THE STEAM-ENGINE. 117 



paddle-shaft. At each reversal of this shaft, the band which 

 had been driving became slack and ceased to drive, while 

 that which had been slack became tight and did the driving. 

 Obviously, the alternate coming into operation of an open 

 and of a crossed band worked from a shaft driven first one 

 way and then the other, would communicate to the paddle- 

 wheel the desired motion in the one direction. 



Further, on the subject of steam navigation, we have in the 

 collection the actual machine (No. 2150) which, in 1788, 

 worked on Dalswinton Lake in Scotland and drove a small 

 boat. We have likewise the actual engine of Bell which 

 drove the first passenger steamer in Europe. 



Leaving navigation, and going to another branch of the 

 subject, namely, steam as applied to land transit, I must 

 refer you to the extremely interesting model (No. 2145) of 

 Cugnot's steam carriage, which, in the year 1769 traversed 

 the streets of Paris carrying passengers. It went but very 

 slowly, only some two or three miles an hour, and it had 

 frequently to stop to get up steam, but nevertheless it was 

 an actual working common road locomotive, and is therefore 

 a model of great interest, looking at the date at which the 

 original engine was put to work. You will see it is a three 

 wheel carriage, the engine and boiler being attached to the 

 front wheel, so that when the front wheel was slewed by the 

 steersman for the purpose of altering the direction, the whole 

 engine and boiler canted with it. Very curiously \it may 

 be in the recollection of some of you) two or three years 

 since, Mr. Perkins exhibited in the grounds attached to these 

 buildings a common road locomotive, which like Cugnot's was 

 a three wheel machine, and like it, had the engine applied to 

 the castor wheel, and turned boiler and all, with the castor 

 wheel as the steersman moved it. 



In Cugnot's engine we have another instance where rotary 

 motion was wanted, and where there was no provision for 

 obtaining it such as we have now-a-days by means of a crank. 

 Cugnot overcame the difficulty by employing two direct 

 acting inverted cylinders which, by means of palls operating 

 on ratchets, kept the driving wheel revolving always in the 

 desired direction. 



Passing from Cugnot, I now come to Trevithick's engine, a 

 model of which we have here. I was about to ask you to 

 look at the Blenkinsop rail, a sample of which, belonging to 



