thought possible before.
4. Lord (1923) gives figures on the number of steam engines produced by Boulton
and Watt between 1775 and 1800, and The Cambridge Economic History of Europe
(1965) provides data on the spread of total horsepower between 1800 and 1815
and the spread of steam power more broadly. However, Kanefsky (1979) has largely
discredited Lord's numbers, which is whywe use figures on machines and horsepower
from Kanefsky and Robey (1980).
Our horsepower calculations are based on 510 steam engines generating about
5,000 horsepower in the United Kingdom in 1760. During the subsequent forty
years, we estimate that about 1,740 engines generating about 30,000 horsepower
were added, which leads to our estimate that the total increased at a rate of roughly
750 horsepower each year. For 1815, we estimate about 100,000 horsepower - that is,
the average of the figures Kanefsky and Robey (1980) give for 1800 and 1830. This,
together with the 35,000 horsepower we estimate for 1800, leads to our estimate that
the total increased at a rate of roughly 4,000 horsepower each year after 1800.
Data on the fuel efficiency, or "duty," of steam engines is from Nuvolari (2004b).
5. Kanefsky and Robey (1980), together with Smith (1977-78), provide a careful historical account of the detrimental impact of Newcomen's, first, and of Watt's patents, later, on the rate of adoption of steam technology. Apart from the books just quoted,
information about Hornblower's engine and its relation to Watt's are widely available
through easily accessible Web sites, such as Encyclopaedia Britannica, Wikipedia, and
so on. Some details of Hornblower's invention may be of interest. It was patented in
1781 and consisted of a steam engine with two cylinders, significantly more efficient
than the Boulton and Watt design. Boulton and Watt challenged his invention and
won, claiming infringement of their patent because the Hornblower engine used a
separate condenser. With the 1799 judicial decision against him, Hornblower had
to pay Boulton and Watt a substantial amount of money for past royalties, while
losing all opportunities to further develop the compound engine. His principle of
the compound steam engine was not revived until 1804, by Arthur Woolf. It became
one of the main ingredients in the efficiency explosion that followed the expiration
of Boulton and Watt's patent.
Watt's low-pressure engines were a dead end for further development; history
shows that high-pressure, noncondensing engines were the way forward. Boulton and
Watt's patent, covering all kinds of steam engines, prevented anyone from working
seriously on the high-pressure version until 1800. This included William Murdoch,
an employee of Boulton and Watt, who had developed a version of the high-pressure
engine in the early 1780s. He named it the "steam carriage" and was legally barred
from developing it by Boulton and Watt's successful addition of the high-pressure
engine to their patent, though Boulton and Watt never spent a cent to develop
it. For the details of this story, readers should see the Web site Cotton Times, at
http://www.cottontimes.co.uk/ (accessed February 23, 2008), or Carnegie (1905),
pp. 140-1. The "William Murdoch" entry in Wikipedia, at http://en.wikipedia.org/
wiki/WilliamJvlurdoch (accessed February 23, 2008), provides a good summary.
More generally, various researchers directly connect Murdoch to Trevithick, who is
now considered the official inventor (in 1802) of the high-pressure engine. Quite
plainly, the evidence suggests that Boulton and Watt's patent retarded development
of the high-pressure steam engine, and thus economic development, by about sixteen
years.
6. The story about Pickard's patent blocking adoption by Watt is told in von Tunzel-
mann (1978).
7. Thompson (1847), p. 110, and quoted in Lord (1923).
8. Scherer (1984), pp. 24-25.
9. NTP, Inc. v. Research in Motion Ltd., Civil Action Number