Steam Engine. ^87 
scjuare inch of the piston ; and the inferiority of that power 
to the known pressure of the atmosphere, was, without 
due consideration, imputed wholly to friction. The bulk 
of water when converted into steam was very erroneously 
computed ; the quantity of fuel necessary to evaporate 
a given quantity of water was not even guessed at ; whe» 
ther the heat of steam is accurately measured by its tem- 
perature was unknown ; and no good experiment had 
been made to determine the quantity of injection water 
necessary for a cylinder of given dimensions. In a word, 
no man of science in this country had considered the sub- 
ject since Desaguliers ; and his writings, in many re- 
spects, tended more to mislead than instruct. 
Such was the state of matters, when, fortunately for 
science and for the arts, Mr. Watt, then a mathematical 
instrument- maker at Glasgow, undertook the repair of 
the model of a steam engine belonging to the university. 
In the course of his trials with it, he found the quantity of 
fuel and injection water it required m.uch greater in pro- 
portion than they were said to be in large engines ; and it 
soon occurred to him, that this must be ov/iiig to the cy- 
linder of this small model exposing a greater surface, in 
proportion to its contents, than larger cylinders did. This 
he endeavoured to remedy, by making his cylinders and 
pistons of substances which conducted heat slowly. He 
employed wood, prepared on purpose, and resorted to 
other expedients, without producing the desired ellect in 
any remarkable degree. He found, also, that all attempts 
to produce a greater degree of exliaiistion, or a more per- 
fect vacuum, occasioned a disproporiionate expenditure 
of steam. In rehecting upon the causes of these pheno- 
mena, the recent discovery, that \vater boiled in an ex- 
hausted receiver at low degrees of heat, (certainly not ex- 
ceeding 100 degrees of Fahrenheit, but probably, wlien 
the ’^^aeuum was perfect, much lower) occurred to him ; 
