35 



side collateral with the upper point of application. 



§ 34. Were the rod laid horizontally on the ground 

 with the lower end at the rear, and the ground appui 

 of the now rear end provided for by some projection, 

 the now anterior C might at its convex side be 

 provided with an independent appui on the ground, 

 which should aid, not in projecting it forward, but 

 in forming the counter-actions which give the 

 spring. Tlds additional appui, by collecting force 

 from the anterior C, would raise it from the ground, 

 and not being on the side first thrust forward by 

 the changing curvature, would leave the ground after 

 the rear point of appui. 



§ 35. We have considered only one S curve, but, 

 if there were to be a succession of them, the ac- 

 tion generated from one " point of application " 

 might be transferred from one S to the other until 

 all were thrown into form. In this case the junc- 

 tions of every two C C (of the S S, marked off fi'om 

 the top downward), even when they belonged to 

 different S S, might all be considered as contrary 

 " cross-lines," in reference to each other, but we 

 shall find it more consonant with facility of explan- 

 ation to suppose fresh "points of application" at the 

 terminal planes of junction of all S S, and " tjross- 

 lines at the junction of their C C." 



