JReirospective Criticism . 229 



of it. If any form is better, it is the inverted cylinder, and its being made 

 to resemble a limekiln in form : this may be worth the consideration of Mr. 

 Rogers and the manufacturer ; but I suggest it with diffidence, doubting whether 

 the theoretic calculation of its advantage would be followed by a correspond- 

 ing practical result. There is one point to which the attention of the manu- 

 facturer should be especially called, which is the cleansing. Mr. Rogers 

 states that the boiler can be taken to pieces, and there is a valve in the plan 

 for the purpose of cleaning it. Both these advantages ought to be rendered 

 as perfect as possible in practice, and to be within the power of any country 

 workman to manage. Were this attained, I see no plan likely to compete 

 with that of Mr. Rogers for simplicity and facility of working, on a small or 

 even on a rather large scale. I should like to see an account of the absolute 

 quantity of water a 15 or 18 inch boiler would heat to 200°, which Mr. Shewen 

 could easily give ; and I agree with the proposition that the 4-inch pipe is the 

 best as a general standard of reference and comparison. — W, Darlington. 

 March 10. 1840. 



Mr. Fen7i's Mode of Ventilating, Sfc. — Why did you not mention your 

 own experiments on heating by hot air ? I know the house in Dickson's nur- 

 sery at Edinburgh, in which you tried these experiments [in 1804], This 

 house is standing there now, or rather I saw it there in 1837. Mr. Rogers's 

 pit is a most excellent one, and is just as useful for keeping up a circulation of 

 cool air in cold pits in damp weather, as for circulating hot air. How singular 

 that I did not find out this system many years since, when I used to lay iron 

 pipes through the centre of bark beds in Mr. Knight's manner, to get out 

 heat from the centre for assisting the heating of the house, and to keep down 

 violent bottom heat. I had one end of the pipe out through the brickwork of 

 the tan bed at the level of the path, and the other end just above the tan at 

 the farthest corner. I recollect the draught of air going in at the lower end 

 would put out a candle in an instant. How stupid of me not to have found 

 out Penn's system by this simple contrivance! I think you might notice in 

 the Gardejier's Magazine how easy it would be to put the air in any pine stove 

 or frame in circulation, by passing two small tubes diagonally through the 

 bark bed; the lower end to be level with the path at each corner of the bed 

 in front, and the pipes to come out or rise through the back corners of the 

 ■bed, and be carried up near to the glass. — D. Beaton, Kingsbiuy, March, 

 -1840, 



The Conservatory at Chatsworth. (p. 103.) — I received 3'our note of 7th 

 January, and in reply I will take its separate " counts," one by one. 

 Jirst, then, you say, " As to a rood of glass covering a rood of ground, that 

 is impossible if any slope is allowed ; and as the most general slope for 

 a large house is, or ought to be, not very far from 45", your calculation 

 will be considerably far from the truth." You will see a perch of glass 

 roofing a perch of ground ; that is, a perch of glass lights, heats, &c., a 

 corresponding perch of plant-stage, or vine-trellis, or bark bed for pines ; in 

 a range of small houses glazed in wood, near the north wall of the kitchen- 

 garden at Syon. [A sketch of this house was sent, but we consider it 

 •unnecessary to engrave it, as the description given below answers ever pur- 

 ,pose required for Mr. Forsyth's object.] " The general slope " (to use your 

 own words) " ought to be not far from 45° :" a fruiting pine-house, when 

 thus constructed, and with upright front glass (as perhaps you will say it 

 " ought to have "), will be not far from bearing the following proportions ; 

 say 12 ft. wide, 15 ft. high, with 3 ft. 6 in. of upright glass in front. The 

 length of glazed work in the cross section of such a house will be nearly 

 20 ft. ; and supposing the house 50 ft. long, it will contain 1000 ft, besides 

 gable-ending ; and supposing this good old-fashioned house to have 3 ft. all 

 j"ound, occupied as fireways and footways, including the walls of the bark 

 bed, it will leave a clear pine bed only 6 ft. wide, which multiplied by 50, the 

 .lengtlf of the house, gives 300, which is not one third, but as 3 to 10, of what 

 I am prepared to show can be got by the following arrangement : — 



