dry measures. Among the various scales belonging to venders, coal dealers, wholesale and retail dealers that have been tested, 57 defective ones were confiscated. Mr. Crotzer states that in his investigations, the dealers are at all times willing to co-operate with him in regulating the weights and measures.

It is Mr. Crozier's suggestion that all consumers furnish themselves with at least a 10-pound scale and weigh all purchases from the dealers. In this way a defective scale may be found with which the inspector is not acquainted. When a mistake is found he has requested a report be sent to him at the City Hall and an investigation will follow at once.

M. T. Toole has been appointed city sealer of Pittston to succeed Michael Jeffries, who resigned several months ago.

Sixteen hundred sacks of flour, in one carload shipment were found to be off weight, by Sealer of Weights and Measures R. A. Pringle. The sacks varied from one to one and a half pounds underweight.

The shipment was the first made to Franklin, by the firm selling the flour and the buyer asked that the sealer be on hand to test the sacks. Mr. Pringle has been busy recently in making tests of incoming shipments, but this is the most serious violation that has come to his attention.

A huckster who gave short measure at West Philadelphia's new curb market at Forty-ninth and Market Streets was sent to jail by Magistrate Harris. The huckster who aroused the magistrate's ire was arrested by Policeman Harry Ferguson, who said that women complained to him that they had been given short meas

ure.

"The huckster or dealer who gives short measure is worse than a highway man who holds up a man and robs him at the point of a revolver," said Magistrate Harris as he sentenced the huckster.

The sixth annual convention of the Pennsylvania Association of Inspectors of Weights and Measures was held in the Senate Chamber of the Capitol, Harrisburg, Pa., Sept. 25-26-27. James Sweeney, chief of the Pennsylvania Bureau of Standards, made the opening address.

"There has not been a time in the history of the weights and measures proposition of Pennsylvania," Chief Sweeney told the hundred or more weight inspectors present at the opening session, "when a conference of the inspectors was more necessary than at the present time, that you might be able to devise methods to prevent the profiteering of certain dealers who take advantage of the conditions

existing as the result of the war." He complimented the inspectors upon the assistance given to the Food Administration and asked that resolutions be adopted to pledge every effort of the inspectors to help win the war.

Governor Brumbaugh in a brief address said that the people relied upon the inspections and that "an impartial investigation of weighing machinery of the State" was something which is necessary to see that people get what they pay for.

The convention began a discussion of the situation in regard to package goods, and a resolution was adopted that all dry commodities be sold by weight. This action was taken at the instance of Max Mayer, of Philadelphia. The convention devoted considerable time to proposed amendments to the law relative to prosecutions and discussed coal, meat, and package sales.

As a result of the Convention, bills will be drafted for presentation to the next legislature requiring sale of all dry commodities by weight instead of measure or count; establishing a standard weight per bushel for various commodities which shall be in uniform with the standards in other states, especially those bordering on Pennsylvania and authorizing counties to pay expenses of inspectors attending conventions to exchange experiences and views. Resolutions will be sent to mercantile and commercial organizations asking co-operation in enforcement of the weights and measures law.

Charles F. Borgel, of York, led a discussion as to the proper way of handling large scales. Following this a discussion as to weighing coal at the mines took place. A resolution was passed requesting the Fuel Administration to have each mine apply the 1,000 pound weight test.

Inspectors Olmstead and Mayer of Philadelphia and Boyer of Harrisburg comprised a committee which prepared a fitting tribute to members of the Association who are in military and naval service.

W. B. Dunn of Altoona was elected president.

Texas.

After January 1 there will be no more cartons of butter less than a pound in weight sold wholesale in Dallas, Texas. The movement is for the conservation both of labor and paper. The rule, however, does not prevent the retailer from cutting the pound and selling any portion of it.

T. J. Tompkins of Dallas presented a petition to the city council asking per

mission to place a scales on East Second street.

Vermont.

A notice has been issued from the State weights and measures department to wood dealers and those who sell wood occasionally, to refrain from illegally selling that fuel "by the load." Wood must be sold by the cord or fraction of a cord-otherwise the seller may get into

trouble.

The notice is signed by Walter F. Scott, Commissioner.

Virginia.

Both wholesale and retail dealers who handle potatoes in Newport have been warned that weight shortage on this commodity has occured a number of times recently, the deficit being found in carload shipments from various parts of the country. In some cases sacks have been found to contain a 25 per cent shortage. It is now compulsory that potatoes, both sweet and white, shall be sold by weight by wholesalers and retailers alike. Therefore they should make certain that potato containers be up to the mark as far as weight is concerned.

West Virginia.

City Sealer Edward A. Brookes of Wheeling has returned from his vacation down-state hunting. He resumed his duties immediately following his return.

Do not let your "Cut Rate" market be a "Cut Weight" market and especially if you live in Huntington, for City Sealer Fizer will "get you, if you don't watch out." One dealer was fined $10 for this offense.

Edward A. Brookes, city and county sealer of weights and measures, received a letter from J. A. Healy, sealer at Worcester, Mass., and secretary of a committee appointed by the national bureau of standards, requesting that sample pages of the local sealers record books for the testing and sealing of various devices of inspection sheets, sample of paper seals, non-seal labels, condemning tags, etc., be mailed to him at once.

This action has been deemed necessary by the national bureau of standards in order that a line may be gotten on the devices used by every dealer in the United States with a view to making one standard of device which will greatly simplify the work. Mr. Brookes complied with the request at once.

Washington.

Numerous complaints that dealers are selling firewood in amounts less than a cord have started City Inspector of Tacoma on a thorough investigation. Scott promises to prosecute all dealers selling short cords.

Come Again and Stay as Long as You Like.

SCALE JOURNAL, Chicago, Ill.

Enclosed please find check for $2.00 for one year's subscription to the SCALE JOURNAL, statement for which came yesterday.

In regard to news items you might give the writer credit in your JOURNAL for having first discovered the Fleckenstein Visible Gasometer, as I gave friend Jack Fleckenstein a little writeup over a year ago, am sure all sealers who inspect pumps as a part of their work will agree with me in what I say about accuracy and speed being the leading feattures of the Fleckenstein measuring device.

In these days of soaring prices in all commodities, especially foodstuffs, it is up to the sealer to see that all weighing and measuring devices are kept up to the authorized standards and there maintained and then to check up on weighings Imade over these scales. A great deal of grain and live stock or I might say pork and beef are demanded now by Uncle Sam for our soldier boys and our Allies and it is with regret that I say a great many dealers who call themselves Americans would use most any old scale for the purchase or sale of these commodi

ties. I wonder how many sealers like to test large hopper scales, wagon and stock scales, etc. Not a great many, perhaps, but these scales should certainly be inspected and tested regularly and I am glad to state that in our section at least any sealer can now get the hearty cooperation of the U. S. Railroad Administration in looking after and maintaining any scales owned by the railroads. Previous to the Government taking over these roads it was sometimes a hard matter to get service from some roads, but am sure that all sealers can now get that service which has often been desired, by applying at the proper place. The writer has been getting splendid results and has had a number of the heavier type of scales replaced in the past year and it is gratifying to say the least in not having so many complaints along that line. Then, too, there are several corporations in my territory who have installed up-todate weighing devices known as Automatic Scales, to replace wornout equipment. Several of these are the large Dormant or Floor Scales and large suspension Tank Scales. The concerns having installed these scales are The Dry Milk Co. of Charlotte, Horner Bros.' Woolen Mills of Eaton Rapids, A. M. Smith Co., creamery, eggs and poultry, of Eaton Rapids, Grand Ledge Condensery of

Grand Ledge, and B. F. Santee, junk dealer, of Charlotte. It's pretty nearly a regular fellow who will install a Dormant Automatic Scale to buy junk on. Will also state that all hardware merchants in my territory excepting five are now using Automatic Computing Scales. The writer is quite proud of weights and measures conditions in Eaton County and feels that he has achieved a great deal in bringing about these results.

I might give you my yearly report as sent to Commissioner Fred L. Woodworth some time ago, but am sure it would not be interesting news to the sealers, so will omit that. This being my first contribution to your valued and interesting "colyumn," I think I have already overstayed, so with best wishes for the success of the SCALE JOURNAL, I am Very truly yours,

FRED A. DOLECKE, County Sealer.

Charlotte, Michigan.

SCALE JOURNAL, Chicago. Ill.

Mr. Editor: I am writing to renew my subscription to the JOURNAL; find check for Two Dollars enclosed.

In the August number I noticed that some one wants to know what methods other scale men are using in setting their stands. I waited for the September number to see what the response would be, and as everybody passed the buck I will just say that we have been setting our stands in pure dry cement, with satisfactory results. One scale installed in 1912 seems as solid and firm as when stands were set. I always set the forms level, fill to the top with the mixture being used, work the rock or gravel down out of the way and finish without using what is usually called a finishing coat or dressing. We then set the stands on about 1/4 or 3 inches of pure dry cement. This will form a resembling fiber when thoroughly set. I might add that we bed our transvere I-beams the same way for rigid deck.

I would like to inquire what, if any, steps have been taken by our executive committee to see that scale men get a square deal in the wage adjustment that is going on. It seems to me as though some one was asleep at the switch: when machinists get 70 cents an hour, rail straighteners make $52.50 per day, and wheelbarrow laborers at the steel works get $5.60 per day, there should be something coming our way.

Can you tell me where to address our secretary for the payment of dues?

A great many complicated physical ; henomena can be expressed in simple mathematical formulas, but what is more often the case many essentially simple ¡hysical phenomena can only be expressed with difficulty, or in such form as to limit seriously or nullify the practical applications to problems which it is Important to solve. In many cases probems of a certain type can be solved and solved easily when they take restricted forms but when our needs make it necessary to consider other cases we are at one confronted with difficulties that too often render us impotent.

One old problem of this character relates to the stresses and twist produced in a prism of irregular cross-section by couples applied to the ends. When we have to deal with a circular cross-section or a hollow circular cross-section it is a simple matter to determine both the twist that will take place with any particular couple, and the magnitude and disposition of the stresses. When we depart from such a simple shape, however, we at once become involved in difficulties of very formidable character.

The solution of what we call the torsion problem has been a matter of practical importance in engineering problems for many years, and various expedients have been adopted to give us imperfectly what we were unable to obtain from our mathematics. Recently the desin of aeroplanes has become of considerable importance, and in which are employed struts, propeller blades, supports, stays, etc., which are subjected to torsion, and in which it is necessary to employ sections of irregular form. The fat that it is of utmost importance to secure the greatest lightness consistent with strength therefore gives the solution of the torsion problem a most peculiar interest. It is, therefore, a rather happy circumstance that just at this time a practical means for solving the torsion problem is presented, which will enable us to obtain all the information necessary and desired in the design of torsion embers having cross-sections of whatever character.

The method employed for getting this information has been developed by Mesars A. A. Griffith and G. I. Taylor, and described by them before the Instition of Mechanical Engineers, and published in Engineering for December 21 and 28, 1917.

The principle upon which the method is based is one that was established several years ago, to the effect that the differential equation for the function from which can be derived the solution of the torsion problem, is the same as the dif ferential equation of the curved surface asured by a uniformly loaded elastic membrane stretched across an opening of the same shape as the cross-section of the

torsion member. This relationship holds when the displacement of the membrane is small.

The authors obtained their solution for the torsion problem from measurements made on soap films in a box suitably arranged and convenient for the purpose. A horizontal plate is provided, in which are cut openings corresponding to the cross-sections it is desired to investigate. A soap film is then stretched across the openings and caused to bulge up slightly by producing a difference in air pressure between the two sides of the film. A glass cover for the box containing the film serves to enclose it and prevent evaporation of the film, thereby prolonging its life.

If the surface tension of the soap film were known and the difference in the air pressure between the sides of the film were measured, the facts necessary for the solution of the problem could be ob

tained from appropriate measurements made on the figure of the film. However, the determination of both the surface tension and difference of pressure are not easily accomplished in practice, and the need is avoided by comparative measurements. In addition to the form of the cross-section to be studied, there is provided in the plate a circular opening which represents the case for which the mathematical solution of the torsion problem is known. The pressure and surface tension is the same for both openings, and the properties of the irregular cross-section are determined by comparing measurements made on its soap film with those obtained from the circular opening.

The slope of the film at any point is proportional to the shearing stresses set An auto-colimating device of simple

up.

OO

construction, using light reflected from the film, is used for this measurement.

The relative stiffness of the member is proportional to the volume contained between the film and horizontal plate. This can be obtained in two ways: first, by budging the films, by displacing the air in the lower compartment of the chamber by introducing a given volume of soap solution; secondly, from the contour lines of the film from which the volume can be obtained by computation.

This latter method, while much more complicated, has some advantages.

The contour lines of the soap film are located by means of a point wet with soap solution, carried downwards from a micrometer spindle passing through the sheet of glass covering the top of the box. The position at which the point makes contact with the film can be determined by inspection, and this can be done with very great accuracy.

In addition to giving the volume, the contour lines serve two other purposes: first, each contour line represents a form of cross-section for which the film above it gives the solution for stress, etc. This can be understood from the fact that each contour is the equivalent of an aperture of that form in the plate. At one time, therefore, the solution for several forms of cross-section can be obtained and cataloged; secondly, the contour lines indicate the direction of the shearing stresses. At the surface of the prism there are no shearing forces across the boundary, or surface traction, and this is true for each contour line. Therefore, a torsion member could be segmented along the contour lines without affecting the strength or stiffness of the mem. ber, if each part were devised to carry its proper share of the torsion couple.

It is a credit to the authors that they developed their apparatus to the extent that they could carry out in a practical manner the various measurements called for and increase the attractiveness of the method.

From a practical standpoint, the method has many important advantages over a mathematical solution of the problem, if such solution were possible. The method appears to be a very flexible one and for routine work can be adapted to give very prompt results. A feature of very great value is that considerable information can be obtained by merely inspecting the form of the soap film. It is easy to see the effect of inwardly directed angles or curves in increasing the stresses of the cross-section, and im provements in design at once suggest themselves. What is of interest also is that when one becomes accustomed to the forms assumed by films a good mental picture can be formed of what will take place without the aid of the apparatus.

This article has been very well received and has excited an unusual amount of interest and favorable comment. In addition to its practical value. it appeals particularly to the imagination and there are, no doubt, many who are taking an interest in the subject not so much for what is done as for the manner of its doing.

Comments.

Solution Applies to Isotropic Materials: What does not appear to be mentioned is that the direct application of the method as given relates only to isotropic substances, while in the construction of aeroplanes wood is used, which is far from being such a material. It may be found desirable, therefore, to modify the method. However, the authors of the papers tried out the method by experiments on a wooden member and found that the twist obtained by a given couple checked very well with that determined by the soap film.

Ambiguity in Certain Cases: There are certain cases of interest in which a wrong conclusion may be formed, and in which the method may be condemned for what would appear to be an obvious failure of

the method. For instance, all of you have twisted in your hands those split brass curtain rods, which can be obtained at any 5 and 10c store. You will also recall that these rods were very easily twisted. The aperture representing this curtain rod is given by figure 1. You will have no difficulty in imagining a soap film stretched across such an opening. Now let us take the curtain rod and secure its edges together so that they can no longer slide past each other when the rod is twisted, such as by soldering them together at several points. It will then be found that the stiffness of the rod is very greatly increased. Let us now cut out the gap from between the two ends of the soap film aperture so that it will represent the case of the stiffer tube or curtain rod as illustrated in figure 2.

On now examining the film stretched across the opening, we fail absolutely to

We have two films, one upwards and one downwards, representing the two sheets required for the solution of the problem. The torsion function is given in this case by the algebraic sum of the ordinates of the two films. We will at once see that the last arrangement of films gives correctly what we know to be the solution for this particular and simple case.

This ambiguity is mentioned for the fact that in going over the scheme mentally, so to speak, I was somewhat upset, after being attracted to the method, to find an apparent failure of it to hold good.

Modifications of the Apparatus.

Location of Lines of Constant Shear: The contour lines give the direction of the surfaces across which there exists no shearing stress. However, a shearing stress along other directions for each

the particular slope for which the light is adjusted. In this manner, by suitably changing the position of the light, all points having the same shearing stresses will be photographed in one figure as a series of lines which can be studied at convenience.

Photographing the Contours: If we take a plate of glass, wet on the lower side with the soap solution, and, holding it in a horizontal position, bring it downwards into contact with the film, the film will make contact with the glass along a contour line. However, the Volume conditions have been disturbed, so that the contour corresponds to the undisturbed film at a different height. If now air is withdrawn until the volume of the remaining part of the film is the same as it was before the plate was used, the contour formed by the film meeting the glass is the same as that present at that position before the glass was brought

get any indication of the great increase in stiffness which we found to be the case with our curtain rod. As stated above, the stiffness of the members subjected to torsion is proportional to the volume inclosed by the film. In joining the two ends of the aperture together, we get no very material increase in the volume of the film.

Furthermore, if we now decrease the diameter of the inner boundary until it is small compared with the outer boundary, we obtain a film somewhat as indicated by figure 3.

This figure indicates that the center hole has a very profound effect on the stresses of the cross-section and that the greatest stresses occur at the center, both of which indications we know to be contrary to fact. What then is the difficulty? This is easily explained.

As soon as we cut the gap from between the ends in the outline representing the split ring we have no longer the appropriate film for our purpose. Before we cut the gap we had one continuous boundary, and the torsion problem could be represented by a surface of one sheet. Just as soon as we have joined the two ends together, however, we have two boundaries and the torsion problem must be represented by a surface of two sheets. The case is therefore given by a film arranged as indicated in figure 4.

point in every contour and the value of the shearing stress will in general vary from point to point along the contour lines. What will be of interest, therefore, will be to locate the lines of equal shearing stress in the cross-section.

I have to suggest a simple arrangement, by means of which this can be accomplished. Let us secure a table or stand of suitable construction, mounted to rotate about a vertical axis, and mount upon it a camera directed downwards towards a support for the box containing the soap film. Also let the axis of the lens coincide with the axis of rotation of the table, and let the film be centered substantially with reference to the axis of rotation; and let a light be placed above the soap film at a fairly great distance from it and so mounted that the light can be adjusted to strike the film at any desired angle with the vertical.

From the position of the camera the light will be seen reflected in the film as a pin-point of light. The point from which the light is reflected is determined by the slope of the film at that point. If now the table carrying the camera and soap film be rotated about the vertical axis, the point of light will appear to move around the film, and the plate of the camera will record a horizontal projection of all points of the film having

into contact; and the desired contour can be photographed if the appropriate illumination is furnished. By placing the cover plate at a different height and making the necessary changes in volume, any series of contour lines can be mapped out and recorded, either by photographing or by drawing on the plate over the outline of the film.

For the required volume adjustments several different schemes suggest themselves. One method would be always to adjust the bubble formed over the circular opening to a fixed height by the use of suitable points or other indicators. However, in this case it would be necessary to arrange so that the glass plate would not make contact with the reference bubble or interfere with the height indicating devices. Another method would be to have the size of the circular opening such that its reference bubble also would make contact with the plate whenever the other film touched it. The volume of the film could then be adjusted by furnishing several templates corresponding to the contour of the circular bubble, for the various heights at which the plate is to be used and then adjusting the pressure until the area of the contour of the circular bubble was of the correct size as shown by the template for that particular position of the plate.