The diameters should in no case be less than those given below:

(f) LENGTH MEASURES.-The yard measure should be made of well-dried wood with metal ends, or entirely of metal, or of other material of which the form and dimensions remain reasonably permanent under normal conditions. It should be subdivided into inches and their fractions, and also into the customary fractional subdivisions of the yard, i. e., halves, quarters, eighths, and sixteenths.

The tape should be of steel, or of wire-woven cloth when such construction gives it sufficient strength and permanency. At least 1 yard of this tape should be subdivided as above.

5. The Use of the Household Set in Checking Amounts of Commodities

The method of using this test set of weights and measures is quite simple, and the proper use will in most cases be evident to the housewife. Only a few suggestions upon the less obvious points, therefore, need be made here.

(a) USE OF SCALE, IN GENERAL.- The scale must be handled carefully and be kept clean and dry. The scale must also be kept in balance, otherwise every indicated weight will be incorrect. The ordinary counter beam scale is in balance if the beam comes to rest midway between the two stops which limit the swing. On a scale with a reading face, such as a spring scale, the indicator should point exactly to a definite and clear zero graduation.

Upon nearly every scale means of adjustment are provided by the manufacturer. Thus, a counter balance of the beam type is adjusted by turning the adjusting screw or balance ball, or by adding weight to or subtracting it from the shot cup under the

pan, or by altering the amount of balancing material wherever placed. Some spring scales may be adjusted as described. Others are adjusted by loosening a screw in a slot in the indicator, turning the latter on its pinion to the proper position, and then tightening the screw. If the scale has a glass face, this must be removed before the adjustment can be made. A properly constructed scale will rarely get out of balance. Therefore these adjustments on such a scale will not often have to be made.

For weighing on a beam scale the commodity should be placed on the pan; the beam should be brought to a balance by moving a sliding poise or by adding and subtracting loose weights. The weights used should be accurately totaled or the exact graduation on the weighing beam read.

In weighing a commodity on a scale which automatically indicates the weight, it is necessary only to make the reading carefully. Fractional parts of ounces are sometimes estimated by noting the exact position of the indicator between divisions.

If when reading the scale different readings are obtained upon moving the eye to the right or left, or up or down, it is necessary to take care that the eye is squarely in front of the point of the scale which is being read.

If a commodity exceeds in weight the capacity of the scale, it may sometimes be divided into two or more parts for checking purposes, each not to exceed the capacity of the scale. By adding the separate weights of the various portions the total weight of the amount purchased is found.

The weight desired and the weight which should be furnished is always the net weight of the commodity, i. e., the commodity itself without wrappings or coverings of any kind. Therefore a commodity should not be weighed in a cardboard carton, or other heavy coverings, or, if weighed with these, the coverings should thereafter be weighed separately and this weight subtracted from the gross or total weight.

(b) MEAT. When shortages are found in the weight of meat purchased, a common excuse of the dealer is that the meat was trimmed after being weighed, and it has often been found that this excuse was used to cover up frauds in such sales. Therefore

the housewife should require that all meat trimmings should be delivered to her by the dealer. This is fair, since they have been purchased and paid for at regular prices; and it is economical since they are often of use to the careful housewife. When such a demand is made, the total weight delivered should never be less than the amount charged for, and if it is so found a shortage may legitimately be claimed.

(c) POULTRY AND FISH.-Poultry drawn and dressed after weighing can not easily be checked as to weight, since the housewife will seldom desire to have the feathers, head, claws, etc., delivered. By observation or experiment one can soon learn what shrinkage in weight is naturally to be expected, and an investigation should be made in cases in which this proper shrinkage is exceeded. In cases in which fish are cleaned after weighing, similar precautions should be observed.

(d) GOODS IN ORIGINAL PACKAGES.—The purchaser should read the labels on the packages of goods purchased and the accuracy of the statements of quantity should be checked. If the contents are not to be removed at once, the package may be weighed gross when purchased and the weight noted on the outside. When empty the container may be weighed, and this weight, subtracted from the gross weight noted previously, gives the net weight. The law requires that this should equal that printed on the label.

(e) LIQUIDS IN BULK.-If a liquid commodity is purchased in bulk and only partly fills the receptacle in which it is delivered, upon pouring it into the measure of the nominal size corresponding to the amount purchased, it should completely fill that measure. If it does not fill the measure the delivery is short. If more of the same liquid of the same grade is at hand, the shortage can be immediately determined by putting a definite, noted quantity of the liquid into the graduate and from this amount completing the filling of the measure. The difference between the quantity of liquid remaining in the tolerance graduate and the original amount put in it is the shortage.

If there is not at hand any more of the commodity, the liquid under test may be poured out of the measure and the measure filled up with water to the same point that the commodity

reached. This can usually be done by observing the top of the wet ring around the measure left by the commodity. Then by completing the filling of the measure with water from the graduate, the shortage

may be found in the manner described above. (LIQUIDS IN CONTAINERS.-Liquid commodities bought in bottles, cans, or other containers may be checked as above described, or the following method may be found easier. The point to which the container is filled is first noted and the contents removed. Then, if the container is of the nominal size of one of the liquid test measures, this measure is filled with water until the water is just level with the top. This is then poured into the container until it is filled to the same point as before. If any water remains in the measure, the delivery is short by this amount. By pouring this into the graduate and noting its amount the

FIG. 7.-Fraudulent 5-gallon measure (with side partially cut away, showing 3-gallon can inside) The purchaser sees only the 5-gallon measure but the 3-gallon measure is the one which is filled-a delivery 40 per cent short resulting.

shortage is determined. (Be sure that the container is filled to the same point in the test that it was when delivered, as the amount actually delivered and not the capacity of the container itself, is of importance in this case.)

When the container is not of the standard size of one of the test measures, the test is made as before, except that one of the

measures must be filled more than once, or various measures and the graduate must be used and the error determined on the last amount added. For example, if a bottle is marked "One gallon," the quart must be filled and poured into the container four times, and the shortage, if any, determined on the last measurefull added.

Again, a bottle may be marked "11⁄2 quarts." In this case the quart measure is filled and poured into the container, and then the pint is filled and poured in. In this case the amount remain

FIG. 8.-Three bottles of extract (front and side views)

This shows the impossibility of correctly estimating the quantity of contents from apparent size of the container. The bottle which is apparently smallest holds the most and vice versa.

ing in the pint measure is the total shortage, and this is determined as before..

A container may be marked "12 fluid ounces." As there are 16 fluid ounces in a pint, or 8 fluid ounces in a half-pint, the 2-pint test measure and the 4-ounce graduate may be employed, or the 4-ounce graduate may be filled three times.

(g) USE OF GRADUATES.-To avoid mistakes in reading cone graduates, it should be noted that these are sometimes more finely subdivided at the base than at the top. For example, a