"*^«ST.„« ,,„^^^,^ LIBRARY Vol. 8 Part 1 MARCH, 1924 Number 43 Bulletin OF THE National Research Council proceedings OF THE THIRD annual MEETING OF THE ADVISORY BOARD ON HIGHWAY RESEARCH, DIVISION OF ENGINEERING, NATIONAL RESEARCH COUNCIL Held at Washington, D. C. November 8-9, 1923 EDITED BY William Kendrick Hatt^ Director, Advisory Board AND E. R. Olbrich^ Assistant to the Director Published by the National Research Council The National Academy of Sciences Washington^ D. C. 1924 Announcement Concerning Publications of the National Research Council The Proceedings of the National Academy of Sciences is partly officially the publii the same Subsc address : Institutio presents < ceedings, existed. The ' price, po! of the " manufact The: renders ; manufact Council. presented open for Duncil on ces. Business ithsonian :han pro- ion have scription ^Jumbers cost of irch cost of ^^esearch Orders for the "Bulletin" or the "Reprints AM Circulars^* of the National Research Council, accompanied by remittance, should be ad- dressed: Publication Office, National Research Council, 1701 Massachusetts Avenue, Washington, D. C. BULLETIN OF THE NATIONAL RESEARCH COUNCIL Vol.8 Parti MARCH, 1924 Number 43 PEOCEEDINGS OF THE THIRD ANNUAL MEETING OF THE AD-VISOEY BOARD ON HIGHWAY RESEARCH, DIVISION OF ENGINEERING, NATIONAL RESEARCH COUNCIL Edited by William Kendrick Hatt, Director, Advisory Board and E. R. Olhrich. Assistant to the Director TABLE OF CONTENTS Part I Page Summary of development of Advisory Board 3 By-laws of Advisory Board 3 Constituent members and representatives 7 Research committees 9 Registration of attendance at meeting . 13 Part II Program of tlie Third Annual Meeting 17 Interim actions of the Executive Committee 18 Nomination of ofiicers 19 Reix)rt of director 19 Report of Committee No. 1, Economic Theory of Highway Improvement 31 Report of Committee No. 2, Structural Design of Roads 66 Report of Committee No. 3, Character and Use of Road Materials . . 97 Objectives in Highway Research, Thomas H. MacDouald 10.5 Highway Research Work of American Association of Land Grant Colleges, Anson Marston 109 Report of Committee No. 6, Highway Finance Ill Research Program of North Carolina State Highway Commission, Charles M. Upham 124 Report of Committee No. 4, Highway Traffic Analysis 131 Report of Committee No. 7, Maintenance 143 Statements of representatives of constituent organizations .... 155 Election of officers 162 Q II vtr. 4.^ 31 2ds> 06^ PART I SUMMARY OF DEVELOPMENT OF ADVISORY BOARD Chronological Statement The National Research Council was established in the spring of 1916, when President Wilson requested the National Academy of Sciences to organize the scientific and engineering forces of the United States for the purpose of defense. The Engineering Com- mittee of the National Research Council did effective work during the war. On May 18, 1918, by executive order, President Wilson perpet- uated the National Research Council and attached it to the National Academy of Sciences, which had been incorporated by congressional charter, approved by President Lincoln, March 3, 1863. The Division of Engineering of the National Research Council occupies offices in the Engineering Societies Building, New York. The Engineering Foundation, which is closely affiliated with the Engineering Division, provides an office for the Division adjoining its own, and contributes a portion of the funds needed for the admin- istrative expenses of the Division. Advisory Board on Highivay Research: On October 26, 1920, the Chairman of the Engineering Division addressed a communication to the governing boards of certain na- tional organizations. Federal and State departments, and educational institutions, stating the need for highway research, outlining projected committee organization, and inviting representatives to a conference for the purpose of completing the organization. At that meeting, held November 11, 1920, an Advisory Board on Highway Research was organized and by-laws adopted. BY-LAWS OF ADVISORY BOARD As Adopted November 11, 1920, and Amended January 16, 1922; August Jf, and March 23, 1923 (1) Name: Advisory Board on Highway Research of National Re- search Council. (2) Purposes: (a) To assist in outlining a comprehensive national program of highway research and coordinating activities there- under. 4 .\/>\ ISfJh'Y BOARD OS HKJinVAY R/-:!ave of absence for the past two years, and PROCEEDINGS OF THIRD ANNUAL MEETING 23 expressing to Director Hatt a sense of deep regret that he was unable longer to continue as active head of the work of the Advisory Board. Amendments to Sections Qa and 7 of the By-Laws were authorized for submission to letter ballot. Upon the authorization of the Executive Committee, Chairman Johnson and Director Hatt recommended to the Executive Com- mittee that Mr. E. R. Olbrich, construction engineer, North Carolina Highway Commission, be appointed as Technical Assistant to the Director. The appointment was made by the Executive Committee. V. AMENDMENTS TO BY-LAWS Increase in Executive Committee On August 4, 1923, the Executive Committee submitted the fol- lowing amendment to the By-Laws: Section Qa, change the word "three" to "five." The amended section will then read: Section 6a. "Executive: Consisting of the chairman, the vice- chairman, and five additional members chosen by the Board, at least two of whom shall also be members of the Division of Engineering." Add the following: ^'Should a vacancy exist in the Executive Committee it shall be filled by a vote of the remaining members." The amendment was adopted by the following vote : Yes, 24 ; no, 0. Term of Service of Members of Research Committees On the authority of the Executive Committee, the following amendment to the By-Laws was submitted to a vote of the Advisory Board, March 10, 1923 : ''Members of the Research Committees of the Advisory Board are appointed for a term of three years ; at the expiration of the term of office they are eligible for reappointment." This amendment was adopted by the following vote: Yes, 20; no, 1. On August 4, 1923, the following amendment was submitted by the Executive Committee to a vote of the Advisory Board. Add to Section 7 the following: ''The term of services of members of Re- search Committees shall be limited to one year, but they are eligible to reappointment." Vote— Yes, 24 ; no, 0. The complete section as amended will, therefore, read as follows: Section 7. "Research Committees: Shall be created by the Board, as deemed necessary, with the approval of the Division of Engineer- 24 ADVISORY BOARD ON HIGHWAY RESEARCH ing. Members shall be appointed by the Chairman with approval of the Executive Committee, and need not be limited to members of the Board. The term of services of members of research committees shall be limited to one year, but they are eligible to reappointment." VI. RELATIONS WITH OTHER ORGANIZATIONS Eno Foundation A resolution was j^assed by the Executive Committee December 2, 1922, by letter ballot: "The Executive Committee of the Advisory Board on Highway Research recommends that the Eno Foundation take up the study of the leading principles of traffic handling and traffic regulations and offers the services of the Advisory Board in a liaison of this work with the constituent organizations, and the publications of the Ad- visory Board as a channel of connnunication for the reports of this study." Highway Education Board The following subjects for research were submitted to the Advisory Board by the Highway Education Board : a. In highway economics and highway sociology, in metropolitan and urban highway and highway transport problems. b. In highway traffic in connection with the Eno Foundation and other organizations. c. Cooperative studies between representatives of motor trucks and the railroads. d. In highway finance. e. In highway legislation with special reference to uniformity of codes and the avoidance of conflict of laws. Amekican Society of Civil Engineers The American Society of Civil Engineers has appointed an Ad- visory Committee on Civil Engineering Research to cooperate with the Engineering Division of the National Research Council. It is understood that this committee of the American Society of Civil Engineers, acting in this capacity, is a sponsor of research in the civil engineering field. The American Society of Civil Engineers has also a Committee on Research and special connnittees on Highway Engineering and on Impact and Highway Bridges. This latter committee has presented a progress report, published in the INIarch, 1923, Proceedings. At the first meeting of the Advisory Board on Highway Research, on October 8, 1919. a recommendation was made for the appoint- PROCEEDINGS OF THIRD ANNUAL MEETING 25 ment of a committee on Highway Bridges and Culverts, and the By- Laws adopted November 11, 1920, specify a research committee of the Advisory Board on Highway Bridges. During the past year con- siderable discussion has taken place between the Advisory Board and the Committee on Impact and Highway Bridges of the American Society of Civil Engineers concerning mutual support in this field of research. The Advisory Board should not organize its Committee on Highway Bridges until the possibility of useless duplication of work is eliminated. The Society of Automotive Engineers After mature discussion of a combined attack upon relations of road and vehicle, a research project has been adopted on the part of representatives of the Automotive Engineers, the Rubber Association and the U. S. Bureau of Public Roads. This project is now under operation at Arlington, Va., and will be discussed in the report of Committee No. 2 on Structural Design of Roads. The Committee on Subgrade of the Former Federal Highway Council Letters have been exchanged between the Director of the Advisory Board and those interested in the work of the Committee on Sub- grade of the Federal Highway Council looking to a mobilization of the activities of these two organizations. No definite result has as yet been attained. National Transportation Institute The National Transportation Institute is devoted to the w^ork of assembling information on problems of national transportation and distributing this information to the public. The Research Council of the National Transportation Institute is located in Washington and has functions described by its title. The Department of Public Relations of the National Transportation Institute will communicate to the public, through the various channels of broadcasting, news- papers and magazines, only such information as is endorsed by the Research Council of the National Transportation Institute. The Director of the Advisory Board has called the attention of the Na- tional Transportation Institute to the work of the Advisory Board on Highway Research in the field of highway transportation. If the Advisory Board on Highway Research occupies its special field with energy, it would seem that the Research Council of the National Transportation Institute might accept the output of the Advisory Board. 26 ADVISORY BOARD ON HIGHWAY RESEARCH American Road Builders Association This friendly organization has appropriated a sum of one thousand doUars for the support of the Advisory Board on Highway Research, for which the Board is grateful. ExGixEERiXG Foundation The Director of the Engineering Foundation, Mr. Alfred D. Flinn, is vice-chairman of the Executive Committee of the Advisory Board. His long-continued interest in the work of the Advisory Board has brought about helpful relations between these two organi- zations. The Engineering Foundation has also appropriated a thou- sand dollars for the support of the Advisory Board. U. S. Bureau of Public Roads, Department of Agriculture The main support for the operations of the Advisory Board is from the U. S. Bureau of Public Roads. Without the sympathetic and active support of the Bureau of Public Roads, through Mr. Thomas H. MacDonald, Chief, the Advisory Board could not have functioned. Reprksentatives of Constituent Organizations The Executi^•e Committee, on March 10, 1923, adopted the follow- ing resolution : 'Tt is the sense of the Executive Committee of the Advisory Board that interest in the work of the Board would be more widely spread if the terms of service of the official representatives of constituent organizations of the Board were limited to three years." VI 1. MEIOTIN(J OF THE COMMITTEE CHAIRMEN A meeting was held at Atlantic City, June 30, 1923. Present: Messrs. Agg, ({oldbeck, Mattimore, Root, Chairman Johnson and Director Hatt. A meeting was held at Richmond, Va., on October 17, 1923. Present: Messrs. Goldbeck, Hamlin, McKay, Mattimore, Chairman Johnson, Director Hatt and Mr. Olbrich. VIII. FUNCTION OF RESEARCH COMMITTEES The following definition of the function and field of activity of Research Committees of the Advisory Board was formulated and ap- proved by the Executive Committee in March, 1923. 1. The entire purpose of the Advisory Board on Highway Re- search is to render service to investigators, and to accelerate progress in the attack upon important problems of highway transportation. PROCEEDINGS OF THIRD ANNUAL MEETING 27 The Board avoids any attitude of proprietorship in the researches which form the subject-matter of the dehberations of its committees, or of any control or direction of the individual research worker, whose initiative and freedom are necessary conditions of his work. Coordination of research is attempted, inspiration given to those who are at work and the broad interests of research advanced by reasonable publicity concerning activities. Information of work under way is spread abroad, so that seekers after knowledge may know where to find it; communication is established between fields of research, as for instance, the vehicle and the road. 2. Research committees do not formulate standards of methods of test or approved practices in construction. 3. Reports of these research committees are not channels for pub- lication of investigations, although a summary of established con- clusions from published investigations should appear in the reports to show the status of research. 4. The unique work of the research committees of the Advisory Board is: a. To bring together research men working in the same field in order to reach a definition of objectives and to bring to each a knowledge of the technique and progress of his fellow-workers. b. To assemble and review the published data of investiga- tions, or the data placed at the disposal of the commit- tees, for the purpose of judging the extent and stability of the research basis for standards, or for the principles and laws underlying the field of the work of these com- mittees. c. To study needed and profitable research activities. d. To stimulate competent agencies to perform research in such fields. e. To prepare working plans for specific researches in cooper- ation with investigators or for submission to research agencies. /. To call the attention of the Executive Committee to re- searches which need financial support and to recommend a means for securing such support. g. To publish information of researches under way, and of the tools of research, as instanced in Bulletin No. 21, entitled "A Census of Highway Research Projects in the United States," published by the National Research Council. 28 ADVISORY BOARD ON HIGHWAY RESEARCH IX. PUBLICATIONS Since the second annual meeting of the Advisory Board the follow- ing publications have been issued: 1. National Research Council Bulletin No. 21, ''Highway Research Projects in the United States," containing a description of 479 highway research projects classified by a decimal system. 2. National Research Council Bulletin No. 32, "Proceedings of the Second Annual meeting of the Advisory Board on Highway Research," containing an account of the development of the Ad- visory Board, its officers and its constituent organizations, and giving reports of research committees. 3. National Research Council Bulletin No. 35, '^Apparatus Used in Highway Research Projects," containing descriptions and draw- ings of apparatus that have been successfully used in the prosecu- tion of highway research. For convenience of reference, a list of 1~) articles and 26 addresses by the Director since August, 1921, is appended. X. WAYS AND MEANS The By-Laws of the Advisory Board provide for a Cominittee on Ways and Means, which has been appointed but has been inactive. Shortly after the formation of the Advisory Board, Vice-Chairman Flinn planned to circularize a number of prospective contributors to the financial budget of the Advisory Board, but the policy of the Board in respect to this matter was not settled. It appeared wise afterwards to look to official sources, such as State Highway Commis- sions and Federal organizations, for the nuiin support. While the Connecticut Highway Commission aided the Board with an appro- jiriation for a period of two years, the desire of other commissions to take similar action was checked by legal l)arriers. These state high- way commissions have heartily cooperated in the research program of the Advisory Board and members of tiicir technical statfs have given good service on our research committees. At the suggestion of the Director, contrilnitions of money and equipment have been made by industries to research projects that were languishing and to which the Advisory Board was asked to give assistance. In such cases the funds are not handled by the Advisory Board. The only responsibility the Director takes is in a matter of judgment as to the probable successful management of the research. In this way tlie Boni'd. without funds of its own, can assist in sup- porting research. A field of worl< (Icmanding direct expenditure of tlic funds of tlic PROCEEDINGS OF THIRD ANNUAL MEETING 29 Board is in holding; conferences of research workers for the purpose of agreement on objectives of research and upon boundaries of fields of action. This fruif'ul service, together with the expenses attend- ing meetings of the researcli committees, could be extended if funds were available. The preparation of Major Ireland's thesis on results of Quarter- master Tractive Resistance Investigation referred to by Professor Agg in Report, of Committee No. 1, has involved an expense out of all proportion to that normally expected to be borne by a candidate for a degree. The Director of the Advisory Board recommends that funds be raised in the amount of $5,000 for the duplication of the thesis and the reimbursement of the accumulated deficit of this research. LIST OF PAPERS AND ADDRESSES, JULT, 1921— NOVEMBER 9, 1923 BY W. K. HATT, DIRECTOR (*) denotes published article. No. Date Title Where PuMished or Delivered 1921 * 1 July 27 Program of Highwaj- Re- Conference on Highway Eco- search. nomics. University of Mary- land. * 2 Sept. 1 The need for Highway Re- Public Roads. search. * 3 Sept. 12 Discussion of Highway Re- Western Society of Engineers. search. Chicago. 4 Oct. 21 Highway Research Engineers Club of Philadelphia, City Paving Conference. * 5 Dec. 8 Coordination of Highway American Association of State Research. Highway Officials. Omaha. * 6 Dec. 11 Research in Highway Oakland, Calif., Tribune. Transport. 1922 * 7 Jan. 16 Highway Research American Road Builders Asso- ciation. 5 Jan. IS Report of Director, Ad- Annual Meeting of Advisory visory Board on High- Board, way Research. * 9 Jan. 20 Discussion on Research in American Society of Civil Engi- Highway Transport. neers. 10 Jan. 24 Discussion on Research in Highway Education Confer- Highway Transport. ence. Lexington, Ky. 11 Jan. 25 Highway Research Projects Illinois Society of Engineers and Surveyors. Decatur, 111. 12 Jan. 27 Highway Problems Purdue University Road School. La Fayette, Ind. *13 Feb. 14 Report of Committee on American Concrete Institute. Research. Report on Ball Cleveland, Ohio. Test for Concrete and on Fatigue. 30 ADVISORY BOARD ON HIGHWAY RESEARCH No. Date Title Where Puhlisthed or IhUrered 1922 14 Feb. 15 Highway Research Boston Society of Engineers. *15 April 1 Organizing Common Sense. The Constructor. 16 April 17 Recent Development in Report of Division of Kngineer- Work of Advisory Board. ing, N. R. C. *17 May — A Survey of the Field of Highways Green Book. 1022. Highway Research. *1S June 8 Review of Highway Re- Automotive Industries. search. *19 June 21 Progress in Highway Re- A. S. C. E. Portsmouth. X. H. search. 20 June 23 Progress in Highway Re- Society of Automotive Engi- search. neers. White Sulphur Springs. 21 Oct. 18 Address Student Chapter, A. S. C. E. Massachusetts Institute of Technology. 22 Oct. 19 Address Yale University. *2.3 Oct. 26 Research in Highway Edu- Second Annual Conference of tion. Highway Educational Board. Washington, D. C. 24 Nov. 10 Address Student Society. University of Pennsylvania, Philadelphia. 25 Nov. 15 Address Student Society, La Fayette College. Easton, Pa. 26 Nov. 16 Address Student Society. Lehigh Uni- versity. South Bethlehem, Pa. *27 Nov. — Census of Highway Re- Bulletin No. 21, National Re- search Projects in U. S. search Council. *28 Nov. 21 Report of Director, Ad- Second Annual Conference of visory Board on Highway the Advisory Board. Research. *29 Nov. 21 Aspects of Research Association of Land Grant Col- leges. Washington, D. C. *30 Dec. 1 Moot Questions in Highway Princeton LTniversity. Reprinted Research. in "Rock Products," March 10, 1923. ^31 1923 Jan. 16 32 Jan. 16 *33 March — *.34 March — *35 May 8 How Individual Organiza- tions Can Fit Their Work into a Broad Highway Research Program. Address on Highway Re- search. Research Foundation for the Economic Theory of Highway Improvement. Highway Research Report on Highway Re- search Projects in U. S. American Road Builders .\sso- ciation. Chicago. 111. Crushed Stone Association. Chi- cago, 111. Journal of S. A. E.. March. 1923. Cornell Civil Engineer. International Road Congress. PROCEEDINGS OF THIRD ANNUAL MEETING 31 No. Date Title Where Fnhlished or Delivered 1923 *o6 June — Proceedings of the Second Bulletin No. 32, N. R. C. Annual Meeting of the Advisory Board on High- way Research. *3T June — Research Committees A. S. T. M., Atlantic City. Re- printed, "Science,'' September 19, 1923. *38 June 28 International Road Con- Engineering News-Record. gress, Seville. 39 July 11 Opening Statement, Division Convention of A. S. C. E. Chi- of Highway Engineering, cago. A. S. C. E. 40 Nov. 5 Scientific Research in the Purdue Branch of Sigma Xi. Highway Field. 41 Nov. — Coordinated Transportation Purdue University Engineering Review. 42 Nov. 8 Report of Director to the Washington. D. C. Third Annual Meeting of the Advisory Board on Highway Research. The report of the Director was accepted. REPORT OF COMiMlTTEE NUMBER 1, ON ECONOMIC THEORY OF HIGHWAY IMFROM^^MENT In its present form the committee is made up of men actually engaged in research activities and of engineers who are able to assist in the organization of research or in the evaluation of data obtained in the progress of the various actvities of the committee. There has never been a meeting of the full conunittee, but various members of the committee have met occasionally in connection with some scientific or technical convention. The committee is really an informal association of men engaged in a certain line of research, and no attempt to supervise the work of the individual members is made by the chairman of the committee, but he does attempt to aid investigators who find their work handi- capped in some way. Generally, financial assistance is needed, but it has not always been available. An attempt is made to interest new personnel from time to time and to expand the research Avork in the field of highway economics. Duplication of effort is elimi- nated by the publicity given the projects that are inaugurated from time to time. There is presented herewith a chart which shows the field the committee seeks to cover and the progress that has been made up to the present time. 32 ADVISORY BOARD ON HIGHWAY RESEARCH PROJECTS COMPLETED The followiiia; projects in the field of Iligliway Economics have been conii)lete(l and the results have been ]tublished or are in prog- ress of publication at the present time: No. 1. Truck Performance ox Grades This research was organized under the direction of Frank F. Rogers, Commissioner of Highways of Michigan, who assigned Victor R. Burton, in his organization, to the project. The work was carried out with the cooperation of the University of Michigan, which was represented by Walter E. Lay, -\ssistant Professor of Me- chanical Engineering.^ No. 2. Rolling Resistance and Related Characteristics of Roadway Surfaces The research was conducted as a cooperative project, in which the U. S. Bureau of Public Roads, the Iowa Highway Commission and the Iowa Engineering Experiment Station participated.^ No. 3. Economics of Higiivay Grades This investigation was conducted by the Iowa Engineering Ex- periment Station, with some cooperation from the Iowa Highway Commission.^ No. 4. Sub-committee on Tractive Resistance The events leading up to the formation of this sub-committee of Committee No. 1 are as follows : In February, 1920. Major Mark L. Ireland, Q. M. C, was ap- pointed a member of Committee No. 1. Shortly after entering Massachusetts Institute of Technology in September, 1920. as a graduate student, he recommended to the Quartermaster General a research project on tractive resistance. In May, 1921, he began the preparation of a bibliography on tractive resistance. In August overtures were made to the Mason Laboratory of Yale University for laboratory tests on vehicles. In September, Prof. Adams, then Chairman of the Division of Engineering, and Mr. A. D. Flinn, ^ The results of the investigation were published in "The Proceedings of the Eighth Annual Conference of Highway Engineering," held at the University of Michigan. Requests for copies may be addressed to Commissioner Rogers or to Professor liay. = Bulletins CA and (!7 of the Iowa Engineering Exi>erinuMit Station, Iowa State College, Ames, Iowa. => Bulletin 05 of the Iowa Engineering Experiment Station. Iowa State Col- lege, Ames, Iowa. PROJECT AND PROGRESS CHART or HIGHWAY INVESTIGATIONS COMMITTEE ON HIGHWAY ECONOMICS -DIVISION of ENGINEERING - NATIONAL RESEARCH COUNCIL ^U^ CORRECT TO rr" ECONOMICS OF GRADES LOCATION TRACTIVE TYPE LOCATION OF AND SURFACE INVESTIGATOR • I SAND-CLAY iDETERMINATlON .^D GRAVEL MACADAM-PLA NiJ"! ■' BRICK Various types" * portd cement concrete sheet asphalt ASPH'C CONCREl RESISTANCE TYPE LOCATION OF AND SURFACE llNVESTICATQ- VEHICLE COSTS HIGHWAY COSTS LOCATION LOCATION PROBLEM '^"D PROBLEM and "■"■ " INVESTIGATOR """""^ INVESTIGATOR SAND-CLAY DETERMINATIOK AND GRAVEL h >! OF MACADAM-PLAIN TOTAL AND TREATED KtilSIABLL BRICK TO VARIOUS TYPES TRANSLATIONJPORT'D CEMENT IliiiDliSniigB OF CONCRETE p— PNEUMATIC SHEET ASPHALT TIRED IasPH'C concrete TRUCKS ESTABLISHING - BASIC ITEMS OF COST FOR MOTOR VEHICLES OF CAPITAL AND OPERATING . COST FOR MOTOR TRUCKS HIGHWAY CONSTRUCTIOM ESTABLISHING ELEMENTS OF COST OF HIGHWAY MAINTENANCE EFFECT OF CURVATURE """m'natiom TRUCKS TRUCKS WITH TRAILERS Qjiiimig OP I SAND-CLAY TOTAL IM" GRAVEL ■tire losses pneumatic laboratory OF VARIOUS TYPESr^ '^^"'^^^°^- '""■' PASSENGER PORT'O CEMENT !.-«»!«4 1 1 ^f]Js()jn' lioAh'i) ox iiiciiwAY in:si:\uvH )f40.000. Suitability of in.^triiinents and toclini(|iie of tractive re- sistance luive been carefully studied. A larine volume of valuable data bas been accumulated, covering; botb rolling; and sliding fric- tion results on some 10 varieties of road surface of varying age and state of maintenance, employing six types of beavy motor vehicles at speeds a.s bigb as 44 miles per bour witb cargo loads from 0% to 150 9( of tbe rated capacity at all seasons of tbe year and under l»ractically all weatber conditions, and utilizing various ty])es and varieties of steel, solid rubber and pneumatic tires, new and worn. Some of these data have appeared in the preliminary reports of tbe three officers of tbe Quartermaster Corps. Considerable difficulty has been experienced in adjusting tbe length of tbe time, program of tests to available funds and the di- vergent needs of the many co-o])erative interests. Tbe usual vicissi- tudes of research work, all of which make for delay, have been en- countered, such as insulficient funds, administrative detail requiring lime and patience to adapt to the situation of a teni])orary co-opera- tive organization on a new" and somewdiat unusual mission which rc- f{uired development of instruments and methods. The progress and the results achieved, however, have been commensurate with the time, energy and funds expended and the thoroughness which .should be observed in so expensive an undertaking. On November 27, 1922, Major Ireland and Captain Maslin were transferred to the University of Michigan. In June, 1923, Major Ireland received from that institution the degree of Doctor of Philosophy, on the basis of study and a thesis embodying about one- half of the data accumulated by him while at the Massachusetts Institute of Technology. His two-volume Preliminary Report and also a short one by Lieutenant Mitchell are before you now. The records have been transferred to Professor Agg, Chairman, Com- mittee No. 1, for completion and final report in co-operation with Major Ireland and Captain Maslin. Major Ireland is complying witb an iVct of Congress which requires a year of service with troops at stated intervals. No. 7). Cost of ]Motoi; VKnici^i: Opkratiox A preliminary study of the cost of vehicle o]>crali(in was under- taken by the Civil Engineering Department. b)\va Sl;\te College, under tbe direction of H. S. Carter, who was at that time a gradu- ate student in highway engineering. His w(-rk was of most value in indicating some of the difliculties involved in such a study and in pointing out the direction in which tbe work could l)est proceed. He did, however, secure considerable valuable data, and it is expcH'tcMl that this material will be published before \\\v end of the year, al- PROCEEDINGS OF THIRD ANNUAL MEETING 35 though I cannot at tliis time announce who will sponsor the publi- cation. PROJECTS IN PROGRESS The following projects have been in progress for longer or shorter periods of time. Progress reports are submitted for several of these projects. No. 1. Wind Resistance on Motor Vehicles Report by Prof. L. E. Conrad, of the Engineering Experiment Station, Kansas State Agricultural College Professor Conrad was unable to be present to report in person. The report is as follows: The investigation of wind resistance of automobiles was under- taken some two years ago by the Engineering Experiment Station of the Kansas State Agricultural College. The problem was under- taken at the instigation of the National Research Councib but from a financial standpoint the only cooperating agency has been the U. S. Bureau of Public Roads. The original apparatus, in addition to a number of small instru- ments, included two circular tanks of heavy galvanized iron. The larger of these tanks, 17 feet in diameter and 0V2 feet deep, is placed in the ground at such a depth that the top edge is level with the ground surface. This tank is partially filled with water, in which the smaller tank floats. The smaller tank, which is 16 feet in diameter and 2 feet deep, has a wooden deck, on the top of which the automobile is placed. The water level is adjusted so that the top of this deck is level with the ground surface. The additional instruments referred to are an anemometer, a stop- watch, a thermometer, a barometer, four weather vanes, and the recording mechanism. The recording mechanism is merely a de- vice for recording on a chart the pull which the wind exerts on the automobile, as measured by the tendency of the float to move away from its initial position. The test is made as follows: (1) The automobile is run on to the float deck and blocked in the proper position. (2) After a careful study of the weather vanes the float is turned until the car heads directly into the wind. (3) The recording mechanism is located on the float directly in front of the automobile. A wire which leads from the recording mechanism is anchored to a shoi't pipe set in concrete near the edge of the large tank. (4) Four stay-ropes are connected to the float at right angles to the 36 ADVISORY BOARD ON HIGHWAY RESEARCH direction of the wind. These ropes prevent side sway of the float, but can exert no force in the direction of the wind. (5) At a given signal, one observer starts the anemometer and the other observer starts the motor which moves the chart. The ob- server at the anemometer records the number of feet of air passing through the anemometer each 30 seconds. At the end of the test, which runs for six minutes, both instruments are shut off and read- ings are taken of the anemometer, barometer and thermometer. If the wind direction shifted materially during any portion of the test, a note is made of the time, and later that portion of the test is dis- carded. Because of the variable intensity of natural winds, it is necessary to average the results of the records over a certain time interval. The determination of this time interval constitutes in itself a research problem. The shorter the interval, the greater will be the accuracy ; but physical limitations in instrument readmg, recording, etc., make any interval less than thirty seconds impracticable. The thirty second interval was therefore adopted. The records are worked up as follows : The charts are divided into twelve sections, each representing thirty seconds of time. These areas are then traced with a plani- meter and the average ordinate is determined by dividing the area by the length of its base line. This ordinate to scale represents the average pull over the thirty second interval. This pull after the temperature and pressure corrections have been applied, together with the corrected wind velocity, locates a single point on the velocity- resistance curve. Due to the gusty character of winds, these points are somewhat scattered, and it wa.'j found necessary to determine, by the method of averages, the equation of the theoretical curve most nearly fitting the points. By applying this method to over 400 points the equation F = 0.0012 A. V.^-383 was obtained. (F^pull in pounds; A ^pro- jected area in square feet; V ^ wind velocity in miles per hour.) The results of aeronautical investigations indicate that the value of the exponent should be two. The results of Riedler on the "Scien- tific Determination of the Merits of Automobiles" indicate for the co- efficient a value of about 0.003. At this stage of the investigation it became evident that a greater refinement would be necessary. For a number of reasons, it was evident that this refinement would be expensive if natural winds were used, and therefore our attention was directed toward some means of producing an artificial wind. After considerable study and investigation, it was decided to PROCEEDINGS OF THIRD ANNUAL MEETING 37 build a wind-tunnel of sufficient size to test full-sized automobiles. This wind-tunnel, which is now nearing completion, differs from the usual type of wind-tunnel in two respects: First, the cross-section is rectangular instead of circular ; second, the testing portion, where the car is placed, is enlarged in order that the air speed in this section will not be increased by the presence of the car. The cross-section of the tunnel ahead of the enlarged portion is approximately 12 feet wide by 10 feet high, while the entire length is about 40 feet. The enlargement, which is made on the two sides and top, is approxi- mately 9 inches, making the test portion of the tunnel approximately 10 feet 9 inches high and 13 feet 6 inches wide. The amount of en- largement was calculated from the average of the projected areas of a large number of different kinds of automobiles. This area was traced with a planimeter from a plot of the car made with a camera lucida. The entrance of the tunnel is flared to admit the air freely, the size of the opening being 121/2 feet high by 17 feet wide. Tlie exit end tapers down to a circular section 10 feet in diameter. An ordinary airplane propeller is used for a fan. The platform on which the car is tested rests on long beams sup- ported by chains from an outside framework. The zero position of the platform is determined before the fan is started. The fan is then started and, as the wind velocity is increased, the pull required to return the platform to its zero position is determined for each incre- ment of velocity. Although, up to the present time, no complete tests have been made, enough data have been taken to indicate that the tunnel may be expected to give satisfactory results. A wind velocity of sixteen miles per hour has been obtained, and a study of streamers placed at various points in the tunnel indicates that the air currents are suffi- ciently uniform for our purpose. The chief difficulty at the present time is the lack of a fan of the proper design. We have at present a combined motor and propeller efficiency of less than 20 per cent. A propeller of considerably less pitch than those used for flying is required, but such propellers, apparently, are not made except as special orders at prohibitive prices. At the present time the roof of the tunnel is being strengthened, so that the cars may be suspended from the roof immediately above the platform to determine the resistance of the platform itself. The problems for the immediate future are: 1. To design and build an adjustable pitch propeller having eight blades. 2. To determine the correction necessary to compensate for the effect of the wind on the platform. 38 ADMHOh'V BOARD ON HIGHWAY HEfiEARCH 3. To design aii for the deceleration of the rotating parts as well as for the linear deceleration of tlie wdiole mass of the vehicle. Unexpected diflHculties met in the attempt to determine the moment of inertia of rotating paits of the vehicle put an end to the work for the time. The following year an improved chronograph was tried out. to- getlier with another from AVorcester Polytechnic, at Sheepshead Bay Speedway, under the auspices of the Research Division of the Stand- ards Committee of the S. A. E. The Worcester chronograph was car- ried in the car and ga\c a record of time, and of the space traveled, tlie latter pen being actuated l)y a series of contacts on a front wheel. From the experience thus gained an imi)rovement of the Worcester chronograi)h was e\olved which used an ignition timer driven like a speedometer from tlie front wheel and discharging a high tension spark through the moving chart every one-fifth revolution of the front wheels. The S. A. E. adopted this as the standard method of determining the performance of a car. In 191. S the j)roblem was again taken nj) and a new chronograph was built, with many <>f the details considerably improv(Ml. and a series of tests were made on a Dodge and a Eord car. One of the drawl)acks of this method was the tedious counting of the small holes made by the spark and the fact that the spark would not jump in a straight line froui one needle ])oint. through tlie pajier. to the other. Early in 1921. with the eoopei'ation of the .Michigan State High- PROCEEDINGS OF THIRD ANNUAL MEETING 39 way Department, the problem of determining the performance of a vehicle was attacked from an entirely different angle, in that the major part of the data was obtained in the laboratory by laboratory methods and merely a check was made on the road. By this method the characteristics of the power jjlant and the power transmission system were determined in the laboratory and only the rolling re- sistance of the running gear was determined on the road. The fuel consumption of the truck was determined on the road as a check on the results obtained in the laboratory. Four 3-ton trucks, all of the same model — gross weight, 16,000 pounds — were used to give a thor- oughly dependable average result. In these tests the greatest diffi- culty was encountered in determining the efficiency of the trans- mission system. A hydraulic dynamometer of the recording type was also developed to measure directly the rolling resistance of the running gear. A formal report of this work was made in the "Pro- ceedings of the Eighth Annual Conference of Highway Engineer- ing," February, 1922. There was considerable criticism among the state highway en- gineers to the effect that the truck was somewhat heavier than the average commercial vehicle on our roads; so that early in the year 1922 work was again begun on two high-speed 1-ton trucks, using methods considerably improved, as a result of the experience in the previous tests. Since the engines in the light trucks were not gov- erned, a much larger range of speed was covered. In testing the transmission system, the engine was removed and the chassis driven by an electric dynamometer. The output at the rear wheels was measured by Prony brakes and checked ])y a reaction dynamometer on the rear axle itself. In calibrating the large hydraulic dynamometer it was found that there was some friction in a stuffing box which might seriously affect our results on a very light pull, so a much smaller dynamometer was built which avoided the use of a stuffing Ijox. Some of this friction still remained, however, and at present we are in the process of de- veloping the reaction dynamometer used on the transmission system tests. Instead of using a piston and cylinder to change the drawbar pull into liquid pressure, a Fulton pressure element is adapted to that purpose. Since it is a combination of a diaphragm and a bellows, no leakage of the liquid can occur, and as there are no sliding surfaces, friction is reduced to a minimum. We intended to use both methods to determine the rolling resistance of the 1-ton truck. The reaction type dynamometer has a distinct advantage in that it is self-con- tained and the truck upon which it is mounted constitutes an excel- lent instrument for comparing the rolling resistance over various 40 ADVISORY BOARD ON HIGHWAY RESEARCH types of road surfaces, rough roads, curves, etc. Since in determin- ing the roUing resistance of the running gear it will be necessary to correct for grade resistance and acceleration or deceleration, a record- ing accelerometer will be used in addition to the recording reaction dynamometer. Table I. — Inertia Tests. 3-ton Truck Group Equivalent moment of inertia, M. ft.2 Actual moment of inertia, M. ft.2 Weight of parts in pounds Radius of gyration, feet Two front wheels 24.06 24.06 494 1.25 Two rear wheels 67.2 67.2 1304 1.28 Transmission main shaft with 1 sliding gears, propeller shaft, 1 universal joints, final drive [ and axle shafts 43.0 1 Clutch drum and driven f discs, clutch shaft and j gear, transmission i counter- shaft [ Gears 4th 3rd 2nd 1st 2.10 4.62 13.84 44.23 1 .025 f All moving parts of the j engine 4th 3rd 2nd 1st 166. 1'l 365.3 2 05 1094.0 ^-^"^ 3491.0 1 • 2 front wheels r = x/24.00 X 32.2 = 1.25 494 ir wheels r = VQ7.2 X 32.2 = 1.28 1304 2 rear Gear ratio: 4th gear= 9.0 3rd gear=13.36 2nd gear = 23. 10 1st gear = 4 1.3 While the truck performance tests were going on we became inter- ested in the effect of storing energy in the rotating parts of a moving vehicle and the relation of this effect to acceleration and deceleration of the vehicle. If a car or truck is accelerated, kinetic energy of translation is stored in the moving vehicle. A large number of the parts have, in addition to their linear motion, rotative motion, and the connecting-rods and pistons have a very high reciprocating ve- locity. Tlius they have an additional quantity of kinetic energy. By suitable methods the capacity of these parts to absorb energy was determined; also a quantity called the equivalent moment of inertia of all these rotating parts. Tt may be defined as the moment of PROCEEDINGS OF THIRD ANNUAL MEETING 41 inertia of such an imaginary mass, located in the rear wheels and revolving with them, that would have at any speed the same kinetic energy as the actual rotating parts of the truck. Thus, when accelerating the 3-ton truck in high gear, we find that 21 per cent of the stored energy will be found in the rotating parts, and in low gear we find over 75 per cent in the rotating parts. Table IT. — Comparison of Kinetic Energy of Translation vs. Kinetic E nergy of Rotation in Various Gears. 3-ton Truck Gear 4th 3rd 2nd 1st Linear speed of truck, miles/hour And of wheel periphery, feet /second (v) Angular velocity of rear wheel radians per second (w) Equivalent moment of inertia of rotating parts, poundal foot:^ (a) Engine to clutch (b) Clutch to neutral (c) Neutral to rear wheels (d) Rear wheels (e) Front wheels (/) Total — when accelerating (g) Total — when coasting Kinetic energy Of translation, foot-pounds % Of rotation, foot-pounds % Kinetic energy Linear speed of truck, M. P. H Energy of translation, foot-pounds % Energy of rotation, foot-pounds Per cent 12 17.64 11.63 8.1 11.88 7.83 365. 4. 43. 67. 24. 504. 134. 35, 69. 15, 30. 4.7 6.87 4.53 1094.0 13.84 43.00 67.20 24.06 1242.10 134.26 166.1 2.10 43.00 67.20 24.06 302.46 134.26 Accel- erating 76,950 79.0% 20,380 21.0% Coast- ing in neutral 12.0 8.1 I 4.7 76,900 35,100 11,830 89.5%i 89.5%^ 89.5% 9,050 4,100 1,389 10.5% 10.5%! 10-5% 10011,830 5%| 48.0% 375 12,800 5% 52.0% 2.6 3.84 2.53 3491.0 44.23 43.00 67.20 24.06 3669.49 134.26 3,664 23.8% 11,720 76.2% 2.6 3,664 89.5% 429 10.5% Kinetic energy of translation =jv v- "2<7 Kinetic energy of rotation =1/2 Iw^ Weight of loaded truck {w), 16,000 pounds. Truck operating at governed speed (1000 R. P. M.) When decelerating, this stored energy is used to drive the truck, so that when the truck is coasting with gears in neutral, 10 per cent of the driving energy is furnished by the rotating parts. When using an accelerometer to determine the rolling resistance by the coasting method, this relation must be known. Of the above 10 per cent, 5 per cent was stored in the rear wheels, 1.8 per cent in 42 ADVISORY BOARD ON HIGHWAY RESEARCH the front wheels and 3.2 per cent in the rotating; parts of tlio truck between the rear wheels and the transmission neutral. The method of determining the ecjuivalent moment of inertia was to jack up the rear end of a truck with the rear wheels over a well. Ropes attached to weights were wound on the wheels. The weights were allowed to drop a given distance, using their potential energy to overcome the friction of the rotating parts, to give the falling weight kinetic energy of motion, and the rotating parts kinetic energy of rotation. If the velocity, distance and time relations are known, equations may be set up and solved for the equivalent moment of inertia of rotating parts. Tables I and II give some in- teresting data obtained from the 8-ton trucks. Table III gives a comparison of the results obtained so far on the 1-ton s])eed truck with those obtained on the o-ton truck. Table III. — Comparison of Resnlts Thus Far Engine charaft eristics 1-ton o-ton Number of cylinders Bore, inches -Stroke, inches Displacement, cubic inches . Maximum horsepower at 1000 R. P. M Speed of maximum torc}ue, R. P. M Maximum torque, jiound, foot Torque per cubic inch displacement, pound, foot Maximum mechanical efficiency, ]^er cent Maximum over-all thermal efficiency, per cent Minimum specific fuel consumption, Ibs./B. H. P./hr Minimum at 1/2 load, Ibs./B. H. P./hr Transmission System Characteristics l-ton ti'uck o-ton ti-uck Gear Speed, M. P. H. Max. High .... 10 85.0 20 84.5 30 83.5 40 80.0 2nd 5 81.3 10 81.5 15 81.5 20 81.5 1st 10 78.9 Full load 84.7 84.2 83.1 78.8 81.2 81.3 81.4 78.4 Half load 83.0 82.0 80.0 75.0 81.0 81.5 80.5 79.0 76.5 Gear Speed, M. P. H. Max. High 3rd . . 2nd . 1st . . 12.0 8.1 4.7 2.G 83 79 78 77 Full load 82 79 78 Half load 82 73.5 72.0 70 0 PROCEEDINGS OF THIRD ANNUAL MEETING 43 No. o. Relation of Road Tyfp: to Tirk Wear Report by Professor W. C. McNown Purpose. — In the selection of a type of surface for a road it is necessary to take into account not only first cost and cost of mainte- nance, but also the effect of that type upon the cost of vehicle opera- tion. Inasmuch as tire wear causes one of the major items in motor vehicle operation cost, it is desirable to know the relation of road surface to it. Such is the purpose of the investigation, the progress of which is here outlined. Revieiv of previous work on the subject of tire wear. — The writer has made a diligent search for records of previous eftorts in the direc- tion of determining the relation of road-surface type to tire wear, not only in engineering literature, but among those research agencies of the country likely to interest themselves in such a problem. If such work has been done, the search made has failed to bring any record of it to light except in one instance, namely, that of three students working on a Bachelor's thesis at the University of Kansas in the spring of 1922. This work of H. A. INIarch, W. B. Wells and F. W. Goodnow, done under the direction of Prof. C. C. Williams, was the forerunner of the present research. The record of this work may be found in the library of the University of Kansas, in the Bachelor's thesis entitled, ''Relative Wear of Automobile Tires on Brick, Concrete, and Asphalt Pavements." l)y March, Wells and Goodnow. A circular track 59 inches outside diameter was constructed and was surfaced successively with the three previously mentioned sur- face types. Upon this track, by suitable mechanism, were run four ten-inch wooden wheels. Around the circumference of each wheel was fastened a strip of rubber casing material one inch wide which had been cut from a standard automobile tire casing. This wearing strip was carefully weighed before and after a 100-mile run on each kind of surface and the loss recorded in per cent of total weight lost. Other details are omitted. It suffices to say that the work was done in a satisfactory manner, considering the attendant circumstances. The results are shown in the following table: Table I. Per Cent Loss in Weight per 100 Miles. Cousidcriufj Wear of Rear Wheels on Concrete as 100 Surface Rear ivheels Front wheels Asphalt 27.2 15.3 Brick 53.3 34.5 Concrete 100.0 80.6 44 ADVISORY BOARD ON HIGHWAY RESEARCH The plan for the present cooperative test. — It was decided to con- duct the test in two parts, one being accomplished on a specially de- vised piece of laboratory equipment and the other on a Dodge touring car especially equipped for the purpose. This plan was carried out, and what follows will be largely a discussion of the purpose, design, construction and testing of these two pieces of equipment. It became clear very early in the summer that practically the whole season would have to be given to such work, and that little could be ac- complished in the way of running a testing program. The objective determined upon has been very nearly completed. Such additional work of getting ready as needs to be done will be pushed during the year and, as far as can be foreseen, all difficulties ironed out, so that a program of testing can be started promptly with the opening of next season. The laboratory apparatus. — This ai»paratiis was designed to handle certain features of the work which could be done only on stationary equipment, or could be done better, cheaper, or more expeditiously in such manner. The following are some of the necessary items of work thought to Ije Ijest done on it : (a) Determining in a standardized situation the individual char- acteristics of the tires to be used as regards wear. {h) Testing the effect of under inflation on wear of tires. (c) Testing effect of moisture on wear of tires. (d) Testing the effect of traction on wear of tires. (e) Testing the relative wear of tires on such surfaces as could be reproduced on such laboratory equipment. Description of the apparatw^^. — General views may be obtained by reference to Figures 1 and 2. A carriage carrying a 12 H. P. motor is supported by a single rear Dodge Bros, motor car wheel and balanced by four coiled springs. This motor drives the auto wheel by belt and pulleys and through it drives the concrete shod wheel which constitutes the road. The ''road'' is connected by belt and pulley to a generator which is wired to load racks. The attendant uses the wattmeter and the load racks to keep the traction Ijetwcen the auto wheel and the "road" at a predetermined constant. The carriage carrying the driving motor is carefully balanced by weights, and its weight plus the balancing weights plus the tension in the four stabilizing springs give a weight on the auto tire equal to the weight on a roar wheel on a Dodge touring cnr when carrying a normal load or 900 pounds. The pulley combinations are such that the speed of the auto wheel is held nearly constant at 25 miles per liour. The shaft of auto PROCEEDINGS OF THIRD ANNUAL MEETING 45 wheel and "road" were each fitted with revohition counters to check speed, and to be used in the determination of the mileage. The running- unit was taken as five hundred miles, and this appears to be about the right amount. Figure 1. Tire testing apparatus used by Professor McNown. The tires. — The tires adopted are standard for a Dodge touring car. The Kelly-Springfield Tire Company entered into cooperation on the 46 Ah] iSOh'Y BOARD OX HIGHWAY RESEARCH test to the extent of funiishiiig Hve of their ^^taiidard BB o2"x4" eord tires. Tlie i)laii wa?^ to measure the wear in terms of the loss in weight for a given distanee. and this, of eonrse, presupposed getting the tires FiGUHK 2. Tire testing .niipJO'ntus used by I'l-nfcssdi- .McXowii. into a coiKhtion of constant weight in all iclations exeepl (hat of wear. Such ])r()ved to he very diHieult — in fact, wc are nut yet satisfied with PROCEEDINGS OF THIRD ANNUAL MEETING 47 our solution of this problem. Eubber tire^ are more or less porous and readily gain or lose in weight by taking on or throwing off moisture. Within certain limits, their weight fluctuaies with the degree of humidity of the air. Table II gives the history of such variation for one tire through a period of two months. An attempt was made to dry the tires to a constant temperature in a specially devised container heated by a Freas electric oven. Table II shows the degree of success attained. It also shows that as soon as the tires were removed from the dryer they began to increase in weight, due to an accession of moisture. Such absorption doubtless continued throughout the twenty-hour period necessary to make a 500-mile-tire-wear run, affected of course both by the heating and the ventilating to which the tire was subjected during the run. The solution adopted was to take the tire from the dryer, expose it to the atmospheric condition of the room in which the test was to be made, ascertain its hourly rate of change in weight, and compute a cor- rection. Considerable thought and effort has been directed toward finding a moisture-proof coating to protect the tires. E. I. du Pont de Nemours & Company has fiu-nished such a coating in their Viscolac. This, however, is a ccHuloid lacquer and not very suitable for use on rubber. It has not yet been given a trial. The laboratory test. — It is well known that traction is an important factor in the wear of the tires on the driving wheels. For that part of the work which is to be done with the car on the various roads the traction will be a variable, chiefly on account of the different grades encountered on the roads and the different tractive efforts required for different road surfaces. For this reason it was thought necessary to plan for the development of a "Wear-Traction" curve through the use of the laboratory apparatus and the wheel shod with concrete, and possibly with brick and asphalt as well. Such work as has been done this season has been on a concrete-shod wheel and at a traction of 50 pounds per ton. This value was taken simply to develop one point on the curve. In order to control the traction on the laboratory apparatus the efficiency curve of the upper or driving motor was determined ; then the carriage was raised so that the auto wheel was clear of the road and the power necessary to run the auto wheel in this condition de- termined. Then the additional power needed to produce a given traction was computed and that traction maintained by the operator by the aid of the w^attmeter and electi'ical resistance in the form of load racks, so called. 48 ADVISOh'Y BOARD OX HIGHWAY I!ESi:\IiCH so "Si S CO 0^ t^ t^ i^t^ i >. >. >. >. >: >-..t; .t; .t; .t: oj iococo(MOxoo^^>oxt^oioioO'-ocioioixxt^t^ir^i>r^xxioiouo:o'CiCTt O 9 Oi t^ OS or lo CO CO iC O CD OJ c t^ CO iC ta lO lO '^ IC CD CD (N cs OJ c^ CN cs I-H c^ CM CM CO o 't r^ a- lO c- Tt< c- CM o c o iM c CO a T( ir CO or CO CD oc "^ CO CO CD iO CD t^ OS (M o: o- cr cr cr ir. c- C-. CD COCO O 1^ CD r— 1 lOiO 00 00 S'^ CO -^ Oi ^ I>l> IC CD «3 . CJ (.N 05 05 t^ ^ iC 1-H t/1 coco o t^ (N Tt<^ '^ M< CO co" CM GO 00 00 t>r CM CO TtH «o CD CO CO I^ 00 ^ O oo o o o <-> O OO (-> o s o CO oo CO CO CO CO CO OO CO CO o o t^t^ 00 1^ r^ r^ r^ CM CM CM CM H o "Ti (M CO CO CO CO CO CO 1-H IC »C iC lO CO CD CO t^ 00 Oi o 1-H CM CM I-H CM i-H '"' I-H 1-H ^^ o ^ o o o <~) o (-> o o r^ r^ lO •O o <~> &'S T— 1 CM 1— 1 1-H CM CM CM ^& (N CM CM CM cm" cm" cm" cm" CM i:|i IC 00 00 CO 00 I-H =J OJ^-^ i5 S-- lO lO CO lO IC »c CD CM CM CM CM CM CM CM s -s ^^■€ CM »o CD Id o iC t^ CO CD CD CD CO CO CD ^ ^ > CM CM CM CM CM CM CM H 2i 00 CO Oi CM .9 »^-3 »o lO CO O cc g (N CM CM CO 05 CO lO lO CO iCi CM CO CO OJ .2 «J D^ a 1-H lO CD iC '^ a> ■r> CO 00 ^ o CO lO CJ Oi 00 00 o (U-^ rt 3 •o Tj* CD lO >o o o nr o ^ CD CM CM 1^. e^. ^-H (M lO in lo lO CD CO CO IC I-H I-H 03 CO ^ o t^r^ CD t^ CO o^ ^ 05 03 00 IC (C . lO Tt* 00 00 Ol iC o O 00 00 t^ 00 00 Tt* 00 00 ^^ o CD r^ 00 CD CD r^ 00 CO IM CO "■; l^ 00 o CM CM I-H I-H I-H »c I-H d ^ t^ ^ 03 o o o lO O -3 o S a. CO lO CO CD CO ?0 CD -D CO s o o OO o o o o o o oo o o o o O o o o o H o o t^ t^ 00 r^ t^ t^ r^ CM CM rM CM o (N CO CO CO CO CO CO ,-H UD iC »c lO CO CD CD l^ 00 05 o in CM CM ^ CM PROCEEDINGS OF THIRD ANNUAL MEETING 51 to CO CD CD >o lO Tf Oi Ci O c^ CO 1— 1 lO CD r- CO r-) o o o O o o O o o CO o CO CO CO CO CO CO CO o fN (N (N ca C^ CM IM (M (M .—1 lO lO lO LO lO lO lO lO •O (M CO ■* lO CD t^ 00 o o ^ '"' (N O t^ o 05 GO CO CO (/) (M CO CO «o T-H 1-H 1—4 (M C2 !M (N (M o o o lO -^ ^^ t^ CO Oi LO ,_, t^ CO CD r^ 03 o (M CO lO t^ 00 o (M 1—1 (M (M c^ (M (N (M CO CO St; ^>^ CD CO OS O ft lO II M 52 AD] L'^OKY BOARD ON HIGHWAY RESEARCH wliom it was purchased, gave it a thorough overhauling, with especial attention to alignment of wheels and mechanical condition of steer- ing gear. Table IV. — Results in Laboratory Tire Wear Apparatus Loss in grams in a 500-mile run Test No. Tire No. Loss Cor- Road m rected wheel grams Remarks Q 513.562 Q 513.562 Q 513 . 562 Q 513 . 562 Q 512.990 Q 509.721 Q 523 . 538 Q M 6550 Q 509 721 17.6 6.0 20.2 24.2 32 9 34.8 31 0 32.0 20.4 22 1 31 2 32.9 42 2* #1 #1 #1 #1 #2 #2 #2 #2 #2 Not previously dried Not previously dried Not previously dried Dried to a low hourly rate of loss Dried to a low hourly rate of loss Dried to a low hourly rate of loss Dried to a low hourly rate of loss Dried to a low hourly rate of loss Brick shod wheel * At the end of 144 miles the brick shod wheel developed weakness which required the discontinuance of the run. The loss in weight of 12.2 grams was raised l)y {proportion to the 500-mile basis for comparison. In order to measure traction, the car had to be equipped with some form of dynamometer or torque meter, together with autographic register. Prof. A. H. Sluss. Director of Fowler Shops at the Uni- versity of Kansas, was perfecting a torque meter, and after seeing it tried out, we decided to adapt it to our puq^ose. An autographic register suitable to our needs, and much less costly than any that could be purchased, was made in Fowler Shops. Figures 3, 4 and 5 show the set-up of both torque meter and autographic register. To adapt the torque meter to the car, it was necessary to cut the propeller shaft and put flanges on the two ends thus exposed. The torque tube also had to be cut and bridged around the space to be occupied by the torque meter. Roller bearings were inserted between torque tube and |)ropeller shaft near each end of the torque meter. A housing over all to keep out dirt will probably be needed. The autographic register is driven from the propeller shaft by suitable gears and a flexible shafting. The paper speed is about ten inches to the mile. Pressures are transmitted to the paper by a steam indicator gauge using a 50-pound spring. A clock-work, solenoid and pencil puts a mark on the paper at fifteen-second intervals. Calibrafion of the tovfjiir meter. — We are at present engaged in calibrating the tor(}ue meler l)y the use of a Proiiy Ijrake. Tlie car PROCEEDINGS OF THIRD ASyVAL MEETIXG 03 is mounted on a platform and the two rear wheels run on two wheels at each end of the Prony brake-shaft, according to the usual set-up. The manv delays incident to ironino- out mechanical difficulties haye Figure 3. Torque meter. SO delayed this work of calibration that no results are available at this writing. It is hoped very soon to establish the relation between trac- tion at the road and the ordinate of the indicator diagram. 54 ADVISORY BOARD OX HIGHWAY RESEARCH No. 4. Cost of Operation ok Motor Vehkj.ks a. Project organized by Aiiierieaii Antoiiiobile Assoeiatioi port not yet availaMe. 1. Ke- Fic.UKE 4. Torque motci'. b. Project organized by Iowa State College. Tliis project involves not only a stndv of tlic various items tliat enter into tlie cost of PROCEEDINGS OF THIRD ANNUAL MEETING 55 vehicle operation, but also a correlation of these costs with type and condition of highway. A small amount of work has already been done, but the project is practically a new one. Figure 5. Torque meter. No. 5. RoLLiXG Resistance and Related Characteristics of Roadway Surfaces It will be noted that a final report has l;een made on a project under this title. There remain a few factors of immediate impor- 56 ADVLSOEl BOARD OX HIGHWAY RESEARCH tance to Ije inves^tigated, and this work is under way at Ames at the present time. Tlie most important is the determination of the change in rolUng resistance with the thrust (or tractive effort) of the driving wheels. In this work we are using the transmission dyna- mometer designed by Mr. Beebe, of the U. S. Bureau of Public Roads, and I think that the desired information will be obtained shortly. Prof. Agg: Prof. Manly, have you anything to re] tort this year? Prof. Manly: We are going on with the work on "The effect of tires and spring suspensions on maintenance of vehicles and roads" and attempting to develop reliable apparatus and methods of tests. The problem gets harder the further one advances. So far, I have no final results to report. Prof. Agg: This is typical of this kind of problem. We work along for several years on a method of solving the problem. When methods axe once determined, results come quickly. Chairman Johnson: Prof. Agg's report is before you for discus- sion. Are there any questions? Chairman Johnson : What factors are to be taken into considera- tion in the evaluation of the roadbed of highways? Prof. Agg : Tliat is a matter wdiich we hope to take up at a future committee meeting — the question of exactly what factors should be included in the evaluation of the highway in the total cost of high- way transportation. I have ventured to prepare a tabulation in w^hich I have included the cost of the right of way, interest on cost of highway surface, cost of maintenance of right of way, cost of the roadway surface and maintenance of roadway surface ; then the usual plan for an annuity whereby the cost of the roadway surface will be amortized at the end of the economic life of the roadway surface. These are items to be included in the cost of the highway itself and similar items for the vehicle. It is largely my own work and will be submitted to the committee before it is published as committee information. It seems logical to put in the cost of the right of way, but most people do not agree with me on this point. I charge the number of acres per mile in the right of way at $2r)() pw acre, and it amounts to about $2,000 a mile. Mr. Shirley: I find the cost runs about $1,000 a mile. Mr. Manly: Do you allow for appreciation of value in the right of way? Chairman Johnson: Is it charged as an expenditure? The tax- payers having spent the money, should there l)e any further interest charge on money spent as cost of highway? There is no current charge in public ledgers. PROCEEDINGS OF THIRD ANNUAL MEETING 57 Prof. Agg: We charge interest every year on $2,000 at 4 per cent as a capital charge for the right of way. There was no charge in the first instance for right of way taken from adjoining lands. Today we do pay hard money for it. Chairman Johnson: How far back should we begin or cease to make charge for interest? Prof. Agg: If the scheme of evaluation I mention is carried out, payment of interest will not cease as long as the highway is used. Prof. Tilden: In the evaluation of $2,000 a mile, what was the original investment in right of way? Prof. Agg: The value of the right of way. Prof. Tilden : Suppose at the present time there is traffic over the right of way of, say, 1,000 vehicles a day. When that increases to 10,000 vehicles a day, is there any change in the capital charge? Prof. Agg: No. Prof. Tilden: The value of that mile of right of way to the com- munity has unquestionably increased. Mr. Blair: The original investment in Iowa roads was practically nothing, and I am wondering if the use of the road or the use of the abutting lands has developed their present worth. Will Prof. Agg evaluate the right of way in accordance with the fluctuating prices of land? Dean Hughes: Does Prof. Agg intend to readjust this value every year or so? I found the cost of Iowa land a few years ago was less than $50 an acre. Prof. Agg: I do hope to get a unit of cost for transportation per ton-mile on public highways of any type that will be applicable at the time it is published. Whether it will bo any good ten days after I cannot tell. It strikes me as being exceedingly important for the highway engineer to know what it costs him per ton-mile to transport over dirt roads, including cost accruing to the vehicle itself and to the road itself. When it has begun to cost him more to transport over a dirt road than over water-bound macadam, he should trans- form the road into water-bound macadam; and when it costs more to transport over bituminous macadam, he should put a high type pavement on the roadway surface. Above 300 vehicles a day, earth roads can be replaced by paved roads and money saved. The road costs are insignificant costs. It is the vehicle costs that count. So far as Iowa is concerned, the total cost of vehicle opera- tion, including capital and operating costs, will be this year about $225,000,000, and the total expenditure for new construction and for maintenance of highways will be $25,000,000. including every kind of highway. A^ehicle costs entirely overshadow the road costs. 58 ADVISORY BOARD ON HIGHWAY RESEARCH Mr. Lemon : I should like to ask whether rate of tire wear is to be determined by tread wear or by carcass failure. Also, after com- pletion of the rotating concrete wheel tests, are tlie results going to be comparable with road service tests? Mr. McNown : The scheme we are using for testing tires is to measure the loss by wear from the tread. ^^^■ do not test carcass failure. That question has not been taken u\k In measuring the wear, work has been done in the laboratory, ])ut nothing as yet on the road. Whether the work done in the laboratory will correlate with that of the road remains to be seen. The report of Committee No. 1 was accepted. Mr. W. L. Holt, of the U. S. Bureau of Standards, presented a pajDer on RESEARCH ON RUBBER TIRES The various departments of the government sjDend in the neigh- borhood of a half million dollars per year for pneumatic tires. In- asmuch as these tires are purchased on a competitive basis, the setting of proper standards to be used as a basis of purchase is a very im- portant consideration. The U. S. Bureau of Standards, in conjunc- tion with the other interested departments and the tire manu- facturers, has been instrumental in establishing a standard for pur- chases, and with the large volume of business it will be seen that a small percentage increase or decrease in the purchase price, or in the quality, means a considerable amount in dollars and cents. I am going to outline some of the work which has been carried on at the Bureau with a view to establishing a better basis for tire purchases. The present specifications, which have been in use for some time, although revised from time to time, cover detailed re- quirements in regard to the material to be used, including certain chemical and physical tests, and also prescribe methods for putting the tires together. While such specifications serve a useful purpose and aid in sorting out the good from the bad, they are not foolproof, as it is possible to build tires of the best of materials while the re- sulting product will not give satisfactory service. On the other hand, tires have been found constructed of materials of mediocre quality which have yielded at least average service. In an attempt to better such specifications a laboratory endurance testing machine has been installed. With this machine the tire is mounted on a wheel, free to revolve, which is carried on a movable carriage and pressed against a 5-foot flat-faced drum with a pressure corresponding to the axle load. The drum is driven at a surface speed of 30 m. p. h. and thus the tire is revolved and subjected to PROCEEDINGS OF THIRD ANNUAL MEETING 59 flexing such as it would receive on the road. Cleats are bolted across the surface of the drum at intervals so as to give the effect of bumps in addition to the flexing. Contrary to what seems to be the popular idea, the problem in tire-building is not to make a tread which will wear, but to make a tire in which the component parts — the plies, the breaker, the tread, etc. — will stay together, and at the same time distribute the various strains so that the strain will not be excessive at any point. As soon as any of the parts separate, chafing immediately sets in, blisters begin to appear at various places, and the life of the tire is about gone. Accordingly, in developing such a test the motive has been to produce primarily a carcass test, the question of tread wear to be taken care of as a separate test. It has been found that the conditions under which a tire is run — ■ that is, the air pressure used, the axle load, and the nature of the cleats — have a marked effect upon the mileage which a tire will give on this machine. By changing these conditions, different features of a tire can be tested. For instance, a high air pressure and small cleats seem to yield a particularly severe action on the tread portion, and failure is apt to occur, due to tread separation, unless that par- ticular feature is especially well designed. Larger cleats seem to cut down the mileage considerably and are especially severe on the carcass, so that failure is apt to occur, due to ply separation or break- ing of the carcass. If a set of, say, ten different tires is run under each of the two different test conditions such as I have mentioned, the order in which the tires will line up with respect to mileage will probably be quite different. Accordingly, the problem has been to so establish the test condi- tions that the method of failure will be the same as would result from average road service. As I have stated before, such a machine does not determine the wearing quality of the tread, but rather is designed to pick out any carcass weakness which may exist. It has also been the aim to set the test conditions so that a test can be run in the shortest possible time. From the present results, it appears that it will be possible to deter- mine whether or not there is any weakness in the tire in about 1,000 miles of running. This means that, by operating a machine day and night, a test could be completed in less than two days. The Rubber Association of America is actively cooperating with the Bureau in this work and it is anticipated that such a test will be adopted in the near future. Aside from the value of such a test directly to the government in making purchases, the numerous requests which are received from 00 AD\ JSOin nOAL'D ox HIGHWAY TiEf^EARf'H manufacturers all over the country show the need for such a stand- ard test for the tire industry in general. In conjunction with such tests as I have outlined, power-loss tests of pneumatic tires have been carried on for the past two years. Over 100 tires of various makes and of sizes ranging from 31/2 to 5 inches have been tested. The equipment consists of two electric aljsorption dynamometers. One was operated as a motor and carried on its shaft a wheel and the tire to be tested, and the other operated as a generator and carried on its shaft a smooth flat-faced drum. The motor is mounted on a movable carriage, the arrangement being such that the tire can be forced against the drum w^ith a pressure corresponding to the axle load. In this way the tire and drum constitute a friction drive by means of which the motor drives the generator. Aside from the tire itself, four principal variables enter into the determinations: speed, tractive effort, air pressure and axle load. The method of determining the power loss is to mount the tire on the wheel on the motor shaft and force it against the drum by means of weights acting through a bell-crank lever. The axle load is varied by changing the weights. The air pressure may be changed in the usual way. In our case the tire can be inflated or deflated while running by means of a slip joint on the hub. This arrangement also aids in maintaining accurate pressures. Speed is taken care of with a variable speed motor and the tractive effort by cutting out generator resistance. The power loss is the difference between the input to the tire l)y the motor and the output to the generator by the tire. It is measured by mechanical rather than electrical means, which eliminates the consideration of motor and generator efficiency, bearing friction, etc. Tests are made by running a tire under varying conditions and measuring the energy input and output under each. The difference between these figures after making a small correction for windage represents the loss in the tire. This loss is manifested l)y Ihe gen- eration of heat and a rise in the tire temperature. Certain conclusions have been drawn as to the eH'cct of cacli of these variables. 1. Speed. — The power loss is very nearly directly proportional to the speed — that is, at 40 m. p. h. the loss is approximately twice that at 20 m. p. h. This means that the resistance which a tire offers to rolling, or what will be referred to as the rolling resistance, is very nearly constant and is inde- pendent of the speed. PROCEEDIKGi^ OF THIRD ANNUAL MEETING 61 2. Axle load. — The axle load has a similar effect on the power loss unless excessive loads are used — that is, there is almost a direct proportion, and doubling the axle load doubles the loss. 3. Air pressure. — No simple relation exists between air pressure and power loss ; there is a rapid increase in power loss under low air pressures. Under high air pressures the loss ap- proaches a constant value. 4. Tractive effort. — The introduction of a moderate tractive effort does not seem to have much effect on the power loss. For instance, if the input to a tire is one horse-powder, and the output zero horse-power, the loss is one horse-power. If there is a five horse-power input, the output will be almost four horse-power. Under higher tractive efforts, the effect on the power loss is more marked. As a simple and quick means of comparing tires, the following were used : ( 1 ) Zero tractive effort. ( 2 ) The air pressure as recom- mended by the S. A. E. for the maximum recommended load. (3) The maximum recommended axle load for each particular size tire, the loss calculated as the rate per 1,000 pounds load. (4) Express- ing the loss as pounds rolling resistance per 1.000 pounds axle load, which eliminates the item of speed. The following figures are a summary of the rolling resistance of different kinds of tires. Table I. RolUnfi Ucsistance per 1.000 I'omids Axle Load in Pounds Si^-incli fabric tires 14 to 20 Average 17 3iA-inch cord tires 8.4 to 13.8 Average 11.5 4-inch fabric tires 14 to 17.5 Average 16 4-inch cord tires 8.1 to 13.1 Average 11.3 5-inch cord tires 9 to 14.5 Average 11 It will be noted that the average rolling resistance of cord tires, irrespective of the size, is a little over 11 pounds, while for fabric tires it lies between 16 and 17 pounds. This indicates that the per- formance of the different sizes of tires is about the same. In round numbers the rolling resistance of tires varies from 8 to 20 pounds 62 ADVISORY BOARD OX HIGHWAY RESEARCH per 1,000 pounds load, with 14 as the dividing Hne between cords and fabrics. At the present time fabric tires are rapidly l)ecoming obsolete, ex- cept for the smaller sizes; but, taking the cord tires only, the one with the lowest rolling resistance is only 56 per cent of that of the highest. The popular idea is that tliese differences are due to the design of the tread. This has some influence, but it is not the deciding factor. Taking two tires, one with a smooth tread and one with a non-skid or bumpy tread, it is not possible to tell from outside appearance which will offer the most rolling resistance, although, other things being equal, the non-skid type of tire will probably show a slightly greater resistance. The real cause for differ- ences lies in the carcass construction. While from appearances the two carcasses may be identical, small dift'erences in the rubber com- pound, the lay of the cords, the amount of rubber, etc., will mate- rially change the internal friction, and hence the rolling resistance. The fact that the carcass construction is the deciding factor is brought out by the differences between cord and fabric tires. It also empha- sizes the statement made in the firet part of this paper, that the prob- lem in tire design is not to make a tread that will wear, but to make a tire that will hold together with strains reduced to a minimum. As a check on laboratory figures, the losses in several sets of tires have been determined in the laboratory and then the same sets put on an automobile one after the other and the difference in the rolling resistance determined by towing behind another car. This test was carried out at various speeds and under different inflation pressures for each set of tires. While the results have not been completely com- piled as yet, the diff'erences in rolling resistance were apparent at once and the results seem to follow closely the laboratory determina- tions. As to the significance which may be attached to such figures, the first thing which comes to mind is the effect on fuel consumption. Comparing the effect of a change in tire equipment on fuel con- sumption is simply a matter of measuring the loss in each case and finding the fuel necessary to generate the excess power. The diffi- culty of making an accurate statement as to what this difference Avill amount to is the l)ig variation in fuel required per horse-] )Ower, not only in different engines, l)ut in the same engine run under differ- ent conditions. Work is being carried on at the present time to make determinations sucli that some fair average values may bo given. A second value which may be attached \(y j)ower loss deterniinatl(ms is their use bv fhc \\iv desii>;ncr to analvzc the influence of details in PR0CEEDIXG8 OF THIRD ANNUAL MEETING 63 the tire construction. By this means it should be possible more closely to coordinate the different parts and reduce friction or wasted energy to a mininnnn, and at the same time prolong tire life. These investigations are being extended to solid and cushion tires. The latter tires in particular are coming cjuite into prominence at the present time, with many new kinds being put upon the market. The balloon, or air-cushion type, is also being taken up, but not enough data is yet available to draw any definite conclusions as to how the power loss compares witli that of the usual type of tire. From the standpoint of absorbing l)umps, there is no question l)ut that balloon tires are superior. In connection with reclaimed ruljber, the use of a larger percent- age in tire treads is being investigated. One hundred tires have been built, the tread of each consisting of sections containing different combinations of new and reclaimed rubber ; these are being tested on post-ofhce cars run over roads selected with a view to giving as many varieties of road wear as possible. In addition to the direct conclusions which may be obtained from these tests, an opportunity is afforded for obtaining some data as to the value of the various types of laboratory abrasion machines in use. Accordingly, a complete study is Ijeing made on the rubber compounds used in the treads, using as many types of abrasion testing machines as possible. These results in connection with the road tests should give valuable informa- tion for interpreting the results of laljoratory tests. Mr. Crane: Has there been enough work accomplished in the way of results to connect in any way the rolling resistance with tire life? Is there any indication that the tire with the least rolling resistance has longer life? Mr. Holt: In that connection what is measured is the total energy lost in the tire which is made up of the sum of the losses in the unit parts. There may be a small part of a tire in which there is a high unit strain which would not show up as excessive compared with the total loss, and until such can be differentiated the tire with the lowest total loss will not necessarily show the greatest road mileage. Prof. E. H. Lockwood. of Yale University, contributed the follow- ing description of a road seismograph for motor vehicles. ROAD SEISMOGRAPH FOR MOTOR VEHICLES This instrument \\as devised originally in an attempt to measure the value of shock absorbers, snubbers, etc., as applied to motor ve- hicles. It proved useful for the purpose mentioned, and since has found applications in measuring the relative roughness of different 64 ADVISORY BOARD ON HIGHWAY RESEARCH stretches of road, the riding quahties of different vehicles over the same road, and the effect of tire inflation and the so-called balloon tires. Figure 1. Lockwood graph. road seismo- A photograph of the instrument shown in Figure 1 contains the usual suspended pendulum, with an attached mechanism for re- cording the relative motion between the pendulum and supporting- frame in a vertical direction only. The recording device consists of PROCEEDINGS OF THIRD ANNUAL MEETING 65 a reciprocating spool, actuated by a linen cord wrapped around the spool, and tied to the ends of a light vertical rod which forms part of the pendulum. A friction ratchet transfers the motion of the re- ciprocating spool, in one direction only, to a revolution counter, which therefore integrates the reciprocating motion by reading on the dial. In the instrument shown in Figure 1, the spool was made about 13-16 inch in diameter, giving a register of one revolution for about 2.6 inches vertical motion. In using an instrument of this type the counter is read at certain stations, or distances, by the speedometer, and comparisons can be made directly from the counter readings as an index of the vertical shaking. One feature of the instilment deserves special mention because it is not a part of the usual seismograph. This feature is a supporting table under the pendulum, whose object is to prevent the periodic 'Mj&Y-Wlf^AH- dart J"^^ Jifi^ff Z^ rnihsLffu havr FiGUKE 2. Record obtained by Lockwood road seismograph. vertical oscillation of a free pendulum when carried on a vehicle. The motion of the supported pendulum consists of a rebound or jump from the table, of height dependent on the intensity of the shock. The motions of the pendulum may be quite small and frequent. In actual use the counter will read steadily, with no visible motion of the pendulum. The sensitiveness of the instrument to small shocks depends on the location of the table. In the instrument described, the table supported about 2 per cent of the pendulum weight, and as thus constructed will not register on a railway coach over a good roadbed, but will give a considerable reading with any automobile over the smoothest road. An application of this seismograph to vehicle measurement is given in Fig-ure 2. Tests A, B, C refer to the riding quality of three 66 Alt] IS<)],'Y liOAh'I) OX HJ (ill WAY RHSKARCH closed cars witli West fonns of shock absorljcrs driven at 20 miles per hour over a roii,t>,h dirt road. Test D uives a similar record of a heavy toiiriii,u car over tlic same road, e(juipi)ed willi tlie usual form of cord tires. Test E shows the effect on car D of ap])lyin,L>, balloon tires with low inflation pressure. Tests F and G are for other forms of hal- loon tires, showing that different forms of balloon tires are about on a par for riding (pialities. This diagram suggests some useful and practical applications that may be made of a simple instrument of this kind. Mr. W. 8. James, of the U. S. Bureau of Standards, gave an illus- trated discussion on "Tests of Braking and 8topi)ing Conditions of Motor A^ehicles." This is not yet available for publication. EEPOHT OFCOIMMnTEE NI^MBER 2. OX STIM CTUHAL DESIGN OF TJOADS The following report is divided into two main portions: first, that dealing with the facts that have been brought out by researches, and, second, that containing suggestions foi- additional research. PART I TUK ]{KSri.TS OF KKSIOAKCH ON HKiHWAY HESICX {L\as of third axxual meetisg 57 Illinois Department of Public AVorks and Buildings A series of soil tests is under way, but no reports have been made uj^on them to date. New apparatus has been devised for testing soils under many subgrade conditions, including a new slaking test and a new method for making soil cylinders. Soil samples have been transplanted into the ground near the laboratory, in order that they might acquire their natural condition after having been disturbed. Results: None reported. Iowa State College Soil investigations have been carried on at Iowa State College since the summer of 1921. No published report has been made to date. The first year's work was reported to the special committee of the American Society of C^ivil Engineers on the Bearing A^aluc of Soils for Foundation.^ This investigation has the following aims : The determination of the various phj^sical properties of soils in their undisturbed state as found in nature, such properties as the tensile, compressive, shearing, transverse and other strengths, these to be determined upon various soils and at various depths and under various conditions. Supplementary tests, such as true and apparent specific gravities, specific weights, capillarity, hygroscopic moisture. |)orosities, me- chanical analyses, coefficients of friction, etc. Chemical analyses. Colloidality of clay. To determine the laws of variation of pressure in earths under various loads and under differing seasonal conditions. Eesults: Tensile tests gave an average value of 211.5 pounds per square foot for yellow clay and 255.0 pounds for blue clay. Tests were made on sections of two square feet area. Shearing tests gave values averaging 312 ])ounds per square foot for sandy loam, 584 pounds per square foot for yellow clay, and 426 pounds per square foot for blue clay. Test pieces had 2 square feet cross-section. Compression tests were made ui:)on pieces from 6 inches to 9 inches in diameter by iy.o inches to 14.5 inches long. Ll)s. per sq. in. Tons per sq.ft. The average for yellow clay 28 . .5 2 . 05 The average for blue clay 17 . 2 1 . 24 ^ See March r'roceedings! of the American Society of Civil Engineers. 1922, page 557. 68 ADVISORY BOARD ON HIGHWAY RESEARCH Some four tests, given in detail, appear to show that soils, or these two clays at least, have a slight elasticity and do recover shape to a small degree after removal of load. No reports of the work of 1922 and 1923 are available. A. S. C. E. Committee on Bearing Power of Soils The Special Committee on Bearing Power of Soils of the American Society of Civil Engineers have gathered from various sources excel- lent information upon colloids in clays. According to Searle, clay contains (1) colloidal silica, (2) col- loidal alumina, (3) colloidal ferric hydroxide, and (4) possibly colloidal organic matter or humus. These colloids may modify the characteristic properties of clay, but Searle thinks those character- istic properties are not due to colloids. Van Bemmelin found that the fine suspensions of clay in water could be easily precipitated by adding a small quantity of acid, salt or alkali to the cloudy sus- pension. In summarizing the discussion on colloids, the conclusion seemed to be that colloids are suspensions of very fine particles, as follows : Suspensions — Particles over 0.1 micron in mean diameter. Colloidal solutions — Particles between 0.1 and 0.001 micron in mean diameter. True solutions — Particles under 0.001 micron in mean diam- eter. The statement was made that the proportion of colloids that can be definitely separated from clays is less than 3.0 per cent. Some properties of clay most satisfactorily explained by assuming the presence of colloids are : 1. The ratio of water absorbed to the plasticity. 2. The hygroscopic quality of clay. 3. The slaking ratio. 4. That to a certain extent pressure plays the same part as water in certain clays in producing plasticity. 5. That clays are remarkably sensitive to the action of electro- lytes. 6. The swelling and shrinkage of clays under increase or de- crease of water content. Rohland suggests that this power of imbibing a definite quantity of water is due to the colloids in the clay, and that as soon as the clay PROCEEDINGS OF THIRD ANNUAL MEETING 69 has absorbed sufficient water to convert its colloids into a colloidal sol, its saturation point has been reached. Mellor, Green and Baugh have analyzed the substances likely to be present in or to be added to clays into five groups, according to their action on the viscosity of the clay. They are as follows : 1. Substances which first make the slip more fluid, while fur- ther additions stiffen the slip : sodium and potassium car- bonates. 2. Small amounts thicken the slip; larger amounts make the slip more fluid: dilute ammonia, copper sulphate, etc. 3. Substances which make the slip thinner: magnesium, so- dium sulphate, etc. 4. Substances which only stiffen the slip: grape sugar, humic acid, ammonium chloride, calcium chloride. 5. Substances which have no appreciable effect on slip : alcohol. Note, then, that the plasticity of clay may be reduced by adding such substances as ammonia, caustic soda, caustic potash, lime, sodium carbonate, borax and water glass. Ashley and Rohland conclude that the plasticity of clay is directly proportional to the amounts of colloids that they contain. The final conclusion of the committee is: "It seems evident, therefore, that ultra clay is the principal bind- ing material of the soil, giving it plasticity, cohesiveness, or hard- ness, according to the moisture content. The recognition of these important properties shows the fundamental relation the material bears to certain engineering problems, including subgrades in road construction." Note. — The Committee on Structural Design of Roads feels that further research is necessary on the effect of mechanical properties of soils due to colloids before a definite conclusion can be drawn. Ohio State University The Ohio State University reports data on thirteen clays from ten different counties in the state. These are given in the following table. One of the promising series of investigations was made upon im- pact upon subsoils. Four shapes of impact heads were used, with the idea of selecting one that would give results the most uniform and the easiest to meas- ure the deformation of accurately. This impact test would then be 70 ADYhSOh') liOMil) ()\ IIJailWAV RE>2 aSnji.mieo a<;( ua\o]) uAvo.iqi ;ou \v.\.) v..m^\ JIOS STUB.lS JIOS SUIB.lS gz ui puBS puu ')(T|n{<;; s-reaA j; -qt^j ui paup .TIT? majuo.) -I'^ii'AV pai.ip iio<^' jtq paq.ios((i; -la^uw -jqSiaM j^.ip ^ua.o jad 'ua>[Tij si3A\ ajdui-BS uaijM ^uajuoa .la^'B^Y Ar.p JO .iapi;.it'qa .lo a.injxai^ /iv.p JO .10[03 ■ CO oi CO ■* CO t^ X -f O^l Ol CO c 0-1 X X LO -t< Tf Tt t> X O c: -t -.C CO 't^ t^ O t^ CO CO t- '^ X ^ Ol l^ iCi-O -- (M -1^ Ol -t C -f C X X X CO X O c: cc X -+■ (M -* Ol ^ O t^ Tfi ^ QC Oi CK :^ CO c. — .— -t -f -^ — -* -f Ol CO ■^■ lO CO CO CO -t -* Ol XiOXcOO: Ct^005C0 "^ X LO -^ CO CO iQ 03 C '^coTt) .^1 ^ „ CO (M t- iC -O (N LO CO 05 !M CO o «-'"^ CT) -H X X O X CO l^ LO CO LO CO CO 05 T-^ ,-H Ol 0-1 Ol 05 CO Ol 01 Ol i !>^. a a o cS T* (K ~ =t _ c — x o C q; ZJ a: iCi?[o JO ON^ aidurcg K^ K-< ^-l ^-l C _0 >H >H ^-l 1^ Q !^ >-l >H PROGEEDl/NGti OF THIRD AXXUAL MEETING 71 calibrated against static bearing power tests, and when calibrated could be used in lieu of the static test, the advantage being in the ease and celerity with which an impact test may be carried on. The shapes used were: 1. The circular fiat head similar to the ones used by the U. S. Bureau of Public Roads. 2. A flat cone 30 degrees with the horizontal. 3. A truncated cone (30 degrees with the horizontal, with a flat face 21/4 inches in diameter. 4. A segment of a sphere. So far as tests have gone, numbers 2 and 3 seem to give the most even results and fulfil the requirements as to measuring deforma- tions. No calibration tests have yet been made. U. S. Bureau of Public Roads The U. S. Bureau of Public Roads tests aim : 1. At standardization of physical tests for subgrade material. 2. At determination of efficiency of admixtures. 3. At standardization of tests to determine the laws of moisture in the subgrade. Physical properties of subgrade soils. — It is thought that the tests in the laboratory for determining the physical properties of sub- grade soils have now been standardized reasonably well, Avith the possible exception of the bearing value test. These tests are discussed in a paper by J. R. Boyd.^ In addition to the laboratory tests for the bearing value of sub- grade materials, a field test has been devised and is described in the recent bulletin on Highway Research Equipment issued by the National Research Council. - Field test for moisture equivalent. — .Vttention should also be called to a field test for moisture equivalent which gives results very closely checking with laboratory results and with very simple ap- paratus. This test was devised by Mr. A. C. Rose, highway engineer of the U. S. Bureau of Public Roads. It is described by Mr. Rose as follows : Procedure: A 300-gram sample of soil is dried at 108° centigrade and prepared in an iron mortar (when no hard rubber is available) to pass a 1-millimeter screen. A SO-gram sample of this is placed in ^ "Physical Properties of Subgrade Materials." See Proceedings of A. S. T. M., 1922, Vol. 22, pp. 344-347. - National Research Council Bulletin No. 35. "Apparatus used in Highway Research Projects in the United States." C. A. Hogentogler. 72 ADVISORY BOARD ON HIGHWAY RESEARCH a steel bowl. Water is added in small quantities from a burette and a tablespoon is used to mix the water and the soil sample. Care is taken to break up all of the lumps that develop. Water is added until the sample becomes uniformly of the consistency of putty and can just be compacted by the spoon without any free water appear- ing on the surface. The moisture equivalent equals (the weight of water added di- vided by the dry weight of the soil) x 100 plus 0.8. A check run is made and the average of the two used as the final result. This test was made on 29 samples of soil taken from Oregon Fed- eral Aid Project No. 51. Portions of the same sample were taken to the Oregon Agricultural College and the moisture equivalent was determined in a centrifuge (1,000 times the force of gravity) by the standard method. Results obtained by the use of the field method outlined above were found to compare very favorably with laboratory results when the constant of 0.8 was added. It has been stated by Mr. Rose that the moisture equivalent value of twenty is somewhere near the dividing line between a poor and good subgrade. This is probably approximately correct, judging from the results of the analyses of a large number of samples of the Bureau of Public Roads Laboratory, but needs further verification. Ejject of admixtures upon so^ne of the physical properties of sub- grade soils. — In many sections of country the subgrade materials are of a very plastic nature, and when road surfaces are built on such subgrades the inadequacy of the subgrade must be compensated for either by thickened design or possibly by alteration of the subgrade material, which will change a soil of plastic nature into one having more granular characteristics and higher supporting value. It was with tliis idea in mind that the following investigation was under- taken : In this investigation a soil was treated in the laboratory with Port- land cement, high calcium and high magnesium hydrated lime. In each case 5 per cent of the aforementioned adulterants was thor- oughly mixed with the soil. Upon each treated sample the capillary moisture test, the volume test under capillary moisture conditions, and the bearing value test under capillary moisture conditions were made. The results of these tests are shown in Chart I. The conclusions obtained from this chart are as follows: 1. The addition of 5 per cent of high calcium and high mag- nesium hydrated lime increased the capillary moisture of the soil PROCEEDINGS OF THIRD ANNUAL MEETING 73 about 4 per cent, whereas the addition of 5 per cent of Portland cement decreased the capillary moisture about 1 per cent. 2. In spite of the slight increase in capillary moisture of the two lime-treated samples, the per cent of volumetric change of these two samples was very much lower than that of the untreated soils (about 10 per cent). The same phenomena took place in the case of the Portland cement treated samples. 3. The bearing value of the three treated samples was materially increased over that of the untreated soil. Admixtures of sand and other granular materials accomplish similar results. Therefore it may be said that, from a purely laboratory point of view, the physical properties of a soil can be materially improved by the addition of hydrated lime or Portland cement and granular materials. Further research looking into the economy of this pro- cedure should be brought about. The effect of heat treatTnent upon the physical properties of sub- grade soils. — In this investigation two different soils were heated at temperatures varying from 200° centigrade to 600° centigrade in steps of 100° centigrade. On the soil resulting from each heat treat- ment the mechanical analysis, volumetric change under capillary moisture conditions, and adsorption number were determined. The results of this investigation are shown on Chart II. The conclusions may be stated briefly as follows: 1. A temperature of 600° centigrade seems to be necessary to materially change the above-mentioned physical properties 'Of the soils treated. 2. If this temperature is used, the mechanical analysis of the soil is materially changed, in that the clay content is de- creased and the sand content increased. 3. The adsorption number (character of clay) is decidedly de- creased, producing a much less active material. 4. The volumetric contraction was decidedly decreased. It may therefore be said that heat-treating a soil at a temperature of 600° centigrade materially improves the physical properties of the soil, but the cost of this process seems to be prohibitive. Status of drainage and moisture in the sub grade. — Experiments on drainage and waterproofing road subgrades were begun during the summer of 1920 at the Arlington Experiment Station for the purpose of determining, if possible, suitable methods which would promote dryness and stability in this part of the roadway. Both lateral and transverse ditches were cut to a depth of 30 inches below the center line of the finished pavement and kept open during the first winter. 74 ADVISORY BOARD 0.\ HIGHWAY RESEARCH Later, these ditches were obstructed and water allowed to stand in them. During the dry season lawn-sprinhlers kept the roadway as- wet as it was possible to make it. Results: During the winter months, and especially near the ap- proach of spring, all of the sections contained rather high percentages of moisture in the subgrade. In some cases the attempts to water- proof, especially by impervious walls, seemed to trap water under the pavement rather than keep it out. Tile drainage did not in this case decrease the amount of water, except at a depth level with the drain. At that depth the decrease was only about 1 per cent. It seems, then, that drainage of the types used was only eftective in removing water in excess of the capillary moisture. The maximum upward movement of the pavement was ai>pi'()xi- mately three-quarters of an inch, which occurred at the time when there was the greatest volume of moisture in the sul)grade. This was at a time when there wa.s the greatest depth of freezing. The minimum movement at this time was under the sc'ti'^n having the greatest depth of gravel base. Influence of vibmfion: Experiments of this sort were made on a 4 feet X 4 feet concrete slab with free water standing at 12-inch and (Vinch depths below the subgrade. For the purpose of i)ermitting the water to have easier access to the subgrade immediately under the pavement, 4-inch pipes were laid all around the slab at a deptli of IS inches under the subgrade. During the sunnner of li)'22 a 120-pouiid nibber-faced lianuncr was allowed to pound almost continuously on the jiavemeut after the subgrade had readied its maxinnmi moisture content by ca])!!- larity, but there Avas no perceptible increase in moisture due to vibra- tion. These experiments will be repeated under heavy pressures. Ejj'ect of temperature on soil capUlaritij and moist are capaciti/: There is every evidence to show that tem])eiature has a very marked influence on the water ca|)aeity of soils. Tbe colder the soils, tbe greater the capacity for holding moisture. Dr. Bouyoucus, of the Michigan Agricultural Experiment Station, has done nmch work in this field, and his results liave been checked at the Arlington Ex- periment Station. Tliese effects are very clearly shown in Figures 1 and 2, which are typical of conditions in many other cliarts and curves on file in tbe Hureaii of Public Ko;ids. From many exam])les available with shallow depths of frost, it is roughly true that frost action affects moisture content at approx- imately twice the deptli of the freezing. There is an increase in moisture in the frozen earth and a decrease in the soil just below the PROCEUDJNOS OF THIRD ANNUAL MEETING 75 frozen section. This is due to the cold soil having a stronger capil- lary pull than the moisture-retaining power of the warmer soil. The following figures show typical capillary moistures for soils indicated at temperatures of from 0 to 5° centigrade and from 25 to 30° centigrade. Soil No. 20 44 61 74 Ottawa Temperature, 0-5° C. Temperature, 25-30° C. Percent Percent 29.2 25.8 34.5 29.7 36.6 32.6 43.6 37.7 25.8 26.1 Difference Per cent 3.4 4.8 4.0 5.9 —0.3 By comparing the curves on Figure 7, it will be observed that much greater differences than these are found mider actual field conditions. Influence of the quality of soil particles on moisture content, bear- ing value, etc.: This influence is particularly well indicated in Fi(;rRE 1 CHART SHOWING EFFECT OF FROST ACTION ON CAPILLARY MOISTURE AT VARIOUS DEPTHS i E 1 i »; 1 or A i.cMrs I I i, .01. ■: i 1 i s t' t r^ s { 1 ' ""', ■i f. — h WIS S i 1 ^ — ^ •" s S — ^1 : "; 1 s r. \ 76 ADVISORY BOARD ON HIGHWAY RESEARCH Figures 3, 4, 5, 6 and 7. The highest moisture content here will be observed in sponge-like materials like cinders and granulated slag; the next highest will be seen in clays havhig admixtures of hygro- FlGURE 2 CHART SHOWING THE CAPILLARY RISE OF WATER THROUGH VARIOUS MATERIALS UNDER FREEZING CONDITIONS S ^0 Moisture > 10 20 , , 3,0 *y2 '^ = 2 -- S. 8 ^- H -^ . -l J MOISTURE SAMPLES TAKEN THROUGH CLAY SECTION MOISTURE SAMPLES TAKEN THROUGH CLEAN SAND AND CLAY SECTION -SAND 4' DEEP 2/^/2! s-..^ CLEAN SAND TH*W£D NOT FROZCN MOISTURE SAMPLES TAKEN THROUGH CLEAN SAND AND CLAY SECTION. SANO 2" DEEP 2/^4/22 n'-^- - 1 ' 1 i MOISTURE SAMPLE TAKEN THROUGH SECTION OF FINE GRAVEL ROAD i/'S/^2 I'lGURK 3 CHART SHOWING MOISTURE CONTENT OF SUB6RADES ON JAN.I8-I9. 1923 AVERAGE TEMPERATURE OF AIR FOR LAST 10 DATS.35°F. TOTAL PRECIPITATION FOR LAST 10 DAYS 2 02 INCHES PROCEEDINGS OF THIRD ANNUAL MEETING 77 scopic materials like lime and Portland cement. Some materials, such as the many types of clays, take up considerable quantities of moisture when it is available and expand in volume, while other ma- terials, like sand, gravel, crushed slag, etc., take up only enough Figure 4 CHART SHOWING MOISTURE CONTENT OF SUB6RADES ON FEB. 12-13, 1923 AVERAGE TEMPERATURE OF AIR FOR LAST 10 DAYS. 33""^ TOTAL PRECIPITATION FOR LAST 10 DAYS, 0.93 INCH II 12 13 SLAB NUMBERS 20 21 22 23 24 Figure 5 CHART SHOWING MOISTURE CONTENT OF SUB6RA0ES ON MARCH 27. 1923 AVERAGE TEMPERATURE OF AIR FOR LAST 10 DAYS 4ST TOTAL PRECIPITATION FOR LAST I0DAYS.L44 INCHES II 12 13 SlAB NUMBERS rs \/>\is(>h') ii(>\i,'i> o.v iiianww h'i:si:\RCH moisture 1(> conl tlic surfiiccs of tlic ]);n'1icl('s and show little eN])an- sioii. ill these charts it is shown that sand- and otlier kindred ma- terials take n]i less moisture than any othei' material studied. These matei'ials also ha\c much liiLiher heariuij, \ahie^. heeause moisture Figure 0 CHART SHOWING nOISTURE CONTENT Of SUBGRADES ON JUNE 1,1923 AVERAGE TEMPERATURE Of AIR TOR LAST 6 DAYS 75T. TOTAL PRECIPITATION FOR LAST 6 OATS 1.75 INCHES 12 13 -SLAB NUMBERS FiGlKK 7 CHART SHOWING MOISTURE CONTENT OF UPPER INCH OF SUB6RADE ON DATES INDICATED 65 60 55 50 I 45 5*0 < I ^° f 25 20 IS L < - 5 ► t to f n < ? s J ^ J 1 a ; 5 12 5 S ; \ 3 s • 3 5 • i c ? J 5 y •-* S S S *< d i 1 ' o " ^ • 5 } 3 0 3 , c S >e js V 2 § ^ ^ S < I I. 5 o s ? y li y _ " _ .•;; « ? U , ^ A* ^ ; 2 i ; SI 2 i i ? g 5 g g g s g 5 g i 5 % J\x : '/ I \ S S ^ ^ 5 2 i 5 5 5 5 i S > d V y O 5 o ,f/l\\ ; / / / \ \ 5 /;• li^'';^ Ni, \» ^. ^ V. A Hi ■•\ ^ i '^ % \ A F fl > \ ij V > ;•••■" % 1 ^^ '^s* 7* — #,\ / \ % F -^ ^ ^ '^ ;^ :V 1 ^ MOTE 1923 ^ '•• AVtMtE TtMCfflATUIIt or»l« P»ECE£OfNS 101 AYS In4>.>;'n"> TOTA pmcip T.TION rOR PtE EtOINfe 10 OATS EiiZ.)t'iiLi4i II 12 13 SLAB NUMBERS 14 15 16 17 18 19 20 21 22 23 2* PROCEEDINGS OF THIRD ANNUAL MEETING 79 (capillary moisture) does not decrease their bearing values to any- marked extent. It seems to be evident that the characteristics of any soil are de- pendent on the relative percentages of the particles of different char- acteristics present, and for this reason clay soils having high expan- sion, high moisture capacity and low bearing value when moist may be greatly improved by sand admixtures, resulting in a soil of lower moisture-holding capacity, lower volume change and higher bearing values. Such admixtures might be used to create an artificial sub- grade over bad soils. The necessary depth of such artificial sub- grades for various road types has yet to be determined. When macadam or broken stone bases are used over a heavy clay subgrade, it often happens that the clay is forced up through the voids in the stone and renders the broken stone layer much less re- sistant to loads and much more liable to damage through the expan- sion action of frost. Experiments, both of a laboratoiy nature^ and by actual use in roads-, show that a blanket of sand or similar ma- terial interposed between the subgrade end the broken stone is effective in preventing the clay from working up through the over- lying stone layer. Tliis statement is borne out by the experience of railroad engineers who have used cinders under broken stone ballast over bad subgrades. Freezing of moisture in soils: The principal apparatus for freez- ing soils artificially is shown in Figure 8. All other apparatus has been abandoned long since and no mention will be made of this ecjuipment. In no case have we been able to check results with any apparatus for freezing except the one shown in Figure 8. With this apparatus we are able to check to within 0.2 per cent. This shows that the method is satisfactory for measuring the expansion of the frozen moisture in the soil being tested. It is now definitely known that the soil itself does not expand on freezing, but the expansion occurs in the moisture alone. This is proved in this way : We determined the expansion of distilled water by actual measurement and found that 1 cubic centimeter of distilled water expands 0.1 cubic centimeter. This is the same value as that obtained by Dr. George J. Bouyoucus, of the INIichigan Agricultural College Experiment Station. Bunsen's data show an expansion of 0.09073 cubic centimeter per gram of water frozen, while Foote and Saxton, of Yale University, obtained a value of 0.09325. The factor of 0.1 is sufficiently accurate for our purposes, as the method for de- ^ Bureau of Public Roads experiments. = Rhode Island practice. 80 ADVISORY BOARD ON HIGHWAY RESEARCH PROCEEDINGS OF THIRD ANNUAL MEETING §1 termining the frozen moisture in soil is vastly more precise than the uniformity of the soil. Checks are made only by the most accurate method of mixing and sampling. We are able to freeze 100 per cent of the added moisture in Stand- ard Ottawa sand, which is clean and free from dirt, but this same sand when reduced to silt which passes a 200-mesh sieve will permit of only 71.47 per cent being frozen. This phenomenon is mentioned to show what it means to have the water existing in a finer capillary state. It was hoped that some relation might exist between the adsorption number and the percentage of moisture frozen, but it seems that this relation is rather remote. It is highly probable that a chemical analysis of the soil w^ould indicate why there seems to be no such direct relation and would possibly furnish "the missing link" in this work. The table on page 82 gives typical comparisons between moisture frozen, adsorption and percentage of clay content for various soils. It will be observed that in no case have we Ijeen able to freeze all the moisture in any soil, and by applying the factor of 0.1 it is possible to determine the volumetric expansion of the moisture frozen. (b) I'RINCIPLES AND METHODS FOR SAND-CLAY, TOP-SOIL AND SEMI- GRAVEL ROADS The following methods and principles are submitted by Dr. Strahan, of Georgia, as the result of a number of years' investigation; with which, after discussion, the Committee on Structural Design agrees. 1. These materials are to be classed as mixtures of loose aggregates wherein the larger sizes and particles (sand and gravel) have a sta- bility due chiefly to mechanical interlocking or bond, and wherein the finer particles (silt and clay) increase the mechanical stability by filling voids and by greater or less adhesion, usually with the aid of moisture, to the larger particles. 2. The adhesion of the silt and clay to other aggregates and their own internal cohesion are due in part to the nature of the silt and clay, but are also largely influenced by the amount of moisture present and acting in a colloidal state on the finely divided particles thereof. 3. The resultant strength under traffic of these materials is de- pendent upon mass action. The slabs must have adequate thickness and monolithic consolidation. Packed in thin layers, weakness and early raveling may be expected. 82 ADVISORY BOARD ON HIGHWAY RESEARCH 4. Mass action in sucli slabs takes advantage of — a. Internal arching of the materials, b. Resistance to direct compression, c. Resistance to internal shears, d. Resistance to diagonal tension when the silt and clay are adhesive. A Comparison between Moisture Frozen and Adsorption in Certain Soils Soil Moisture Adsorp- Per cent No. frozen tion clay 290 84.0 6.3 16.0 293 81.0 7.5 14.4 158 75.6 2.5 5.2 188 75.0 15.0 15.9 29 74.5 25.0 29.3 13 74.1 17.5 46.0 5 70.0 7.5 11.6 200 70.0 57.5 40.4 201 70.0 15.0 20.4 59 69.3 40.0 34.8 56 69.4 18.8 28.0 294 69.0 7.5 13.6 202 67.2 25.0 18.0 28 67.0 45.0 34.0 119 66.3 20.0 45.6 27 63.8 127.5 52.0 60 63.5 87.5 80 62.3 22.5 41^8 36 61.7 31.3 36.4 14 61.4 55.0 48.8 19 61.4 15.0 29.6 25 60.8 150.0 45.7 11 60.0 8.3 22.4 6 60.0 22.5 32.0 12 59.3 17.5 36.8 291 59.1 7.5 19.2 120 57.5 50.0 76.5 55 57.2 55.0 36.0 92 57.0 35.0 32.0 161 56.7 * 50 55.0 38'8 38.8 54 55.0 52.5 24 54.3 90.0 52^8 186 53.8 35.0 35.0 15 53.8 22.5 41.6 67 53.4 130.0 157 52.5 110.0 45.6 23 50.7 57.5 40.8 26 50.7 154.8 60.4 22 47.7 90.0 56.8 392 47.7 32.5 32 47.5 42.5 33^2 42 43.0 120.0 82.0 44 33.1 140.0 71.6 161 22.9 * * Burned. PROCEEDINGS OF THIRD ANNUAL MEETING 83 5. Full bearing and impact resistances are not developed near the edges of the uncurbed slabs. Weakness may be expected for a dis- tance about twice the thickness of the slab from the outside edges if unsupported. Usually these materials are deposited in trenches with supporting shoulders. 6. Such slabs show some degree of elastic reaction, but not accord- ing to quantitative laws. Laboratory specimens for transverse break yield by tension for spans longer than twice the vertical depth of the specimen. For shorter spans the yield is by diagonal tension from the load point downward toward the point of support. 7. Successful roads require — a. Graded mixtures capable of making a dense mass on consolidation. The best mixes run as low as 20 per cent porosity. b. Uniform composition, secured by thorough mixing. c. Heavy compacting from the bottom upward. This is usually done by traffic. There is need for a type of multiple-rim roller to hasten and unify the degree of compactness imparted to these roads during con- struction. d. Mixtures with a liberal amount of coarse material above No. 60 sieve. 8. The detailed classification, percentage limits, and methods of laboratory analysis and tests are to be found in U. S. Department of Agriculture Bulletin No. 555, issued by the Bureau of Public Roads ; and in Univereity of Georgia Bulletin Vol. XXII, No. 5a, June, 1922. The procedure is to separate the original sample as follows : (1) Tlie well-pulverized sample is weighed and passed through a No. 10 standard sieve. The residue on the sieve is called "coarse material." (2) The material passing through the sieve is regarded as a "soil mortar" and is analyzed as follows: Fifty grams are thoroughly shaken and rubbed with about 250 millimeters of water, adding 5 millimeters of a 4 per cent solution of ammonia; allowed to stand 8 minutes and siphoned to a depth of 8 centimeters from the surface. Operation repeated until a clear super- natant liquid is obtained. The fine sediment thus removed is called "clay" or "total clay." The total clay can be further separated by centrifuge into "suspen- sion clay" and coarse clay. The washed residue is dried on a water bath and separated through a nest of sieves — Nos. 20, 60 and 200. All residue above the No. 200 sieve is classed as "sand." The residue 84 ADVISORY BOARD ON HIGHWAY RESEARCH passing No. 200 and caiiglit in the pan i< called "silt."' The sand coarser than No. (30 sieve is a most impoj'tant criterion. Unless it reaches or exceeds 80 per cent, it is hopeless to expect a satisfactory soil mortar. Classificatiox of Soil Mortaks Class A Class B Class C Percent Percent Percent Clay 9 to IS 15 to 25 10 to 25 Silt 0 to 15 10 to 20 10 to 20 Total sand 65 to 80 60 to 70 55 to SO Sand above No. 60 sieve 45 to 60 30 to 45 20 to 30 Hard Medium Soft For further details see bulletins referred to above. Coarse material, above No. 10 sieve, has a decided influence on the durability, hardness and smoothness of the surface. Ten per cent of such hard, gravelly material, with a Class B soil mortar, justifies reporting the whole as a Class A material. 9. Extensive experience in the field and in the laboratory support the following comments: (1) As to soils having no coarse material above No. 10 sieve: a. Classes A and B give respectively admirable and adequate traffic service for the expense involved. It is doubtful whether Class C soil mortars should be used except on secondary roads with very light traffic and when the ma- terial is close at hand. h. Tlie most important factor is the presence of coarse sand in liberal amount. Of course, clays and silts differ markedly, but roads built of many dift'erent clays have shown ex- cellent durahility. It will be noted that the total clay present in the best samples rarely exceeds 18 per cent. Efforts are, however, being made to differentiate between clays by laboratory tests, and some tentative progress has been made. c. It may be accepted that the sand lie.ow No. 00 sieve lias little supporting value. It has some xahie in lilling voids when coarser sand is present. d. Highly organic silts have binding value and strengthen the mixture, so long as the organic niatlcr persists, but such silts lose the oi-ganic matter ([uiti rapidly by decay and dusting. e. The more adhesive or highly colloidal clays give effective bond with as little as 9 per cent present. Eess plastic clays PROCEEDINGS OF THIRD ANNUAL MEETING §5 are needed in larger amount np to a maximum of 25 per cent, but the mixtures with 20 to 25 per cent clay are apt to be slippery and cut into mud in long spells of wet weather. /. Laboratory tests indicate that the adhesive values of clays are largely dependent upon the colloidal state of the material and the moisture present. Metal plates cemented to- gether with thin films of clay show markedly greater ad- hesion when containing 5 to 10 per cent of water than when dried at 212°. g. In general, it is believed that inorganic silts tend to diminish the strength of road soils, and noticeably so when present above 15 to 20 per cent of the soil mortar. (2) As to soils containing 10 per cent or more of hard coarse material above No. 10 sieve: a. It is a matter of assured experience that such coarse material adds greatly to the durability of the mixtures. Such soils are found in many places and give rise to a series of semi- gravels reaching from 10 per cent up to the 50 or 60 per cent in a full gravel road. b. A good Class A soil mortar w^itli from 20 to 30 per cent coarse material gives a fine surface of marked durability. This is especially true of the iron-silica gravels in southwest Georgia, with from 25 to 40 per cent of a nodular iron- impregnated gravel, ranging from one inch downward. These nodules are not nearly so hard as true silica gravel, but are sufficiently strong to give most excellent service and very smooth travel. The effect of wet loeatlier a. Moderate ground moisture increases the strength of all classes of these roads. b. Long and heavy rains influence them as follows: Class C. Soften the surface so that wheel ruts two or more inches deep may appear under heavy vehicles, and when the sand is fine considerable material is washed down the steeper grades and into the side ditches. They do not show deep mud and are not slippery. Repairs are readily made with drags and road niachines. Class B. Soften only after very long and continued wet Aveather, wheel ruts rarely as deep as two inches, not much 86 ADVIf^ORY BOARD ON HIGHWAY RESEARCH washing into side ditches. Easily repaired with road machines. Dry out promptly and are usually too hard for any but the heaviest metal drags and light road machines. Class lA. Do not materially soften except where water stands for a long time in depressions. Wheel ruts rarely deeper than one inch. The slab below this surface nnid is generally hard and firm, and will require the heaviest road machines to reshape the surface. Dries off rapidly and soon becomes too hard for the heavi- est road machines to cut. c. Laboratory tests show that the road soils are of marked density, are not readily penetrated by rain on the surface, and when made into briquettes with varying amounts of w^ater they retain the larger part of their tenacity until 15 to 25 per cent of water is added. Thickness and crown The thickness used is designed to supply mass action and to pro- vide ample shearing resistance against traffic pressures and impacts. The usual specification has been ten inches when packed in place. On weak subgrades, especially those subject to ground water satura- tion or capillary lift from below, this should be increased to 12 or 14 inches. There seems little or no advantage in exceeding the above, for the following reasons: 1. Examination of the packed road slal) shows that the hard pack rarely extends more than 6 inches deep. The material below this, while firm, is easily penetrated by a pick and comes up in small pieces rather than as a large clod. 2. The lower portion is thus functioning less actively in bearing resistance, but most probably it serves usefully as a drainage layer and also as a cut-off plane against capillary lift of water from the subgrade. 3. The surface wear of the better soils is not so rapid as to demand a large margin of extra thickness. The roughness which comes under wear aft^r five to eight years will require scarifying and re- shaping. The most expensive part of this process is pulverizing the clods, and the addition of new soil saves the necessity of much fine pulverization. It also restores the thickness of the original slab. As to the proper crown, the earlier view advocated a high crown, even as much as 1 inch per foot, but later experience has shown that one-fourth inch j)er foot is adequate and much more acceptable to the traffic. PROCEEDINGS OF THIRD ANNUAL MEETING 87 Research investigations With the vakied cooperation of the U. S. Bureau of PubUc Roads, there is in progress in Georgia, under its State Highway Department, an important and state-wide study of the effective life and economic results of 29 Federal Aid Projects built of these materials and under constant maintenance. The program embodies: a. Samples taken at frequent intervals on each project at speci- fied points and careful analysis of same. New samples are taken each year, and will continue to be taken until the surface must be reconstructed. b. Traffic counts 8 days per month for one year, repeated in alternate years. c. Careful field inspection (with photographs taken) made in summer and winter. d. Taking of cost data on maintenance. Thus far. after 18 months, all but two of the projects show very satisfactory stability. These two were constructed of poor soils, not recommended by laboratory tests. Some of the projects have been under traffic for more than four years. Incident to the above economic research, the laboratory is seek- ing to solve the very intricate problem of the specific interaction and influences of the gravel, sand, silt and clay in such mixtures. A great variety of tentative tests have been tried out. Attention is called to several of the more promising one^. as follows: a. The Adsorption Test for Distinguishing Clays: This test was first devised and promulgated by the Bureau of Public Roads, using standard solutions of ci-ystal violet filtered through and adsorbed by the material. Special filter tubes are used. (See Proc. A. S. T. M., 1922, Vol. 22, Part II, pp. 344-7.) The possible value of this test as applied to original samples, to the separated ingredients, to the clay in particular, is easily seen. Further, the total clay itself can be further fractionally separated in a centrifuge and the adsorption of these fractions studied. This test has been followed on the soil samples from the 29 Federal Aid Projects with interest in this laboratory. It is noted that the soils and clays of this state show markedly lower adsorption values than the subgrade materials reported by the Bureau. Change of the color medium to methylene blue has been tried out. In general, samples at this laboratory adsorb somewhat larger amounts of the latter than of crvstal violet. Methylene blue also 88 ADVISORY BOARD ON HIGHWAY RESEARCH has certain advantages of stability of rolor. and is not apparently influenced by acid or basic condition of the sample. Crystal violet is thus influenced in color. Further, the pure blue color is free from pronounced overtones, and a test by adding excess color, agitating vigorously or boiling can be applied and the excess determined in a colorimeter. Less time is required for this test than for the very slow filtration incident to the filter-tube method. This modification of the test is being fully tried out. b. The Disk Shear Test : Results fairly consistent have been ob- tained by this test, conducted as follows: Three disks, 2 inches in diameter and 1 inch high are prepared from each sample. The soil is moistened to stiff mud consistency and compressed under 3,000 pounds. A thin tin collar is used inside the mold, which aids the removal of the disk. The disks are dried in air or on the water bath. For test, the disk is inserted into the test cylinder and broken through in shear by the 1-inch steel plunger. ConclusioTis by Comrrhittee. — 1. Laboratory tests and procedure for determining physical properties of subgradc materials seem to be well developed. 2. Tile drainage and ditches remove only moisture in excess of capillary moisture in the soil. 3. Dift'erence of temperature in soils tends to cause motion of moisture from the warmer to the colder layers. 4. Undesirable, plastic and finely divided soils are improved by admixtures of granular materials and small percentages of Portland cement and hydrated lime. 5. A blanket layer of granular material beneath the base course of macadam or gravel roads is effective in preventing settlement and penetration of the subgrade into the base course and tends in general to reduce heaving and displacement of subgrades. (c) FORCES ON PAVEMENTS RESULTING FROM TRAFFIC The following is reported by Mr. E. B. Smith as a result of the Bureau of Public lioads' tests. They are to be regarded as pre- liminary results and are now being verified in a large series of tests in cooperation with the Rubber Association and the Society of Automotive Engineers. Before strength designs may be definitely made, it is first necessary to determine the loads or forces which must be sustained. This applies no less to the design of a concrete road slab than to any other structure. The loads and forces to be sustained by a road are those PROCEEDINGS OF THIRD ANNUAL MEETING 89 resulting from the action of traffic, and those most severe are mainly from heavy motor trucks. A former series of tests were run about two years ago to determine the value of impact of different trucks, and these investigations indi- cated quite a wide range of values. (See Mar. and Dec, 1921, "Public Roads.") The tests were run over a selected stretch of road on which were placed different types of obstructions. Under the condition of having an obstruction 1 inch high and with a 7%-ton truck fully loaded, running at 15 miles per hour, an impact value of 28,000 pounds was secured when using worn solid tires. Higher impacts were obtained when using higher obstructions. A later series of tests was run to determine the average impact that may be expected from different trucks traveling over a certain selected stretch of road. The trucks selected were a 2-ton Mack and a 5-ton Pierce-Arrow, both trucks being constructed with the usual rear-axle differential. An accelerometer of the new type was at- tached to the rear axle, which gave deceleration values of the un- sprung weight, to which should be added the sprung weight on one wheel. The conditions of the tests and results are as follows : 1. 2-ton Mack Truck, No Cargo; Static Rear Wheel Load, 2,285 Pounds a. Gravel road — bad ruts — high speed (15 to 18 m. p. h.), 12,500- 18,500 pounds maximum impact force. b. Gravel road — smooth — slow speed (5 m. p. h.), %-inch ob- struction, 3,000 pounds maximum impact force. Gravel road — smooth — slow speed, 2-inch obstruction, 5,600 pounds maximum impact force. Gravel road — smooth — moderate speed, %-inch obstruction. 3,300 pounds maximum impact force. Gravel road — smooth — moderate speed, 2-inch obstruction, 8,000 pounds maximum impact force. 2. 2-ton Mack Truck, 3,000-pound Cargo; Static Rear Wheel Load, 3,735 Pounds a. Concrete road (Columbia pike) running over one experimental section at a time at about 8 miles per hour, the impact values varying from 6,600 pounds to 13,235 pounds, most of the thirty-two sections giving impact values around 7,500 pounds. b. Concrete road — maximum impact value at maximum speed (about 15 m. p. h.), 20,100 pounds. 90 ADVISORY BOARD ON HIGHWAY RESEARCH 3. 2-ton Mack Truck, 4,000-pound Cargo ; Static Rear Wheel Load, 4,285 Pounds The increase in cargo seemed to reduce the tendency of the rear end to jump, so that the accelerations obtained for this cargo load were consistently a little lower than for the smaller cargo loads and gave impact values which were slightly lower. 4. 5-ton Pierce-Arrow, 10,000-pound Cargo; Static Rear Wheel Load, 8,875 Pounds a. Gravel road — smooth — ys-inch obstruction. 8 miles per hour — 13,000 pounds maximum impact force. 10 do. 15,000 do. 12 do. 16,800 do. Gravel road — smooth — 2-inch obstruction. 8 miles per hour — 22,500 pounds maximum impact force. 12 do. 28,700 do. 15 do. 33,700 do. b. Concrete road — %-inch obstruction. 8 miles per hour — 16,800 pounds maximum impact force. 10 do. 19,600 do. 12 do. 21,250 do. 15 do. 27,800 do. Concrete road — 2-inch obstruction. 6 miles per hour — 22,600 pounds maximum impact force. 10 do. 34,000 do. 12 do. 36,800 do. c. When running at full speed over i^-mile to 2-mile stretches of typical concrete pavement at maximum speed (sometimes reaching 20 miles per hour on downgrades), impact values ranging from 28,300 pounds to 38,700 j^ounds were ob- tained. These preliminary tests indicate that when driving over a typical concrete road at high speed with a fully loaded 2-ton truck, we may expect impact values of approximately 20,000 pounds, or over five times the static load on one rear wheel. With a fully loaded 5-ton truck at high speed, we may expect forces of 35,000 pounds to 40,000 pounds, or over four times the static load on one rear wheel. The impacts as shown in tlie table represent the values usually ob- PROCEEDINGS OF THIRD ANNUAL MEETING 91 tained, but the maximum value of the force sustained by the road may at times be much greater, owing to the oscillating action of the truck body. If a downward movement of the loaded truck body occurs at the same instant as an impact, the two forces are, of course, additive, and the result may be possibly double the impact. There- fore, under certain bad surface conditions a total force of 40,000 to 60,000 pounds may be obtained at each rear truck wheel. The com- mittee feels that the facts above stated were ascertained by careful test methods and concurs in the tentative conclusions. (d) GENERAL CONCLUSIONS FROM FIRST SERIES OF TESTS ON PAVEMENT SLABS AT THE BUREAU OF PUBLIC ROADS The following report is submitted by Mr. C. A. Hogentogler, of the Bureau of Public Roads : These tests were conducted on slabs 7 feet square laid directly on the subgrade. They were subjected to impact of a specially con- structed machine designed to simulate the blow struck by the rear wheel of a motor truck. Essentially, the machine consisted of an unsprung weight shod with a rubber tire and a sprung weight sup- ported on the unsprung weight by means of a 5-ton truck spring. Repeated blows were struck in the center of the test specimens, in- creasing impacts being obtained by raising the height of fall. First tests of impact on pavements of the Bureau of Public Roads, carried on at the Arlington Experimental Station in 1920, a report of which was published in ''Public Roads" for October and Novem- ber, 1921, indicated that: 1. Resistance of 1: l^^: 3 concrete slabs 7 feet square to impact blows increased consistently with increase in thickness. Beginning with 4-inch thickness, which resisted 14,000 pounds equivalent static load, the resistance of slabs increased 5,000 pounds with each inch of additional thickness. Specimens tested as beams, however, indi- cated that resistance to static loads varies as the squares of the depths. 2. The monolithic brick slabs (concrete base, i/4-iiich sand cement cushion, cement-grout-filled brick tops) in most cases showed less resistance than 1: l^l;: 3 concrete slabs of the same depth. The re- sistance of the eleven slabs on this particular subgrade was equiva- lent, on the average, to that of the concrete slabs having thicknesses of but 82 per cent of those of the monoliths. 3. Semi-monolithic slabs (concrete base, 1-inch sand cement cush- ion, cement-grout-filled tops) showed less resistance than the mono- liths. The average of the resistances of four slabs was equivalent to that of concrete slabs having thicknesses of but 68 per cent of those of the semi-monoliths. 92 ADVISORY BOARD ON HIGHWAY RESEARCH A second large series of impact tests is just being completed by the Bureau of Public Koads, but results are not available for the present report. (e) RESULTS ESTABLISHED BY THE PITTSBUUG TEST ROAD INVESTIGATION The following conclusions are stated by Mr. Lloyd Aldrich as a result of tlie studies of tlie Pittsl)urg Test Uoiul investigations: Note. — Reference should be made to page numbers in ''Re- port of Highway Research at Pittsburg, California, 1921-1022." 1. The method of constructing the subgrade on adobe soil, as de- scribed, SO reduced the objectionable features of this material that a reliable foundation for the pavements resulted. Reference (page 141; title, ''Discussion"; subtitle, "Subgrade"). 2. This subgrade was not injured by the water which filled the side ditches for three months. Reference (page 141: title. "Discussion": subtitle. "Subgrade"'). 3. The unsurfaced concrete completely resisted all surface wear due to solid rubber-tired traffic. The limited metal-tired travel indi- cated that it would cause, with sufficiently heavy loads, an early fail- ure of surface. Reference (page 141; title, "Discussion"; subtitle, "Surface Wear"). 4. Early morning travel was more injurious to the pavement than the day traffic. Reference (page 143; title, "Discussion"': subtitle. "Deflections under traffic loads"). 5. The observed deflections of the pavements are very nearly di- rectly proportional to the loads. Reference (page 95; title, "Special Tunnel Studies"; also, page 94, Plate 36). 6. Those sections that had steel placed in such positions as enabled it to resist the tension flexural stresses were more durable than those sections of the same dimensions that did not contain steel so placed. Reference (page 145; title, "Discussion"; subtitle, "Deflections under traffic loads"; also, Table 17. i)age 143; also, page 146, sub- title, "Effect of traffic on corners of plain and reinforced concrete"). 7. The rock ballast, as constructed under Section A, was less effi- cient than the earth subgrade. Reference (page 145; title, "Discussion"; subtitle, "Deflections under traffic loads"). Note. — Furthci' reference in entire body of report. PROCEEDINGS OF THIRD ANNUAL MEETING 93 (/) GENERAL CONCLUSIONS FROM THE BATES ROAD TEST Mr. Clifford Older has submitted the following conclusions as a result of the Bates Test at Bates, Illinois, conducted by the Illinois State Highway Department: 1. The endurance limit of concrete when subjected to repeated tensile stresses may safely be taken as not less than 50 per cent of the modulus of rupture of the material. 2. That the resistance to structural failure of plain concrete pave- ment slabs of uniform thickness and laid on a uniform subgrade, when subjected to highway traffic loads, is in proportion to the square of the thickness of the slab. 3. That in plain concrete pavement slabs overloaded by ordinary traffic, structural failure normally first appears at corners formed by open cracks or joints, which either intersect each other or the edge of the pavement slab. 4. That corners formed by intersecting cracks or joints which are held in close contact, either by compression set up by temperature expansion or artificial means, may be expected to resist structural failure under loads much greater than those required to cause failure where the cracks or joints are open. 5. That in concrete pavements having widths up to 18 feet, a center joint may be so designed and constructed as practically to ehminate the formation of longitudinal cracks and to insure a close contact along at least one edge of all interior corners. 6. That the contraction of concrete pavement slabs concentrated at transverse joints or cracks, thus causing the formation of corners unsupported along either edge, makes necessary the strengthening of the edges of ordinary rural pavements in order that such corners, as well as the edge of the pavement, may afford at least approxi- mately the same resistance to structural failure as the interior por- tion of the slab. Another conclusion reached was that soft tops, such as sheet asphalt, asphaltic concrete and bituminous-filled brick, may add slightly to the resistance of the rigid pavements to structural failure when laid on good subgrades by minimizing the effect the daily warping of the slab has on subgrade support ; but that this advantage over uncovered concrete pavement slabs which are divided by center joints is so slight that it can be practically neglected. The committee has discussed with interest and submits the con- clusions suggested, respectively, by Mr. Aldrich on the Pittsburg road and Mr. Older on the Bates road. The committee will suggest under Part II further research endeavors, which include some prob- lems arising in connection with the above reports. 94 ADVISORY BOARD ON HIGHWAY RESEARCH (g) THEORETICAL ANALYSIS OF STRESSES IN ROAD SLABS A theoretical investigation of the mechanics of concrete slabs is now being undertaken by Dr. H. M. Westergaard for the Bureau of Public Eoads. The following note is contributed by him : "A contribution to the understanding of the mechanical behavior of concrete roads should be obtained by application of the mathe- matical theory of elasticity. This theory has methods available for determining the distiibution of stresses in an elastic slab resting on an elastic subgrade and carrying given loads. The resemblance of a structure of this kind to the road slab, with its subgrade, is suffi- cient to warrant the assumption that exact information concerning the perfectly elastic slab, when its dimensions and physical constants are chosen properly, may serve as approximate information concern- ing the road slab. Since it is possible by the theory of elasticity to investigate rather completely a wide range of cases, it appears to be a reasonable plan to obtain theoretical data for use as a basis for in- terpretation of experimental data and as a means for interpolating between the results of one test and another. "A rather extensive literature deals with the theory of flexure of elastic slabs. Two papers deal with the subject of slabs on elastic support : one, by Hertz, was published in 1884 ; the other, by Hap- pel, appeared in 1920.^ Notable contributions to the general theory of elastic plates have been made during recent years, especially by Nadai, Mesnager, and Galerkin.^ "A theoretical investigation of the mechanics of stiff road slabs is in progress at present at the Bureau of Public Roads. So far, the assumption has been used that at each point the subgrade reaction is proportional to the deflection of the slab. Diagrams of deflections and bending moments have been obtained for various cases." ^ H. Hertz : tJber das Gleichgewiclit schwimmender elastischer Flatten, Wiedemann's Annalen der Physik uud Chemie, Vol. 22, 1884. pp. 449-455, or in the Complete Works, Vol. 1, 1895, pp. 288-294. H. Happel : tJber das Gleichge- wicht von elastisclien Flatten unter einer Einsellast, Mathematische Zeitschrift, Vol. 6, 1920, pp. 203-218. ^A. Niidai : Forschungsarbeiten auf dem Gebiete des Ingenieurwesens, 170- 171, 1915 ; Zeitschrift des Vereines deiitscher Ingenieure. 1914, p. 487 ; 1915, p. 169; 1922, p. S48; Schweizerische Banzeitung. Vol. 7G, 1920, No. 23; Der Bauingenieur, 1921. p. 11, p. 299; Zeitschrift fiir angewandte Mathematik und Mechanik, 1922, p. 1, p. 381; Fhysikalische Zeitschrift. 1922, p. 366. A. Mes- nager: Annales des Fonts et Chauss^es, Memoires. 1916, III, p. 313, and a series of short papers in Comptes Rendus, 1916-1919. Vols. 162-165, 168, and 169. B. G. Galerkin : a series of comprehensive papers in Russian, Annals of the Polytechnic Institute of Fetrograd, 1915-1910, Vestnik Engeneroff, 1915 and 1917, and Bulletin of the Russian Academy of Science, 1919. PR0CEEDIXG8 OF THIRD ANNUAL MEETING 95 PART II PROBLEMS FOR FURTHER RESEARCH It is suggested by your committee that research should be con- ducted on the following problems in order to obtain further infor- mation leading to the establishment of principles of design : 1. Study of elastic i^roperties of subgrade soils. Investigations should include observations on the flow of soils if they are only partly elastic. Research should also take into account that highway loads are moving loads of short duration many times repeated. 2. Tile transmission of |)ressnre l)y various thicknesses of mac- adaiii and gravel to subgrade soils. Such an investigation would call for an observation of pressures for a period of several years fol- lowing the original construction of the surface. Such an investiga- tion should cover a wide variety of methods of construction and materials on different types of subgrade. 3. The mathematical analysis of stresses in rigid slabs (now under way). 4. Exhaustive measurement of stresses in rigid slabs to confirm the mathematical analyses (now under way) . 0. Ascertain if the same or similar treatment of other heavy soils will make them impervious and good subgrade material as it did the heavy black adobe at the Pittsburg tests; also, determine if such treatment will retard or eliminate frost action. 6. Determine the comparison or relationship, in a pavement slab, of deflections and stresses under different loading conditions, with obserA'ations taken longitudinally versus transversely. 7. Comparison of vertical and horizontal movements of a concrete pavement due to moisture versus temperature. 8. Establishment of a basis of comparison of various asphaltic pavements versus Portland cement concrete. Conditions: (0) Sub- grade as nearly uniform in every respect as is possible to attain; (b) Pavement slabs of equal dimensions; (c) Identical atmospheric tem- perature and moisture conditions; (d) Identical loading and traffic, and (e) The same installation of recording apparatus to measure the direct comparison. 9. The effect of longitudinal reinforcing on transverse tempera- ture cracks in a Portland cement concrete slab. 10. The effect of an asphaltic top course on a Portland cement concrete base in changing the magnitude or rate of vertical motion due to temperature changes. 11. To determine the proper ratio of transverse steel to longitvi- dinal steel and the maximum amount of each necessary in a rein- forced concrete pavement slab to safely withstand unlimited repeated 96 ADVISORY BOARD ON HIGHWAY RESEARCH applications of, say, single axle loads of ten tons, without impact (this in conjunction, of course, with fair average subgrades), 12. To conduct further practical studies of the effects on pave- ments of various types of motor vehicles, including tires, springs and various accessories now on the market. 13. To evolve some simple means of predetermining the useful- ness of a finished subgrade made of the various soils, if it be properly prepared to get the best known results. (This is quite aside from the present method of determining bearing power.) 14. To confine our highway research, for the present at least, to pavement slabs between the limits of five and seven inches in thick- ness. The balance of the study can well be spent on the preparation of soils in securing proper subgrades for such pavements which would then safely carry any present-day traffic. 15. Need of dowels in two-section pavement. 16. The increment of strength added to pavements by thickened edges. 16a. Are thickened edges along center joints necessary? 17. The effect of crown subgrade on drainage. 18. To devise a method of surfacing sand-clay, top-soil and gravel roads similar to veneer surface as being carried on by North Carolina. 19. More definite information as to the average and maximum value of the load and impact force that have to be considered for each type of road. 20. The relative destructive effect of different types of vehicles (this with reference to types and weights of axles, chassis, and load distribution). 21. Importance of proper tire equipment. 22. A simple and definite method of testing tires to determine relative cushioning quality. 23. Relative stress effect of static and impact forces on materials. 24. Deformation and modulus of elasticity of materials under impact. 25. Investigation of merits of reinforcement and special joints in carrying loads across adjacent slab. 26. Study of methods of compaction of fills of various kinds of materials. 27. Distribution of pressure intensities under road slabs. 28. Further studies on effect of climate and snow on moisture in the subgrade. 29. Further study of beneficial effect of admixtures on plastic subgrade and the depth required. Note. — Subgrade treatment, 8 inches tliick. PROCEEDINGS OF THIRD ANNUAL MEETING 97 Dr. H. M. Westergaard, University of Illinois, gave an illustrated progress report of investigations on the ''Mechanics of Stiff Road Slabs." This report is not yet available for publication. A brief summary by Dr. Westergaard is given : This work was carried on during the summer of 1923 by the Bu- reau of Public Roads and is being continued at present at the Uni- versity of Illinois for the Bureau of Public Roads. The method of the investigation is that of the mathematical theory of elasticity. The nature of the problem is as stated in the note on theoretical analysis, embodied in the report of the Committee on Structural Design. The problems yet to be solved are the problem of local stresses in the immediate neighborhood of a wheel load, and the problem of the effect of a non-uniform subgrade. When the results of the investigations are plotted in the manner of test data, they should furnish a means for interpreting the results of tests. REPORT OF COMMITTEE NUMBER 3, ON CHARACTER AND USE OF ROAD MATERIALS In the absence of the chairman, Dr. Hatt gave an outline of Mr. Mattimore's report. The report follows: A meeting of this committee was held at Harrisburg on October 4 for the purpose of receiving and acting on reports pertaining to prob- lems which had been assigned to individual members of the com- mittee. Those present were B. A. Anderton, R. W. Crum, F. C. Lang, C. S. Reeve and H. S. Mattimore ; or an attendance of five out of a total membership of seven. The twelve problems assigned to individual members of the com- mittee for investigation were considered to be subjects upon which no definite conclusions have been drawn by research workers in the highway field. Their importance is recognized by highway research workers and it is considered that a solution of many of them would be a step forward in quality control of both concrete and bituminous surfacing, and may result in correcting many defects in both types of highway surfacing. These research problems are as follows : Problem Assigned to 1. Soundness Test for xlggregates in Concrete. M. 0. Withey. 2. Soundness Test for Concrete. 3. Effect of Grading of Fine and Coarse Aggre- R. W. Crum. gate in Concrete. 4. Determination of the Amount of Cement in Hardened Concrete. 98 ADVISORY HOAN!) ON HIGHWAY BESlJAh'CH 5. Significance of Modulus of Kupture Tests for H. H. Scofield. Concrete in Road Slabs. 6. Significance of the Jones Talbott Rattler as Tests for Concrete in Road Slabs. 7. Absorption of Concrete as Effected by Aggre- F. C. Lang. gate, the Ultimate Effect of this in Expan- sion of Road Slabs. 8. Significance of Ball Test (Penetration of M-- inch Steel Ball inider Pres^sure in Concrete and Mortar) . 9. Strength of Cast Concrete versus Drilled H. S. Mattimore. Specimens. 10. Strength of Hardened Concrete in Wet and Dry Condition. 11. Effect of Grading of Mineral Aggregates on B. A. Anderton. Sheet Asphalt and Bituminous Concrete Construction Relative to Deformation of Surface under Traffic. 12. Recovery of Bituminous Materials from C. S. Reeve. Road Surface Mixtures in Original Condi- tion. Problem iNo. 1, Soi ^dxess Test for Aggregates ix Concrete Problem No. 2, Soundness Test for Concrete So far as the committee can ascertain, very little research work has been done on these problems. The committee is of the opinion that accelerated tests, both on aggregates and concrete, endeavoring to duplicate atmospheric action, are one of the vital needs in this field. The Sodium Sulphate Test^ might prove to be valuable for this puipose, and it is recommended that research workers give con- sideration to it. Problem No. 3, Effect of Grading of Fine and Coarse Aggregate in Concrete The report submitted gives the present status of research on this subject. Several theories for proportioning concrete, devised by re- search workers, have been compared, and the following are some of the conclusions taken from the full report: Established facts and principles. — "As a result of the work of the various authorities, the following fundamental fact has been practi- * "Propo.sed Soundness Test f(n- Concrete Aggregate," United States Depart- ment of Agriculture, Rulletin No. 040. page 62. PROCEEDINGS OF THIRD ANNUAL MEETING 99 cally established. The strength of Portland cement concrete varies with the amount of actual solid material present in a given volume and with the relative parts of this volume that are cement and aggre- gate. Different experimenters have expressed this relation in differ- ent ways, but appear to be in substantial agreement as to the under- lying principle." "The fact having been established, that for the same amounts of cement and aggregate, other conditions remaining the same, the strength of concrete varies with the amount of mixing water, it readily follows that concrete should be mixed with as little water as will yield a workable mixture for the use at hand. It has also been reasonably well established, other conditions being the same, that the grading of the aggregate has a decided effect upon the amount of water that may be used to yield a workable mixture. We have here, therefore, an explanation of the benefit of using well-graded aggre- gates." Controversial factors. — ''The various suggested methods of evalu- ating the sieve analyses of aggregates have not yet been reconciled or the fundamental laws fully established. Concrete has been designed and made both experimentally and in actual construction, with suc- cess, by each of the various methods. Tlio 'Fineness Modulus,' ^ and the 'Surface area' - methods depend upon a complete sieve analysis. Prof. Talbot's mortar void method'' tlepends upon the characteristics of the mortar portion of the concrete. The Iowa Highway Conunis- sion* has used mixtures Ijased upon the ratio of fine to coarse aggre- gate. "It can be shown that with materials from a given source these various functions have definite relations with each other, but that with random materials froui different sources no such definite rela- tion obtains. It is probable that equivalent concrete mixtures can be designed by each method for aggregates generally similar, Ijut varying in grading." Recommendations. — "The committee recommends that research tending to coordinate the data now available be encouraged, and that research to develop methods of making more uniform concrete upon a large scale be strongly urged upon all interested agencies as pre- requisite to making eft'ective use of the information already avail- able on the design of concrete mixtures." ^ Bulletin No. 1, Structural Materials Research Laboratory, Lewis Institute. = Proceedings A. S. T. M., Vol. 18. 1918. page 230. = Proceedings A. S. T. M., Vol. 21. 1921. page 940. * Bulletin No. 60. Eng. Exp. Station, Iowa State College. 100 ADVISORY BOARD ON HIGHWAY RESEARCH Problem No. 4, Determination or the Amount or Cement in Hardened Concrete This report should be a valuable guide to research workers in this field, as it takes up the present status of the problem, with tabulation of the past work and the work at present under way. No report has been submitted on Problems Nos. 5 and 6, but a sub- committee is engaged in this work and expects to submit some data at the next meeting. Problem No. 5, Significance of Modulus of Rupture Test for Concrete in Road Slabs The committee considers Problem 5 a very important subject. There are considerable data at the present time on this subject, the importance of which for the testing of concrete road slabs is recog- nized. Problem No. 6, Significance of the Jones-Talbott Rattler as Tests for Concrete in Road Slabs The committee does not consider that data obtained at the present time indicate that this is a reliable test, and they question that it should be placed in the status of an important or essential test for concrete when used in road slabs. Problem No. 7, Absorption of Concrete as Affected by Aggre- gate: the Ultimate Effect of this in Expansion of Road Slabs The full report on this subject gives the status of the research to date. The committee considers that this is a vei*y important problem and that research along this line should be encouraged. An investi- gation of the absolution of concrete, combined with temperature changes to which it is exposed, may result in some solution of the cracking and rupturing of concrete bases and road surfaces. Problem No. 8, Significance of Ball Test (Penetration of %- inch Steel Ball Under Pressure in Concrete and Mortar) The research on this problem, so far as the committee is informed, has been limited — in fact, confined to one agency;^ but it is the opinion of the committee that if the test proves to be a reliable indi- cation of the compressive strength and effect of fatigue on concrete slabs, further development should be encouraged. The weak feature * "Ball Test Applied to Cement Mortar and Concrete." Bulletin No. 12, Engineering Experiment Station. Purdue TTniversity, Lafayette, Indiana. PROCEEDINGS OF THIRD ANNUAL MEETING 101 of this test in its present state seems to be that the penetration of the ball in a concrete mix is affected when coarse aggregate is encountered. Problem No. 9, Strength of Cast Concrete versus Drilled Specimens Considerable data on this subject have been secured and compiled. The data to date have not shown any direct relation between cast and drilled specimens, but it is the opinion of the committee that, while some such relation should exist, lack of control and absence of detailed data on the field concrete make it difficult to trace the relation. For guidance of research workers on this problem, a research outline is given ('^Strength of Concrete as Determined from Moulded and Drilled Specimens"). Problem No. 10, Strength of Hardened Concrete in Wet AND Dry Condition The general conclusion that could be drawn from this report is that concrete when tested in a wet condition shows a decrease in strength over that which is tested dry. The moisture content of concrete specimens when tested may be a large factor in the variables present in concrete testing; therefore, it is considered that this is a factor that should be taken into account by research workers in the concrete field. Problem No. 11, Effect of Grading of Mineral Aggregates on Sheet Asphalt and Bituminous Concrete Construction, Relative to Deformation of Surface under Traffic The report gives the present status of this problem. It is sui*pris- ing that so little conclusive research has been carried on in this field. All investigators emphasize the need of more information. The com- mittee considers this a very important field for research. Problem No. 12, Recovery of Bituminous Materials from Road Surface Mixtures in Original Condition The full report on this problem discusses several methods now in use for the recovery of bituminous materials from mixes. Conclu- sions are also drawn regarding the efficiency of the various methods, and a suggested line for further research is presented. The commit- tee considers this an extremely important subject, in that if we had an exact method for recovering bituminous materials in their original condition we might be in a position to determine some of the factors which cause the pushing and cracking of bituminous road surfaces. 102 ADVLSOllY BOARD OX HWHWAY REtvis mixes, the committee found that very little work of a broad research nature had been carried on. This field should \)Q a fruitful one for high- way research workers. The following concise statement is given in an endeavor to im})ress upon highway research workers just how im- portant this field is : "Research upon bituminous materials, particularly as used in ])av- ing mixtures, includes a number of problems upon which it would be desirable to have much more comprehensive information than is available at present. Discussions in recent years as to the causes of instabilit}' of Ijitnminous pavements em])hasize strongly the need for investigating the properties of mixtures of this nature. The prob- lem is difficult of attack on account of the great variety of materials, proportions, methods of preparation and construction, and because of the interdependence of many factors to be controlled and measured simultaneously. Such research requires special equipment, develop- ment of testing apparatus and methods, thorough investigation in a l)road field of possible comljinations of materials, and will probably necessitate considerable time and expenditure for the obtaining of comprehensive data. It is worthy of the most extensive and l)est thought which may be Iji'ought to bear on in\'estigatioH of road ma- terials." Research outlines, entitled "Outline Investigation of Stability of Bituminous Surface Mixes" and "Strength of Concrete Determined from Moulded and Drilled Specimens," are attached for the guidance of the workers in this field. Chairman Johnson gave the meeting over to a general discussion of the various reports and ])apers. jNIr. Older: Results olvtained by repeated soil-l)earing tests on the Bates Road subgradc may be of interest in coiniection with Dr. Westergaard's rej^ort. Loads of varjnng amounts were apidied to concrete slabs, cast in ])lace on the subgrade, ten to fifteen times at short intervals, followed by rest periods. The first load a|>plii'(l after a rest period invariably resulted in a marked j^ermanent dejjression of the subgrade soil. This permanent depression always seemed to appear, even although the rest period amounted to only a few minutes. Further, the repeating of a number of loads at frequent intervals seemed to result in continued permanent depressions, although apparently on a decreasing scale as the numl>er of repeti- tions of load increased. As might be expected, there is some indica- tion that as the moisture content of the soil increases ai)proximati>ly to the i)oint of saturation, the permanent depression caused l)y each application ol' load increases (|uite rapidly. PROCEEDINGS OF THIRD ANNUAL MEETING 103 3 a h ^ a o 04 u & o o w. 2 .5 o I! -:w ,(10 • o I S 104 ADVISORY BOARD ON HIGHWAY RESEARCH ft 0- a ♦> S " is o tf o *> -3 X a (3 ■»» O ts o •pis ::!?:2 ^»-3:: a- s c ^ p. ^ f" J! ^ " ^ fi P.-1 « 4J (3 ►j O O •^ ► a a ■ o ■ IS o o • «l • P- « « -; ■ h g a r-t >< p.« x> o O O (/I «< < tr> o PROCEEDINGS OF THIRD' ANNUAL MEETING IQS The observations so far made indicate that at least one other item may have an extremely important bearing upon the behavior of a clay subgrade soil when subjected to repeated loads. This relates to the effect produced by freezing and thawing of the soil. During the summer and fall months, when the moisture content was about 28 per cent, ten repetitions of a load amounting to 6 pounds per square inch might result in a total permanent depression of perhaps not more than 1-100 inch. Later, when the moisture content had in- creased under the same slah to about 33 per cent, in the meantime the ground having frozen and thawed a few times, ten applications of a load of the same magnitude produced a permanent depression of 16-100 inch. Still later, after another period of freezing and thawing, during which the moisture content had again decreased to about 28 per cent, ten applications of the same load produced a permanent depression of about 6-1000 inch. It would seem that north of the latitude where freezing and thawing of the subgrade may be expected this phenomenon would have an important bearing upon the subgrade problem. Dr. Westergaard : It is possible that ruts can form under the pave- ment, provided the traffic follows definite tracks. I believe, then, that one of the particular problems that should be dealt with theoreti- cally is the matter of the flexure of the slab having less support along two different lines. I believe that problem can be handled. Now, as to the matter of permanent sets in general, a committee on stresses in railroad tracks has been confronted with that same thing. Second Session The meeting was called to order at 7 :30 P. M. The Chairman in- troduced Thomas H. MacDonald, chief of the U. S. Bureau of Pub- lic Roads, who made an address on THE OBJECTIVES IN HIGHWAY RESEARCH The United States is now carrying on a highway improvement program which, measured in terms of expenditures, approximates a billion dollars annually. As in nearly everything else of concern to the general public, we have gone in for quantity production. A large yearly mileage of new roadways has been demanded, that a place might be found upon which to operate the quantity production of the motor vehicle. So large are the annual progi*ams of new con- struction in many states that even minor modifications in specifica- tions or designs have a very large financial aspect. In the face of large annual increases in the mileage of improved 106 ADMSORY BOARD ON HIGHWAY REtrei)arati()n and ex))erience re- quired to fit one adequately to practice the profession of engineer- ing, the hard work, both mental and physical, usually involved in the practice of the several branches of engineering, the sincerity and persistency of ])ur]x>se required for any successful accom]dishnient. PROCEEDINGS OF THIRD ANNUAL 2IEETING 107 and the responsibilities of all kinds imposed, why has not the en- gineer been accorded more readily and more consistently leadership and the rewards of leadership, not necessarily material, which gen- erally accrue to recognized authority? And the answer, which is reasonably suflficient to me, but which I do not insist that any one else shall accept, is this: the lack of engineering research. As we review engineering history, this lack and often entire absence of anything approaching adequate research ha8 been until very recent years characteristic of both individuals and institutions. For example, agricultural experiment stations have been estab- lished through federal and state legislation and have been supported for many years to promote the development of the science of agri- culture. These experiment stations are established in every state, but only a very few educational institutions have established en- gineering experiment stations, and most of these have had serious trouble to maintain a meager existence. Why research in engineer- ing has not been more fully developed is most difficult to understand. Undoubtedly our great natural resources have lulled us into the be- lief that it was not necessary to use them with the economy which ought to follow proper research ; yet this certainly cannot be the only reason, when we consider the tremendous fertility of the lands which have been ours for the taking; and yet we find a widespread and persistent extension of agricultural research in the states where there has been no engineering research developed. I am fully convinced that engineers have been so busy with the day's work that they have not insisted upon the development of an adequate engineering research until the fact was accomplished, and that this reason is, to a greater degree than anj^ other, responsible for the failure of engineers up to this time to secure that degree of recognition justified. Very hopeful indications now exist that this recognition is coming slowly but surely; and without doul)t the motivating reason behind this changing attitude of the public toward the engineer is the greater interest, activitj^ and insistence of the engineers in the establishment and conduct of engineering research. You will find in this statement no striving for eft\^ct, no unfriendly criticism, but rather an expression of the deepest concern that the great needs and great opportunities for research that are presented shall receive full recognition from the engineer. And what is true of the necessities and fruitful potentialities in the field of engineer- ing research is equally true in that of other professions; for ex- ample, the economist, who is concerned with the economic and social development of tliis nation. 108 ADVISORY BOARD ON HIGHWAY RESEARCH Consider the hi,2,h\vay transport field. Tlie civil, the mechanical, the automotive, the transportation engineer, the economist, and the chemist — all have presented to them broad opportunities for direct research, while, although more indirectly and perhaps somewhat more removed in time, there are other professions that may veiy conceivably find worthy phases for research endeavor. Highway research offers two major objectives: First, the isola- tion and definition of each problem; and, second, the solution of the problem. These objectives are supplemental and sequential. All too frequently the first is neglected to the undoing of the second. Perhaps I may even be justified in saying that engineers have fre- quently in the past made grave errors, and, judging the future by the past, will continue to make grave errors in big enterprises or in the solution of big problems in one of the following ways: a. Insufficient and inaccurate fundamental data. b. Limitations in scope of data secured. c. Limitations imposed by legislation or other authority. d. The misapplication or maladjustment of data in itself correct. Examples might be multiplied of failure or lack of obtainable success in engineering projects through an incorrect or inadequate analysis of the whole problem involved. More specifically, failures might be catalogued under one of these four reasons. It might be urged that engineers are not responsible for the limitations imposed by legislation or other authority ; but are they not responsible when, as is frequently true, they cut and trim estimates and designs below safe practice to meet these limitations? More than this, it matters not what the reasons are or what handicaps exist, the engineer in charge will be held in the court of public opinion for any failures. There are these two groups of highway research objectives: First, those having for their purpose the correct analysis of each problem, and, second, those having for their purpose the finding of the remedy or the solution when the problem shall have been correctly isolated and analyzed. Just at this time, of tlie two, the first is the more important. Real progress is always slow. Consider the years of research required to isolate the yellow-fever germ, and in the mean- time the lives that were lost from this cause, unknown and unchecked. The cause once known, the remedy has followed swiftly — to the everlasting credit of the medical profession. So, too, the oppor- tunity is offered in the highway field to do a big work now in re- search. Much progress has been made, but we have only started. Nor can the research worker stop when he shall have reached both of these objectives, for he must then insist upon the practical appli- PROCEEDINGS OF THIRD' ANNUAL MEETING 109 cation of the solutions he finds in the field of highway transport. The most perfect solution is only valuable when applied. After all, the real objective of highway research is the most economical, in every sense of the word, and the most efficient highway transport sendee possible for the use of the public as a w^hole. The oppor- tunities are unlimited. There is not a single phase in which there is not the necessity for major research. In the field of finance, no rational policies of providing the funds have been worked out. We have only fairly started upon the improvements in the design of road surfaces. We need to develop better materials — cement, as- phalt, aggregates, and particularly sands. Field control of processes to insure uniformity of product is still almost untouched. Adequate provision for, and the control and regulation of, traffic must be solved. These unsolved problems, and many more, present a wide, almost unexplored, field for highway research. Anson Marston, Dean of Engineering, Iowa State College, was unable to be present. Dr. Hatt read Dean Marston's paper on HIGHWAY RESEARCH WORK OF AMERICAN ASSOCIA- TION OF LAND-GRANT COLLEGES In another report the writer of this paper is giving a summary ac- count of engineering research at the land-giant colleges of the United States. In a word, they constitute the greatest single organized agency for engineering research in the country. The status of the land-grant colleges is especially favorable for organized highway research conforming to a national program, such as that of the Advisory Board on Highway Research. a. The land-grant colleges are the only nationally endowed and supported educational institutionp of higher learning in the country. Hence, each of these land-grant colleges should cooperate with the National Government in its highway research (conducted by the Bureau of Public Roads) and with the Advisory Board on Highway Re- search, itself (through the National Research Council) a creation of the National Government. h. Each of these land-grant colleges also is officially established and supported by the government of the state in which it is situated. Hence, it is especiidly fitting that the land- grant college in each state should cooperate officially and systematically Avith the State Highway Commission in the same state, since they constitute two branches of the state government. 110 AD] ISORY BOARD ON HIGHWAY RESEARCH Furtlieriiiorc, tlic l;',ii(l-,L!;raiit colleges uk- by organic law devoted to education an 1 icsc irch in the industries, including especially agriculture and uic.liaMic arts. No line of research, therefore, could be more fittingly dc\(Io]i;'l at each land-grant college than highway research. The jilanning, the construction and the maintenance of good roads, tbcir use and the economics of highway transpor- tation have the most vital relation to agriculture and to every branch of meclianic arts; in fact, to every line of work for whicli the land-grant colleges we^'- established. Under these conditions it is not surprising that the land-grant colleges, without any |)articular organized effort, already have en- gaged extensively in highway research. It appears that in 29 of the 4()pulated. rich agricultural county, with no highway bond issues. Madison. th(> ca])ital city of the state, is located in the center of this county. Outogaiuie Conufji is a good agricullural county. Tt has financed a large portion of the con^lructiou of its highway system through bond issues since li'H). Rusk County is located in the newer section of the state. It is a purely agricultural county, with soil of average fertility, but as yet largely undeveloped. No highway bond is.sues have been made and its highways are, with the exception of the state trunk highways, largely unini])i-o\-ed. PB0CEEDING8 OF THIR,D ANNUAL MEETING 117 Waukesha Count ij lies west of the city of Milwaukee and is a rich dairying and manufacturing county. It has floated large bond issues during the past three years. To determine the total expenditures for higliAvays in each of the counties in which the analysis was made, it was necessary to con- solidate the expenditures made by the local units within the county, the expenditures made by the county, and the expenditures of state and federal funds within the county. The expenditures by the different governmental units have been carefully analyzed and all expenditures appearing in the records of more than one unit have been eliminated in all except the unit which actually raised the funds for highway construction or maintenance. In the local units the expenditures made by cities have been eliminated, as their expendi- tures are almost exclusively for streets rather than for highways. The following tables show the total expenditures for highways w^ithin each county — Tables 1(a) to l{d) . From these tables it is evident that the proportion of the burden of all highway expendi- tures which rests upon real property varies from 55 per cent to 70 per cent in the four counties. The highest proportion is found in Rusk County, a relatively undeveloped county, while the lowest pro- portion is found in AVaukesha County, a very rich section, both agriculturally and industrially. This variation is due largely to differences in revenue derived from the income tax. In the richer counties the income tax furnishes a larger portion of the total rev- enue, and, as a result, in so far as funds for highway expenditures are taken from general revenues, the income tax furnishes a larger por- tion of revenue for highway expenditures. It will be noticed that real property taxation and "other revenue" taken together furnish almost the same proportion of revenue for highway expenditures in all counties, the figures being as follows: Dane County, 76.82 per cent. Outagamie County, 79.86 per cent. Rusk County, 81.35 per cent. Waukesha County, 78.68 per cent. A comparison of the proportions furnished by each source of rev- enue in any one county for the different years substantiates the con- clusion that the contribution of real property varies inversely as the contribution of "other revenue." "Other Revenue," as the name indicates, is made up of all sources of revenue except the sources listed separately in Table I, but the principal part of it is made up of revenue derived from the income tax. The other sources of revenue 1^ &3 05 ^ "^ o o s o o QJ 'T3 o o t^ »o o • O t^ GO rti 05 • in O CO ^ • m T— ( >> O Oi ^ CO •* "c O O^'O'M (M OP :3 O X! "^ ~ o OCO ^r-l o Q »-H u 1 c^ f^ , OCOCOt^ 1— 1 cj OOOTt< -^ o o O"^ C^5 O) Tt^' h-1 Ot-.-H OOiOCOOOlMOO "^ O (M iM »0 O 00 O "o O O --H CO o ^ H O CO 1— • 1— 1 .— 1 lO »o ^ CO CO c3 fci ■M (M OJ 't' rji TS CO CO fe CO CO CO CO s^ • • COC^ 00O5l> l>C0 05(M 00 oO'* 'f "oco 00>OiM -:*< CO CO lO lO CO 05 (M O 00 0. lOC^ 'M CO CO m 3 (M moi o ^ CO r-< C2 00 ^„ 3 o C^l CO'l^'t-- CO o o 'ti COCi t-- a M< 05 rt C^l o c3 C^l lO t^l^ C<) o ^ lOOOt^ •o^ O I-:) 1,814 1,343 218 171 05 m CO > ss§ 03 -f ir; ^H C (B O O "5; — S5 oj o b 1^ *- 03 J2-G ti S i£ o aj S +^ :3 cLun e- ff :;a ^ >> COl^'* "r^ O OO'-H o ^ 3 O -H t^ ^ CJ o C t^l-H ^ ^ o Ah , or^c^i CO X tS O lO iQ lO CO o o O --0 r^ O I— 1 H^l O --O ^(M Ol>^ O5C0 00 "3 O C o3 ^ 00 ^ CO o3 I> CO »0 lO 05 ^cocoio-H 00(MO»00 m ooio ^ Tt< 00 CO coco --1 00 CO rH ,-1 s© t^COl^ tH lo au> 00 >. CO CO (^00 a; "tH 00 Tjl GC* C 3 Oic^ coco =! O Tt* LO r-H 00 O o (M 05 coco a •O 0(M i-H < .— 1 .—1 CO-N CO ■* 'J 00 05 QCiO •^ ^H CO CO (M "3 1^05 00 00 a 1>- "^ "^ 05 00 O 1^ ^ COCO oi Tl^" t^ — i(M —< O t^-H^J ,_! ^ C^ CO lO o o ^ ooco -* coco IC 1 ^_^ t^ lO 1> C^ 02 Co a: 00 ■* t^Tf< rt O X ^ O^Ol'f lO^ CO —• ^ ! • c .'.2 ! '-3 ^ 03 OQ ••p 0) ■"^ T o;^ X >. 5S c s^a 1^ cc . -p o •=;' ' cc . ri t^ C Oj 0) S^ S3 « 03'— 5S > O00G0-:t< • OO'* 00 ■ ;3 o .-1 CO ■* • O O OJ'-i !-i o 1—1 i^ t^ CO (D coo O i-H o "S oocoicco-^ !>■ CO lO 1> CO m iM 00»O-*CO OJCO rH O T-l I— 1 CO t^rH 00 0:>fM OCD >> CO >ocooo 02 G s iO-^00(M 0o ^ •a> o -. -. .^ >. ^ h^ T^ CO 1— I lO • 1— t O Oi CO '^ CC* m Tt<^i> ooco>o CO 00 ■* CO CO o OCOO --H ooo 13 CO 1-H CO CO -*i 00 o 03 CO "* >0 CO 03 ■* CO 00 ■* CO ^^ lOt^CO '^lO (M 00 --H --< >-( 1— 1 _^ 1 -.2 ri cS cc •pH CO 2 c a 1 o3 -g a s'j t^ o B 02 ri in C q; 1> --1 • oS O 000003 o • ^ OCO lO • >> ooooco ■ ■4^ O 00 05 05 • ;3 3 OCO CO CO • o 01> --1 --1 • Ph O I— 1 1 o 00 c» CO t- o3 O --1 "OCO lO CJ O CD-^OtH 1-1 O CO rH CO I— 1 ocnoicsoco Is O -03 ^ CO 05 CO -M o O lOOCOO H O lO i-( CO 1— c 1—1 CO C !>• cS ^ I>^ t^ OJ iC lO -« 00 00 Ci m O 1> CO ^ o -*OTt<^l:^ »OC0 CO m- OOOCOCO CO coco CO h -TtHcocJo 02 lO 02 ^ "O " coi-i T-^o 3 o Tft lOCO l^ o o COCOO C75 a < t-iOi-i ooooco^ •o COCOi-H-^ :^ 00 c^ GOO • t^ 'S lO^ CO '^l^ • t^ u 005^Tt< •lO o I-:; coo TfH CO •»o oocoo^ OCO 1-1 CO €# O I^~ O CO 1-1 O »OC0 OO^'O'^ 1-1 Tt< »o --( C^ 1> "^ CO CO CO CO CO t^ ■+^ 00 O CO r-H O UO o »,•.».*.*.*. H tJ< CO CO rH O kO QOt-iocot^ ■* coco >oco CO 1-1 ®& ;.o '. '-(3 fl ^ .2$^ a; a; 02 <^ .. oj 42 X >i 23 G o3 -e CI S t.yt oper ue. lice ssmi iH tn K 0) OJ cc ] prope 3onal p er reve omobil cial ass -t^ 03 j: ^ -j^ Qj O a) (D -u 3 CL 1 r- P '^^C bleni is not neces- sarily a problem of research alone. l)ut also a ])roblem of ai>])lieation. Probably there is nothing new in the fact that North Carolina has a research program, because eveiy state, county or nuuiicipality that is carrjang on the construction of liighways has, to some degree, at least, a research program. In addition to this, laboratories scattered throughout the United States are also carrying on various researches. A survey by the Highway Research Connnittee disclosed the fact that many hundreds of projects are in o])eration. Some of them, of course, were duplicates, Ijut the fact was disclosed that a great amount of research is being carried on. The difficulties of a research program are numerous, but ]>robably no more important problem exists than that of keeping research within practical limits and carrying it on in such a manner that when completed it may have a practical api)lication and serve as a PROCEEDINGS OF THIRD ANNUAL MEETING 125 step in working out economically some of the highway problems. Already a vast amount of research knowledge has been acquired, and this information is written into bulletins and distributed; but, un- fortunately, many times the bulletins are of little practical value, because they do not reach the proper persons, nor are they written in a manner that will serve the busy official who generally organ- izes programs. Our research problems are still unsolved because the important step^ — the application of highway research work — has not been accomplished. The problem of research seems to divide itself into two minor problems — research of the problem itself and the application of the research results. Tlie application of the results obtained is such an important step that it could almost be considered a research problem in itself, and a proper study of this application would be well worth while. We find that a great amount of research has been caiTied on, but as yet a vast amount of the knowledge gained has not been applied to road-building, and consequently the economical value of the research has been lost. In North Carolina every attempt has been made immediately to apply the results of research. Almost every step in construction is being studied in detail, and when anything of value is discovered it is immediately applied to construction on a large scale. Not only does research include problems that may be carried on by the laboratory, but it includes transportation problems, problems that will ultimately render road service to the user of the highway. The great research problem is to render road service to the public so that transportation may be more economically carried on. These major problems of- transportation and road service are, of course, influenced by other problems, such as the problems of construction, maintenance and operation. In all highway research there should be constantly kept in mind the question of how the results of this research can influence the road service rendered to the user of the highway. Sand asphalt pavement. — One research that has recently been carried on has been the development of progressive type roads. Cer- tain localities in North Carolina furnish practically no road-building material, with the exception of sand. The problem in this instance was to devise some means whereby these large quantities of sand could be used in road construction. The answer came in the construction of the sand asphalt road, which is made up of approximately 88 per cent sand and 12 per cent asphalt. These roads may be con- structed in either single or double track, according to the needs of the traffic, and are generally constructed 4^ inches in depth. This method of construction aft'ords a moderatelv low cost road, which 126 ADVISORY BOARD ON HIGHWAY RESEARCH renders very satisfactory road service to the locality. It is not ex- pected that this type of road will stand up under a large amount of heavy truck traffic, but it is a development road and is satisfactorily rendering road service to the user of the highway. Though this type of road is, of course, the result of laboratory research, it is more par- ticularly the result of the proper application of laboratory research. 3Iarl rock ha-^e. — Another locality in North Carolina furnishes no stone for road aggregate, but it does have an underlying strata of mai'l rock, the result of a large deposit of shells. Although at some time this deposit must have been on the seashore, at the present time it lies some 8 or 10 miles back from the ocean and about 4 or 5 feet underneath a swamp. This marl rock is being quarried and crushed and is furnishing an excellent base for a sand asphalt surface. While the details of this construction were taken from earlier road work, still the use of this marl rock means the practical application of in- vestigative research. Subgrades. — The studies of subgrades have been demanding con- siderable attention during the past few years, but there seems, even at this time, to be a lack of proper application of the findings. It has already been determined that different soils are affected differently by the varying amounts of moisture. In some cases, however, no at- tempt is made to select the best soils for subgrade purposes. In many localities a proper selection of soils will not only furnish an ideal subgrade, but M'ill serve as a road, a subgrade highway, to take traffic up to as high as 400 vehicles a day. These selected soil roads may be maintained for a period and then used as an excellent sub- grade for the next better step in construction. Capilhrity. — Recent research has brought to light considerable information regarding the capillarity of different soils. For a long period it seemed as though a poor subgrade or a clay condition might be best remedied by the construction of a Telford foundation. Now it appears that the Telford foundation served as a drain to carry away the free water, but was of little, if any, value in taking care of the capillary moisture. Recent experiments have shown that a layer of material similar to sand is more effective as a means of cutting off capillary moisture. This explains why excellent results have been obtained in the construction of macadam roads on clay sub- grades when screenings have been spread on the subgrade. It also explains the excellent results obtained with bituminous roads con- structed on a gi'avel and sand foundation. Research has shown that capillary moisture in the subgrade is not cut off from the overlying road surface as efficiently by coarse ma- terial as it is by ordinary sand. The reason probably is that the PROCEEDINGS OF THIRD ANNUAL MEETING 127 subgrade material findn its way into the interstices of the coarse ma- terial and the capillary tubes of the subgrade material continue to act through the interstices of this coarse material; whereas, in the case of the sandy material, the subgrade material of high capil- larity is excluded, and therefore the capillary action of the sand is not so great. This is an important point and one iliat can be taken advantage of and easily and economically ap])]ied to road work on a large scale much more so than is commonly done. Veneer surface. — Anotlur important ai)plicati(in of research work is in the construction of stone veneer on earth roads. In the south there is an exceptionally large mileage of earth roads, consist- ing mainly of the sand clay, topsoil and gravel ty])es. The bear- ing power of these roads is higli and the strength is great; yet main- taining these roads is a proljlem, since they offer little resistance to abrasion. Thus the problem in this particular instance is to secure some means of ]:)rotecting the surface against abrasion and to abate the resulting dust nuisance. The answer seems to be the stone veneer surface in which the quality of an asphalt wearing surface is com- bined with the strength of an earth road. Several attemi)ts have been made in applying liituminous material to earth roads. The results, in general, have been unsuccessful because, even though the bitumin- ous material was sufficiently light to penetrate the surface, it had no binding power or strength Avhatever, and if the bituminous material was sufficiently heavy to have a binding value, it would congeal on the surface and peel off, owing to a dust mat that formed underneath the bituminous material. With the veneer surface, a layer of stone of approximately three inclies in size is applied to the earth road and rolled partly into the surface, after the surface has been scari- fied or loosened. This veneer surface of stone is then penetrated with a bituminous material having a consistency to give sufficient strength to hold the stone in place and afford resistance to wear. This bi- tuminous material is then covered with proper sized stone, is rolled, and then opened to traffic. The stone is held in place from below by the eaith road and has become an integTal part of it. The bi- tuminous material holds the stone in place on the surface and affords resistance to traffic abrasion. Thus we have a combination of the strength of an earth road and the wearing (jualities of an asphalt pavement — another instance of the practical a]iplication of research work. Hard surface types — coarse aggregate. — Another problem in con- struction is the ideal construction of hard-surfaced roads. Much laboratory research has determined that the aggregates must be of certain qualities and ])r()])C'rly graded. Recent experiments in con- X28 ADThSORY BOARD ON HIGHWAY RESEARCH Crete work have .shown that much depends upon the grading of the aggregate. Although the early tests indicate that a large aggregate gives a higher strengtl? value, long-time tests seem to show that the aggregates below an inch and a half give higher strength values. In this instance we have a research giving one result, yet it is impossible to apply this result in its entirety, since the present arrangement of crushing plants and aggregate production machinery does not furnish economically an aggregate that gives the highest ultimate strength. Therefore the proper application of this research demands a com- promise approaching the ideal conditions. Core drill. — Considerable research is being carried on attempting to correlate the laboratory tests of materials to the final product as found in the pavement by the core drill. At the present time it seems there are so many variable factors that it is impossible to control all of them, and, as a consequence, the final product varies considerably. Not only have studies of the coai-se aggregate Ijeen made, but a new test for fine aggregate has been developed, which seems to permit a wider range of fine aggregate to be used, or, in other words, aggre- gate that would be condemned under the standard test can now be used with safety. This test consists essentially of testing the aggre- gate in compression and transversely rather than in tension, as is done in many of the present-day standards. Investigations in cement also show that this material varies con- siderably, and that the final strength of the road depends primarily upon the quality of the cement used. It seems that individual brands may vary considerabl}-. This may not be only a question of manu- facture, but also one of handling and storage until the cement is used. Surface finish. — Possibly no single demand on a pavement is so great as that of impact of traffic. Impact dcjiends entirely upon the smoothness of the road surface ; therefore considerable study has been given to obtaining smooth surfaces. This has been put into practice by devising various means of finishing the surface. In the case of concrete, it is found that the smoothness of the surface depends upon many factors, chief among them being the consistency of the concrete and the character of the subgrade. The subgrade, after being wet by the concrete, expands or contracts and the green concrete in the road surface does the same. Various types of check templates are being used to check the surface before it sets up, so that it may be corrected while still plastic. In the case of the bituminous roads, test-boards and straight-edges are used during con- struction. After the road has been opened to traffic it is again PROCEEDINGS OF THIRD AXXUAL MEETING 129 tested. Recently, experiments have been carried on with the vialog, an instrument devised for measuring unevenness of road surfaces. There has been much discussion relative to the detailed construc- tion and maintenance research problems, but probably the most press- ing problem and the most immediate need is a solution of the proper methods for taking the research work that has already been done and giving it practical application. This leads us to the problem of organization and personnel ; the varying results obtained in the work show the influence that personnel has on any undertaking. Re- search in construction and maintenance problems deals with definite quantities, but research in organization and personnel, since it deals with varying factors of the human mind, is more difficult so solve. The proper and economical application of research Avork still in- volves research in the matters of personnel and organization. It is useless to spend time, money and effort in carrying on research unless the results are applied in such a way that our highway prob- lems will be more economically solved than they otherwise would be. In order that the facts may be used to the greatest advantage, they must be presented to the proper persons in a way that is definite, concise and easily understandable. Third S ess-ion The meeting was called to order at 9:30, Friday morning. Chairman Johnson: The Board has received a report of coopera- tive work being carried on in Iowa on "Impact in Highway Bridges," by Prof. Fuller, Chairman of the Subcommittee of the American So- ciety of Civil Engineers on Impact in Highway Bridges. His re- port is upon the cooperative program between the U. S. Bureau of Public Roads, Iowa State College, and the Engineering Experiment Station of Iowa State College. Dr. Hatt will read this report. Dr. Hatt: Mr. Fuller's report is as follows: Cooperative work was begun in the summer of 1922. An im- portant item at that time was the adaptation of existing instruments or the development of new instruments for measuring dynamic stresses. A number of instruments were tried out, with the result that only those best adapted to the purpose were used in 1923. A description of the 1922 work, discussion of the various instru- ments used, and a portion of the results have been published as Bulletin No. 63 of the Engineering Experiment Station of Iowa State College. The greater portion of this bulletin was included in the 1922 report of the Committee on Impact in Highway Bridges of the American Society of Civil Engineers and published in its Proceed- ings for March. 1923. 180 Ainisoh'Y i{()\h-i) <)\ iiKiuwAY Ni:si:.\h'CH Tile general s(m)i( n^i tlic work for llic suiiiincr of \'i)'l-\ \< iiulicatt'd in the followiiit;: Sfr}(rfu !'('>< ill !•( Migafcd : 1. ()ii('-]iiiii(li-('(l-an(]-Hfly-foot sj)an, 2(l-foo1 loadwav. tlirou,i;li curved cliord .-tccl trus?;. with concrt'te floor on stet'l ,4rin_i;ers, known as the Skunk Kivcr Hrid,iL»c\ on the Lineohi ni,i>hway, located about tlu'ee-fourths of a mile east of Ames. 2. Thirty-tlu'ee-foot span, 20-foot roadway, consisting of concrete floor on steel stringers. This is an approach span to the main Skunk River Bridge. 3. Seventy-foot span, 24-foot roadway, through plate girder, Avith concrete floor on steel stringers. This is known as the Squaw Creek Bridge, on the Lincoln Highway, and is located within the city- limits of Ames. 4. Forty-foot Pony truss bridge, 18-foot roadway, with reinforced concrete slab resting directly upon transverse floor beams, located on county road near Koland. Iowa, about twenty miles from Ames. Loads: 1. Two 15-ton trucks with solid rubber tires, with about twelve tons on the rear axles and three tons on the forward axles. 2. A 10-ton Holt caterpillar tractor (on structures 1 and 2 only and only for about two hours) . Range of work: Floor. — Rather complete work was done on the stringers of struc- tures 1, 2 and 3 and on the floor beams of 1, 3 and 4. This includes impact on smooth floor, over one-inch obstructions and over two-inch obstructions, and also distribution data for static loads which show the actual unit stresses on all of the stringers for one and for two trucks. Trusses. — Readings have been taken upon the greater portions of the web members of structures one and four, of the chords in struc- ture four, and of the girder flanges in structure three. Time. — Field-work was started on July 1 and ended September 15. Insfriniients: 1. Three Turneaure recording extensometers. 2. Five West direct reading extensometers. 3. One ])h()t()gra])hi(' miri'or recording extensometer developed by the U. S. Bureau of Public Roads: available after Auuust 1. PROCEEDINGS OF THIRD ANNUAL MEETING 131 4. Six electrical remote reading and recording strain gages, in which records of all six instruments are photographed on one roll of paper — developed by U. S. Bureau of Standards and made available for one week in September through courtesy of that Bureau. Calibration, of instniments: Before the field-work was begun the instruments were calibrated for static stresses. No suitable equipment was then available for dynamic calibrations. Two more or less impromptu devices have since been developed which have given, within reasonable limits, positive as well as comparative calibrations under rapidly changing conditions. These have been, first, an impact machine in which a weight was dropped upon a tension bar to which the instruments were attached ; and, second, a vibrating device which produced known changes in length in very short but known increments of time. The work of calibration, while yet not complete, has been carried far enough to insure a reasonably accurate interpretation of results, which will be comi)iled as soon as practicable. Chairman Geo. E. Hamlin presented the RP^PORT OF COMMITTEE NUMBER 4, ON HIGHWAY TRAFFIC ANALYSIS During the year many states have carried on, to a greater or less extent, highway traJSic censuses. Attached to this rejDort is an analysis of the characteristics of such surveys, Avith data obtained. The detailing of this data shows that the various states realize the im]3ortance and value of traffic information and are each year in- creasing the census records to specify additional information of movement, conunodity and length of haul. In this connection, the chairman wishes to call particular atten- tion to Prof. Blanchard's recommendation, that this committee strongly differentiate between a highway traffic census and a high- way transport survey. The traffic census will give information per-, taining to the traffic using the highway at the time the census is taken ; the purpose of the highway transport survey is to determine the probable amount and character of the future traffic which will use a given highway during the lives of its several component parts. Up to the present time, the information collected by the various states has given traffic census information, Ijut as far as the chairman has been able to determine, no state has extensively taken up the highway transport survey. 132 ADVISORY BOARD ON HIGHWAY RESEARCH A highway traffic census is of value only for determining condi- tions which exist at the time the census is taken. It may be that by the construction of a new section of road the general trend of traffic may be radically changed in any particular locality. After an ex- tended highway traffic census, it is recommended that additional counts be taken at critical periods of the year in succeeding years, from which, after a number of counts have been taken, curves of natural increase can be plotted from which in turn an estimate of increased traffic for a reasonable period of years can be roughly de- termined. This estimate should also include the curve of increase of motor vehicle registration which can be determined at the present time in every state. It is questionable if such a curve can be applied to other than the locality in which it is developed. In fact, even in a state of small area, different curves will have to be developed for different sections of the state, rather than to utilize a general curve for all of the main highways. Up to the present time, it has been found impracticable, with the data available, to develop a formula of this character. The value of an extended traffic survey in determining the alloca- tion of construction and maintenance funds in the development of a highway system is unquestioned. Such traffic survey, however, should be utilized for type and strength of surface rather than for location of expenditures, inasmuch as the development of new terri- tory within a state is as much a demand upon the expenditure of highway funds as is the taking care of traffic already developed. This point cannot be stressed too strongly, for, if the allocation of funds depends wholly upon the volume of traffic, only a few of the roads in any particular state would ever receive a construction allot- ment. This is where the decision of the engineer in charge is valu- able, and this decision must be based on potential as well as present traffic conditions. In other words, he must minimize any particular locality to consider the value of transportation in the whole state, and each construction unit must be based on his vision of the value of such unit. To carry out a traffic survey successfully, much planning must be done preliminary to the actual field-work. Stations must be chosen which will give the average condition along each highway, and these must be located at a sufficient distance from the congested centers to eliminate as far as possible the strictly local traffic which will not enter into the construction program. Care should also be taken to establish stations where traffic is divided, so that the value of each section as well as each road may be determined. Each station should be occupied at least one day each month, for an eight to twelve-hour PROCEEDINGS OF THIRD ANNUAL MEETING 133 period. This should be arranged so that tlie same station will be occupied successively on different days of the Aveek, and on a differ- ent hourly basis, to determine daily as well as seasonal variation of traffic. Blank forms where check-marks may be used will reduce the time required for gaining information and names of cities ; char- acter of loading and make of truck or car may be coded, thus sim- plifying the actual compilation of the records in the office. The personnel require a certain amount of special training, al- though courtesy has been found to be the largest asset. A special sign at the census station giving the reason for the questioning is the greatest factor in reducing the time required to pass a car or truck through the station. The enumerator can check many of the ques- tions on the card by observation — the number of questions asked the driver in the case of the pleasure car covering only origin, destina- tion and business or pleasure use. In the truck census other neces- sary information will require a longer period and larger party, where weights are actually taken on a platform scale installed for this pur- pose. A platform pit type scale of 25-ton capacity can be installed, with automatic dial and the necessary covering house, for $2,500 to $2,800. The chief of survey should be thoroughly acquainted with the entire district, both in regard to road and traffic conditions, and be able to detect any seeming discrepancy in the records sent in. Ab- normal conditions due to a particular event, such as a football game, must be previously discounted, so as not to congest traffic on that day, even if the stopping of the cars for questioning is omitted. Each party may consist of from three to six men, the number vary- ing with the importance of the station occupied, each party having a designated head, who reports to the chief of survey and under whose direction the party operates. In the State of Connecticut the cost of three parties operating over a year's period with varying numbers, including necessary automo- bile transportation, board allowance, etc., has been about $27,000. It will not be a function of this report to determine results ob- tained from the Connecticut traffic census, which are being compiled and tabulated under the direction of J. G. McKay in the U. S. Bu- reau of Public Roads. The survey was carried on for a full year, ending in September, 1923, with 56 stations and operated by three parties. The tabulation is still in progress and the full report is being prepared, so that the entire cost cannot be given at this time. It is, however, believed that the tabulation cost, including the print- ing of the reports, will not exceed $10,000. Dean Johnson, of this committee, in addition to continuing the 134 AU] hSONY BOARD OX HIOHWAY RE>iEARCH traffic count of the ^laiyland highway syt^tein, from whicli a revised traffic map of the state will be compiled, has, in cooperation with the U. S. Bureau of Public Roads, established an experimental traffic counting station where weights and numbers of vehicles are auto- matically recorded. This installaton was in ()})eration for two months during the late sunnner, but progress was not made past the experimental stage, and in September the apparatus was removed for changes. Dean Johnson notes, in commenting upon this work, that there is ''sufficient promise of results to make it worth while to spend considerable more time experimenting." A self-recording device of this kind, which can be installed and operated at minimum cost, should solve, to a large extent, the objec- tion in many localities to the expense of a traffic census for a period of time sufficiently long to obtain accurate results. The committee believes that, while a traffic count is of value, more details should be considered than has been customary in the past, so that an analysis of present and probable future traffic may be made, as stated in a previous paragraph. This census nuist include enough detail and be taken over an extended enough period of time to deter- mine the seasonal traffic, which varies greatly in certain parts of the country and which is a maintenance rather than a construction prob- lem. The construction type must be based on the heaviest travel rather than on average or seasonal travel. A determination should also include the effect of slow-moving vehicles upon the general situation and the consideration of the possibility of legislation which will remove such vehicles to secondary roads during high peaks of traffic. This may include a recommendation for legislative act to govern the minimum speed permitted u])on Class A roads during certain hours of the day or times of the year. It is believed that the slow-moving vehicle is often the cause of congestion upon a main highway, that the importance of this fact is not at the present time understood, and that tlie capacity of the road can l)e largely increased by restrictive regulation. Mr. Hamlin: Widioul (piestion, the design of road must cover ])eak conditions rather than average conditions. Probably the most abnormal conditions we have in Connecticut are during the footl)all games at New Raven. Two years ago we took a census of xchicles at the Yale-Princeton game. Usually about 77,000 i)eople attend the most popular games, and several yeai's ago many s])ecial trains were run. There may have been si)ecial trains at the ^'ale-Anny gaiue this 3^ear; but, if so, they were not extensively advertised. Quite a number of people come the night before, stay at New Haven, and go back the second day; so this figure does not represent the entire PROCEEDIXGS OF THIRD ANNUAL MEETING I35 traffic. Two years ago we took a census from 8 :00 a. m. to 8 :00 p. m. At that time there were counted 26,016 cars on seven different roads which cover, to a large extent, the highways into New Haven. This year there were counted for the Yale- Army game 35,981, or approxi- mately a 3,000 average per hour. The maximum recorded traffic for one hour, at Station 5, on the Derby Road, was 1,393 vehicles between 6 and 7 p. m., or 23 vehicles per minute — more than one vehicle every three seconds. At Station 5 the maximum passenger- car traffic was 1,385, showing that out of 1,393 cars only eight were other than passenger motor cars. The figures show an increase of about 40 per cent in all traffic in two years attending one of the major games. This is one of the conditions to be met on the Boston Post Road. AVe are building a 24-foot highway of 10-inch concrete, with shoulders varying from 8 to 10 feet — the maximum shoulder we can obtain. The plans for the future comprehend the construc- tion of 36 feet of concrete, or four 9-foot travel patlis, giving two lines of travel in both directions. In the material received at the committee's request from the vari- ous states, there was a detailed report from the State of New Jersey. Mr. Sloan, of New -Jersey, has develo])ed a factor of 12.2 as repre- senting the relation between the maxinnun daily and hourly travel. 1 have been unable to clieck tliis factor Avith the results we have ob- tained in Connecticut, but. roughly, he estimates that the maximum of travel on any road is double the average liourly travel. I would be interested if any of you can clieck these conclusions. In regard to foreign cars, the average of 14 traffic stations extending to all parts of the state shows a percentage of foreign cars of 20.8. He has statistics regarding draw-bridges and he draws some very interesting- conclusions in regard to the value of lost time to vehicular traffic due to their opening. According to Mr. Sloan, the number of licenses will depend on growth in population, and within the next decade will reach a saturation point at about one vehicle for each four per- sons. By 1950 the population of the state will have increased not quite 50 per cent over that of 1920. By 1950 it is expected that the number of licenses will be 2.7 times that of the present year. It cannot be assumed, however, that highway traffic will increase in direct proportion to the number of licensed motor vehicles. It has been found that traffic has increased in New Jersey as the square root of the increase in the number of licenses. Assuming that licenses will increase 2.7 times, it is seen that highway traffic will increase slightly more than 1 .5 times. Dean A. N. Johnson: The results of the Marvland traffic census 136 ADVISORY BOARD ON HIGHWAY RESEARCH were compiled too late to forward to Mr. Hamlin and are now pre- sented. The State Roads Commission of Maryland has conducted a traffic census, the results of which now cover six complete years, 1917 to 1922, inclusive. The counts were taken at approximately 200 sta- tions throughout the state highway system. The mileage to which the traffic counts apply is approximately 1,600 miles. The counts STATE ROAC UNIVERSITY OF MARYLAND TRAFFIC STUDIES MADE IN COOPERATION WITH S COMMISSION AND US- BUREAU PUBLIC ROAD* "l> ^^--^ '> '^^^ J.. ^277, 258 ^'-1°'' i y K-^y^'-^ .^:^ < / <^ — Too ^0-^ .-^'' «<--■,-- 109. lOOOOO ^ \'x~-~ap ,-' 95.J ■^ ,6l\ \>j- r' 80 ^^.--^ \Vo - / ^ _*S4 _^.^ -i^Ojsj^^ DRAWN \ '^ y i. 61 59 42 ^•A ^-^ / ao ^6 DAILY AVERAGE NUMBER OF VEHICLES PER MILE OVER MO. STATE. HIGHWAY SYSTEM SHOWING RELATIVE RATE OF IN- CREASE FOR YEARS AS INDICATED. were taken at the same stations during the six-year period, one day per month, on different days of the week. The results are best seen by reference to Figure "A." First, attention is called to the increase in the average daily num- ber of vehicles per mile over the state highway system — from 258 vehicles in 1917 to 579 vehicles in 1922. It is noticed, however, that the rate of increase is flattening (the diagrams being drawn to semi- logarithmic platting, the slopes of the lines indicate the true relative rate of increase). Whereas the rate of increase for the period 1918 to 1921 is approximately 25 per cent, from 1921 to 1922 it is but little over 10 per cent. It is of interest to compare this rate of in- crease with the rate of increase of motor vehicle registration, shown by the dotted line. It is seen that for the four years the increase in motor vehicle registration is nearly constant (about 20 per cent per PROCEEDINGS OF THIRD ANNUAL MEETING 137 year), and that the rate of increase in total traffic has lessened some- what in advance of what we may expect for motor vehicle registra- tion. Both of these curves, it would be expected, ultimately would follow closely the population curve. The traffic counts show the number of horse-drawn vehicles and the number of trucks. It is to be noted that the amount of horse- drawn traffic has varied but little in the past five years, being almost constant in amount, while the rate of increase in truck traffic has been rapid, averaging for four years, 1917 to 1920, nearly 50 per cent a year. The curve then shows a somewhat marked lessening of l-non STATE ROADS COMM.SS 0~ AHOUS BUREAU OF PUBUC P0.05, 1500 1000 900 600 700 600 500 400 300 £00 ■00 BO 1666 i ROAD 1756 c-^"" M02Z . 9_58 .^^J^' 645^ '^9 -J I > ^'V 365_, ^^Oj^ ^214 166 195 ei UNIVLRSITV OF MAR^UANO TRAFFIC STUDIES MADE IN COOPERATION WITH STATE ROADS COMMISSION AND US BURtAU OF PUBUlC ROADS. 998 ALL vV >~^o^(e 1215. 997 ^^"'C o/ 1016 881 ^re may have been some decrease in the traffic, it is doubtful if it was as marked as has been indicated. In general, it is to be observed that there has been a somewhat greater increase in the amount of traffic in the second ten-miU' zone as compared witli the first ten-mile zone. From a study of these records, it would seem that we are now able to forecast somewhat closely the traffic, as a whole, that the state highways of Maryland will be called upon to carry for a number of years hence, as it is evident that the rate of increase must ultimately conform to the population curve, which has an increase, perhaps, of 1 or 2 per cent a year. As the present traffic increase is somewhat in excess of 10 })er cent a year, for a few years hence the increase will continue somewhat more rapidly than that for the population. The results, as here developed from the traffic census on state high- ways of Maryland, it is seen, coincide very closely with the results as pointed out by Mr. Hamlin from the New Jersey reports on traffic conditions, so that we may expect, in most well-settled connuuni- ties, to find our traffic conditions ap])roaching, within a very few years, more nearly the normal and lixed rates of increase, comparable with that of population. Mr. Duchastel : 1 remember seeing a frw weeks ago, in Quebec, at the Highway Department, the figures (in the King Kihvard lligh- way, between Kosse^ui Point and Ivlward's Highway. Tlic total number of pleasure cais was between 1.100 and 1,200. The f.ii'eign cars, and we call foreign eai's the .\iuerican ears, uumbered oxer SOO ; so you see our proportion of foi'eign ears is 70 per eei\t, which is far more than Mr. Hamlin's count. Chairman Johnson: Dr. J. (i. M(dva\-. of the I'. S. Bureau of Public Roads, will give a talk on PROCEEDINGS OF THIRD ANKUAL MEETING 139 ANALYSIS OF MOTOR TRANSPORT IN NEW ENGLAND Three groujjs of agencies in New England are engaged in the tran.'^l)ortation of })eople and connnodities — steam and electric rail lines, boat lines, and highway transportation companies. The ulti- mate i)urpose of each agency is to produce the highest type of coor- dinated transportation service for New England producers and con- sumers. The highway as an agency for the mass movement of people and connnodities is a modern development of transportation. From September to December, 1922, over a million net tons of freight were transported by motor trucks over the Connecticut Highway System. Rail service during this period was Ijelow par. owing to labor difh- culties, and naturally this situation increased the volume of highway transportation above the normal level. The largest portion of motor truck net tonnage movement was limited to the short-haul zone: 67.4 per cent was hauled less than 30 miles, 1()n. As it is generally conceded that corrugations form with little regard to construction or maintenance metliods, it would seem, then, tliat their prevention is a matter of traffic restriction or regulation. A\'e do not believe that this is feasible. 2. Dust preveution and .surface tiratiiient of (jvai'd roixh iiichi(Uitg sizes of mulch gravel. — Last year's maintenance report eoutained con- clusions from the preliminary report of the joint research project of the University of ^Michigan and the Michigan State Highway Depart- ment. Their final report has now been published and distributed. The Michigan Highway Department has continued this research work through the yoar^ and they have surface-treated with refined tar approximately 45 miles of well-constructed gravel roads. These roads carry a trafiic of from SOO to I.-IOO vehicles a day. PROCEEDINGS OF THIRD ANNUAL MEETING 145 The method which they have adopted this year is, briefly, as fol- lows : The work is started early in the spring, when the frost is com- ing out of the ground. The road is first shaped up and the gravel allowed to consolidate. As soon as it is fairly w^ell dried out, it is swept and an application of one-sixth of a gallon of refined tar per square yard is applied and allowed to penetrate into the surface. As soon as the surface takes on a brown a|)pearance, approximately one- third of a gallon per square yard of the same cold material is ap- plied and immediately given a top dressing of pea gravel or slag screenings. The Michigan Highw^ay Department emphasizes the fact that the success of this kind of work depends largely upon the maintenance given the road, both before and after the treatment. Surface breaks are taken care of as fast as they occur. They are filled with a cold mixture of refined tar and coarse sand. These patches are struck off and smoothed with a flat-bottom shovel. The Wisconsin Highway Commission has also treated a large mileage of gravel roads with refined tar, and they report satisfactory results. The Wisconsin method of treatment is similar to that used in Michigan. The road is first scarified and smoothed and allowed to compact. It is then swept and an application of cold tar applied from shoulder to shoulder, one-sixth of a gallon per square yard. The road is then opened to traffic for 24 hours and one-third of a gallon of cold tar per square yard api3lied and lightly sanded, using about 15 cubic yards of sand per mile for a 20-inch width. The travel is then allowed to use the road, and in a few days a second sand coat is applied, using about 35 cubic yards per mile for a 20-inch width. For patching, a tar mastic is prepared and allowed to stand in piles along the side of the road. Seventeen gallons of cold tar per cubic yard of sand is used in preparing this mastic. The Michigan Highway Department has experimented with cal- cium chloride and light oils for the purpase of dust prevention on their gravel roads. Their results with both have been satisfactory. They report the following points to be of great importance in calcium- chloride treatment; The chloride should be applied while there is still some moisture in the gravel and the application should be in the neighborhood of from one-half to one pound per square yard, giving a light treatment which will help avoid future trouble. The application should be uniform over the surface. The road should be left free from all lumps or piles of chloride. Ordinarily, three appli- cations during the season is sufficient, but the second or third appli- cation should be applied before the former application of material is entirely gone. In other words, there should be a little overlap rather than a dust interval. Care should be exercised in varying the amount to be applied to the different roads or sections of roads, inas- 146 ADVISORY BOARD ON HIGHWAY RESEARCH much as shaded porlioiis as well as sand or clay gravel have much to do with the quantity of chloride that should be used. Nearly everj^ state is now maintaining most of its gravel road mile- age with a mulch treatment of gravel. Various sizes and amounts are used. Little published data is available on this subject. The Michi- gan Department recommends the following: Replacement material should be limited to not exceeding one inch in size, preferably three- quarters inch, with around 60 to 75 per cent of the material being retained on one-eighth-inch mesh. The clay content should be held down to not over five per cent and the filler should be substantial ma- terial, which will help get away from the dust nuisance. The mulch should be applied in very light courses, not exceeding one inch in depth at any time, and should lie applied preferably late in the season or during the spring months. For future research work on this subject the committee would suggest a study of methods of maintaining tar-treated surfaces with special reference to the edges. A study should also be made to over- come as far as possible breaks that occur in the spring. 3. Crack-fillers for concrete pavement. — Nearly every state and numerous municipalities are experimenting more or less with various tars and asphaltic fillers for the repair of cracks in concrete pave- ments. A few y^ars ago there seemed to be a demand for a light- colored filler which would harmonize with the color of concrete pave- ment and thus be less noticeable than the black tars or asphalts. A satisfactory light-colored filler at a reasonable price has not been de- veloped. The committee is inclined to believe that public sentiment on this question is changing. As the mileage of concrete roads in- creases, the public becomes familiar with maintenance methods and does not look upon the black streaks with the horror that it did a few years ago. Furthermore, on heavily traveled pavements the oil and grease dropping-s from cars soon stain the pavement until the black crack-fillers are not conspicuous. One of the most extensive tests of crack-fillers is being carried on by the Iowa Highway Commission on a concrete road imme- diately outside of the city of Des Moines. On this experiment three tars, nine asphalts and blown oils, one emulsified asphalt and one light-colored material were used. The experiment has just been started. One barrel of each material has been applied and is now being tried out in service. The Iowa Highway Commission will issue periodic reports on this test; the first one will be available next spring, after the road has gone through the winter. The committee does not believe that further search for a light- colored filler is necessary. We do believe, however, that further re- search to determine the best specification for asplialt and tar fillers is desirable. PROCEEDINGS OF THIRD ANNUAL MEETING 147 4. Snow removal and snow removal equipment. — The Bureau of Public Roads is now making a study of this subject in a number of states which are engaging in snow removal. When this investiga- tion is completed a report will be prepared and probably made avail- able for distribution. The committee believes that further research is necessary on this subject, particularly with the end in view of de- veloping snow-removal equipment especially adapted to highway work. 5. Guide, caution and danger signs. — The rapid increase in recent years of intercounty, interstate, and even international traffic has brought a demand for standardization of guide, caution and danger signs. The idea of standardization of signs is that one can familiar- ize himself with the signs in his own locality, and can then proceed to travel in other parts of the state or in other parts of the United States with assurance that he need not learn a new set of signs in every county or state through which he may pass. The most comprehensive study of this subject during 1923 was undertaken by a committee of the Mississippi Valley Association at their annual meeting in Chicago in January, 1923. Some of the outstanding features of the report are : Highway signs are classified into three groups: (1) Route mark- ers, including road numbers and the distinguishing symbols accom- panying same; (2) Warning signs; (3) Guide and information signs. A. Route markers. Every state should adopt a distinctive symbol for its route markers. B. Warning signs. Should be uniform in all states and should not be used except where necessity is obvious. (1) Color. Warning and guide signs should be painted in black and white. (2) Location. Warning signs should be placed with the center 3 to 4 feet above the center of road elevation and 1 foot outside of the shoulder line. They should be placed from 300 to 500 feet in advance of the point of danger. (3) Shape and size. Warning signs should be 2 feet in diame- ter and the shape should indicate the nature of warning, as follows: Round Railroad crossings Octagonal Stop Diamond Slow Square Caution Rectangular For 2,uide and information signs 148 ADVISORY BOARD OX HJdHWAY RESEARCH C. Guide and information signs should be rectangular. All ad- vertising signs should be forl)idden on the highway right of way. The comnnttee feels that the work of tho Mis-sissippi Valley Asso- ciation is a step in the right directi(>n. The next step is to make nation-wide rules, so that uniform signs may be had in all parts of the United States. VCe would reconnnend that some national or- ganization, snch as the American Association of State Highway Otlicials, make further study of this problem and standardize as far as possible all highway guide, caution and danger signs. (). Maintenance accounting. — ^Last year your committee recom- mended that research work be undertaken to develop the most feas- ible uniform system of maintenance accounts, so that maintenance reports of various states might be put on a comparable basis. We are glad to report this year that both the Mississippi Valley and the American Associations of State Highway OfRcials have committees at work on this problem. The following report was adopted by the Mississippi A^'alley Association, and these states are attempting to make their accounting conform to the general classification. ''Expenditures by the states, through their maintenance organiza- tions, shall be classified in their annual reports as follows: 1. Maintenance Administration. 2. Maintenance Proper. 3. Additions and Betterments. 4. Parks, Camping-grounds and Roadside Beautification. "The last three elassificati(»ns shall be charged to a specific section of road. ''Additions and betterments shall include expenditures for the in- creased permanent value of the road to aceonnnodate traffic. This will include the cost of — 1. Increased width of surface or roadbed. 2. Increased depth of wearing surface over any previous maxi- mum depth. 3. 'New guard rails. 4. New walls. 5. New drainage struetnres and watercourses. 6. New subdrains. 7. Improving grades, alignment and vi^■ion."' Wisconsin has ad()i)ted ibc following classi Ileal i( tn : ]. Maintenance administration, covering general oiliet' over- head. PROCEEDINGS OF THIRD ANNUAL MEETING I49 2. Mainleuaneo supervision, covering district office overhead. 3. General maintenance, covering patrol and repair work. This cost is ke]3t bv patrol sections. 4. Eeconst ruction ; covers replacement to restore to original condition. Cost kept by patrol sections. 5. Additions and betterments; covers expenditures which change the section of the road and produce increased capital investment. 6. Marking and signing ; covers maintenance of all traffic guides on state system. It will be noted that these classifications can Ije consolidated to con- form to the four main divisions of the Mississippi Valley report. AVhen maintenance costs are kept in a uniform manner by the various states, it will be possible to make comparisons of snch costs. This, however, is only the first step toward uniform maintenance reports. Maintenance reports, even though summarized from uni- form records, are not of the greatest value as long as they are bashed on a per-mile cost. The next step is to reduce such costs to a ton- mile unit ; then useful comparisons between states and between type.s can be undertaken. This, of course, involves traffic census data and is partly within the scope of another Advisory Board committee. We would recommend continued efforts, in view of producing a standard basis of maintenance accounts for all state highway depart- ments. We would then recommend that, for comparative purposes, maintenance costs on a representative number of roads in each state be reduced to cost per ton-uiile of traflic on such roads. 7. Maintemtnce coxts as affected Inj the life of the road. — Little matter is availal^lc on this suljject. Constantly increasing trathc makes it almost impossible to determine accurately the efTect of age on maintenance costs. We believe, howcvei'. tiiat the subject is worthy of further consideration. The Maintenance Connnittee has been handicai)ped in its work because funds have not lieen available for connnittee meetings. All of the work has been handled by correspondence, the committee never having held a meeting since its appointment. We believe that two committee meetings a year are necessary for satisfactory work. One committee meeting should be held soon after the annual meeting of the Advisory Board, for the purpose of outlining the year's work. The second committee meeting should be held just prior to the annual meeting of the Advison^ Board, for the purpose of presenting and discussing subcommittee reports and formulating the annual report of the committee. 150 AJ91"/,s'0/?r BOARD ON HIGHWAY RESEARCH Mr. Root: I would like to say, in conclusion, that I concur in what Mr. MacDonald said last night. The biggest proljlem of a research organization is to lay out its program. I believe that if the mainte- nance committee can once lay out a definite program for the next year it will have gone a long way. O'nce the program is laid out, Ave shall have no trouble in securing cooperation of state highway depart- ments. The American Association of State Highway Officials has a maintenance committee A\hich has a subconnnittee on cooperation in research. The purpose of this body is to cooperate with that com- mittee on research problems. Mr. H. G. Shirley. State Highway Engineer of Virginia: My ob- servations have been that these corrugations in gravel and soil roads occur in dry weather more than in wet. In Mr. Root's report he re- ferred to kicking back of water and mud which built up these corru- gations. My experience has been that these corrugations occurred during unusually dry weather, when we were suffering from dust. Mr. Upham: I think this report brought out some excellent ideas. I do not agree, however, that corrugations should not be continued as a research subject. As Mr. Shirley says, we have more corruga- tions in dry weather than in wet weather. One of the recommenda- tions of the committee is as to scarifying the road. From a practical standpoint, a sand-clay or top-soil road will stand just so much scari- fying and no more. From another point of view, scarification not followed by proper weather gives a result almost as bad after scarifi- cation as before. Just one other point in connection with these cor- rugations. The committee has recommended scarifying. If weather conditions are favorable, scarifying is satisfactory for the time being. In some cases we have lightly scarified the surface and added an unusually strong binding clay, and then rescarified and redisked it, and we have found by addition of the binding clay that we have in- creased the surface tension of that road and the corrugations do not develop so quickly after that. One cannot always get strong bind- ing clays. Diu'ing the diy spells the corrugations are quite often corrected by dry dragging. Mr. Blair: Much complaint over the country is directed against the coarse material used in repairing corrugations. In leveling up the road, either l)y dragging or otherwise, what method have you used to reduce to a niinimiun the effect of coarse material in corruga- tions causing discomfort to travel and injury to tire? Mr. Upham : If we could get a homogeneous surface we should not have troul)lc with corrugations, but in soil roads it is iini)0ssible to get homogeneous materials. The material varies with almost every PROCEEDINGS OF THIRD ANNUAL MEETING 151 foot of road. We have to treat almost every section of the road differently. We carried on a census in North Carolina in connection with the Bureau of Public Eoads to determine, if possible, the limitations of the sand-clay and top-soil roads. We found that the figures were somewhere around 400 vehicles per day. This count included mostly passenger cars, a very small percentage of heavy trucks, a few light trucks and some horse-drawn vehicles. The particular road on which this investigation was made was an average sand-clay and top-soil road which satisfactorily withstood 400 vehicles. On an- other road, made of practically the same material, it was found that when the number of vehicles reached 500 or 600 the maintenance became somewhat high and the road corrugated badly. It seems that the limitation of sand-clay and top-soil road is somewhere around 400 vehicles per day. Of course, there are Avide variations of that type of road, depending on the quality of the material in the road surface. Mr. Manly: I cannot see why the road should carry 400 cars per day satisfactorily and then go to pieces on 500 or 600 per day, unless there is some heating or fatigue effect. Mr. Upham: After the corrugations once start, they develop rap- idly. This may possibly be a factor if not the reason. AVhen there is a small number of vehicles, the maintenance forces get out in time to catch corrugations in between the vehicles. When there is a large number of vehicles, there is not as intensive maintenance per vehicle. On tourist routes that pass through the state, we get as high as 800 to 1,000 vehicles per day. The particular road which is used mostly by these tourist passenger cars is corrugated very badly, and the east and west roads, made of practically the same material, that do not carry a traffic of more than half of these other roads, will be in very good shape and not corrugated and the maintenance remain quite low. These same north and south tourist lanes remain in good condition during almost the entire year, except when the tourist traffic is going north or south, and then these roads start corrugating. I do not know whether I can give you the reason why a larger num- ber of vehicles would l)e harder, in proportion, on the road than the lower number of vehicles, unless it is due to the fact that they do not receive as much maintenance in relation to the nvimber of vehicles. Prof. Eno : Are the tourist cars heavier and driven at a more rapid rate of speed than the local cars, as a rule? Mr. Upham : I believe not. In general, many of the touring cars are Fords. Local cars have little respect for speed laws. 152 ADVhSORY BOARD OX HIGHWAY UHtiEARCH Prof. Lay: We were getting ready to make some investigations on the action of the vehicle when travehng over some bumps and wanted to get a general idea of this ; so we made a simple experiment with a Ford car, a rail and a pile of sand. We set the rail on a con- crete road and sprinkled some sand on the road on the other side of it. We drove the Ford over this rail at various speeds. It had tires with a non-skid tread, so that we could observe any skidding or slipping of the wheels on the sand. At low speed, when the front wheels first struck the pavement they rolled along without any slip- ping, as we could observe the clear imprint of the non-skid tires. At higher speeds they bounded up and struck again several times. The rear wheels always wiped the sand off the spots which they struck the first time, and as we increased the speed the rear wheels jumped a little farther and perhaps would bound and strike again. When driving the car at much higher speeds the whole body was raised so high that while the springs did force the wheels and axle down, the wheels did not toucli the road until at the end of the second ''hop" of the rear wheels, when they struck with great force, clearing away the sand as before. This simple test throAvs considerable light on the corrugations formed on a dry gravel road. It shows that there can be a relation between the period of the springing system, the speed of the car and the distance between corrugations. It shows clearly how the driving wheels gain in speed while off the ground and abrade the road surface when they do come down on it. Mr. Crandell : I should like to suggest that some one on the com- mittee formulate a definition of gravel, because throughout the coun- try no one seems to know just what gravel is. I have tried experi- ments of getting students to duplicate on a blackboard in actual size their idea of gravel. I had seventeen students at the board and got seventeen different kinds of gravel. One man thinks sand is gravel and another man thinks crushed rock or large pebbles is gravel. AVe should have a clear definition of what constitutes gravel. Chairman Johnson: This was always an interesting subject for discussion. No one seems to get much further than the definition of the ^lassachusetts Highway Commission, that gravel is stone broken by nature. Mr. Manly: On the matter of corrugations of gravel roads, Mr. Upham brought out the point that maintenance cost has been on a per-mile basis rather than per vehicle passing. Corrugations, there- fore, have possibly formed on tlie basis of munber of vehicles passing in any given time, but within the re])air periods; so that if that is the whole answer, that is all right. On the other hand, if there is any- thing else in it, like the question of fatigue or the time element in PROCEEDINGS OF THIRD ANNUAL MEETING 153 settling of the subsoil or the top-dressing, it seems that it would be rather important to determine it. It could be determined if two pieces of road as nearly alike as possible were maintained in direct comparison, the one on the present system and the other on the basis of the number of vehicles passing rather than on the mileage basis. It could be settled if this type of road is limited to so many vehicles per day, or if it were simply limited to so many vehicles per unit of maintenance expense. Chairman Johnson : I understand the number of vehicles in each instance amounts to 400 or 500 per clay. Mr. Manly : But if it is 400 vehicles per day, and you maintain that road once a month or once in six months, it makes a difference. The question I brought up was : Are those corrugations due to pass- ing a critical point in the expense of maintenance of that road per vehicle passing, or is there some element of the frequency of passing that is a criterion beyond which you cannot go without enormously adding to the expense? Mr. Hubbard: It seems to me that it is not unreasonable to sup- pose that there is a limit to the number of vehicles that can pass over a gravel road. The suggestion has occurred to me that the resistance to displacement of surface of gravel road or top-soil road is more or less dependent upon the amount of moisture which the surface holds. During the day the moisture which is absorbed during the night gradually disappears if the weather is dry. The resistance has there- fore become less as the day progresses. It is not unreasonable to sup- pose that if the number of vehicles is greater the punishment will be more severe and the formation of corrugations more rapid. Mr. Bishop: I think that Mr. Upham and Mr. Shirley both dis- agree with conditions reported, and while I want to confess I am not an expert on this subject, I have studied with Dr. Ladd on gravel roads and I think that perhaps a part of the difference of opinion may be explained by the fact that Dr. Ladd's report covered the New England States and some of the Northwest States on gravel alone and did not extend to sand clay. The statement made, that the corrugations occur when the num- ber of vehicles reaches about 500 or 600, is merely a summation of statements made by state highway departments that they actually did occur between those limits. The question of maintenance, or the scarifying or dragging, is the summation of consensus of opinion of most of the state highway departments which reported on corru- gations of gravel roads. His investigation did not cover the sand clay. Mr. Older: I understand the committee recommends that main- 154 AD]WOHY BOARD OX HIGHWAY RESEARCH tenance costs be reported on the basis of ton mileage. I doubt very seriously the value of such a basis for reporting maintenance costs. For example : It might readily be conceived that a 4-inch or o-inch concrete road might sustain 100,000 ton-miles or more of passenger car traffic without damage of any kind, and yet be practically de- stroyed by 1,000 or 2,000 ton-miles of freight traffic carried in units imposing wheel loads of four or five tons. To report maintenance costs on the basis of ton mileage would add enormously to the cost of keeping the records without necessarily adding to their value. Prof. Agg: One of the things Mr. Root had in mind is suggested by the problems that are before Committee No. 1 in the determina- tion of transportation costs. Certainly we ought to get some figure to represent transportation costs on highways, and there must be some way to correlate those costs with the volume of traffic on the highway, because we know maintenance costs increase with traffic. I would suggest that, whatever unit might be adopted, it is higlily desirable that there be adopted some systematic way of reporting maintenance costs which will correlate those costs with the amount, and probably the type, of traffic that occasions those maintenance costs. Possibly a flat ton-mile is not the correct basis, but there should be some equivalent way of reporting costs. In attempting to get at the costs that have been reported, the committee has found a great deal of difficulty in interpreting the costs in such a way that those that are reported from some of the states — such, for instance, as New York State — can be compared with reports that come from Illinois, because of differences in practices in that respect. We do not use the same unit in reporting the traffic of the highway, neither do we use the same unit in reporting maintenance costs. I think that these committees should agree upon a system of units. Mr. Hamlin's committee and Mr. Root's committee might submit a unit for reporting traffic — they might work out an agreement as to a method of reporting traffic, a system of units that all would use — and likewise if the maintenance costs could be reported with some statement as to volume of traffic it would help. Mr. Shirley: I believe that some such distribution has got to be made of highway traffic. We know that with the passenger vehicle it is the number rather than the weight, and with the freight vehicle it is the weight rather than the number. PROCEEDINGS OF THIRD ANNUAL MEETING 155 STATEMENTS FROM REPRESENTATIVES OF CONSTIT- UENT ORGANIZATIONS Mr. Prevost Hubbard, representing the American Society for Test- ing Materials: During the past year the xA.merican Society for Test- ing Materials has revised, or adopted as new, a total of forty methods of tests and specifications for highway materials, and has submitted six or more standard specifications and tests to the American Engi- neering Standards Committee for adoption as Tentative American Engineering Standards. Upon recommendation of Committee D-4, the Society has adopted four standard specifications for slag and stone block. It has revised three standard tests for bituminous ma- terials and one standard specification for materials for cement mortar bed for brick and block pavements. It has revised two tentative methods for testing subgrade soils and bituminous materials, and eleven tentative specifications, one for non-bituminous, and ten for bituminous materials. In addition, it has adopted as tentative one new test for bituminous material, two new specifications for bitumi- nous materials, and five new specifications for non-bituminous ma- terial. It is expected that at the next annual meeting of the Society a considerable number of new specifications and methods of test for highway materials will be presented, and that many existing tenta- tive standards will be recommended for advancement to standard. It is believed that quite a number of the new tests will be of value to research workers on highway problems as a means of securing data which is necessary to the solution of such problems, even though the problems themselves are not principally concerned with materials of construction. Mr. Ward P. Christie, Research Engineer of the Associated Gen- eral Contractors of America: In undertaking the investigation of certain subjects concerned with highway development, the Associ- ated General Contractors have proceeded on the following assump- tions : 1. That problems of design and engineering will be successfully solved by the country's technical and professional minds. 2. That the development of local and state highway systems to meet the needs of agriculture, industrs^ and pleasure will be devel- oped by highway engineers in cooperation with representatives of these various interests and in spite of adverse political influence. 3. That it is the duty of construction companies to study the means of building these systems properly and economically. With these assumptions in mind, therefore, the association has concentrated its investigations upon elements that affect the cost of 156 ADVISONY HOARD OX JI Kill WAY BEHEMU'H construction. Though these subjects may, at casual glance, appear remote from the problems of design and engineering, yet they are, in the last analysis, the controlling factors in determining whether the engineer shall be able to execute his conceptions and render his utmost service, and whether the public, which he is attempting to serve, will appreciate his efforts and receive a satisfactory mileage for its appropriation. The field of research and investigation that may properly be as- signed to construction companies is practically virgin in so far as recorded data and authentic information is concerned. Text-books and treatises have been written in great numbers, but they deal ])ri- marily with those factors that are more or less definite in character and fail to touch the less tangible problems involved in the inexact science of construction economics. It is to problems of this nature, concerned more directly with the business end of construction, but also to a great extent with the business of the designer, that the Asso- ciated General Contractors are devoting their efforts. The principal subjects under investigation may be outlined as fol- lows: (1) Material supplies; (2) The field working force; (3) Con- struction methods; (4) Equipment development; (5) Seasonal sus- pension of work; (6) Preventable hazards in the industry; (7) Com- parative statistical tabulations. The following is an outline of the work which the Associated Gen- eral Contractors are carrying on in their individual field: 1. Matenal supplies. — Data is being compiled from government and industrial sources to show the quantities of the principal con- struction materials necessary to serve the industry, so that engineers and construction companies in laying their programs may have in- formation on production that will assist them in fitting their demand to the production capacity. 2. Field working force. — Efforts are being made to induce young men to join the ranks of construction labor and to train them in the use of specialized machines that are rapidly increasing in highway construction. Incident to this work is the tabulation of estimated shortages and the study of wages and cost of living. 3. Methods. — A study of methods is being carried on l)y indi- vidual companies, as the Associated General Contractors as an or- ganization has not yet obtained the funds sufficient to carry on this work. Reports of progress and new methods are furnished, however, by individual companies to the association and to the trade maga- zines. 4. Equipment development. — In coo])oration with manufacturers PROGEFA)rNGti OF THIRD ANXl'AL MEETING 157 of construction equipment, the improvement and development of machines is being handled through the Joint Committee on Con- struction Equipment, which has, to date, practically completed work in two lines, namely, Avheelbarrows and concrete mixers. These items of plant have been standardized within the past year and a time limit of January 1, 1925, has been set for the elimination of non-standard types. The wheelbarrow standards have already been put into effect. 5. Seasonal suspension of work. — The saving of construction costs that may be produced by continuous operation in the construc- tion industry has never been estimated, but the tendency of this continuity is well known. The Associated General Contractors for the past two years has devoted considerable effort and money in seek- ing to interest other elements of the industry in this question. Its policy has been that, rather than wait for quantitative estimates and exact data, it could more profitably urge adoption of the policy know- ing that the energy extended would result in great economy. With this question is closely linked the problem of winter construction methods, in which the cooperation of the engineering profession in experimental work is urgently needed. How far the seasonal depres- sion can be eliminated in the highway field is uncertain, but the consensus of opinion is that, where it can be brought about, the efforts will effect considerable saving. 6. Removable hazards of the industry. — As one of the important elements of cost in the construction project is the assumption of cer- tain hazards incident to the business world and to natural causes, an effort is being made to induce the owners of public construction to assume those hazards which such owners can most economically assume rather than incorporate them into the cost of construction. Many of these hazards, occurring but once in the lifetime of a con- struction company, are sufficient to throw it into bankruptcy, and it must obviously protect itself by insurance or contingency charges. The infrequency of such event makes it considerably more econom- ical for the owner to accept such hazards. Work on this subject is being continued in contingent cooperation of the American Associa- tion of State Highway Officials and various architectural and engi- neering organizations. The report upon the subject was issued last December by the joint committees of the first-named organization and the Associated General Contractors. 7. Comparative statistical relations. — In order that construction companies and others interested may keep track of the trend of vari- ous factors influencing the industry, graphical analyses comparing these factors are beina carried in the magazine of the Associated 158 ADVISORY BOARD ON HIGHWAY RESEARCH General Contractors. Among the statistical compilations carried from month to month are the construction cost index, the construc- tion volume index, based on the quantity of material shipped, index numbers of wages, cost of living and material prices, contracts awarded, volume of transportation and other data of a similar nature. Statistics pertaining to the construction industry have been meager in past years, but new tabulations are being started, and within the course of a few years sufficient data will be available to intelligently interpret the trend of economic factors within the industry. The Associated General Contractors, in common with other na- tional organizations, is restricted in its research work by lack of personnel, but the anticipation is that work of this nature will be expanded materially within the next two or three years. One of the most important subjects upon which the construction companies of the country would like to see a definite determination made by the engineering profession is the question of hardness of aggregates and their suitability as road materials. At present, aggre- gates of different French coefficient are specified in particular states, apparently for highways of the same general type. Conclusive in- formation upon this subject will doubtless make possible the utiliza- tion of many local sources of materials and result in a decrease of unit costs. Mr. A. D. Flinn, representing the Engineering Foundation, gave assurance of the friendly support of the Engineering Foundation. Prof. C. J. Tilden, speaking for Mr. William P. Eno, of the Eno Foundation, expressed the continued interest and cooperation of the Eno Foundation in the work of the Advisory Board on Highway Research. Mr. Pyke Johnson, of the National Automobile CJiamber of Com- merce: The most interesting bit of research undertaken by the Na- tional Automobile Chamber of Commerce in recent months is an analysis of the financial aspects of the highway program in its rela- tion to the total tax bill of the nation, by John E. Walker, former tax adviser to the United States Treasury, Mr, Walker first determined the total national, state and county taxes, then found what percentage of these were devoted to highway work, where the funds so expended were obtained, and to what uses they were put. As just one outstanding fact of interest, his study shows that while $1,000,000,000 in round figures were expended for highways in 1921, and while the total taxes collected in that year were $9,000,- 000,000, the actual effect on the general taxpayer was a contribution PROCEEDINGS OF THIRD ANNUAL MEETING 159 of less than six cents of each tax dollar for highway purposes due to the many special taxes levied for highway work. The study has only recently been completed, but will be ready for distribution within the next month. Another survey of general interest, now nearing completion, is that of the Chamber of Commerce of the United States, which has had committees working for nearly a year on the broad question of transportation. One of these, headed by Alfred H. Swayne, Vice- President of General Motors, and including in its membership agri- cultural, electric and steam railway, waterway, motor and public representatives, has been engaged in studying the relation of high- ways and highway transport to other transportation agencies. An- other committee, including A. J. Brosseau, of Mack Trucks, Inc., and Roy D. Chapin, President of the Hudson Motor Car Company, has been going into the problem of transportation taxation. Waterways, railroad consolidations, governmental relations and the freight rate structure are other problems under review. These reports will be issued this month. Mr. H. W. Alden, representing the Society of Automotive En- gineers: The Society is continuing its work on a number of different projects. Some of these are investigations by the Bureau of Stand- ards. Tests are to be undertaken in cooperation with the Bureau of Public Roads and the Rubber Association. This work covers impact tests, automobile brake tests and headlight specifications. Dean Anson Marston, representing the Association of Land Grant Colleges: The members of the x4.ssociation of Land Grant Colleges are the land-grant colleges of the United States, including Alaska, Hawaii, and Porto Rico. These colleges were organized in accord- ance with the Morrill Land Grant Act of 1862, amended by later legislation. Each of these institutions has some endowment provided by the United States Government, and in addition to the income from these endowments, the institutions also receive considerable annual sums from the central government, provided by Congress through continuing appropriations. The United States Government retains some general authority over all of these land-grant colleges, and in a general way it may be stated that they constitute the only educa- tional institutions in the United States having any such official na- tional relation. In addition to the funds received from the national government, each of these institutions has state support, and in most of the insti- tutions the state funds are several times those provided by the national government ; hence each land-grant college has actual official relation to the nation and to its state. They are unique in this respect. 160 ADVISORY BOARD ON HIGHWAY RESEARCH The United States Government appropriates very large sums of money annually to the land-grant colleges for the support of agri- cultural research, but as yet makes no corresponding appropriation for the support of engineering research. However, the land-grant colleges, by legislative action on the part of some states and without such action in others, have proceeded to organize engineering experi- ment stations in twenty-four different states. In addition, engineer- ing research is carried on actively in many of the land-grand colleges which do not have formally organized experiment stations. The Association of Land Grant Colleges is promoting the organi- zation of such stations and maintains a Standing Committee on En- gineering Experiment Stations. Through this committee the Asso- ciation publishes quarterly the "Engineering Experiment Station Record (in mimeographed form), which is sent to all of the land- grant colleges. Through the "Engineering Experiment Station Record" the engineering experiment stations and other engineering research organizations of the land-grant colleges in the several states are kept informed definitely of all engineering researches in prog- ress in such institutions throughout the United States. The several institutions thus can cooperate in research and can avoid undesirable duplication. The cooperation secured to date has been of most satis- factory character. In general, each college organizes its researches on the project basis, with definite assignment of research staff and of funds to each project. Some of these projects continue for several yeai-s, in which case progress reports are required from time to time. Eventually each project normally results in the publication of one or more bulletins. From the forthcoming annual rejiort of the Engineering Experi- ment Station Committee (which for such report collects definite sta- tistics from every state at al)out this time each year) , it appears that the land-grant colleges of the continental United States expended about $589,580 in engineering rcvsearch in the year 1923 ; that in that year their engineering research staffs included about 95 members full time, about 88 members part time paid, and more than 241 mem- bers part time with no extra pay ; also, it aj^pears that up to Decem- ber 31, 1923, the colleges will have published in excess of 469 en- gineering research bulletins, of which approximately 53 are for the year 1923. The magnitude of the engineering research work carried on by the land-grant colleges may be understood better by comparing it with that promoted by the Engineering Foundation. The land-grant colleges spend more annually for engineering research than the entire PROCEEDINGS OF THIRD ANNUAL MEETING IQl endowment of Engineering Foundation. Undoubtedly the land- grant colleges together constitute the largest organized engineering research agency in the country. The land-grant colleges are especially interested in highway re- search, and are especially qualified to conduct such research, and they are especially desirous of joining in a nation-wide program of highway research. a. They are the only nationally endowed educational and re- search institutions in the several states. b. Each of them is also a state institution, and for that reason is especially well qualified to cooperate with the state high- way commission. For these reasons the Association of Land Grant Colleges made ap- plication in 1922 for admission to the Advisory Board on. High way Research, which application was granted immediately. The youngest member of the Advisory Board on Highway Re- search submits this report with its verj^^ best good wishes for the Board. Prof. A. II. Bhrnchard, representing the National Highway Trajfic Association, sent in the following written report: In the 1922 Report to the Advisory Board on Highway Research, the titles of about thirty-five national committees devoted to research problems were included. Although all of these committees are ac- tively engaged in research work, in order to avoid repetition, in this report only the following titles of those committees wdiich will report, at meetings during the current year of 1923-1924 Avill be mentioned. ''License Fees and Motor Vehicle Taxation." "Regulations Covering Speeds, Weights and Dimensions of Trailers." "Equitable Distribution of Cost of Construction, Interest on Bonds, Replacements and Maintenance of State Llighways." "Development of Transportation." "Interrelationship and Coordination of Raihvay, Waterway, Air- way and Highway Transportation." "Traffic Capacity and Widths of Highways Outside of Munici- palities." "Widths of Roadways and Sidewalks in jMunicipalities." "Highway Transport Franchises." "Highway Transport Clearing Houses." "Uniform Highway Signs." "Traffic Center Lines on Roadways." "Safety Regulations at Railroad Grade Crossings." 162 ADVISORY BOARD ON HIGHWAY RESEARCH "Mechanical Devices for Highway Traffic Regulation." ''Relative Efhciency of Different Types of Car-stop Safety Zones and Their Relation to Parking and Ranking Regulations." "Segregated and O'ne-way Streets in Municipalities." The national committees listed in this report, as well as our other committees, are composed of from three to fifteen specialists. Our reports are mimeographed or multi graphed and are available for distribution to those who are particularly interested, and hence are available for review and reference by any committee of the Ad- visory Board on Highway Research. ELECTION OF OFFICERS Report of Committee on Nominations: T'he following nominations for officers of the Advisory Board were reported by the Committee Chairman, Mv. A. T. Goldbeck: Chairman, Dean A. N. Johnson (renominated). Vice-Chairman, Mr. A. D. Fhnn (renominated). Executive Committee : Mr. Thos. H. MacDonald (renominated). Prof. T. R. Agg (renominated). Dr. II. C. Dickinson. Mr. C. M. Upham. Mr. A. J. Brosseau. The report of the committee was accepted and the Director cast the ballot of the Board for the nominees. Chairman Johnson: A meeting of the Executive Committee is called immediately following adjournment. Adjourned at 10 p. m. Publications of the National Research Council Bulletin Series Volume 1 Number 1. The national importance of scientific and industrial research. By George Ellery Hale and others. October, 1919. Pages 43. Price $0.50. Number 2. Research laboratories in industrial establishments of the United States of America. Compiled by Alfred D. Flinn. March, 1920. Pages 85. Price $1.00. [Out of print. See Number 16.] Number 3. Periodical bibhographies and abstracts for the scientific and technolog- ical journals of the world. Compiled by Ruth Cobb. June, 1920. Pages 24. Price $0.40. Number 4. North American forest research. Compiled by the Committee on Amer- ican Forest Research, Society of American Foresters. August, 1920. Pages 146. Price $2.00. Number 5. The quantum theory. By Edwin Plimpton Adams. October, 1920. Pages 81. Price $1.00. [Out of print. See Number 39.] Number (5. Data relating to X-ray spectra. By William Duane. November, 1920. Pages 26. Price $0.50. Number 7. Intensity of emission of X-rays and their reflection from crystals. By Bergen Davis. Problems of X-ray emission. By David L,. Webster. December. 1920. Pages 47. Price $0.60. Number 8. Intellectual and educational status of the medical profession as repre- sented in the United States Army. By Margaret V. Cobb and Robert M. Yerkes. February, 1921. Pages 76. Price $1.00. Volume 2 Number 9. Funds available in 1920 in the United States of America for the en- couragement of scientific research. Compiled by Callie Hull. March, 1921. Pages 81. Price $i.oc. Number 10. Report on photo-electricity including ionizing and radiating potentials and related effects. By Arthur Llewelyn Hughes. April, 1921. Pages Sj. Price $1.00. Number 11. The scale of the universe. Part I by Harlow Shapley. Part 1 1 by Heber D. Curtis. May, 1921. Pages 47. Price $0.60. Number 12. Cooperative experiments upon the protein requirements for the growth of cattle. First report of the Subcommittee on Protein Metabolism in Animal l-eeding. By Henry Prentiss Armsby, Chairman. June, 1921. Pages 70. Price $1.00. Number 13. The research activities of departments of the State government of California in relation to the movement for reorganization. By James R. Douglas. June, 1921. Pages 46. Price $0.60. Number 14. A general survey of the present status of the atomic structure problem. Report of the Committee on Atomic Structure of the National Research Coinicil. By David L. Webster and Leigh Page. July, 1921. Pages 61. Price $0.75. Number 15. A list of seismologic stations of the world. Compiled by Harry O. Wood. July, 1921. Pages 142. Price $2.00. Volume 3 Number 16. Research laboratories in industrial establishments of the United States, including consulting research laboratories. Originally compiled by Alfred D. Flinn; revised and enlarged by Ruth Cobb. December, 1921. Panes 135. Price $2.00. Number 17. Scientific papers presented before tiie American Geophysical Union at its second annual meeting. March. 1922. Pages 108. Price $1.50. Niiitiber 18. Theories of magnetism. By members of the Committee on Tlieories of Magnetism of the National Research Council. A. P. Wills, S. J. Barnett, L. R. Ingersoll. J. Kunz. S. L. Quimby, E. M. Terry, S. R. Williams. August, 1922. Pages 261. Price $3.00. Volume 4 Numl)er 19. Celestial mechanics. Report of tlie Committee on Celestial Mechanics of the National Research " Council. E. W. Brown, G. D. Birkoff, A. O. Leuscliner, H. N. Russell. September, 1922. Pages 22. Price $0.40. Number 20. Secondary radiations produced by X-rays, and some of their ai)i)lica- tidus to physical problems. Arthur H. Compton. October, 1922. Pages 56. Price $1.00. Number 21. Highway research in the United States. Results of census by Ad- visory Board on Highway Research, Division of Engineering, National Research Council, in cooperation with the Bureau of Public Roads, United States De- partment of Agriculture. William Kendrick Hatt. October, 1922. Pages 102. Price $i.SO. Number 32. Mechanical aids for the classification of American investigators, with ilhislrations in the field of psychology. Harold C. Bingham. November, 1922. Pages 50. Price $0.75. Number 2H. Certain problems in acoustics. Compiled by the Committee on Acous- tics, National Research Council. November, 1922. Pages 31. Price .$0.50. Number 24. Electrodynamics of moving media. Report of the National Research Council Committee on Electrodynamics of Moving Media. W. l\ G. Svvann, lohn T. Tate, H. Batcman, and E. H. Kennard. December, 1922. Pages 172. Price $1.50. Number 25. Celestial mechanics. A survey of the status of the determination of the general perturbations of the minor planets. Appendix to the report of the Committee on Celestial Mechanics, National Research Council. A. O. Leuscbiier. December, 1922. Pages 73. Price $1.00. Volume 5 Number 2(>. Cooperation with the Federal Government in scientific work. E. VV. Allen. December, 1922. Pages 27. Price $0.50. Number 27. The present status of visual science. Leonard Thompson Troland. December, 1922. Pages 120. Price $1.50. Number 28. Algebraic numbers. Report of the Committee on Algebraic Numbers, National Research Council. L. E. Dickson, H. H. Mitchell, H. S. Vaudiver, G. E. Wahlin. February, 1923. Pages 96. Price $1.50. Number 29. The status of scientific research in Illinois by state agencies otiier than the University of Illinois. Leonard D. White. March, 1923. Pages 7- Price $0.50. [Replaced by Number 44.] Number 36. List of manuscript bibliographies in chemistry and chemical technol- ogy. By Clarence J. West and Callie Hull. December, 1922. Pages 17. Price $0.25. Number 37. Recent geographical work in Europe. By W. L. G. Joerg. July, 1922. Pages 54. Price $0.50. Number 38. The abstracting and indexing of biological literature. J. R. Schramm. November, 1922. Pages 14. Price $0.25. Number 39. A national focus of science and research. George Ellery Hale. November, 1922. Pages 16. Price $0.25. Number 40. The usefulness of analytic abstracts. Gordon S. Fulcher. December, 1922. Pages 6. Price $0.15. Number 41. List of manuscript bibliographies in astronomy, mathematics, and physics. Clarence J. West and Calhe Hull. March, 1923. Pages 14 Price $0.25. Number 42. Doctorates conferred in the arts and the sciences by American uni- versities, 1921-1922. Clarence J. West and Callie Hull. March, 1923. Pages 14 Price $0.25. Number 43. Functions of the Division of Geology and Geography of the National Research Council. Nevin M. Fenneman. December, 1922. Pages 7. Price $0.20. Number 44. Fine and research chemicals. Second revision. Clarence J. West May, 1923. Pages 45. Price $0.50. Number 45. List of manuscript bibliographies in the biological sciences. Clarence J. West and Gallic Hull. June, 1923. Pages 51, Price $0.50. Number 46. Problems in the field of animal nutrition. Sub-committee on Animal Nutrition. May, 1923. Pages 9. Price $0.15. Number 47. A statistical study of tuberculosis mortality in Colorado for the thirteen years 1908 to 1920. Henry Sewall. August, 1923. Pages 33. Price $0.50. Number 48. Psychological work of the National Research Council. Robert M. Yerkes. November, 1923. Pages 7. Price $0.20. Number 49. Statement of activities of the National Research Council for the year, July i, 1922-June 30, 1923. Vernon Kellogg. November, 1923. Pages 16. Price $0.25. Number 50. Second report of the Committee on Contact Catalysis. Wilder D. Bancroft, Chairman. In collaboration with other members of the committee. December, 1923. Pages 141. Price $0.50. Number 51. The higher agricultural education of the future. E. Marchal. April- June, 1923. Pages 6. Price $0.20. Orders, accompanied by remittance, should be addressed to Publications Office NATIONAL RESEARCH COUNCIL Washington, D. C 2692 nS NORTHEASTERN UNIVERSITY LIBRARIES 3 9358 00865395 5 DUE DATE Printed in USA