ciently extended, each pair of curves that cross in this way must cross again.

The application of the facts shown by figure 19 to actual measurement, using only vapor pressure as a means of correction, can perhaps be best shown by an example. It will be assumed that a mixture supposed to contain 70 percent of normal butane and 30 percent of propane actually contains the two hydrocarbons in the inverse ratio. The effect of measuring the fuel at various temperatures and by various methods is shown in table 7. The method of correction applied to the measured liquid volumes is simply that of correcting the heating value of a mixture containing 70 percent of butane for thermal expansion from 60° F to the temperature indicated by the vapor pressure.

TABLE 7.—Effects of composition and temperature on measurement of fuel by various

methods 1

[It is assumed that fuel supposed to have the molecular composition, 70 percent normal butane and 30 percent propane, is actually 30 percent normal butane and 70 percent propane. It is further assumed that the heating value of the fuel is computed at 60° F. This table gives the ratio of the number of Btu actually available from the fuel to the number that would be computed by various methods of measurement at stated temperatures.]

1 The irregularities in the table result from carrying the computation to more significant figures than the original data (or a graphical interpolation among them) would justify, except to show their degree of uncertainty.

Obviously, if this very simple method of correction, which requires no new instruments, not even a thermometer, is applied to measured liquid volumes, the amount of fuel can be determined without regard to temperature or composition with nearly the same accuracy as by weighing. Variations of composition which might occur through error will be entirely negligible, and any temptation on the part of the seller to change the composition of the fuel for the sake of its effect on consumption is eliminated. If gage readings are taken, as would usually be the case, against barometric pressure and not against a vacuum, the liquid measurement, when corrected on the assumption that the gage reading plus 14.7 pounds gives the absolute vapor pressure, would be slightly affected by altitude. To show the magnitude of this effect and also to show the combined effects of the several sources of error on uncorrected measurement with a gas meter, it is assumed that the measurements are made at an altitude of 4,000 feet. Conditions are otherwise the same as for table 7. The results are given in table 8.

TABLE 8.-Combined effects of composition, temperature, and altitude on measurement of fuel by various methods

[Assumptions as in table 7, except that measurement occurs at an altitude of 4,000 feet]

Even the small uncertainty remaining because of differences in composition of fuel can be eliminated almost completely by making a simultaneous reading of temperature and vapor pressure, estimating the composition of the liquid, and selecting the curve, from a group of curves like those of figure 19, from which the heating value corresponding to the observed pressure may be read. The most practical way

to make the corrections would be to use a table of correction factors for temperature and vapor pressure, just as gas volumes are ordinarily corrected for temperature and pressure.

Table 9 is such a table. It was prepared by interpolating with the aid of the assumption that the density of a mixture of two pure hydrocarbons is intermediate between the densities of the two constituents when each has a vapor pressure equal to that of the mixture, and that the difference between the density of the mixture and that of one constituent is proportional to the percentage of the other constituent. The table was based on the properties of binary mixtures of the normal saturated hydrocarbons only. Other compounds will affect the figures a little if present in large amount. Time did not permit as careful study of the data as would have been desirable, and the table is subject to revisions with respect to even the hydrocarbons from which it was derived. Nevertheless, the table is probably applicable to the measurement of any mixture that would be considered a liquefied petroleum gas under any ordinary conditions of measurement, with an error of not more than a few tenths of 1 percent.

The ACTING CHAIRMAN. I will ask Mr. Pisciotta to take the chair. (At this point Alex Pisciotta, Vice President of the Conference, assumed the chair.)

REPORT OF SPECIAL RESOLUTION COMMITTEE, PRESENTED BY C. E. TUCKER, CHAIRMAN

The ACTING CHAIRMAN. I understand that Mr. Tucker, Chairman of the Special Resolution Committee to Memorialize Fay Stanley Holbrook, is ready to make his report.

Eminent and distinguished physicist and metrologist, leader in the field of weights and measures, and a gentle, kind friend.

Whereas our Gracious God, the Almighty Ruler of the Universe, has seen fit in His infinite wisdom to call Fay Stanley Holbrook from our midst on February 4, 1940; and

Whereas this National Conference on Weights and Measures of the United States of America, assembled in its thirtieth meeting in Washington, D. C., June 4 to 7, 1940, profoundly realizes the deep loss which has been sustained in the death of Fay Stanley Holbrook, who served as its beloved and respected Secretary from 1922 to the time of his death, and who by his exemplary charac ter and achievements as an engineer, lawyer, chief adjuster of weights and meas ures, inspector of weights and measures for the State of New York, Co-Chief of the Division of Weights and Measures of the National Bureau of Standards, organizer and founder of the Department of Weights and Measures for the Territory of Puerto Rico, Chairman of the Conference Committee on Specifica tions and Tolerances, endeared himself to all with whom he came in contact; and

Whereas it is the earnest and expressed desire of this Conference, at this, its earliest opportunity, and before entering upon its regular business duties, to properly acknowledge and record his notable and splendid achievements in the science of metrology and in furthering the cause of weights and measures, to convey its deepest sympathy to his surviving wife and children, to express its deep sense of grief at the loss of one who was always ready and eager to serve, who devoted himself to the advancement of the public good and freely gave of himself to those who desired his valuable assistance, and to indicate its love and esteem and respect for the sterling qualities of the man himself; and

Whereas it is the wish of this Conference to perform some act which will give permanent expression to the feelings and sentiments herein set forth: Therefore be it

Resolved, That this Conference hereby express its heartfelt sympathy to Mrs. Ann S. Holbrook and children in their bereavement; and be it further

Resolved, That this Conference, through a committee to be appointed by the Chair, procure a suitable memorial plaque to be presented to the National Bureau of Standards and to be hung in the office of the Division of Weights and Measures; and be it further

Resolved, That a copy of these resolutions be written and presented to Mrs. Holbrook, as a token to her of the respect of this Conference.