Skip to main content <#maincontent> We will keep fighting for all libraries - stand with us! Internet Archive logo A line drawing of the Internet Archive headquarters building façade. Search icon An illustration of a magnifying glass. Search icon An illustration of a magnifying glass. Upload icon An illustration of a horizontal line over an up pointing arrow. Upload User icon An illustration of a person's head and chest. Sign up | Log in Web icon An illustration of a computer application window Wayback Machine Texts icon An illustration of an open book. Books Video icon An illustration of two cells of a film strip. Video Audio icon An illustration of an audio speaker. Audio Software icon An illustration of a 3.5" floppy disk. Software Images icon An illustration of two photographs. Images Donate icon An illustration of a heart shape Donate Ellipses icon An illustration of text ellipses. More Hamburger icon An icon used to represent a menu that can be toggled by interacting with this icon. Internet Archive Audio Live Music Archive Librivox Free Audio Featured * All Audio * This Just In * Grateful Dead * Netlabels * Old Time Radio * 78 RPMs and Cylinder Recordings Top * Audio Books & Poetry * Computers, Technology and Science * Music, Arts & Culture * News & Public Affairs * Spirituality & Religion * Podcasts * Radio News Archive Images Metropolitan Museum Cleveland Museum of Art Featured * All Images * This Just In * Flickr Commons * Occupy Wall Street Flickr * Cover Art * USGS Maps Top * NASA Images * Solar System Collection * Ames Research Center Software Internet Arcade Console Living Room Featured * All Software * This Just In * Old School Emulation * MS-DOS Games * Historical Software * Classic PC Games * Software Library Top * Kodi Archive and Support File * Vintage Software * APK * MS-DOS * CD-ROM Software * CD-ROM Software Library * Software Sites * Tucows Software Library * Shareware CD-ROMs * Software Capsules Compilation * CD-ROM Images * ZX Spectrum * DOOM Level CD Books Books to Borrow Open Library Featured * All Books * All Texts * This Just In * Smithsonian Libraries * FEDLINK (US) * Genealogy * Lincoln Collection Top * American Libraries * Canadian Libraries * Universal Library * Project Gutenberg * Children's Library * Biodiversity Heritage Library * Books by Language * Additional Collections Video TV News Understanding 9/11 Featured * All Video * This Just In * Prelinger Archives * Democracy Now! * Occupy Wall Street * TV NSA Clip Library Top * Animation & Cartoons * Arts & Music * Computers & Technology * Cultural & Academic Films * Ephemeral Films * Movies * News & Public Affairs * Spirituality & Religion * Sports Videos * Television * Videogame Videos * Vlogs * Youth Media Search the history of over 835 billion web pages on the Internet. Search the Wayback Machine Search icon An illustration of a magnifying glass. Mobile Apps * Wayback Machine (iOS) * Wayback Machine (Android) Browser Extensions * Chrome * Firefox * Safari * Edge Archive-It Subscription * Explore the Collections * Learn More * Build Collections Save Page Now Capture a web page as it appears now for use as a trusted citation in the future. Please enter a valid web address * About * Blog * Projects * Help * Donate * Contact * Jobs * Volunteer * People * Sign up for free * Log in Search metadata Search text contents Search TV news captions Search radio transcripts Search archived web sites Advanced Search * About * Blog * Projects * Help * Donate Donate icon An illustration of a heart shape * Contact * Jobs * Volunteer * People Full text of "Recent Actions Taken By The Symbols, Units, And Nomenclature " See other formats Recent Actions Taken By the Symbols, Units, and Nomenclature (S.U.N.) Committee Of the International Union of Pure and Applied Physics (I.P.U.) in Reference to C.G.S. Magnetic Units BY ARTHUR E. KENNELLY* Fellow and Past President of the A.I.E.E. Synopsis.—Iin reference to the C.G.S. magnetic units, a brief historical outline is presented of the actions taken by international electrical congresses from 18938 to 1904, and by international tech- nical and scientific unions from 1904 to 1982. Emphasis is laid on the results of the S.U.N. Committee meeting of July, 1932, as the latest development. l The present status of international agreement on C.G.S. magnetic units is unofficially presented, in the light of the historical back- ground. | Turning to future possible developments, certain questions are PURPOSE AND SCOPE HE aim of this paper is to offer a brief unofficial T account of what international actions have recently been taken in the matter of C.G.S. magnetic units, tending to eradicate some of the ambiguities that have pervaded magnetic literature during the last thirty years, and especially the actions in this direction taken during 1931-1982, by the International Electrotechnical” Commission (I.E.C.) and by the International Physical Union (I.P.U.), together with the present nomenclature status of these C.G.S. units, and what little yet remains to be done, in order to secure general international agreement. It will be seen that this is essentially a mat- ter of names, definitions and nomenclature; since no changes have been suggested in the C.G.S. units themselves. The second part of the paper relates to magnetic units in the practical, or volt-ohm-ampere series. Here rela- tively little has yet been accomplished. In order to extend this series into a complete comprehensive sys- tem, it appears to be necessary to secure agreement upon two very important questions: (1) Shall the mmf of a coil of NI ampere-turns be 47NTI practical units, or NI ampere-turns simply? (2) What unit of length shall be embodied in the system? Suggestions are advocated upon both questions. HISTORICAL OUTLINE SINCE 1927 In a paper entitled Magnetic-Circuit Units, read before the Midwinter Convention of the A.I.E.E., January 29th, 1930, the writer had the privilege of presenting a historical outline of magnetic units from *Harvard University and Mass. Institute of Technology. 17. See bibliography for numbered references. Presented at the Winter Convention of the AJI.E.E., New York, N. Y., January 23-27, 1938. pending for consideration, by either or both committees, in order to clear up any outstanding differences of international opinion over magnetic units in the classical C.G.S. system. In order to convert the existing series of practical electromagnetic units (ohm, volt, ampere, coulomb, farad, joule, watt, henry and pramazwell) into a complete system, international agreement must first be reached upon the two outstanding questions: (a) Subrationalising (4T NI or NI for mmf). (b) The unit of length to be adopted (meter or cm). Suggestions are offered along both these lines. 1861 up to that date. It will therefore not be necessary here to reconsider the history of the subject prior to 1927, when the International Electrotechnical Com- mission (I.E.C.) prominently entered the fidld. The I.E.C. was projected in 1904, at the International Electrical Congress of St. Louis, and was inaugurated at London in 1906. It has been continuously in existence since that time. It has national committees in some twenty countries. Each such national committee is composed of representatives of its national government, local scientific and technical bodies, as well as various electrical industries. The general secretariat is in London; but the presidents, of which there have been eight up to this date, are elected successively from the various countries. Each national committee has its own president, secretary, and other officers. The work of the I.E.C. is divided among nineteen separate advisory committees, each with its own presi- dent, and with a secretariat located in one country. Some advisory committees have delegates from a rela- tively small number of countries. Others have delegates from all or nearly all countries. Through the London central office, the different advisory committees convene at suitable times and places, to work out their own problems. As and when they arrive at agreements, they notify the central office, and report to the next plenary meeting of the I.E.C. asa whole. The plenary meetings are held about once in two years, and there have been nineteen in all thus far, distributed in time and place among twelve cities. The work of a plenary meeting consists largely of receiving reports from the various advisory committees, and of adopting them, as a body, by vote. The I.E.C. also receives and passes upon applications for the holding of international electrical congresses, in countries which desire to hold them. The tasks undertaken by the various advisory com- mittees relate broadly to electrical engineering, and 647 33-52 648 especially to the standardisation of machinery, appara- tus, specifications, methods and tests entering into electrical applications. In advisory committee No. 1 on “Nomenclature,” definitions, symbols and units have been dealt with at various times. At the plenary meeting of 1927, in Bellagio and Rome, a proposal was introduced by the Italian national committee, relating to magnetic units; both C.G.S. and practical. After some unsuccessful discussion, it was referred to a special subcommittee on “Magnetic Units” of the Advisory Committee on Nomenclature, for study and report, with a personnel group of rep- resentatives from seven countries. This subcommittee worked over the subject until 1929, without being able to reach an agreement, owing to the marked differences of opinion presented among the various national com- mittees involved. The subcommittee reported in 1929, recommending that, in view of the large range of differences in opinion, the subject should be discussed in the national committees and placed upon the agenda for the next plenary meeting at Oslo in 1930. The subject was, in fact, ventilated in the technical press of several countries, in advance of the meeting of 1930. The main difficulties in reaching agreement centered around the unit? name gauss, which had been adopted in 1900, at Paris. Magnetic literature all over the world, between 1900 and 1930, showed much confusion and ambiguity as to the meaning of this name. Some writers used it exclusively for the unit of flux density, or induction density B. Others used it exclusively for the unit of magnetising force, or magnetic intensity H; while yet others used it for both B and H indiscrimi- nately. Moreover, some authoritative text books held that the quantities B and H were physically differ- ent, in which case, of course, the unit name gauss could only be applied to one; but to which was not certain. Other equally authoritative text books held, on the contrary, that the quantities B and H were physically identical; so that the gauss if applied to one, would also be applicable to the other. There was no middle ground between these two positions, and the first aim of the international subcommittee dealing with the subject, was to decide which of the two contentions should be supported. When the magnetic units subcommittee met (June, 1930) in Scandinavia, its numbers were increased to include twelve countries, and its name was changed to Section B of the Nomenclature Committee, dealing with Electric and Magnetic Magnitudes and Units, (E.M.M.U.). After discussion, it was voted unani- mously to adopt the convention, that for electro- technical purposes, the quantities B and H should be regarded as being physically distinct; so that in a vacuum, or free space, B = uH, where us is the space permeability, and is not a mere number, but a physical quantity having dimensions; furthermore that in the case of magnetic substances, B = uH, where u is the absolute permeability of the medium, with the same KENNELLY Transactions A.I.E.E. dimensions as uo; but the relative permeability of the medium would be u/uo a mere number. These conventions having been adopted, the E.M.- M.U. committee was able to deal speedily with other C.G.S. magnetic units which had offered difficulty. The following list of units and names was agreed to, either unanimously or by considerable majorities.* (a) For the C.G.S. Magnetic Unit of magnetic flux 6, the name maxwell (b) For the C.G.S. Magnetic Unit of flux density B, the name gauss (c) For the C.G.S. Magnetic Unit of magnetising force H, the name oersted (d) For the C.G.S. Magnetic Unit of mmf the name gilbert. The name maxwell had already been adopted in 1900 at Paris, for magnetic flux 6, as reaffirmed here under (a). The name oersted was applied to the unit of H, partly to indicate that, by convention, the name gauss being assigned to the unit of B, was not applicable to the unit of H. In America, the name oersted had been applied provisionally to the unit of reluctance @, since 1900; but without international authorisation. The assignment of the name oersted to the unit of H as in (c), leaves the C.G.S. magnetic unit of reluctance with- out a name, for the present. The E.M.M.U. Committee also adopted the name pramaxwell for the unit of magnetic flux © in the practical series (108 maxwells) and the prefix pra as a general means of designating a practical unit derived from a C.G.S. unit. The preceding resolutions, and others not directly within the scope of the C.G.S. units, were reported to the plenary convention of the I.E.C. at Oslo, and by it unanimously adopted, July 9th, 1930. Following upon the publication of the I.E.C. actions on the C.G.S. magnetic units, comments appeared in the scientific and technical press of various countries. In general, these comments were favorable; but some adverse criticisms also appeared, mainly directed to two points: (a) That names of savants should not have been given to C.G.S. units. (b) That the name gauss should not have been given to the unit of B; but to that of H, in conformity with the usage of terrestrial magnetic observatories. In regard to (a), the Paris Congress of 1900 had assigned the names maxwell and gauss to C.G.S. units, in view of the decision of the Congress of 1893, at Chicago, that working magnetic units should be kept within the C.G.S. system. In regard to (b), it was pointed out that in view of the confusion in literature over the use of the name gauss, the I.E.C. convention necessarily brought em- barrassment to some group of workers; but that the number who were in the habit of using the gauss for B seemed to be much greater than those accustomed to using the gauss for H. Moreover, the earth’s magnetic June 1933 field is measured in air, which being a feebly magnetic medium, may be regarded as developing a flux density B under the action of the magnetic intensity H al- though, from a numerical standpoint, there would be no appreciable difference between the air values of H and B. . Terrestrial magnetic observatories are also accustomed to report their magnetic measures in terms of a special unit the gamma or y, equal to 1/10° oersted, and there had been no suggestion of disturbing this usage. In view of any or all differences of opinion elicited by the publication of the I.E.C. Oslo resolutions, ar- rangements were made and announced early in 1931, to hold a special meeting of the E.M.M.U. committee at London, in September 1931, to review these C.G.S. recommendations. ESTABLISHMENT OF THE S.U.N. COMMITTEE BY THE INTERNATIONAL PHYSICAL UNION The International Union of Pure and Applied Physics (I.P.U.), met at Brussels in July 1931, and established for the first time, an international com- mittee on “Symbols, Units and Nomenclature,” (S.U.N.). The 8.U.N. committee elected as its Chair- man Sir Richard Glazebrook, and as its Secretary, Dr. E. Griffiths. Since electro-technical units fall within the scope of the 8.U.N. Committee, the I.E.C. extended an invitation to the I.P.U., to aid in the task of seeking international agreement on electric and magnetic units, through the coöperation of the 8.U.N. and E.M.M.U. committees. This invitation was ac- cepted, and the 8.U.N. Committee was invited to have its members attend and take part in the London E.M.M.U. meeting, already scheduled for September 1931. The S.U.N. Chairman and Secretary were able to attend, and also the General Secretary of the I.P.U., Professor H. A. Abraham. ACTIONS OF THE LONDON MEETING oF E.M.M.U. COMMITTEE, SEPTEMBER 1931 At the opening of the E.M.M.U. meeting in London, September 18th, 1981, the President of the I.E.C., Dr. A. F. Enström, and the General Secretary, Mr. C. LeMaistre, took part.” The national committees of ten countries were represented:—Denmark, France, Germany, Great Britain, Holland, Italy, Norway, Poland, Sweden and U.S.A. The agenda proposed by the American committee in February 1931, were then adopted as basis for procedure. The first proposal on these agenda: (la) was con- sidered: “that the C.G.S. magnetic units and their ac- cepted names, as fixed at the Oslo plenary meeting, should remain undisturbed”’ This proposal was voted unanimously, two countries, however, abstaining, (Holland and Italy). The next proposal (1b), related to a series of practical SYMBOLS, UNITS, AND NOMENCLATURE COMMITTEE 649 magnetic units, based on the pramaxwell, or volt-second as adopted at the Oslo meeting, with the ampere-turn as unit of mmf. After a long discussion, during which considerable difference of opinion was revealed, it was voted to defer action on these practical units until the national committees should have had further oppor- tunity of examining the question. On the particular question, however, of a practical unit of mmf F, and whether a coil of N turns, carrying a steady current of I amperes, should be taken as § = 4r (NI) units, or merely 5 = (NI), a vote was called for on the proposition that it should be § = 4 7 (NI). Three countries were in favor of the proposition (France, Holland and Italy); while four voted against the proposition: (Germany, Great Britain, Sweden and U.S.A.). Two countries, Norway and Poland, ab- stained from voting. It is evident that there is much difference of opinion over this important question. Item II of the agenda, on a conventional interpre- tation of reactive power in unspecified, alternating- current right-angle triangles, was next considered; t.e., whether + 7 vars should be taken as signifying induc- tively reactive power, or the reverse, when the basis of the diagram is not specified. It was unanimously voted that a convention should be adopted; but that decision as to the direction, should be deferred until a new and definite American recom- mendation should be presented. Some difference of opinion had been raised as to whether the particular direction recommended in the agenda (+ 7 P, for inductively reactive power), was the more advantageous choice. Item III of the agenda related to a proposed inter- national adoption of the prefixed “ab”? and “stat,” applied to a practical unit, for the designation of a corresponding unit in the C.G.S. magnetic, or in the C.G.S. electric system, respectively. These prefixes have been and are in use, to some extent, in America; but have never been internationally adopted. For lack of support to the proposal, no action was taken and the meeting adjourned. MEETING OF THE S.U.N. COMMITTEE IN PARIS The 8.U.N. Committee issued a questionnaire, in December 1981, to physicists and physical societies in different countries, on “Electrical Units,” calling attention to certain ambiguities in the definitions of fundamental electric and magnetic quantities entering into the C.G.S. system, and inviting opinions as to how these ambiguities might be eliminated, in order to arrive at international agreement. The replies to this questionnaire were collected. In view of the con- siderable number of physicists assembling in Paris during the week July 5th-12th, to attend the Inter- national Electrical Congress of 1932, the S.U.N. Com- mittee called an informal meeting in Paris on July 9th, to discuss these questions, and especially the C.G.S. magnetic units. 650 The meeting was held under the chairmanship of Sir Richard Glazebrook, Dr. E. Griffiths of the British National Physical Laboratory (N.P.L.) being the Secretary. Nineteen persons, from eight countries, attended. The actions taken were informal, in the sense that the voting was not by countries; but by individuals. The following propositions were adopted, either unanimously, or by considerable majorities: 1. Any system of units recommended must retain the eight internationally recognised practical units: joule, watt, coulomb, ampere, ohm, volt, farad, henry. 2. The C.G.S. system of units is suitable for the physicist. 3. A system of practical units ineluding the above eight quantities, is derived from these by multiplying the C.G.S. unit by appropriate powers of 10. 4. That M. Abraham at the request of the conference, should prepare for consideration by the Commission, a preliminary report on the fundamental units. 5. The system of magnetic units may be based on the follow- ing two methods as alternatives: (a) The force between two elementary magnet poles, (Coulomb). (b) The force between two elements of current,(Ampére). 6. B and H are quantities of different nature. 7. The factor 47/10 should be retained in the definition of magnetomotive force. The following table of electromagnetic quantities, which in accordance with suggestions received, has KENNELLY in electrotechnical journals. Transactions A.I.E.E. and that the unit names maxwell, gauss, gilbert and oersted for #, B, § and H are respectively satisfactory. If these conventions continue to receive support, the ambiguities and confusion which have pervaded inter- national magnetic literature during the past thirty or forty years, should disappear. The only group whose current practise seems to be embarrassed by the con- ventions, is the terrestrial magnetic group, who have long been accustomed to employ the gauss for H. As has already been pointed out, however, these important scientific researches in terrestrial magnetism deal almost exclusively with magnetic phenomena in air, at ordinary temperatures and pressures. The numerical difference, between B and H in air under such conditions, is ad- mitted to be only about one part in three millions; which is insignificant in the existing stage of measure- ment precision in terrestrial magnetism. There can be no present confusion in magnitudes, when the gauss in air is mentioned. Moreover the use of the gamma in terrestrial magnetic work is not interfered with. In the United States, The I.E.C.-I.P.U. Conventions have been accepted by the National Bureau of Stand- ards, the Society for Testing Materials, various committees of the A.I.E.E., and the U.S. National Committee of the I.E.C. Endorsements have appeared They are also used, to the exclusion of all others, in the “Report on Proposed TABLE I—ELECTROMAGNETIC QUANTITIES C.G.S. units Volt-ampere units Quantity designated Symbol Equation Name Unit Defining units dë TN 5 seg ek e a e ee oe ee: a —E...... Maxwell........... -10° MIRWEUS: 24% 24 25 e4 we Volt-second Induction or flux density..................00 2 re J BMS 58o. Gauss............0-. 10° gauss................. Volt-sec /em? l 1 Mmf round a circuit 2.2... 0.0... 00. ees f a ere F =4rNI......... Gilbert or oersted-cm. To gilbert or vis ampere-turn......... Amperes Magnetizing force or field strength..... ...... i; ere S B . cos e.ds = F ..Oersted.............. 2: oersted or =472NI 10 amp turn percm.....Amperes per cm T 10° gauss lt- 2 Permeability... ....s esunsun a nno Heas u =., eee Permeability......... ee ee en eg O O H oersted been revised since the meeting, is now proposed for adoption by the British representatives. PRESENT STATUS OF THE C.G.S. MAGNETIC UNITS AS THE RESULT OF I.E.C. AND I.P.U. ACTIONS It will be seen from the preceding 8.U.N. recom- mendations and table, that there is agreement in all essentials between the findings of the I.E.C. and I.P.U. committees, so far as concerns the C.G.S. magnetic units. Both have recommended the convention that the quantities B and H should be taken as different, amp turn/cm mw American Standard Definitions of Electrical Terms,’ sponsored by the A.I.E.E., and recently published (August 1932). Up to the present date (November 1932), no objections seem to have appeared in North American literature to the I.E.C.-I.P.U.-C.G.S. con- ventions, and the only complaints raised have been that the previously existing C.G.S. unit name for re- luctance has been removed, without designating any substitute. As a minor detail of symbology, however, an objection has been raised to the use by the I.E.C. of the symbol uo for space permeability. This symbol June 1933 has been used to some extent in America for ‘‘initial permeability;” 7. e., the “normal permeability when both the magnetising force (H) and the induction (B) are vanishingly small.” If it should be desired to retain the symbol yu, for initial permeability, space permea- bility, or the permeability of vacuum, might be represented by some other symbol, such as w. MATTERS OUTSTANDING FOR CONSIDERATION IN RELATION TO C.G.S. UNITS Up to this point in the paper, the writer has en- deavored to present an unbiased historical picture of what the I.E.C. and I.P.U. have been able to do for clearing international understanding of magnetic uni- tology. In what follows, opinions and suggestions are submitted, in the role of an advocate rather than of a historian. | It is often supposed that the well known relation B = uH gausses, which connects magnetic flux density with permeability and magnetising force, extends only to magnetic media, and has neither theoretical nor practical interest when free space, or non-magnetic media, are considered. Further examination shows, however, that this is too narrow a view to fit the facts. In a vacuum, free space, or non-magnetic media, the equation becomes B, = uH gausses; where B, is the space flux density, and uo the space permeability. In the C.G.S. magnetic system, the numerical value of uo is arbitrarily made equal to unity, and for purposes of mere computation, the equation might just as well be B = H, in which form it sometimes appears in text books and which would justify the view that B and H are physically identical. But in the C.G.S. electric system (commonly called the electrostatic system), the - value of u, necessary to satisfy the equation is uo = 1/c?, where c is the velocity of electromagnetic wave propa- gation, or nearly 3 X 108 meters per second. In any other system of electromagnetic units, a value of u, differing from unity must be used, even for purposes of computation, except in the Maxwell Q.E.S. (quadrant- eleventh gram-second) system. A few cases appear in the following table: TABLE II Values of uo Satisfying the Space Relation Bo = uoH for Various Systems No. System of units Value of po 1. .Classical or Maxwellian C.G.S. magnetic unrational- 1.0 TSO can thaia Gey en oe a ee ae 1/(0.9) X 10-2! 3..M.K.S. (meter-kilogram-second) unrationalised ..... 10-7 4..M.K.S. (meter-kilogram-second) rationalised....... . 4r X 1077 5..0.G-S.S. (em-gram-seven-second) unrationalised .... 10-9 6..C.G-8.S. (cm-gram-seven-second) rationalised...... 4r x 10-9 7..Maxwellian Q.E.S. (quadrant-eleventh-gm-second) unrationalised................ ee ee re eee 1.0 It has been pointed out in several countries that there are several classical C.G.S. magnetic formulas which should include uo, in order to acquire general applica- SYMBOLS, UNITS, AND NOMENCLATURE COMMITTEE 651 tion. One of these is B, = uH, often written B = H for nonmagnetic media. This question of inserting uo in certain classical C.G.S. magnetic formulas where it generically belongs, ought to be considered by the E.M.M.U. Committee, independently of the I.E.C. Convention that u. has dimensions of some kind, other than a mere number. It has been claimed that uo as space permeability, is a fundamental physical constant. The Ausschuss fiir Einheiten und Formelgréssen has named it the “Induction Constant.” QUESTIONS TO CONSIDER IN RELATION TO THE PRACTICAL UNITS It will probably be generally admitted that the difficulties in the way of rectifying the C.G.S. magnetic units internationally are small compared with the difficulties of reaching agreement over the extension of the practical series of units into a complete compre- hensive system. No changes have been made in the classical C.G.S. Magnetic system, which was adopted by the Paris Congress of 1881, endorsed, beyond any question of modification, by the Paris Congress of 1900, accepted by the Electrical Committee of the Inter- national Committee of Weights and Measures and was also endorsed by the S.U.N. Committee at the Paris meeting last July, under item (2) above referred to. On the other hand, the practical units are now nine in number: the ohm, volt, ampere, coulomb, farad, watt, joule, henry and pramaxwell or volt-second. These form a one-to-one series; but not a complete system. We may define a comprehensive practical system as any system of units susceptible of universal scientific measurements and computations, which includes the nine practical units as constituent elements. In order to complete the practical series into a practical system, certain additions must be agreed upon. These missing elements are of great importance. The first elements of the series are now more than fifty years old, and the completed system may last for centuries. Great care is therefore needed in choosing the best new elements; because while the choice is pending, we. can carry on all our work in the C.G.S. system. QUESTION OF PRACTICAL UNIT OF MMF AND SUBRATIONALISATION Probably the first question to consider is the defini- tion of the practical unit of mmf. Shall it be the 1/47th part of an ampere-turn or just an ampere-turn? This is the old rationalisation question first raised by Heaviside in the seventies. He showed that ration- alised C.G.S. units would be much simpler to think with, and work with, than those of the classical C.G.S. system; although the method by which he advocated rationalising, would have meant changing the magni- tudes of nearly all the existing practical units and standards, and was quite unacceptable. Later on, it became known how to subrationalise a system; 7. e., to obtain the advantage of rationalisation through in- 652 creasing the unit of space permeability 47 times, and diminishing the unit of space inductivity 47 times. Karapetoff has recently presented a simple means of analysing any system of electromagnetic units, (setting aside the units of length and mass), in terms of five significant constants: namely, (n) the ratio of electric flux to the charge producing it; (p) the corresponding ratio of magnetic flux to the pole producing it; (k) the ratio of themmf of a coil to the current-turns producing it; (€o) the space inductivity and (uo) the space permea- bility. He proceeds to show that a certain simple relation must hold among these five parameters in any n workable system €o uo È = een E The three parameters n, p, and k are indicated to be arbitrary numerics; while the product e€, uo has the same dimen- sions as the inverse square of c the light velocity. On the basis of this analysis, we may define a system as subrationalised, when the ratios n and p are each equal to unity. If k is also unity, so much the better. In both classical C.G.S. systems (magnetic and electric), as well as in the Q.E.S. system, the ratios n and p are each equal to 47. This makes unit pole produce 4r maxwells, and a corresponding numerical relation be- tween electric charge and flux; whereas the basic idea should be that unit pole is a maxwell, and a unit charge is unit flux. So great is the advantage of this simplifica- tion that, as we all know, a large part of modern electro- magnetic theory is carried on in the Heaviside-Lorentz rationalised C.G.S. system, and not in the classical C.G.S. system. Various writers have indicated the simplification in working formulas brought about by subrationalisation. The books and papers thus far written in systematised practical units (Giorgi, Bennett, Karapetoff) have used n = p = k = 1. The proposition to adopt 4r NI as the mmf ofa coil in the practical system was raised in the E.M.M.U. London meeting of September 1931. At that time, the vote was 4 to 3 against the proposition, (two countries not voting). At the S.U.N. Paris meeting of July 1932, a similar proposition was carried by a vote of 10 to 3; so that the E.M.M.U. was in favor of rationalising, by a small majority; while the S.U.N. was opposed to rationalising, by a large majority. The question should be thoroughly examined in all KENNELLY Transactions A.I.E.E. countries and only finally settled, one way or the other, by a large majority. There is a danger that if the vote is not nearly unanimous, there may be a schism in the practical system, just as there has already been a schism in the C.G.S. system. We should thus face the danger of having a classical practical sect and a rationalised practical sect. The arguments raised against practical rationalisa- tion seem to be weak. One has been that certain well known relations now existing between the C.G.S. mag- netic and C.G.S. electric systems would be lost if rationalisation were adopted. Granting that this is true, there can be no bad result in the case considered, because there will be a practical magnetic system; but no practical electric system. The duality of the present classical C.G.S. system will fortunately disappear from the proposed completed practical system. It has also been urged that the product €o uo which is 1/c? in the C.G.S. system, will have some other value in the prac- tical system, when rationalised. This may be true in the general case; but is not true in either of the pro- posed completed practical systems. See Table III, Nos. 6 and 8. Table III, prepared in accordance with the Karapetoff analysis, brings a number of different systems into comparison. Nos. 1 and 2 are the classical C.G.S. systems. No. 3 is the rationalised Heaviside-Lorentz system, No. 4 is the Maxwell practical system, as drawn up by its author and unrationalised. No. 5 indicates how that system would appear if rationalised. No. 6 is the Giorgi system, in the rationalised form presented by its author; while No. 7 is how it would appear unrationalised. No. 8 is the system for which there are a number of claimants, among them Dellinger, Bennett and Karapetoff. No. 9 is the same in unration- alised form, first suggested by the Committee on Units and Standards of the A.I.E.E. in 1891; although the implications as to the unit of mass (10’ gm) were not then apprehended. It has always been urged against rationalisation that if the factor 47 is taken out of mmf it will have to be inserted in permeance and reluctance in order to main- tain the magnetic flux @ unchanged in magnetic-circuit Ohm’s law. The argument is that no advantage would be gained by the process. It must be remembered, however, that in the great majority of practical mag- netic-circuits, magnetic material such as iron or steel TABLE III L M T No. System cm gm sec n p k €o Mo | nee ene eee Classical C.G.S. electric unrat................2.. ccc eee eee | re L- E | Pr r E 4m... 4r.. 1 .1/c? PEE Classical C.G.S. magnetic unrat........... teeta eg eens aie L... 1 1l.. 4m... Am... 4x.. 1/c? 1 e RO EES Heaviside-Lorentz rat... 0.00000 ee eee 1 A 1 Pee eee L eeu aera Ea 1 1 re ae Maxwell practical Q.E.S. unrat..........0. 00.00.00. cee eee 10?..... 10-1! oe 4q...... 4m... 4r.. 1/c? 1 EREE Maxwell practical Q.E.S. rat......... cece ee ee ee a 108... 107 ces | ee eee D can ces | ee 1 1/4 xc? 4r Go caedeens Giorgi M.K.S. fab o5.9'sd one ado ah urranna ee eee 107..... 10° Seuec2 | EIEE 1 T nas 1 . 107/4 r c?...4 7/10" Toen ORE E E E E E E 10?..... 108 oe oe Agr... 4r Aw... 107/2... 107 Se eae Dell. Benn. Kar. C.G.S.S. rat... 0. cc ees Lee O'n we aatuce s rene 1 eee 1 109/4 r c?...4 7/109 9 CS aod ob cores Fans koe Sess He bake es Sees l See. 107 1l.. 4T... E 4r 5 a 10-° June 1933 has to be employed, and that the relative permeability of this material has to be included in the computation, usually through the use of a table or curve sheet. Since the relative permeability factor, has, therefore, ordi- narily to be inserted, it is easy to include the 47 factor in the table or curve sheet, so as to have no necessity for introducing the 4r term separately. There is thus an economy of time and effort involved in the ration- alised process of magnetic-circuit Ohm’s law. The Ohm’s law formula is, moreover, only one of many in which the 47 factor appears. In nearly all of them the rationalised formula is distinctly simpler than the un- rationalised formula. | A real objection to rationalisation would be the divergence of the practical system from the unration- alised basic C.G.S. magnetic system. The main question to be decided would be whether the very real advantages of rationalisation are counter-balanced by the disadvantage of departure from the corresponding classical C.G.S. formulas. QUESTION OF THE UNIT OF LENGTH IN THE PRACTICAL SYSTEM In addition to, but independent of the question of rationalisation, comes the question of the unit of length to be used. The choice of the length unit logically | determines the units of area and volume. It has been shown by various writers, that in order to conform with energy and power requirements, (the joule and wait), there are only two units within the range of serious consideration; namely, the meter and the centimeter. Theoretically the quadrant is also included; but it is so large as to be out of discussion. The quadrant as a unit of length, exceeds the earth’s radius. The square quadrant, as a unit of area, is about one-fifth of the surface of the globe, and the cubic quadrant, as a unit of volume, is of the same order as the earth’s volume. No one has had the courage to propose these units seriously, and it does not appear that Clark Maxwell, who discovered the relation of the units in this system, regarded them as of a workable character. The centimeter is a more convenient and familiar unit of length than the meter for most electrical engi- neering purposes. On the other hand, however, it is linked with a large and awkward unit of mass: 10’ grams, or ten metric tons. It is a question, therefore, whether in the general use of the practical system, for purposes other than electromagnetic, the disadvantage of this large mass unit might not outweigk the advan- tages of the centimeter. It has been contended that neither the M.K.S. nor the C.G.S.S. system is permissible as a scientific unitary © structure, for the reason that in neither of them is unit density the same as that of water. The unit density in the M.K.S. system is the kilogram per cubic meter, which is 1,000 times less than that of water. The unit density in the C.G.S.S. system is 10 tons per cubic centimeter, which is 10’ times the density of water. It SYMBOLS, UNITS, AND NOMENCLATURE COMMITTEE 653 is of course desirable to have the unit of density iden- tical with that of water, from a practical point of view; but the failure to meet this condition probably is not so serious as some have claimed. For practical pur- poses, what we need is a table of specific gravities, or the densities of substances divided by the density of water. Once such a table has been prepared, the absolute density of water becomes of relatively small importance. It seems doubtful whether any practical system, otherwise acceptable, should be rejected be- cause its unit of density is not that of any familiar substance. Bibliography of Magnetic-Circuit Units Subsequent to that contained in author’s A.J.E.E. paper of January 1931. (Magnetic-Circuit Units Adopted by the I.E.C.) 1. Bennett, Ed., Comprehensive Systems of Units for Universal Use. Trans. A.I.E.E., April 1930. 2. Lombardi, L. “Comitato Consultivo di elettricità annesso al Comitato Internatzionale dei pesi e misure.” L’ Hlettrotecnica A.E.I., Dee. 15, 1930, Vol. 17, No. 35. 3. Mailloux, C. O. “La question du ‘gauss’. La question de la Responsabilité-Les diverses définitions de l'expression ‘flux magnétique’.” Revue Gén. de l’El., Paris, Vol. 29, Jan. 24, 1931, pp. 125-135. 4. Kennelly, A. E., Magnetic-Circuit Units Adopted by the I.E.C., ELECTRICAL Ence., N. Y., Feb. 1931, p. 133. 5. Brylinski, E., “Sur le Système d’ Unités,” Revue Gén. de l’ Elect., Paris, Vol. 29, pp. 209-213, Feb. 7, 1931. 6. Budeaunu, C. I., “La question des Grandeurs et Unités Electriques et Magnétiques,” Inst. Nat. Roumain, No. 11, Feb. 1931, pp. 23. 7. Kennelly, A. E., “The Convention of Equidimensional Electric and Magnetic Units.” Proc. Nat. Ac. Sc., Washington, D. C., Vol. 17, No. 3, pp. 147-153, March 1931. 8. Van de Well, G. J., “De Magnetische Einheden in 1900 te Paris en in 1930 te Stockholm,” De Ingenieur, Holland, 1931, No. 35, pp. 8. 9. Kennelly, A. E., ‘“‘Rationalised versus Unrationalised Practical Electromagnetic Units,’ Proc. Am. Phil. Soc., Vol. 70, No. 2, pp. 1038-119, Apr. 1931. 10. Kennelly, A. E., “The Oersted Considered as a New International Magnetic Unit.” The Scientific Monthly, N. Y., April 1931. 11. Brylinski, E., “Sur le Système M.T.S.V. de Mesures,” Rev. Gén. de l Elec. Vol. 29, pp. 733-740, May 9, 1931. 12. Blondel, André, “Comparaison entre les Systémes prati- ques d’Unités électromagnetiques,”’ Rer. Gén. de VEL, May 16, 1931, Vol. 29, pp. 771-783 and May 23. 1931, pp. 814-825. 13. Brooks, H. B., The Unit of Electrical Resistance. Past History and Impending Change.” Proc. A.J.H.E#., June 1931. 14. Blondel, A., “Remarques sur la subrationalisation des unités pratiques,” Rev. Gén. del’ Elec. Paris, Vol. 29, June 1931. 15. Spooner, Thomas, “C.G.S. Magnetic Units Should Re Standard,” The Elec. Journal, Aug. 1931, pp. 475-476. 16. Mitkevitch V., “Sur les Unités Magnétiques Pratiques.” Proc. Ac. des Sc. de ’ U.S.S.R., Leningrad, Aug. 1931, pp. 1-9. 17. I.E.C. Document R. M. 97. Unconfirmed Minutes of Meeting of E.M.M.U. Committee, Sec. B Av. Com. No. 1 on Nomenclature, held in London, Sept. 18, 1931, pp. 4. 18. Howe, G. W. O., “The Nomenclature of the Fundamental Concepts of Electrical Engineering.” J.H.E. Journal Scottish Centre, Chr’s. Address, Vol. 70, No. 420, Dee. 1931, pp. 54-61. 19. Questionnaire issued on ‘‘Electrical Units,” by the S.U.N. Committee of the I.P.U., Dee. 1931, pp. 2. 20. Kennelly, A. E., “The Present Status of Magnetie-Circuit 654 Units,” Prac. Jap. Inst. March 1932, pp. 197-209. 21. Page, Leigh, “Electromagnetic Equations and Systems of Units,” Physics., Vol. 2, No. 4, pp. 289-302, April 1932. 22. Wennerberg, John, “A Study of Physical Quantities in Mechanical and Electrical Engineering,’ Swedish Elect. Com. 1932, pp. 38. 23. Germani, D., “Choix d'un Système d'unités Electromag- nétiques.” Rev. Gén. del El., Vol. 32, pp. 39-50, July 9, 1932. 24. Andronesen, P., ‘‘Notes sur la question des Grandeurs et Unités électriques et magnétiques.” Com. Elect. Roumain, No. 13, 1932. El. Engrs. Tokyo. Vol. 52, No. 524, 25. Bodea, E., “Un Système d’Unités pratiques doté de 4 Dimensions Fondamentales.” Com. Elect. Roumain, No. 14, 1932, pp. 35. 26. Vasilesco-Karpen, N., “Sur une nouvelle Exposition des Phénomènes Electromagnétiques. Inutilité de la Notion de Masse Magnétique,” Com. Elect. Roumain, No. 15, 1932, pp. 9. 27. Budeanu, C. I., “Sur les Principes Fondamentaux con- cernant la Structure des Systèmes d’Unités et les Définitions des Grandeurs en Electrotechnique.” Comité Elect. Roumain, No. 16, 1932, pp. 16. 28. Karapetoff, V., A General Theory of Systems of Electric and Magnetic Units. A.J.E.H. Trans, Sept. 1932, pp. 715. 29. Kennelly, A. E., Recent Developments in Magnetic Units, ELECTRICAL ENGINEERING, N. Y., May 1932. 30. Kennelly, A. E., “Actions of Electrotechnical Com- mission on Names for Meenstic Units.” A.S.A. Bulletin, N. Y., June 1932, pp. 183-184. 31. Cotton, A., “Les Noms des Unités Magnétiques.” Int. € Elect., Paris, 2nd Sec., July 1932, pp. 7. 32. Joly, Louis, ‘“L’Etat actuel de la Question des Unités Electriques et Magnétiques,” Cong. Int. d Elect., Paris, July 1932, pp. 17. 33. “Report on Proposed American Standard Definitions of Electrical Terms.” Sponsor A.I.E.E. Glossary of 3,300 Terms, pp. 204, Aug. 1932. 34. Kennelly, A. E., “Standardisation of Metric Magnetic Units.” Industrial Standardisation, Nov. 1932, Vol. 3, No. 11, pp. 284-285. Cong. Discussion George A. Campbell: I should like to call attention to the practical results of the conventions set forth in this paper, as shown by a document which the paper says has used these con- ventions to the exclusion of all others. I refer to the “Report on Proposed American Standard Definitions of Electrical Terms”? sponsored by the A.I.E.E., with Doctor Kennelly as chairman of the committee in charge. It was published last August with a request for comments and suggestions. Section 35 on “Units and Systems of Measurement” of the Group 05 on “General Terms” defines 5 ¢.g.s. systems with 27 of the c.g.s. units, next a practical system with 9 practical units, and finally an international system with 10 units. From these definitions the engineer can learn, for example, about the ohm, the abohm, the absolute ohm and the international ohm, but nowhere is he told what ohm is actually used in our shops. The engineer may be impressed with the multiplicity of ¢.g.s. systems but he cannot fail to find them confusing. Even if he is wise enough to skip the first four-fifths of the text and start in with the international system, he finds that every international unit definition refers him, in the end, to the corresponding absolute unit. This first leads him back to the definitions of the practical units, and then on back to the long, involved definitions of the e.g.s. units and ¢.g.s. systems. Would it not be far better, in definitions intended for the electrical engineer, to define first, as briefly and directly as 1. Report No. 2 of A.I.E.E. Standards. KENNELLY Transactions A.I.E.E. possible, the units which actually are used by the machine shop in the manufacture of electrical equipment, by the retail shop in the sale of electrical equipment, and in the shop talk at all A.I.E.E. conventions? The engineer could then get what he really needs without encountering the artificial difficulties which were invented when the leading scientists hardly had the grasp of electrical units which the youthful radio amateur now seems to have. It would, however, not be feasible to omit all reference to ¢.g.s. units because the electrical engineer will unfortunately encounter them in the literature of physics. This calls for the familiar conversion table to enable him, upon encountering any ¢.g.s. unit, to translate it into the shop unit which he really understands. | What I am suggesting is that in this A.I.E.E. glossary of terms relating to units both the order and the emphasis should be inverted. Put the shop ohm, the shop volt, the shop ampere- turn and the shop volt-second first and not the c.g.s. units. There would be no reference to ¢.g.s. units before the table at the end of the glossary. The shop units would be the basie units and the ¢.g.s. units would appear as derived units. If you agree to this, it seems to follow that in his training the engineering student should start at once with the shop units. Then much later he would be introduced to the conversion table for use when, and only when, he encounters a ¢.g.s. unit in the literature. This plan should work perfectly with boys who have become familiar with the shop electric units as radio amateurs and household electricians. If this is the best procedure with students of engineering, it is, I believe, also the best procedure with high school and college students of physies. It seems quite. certain that if a generation of engineers and physicists were trained in this way, they would continue to think and write in terms of shop units, and ¢.g.s. units would soon become of his- torical interest only. This, it seems to me, is a goal devoutly to be worked towards, and the present an opportune time for ener neers to assist in accelerating the process of education. Referring again to the A.I.E.E. electrical definitions: under the “International System of Electric and Magnetic Units,” on page 34, occurs thestatement, ‘‘The International Committee of Weights and Measures has decided to discard these units in the near future” in favor, it is implied, of the absolute practical units, as announced in the technical literature. This also calls for inversion, in my opinion.2 Why? One among several reasons is this: the absolute practical units and all ¢.g.s. units lead to formulas which are in- fested with 47°’s and the 47’s are an unmitigated nuisance in the shop, laboratory and study. The international ohm, am- pere-turn, mechanical watt shop system is necessarily free from these 47’s and such a system is ‘‘much simpler to think with and work with,” as Doctor Kennelly says. I wish there were time to take up Doctor Kennelly’s paper point by point, but in the in- verse order of presentation, and show how this inversion simplifies the subject and automatically eliminates one by one the diffieul- ties which he sets forth. I refer especially to: “. . . the difficulties of reaching agreement over the extension of the practical series of units into a complete, comprehensive system,” including the meter and ampere-turn, because it requires dropping the density unit and the 47 of the c.g.s. electromagnetic system; “. . . the ambiguities and confusion which have pervaded international magnetic literature during the last thirty or forty years.” . certain ambiguities in the definitions of fundamental electric a magnetic quantities.” To speed the adoption of the shop units as the dermal: pre- ferred, universal system, the legalizing acts for metric and elec- tric units, in the United States and other countries, should be 2. I cannot see, from the paper under discussion or from other reports of the deliberations of international committees, that the great advantages of relegating the c.g.s. systems to history in favor of a single, preferred system of units, based on the meter, kilogram, second and shop ohm, have ade- quately been considered by, or even clearly presented to, any of these committees. Many of their votes, therefore. seem to me to lack finality. June 1933 consolidated, revised and amplified. That the required changes are small is illustrated by the accompanying table.’ A DEFINITIVE CHOICE OF MECHANICAL, ELECTRIC, AND MAGNETIC UNITS SUITABLE FOR A LEGALIZED SYSTEM Symbol and . Unit name defining Quantity (bracketed names to be shortened) equation 1 length............... INV sa cckateesa beware a aa L 2 OQCbss ede cd seen ees [square meter] = square meter...... S = alo 3 volume.............. stere = cubic meter............... V =LS 4 055 |: ae ee enano en SOCONG ss ebaee dep eeoe iva eb 6 OES T 5 POWER... anaana Wati 05 2 ond he hp ek er r we P 6 energy...... SIOU Oost nS ie ee Scobie a de ee eae W =PT T POPC 6 oss cewek cbeaws [joule per meter].................. F =WL-1 8 pressure............. [joule per meter per square meter]..p = FS-t 9 mass................ [kilogram]. ccs ccd kas ecwsaaee a iveen M = Fa" 10 density.............. [kilogram] per stere................ d =My-} 11 electricity............ COMMOMID® 24:2.4.24 6 oka s 66 S94 bE 12 electric current....... ampere*..... 0.0.20. .00 cece eee eee I =QT" 13 electric potential...... VOLT eo cae ene pee eos E = Wọ- 14 electric capacity...... PAPA Seah aie de ded anann n 6 dl a etd C =QE-! 15 electricresistance.....0hm..............0 00 cee cee eee R =EI-! 16 electric conductance. .mho................ 0000 cc eee ees G =IE- 17 electric inductance....henry...............00 000 cece £ =RT 18 magnetomotive force. .ampere-turn*...............-..00- Ff = NI 19 magnetic flux......... volt-second*..... a Gos a Rs ened 6 =HT 20 reluctance............[ampere-turn per volt-second]....... R = Fp- a = acceleration in meters per second per second. *The five starred units are to be about one-fiftieth of one per cent smaller than the corresponding international units. The general adoption of these units should make it unnecessary to become familiar with more than one system of units. This would release human effort and soon eliminate the waste and confusion which are a necessary consequence of our present multiplicity of systems. Although the proposed single system of units merely is a selection. and a re-alignment of present units, it is not a patchwork expedient. It would sweep away at one stroke all of the major difficulties set forth in Doctor Kennelly’s paper. Doctor Kennelly follows the international committees in con- sidering that the classical ¢.g.s. magnetic system has been “‘en- dorsed, beyond any question of modification, by the Paris Con- gress of 1900.” The literture of physics actually shows, however, that physicists do not consider the c.g.s. electromagnetic system of units suitable for all their needs. Theoretical physicists have substituted the Heaviside-Lorentz systems. Experimental physicists actually make great use of wattmeters, ammeters and voltmeters, together with the farad, ohm and henry. Funda- mentally, the choice today is between ¢.g.s. authority and free- dom to use the units which actual experience has shown to be “simpler to think with and work with than those of the classical c.g.s.”’ C. L. Dawes: Professor Kennelly’s paper gives a very concise and yet comprehensive presentation of the developments and the present status of the I.E.C. and associated committees in the matter of the standardization of the c.g.s. magnetic units, both in their dimensional relations and in nomenelature. Likewise, the paper states that although practically unanimous agreement has been reached in the retention of eight of the practical units, yet names and even definitions have not as yet been assigned to the other practical units such for example as mmf. Progress has been made in that definite names have been assigned to the c.g.s. magnetice units of flux, flux density, mag- netizing force and magnetomotive force. Moreover, distinction 3. This table is taken from the manuscript of a paper under preparation. The same units have been discussed in ‘“‘Three Superfluous Systems of Electromagnetic Units” Physics, Vol. 3, No. 5, November 1932, pp. 230-239, and “A System of ‘Definitive Units’ Proposed for Universal Use” Science, Vol. 61, April 3, 1925, pp. 353-357. l SYMBOLS, UNITS, AND NOMENCLATURE COMMITTEE 655 between the controversial units of B and H decisively is indicated in their being assigned the respective names of gauss and oersted. Also the troublesome question of the dimensions of permeability u appears to be settled by convention in a very rational manner. However, so far the c.g.s. unit of reluctance is not named; it would seem highly desirable that a name be assigned to this unit, not only to complete the nomenclature of these related units, but also it is convenient to refer to a unit by name. The name itself is not material so long as it is in general harmony with electrical and magnetic nomenclature. It is the appro- priate custom to name the units for savants and the name Weber appears to be the best adapted to this magnetic unit. In the very early days the name Weber was used as the name of the practical unit of quantity and later as the name of the practical unit of current; in more recent years it has been used unofficially in this country as the name of the practical unit of flux, but the names ‘“‘voltsecond” and ‘‘pramaxwell,’’? make the name Weber no longer necessary for the unit of flux. Hence the name Weber appears to be available and could be well recommended as the international unit of reluctance in the ¢c.g.s. system. It also would seem highly desirable that all the remaining magnetic and electrical units of the practical system be defined and named. This system is the most used by engineers and yet there has been little or no attempt to give names and defini- tions to the magnetic and electrostatic units (except henry and farad). In this respect I am in substantial agreement with Doctor Campbell. I believe, however, that the factor 47 should be retained in the practical unit of mmf. This is in accord with proposition (7) adopted by the S.U.N. committee in Paris in 1932. Al- though a practical system should lean towards usefulness rather than toward a rational and fundamental derivation, it is ques- tionable if the factor 477 does produce any more inconvenience when it appears in the numerator than it does when it appears in the denominator either in the physical dimensions of the magnetic circuit or in the permeability of vacuum. The author recommends the ampere-turn as the unit of mmf, stating that in engineering uses of magnetic materials their properties are given as data and curves in which the permeability factor is included, so that the 47 term in the denominator causes no inconvenience. This same condition has existed for a long time even under present conditions where the ¢.g.s. unit of mmf having 47 in the numerator is used. As is well known it has long been customary in scientific work of all kinds to plot the flux density as a function of the ampere-turn per unit length. Hence so far as such data or curves are concerned, it seems im- material whether 477 appears in the numerator or in the denomi- nator. Furthermore all the present units of the practical system have a definite relation to the fundamental ¢.g.s. system, the unit of length being 10° em and the unit of mass 10-" gram. Elimina- tion of the 47 factor in the unit of mmf will break this rational relationship between these two systems, which I believe increases confusion and is thus highly undesirable. With several years’ experience, it appears to me that the so- called confusion which the 47 factor produces, even among students, is exaggerated greatly and is far out of proportion to the efforts being made to eliminate it. The physical picture of a unit charge or unit pole emitting 477 lines in a medium whose absolute permittivity or whose permeability is unity is simplicity in itself. In fact, to my mind this picture is a much more rational one than attempting to visualize a unit charge or a unit pole emitting only a single line and in a medium whose permeability is 477. Moreover, the unit charge and the accompanying 47 lines lend themselves very readily to the very fundamental work and potential relationships which are necessary in the derivations of electrostatic potentials and capacitances. Only very recently I asked two of my classes one of which was composed of students who had only just begun seriously 656 the study of magnetic and electrostatic units, whether or not the 47r lines emanating from unit poles or charges, was at all bothersome. They unanimously stated that this convention was very clear to them, in fact was much clearer than the unit line per pole or unit charge. In fact, it has long been our experience that students who are beginning electrostatic and electro- magnetic studies have become greatly confused in attempting to use text books where the ampere-turn and the unit line per unit charge conventions are used. Hence, it is in accord with my experience that the retention of the 4r term in the practical unit of mmf would simplify the understanding of magnetic relationships rather than confuse them. I am heartily in sympathy with Doctor Camphbell’s ‘‘defini- tive” system and the simplicity which it implies, with the possible exception of the unit of mmf just discussed. However, I do not believe that the two fundamental c.g.s. systems should be discarded or even thrust too far into the background, since their units are derived in a simple manner from very fundamental postulates; moreover, even if the “‘definitive’’ or an equivalent system were legally adopted, I believe that these two funda- mental systems still would be found to be very useful in certain types of scientific work. S. L. Gokhale: The paper divides itself into two parts: 1. A summarized history of the subject up to the present date. 2. Recommendations by the author as to further changes. The first part contains a table of symbols and units (Table I) suggested by the S.U.N. Committee, and recommended by them for adoption. Doctor Kennelly agrees with the opinions of this committee except on the question of rationalization; that is, Doctor Kennelly stands for omission of the factor 47 in the equa- tion F = 47 NI, while the S.U.N. Committee has decided to retain it by a vote of 10 to 3. Table I, column 3, contains what seems to be the defining equations of the concepts in column 1, although they are not specifically called by that name. Construing them as defining equations, we find that their sequence is the reverse of that of the scheme adopted by the I.E.C. and A.S.A. The I.E.C. and A.S.A. first define m the magnetic pole strength. From this they derive H and from H, F and from 3, J. The S.U.N. on the other hand first define J and from this they derive F, and from 5, H. Therefore, the agreement between the I.E.C. and the I.U.P. (International Union of Physics represented by the S.U.N.) as stated by Doctor Kennelly appears to apply to the names of the units and their magnitudes. I hope that it will be possible to work further with a view of arriving at an agreement as to the sequence of the definitions. With reference to the omission of 47 in the equation F = 4r], and its transfer to the equation By = 47H, it may be conceded that a large number of scientists favor the omission on the basis of the dimensional conception of permeability, but the point still is unproved. The practical advantages of omitting 47 are more apparent than real. For instance, in machine design where the magnetic circuit is partly air and partly iron, the 477 cannot be eliminated from the computation, for if it is suppressed in the calculation of the iron part of the path, it will reappear in the air part. In only a few instances, such as design of transformers without leakage, can the 47 effectively be suppressed. There are other cases in which, though the 47 does not enter into the numerical computations, has yet to be ineluded in the new form of the equation, and it is possible that it may be for- gotten, until the old practices are forgotten and new practices formed. For example, in transmission line calculation, engineers would have to use the formula 2ul1 Ar d log ——- (where u = 4T) r KENNELLY Transactions A.I.E.E. in place of the present formula d 2u1 log —— (where u = 1) r without the 47r. The best policy under the present situation, is a policy of compromise, which permits both parties to hold to their re- spective views, without interference from the other party. The decision of the I.E.C. in 1930 was the best possible compromise under the circumstances. It permitted one school of thought to assume the physical identity of B and H and to use the equation B = H for space. This equation, though not obligatory to the opposite school, is permissible for them on the hypothesis that Ho has been given arbitrarily the value of 1. The compromise also permitted followers of the dimensional conception of per- meability to use separate names for the units of B and H. Such a policy, though not obligatory to the non-dimensional scheme is quite permissible. This situation might have continued indefinitely without a conflict between the two schools until the question of dimensional character of permeability and the conse- quent difference in the physical nature of B and H could definitely be settled. An attempt to introduce 47 into the equation B, = 47H at this time would compel the followers of the non- dimensional school to abandon the idea of B and H being physi- cally identical, together with the equation B = H. It seems to me that in proposing the change Doctor Kennelly is now going against the spirit of that compromise, which governed the decision of the I.E.C. in 19380. He admits, however, that the change should not be made on the basis of a narrow margin of votes as in the I.E.C. resolution of 1931, (R.M. 97).4 With reference to the substitution of the meter for the unit of length in the practical system, while there may be theoretical reasons for advocating this, I feel that in practise it only would add further confusion to the present condition by ending in giving us a third system in addition to the present two. Hans Lippelt: During the discussion of the paper the use of “shop units” in preference to scientific units has been advocated. We do have shop units already, and I think this is a good op- portunity to point out that some of them are bad. For illustra- tion I wish to refer to the lb, which is often used in the shop as the unit of the mass. But it is utterly wrong to do so. The unit of the mass is lb-ft-! see?. If that term is too cumbersome for practical use, then let us find a short name for it and a symbol. In going over the whole list of shop units, others also will be discovered which cannot bear scrutiny in the light of logic and accuracy. (For instance secondfeet, secft, is used for cubic feet per second.) IJ think that all such faulty units should be amended to become a true expression of their physical meaning. When short and new names for such units become necessary, exact definitions should accompany them. Such a purging process should, indeed, be extended also to our technical seript. It is customary to write sin-!z* to express an angle œ whose sine is x, a method which is not compatible with the definition of a negative exponent as given in mathe- matics. Under this mathematical definition sin-'z is equivalent to ine x and to nothing else. If internationalism in the language of science and engineering is not objectionable, but striven for, then we might just as well accept the method as used in Europe. 4. Unconfirmed minutes of Meeting Section B on Elec. and Magnetic Units of Comm. No. 1, held in London, Sept. 18, 1931. *The reader may answer that I have myself used this wrong method of writing. See A.I.E.E. Trans., Feb. 1926, p. 408, col. 2, last line. My manuscript was correctly written. I had to change these mathematical terms to the customary form, to render my paper acceptable to the committee. June 1933 They write ares in z for an angle (arc), œŒ, whose sine is z. Ar Sin y or Ar Sin y W. I. Slichter: In this paper there are two matters of very great interest to teachers of electrical engineering because they involve the setting up of two new conventions, different from previous practise and because it is the duty of the teacher to prepare the future generations of engineers to accept and use these conventions. The first is the question of the algebraic sign and the vector representation of reactive power and explicitly: shall inductive reactive power be designated by + j and drawn upward from the horizontal axis in vector diagrams? Either sign or direction ‘may be justified mathematically and scientifically depending upon whether we take current or voltage as our reference vector. for the hyperbolic angle (area), V whose hyper- bolic sine is y. | The paper indicates that the international committee prefers a positive sign and upward direction for inductive reactive power, which means taking current as reference vector. This is probably quite satisfactory to those engaged in communication engineer- ing. But those engaged in power engineering are dealing almost entirely with nominally constant potential circuits with one voltage common to all the circuits of the network and they therefore naturally think of voltage as the reference vector. In this case an inductive load means a lagging current and this is so common that we frequently hear the terms “lagging load” and “‘lagging power factor” where lagging means inductive. Thus these people would naturally prefer the negative sign and the downward vector for inductive power and the converse for anti-inductive or capacitative reactive power. The final choice primarily is one of suiting the habits of the majority of users or that which will involve overcoming the least amount of mental inertia. The other question involves taking the constant 47 or 0.47 out of the magnetomotive force and putting it into the reluctance in the formula for the flux in a magnetie circuit. This again easily is justified scientifically if we ignore the very early de- velopment of the art, but more important, this change would put the law for the magnetic circuit in a form similar to that for the electric circuit and the dielectric circuit and make for uni- formity and therefore for clarity, logic, and ease of teaching. Moreover, this is done already in most practical applications for every one of our magnetization curves of magnetic steels is plotted between flux density and ampere-turns and not gilberts or 0.477 ampere-turns. Here the mu and the 0.47 are made part of the reluctance or reluctivity of the material and the curves give us immediately the ampere-turns which is the quantity of practical interest, and not gilberts. If this change were adopted we would have the three outstanding relations in electrical engineering expressed in a similar manner: NI 1 Qo = — = ——_—_ R 0.4 THA E pl I = P R = 7 E 1 e 8 = -KA A. C. Seletzky: Whether or not + 7 vars or — j vars should be considered as representing inductively reactive power in unspecified alternating current right-angle triangles will probably be decided on the basis of whichever usage predominates in present practise. Both forms may be justified from the view- point of mathematical consistency. In view of the fact that + jz has been internationally accepted for some years to repre- sent inductive reactance, inductively reactive power in the form of /?(j7z) would call for a positive sign, current being considered SYMBOLS, UNITS, AND NOMENCLATURE COMMITTEE 657 the axis of reference. On the other hand if the voltage were used as the reference axis, inductively reactive power in the form, E? (— 7b), would require a negative sign. In spite of the fact that capacity loads are encountered in practise, I have observed that engineers generally think sub- consciously of reactive power as inductive when it is unspecified, by sheer weight of its more frequent occurrence. Since we generally label quantities, which occur more frequently on one side of an arbitrary line of demarcation as positive, the use of the plus sign to represent inductively reactive power would seem to be the more convenient form. An analogous situation is present in the consideration of positive angles as those which are drawn counter-clockwise from the positive axis of abscissas. Invariably when one is demonstrating anything involving an angle in space, one begins by drawing an angle in the first quadrant counter- clockwise from the real axis: it is very probable that this usage grew from manual convenience in drawing angles in this manner and then led to the general acceptance of such angles as positive. It seems that a similar condition obtains in the case of reactive power and that inductively reactive power, from the standpoint of convenience alone, should be considered positive. A. E. Kennelly: Doctor Campbell has well pointed out the advantages of a single system of units over the existing multitude of systems. He advocates extending the present series of nine international practical units (ohm, volt, ampere, ete.) into a com- plete comprehensive and so-called absolute system, embracing the meter, kilogram and second, with no separate electrostatic system, and with the ampere-turn as the official unit of mmf.; 1. €., With the system subrationalised. His “Definitive System,” substantially the same as the Giorgi system, would be a great boon, if it could be used both in physics and in engineering, to the exclusion of all other systems. In detail, there is some ques- tion whether the international ohm should be taken at its existing value; or whether it should be first brought more nearly into agreement with the so-called absolute ohm of 10° abohms; but this is a subsidiary question, outside of immediate discussion. Professor Dawes points out that there is need for a name to be assigned to the c.g.s. magnetic unit of reluctance, now that the oersted has been taken for the unit of H. There are many who agree with him. It is to be hoped that a name for the c.g.s. magnetic unit of reluctance may be internationally agreed upon and adopted. Professor Dawes supports Doctor Campbell’s definitive sys- tem; except that he wants it left unrationalised, like the existing c.g.s. magnetic system. It was Heaviside and Lorentz who showed the marked advantages of rationalisation; so that a large part of electromagnetic theory is today written in the Heavi- side-Lorentz ¢.g.s. system, instead of in the classical ¢.g.s. sys- tem, owing to the simplification thereby effected. There are thus two sects among basic electromagneticians; namely, the e.g.s. classicists and the c.g.s. rationalists. The question will have to be settled internationally, sooner or later, whether the applied electromagneticians, using practical magnetic units, are to be classicists or rationalists. It is not a question of what is right or wrong; but as to the relative balance of advantages. No classicist should be foreed to use rationalised units against his will, and reciprocally. In view of the existing schism in the basic literature, and of the considerable difference of opinion that exists among engineers and technicians in regard to the practical units, it looks at present, as though we might have to recognize the existence of both classical and rational practical units; 2. e., both of the ampere-turn and of the ampere-turn/47, and allow both these sub-systems to continue side by side, until inter- national opinion shall have crystallized in favor of one of them; when the other can be formally discharged. This is virtually the state of affairs today in the basic c.g.s. world literature. Not only are opinions on this question of subrationalisation different among electrotechnicians in different countries; but there appear also to be differences of opinion in each of nearly all countries. 658 Mr. Gokhale points out that we still are unable to decide on scientific grounds, whether space permeability Ho has or has not dimensions. We probably will all agree with him on this. There are good authorities on both sides of this vexed question. Ulti- mately, of course, there can be only one answer. Space permea- bility Mo either is a mere number, or it is not a mere number. The question is of less importance to the engineer than to the physicist; but neither engineers nor physicists can use their symbols, units, or definitions logically, without coming to a con- ventional agreement upon the question. In the present state of our knowledge of electromagnetism, it probably is better to adopt an international convention, even if it should later be proved wrong, than to carry on indefinitely the confusion of the last thirty years. Thus far, we have had an agreement both in the I.E.C. and in the I.P.U. on the convention that Mo has dimensions, and is not a mere numeric. Mr. Gokhale also properly draws attention to the need of standardizing internationally the definitions of our magnetic quantities. It is very desirable to have this done. The defini- KENNELLY Transactions A.I.E.E. tions we still cling to are based upon the classical conceptions of magnetic poles, consisting of magnetic point sources. Most physicists admit the validity of these concepts and of the defi- nitions resulting therefrom; but less artificial and more acceptable definitions might be arrived at on the basis of concepts of mag- netic flux, obtained from the researches of Ampère, Faraday and Helmholtz. Professor Slichter has drawn attention to the desirability of having an international convention as to the interpretation of right-angle power triangles when they do not specify the meaning of lagging power. Should lagging power, or — j power, be taken to mean inductively reactive power, or the reverse? Here again is a need for a convention, and not a decision as to right or wrong. He favors lagging power going along with lagging current. General opinion among engineers seems to agree with him. Mr. Seletzky, however, seems to be of the opposite opinion. There probably is no large majority of opinion on this question; but it should be settled as a convention, for general convenience among engineers, and to avoid confusion.