COMMON MODE OF GLAZING ROOFS.
"In the construction of glass-houses requiring much light, there always appeared to me one important objection, which no person seemed to have taken up or obviated; it was this. In plain lean-to or shed roofs, the morning and evening sun, which is on many accounts of the greatest importance in forcing fruits, presented its direct rays at a low angle, and, consequently, very obliquely to the glass. At those periods most of the rays of light and heat were obstructed by the position of the glass and heavy rafters, so that a considerable portion of time was lost both morning and evening; it consequently became evident that a system by which the glass would be more at right angles to the morning and evening rays of the sun would obviate the difficulty, and remove the obstruction to rays of light entering the house at an early and late hour of the day.
METHOD BY RIDGE-AND-FURROW.
"This led me to the adoption of the ridge-and-furrow principle for glass roofs, which places the glass in such a position that the rays of light in the mornings and evenings enter the house without obstruction, and present themselves more perpendicularly to the glass at those times when they are the least powerful; whereas at mid-day, when they are most powerful, they present themselves more obliquely to the glass. Having had this principle fixed in my mind, and being convinced of its importance, I constructed a pine-house in 1833 as an experiment, which still exists unimpaired, and has been found fully to answer the purpose.
"In 1834 I resolved to try a further experiment on a larger scale, on the ridge-and-furrow principle, in the construction of a green-house of considerable dimensions, which also remains and answers admirably. For this building I made a still lighter sash-bar than any I had previously used; on which account the house, when completed (although possessing all the advantages of wood), was as light as if constructed of metal. The whole length of this structure is 97½ feet, and its breadth 26 feet; the height at the back is 16 feet 9 inches, and in the front 12 feet 3 inches. A span so large as 26 feet could not be safely covered with a roof constructed in the ordinary way, unless the sash-bars were stronger, and the assistance of heavy rafters and numerous supports was afforded. The house presents a neat and light appearance, and consists of 15 bays, and pediments in front, supported by 16 slender reeded cast-iron columns. Whilst it makes an admirable green-house, it is also an economical building; for, at the period of its construction, notwithstanding the heavy tax on glass (since removed), it only cost at the rate of twopence and a fraction per cubic foot. At the present time, considering the change in the price of material, and the removal of the glass-tax, it could be constructed at a considerably smaller amount.
"Having in contemplation the erection of the Great Conservatory in its present form, it was determined, in 1836, to erect a new curvilinear hot-house 60 feet in length and 26 feet in width, with the elliptical roof on the ridge-and-furrow principle, to be constructed entirely of wood, for the purpose of exhibiting how roofs of this kind could be supported. The plan adopted was this: the curved rafters were composed of several boards securely nailed together on templets of wood cut to the exact curve; by this means a strength and firmness were obtained sufficient to support an enormous weight.
"In 1837 the foundations of the Great Conservatory were commenced; and in constructing so great a building it was found desirable to contrive some means for abridging the great amount of manual labour that would be required in making the immense number of sash-bars requisite for the purpose. Accordingly, I visited all the great workshops in London, Manchester, and Birmingham, to see if anything had been invented that would afford the facilities I required. The only apparatus met with was a grooving-machine, which I had at once connected with a steam-engine at Chatsworth, and which was subsequently so improved as to make the sash-bar complete.
