Star Formation in the Ophiuchus Molecular Cloud Complex ABSTRACT The nearby Ophiuchus complex extends for tens of parsecs in the plane of the sky and is comprised of filamentary clouds of low density molecular gas interspersed with dense molecular cores. What distinguishes the Ophiuchus clouds from other dark cloud complexes is the large (550 Msun) centrally condensed core in the westernmost cloud and the associated high density of young stellar objects. The 1 pc x 2 pc core region is characterized by high gas column densities corresponding to visual extinctions of 50-100 mag. The relatively high efficiency of star formation in the core (SFE 20%) suggests it is the formation site of a gravitationally bound open cluster. The distribution of both high and low density molecular gas in the Ophiuchus complex will be reviewed with particular emphasis on the westernmost or  Ophiuchi cloud. The corresponding distribution of young stellar objects as revealed by infrared observations and Ha surveys of the complex will also be presented. Combining data from IRAS with ground-based infrared observations has led to the identification of 78 cluster members in the Rho Oph cloud. The evolutionary state of these objects inferred from their 1-100 um spectral energy distributions and the luminosity function of the cluster will be discussed. A major conclusion from this analysis is that the duration of star formation in the Rho Oph core is less than 3.5 x 10^6 years, suggesting that stars have formed in a relatively efficient burst. Most recently, attention has been turned to star formation in the L1689 cloud which lies about 1 degree east of the Rho Oph cloud. Interferometric observations of the protostellar object IRAS 16293-2422, which have revealed a rotating molecular disk, H2O masers, and several compact regions of ionized gas, will be reviewed. I. Introduction Lying in the Scorpius-Centaurus OB association, the filamentary system of dark clouds in Ophiuchus extends for tens of parsecs. Their distance from the Sun has been estimated to be about 160 pc (Bertiau 1958, Whittet 1974, Chini 1981), although a more recent evaluation suggests a somewhat lower value of 125+-25 pc (de Geus, de Zeeuw, and Lub 1989). The westernmost cloud in the complex, or  Ophiuchi cloud, is comprised of the L1681, L1686, and L1688 clouds and contains a large visually opaque core (Bok 1956). The core region lies adjacent to a reflection nebula illuminated by the B2 V star HD147889. The L1689 and L1709 clouds form the bases of two filamentary streamers which extend to the east (L1689, L1712, L1729) and northeast (L1709, L1740, L1744, L1755, and L1765) from the  Oph cloud. Early spectroscopic surveys of the area revealed numerous emission-line stars (Struve and Rudkjobing 1949, Haro 1949); the realization that these were pre-main sequence stars pointed to the  Oph cloud as a region of recent star formation. This was confirmed by near infrared surveys which unveiled a large population of embedded objects (Grasdalen, Strom, and Strom 1973; Vrba et al. 1975). This review will focus on star formation in the  Ophiuchi cloud but will discuss observations of other regions of the complex. It complements and expands upon a recent overview of the  Oph cloud by Klose (1986). In Section II, the distribution of low density molecular gas in the Ophiuchus complex, as traced by emission from CO and its isotopes, will be discussed. In Sec. III, observations of high density molecular gas are reviewed, focusing on cold cores revealed by a recent study of DCO emission. The distribution of lightly obscured emission-line stars and embedded infrared sources and their relationship to the molecular gas is the topic of Sec. IV. Evidence for mass loss among the population of young objects and the cloud energetics is the subject of Sec. V. An analysis of star formation in the  Oph cloud which combines the molecular-line, near-infrared, and far-infrared data of the region is presented in Sec. VI. Finally in Sec. VII, recent observations of the heavily obscured source IRAS 16293-2422 in the L1689 dark cloud are discussed. II. The Distribution of Low Density Molecular Gas Widespread self-absorption and the high optical depths of CO emission lines make CO the best tracer of column density in the Ophiuchus clouds (e.g., Encrenaz, Falgarone, and Lucas 1975). The distribution of CO(J=10) emission, mapped with 2.4 arcmin resolution, is shown in Figs. 1a and 1b over a 30 square degree area which includes the  Oph cloud and the eastern filaments (Loren 1989a). The total mass of the complex is determined to be 3050 M0 with about half of this contained in the  Oph cloud. These maps not only underscore the filamentary structure of the clouds, also evident in lower resolution maps of CO emission (de Geus, Bronfman, and Thaddeus 1989), but also demonstrate the clumpy nature of the gas. From these data, Loren (1989a,b) identifies 89 molecular clumps which are either kinematically or spatially distinct. The overall cloud morphology is suggestive of the passage of a shock, however, the expected streaming motions of the gas are not well established (e.g., Vrba 1977; Loren and Wootten 1986; Loren 1989a,b; de Geus, Bronfman, and Thaddeus 1989). In the core of the  Oph cloud, CO (as well as CO) emission lines are self-absorbed. Hence, optically-thin CO emission best delineates the column density distribution in the core (Lada and Wilking 1980). CO(J=10) observations with a resolution of 1.1 arcmin have revealed a 1 pc x 2 pc ridge of high column density gas which forms a centrally condensed core containing about 550 M0 (Wilking and Lada 1983). It is this core that distinguishes the Ophiuchus cloud from other clouds forming low mass stars such as those in Taurus-Auriga. Hydrogen column densities in the core range from 0.4-1.4 x 10 cm which corresponds to 25-100 mag of visual extinction. In Fig. 2, a previously unpublished 1.4 arcmin resolution map of the integrated intensity of CO(J=21) emission lines is shown. The distribution of high column density gas is similar to that revealed by the CO(J=10) maps, i.e., it lies in a northwest-southeast ridge with a very sharp column density gradient on the western edge. In contrast to the CO (J=10) map, the emission from regions of high temperature and/or density are clearly enhanced in the 21 data (see following section). A large velocity gradient code has been used to model the observed 21 and 10 CO line intensities; spatial densities of n(H) = 10 cm are inferred for the core with the larger values in the northwest end of the ridge (Fig. 3). The morphology of the CO ridge and its orientation perpendicular to the cloud filaments has led to the suggestion the ridge has been swept up by a shock propagating from the southwest in the direction of the Sco OB2 association (Loren and Wootten 1986). III. The Distribution of High Density Molecular Gas Observations of density-sensitive molecules, such as SO, HCO, NH, HCO, and DCO, have shown that the high density gas is closely associated with the high column density gas. In the core of the  Oph cloud, observations of SO and HCO (=2mm) emission have located two distinct concentrations of dense gas separated by about 0.6 pc and connected by a plateau of dense material:  Oph A at the northwestern end of the CO ridge and  Oph B just northeast of the ridge center (Gottlieb et al. 1978; Loren, Sandqvist and Wootten 1983). Modeling of HCO emission lines at =2 mm and 2 cm imply peak densities of n(H) = 3 x 10 cm in the 50 M0  Oph A core, densities greater than 10 cm in the 110 M0  Oph B core, and densities of about 5 x 10 cm in the plateau (Loren, Sandqvist and Wootten 1983). The  Oph B core is unique in that it displays rare 2 cm HCO emission, originating in several rotating fragments (Loren et al. 1980; Martin-Pintado et al. 1983; Wadiak et al. 1985). In addition, the  Oph B core is colder than  Oph A; observations of NH suggest kinetic temperatures of 19 K at  Oph B and 45 K for  Oph A (Zeng, Batrla, and Wilson 1984). This temperature difference is confirmed by the dearth of temperature-sensitive deuterated molecules in  Oph A relative to  Oph B (Loren and Wootten 1986). The most extensive study of high density gas in the  Oph complex has been made using multi-transition observations of the DCO molecule (Loren, Wootten, and Wilking 1990). The nature of the DCO chemistry and excitation requires that the emitting regions be simultaneously dense and cold. Twelve dense cores have been observed, the majority in the  Oph cloud but several in L1709 and L1689. They range in mass from 8-44 M0 and in density from 10 cm. The kinetic temperature is less than 15 K in most cores but may reach 25 K in the core associated with  Oph A. The distribution of dense gas (cores A-F) in the CO ridge of the  Oph core is shown in Fig. 4. Submillimeter emission from cold dust (155 K) has been detected from a compact region (0.5' x 1.0') which coincides with the center of the  Oph A core (Ward-Thompson et al. 1989; see also Schwartz, Snell, and Schloerb 1989). The relationship of the cold DCO cores with submillimeter emission and with heavily obscured IRAS sources has led to the suggestion that they are future sites for star formation. It is interesting to note that the masses, temperatures, and velocity dispersions of these cores lie between those derived for cold cores in the Taurus-Auriga complex and for giant molecular cloud cores (e.g., Table 1, Wilking 1989). To the extent that we can compare core properties determined from different molecules, the implication is that cores in Ophiuchus may ultimately form stars of intermediate mass. IV. The Embedded Population A. Emission-Line Stars and X-Ray Sources The presence of emission lines, notably Ha, in the spectrum of stars in regions of star formation is characteristic of objects in the T Tauri phase of pre-main-sequence evolution. Spectroscopic observations, and later objective prism surveys, identified about 75 emission-line stars in the complex, with the majority concentrated in the  Oph cloud (Struve and Rudkjobing 1949, Haro 1949, Dolidze and Arakelyan 1959). More recently, an H objective prism survey of the complex by Wilking, Schwartz, and Blackwell (1987) has revealed a total of 65 emission-line stars; a table of accurate positions and relative H line strengths can be found for these sources in their paper. Only 32 of these objects had been identified in previous H surveys, suggesting that variability may be a major factor in the detection of such objects. Among these 65 objects, 18 have thus far been confirmed as T Tauri stars (e.g., Rydgren, Strom, and Strom 1976, Cohen and Kuhi 1979, Rydgren 1980). The tendency for the remaining H objects to be associated with molecular gas, x ray emission, and/or far-infrared emission implies many are young stellar objects in a T Tauri phase of evolution. The distribution of H emission-line stars relative to the  Oph, L1689, and L1709 molecular clouds is shown in Fig. 5. The majority of objects are associated with the western half of the  Oph cloud yet few are observed directly toward the densest gas, an effect created by the large visual extinctions. This concentration of emission-line stars near the large centrally condensed core may imply either a critical density for the formation of stars or a progression of star formation through the complex from west to east. The latter would be favored by proponents of a shock wave origin for the cloud morphology. Highly variable soft x-ray emission is also associated with pre-main sequence stars which have dissipated most of their circumstellar envelopes. Thought to be connected with surface flare activity, the x-ray emission has also been found in a new class of young pre-main-sequence objects with virtually no circumstellar dust called the naked T Tauri stars (Walter et al. 1988). Einstein observations of the Ophiuchus region have revealed 50 x-ray sources with nearly half known to be pre-main sequence stars (Montmerle et al. 1983). A fraction of these also display variable radio continuum emission (e.g., Andre, Montmerle, and Feigelson 1987; Stine et al. 1988). The distribution of x-ray sources is reminiscent of the H stars as they are found primarily at the periphery of the dense molecular gas. B. The Embedded Population: Infrared Observations The majority of young stellar objects (YSOs) which have formed in the  Oph cloud are rendered invisible both by extinction from circumstellar dust and from the dark cloud itself. The first infrared surveys of the  Oph cloud core unveiled a cluster of 2 m sources with a stellar density higher than expected from background sources (Grasdalen, Strom, and Strom 1973; Vrba et al. 1975). Balloon-borne far-infrared observations provided evidence that most of the embedded objects were low-luminosity YSOs (Fazio et al. 1976, Cudlip et al. 1984). Subsequent near-infrared surveys expanded the range and sensitivity of previous work while attempting to distinguish between association members and background field stars (Elias 1978, Wilking and Lada 1983). Criteria to identify association members include the absence of 2.3 m CO emission, a relationship with large columns of gas and dust, and the presence of an infrared excess in the 3.4-20 m spectral region (e.g., Elias 1978, Lada and Wilking 1984). In the pre-infrared camera era, it was impractical to conduct a 2 m survey of an entire star-forming cloud such as  Oph to a high degree of sensitivity. One of the best guides to the embedded population has been provided by IRAS 12 and 25 m observations which completely sampled the Ophiuchus complex. A comprehensive study of more than one thousand sources from the IRAS Point Source Catalog has been made over a 170 sq. degree area of the Ophiuchus molecular complex by Ichikawa and Nishida (1989). They have found that the distribution of cooler sources (S(12m)Tdust) in the core, but this may be a selection effect caused by the presence of internal heat sources. This situation is in contrast to the L1689 cloud where the interstellar radiation field from the Sco OB2 association is probably the dominant heating source and the dust and gas are well-coupled in the outer cloud layers (Jarrett, Dickman, and Herbst 1989). VI. Star Formation in the Rho Oph Cloud A. The Utility of Spectral Energy Distributions Clues to the evolutionary state and mass of a YSO lie with its spectral energy distribution (SED). The shape of the emergent 1-100 m SED of a young stellar object depends upon the distribution of circumstellar dust which, in turn, is determined by its evolutionary state. For example, we expect a protostar which is surrounded by large amounts of circumstellar material to have a different infrared signature than an optically visible T Tauri star whose main infall stage has ended. Indeed broad band infrared photometry of association members in  Oph show that their SEDs fall into well-defined classes with systematic variations in shape (Lada and Wilking 1984; Wilking, Lada, and Young 1989). Theoretical models have been successful in describing these variations in SED shapes and suggest they form a quasi-continuous evolutionary sequence (e.g., Adams and Shu 1986, Adams, Lada, and Shu 1987, Myers et al. 1987, Lada 1987). SEDs which rise steeply into far-infrared wavelengths (Class I) are associated with heavily obscured objects and are modeled as accreting protostars. SEDs with peaks in the near and far-infrared (Class IID) are modeled as objects which have developed a strong wind and cleared away significant amounts of original infalling material. SEDs with small infrared excesses (Class II) are associated with T Tauri stars surrounded by disks and those with no excess (Class III) with post T Tauri stars or naked T Tauri stars. Examples of Class I and Class II SEDs for sources in  Oph are shown in Figs. 9a and 9b. Theoretical evolutionary tracks for pre-main-sequence objects indicate they evolve toward the main sequence from a higher luminosity regime (e.g., Iben 1965, Stahler, Shu, and Taam 1980). Therefore, the bolometric luminosity of a YSO gives an upper limit to its ultimate main sequence luminosity and mass. Reliable estimates for the bolometric luminosity of a YSO which has been observed over a broad range of wavelengths can be obtained by simply integrating its SED provided the following assumptions are valid: (1) the source's luminosity is radiated isotropically and either (2) there is no extinction toward the source, or (3) all of the extinction is produced by a shell of circumstellar dust which reradiates the absorbed light in the near- to far-infrared spectral region. Modeling of Class I sources suggests that assumptions (1) and (3) are well satisfied. For Class II objects, however, the assumption of no extinction underestimates the true luminosity while the presence of an anisotropically radiating disk will lead to overestimates (see Wilking, Lada, and Young 1989 for detailed discussion). These two effects introduce at most a factor of two uncertainty but tend to cancel. B. The Youth of the Ophiuchus Cluster From the collection of over 50 SEDs for objects in the  Oph cloud, the duration of the Class I phase and the age of the infrared cluster have been estimated (Wilking, Lada, and Young 1989). Since the number of Class I and Class II objects are roughly equal, the duration of the Class I phase is estimated to be approximately equal to the average age of the  Oph T Tauri stars, or 4 x 10 years. This translates into a mass accretion rate of 2.5 x 10 M0/yr for a 1 M0 star. The estimate for the length of the embedded state would drop by a factor of 4 if allowances were made for the luminosity evolution of Class I objects and for a population of naked T Tauri stars equal in age and number to the Class II population. The resulting mass accretion of 10 M0/yr is similar to that estimated for the  Oph cloud from protostar theory (Adams, Lada, and Shu 1987). The relative number of Class II to Class III objects in the cloud sets an upper limit to the duration of star formation of 3.5 x 10 years. This upper limit is consistent with the estimate of 1.5 x 10 years for both the age of the oldest  Oph T Tauri star (SR-22) and the contraction time for the least massive main sequence star, the B9-A0 V star SR-3. C. A Deficiency of Intermediate Mass Stars? Combining ground-based infrared photometry with IRAS observations, bolometric luminosities have been estimated for 58 association members and upper limits to Lbol for an additional 16 (Wilking, Lada, and Young 1989). The luminosity function, shown in Fig. 10, underscores the low-luminosity nature of the infrared cluster; nearly half of the objects have luminosities less than 1.8 L0 and 76% less than 5.6 L0. A remarkable feature of the luminosity function is that sources are segregated in luminosity by their SED class; 82% of the sources with 5.6 L0 < Lbol < 56 L0 are Class I while 67% of the lower luminosity sources are Class II. This segregation suggests that either low mass stars in  Oph are undergoing luminosity evolution as they progress from the Class I to Class II phase or that the most recent episode of star formation has produced predominately intermediate mass stars. The former possibility would point to a deficiency of intermediate mass stars in the cloud relative to the Initial Luminosity Function. D. A Starburst in the Rho Oph Cloud A high stellar density of YSOs in the large centrally-condensed core sets the  Oph cloud apart from other regions of low-mass star formation. This feature has been quantified by estimations for the star formation efficiency (SFE): the ratio of total stellar mass to the total mass of stars plus gas. While determinations of the SFE are subject to large uncertainties, conservative estimates for the  Oph core imply a SFE > 20% (Wilking and Lada 1983; Wilking, Lada, and Young 1989). The absence of massive stars in the cloud, which maintains the quiescent conditions of the core gas, insures the continued conversion of gas into stars and the slow release of gas from the cluster. These factors will ultimately lead to the emergence of a gravitationally bound cluster from the core (Wilking and Lada 1983; Lada, Margulis, and Dearborn 1984). In principle, an extended episode of star formation could lead to the relatively high values observed for the SFE. However, the youth of the  Oph cluster precludes this scenario. Instead, it appears that the high SFE is the result of an efficient burst of star-forming activity which has occurred over the last few million years. VII. Star Formation in L1689: IRAS 16293-2422 A smattering of H stars, x-ray sources, and IRAS sources give testimony to recent star formation in the L1689 cloud. Although near-infrared studies of the cloud are only beginning, array camera images suggest that the star formation process has not been as prolific as in the  Oph cloud (Greene, Young, and Meyers-Rice 1990). A dense molecular core of about 20 M0 lies in the northern region of the cloud and on its western edge sits an extremely cold infrared source IRAS 16293-2422 (Wootten and Loren 1987, Walker et al. 1986). The source has yet to be detected in the near-infrared; dust emission characteristic of a 40 K blackbody with a to emissivity law has been detected from the source over the 25 m - 2.7 mm spectral region, radiating a total luminosity of 27 L0. As mentioned earlier, IRAS 16293-2422 lies at the centroid of a quadrapolar distribution of high velocity molecular gas (Walker et al. 1988). Under the scrutiny of interferometric observations, IRAS 16293-2422 has displayed an increasing amount of complexity. Initial 4" x 6" resolution observations with the Owens Valley Radio Observatory (OVRO) Millimeter-Wave Interferometer of CO(J=10) and the 2.7 mm continuum have shown the IRAS source to be an elongated structure (1800 AU x <800 AU) of gas and dust, rotating about its minor axis (Fig. 11; Mundy et al. 1986; Mundy, Wootten, and Wilking 1990). The orientation of the structure is roughly perpendicular to the axis of the molecular outflow and the direction of the local magnetic field. The 2.4 km s rotation implies an included mass of 1.5-1.7 M0. This disklike structure lies near the center of a flattened region of NH3 emission observed with the VLA which extends for 8000 AU at roughly the same position angle (Mundy, Wootten, and Wilking 1990). VLA radio continuum observations at = 2 cm and 6 cm have detected two sources within the CO emission region, separated by 750 AU along its major axis (Wootten 1989). The southernmost radio source appears to be responsible for the molecular outflow since it is associated with a clump of HO masers, extended 2 cm emission, and compact SO emission (Wilking and Claussen 1987, Wootten 1989, Mundy et al. 1990). The model proposed for IRAS 16293-2422 is one where the compact gas and dust emission arise from a circumbinary disk or a combination of two circumstellar disks in a binary star system. A schematic for this proposed model is shown in Fig. 12. This simple model is complicated by new 4.3" x 2.4" observations of IRAS 16293-2422 at 2.7 mm with the OVRO interferometer (Mundy et al. 1990). At this higher resolution, the cold dust emission is resolved into two sources which correspond to the weak HII regions observed by Wootten (1989). It will be most interesting to see if higher resolution CO observations of IRAS 16293-2422 also resolve the gas component into two sources or if the binary system shares a common gas disk. VIII. Summary and Future Work Throughout much of the Ophiuchus complex, the distribution of gas and young stars resembles most other regions of low mass star formation. What distinguishes the Ophiuchus complex from these other regions is the large centrally condensed core in the  Oph cloud which appears to be the formation site of a gravitationally bound open cluster. The high stellar density of YSOs in the core has allowed us to look beyond the study of individual objects and to investigate the more global aspects of star formation. The population of YSOs in the core appears to have been formed by an efficient burst of star forming activity over the past few million years. The most heavily obscured, and presumably youngest, objects have higher luminosities than their more evolved counterparts, suggesting they may have an additional source of energy such as accretion. An outstanding question is whether the mass function of this infrared cluster varies significantly from the Initial Mass Function. Infrared array cameras which can probe more deeply into the dense cloud and sample the lowest luminosity YSOs may be able to address this problem with the help of a better theoretical understanding of how YSOs evolve in luminosity as they approach the main sequence. Another unresolved problem is the role of shocks in molding the present cloud morphology and triggering the formation of stars. Perhaps high resolution molecular-line observations will reveal the detailed kinematics expected from a shock. Because of the large number of embedded sources with well-determined spectral energy distributions and luminosities, the  Oph infrared cluster can serve as a standard to which other regions of low mass star formation can be compared. Detailed studies of the embedded populations in adjacent clouds in the Ophiuchus complex, such as L1689, will provide important data for comparisons with the  Oph cluster and give us insight into how the star formation rate and mass function may vary over a given complex. The proximity of the Ophiuchus complex and the large population of YSOs in a wide variety of evolutionary states makes the cloud an ideal target for high resolution infrared, millimeter, and radio wavelength observations. Infrared observations of YSOs with milliarcsecond to arcsecond resolution are possible using speckle interferometry, lunar occultations, and infrared array cameras. These observations have already begun to reveal the distribution of circumstellar dust and the presence of pre-main-sequence binary systems on size scales of 20-1000 AU (see chapter by Zinnecker for review). The dynamics and distribution of circumstellar gas and dust on scales of tens to hundreds of AU are within the reach of observations with the VLA, millimeter-wave interferometers, and submillimeter arrays planned for the future. These observations will yield important constraints to theoretical models for the early evolution of pre-main sequence stars and ultimately lead to a better understanding of the role of accretion, circumstellar disks, and mass outflows in this evolution. Figure Captions Figure 1 - Low density molecular gas over a 30 square degree area of the Ophiuchus complex as traced by 2.4 arcmin resolution observations of CO(J=10) emission lines (Loren 1989a). The contours are in units of TR*; the lowest (dashed) contours represent values of 2 K and 3 K, followed by contours of 4, 5, 6, 7, 8, 10 (bold), 14, 16, 18, and 20 K (bold). Also shown for reference are the locations of several early B stars in the region. Figure 2 - A previously unpublished map of CO(J=21) integrated intensity in the  Oph core. The 1.4 arcmin resolution observations were obtained with the University of Texas Millimeter-Wave Observatory. The contours levels are 2.5, 3.5, 4.5, 6, 7.5, and 9 K km s. The 225 positions in the cloud observed to construct this map are shown by dots. A cross marks the (0,0) for the map which is the position of the YSO Source 1 (E25): 16h23m32.8s -2416'44". The nominal locations of the high density regions  Oph A and B would be (-2,1) and (8,-5), respectively. Figure 3 - Spatial gas density in the  Oph core from CO data. The values for log n(H) were obtained for selected points combining the CO J=21 observations presented in Fig. 2 with CO J=10 observations of similar angular resolution (Wilking and Lada 1983) and using a large velocity gradient code. Varying the assumed Tex of 35K by 10K results in a change in log n(H) of only 0.1. As in Fig. 2, the cross marks the (0,0) for the map and the position of Source 1. Figure 4 - The distribution of DCO(J=21) in the core of the  Oph cloud (Loren, Wootten, and Wilking 1990). Solid contours show the intensity of TA* of DCO at the levels of 0.4, 0.8, 1.2, 1.6, and 2.0 K. The positions sampled are shown by filled circles. The dense cores are labeled A-F. The ridge of lower density CO gas is shown by short dashed contours. The boundaries of the  Oph cloud are outlined by two long dashed contours and are the 5 K and 10 K contours of TR* of CO. The positions of the four most luminous objects in the cloud are labeled by crosses. Figure 5 - The distribution of H emission line stars relative to the major concentrations of molecular gas in the Ophiuchus complex (Wilking, Schwartz, and Blackwell 1987). The clouds are outlined by contours of CO emission where TA* = 3 K (Loren and Wootten 1986). Figure 6 - The locations of point and small extended 12 m sources in the  Oph cloud observed by IRAS, superposed on the red photograph from the Palomar Sky Survey (Wilking, Lada, and Young 1989). Large crosses mark the positions of sources with 12 m flux densities greater than 0.25 Jy and small crosses less than this value. The solid contour outlining the molecular cloud represents a CO emission line strength of TR* = 6 K (Loren 1989a). Figure 7 - The distribution of 78 association members relative the cold, dense gas in the  Oph core (adapted from Loren, Wootten, and Wilking 1990). Larger symbols correspond to the more luminous YSOs. The boundary of the molecular cloud is marked by the TR* = 10 K contour for CO. This area of the cloud has been completely sampled by IRAS and H emission-line surveys but not near-infrared surveys. The contours and labels for the DCO emission are the same as Fig. 4. The large box outlines the area imaged by sensitive near infrared camera observations which are presented in Fig. 8. Figure 8 - A mosaic of 156 images at 2.2 m of a 650 square arcmin region of the  Oph core (Greene, Young, and Meyers-Rice 1990). North in declination is up and east in right ascension is to the left. The center of the 24.3' x 26.8' region is 16h24m15.0s, -2426'30". Over 200 sources are present in this mosaic; the faintest objects are between K=14.5-15.0 mag. The area displayed is shown in reference to the molecular gas and known association members in Fig. 7. Figure 9 - Spectral energy distributions, grouped by morphology, for selected YSOs in the  Oph cloud displaying far-infrared emission (Wilking, Lada, and Young 1989). The power of ten used to scale the SED is given in parentheses below the source name. Steeply rising SEDs characteristic of Class I sources are shown in Fig. 9a and Class II SEDs with small infrared excesses are shown in Fig. 9b. For comparison, the SED for a 2300 K blackbody is shown in Fig. 9b. Figure 10 - The luminosity function of 74 sources embedded in the  Oph cloud (Wilking, Lada, and Young 1989). For comparison, the luminosity function derived from the Initial Mass Function (the ILF) is shown, normalized to the number of sources with well-determined luminosities in the -0.25 < log L < 0.75 range. The YSOs in each luminosity bin are grouped according to the shapes of their SEDs. Luminosity estimates for sources with no detectable IRAS flux are shown as upper limits and are not included in the normalization of the ILF. Figure 11 - Interferometric observation of IRAS 16293-2422 in CO(J=10) and the 2.7 mm continuum (Mundy, Wilking, and Wootten 1990). The synthesized beam for the maps is 6.3" x 4.5", P.A.=0. Crosses mark the positions of the two radio sources found by Wootten (1989). Contour levels for the CO integrated intensity are 1, 2, 3, 4, and, 5 Jy beam-1 km s-1; for the continuum map they are -25, 25, 50, 75, 100, 150, 200, 250, and 300 mJy beam-1. Figure 12 - A schematic of a model for the IRAS 16293-2422 region including the inner disk (defined by interferometric CO observations), the outer disk (interferometric NH3 observations), and the static core (single dish CO and CO observations) (Mundy, Wootten, and Wilking 1990). The plusses within the disk region represent the radio continuum sources observed by (Wootten 1989). The long arrows above and below the outer disk represent the molecular outflow.