Further Remarks on the "Crucial-Test " Argument. 489 



The mean of these determinations is 



4-031, 



which agrees closely with the value calculated from the experi- 

 ments with the disks. 



55. It seems proper to point out in conclusion, that Professor 

 J. Clerk Maxwell indicated so long ago as 1856 (Brit. -Assoc. 

 Report, 1856, Trans, of Sect. p. 12; see also, for fuller develop- 

 ments, Camb. Phil. Trans, vol. x. p. 27) the fundamental idea 

 of the method of treating questions relating to the flow of elec- 

 tricity in conductors of more than one dimension which was used 

 in the first Part of this paper, — and also to mention that problems 

 presented by finite plane conductors, in cases which become un- 

 manageable by the simple geometrical processes employed by us 

 (as, for instance, the case of rectangular conducting sheets), have 

 been solved by Jochmann (Schlomilch's Zeitschr. f. Mathemat. 

 1865, p. 48) by a method founded on the superposition of the 

 effects of separate poles, and by Heine (CrehVs Journal, 1874, 

 vol. lxxix. p. 1) by an application of the geometrical principle of 

 point- point correspondence (Abbildung) to Kirch hofPs results. 



LVIL Further Remarks on the " Crucial-Test" Argument. 

 By James Croll, of H. M. Geological Survey*. 



AS there is a slight discrepancy between Muncke's Table of 

 the expansion of sea-water and that of Professor Hubbard, 

 especially in reference to high temperatures, I have been induced 

 to compare the results obtained by the use of the latter Table 

 with that by the former ; and I find that they are almost identical. 

 Both give 2 feet 6 inches as the amount by which the expansion 

 of column B (lat. 23° N.) exceeds that of the equatorial column. 

 Muncke's Table, as we have already seen, gives 3 feet 6 inches 

 as the amount by which column A (lat. 38° N.) exceeds that of 

 the equatorial, while according to Professor Hubbard's Table it 

 is 3 feet 4 inches, being a difference of only 2 inches. 



Dr. Carpenter objects f to my result on the ground that I 

 have omitted the consideration of the inferior salinity of the 

 equatorial column. Had I taken this into account, he thinks I 

 should have found that it makes a difference in the opposite direc- 

 tion of about 1 foot in 1026, which would more than neutralize 



* Communicated by the Author. 



t Philosophical Magazine for November, p. 403. 



