On Sal Nitriim and Nitro- Aerial Spirit 59 



from e to i — while the rigid body is thus bending 

 there is no need that the convex surface should move 

 inwards, and so the matter of a rigid body thus 

 bent will suffer no compression. And yet the convex 

 surface is diminished a half less in this case than in 

 the preceding. 



And now we remark that in the case of a very 

 slender rigid body such as glass threads, whose surfaces 

 are much nearer each other than in the figure, the 

 contraction of the concave surface and the elongation 

 of the convex are extremely small. And hence it is 

 that the more slender rigid bodies are very easily bent ; 

 for if their sides, as is the case here, be but a little 

 lengthened or shortened their matter will scarcely be 

 compressed at all. On the other hand, let a, c^f^ g^ be 

 a bent rigid body twice as thick as the former. If in 

 bending this body, its convex surface is not to be 

 drawn inwards, nor its matter compressed, the elonga- 

 tion of the convex surface and the shortening of the 

 concave must be much greater than in the previous 

 case. For the elongation of the convex surface has to 

 be as great at each end as is the distance/, o, and g^ o, 

 or at least half that distance. For we suppose the line 

 between the limits o, o, to be equal to the line ^, ^, or 

 what is the same thing to the length of the rigid body 

 before inflexion. Now this can be easily illustrated 

 by means of the instrument delineated in Fig. 8. For 

 if the string of that instrument be placed near the rod, 

 so that the apparatus may represent a somewhat 

 slender rigid body, then if the string be lengthened out 

 a little, while the rod is bent, it will remain always 

 parallel to the rod and will not be constrained to move 

 inwards, and yet if the string be at a greater distance 

 from the rod you will see that the string, unless it be 

 drawn out much more while the rod is bending, will 



