n a t i o n a l a c a d e m y o f s c i e n c e s
Any opinions expressed in this memoir are those of the author(s)and do not necessarily reflect the views of the
National Academy of Sciences.
H a r o l d l e s t e r J o H n s o n
A Biographical Memoir by
Grard H. de vacouleurs
Copyright 1995NatioNal aCademies press
HAROLD LESTER JOHNSON
April 17, 1921April 2, 1980
B Y G R A R D H . D E V A U C O U L E U R S
HAROLD JOHNSON, ONE OF the most productive and influ-ential observational astrophysicists of this century, wasborn in Denver, Colorado, on April 17, 1921, the son ofAverill C. and Marie (Sallach) Johnson. He received hiselementary and secondary education in Denver schools andwent on to the University of Denver, receiving the B.S. de-gree in mathematics in 1942. His correspondence makes itclear, however, that his mind was already set on becomingan astronomer.
WAR YEARS AND GRADUATE STUDIES, 1942-48
Graduating with a strong physics background shortly af-ter the entry of the United States in the Second World War,Johnson was immediately recruited by the Radiation Labo-ratory at the Massachusetts Institute of Technology, wherehe worked on radar interference techniques. Here he metAlbert Whitford, an astronomer then applying electronictechniques to photoelectric measurements of the light ofstars. Toward the end of the war years Johnson moved tothe Naval Ordnance Test Station, Inyokern, California, wherehe worked with Gerald Kron, also an astronomer engagedin the photoelectric photometry of stars.
244 B I O G R A P H I C A L M E M O I R S
After the war, with Krons support and encouragement,Johnson began graduate studies in astronomy at the Uni-versity of California in Berkeley. He completed his thesiswork in the remarkably short time of two years and re-ceived his Ph.D. in 1948. His thesis adviser was Harold Weaver,and his thesis project involved the development of an elec-tronic plate measuring machine. Most of the work was doneat Lick Observatory on Mount Hamilton, where further as-sociation with Kron turned Johnsons attention to photo-electric photometry. His first two papers, published in 1948-49 in The Astrophysical Journal and the Publications of theAstronomical Society of the Pacific, dealt with electronic cir-cuitry and with the ultimate sensitivity limit set by quantumnoise. These marked the beginning of a lifetime of dedica-tion to high-precision astronomical photometry, a field ofwhich he was to become the leading practitioner.
RESEARCH AT LOWELL, WASHBURN, AND YERKES OBSERVATORIES,
Upon completion of his work for the Ph.D., Johnson ac-cepted an offer to join the staff of Lowell Observatory, Flag-staff, Arizona, where he spent the second half of 1948, build-ing an AC electronic amplifier for the observatory solarvariation project under a Weather Bureau contract. Thisfirst AC amplifier did not work well, however, and it waseventually replaced by a standard DC amplifier. This initialperiod at Lowell Observatory did not measure up to Johnsonsexpectations either, and at the end of 1948 he moved toWashburn Observatory as assistant professor at the Univer-sity of Wisconsin in Madison. There he joined the projectstarted by Joel Stebbins and Albert Whitford to establish aphotoelectric calibration for the stars in selected areas thenused as standards for the magnitudes of Cepheid variablesin local group galaxies. Johnsons expertise was important
245H A R O L D L E S T E R J O H N S O N
in extending the sequences to fainter limits. This study showedconclusively that the previous photographic calibration erredby nearly a magnitude (a factor of more than 2 in intensity)at the faint end. The corrections found necessary enteredinto the ensuing revisions of the cosmic distance scale, whichput external galaxies considerably farther away than theyhad appeared to be in the first estimates. This work con-firmed Johnsons determination to push the photoelectrictechnique to its extreme limits, which he did very success-fully over the next ten years.
Johnson moved then to the Yerkes Observatory of theUniversity of Chicago, where he was an assistant professorfor two years (1950-52). There he met William W. Morgan,then the leading expert in stellar spectral classification andwith whom he would soon begin a momentous research onthe combined photometric and spectroscopic properties ofstars which, as will be explained below, led to revolutionaryadvances in astrophysics. Neither the teaching professionnor the murky skies of the Chicago area were much toJohnsons liking, however, although he had access for hisobservations to the clear skies of McDonald Observatory inWest Texas, which was then under the management of theChicago astronomers. At any rate, upon learning that vet-eran Lowell astronomer C. O. Lampland (1873-1951) hadpassed away, Johnson revisited Flagstaff in early 1952 andconvinced the elderly director, Vesto M. Slipher, that a sec-ond try at using his talents would be in the best interest ofthe observatory. This was agreed and in August 1952 HaroldJohnson returned as staff astronomer to Lowell Observa-tory, where free of teaching duties and under a favorablesky, he could devote full time to his efforts to push stellarphotoelectric photometry to its ultimate limits.
Prior to 1950 photoelectric photometry had been a diffi-cult technique reserved to a few skilled practitioners. In
246 B I O G R A P H I C A L M E M O I R S
these early years the charge generated over a specified timeinterval was measured with an electrometer, a delicate de-vice transplanted from the physics laboratory to the focalplane of a moving telescope. Paul Guthnick in Germanyand Joel Stebbins in the United States were among the fewto have successfully mastered the technique. The chargesensitivity fell far short of reaching the quantum noise limit.The thermionic amplifier introduced by Whitford in 1930considerably increased the sensitivity of photoelectric ob-servations. The development of the electrostatically focusedmultiplier phototube by RCA in 1940 eliminated the needfor a low-level amplifier and provided better cathode re-sponse. The greatly simplified technique encouraged a largernumber of astronomers to enter the field, including thosewith smaller telescopes.
Photoelectric photometry was thus just coming of agewhen Johnson began his two-year period at Yerkes Observa-tory in 1950. He realized that the inherent linearity of thephotoelectric process (favorable for direct subtraction ofthe superimposed light of the night skya large correctionwhen observing faint stars) had to be preserved and that allfluctuations introduced by the equipment had to be re-duced to a level below that of the unavoidable statisticalfluctuations in the number of photoelectrons released bystarlight. He also realized the need for better definition ofthe measured quantity, the stellar magnitude of the observedobject. This required knowing the spectral sensitivity func-tion of the system, including all the optics in front of thedetector (atmosphere, telescope, filter).
Early photoelectric observers had attempted to match theirmeasured magnitudes to the international photographic andphotovisual systems adopted by the International Astronomi-cal Union in 1922, following extensive measures of a set ofstars near the north celestial pole, the North Pole Sequence,
247H A R O L D L E S T E R J O H N S O N
by many astronomers at several observatories and especiallyby Frederick H. Seares at Mount Wilson Observatory. Thephotovisual system, based on orthochromatic emulsions anda yellow filter, was intended to match the average sensitivityof the human eye, the first instrument used to estimatestellar magnitudes. With a suitable color filter in front ofthe photocell, a reasonably satisfactory match could be ob-tained.
Attempts to match the international photographic mag-nitudes were complicated by the wide spectral range cov-ered by blue-sensitive photographic emulsions, a difficultythat had led to poor agreement between magnitudes mea-sured with different telescopes; the ultraviolet cutoff wasdependent on absorption by the flint glass component inrefractors or on the falling reflectivity of the silvered mir-rors in reflectors. After 1933 the change from silver to alu-minum coatings, with their higher UV reflectivity, led tofurther complications.
Johnson recognized that these difficulties arose from in-clusion to a varying degree of the near ultraviolet around 0.37-0.38 microns, where the energy distribution in stellarspectra varies rapidly. He demonstrated that a reproduciblemagnitude system could be established by placing a suit-able filter before the photographic plate or photocell toblock all wavelengths shorter than 0.38 microns. Also, theoverall shape of the spectral energy distribution could bebetter defined by using, in addition to the yellow and bluebands, a third color in the ultraviolet near the head of theBalmer series of hydrogen and thus sensitive to the size ofthe Balmer jump, a measure of the temperature and den-sity in stellar atmospheres. In collaboration with W. W. Morganand D. L. Harris, Johnson thus introduced a new standardsystem of stellar photometry, the U, B, V system, based onten primary standard stars and, initially, 108 secondary stan-
248 B I O G R A P H I C A L M E M O I R S
dards well distributed around the northern and equatorialzones of the sky accessible from the McDonald and Lowellobservatories where his observations were made. The newsystem was published in epoch-making papers in The Astro-physical Journal in 1953-54 and in Annales dAstrophysique in1955.
The system was rapidly adopted and quickly became thede facto international standard of stellar (and, later, gal-axy) photometry, a role it has retained to this day. Johnson,alone or in collaboration with Allan Sandage, RichardMitchell, Braulio Iriarte, and others, made massive applica-tions of this system to galactic clusters, producing well-de-fined, precise color-magnitude and color-color diagrams,that is, plots of apparent V magnitude versus B-V color indi-ces (resembling the Hertzsprung-Russell diagram of lumi-nosity versus effective temperature) and of U-B color versusB-V. He established the fundamental properties of thesediagrams and showed how to use them to disentangle theeffects of temperature (intrinsic) and interstellar redden-ing (extrinsic). After his move to Flagstaff he demonstratedhow clusters of different ages have characteristically differ-ent color-magnitude diagrams and, in particular, how theturn-off point where the stars begin to depart from thezero-age main sequence can be used to estimate fairlyprecisely the ages of the clusters. This striking observationalconfirmation of theoretical modeling, initially by MartinSchwarzschild, of the paths that stars follow on the color-magnitude diagram in their post-main-sequence evolutionopened the way to many investigations of stellar and clusterages that are continuing to this day. For this work HaroldJohnson was awarded, in 1956, the Helen B. Warner Prizeby the American Astronomical Society.
During this period in Flagstaff, Johnson was also one ofthe first (simultaneously with William Baum at Palomar) to
249H A R O L D L E S T E R J O H N S O N
push the sensitivity of photoelectric photometry to the quan-tum noise limit by developing pulse-counting photometers,essentially counting photoelectrons one by one. In his pur-suit of the ultimate precision, he also built what was prob-ably the first two-channel photometer, one measuring thestar under study and the second simultaneously measuringthrough an identical aperture the variable luminosity ofthe night sky in a nearby spot, thus eliminating, by differ-ence, the troublesome fluctuations of the night airglow. Byrapidly reversing the roles of the two channels, any system-atic difference between the two optical trains and photo-multipliers was neatly eliminated. Johnson developed con-venient forms to facilitate (in those precomputer days) thereduction of photoelectric observations, designed an inge-nious analog-to-digital device to measure the star and skydeflections on chart records of the photo-current, and de-fined the rigorous procedures to be followed to obtain thehighest precision in this type of observation.
During his second period on the Lowell staff, Johnsonwas also actively engaged with Aden B. Meinel in the sitesurvey for the future National Optical Astronomical Obser-vatory (NOAO), which was eventually built on Kitt Peaknear Tucson.1 In 1956 he actually spent six months in Phoe-nix, where the initial NOAO office was located at the time.He was even considered by the first director, Leo Goldberg,for a top research position in the new organization buteventually decided to return full time to Flagstaff at theend of 1956.
The author of this memoir was fortunate to becomeJohnsons colleague and friend during his stay at LowellObservatory in 1957-58, when he began a long-term pro-gram of galaxy photometry in the U, B, V system, initiallywith Johnsons photometer attached to the 21-inch reflec-tor. This program, since continued at McDonald Observa-
250 B I O G R A P H I C A L M E M O I R S
tory (and elsewhere by many others), has provided the ba-sis for the most generally used systems of total magnitudesand colors of galaxies.
AT THE UNIVERSITY OF TEXAS, 1959-62
Toward the end of 1959 Johnson accepted an invitationof Grard Kuiper, director of the Yerkes and McDonaldobservatories, to join, as professor of astronomy, the newlyformed Department of Astronomy of the University of Texasat Austin. He became briefly its chairman in 1961-62 afterKuiper relinquished his directorship and the chairmanshipof the joint Chicago-Texas department to move to the Uni-versity of Arizona in Tucson, where Johnson was soon tofollow him.
Johnsons years at Texas were very productive in the sensethat he developed and used much new equipment atMcDonald Observatory, but frustrating because he failed toreceive from the university administration whole-heartedsupport for the kind of research and development he wantedto give to the department and, especially, the observatory.He was more interested in the directorship of the observa-tory than the chairmanship of the department, but the ad-ministration saw things differently.
It is during this period in Texas that Johnson first camein contact with Frank Low, who was then a physicist withTexas Instruments in Dallas, where some very sensitive in-frared detectors were being developed. Johnson immedi-ately seized on this opportunity to build a photometer ex-tending the U, B, V system to the longer wavelengths of thenear infrared, the R, I, J, K, and L bands out to 4 microns.With Lows germanium bolometer, this was later extendedto the N band at 10.2 microns. The ability of the longerwavelengths to better penetrate the selectively obscuringinterstellar haze opened new avenues of research. The longer
251H A R O L D L E S T E R J O H N S O N
wavelength also gave more information on the radiationfrom cool stars, which, by Wiens displacement law, is mainlyemitted in the infrared and is heavily blanketed by molecu-lar absorption bands in the visible. The results were firstreported in The Astrophysical Journal in 1962.
It was also during his stay in Austin that Harold Johnsonbecame aware of the remarkable work of Larry Mertz, atHarvard Observatory, where he had built the first experi-mental Fourier-transform interference spectrometer (1958-59). This author, who was there at the time, remembersvividly how the Harvard faculty failed to appreciate the sig-nificance of what Mertz was doing and dismissed him as amere tinkerer. Made aware of Mertzs work, Johnson im-mediately grasped the enormous potential of interferencespectrometry, particularly for the infrared, and before leav-ing Texas proceeded to build the first successful Fourier-transform stellar interferometer working in the near infra-red. He was to greatly improve and develop this techniqueafter his move to Arizona.
AT THE LUNAR AND PLANETARY LABORATORY,
UNIVERSITY OF ARIZONA, 1962-69
In February 1962 Johnson accepted Grard Kuipers invi-tation to follow him to join the newly created Lunar andPlanetary Laboratory (LPL) at the University of Arizona inTucson, where he served as a research professor (1962-67)and then associate director (1967-69). Although Johnsondid make some applications of his early version of the Fou-rier-transform spectrometer to the infrared spectra of themajor planets, he was free to pursue his main line of inter-estnamely, the infrared photometry and interference spec-troscopy of stars. He used to joke that he was the stellardivision of the Lunar and Planetary Laboratory. It is fortu-nate for astronomy that Kuiper was a far-seeing scientist
252 B I O G R A P H I C A L M E M O I R S
able to accept on his staff a gifted and productive indi-vidual, even if he was not working on the main line ofinterest to the institution.2
Among the many contributions made by Johnson in thisatmosphere of freedom under the favorable skies of south-ern Arizona, where Kuiper had established a fine high-alti-tude observatory in the Catalina Mountains northeast ofTucson, we may mention major papers on the infrared pho-tometry of late-type and carbon stars, studies of atmosphericand interstellar extinction, massive catalogs of eight- andlater thirteen-color photometr y of bright stars (in collabo-ration with R. I. Mitchell, W. Z. Wisniewski, and B. Iriarte)published in the Astrophysical Journal and Communications ofthe Lunar and Planetary Laboratory between 1962 and 1969.Johnson, not satisfied with the differential measurementsof stellar magnitudes, also undertook the more difficult taskof performing an absolute calibration of stellar magnitudesin terms of energy fluxes. He was thus able to produceabsolute energy curves, eventually extended up to 20 mi-crons, for all sorts of important categories of stars: subdwarfstars, cepheids, M stars, carbon stars, infrared objects, andeven circumstellar shells.
One of the most important discoveries made by Johnsonduring this period was the great intensity of the infraredemission of the prototype quasar, 3C273, surpassing evenits visible emission. This result was soon found to be a gen-eral property of the radiation from quasars and other activenuclei of galaxies.
During the same period Johnson utilized the accumu-lated observational data on the spectral energy distribu-tions of stars of all types to derive a new set of bolometriccorrections to their visual magnitudes and then to establisha revised temperature scale more directly based on observed
253H A R O L D L E S T E R J O H N S O N
energy distributions. He also used observations of reddenedstars in galactic clusters to discuss the wavelength depen-dence of extinction by interstellar dust. The infrared excessat the longest wavelengths in certain clusters over the nor-mal absorption was, in the end, identified as radiation fromwarm circumstellar dust shells, rather than abnormal prop-erties of the interstellar dust particles. The fundamentalcontributions were summarized in classical review papers inAnnual Reviews and in the standard compendium Stars andStellar Systems, where Johnson also described in some detailthe construction of his stellar photometers. Many photom-eters built after Johnsons design are still in use around theworld. Johnsons election to the National Academy of Sci-ences in 1969 recognized his major influence on the progressof astronomy during the previous twenty years.
During this same period also Johnson began to be inter-ested in the design and construction of low-cost medium-size reflectors, specially designed for infrared photometryand Fourier-transform spectroscopy. An experimental 60-inch reflector made of spun aluminum was successfully builtand tested under his direction at the Catalina station. Itwas later transferred to the Mexican National Observatoryin Baja California. The poor optical quality (by ordinarystandards) was quite good enough to feed most of the en-ergy of the infrared image of a star into the rather largeentrance aperture of the photometer or spectrometer. Hisideas of building cheap telescopes for specialized tasks werenot without their detractors, but Johnson was more inter-ested in doing great science than big (meaning expen-sive) science. If results of equal significance could be got-ten more cheaply, he would prefer the latter, a case of brainversus brawn, which proved itself a decade later in his con-tributions to Mexican astronomy.
254 B I O G R A P H I C A L M E M O I R S
THE ARIZONA-MEXICO CONNECTION, 1969-80
Harold Johnson had for many years collaborated withMexican astronomers Braulio Iriarte and Eugenio Mendoza,among others, informally assisting them in their researchwith advice and the loan of photometers. In 1969 this asso-ciation was formalized when he became a part-time mem-ber of the scientific staff of the Institute of Astronomy ofthe National University of Mexico, eventually becoming afull-time professor in 1979 when he actually moved to livein Mexico City. In 1973 he was one of the founders and in1975 became head of the Department of Applied Physics ofthe newly created Center for Scientific Research and HigherEducation in Ensenada, Baja California Norte. He main-tained, however, his ties to Arizona, where in 1969 he trans-ferred to the Optical Sciences Center (then headed by AdenB. Meinel) as research professor and to the Steward Obser-vatory (then headed by Bart J. Bok, 1906-83) as astrono-mer. He maintained this dual connection with Mexico andArizona until the end.
Johnsons active involvement in Mexican astronomy be-gan in 1964 with his participation in multicolor observa-tions of bright stars with the 1-meter (40-inch) reflector ofthe Tonantzintla Observatory near Puebla. He helped withthe search for the best location for the new national obser-vatory. His support of a proposal by E. E. Mendoza to thendirector G. Haro, identifying a peak in the Sierra de SanPedro Mrtir, Baja California, at an altitude of 2,800 meters(9,200 feet) among the pine trees of a protected nationalforest, as suitable for the proposed observatory, was cru-cialaccording to Mendoza, Without his help, no SPMobservatory, most likely. Of special importance and of in-terest to Johnson was the low water vapor content of theatmosphere, making it very suitable for work in the infra-
255H A R O L D L E S T E R J O H N S O N
red. The main telescope, with a 2.1-meter main mirror, em-bodying many of Johnsons ideas, was dedicated and beganoperation in 1979. In grateful recognition of his role in thisproject and in the development of Mexican astronomy ingeneral, the National University of Mexico conferred himthe degree of doctor honoris causa in 1979 and, after hisuntimely death the next year, the Universities of Mexicoand Arizona named after him the 1.5-meter aluminum-mir-ror infrared telescope Johnson had brought from Arizonato Mexico. Symposium 96 of the International Astronomi-cal Union on Infrared Astronomy held in 1980 was dedi-cated to Harold Johnsons memory.
Building on the laboratory facilities he had in Tucson,where he remained active during his Mexico years, Johnsonperfected a high-resolution Fourier-transform infrared spec-trometer, using as its core a Michelson interferometer builtby the Block Engineering Company of Boston, under thedirection of Larry Mertz. Johnson and his associates, F. F.Forbes, R. I. Thompson, and D. L. Steinmetz of the Univer-sity of Arizona and O. Harris of the National University ofMexico, used it on several telescopes in Arizona and on theNASA Lear jet stratospheric observatory to produce high-resolution spectra of the sun and bright stars in the spec-tral range of 1.0 to 4.0 microns (stars) and 5.6 microns(sun). The results were first reported in the Publications ofthe Astronomical Society of the Pacific in 1973. Later the resultswere collected in a comprehensive Atlas of Stellar Spectra.The resolution of 0.5 centimeters1 corresponds to about0.1 angstrm unit in the middle of the range, possibly thehighest resolution ever achieved on astronomical sourcesin this spectral region. The tracings of the infrared spectraof bright stars and planets, displayed in Johnsons Tucsonlaboratory, covered several tens of meters on the walls!
Toward the end of his life Johnson became increasingly
256 B I O G R A P H I C A L M E M O I R S
interested in the design and construction of mirror arrays,that is, multiple-mirror telescopes, as a way to realize largeapertures at a fraction of the cost of solid monolithic mir-ror telescopes. In this field too he was a pioneer, and heproposed to build for Mexico an array of twenty 2-metertelescopesMextels, as he called themto provide thelight-gathering power of a 10-meter telescope at half thecost. A prototype was built and installed at the nationalobservatory site. The multimirror scheme has now beenwidely accepted and implemented around the world as theway to build super-large telescopes.
HAROLD JOHNSON AS FRIEND AND COLLEAGUE
Little is known of the private life of Harold Johnson. Hewas married to Mary Elizabeth Jones in 1954. They had twochildren: August Harold and Selma Marie. After Haroldsuntimely death in 1980 in Mexico City, Mary returned toTucson, where she lives in retirement.
To those who did not know him well, Harold Johnsonmay have seemed to be often blunt, brusque, and lackingin the suave polish necessary to become a successful aca-demic. He did not care much for formal teaching. It maybe true that as a colleague Johnson was occasionally diffi-cult to live with, but it was well worth the effort. He suf-fered all his life from breathing problems that got worsewith time and may well have influenced his personality. Buthe was a fundamentally honest man, with a strong religiousbent (once he even attended a revival meeting) and a pro-found dedication to the truth in science as well as everydaylife. He was impatient with mediocrity, and all his life wasdedicated to striving for the ultimate precision and exacti-tude in his several fields of endeavor, in each of which hemade fundamental contributions. He was always willing, eveneager, to share his profound knowledge of photometry and
257H A R O L D L E S T E R J O H N S O N
electronics with students, associates, and colleagues. He hadan extraordinary skill in designing and building new instru-ments; he had electrons in his finger, as an envious andadmiring competitor remarked once. He knew better thananyone how to build amplifiers whose response was linearover an enormous range of intensities; he built one atMcDonald Observatory with which he could measure withthe 2.05-meter reflector without loss of linearity from Siriusto 22-nd magnitude starsthat is, an interval of 2.5 billionsto 1 in light flux.
Harold Johnson brought extraordinary instrumental andelectronic talents to devising equipment that utilized tomaximum advantage newly developed photoelectric detec-tors as they became available. His measurements of the col-ors and magnitudes of stars in galactic clusters on the pre-cisely defined system he devised in collaboration with W.W. Morgan led to age determinations that opened the wayto exploitation of the color-magnitude diagram as a diag-nostic in studies of stellar evolution. He had a leading rolein the use of new infrared detectors in the photometry ofstars and galaxies. With his colleagues he measured thou-sands of stars that became reference standards.
Johnson applied these measurements to calibrate spec-tral energy distributions of stars and thus provide an im-proved observational basis for the stellar temperature scaleand the bolometric corrections to visual magnitudes. Hewas the first to apply Mertzs concepts to build practicalstellar Fourier-transform spectrometers; for cool stars, inparticular, these gave unsurpassed resolution in the infra-red. These fundamental contributions to observational as-trophysics constitute Harold Johnsons enduring scientificlegacy.
258 B I O G R A P H I C A L M E M O I R S
The twenty-five titles in this memoirs selected bibliogra-phy are among the most important of some 135 paperspublished by Johnson between 1948 and 1980, but many ofthe others were no less significant and influential. His lasttwo papers appeared posthumously in 1981 in the proceed-ings of the symposium Recent Advances in ObservationalAstronomy (UNAM, 1981), which he helped organize. Hedied of a heart attack in Mexico City on April 2, 1980.
THIS MEMOIR HAS BENEFITED ENORMOUSLY from the generous col-laboration of Albert Whitford, who not only provided his own remi-niscences of Harold Johnsons early scientific contributions but alsocommunicated copies of letters in the Lick Observatory Mary LeaShane archives and secured a valuable testimony from Gerald E.Kron, all referring to Johnsons period as a graduate student atLick and Berkeley. Whitford kindly revised, corrected, and enlargedseveral sections of this memoir but modestly declined to be namedas a coauthor. I am deeply grateful for his contribution. I also ac-knowledge valuable communications from H. C. Giclas, Lowell Ob-servatory; E. E. Mendoza V, University of Mexico; and W. Z. Wisniewski,University of Arizona at Tucson. The frontispiece photograph ofHarold Johnson, taken in 1965 at LPL by D. Milon, was kindlyprovided by E. A. Whitaker, University of Arizona at Tucson.
1. The writer remembers that Johnsons own preference for thesite of the new observatory was an isolated peak, Slate Mountain, inthe desert northwest of Flagstaff as a better, darker, and dryer sitethan Kitt Peak and likely to remain free of light pollution for manyyears, but practical considerations of accessibility, development costs,and living convenience for the staff prevailed in the end.
2. A very readable account of the founding and early years ofLPL written by Ewen A. Whitaker (University of Arizona, 1985)includes a section on Johnsons contributions.
259H A R O L D L E S T E R J O H N S O N
S E L E C T E D B I B L I O G R A P H Y
A theoretical discussion of the ultimate limit of astronomical photo-electric photometers. Astrophys. J. 107:34-47.
With M. Schwarzschild. On the color-magnitude diagram for M 15.Astrophys. J. 113:630-36.
With W. W. Morgan. On the color-magnitude diagram for the Pleia-des. Astrophys. J. 114:522-43.
On magnitude systems. Astrophys. J. 115:272-82.
With W. W. Morgan. Fundamental stellar photometry for standardsof spectral type on the revised system of the Yerkes spectral atlas.Astrophys. J. 117:313-52.
With D. L. Harris III. Three-color observations of 108 stars intendedfor use as photometric standards. Astrophys. J. 120:196-99.
Galactic clusters and stellar evolution. Astrophys. J. 120:325-31.
With A. R. Sandage. The galactic cluster M 67 and its significancefor stellar evolution. Astrophys. J. 121:616-27.
With W. A. Hiltner. Observational confirmation of a theory of stel-lar evolution. Astrophys. J. 123:267-77.
Photometric distances of galactic clusters. Astrophys. J. 126:121-33.
260 B I O G R A P H I C A L M E M O I R S
With R. I. Mitchell. The color-magnitude diagram of the Pleiadescluster II. Astrophys. J. 128:31-40.
With S. N. Svolopoulos. Galactic rotation determined from radialvelocities and photometric distances of galactic clusters. Astrophys.J. 134:868-73.
Infrared stellar photometry. Astrophys. J. 135:69-77.
With R. I. Mitchell, B. Iriarte, and D. L. Steinmetz. Photoelectricphotometry of cepheid variables. Bol. Obs. Tonantzintla Tacubaya3:153-304.
The colors, bolometric corrections and effective temperatures ofthe bright stars. Bol. Obs. Tonantzintla Tacubaya No. 25.
With R. I. Mitchell, W. Z. Wisniewski, and B. Iriarte. UBVRIJKLphotometry of bright stars. Commun. Lunar Planet. Lab., Univ.Ariz. 63:99-110 plus 130-page catalog.
Astronomical measurements in the infrared. In Annu. Rev. Astron.Astrofis. 4:193-206.
Interstellar extinctions. In Stars and Stellar Systems, vol. VII: Nebulaeand Interstellar Matter, pp. 167-220. Chicago: University of Chi-cago Press.
With I. Coleman, R. I. Mitchell, and D. L. Steinmetz. Stellar spec-troscopy, 1.2 microns to 2.6 microns. Commun. Lunar Planet. Lab.,Univ. Ariz. 113:83-103.
With F. F. Forbes and W. F. Stonaker. Stellar and planetary spectrain the infrared from 1.35 to 4.2 microns. Astrophys. J. 75:158-64.
261H A R O L D L E S T E R J O H N S O N
With F. F. Forbes, R. I. Thompson, D. L. Steinmetz, and O. Harris.A high-resolution Fourier-transform spectrometer. Publ. Astron.Soc. Pac. 85:458-67.
With R. I. Mitchell. Thirteen-color photometry of 1380 bright starsover the entire sky. Rev. Mex. Astron. Astrofis. 2:299-324.
A new Michelson spectrophotometer system. Rev. Mex. Astron. Astrofis.2:219-30.
An atlas of stellar spectra, I-II. Rev. Mex. Astron. Astrofis. 2:71-170,4:3-201.
The absolute calibration of stellar spectrophotometry. Rev. Mex. Astron.Astrofis. 5:25-30.