CALENDARIAL MEMORANDA FOR DECEMBER. 
477 
built at the bottom of the house, whence a drain is laid to carry away 
the water to some lower level. This drain has a water-trap formed in 
some part of it, to prevent the ascent of air to the ice. 
The entrance to the house is a brick-built arched passage, about five 
feet high, opening into the house a little below the arched crown, 
where an inner door is placed. The length of the passage is more or 
less, according to the declivity of the ground ; but this, as well as the 
crown of the house itself, is covered with earth three or four feet thick, 
and usually planted with evergreen shrubs. 
Immediately in front of the outer door there is an enclosed area, 
fifteen feet square, walled round, and paved level with the floor of the 
passage. This is for breaking the ice in ; and in order that the carts 
may draw up close, to shoot the ice into the area, the surface on the 
outside is raised nearly level with the walls. 
Such is a brief account of the construction and manner of filling ice- 
houses; and it may be remarked of them that, though they preserve 
the ice pretty well, they are very inconvenient. When a couple of 
pailfuls of ice are wanted, nearly the whole of the straw in the pas¬ 
sage must be removed, especially if the inner door be on hinges; and, 
after midsummer, a ladder must be used to get down to the ice, which 
then stands like a frustrum of a cone in the centre of the building. 
On the top of this the man must stand to get the ice, and must be 
cautious lest he slip off among the loose straw round between the ice 
and wall, which, if he did, he would have some difficulty to extricate 
himself from. When he has got what ice he wants, the whole must 
be closed again ; and this labour must be repeated every morning, 
perhaps, for months together. 
Having seen ice-houses in many situations, and having often felt the 
inconvenience of their construction, we have long entertained the idea 
that they might be considerably improved, without much, if any, addi¬ 
tional expense in building. We have always thought them too deep, 
not only from the difficulty of getting out the ice, but because the 
deeper the excavation, the more liable is it to be affected by land- 
springs, as well as the heat of the earth, which always increases the 
deeper we descend. This is the more probable, inasmuch as ice-houses 
formed on the surface, yea, even on an elevated surface, if sufficiently 
defended from the influence of air, preserve ice as well as, if not better 
than, the subterranean houses. 
We would, therefore, propose the following plan of an ice-house, 
which we think would combine the facilities of management with its 
efficiency as an ice-store. Choose the north side of a hill or wood, as 
