All night long and all the next morning men were continuously at work pouring melted fat over the surface of the slowly-rotating carcass of the Warwick ox, skilfully directing a ladleful to any part that indicated undue dryness. By this device the meat is more or less completely enveloped in a varnish of hot melted fat, which assists in the communication of heat, while it checks the evaporation of the juices. In such roasting the heat is partially communicated by convection through the medium of a fat-bath, as in stewing it is all supplied by a water-bath. I have made some experiments wherein this principle is fully carried out. In a suitably-sized saucepan I melted a sufficient quantity of mutton-dripping to form a bath, wherein a small joint of mutton could be completely immersed. The fat was then raised to a high temperature, 350° (as shown by Davis’ _tryometer_, presently to be described). Then I immersed the joint in this, keeping up the high temperature for a few minutes. Afterwards I allowed it to fall below 200°, and thus cooked the joint. It was good and juicy, though a little of the gravy had escaped and was found in the fat after cooling. The experiment was repeated with variations of temperature; the best result obtained when it was about 400° at the beginning, and kept up to above 200° afterwards. I used loins and half-legs of mutton, exposing considerable surface. I find that Sir Henry Thompson, in a lecture delivered at the Fisheries Exhibition, and now reprinted, has invaded my subject, and has done this so well that I shall retaliate by annexing his suggestion, which is that fish should be _roasted_. He says that this mode of cooking fish should be general, since it is applicable to all varieties. I fully agree with him, but go a little further in the same direction by including, not only roasting in a Dutch or American oven _before_ the fire, but also in the side-ovens of kitcheners and in gas-ovens, which, when used as I have explained, are roasters—_i.e._ they cook by radiation, without any of the drying anticipated by Sir Henry. The practical housewife will probably say this is not new, seeing that people who know what is good have long been in the habit of enjoying mackerel and haddocks (especially Dublin Bay haddocks) stuffed and baked, and cods’ heads similarly treated. The Jews do something of the kind with halibut’s head, which they prize as the greatest of all piscine delicacies. The John Dory is commonly stuffed and cooked in an oven by those who understand his merits. The excellence of Sir Henry Thompson’s idea consists in its breadth as applicable to _all fish_, on the basis of that fundamental principle of scientific cookery on which I have so continually and variously insisted, viz. the retention and concentration of the natural juices of the viands.