between a Machine and its Model 147 



facts, the most important which geometry presents, my after 

 remarks are mostly to be founded. 



All machines consist of moveable parts, sliding or turning on 

 others, which are bound together by bands, or supported by 

 props. To the frame-work I shall first direct my attention. 



In the case of a simple prop, destined to sustain the mere 

 weight of some part of the machine, the strength is estimated 

 at so many hundredweights per square inch of cross section. 

 Suppose that, in the model, the strength of the prop is sufficient 

 for double the load put on it, and let us examine the effect; of 

 an enlargement, ten-fold, of the scale according to which the in- 

 strument is constructed. By such an enlargement, the strength 

 of the prop would be augmented 100 times; it would be able 

 to bear 200 loads such as that of the model, but then the weight 

 to be put on it would be 1000 times that of the small machine, 

 so that the prop in the large machine would be able to bear only 

 the fifth part of the load to be put on it. The machine, then, 

 would fall to pieces by its own weight. 



Here we have one example of the erroneous manner in which 

 a model represents the performance of a large instrument. The 

 supports of small objects ought clearly to be smaller in propor- 

 tion than the supports of large ones. Architects, to be sure, 

 are accustomed to enlarge and to reduce in proportion; but Na- 

 ture, whose structures possess infinitely more symmetry, beauty, 

 and variety, than those of which Art can boast, is content to 

 change her proportions at each change of size. Let us conceive 

 an animal having the proportions of an elephant and only the 

 size of a mouse ; not only would the limbs of such an animal be 

 too strong for it, they would also be so unwieldy that it would 

 have no chance among the more nimble and better proportioned 

 creatures of that size. Reverse the process, and enlarge the 

 mouse to the size of an elephant, and its limb?, totally unable to 

 sustain the weight of its immense body, would scarcely have 

 strength to disturb its position even when recumbent. 



The very same remarks apply to that case in which the 

 weight, instead of compressing, distends the support. The 

 chains of Trinity Pier are computed to be able to bear nine 

 times the load put on them. But if a similar structure were 

 formed of ten times the linear dimensions, the strength of tlw 



K 2 



