Search

We have newly identified a substantial number of type 1 active galactic nuclei (AGN) featuring weak broad-line regions (BLRs) at z < 0.2 from detailed analysis of galaxy spectra in the Sloan Digital Sky Survey Data Release 7. These objects predominantly show a stellar continuum but also a broad H-alpha emission line, indicating the presence of a low-luminosity AGN oriented so that we are viewing the central engine directly without significant obscuration. These accreting black holes have previously eluded detection due to their weak nature. The new BLR AGNs we found increased the number of known type 1 AGNs by 49%. Some of these new BLR AGNs were detected at the Chandra X-ray Observatory, and their X-ray properties confirm that they are indeed type 1 AGN. Based on our new and more complete catalogue of type 1 AGNs, we derived the type 1 fraction of AGNs as a function of [OIII] 5007 emission luminosity and explored the possible dilution effect on the obscured AGN due to star-formation. The new type 1 AGN fraction shows much more complex behavior with respect to black hole mass and bolometric luminosity than suggested by the existing receding torus model. The type 1 AGN fraction is sensitive to both of these factors, and there seems to be a sweet spot (ridge) in the diagram of black hole mass and bolometric luminosity. Furthermore, we present a hint that the Eddington ratio plays a role in determining the opening angles., Comment: Accepted for publication in ApJS. 17 pages including 17 figures and 9 tables. 2 machine-readable tables to appear online

We investigate early black hole (BH) growth through the methodical search for $z\gtrsim5$ AGN in the $Chandra$ Deep Field South. We base our search on the $Chandra$ 4-Ms data with flux limits of $9.1\times\ 10^{-18}$ (soft, 0.5 - 2 keV) and $5.5\times\ 10^{-17}\ \mathrm{erg}\ \mathrm{s}^{-1}\ \mathrm{cm}^{-2}$ (hard, 2 - 8 keV). At $z\sim5$ this corresponds to luminosities as low as $\sim10^{42}$ ($\sim10^{43}$) $\mathrm{erg}\ \mathrm{s}^{-1}$ in the soft (hard) band and should allow us to detect Compton-thin AGN with $M_\mathrm{BH}>10^7 M_{\odot}$ and Eddington ratios > 0.1. Our field ($0.03~\mathrm{deg}^2$) contains over 600 $z\sim5$ Lyman Break Galaxies. Based on lower redshift relations we would expect $\sim20$ of them to host AGN. After combining the $Chandra$ data with GOODS/ACS, CANDELS/WFC3 and $Spitzer$/IRAC data, the sample consists of 58 high-redshift candidates. We run a photometric redshift code, stack the GOODS/ACS data, apply colour criteria and the Lyman Break Technique and use the X-ray Hardness Ratio. We combine our tests and using additional data find that all sources are most likely at low redshift. We also find five X-ray sources without a counterpart in the optical or infrared which might be spurious detections. We conclude that our field does not contain any convincing $z\gtrsim5$ AGN. Explanations for this result include a low BH occupation fraction, a low AGN fraction, short, super-Eddington growth modes, BH growth through BH-BH mergers or in optically faint galaxies. By searching for $z\gtrsim5$ AGN we are setting the foundation for constraining early BH growth and seed formation scenarios., Comment: Accepted for publication in MNRAS, 22 pages, 15 figures, for an overview of our analysis see figure 2

We report on observations of NGC1068 with NuSTAR, which provide the best constraints to date on its $>10$~keV spectral shape. We find no strong variability over the past two decades, consistent with its Compton-thick AGN classification. The combined NuSTAR, Chandra, XMM-Newton, and Swift-BAT spectral dataset offers new insights into the complex reflected emission. The critical combination of the high signal-to-noise NuSTAR data and a spatial decomposition with Chandra allow us to break several model degeneracies and greatly aid physical interpretation. When modeled as a monolithic (i.e., a single N_H) reflector, none of the common Compton-reflection models are able to match the neutral fluorescence lines and broad spectral shape of the Compton reflection. A multi-component reflector with three distinct column densities (e.g., N_H~1.5e23, 5e24, and 1e25 cm^{-2}) provides a more reasonable fit to the spectral lines and Compton hump, with near-solar Fe abundances. In this model, the higher N_H components provide the bulk of the Compton hump flux while the lower N_H component produces much of the line emission, effectively decoupling two key features of Compton reflection. We note that ~30% of the neutral Fe Kalpha line flux arises from >2" (~140 pc), implying that a significant fraction of the <10 keV reflected component arises from regions well outside of a parsec-scale torus. These results likely have ramifications for the interpretation of poorer signal-to-noise observations and/or more distant objects [Abridged]., Comment: Submitted to ApJ; 23 pages (ApJ format); 11 figures and 3 tables; Comments welcomed!

We present near-infrared (NIR) spectra of four optically-elusive AGN and four X- ray bright, optically normal galaxies (XBONGs) from the Swift-BAT survey. With archival observations from SDSS, 2MASS, Spitzer and WISE, we test a number of AGN indicators in the NIR and MIR; namely, NIR emission line diagnostic ratios, the presence of coronal high-ionization lines, and infrared photometry. Of our eight hard X-ray selected AGN, we find that optical normalcy has a variety of causes from object to object, and no one explanation applies. Our objects have normal Eddington ratios, and so are unlikely to host radiatively-inefficient accretion flows (RIAFs). It is unlikely that star formation in the host or starlight dilution is contributing to their failure of optical diagnostics, except perhaps in two cases. The NIR continua are well-fit by two blackbodies: one at the stellar temperature, and a hot dust component near the dust sublimation temperature. XBONGs are more likely to have significant hot dust components, while these components are small relative to starlight in the optically-elusive AGN. Some of our sample have NIR line ratios typical of AGN, but NIR diagnostics are unsuccessful in distinguishing HII regions from AGN in general. In one object, we discover a hidden broad line region in the NIR. These results have strong relevance to the origin of optically normal AGN in deep X-ray surveys., Comment: 38 pages including 3 tables and 11 figures, Accepted for publication in ApJ

We present a broadband (~0.5-79 keV) spectral and temporal analysis of multiple NuSTAR observations combined with archival Suzaku and Chandra data of NGC4945, the brightest extragalactic source at 100 keV. We observe hard X-ray (> 10 keV) flux and spectral variability, with flux variations of a factor 2 on timescales of 20 ksec. A variable primary continuum dominates the high energy spectrum (>10 keV) in all the states, while the reflected/scattered flux which dominates at E< 10 keV stays approximately constant. From modelling the complex reflection/transmission spectrum we derive a Compton depth along the line of sight of tau_Thomson ~ 2.9, and a global covering factor for the circumnuclear gas of ~ 0.15. This agrees with the constraints derived from the high energy variability, which implies that most of the high energy flux is transmitted, rather that Compton-scattered. This demonstrates the effectiveness of spectral analysis in constraining the geometric properties of the circumnuclear gas, and validates similar methods used for analyzing the spectra of other bright, Compton-thick AGN. The lower limits on the e-folding energy are between 200-300 keV, consistent with previous BeppoSAX, Suzaku and Swift BAT observations. The accretion rate, estimated from the X-ray luminosity and assuming a bolometric correction typical of type 2 AGN, is in the range ~0.1-0.3 lambda_Edd depending on the flux state. The substantial observed X-ray luminosity variability of NGC4945 implies that large errors can arise from using single-epoch X-ray data to derive L/L_Edd values for obscured AGNs., Comment: Accepted for publication in ApJ

Ultraluminous and luminous infrared galaxies (ULIRGs and LIRGs) are the most extreme star-forming galaxies in the universe, and dominate the total star formation rate density at z>1. In the local universe (z<0.3), the majority of ULIRGs and a significant portion of LIRGs are triggered by interactions between gas-rich spiral galaxies, yet it is unclear if this is still the case at high-z. To investigate the relative importance of galaxy interactions in infrared luminous galaxies, we carry out a comparison of optical morphological properties between local (U)LIRGs and (U)LIRGs at z=0.5-1.5 based on the same sample selection, morphology classification scheme, and optical morphology at similar rest-frame wavelengths. In addition, we quantify the systematics in comparing local and high-z datasets by constructing a redshifted dataset from local (U)LIRGs, in which its data quality mimics the high-z dataset. Based on the Gini-M20 classification scheme, we find that the fraction of interacting systems decreases by ~8% from local to z<~1, and it is consistent with the reduction between local and redshifted datasets (6(+14-6)%). Based on visual classifications, the merger fraction of local ULIRGs is found to be ~20% lower compared to published results, and the reduction due to redshifiting is 15(+10-8)%. Consequently, the differences of merger fractions between local and z<~1 (U)LIRGs is only ~17%. These results demonstrate that there is no strong evolution in the fraction of (U)LIRGs classified as mergers at least out to z~1. At z>1, the morphology types of ~30% of (U)LIRGs can not be determined due to their faintness in the F814W-band, and thus the merger fraction measured at z>1 suffers from large uncertainties., Comment: 10 pages, 6 figures, ApJ accepted

While performing a survey to detect recoiling supermassive black holes, we have identified an unusual source having a projected offset of 800 pc from a nearby dwarf galaxy. The object, SDSS J113323.97+550415.8, exhibits broad emission lines and strong variability. While originally classified as a supernova (SN) because of its nondetection in 2005, we detect it in recent and past observations over 63 yr and find over a magnitude of rebrightening in the last 2 years. Using high-resolution adaptive optics observations, we constrain the source emission region to be <12 pc and find a disturbed host-galaxy morphology indicative of recent merger activity. Observations taken over more than a decade show narrow [O III] lines, constant ultraviolet emission, broad Balmer lines, a constant putative black hole mass over a decade of observations despite changes in the continuum, and optical emission-line diagnostics consistent with an active galactic nucleus (AGN). However, the optical spectra exhibit blueshifted absorption, and eventually narrow Fe II and [Ca II] emission, each of which is rarely found in AGN spectra. While this peculiar source displays many of the observational properties expected of a potential black hole recoil candidate, some of the properties could also be explained by a luminous blue variable star (LBV) erupting for decades since 1950, followed by a Type IIn SN in 2001. Interpreted as an LBV followed by a SN analogous to SN 2009ip, the multi-decade LBV eruptions would be the longest ever observed, and the broad Halpha emission would be the most luminous ever observed at late times (>10 yr), larger than that of unusually luminous supernovae such as SN 1988Z, suggesting one of the most extreme episodes of pre-SN mass loss ever discovered., Comment: Accepted for publication in MNRAS

We present early results of the Herschel PACS (70 and 160 \micron{}) and SPIRE (250, 350, and 500 \micron{}) survey of 313 low redshift ($\rm{z} < 0.05$), ultra-hard X-ray (14--195 keV) selected AGN from the 58 month Swift/BAT catalog. Selection of AGN from ultra-hard X-rays avoids bias from obscuration providing a complete sample of AGN to study the connection between nuclear activity and star formation in host galaxies. With the high angular resolution of PACS, we find that $>$35\%$ and $>$20\%$ of the sources are "point-like" at 70 and 160 \micron{} respectively and many more that have their flux dominated by a point source located at the nucleus. The inferred star formation rates (SFR) of 0.1 - 100 M$_{\sun}$ yr$^{-1}$ using the 70 and 160 \micron{} flux densities as SFR indicators are consistent with those inferred from Spitzer NeII fluxes, but we find that 11.25 \micron{} PAH data give $\sim$3x lower SFR. Using GALFIT to measure the size of the FIR emitting regions, we determined the SFR surface density [M$_{\sun}$ yr$^{-1}$ kpc$^{-2}$] for our sample, finding a significant fraction of these sources exceed the threshold for star formation driven winds (0.1 M$_{\sun}$ yr$^{-1}$ kpc$^{-2}$)., Comment: 5 pages, 5 figures, accepted for publication in ApJ Letters

Galaxy interactions/mergers have been shown to dominate the population of IR luminous galaxies (log(LIR)>11.6Lsun) in the local Universe (z<0.25). Recent studies based on the relation between galaxies' star formation rates and stellar mass (the SFR-M relation or the galaxy main sequence (MS)) have suggested that galaxy interaction/mergers may only become significant when galaxies fall well above the galaxy MS. Since the typical SFR at given M increases with redshift, the existence of galaxy MS implies that massive, IR-luminous galaxies at high-z may not necessarily be driven by galaxy interactions. We examine the role of galaxy interactions in the SFR-M relation by carrying out a morphological analysis of 2084 Herschel-selected galaxies at 0.2 < z < 1.5 in the COSMOS field. Herschel-PACS and -SPIRE observations covering the full 2-deg^2 COSMOS field provide one of the largest far-IR selected samples of high-redshift galaxies with well-determined redshifts to date, with sufficient sensitivity at z ~ 1, to sample objects lying on and above the galaxy MS. Using a detailed visual classification scheme, we show that the fraction of "disk galaxies" decreases and the fraction of "irregular" galaxies increases systematically with increasing LIR out to z ~ 1.5 and z ~ 1.0, respectively. At log(LIR) > 11.5 Lsun, >50% of the objects show evident features of strongly interacting/merger systems, where this percentage is similar to the studies of local IR-luminous galaxies. The fraction of interacting/merger systems also systematically increases with the deviation from the SFR-M relation, supporting the view that galaxies fall above the MS are more dominated by mergers than the MS galaxies. Meanwhile, we find that ~18% of massive IR-luminous MS galaxies are classified as interacting systems, where this population may not evolve through the evolutionary track predicted by a simple gas exhaustion model., Comment: 14 pages, 11 figures, ApJ accepted

We present the results from a large 850 micron survey of the sigma Orionis cluster using the SCUBA-2 camera on the James Clerk Maxwell Telescope. The 0.5-degree diameter circular region we surveyed contains 297 young stellar objects with an age estimated at about 3Myr. We detect 9 of these objects, 8 of which have infrared excesses from an inner disc. We also serendipitously detect 3 non-stellar sources at > 5sigma that are likely background submillimetre galaxies. The 9 detected stars have inferred disc masses ranging from 5 to about 17MJup, assuming similar dust properties as Taurus discs and an ISM gas-to-dust ratio of 100. There is a net positive signal toward the positions of the individually undetected infrared excess sources indicating a mean disc mass of 0.5 MJup . Stacking the emission toward those stars without infrared excesses constrains their mean disc mass to less than 0.3MJup, or an equivalent Earth mass in dust. The submillimetre luminosity distribution is significantly different from that in the younger Taurus region, indicating disc mass evolution as star forming regions age and the infrared excess fraction decreases. Submillimeter Array observations reveal CO emission toward 4 sources demonstrating that some, but probably not much, molecular gas remains in these relatively evolved discs. These observations provide new constraints on the dust and gas mass of protoplanetary discs during the giant planet building phase and provide a reference level for future studies of disc evolution., Comment: 10 pages, 6 figures, accepted for publication in MNRAS

We present deep 450um and 850um observations of a large, uniformly covered 394arcmin^2 area in the COSMOS field obtained with the SCUBA-2 instrument on the James Clerk Maxwell Telescope (JCMT). We achieve root-mean-square noise values of 4.13mJy at 450um and 0.80mJy at 850um. The differential and cumulative number counts are presented and compared to similar previous works. Individual point sources are identified at >3.6sigma significance, a threshold corresponding to a 3-5% sample contamination rate. We identify 78 sources at 450um and 99 at 850um, with flux densities S450=13-37mJy and S850=2-16mJy. Only 62-76% of 450um sources are 850um detected and 61-81% of 850um sources are 450um detected. The positional uncertainties at 450um are small (1-2.5") and therefore allow a precise identification of multiwavelength counterparts without reliance on detection at 24um or radio wavelengths; we find that only 44% of 450um-selected galaxies and 60% of 850um-sources have 24um or radio counterparts. 450um-selected galaxies peak at =1.95+-0.19 and 850um=selected galaxies peak at =2.16+-0.11. The two samples occupy similar parameter space in redshift and luminosity, while their median SED peak wavelengths differ by ~10-50um (translating to deltaTdust =8-12K, where 450um-selected galaxies are warmer). The similarities of the 450um and 850um populations, yet lack of direct overlap between them, suggests that submillimeter surveys conducted at any single far-infrared wavelength will be significantly incomplete (~>30%) at censusing infrared-luminous star formation at high-z., Comment: 37 pages (21 of text),14 figures, updated to MNRAS accepted version

We present the first analysis of the all-sky Swift BAT ultra hard X-ray (14-195 keV) data for a targeted list of objects. We find the BAT data can be studied at 3x fainter limits than in previous blind detection catalogs based on prior knowledge of source positions and using smaller energy ranges for source detection. We determine the AGN fraction in 134 nearby (z<0.05) luminous infrared galaxies (LIRGS) from the GOALS sample. We find that LIRGs have a higher detection frequency than galaxies matched in stellar mass and redshift at 14-195 keV and 24-35 keV. In agreement with work at other wavelengths, the AGN detection fraction increases strongly at high IR luminosity with half of high luminosity LIRGs (50%, 6/12, log L_IR/L_sun>11.8) detected. The BAT AGN classification shows 97% (37/38) agreement with Chandra and XMM AGN classification using hardness ratios or detection of a iron K-alpha line. This confirms our statistical analysis and supports the use of the Swift BAT all-sky survey to study fainter populations of any category of sources in the ultra hard X-ray band. BAT AGN in LIRGs tend to show higher column densities with 40\pm9% showing 14-195 keV/2-10 keV hardness flux ratios suggestive of high or Compton-thick column densities (log N_H>24 cm^-2), compared to only 12\pm5% of non-LIRG BAT AGN. We also find that using specific energy ranges of the BAT detector can yield additional sources over total band detections with 24% (5/21) of detections in LIRGs at 24-35 keV not detected at 14-195 keV., Comment: Accepted for publication in ApJL

We study the fraction of dual AGN in a sample of 167 nearby (z<0.05), moderate luminosity, ultra hard X-ray selected AGN from the all-sky Swift BAT survey. Combining new Chandra and Gemini observations together with optical and X-ray observations, we find that the dual AGN frequency at scales <100 kpc is 10% (16/167). Of the 16 dual AGN, 3 (19%) were detected using X-ray spectroscopy and were not detected using emission line diagnostics. Close dual AGN (<30 kpc) tend to be more common among the most X-ray luminous systems. In dual AGN, the X-ray luminosity of both AGN increases strongly with decreasing galaxy separation, suggesting that the merging event is key in powering both AGN. 50% of the AGN with a very close companion (<15 kpc), are dual AGN. We also find that dual AGN are more likely to occur in major mergers and tend to avoid absorption line galaxies with elliptical morphologies. Finally, we find SDSS Seyferts are much less likely than BAT AGN (0.25% vs. 7.8%) to be found in dual AGN at scales <30 kpc because of a smaller number of companions galaxies, fiber collision limits, a tendency for AGN at small separations to be detected only in X-rays, and a higher fraction of dual AGN companions with increasing AGN luminosity., Comment: Accepted for publication in ApJL

We have assembled the largest sample of ultra hard X-ray selected (14-195 keV) AGN with host galaxy optical data to date, with 185 nearby (z<0.05), moderate luminosity AGN from the Swift BAT sample. The BAT AGN host galaxies have intermediate optical colors (u-r and g-r) that are bluer than a comparison sample of inactive galaxies and optically selected AGN from the Sloan Digital Sky Survey (SDSS) which are chosen to have the same stellar mass. Based on morphological classifications from the RC3 and the Galaxy Zoo, the bluer colors of BAT AGN are mainly due to a higher fraction of mergers and massive spirals than in the comparison samples. BAT AGN in massive galaxies (log M_*>10.5) have a 5 to 10 times higher rate of spiral morphologies than in SDSS AGN or inactive galaxies. We also see enhanced far-IR emission in BAT AGN suggestive of higher levels of star formation compared to the comparison samples. BAT AGN are preferentially found in the most massive host galaxies with high concentration indexes indicative of large bulge-to-disk ratios and large supermassive black holes. The narrow-line BAT AGN have similar intrinsic luminosities as the SDSS NL Seyferts based on measurements of [O III]. There is also a correlation between the stellar mass and X-ray emission. The BAT AGN in mergers have bluer colors and greater ultra hard X-ray emission compared to the BAT sample as whole. In agreement with the Unified Model of AGN, and the relatively unbiased nature of the BAT sources, the host galaxy colors and morphologies are independent of measures of obscuration such as X-ray column density or Seyfert type. The high fraction of massive spiral galaxies and galaxy mergers in BAT AGN suggest that host galaxy morphology is related to the activation and fueling of local AGN., Comment: 30 pages, 30 figures, accepted for publication in ApJ

We have discovered a binary AGN in the galaxy Mrk 739 using Chandra and Swift BAT. We find two luminous L_2-10 keV=1.1*10^43 and 1.0*10^42 erg/s, unresolved nuclei with a projected separation of 3.4 kpc (5.8\pm0.1) coincident with two bulge components in the optical image. The western X-ray source (Mrk 739W) is highly variable (2.5x) during the 4-hour Chandra observation and has a very hard spectrum consistent with an AGN. While the eastern component was already known to be an AGN based on the presence of broad optical recombination lines, Mrk 739W shows no evidence of being an AGN in optical, UV, and radio observations, suggesting the critical importance of high spatial resolution hard X-ray observations (>2 keV) in finding these binary AGN. A high level of star formation combined with a very low L_[O III]/L_2-10 keV ratio cause the AGN to be missed in optical observations. CO observations of the (3-2) and (2-1) lines indicate large amounts of molecular gas in the system that could be driven towards the black holes during the violent galaxy collision and be key to fueling the binary AGN. Mrk 739E has a high Eddington ratio of 0.71 and a small black hole (log M_BH=7.05\pm0.3) consistent with an efficiently accreting AGN. Other than NGC 6240, this stands as the nearest case of a binary AGN discovered to date., Comment: 7 pages, 5 figures, Published in Astrophysical Journal Letters. Reformatted and fixed typos in v1

We discuss the merger rate, close galaxy environment, and clustering on scales up to a Mpc of the SWIFT BAT hard X-ray sample of nearby (z<0.05), moderate-luminosity active galactic nuclei (AGN). We find a higher incidence of galaxies with signs of disruption compared to a matched control sample (18% versus 1%) and of close pairs within 30 kpc (24% versus 1%). We also find a larger fraction with companions compared to normal galaxies and optical emission line selected AGN at scales up to 250 kpc. We hypothesize that these merging AGN may not be identified using optical emission line diagnostics because of optical extinction and dilution by star formation. In support of this hypothesis, in merging systems we find a higher hard X-ray to [OIII] flux ratio, as well as emission line diagnostics characteristic of composite or star-forming galaxies, and a larger IRAS 60 um to stellar mass ratio., Comment: 15 pages, 4 figures, published in the Astrophysical Journal Letters

We construct a new sample of 32 obscured active galactic nuclei (AGNs) selected from the Second XMM-Newton Serendipitous Source Catalogue to investigate their multiwavelength properties in relation to the "scattering fraction", the ratio of the soft X-ray flux to the absorption-corrected direct emission. The sample covers a broad range of the scattering fraction (0.1%-10%). A quarter of the 32 AGNs have a very low scattering fraction (smaller than 0.5%), which suggests that they are buried in a geometrically thick torus with a very small opening angle. We investigate correlations between the scattering fraction and multiwavelength properties. We find that AGNs with a small scattering fraction tend to have low [O III]lambda5007/X-ray luminosity ratios. This result agrees with the expectation that the extent of the narrow-line region is small because of the small opening angle of the torus. There is no significant correlation between scattering fraction and far-infrared luminosity. This implies that a scale height of the torus is not primarily determined by starburst activity. We also compare scattering fraction with black hole mass or Eddington ratio and find a weak anti-correlation between the Eddington ratio and scattering fraction. This implies that more rapidly growing supermassive black holes tend to have thicker tori., Comment: 13 pages, 10 figures, accepted for publication in ApJ

The Swift Burst Alert Telescope (BAT) survey of Active Galactic Nuclei (AGN) is providing an unprecedented view of local AGNs ( = 0.03) and their host galaxy properties. In this paper, we present an analysis of the optical spectra of a sample of 64 AGNs from the 9-month survey, detected solely based on their 14-195 keV flux. Our analysis includes both archived spectra from the Sloan Digital Sky Survey and our own observations from the 2.1-m Kitt Peak National Observatory telescope. Among our results, we include line ratio classifications utilizing standard emission line diagnostic plots, [O III] 5007 A luminosities, and H-beta derived black hole masses. As in our X-ray study, we find the type 2 sources to be less luminous (in [O III] 5007 A and 14-195 keV luminosities) with lower accretion rates than the type 1 sources. We find that the optically classified LINERs, H II/composite galaxies, and ambiguous sources have the lowest luminosities, while both broad line and narrow line Seyferts have similar luminosities. From a comparison of the hard X-ray (14-195 keV) and [O III] luminosities, we find that both the observed and extinction-corrected [O III] luminosities are weakly correlated with X-ray luminosity. In a study of the host galaxy properties from both continuum fits and measurements of the stellar absorption indices, we find that the hosts of the narrow line sources have properties consistent with late type galaxies., Comment: 84 pages, 20 figures, 17 tables, accepted in ApJ