197 8ApJ. . .219. .12IB The AstrophysicalJournal,219:121-128,1978January1 © 1978.TheAmericanAstronomicalSociety.Allrightsreserved.PrintedinU.S.A. tended backgroundisthoughttobecomingfrom from anextendedbackground.Severalofthediscrete resolution (RiekeandLow1973,hereafterRL73; 2-20 ¿cmradiation(BecklinandNeugebauer1968, radiation isalsouncertain. compact Hiiregionsorplanetarynebulae(RL73; uncertain, althoughsomehavepropertiessimilarto comes fromatleast19discretesourcesaswell Becklin andNeugebauer1975,hereafterBN75)have sources areseenbestat10¿cm;thenatureoftheseis (BecklinandNeugebauer1968).Severaldiscrete sources havebeenshowntobestars(Neugebaueretal. shown that,ina40"diameterregion,theradiation the diaphragmsusedweresmallenoughtoseparate farther totheeast;andindependentpositionswere the 10¿cmmapwithsameresolutionwasextended 20 ¿cm,bothdiscreteandextended.First,anew¿cm study thenatureofsourcesobservedfrom1to the sourcebeingobservedfromothercompactsources determined forthediscrete10¿cmsources.Second, BN 75).Theoriginoftheextended10and20¿cm broad-band photometricobservationsfrom1.2to map oftheregionwasobtainedwith2"3resolution; the primaryresultsof spectrophotometry areto diaphragms weremadefrom 8to13¿cmatpositions; observations with1%spectral resolutionand5" served atallwavelengths.Third, spectrophotometric and fromtheextendedbackground;14sourceswere observed inthismanner,although notallwereob- 1976), andmostofthe2-3¿cmradiationinex- 1969; Lowetal1969).Mapswith2"-5"spatial 1976; Soifer,Russell,andMerrillTreffersetal. 12.5 ¿cmweremadeofindividualcompactsources; The galacticcenterisknownasabrightsourceof Three kindsofobservationshavebeenmadeto © American Astronomical Society • Provided by theNASA Astrophysics Data System 6-3 emit mostoftheirluminosityatwavelengths10¿onandgreater;thisemissionisprobably the brightestsourceat2¿onisprobablyasupergiant,andinfrarednearnonthermal maps with2''3resolutionaregiven. from heateddust.OneofthesourcesisobservationallysimilartoextremelyredOH/infraredstars. sources foundinthegalactic-centerregionarereported.Inaddition,revised10andnew20¿an from opticallythinsilicatedustisseeninthese. Other sourceshaveluminositiesandlinearsizessimilartothoseofcompactHnregions;emission radio sourceisprobablyastellarclusterwithdensitygreaterthan10M©pc.Othersources Subject headings:galaxies:MilkyWay—nucleiinfrared:sources Photometry from1.25to12/¿mandspectrophotometry813jumofthecompact The natureofthecompactsourcesisdiscussed.Somearebestidentifiedasstarsorstarclusters; Hale Observatories,CaliforniaInstituteofTechnologyandCarnegieInstitutionWashington I. INTRODUCTION INFRARED OBSERVATIONSOFTHEGALACTICCENTER.I. E. Becklin,K.Matthews,G.Neugebauer,andS.P.Willner NATURE OFTHECOMPACTSOURCES Received 1977March28;acceptedJune17 ABSTRACT 121 used inPaperIII(Neugebaueretal.1978)tomapthe in thegalacticcenter.Spectroscopynear2.2¿cmis used asanindicatorofthedistributionionizedgas papers, dividedbysubjectratherthantypeofob- to measurethestrengthof[Nen]emissionline the sameasthatofBN75.Newrelativepositions were obtainedon1975July1the5mHaletele- center isdiscussedinPaperIV(Becklinetal.1978). determine thenatureofsources.InPaperII at 12.8¿cmasafunctionofposition. the mostsignificantchangeisthat1RS3closerto new positionshavebeenincorporatedintoFigure1; By line.Theinterstellarextinctiontothegalactic (Willner 1978),theobservationsof[Nen]lineare confirm thestrengthandidentificationofsilicate the opticalfieldstarindicatedbyanX.Thenew 1RS 7thanwasindicatedbyBN75.Theabsolute scope. Theobservationandreductionprocedurewas sisting ofscanstotheeast1RS1with2"3resolution, of BN75ispresentedinFigure1.Newdata,con- servation. Inthispaper,theobservationsareusedto absorption observedwiththebroad-bandfiltersand present positions,althoughbased onfourindependent differential refraction.Itshould benotedthatthe positions showninFigure1differfromthoseofBN coordinates ofthemapwerealsoremeasured,using uncertainty ofabout±2"in bothcoordinates. due tothedifficultyofaccounting fortheeffectsof all thediscrete10¿cmsourceswerealsomeasured.The measurements, includingthose ofBN75,stillhavean of thosegivenbyBN75.Thedifferenceisprobably 75 by2"inthesensethatnewpositionsaresouth A revisedandextendedversionofthe10¿cmmap The resultsofthepresentstudyaregiveninfour II. OBSERVATIONS 197 8ApJ. . .219. .12IB 10-1hm3 ; hm 122 are shown.Thecontourintervalis4x10Jysr(4perbeam).portionofthemapwest174230fromdata of The beamsizeisshownbythehatchedcircleinupperright-handcorner.Xdenotesavisiblefieldstar.1RSnumbers The absolutefluxisuncertaintoafactorof1.5because made in1975Marchonthe5mHaletelescopeis made ontheHooker2.5mtelescopein1972June. BN 75,withrecalibratedpositionsandbrightnesses(seetext). shown inFigure2.Theportionofthemapeast at 20/¿m. in 1975JuneandJuly,withthewavelengthbandsat were madeatthe5mHaletelescopeonfournights (AA =1.5/xm),10.1/xm511.2 (AA =0.6/xm),8.7/xm1.29.5 (AA =0.4/xm),3.5/xm0.64.8 of thelargeandpoorlyknownatmosphericextinction (AA =1.5/xm),and12.5/xm1.3/xm).The in .The8.7-12.5 /xmobservationsuseda focal planechopperproducinga7''5beamseparation InSb detector,eithera3'8or5"0diaphragm,and absolute calibrationisgivenbyBeckwithetal(1976). 174230?2 iscompletedwith5"resolutionscans in declination.Thevisualseeing wasbetterthan2"for bolometer detector,a2''3diaphragm, andawobbling secondary chopperproducing a10"beamseparation all oftheobservations.The position ofeachsourceto 1.25 /xm(AA=0.3/xm),1.652.2/xm A20/xm (AA=7^m)mapwith2''3resolution Fig. 1.—Mapofthegalactic-centerregionat10/¿m(AA=5/um)madewitha2"3circulardiaphragmonHalemtelescope. Broad-band observationsofindividuallocations The 1.25-4.8/xmobservationsweremadewithan © American Astronomical Society • Provided by theNASA Astrophysics Data System BECKLIN ETAL. infrared signaloftheindividualobjectsateither2.2or employed byBN75tomakethe2.2and10/xmmaps be measuredwasdeterminedonthebasisof paper havebeencorrectedforinterstellarextinction, the backgroundfluxcomesfromreddenedgiantand of thegalacticcenter. as willbediscussedinPaperIV.Theextinctionthe based ontheobservationsof1RS7(seebelow)and in thatwavelengthregion.Inparticular,itisassumed the assumptionthatitsenergydistributionissmooth supergiant starsandagreeswithcurve15ofvande that 1RS7,whichhasbeenidentifiedasalate-type Hulst (1946).Theextinctioncurvenear10/xmis 10 /xm.Thespatialresolutionwassimilartothat tions ofVICygNo.12(Rieke1974;Gillettetal. derived extinctiondiffersfromthatbasedonobserva- (Treffersetal.1976),hasnostrongintrinsic resultant magnitudesaregiven inTable1. as muchsilicateabsorption forthesameamountof silicate emissionorabsorptionnear9.5/xm.The 1.2-3 /xmregionisderivedfromtheassumptionthat 1-3/xm reddening.Thecorrections appliedandthe 19756) inthesensethatgalactic centershowstwice All oftheinfraredobservationspresentedinthis The measuredfluxdensities fromthephotometry 197 8ApJ. . .219. .12IB 11_1 The contourintervalis1.2x10Jysr(12perbeam).calibrationuncertaintoafactorof1.5dueatmos- pheric extinction.ThedashedcontourswereobtainedfromscansmadewithaS"circulardiaphragmontheHooker2.5mtelescope. Fig. 2.—Mapofthegalactic-centerregionat20(AA=7/xm)madewitha2''3circulardiaphragmonHale5mtelescope. © American Astronomical Society • Provided by theNASA Astrophysics Data System 20 Extinction. 10 19 16 12 11 4 2 3 7 9 6 5 8 1 : SeetextandPaperIVfordiscussion ofuncertainties. Source 1.25/xm1.652.23.54.88.7 6.1 7.7 De-reddened IntrinsicMagnitudesofGalactic-CenterSources* >10.6 4.6 6.4 4.7 6.2 6.4 5.9 5.7 5.6 4.0 5.7 7.4 2.7 5.4 6.0 5.6 6.1 5.4 4.0 4.7 3/7 3.1 5.1 4.3 5.0 5.0 5.8 1.4 TABLE 1 2.7 3.0 Ï/7 0.8 5.2 2.9 3.5 3.1 3.6 5.4 -1.6 -2.1 + 0.7 -0.4 + 1.0 -0.7 -0.4 -1.4 -0.5 -0.8 -1.0 + 2.4 -1.3 + 0.3 -1.6 -1.1 -3.1 -1.2 -2.2 -1.4 + 0.3 -1.1 -1.5 + 3.8 -2.0 -2.3 -1.8 -1.1 -3.4 -2.6 -1.8 -0.2 -1.8 -2.1 -2.3 + 2.8 -2.2 -2.1 -1.2 -1.0 -2.8 -2.1 -0.5 -1.8 -1.9 -2.1 -1.5 + 1.3 -1.5 -2.3 -3.8 -1.1 -2.7 -2.1 -2.1 -3.1 -2.5 -3.0 + 0.9 124 BECKLIN ET AL. Vol. 219 are given in Table 1 of Paper IV. For all measure- X(/x) ments, except those of 1RS 11 and 1RS 19 at 4.8 /¿m, the statistical uncertainties are less than the probable systematic errors that are due to imperfect centering of the source in the diaphragm, loss of radiation due to seeing effects, and flux in the reference beam. At 20/¿m the large and poorly known correction for atmospheric extinction is the dominant source of uncertainty. Except for one source, 1RS 16, con- tamination in the reference beam was less than 20% ; in the case of 1RS 16, it was necessary to measure the flux in the reference position independently and to correct the data appropriately. As determined from remeasurement of a number of the sources on dif- ferent nights, centering plus seeing errors were less than 0.2 mag. Overall, the infrared fluxes of individual sources should be accurate to better than 0.3 mag, except at 20 /xm. Spectrophotometric observations in the 8-13/xm range with 1% spectral resolution were made on seven nights in 1975 May and June. The observations were made at the Mount Wilson 2.5 m Hooker telescope and used a 5" diaphragm and a 17''5 beam separation in declination. The positions chosen are shown in Figure 1 of Paper II. It is seen there that, in general, the positions isolate individual sources as determined by the 2''3 diaphragm. The seeing disk was typically 1". The observation and reduction techniques were the same as described previously (Willner 1976); relative fluxes were obtained by assuming that the bright limb of the Moon radiates as a 350 K blackbody, and all fluxes were normalized to the 12.5 /xm flux of a Sco, which was assumed to be 1.80 x 10-23 W m-2 Hz-1. The results of the ob- servations at seven positions are given in Figure 3.

m. DISCUSSION—SOURCES BRIGHT AT 2 MICRONS a) 1RS 77, 72, and 19 Neugebauer et al. (1976) observed a CO absorption band near 2.3 /xm in sources 1RS 11 and 12 which they interpreted as originating in the photosphere of late- type stars. The 1-3 /xm energy distributions of 1RS 11 and 12, as well as of 1RS 19, shown in Figure 4, are consistent with the interpretation that the sources are late-type stars. The rise in 1RS 12 at 3.5 and 5 /xm is probably due to contamination from the nearby source 1RS 2, which is bright at 3.5 /xm and longer wavelengths. The question of whether these discrete sources are single stars or clusters is still open. The Fig. 3.—Observed fluxes from 7.5 to 13.5 /tm obtained with sources are unresolved and less than 2" in diameter. a 17o spectral resolution and a 5" circular diaphragm. Error The absolute 2.2 /xm magnitudes of all three sources bars are given for points whose statistical error exceeds 107o- The continuous lines represent fits (Willner 1976) to the are about — 9. For class III, this corresponds spectrometer observations. to one M7 star, 20 M4 stars, or 100 K5 stars (Lee 4 1970). The extreme depth of the CO absorption in sources range from 1.5 to 3 x 10 Lo, if 2500 K 1RS 11 suggests that it could be an M8 star (Baldwin, (spectral type M7 [Lee 1970]) is assumed. Frogel, and Persson 1973). If the sources are super- giants, they could contain at most six stars, even if the b) 1RS 7 stars are as early as K5 lb. The total luminosity of At 2.2 /xm, 1RS 7 is 4 times brighter than any other each source depends on the ; the source within 1.5 pc of the galactic center (BN 75).

© American Astronomical Society • Provided by the NASA Astrophysics Data System No. 1, 1978 IR OBSERVATIONS OF GALACTIC CENTER 125

X (/x) X(/x) 5 4 3 2 1.5 20 10 8 6 4 2

Fig. S.—The 1-20 ¿un energy distribution of 1RS 7 cor- rected for interstellar extinction. The ordinate is in arbitrary units. Also shown are the energy distributions of two stars; measurements and interstellar reddening corrections of HD 143183 are from Humphreys and Ney (1974), and the measure- ments of AC Her are from Gehrz (1972).

In a general sense, all three sources plotted in Figure 5 have similar energy distributions. If the additional extinction of 15% is applied to 1RS 7, the spectra rise from 3.5 to 2.2 ¿¿m, where the flux is Fig. 4.—Energy distributions of 1RS 12, 16, 11, 19, and 7 emitted by the stellar photospheres. The spectra also corrected for interstellar extinction. Uncertainties are dis- rise toward wavelengths longer than 5 /¿m. The 10 ¿¿m cussed in the text. emission seen in all three stars in Figure 5 is probably due to circumstellar dust shells. The excess in 1RS 7 is not an artifact of the extinction correction; even The presence of CO in absorption at 2.3>m with no correction near 10/xm, there is still excess (Neugebauer et al. 1976; Soifer, Russell, and Merrill emission. The apparent absence of a silicate emission 1976; Treffers et al. 1976) suggests that the source is a feature in 1RS 7 similar to that seen in HD 143183 star. The luminosity, the depth of the CO absorption, might be due to an underestimate of the silicate extinction, or the dust shell might lack silicates. In and the absence of 02 absorption (Treffers et al. 1976) require that the object be similar to a late-type, summary, it appears that 1RS 7 has the properties of a oxygen-rich supergiant star. late-type supergiant such as HD 143183 situated at the Figure 5 shows the 1-20 /¿m energy distribution of galactic center. 1RS 7 corrected for interstellar extinction as described in Paper IV. The energy distributions of HD 143183 c) 1RS 16 (Humphreys and Ney 1974) and AC Her (Gehrz 1972) One of the most interesting sources in the galactic are also shown for comparison. HD 143183 is an M3 center is 1RS 16. It has an energy distribution similar la supergiant, while AC Her is an RV Tauri star. An to the stellar source 1RS 12 (Fig. 4), and its absolute additional reddening correction to the observations of 2.2 /xm magnitude is about —9. The source is near the 1RS 7 amounting to 15% of the total would make the 10 /xm ridge between 1RS 1 and 1RS 2, and the rise in intrinsic 1.2-5/¿m energy distribution of 1RS 7 the flux at 5 /xm is probably due to contamination. same as that of HD 143183 within the uncertainties Neugebauer ei al. (1976) found little or no CO of the measurements and extinction corrections. 1RS 7 absorption in this source; it is extended on a scale of would then be intrinsically fainter than HD 143183 -2" (0.1 pc) (BN 75). The map of BN 75 showed by a factor of 3. If it is of the same spectral class as 1RS 16 to be coincident with a nonthermal radio HD 143183, then its luminosity is about 105L©. In source (Balick and Brown 1974; Lo et al. 1975); the addition to HD 143183, many other M supergiants new position measurements presented in this paper 5 are known to have intrinsic of ~10 Lo indicate that the peak of 1RS 16 may be l"-2" south (Lee 1970). of the nonthermal source.

© American Astronomical Society • Provided by the NASA Astrophysics Data System 126 BECKLIN ET AL. Vol. 219

One explanation of the nature of 1RS 16, consistent x(M) with both its lack of strong CO absorption and its extended size, is that it is a whose members are giants considerably earlier than MO. For any reasonable population distribution, the stars pro- ducing most of the 2 jum radiation are the latest stars in the cluster, in this case K5-M0. About 100 giant 3 stars, or 10 M0, are required; this implies a density of 106 M© pc-3. Another possibility is that the stellar population is dominated by dwarfs, which show no CO absorption band (Baldwin, Frogel, and Persson 1973) , regardless of spectral type. As an example, if the stars are M4 dwarfs, 106 stars—or 105 M©—are required; the density is then 108 M© pc-3. For com- parison, the centrally concentrated globular cluster NGC 6388 has a density of about 105 Af© pc“3 within the central 0.6 pc radius (Illingworth and Freeman 1974) . Thus, if either giants or dwarfs dominate the radiation, 1RS 16 has a stellar density greater than a dense globular cluster. It should be pointed out that, other than the 1.64-3.5 /¿m colors, there is no concrete evidence that 1RS 16 consists of stars.

d) Extended 2.2 Micron Emission Because of its spatial similarity to the nucleus of M31, Becklin and Neugebauer (1968) attributed the extended 2.2 /¿m emission to radiation from unresolved stars. The present observations are consistent with Fig. 6.—The 1-20 ¿un energy distributions of 1RS 1, 8, 3, this view. The average observed 1.65-2.2/xm color and 9 corrected for interstellar extinction. for the three discrete sources in which CO absorption has been detected is 2.3 mag, which agrees with the energy distributions of these discrete sources are average color of the background over a 1Í8 diaphragm. considerably hotter than the background; see also The known discrete sources contribute only about RL 73. This indicates that each contains its own source 10% of the 2.2 iim radiation which is observed in the of luminosity. F8 area. Thus the 1.65-2.2 ¿xm color of the extended The luminosity of 1RS 1, estimated from the 1-20/xm background is similar to that of discrete sources flux observation, is about 5 x 105L©; the other known to be stars. Since the 1.65-2.2/xm color of all sources radiate between 1 and 2 x 105L©. BN 75 stars is nearly the same, it can be concluded that the suggested that these sources had linear sizes and observed color of the extended 2.2/xm emission is luminosities similar to those of the planetary consistent with its being produced by unresolved NGC 7027. Because of the extinction correction, stars. particularly in the 10 /xm silicate band, the luminosities found here are about 10 times larger than those quoted rv. DISCUSSION—SOURCES BRIGHT AT by BN 75. Furthermore, NGC 7027 may be nearer 10 MICRONS (Cudworth 1974) and thus less luminous than assumed by BN 75. For these reasons, it appears that the ridge a) Ridge Sources—1RS 7, 2, 5, 6, 9, and 10 sources are too luminous to be normal planetary The map of Figure 1 shows these six sources strung nebulae. Further reasons for doubting that the sources in an arc along a ridge of 10 /xm emission. The energy are normal planetaries are given by Alloin, Cruz- distributions of these sources resemble those of González, and Peimbert (1976). From a luminosity compact H ii regions or planetary nebulae in that they standpoint, the sources are more similar to compact are bright at 10/xm but relatively faint at 2/xm galactic H n regions. They are different from most (BN 75). The sizes of about 0.1 pc are also comparable compact H ii regions in three ways. First, there is no to the sizes of dense H n regions such as the Orion indication of enhanced ionized gas at the position Nebula or planetary nebulae. Figure 6 shows the of individual compact sources, as determined from extinction-corrected energy distributions of 1RS 1 and the Ne n emission line at 12.8 /xm (Paper II). Recent 1RS 9 from 1.65 to 20/xm, and Figure 7 shows observations of the By line at 2.17 /xm tend to confirm the 8.7-12.5/xm fluxes of all of the ridge sources. this result (Paper III). Second, as shown in Figure 6, None of the sources seems to have any silicate ab- the flux density at 20 /xm is less than or about equal sorption, but if the extinction corrections that were to that at 10 /xm, whereas in galactic H n regions there applied are correct, then all show intrinsic silicate is a large increase in flux density from 10 to 20/xm emission typical of optically thin silicate dust. The (see, e.g., Wynn-Williams and Becklin 1974). Third,

© American Astronomical Society • Provided by the NASA Astrophysics Data System 197 8ApJ. . .219. .12IB 2 No. 1,1978 in compactHnregions(Gillettetal.1975a;Willner there isnosilicateabsorption,whichoftenpresent composite, showingevidenceofemissionbystars. be uniqueintheGalaxy. corrected forinterstellarextinction. redder thansourcesinterpretedasbeingstellar,but sources arenottypicalcompactHnregionsandcould Figure 6showsthatthefluxof1RS9risesfrom2.2to present inornearthesource.Itisthuspossiblethat marginal evidenceforCOabsorptionat2.3/¿m sources thatareinterpretedasbeingstars.1RS1is clusters. (Neugebauer etal.1976)indicatesthatstarsmaybe the region2x2pcisduetolow-surface-brightness from themapbecauseofinstrumental effects.There some oftheridgesourcesareassociatedwithstellar 1976). Theaboveevidencewouldindicatethatthese is nosmall-scalecorrelation betweentheextended low-surface-brightness 10^m radiationismissing emission whichisnotspatiallyresolved;additional 1.65/xm; thecoloriswithinrangeofcolors extended emissiontobemuch redderinthe5-20/xm 2.2 and10^memission.RL 73foundthecolorof At leasttwooftheridgesourcesappeartobe Fig. 7.—The8-13pmenergydistributionsof11sources Approximately one-fourthofthe10^mfluxfrom © American Astronomical Society • Provided by theNASA Astrophysics Data System 13 12II1098 X(/x) log i/(Hz)v b) ExtendedEmission IR OBSERVATIONSOFGALACTICCENTER 4 the previoussections. region thanthatofthecompactsourcesdiscussedin resolution (Harvey,Campbell,andHoffmann1976) whose energydistributionpeaksnear50/xm.The extended sourcewhichiscenteredneartheridgeand extended backgroundobservedat10and20/xm, and 1'resolution(Gatleyeial.1977)haveshownan According toGatleyetal.(1977),thefar-infrared wavelength manifestationofthefar-infraredsource. because ofitsveryredcolor,islikelytobetheshort- formly distributeddustatthegalacticcenterheated region. Thissameultravioletradiationisbelievedto which comefromthelargenumberofstarswithin is givenbyOort(1977). produce thethermalradiosourceSgrA(Ekersetal. by thehighdensityofultravioletandvisualphotons source isthoughttobethermalemissionfromuni- tended backgroundareshowninFigure4ofRL73. the otherdiscretesourcesandhasanenergydistri- It isseentherethat1RS4significantlycoolerthan 1RS 4,whichisitselfextendedontheorderof5",may bution likethatoftheextendedbackground.Infact, energy distributionof1RS20,whichisseencon- be acondensationinthegeneralbackground.The 1975). Areviewoftheradioandinfraredobservations although itisnotasredtheextendedbackground. spicuously onlyonthe20/xmmap,isincludedin is theonlysourcewithintrinsicsilicateabsorption complex, becauseitisanisolated,unresolvedobject Figure 7.Itisredderthantheotherdiscretesources, energy distributionatwavelengthsbetween5and least 1.3.AsshowninFigure6,1RS3hasabluer other source(Fig.3).Infact,whentheobserved and hasalargerobservedsilicateabsorptionthanany (Becklin, Neugebauer,andWynn-Williams1973)or feature, 1RS3issimilartoothergalacticinfrared (Fig. 7);theopticaldepthofthisabsorptionisat spectra arecorrectedforinterstellarextinction,1RS3 20 /xmthandotheridgesources1RS1and9. the 1612MHzOHsourcesstudiedbyEvansand distribution, andthepresenceofasilicateabsorption the region(PaperIV)makessuchapossibilityun- presence ofamolecularcloudatthegalacticcenter sources, suchastheBNsourceinOrionNebula likely. TheOHsourcesare thoughttobelate-type uniformity ofthe2/xmextinctionacrossface neither hasbeenestablishednorcanberuledout,the associated withgiantmolecularclouds.Althoughthe Beckwith (1977).Sourcesliketheformerobjectare very similartothe1612MHz emittinginfraredstars. mass loss(EvansandBeckwith 1977).Theluminosity Mira starswhichhaveundergone alargeamountof of 1RS3correctedforextinction isabout3x10L, o Far-infrared mapsofthegalacticcenterwith0!5 The energydistributionsof1RS4andtheex- 1RS 3isauniquesourceinthegalactic-center In termsofitssmallsize,relativelywarmenergy c) 1RS4and20 d) 1RS3and8 127 197 8ApJ. . .219. .12IB 63 resolved (BN75). type giantswhichhavelowerluminosity.1RS16, center sourcesarestars.Themostprominentofthese, like thoseoftheridgesources;ithasbeenspatially 1RS 7,isinterpretedasbeingalate-typesupergiant; distribution (Fig.6)andasilicateemission7) which isthesourcenearestnonthermalradio other assumedstellarsourcesappeartobeverylate- This mightthereforebethepositionofhigheststellar spectral-type giantsorofdwarfs.Ineithercase,the source, isproposedtobeaclusterofrelativelyearly- emission inthosesourceswhichradiatemostpromi- density intheGalaxy. mass densityimpliedisgreaterthan10Mpc“. 128 nently at10¿anandlongerwavelengths.Manyofthe .1975,Ap.J.{Letters),200,L71(BN75). ——. 1969,Ap.J.{Letters),157,L31. Allein, D.,Cruz-González,C.,andPeimbert,M.1976, Becklin, E.E.,andNeugebauer,G.1968,Ap./.,151,145. Baldwin, J.R.,Frogel,A.,andPersson,S.E.1973,Ap. Becklin, E.E.,Neugebauer,G.,andWynn-Williams,C.G. Becklin, E.E.,Matthews,K.,Neugebauer,G.,andWillner, Balick, B.,andBrown,R.L.1974,Ap.J.,194,265. o Cudworth, K.M.1974,A.J.,79,1384. Beckwith, S.,Evans,N.J.,II,Becklin,E.E.,andNeugebauer, Gehrz, R.D.1972,Ap.J.,178,715. Gatley, I.,Becklin,E.E.,Werner,M.W.,andWynn-Williams, Evans, N.J.,II,andBeckwith,S.1977,Ap.217,729. Ekers, R.D.,Goss,W.M.,Schwarz,U.J.,Downes,and S. P.Willner:DepartmentofPhysics,C-011,University ofCalifornia,SanDiego,LaJolla,CA92093 Gillett, F.C,Forrest,W.J.,Merrill,K.M.,Capps,R.W., Harvey, P.M.,Campbell,M.F.,andHoffmann,W.F. Gillett, F.C.,Jones,T.W.,Merrill,K.M.,andStein,W.A. of Technology,Pasadena,CA91125 E. Becklin,K.Matthews,andG.Neugebauer:Downs LaboratoryofPhysics(320-47),CaliforniaInstitute The simplestobjectstoidentifyamongthegalactic- The natureof1RS8isunknown.Ithasanenergy Heated dustislikelytoberesponsibleforthe Ap. J.,205,74. 184,427. S. P.1978,Ap.J.,220,inpress(PaperIV). G. 1976,Ap.J.,208,390. Rogstad, D.1975,Astr.Ap.,43,159. C. G.1977,Ap.J.,216,277. and Soifer,B.T.1975a,Ap.J.,200,609. 1973, Ap.J.{Letters),182,L7. 19756, Astr.Ap.,45,77. 1976, Ap.J.{Letters),205,L69. © American Astronomical Society • Provided by theNASA Astrophysics Data System V. CONCLUSIONS BECKLIN ETAL, REFERENCES thin. Astellarcomponentappearstobepresentin those ofextremelyredOH-emittinginfraredstars. properties oftheisolatedsource1RS3aresimilarto dust producingtheinfraredradiationisoptically Other sourceshavesome,butnotall,oftheproperties 20 /onisbelievedtobetheshort-wavelengthextension of compactHnregionsorplanetarynebulae.The 05-002-207 andNationalScienceFoundationgrant frared groupfortheirsupport—inparticular,S. star clusters.Theextendedemissionseenat10and some ofthesources;theymightthereforebedusty of theradiationseenat30-100/on. Aeronautics andSpaceAdministrationgrantNGL and J.CarrascotheMountWilsonnightassistant AST 74-18555A2. Werner. ThisworkwassupportedbyNational Beckwith, J.Bennett,T.A.Boroson,Elias,G. E. Hancock.WealsothanktheentireCaltechin- Forrester, I.Gatley,S.Hage,A.Pruett,andM. Humphreys, R.M.,andNey,E.P.1974,Ap.J.,194,623. Illingworth, G.,andFreeman,K.C.1974,Ap.J.{Letters), Neugebauer, G.,Becklin,E.E.,Beckwith,S.,Matthews,K., Low, F.J.,Kleinmann,D.E.,Forbes,F.,andAumann, Lo, K.Y.,Schilizzi,R.T.,Cohen,M.H.,andRoss,H.N. Lee, T.A.1970,Ap.J.,162,217. Neugebauer, G.,Becklin,E.E.,Matthews,K.,andWynn- — .1978,Ap.J.,219,inpress(PaperII). van deHulst,H.C.1946,RecherchesAstr.Obs.(TUtrecht, Treffers, R.R.,Fink,U.,Larson,H.P.,andGautier,T.N., Rieke, G.H.,andLow,F.J.1973,Ap.J.,184,415(RL73). Rieke, G.H.1974,Ap.J.{Letters),193,L81. Oort, J.H.1977,Ann.Rev.Astr.Ap.,15,inpress. Soifer, B.T.,Russell,R.W.,andMerrill,K.M.1976,Ap.J. Willner, S.P.1976,Ap.J.,206,728. Wynn-Williams, C.G.,andBecklin,E.1974,Pub.A.S.P., We thankthePalomarnightassistantsG.Tuton and Wynn-Williams,C.G.1976,Ap.J.{Letters),205, H. 1969,Ap.J.{Letters),157,L97. 188, L83. L139. {Letters), 207,L83. Williams, C.G.1978,Ap.J.,220,inpress(PaperIII). III. 1976,Ap.J.{Letters),209,LI15. 1975, Ap.J.{Letters),202,L63. Vol. 11,part1. 86, 5.