off. On March 24 all the animals were weighed, after a limited morning feed, their fasted live weights were taken next morning, and they were slaughtered, the carcass weights being recorded by Mr. Forrester, the farm manager.

It will be noticed that No. 1 (Lot I.) did not gain as much as the others of this set. In explanation, it may be pointed out that when put up he was (along with Nos. 7 and 10, i.e. one in each lot) one of those most forward in condition, but throughout the experiment he had always shown a nervous, restless disposition, and when killed ultimately he was found to have an abscess on the diaphragm, while imbedded in this was a piece of wire, and a couple of 2-inch nails were also found in the second stomach. Nevertheless, he was the "ripest "bullock of the four of his lot. Whether this animal be taken into consideration or not, the general conclusion is the same, viz., that of the additional foods bean meal undoubtedly did the best in respect of fattening. During the second period, however, the increase from the giving of decorticated cotton cake with linseed cake was very satisfactory. Bullock No. 12 (hay lot), which had been thrown out of the experiment after the first weighing on January 28, was closely examined after being slaughtered, to find, if possible, the cause of his not having done well. Beyond ascertaining that the liver contained a considerable number of flukes, there was no unsoundness apparent.

As regards the influence of the respective foods, undecorticated cotton cake, bean meal, and long hay, in checking tendency to "scour," close observation throughout the experiment led to the following conclusions :—

1. That well-ripened mangels given in moderate quantity, say, 28 lb. to 30 lb. per head daily, can be quite well fed to fattening bullocks in the early stages, in place of swedes, if along with them be given either common cotton cake, bean meal, or a plentiful supply of long hay.

2. That as soon as the quantity of mangels reaches 35 to 40lb. per head daily, "scouring " will probably appear.

3. That when this is the case the giving of long hay in quantity will prove no palliative.

4. That both undecorticated cotton cake and bean meal will soon check the tendency to "scour," and that, of the two, bean meal proves the more satisfactory by giving the larger increase in live weight.

From general observation made during this experiment it was noticeable that the feeding with mangels produced, in the early stages at least, a feverishness about the animals which is not the case when swedes are similarly fed. The animals, however, always took the quantity supplied to them with much avidity. It was clear, moreover, that mangels can, if there is necessity, be quite well substituted for swedes even in the early stages of feeding, provided a small quantity only-say, not exceeding 30 lb. per head daily for 3-year old bullocks-be given at first, and that along with these cotton cake or bean meal be used, Further, this quantity of mangels may

Carcass

Average gain per head daily, 1st period (14 sheep)

2nd

1 12 sheep only.

be increased gradually up to 45 lb. or 50 lb. as feeding goes on, without doing any harm.

Table I., p. 563, sets out in detail the weights recorded at the different periods, the gains, carcass weights, &c.

SUPPLEMENTARY EXPERIMENT WITH SHEEP ON MANGELS.

While the above experiment with bullocks was in progress a pen of 14 sheep-Hampshire and Oxford cross-ten months old, and similar to those used in the sheep-feeding experiment with gorse recorded subsequently (p. 567), was set aside on December 20, and to these sheep mangels were given, the sheep having until then been on grass with just a few swedes thrown out to them. From December 20, when they were weighed and penned on the arable land, the sheep received mangels only as succulent food in place of swedes. The additional dry foods given them were lb. linseed cake and lb. undecorticated cotton cake per head daily, with hay chaff ad libitum. The sheep were kept on mangels until April 9, and then killed. Two sheep of the number were removed in the course of the trial, one (No. 2) dying on March 26 from inflammation of the lungs, and another (No. 8) was killed on April 6 and found to have diseased kidneys. There was, however, no "scouring" effect whatever produced by the feeding of the mangels, and it would appear, therefore, that mangels can quite well be fed to sheep from the commencement of root feeding, in place of swedes, if about lb. per head daily of undecorticated cotton cake be given along with the food.

Attention has been directed to the harm that may be caused to male sheep by the feeding of mangels freely, there being a tendency to act on the kidneys, produce increased secretion of urine, and, in extreme cases, the formation of crystals in the urethra. In the above experiment one-half of the fourteen sheep were male, the other half female. The sheep that died on April 6 was a male sheep, and the kidneys were, as stated above, certainly affected, though the bladder was not.

Table II. (p. 565) gives the weights of the sheep at different periods, and may be taken in conjunction with the similar table (p. 573) of the experiment on sheep feeding with gorse. It will be noticed that there was much greater irregularity at first in the way that the sheep n the mangels did, as compared with those fed on swedes.

13 Hanover Square, W.

J. AUGUSTUS Voelcker.

567

GORSE AS A FOOD FOR SHEEP.

AT the Woburn Experimental Farm there is, on the hillside of one of the fields of the farm, a very poor and sandy piece of land upon which no satisfactory crop could be got to grow. On this, as an experiment, gorse was drilled in May 1897, the variety tried being that known as "French" gorse. No manure was used, and a barley crop was first put in, gorse being drilled between alternate rows of the barley, and thus in rows about 18 in. apart. The barley was once more a poor crop, but the gorse came up fairly regularly; the plant was just visible, about 2 to 3 in. high, through the winter, and it began to shoot out at the end of April 1898. Two horse-hoeings during summer were the only cultivation required, and by October the gorse was ready for cutting. The crop cut during the winter weighed 11 tons per acre.

It was decided, as the gorse came all right, to see how it would do for sheep-feeding, and, chiefly, how far it would replace roots, and so come in usefully in the event of a failure of the root crop or a short supply of roots. There being no machine on the farm for preparing the gorse, it had, after cutting, to be sent some distance off to a farm where there was a proper gorse-crushing machine. In consequence of the distance the gorse was only sent twice a week.

Two lots of sheep, Hampshire and Oxford cross, about ten months old, fourteen in each pen, were selected, and to both lots linseed cake and hay chaff were to be given. To one lot roots were intended to be supplied, as much as the sheep would take, while to the other gorse was to be fed ad libitum, and the rest of the diet made up, if necessary, with roots.

The sheep were weighed on November 30, 1898, about 10 A.M., after receiving, first thing in the morning, a limited ration of 6 lb. per head of swedes and lb. of mixed linseed cake and hay. The sheep were at this date valued at 37s. 6d. apiece. The experiment then began, and the sheep in Pen I. had at first about 20 lb. of swedes per head daily, while those in Pen II. took about 11⁄2 lb. of gorse per head daily, and the additional swedes were restricted in amount to 11 lb. per head daily. It was found, however, that the sheep did not eat nearly as much gorse as was expected, and the additional roots did not prove enough. So, after a week, the roots had to be increased to 14 lb. per nead daily. Still this was not

enough, and, the sheep eating less than 2 lb. of gorse, the roots, after a fortnight, had to be further increased to 20 lb.

The gorse when fresh was taken very readily, but, the weather being very windy, the gorse dried quickly, and then seemed to get too dry and prickly for the sheep. About January 1, however, it was found that, by putting the gorse through the machine twice instead of once as previously, it was much better and was not nearly so prickly when dry. When brought in fresh the sheep would eat as much as 2 lb. each daily of the gorse, but as it got staler they would not consume it so readily. The best way of dealing with it was found to be to spread it, when brought fresh from the machine, to a depth of 3 to 4 in. on a stone floor. This kept it fairly moist. Damping the gorse with water before using it was tried, but was not found to answer well.

On January 25 the two lots of sheep were weighed, and the weights with respective gains are given in Tables II. and III., on pages 572 and 573.

Summarised the results at this time were :

Thus, practically, 1 lb. of gorse took the place of 3 lb. of swedes. By looking at the details in Table II. (p. 572) it will be noticed that one sheep (No. 10) in Pen I. gave no increase in live weight at the end of the first period. This sheep had a bad cold, which afterwards developed into a kind of malignant catarrh; the subsequent weighing of March 13 gave an actual loss in the case of this sheep,