THE MECHANICS 
34 7 
the first chapter or chapters of the real Mechanics which had 
been lost. The treatise would doubtless begin with generalities 
introductory to mechanics such as we find in the (much 
interpolated) beginning of Pappus, Book YIII. It must then 
apparently have dealt with the properties of circles, cylinders, 
and spheres with reference to their importance in mechanics ; 
for in Book II. 21 Heron says that the circle is of all figures 
the most movable and most easily moved, the same thing 
applying also to the cylinder and sphere, and he adds in 
support of this a reference to a proof £ in the preceding Book b 
This reference may be to I. 21, but at the end of that chapter 
he says that ‘cylinders, even when heavy, if placed on the 
ground so that they touch it in one line only, are easily 
moved, and the same is true of spheres also, a matter which 
ice have already discussed ’; the discussion may have come 
earlier in the Book, in a chapter now lost. 
The treatise, beginning with chap. 2 after the passage 
interpolated from the Bapov\KO<s, is curiously disconnected. 
Chaps. 2-7 discuss the motion of circles or wheels, equal or 
unequal, moving on different axes (e.g. interacting toothed 
wheels), or fixed dn the same axis, much after the fashion of 
the Aristotelian Mechanical 'problems. 
Aristotle's Wheel. 
In particular (chap. 7) Heron attempts to explain the puzzle 
of the ‘ Wheel of Aristotle ’, which remained a puzzle up to quite 
modern times, and gave rise to the proverb, ‘ rotam Aristotelis 
magis torquere, quo magis torqueretur A ‘ The question is ’, says 
the Aristotelian problem 24, ‘ why does the greater circle roll an 
equal distance with the lesser circle when they are placed about 
the same centre, whereas, when they roll separately, as the 
size of one is to the size of the other, so are the straight lines 
traversed by them to one another V 2 Let AC, BD be quadrants 
of circles with centre 0 bounded by the same radii, and draw 
tangents AE, BF at A and B. In the first case suppose the 
circle BD to roll along BF till D takes the position H; then 
the radius ODG will be at right angles to AE, and G will be 
at G, a point such that AG is equal to BH. In the second 
1 See Yan Capelle, Aristotelis quaestiones mechanicae, 1812, p. 268 sq. 
2 Arist. Mechanica, 855 a 28.