% v1.6 for accepted version for astro-ph % response to referee and other fixes % -X- revised values in table 1 % % v1.4 has % -X- plot H2O luminosity against core radio luminosity? % -X- put in galaxy classification into Table 1. % -X- comments from Chris & Stefi, Heino, Alison, Christian, Gallimore. % -X- modifications to discussion by Alison % -X- GBT with comments by Falcke, Henkel, and Mundell % -X- ref to Ylva's thesis % -X- add to NGC4151 % -X- ask Rene about NGC1052 (got draft from Ken) % -X- expand discussion % -X- write conclusions % -X- write abstract % % Draft v1.1 -- ABP % reorganized intro a bit, added table 1 and refs. % % TXS2226-184 Draft v1.0 on 5/01/2001 % todo for paper: % -X- spectrum 1.4 to 22 GHz, describe observations, table % -X- incorporate H2O non-detections % -X- fit guassians to figure 1 and 2 % -X- make table of gaussian parameters % -X- discussion section with table of H2O and HI properties in % all H2O masers. Examine parameter space and luminosity % ratios. % --- paragraph about all sources being different but compare with % unified model prediction? luminosity? accretion rate? % -X- analyze 22 GHz image in Difmap. Is it resolved? Yes. Compare % with earlier claim by Heino. % %\documentclass[manuscript]{aastex} %\documentclass[12pt,preprint]{aastex} \documentclass[preprint]{aastex} \usepackage{psfig} \def\HI {H\kern0.1em{\sc i}} \def\water {H$_{\rm 2}$O} \def\txs {TXS\,2226{\tt -}184} \def\pks {PKS\,2322{\tt -}123} \def\radm {rad m$^{-2}$} \def\etal {{\sl et~al.\ }} \def\dg{$^{\circ}$} \def\deg{$^{\circ}$} \def\kms{km s$^{-1}$} \begin{document} \title{~~\\ ~~\\ \HI\ Absorption in the Gigamaser Galaxy TXS\,2226{\tt -}184 and the Relation between \HI\ Absorption and Water Emission} \shorttitle{\HI in \txs} \shortauthors{Taylor \& Peck} \author{G. B. Taylor\altaffilmark{1}, A. B. Peck\altaffilmark{2,3}, C. Henkel\altaffilmark{2}, H. Falcke\altaffilmark{2}, C. G. Mundell\altaffilmark{4}, C. P. O'Dea\altaffilmark{5}, S.A. Baum\altaffilmark{5}, \& J. F. Gallimore\altaffilmark{6}} %\email{gtaylor@nrao.edu} \altaffiltext{1}{National Radio Astronomy Observatory, P. O. Box 0, Socorro, NM 87801, USA} \altaffiltext{2}{Max-Planck-Institut f\"ur Radioastronomie, Auf dem H\"ugel 69, D-53121 Bonn, Germany} \altaffiltext{3}{{\it Current address:} Harvard Smithsonian Center for Astrophysics, SAO/SMA Project, P.O. Box 824, Hilo, HI 96721, USA} \altaffiltext{4}{Astrophysics Research Institute, Liverpool John Moores University, Twelve Quays House, Egerton Wharf, Birkenhead, CH41 1lD, UK} % cgm@astro.umd.edu. \altaffiltext{5}{Space Telescope Science Institute, Baltimore, MD 21218, USA} \altaffiltext{6}{Department of Physics, Bucknell University, Lewisburg, PA 17837, USA} % jgallimo@bucknell.edu %\received{2000 March 24} %\accepted{2000 May 9} %\journalid{337}{15 January 1989} %\articleid{11}{14} \slugcomment{As Accepted by the Astrophysical Journal} \begin{abstract} We report on the discovery of \HI\ in absorption toward the gigamaser galaxy \txs\ using the VLA. The absorption appears to consist of two components -- one with a width of 125 \kms, and one broader (420 \kms), both toward the compact radio source in the nucleus of the galaxy. Based on these large velocity widths we suggest that the \HI\ absorption is produced in the central parsecs of the galaxy, on a similar scale to that which gives rise to the water maser emission. This brings to eight the number of galaxies known to exhibit both water masers and \HI\ absorption. We explore the relationship between these two phenomena, and present a physically motivated (but unfruitful) search for water maser emission in five radio galaxies known to exhibit strong \HI\ absorption. \end{abstract} \keywords{galaxies: active -- galaxies: individual (\txs) -- galaxies: nuclei -- radio lines: galaxies -- masers} \section{Introduction} A number of studies (e.g., van Gorkom \etal\ 1989, Pihlstr\"om 2001) have shown that the rate of detection of \HI\ absorption in galaxies is highest in Compact Symmetric Objects (CSOs) and Compact Steep Spectrum (CSS) objects. Of the theories proposed to explain this phenomenon, the existence of a circumnuclear disk or torus structure seems the most likely. In this scenario, it is the orientation and geometry of the source which increases the likelihood of detecting \HI\ in absorption. In the CSOs and CSS objects the emission is dominated by the hot spots and lobes which are not Doppler boosted. Thus, in the CSOs we have bright symmetric emission against which to search for \HI\ absorption on parsec scales at a large range of source inclination angles. In large sources, however, the parsec scale emission is dominated by the relativistic jets. It is much harder to see \HI\ absorption in these sources since when the source is oriented towards us, the counter jet is deboosted and any disk which is perpendicular to the jet is at an unfavorable angle for absorption. %The core-dominated sources are %oriented close to the line of sight and the jets are comprised of %extremely high velocity outflows. This causes the approaching jet to be %strongly Doppler boosted, while the counterjet is Doppler dimmed. The %more symmetric sources, however, are randomly oriented with respect to %the line of sight, and so the continuum emission is not strongly %beamed. In sources with radio axes close to the plane of the sky, %such as CSOs, the counterjet can contain up to half of the flux %density. Obscuration by a circumnuclear torus can then be seen %against the core and the inner jets, and if the structure has a %significant fraction of atomic gas, \HI\ absorption can be expected %(Conway \& Blanco 1995). Water megamasers are found predominantly in Seyfert 2 (or obscured) galaxies for similar reasons (Braatz, Wilson \& Henkel 1997). The masing gas is usually thought to be located in a circumnuclear accretion disk ($e.g.$ NGC\,4258, Miyoshi et al.\ 1995; NGC\,1068, Greenhill \& Gwinn 1997). In order for the pathlength through the disk to be long enough to obtain significant amplification, the disk must be viewed nearly edge-on. Neufeld \& Maloney (1995) predict that in sources that have a molecular accretion disk sufficiently dense in the central parsec to generate stimulated emission, atomic gas should be present above and below the plane of this disk where the pressure is slightly lower, as well as at larger radii. Although this atomic gas is expected to have a temperature as high as 8000 K, resulting in very low optical depths, detection is possible if the continuum source is bright enough. Thus H$_2$O megamaser sources which have sufficient continuum flux at 1.4 GHz are prime candidates to search for \HI\ absorption (see NGC\,1068, Gallimore, Baum \& O'Dea 1996; NGC\,3079, Satoh et al.\ 1997). Several symmetric radio sources which exhibit \HI\ absorption have been studied with the VLA and the VLBA. A few of these are consistent with the parsec-scale circumnuclear torus model ({\it i.e.} 1946+708, Peck, Taylor \& Conway 1999; \pks, Taylor et al.\ 1999; Hydra~A, Taylor 1996). These sources exhibit very broad (several hundred \kms) shallow systems of multiple blended lines, usually slightly redshifted with respect to the systemic velocity of the galaxy. Not all \HI\ absorption systems are interpreted as the result of a parsec scale torus however. Another class of radio sources toward which \HI\ absorption is seen have very distinct types of absorption systems which are also close to the systemic velocity of the galaxy. These systems are comprised of only a few strong, very narrow (3--20 \kms) lines, and vary little in FWHM or optical depth across the central parsecs of the continuum source ({\it i.e.,} Cen~A, Peck \& Taylor 1998a). %%; NGC 315, Peck 1999). This is thought to be due to a larger, kiloparsec scale disk or dust lane in the host galaxy. In both cases, only a few sources have been well studied. Thus both \HI\ and H$_2$O can be tracers of circumnuclear material. By combining observations of both types, we are attempting to better constrain the distribution of material in the central parsecs of AGN. Our survey to date encompasses several types of radio sources (Seyfert 2, LINER, CSO, etc.) with a large range of radio powers, central masses and estimated accretion rates. The standard model of the central few parsecs should scale accordingly. It might be possible for a radio galaxy with a central mass of 10$^9$ $M_\odot$ to have a molecular disk of radius $\sim$20 pc and maser emission with apparent $\sim$10$^5$ $L_\odot$ isotropic luminosity (Neufeld \& Maloney 1995). Here we present a search for H$_2$O maser emission towards symmetric sources in which the \HI\ absorption is thought to arise in a circumnuclear torus, and a search for \HI\ absorption towards the H$_2$O gigamaser galaxy \txs. In \S 2 we describe the search for water masers toward ``\HI\ torus'' sources carried out with the Effelsberg 100-m telescope, and in \S 3 we describe VLA observations of \HI\ absorption toward \txs. A discussion of our current understanding of the circumnuclear environment is presented in \S 4. We assume H$_0 = 75$ km s$^{-1}$ Mpc$^{-1}$ and q$_0$=0.5 throughout. \section{H$_2$O Maser Observations and Results } The Effelsberg 100-m telescope\footnote{The 100-m telescope at Effelsberg is operated by the Max-Planck-Institut f{\"u}r Radioastronomie in Bonn.} of the MPIfR, equipped with a dual channel K-band HEMT receiver, was used to observe five sources known to have high columns of \HI\ in their central parsecs. The system temperature was 150--250 K on a main beam brightness temperature scale. The beam size of the telescope at 22 GHz is $\sim$40\arcsec. The data were obtained with an autocorrelator with 8$\times$512 channels and a bandwidth of 80 MHz (except on July 6, 2000; see below) for each of the eight backends, leading to channel spacings of 4-5 \kms. The measurements were carried out in a dual beam switching mode with a switching frequency of 1 Hz and a beam throw of 2\arcmin\ in azimuth. Pointing was checked every hour on nearby continuum sources. Linear baselines were removed from the summed spectra and we estimate the flux calibration to be accurate to $\pm$15\%. Whenever not specifically mentioned, W3(OH) (3.2 Jy; Mauersberger et al. 1988) was used for amplitude calibration. No \water\ emission was detected in any of the sources down to the rms levels indicated below. Upper limits for maser luminosities (with isotropic emission assumed here and throughout this paper) are given in Table~1. In some cases continuum detections are reported using the entire 0.5 GHz bandwidth available to the K-band HEMT receiver. %Observations: \subsection{Hydra~A} Hydra~A is a high-luminosity FR I radio galaxy at a redshift of $z$ = 0.05384. \HI\ absorption with a width of 80 \kms\ is detected in this source and thought to lie in a circumnuclear disk or torus with a scale height of $<$30 pc (Taylor 1996). This source was observed at 21.09908 GHz on March 17, 2000. The continuum flux density at this frequency over a 0.5 GHz wide band was 2.0$\pm$0.3 Jy. The H$_2$O noise level obtained with 4.4 \kms\ channel spacing (the channel resolution is about 20\% wider) was 8 mJy for a $-$600 \kms $<$V$<+$580 \kms\ band relative to 21.09908 GHz. \subsection{\pks} \pks\ is the central cD galaxy in the cluster Abell 2597. \HI\ is detected on the kiloparsec scale (O'Dea \etal\ 1994), as well as on smaller scales (Taylor \etal\ 1999). On the parsec scale, two distinct \HI\ absorption lines were detected, one narrow line (110 \kms\ FWHM) is seen toward the core and one of the inner jet components, and a very broad line (735 \kms\ FWHM) appears toward only the core. This broad line is consistent with a compact, rapidly rotating circumnuclear disk with a scale height of $<$20 pc (Taylor \etal\ 1999). This source was observed at 20.546 GHz ($z$ = 0.08220 for the H$_2$O rest frequency of 22.23508 GHz) on March 18, 2000. We did not try to measure a continuum flux. Calibration at the sky frequency was obtained towards NGC\,7027 (5.55 Jy according to Ott et al. 1994). With 4.6 \kms\ channel separation we obtained an rms noise level of 4 mJy for $-$1050 \kms $<$V$<+$1150 \kms\ relative to 20.546 GHz. \subsection{NGC\,3894} NGC\,3894 is a nearby ($z$=0.01068) elliptical galaxy with twin relativistic parsec-scale jets (Taylor, Wrobel \& Vermeulen 1998). The structure of the \HI\ detected in absorption in this source is substantially more complicated than in Hydra~A or \pks. High resolution observations (Peck \& Taylor 1998b) indicate that the absorbing gas might lie in a circumnuclear torus, a large number of clouds along the line of sight to the radio continuum source, or most likely, a combination of both. This source was observed on 2000 March 19 and 2000 July 6, in the latter case with 40 MHz backends. Rms limits are 4 mJy for 2800 \kms\ to 4300 \kms\ and 6 mJy for 2050 \kms\ to 2800 \kms\ LSR in March. In July, the rms was 5 mJy for 2620 \kms\ to 3880 \kms. In both cases, the channel separation was 4.2 \kms. \subsection{NGC\,4151} NGC\,4151 is a nearby ($z$=0.00332; distance=16.5 Mpc) Seyfert 1.5 galaxy which has been well-studied at most wavebands. MERLIN observations indicate that the \HI\ absorption in this source occurs only toward the core with a linewidth of 90 \kms\ in the strongest component. If this gas is in a toroidal structure, it must be less than 50 pc in height (Mundell \etal\ 1995). This source was the nearest target in our sample. NGC\,4151 was observed on March 20, 2000. We obtained 3 mJy rms noise levels for 0 \kms\ to 2000 \kms\ LSR with a channel spacing of 4.2 \kms. \subsection{1946+708} 1946+708 is a CSO at a redshift of $z$=0.101. This source was initially found to exhibit \HI\ absorption in VLA observations in 1995 (Peck, Taylor \& Conway 1999). This absorption was then imaged at very high angular resolution with the Global VLBI Network, showing the region of broadest absorption (FWHM$\simeq$400 \kms) toward the core, and several narrower components toward both the approaching and receding jets. The absorbing gas in 1946+708 is thought to occur in a circumnuclear torus with an outer radius of at least 80 pc and a scale height of $<$10 pc. Further evidence for a torus in this source is found in free-free absorption (Peck \& Taylor 2001). This source was observed with the Effelsberg 100-m telescope three times, on March 19, April 13, and July 6. As with NGC\,3894, the July data were obtained with 40 MHz backends. The continuum flux was measured on April 13: 0.27$\pm$0.04 Jy at 20.19535 GHz (with a bandwidth of 0.5 GHz). With a channel separation of 4.6 \kms\ we obtained 2-3 mJy rms values between $-$1050 $<$ V $<$ 1180 \kms\ in March, 5 mJy between $-$900 $<$ V $<$1000 \kms\ in April, and 4 mJy between $-$550$<$ V $<$ 730 \kms\ in July (all velocities with respect to the central observing frequency of 20.19535~GHz). %GBT: Can these results on 1946 be combined to get a lower limit? % CH: No. Combining all data, the total % rms becomes 2 mJy. This is not a significant gain % w.r.t. the 2-3 mJy in March. Because of this % and potential H2O time variability, it makes sense % to present the noise levels separately. \section{\txs\ \HI\ Observations and Results} The observations were made with Very Large Array\footnote{The National Radio Astronomy Observatory is operated by Associated Universities, Inc., under cooperative agreement with the National Science Foundation.} at a center frequency of 1385 MHz on 2001 April 19 and 20. The array was in the 'B' configuration, which provided an angular resolution of $\sim$6\arcsec. A total of 3.6 hours were obtained on source on April 19th using 32 channels across a 6.25 MHz band to provide a resolution of 43 km/s. Both right and left circular polarizations were observed. Phase and bandpass calibration was obtained by short (1 min) observations of the strong (2.1 Jy) calibrator J2246{\tt -}1206 every $\sim$20 minutes. The data were loaded in real time and a rough reduction revealed the \HI\ detection. For the observations on the following day the velocity resolution was doubled (to 22 km/s) at the expense of only recording in right circular polarization. A total of 2.3 hours were obtained on source. %The last hour of observations on April 20 was lost due to high winds. Following standard VLA calibration practices, a continuum visibility data set and image were generated from the line data for each epoch by averaging the line-free channels. These data were then self-calibrated to remove small residual phase errors. The rms noise in the final continuum image of \txs\ is 0.10 mJy. The continuum emission from \txs\ is unresolved (size $<$3 kpc) and there are no other sources stronger than 0.3 mJy within one arcminute. The self-calibration solutions were then transferred to the line data. Spectral line cubes were made for \txs\ and all the sources in the field (see Fig.~1) stronger than 2 mJy simultaneously using the AIPS task IMAGR. The rms noise in a single channel from the combined data set is 0.25 mJy/beam, while in the higher spectral resolution data the rms noise in a channel is 0.66 mJy/beam. The higher noise is the result both narrower channels (by factor 0.5) and less data (factor 0.24). In Fig.~2 we present the 43 km/s resolution spectrum of \txs. In addition to a component with width $\sim$125 km/s, there appear substantial wings to the line, indicative of a broader (420 km/s) component. Two Gaussian components have been fit to the data and are drawn individually in Fig.~2 along with their sum. Parameters of the fits are given in Table~2. The higher resolution (22 km/s) spectrum is shown in Fig.~3 with the Gaussian fits overlaid. \subsection{\txs\ Continuum Observations and Results} The continuum observations were made with Very Large Array on 2001 April 24 at C, X, U, and K bands. A total of 1 hour was used to obtain the measurements presented in Fig.~4 and Table~3. The observations at frequencies of 15 GHz and below consisted of a single scan on \txs\ of 1--3 minutes duration bracketed by a 1 minute integration on the strong calibrator J2246{\tt -}1206. At 22 GHz we employed the fast switching mode using a 3 minute cycle time for 18 minutes and the nearby calibrator J2236{\tt -}1433. From 1.4 to 5 GHz the spectral index, $\alpha$, where $S_\nu \propto \nu^\alpha$, is $\alpha = -0.66$. Between 5 and 22 GHz the spectrum steepens to a power law of slope $-$0.92. The highest resolution, obtained at 22 GHz, is 0.39 $\times$ 0.21\arcsec\ in position angle $-$12\deg. At this resolution \txs\ is found to be slightly resolved with an extension along PA 145.3\deg. This is in good agreement with the 'A' configuration VLA image presented by Falcke \etal\ (2000) at 8.46 GHz. \section{Discussion} \txs\ contains the most luminous known H$_2$O megamaser with a luminosity in the 1.3 cm line of 6100 $L_\odot$ (Koekemoer et al.\ 1995). The water maser emission from \txs\ is fairly broad, with a FWHM of 88 km s$^{-1}$. Recent HST observations by Falcke et al.\ (2000) classify the galaxy as a spiral and reveal a dust lane cutting across the nucleus. Falcke et al.\ (2000) also present VLA observations at 8.4 GHz showing that the radio emission is compact ($<$ 1\arcsec), symmetric, and has an axis perpendicular to the dust lane. No larger-scale diffuse emission is present to the sensitivity limits of the Northern VLA Sky Survey (NVSS) at about 2 mJy/beam (Fig.~1; Condon et al.\ 1998). Although the $\sim$420 \kms\ absorption probably results from neutral material associated with the atomic and molecular torus thought to feed the active nucleus, the deeper $\sim$125 \kms\ line in \txs\ could be indicative of an interaction between the radio jet and surrounding material. The water maser emission is also fairly broad (88 km/s) in \txs, and could likewise originate from the central torus or from a jet-cloud interaction. VLBA observations of both the \water\ emission and \HI\ absorption could help to discriminate between these two models. Both the \water\ megamaser emission and the \HI\ absorption are thought to arise from the central 0.1 to 100 parsecs of the molecular and atomic torus around an AGN. In searching for these two processes, however, it is important to keep in mind that the observed flux density of the maser emission falls off with distance (although this may be less steep than the inverse square-law if the water maser emission is amplifying a beamed nuclear continuum), while the \HI\ absorption relies on the presence of suitably bright radio emission, but is independent of distance. Perhaps because of this difference, maser emission and \HI\ absorption studies thus far have focused on different classes of AGN. The majority of megamaser surveys have targeted low luminosity AGN, predominantly optically selected and relatively nearby Seyferts and LINER galaxies, while the \HI\ surveys, encompassing a much smaller number of sources but usually obtaining longer integrations, have mainly targeted powerful radio galaxies. There has been one megamaser survey directed toward FRI radio galaxies (Henkel et al. 1998) and a survey of FRII galaxies and BL Lac objects has also been made (E. Ros, A. Tarchi, priv. communication), but no detections were found and the upper limits of the two latter studies have yet to be determined. %A megamaser survey %has been directed toward FRI radio galaxies (Henkel \etal\ 1998) and %even BL Lac objects (unpublished), but the integration times have been %short and the upper limits found in these are %$\sim10^3$ L$_{\odot}$. In Table~1 we summarize the properties of all known \water\ megamasers and the \HI\ absorption detections in these systems to date. This table illustrates a fair amount of overlap with 8/19 (42\%) of the sources having both \water\ megamaser emission and \HI\ absorption. Not all megamaser sources shown have been searched carefully for \HI\ absorption, so the actual overlap could be somewhat larger. Given the fairly high incidence of \HI\ absorption in systems with \water\ megamaser emission previously reported in the literature, we carried out a moderately deep search for \water\ megamaser emission in five sources with optical depths in \HI\ of 0.1 or more. None of these sources were detected (see Table~1 for the luminosity limits derived from the observations). The luminosity limits for the two closest sources, NGC\,4151 and NGC\,3894 are below those of all known \water\ megamasers, so in the case of isotropic emission we would have expected to detect those sources (see Fig.~5). For the other, more-distant, systems the luminosity limits are fairly high so it is still possible that faint \water\ megamasers reside there, or that these systems are also beamed away from us. As evident from Fig.~5, there is only a weak correlation between radio power and \water\ maser luminosity, with a Spearman's rank-order correlation coefficient of 0.465 leading to a probability that the two quantities are correlated at the 97.5\% level. The \HI\ absorbers detected so far appear to be distributed in a similar way as the megamasers without any \HI\ absorption. There are several possible reasons for not detecting \water\ emission from the nearby sources, even if the interpretation of the \HI\ is correct, and a pc scale molecular disk exists within the torus, as in the scenario proposed by Neufeld \& Maloney (1995) based on the prototypical source NGC~4258. First, the molecular disk is expected to be extremely thin and warped by radiation pressure (Pringle 1996), making the detection of \water\ much more dependent on the geometry and orientation of the disk than the detection of \HI\ absorption is on the orientation of the larger torus. In some sources, the radio axis might be close to the plane of the sky, but the regions in the warped disk which have suitable temperatures and pressures to obtain population inversion do not lie along our line of sight to a continuum source nor have a long enough path length to generate self-amplification. Another possibility is that the model wherein a molecular disk is embedded in the \HI\ torus simply does not scale to higher powered AGN. A higher accretion rate, a stronger gravitational potential, a stronger central x-ray source, or a combination of these factors might alter the temperature and pressure of the mid-plane of the torus and thus the conditions for establishing \water\ masers is not met in these systems. Conversely, the non-detection of \HI\ absorption in a few megamaser galaxies might indicate that the \water\ emission is the result of an interaction between the jet and a nearby molecular cloud, rather than arising in an accretion disk, as is thought to be the case in Mrk~348 (Peck \etal\ 2001, in preparation), or could possibly be the result of very high spin temperatures in the atomic part of the torus, yielding extremely low \HI\ optical depths. Braatz, Wilson \& Henkel (1997) find a detection rate for extragalactic \water\ masers of 7.2\% in Seyferts 2's and LINERs in a distance-limited sample. This is a greater success rate than we achieved in selecting systems with \HI\ absorption, but our sample is still too small to say anything meaningful about the difference. However, the number of sources with known \HI\ absorption is increasing dramatically thanks to new facilities at the Giant Meterwave Radio Telescope (GMRT) and Westerbork Synthesis Radio Telescope (WSRT). The detection of a torus signature in either \HI\ or \water\ masers is of interest since it can then be followed up with VLBA studies which can provide important information about the spatial distribution and velocity field of the accreting gas and the mass of the central engine. %The detection of a dust lane signature would %contribute to our knowledge of the host galaxies of radio sources. \section{Conclusions} We report on the detection of broad \HI\ absorption in the gigamaser galaxy \txs. Sensitive VLBI observations are being planned in order to investigate the location and kinematics of the \HI\ gas. It will also be of interest to learn more about the parsec-scale morphology of this exceptional object. Among known \water\ masers the incidence of \HI\ absorption is fairly common ($>$42\%). Although a moderately deep survey of five radio galaxies failed to yield any new \water\ megamasers, there is still some promise to a physically motivated search. In the near future it should be possible to start with larger samples of \HI\ absorbing systems and refine our understanding of the connection between the \HI\ absorbers and the \water\ emitters. \acknowledgments We thank the referee for insightful comments on an earlier draft of the manuscript. This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, Caltech, under contract with NASA. This research has also made use of NASA's Astrophysics Data System Abstract Service. \clearpage \begin{references} \reference{bra97}Braatz, J. A., Wilson, A. S. \& Henkel, C. 1997, ApJS, 110, 321 \reference{bra96}Braatz, J. A., Wilson, A. S. \& Henkel, C. 1996, ApJS, 106, 51 \reference{con98}Condon, J.J., Cotton, W.D., Greisen, E.W., Yin, Q.F., Perley, R.A., Taylor, G.B., \& Broderick, J.J.\ 1998, AJ, 115, 1693 \reference{con96}Conway, J. E. 1996 in {\it The Second Workshop on Gigahertz Peaked Spectrum and Compact Steep Spectrum Radio Sources}, ed. I. A. G. Snellen, R. T. Schilizzi, H. J. A. Rottgering and M. N. Bremer [Leiden Observatory], 198 \reference{con95}Conway, J. E \& Blanco, P. R. 1995 ApJ, 449, L131 \reference{fal00}Falcke, H., Wilson, A.S., Henkel, C., Brunthaler, A., \& Braatz, J.A.\ 2000, ApJ, 530, L13 \reference{gal96}Gallimore, J. F., Baum, S. A. and O'Dea, C. P. 1996 ApJ, 464, 198 \reference{gal99}Gallimore, J. F., Baum, S. A., O'Dea, C. P., Pedlar, A. \& Brinks, E. 1999, ApJ, 524, 684 \reference{jvg86}van Gorkom, J. H., Knapp, G. R., Raimond, E., Faber, S. M., \& Gallagher, J. S.\ 1986, AJ, 91, 791 \reference{jvg89}van Gorkom, J. H., Knapp, G. R., Ekers, R. D., Ekers, D. D., Laing, R. A., \& Polk, K. S. 1989, AJ, 97, 708 \reference{gre97}Greenhill, L. J. \& Gwinn, C. R. 1997, Ap\&SS, 248, 261 \reference{gr297}Greenhill, L. J., Herrnstein, J. R., Moran, J. M., Menten, K. M. \& Velusamy, T. 1997, ApJ, 486, L15 \reference{gre95}Greenhill, L. J., Henkel, C., Becker, R., Wilson, T. L. \& Wouterloot, J. G. A. 1995, A\&A, 304, 21 \reference{hag00}Hagiwara Y., Diamond, P. J., Nakai, N. 2000, in {\it Proceedings of the 5th EVN Symposium} Eds. J. Conway, A. Polatidis, R.Booth \& Y. Pihlstr\"om, Onsala Space Observatory, Chalmers Technical University, Gothenburg, Sweden \reference{hen98}Henkel, C., Wang, Y.~P., Falcke, H., Wilson, A.~S. \& Braatz, J.~A. 1998, A\&A, 335, 463 \reference{koe95}Koekemoer, A. M., Henkel, C., Greenhill, L.J., Dey, A., van Breugel, W., Codella, C., \& Antonucci, R.\ 1995, Nature, 378, 697 \reference{kor96}Koribalski B., 1996, in {\it The Minnesota Lectures on Extragalactic Neutral Hydrogen} ASP Conf. Ser. Vol. 106, ed. Skillman E. D., [ASP: San Francisco], p.238 %\reference{mar98}Martel, A. R., Baum, S. A., Sparks, W. B., Wyckoff, % E., Biretta, J. A., Golombek, D., Macchetto, F. D., McCarthy, P. J., % De Koff, S. \& Miley, G. K. 1998, BAAS, 192, 5204 \reference{mau88}Mauersberger, R., Wilson, T. L. \& Henkel, C. 1988, A\&A 201, 123 \reference{miy95}Miyoshi, M., Moran, J., Herrnstein, J., Greenhill, L., Nakai, N., Diamond, P. \& Inoue, M. 1995, Nature, 373, 127 \reference{mun95}Mundell, C. G., Pedlar, A., Baum, S. A., O'Dea, C. P., Gallimore, J. F., \& Brinks, E. 1995, MNRAS, 272, 355 \reference{neu95} Neufeld, D. A. \& Maloney, P. R. 1995, ApJ, 447, L17 %\reference{ode86}O'Dea, C. P. and Owen, F. N. 1986, ApJ, 301, 841 %\reference{ode94}O'Dea, C. P., Baum, S. A., \& Gallimore, J. F. 1994, ApJ, 436, 669 \reference{ott01}Ott, M., Whiteoak. J. B., Henkel, C. \& Wielebinski, R. 2001, A\&A, 372, 463 \reference{ott94}Ott, M., Witzel, A., Quirrenbach, A., Krichbaum, T. P., Standke, K. J., Schalinski, C. J. \& Hummel, C. A. 1994, A\&A 284, 331 \reference{pec98}Peck, A. B. \& Taylor, G. B. 1998a, BAAS, 193, 620 \reference{pec98}Peck, A. B., \& Taylor, G. B. 1998b, ApJ, 502, L23 %\reference{peb99}Peck, A. B. 1999, Ph.D. Thesis, New Mexico Institute of Mining and Technology \reference{pec99}Peck, A. B., Taylor, G. B. \& Conway, J. E. 1999, ApJ, 521, 103 \reference{pec98}Peck, A. B., \& Taylor, G. B. 2001, ApJ, 554, L147 \reference{pih00} Pihlstr\"om, Y.M., Conway, J.E., Booth, R.S., Diamond, P.J. \&Koribalski, B.S. 2000 A\&A, 357, 7 \reference{pih01} Pihlstr\"om, Y.M. 2001, Ph.D. Thesis, Chalmers University \reference{pri96}Pringle, J. E. 1996, MNRAS, 281, 357 \reference{sat00}Sawada-Satoh, S., Inoue, M., Shibata, K. M., Kameno, S., Migenes, V., Nakai, N. \& Diamond, P. J. 2000, PASJ, 52, 421 \reference{sat97}Satoh, S., Inoue, M., Nakai, N., Shibata, K. M., Kameno, S., Migenes, V., Diamond, P. J. 1997, in {\it The Central Regions of the Galaxy and Galaxies}, IAU Symposium 184, 208 \reference{tay96}Taylor, G. B. 1996, ApJ, 470, 394 \reference{tay98}Taylor, G. B., Wrobel, J. M., \& Vermeulen, R. C.\ 1998, ApJ, 498, 619 \reference{tay99}Taylor, G. B., O'Dea, C. P., Peck, A. B.\ \& Koekemoer, A. M. 1999, ApJ, 512, L27 \end{references} \clearpage %\baselineskip=5pt \tightenlines \singlespace %\setcounter{table}{2} \begin{center} \begin{deluxetable}{l c c c c c r r c c c c} \tablecolumns{10} \tablewidth{-10pc} \tablecaption{Megamaser and \HI\ Detection Summary} \tabletypesize{\scriptsize} \tablehead{ & Host & Isotropic & H$_2$O &H$_2$O &Maser & Isotropic & Peak \HI\ & Peak \HI\ &\HI\ \\ Megamaser & Galaxy & Lum. & FWHM &V$_{\rm LSR}$& Ref. & $P_{\rm 1.4 GHz}$ & Opt. Depth &FWHM & ref. \\ Source & $\bullet$ & (L$_{\odot}$) & (\kms) &(\kms) & & (W Hz$^{-1}$) & ($\tau$) &(\kms) &} \startdata Circinus & Sy2 & 24 & 1& 274--562& 1 & $6.8 \times 10^{21}$ & \nodata & $\sim$180 & 2 \\ %%% 1.5 Jy at 449 km/s redshift 0.0015 1.9e25 cm ESO~103-G35 & Sy1/2 & 360 &8&4073& 1 & \nodata &\nodata&\nodata&\nodata\\ %%% no L band at 4073 km/s IC~2560 & Sy2 & 130 &25&2919 & 1 & $5.6 \times 10^{21}$ &\nodata&\nodata&\nodata \\ %%% 0.031 Jy at 2925 km/s z=0.0098 1.2e26 IC~1481 & L & 320 &2&6235--6237 & 1 & $2.9 \times 10^{22}$ &\nodata&\nodata&\nodata \\ %%% 0.036 at 6118 km/s z=0.0204 2.5e26 Mrk~1 & Sy2 & 64 & 3 -- 6&4868--4869 & 1 & $3.8 \times 10^{22}$ &\nodata&\nodata&\nodata\\ %%% 0.076 Jy at 4780 km/s z=0.0159 2.0e26 Mrk~348 & Sy2 & 450 & 133&4610--4678& 3 & $1.4 \times 10^{23}$ &$<$0.004& & 4\\ %%% 0.293 at 4507 km/s 0.0150 1.9e26 Mrk~1210 & Sy2 & 99 &0.8 -- 11& 4213--4314& 1 & $4.2 \times 10^{22}$ &\nodata&\nodata&\nodata\\ %%% 0.114 at 4046 km/s z=0.0135 1.7e26 NGC\,1052 & L/Sy2 & 140 &80 -- 120&1605--1668& 1 & $4.0 \times 10^{22}$ & 0.021 & 35 & 5\\ %%% 0.850 at 1470 km/s z=0.0049 6.1e25 NGC\,1068 & Sy1/2 & 170& &800--1500& 1 & $1.4 \times 10^{23}$ &0.078 & 161 & 4\\ %%% 4.845 Jy at 1137 km/s z=0.0038 4.7e25 NGC\,1386 & Sy2 & 120 &4 &969& 1 & $6.1 \times 10^{20}$ &\nodata&\nodata&\nodata\\ %%% 0.0371 Jy at 868 km/s z=0.0029 3.6e25 NGC\,2639 & Sy1.9 & 71 &1 -- 7 &3300--3323& 1 & $2.6 \times 10^{22}$ & $<$0.003 & & 4\\ %%% 0.103 at 3336 km/s z=0.0111 1.4e26 NGC\,3079 & L/Sy2 & 520 &10 &926--1190& 1,6 & $2.3 \times 10^{22}$ &0.8 & 133 & 4,6\\ %%% 0.845 Jy at 1125 km/s z= 0.0038 4.7e25 NGC\,3735 & Sy2 & 12 & 1.4--3&2262--2268& 7 & $1.3 \times 10^{22}$ &\nodata&\nodata&\nodata\\ %%% 0.086 at 2696 km/s z=0.0090 1.1e26 NGC\,4258 & L/Sy1.9 &85& 1 -- 7 &-365--1460& 1,8 & $3.6 \times 10^{21}$ &$<$1.4&\nodata& 9 \\ %%% 0.792 at at 448 km/s z=0.0015 1.9e25 NGC\,4945 & Sy2 & 57 &3& 694& 1 & $4.4 \times 10^{22}$ &0.29&$\sim$250&10 \\ %%% 6.60 Jy at 560 km/s z=0.0019 2.3e25 NGC\,5347 & Sy2 & 32 &15 &2376--2401& 1 & $6.5 \times 10^{20}$ & \nodata&\nodata&\nodata\\ %%% 0.0056 at 2335 km/s z=0.0078 9.6e25 NGC\,5506 & Sy1.9 & 61 &0.7 -- 6&1731--1821 & 1 & $2.5 \times 10^{21}$ &0.109 & 95 & 4\\ %%% 0.0355 Jy at 1853 km/s z=0.0061 7.5e25 NGC\,5793 & Sy2 & 125 &1.3, 14 &3190--3677 &11 & $3.0 \times 10^{23}$ &3& 16, 50 & 12\\ %%% 1.20 Jy at 3491 km/s z=0.0116 1.4e26 TXS~2226{\tt -}184 & E/S0 & 6100 &88 & 7570 & 1 & $8.7 \times 10^{22}$ &0.1& 125, 418& 13\\ %%% 0.0717 at 7495 km/s z=0.0250 3.1e26 \tableline & Host & 3$\sigma$ H$_2$O &Lum. &V$_{\rm LSR}$ &Maser & Isotropic & Peak \HI\ & Peak \HI\ &\HI\\\ \HI\ Torus & Galaxy & Upper lim. &limit\tablenotemark{\dag} &Searched\tablenotemark{*} &Ref. & $P_{\rm 1.4 GHz}$ & Opt. Depth & FWHM & ref. \\ Source & & (mJy) &(L$_{\odot}$) & (\kms) & & (W Hz$^{-1}$) & ($\tau$) &(\kms) & \\ \tableline Hydra~A & E & 13& $<$140 &15552--16732 &13 & $2.3 \times 10^{26}$ &0.99 &80 &14 \\ %%% 40.8 Jy at z=0.05384 6.7e26 PKS~2322{\tt -}123& E & 12 & $<$310 &23610--25811 &13 & $2.5 \times 10^{25}$ &0.3,0.6 &735, 110 &15 \\ %%% 2.0 Jy at z=0.08220 1.0e27 NGC\,3894& E & 12 & $<$ 5 &2050--4300 &13 & $8.7 \times 10^{22}$ &0.1 &80 &16 \\ %%% 0.415 Jy at 3223 km/s z=0.0107 1.3e26 NGC\,4151& Sy1.5 & 9 & $<$0.6 &0--2000 &13 & $7.0 \times 10^{21}$ &0.23 &90 &17 \\ %%% 0.333 Jy at 995 km/s z=0.0033 4.1e25 1946+708& E & 12 & $<$470 &29250--31480 &13 & $2.0 \times 10^{25}$ &0.1 &375 &18 \\ %%% 0.953 Jy at z=0.101 1.3e27 \tableline \enddata \tablecomments{References: (1) Braatz et al 1996 and references therein, (2) Koribalski 1996, (3) Falcke \etal\ 2000, (4) Gallimore et al 1999, (5) van Gorkom, \etal\ 1986, (6) Sawada-Satoh \etal\ 2000, (7) Greenhill \etal\ 1997, (8) Greenhill \etal\ 1995, (9) Mundell \etal\, in preparation, (10) Ott \etal\ 2001, (11) Hagiwara \etal\ 2000, (12) Pihlstr\"om \etal\ 2000, (13) this paper, (14) Taylor 1996, (15) Taylor \etal\ 1999, (16) Peck \& Taylor 1998b, (17) Mundell \etal\ 1995, (18) Peck, Taylor \& Conway 1999 \\ \noindent $^{\bullet}$ Sy=Seyfert, L=LINER, E=Elliptical. \\ \noindent $^{\dag}$ calculated assuming a linewidth of 10 \kms. \\ \noindent $^{*}$ optical velocity convention (cz).} \end{deluxetable} \end{center} \clearpage \begin{center} %\baselineskip=5pt \tightenlines \singlespace TABLE 2 \\ \smallskip \txs\ \HI\ C{\sc olumn} D{\sc ensities} \smallskip \begin{tabular}{l r r r r r r r r} \hline \hline & $S_{\rm 1.4 GHz}$ & $S_{\rm line}$ & $V$ & $\Delta V$ & $\tau$ & $N_{\rm H}$ \\ Component & (mJy) & (mJy) & (km s$^{-1}$) & (km s$^{-1}$) & & (cm$^{-2}$) \\ (1) & (2) & (3) & (4) & (5) & (6) & (7) \\ \hline \noalign{\vskip2pt} narrow & 73.5 & 12.3 $\pm$ 0.98 & 7491 $\pm$ 3 & 125 $\pm$ 10 & 0.183 & 3.3 $\times 10^{23}$ \\ broad & 73.5 & 5.6 $\pm$ 0.74 & 7454 $\pm$ 18 & 418 $\pm$ 72 & 0.079 & 4.8 $\times 10^{23}$ \\ \hline \end{tabular} \end{center} \begin{center} {\sc Notes to Table 2} \end{center} \noindent Col.(1).---Component name. Col.(2).---Continuum flux density at 1.4 GHz in mJy. Col.(3).---Depth of the line in mJy. Col.(4).---Central velocity (c$z$) in km s$^{-1}$. Col.(5).---FWHM in velocity in km s$^{-1}$. Col.(6).---Optical depth, assuming a continuum source covering factor of unity. Col.(7).---The column density in units of cm$^{-2}$ calculated assuming a spin temperature of 8000 K (Conway \& Blanco 1995). If the \HI\ absorption results from gas outside the nuclear region then a more reasonable assumption for the spin temperature would be 100 K, reducing the column densities quoted here by a factor of 80. \bigskip \begin{center} TABLE 3 \\ \smallskip C{\sc ontinuum} O{\sc bservations} \smallskip \begin{tabular}{r r r r } \hline \hline $\nu$ & $S_{\rm VLA}$ & time & $\Delta\nu$ \\ (GHz) & (mJy) & (min) & (MHz) \\ \hline \noalign{\vskip2pt} 1.385 & 73.3 $\pm$ 2.2\phantom{0} & 350\phantom{.0} & 5 \\ 4.860 & 32.2 $\pm$ 0.98 & 1.4 & 50 \\ 8.460 & 20.0 $\pm$ 0.60 & 0.8 & 50 \\ 14.940 & 11.7 $\pm$ 0.38 & 2.0 & 50 \\ 22.460 & 7.87 $\pm$ 0.41 & 10.2 & 50 \\ \hline \end{tabular} \end{center} \smallskip \clearpage \bigskip \clearpage % Figure 1 -- NVSS field \begin{figure} \plotone{f1.eps} \caption{The NVSS (Northern VLA Sky Survey) image from Condon et al.\ 1998. Spatial resolution is 45\arcsec. Contours are drawn at $-$1, 1, 1.4, 2, ..., 181 mJy/beam by $\sqrt{2}$ intervals with negative contours shown dashed. The cross marks the position of \txs. \label{fig1}} \end{figure} % Figure 2 -- Low resolution HI spectrum \begin{figure} \plotone{f2.eps} \caption{The low resolution \HI\ spectrum towards \txs\ derived from combined observations on 2001 April 19 and 20. The velocity frame is LSR (Local Standard of Rest) and the velocity resolution is 43 \kms. The solid line represents the best fit of two Gaussians (dotted and long-dash lines) to the data. \label{fig2}} \end{figure} % Figure 3 -- High resolution HI spectrum \begin{figure} \plotone{f3.eps} \caption{The high resolution \HI\ spectrum towards \txs\ derived from observations on 2001 April 20. The velocity frame used is LSR and the velocity resolution is 22 \kms. The solid line represents the best fit of two Gaussians (dotted and long-dash lines) to the data. \label{fig3}} \end{figure} % Figure 4 -- Continuum emission spectrum \begin{figure} \plotone{f4.eps} \caption{The broadband continuum emission spectrum towards \txs\ derived from observations in April 2001. \label{fig4}} \end{figure} % Figure 5 -- H2O lum vs radio power \begin{figure} \plotone{f5.eps} \caption{A plot of \water\ Maser Luminosity against the Radio Power at an observed frequency of 1.4 GHz for all the sources in Table~1. Sources plotted with filled circles have \HI\ detected in absorption, while those plotted with empty circles have either not been detected or not been observed. \label{fig5}} \end{figure} \end{document}