Making medicine a business in Japan : Shimadzu Co . and the diffusion of radiology ( 1900 – 1960 ) *

This contribution focuses on the role of the firm Shimadzu in the marketing of X-ray machines in Japan during the first part of the 20th century, viewed from a business history perspective. It attempts to further understanding of the process of technology diffusion in medicine. In a global market controlled by American and German multinational enterprises, Japan appears to have been a particular country, where a domestic independent firm, Shimadzu, succeeded in establishing itself as a competitive company. This success is the result of a strategy based on both the internalisation of technological capabilities (recruitment of university graduate engineers, subcontracting of research and development activities) and an original communication policy towards the medical world.Finally, the specific structure of the Japanese medical market, composed of numerous and largely privatised small healthcare centres, facilitated the rapid diffusion of X-ray machines, a new technology which conferred a comparative advantage on its holders.


Introduction
Radiologyplayed akey roleinthe processwhich made medicine amoneymaking business duringthe firstpart of the20th century 1 .Onthe one hand, it contributed, together withdrugs produced by thepharmaceuticalindustry and the firstmedical laboratories,tomaking medicine an innovation-based andsteadily growing economical activity 2 .Onthe otherhand, the particular importanceofradiology lies in thefact that it is an infrastructure with amajor impact on the organisation and managementofhospitals.Research carried out, especially on American and Swiss cases,highlights the essential role of radiology equipmenti nt he transformation of hospitals into organisations managedl ike modern enterprises and theiri nsertion in the heart of the healthcare market 3 .
In thecase of Japan, this change took place in aspecificcontext -thatof the transformation of the domestic healthcare system as aresult of the spread of Western medicine 4 .When the firstX-raymachinesappeared in Japanese hospitals during the 1900s, the transition of the medical system to the Western modelw as well under way.T he opening of the Faculty of Medicine of Tokyo University (1877),the founding of agrowing numberofhospitals (63 in 1875 and 783 in 1900) 5 ,aswell as theadoption of alaw on the practice of medicine which legalised theh ierarchy betweent he differentk inds of practitioners (1907) 6 madei tp ossible for the Western model, essentially German, to take root in Meiji Japan.To date,only alimitednumberofworks have been devoted to thetransformation of theJ apanese healthcare system tackled fromatechnologyhistoryand business historyperspective.The question of drugs,animportant partofthe business of medicine,has been covered by scholars focusing on theshift from the production of traditionalmedicine towards Western industriallymanufactured products andthe permanence of Japanese pharmaceutical enterpriseswithin this process 7 .Asfor the problem of infrastructurea nd equipments,ith ardlyc aughtt he attention of researchers,except the worksofIkai on the structureofthe hospitalmarket 8 and of Tsukisawa on theintroduction of German medical instruments into Japan during the 1870s 9 .
Theo bjective of this paperi st oa nalyse,throught he example of X-ray machines, the way new Western medical technologieswere introduced,diffused and developed in Japan.As with drugs, therea re af ew rare cases of 2B lume 1992;L öwy 1993;S tanton 1999;P ickstone 2000;S tanton 2002 (particularly her introduction, 1-18).3R osner1982; Howell 1995; Donzé 2007.4S hinmura 2006.5 Nihon choki tokei 2006,vol.5, 170.6T his law (ishiho) classified the different kinds of doctorsi nto threec ategories,w ith the one holding aJ apanese doctorate degree (hakasego) or af oreign PhD on the top of the hierarchy,and doctorspracticingtraditional medicine at thebottom.Shinmura 2006,255.7Y ongue 2005;Yamashita2010.8I kai 2010.9T sukisawa 2009.continuity from the Edo period onto the 20th century,with the shift from the production of traditional medicines to them anufacturing of medical instruments, including X-ray machines,ascan be seen fromthe example of the Osaka-based companyShiraimatsu.But theseexamplesare too uncommon to be representativeofageneral trend 10 .Rather,new enterprises appear to havebeen the main producers of these machines whichchanged modern medicine,especially American (General Electric,WestinghouseElectric) and German (Siemens,AEG) multinational enterprises that very soon dominated the worldmarket,including Japan.Yetbesidetheseglobalfirms,there were also several independent domesticfirms that contributed to the diffusion of X-raymachines and radiology in thecountry,among whichthe biggest was ShimadzuCorporation, the subject of this contribution.Thefocus is on the strategies adopted by this enterprise in order to establishitself alongside the powerful foreignmultinationals as akey actorofmedicine business in Japan during the first part of the20th century.

The beginnings(1900-1914)
Thec ompany ShimadzuC orporation (Shimadzus eisakujo) was founded at Kyoto in 1875 by ShimadzuGenjo(1839-1894) as aworkshop specialised in the manufactureo fs cientifici nstruments primarily meantf or schools11 .This industrial activity resultedfrom achangeineducation policy, with the spread of aW estern-stylee ducational systemt hatl ed to newn eeds.For launching outinto this business,Shimadzubenefittedfromtraditionalcrafts know-how,a sh is fathero wned aw orkshop for manufacturing Buddhist worship objectswhereheworkedfor severalyears,onthe one hand, andfrom the support of achemical and physics research institute (Kyoto seimikyoku) setupin1870 by the authorities of Kyoto prefecture12 .This external backing, particularlythroughaGerman engineer who taughtthere,enabledShimadzu to embark upon the production of aw hole serieso fn ew instruments and devices for schools.
Nevertheless,i tw as especiallyt he second generation homonym owner, ShimadzuGenzo II (1869-1951), who madethe family workshop an industrialf irmd iversifiedi nn ew sectors,i ncluding medicine.T wo years after havingtaken over themanagement of thefirm following his father's death, he took the first X-rayinJapan ( 1896) 13 .This wasmadepossible thanks to his collaboration with aphysicist, Muraoka Hanichi (1853-1929).Between 1878 and1881, Muraoka studied physics at theUniversity of Strasbourg, where he obtainedaPhD and where he met with Wilhelm Roentgen, assistantprofessor in the university at the time.After returning to Japan, Muraoka held several academic positions,including that of physics teacheratK yoto Third HighS chool from 1893 onwards 14 .The realisation of this first X-ray image was most certainlyboth ascientificand apromotionalsuccess forShimadzu.Shortly thereafter,hedeveloped afirstX-raymachine for teaching (1897) 15 .Subsequently,thesedevices were designed for the traditionalmarket of the firm-schools-for physics teaching 16 .Ittook some ten more yearstoshift from as cientific device to am edical appliance.Indeed,itw as not until the end of the 1900s that Shimadzu produced itsfirst X-ray machinesfor medical practice.First, German companies,mainly Siemens 17 ,enjoyed av irtual monopoly in terms of X-ray machines forhospitals.Despite this stiff competition, Shimadzudeveloped an X-ray machinefor medical purposes in 1908 18 , subsequently deliverings uch devicest oK onodaiM ilitary Hospital ( 1909)  and Otsu Japanese Red Cross Hospital(1911)  19 .Inaddition, the first private hospitals ordered similar appliances in 1910 20 .H owever,S himadzut ruly establisheditself on this newmarket duringWorld WarI.
Next,o ne should mention am ajort echnological innovation realised in the UnitedStates, in the laboratory of themultinational enterpriseGeneral Electric Co (GE): the developmentofthe CoolidgeX-ray tube (1913) 23 .The particularities of this tube (high vacuum; tungsten spiral filaments for heating) allowed for easy controlofthe dosage of rays by making their volume (current) independent from their penetrative power( voltage).This breakthrough made possible an excellent application of radiologytomedicine and facilitated its speedyd iffusion.In Japan, this technology wasc ontrolledb y TokyoE lectricC o, aG E-affiliate founded in 1905.It hadamonopoly on imports and sales of Coolidget ubes in Japan, thent heir manufacture from 1920 onwards24 .Yet up untilthe early 1930s, GE did not producecomplete X-ray appliances butonly tubes,which were sold to appliance makers.However,the Americanmultinationalwas notthe sole tube producer, even if its tubeswere of top quality.GE'smain rivalswere German companies, as well as several Japanese manufacturersproducing lower-quality copies(Kinzawa Medical Appliances Co,S hibuya Roentgen Co) 25 .S himadzua pproached TokyoElectricand usedCoolidge tubes for its machines from 1917 onwards26 .Thetwo firmseven signed an agreementin1922, therebyensuring the supply for Shimadzu27 .This situation lasted until the beginning of the 1930s.As the patent for Coolidgetubes expiredin1934,GE andTokyoElectricplanned at the time to diversifyinto the production of complete X-ray equipment with the launchin1932 of the Giba 75 appliance,the mainstay of TokyoElectric's competitiveness on this market28 .Asfor tube production, many newcomers enteredthe market after1934 29 .
Accordingly, for somet wentyy ears,from WorldW ar It ot he beginning of the 1930s,Shimadzutook advantage of favourable conditions to establish itself as an X-raymachine producer in Japan.Theabsence of German competitors, followed by the collaborationwith GE, boostedthe firm'sgrowth.However,thereisnoinformation on eitherthe production volume of X-ray machines or the share of this division of total gross company sales until 1935, ayear in which the valueofX-rayappliances sold amounted to 1.7million yen, thati s1 9.2% of gross sales 30 .S upported by the Mitsubishi zaibatsu capital, Shimadzuw as an enterprise that diversified extensivelyd uring the interwar period (machines,appliances,instruments,etc.),postingvery high growthafterits transformation into ajoint stock company (1917) 31 .Its total gross sales rosefrom 2.8 millionyen in 1920 to 8.9 million yen in 1935, while the numbero fe mployees went from 428t o1 318 persons over thes ame period 32 .The firm'sexpansionwas essentially driven by the domestic market andcolonies,with the opening of sales subsidiaries at Dairen (1920), Seoul  (1924)  and Taipei(1931) 33 .The volume and natureofexports are not known, but Shimadzud id export some X-raye quipment to East Asia andL atin America during theinterwaryears 34 .
Thus,during the interwarp eriod, Shimadzu succeeded in establishing itselfa longside foreign multinationals as ak ey actoro nt he X-ray machine market.This wasn ot only the result of particularc onditions conducivet o the firm's growth.Its success also stemsfrom an expansion strategy that combined the internalisation of technological resources with an intensecommunication policytargeting medical circles.

Shimadzu'sR&D policy
In ordert od evelopt he production of X-raym achinest hatc ould compete with appliances imported fromG ermanya nd the United States, Shimadzu was obliged to internalise certaint echnological skills,aprocess essentially based on collaborationw ith external research centres,p articularlyK yoto University.B yh iring engineers who were graduates of technicalc olleges and universitiesand subcontracting some research activitiestothese centres, Shimadzuw as ablet ob enefit froma dvancedt echnologiest hatm adei t possible to develop new kinds of X-raye quipment.This internalisation of scientific and technical knowledge paido ff in the early1 930s,whenT okyo Electricjumped into the production of complete X-ray devices.
In the 1920s, Shimadzub eefed up itso rganisationalR &D capability by hiring engineers in chargeo fd eveloping new X-ray devices.The first such engineer, Fukuda Shunichi, was aprofessor of physics at Kumamoto Indus-trialH ighS chool (Kumamoto koto kogyo gakko) when he was poached (1920) 38 .Upuntil the time of his death in 1930,hesupervised the set-up of new equipments,especially the launching in 1922 of the firstr adiotherapy machine built in Japan (Jupiter model) 39 ,a sw ella sad evice designed for dentists (1928).P roductionv olume wasr amped up from some 100X -ray machines in 1924 to 250in1926 40 .Yet despitethis growth, Shimadzurelied on the outside for some specialp arts.F or example,itd id not producea ny X-ray tubes, importedfrom theUnited States(GE) and Germany (Müller) 41 .However,the 1922 agreement signed with Tokyo Electricguaranteedasupply of Coolidgetubes 42 .Inaddition, Shimadzudependedonforeign firms for others pecific parts.During his business trip to Europe in 1920,Shimadzu Tsunesaburo,brothert ot he president and membero ft he firm'sb oard of directors,signeda greements witht wo producers of partsf or X-ray equipment, CAF-Kahlbaum AG (Germany) forX -rays creens, andG ebrüder Miller GmbH (Austria) forX-rayspectrometers 43 .Underthese conditions,it is possible that the subsidiaryo peneda tB erlini n1 922, whichr emained operational until1 932, was tasked witho btaining some parts or technical assistancefor the production of X-raymachines fromGerman companies 44 .In particular,Shimadzuhad business dealings withthe multinationalSiemens &Halske, whose X-ray counters are imported to Japan at the end of the 1920s 45 .1921-1930, vol.2,255-256;Goto 1970, 118.38 Shimadzu 1967, 45.  39 Shimadzu 1967, 354.40 Shimadzu 1967, 356.41 Shimadzu 1967, 378.42 Shimadzu 1967, 45.  43 Shimadzu 1967, 47.  44 Shimadzu 1967, 48.45 Goto 1969, 22.  Thus,u ntil 1930 Shimadzuw as an assemblero fX -ray equipments,h ighly developed of course,but dependent on the outside for technology.Whati s more,the company obtainedonly four patents in the field of radiologyduring the 1920s,that is,7.0% of thetotal amount of patentsregistered in Japanin this field 46 .In the early1930s,the market became very competitive whenTokyoElectric embarkedu pon the production of complete X-ray machines and Siemens embarkedupon ajoint venture to manufacturetheminJapan (Goto Fuundo  Manufacturing Co,1933)  47 .Shimadzureacted by strengthening its organisationalcapabilities.Afterthe death of Fukuda in 1930,the managementofthe X-ray machine division was entrusted to agraduate of the Faculty of Sciences of Tohoku University,S ekito Kikuchi48 .A fter completing his studies, he  As ar esulto fS himadzu'st echnological development in the first part of the 1930s,the firm obtainedag rowing number of patents:itr egistered 23 during this decade, that is,21.7% of all thep atents registered in Japan during thatperiod for X-ray equipment.Shimadzu became one of the main players in this newindustry.As for the products marketed,Shimadzuappears to have been very active here as well,launching 19 newappliances during the 1930s 56 .

Making its machines known
Together with the developmentofits technological competences,Shimadzu did adopt in the interwar period an active strategy of supporting the dissemination of its X-raym achines withint he medical world.This strategy was 49 Goto 1970, 114.50 Goto 1970, 312.51 Yamada 2005,2.52 Shimadzu 1967, 79.53 Dainihonhakushi 1921-1930, vol.5,108f.54 Shimadzu 1967, 378.In fact,the tubes of the firm Müllerwere madeatthat time by the Dutch company Philips,towhich Müllerbelonged since 1927.Kees 2003, 82.55 Shimadzu 1967, 80.  56 Shimadzu 1967, 79.  at the crossroadso fs cientificc ommunication related to the utilisation and usefulnessofX-raymachines and apure advertisement for its own products.Thedual dimension of this strategy canbeexplainedbythe fact that X-ray equipment was new and it was necessary to explain its properuse,onthe one hand, and because themarket was extremely competitive,with the presence of multinationalf irms sucha sS iemens and GE, on the other hand.The objective of this policyw as thus twofold:encouraging the consumption of X-ray equipment, and if possible Shimadzu-made material.
Thee arlyc ommunication policyw as initially consistent with ac ompany producing scientific instruments foreducation, that is,Shimadzu'straditional market.Until 1908,the X-raym achines madeb yS himadzuw ere destined for schoolsrather than doctors.Siemens' quasi-monopolistic domination of this market in Japan doubtlessdelayed Shimadzu's diversification into anew market,the medical sector.F or some ten years, Shimadzuv iewed its X-ray machinesasscientificinstruments suitedtophysics classes.Thus,soon after producing the first X-raye quipment for teaching (1897), ShimadzuG enzo made public presentationsa tc onferences in the Kansai region (Kyoto, Osaka, Kobe) 57 .In1908, at aK yotonationwide meeting of secondary school directors,hedemonstrated theworking of his appliances 58 .After WorldWar I, when hisX-raymachines becameinstruments designedfor medical practice, Shimadzup ursued and developedt his communication policy, which consisted of threemain thrusts: establishing networks in the medical world, organising scientific eventsand setting up schools for radiologytechnicians.
First of all, Shimadzuestablished ties withmedical circles throughdoctors.This approach appears as ak ey element of thee xpansion strategy towards the medical field adopted by thec ompany in the 1910s.When Shimadzu was turned into ajoint stockcompany in 1917,ShimadzuGenzo comprised 42 persons,ofwhom 12 were doctors(see table 1).This proximity between doctors and engineers may bring to mind collaboration aimed at thedevelopment of newm edical devices,a sc ould be seen in Europe in the 1950s and 1960s 59 .Inthe case of Shimadzu, however, the relationship with doctors was different.Their presence cannot be explainedb yad esire to work together to develop X-ray machines.E xceptf or the surgeonH ida Shijiro, the ideam an behindt he firstX -rayd evice form edical practice produced by Shimadzu(1915) and co-founder of theT okyo-based firm Hida Electric Industry(1915) 60 ,none of thesedoctorswere engaged in the development of 57 Goto 1969, 18.  58 Goto 1969, 64.59  radiologyand medical devices.Forexample,noone,not even Hida,heldany patents in the field of radiology 61 .More than as ource of know-how,these doctorswere akey resource forthe commercial opportunitiesofthe firm.The nine doctorswhich career is known havesimilar profiles.They were allpart of the "nameless practitioners" who transformed theJ apanese medical system during the Meijiperiod 62 .Alongside the greatmedical figures of the UniversitiesofTokyo and Kyoto who capturedthe interest of scholars,there was also aspecific medical elite who stayedinthe shadow of historiography but played akey roleinthe modernisation of Japanese medicine:private practitioners trainedabroadontheir own.Astay in aWestern university,essentially in German-speakingc ountries, was not integrated into an academic career butviewed as along-term investment.On avery competitive Japanese hospitalmarket -therewere 808 hospitalsin1910, of which88.9% were private 63 -G erman training conferred an edge,aso nly as mall minority of doctorsbenefitedfrom this kind of background 64 .Thed octors who helpedt urnedS himadzui nto aj oint stockc ompany in 1917 allbelonged to this groupof"namelesspractitioners".They enjoyed excellent training: apart fromS akura, all held doctoral degrees fromt he University of Tokyo or the University of Kyoto,t hati s, the bestm edical schoolsinthe country.What is more,seven of them trainedinEurope,mainly in German andA ustrian faculties.Theset raining trips were carried out at their owne xpense,i.e.asl ong-term investments,exceptf or Kusunoki and Sakura.In 1917, when Shimadzu Corporation Ltdw as founded, theyw ere running private hospitalst hey owned, in addition to whichs everal were engaged in professional associations.Thecareer of Ogata Masakyo perfectly exemplifiesthese various aspects.Born in 1864, himselfthe sonofadoctor, he studied medicine at theU niversity of Tokyob efore training at his own expense in Germanya nd Austria, where he specialised in gynaecologyobstetrics (1888-1892).Upon hisr eturn to Japan, he joinedt he hospital foundedb yh is fathera tO saka, where he took over them anagement.Obtaining his PhD in medicine from the University of Tokyoin1905, he was anotable of Osaka in the early 20th century.President of the OsakaPhysicians' Association and of the Osaka MidwiferyAssociation, as wellasVice-Presidentofthe Dai-Nihon Physicians' Association, he alsoserved as Director of the Osaka Charity Hospital's College of Medicine andwas amember of the Osaka Municipal Assembly.Finally,heran amedical laboratory in the city of Osaka 65 .Like him, the doctorswho joined Shimadzuwere entrepreneurs who helped turn medicine into ab usiness.F or these private practitioners,new technologies such as X-ray machineswere destinedtonot only improvet heir medical practiceb ut alsos trengthen theirp osition on the medical market.This wasp robablyt heir main interestw hent heyc ame on board in 1917.As for the firmi tself,the purpose of this collaboration with suchdoctors wasobvious:itallowedthe company to access networks facilitating the salea nd distribution of new devices, whichw ere then launched on the market.
Thecontent of the firstone-weekcourseorganised in 1921 onceagainshed light on the collaborationb etween Shimadzu, Kyoto University,and some private practitioners.Besides Fukuda, theh ead of the X-ray machine division at Shimadzu, there were indeed three professors fromKyoto University and onef rom Kyoto Third High School, together witht hree practitioners, including Urano, who leftK ansei Hospital in 1933 to openo ne of the first private radiology instituteso fJ apan 68 .Yet organising such courses was not an original ideah atched by Shimadzu.I ndeed, doctorsi nvolved in the beginningso fr adiology staged similar demonstration courses in some hospitals and faculties of medicine starting in the early1 910s.A sf or X-ray machinesmanufacturers and sellers,Goto Fuundo,aT okyo-based competitor of Shimadzut hat importedG erman X-ray machinesi nto Japan, first organised suchacourse in 1908 at ameeting of the Japanese Association of DigestiveO rgan Diseases (Nihons hokakibyo gakkai) 69 .In1 918, theX -ray tube makerT okyoE lectric alsor an as imilar course.Like Shimadzut hree years later, the lectures were given by thee ngineers of the firma sw ella s somep rofessors of medicine.They aimeda tp assing on theoretical knowledge on the physical principles of radiology andalsoontheir medical applications using the firm'sproducts 70 .
In addition to this one-week course,S himadzua ctivelye ngaged in the scientific life of medicine from the mid-1920s onwards, mainly throught he engineer Fukuda,who participated in the activitieso ft he Japanese Roentgen Association, presentingp apers at annualc onferences andp ublishing articlesi nt he society's journal 71 .Shimadzui tself organised some scientific events.InFebruary 1924, to mark thefirstanniversary of Wilhelm Roentgen's death, the enterprise setu paR oentgen Festival (rentogensai), gathering together in Kyoto more than 200p ersonsw ho heard as eries of scientific communications 72 .A similar event has taken place sincet hene very year,a practicewhich still exists today.Moreover, Shimadzuhas also publishedsince 1925 its ownscientificjournal, the ShimadzuRadiologyBulletin (Shimadzu rentogen jiho), which largelyconsists of articlesbythe firm'semployees 73 .

Conclusion
Thed iffusion of newm edical technologies in Japan during the first part of the 20thc entury,approachedt hrought he example of X-ray machines producedb yS himadzu, underscorest he key influenceo ft he market structure in this process,int his case ac ompetitivea nd relatively liberalised market.Western historiographyusually tends to stress theimportance of social networks, especiallydoctors' ones,inasocialconstruction of technology (SCOT) approach,a nd this of policyc ontrol by the State, to explain the diffusion of medical technologies 99 .The rare analyses that incorporate the economic dimension area bout regulated markets,inw hicht he State plays ad ecisive roleintermsoftechnical equipment, as can be observed forexample in the cases of osteosynthesis and dialysisdevices after WorldWar II 100 .Thus,upto thatp oint, the attention wasf ocused on the users andc onsumers of technologies(State,doctors, hospitals and patients)rather than on their producers,t hat is the medical device manufacturers,asi ft heyh ad played only a passive rolei nt he diffusion of technologies 101 .T his paperh as adopteda complementary approach -namelythat of business history.
Thed iffusion of radiology in Japan occurred in ap articulare conomic context,that of af ast-growing,largely privatised andc ompetitiveh ospital market.This market structurehad two mainimpacts on the firm'smanagement.First, it ledShimadzutoadopt very earlyonastrategy of insertion into the medical networks, withthe collaboration of privatedoctors, involvement in scientific activities,and the creation of atraining school for X-ray technicians.S econd, owing to itsf ast growtha nd keenc ompetition, the medical market required at echnological developments trategy based on the internalisation of capabilitiesfrom universities,especially Kyoto University,and external R&D centres.F inally,o nb oth sides, access to domestic external resources,such as medical networks and technological know-how developed within universities,enabledShimadzutoestablishitself as the firstJ apanese producer of X-ray machines and to competeeffectively with Americanand German multinationale nterprises.Viewed from this perspective, the structure of the Japanese medical market had apositiveimpactonthe diffusion of newt echnologiesa nd the growth of ad omestic medical device industry.It is difficult to make acomparison with other medical devicesdue to alack of research in this field, but aq uick glance at the Japanese foreign trade statistic showst hat the interwary ears were also as ap ivotalp eriod for the surgical instrument industry.Afterpostedsteady growth in the yearsfollowing WorldWar I, peakingat1.1 millionyen in 1924, imports indeed entered ap haseo fcontinuous decline,due to the development of nationalp roduction.Exportseven overtookimports in 1928 102 .
Acomplete business historyapproach would alsorequiredue consideration of the perspective of customers -hospitals,clinics-inorder to analyse the impactofnew technologiesonthe management of theseinstitutions,the evolution of the medical market structure and thediffusion of technologies towardsp atients.T he liberalised and atomised structureo ft he Japanese medical market ledt ok een competitionb etweenh ealthcare institutions whichobviously favouredthe diffusion of new technologies that could have beenu sed as distinguishing features and comparative advantages.I ndeed, the Japanese hospitalsystem is madeofnumerous,mainly private andsmall health centres.In 1960, there were atotal of 6,094hospitals (byoin) -ofwhich 74.2% were private-with 113beds on average,together with 59,008 clinics (shinryojo) -n earlya ll private -w ith 3b eds on average 103 .A tt he time, 90.1% of hospitals and 56.6% of clinics were equippedwith X-ray installations 104 .U nder these conditions,h ospitals and especially clinicsp robably made frenzied use of these devices to ensure their amortisation, whichwould largely explainwhy Japan is even today the countrywith the highest density of medical imaging devices in theworld 105 .

35
Shimadzu 1967, 27.36 Database of theI ndustrial Property DigitalL ibrary,w ww.ipdl.inpit.go.jp/homepg.ipdl(accessed in February and March2010).37 Dainihonhakushi echnicalc ollegea tH amamatsu beforer eturning to Tohoku University,whereheworkedasaresearcher and obtainedhis PhD in 193049.Othere ngineersw ere engaged to work underK ikuchi, among whom was NakaboriK ouji 50 .After graduating from the Department of Electrical Engineering at the Faculty of Engineering of Kyoto University in 1935, he enteredShimadzuimmediately thereafter,wherehespent his entire career as aresearcher,obtaining his PhD at Kyoto University in 1956 and becoming amember of the boardofdirectorsin196351.In addition to this strategy of employingengineerswith university degrees, in the early1930s Shimadzu cooperated withoutside institutions withregard to R&D on Coolidgetubes 52 .In1932, it commissionedthe Aoyagi Research Laboratory,which washeaded by aprofessor of Kyoto University specialised in research on vacuumtubes,todoresearch forit 53 .Thatsame year,Shimadzu also subcontracted someR&D to an Osaka-basedfirminwhichitinvested, the Japan Quartz Industry Co (Nihon sekiei kogyo).In1 934, this company produced aprototypeofaCoolidge tube,based on amodelfromthe German company Müller, which it delivered to Shimadzu 54 .Subsequently,Shimadzu began in-houseproductionofthesetubes in 1940 55 .
and still existst oday.S ee www.shimadzu.co.jp/ tec_news/index.html(websiteaccessed19March2010).X-raymachines in hospitals and clinics,the firm not only ensuredthe training of anew generation of technicians abletouse X-ray equipment effectively but also backed the diffusion of this new technology into Japanese healthcare institutions ande ncouraged the use of its own apparatuses 74 .A fter .3.Lecturers at theK yotoT rainingCenter forX-Ray Te chnology,1927 1 .TheT raining Centre forX-Ray Te chnology (Rentogen gijutsukoshujo) was openeda tK yoto, close to theS himadzuf actory,i n1 927.Once again, the composition of theteaching staffreflected Shimadzu'sclosecooperation ties with universitiesa nd private practitioners (see table3).Thec enter,run by Fukuda,alsoe mployed four other engineers.Asf or the three external lecturers,t hey were Professor Aoyagi and Doctor Urano,both of whom also had beensold in greatquantitiesuntil the war 90 .These salesmadeitpossible to restart production in the Kyoto factory aftert he war,b ut growthw as primarily driven by domestic demand after1950.Sales of X-ray equipment by Shimadzusoared from 222million yenin1949 to 1.2 billion yen in 1960, enabling Shimadzutoremainamong the leading Japanese producers of X-ray machines.Ithad an average market share of 44.5% for 1954-1960 (see fig.2).As wasthe case beforethe war, this development wasbased on the model of the internalisationo fe xternal capabilitiesa nd an active R&D policy.Cooperation tieswith Kyoto University and university researchers continued during and after thew ar.T hus,K ato Noboyoshi, ap rofessoro fe lectrical engineering at Kyoto University,was engaged as aconsultant in 1944 91 .More- Tab , between1946 and 1960, four Shimadzu researchers earned PhDs,three from Kyoto University andone from JikeiUniversity,for research on X-ray technology92.Finally,Shimadzubenefited frompublic supportfor R&D.Six research projectsw ere funded by theM inistryo fE ducation in 1949-1950, followed by twoothersbythe Ministry of International Tr adeand Industry (MITI) in 1954-1955 93 .After the war, an engineer hired around the mid-1920s, Fujimoto Keiji, became directoro ft he X-rayd ivision.He was as cientist actively engagedinthe Japanese Roentgen Association 94 ,and taught in the early 1940s at the Faculty of MedicineofK okkaido University 95 .Paralleltothe pursuit of collaboration with academic research,Shimadzu adoptedas econdt echnological developments trategy: cooperation with a foreignm ultinationalc ompany.T he firm'si ntegration into the Mitsubishi group ledt oi ts inclusion in the negotiationso nageneral technologyc ooperation agreement signed in 1951 witht he Americanf irmW estinghouse ElectricCo.,which had beenworking together with Mitsubishi Electricsince 1923 96 .The 1951 agreement enabled Shimadzutobenefitfromthe technologies of the American companya nd stayc ompetitivew ith its rivalso nt he Japanese market. over