32 The Contrast of Practice and Theory. 



to be able to follow out its consequences in the things it does 

 determine. This is particularly likely to be the case where a 

 choice must be made among different sets of data, selected 

 among the same set of facts, for our starting point of theoretical 

 deduction. 



It has been pointed ont that where, as in electrical machinery, 

 one set of theoretical rules governs the great body of results, 

 leaving comparatively small outstanding differences, much more 

 confidence is likely to be placed in theoretical calculations than 

 where the reverse holds good, as in the case of the steam 

 engine ; and here ability to follow out the consequences which 

 flow from assumed data comes into play, enabling experiment 

 to be dispensed with, as in the case of bridge design, to a con- 

 siderable extent. At the same time, increase in the scale on 

 which operations are carried on introduces always, more or less, 

 new practical questions, some capable and some incapable of 

 being settled, or put in the way of settlement, by theory. In 

 such a subject as this, then, it is desirable to understand the 

 method of reasoning to be employed, in its general features at 

 least, in order to be able to see what choice of means can be 

 suggested by theory, from which in practice that best suited to 

 the special circumstances of each case may be selected. 



The line of thought I have taken up leads, in a general way, 

 to looking at Practice as something wherein all the facts of a 

 case are known, down to the minutest microscopic particular, 

 and every possible consequence of every possible influence works 

 itself out in due proportion ; while Theory always, and una- 

 voidably, starts with facts more or less wrong, and data consisting 

 of a selection only, generally a very small one, of the real cir- 

 cumstances, with only a few of the consequences completely 

 followed out, and the rest either ignored altogether, roughly 

 guessed at, or merely fixed as lying within certain limits. 



Improvements in Practice resulting from theoretical con- 

 siderations are, from this point of view, best insured by a careful 

 examination of the points that the existing theory leaves out of 

 account. The importance of knowing how to follow out the 



