322 FORCING GARDEN. 
the house proposed by Sir George Mackenzie was too high, 
in proportion to its length and breadth, and therefore 
recommended a smaller section of a sphere, with a greater 
radius. His dimensions are forty feet long, fourteen wide 
in the centre, and, including the front parapet, twelve feet 
high. The late Mr. Loudon, who, it is believed, was the 
first that actually erected hot-houses on this principle, pro- 
posed several subvarieties of form. He describes (Eincye. 
of Gard.) the acuminated semidome, the acuminated semi- 
globe, the semiellipse, and the parallelogram with curved 
roof and ends. With Mr. Loudon, we should certainly 
prefer the last mentioned. A considerable number of 
curvilinear houses have been erected in the southern part of 
the island, particularly as repositories for ornamental plants, 
such as in the Royal Gardens at Kew, Loddiges’ nurseries 
at Hackney, the London Horticultural Society’s Garden, 
the Manchester Botanic Garden, the Duke of Northum- 
berland’s at Syon House, and in various other private 
gardens. 
As far as we are aware, no extensive experimental in- 
vestigation of the comparative merits of curvilinear houses 
has hitherto been made. A writer in the Gardener's 
Magazine (vol. ii.) states that he found it necessary, dur- 
ing the summer months, to shade his pine-apples growing 
in such a house, from nine or ten o’clock in the morning to 
three or four in the afternoon, in order to prevent the plants 
from assuming a rusty tinge and unhealthy appearance. 
Another practical gardener complains (vol. v.) that “the 
circular roof concentrated the sun’s rays so immoderately 
that the tops of the grape-vines were scorched, even when 
the doors and ventilators at the back were open.” This, 
he says, was always the case in summer; and in winter, it 
was with difficulty, and only with the assistance of bast 
