90 



machines depends greatly on the attention of the fireman ; the sup- 

 ply of fuel should be nearly uniform, but if the attendant, as is usu- 

 al, waits till the fiirnace is nearly empty, and then throws in a large 

 quantity of coal, we may indeed burn the smoke by opening the re- 

 gister so as to admit more air, but it is not possible to do this with- 

 out losing much heat and such alternations should be avoided ; if a 

 steam engine were at hand it would perhaps be advisable to feed the 

 fire by machinery, and leave to the workmen only the care of keep- 

 ing the grate open, as it is liable to be choaked with slag. 



I will now conclude this long discussion, with the hope that some 

 person may resume the subject whose situation can enable him to 

 j)rosecute these experiments, and give them that publicity which is 

 necessary to make them useful ; if it be found that the fumivore is 

 as effective when applied to a large fijrnace as we found it, the use 

 of it should be enforced, and not left to the mere option of the ma- 

 nufacturer. It is not a diminution of the liberty of the people to say 

 that no establishment shall be tolerated which incommodes its neigh- 

 bours, at least when the inconvenience can be prevented. Interest 

 is not always adequate to overcome prejudices, but it is not to be ex- 

 pected that they should be gratified at the expense of the health 

 and comfort of thousands. We hear every day some proof of the 

 necessity of legislative interference on this head ; in fact where a ma- 

 nufacturer does not avail himself of the resources which the wide 

 diffusion of knowledge places within his reach, he is to a certain de. 

 gree criminal, but he is doubly so where his neglect endangers 

 others. 



conds : the fact confirms my opinion of the defect of Roberton's apparatus, and I have ob- 

 served another to the same elTect ; all our fumivorcs smoke when the fire it kindling and, un- 

 til the roof of ihc fire-place is red hot. 



