sistent and reliable results than we could possibly secure by attempting to make cheese with the small amounts of milk at hand."

The average data for each breed, based on 1 cow and a period of lactation of 10 months, are given in the following table:

Average yield and value of cheese per cow for one period of lactation.

The relation of the fat in the milk to the yield of green cheese is shown separately for each cow in each period of lactation. This shows "that there is more or less variation in the relation of fat to cheese yield, but that the results are fairly uniform within certain limits." To show this relation more plainly the cows are grouped according to fat content of the milk in the following table:

"An examination of the foregoing tabulated summary reveals some very interesting facts:

"(1) In milk containing less than 3 per cent of fat the cheese yield for 1 lb. of fat is much higher than the average.

"(2) In milk containing from 3 to 4.5 per cent of fat the amount of cheese made for 1 lb. of fat is quite uniform, and the average is nearly the same as that secured in all our previous work at cheese factories. These are the limits within which the per cent of fat of most cheese factory milks fall. These results agree in a surprising way with those secured in our most extended investigations, especially when we consider that these are results obtained with individual cows instead of herds.

“(3) In milk containing 5 to 6 per cent of fat or more the amount of cheese made for 1 lb. of fat was noticeably less than in milk containing 4.5 per cent of fat or less, but in milk containing 5 per cent of fat or more up to 6.13 per cent of fat we have very uniform results, so that these data point to the fact that milk containing 6 per cent or more of fat will yield as much cheese per pound of fat as will milk containing less than 6 per cent of fat down as low as 5 per cent."

Referring to the first table above it will be seen that the amount of milk required per pound of cheese was least for the Jersey and greatest for the Holstein-Friesian milk. Taking these extremes the effect of paying for the milk by weight and on the basis of fat content are compared.

Results of paying for milk by different methods.

"A study of the above figures leads to the following statements: "(1) Comparing results of paying on basis of fat and on basis of actual cheese yield with these two herds, whose milk differs as the milk of no two herds of cheese factory cows differs for a whole season's average, we see that the poorer milk receives 8 cts. more per 100 lbs. on the basis of actual cheese than on basis of fat, while the milk richer in fat receives that much less.

"(2) If the cheese made from the richer milk sold for one-half cent more per pound than the other, then the fat basis would be absolutely fair. Would the cheese made from the richer milk differ in composition enough to make the difference of one-half cent per pound? The following would represent the composition of the cheese made from these two milks:

"It is clear from these figures that the cheese made from the two kinds of milk differs much in composition, since that made from the richer milk contains nearly 6 per cent more of fat. Allowing that the cheese was made with equal skill, it is safe to say that the richer cheese would sell easily for a half cent more per pound.

"(3) If the milk of each herd was paid for by weight alone, then each would receive the same amount, or one-half the total receipts, which would give $1.12 to each milk.

"(4) If the cheese sold for different prices, as indicated above, then the fat basis would be strictly just and, as between this method and the method of paying by weight alone, the poorer milk would receive 28 cts. more and the richer milk 28 cts. less than each ought per 100 lbs. of milk. But, granting that the cheese would all

sell at one price, what would be the comparative justice of the fat basis and the basis of weight alone? On the basis of the fat the richer milk would receive 8 cts. more per hundred than its cheese yield would entitle it to. On the other hand, on the basis of weight alone, the poorer milk would receive 20 cts. more than it onght. "Where the differences in composition of milk are less the chance of doing any possible injustice in paying on basis of fat becomes proportionately less or practically disappears."

The average composition of the cheese made from the milk of each breed is tabulated.

Investigations of the milk of 16 thoroughbred Dutch cows during one period of lactation, K. HITTCHER (Landw. Jahrb., 23 (1894), No. 6, pp. 873-967).-In 1891 Prof. W. Fleischmann published a record of observations on a herd of 145 Dutch cows bred in East Prussia, covering one year with special reference to the milk of 16 cows which were in the same stage of the milking period (E. S. R., 3, p. 424). The present paper reports a continuation of these studies under the advice. of Professor Fleischmann by his assistant in the former investigation, the period covered being Notion. As in the previous year, samples of the milk were sent three nes a week to Professor Fleischmann's laboratory at Königsberg for analysis.

Observations on the entire herd. There were at the royal domain of Kleinhof-Tapian during the year 144 cows, the average number in milk at all times being 124. The BR Ruth of the period of lactation was 314 days. The average yield of milk per cow was 2,786 kg. during the year, or 8.81 kg. daily.

The cows were placed on stall feeding with the beginning of the investigation, October 1, and, as was customary, were divided into 2: general lots, lot A including all the cows which had not reached the last quarter of their period of lactation, and lot B those which were old in milk or dry or were to be fattened. Lot A received per cow daily 4.25 kg. of grain, consisting of 1 kg. of meal (rye, oats, barley, and peas), 1 kg. of coarse wheat bran, 1 kg. of malt sprouts, 0.75 kg. of sunflower cake, 0.25 kg. of peanut meal, and 0.25 kg. of palm-nut cake, to which was added at different times hay, straw, potato slump, and grass and serradella silage. Besides this, each cow had 30 gm. of salt and 30 gm. of phosphate of lime per day. Lot B received per head 5 kg. of hay, 2 kg. of straw, 1.25 kg. of wheat bran, and 0.75 kgof sunflower cake. Early in May both lots were turned to pasture and remained there to the close of the period covered.

The milking commenced at 4 in the morning and between 4.30 and 5.30 at night, so that the longest interval was always during the day. At all seasons of the year the milk yield was larger at night, the average difference being 0.4 kg. per cow daily. The morning's milk was 0.065 per cent richer in fat and 0.046 per cent richer in total solids than the night's milk, and the percentage of fat in the total solids was 0.43 higher; but the specific gravity and the solids-not-fat were both slightly lower in the morning's milk.

The following table shows the average composition of the mixed milk of the whole herd for the year, and the ranges in composition of the herd milk for the day (mixed morning's and night's milk), and for separate milkings (morning's or night's):

Average and range of composition of herd milk for 1 year.

The minima observed for the mixed herd milk for 3 years at this estate, always on thoroughbred Dutch cows, were as follows:

In 1889-90 the variation in composition was greater during the winter than during the summer, which is the reverse of the case in the two preceding years. For 10 days before the cows were turned to pasture the average per cent was 9.48 kg. of milk with 3.23 per cent of fat, and for 2 weeks after the change it was 11.35 kg. of milk with 3.199 per cent of fat. That is, when the cows were turned to pasture the milk yield increased perceptibly, but the fat content decreased slightly. The latter was relative as well as absolute, since the percentage of fat in the total solids was 27.38 before and 27.08 after pasturage began. In the two preceding years not only the milk yield, but both the absolute and relative fat content increased in changing from stall to pasture feeding. It is suggested that the difference may have been due to the cows receiving a richer stall feeding in 1890. The maximum milk yield was reached May 28, being 12.24 kg. per cow.

Observations on 16 cows.-As in the previous year, cows were selected which calved at nearly the same time, most of them in March or early in April. Five of the cows had been used the previous year; 6 were prize cows. The length of the milking period of these cows ranged from 270 to 461 days, but for most of them it was not far from -300 days.

The quantity of colostrum given at the first milking after calving varied widely, ranging from 1 to 7.9 kg., and the fat content equally widely, from 1.73 to 7.75 per cent. In the following table the cows are

divided into 2 groups, according to the fat content of the colostrum, and this compared with the fat content of the milk afterwards given (average for whole period of lactation):

Fat content of colostrum and of milk of 13 cows.

Where calves were allowed to run with the cows it was expected that the little milk that was left would become rich in fat, as is the case with the last part of the milking; but quite the contrary was often observed, which suggests to the author that the calf can get the fat from the udder more thoroughly than hand milking.

Striking instances are mentioned of the disturbing influence of removing the calf. A cow whose calf was taken away in the forenoon gave that night only 1.7 kg. of milk with 0.45 per cent of fat, and the the following morning 5.9 kg. of milk with 0.72 per cent of fat; 24 hours later the yield was 7.5 kg. with 5.6 per cent of fat. Another cow which had been giving milk with 4.67 per cent of fat, gave 2.4 kg. of milk with only 0.65 per cent of fat the night after the calf was taken away, and 7.1 kg. with 1.27 per cent of fat the next morning; but the following evening the milk was normal. This was noticed with several cows, but in the case of one the milk was richer. Cows in heat frequently decreased in milk, and the milk was very poor in fat (0.17 and 0.47 per cent), though in one case the milk became richer. The composition of the milk at the end of the milking period, when the cows were being milked only once in 3 or 4 days, is given as follows:

Milk produced when the cows were practically dry.