A comparison of these tables with the results given for each pen in Table No. 3, will explain some of the variations from week to week, that would otherwise appear anomalous. For instance, in Table No. 3 the amount of food required to produce a pound of increase in Pen A, is 4.85 lbs. for the 22d week, 10.44 lbs. for the 23d, and 6.94 lbs. for the 24th week. Table No. 5 shows that Pig No. 1 gained 2 lbs. the 22d week, nothing the 23d week, and 2 lbs. the 24th week. Pig No. 2 gained 3 lbs. the 22d week, nothing the 23d week, and 1 lb. the 24th week. Pig No. 3 gained 5 lbs. the 22d week, 4.50 lbs. the 23d week, and 5 lbs. the 24th week.

Table No. 7 shows that Pig No. 3 made 1 lb. of increase from 4.40 lbs. of meal the 22d week, from 5.44 lbs. of meal the 23d week, and from 5.60 lbs. of meal the 24th week.

From this it is seen that the unusual amount of feed required to produce a pound of increase in Pen A, the 23d week, as recorded in Table No. 3, is caused by the irregular growth of Pigs Nos. 1 and 2.

The following Table No. 8, shows the return in increase of live weight for the meal consumed in Pen B, for five periods of four weeks each, beginning with the 9th week, and ending with the 28th week. The 29th week is not included, as it was thought best to make the periods of uniform length.

The derangement of the digestive organs of the pigs in Pen A, produced by the change of feed from milk to meal, as already noticed, would of course prevent any satisfactory comparison of results for the different periods, and they have accordingly been omitted from the table. From the 5th to the 8th week, inclusive, the feed in both pens consisted of milk and meal, in varying proportions, so that the results of the feeding during this period, cannot well be compared with the periods in which the feed was limited to a single article of diet.

TABLE No. 8.

TOTAL INCREASE, TOTAL FEED CONSUMED, FEED CONSUMED FOR EACH ONE HUNDRED LBS. OF LIVE WEIGHT, FEED REQUIRED TO PRODUCE ONE LB. OF INCREASE, AND THE GAIN PER CENT. IN PEN B, IN PERIODS OF FOUR WEEKS, FROM THE NINTH TO TWENTY-EIGHTH, INCLUSIVE.

The third column shows a marked decrease in the amount of feed consumed for each 100 lbs. of live weight, as the pigs approach maturity. Under favorable circumstances the decrease in consumption for each period, should be uniformly shown throughout the experiment. It will be seen that the total amount of feed consumed the second period, is over fifty pounds less than was consumed in either the first or third periods. The amount of feed consumed for each 100 lbs. of live weight, during the 2d period, as shown in the table, is therefore probably lower than it would be, with an average consumption of food for the period.

An examination of Table No. 1 will show that Pig No. 5 lost weight the 17th week, so that the amount of feed consumed for each 100 lbs. of live weight, for the 3d period, as given in the table, is probably higher than it would be, with a uniform rate of increase. The fourth column of Table No. 8, shows a rapid increase in the amount of feed required to produce a pound of live weight, as the pigs improve in condition. The average amount of meal required to produce a pound of increase in live weight, is 4.55 lbs., for the whole period of twenty weeks; but 3.81 lbs. of meal is required to produce a pound of increase during the 1st period, while in the 5th period 5.98 lbs. of meal is required to produce the same result.

During the first four weeks of the experiments, the milk consumed for each 100 lbs. of live weight, diminishes rapidly as the pigs increase in size, while the milk required to produce a pound of increase as constantly increases.

In the following tables, (Nos. 9 and 10,) are given the amount of milk consumed per week, for each pound of live weight, and the amount of milk required to produce a pound of increase per week, for the first four weeks of the experiment.

The results of the experiments in pig feeding in 1866, (Report of the Secretary of Michigan State Board of Agriculture, pp. 59 to 62,) are likewise included in the tables, for convenience of comparison. The average results of the two experiments are also given.

3.94 80.93 3.67 86.75 2.93 27.49 2.45 27.85 3.07 50.76

AVERAGE AMOUNT OF MILK CONSUMED FOR ONE POUND OF INCREASE.

In addition to the milk consumed the 4th week, in the experiment of 1868, one pound of oats was fed in the course of the week. As this extra feed was not sufficient to materially change the results, it was not taken into account in making the tables.

The similarity of the results in these experiments is very striking. Some exceptional variations, however, require an explanation that is readily suggested, by reference to the history of the experiments. It will be seen that more milk was consumed, on the average, for each pound of live weight, during the 2d, 3d and 4th weeks of the experiment of 1868, than in the same periods in the experiment of 1866, while during the first week, a larger amount was consumed in the experiment of 1866. This variation is undoubtedly owing to the mode of feeding practiced in the different experiments during the first week. In 1866 the pigs were fed from a can furnished with a "rubber nipple," which was apparently a satisfactory arrangement, so far as the pigs were concerned. In 1868, the pigs were fed from a trough during the first week, and as they were just learning to drink, they probably did not consume as much in proportion to their live weight as they otherwise would have done.

Had the same method of feeding been adopted in 1868 that was practiced in 1866, I have no doubt that the quantity of milk consumed for each pound of live weight during the first week, would have been greater than it was during the second week, as shown in table No. 9.

It will be observed that the amount of milk required to produce a pound of live weight, in each of the experiments, is greater during the 3d week than it is during the 4th week. This exceptional result is easily explained in both cases. In 1866, a "derangement of the digestive organs was observed during this week, as shown in a tendency to constipation," a larger amount of feed was thus required to produce a pound of increase, than would otherwise have been necessary. In