sideration of these devices, and bring back to the conference a report thereon for its consideration. The first year the committee assembled the devices at the Bureau of Standards and studied them, and at the Fifteenth Annual Conference tolerances on these devices were proposed to you with the suggestion that they be tentatively adopted, and this action was taken. At the same time Ralph W. Smith, of the bureau, gave a very excellent paper describing a method of testing these devices, which was adopted by the conference as the standard method of test. Your committee continued its work and last year, at the Sixteenth Annual Conference, specifications having been developed by the committee, these were proposed to you and again it was suggested that a tentative adoption be had in order that you and the manufacturers might study the recommendations in detail at your leisure and determine their practicability and workability. It is these tentative specifications and tolerances which the committee brings before you to-day for action. In the meantime your committee has been examining and testing devices of this character on its own account and has also been receiving and studying suggestions made in connection with the tentative code adopted. As a result of this consideration your committee has to report that only one amendment to the tentative code is suggested. This amendment is to a tolerance figure. In general, in the tolerance table, the tolerances on excess deliveries have been made somewhat greater than the tolerances on deficient deliveries except in one case, the tolerance on the yard, in which the tolerance in excess and the tolerance in deficiency were made the same figure; that is, one-quarter of an inch. The committee now desires to suggest that the tolerance in excess on the 1-yard graduation be increased. In the case of used machines in the field we recommend that the tolerance in excess be increased from one-fourth inch to three-eighths inch on registrations of 1 yard or less. That will increase the tolerance on new machines from one-eighth inch to threesixteenths inch, and will result, if it has any effect at all, in a slightly greater average delivery to the customer. So far as we have learned, users of the machines do not object to this increase in tolerance, and the change has the decided advantage of making it somewhat easier for manufacturers, who have had some difficulty in some cases in controlling backlash within the old limits, to produce the machines. On these accounts, and also since the new value can not be considered an unjustifiably large one, it seems that the amendment suggested is an entirely reasonable one. As we have stated, these tolerances have been before you for two years and the specifications for one year. Last year they were distributed to the conference in mimeographed form. Following the conference a large number of copies were distributed by mail, and they were also included in the printed proceedings of last year's conference. We therefore have no doubt that you are entirely familiar with these specifications and tolerances, copies of which are now again placed in your hands for reference. With the amendment suggested above these will now read as follows: SPECIFICATIONS AND TOLERANCES FOR RETAIL FABRIC-MEASURING DEVICES DEFINITION.-A fabric-measuring device is a mechanism or machine adapted to measure and automatically indicate the length of fabric passed through it. A computing fabric-measuring device is a fabric-measuring device which automatically indicates the total price of the amount of material measured, for a series of unit prices. Unless otherwise specifically indicated by the context, the term "fabric-measuring device" as used herein shall be understood to include computing fabric-measuring devices. NOTE. The following specifications shall be applied only to fabric-measuring devices designed and constructed to measure fabrics at retail. SPECIFICATIONS.-1. Fabric-measuring devices shall be graduated in units of the customary system and its usual subdivisions. The maximum value of the minimum length graduations on fabricmeasuring devices shall be one-eighth yard. NOTE. This shall not be understood to exclude fabric-measuring devices graduated in the metric system. (See Specification No. 12.) 2. The length graduations and the value graduations on all fabricmeasuring devices shall be clear and distinct and their length shall be so varied or they shall be so arranged that their meaning or value is readily apparent and their indications may be conveniently read. The width of any graduation mark shall in no case be less than 0.008 inch. 3. The clear interval between one-eighth yard graduation marks on fabric-measuring devices shall not be less than eleven-sixteenths inch (0.6875 inch); if inch graduation marks are employed the clear interval between such inch graduation marks shall not be less than one-eighth inch (0.125 inch). These values shall be applied to the most sensitive indicating element with which the device is equipped. The clear interval between value graduation marks on fabricmeasuring devices shall not be less than 0.02 inch. 4. Value charts may be made in accordance with either of the following principles: (a) If the device is so designed and constructed that it purports automatically to compute for a series of unit prices the total price for every length within the range of the device, then the device shall be equipped with a value pointer or indicator and value graduation marks; the value graduation marks shall be correctly placed; and in any position which the indicator or pointer and the chart may assume there will be exposed to view a sufficient number of value figures and graduations to permit the value indications of the device to be read correctly. The value graduations shall not exceed 1 cent at all prices per yard up to and including 30 cents. At any higher price per yard the value graduations shall not exceed 2 cents: Provided, however, That nothing in the above shall be construed to prevent the placing of a special value graduation to represent each 5-cent interval. These special graduations may take the form of dots, staggered graduations, or similar forms. They shall be so placed that their meaning and value may be clearly understood, but they shall not be placed in the space between the regular graduations. (b) If the device is so designed and constructed that it purports automatically to compute only for lengths corresponding to a definite series of length graduations, then there shall be no value graduation marks and at no position which the chart may assume shall two value figures for the same unit price be completely and clearly exposed to view at the same time. One of the following alternatives shall also be complied with: (1) There shall be a value computation for each length graduation throughout the range of the device; or (2) no value indication may be exposed to view except at such times that the device registers a length indication for which a correct value indication is provided; or (3) each column or row of value graduations shall be clearly and conspicuously marked with the length graduation to which the values correspond, the device shall be marked with the character and limitations of the computations made, and there shall be a computation for at least each one-eighth yard, throughout the range of the device. All money values corresponding to definite length graduations must be mathematically correct except as follows: If the mathematically correct amount includes a fractional part of a cent, the fraction shall be dropped if it is less than one-half, but if the fraction is one-half or more the next higher cent may be shown. 5. Each pointer or indicator used in a fabric-measuring device shall be so designed and constructed that a clear, distinct, and accurate reading is given. All pointers or indicators shall be symmetrical about the graduation marks at which they may stand and shall reach to all such graduation marks. The width of the pointer or indicator, or of the end thereof, shall not exceed the width of the smallest graduation marks on the scale with which it is used and in no case shall such width exceed 0.015 inch. The distance between the pointer or indicator and its scale shall not exceed 0.06 inch. 6. Fabric-measuring devices shall be so designed and constructed that in any position which the indicator or pointer and the chart may assume in its operation there will be exposed to view a sufficient number of figures and graduations readily to permit the length indications of the device to be read correctly. 7. Fabric-measuring devices shall be so designed and constructed that the indicating elements used in registering lengths or prices of deliveries to individual purchasers are returnable readily to a definite and clear zero reading before the next measuring operation is begun. 8. All fabric-measuring devices shall be correct in their length and value indications whether the indications are being increased or decreased. 9. If a fabric-measuring device will not give correct results when used for the measurement of all fabrics, then the device shall be so marked as clearly to indicate its limitations. 10. All markings, instructions, figures, and graduations required under these specifications shall be of such size, design, material, and location and shall be so applied or affixed that they will not tend easily to become obliterated or illegible. 11. All fabric-measuring devices and all devices designed to be attached thereto and used in connection therewith, shall be of such construction that they are not designed to and may not be used to facilitate the perpetration of fraud. 12. Nothing contained in the above specifications shall be understood or construed to prohibit the sale or use of fabric-measuring devices constructed or graduated in units of the metric system. The tolerances to be allowed on fabric-measuring devices constructed or graduated in units of the metric system, shall be the same as those specified on similar devices of an equivalent size in the customary system. TOLERANCES.-The tolerances to be allowed on the delivery of retail fabric-measuring devices in excess (underregistration of device) and in deficiency (overregistration of device) to be applied on both increasing and decreasing registrations of the machine shall be the values shown in the following table: Provided, however, That the manufacturers' tolerances or the tolerances on all new retail fabric-measuring devices shall be one-half of the values given: 1 "Tolerances on delivery" refers to the variations between the indications of the fabric-measuring device and the corresponding actual lengths of testing strip passed through the device, and does not apply to check measurements made upon lengths of fabrics which have been commercially measured. 'Or less. For machine indications of more than 15 yards add one-sixteenth inch in deficiency and one-eighth inch in excess per indicated yard. Respectfully, (Signed) F. S. HOLBROOK, Chairman, WM. B. MCGRADY, Committee on Specifications and Tolerances. DISCUSSION OF ABOVE REPORT Mr. HOLBROOK. The recommendation of the committee is that these specifications and tolerances be now brought up for final adoption that they may be put into force and effect in the various States. I do not know how you would prefer to handle this. The report might be discussed in detail if you desire to handle it that way. Mr. SCHWARTZ. Mr. Chairman, I move that the specifications and tolerances, with the amendment suggested, be finally adopted. The CHAIRMAN. Does any member of the conference wish to raise any question on this report? Mr. SCHWARTZ. Mr. Chairman, as Mr. Holbrook has explained, this has been under consideration for two years. The only change made this year is in one tolerance figure. I would suggest that it is the desire of the conference to adopt the report as a whole. I do not see anything here that we should change. The CHAIRMAN. Are you ready for the question? Mr. EMERY. I understand that Mr. Holbrook made a recommendation that a certain change be made. I move that this amendment be adopted first. Mr. SCHWARTZ. I second the motion. (The question was taken, and the motion was agreed to.) The CHAIRMAN. Now we will consider the report as a whole. Mr. SCHWARTZ. Mr. Chairman, I now renew my motion that the report as a whole be finally adopted. (The motion was seconded, the question was taken, and the motion was agreed to.) UNIFORMITY OF REGULATIONS FOR MILK BOTTLES By RALPH W. SMITH, Bureau of Standards At the first meeting of the Annual Conference on Weights and Measures in 1905, Doctor Stratton, in his opening remarks, stated that the reason for calling the conference was the existing lack of uniformity in the weights and measures laws and regulations of the different States and the obvious need for cooperative effort on the part of the officials concerned with the enforcement of these laws and regulations if this condition were to be remedied. Since that time the constant effort of the conference has been to promote uniformity throughout the United States in the statutes, the specifications, and the tolerances affecting weighing and measuring devices, and to coordinate the activities of the officials charged with the duty of giving effect to these regulatory measures. The advantages of uniformity and proper coordination of effort are too well understood by all of you to need elaboration at this time. The notable achievements of the annual conferences in this direction constitute the best evidence of your lively appreciation of the value and desirability of realizing these advantages in the country-wide administration of weights and measures supervision. Without uniformity of written rules, without cooperation of administrative officers, the work in each State is made more difficult, the general efficiency of the separate departments is lowered, industry is needlessly hampered, and the ultimate service rendered to the people of your several communities is lessened. And these unfortunate results, with no compensating advantages, come about notwithstanding the individual excellence of the regulations which may be in force in any given jurisdiction, and notwithstanding the individual zeal of the several weights and measures officers. But, however much has been accomplished already, the need for continued effort along these lines still exists. A conspicuous example of the divergence which still prevails in the requirements governing |