I do not think an incubator can ever be complete unless it is a double-cased machine. It requires that to effectually resist thermal changes. Years of careful experiment, and of experience in the competitive show room have convinced me of the truth of this. Extreme cold will affect the uniformity of heat in the egg-chamber of single-cased machines. Imagine if you can a single-cased machine constructed of five-eighth inch stuff, with a temperature of 103 degrees inside, and that of freezing outside. How can the eggs at the extreme corners and the thin cold sides of that machine be as warm as those in the centre? Of course that difference does not exist in warm weather, but that is the time when incubators are usually let alone and the business is given up to the old hen. Now, I say this frankly, as much for the benefit of incubator manufacturers as for their customers. I have letters every day from parties ordering circulars and saying that they have used from one to three machines of different makes, denouncing the machines and their makers in the most emphatic terms as frauds. Now, this is all wrong; one-half of the time you will find that it is the purchasers, not the makers, who are at fault. There are probably just as many honest incubator makers as there are in any other branch of business. But there is such a thing as a man being honest and yet ignorant. Many of the manufacturers of incubators know very little about the first principles of artificial incubation. They have the idea that a simple heat regulator is all that is necessary to insure the success of an incubator, when in reality it is only one of the many requirements. I will enumerate some of the most essential points, viz.: heat regulation; uniformity of heat in egg-chamber; absolute control of heat by the operator on any given egg-tray; automatic moisture supply; accurate thermometers; thorough construction and good material to avoid warping and shrinkage, together with a safe lamp adjustment. There are many other minor points which will come up with care of machines. I am often asked, "Why do so many fail to hatch with incubators?" I will answer by saying: "Not because it is difficult; for I have always found it a far more difficult thing to grow ducks and chicks successfully after they are hatched, than it is to hatch them." Doubtless everyone knows that an incubator, different from other machines, must run three weeks continually night and day, (and when filled with duck eggs, four weeks,) and preserve an even temperature all the while.