II EUGENE GARFIELD

I INSTITUTE FOR SCIENTIFIC lNFORMATION@3S01 MARKET ST. Pl+t LADE LPHIA. PA 19104

More Than a Mere Physical Attraction:The 100 Most-Cited Papem from the

Reviews of Modern Physics

Number 26 June 27, 1988

Incelebrating the periodical’s 60tlranniversary, the 100most-citedpapers fmm the Revkws of W&mPhysics are examined. By using ISI@ data from the Science Ci@”on Index” and Journal CitationReports”, the journal’s centralityto the physicsliterature is detailed. Someof the topics coveredincludea historicalsynopsisof thejournal’soriginsandits editors, informationon Nobelistauthors,as well as multipleauthorshipand scientificinstitutionsthat are representedamongthe 100papers.

In October 1928 a letter was sent to 53prominent American physicists from JohnTorrence Tate, then editor of the AmericanPhysical Society’s Physical Review. Tatewanted their views on starting a supplementto the journal, which would focus on reviewpapers in the field of physics. Of the 48 cor-respondents who replied, 46 agreed thatsuch a review publication was necessary, IThose who replied included Percy WilliamsBridgrnan (Nobel Prize, physics, 1946),Edward UMer Condon (codiseoverer of the1926 Franek-Condon principle of molecularspectroscopy), Irving Lat2gmti (NobelPrize, chemistry, 1932), Robert A. Millikan(Nobel Prize, physics, 1923), Robert S.Mulliken (Nobel Prize, chemistry, 1966),and J. Robert Oppenheimer (bat known forhis role in the development of the atomicbomb).

In July 1929 the first issue of the Physics/Review Supplement was launched. In 1930it was renamed the Rm”ewsofMo&m Phys-ics. For nearly 60 years the publication hasbeen one of the best sources of broad-rang-ing review articles ht physics. According toinformation scientist and physicist MichaelMoravcsik:

[Reviews] has quite frequently during itslongexiste~trmst certainlysinceI haveken following it since the 1950s–pub-lished “classic” review articles that were

instrumentedto the development of manyof the present high-interest and high-ac-tivity physics fields. . . . [Reviews] is defi-nitelyone of the top five physics reviewjournals in the world.2

The original concept of what a reviewjournal should be has evolved somewhatsince the founding of Reviews. In the earlyyears, the primary function of the journalwas archival, Fields of physics in whichthere were questions, controversies, or con-flicting theories yet to be tested were con-sidered not “ripe” for review and had noplace in the journal. However, the explo-sion of knowledge and the emergence of newand exciting fields of research in the late1950s and 1960s-too new to be consideredproper for an old-style scholarly review—were powerfid lures that drew both readersand potential writers away from archivalpublications.

The size and readership of Reviews werein serious decline when the present editor,David Pines, took over the editorship of thejournal in 1973. When the journal’s boardmembers first invited him to consider theposition, he refused, saying that he couldhardly be editor of a journal to which he hadjust cancelled his own subscription. But themembers of the board who had invited himconvinced him that this was exactly why hewas needed. J A new policy dealing with. .

207

publishing scholarly reviews was instituted.In an editorial in ApriI 1974, Pines an-nounced that the editorial board wotdd solicit

articles on frontier topics inphysicswhichare intendedto conveyto graduate stu-dents, and to physicistsin other fields,asenseof whythattopicis of greatcurrentinterest,whatprogresshasbeenmadere-cently, and what are its likely futuredirections.4

Today, over 10,000 subscribers (nearly2,600 are library subscriptions) on six con-tinents receive the quarterly journal. 5 Thejournal has published over 2,000 reviewpapers. On occasion, topics of interest toreaders who are not physicists are covered.These have included the safety of light-waternuclear reactors,b the feasibility of theStrategic Defense Irtitiative,T and studiesrelevant to DNA. 8

Journal Editors

in its almost six-decade history the jour-nal has had a number of influential editors.The first, Tate (1929-1941, 1947), headedthe Office of Scientific Research and Devel-opment division on antisubmarine warfareduring World War II. J. William Buchta(1941-1946, 1948-1951), an expert on me-tallic crystals, served on the Advisory Com-mittee on Govemtnent-University Relationsfor the National Science Foundation in1954-1955. Samuel Abraham Goudsmit(1951-1957) was a co&coverer of the spinof the electron in 1925 and, as chief of the%lence Intelligence Mission in Europe, ap-praised German nuclear-weapon capabilitiesin 1944-1945. Condon (1957-1968) workedon the Manhattan atomic bomb project andduring the rnid-1960s worked on a US AirForce project to investigate the UFOphenomenon. Lewis McAdory Branscomb(1968- 1973), an atomic and molecularphysicist, was active in the late 1960s andearly 1970s with CODATA (the Comtnit-tee on Data for Science and Technology),an International Council of %ientilic Unionaagency for worldwide data evaluation.Pines, whose career has inchtded pioneer-

ing work in quantum plasma theory, con-densed matter, neutron stars, and compactX-ray sources, as well as being a past chair-man of the National Academy of Sciences(NAS)-Nuclear Regulatory CommissionCommittee on US-USSR Cooperation inPhysics, has been editor since 1973.

Branscomb is well known in the field ofinformation science and science policy. Hehas indeed had a distinguished career. Hiscum’culum vitae includes a 19-year stint atthe Nationaf Bureau of Standards (NBS),Washington, DC, which included 4 years asdirector (1969- 1972). Other positions in-clude vice-president of research at IBM Cor-poration, membsr of the President’s ScienceAdvisory Committee (1965-1968), andchairman of the Commission on ScholarlyCommunications with the People’s Republicof China (1977-1980) for the NAS. He nowserves as professor of science, technology,and public policy at the John F. KemedySchool of Govemtnent at Harvard Univer-sity, Cambridge, Massachusetts.

In addition to Brarracomb,one of the morecolorful of the editors was Condon. Hebegan his career as a journalist in 1918, re-porting on the development of organized la-bor for the Oakkmd [California] Enquirer.In 1921 he entered college at the Universi-ty of California, Berkeley, and five yearslater earned his doctorate in physics.g

In the early postwar years Condon was amajor influence on, as well as an outspokenadvocate for, the civilian control of atornic-energy development. His outspokennessmade him the target of the House Un-Amer-ican Activities Committee (HUAC) begitt-tting in 1948, leading to several investiga-tions into his security clearance. Americanscientists-in a show of support-voted Con-don president of the American Associationfor the Advancement of Science in 1953. Inlate October 1954 he lost his security clear-ance because of the activities of HUAC andits partisans, among them then Vice-Presi-dent Richard M. Nixon. 10The US Depart-ment of Defense reinstated Condon’s secur-ity clearance in 1965.1I

in 1957 Condon was appointed editor ofReviews, a position that he held for 11years.In the last issue of Reviews for which he was

208

editor, Condon expressed his strongly heldbeliefs on the importance of review journals:

There is a special art to the writing of agod titeratrrrereview paper. I believethatit is not being cultivated as much as itshould be in our graduate schools. Everystudent when he starts on his thesis workought to have to search the literature, readthe papers he so retrieves (ridiculousword!) and prepare a critical summary ofthe present state of the field in which hisresearch is to be done, Later when hereaches ~stdoctoral status or becomes ayoung assistant professor he ought to un-dertake a wider and more sophisticatedre-view of his field. From the best of suchmateriafs would be generated a stream ofpapers which would make the ReviewsC#Modern Physicsa muchmoreusetidjour-rudthanin thepast.This..must be regard-ed as a persortat responsibility to beworked on by every physicist who wouldlike to be regarded as part of the researchteam of the world. Iz

It is hard for me to understand why I wasunaware of Condon’s strong feelings aboutreviews—not only when I discussed the im-portance of reviews in Current Contents@10 years ago, ISbut also in the 1950s whenI first explored their special role in knowl-edge accumulation. I remember well howthis “physics” journal was singled out bythe advisory committee to the Genetics Ci-

tation Index project as being important tothe emerging field of molecular biology.This multidisciplinary aspect of modem ge-netics epitomized the policies for journalcoverage in the Science Citation Index@(SCF ).

Impact and Half-Life Data ConfirmJournal’s Preeminence

Since 1929 Reviews has published about2,000 papers. For this study, we have iden-tified and examined its 100 most-cited pa-pers as reflected in the SCI from 1955 to1986. (See Bibliography.) Their citation fre-quencies range from 289 to 2,506, with amedian of 527—making them all CitationClassicsB. Five percent is indeed a verylarge number of papers to be classified in

Figure 1: Year-by-yeas- cit2rtiom to the four most martpapers in the 100 most-cited Reviews of W&nPhysics papers, 1955-1986 SCF. Articles graphedare: Hohenberg P C. Rev. Mod. Phys. 49:435-79,1977 (522 cites 1955-1986 SCI, 65 cites 1987 SCf);Jsckiw R. Rev. Mod. Php. 49:681-706, 1977 (341cites 1955-1986 SCI, 15 cites 1987 SCf); Kogut J B.Rev. Mod. Phvs, 51:659-713. 1979 (295 cites

1955-1986 SCI, 35 cites 1987 S(X); and&PA. Rev.Mod. Phys. 53:769-806, 1981 (308 cites 1955-1986SCI, 78 cites 1987 SCff.

90-—Le13P A

80 “ .-- Kc.gut J B

70- . . . Hohanberg P C

— . J.cklw R

.

10. !It

oIt

I I I 1 I I I I1

1977 soYear

85

this category. Ordinarily, one wotdd expeetabout 1 percent of a journal’s output toachieve this distinction. However, it is notunusual for review journals to produce abetter-than- average number.

The oldest article in the Bibliography isby Hans A. Bethe, Department of Physics,Cornell University, Ithaca, New York, whowon the Nobel Prize in physics in 1%7. His1937 article, “Nuclear physics. B. Nucleardynamics, theoretical, ” has been cited near-ly 350 times since 1955. (He is, of course,the author of many other Cikztiorr Classics.)

The most recent paper in the list, ‘‘Ex-tended x-ray absorption fme structure-itsstrengths and limitations as a structuraltool, ” is by Patrick A. Lee and colleagues,Bell Laboratories, Murray Hill, New Jersey.It appeared in 1981 and is the only paperin the Bibliography published in the 1980s.Figure 1, a graph of annual citation accu-mulations for the four most recent papersin the list, includes this review article.

The citation-frequency distribution for the655 Reviews articles cited at least 50 timesin the SCI from 1955 to 1986 is shown in

209

Table 1: Citation-fkeqrsesrcy distribution for the 655articles published in the Rew”rrvs of Modem Physicswith 50 or mom citstions, 1955-1986 SCF.

Citation Number of Items Percent ofLevel at Level Total Items

22,500 I 0.21,250-2,499 2 0.4

950-1,249 3 0.5750-949 6 0.9Ws3-749 12 1.845(P599 21 3.2350-449 22 3.4300-349 29 4.4250-299 32 4.92C0249 52 7.9150-199 61 9.3100-149 138 21.1

50-99 276 42.1

Table 1. Since the papers listed are baaedon only 32 years of cumulated SCI data,some older papxs that were mainly citedprior to 1955 did not turn up. (The SC1for1945-1954 will be published later in 1988.)Nevertheless, 24 percent of the papers list-ed were published in the two decades be-fore 1955. Table 2 provides a chronologi-cal breakdown. The peak is between 1960and 1964.

Another way of viewing the long-rangeinfluence of a journal is to examine impactand half-life. Impact is a meaaure of the fre-quency with which the “average article” ina journrd is cited during a specified periodof time. The 1986 impact factor for Reviewsis 27.0, the second highest in the SCI. (TheAnnual Review ofBiochemistry is fist, withan impactof31. 6.) The 1986 impact factor

Table 2: Ctsrrasdogk dMriWks of publkation datesfor the 103 most-cited articles from the Reviews ofMdem Physics, 1955-1986 SCP.

Publkatlou Number ofYear Papers

1935-1939 21940-1944 1194s- 1949 81950-1954 131955-1959 141901964 211965-1969 141970-1974 151975-1979 111980-1985 1

meaattres how often the average article pub-lished in 1984 and 1985 was cited in 1986.

While a journal’s impact is an indicationof its current inflttenee in the literature, itshalf-life reflects the pace of obsolescence.Since Reviews publishes reviews, one wotddexpect continued citationa to its articles overalong period, and thus a relatively long half-life for this journal would not be surprising.

Indeed, in the 1986 Journal Citation Re-ports”, we do find this to be the case.Reviews has a long half-life: both with re-spect to cited and citing papers (> 10 and7.4, respectively)-longer than the cited andciting average for the entire SCI (6.8). Incomparison, Physical Review Letters-ajournal that publishes brief communicationsquickly-had 1986 cited and citing half-livesof 4.4, and the Annual Review ofBiochem-ist~, 5.6 and 3.9 years, respectively. Citinghalf-life refers to the median age of papersthat the review articles cite; cited half-liferefers to the median age of the review paperscited by other journals.

Concerning the half-life data for Reviews,Branscomb expresses the opinion that

Your data (7 to 10 years) suggest thatsomething less than itnmortality but con-siderably more than fleeting notice carsbehad by authoring a fust-class review. Per-haps scientistsshtstddremember that when[Reviews] editors are pleading with themto write. [4

Research Front8

Probably the most remarkable aspect ofthese review papera is the continuing rolethey play in identifying current research-front activity. I cannot help but think of the‘‘Daedalus effect” that Barttch S. Bhtmberg,1976 Nobel Prize winner in physiology ormedicine, discusses in a recent interview iniWE SCIEN77S~. 15 As Bhsrnberg notes,science often raises more questions than itanswers. This is perhaps most evident inNobel-class work and in a critical, provoc-ative review article. So it would not be sur-prising for these classic articles to be citeddecades after publication. In Table 3 wehave listed the research fronts having three

.

210

Table 3 The 19S7 SCPNSCP reaearcfr fronts that include tlrrczor more citing papers pubfished in tk Reviewsof W&m Physics. fTbe relatively low citing threshold is due to the malt numlmr of papers k jounraf publisheseach year. ) Over 400 other research fronts were cited by Reviews in 1987. A= number of Reviews articles citingthe core of each frmrt. B =total number of citing documents. C =total number of core documents.

Number Name ABC

87-0329 Universaf conductance fluctuations, quantum interference in smsfl metattic rings, 4 353 32Abaronov-B&mr oscitfations, and disordered conductors

87-0665 Excited weak vector trosons, particle physics at hadron cotfiders, and standard 4 218 16model

87-0593 Heavy-fermion systems, Anderson Mice rncdel, anisotropic supercorrdncdng state, 3 526 49large-N expansion for difute magnelic atloys, Kondo resonarr-ce

87-1525 Neutrino nscillatiorts, expmential solar density profde, and nonadiatmtic resonant 3 237 2Sconversion

87-1757 Lattice QCD, dynamicsl Wilson fermions, and strdmstic qurmdr.ation simulation of 3 244 22.$4 rheory

or more citing papers published in Reviewsin 1987.

Clearly, even a key journal like Reviewscould not cover in one year all the myriadtopics that physicists investigate. Indeed, itis one of the stated objwtivea of the ZSZAhso~Science@’series to use our objective data,in combination with expert opinion, to helpour own and other review-journaf editorsidentify the areas most in ned of up-to-datereviews.

For it was that special intellectual chal-lenge that my colleagues and I took up over20 years ago-before it was called artificialintelligence-to combine the use of the hu-man brain with the machine to create not justautomatic indexing of the individual docu-ment but also a synthesis of the collediveoutput of the invisible colleges. Word pro-cessing and electronic publishing have rev-olutionized the way in which all journalshave been produced. But the creative use ofthe artificially intelligent machine representsthe real challenge to and opportunity for thereview editor of the futmre. Two essentialingredients in the selection and creation oftimely and relevant topics to review are ourcompact disc read-only-memory (CD-RCXV9)version of the SCZ(SCYCD .?ikiition)16and

the research-front database derived from theSCI each year.

The Five Mo5t-CitedPapers

The 5 most-cited papers from Reviewshave already been identified in our study ofthe 1,000 most-cited articles in the SC1,

1961-1982,17.18 and include 1 by Chan-hsekhar (1943),2 by Roothaan (1951 and1960), as well as a paper by Alder and col-leagues (1956) and 1 by Lane and Thomas:1958). Five other papers in the Bibliogra-phy also appeared in that study-namely,hose by Barnbynek and colleagues (1972),Eteardenand Burr (1%7), Burbidge and col-Ikagues (1957), Kadanoff and colleagues[1%’7),and Kisslinger and Sorensen (1%3).

Chandrasek.har

The most-cited paper in the Bibliography,with just over 2,500 citations, is by Subrah-manyan Chandrasekhar, Laboratory for As-trophysics and Space Research, Universityof Chicago, Illinois. The paper deseribes“Stochastic problems in physics and astron-omy. ” It is now over 40 years old but cort-tinues to be cited ilequently. It receivedabout 100 citations in 1987. (Cbdraaekharwas awarded the 1983 Nobel Prize in phys-ics, an event we covered in a previous es-say. 19)In a letter commenting on his classicpaper, Chandrasekhar states that:

In 1941I becamecmvineedtfrst a proper treatment of stellar encounters must bealong the lines of the theory of Browrtianmotion. WM this application in view, Ibqgan an earnest study of the claasieal pa-pers of Smoluctrowski, Einstein, Plnnck,Lartgevin, and othera. And since I couldfind no extant account deafirtg with aflthese matters coherentlyand systernatieal-ly, I wrote an account for myselfi andwhile writing the account I thought that

211

I was able to generalizeexistingtreatmentsand discover new relations-rdbeit minor.I was not sure whether my account shoufdbe published; but John von Neumann, towhom I showed the account, strongly en-couraged me to do so. And that was howthe paper came into being. And it is sat-isfying to know that the efforts one takesin science to learn for one’s self can alsobe helpful to others.zo

Roothaan

Clemens Carel Johannes Roothaan, De-partment of physics, University of Chicago,wrote “New developments in molecular or-bital theory” in 1951. Since 1955 the paperhas been cited itsover 2,400 papers and con-tinues to be quoted today45 times in 1987.

A graph of the citations to Chandrasek-har’s and Rootttaan’s most-cited papers ap-pears in Figure 2. Roothaan’s other paper,the fifth most cited in the Iist, “Self-con-sistent field theory for open shells of elec-tronic systems, ” has been cited in 1,165 ar-ticles since its publication in 1960.

Alder and Colleagues

The third most-cited paper is a multiau-thored work that was published in 1956.Kurt Alder, Aage Bohr (1975 Nobelist inphysics), Torberr Huus, Ben Mottelson(1975 Nobel laureate, physics), and A. Win-ther, CERN Theoretical Study Division,Geneva, Switzerland, and Institute for The-oretical Physics, University of Copenhagen,Denmark, authored a “Study of nuclearstructure by electromagnetic excitation withaccelerated ions. ” Citations to this paperhave reached 1,320.

1.4ne and Thomas

The fourth most-cited paper, “R-matrixtheory of nuclear reactions, ” is by AnthonyM. Lane, Theoretical Physics Division,Atomic Energy Research Establishment,Harwell, UK, and R.G. Thornas, formerlyof the Los Alarnos Scientific Laboratory,New Mexico. Published in 1958, it has bexmcited in over 1,230 subsequent publications.

F-e 2: Year-by-year citatiorks to the two most<itedRm”ervs of Modsm Physics papers, 1955-1986 SCP,Solid Lbte=Chandrasekhsr S. Rev, Mod Phys.15:1-89, 1943 (2,506 cites 1955-1986 S(Y, 102 cites1987 SCJ). Broken Lbte=RoothsarrCCJ. Rev. Med.Phys, 23 fi9-89, 1951 (2,430 cites 1955-1986 SCI, 45cit& 1987 SCf’).

lEO-.

—Chandrasekhsr S140- . . . . RoothaanCCJ

...120- ...... . .. ;

10o-. .. :

zZ ao-:

ao - . ..

>40.

0 1 lr’’’r T’’’rT—Tf—T1955 60 65 70 75 ao ss

Yell

Nobel Author Information

Studies at 1S1have demonstrated that No-bel Prize winners consistently publish oneor more classic papers. The record for Re-views confirms this. Among the authors ofthese 100 works are 15 Nobelists— 11 fromthe US, 2 from Denmark, and 1 each fromthe People’s Republic of China and the UK.The oldest paper by a laureate in the jour-nal (which is also in the Bibliography) is the1937 piece by Bethe, mentioned earlier. Themost recent paper by a Nobelist is by Ken-neth G. Wilson (physics, 1982), CornellUniversity, which was published in 1975.It has been explicitly cited in over 435 pub-lications. Richard P. Feynman (physics,1%5), who died earlier this year, was anauthor of two of the papers in the Bibliog-raphy. He was also a member of the com-mission that investigated the space shuttleChallenger accident in 1986. It should beadded that Reviews publishes the acceptanceaddress of every Nobel laureate in physics,a practice that was instituted by the currenteditor in 1973.

212

Several readers have written to me tocomplain that our Citation Cfassics selec-tions are biased in favor of older papers. Butthe fact is that, more often than not, the yearof publication is biased agaz”mrtthe selectionof older papers. As the literature grows, thenumber of papers that reach a given thresh-old increases. A modem’ ‘equivalent” to thepapers published by Bethe and others willhave to be cited four or five times as often.Back in 1950, a Citation Classic mighthavebeen cited only 100 times, whereas todaywe fmd thousands of papers at that frequen-cy. It is not unusual that some of the’ ‘goldenoldies” will not be cited even 100 times inthe SC&—notonly because we do not yethave the citation information for the postwaryears, but also bmause wme of these paperswere aff@ed by the “obliteration phenom-enon ~Z?I In o~er words, they became Pm

of the common wisdom and did not requireformal citation each time the phenomenonwas mentioned or discussed.

Multiple Authorship and Age of Papers

In the past 20 years various scholars havenoted a growing trend toward multiple au-

TabSe 4: The ntrmber of authors per paper for tbe 100mostded articles from tbe Rew”ews of ModemPhysics, 1955-1986 SCP.

Number of NumberAutbora of

per Paper Papers

10 18 26 15 14 33 82 331 51

thorship of scientific papers. In the field ofphysics, multiple authorship is becomingmore and more common as scientific proj-ects involve larger and larger numbers ofresearchers. For example, the average num-ber of authors for the most-cited articles inthe physical sciences for 1983 and 1984 was11.322and 8.923 per paper, respectively. Inthis study, however, the average is less thantwo-5 1 papers had one author. Only a fewof the rest had more than five or six authors.(See Table 4.) The average of 1.9 for reviewpapers is significantly lower than the typical

- * y~-bY-Y@ wb~rm OfR~”* O~M*m phy~~ Pspm. Solid ~e=nwk of meat-cited PSKWN.Broken Une=totai number of items tmbtiskd in Reviews. Conference wceedka awxnt fm the high publicationrates in @ years, such as 1958 ~141 items), 1960 (117 items),~ 1964 {425 items).

213

research paper in physics and other fields.This is because the aim of a review is to givea unified overview of a subject, a goal bestaccomplished by one person. For Reviews,the policy is that, generally, papers are so-licited from a single author, who is left freeto choose collaborators if desired. Fre-quently, a theorist wifl choose an experimen-talist, or vice versa, to cover the side of theliterature that he or she is less familiar with.

Publication dates for the majority of thepapers in the Bibliography are clusteredrather heavily into just 15 years, with near-ly half (49) of the articles published between1955 and 1969. A graphic depiction of thedistribution of the 100papers by year againstthe totaf number of papers published an-nually by Reviews appears in Figure 3.

There are 73 unique institutions listed asaffiliations by the authors of the 100 most-cited papers. The top six are, in descendingorder (with the number of occurrences inparentheses): the Murray Hill facility of BellLatmatories (10); the University of Chicago(10); Cornell University (6); the ArgomeNational Laboratory, Larnont, Illinois (5);the NBS (5); and the University of Califor-nia (represented by three casnpuaes-Berke-ley, 3; Irvine, 1; and San Diego, 1). Ac-cording to Branscomb, the fact that theArgome and Bell laboratories are in the topSiX indicates that scientists there believe itis either their moral duty or it is importantto their work to write good reviews. 14Hecomments further that

I believe both are true. Much universitywork is forced into a series of quick re-

connaissance missions, about the size of

a PhD thesis. With a brilJianttiwultymem-ber and a clever student, wondertid thingacan happen. But the great new synthesesof knowledge do not often happen thatway... . At places fike the ~S] scientistsare encouraged to be very thorough andvery careful, ad sel.xt long-range goalsworthy of the effort. A review may playa significant role in such a strategy. 14

Tabte 5: Geqraptdc areas represented by tkeirw.tiNtiOd affllations given by authors in theBibliography, listed in descending order by the nuenberof papers pmdueed (eobunn A). B= number of paperscoauthored with researchers affiliated with institutionsin other Colurtries.c= national locations of imtimionalisted by coauthors.

Geographic Imcationof InstItutiorsa A B c

United States 85 8 Belgium, Canada,FRG, Japan, UK,Switzerland

Illiiois 19NewJersey 15New York 15California 13Massachusetts 7Wasbimton. DC 6cOnne&ut’MarylandPemsylvsniaColoradoOhioOregonTemes~ArizonaGeorgiaIowaMichiganMinnesotaMissouriNebraskaNew MexicoOklahomaRhode IslandTexssWisconsin

UKCanadaDenmarkFRGSwitzerlandJapanAustrrdiaBelgiumFrance

55322221

8 3 AustiirI, US31US3032US31US21USllUKlIUS10

The 73 institutions represented are locatedin 10 countries. (See Table 5.) The US ac-counts for the vast majority of papers-85.However, the gap between it and the twonext most-listed countries-the UK and Cart-ada—is not consistent with population andother factors. About 35 percent of the re-views published today are from non-US au-thors, and the geographical spread is broadertitan that of the 100 most-cited papers, in-cluding good representation from theUSSR.24

214

Co5clnsion

The Reviews of Modern Physics hasclearly played a major role in the develop-ment of modem science. While it wouldseem that fewer experimentalists these daysfind the time and energy to write for reviewjournals, writing reviews continues to be anincreasingly important activity for the sci-

;:3.4.

5,6.

7.

8.

9.10.

11.12.

13,

14,15,16,

17.

18,

19,

20.21,

22,

23,

24.25,

entist with a more global view of his or herdiscipline.zs

*****

My thanks to Peter Pesavento and Erici%rschwell for their help in the prepara-tion of this essay.

‘c[s,19M

REFERENCES

Elklnd E. Peraonrd communication. 12 Februsry 1988.Moravcsik M. Personal communication. 19 May 1988.Pines D. Peramud communication. 17 May 1988.Pbtea D, Gould R W, Hafperirr B I, Landauer R, Fhelpa A V, Setpeter E E, Scldf’fer J &

lkeirnarr S B. Editoria: amrrenew directions. Rev. Mod. Phys 4&235, 1974.Pfnes D. Peraomd commmdration. 18 Msrch 1988.Lewis H W, Brrdrdta R J, Cmtkmmn A W, DQrfan D E, Finkyaon F C, Gauwin R L, Hebel L C,

Keeny S M, Mutter R A, Taylur T B, .%neot G F & von Hfppel F. Rept te the AmericanPhysical Society by the study greup on light-water reactor safety. (Whole issue. )Rev. ~mi. Pfrys. 47( SupfJ,1), 1975. f23 p.

Bluernbergen N, F’atel C K N, Avfaenis P, Clem R G, Herfaberg A, JolmaunT H, Mwahafl T,Milk R B, Morrow W E, Srdpeter E E, Seder A M, SrrUivM J D, WycrntJ C, Yadv A,ZsrreR N, Gtam A J & Hebel L C. Repert m the American Physicrd Seciety of the study group onscience and technology of diremd energy weapmrs. (Whole issue.)Rev. Mod. f%ys. 59(SUPP.3; Part 2), 1987.201 p.

L6wdfn P-O. Protonmmeling in DNA end its hiologicaf implications.Rev. Med Phys. 35;724-32, 1%3.

Morse PM. Edward Ufder Condon, 1902-1974. Rev. Mod. Phys. 47:1-6, 1975.Britthr W E & Odabavt H, eda. Topics ie modem physics: a tribute to Ei&rrd U. Conabn.

Bmdder, CO Colorado Asaeciatcd University Press, 1971. p. xviii.An outaprrkenscientist: Ekfwsrd Ulder Coadon. New York 7imes 8 October 1966. p. 15.Condon E U. The past mrd the fiturc of the Reviews of Modem Physics,

Rsv. Mod. Phys. 4&87&8, 1%8.Garfield E. Proposal for a new profession aeientific reviewer. Essays of rrn information scienrist

Philadelphia:1S1Press, 1980.Vol. 3. p. 84-7.BraneconrbL M. Persend commmdcntimr.16May 1988.BamchBlumberg:scienceon TV. 7YiESCIENTIST2(8):14-5,2 May 1988.GarfIeME. Annemrciug tfre SC1 Compact Disc Edition: CD-ROM gigabyte storage technology, novel

aeftware, muf bibliographic ceopling matre desktep research and discevery a reatity.Current Contents (22):3- 13, 30 May 1988.

—. The articles moat cited in 1%1-1982. 2. Another 100 Citation C&rssics highfight thetechnology of science. Essays of an inforrnakrn scisntist: the awards of scienceand otheressays,Pfriladelphk: 1S1 Press, 1985. Vol. 7. p. 218-27.

—-—--=. The srticles most cited in 1%1-1982. 5. Anetfrer 100 Citation Cfu.ssics and a summary ofthe 300 pspers identified to date. fbid. p. 326-35.

——. Tbe 1983 Nobel Prizes. Part 1. Physics and chemistry awards go m Chandmaekhar,Powler & Taube. Ibid. p. 420-9.

Chandraaekhar S. Peraonsl communication. 12 Jamrury 1977.Garffeld E. The 1983 srticles most cited in 1983 and 1984.2. Physiral sciences.

@r. cit., 1986. Vol. 8. p. 480-%,-—-=. The meat+ited 1984 physical-wiencea ardcles-high+rergy physics dominates.

Ibid., 1988. Vol. 9. p. 366-80.—.-— . The ‘obliteration phenomenon’ in science—snd the advantage of being obliterated!

bid., 1977. Vol. 2. p. 3%8.Frkdmrm K. Peramrrdcommunication. 18 March 1988.Garfteld E. Reviewing review literature. Part 1. Definitions and uses of reviews.

Current Contents fi8):3-6, 4 May 1987.

215

Bibliography: The 100 most-cited articles from the Reviews of Modem Physics based on the 1955-1986 SCF,listed in alphabetic order by first author. Numbers following the bibliographic entry indimte the most recentscImcP research-front specialties for which these are core papers. A= 1955-1986 citstiom. Asterisks (*) indiatcarticles with Cikuion CkISSiC@commentaries. The issue numbers, years, and editions of Curreru Corrtents@inwhich these commentaries appeared are in parentheses.

A

4541,320

474

590

596

877

573

328

643833

334

418

563

568

512

590

946

305

2,506

321

320

289291

310477

356591

684

644

716

377

Bibliographic Data

Ajzenberg F & L+mritsen T. Ener&ylevelsof lightnuclei.V. Rev. Mod. Phys, 27:77-166, 1955,A3dec K, Bohc A, Hrms T, Mottelscm B & Wlnther A. Study of nuclear structure by

electromagnetic excitation with accelerated ions Rev. Mod. Phys. 28:432-542, 1956. 85-0463AllisonS K. Experimental results on charge-changing collisions nf hydrogen snd helium atoms and

ions st kinetic energies above 0.2 kev. Rev. Mod Phys. 30:1137-68, 1958. 87-0633Anderson P W & Weiss P R. Exchange narrowing in paramagnetic resnnance. Rev. Mod. Phys.

25:269-76, 1953.874884“Antlo T, FowlerA B & Stern F. Ekxtronic properties of twodimensionrd systems. Rev. Mod

Phys. 54:437-672, 1982. (26/88/ET&AS; 26188/PC&ES) 86-0924“Bambynek W, CrasemrmcI B, Fink R W, Freurrd H-U, Mark H, Swlfk C D, Price R E & Rao

P V. X-ray fluorescence yields, Auger, and Coster-Xronig transition probabilities. Rev. Mod.Phys, 44:716-813, 1972. (41/81/PC&ES) 85-1663

Barker J A & Henderson D. What is “liquid”? Understarrdirrgthe states of matter. Rev. Mod.Phys. 48:587-671, 1976.87-0798

Battecrnan B W & Cole H. Dynamical diffmction of x rays by Wrfect crystals. Rev. Mod. Phys.36681-717, 1964.86-0287

Bencden J A. X-ray wavelengths. Rev. Mod. Phys. 39:78-124, 1967.85-0145BenrdenJ A & Burr A F. Reevaluation of x-ray atomic energy levels. Rev. Mod. Pfrys.

39:125-42, 1%7. 75-0274Bethe H A. Nuclear physics. B. Nuclear dyrrmnics, theoretical. Rev. Mod. Phys. 9:69-244, 1937.

70-0257Bets H-D. Charge states and cluwgedmrrgirrg cross sections nf fast heavy ions penetrating throu8h

gaseous and snlid media. Rev. Mod. Phys. 44:465-539, 19’72, 87-0382Biedenfmcm L C & R- M E. Theorv of anmdm correlation of nuclear radiations. Rev. Mod..-

Phys. 25:729-77, 1953.Blatt J M & Biederrharn L C. The angular distribution of scattering and reaction crnss srxtions.

Rev. Mod. Phys. 24:258-72, 1952.85-2709Brack M, Dnmgaard J, Jencen A S, Pauli H C, Strutfnsky V M & Wong C Y. Fmmy bills: the

shell-correction apprnach to nuclear sheU effects arrdits applications to the fission prncess. Rev.Mod. Phys. 44:320-405, 1972.87-0514

● Brandow B H. Ltieckluster expimsions for the nuclear nrany-bmiy problem. Rev. Mod. Phys.39:771-828, 1%7. (45181/PC&ES) 87-3162

Burbidge E M, Berrbidge G R, Fowler W A & Hoyle F. Synthesis of the elements in stars. Rev.Mod. Phys. 29:547-650, 1957.87-3527

● Cac’ruthers P. Theory nf thermal conductivity of solids at low temperatures. Rev. Mod. Phys.33:92-138, 1%1 (451801FC&ES) 74-0558

Chanrfrasefdmr S. Stochastic problems in physics rmd astronomy. Rev. Mod. Phys. 15:1-89, 1943.86-3947

Chen S-Y & Taken M. Brccdening and shift of spectral lines due to the preserrce of foreign gases.Rev, Mod. Phys. 29:20-73, 1957.

“Cohen E R & DuMond J W M. Gur knowledge of the fundamental constants of physics andchemistry in 1%5. Rev. Mod. Phys. 37:537-94, 1%5. (35/81 /PCSiES)

Coleman A J. Structure of fenrrion density matrices. Rev. Mod. Phys, 35:668-87, 1%3. 86-0769Cokrnarr B D & Nell W. Foundations nf lima viscndasticity. Rev. Mod. Phys. 33:23949, 1%1.

87-2261De Gemes P G. Boundary effects in su~rconductors. Rev. Mod. Phys. 36:225-37, 1964.87-4789De Swael J J. The @et mndel and its Clebscb-Gordarr coefficients. Rev. Mod. Phys. 35:916-39,

1%3,Dime P A M. Forms nf relativistic dynamics. Rev. Mod. Phys. 21:392-9, 1949. 87-t%3

“EdrnMon C & Ruedenberg K. I@alixed stomic rmd molecular orbitals. Rev. Mod. Phys.35:457-65, 1%3. (9181/FC&ES) 87-5216

Elliot R J, Krumhad J A & Lath P L. The theory and properties of randomly disorderedcrystals arrdrelated physical systems. Rev. Mod. Phys. 46:465-543, 1974. 87-52%

Fano U. Description of states irrquantum ntcchanics by density matrix and operator techniques.Rev. Mod. Phys, 29:74-93, 1957.87-0486

Fano U & Cnoper J W. Spectral dislributinn nf atomic oscillator strengths. Rev. Mod. Phys.40441-507, 1%8, 87-f434

Fano U & Macek J H. fmpact excitation and polarization nf the emitted light, Rev. Mod. Phys,45:553-73. 1973.87-0948

216

A

291584

532

316

635

435

411

417

346

333393

407711

522

381

322

637

341934

321

379

347736

1,147

566357

295

480

331

1,236

347330308

474

337

Bibliographic Data

FerreU R A. Theory of pesitron annihilation in solids. Rev. Mod. Phys. 28:308-37, 1956.Feynman R P. Space-time appmsch to non-relativisticqmumunrrrechanks.Rev.Mod.Phys.

20:367-87,1948.87-0470FisherM E. The renormaliition groupin the theory of critical behavior.Rev.Mod.Phy$.

46:597416, 1974.86-5512Foster J M & Ihys S F. Canonical configurational interaction precedure. Rev. Mod. Phys.

32:3002, 1960.87-5216*Geiflard M K, Lee B W & RawrerJ L. Searchfor charm.Rev. Mod. Phys. 47:277-310, 1975.

(10/81/PC&ES) 83-0120Garcbt J D, Fortner R J & Kavanagh T M. Jnner-shell vacancy production in ion-atom coUisions.

Rev, Mod. Phys. 45:111-77, 1973.86-1679Gemmell D S. Channeling and related effects in the motion of charged particles thrmrgh crystols.

Rev. Mrrrf. Phys. 46; 129-227, 1974.87-0422f-fakea IL Ceopcrative phenomena in systems far from thermaf equilibrium end in nonphysical

systems. Rev. Mod. Phys. 47:67-121, 1975. 87-C091● F%ndltorI J & Wookock W S. Determination of pion-nucleonparametersend phaseshitlsby

dispersion relations. Rev. Mod. Phys. 35:737-87, 1963. (7186iET&AS; 7/861PC&ES)

I-lapper W. Opdwd pumping. Rev. Med. Phys. 44:169-249, 1972.86-6341Held F W, von der Heyde P & Kerllc.k G D. General relativity with spin and tnrsion:

foundations and prospects. Rev. Mod. Phys. 48:393-416, 1976.87-6572Herring C & Nichok M IL ‘flennioeic emission. Rev. Mod. Phys. 21:185-270,1949.87-2169Hofetadter R. Electron wwtcring and nuclear SOUCNE Rev. Mod. Phys. 28:214-54, 1956.

86-6617Hohartbwg P C & Hefpmin B L Theory of dymmdc criticrd phenomena. Rev. Mod. Phys.

49435-79, 1977, S65758Hofbmder J M, Perbnsn 1 & Se@mqj G T. Table of isetopes. Rev. Mod Phys. 25:469-651,

1953.I@s D R. The energy levels and the structure of tight nuclei. Rev. Mod. Phys, 25:390-450,

1953.“Irtokuti M. Inelastic colhsions of fast charged particles with atoms and molecules-the Bethc

theory revisited. Rev. Mod. Phys. 43:297-347, 1971. (42/80/PC&ES) 87-2502Jscidw R. Quantum meaning of classical field theory. Rev. Mod. Phys 49:681-706, 1977. 87-W385Kdanoff L P, Gotza W, Homblen D, Hecht R, Lewfs E AS, Pakimrskas V V, Rayl M, Switl

J, Aspnes D & Karre J. Static phenomena near critical points: theory and experiment. Rsv.Mod. Phys. 39:395-431, 1%7. gl-1960

Katz L & Penfold A S. Range-energy relations for electrons and the deterrrdoation of bets-rayend-point energies by absorption. Rev. Morf. Phys. 24:28-44, 1952, 87-1087

Kieffer L J & Dunn G H. Electron impact ionization crnas-section data for atoms, atemic ions,and diatomic molecules: 1. Experimentaldata. Rev. Mod. P/rys. 38:1-35, 1966.87-7165

Kfhara T. Virial coefficients and models of molccuies in gases. Rev. Mod. Phys. 25:831-43, 1953.Kfrkpetrkk S. Percolation end conduction. Rev. Mod. Phys. 45:574-88, 1973. 87-&$79Klsslbrger L S & Sorensen R A. Spherical nuclei with simple residurd forces, Rev. Mod Phys.

35:853-915, 1%3. S6-7141KMef C. Physical tbewy of ferromagnetic domains. Rev. Mod. Phys. 21:541-83, 1949.874019Keeh H W & Motz J W. Bremsstraldungcross-sectionformulasand relateddata. Rev. Mcmf.

Phys. 31:920-55, 1959.86-2472Kogut J B. An introduction te lattice gauge theory and spin systems. Rev. Mod. Phys. 5 I :659-713,

1979.86-1689Kokre W & RwtbmM C C J. Accurste electronic wave thctions for the H2 molecule. Rev, Mod.

phy$. 32:219-32, 1960.76-1394Lomb G L. Arrafytical descriptions of ultrashort optical puls-s propagation in a reamrant medium.

Rev. Mod. Phys. 4399-124, 1971.87-7353Lnne A M & Thomas R G. R-matrix theory of nuclear rcoctiens. Rev. Mod. Phys, 30:257-353,

1958. 87-6S19Lax M. Multiple scattering of waves. Rev. Mod. Phys. 23:2W-3 10, 1951. g7-5662Lax M. IWctuations tlom the rmncquifibrimrrsteady state. Rev. Mod. Phys, 32:25-64, 1960.Lea PA, Citrin P H, E?@nbergcr P & Kfndrf B M. Extended x-ray absorption fine

structure-its strengths and Ondtetions es a structural tool. Rev. Mod. Phys. 53:769-806, 1981.87-2145

Leggetf A J. A theoretical description of the new pfrases of liquid %fe. Rev. Mod. Phy$.47:331-414, 1975, 87~31

Lin C C & Swnlen J D. Internal rotation and micrewave spectroscopy. Rev. Med. Phys.31:841-92, 1959.87-3519

217

A

597

555

435417

319

310

526

726

391

3U2401

2,430

1,165

377

650

438

333318384

643932

446611

485

365

302

326

325

339833

368

317

437

320

BibfingraplrfcData

MacFarlarre M H & French J B. Stripping reactions and the structure of light and intcmrediatcnuclei. Rev. Mod. Phys. 32:567-691, 1960,

“Mandel L & Wolf E. Coherence properties of nptical fields. Rev. Med. Phys. 37:231-87, 1%5.(47/801FC&ES) 86-2023

Mergenau H. Ven der Wads forces. Rev. Mc-4. P/Iys, 11:1-35, 1939.86-7473McWeeny R. SOW rrxent advances in density matrix theory. Rev. Mod. Phys. 32:335.69, 1W3.

86-0769“Mormtaln R D. Spectraf distribution of scattered light in a simple fluid. Rev. Mod. Phys.

38:205-14, 1966. (48/80/PC&ES) 84-7258Newton T D & Wigner E P. Lncalii states for elementary systems. Rev. Med. Phys. 21:40-6,

1949.87-7590Niafasm H H. The vibration-rotation ener8ics of molecules. Rev. Mod. Phys. 23:90-136, 1951,

87-0728“PhRffpa J C. Iorrkity of the chemical bnnd in crystals. Rev. Mod. Phys, 42:317-56, 1970,

( l/81/PC&ES) 87-6762“Pdfnck G L. The sofid state of t-aregaacs. Rev. Mod Phys. 36:748-91, 1964. (40/8 I/PC&ES)

75-0983Primekoff H. ‘fbeory of muon capture. Rev, Mod. Phys. 31:802-22, 1959, 85-2036

● Ranaif B J. Studies in mnlecular sh’ucture. U. LCAO-MO-SCF wave functions fnr aclected fkst-row diatomic molecules. Rev. Mod. Phys. 32:245-54, 1960. (35/84/ET&AS; 35/84/PC&ES)74-1404

Rnothaan C C J. New developments in nrdecolar orbital thcmy, Rev. Mod. FYtys. 23:69-89,1951, 87-8095

Rtmttm$m C C J. Self-cmraiatcnt tield tbexny for open shells of electronic syatcma. Rev. Mcwf.Phys. 32:179-85, 19W, 87-8W5

Rnnthaan C C J, Sacfra L M & Wefsa A W. Analytical self-consistent field functions for theatomic configurations ld, 1#2.c, and 1s2’2s2.Rev. Mod. Phys. 32:186-94, 1960.

Rose H J & Brink D M. Angular distributiorra of gamma rays in terms of phaae-detined reducedmatrix elenrenta. Rev. Mod. Phys. 39:306-47, 1%7. 78-1288

Rosmtfeld A H, Bttrbaro-Gaftierf A, Podofeky W J, Prfce L R, Sorfing P, Wnfd C G, RenaM& Wiflfa W J. Data on particles and reannant states. Rev. Mod. Phys. 39:1-51, 1%7.

Ruedertberg K. The physicnf rramreof the chemical bond. Rev. Mod. Phys. 34:326-76, )%2.Schufz G J. Reamrances in electron impact on atoms. Rev. Mod. Phys. 45:378-422, 1973.87-3946SchUkt G J. Resonances in electron impact on diatomic molecules. Rev. Mod Phys. 45:423-86,

1973.87-3946SeItz F. Color centers in alkah bahde crystals. II. Rev. Mod. Phy$. 26:7-94, 1954.86-3067Strnrrdnger D, Hollmtder J M & .$eahorg G T. Table of isotopes. Rev. Mod. Phys. 30:585-97,

1958.Stress H E & Urey H C. Abundances of the elements. Rev. Mini. Phys, 28:53-74, 1956,Tanaura T. Anrdyses of the scattering of nuclear particles by collective nuclei in terms nf the

coupled-channel cafcrdation. Rev. Mod. Phys. 37:679-708, 1965. 87-8151“Taylnr B N, Parker W H & Lrmgrmberg D N. Determination of e/h, using macroscopic quantum

phase coherence in supercrmductnrs: implications for quantum elcdrodynamics and thefundarnentaf physical constants. Rev, Mod. Phys. 41 :3754%, 1969. (38/87/ET&AS;38/87/PWES) 87-4221

“Tayfnr R L & Bittermarr S. Survey of vibrational relaxation data for prncesaea impnrtant in theC02-N2 laser system. Rev. Mod. Phys. 41:26-47, 1%9. (40/80/PC&ES) 85-7788

VarI Hove L. High-energy collisions of strongly interacting particles. Rev. Mod. Phys. 36:655-65,1964.86-0771

Van Vleck J H & Weiaakopf V F. On the shape of colliiion-brnadened li. Rsv. Mod. Phys,17:227-36, 1945.

Van Vfeck J H. ‘TIRcoupfing of angular momentum veztors in molecules, Rev. Mod. Phys.23:213-27, 1951.70-1186

Varnra C M. Mixed-valencecompnunda.Rev, Mod. Phys. 48:219-38, 1976, 85-7305Wring M C & Uhfenbeck G E. t% the theory of the Brnwnian mntion fI. Rev. Mud. Plrys.

17:323-42, 1945, 86-3947Wbeatley J C. Experimental properties of superfluid ‘He. Rev. Mod. Phys. 47:415-70, 1975.

86-0707Wheeler J A & FeyarmaaaR P. fnteractinn with the absnrbcr as the mechenism of radiation. Rev,

Mod. Phys. 17:157-81, 1945. 86-84%Wilson K G. The renormalization grnup: criticaf phenomena ad the Korvdo problem. Rev. Mod.

Phys. 47:773-840, 1975.87-0593Yarw C N. Conceot of offdiagorraf Ionmarrae order and the muurtum Dhaaes of liauid He and of

su~rconductors ~ Rev. MM. >hys. 34:-@4-%4, 1%2. 70-102> “

218