



For orifices nos. 8-10, the value of the index increases 

 as the diameter of the orifice diminishes. This result, 

 together with the very approximate constancy of the 

 index a for disks nos. 1-6 shown in column 4 of the 

 table, indicates that these disks only of the series employed 

 can be regarded as thin disks. Disks 8, 9, and 10 are 

 to be regarded rather as orifices in thick disks, disk no. 7 

 affording a transition from one class to the other. The 

 results indicate that an orificed disk is to be regarded as 

 thin if the diameter of the orifice is not less than about 

 three times the thickness of the plate. In the recent paper 

 of Buckingham and Edwards * on the efflux of gases through 

 small orifices, the diameters of the orifices employed were 

 considerably smaller than any employed in the present 

 work, and moreover, were in no case greater than 1*7 times 

 the thickness of the disk, being in three cases out of four 

 very much less than this. Such disks would, in the light 

 of the present work, be characterized as thick disks. 



The average value of the contraction coefficient in the 

 present work was found to be 0674. For orifices of 

 diameter d cm. in thin disks (nos. 1-6) of thickness 0'0229 cm., 

 the discharge of air is given by the formulae 



V ' = 509^ 8 92 e 0-482±0-004 ( y i.) 



(2^)1/2 



0-674S 



N 



(1-2-9 J) , • • (vii.) 



the symbols having the significance given in the text. 



Of these, formula (vi.) is of a type which has some 

 physical justification. In addition, (vii.) is correct from 

 the point of view of physical dimensions. 



For purposes of comparison between orifices in thin and 

 thick " disks," experiments were in like manner carried out 

 with very carefully prepared short channels bored in brass. 

 Particulars of the various channels are given in Table III., 

 together with empirical formulae, the detailed results being 

 plotted in fig. 5 and the empirical formulas deduced as 

 already explained. 



* Scientific Papers, Bureau ot Standards, vol. xv. p. 584 (1919-20). 



