DESIGN. • 37 



without regard to anything excepting the caprice or fancy of the 

 architect or builder. But, as general principles, we regard them 

 as unsafe and dangerous, were they to be practicall}?- acted upon 

 in the Southern States. Suppose, for instance, the roof should 

 be laid at right angles to the sun in mid-summer, — as is some- 

 times done in England, upon this principle, — then the conse- 

 quence would be that his rays would be unendurable by any 

 species of vegetation. The mid-summer sun, even in the lati- 

 tude of Baltimore, (39° 45',) falling on a transparent surface, 

 at right angles to the impinging rays, would scorch vegetable 

 forms, and dry them up in a few hours. 



It is, therefore, absolutely necessary that the exercise of this 

 principle be limited to northern latitudes, where it is indispensa- 

 ble to economize the sun's heat, for the purpose of accelerating 

 the maturation of fruits. It may, also, be applied in more 

 southern latitudes, when all the warmth of the sun's rays is 

 required early in spring ; and, therefore, if the principle be 

 applied south of the 40° of latitude, it should be taken when 

 the sun is at its very lowest altitude, otherwise the pitch of the 

 roof will be too flat for the months of summer. 



We are decidedly of opinion — and this opinion is fully con- 

 curred in by some of the most learned and skilful gardeners in 

 the country — that a great deal of error is committed in the 

 pitch of hot-house roofs ; probably more than four fifths of them 

 are made too flat; their angles of elevation are much too small 

 for the climate ; and yet, notwithstanding the fierce heat of our 

 perpendicular sun in summer, this practice is daily persisted in. 

 One would suppose that the scorching of vine-leaves, peaches, 

 and other plants, would convince people of the impropriety of 

 erecting their hot-house roofs at right angles to the sun's rays 

 in any of the summer months : and yet we know some of the 

 finest graperies in this country on angles of about 20^. If we 

 consider how very few of the rays are reflected by the glass, as 

 its plane approaches a perpendicular to the sun's altitude, and 

 how many are reflected as the angle of incidence is increased, 

 we will then have some notion of the advantage of increasing 

 the obliquity of the roof. 



The annexed table will show the number of rays reflected 



