On Machines in Generali \i 



seem to have any connection with them. It is in vain to tell 

 them that every machine may be reduced to the lever ; thiii 

 assertion is too vague and too wire-drawn to be admitted . 

 without a profound examination ; they cannot persuade 

 theinstlvcs that machines which appear to have notliing in 

 common with those denominated simple ones, arc subject 

 to the same law, nor that we can pronounce upon the in- 

 utility of a secret winch has not been connriunicatcd to any 

 person : thence it happens that, the most absurb ideas, and 

 the furthest removed from the simplicity so advantageous to 

 machines, arc those which furnish them the most hopes. 



The method of rooting out this error is, certainly to attack 

 it in its very source, by showing that not only in all the 

 machines known, but also in all possible machines, it is an. 

 invariable law — that we always lose in time or in velocity 

 what we gain in power, — and to explain cleady what this 

 law signifies ; but to thi§ effect we must raise ourselves to 

 the greatest generality possible, and not stop at any parti- 

 cular machine, or resort to any analogy^. In the last place, 

 there must be a general demonstration, 'deduced imme-, 

 dialely and geometrically from the first axioms in ifiecha- 

 nics : this is what I have atten^pted in this Essay. I have 

 strongly insisted upon this fundamental point, and I do not 

 know' if f have succeeded in placing it in a sufficiently clear, 

 light ; but on attacking error we are compelled to substitute 

 truth in its place; — 1 have shown what is the true end of 

 machinery: if it be unreasonable to expect prodigies from- 

 them beyond all probability, we shall still find there is 

 plenty of utility in them for exercising the most lively ima- 

 gination. 



The reflexions T propose upon this law lead me to say a. 

 word of perpetual motion : and I have shown not only that, 

 every machine abandoned to itself must infallibly stop, but 

 I assign the- verv instant when this must happen. 



There will also be found among these reflections one of 

 the most interesting properties of machines, which I think 

 has not yet been remarked ; it is, that in order to make them 

 produce the greatest possible eflTect, it must necessarily 

 happen that there be no percussion, i.e. that the move- 

 ment 



