A LAW OF NATURE HAS NO EXCEPTIONS. 197 



extensive research, after we have drawn all the conclusions 

 that we are able, there remain a few outstanding cases 

 which do not conform to any of the general truths we 

 have found, and which also we cannot explain by them. 

 In those exceptional cases, as there is a difference of 

 result from that in the ordinary ones, there must be a 

 cause for that difference, and the cause should be dis- 

 covered. 



A fact which cannot be explained by any known 

 law or cause is probably an instance of the operation of 

 some new law or cause, and therefore important, A 

 single exception points to the existence of a more general 

 law ; for instance, the expansion of water, iron, iodide of 

 silver, fusible alloy, &c., during the act of cooling, at par- 

 ticular temperatures, points to a more general law than 

 the commonly-received and erroneously-stated superficial 

 one, ' all bodies expand whilst being heated.' The more 

 general truths are made manifest only by the exceptional 

 circumstances, and a truth of the most general kind is 

 unity in the widest diversity, and is one which is capable 

 of harmonising with the greatest variety of phenomena, 

 Whilst investigating the electrical relations of unequally 

 heated metals in liquids, I found in a large number of 

 cases that, provided chemical action and all other inter- 

 ferences were absent, hot platinum was negative to cold 

 platinum in solutions which were acid to test-paper, and 

 positive in those which were alkaline, and I therefore con- 

 cluded that the direction of the electric-current was de- 

 termined by the chemical nature of the liquid ; but by 

 examining a still more extensive number of liquids, I met 

 with a few decided exceptions (selenious acid, chrome- 

 alum, &c.), and was therefore compelled to consider the 

 law I had discovered was only an apparent one, and to 

 draw the more general conclusion that the direction of the 



