148 JAMES PRE3COTT JOULE
not generally seen unless looked for. The locomotive is
obtrusive; it will be seen: and by 1842 locomotives had
obtruded themselves pretty well all over;, Europe. They
immediately took their places as objects of as much wonder
and interest to the grown people who saw them for the first
time as they are still to the young; demanding the attention
even of philosophers who had previously studied nothing
lower than the planets."
Figures from which the mechanical equivalent of heat can
be roughly calculated are given in a book by M. Seguin
which appeared in 1839, and is named Etude sur LJ Influence
des Chemins de Per. Seguin was a nephew of Montgolfier,
the French paper manufacturer and inventor of the balloon.
He told W. R. Grove that Montgolfier believed force was
indestructible, an opinion reinforced by his consideration of
the principle of the hydraulic ram. Following the sugges-
tion of the indestructibility of force, Seguin contended that
power could not be derived from a mere transfer of heat.
He assumed that the expanding steam in the cylinder of a
steam engine did work equal to the amount of heat it lost.
He attempted to measure the amount of heat lost by
measuring the heat taken from the boiler and given to the
condenser, but his experimental methods were inadequate.
On the assumption that the subtraction of one unit of heat
from water vapour at some temperature between 100° C.
and 150° C., say 120° C., reduced the temperature by one
degree, say to 119° C., he calculated from the figures he had
published in 1839, the quantity of mechanical work
obtained by a given loss of heat. As the properties of water-
vapour in this range of temperature depart widely from those
of a perfect gas, S^guin's method could not give an accurate
result. One unit of heat reduces the temperature of a unit
of water vapour only three-tenths of a degree, and not one
degree. But this incorrect datum accidentally made his
result nearly correct. He concluded that one gramme of
water which lost one degree centigrade would provide
enough work to raise 500 grammes through one metre.
The correct result is 427. Thus the first material for an
estimate of the mechanical equivalent of heat seems to have