University of P.!Innesotr? 116 Church Street, SE Minneapolis, Rli?

University of P.!innesotr? School of Physics znC Ast-ronony 116 Church Street, S.E. Minneapolis, rli?nc?soLa 55455

Invited Rcview Papr-r at I.A.U. Syn~osiun?42

nambera, F!?G Scptcrber 1377

KEY BJORDS: Lntc Typc Stars

Infrared Spec trophotonetry

Circunstellnr Dust

Address for proofs: Uni vcrsit:; of :!i.n~.nsota School of 2hys.i.c:; an? Astroncny 11G Church Street, S.E. P.!innca?olis, ?,',inrl~~;ot~.554'-55 ABSTRACT Substantative mass-loss resulting in appreciable circumstellar dust envelopes is common in late-type stars. The evolutionary history and physical state of a cool star determine the chentistry within the outer stellar atmosphere mirrored by the molecular and partic:~latematerial present in the envelope. The 9. b observational consequences of this debri determined by moderate spectral resolution (X/AX - 50-100) infrared spectrophotonetry are reviewed. Significant informtion is provided by observations of the emergent energy flux of both the cool stellar photosphere and of the

L circumstellar dust envelope. While most of the infrared features observed in late-type stars can be plausibly identified, an as yet unknown emission mechanism marked by strong infrared band structure is known to be opera- tive in a variety of infrared sources. The observed range in &ell optical depth in both M and C stars b suggests that mass-loss occurs to some degree throughout late stellar evolutionary phases and that occasional periods of high mass-loss are not uncommon. 1 +- I. INTRODUCTION The mapping of pathways followed by stars undergoing post-main-sequence evolution is a task vital to an understanding of the evolutionary history of the matter in our galaxy - whether to be presently aggregated into stars or distributed in the form of gas ancl dust. During the red giant phaae the outer atmosphere becomes greatly distended in response to conditions Jeep within the stellar interior. Such cool, tenuous, extended envelopes provide ideal . conditions for the formation of gaseous molecules and particulate condensates. Apparently appreciable mass is lost by stars throughout post-main-sequence phases, whether by the quasi-steady outflow from both giants and supergiants, the impulse driven expanding shells of planetary nebulae, or the explosive events seen in novae and supernovae. The final evolutionary state of a star - be it white dwarf, neutron star or even black hole - may well be. determined by the degree to which the .star is able to return mass to the interstellar medium i

: from which it came. The material is returned metal- i enriched by nuclei synthesized in the interior and later j important mixed out, so that the debri provide clues 4t about stellar structure and nucleosynthesis. Not too j surprisingly, these later stages of stellar evolution

* are often veiled from exploration by standart! optical techniques when dust condensation in the ejected material modifies the radiation field of the centraS star through extirlction and thermal re-emission. Stars undergoing extensive mass J?ss tend to become infrared rather than optical sources. Radio maser line and continuum emission from the ejected molecular clouds is not uncommon. Thus the study of mass loss, late stellar evolution and nucleosynthesis as interdependent phenomena requires definitive infrared observations. Present difficulties in developing a cohesive : theoretical understanding of the process of late stellar evolution (including possible alternative pathways) have been exacerbated by a dearth of hard observational evidence to provide the necessary guideposts. However, recent advances in detector technology have permitted high spatial, temporal and spectral resolution observations with sensitivities sufficient to include a large and hopefully representative sample of late-type stars. Late-type variable stars are seen to be strongly interacting with their immediate environment by way of net mass outflow. What, then, is the nature of the matter being returned to the interstellar medium? What is the nature of the mass loss mechanism? Why do sone objects appear to be almost literally choked in their

-- own dust? The present paper will attempt to provide a brief exploration of tho observational data which have an impact on these questions. Since moderato resolution spectrophotometry (X/AX - 50-100) has been obtained for a fairly large (and hopefully representative) sample of late-type stars (see Merrill and Stein, 1976a,b,c (hereinafter WSI, 11, 111) for a recent summary), and 6 &) is most readily available to the author for illustrative purposes, the present discussion will deal primarily with such data. Except whexe noted, the narrow bandpass (X/AX - 65) spectra ehow were obtained at the University of Minnesota-UCSD Mt. Lemon Infrared Observing Facility.

COOL STELLAR PHOTOSPHERES Observations of cool stellar photospheres provide I a window into the interiors of late-type stars and reveal the nature of the material available for mass outflow. infomation on opacity sources and energy

b mechanisms should enable a better understanding of stars undergoing extensive mass loss. The rapid increase in molecular abundances with decreasing effective temperature results in complicated optical ssectra which are not yet understood in detail (Fujita, 1970).

Molecular dissociative equilibrium calculations (eg. Tsuiji, 1973; Scalo, 1974; Johnson, Beebe and Sneden, 4

1975) suggest that affective temperature, luminosity class and relative atomic abundances should produce observable differences in the emergent photospheric emission in late-type stars. In particular, the lowering of the oxygen/carbon ratio differeltiatas red giants

into spectral types M, KS, S, SC, and C. Polyatomic species are incraasing1.y important at temperatures below 4. -1 2800K in comparison to the better studied diatomic species. Molecules provide the dominant opacity source. The observational picture is summarized by the spectra in Figure 1. (R Lyn and Y Lyn were taken at Mt.. Lemon by B. Jones and S. Willner. The rest are in MSI.) AS has been demonstrated by Baldwin, Frogel and Perssoa (1973) the strengths of the absorption bands due to H20 at 1.9p and CO at 2.3~correlate with spectral type .(both deepening with decreasing temperature) and luminosity in normal abundance ("oxygen-rich") stars. Clearly the 2.7~water absorption band, noticeably present at low resolution by type M3II1, dominates the spectra of cool semi-regular and Mira M type variables. Variation of the strength of the infrared steam bands with phase has been confirmed in detail in the 1.2-4.0~ spectra of R Cas (Strecker, Ericson, and Witteborn, 1976 aboard the NASA Kuiper Airborne Observatory - KRO) in agreement with the optical results of Spinrad --et al, (1966), Watt-r krbss-rptdiairin tire! itp per giant. VS Sgr (Figure 5) is quite pros~ounct*d and is clt~~trlyscbt*n

1,9p it1 S t1es (St r~l~k~f:~k nl 1976) , Mlrr -* '

For aksrs wf th 06% itoar utrity , t llcz Cc) b.~nd is

incredibly etrorrg 4S1nyt.h .I*urat: ale,-. 1971). Ubaervafion,ll

coveraga of S stars fa f.lr from complcko (prompting ,\

recant plea fur dntn by Wing and Yorka (1977)), ft hi~:t 'i bean suggastad an tha baais of near infrared optical spaetra that B stars nra water dsf icient: cornparad to M giants a.P the same tampouattrru (Spinrlsd ut a1 Isf"c<)). rwr -* , Certdinly tho MS stars Y Lpn (Ffgura 1) n1k1 Hs Cnc

(Figure 5) show w.xt-@r absorption, wllcri~as the stror~gfi

eittrx: R Lyn (Figure 1) olrsws al11y cnttrrut~-rICO nb::or~~Liot~

at 2.3i1,

Tltrer sy;recttra of cnrbc.r~~-ricI~stars irro market\ by diskinetivtl? absorption bat~da (Figuru 13 , Tl~c: 3.07tr b.~nt\ (first not-e?d by Jolln~on c& 2L. , 19ti.8) c,xn bt~ccamt*quite ilat"p. ~st:'r?nuliv@3p obs~2xvi1tfons(Nnguchi --pt ax., 1977: Fay arrd Ridgway, 19761 EtSl; ElSXI; PISIIX) rcinforcc Cht*

suggestion og l~~wat. t\L. (1970) tlrak Clris k.\n~ll i~; - *- clistincCivl~lin cc.\ul carbon :sti\rS in goxltbr:rl. TIIL* Ll.tn~1

is not.icc.ably cl~tihlr*in Ori, L\co~dilt~il Cl.ikt. bottr~mtxi

Zikn tt~o2. 3t1 CU b*~triiStr st~trti-r1lgul.l: v,\z-int'lo:; (till Aur)

a11d LILX~~i\rrd n.1 rxrtnw .it\ $1 i r.1 v.1 r i,1).8115:: f S:; \;1~-).

I-3rt%liminnryhigh rasoltlt i ~~rrc~?~:ii.e\.,tt.i~~l\s by i;.C~iq\s.~y t3t ,I l .-... .*...- . (1976u,b) s~tggost t110 bl\nd is cuc\p~situ-- inclutlin~; varying amoutlto or liCN and C2112. Cg nay also bc logically prlzsont: (1lSI) . 1'110 systemat Cc wotrkcning af tho 2.3~CO band in Mirn varii~bloshas teen noted in detail (McC;l~r\msn, E!t\nch and N~ucjcbrruar, 1976; Frogol and Myland, 1972) and is gonarnlly attributed to veiling - possibly by thormal amisnion in n circu~nstall~rshall. h $ Other bands nra also noticeably yrc.st?nt at low rusolu-

tion in carbon Mfraa. band at X 2 ,Sir is most

pr~beblydua to tirc Av = -1 suqucnco of the C2 D~llick- Ramsay aystam (cf. the synthetic! C sttar spectrum of Quarci --ot al., 1974), whall known in c.trbon stars at shorter t~nvslongtlra (MSI; Cilrn, 1977). Tim 3.9~ba~.rcA

nay be duo to SicZ (NSX; Gilro, 1977) since tho blerrill-

Sanford electronic absorption bands of SACZ as well an

particttlt~taSIC in cmission (Gilrn, 1372 ; TraEfcra and Cohen, 1974) are alraady known to be prosent.

Some important rrspccts of tl~c2-411 spectral nbsorhitnce of circ~unst~.llarmaterial arc? s\~nut\;lrizc~lin

Figure 2 (MSX), where. the spectra of ,\n bl Elira (o Cct),

n C blircl (SS Vir) and il compact: infracr-d sourcu in a

molecular cloud (AGL 989) have been divi~lcdb.kr appropsiaka blackbody spac tra (2200K, 2150K, and 8001i uasyoctivnly) . sviclont . TI;@ obs~rvodatrangth of such bands cantrains the degrco of veiling yrosont due to cireumstallar emission. Such bands arc dotoetabla cven fn the praeenca of substantial circumst~llarthermal dust emission and provide n means of identifying late otars in denea

envelopes (&!Sf, MSII, MSIII). Note in particular that although the carbon star band and the "icen band t k (apparently seen primarily in objects embedded in dense molecular clouds; see Morrill, R\rseell, Soifer 1976) appear at the same wavelength, their spectral shapes are demonstrably different at modcratc rssolution. It is interesting to note that the long wavelength wings of the steam and "ice" bands are similar. Observations of the XX4-8p spectra of late-type c and M stars from the ItnO have been shown the prosence of the CO fundamental at 4.6vm (Russell, Soifar, Farrast

1975; Puetter et &. , 1977; Goebel -ct -*a1 ' 1977). Photospheric blackbody omission is seen in the M super- * giants a Ori and )1 Cep and the M Mira R Cas near maximum

(Russell -et *Ial 1975; Puetter ot- a1-* 1977). The 6.21t steam absoxption band is seen in the b¶ blirn R AQ~noar

imum (Soifax, p.c. ) s~iggestingthat thc non-dctcution

in R Ca8 was probably LIUQ LO th~?known strang phase dcpendcncc in band strertgth, Thc. snroo th spcctrura long

of the 4,Gp CO Eunclarnt?r.tnl for t'tlc C star V Cyg (Ducttcr et s., 1977) is consiotont with vailing by circw,stollar emission (known to be largo, Forreat (1974)), whorcas

the possible detection of Cj .tbaorption long of Sp in Y CVn (Goebel et aJ., 1977) (with minimal infrared excess and infrared spectrum like UU Aur in Figure 9) is consistent with detection of the stellar photosphere. The ob'served weakness of the molecular bands long of 2~1 $1 nay #imply arise as a result of radiative transfer effects in cool stellar photoepharos or be due to veiling by chromospheric (Lambert and Snell, 1975) , molecular or thermal dust emission. Ultimately hillla spectral resolution observations will enable the identification of weaker molecular features in emission and absorption such as SiO in M stars

(Hinkh et e.,.I9761 Cudaback et 1971, Wollnan -et -*a1 ' 1973) sand determine the isotopic abundance ratios in evolved stars (Rank, Geballe and IJollman 1974; Hall ot a1 - -* ' 1977; Barnes p-$ ale,1977).

311, CXRCUMSTELLAR DUST ENVELOPES IN LATE-TYPE STARS

A* OrnWIEW Excess radiation above that expected fzom a cool stellar photosphere of the temperature indicated by optical spectra is common among late type variable stars.

Although it is generally believed that much of this cxct3un in ittt.ributabla to tlrormal rc-omidsian by c t rcum-

titulhr dtrat (Woalf, 1373) , pl.~erna proct.u8t*s it\ tht.

srrvallopc) (Gil~nnn1974 ) ar clrromrlupl~erfc amission (I,a~;bart and Snall 1975) may also contribute aubstnnti-*lly to tho omargent flux.

Optically thin omiaaian from silicates in N stars (~oolf'and Nay, 1969) and silicon c~lrbidain cartan

%\ etara (Cilrn 1972t Traffers and Cohsn 1974) nro gonerally recagnized and the spectral shapa of the oxcsss onissfon has been examined in dotnJ 1 at rrwilurala rosalutiun in u large number sf eool variable stars (Forrest, Gillctt and Stain 1975). The ncud far a festuraloas component such

8s carbon in C stake has nlbo been racagnizod (TraEf,-re and Cohen 1974, Forrust --at al. 1975). Explaration of the optically peculiar or otlrc?n~ise3 uniduntifiod objects in spatially unbiased infrared survey catslaga has pushed tho samplc of Mira class 1uminosit.y variablo stars out to 2 kpc and cni~bl~dtho idantif ication

sf n tviclo range in apparent she31 optical dcpth in tlia circu~tstallnrdust cnv~~laycsin 1at~typcstars.

Systematic studies OE ~bj~ctsill the 211 S!cy Survt-y-- (Neugeb.~uarand Laighton, 1969) by Iiyland --.ct a1 . (1972), Strpckcr arrd Ncy (1974;1,b) , PIST J un~lothe~:a ha-l\v citn f irn!txl

nturteronu clc..~rlyidt~~ t i f i.lbl c? (tra:;~x3 '711 og t. J.c.11 .~nct n~*=\r

inft-arc4 t.c~:lini.qtw:;) I.i~t.c*-t~ilt*st.,lr:; tt'it.11 '?r?n:;c* circumstallar ee.valopes, For the objoctn in tho rocket- borne APCRt/AGL Xnfrilrud Sky- Surwy (Uulkce and Price, 19751 Prico and Walker, 197G), although optical techniques are often useful, infrared classif icntion techniquoa arc necessary to identify the staro in the densaat envskopes (eree Low a.,1976; Gehrz and liackwall 1976; Lobofsky et a1 1976; Lobofsky and Kleinmann 1976; Coh~nand Ruhi, rrll *' t 1976 and 1977, Allan st -*a1 1977 and MSIII). I) - ' As a persepctive on what follows, the broad bandpaso energy distributions of two extreme classes of infrared objscts found to be late-type stars are shown aubjetlt to arbitrary normalization in Figure 3 Typical cool giants (FLgure 3a) emit nearly like blackbodies (salid curves) regardless of O/C ratio. The infrared excess in such stars contains only a fraction of the total flux and photoepheric emission dominates the emergent flux. In sharp contrast, stars which resemble the M stars IK (Nbn) Tau, V1489 (NMt) Cyg or tho carbon star CW Leo (IRC $10216)

(see eg. Strecker and' Ney 1974a,b) have their energy distribution over a wide spectral range or resemble cold blackbodies, Thermalized enorgy rc-distribution by

he energy flux XFA = vFv (W/m 2 ) is also proportional to photons/unit wavelength. Total blackbody emission is 1.36 times the peak XFA and tho total 10p silicate flux for optically thin emission is ,343 tines the peak MA. radiative transfer in cxtanuivo dust env~?lop~asor othax omionion yroceseao uro clanrly prcsant. Such bra14 band-

pass spectra hnvo reccrttlp s\~gy~stcdthe promr\Cc of it11 additional circumstallar (silicato?) ornloaion b.-ind at 33y in 1.i atcrro (Ilagen, Simon and Dyck 1975) and ttrca possibility of non-silicate omitrsian in S stars (Tiromas, Robinson and Hyland 1976). t \ Figure 4 (MSIXI) illustrates th~energy distributions of abjocts from the AGL Survey. (Inclividuol spectra arc identified in Table 5 of MSfII.) Tho objects in Figure 4a exhibit black body-like spectra and are primarily cart-.?r,stars while those in Figure 4b sllow evidence of 9.71) silicate emi,seion or at~orptionand arc primarily M stars. The objects in Figure 4c form a class of objcct~ embedded in dense circur;lstallar discs of uncertain ~volutionarystate and will not be discussed hsrc.

8 SUPERGIANTS As examples of optically tkin shc13 cm~ssian,thc

AX 2-14~1spectra aE luminous cool aupcrgiantn ilre shown in Figure 5. Characteristically broad 2.3~ CO band absorption is soen in all. Stcam absorption is present

in the cooler objects (S Per, VS Sgr, RS Cnc). For the

superpovli tion of n, cool blackbocly-1 ikc plro tosp!li?r,? ant? optically thin silicate emission closely akin to that seen in emis~ionnear the Trapezium stars in Orion (Ney and Allen, 1969; Fcrrest et &., 1975). This excess is strongest for luminosity class fa whereas les~luminous etars uuch as CE Tau (M2Ib) show little excess. Wool f (1973) has discussed the onset of eilicate emission with 8pectral and luminosity class. IRC +60370, a highly luminous K star (Humphrey. and Ney 19741, has onG of the mtrongest 9.7~ silicate bands known*

The spectra of S Per and VX Sgr (Figure 5) are qualitatively different. The overall trend in their energy distributions toughly parallels a free-free slope. Not only are the optical lines veiled in these stars (Hui. '-reye1974) but at least in S Per (M3eIa) the 2.3p CO band appears to be weak in comparison with RW Cyg (M31a). The origin of these differences, whether duo to radiative tracvfer in dust or some plasma emission process (~ilman1974) merits further study. The sharp upturn near 4pm in the spectra of S Per and IRC +60370 (discussed in MSIII) remains unidentified but might be due to SiO emission.

C. M STARS The observed range in apparent shell optical depth in M stars is summarized in Figure 6. (AGL 2205 is from P'orrr*t;t pk f~l*,, L977. tt'l\rt !;pa~:t-J'ntrot prulat.t~:tr Lrt PIST,

'~lleco;~binvrl CO-1120 abnorpt ion b.\ndu eiyonr iir rt ll t hu

unambSguauaXy ddanthfiad n@ M typ3 vz~riablastars on tha 1 baais at! optical or nunr inf rnsd clnssif Sathan txchnhquos,

prassnco of the CO-H20 band sboorptlon. The 9.714 oilicnkg band incrdassm Ersm nlodsra ta

(o Cst) to atrang rzmisaisn in tha Mira '1% Car,\ (or tlm aupargiank VX Sgr of F*kgura 5) and sfnrts into posudbla aali absorption in XK (NMC) Tau. Aa tho nyparant 1 color turnp~raturndaaroawas, the 9,711 band undergoes I distf net usXf -absscptf on in the bnc3 cantar while still

itr net amiseion in tlza tvhnght in NV Aur (IRC +!50137) and

V1189 (NU) 6yg (alaa in XRC -10529 and IVS Puo (XRC +3QObl)

(EISXX) sot ahown ham\. Xn ACL 2005 and AOL 2290 the 9.7)1 band is daaply into n~tabsorption. PAnnXXy in

the apparent 2-lit solar tawyurnk.ura n chilly 375K. T~G?

Ebuv from all t.tltrt3~ objoet'o vnrit~ttt~ith El.Lra eldsta tilt\@

scales. AX1 bttt AGE 2t385 nra re-cngnizc)d Qlf rtiasar rndio The 2-40~spectrum 02 .PGL 2205 is shown in Figure 7 (Forrest ot al., 1977). In a situation analogous to that of the 9.711 eilicate band in IJtt Aur the 18~silicate band appears near the point of net absorption at the band canter. The total flux of this star varies on Mira time rcales. The color temperature decreases and the depth of the 9.7~band absorption increases near minimum 4 *) suggesting the overall cooling of dust in the envelope in rerponse to variation8 in luminosity of the central .tar. As an OH Tye I1 (AnBersson et s., 1974) maser undergoing cyclic bolometric lumfnosity changes, and possessing a silicate dust envelope through which remnant CO-H20 absorption can still be seen, AGL 2205 is almost certainly an M type variable star undergoing extensive mass loss. The 1800K curve results from a simple de-reddening of the observed flux by the observed 2-4p extinction curve determined towards VI Cyg 812 (Gillett it al. 1976) and is plotted to illustrate the residual' presence of the CO-H20 band. The overall energy distribution is psedictz3ly not well fit by a blackbody. Recent infrared surveys of OH sources by Schultz, Kreysa and Sherwood (1976) and Evans and Beckwith (1977) suggest such OH/IR stars in dense dust envelopes are not rare. Other AGL candidates include AGL 230, 1822, 1992, 2019, 2086, 2174, and 2188 to name but a few (see, eg., Lebofsky $it.>, r?lt arld l!t.,il to t kt? int e~rtt01 tar l\\w~iunl,

propertias of minaunt aiLicfit~~haw grn~lunltyh~d to a

*& \ battar understanding of ths circwi\atallux ailicaka barld

tltrPBman 1974 ; tlunt and Salisbury 1974, 1976: Rtiikowsli2 and Knacka 1975; Railkawskl, Knaekt? and Parco 1975:

A ganarably nr\torghaua, rnnd~migodlow ardcr crystal

uCruc tura ae(fbn\~it.a CCIIRF~ cle~slt?sttm cl\\pkic;z king tl~r?9.711 nbsorptiun band, Recnn tly Rase (1977a) haa adtQmpt~*d to &uplicn:.e the infrared ~missiansyt3ctrn of coiiwts by

diuoct lahnr;\tory tttacrsi\rr?~11ent:8of qrain t?lq\i:3~io11.

%'ypir=:llr@%u'Lts for *d00# grains arcl t;trawr~ in P.i.rl\\~~&I.

TIla 9,711 r?IRis~i~nhatld :;am .in tho OriOtl T.~^it;h.*%i\~t\t"~*qir?;l

sl~ofsrrhara for compauisan (riguk*t3 8) uc7a:;st1;tbly maLL-han

tka- ubs~*r\*t?d~~~e9k.s V~?N~UY~C~H in El st,^^:; (k'drrc*:qk a*c*t - A. G.,

197fi). Mxr~3\-tlcst\t,,fits td ttllt* ~>hsel:t*rsrl9.711 ;ib:;orpt.i.t>l\

.in\ C\ vnri~?tyuS j,nflra;r\c? sc3tti:~:c~sitrt? pc~s:fib.l~?f:0r ,111 assumed 250IC Trapezium omissfan tcmporrrturo (Gillott, Forrest, ot s., 1975). Aesuming such an isothermal cloud as a firat approximation (an attempt to allow fox I. r temperature and density gradients in the Trapezium would

yiold somewhat dif feront results (Dadijn, 1977)) , tho 9.7p band translates as indicated from grains at 250K to 350K and 600K. The 0-2p sized Montmorillonito emission is a very good match. Zaikowski --et al. (1975) have previously commented on the close match of such phyllosiPlcates to observed absorption spectra in n variety of galactic sources. Rose (1977b) concludes that several reasonable mixtures of high and low temperature condensates as amorphous high temperature condensates provide reasonable fit to the observed comet spectra. Such ensembles of varied silicate grain size, shape, and I composition also fit the 9.7~emission band (see also 3. Forrest 1974) as can be inferred from Figure 8.

E. C STARS b The observed range in slw3.1 optical depth in carbon

stars is shown in Figure 9. (AGL 3099 was taken at KPNO by Gehrz (p.c.), the rest are from MSI, MSXI, MSXII.)

Cool carbon stars such as UU Aur have black body like continua. Strong SIC emission is seen near 11.2~in UU Aur and the Mira R Lep. The 2-4~color temperature r~ddaneae tha energy distribution gradually thurmalizas

to a cold SGOOK npparutlt: blackbody tonrpaxntx~ra. T~IQ3k1 absorption bande are clearly viaibla. Unlike the M stars, no new ~bsorptionbande appear with increasing ,shell optical depth. SIC is clearly sosn in all cnoes shown (AGL 865 data are too noisy to tell). The primary constituent of the dust in carbon-rich envelopes is t "1 relatively featureless (Forrest st;-- al. 1975) ancl is generally considered to be carbon (implying an emissity 2 oP &/A for mall grains). ! The propertiara of rich tnolacular envsl~peaof stars i like CN Leo (IRC +1021G) have been recently summarized by M~rria (1975). High resolution 2p spectroscopy of CW Lao (EIaLl --at al. 1977) has datuctad CE) in the shell I and along with data at 4.7~Rank at al. (1974) and radio - I- data are beginning ta establish isotope ratios in tho evolved material. A mass of ~O'~M~ is suggested for the t molecular cloud (Hall --et al. 1977). Again, such dense * expanding CO clouds axe not rare (Zuckamnn --et nl. 1977). A significant fraction of tho AGL Survey sourcas are such carbon stars. Such carbon stars are returning

condensed carbon and Sic, gaseous C0, and other metallic nmlecules Lo the interstellar medium. ZV. INVIMIa:O gMXS820N DUIIXNG LATG EVU1,UTIONARY PIIASES

A. NOVA (NO) VULPft:CUW\ 1976 %'ha 2-8rtm wgactrum dusing an early stage in the development of Nova (NQ) Vul 1976 is ohown in Fiyuro 30.

the XX 2-411 (Prom Mt. Lclriunan, 31 Oct. 1976) and t21a AX 4-8p (Soifer, p,c, from tha KAO, 5 Nov. 1976) spectra hava bean matchad according to the appropriate over- h b lapping broad bandpass data. During thio etago the overall ansEgy distribution showed variable, approximately frae-free infrared amiaeian and strong viaual flus (Nay and Makfiald, 1977; Nougabauer, Gehrz, Soifer, p.6.). Nun\eroue hydragem lines (probably blanded at this rasalu- tion (X~AX- 65) with helium lines) ate seen, including the Pfund serics limit The line at 3.37ki remains unidentified. As was notad in Nsvn FN Ssr 1970 (Gaiaal,

-Kluinmann and IBWr 1970) the 14.9pI band was clearly stranger than could be acco~lnted for by l~ydragon lines. The XX 4-8p spectrum cortununicatcd by Soifor h (p.c.) clearly exhibits a atrong, braad emission feature Eram the atmospheric cut-sn at 4.5k1nt t~ 5.4~and beyond.

Groundbased Sy spectra by Gahrz (p.c.) confirm tho band emission. Ovaxlupping broad bandpaas absarvatians an the twa days suggest the spectra are proporly narrniklizad to each other to within about 215%. By comparison wikh

llumplrreys or(M 6-5) the lrydragan lines ara n minor paturbation an. the band. a. Black (p.c.) has suggested Cj e~nissionas baing rosporrsiblcs for this brrt\d. The band diorrpponrod in tho broad bandpass dakn just prior to the onset of grain formation near day 64 (Nay, 1977; Nay and Hatfield 1977). Later spectra (Gehrz, p.c.) show a cool blackbody and are consistent with subsequent broadband data which indicate approximately 900-1000K blackbody emission.

It is tempting to suggest that a super-sat@*~atedCj vapor rapidly concensed into carbon.

B. IIM SAGITIPAE The remarkable omirzuion-line object WM Sagittae, first reported by Dakuchaera (1976) as an object which had brightened over 5 magnitudes in two years, has been suggested as a planetary nebula in the early stages of Formation akin to V1016 Cyg (Stover and Sivartsen, 1977; \ (i Ciarti --ot ale 1977). The broad bandpass onaxgy distwibu- tion during June 1977 is shawn in Figure 11 (Davidson, Humphreys, and Merxill, 1977) along with that of VlOlG

Cyg and CRL/AGL 1274, a possibly related object which is still visually faint (Allen at-- ale 1977 and Cohen and Kuhi. 1977). The 9.7~ silicate band (confirmed by selected narrow bandpass points) is apparently seen in emission above a lOOK blackbody-like amission. It is parhnps significant that the nova NO Vul and HI\I Sge differ in the absence and tho presence of milicatas respectively but are alike in the apparent color temperature of the bulk of the obsezved emission. Apparently temperatures near lOOOK permit substantial grain condensation.

C, UNXDENTIFXED BAND EMISSION Having hopefully impressed everyone with our t \ aollective wisdom concerning the state of the material returning to the interstellar medium through mass loss from late-type stars, it is now necessary to impress upon everyone the profound depths of our collective ignorance. The concept of assorted simple molecules coexisting with refractory materials such as silicates, graphite, and silicon carbide (or possibly even poly- saccharides as has been suggested by Hoyle and Wickramasinghe, 1977) is quite appealing. However we are unable to account for what is gradually becoming known as one of the more ubiquitous emission processes observed in the infrakedl Infrared spectra of the nuclei of the galaxies M82 and NGC 253, assorted planetary nebulae and galactic HI1 regions and certain peculiar stars are marked by strong, unidentified band structure which often dominates the observed energy flux. (cf. Gillett et-- al. 1973; Gillett, Kleinmann et-- al. 1975; Grasdalen and Joyce 1976; b1srri11, Soifsr, Russ011 1975; Cohan st al. 19751 II- MSIIZI Kleinmann at &la 19378 Russell, Baifar, Marrill LI- 1977). New sources are continually baing confirmed, representing apparently divereg evolutionary atatee,

tsmperaturee, daneitias and axcitation conditione. Yet, as is true for the 9.7~band, tho band cantroide and opectral shapee appear to be remarkably constant, Thia challanging state of affaire is summarized in %\ Figure 12 where the 2-13p spectra of representative objects axe ahown. (Data shown originated as follows in the 2-4p, 4-8~and 8-13p regions? HD 44179 (Rueeell, Soifer, Merrill 1977; Russell --et al. 1977b; Cohen --at al. 1975),-NGC 7027 (Merrill, Soifer, Russell 1975; Russell et-- al. 1977a'; Russell et-- al. 1977a), and ADL 3053 (~oyce (pc, at KPNO), Soifer (pacefrom the KAO); Marxill, (unpublished).) The nebula NGC 7027 has always been a challenge. The observed spectrum is nearly orthogonal to the anticipated free-free continuum. Extensive far infrared (Telesco and' Harper 1977) and CO radio emission (Mufson, Lyon and Marionni 1975) are also prcsent. HD 44379 (Cohen --et al. 1975) is an apparent low excitation nebular binary system imbedded in a dense circumstellar disc. AGL 3053 is a region of extended infrared emission (Joyce --et al. 1977) near the compact radio source surveyed by Israel, Habing and do Jong (1373) in Sharpless 159. Excltid.incj Erain this discueniota the aa?lsar.'ttbdat rtn*ic forbidtton and poxmi t tcd lin~!r;nrrd tl~al~cul'rrhyJrorjc*n lin~u seen in NGC 7027, I draw your attontian ta bmds :it 3.3ri

(with ahouldax or soparato band at: 3.411), 6.211, 7.8tif 8.611, and 11.211, Although tho origindl sugg~ction (Gillatt --ot a1. 1973) that tho 11,211 Pwatura might be due to mineral cnrbnakas has boon suypcrted by laboratory investigations (Drogman nird nntrk 1975: Pcr~man l976b) t91o absence of the strong v3 fundnmcntal nonr 7ktm is hard to explain and tho association of the 8.6\1 feature with mineral su:.fatcjs (Calran st ale 1975) irr sirnilarily -."I-- difficult to support. At: sasoltrtion X/AX p 200 botlr the 8, Gu and 11.3~Ecaturoa remain coritinuous (Brocgnan 1977, Dregmatr and Rank 1975). T!ra 3,311 Pcclturc dizr;cusr;sd at length by Russell, Soifar and Morrill (1977) and Gradslen and Joyce (1976) ramains continuous at higher rasolukion (X/AX * 300 and possibly even at 300Q). This lack of structure, along with the observed uniformity of band shape and csxrtroid far an apparently t~idcrango oE physical condi tians, argue agnina t idsntifiea tion wi tla a molecular band (such as CIVCH' ~~uggestodby Crasdalcn and Joyco (1976)). Observations aE NGC 7027 and EID 64t179 from the NASA ICAO (Itlurtscll, Saifer, and IVillner 1977i1,h) have added tha bands at 6,2p and 7,8\1 ta the list, The feature6 remain tantalf zingly unidentif ied. Although the present state of affairs is hardly com~rrroble, one cannothalp being reminded of tho situation that

exi~tednot 80 very long ago with the discovery of *nebuliumu in planetary nebulae. Any ~ucce8sfuldezscrip- tion of the observed phenomona must account for all the data - thr bandr which are not prorent as well aa those t which are present.

V. SUMMARY Table I sunu~arizeothe principal well establish&i infrared rignature8 currently reaognized at moderate spectral resolution ()i/AA - 50-100). Approximate centroid., widths (for emission), whether seen in absorption or emiaeion, po88ible cause, and example sources are liuted. Bands seen only .at high reoolution, as for example the 4p SiO bandr (Hinkle et-- ale 1977) at6 not included here. Even at such moderate resolution b the number of observed bands has continued to grow. Besides providing information on the materials in the

a circumstellar environment of late-type stare, such band structure often facilitates the meaningful claseification of otherwioe unidentifiable infrared sources (MSIII). Large numbers of late-type stars are known to be undergoing extensive mass-loss, irrespective of photospheric chemistry na determined by tho pxoducta of nuclsosyntheaiu.

Tharo ia as yot no roauan to asfltrtl\e that the upper limit to sholl optical depth and inferrod mass-lose hno yet been observed. It would appear that mnas-lose occurs throughout the lntor phases of atellar evolution. Periods of high maas-loes, markod by dense circumstallar dust envelopes, may well ba andemic as an evolved star t adjusts to the changing conditions deep within its \ interior. The observational consequences of the recycled material are dibtinctly different both in the photosphere and in the circumstellnr duet envelope as determined by impact of the physical etate and the evolutionary history of the parent star on the chemistry in tho sholl. While the compoeition of much of this matter is identifiable ! with at least some degree of confidence, the identity of a. n eignificant fraction of the material is presently un- clear. An as yat unidentified material of unknown origin provides the dominant infrared omission mechanism in a variety of infrared sources. Much can be learned about the mass-loss phenomenon in late-type stars on the basis of moderate resolution infrared spectrophotometry. However, implementation of new technology will enatlle high spectral resolution infrared observations to be used in conjunction with improved radio data to identify molecular bands, establish isotopic ratio8, and monitor rhell dynamics. Lunar occulation (Toombm --at ale 1972) and spatial interferometry (McCarthy, Low and Howell 1977; Button --et al. 1976) can determine the angular diamcrrters of circumotellar durt anveloper. Detailed radiative tranmfor calaulationr a 4 ? *in4 kinematic rhell model8 (much ao ~edijn1977; Olnon 1977; Pinn and Sbn1977; Apruzese 1976; Goldreich and Scoville 1976; Joner and Merrill 1976; Taam and Schwaztz 1976; Schmidt-Burgk and Scholz 1976; Dyck and Simon 1975; Kwok 1975; Salpeter 1974) will be required to fully realize the physical state of these dusty circumstellar snvelopeo. We have truely only just begun. ACKNOWLEDGEMENTS Thanks to B.T. Soiferr R. Gehrz, be 'Strepker, G, Neugebauer, E.P. Ney and their colleaguer for com- municating data in advance of publication. Special thanks go to R,fV. Rusrell and SOP. Willner for tracking down at ths elusive data sheet#. A very special note of gratitude goes to E.P. Ney and W.A, Stein for their t continued rupport and encouragement, Rerearch in infrared astrophysics at the Univereity of Minnesota ir ruppouted by the National Science Founda- tioar the National Aeronautics and Space Administration and the Air Force Geophysical Laboratory. Figure 1 - Relative XA 2-41.1 spectrophotomatry of late- typo stars illustrating the variation in photospheric molecular band absorptions with temperature , luminosity, oxygen/carbon ratio and variabiliky claes. Shown (top to bottom) are: a Boo, 6 Vir, RZ Air, RX Boo, o Cot, Y Lyn, R Lyn, GP Ori, UU Aur and SS Vir, identified by spectral type and "ariobi.1 ity class . Figure 2 Observed AX 2-4v absorptance in circumstellar 1 - i ent-elopes. Spectra (top to bottom) have been divided by appropriate blackbodies to better' define band structure: o Cet (2200K) an M star8 SS Vir (200K), a C star; and AGL 989 (800X) a compact infrared source in I \. a molecular cloud with the "interstellar ice" band. Figure 3 ' icelative energy flux of late-type stars with minimal infrared excess consistent with photospheric emission (Panel a) and with large infrared excess indicative of thermal-. ized emission from an extensive dusty circumstellas envelope [Panel b) . Stars shown (top to bottom) in Panel a: a Boo

(K2111), a Ori (M2Ia), UU Aur (C5,4), R Lep (C7,6t.?), Cct (gbl'lrt), 10tOOOK dt\J 24001~ blackbodies; Pat\t?l b: CW Lea (apcrr circlas) ,

IK Tau (fillccl circSas) and V1489 Cyg (squarae) , 10,0001; and 6001t bln~kbadi~~s, Figuro 4 - Halativo anorgy flux of raprosontb%tivoobjocts in the AGL InPrnrud Sky Sti~xfrom $!SIX1 (individual objocto wrs idanti%iedin Tabla I of MGIXX.) In Panel a nxc? pritrtnrily C stars; in Panal b, yrim&rily M stars (or nt

objects with O/C normal) t and in Panel C, objects sf uncertain avolutisnnry atxta surroundad by danea circumstollar disc-liko anvelopes.

raprasantntivc ltunir\ous cool. stars (M and K

suparginnts). A 33QOK blackbody and a fx~3s- 8 f rna el~prsnra included fox: con\parisorr.

total optical depth in dusky circumatullclr envelopes. Figuro 7 - Qbsarved XX 2-9011 spactrum AGL 2205 (QN26.5 + 0.6) from Forrast uk nl. (1977). - ."=-" A 375K blackbody curvo nad tht* 2-1u sy~?ctrum de-rcddaned using tho axtinc t ion law abaorv~~l

towards VX Cyg X12 nro also st~awn. .I Figure 8 - Observed relative emission spectra for small (0-2p) grains at -6001{ from the laboratory meaaurements of Rose (1977) for Lunar sol,l 15531 (L), a melted anorthite (A) ,, Montmorillonite (M) , Columbia River Basalt (8) , Dunite (Dl, and amorphous silicate (S) . The observed emission from the Orion Trapezium ib shown (Forrest --et al. 1975) renormalized from assumed 250K grains to 350s and 600K grains for compariaon. Figure 9 - Relative XA 2-13~spectrophotometry of C stars illustrating the observed range in total optical depth in circumstellar dusty envelopes. Figure 10 - Spectrophotometry of Nova (NO) Vulpecula 1976 during the f ree-f ree emission.phase,

>. taken 31 Oct. 1976 (2-4~) and 5 Nov. 1976 (4-8~). Overlapping broad bandpass data " agree to +15%.- Positions of hydrogen lines are indfiated. Energy flux at 3.0~is 3.0 x 10-I5 PI/cm2. Figure 11 - Energy flux from HM Sge during June 1977 together with possibly similar objects from Davidson, Humphreys, and l-lerrill (1977). 30

Figure 11 - Relative AA 2-13~spectrophotometry of representative infrared sources with unidentified emiesion feoturoe at 3.3~/3.4p, 6,2p, 7.9p, 8.6~ and 11.2~. Points with broad horizontal bar8 are broad bandpilss observations 1 ..r 4r - * ? .I *j 'a .I. * ' 4 k &.mMwm!'LIpII/. --."------lid&------we- - -

TABLE I

OBSERVED XNFRAlUD BANDS AT btODERRTE SPECTRAL RESOLUTION

BAND BAND CENTER WIDTH (W) (urn) abs om CAUSE tJHERE SEEN 1.4,1.9,2.71 6.2 X g:H20 M stars cool stars carbon stars g:C3rC2H2,HCN~ carbon stars "ices" molecular clouds carbon stars lines nebula -0 .05 nebulae, pec. stars 0.2 nebulae, pee. stars "0.8 carbon stars, Nova Vul 0.3 nebulae, pec. stars 0.8 ?. nebulae, pec. ctars 0.3 3 nebulae, pec. stars - 3+ silicates oxygen rich stars, ISM -1.7 silicon carbide carbon rich stars 0.4 3 nebulae, pec. stars -0.3 3 nebulae Allen, D.A., Hyland, A.R., Longmore, A.J., Cnswall,

Anderason, C., Johnnsson, G.E.D., Goss, W.M., Winnbarg,

A., Nguyen-Qunng-Riaw 1974, Astrun. & Astraphys.,

Apruzsse, J.P. 1976, Ap.J., 207, 799. UI Bnldwfn, J.R., Frogal, J.A., and Persson, S.E. 1973, ' \ AJ~J~,-184, 427. Barnes, T.G., Beer, R., Hinkla, K.H. and Lambert, D.L.

Bedijn, P.J. 1977, Ph,D. Thesis, University of Leidon. Bregman, J.D. 1977, Publ. A.S.P., -89, 335. 7

Bregman, J.D., and Rank,.D.M. 1975, Ap.J. (Letters),- -_I_195

Ciatti, F., Mammano, A,, and Vittono, A. 1977, Astron.

& Astraphys. (in press).

Cohen, M., Anderson, C.M., Cowley, A., Coynw, G.V.,

Pawley, W., Gull, T.R., Hnrlnn, E.A., Herbig, G.Ii., Nolden, F., Nudson, M.S., Jnkoubek, R.O., Johnson,

H.M., Morsill, K.M., SchiEEar, F.H., Soifar, B.T., and Zuckcxman, B. 1975, ~AJ.Y., ---196, 179. Cohen, M., and Ruhi, L.V. 1976, Publ. A.S.P., -88, 535. 1977, proprint. Cudaback, D.D., Gauskad, J.E., and Knncko, R.F. 1971, 33

Davidson, K., liwnphreys, R.M., and Merrill, K.M. 1977, A~.J.(in press). Day, K.L. 1974, A(Letter.), -192, LlS.< 1976, Icarur, -27, 561, Day, K.L., Steyer, T.R., and Huffmn, D.R. 1974, AP.J.,

Dokucharrva, OwDw 1976, 1,AoUw Information Bull. Variable t ? -Stars, No. 1189. Dorlrchner, J. 1971, Astron. Wachr., -293, 53. Dorschner, J., GQrtler, J., and Friedemann, C. 1976, Astron. Nachr., -293, 159. Dyck, H.M., and Simon, T. 1975, Ap.J., -195, 689. Evans, N.J., and Beckwith, 8. 1977, Ap.J. (in press). Fay, T.D., and Ridgway, 8.T. 1976, Ap.J., -203, 600. Finn, G.D., and Simon, T. 1977, Ap.J., -222, 472. Forrest, W.J. 1974, Ph.D. Thesis, University of California,

Forrest, W.J., Gillett, F.C. and Stein, W.A. 1975,

Forrest, W0J.p Gillett, F.C., HOUC~, J.R., McCarthy, J., Merrill, K.M., Pipher, J.L., Puetter, R.C., Russell, R.W., Soifer, B.T., and Willnet,' S.P. 1977, Ap.J. (in press) , Frogel, J.A., and Hyland, A.R. 1972, Mem.-- Soc. Royal Sciences Liege -111, 111. Pujito, Y. 1970, --Intarprotation of Spectra and Atmospheric Structure in Cool Stars (Daltimorer University Park Preea) . Gehrz, R.D. and Hackwcrll, J.A. 1976, AhJ. (Letters), -206, L161m Geisel, B.L., Kleinmnn, D.E., and Low, F.J. 1970, Ap.J. (Letters), -161, L101. t - b Gillett, P.C., Forrest, Y.J., and Merrill, K.M. 1973, ApmJm, -183, 87. Gillett, F.C., Forreat, N.J., Merrill, K.M., Capps, R.P., and Soifer, B.T. 1975, Ap.J., 200, 609. i - I i Gillett, F.C., Jones, T.W., Merrill, K.M., and Stein, W.A., 1976, Astron. & Astrophys., -45, 77. it Gillett, F.C., Kleinmann, D.E., Wright, E.L., and Capps, R.W. 1975, Ap.J. (Letters), -198, 165. 3 ; - i Oilman, R.C. 1974, Ap.J., 188, 87. f - 1; Gilra, D.P. 1972, in The Scientific Results irom the I' orbiting Astronomical Observatory (OAO-2) , ed. t A.D. Code (NASA SP-310) . 1977, in preparation. i Goebel, J.H., Bregman, J.D., Strecker, D.W., Witteborn, F.C., and Erickson, E.F. 1977, Bull. A.A.S., 9, 364. I- Goldreich, P., and Scoville, N. 1976, Ap.J., 205, 144. -I Grasdalen, G.L., and Joyce, R.R. 1976, Ap.J. (Letters), Hagen, W., Simon, T., and Dyck, H.M. 1975, Ap.J. (Letters), - =201, L81m Hall, D.N.B., Kleinmann, S.G., Ridgway, S.T., Weinberger, D.A., and lVojslaw, R.S. 1977 (preprint) . Hinkle, K.H., Barnes, T.G., Lambert, D.L. and Beer, R. 1976, AP:J. (Letters), -210, L141. noyle, P., and Wickramasinghe, N.C. 1977, Nature- --I268

Humphreys, R.M. Humphreys, R.M., and Ney, E.P.

Hunt, G.R. and Salisbury, JmHm 1976, Mid-Infrared Spectral Behavior of Metamorphic Rocks, AFCRL-TR-0003. 1974, Mid Infrared Spectral Behavior of Igneous Rocks,, AFCRL-TR-0625.

Hyland, A.R., Becklin, E.E., Frogel, J.A., Neugebauer, G. 1972, Astron. & Astrophyu., -16, 204. Israel, FOP,, Habing, H.J., and de Song, T. 1973, Astron. & ~strophys., -27, 143. Johnson, HmLm, Mitchell, R.I., Iriarte, B., and Wisniewski, W.Z. 1968, Comm. Lunar Planetary Lab., -4, 99. Johnson, N.R., Beebe, R.F., and Snsden, C. 1975, Ap.J.

Janes, T.W. and Merrill, K.M. 1976, Ap.J., -209, 509, 36

Joyce, R.R., Cappa, R.W., Gillett, F.C., Orasdalen, G., Kelinmann, S.G. and Sargent, D.G. 1977, Ap.J. (Letterr?), -213, L125. Rleintnann, S.G., Sargent, D.G., Gillett, P.C., Grasdalen, G.L., and Joyce, R.R. 1977, Ap.J. (Letters), 215,- L79 . Knacke, R.F., and Thompson, R.E. 1973, Publ. A.S,P., '5 b -85, 341.

Lsmbert, D.L. and Snell, R.L. 1975, -aMNRAS, 172, 277. Lebofsky, M.J. and Kleinmann, S.G. 1976, A.J.,- 81,- 534. Lebofsky, M.3., Kleinmann, S.G., Rieke, G.H. and Low, F,J. 1976, Ap.J. (Letters), 206,- L157. fdw, F.J., Johnson, H.L., Kleinmann, D.E., Latham, A.S., and Geisel, S.L. 1970, Ap.J., -160, 531. Low, F,J., Kurtz, R.F., Vrba, F.3. and Rieke, G.H. 1976, Ap.J. (Letters), 206,- L153.

McCmmon, D. , Mitnch, G.,8 and Neugebauer , G. 1967, ApeJ, , -147, 575. McCarthy, D.W., Low, F.J., and Howell, R. 1977, Ap.J. --'(Letters) -214, L85. Merrill, K.M., Russell, R.W., and Soifer, B.T. 1976, Ap.J., -207, 763. Merrill, K.M., Soifer, B.T. and Russell, R.W. 1975, Ap.J. (~etters),-200, L37. Me(erri3.1, K.M., and Stain, Y1.A. :976a, fxbl. A.S.P., 88, 205. (M81). II- 1976b, m, - 294 (MSII) . 1976~,$bidet 874 (MSIII) Morris, Me 1975, Ap.J., z, 603. CI- Muison, S.L., Lyon, J., and Marionni, P.A. 1975, Ap.3. (Letters) -201, L85m t leugebauer, 0.. and Loighton, R.B. 1969, Wo-Nicron b Sky Survey (NASA SF-3047, Washington, D.C.) .

Nay, E.P. 1977, Bull. h.A.8.. ma9, 316.

q NayI E.P., and Allon, D.A. 1969, Ap.J. (Lcttcra) , E,- L19tm f i Ney, E.P., and Batfield, B.F. L977, &.J. (Letters) i (eubmitted) f . i Noguchi, K., blaihara, T., Okuda, He, Snta, S., and Mukni, T. 1976, Bull. AeAmS*, -8, 526. Olnon, F.M. 1977, Ph.D. Thceis, University of Leiden. Pennan, J.M. 1976a. M.N.R.A.S.. -z, 149. 1976bI M:N.R.A*S,, -176, 539. Perry, C.H., ArwlD.K., Anaetoesakis, E., Lowndus, Rope, Rastogi, R., and Tornberg, N.E. 1972, The 4, 315, .-- Pollack, J.D. , Toon, O.B. and Khare, B.N. 1973, ;cart~~, 19, 372. --.-.-).- Price, S.D., and Walker, R.G. 1976, Tho AFGL Four Color

I Infrared Sky SU~V~J!~AFGL-TR-0208. Puetter, R.C., Rurrell, R.fV., Sellgren, K. ,-and Soifer, 8.T. 1977, Publa AeS*P*, -89, 320, Quarci, P., Querci, M., and Tmuiji, T. 1974, Astron, C

Rank, D.M., Geballe, T.R., and Wollmn, E.R. 1974, t b Ap.J. (Letterr), -x, L111. Ridgway, S.T., Carbon, D.F., and Hall, D.N.B. 1976,

I Ridgway, S.T., Hall, D.N.B., Kleinmann, S.G., Weinberger, t D.A., and Wojrlaw, R.S. 1976, Nature, 264, 345. 1 a i Rorre, A: 1977a, Ph.D. Theris, University of Minnerota, i t i Minneapolis. 1977b, .-Icarur (submitted) . Ruasell, R.W., Soifer, B.T., and Forrest, 1V.J. 1975, Ap.J. (Letterr), -198, L41. Rus8el1, R.W., Soifer, B.T. and Merrill, K.M. 1977,

1 Ap.J., 213,- 66. Russell, Raw., Soifer, B.T. and Willner, s'.P. 1977a, 1 A~.J. (Letters), (in press).

Russell, R.W., Soifer, B.T. and Willner, S.P. 1977b, i Ap. J. (submitted) . Salpeter, E.E. 1974, Ap.J., -193, 585. Scale, J.M. 1974, Ap.J., -194, 361. Schmid-Burgk, J. and Scholz, M. 1976, Ast~:on.- we- & Astrophys., 51, 209. 'lul Schultz, G.V., Kreysa, E. and Sherwood, W.A. 1976, Astron. L Aatrophys., -50, 171. 1 Smyth, M.J., Cork, G.M.W., Harris, J., and Wallace, T. 1971, Nature (Physical Science), -231, 104. Bpinrad, H., Pyper, D.M., Newburn, R.L., and Younkin,

I b R.L. 1966, ApmJ., w,- 291. Steyer, T.R., 1974, Ph.D. Thesis, University of Arizona. Steyer, T.R., Day, K.L., and Hufiman, D.R. 1974, Applied Optics, =13, 1586. Stover, R.J. and Sivertsen, S. 1977, Ap.J. (Lettere),

Strecker, D.W., Erickson, E.F., and Witteborn, F.C.

Strecker, D.W. and Ney, E.P. 1974a, Astron. J., -79, 797.

Button, E.C., Betz, A.L., Townes, C.H., and Spears, D.L. 1976, Bull. A.A:s., -8, 525. Taam, R.E., and Schwartz, R.D. 1976, Ap.J., 204,- 842. Tolesco, C.M., and Harper, D.A. %Bhomas,J.A., Robinson, G., and Hyland, A.R. 1976, M.N.R.A.S., -174, 711. Toombs, R.I., Becklin, E.E., Frogel, J.A., Law, S.K.,

Porter, F.C., and Westphal, J.A. 1972, Ap.J. (Letters), -173, L71. Treffors, R., and Cohen, b!. 1974, Ap.J., -188, 545. Tsuiji, T. 1973, Astron. L Astrophys., -23, 411. Walker, R.G., and Price, 8.D. 1975, AFCRt Infrared Sky Survey, Vol. 1, AFCR&-TR-0373. Wing, R.P., andYorka, 8.8. 1977, M.N.R.A.S., 138,- 383, Wollman, E.R., Geballe, TOR,, Greenberg, L.T., Holtz, J.Z., and Rank, D.M. 1973, Ap.J, (Letters), -e, t85, Woolf, N.J. 1973, in Interstellar Dust and Related Topic., Greenberg and Van de Hulst (eds. ) , (Reidel Pub1 . Co. , Dorbrscht) p. 485.

Woolf, N.Jb and Ney, E.P. 1969, Ap.J. (Letters) 1-155, 7

Zaikowski, A., and Knacke, R.P. 1975, Astrophys. & Space -sic., 37,- 3. Zaiko~ski~A., 'Knacke, R.F., and Porco, C.C. 1975, c. Astrophys. & Space SCL~~-x, 7. Zuckerman, B., Palmer, P., Morris, M., Turner, B.E., Gilra, D.P., Bowers,,P.F., Cilmore, W. 1977, Ap.J. (Letters), -211, L97. . ------. FIGURE 1 FIGURE 2 RELATIVE XF~(ergs ~rn'~sol)

FIGURE 5

6, I I I 1 I I I I "IP

.c . ++.*+ + +++* IRC+68370 +KBIa+ + . .

H3IcrIab I.

M Tau *+ H2Iab-Ib n

6 +. + cn ++* 0 ., *** - 2..

0, I I I I I I I I 111, 2 4 6 8 10 12 14 X (pm)

-.rr FIGURE. 6

FIGURE 10 2. I l illll

I b I b

a 8 I b 5 4 b . Nova Vul 1976 I

I I I .

I B

a b

a B

a b

4889K blw- .

4 b .. ..

.b I I I I Ill I I I I I 118

r*, 4 - /:2 a 0

C Id8 - looo Y: - m ./: + -

LI .I I 10-9- + -

m JJ * 0 HM Sge 4 % CRL 1274 + V1016 Cyg - k .I 16'~- *I 8 t' C - I I I I III t I I I t 111 J 0.3 (X4Q5 I 2 3 45 10 X(pm)