496 JAMES CLERK MAXWELL. 



stant, as Poisson supposed, may be almost any whatever, 

 cork having great rigidity in comparison with its power to 

 resist compression, while caoutchouc, on the other hand, has 

 extremely little. 



3. While a bachelor-scholar of Trinity College, on March 

 13, 1854, Clerk Maxwell read a paper before the Cambridge 

 Philosophical Society " On the Transformation of Surfaces 

 by Bending." This paper, which embodies a great deal of 

 thought, and indicates that its author possessed an exten- 

 sive acquaintance with the geometrical works of Gauss, 

 Monge, Liouville, and others, is of more interest to the pure 

 mathematician than to the general reader, but in connection 

 with it we may mention a surface which was prepared for Pro- 

 fessor Maxwell many years afterwards, and coated by his own 

 hands, as described below, and now preserved in the Caven- 

 dish Laboratory. If a heavy uniform string be suspended 

 from two points, and allowed to hang freely, it assumes the 

 form of a curve called a catenary. If a board be cut out in 

 this shape, and the string be stretched around its edge, then 

 cut at the lowest point and one-half of the string unwrapped 

 so as always to be kept stretched, and therefore the free por- 

 tion of the string a tangent to the board at the point where 

 it leaves it, the extremity of the string will describe a curve 

 called a tractory, because it is the curve traced by a particle 

 lying on a rough horizontal plane, when the end of a string 

 attached to it describes a straight line on the plane, the 

 particle starting from a point outside the straight line. As 

 we go farther along the curve traced by the particle, we 

 of course continually approach the straight line described 

 by the end of the string, but never reach it. This straight 

 line is the " asymptote " of the tractory, and is also the 

 " directrix " of the catenary above referred to. ' If the trac- 

 tory be made to revolve about ' its asymptote it will trace 

 out a surface which has the peculiarity that the intrinsic 

 curvature is the same at every point of the surface and is 

 negative, the two principal radii of curvature being opposite 

 in direction and inversely proportional the one to the 



