gator of the division of chemistry of the Department of Agriculture, studied the weight of flour with respect to the relative humidity of the air. He exposed five lots of flour for 18 days and made at intervals 15 moisture tests on each flour. The flours varied in moisture content when the test began, from 7.80 to 13.68 per cent. The flour was exposed in a room with free access to the air, but protected by a screen from other influences than air.

The tests showed that the weight of the flour was dependent upon the relative humidity of the air. During the 18 days the relative humidity of the air varied from 34, the lowest, to 66.9, the highest. Taking 100 pounds as the weight at the commencement of the test, the weight of one sample, for instance, was 99.88 pounds when the humidity of the air was at 34, and 102.88 pounds when the humidity stood at 66.9. The gain in one sample of flour at the end of the test, when the humidity was 66.9, was 5.95 pounds, or 5.95 per cent. Others gained less, but all gained except the one with the 13.68 per cent of moisture, which weighed 99.35 pounds, having lost 0.65 of 1 per cent in the 18 days. All the flours except this last were exceptionally small in moisture content, while this one sample was unusually large in moisture content. The tests conclusively showed that flour will take on and part with moisture as the humidity of the air rises and falls.

J. T. Willard, analyst of the Kansas State Board of Health, in 1911, studied the variation in moisture content of sacked flour. In the experiment twenty-seven 14-barrel sacks were piled as closely together as possible in three layers of nine sacks each in an airy room, which during the winter months was heated to ordinary room temperature. Each of the sacks was weighed monthly for a period of 12 months. Very little loss was observed during the first two months. When the room was heated during the winter there was a steady loss which, however, was in part made up by slight gains during summer months. At the end of the period during which the room was heated, the average loss was 0.79 pound per 49-pound

sack.

The results of these experiments have been confirmed thousands of times by millers and by flour buyers who have noted decided difference in the weight of flour on arrival from the weight when shipped. In the enforcement of the food and drugs act many cases have been filed against millers for shipping short-weight flour. In some cases there has been no doubt as to the guilt of the miller, but in other cases the shortage was undoubtedly due to the loss of moisture during the period which elapsed between the packing of the flour and its weighing by the customer and inspector.

The tendency of flour to take on and lose moisture has recently been studied by C. H. Bailey, who, in a paper on "The hygroscopic moisture of flour exposed to atmospheres of different relative humidity," shows conclusively that flour responds readily to changes in the humidity of the surrounding air, the rate at which equilibrium in moisture content is approached depending apparently upon conditions of exposure. The author quotes other investigators, including Willard, Neumann, Guthrie and Norris, Sanderson, Swanson, Willard and Fitz, and Stockham, who have studied the changes in weight and moisture content of stored flour.

6379°-24- -6

Some idea of the variation in moisture content of flour milled from different varieties of wheat in different sections of the country is shown by the figures below, which are averages of some 800 cars of flour shipped during the first four months of this year.

The wide ranges in the moisture content of these flours, though they were milled under carefully controlled conditions, show most definitely the need for taking the moisture content of flour into consideration when determining flour weights. To fail to do so introduces the element of guesswork into the transaction. Fluctuating moisture contents have exactly the same effect on business as fluctuating money values or unstabilized exchange.

Stockham reports the moisture content of wheat, bran, shorts, and flour exposed in a "saturated" and "dry" atmosphere, but he did not employ any degrees of atmospheric humidity between these extremes. He found that a composite sample of flour exposed in a "still, saturated" atmosphere at a temperature of 23° C. reached a maximum moisture content of 28.74 per cent in 9.12 days, at which time it was moldy. In a saturated atmosphere at 0° C. a moisture content of 34.78 per cent was reached in 17 days, which he states was not the maximum.

Schollenberger, in Bulletin No. 1013 of the United States Department of Agriculture, refers to the "well-known fact that the normal moisture content of air-dry wheat is higher when stored in moist climates than when stored in dry climates." The author further defines the term "normal" as that point at which equilibrium is established between the moisture content of the wheat and the humidity of the air.

North Dakota Experiment Station Bulletin No. 120, 1917, on the capacity of wheat and mill products for moisture, says in part, "One of the well-known relationships of the moisture content of wheat is its approximate parallelism to the humidity of the atmosphere. The moisture problem would be relatively simple if all samples responded the same under like conditions, but unfortunately they do not, as they differ in their rate of change, or natural capacity. The capacity of wheat and its products for atmospheric moisture and water increases as the physical equilibrium between the component particles is approached."

Shelled corn loses weight in storage. In experiments conducted by the office of grain standardization, United States Department of Agriculture, corn held in the hopper of an elevator scale for 147 days

lost 5.6 per cent. J. W. T. Duvel and Laurel Duvel, in United States Department of Agriculture Bulletin No. 48, 1913, commenting on these tests, say, in part," There is unquestionably a natural shrinkage in commercial corn during transit and while in storage, which varies with the moisture content of the corn and the atmospheric conditions to which it is exposed."

In a personal communication from the Bureau of Plant Industry of the United States Department of Agriculture, my correspondent points out that "there is a loss in weight of apples, corn, hay, etc., when held in storage. This is due, to a large extent, to the loss of water. At the same time in many fruits and vegetables which are alive in storage there are changes, mostly catabolic, which take place more or less rapidly according to the character of the product or temperature at which it is held. The changes according to Van't Hoff are doubled or trebled with each 10° rise in centigrade. There are, of course, a great many changes in the physical conditions which are progressive and which are apparently influenced by the temperature conditions."

Bread as commercially sold to-day is either manufactured in standard sizes or sold under a declaration of its weight. It is probable that no food commodity on the market is so subject to fluctuations in weight as is bread. This is due to the fact that the moisture content of freshly baked bread is high, as well as to its composition, which is so largely starch, itself very hygroscopic.

In an address made to the weights and measures officials of Indiana at their 1924 conference, which was later printed in Baking Technology, I showed the loss of moisture in commercial bread baked and analyzed at the laboratories of the American Institute of Baking, after having been exposed to the usual temperature and humidity conditions of the room. The average moisture content of the bread one hour after baking of a series of 13 loaves which were exposed unwrapped, was 36.23 per cent; 24 hours after baking the moisture content was 29.20 per cent; 48 hours after baking the average moisture content was 24.93 per cent; 72 hours after baking the average moisture content was 21.59 per cent.

65.0 1.70 1.50 2.00

888888888888888

1.0 8.30

Similar studies were made of the loss in weight of different types of bread made with varying amounts of water; that is, with flour of varying absorption. The average loss in 24 hours of unwrapped bread varied from 1.03 to 1.53 ounces.

Bread wrapped in paraffin paper lost very little moisture. When heavily waxed paper was used the loss in 24 hours time on a 17ounce loaf was but 0.41 ounce. When wrapped in paper with a less dense paraffin coat, the loss was approximately twice as high, or 0.82 ounce.

From unpublished data furnished me by the chemist of a large bakery organization I quote the following figures: Loaves made from a lean formula, that is, containing no shortening or sweetened condensed milk, contained when taken from the oven an average moisture content of 42.91 per cent; 72 hours later, at the conclusion of the experiment, they contained but 34.97 per cent of moisture, a loss of 7.94 per cent. The average weight of the loaves when taken from the oven was 768.5 grams. At the conclusion of the experiment the average weight was 674.5 grams.

Similar studies made on bread with a rich formula containing 7 per cent of sweetened condensed milk and from 2 to 3 per cent of shortening showed an average moisture content on the freshly baked bread of 41 per cent, whereas 72 hours later at the end of the experiment the moisture content was but 33.08 per cent, a loss of 7.92 per cent of moisture. The average weight of these loaves when freshly baked was 806 grams. At the conclusion of the experiment the average weight was 710.5 grams.

But whatever the condition or weight of the bread (and this statement is true of all foodstuffs which gain or lose in weight under varying conditions of humidity and temperature) the change in weight is always due to loss or gain in water content. There is no change in the content of the food ingredients. A loaf of bread containing 38 per cent of water and weighing 16 ounces will contain just as much food for the maintenance of bodily efficiency after its moisture content has been reduced to 20 per cent, to 10 per cent, or to 0. The fuel value of a ton of coke is determined not by the weight of the coke but by the weight of the moisture-free matter. This is not entirely a true statement, for in the case of fuels containing water a certain number of British thermal units will be consumed in evaporating the moisture content. When, however, foods are burned in human metabolism it is not necessary to take this fact into account. The only point which interests us is the absolute quantity of food obtained.

In a bulletin of the Kansas State Board of Health, J. T. Willard very pertinently points out the necessity for an accurate basis for the determination of legal weights. He says, in part:

The change of weight to which commodities sold by weight are subject is a very important factor in the commercial world. The honest farmer who salts the cattle well that they may be encouraged to drink freely of water just before they go on the buyer's scales, and the milkman whose most effective ally is the pump, are but crude workers in a field of practice in which certain other producers are more skillful even if no more honest. It is obvious, too, that a dealer might suffer considerable loss by the natural drying of a commodity kept in bulk and sold by weight. On the other hand, it is not impossible that if he holds it under more humid conditions he may profit from an unearned increment due to the absorption of moisture.