sticks for this purpose, but we do know they are so used in many instances, and we have never undertaken to stop that practice or to prosecute anyone for so using them. This is because we feel that the stick is a measure in a sense, and since we have used such care in calibrating them we feel that they are just as accurate a measurement as we could get.

Mr. MATHEWS. I would like to say we have had some experience in the gauging of tank trucks. We find that all filling at the stations is done with the truck standing on a level surface, so in calibrating the vehicle tanks of the various oil companies we do it on the level. Some of them want us to jack up the front end to drain the pipes during the test. Of course, we are not willing to do this, because they do not do that in filling the dealer's underground tank. In making a delivery the truck is ordinarily level, and consequently they would not empty all of the fuel oil or "gas" which a tank is supposed to deliver if in the calibration one end of the tank had been raised. The pipes may hold several gallons that will not drain out on a level. After calibration we mark on the tank the capacity load of each compartment, so that if a dealer buys a solid load he can be advised what each compartment holds.

Mr. AUSTIN. I might say in connection with that that we have had one or two requests to regauge tanks owing to the fact that the companies claimed they were unable to get an entire emptying of the tanks. We gauge with the truck standing on the level. They claimed that their customers were wise enough to insist upon their jacking up the front end of the truck or driving on an incline, whereby they would get a complete drainage of the tank, and as a result they were losing whatever volume of gasoline they were able to get out in that manner. They requested the retesting of the tanks using that process. My reply was that I would defer reinspection until they had rebuilt the tanks so as to permit proper draining on a level surface.

The ACTING CHAIRMAN. Are these tanks used in the dispensing of gasoline to parties who have underground tanks?

Mr. AUSTIN. Yes, sir; they are used by the oil companies in furnishing their own stations and likewise all the filling stations that are within the city.

The ACTING CHAIRMAN. In furnishing their own stations, why should they want to know the capacity of the tanks?

Mr. AUSTIN. They want to keep as close a check on their gas stations as possible, and it is necessary to do this if they want to remain in business very long.

Mr. SIREN. Mr. Chairman, most of the tanks are either cylindrical or oval. Would it not be possible in putting the stick down to put it in a sloping position, so that there would appear to be a greater depth of gasoline in the tank than is actually the case?

Mr. AUSTIN. We anticipate that the average inspector detailed to the work of inspecting these tanks will ascertain where he can get a permanent and a satisfactory position for the stick and mark this position on the openings.

Mr. SIREN. If you let the stick remain too long in the tank, the gasoline will "creep " up the stick.

The ACTING CHAIRMAN. I would like to ask you about what the accuracy is. Suppose you have a stick that is accurately graduated, and several individuals make the same measurements; how nearly would they be alike?

Mr. AUSTIN. At the full capacity you would not be more than one-half gallon in error, but at an intermediate point, such as at the center of the load where the cross section of the tank is largest, even one-sixteenth of an inch in measurement might result in 1 gallon difference from the actual measurements, and you might get such an error. For instance, I have a gauge stick calibration record here which shows that on this particular compartment, which had a total capacity of 202 gallons, 5 gallons was represented by a height of one-half inch at the middle of the tank, while near the top 5 gallons made a difference of 17 inches in the liquid level.

Mr. MATHEWS. We find it is very difficult to estimate accurately in the case of large tanks at the point of maximum cross section. In some tanks, at the middle, one-quarter inch represents as much. as 5 gallons. The error on compartments of 200 to 300 gallons capacity would not be more than 1 gallon.

The ACTING CHAIRMAN. That is more nearly accurate than I would have thought.

Mr. KANZER. Have you had any experience with drivers who, having withdrawn a certain number of gallons from the tank, delivered the balance for the full capacity?

Mr. AUSTIN. Yes; we have. It not infrequently happens that our inspectors upon intercepting wagons on the street find a driver who has his memorandum of sale for an entire load, but who still has a quantity of the load remaining in his tank. We are able to deal with him. In one such instance a driver had a receipted bill for the entire contents of his truck and had a full compartment left. It transpired, when we took the matter up with the purchaser, that the firm had a discrepancy in measurement of something like 10,000 gallons in the previous six months. Our method enables us to check up such cases and assist the business people in securing an honest

measure.

Mr. WINCHESTER. This discussion has brought out this fact, that the errors mentioned are based on 1,000-gallon tanks. The average compartment of a 1,000-gallon tank would hold about 300 gallons. I think it would be well for us to take into consideration the number of 1,000-gallon tanks in the country as compared with the number of smaller tanks ranging from 270 to 650 gallons capacity. The percentage is very small; therefore, the percentage of error in the smaller tanks must be materially smaller than the percentage of error stated in this case.

Mr. MATHEWS. It has been said that the driver may sell off possibly 10 or 15 gallons and then sell the balance to another dealer as a full compartment. In this relation the purchaser must do his part. These truck drivers are required to carry delivery sheets showing every delivery. If the driver says he has a full compartment, the purchaser can look in the compartment to see that it is full, and it is his duty and his privilege to do this. He should also ask to see the sheet for the compartment, and this will show whether some has been drawn off.

REPORT OF COMMITTEE ON SPECIFICATIONS AND TOLERANCES ON SPECIFICATIONS AND TOLERANCES FOR VEHICLE TANKS, PRESENTED BY F. S. HOLBROOK, CHAIRMAN

Pursuant to the instructions of the conference, your committee on specifications and tolerances last year brought in a report on specifications and tolerances for vehicle tanks. On account of the press of other business this report was not discussed in full.

In view of this your committee suggested that in order to get the report formally before this body the specifications and tolerances as proposed be tentatively adopted by the conference with the understanding that they were then to be negotiated with the various officials, with users of vehicle tanks, and with manufacturers of vehicle tanks, and the whole code put into final form, if possible, for presentation for final adoption by the present conference.

Shortly after the last conference adjourned these specifications and tolerances were very widely circulated among the officials, users, and manufacturers throughout the United States, and requests were made for criticisms and suggested amendments. The committee received a large number of replies in which the tentative specifications and tolerances were criticized, and a study of the criticisms seemed to indicate that many of the objections were reasonable ones deserving careful consideration. In view of this your committee requested, by letter, that the manufacturers and users of vehicle tanks and officials come to the conference prepared to discuss these specifications and tolerances on the floor, and fully to present any criticisms or suggestions which they had to make to the tentative report. It was suggested that the matter would doubtless be brought up by the reading of last year's report, section by section, for comment.

Inasmuch as we anticipate, in view of our experience with this matter, that a large number of criticisms will be made, and inasmuch as we recognize the impossibility in the time available of making satisfactory amendments and completing the code this year, the committee on specifications and tolerances will be prepared to propose a motion to the effect that the specifications and tolerances, together with all criticisms of record, be referred back to the committee. It is our expectation that during the coming year the committee will be able to prepare a code which will be more favorably received by the various parties in interest than the present tentative code has been, for consideration at the next meeting. Respectfully submitted.

(Signed)

F. S. HOLBROOK, Chairman.
WM. F. CLUETT,

R. F. BARRON,

C. M. FULLER,

Wм. B. MCGRADY,

Committee on Specifications and Tolerances.

CONSIDERATION OF TENTATIVE SPECIFICATIONS AND TOLERANCES FOR VEHICLE TANKS

Mr. HOLBROOK. We have here a number of copies of the tentative code. Therefore, we can place a copy in the hands of each of you, so that during the reading you can follow the text very closely. We would suggest that after the reading of each section discussion be had

on that section. If that procedure is satisfactory to the conference, it will be in order to distribute these specifications.

(Copies of the tentative specifications and tolerances were distributed.)

The ACTING CHAIRMAN. The secretary will read these specifications section by section.

Mr. HOLBROOK. I hope, Mr. Chairman, that it is well understood that the meeting is open not only to officials, but to users and manufacturers, guests of the conference, and anyone else who is interested; that they have the full privilege of the floor in this discussion. [Reading:]

TENTATIVE SPECIFICATIONS AND TOLERANCES FOR VEHICLE TANKS (WHEN USED AS MEASURES)

[Adopted by the Sixteenth Annual Conference on Weights and Measures, May, 1923] NOTES.-The following specifications and tolerances shall apply to vehicle tanks and their accessory piping, valves, etc., in those cases in which the tanks or the compartments thereof are used or to be used as measures to determine the amount of liquid delivered, and such use shall be permitted only when these specifications and tolerances are complied with.

If a tank is damaged in any way (as from collision, etc.), or if repairs which might in any way affect the accuracy of measurement are made, such tank shall not again be used as a measure until inspected and, if deemed necessary, tested by the weights and measures official.

In determining or checking the capacity of tanks, water is recommended as a testing medium. Mineral oil should never be used, because the change in volume due to temperature variations is large, and evaporation of the oil during the test may result in very serious inaccuracies in the calibration.

It seems inadvisable for the committee to defend these specifications, at least to the extent of starting a long argument, because we will not progress so rapidly in that way, and the object of this discussion is to bring out objections to them. Of course, the committee is prepared to state what they have in mind in reference to the specifications whenever such a request is made.

The ACTING CHAIRMAN. You have heard the reading of the "Notes." Is there any objection or any suggestion? If not, we will proceed.

Mr. HOLBROOK (reading):

DEFINITION.-A vehicle tank, hereinafter referred to as a

tank," shall,

for the purpose of these specifications and tolerances, mean a container, which may or may not be subdivided into two or more compartments, mounted upon a wagon or automobile truck and used for the delivery of liquids. The term compartment" shall mean the entire tank whenever the tank is not subdivided; otherwise it shall mean any one of those subdivided portions of the tank which is designed to hold liquid.

66

SPECIFICATIONS.-1. All tanks and all indicators, piping, valves, etc., attached thereto and used in connection therewith, shall be of such design, construction, and material that they may reasonably be expected to withstand ordinary usage without impairment of the accuracy of measurement.

2. An indicator shall be provided which shall be centrally located with respect to the longitudinal axis of the compartment. The lowest part of this indicator shall clearly and distinctly define the height to which the compartment must be filled in order to contain its marked capacity. If this indicator is adjustable, it shall be so constructed that it can be sealed in such a manner that its position can not be changed without destroying the seal.

The ACTING CHAIRMAN. You have heard those items. Are there any suggestions or objections or any questions you care to ask the committee?

Mr. ESTES. Mr. Chairman, are these specifications retroactive in their effect?

Mr. HOLBROOK. Not as a rule. In the final report which the committee will make it will be recommended that certain specifications certainly those designed to prevent the perpetration of fraud-be made retroactive. As a rule specifications will not apply to those tanks which are already in use.

Mr. MARONEY. Would the filler opening have to be in the center of each compartment?

Mr. HOLBROOK. It is suggested that the indicator shall be centrally located with respect to the longitudinal axis of the compartment, and this would require the filler opening at some part of its diameter to be in the center. This is for the reason that accuracy can be more nearly obtained by this arrangement when the tank is not in level. Any type of indicator may be used which is strong enough to withstand ordinary damage and which is definite enough to establish a definite filling point.

Mr. MATHEWS. I feel that the first thread in the dome is probably as accurate an indicator as can be obtained in any tank wagon, and I have seen many tanks calibrated. I think a better opportunity is afforded for checking the amount of liquid in the tank by employing this method than by employing an indicator, since I feel that under certain conditions there would be the possibility of the indicator being damaged on account of the position in which it must necessarily be attached.

Mr. AUSTIN. Mr. President, I might say that many of our regaugings have been made necessary by the fact that in our first inspections the tanks were filled to the bottom of the dome. In a large number of cases we have found that where the domes were small-6 or 8 inch collars 2 or 3 inches high, for instance there was not sufficient space for expansion, which caused the tanks when properly filled to overflow their contents along the street. To have such a condition exist is very objectionable from a safety standpoint and very objectionable from the merchants' standpoint. Some companies having that class of tank have had the domes reconstructed and have extended the collar of the dome down far enough into the shell of the tank to allow sufficient room for expansion. Then we regauged those tanks, measuring to the bottom of this dome-collar extension. Technically that does not comply with our ordinance because the tank is supposed to be shell full, but in those instances where we found the size of the dome to be insufficient to take care of expansion we have permitted the reconstructed collar to provide room for expansion. Unless the tank has a large enough dome it would seem that some definite indication mark should be provided.

Mr. WINCHESTER. Mr. Chairman, in a city like Detroit the oil manufacturers have changed the construction of their tanks so as to allow for the expansion, and in the city of Detroit we have very close supervision over the deliveries that are made. When a driver pulls up to a filler tank, he gets the correct amount in his tank because we have a man to check the filling. It is to be regretted that in all the smaller stations the filling can not be checked in the same way, but there are places where the seller and the buyer are entirely