%%Beginning of first draft by jsu, 31dec98 % Almost completed, jan 2-4/99 % 2nd draft, 18jan99--22jan99 %\documentstyle[12pt,amssym,aasms4]{article} \documentstyle[11pt,amssym,aaspp4]{article} \received{DRAFT 2 - 22 January 1999} \accepted{} \journalid{}{} \articleid{}{} \lefthead{Ulvestad et al.} \righthead{Non-relativistic Seyfert Jets} \def\H2{\ion{H}{2}} \def\hbeta{H$\beta$} \begin{document} \title{Non-Relativistic Radio Jets and Parsec-Scale \\ Absorption in Two Seyfert Galaxies } \author{J.S.~Ulvestad\altaffilmark{1}, J.M.~Wrobel\altaffilmark{1}, A.L.~Roy\altaffilmark{1,2}, A.S.~Wilson\altaffilmark{3}, \\ H.~Falcke\altaffilmark{2}, \& T.~Krichbaum\altaffilmark{2} } %G.~Bower\altaffilmark{1,2}, A.~Zensus\altaffilmark{2} } \altaffiltext{1}{National Radio Astronomy Observatory, P.O. Box O, Socorro, NM 87801; julvesta,jwrobel@nrao.edu} \altaffiltext{2}{Max Planck Institut f\"ur Radioastronomie, Auf dem H\"ugel 69, D-53121 Bonn, Germany; aroy,hfalcke,tkrichbaum, @mpifr-bonn.mpg.de} \altaffiltext{3}{Department of Astronomy, University of Maryland, College Park, MD 20742; wilson@astro.umd.edu} \begin{abstract} The Very Long Baseline Array has been used to image the milliarcsecond-scale 15-GHz radio emission from the Seyfert galaxies Mrk~231 and Mrk~348 at two epochs separated by about 1.7~yr. Both galaxies contain parsec-scale double radio sources, whose components have brightness temperatures of $10^9$--$10^{11}$~K, indicating that they are generated by synchrotron emission and probably associated with radio jets. The nuclear components of both galaxies are identified by strong variability between epochs, implying that the double sources are apparently one-sided jets. In Mrk~348, the relative component speed is $0.074\pm 0.035c$ at a separation of 0.5~pc, while the apparent speed in Mrk~231 is $0.139\pm 0.052c$ at a separation of 1.1~pc (using $H_0=65$~km~s$^{-1}$~Mpc$^{-1}$). The lack of observed counterjets is probably due to free-free absorption by torus gas having ionized densities $n_e > 2\times 10^5$~cm$^{-3}$ within a parsec of the nuclei. This gas density is consistent with X-ray absorption measurements for both galaxies, and also is similar to that inferred in H$_2$O megamaser galaxies. Except for radio power, the overall properties of the Seyfert radio sources are similar to compact symmetric objects, whose one-sided jets also could be due to free-free absorption by gas having $n_e\sim 5\times 10^4$~cm$^{-3}$ on scales of 5--10~pc. \end{abstract} \keywords{galaxies: active --- galaxies: individual (Mrk~231, Mrk~348=NGC~262) --- galaxies: jets --- galaxies: nuclei --- galaxies: Seyfert --- radio continuum: galaxies} \section{Introduction} Seyfert galaxies have weak and small radio sources relative to radio galaxies and quasars, with typical sizes $\leq 500$~pc and typical centimeter-wavelength powers of $\leq 10^{23}$~W~Hz$^{-1}$ (\cite{ulv89}). These sources are apparently produced by weak jets propagating within ionization cones (whose axes are determined by the galaxies' central obscuring disks) and interacting with thermal gas in those cones (\cite{fal98}). The presence of disks is strongly supported by the detection of nuclear H$_2$O masers in a number of Seyfert~2 galaxies (\cite{bra96}), by VLBI imaging of those masers in galaxies such as NGC~4258 (\cite{her97}), and by HI absorption seen on parsec scales in weakly active galaxies (\cite{pec98}). %The weak radio emission in such %radio-quiet active galactic nuclei has been %variously attributed to low accretion rates (\cite{fal95}), %low black-hole spin rates (\cite{wil95}; \cite{mei97}), %offsets between the angular momentum axes of the central black %hole and the accreting gas (\cite{cal97}), %or strong interactions between the jet and the surrounding %gaseous medium (\cite{bic98}). It is of considerable interest to measure the speeds of Seyfert radio jets as close to the nucleus as possible, since such measurements present the possibility of differentiating between ``intrinsic'' and ``environmental'' effects in Seyferts. Previous measurements of jet speeds in Seyfert cores are rare. In NGC~4151, upper limits of $0.14c$ and $0.25c$ have been measured on scales of 7 and 36~pc (\cite{ulv98}). In NGC~1068, component positions measured with the Very Large Array and with the Very Long Baseline Array (VLBA) 12.5~yr later imply an upper limit of $\sim 0.5c$ for components $\sim 20$~pc apart (\cite{ulv87}; \cite{roy98}). The galaxies Mrk~231 (UGC~08058) and Mrk~348 (NGC~262) contain two of the strongest radio sources found in Seyferts, and are prime candidates for the measurement of motions in their cores. Within larger-scale VLBI structures, each galaxy contains a double radio source with a total size of $\sim 1$~pc or less (\cite{hal97}; Ulvestad, Wrobel, \& Carilli 1999, hereafter UWC). These central sources have now been imaged at two epochs using the VLBA;\footnote{The VLBA is part of the National Radio Astronomy Observatory, a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.} this {\it Letter} reports measurements of the component separation speeds on scales of $\leq 1.1$~pc. \section{VLBA Observations \& Data Analysis} \label{vlbaobs} The 10 telescopes of the VLBA (\cite{nap94}) have been used to observe Mrk~231 and Mrk~348 at left-circular polarization at two epochs separated by $\sim 1.7$~yr, between 1996 and 1998. The images at 15~GHz have the best combination of resolution and sensitivity for the compact components, and are the subject of this {\it Letter}. In all cases, 2-bit sampling was employed for four 8-MHz channels, and the central frequency was 15.365~GHz; on-source integration times ranged from 72 to 156~minutes. %Table~\ref{tab:obs} contains a journal of the observations. The first-epoch observations were affected somewhat by poor weather at several stations. All data were initially calibrated by using measured gains and system temperatures. Residual delays and rates were removed in AIPS (\cite{van96}) by fitting to the program galaxies or to phase-referencing sources. (Since the first-epoch observations were not phase-referenced, absolute positions have not been derived.) All data were iteratively imaged and self-calibrated in DIFMAP (\cite{she97}). %In Mrk~231, this process removed the possible weak southwestern %source seen in the first epoch (\cite{ulv99}), %confirming its spurious nature. Final images, made with uniform weighting, had r.m.s. noises of 0.4--0.5~mJy~beam$^{-1}$, except for the second-epoch image of Mrk~348, where dynamic-range limitations caused an effective noise of 1.1~mJy~beam$^{-1}$. Two-component Gaussian models were fitted both in the ({\it u,v}\/) plane and in the image plane, using both DIFMAP and AIPS. Results from the different fitting procedures were generally consistent with each other; we use the image-plane fits here. Flux-density errors (all errors are $1\sigma$) are derived by combining a 5\% scale uncertainty in quadrature with the fitting error (which includes noise). Estimated size errors are 20\% in each axis, %spanning the range found by the different fitting programs, with point sources taken to have upper limits of half the beam size. %The resulting errors in brightness temperatures are about 29\%. Error estimates for relative component positions are generally derived from the image-plane fits; they are consistent with the ranges given by the other methods, and with expectations from the measured noises. For the first-epoch map of Mrk~231, the total range of $\pm 0.03$~mas found by the different procedures is greater than the nominal uncertainty of $\pm 0.009$~mas, so the larger error is used. \section{Results} \subsection{Mrk 348} Mrk 348 has a redshift $z=0.014$ relative to the microwave background (\cite{dev91}), yielding a scale of 0.31~pc~mas$^{-1}$ for $H_0=65$~km~s$^{-1}$~Mpc$^{-1}$ (used throughout). This galaxy is a type 2 Seyfert with a hidden broad-line region (\cite{mil90}), implying that its disk/torus is seen nearly edge-on. It also contains a 200-mas (60-pc) triple radio source (\cite{nef83}), which coincides with optical [O III] emission imaged by Capetti et al. (1996). Our VLBA images (Figure~\ref{fig:348-im}) show a small-scale double source that is well-aligned with the larger scale radio and optical emission; measured component properties are given in Table~\ref{tab:prop}. The relative component separation increased from 1.46~mas (0.46~pc) to 1.58~mas (0.50~pc) in 1.65~yr. Since the stronger component was resolved at the second epoch (see below), its centroid may have shifted. We estimate a $3\sigma$ upper limit for the shift that is equal to the component size at the second epoch, giving a total $1\sigma$ error of 57~$\mu$as for the position of the weaker component relative to a fiducial position in the stronger component. The proper motion is then $0.073\pm 0.035$~mas~yr$^{-1}$, corresponding to an apparent speed of $\beta_{\rm app} = 0.074\pm 0.035$. The epoch of zero separation at the current speed is $1977^{+7}_{-20}$, so the secondary could have had its genesis during a strong flux outburst in early 1982 (\cite{nef83}). Assuming an optically thin spectrum between 10~MHz and 100~GHz, and equal energies in electrons and protons, the synchrotron lifetime of the secondary component is $\sim 50$~yr, so no particle acceleration is required since the component ejection. The total flux density of $122\pm 6$~mJy in 1997.10 was slightly smaller than the total of $169\pm 9$~mJy measured at 1995.26 (\cite{bar98}), but Mrk~348 since has undergone a major radio flare. The southern component increased by a factor of 5.5 in 1.65~yr, giving a brightness temperature $T_b> 10^{11}$~K and strongly suggesting that it is the galaxy nucleus. The slight resolution at the second epoch, in a position angle (PA) similar to the component separation, indicates that a new component might have been ejected along the same axis. \subsection{Mrk 231} The redshift of Mrk~231 relative to the microwave background is $z=0.042$ (\cite{dev91}), and the corresponding scale is 0.93~pc~mas$^{-1}$. Mrk~231 is a Seyfert 1/starburst galaxy with a heavily obscured nucleus and a total infrared luminosity of $\sim 3\times 10^{12}L_\odot$ (\cite{soi89}). It contains a 40-pc north-south radio source (\cite{nef88}; \cite{ulv99}) embedded within a starburst several hundred parsecs in extent (\cite{bry96}; \cite{car98}). Our VLBA images (Figure~\ref{fig:231-im}) show a double source with a position angle differing by about 65\arcdeg\ from the larger scale source, as first reported by \cite{ulv99}. The component separation (see Table~\ref{tab:prop}) increased from 1.08~mas to 1.16~mas in 1.77~yr, for a measured proper motion of $0.046\pm 0.017$~mas~yr$^{-1}$, corresponding to $\beta_{\rm app}=0.139\pm 0.052$. The epoch of zero separation implied by the measured speed is $1973^{+7}_{-15}$. Variability by a factor of 2.5 between epochs implies that the weaker, eastern component is the actual nucleus of the galaxy. \section{Nature of the One-Sided Sources} Parsec-scale radio sources are known in several other Seyfert galaxies, and are typically associated with radio jets. In NGC~1068, the parsec-scale source is identified with the accretion torus (\cite{gal97}). However, the brightness temperatures in Mrk~348 and Mrk~231 are too high for thermal emission, and instead suggest association of the radio components with outflowing jets. Strong variability in one component in each galaxy indicates that it is close to the active nucleus, and that the double sources represent one-sided jets rather than straddling the nucleus. \subsection{Relativistic Beaming?} The jet/counterjet ratios are $R>17$ for Mrk~348 and $R>45$ for Mrk~231. One-sided structures often are associated with relativistic jets having speed $\beta c$ that are pointing nearly toward the observer, at an angle $\theta$ with respect to the line of sight. In that case, (cf. \cite{pea87}) \begin{equation} \beta_{\rm app}\ =\ {\beta\sin\theta\over 1 - \beta\cos\theta}\ . \end{equation} The ratio between the flux densities of approaching and receding components is given by \begin{equation} R = \bigl[(1+\beta\cos\theta)/(1-\beta\cos\theta)\bigr]^{3-\alpha}\ , \end{equation} where the flux density $S(\nu)$ at frequency $\nu$ is $S(\nu)=S(\nu_0)(\nu/\nu_0)^{+\alpha}$. In Mrk~348, if the jet components are optically thin, then $\alpha\approx 0.7$. The observed values of $R>17$ and $\beta_{\rm app}\approx 0.08$ can be caused by relativistic beaming only if $\beta\approx 0.55$ and $\theta \lesssim 5^\circ$. However, the Mrk~348 disk/torus is edge-on, and the half-angle of the ionization cone is $\sim$45\arcdeg\ (\cite{sim96}). If the radio jet were inside the ionization cone, this would imply $\theta > 45^\circ$, inconsistent with relativistic motion near our line of sight. In Mrk~231, the slightly higher observed speed (consistent with Mrk~348 within the errors) and higher upper limit on $R$ translate to $\beta\approx 0.69$ and $\theta \lesssim 8^\circ$, whereas the large-scale disk is $\sim$45\arcdeg\ from the line of sight (\cite{bry96}; Carilli et al.~1998). If the radio axis is perpendicular to that disk, then $\theta\approx 45^\circ$, again inconsistent with explaining the one-sided source due to relativistic motion at a small viewing angle. We conclude that relativistic boosting is not sufficient to account for the observed one-sided radio sources unless the underlying jet flow is much faster than the proper motions of the components, which might then be quasi-stationary shocks. Such an interpretation has been made for the VLBI jet in Centaurus~A, which has $\beta_{\rm app}\sim 0.1$, based on the internal evolution of radio components (\cite{tin98}). In Mrk~231 and Mrk~348, no such evidence exists at present, since the structures of the jet components are consistent between the two epochs, and the lifetimes do not require particle re-acceleration in shocks. \subsection{Free-Free Absorption} A likely explanation for the one-sided sources is that the ``missing'' components are in the receding jet and are free-free absorbed by ionized gas. The free-free optical depth at frequency $\nu$ is given by \begin{equation} \tau_{\rm ff}(\nu)\ \approx\ 8.2\times 10^{-2}\ T^{-1.35}\ (\nu/{\rm GHz})^{-2.1}\ E\ \end{equation} (\cite{ost89}), where $T$ is the temperature in Kelvin and $E$ is the emission measure in cm$^{-6}$~pc. Optical depths of $\tau_{\rm ff}(15\ {\rm GHz}) \gtrsim 4$ are required to account for the jet/counterjet ratio, implying $E \gtrsim 3\times10^9$~cm$^{-6}$~pc. For a line of sight distance of 0.1~pc and a temperature $T\sim 8000$~K, the average density of ionized material must be $n_e \gtrsim 2\times 10^5$~cm$^{-3}$ at 0.5--1.0~pc from the galaxy nuclei, about 100 times higher than that required 15--20~pc from the nucleus of Mrk~231 (\cite{ulv99}). This density is also a factor of 10 higher than is required for the free-free absorption $\sim 2$~pc from the nucleus of 3C~84 (\cite{lev95}). Thus, the current results are consistent with weak active galactic nuclei having disks or tori with $n_e\sim 10^5$~cm$^{-3}$ within a parsec of the nucleus, $n_e\sim 10^4$~cm$^{-3}$ at a few parsecs, and a typical narrow-line region density of $n_e\sim 10^{3}$~cm$^{-3}$ at 20~pc from the nucleus. Mrk~348 has an X-ray absorption column density of $10^{23}$~cm$^{-2}$ (\cite{smi96}), while that in Mrk~231 is $6\times 10^{22}$~cm$^{-2}$ (\cite{nak97}). For a 0.1-pc path length, these values correspond to average densities of $2-3\times 10^5$~cm$^{-3}$, remarkably consistent with the values required for counterjet absorption. The densities also are consistent with those found from the spectral turnover of the VLBI core in the H$_2$O-maser galaxy NGC~2639, assuming that the absorption takes place in a warm, weakly ionized medium (\cite{wil98}). Therefore, even though no H$_2$O maser emission has been detected in Mrk~231 or Mrk~348 (\cite{bra96}), the one-sided non-relativistic jets support the presence of a torus whose physical properties are similar to those in the megamaser galaxies. If $\tau_{\rm ff}(15\ {\rm GHz})\approx 4$, then the ionized medium should become optically thin near 30~GHz, and the counterjets might be detectable in very sensitive VLBI observations at 43~GHz. \subsection{Comparison to Other Parsec-Scale Sources} The low jet speeds in Mrk~231 and Mrk~348 are similar to those measured on parsec scales in weak Fanaroff-Riley-I (\cite{fan74}) radio galaxies such as Cen~A (\cite{tin98}) and 3C~84 (\cite{dha98}). The receding jet in 3C~84 is known to be free-free absorbed (e.g., Levinson et al. 1995), and the same may be true of the Cen~A counterjet. Since the Seyfert jets are non-relativistic when they emerge from the broad-line region, their initial speeds can be highly relativistic only if they are slowed by interactions with the gas in that region. Physics directly related to the central black holes, such as low spin rates (e.g., \cite{ree82}; \cite{wil95}), is another promising explanation for the slow speeds of the Seyfert jets. Both Mrk~348 and Mrk~231, as well as NGC~2639 (\cite{wil98}), have spectra that peak at frequencies of 10--20~GHz. The spectral turnovers are consistent with either synchrotron self-absorption or free-free absorption in their cores. The evidence presented in this {\it Letter} strengthens the case for free-free absorption in the component at the nucleus, since the gas properties required for the spectral turnovers of the cores are similar to the values measured by X-ray absorption and deduced for the obscuration of the counterjets. The Seyfert radio sources have general characteristics similar to the compact symmetric objects (CSOs), which display gigahertz-peaked spectra, symmetric radio sources on scales of $\sim 100$~pc, and one-sided jets on scales of 10~pc (\cite{tay96}). In addition, the outer components of CSOs have speeds of $\sim 0.15c$ (for $H_0=65$) relative to their nuclei (\cite{ows98a}; \cite{ows98b}). The outer VLBI lobes of Mrk~231 and Mrk~348 are $\sim 20$~pc from their nuclei, corresponding to ages of $\sim 700$~yr if they have steady advance speeds of $\sim 0.1c$. These ages can be compared to typical ages of $\sim 10^3$--$10^4$~yr inferred for CSOs (\cite{rea96}; \cite{ows98a}; Owsianik et al. 1998). Hence the Seyfert galaxies seem to be somewhat smaller and much less powerful versions of the CSOs. The relation between Seyfert and CSO jets is of considerable interest. The CSO jets are one-sided at 15~GHz (\cite{tay96}), and even at 43~GHz (Taylor, private communication), on scales of 10~pc. At least one CSO shows patchy H{\sc I} absorption with a column density of $2\times 10^{23}$~cm$^{-2}$, on scales of tens of parsecs, and also may show low-frequency free-free absorption of the receding jet (\cite{pec99}). If the high-frequency one-sidedness in other CSOs were due to free-free absorption, as we infer for the Seyferts, then $\tau_{\rm ff}(43\ {\rm GHz})\geq 2$, implying an average ionized density $n_e\geq 5\times 10^{4}$~cm$^{-3}$ for a path length of 5~pc; the corresponding column density would be $\sim 8\times 10^{23}$~cm$^{-2}$. If free-free absorption is the cause of the one-sidedness in CSOs, this indicates that the tori/disks in the CSOs may be somewhat larger and more dense than in Seyferts. X-ray spectral studies at moderately high energies could be used to search for the absorption due to this gas. \acknowledgments We thank Greg Taylor, Alison Peck, and Jack Gallimore for useful discussions, and Daria Halkides for assistance with the data reduction for the first epoch of Mrk~348. This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. \clearpage \begin{thebibliography}{} %\bibitem[Armus et al.\ 1994]{arm94} Armus, L., Surace, J. A., Soifer, %B. T., Matthews, K., Graham, J. R., Larkin, J. E. 1994, \aj, 108, 76 \bibitem[Barvainis \& Lonsdale 1998]{bar98} Barvainis, R., \& Lonsdale, C. 1998, \aj, 115, 885 %\bibitem[Baum, Zirbel, \& O'Dea 1995]{bau95} Baum, S. A., Zirbel, E. L., %\& O'Dea, C. P. 1995, \apj, 451, 88 % %\bibitem[Bicknell et al. 1998]{bic98} Bicknell, G. V., %Dopita, M. A., Tsvetanov, Z. I., \& Sutherland, R. S. 1998, %\apj, 495, 680 % % %\bibitem[Boksenberg et al.\ 1977]{bok77} Boksenberg, A., Carswell, %R. F., Allen, D. A., Fosbury, R. A. E., Penston, M. V., \& Sargent, %W. L. W. 1977, \mnras, 178, 451 %\bibitem[Bonatto \& Pastoriza 1997]{bon97} Bonatto, C. J., \& %Pastoriza, M. G. 1997, \apj, 486, 132 % \bibitem[Braatz, Wilson, \& Henkel 1996]{bra96} Braatz, J. A., Wilson, A. S., \& Henkel, C. 1996, \apjs, 106, 51 \bibitem[Bryant \& Scoville 1996]{bry96} Bryant, P. M., \& Scoville, N. Z. 1996, \apj, 457, 678 %\bibitem[Calvani, Marziani, \& Sulentic 1997]{cal97} Calvani, M., %Marziani, P., \& Sulentic, J. 1997, Mem. S.A.It., 68, 93 % \bibitem[Capetti et al. 1996]{cap96} Capetti, A., Axon, D. J., Macchetto, F., Sparks, W. B., \& Boksenberg, A. 1996, \apj, 469, 554 \bibitem[Carilli, Wrobel, \& Ulvestad 1998]{car98} Carilli, C. L., Wrobel, J. M., \& Ulvestad, J. S. 1998, \aj, 115, 928 \bibitem[de Vaucouleurs et al.\ 1991]{dev91} de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H. G., Jr., Buta, R. J., Paturel, G., \& Fouqu\'e, P. 1991, Third Reference Catalogue of Bright Galaxies (New York: Springer) \bibitem[Dhawan, Kellermann, \& Romney 1998]{dha98} Dhawan, V., Kellermann, K. I., \& Romney, J. D. 1998, \apjl, 498, L111 %\bibitem[Falcke \& Biermann 1995]{fal95} Falcke, H., \& %Biermann, P. L. 1995, \aap, 293, 665 % \bibitem[Falcke, Wilson, \& Simpson 1998]{fal98} Falcke, H., Wilson, A. S., \& Simpson, S. 1998, \apj, 502, 199 \bibitem[Fanaroff and Riley 1974]{fan74} Fanaroff, B. L., \& Riley, J. M. 1974, \mnras, 167, 31P \bibitem[Gallimore, Baum, \& O'Dea 1997]{gal97} Gallimore, J. F., Baum, S. A., \& O'Dea, C. P. 1997, \nat, 388, 852 \bibitem[Halkides, Ulvestad, \& Roy 1997]{hal97} Halkides, D. J., Ulvestad, J. S., \& Roy, A. L. 1997, \baas, 29, 1375 \bibitem[Herrnstein et al.\ 1997]{her97} Herrnstein, J. R., Moran, J. M., Greenhill, L. J., Diamond, P. J., Miyoshi, M., Nakai, N., \& Inoue, M. 1997, \apjl, 475, L17 %\bibitem[Hutchings \& Neff 1987]{hut87} Hutchings, J. B., \& Neff, %S. G.~1987, \aj, 93, 14 % %\bibitem[Kukula et al.\ 1995]{kuk95} Kukula, M. J., Pedlar, A., Baum, %S. A., \& O'Dea, C. P. 1995, \mnras, 276, 1262 % \bibitem[Levinson, Laor, \& Vermeulen 1995]{lev95} Levinson, A., Laor, A., \& Vermeulen, R. C. 1995, \apj, 448, 589 %\bibitem[McCutcheon \& Gregory 1978]{mcc78} McCutcheon, W. H., \& %Gregory, P. C. 1978, \aj, 83, 566 % %\bibitem[Meier 1999]{mei99} Meier, D. L. 1999, \apj, submitted % %\bibitem[Meier et al. 1997]{mei97} Meier, D. L., Edgington, S., Dodon, P., %Payne, D. G., \& Lind, K. R. 1997, \nat, 388, 350 \bibitem[Miller \& Goodrich 1990]{mil90} Miller, J. S., \& Goodrich, R. W. 1990, \apj, 355, 456 \bibitem[Nakagawa et al. 1997]{nak97} Nakagawa, T., Kii, T., Fujimoto, R., Miyazaki, T., Inoue, H., Ogasaka, Y., Arnaud, K., \& Kawabe, R. 1997, IAU Symposium 186, Galaxy Interactions at Low and High Redshift, 103 \bibitem[Napier et al.\ 1994]{nap94} Napier, P. J., Bagri, D. S., Clark, B. G., Rogers, A. E. E., Romney, J. D., Thompson, A. R., \& Walker, R. C. 1994, Proc IEEE, 82, 658 \bibitem[Neff \& de Bruyn 1983]{nef83} Neff, S. G., \& de Bruyn, A. G. 1983, \aap, 128, 318 \bibitem[Neff \& Ulvestad 1988]{nef88} Neff, S. G., \& Ulvestad, J. S. 1988, \aj, 96, 841 \bibitem[Osterbrock 1989]{ost89} Osterbrock, D. E. 1989, Astrophysics of Gaseous Nebulae and Active Galactic Nuclei (Mill Valley: University Science Books) \bibitem[Owsianik \& Conway 1998]{ows98a} Owsianik, I. \& Conway, J. E. 1998, \aap, 337, 690 \bibitem[Owsianik, Conway, \& Polatidis 1998]{ows98b} Owsianik, I., Conway, J. E., \& Polatidis, A. G. 1998, \aap, 336, L370 %\bibitem[Pacholczyk 1970]{pac70} Pacholczyk, A. G. 1970, Radio %Astrophysics (San Francisco: W. H. Freeman) % %\bibitem[Patnaik et al. 1992]{pat92} Patnaik, A. R., Browne, I. W. A., %Wilkinson, P. N., \& Wrobel, J. M. 1992, \mnras, 254, 655 % \bibitem[Pearson \& Zensus 1987]{pea87} Pearson, T. J., \& Zensus, J. A. 1987, Superluminal Radio Sources, eds. J. A. Zensus \& T. J. Pearson (Cambridge: Cambridge University Press), 1 \bibitem[Peck \& Taylor 1998]{pec98} Peck, A. B., \& Taylor, G. B. 1998, \apjl, 502, L23 \bibitem[Peck, Taylor, \& Conway 1999]{pec99} Peck, A. B., Taylor, G. B., \& Conway, J. E. 1999, \apj, submitted %\bibitem[Pringle 1996]{pri96} Pringle, J. E. 1996, \mnras, 281, 357 % %\bibitem[Pringle 1997]{pri97} Pringle, J. E. 1997, \mnras, 292, 136 % \bibitem[Readhead et al.\ 1996]{rea96} Readhead, A. C. S., Taylor, G. B., Xu, W., Pearson, T. J., Wilkinson, P. N., \& Polatidis, A. G. 1996, \apj, 460, 612 \bibitem[Rees et al.\ 1982]{ree82} Rees, M. J., Begelman, M. C., Blandford, R. D., \& Phinney, E. S. 1982, \nat, 295, 17 \bibitem[Roy et al.\ 1998]{roy98} Roy, A. L., Colbert, E. J. M., Wilson, A. S., \& Ulvestad, J. S. 1998, \apj, 504, 147 \bibitem[Shepherd 1997]{she97} Shepherd, M. C. 1997, Astronomical Data Analysis Software and Systems VI, eds. G. Hunt \& H. E. Payne (San Francisco: ASP), 77 \bibitem[Simpson et al.\ 1996]{sim96} Simpson, C., Mulchaey, J. S., Wilson, A. S., Ward, M. J., \& Alonso-Herrero, A. 1996, \apjl, 457, L19 \bibitem[Smith \& Done 1996]{smi96} Smith, D. A., \& Done, C. 1996, \mnras, 280, 355 \bibitem[Soifer et al.\ 1989]{soi89} Soifer, B. T., Boehmer, L., Neugebauer, G., \& Sanders, D. B. 1989, \aj, 98, 766 %\bibitem[Surace et al.\ 1998]{sur98} Surace, J. A., Sanders, D. B., %Vacca, W. D., Veilleux, S., \& Mazzarella, J. M. 1998, \apj, 492, 116 % %\bibitem[Taylor et al.\ 1994]{tay94} Taylor, G. B., Vermeulen, R. C., %Pearson, T. J., Readhead, A. C. S., Henstock, D. R., Browne, I. W. A., %\& Wilkinson, P. N. 1994, \apjs, 95, 345 % \bibitem[Taylor, Readhead, \& Pearson 1996]{tay96} Taylor, G. B., Readhead, A. C. S., \& Pearson, T. J. 1996, \apj, 463, 95 \bibitem[Tingay et al.\ 1998]{tin98} Tingay, S. J., et al.\ 1998, \aj, 115, 960 \bibitem[Ulvestad, Neff, \& Wilson 1987]{ulv87} Ulvestad, J. S., Neff, S. G., \& Wilson, A. S. 1987, \aj, 93, 22 \bibitem[Ulvestad et al.\ 1998]{ulv98} Ulvestad, J. S., Roy, A. L., Colbert, E. J. M., \& Wilson, A. S. 1998a, \apj, 496, 196 \bibitem[Ulvestad \& Wilson 1989]{ulv89} Ulvestad, J. S., \& Wilson, A. S. 1989, \apj, 343, 659 \bibitem[UWC]{ulv99} Ulvestad, J. S., Wrobel, J. M., \& Carilli, C. L. 1999, \apj, 516, in press (UWC) \bibitem[van Moorsel, Kemball, \& Greisen 1996]{van96} van Moorsel, G., Kemball, A., \& Greisen, E.\ 1996, Astronomical Data Analysis Software and Systems V, eds. G. H. Jacoby \& J. Barnes (San Francisco: ASP), 37 \bibitem[Wilson \& Colbert 1995]{wil95} Wilson, A. S., \& Colbert, E. J. M. 1995, \apj, 438, 62 \bibitem[Wilson et al. 1998]{wil98} Wilson, A. S., et al. 1998, \apj, 505, 587 \end{thebibliography} \clearpage %\figcaption{Two-epoch 15-GHz images of Mrk~348. %\label{fig:348-im}} % %\figcaption{Two-epoch 15-GHz images of Mrk~231. %\label{fig:231-im}} % %\clearpage %\begin{deluxetable}{lcc} %\tablecolumns{3} %\tablewidth{0pc} %\tablecaption{Log of 15-GHz VLBA Observations} %\tablehead{ %\colhead{Parameter}& %\colhead{Epoch 1}& %\colhead{Epoch 2} } %\startdata %\multicolumn{3}{c}{\underbar{Mrk 348}} \\ %Date&1997FEB05&1998SEP30 \\ %Phase-ref. source&None&J0112+35 \\ %Integration (min.)&110& 72 \\ %Flagged Antennas\tablenotemark{a}&NL,HN,SC&PT \\ %Beam Size (mas)&$0.88\times 0.49$, PA $-30^\circ$& %$0.69\times 0.42$, PA $-2^\circ$ \\ %\multicolumn{3}{c}{\underbar{Mrk 231}} \\ %Date&1996DEC08\tablenotemark{b}&1998SEP15 \\ %Phase-ref. source&None&J1302+57 \\ %Integration (min)&156 &75 \\ %Flagged Antennas\tablenotemark{a}&BR,HN&Some PT,HN \\ %Beam Size&$0.57\times 0.42$, PA $0^\circ$& %$0.54\times 0.42$, PA $-2^\circ$ \\ %\enddata %\tablenotetext{a}{Flagged antennas, in most case due to winter %weather, are denoted as follows: BR=Brewster, Washington; %PT=Pie Town, New Mexico; NL=North Liberty, Iowa; %HN=Hancock, New Hampshire; SC=Saint Croix, Virgin Islands} %\tablenotetext{b}{The first observation of Mrk~231 used a single %25-m antenna at the Very Large Array in addition to the VLBA.} %\label{tab:obs} %\end{deluxetable} %\clearpage \begin{deluxetable}{cccccc} \tablecolumns{6} \tablewidth{0pc} \tablecaption{Results of 2-Component Gaussian Fits} \tablehead{ \colhead{No.}&\colhead{Flux Density}&\colhead{Size}&\colhead{$T_b$\tablenotemark{a}}& \colhead{Offset}&\colhead{PA} \\ &\colhead{(mJy)}&\colhead{(mas$\times$mas, deg)}&\colhead{($10^9$K)}& \colhead{(mas)}&\colhead{(deg)} } \startdata \multicolumn{6}{c}{\underbar{Mrk 348, 1997.10}} \\ 1 (S)&$96\pm 5$&Unresolved&$>6.7\pm 1.9$&0.000&\nodata \\ 2 (N)&$26\pm 2$&Unresolved&$>1.8\pm 0.5$&$1.460\pm 0.009$&$-16\pm 1$ \\ \multicolumn{6}{c}{\underbar{Mrk 348, 1998.75}} \\ 1 (S)&$552\pm 28$&$0.16\times 0.11$, PA $-9\pm 4$&$238\pm 68$&0.000&\nodata \\ 2 (N)&$17\pm 1$&Unresolved&$>1.8\pm 0.5$&$1.581\pm 0.021$&$-15\pm 2$ \\ \multicolumn{6}{c}{\underbar{Mrk 231, 1996.94}} \\ 1 (E)&$17\pm 1$&Unresolved&$>2.2\pm 0.6$&$1.081\pm 0.030$&$65\pm 1$ \\ 2 (W)&$51\pm 3$&$0.32\times 0.18$, PA $-74\pm 4$&$6.9\pm 2.0$&0.000&\nodata \\ \multicolumn{6}{c}{\underbar{Mrk 231, 1998.71}} \\ 1 (E)&$44\pm 3$&Unresolved&$>6.0\pm 1.7$&$1.162\pm 0.004$&$68\pm 1$ \\ 2 (W)&$60\pm 3$&$0.30\times 0.13$, PA $-88\pm 4$&$12.0\pm 3.4$&0.000&\nodata \\ \enddata \tablenotetext{a}{All brightness temperatures are expressed in the source rest frames, by multiplying the observed brightness temperature by $(1+z)$.} \label{tab:prop} \end{deluxetable} \clearpage \begin{figure}[htp] \vspace{14cm} %\special{psfile=/home/arana/julvesta/astronomy/aaslatex/submit/m348/U348OLD-RS.PS %hoffset=0 voffset=-100 hscale=75.0 vscale=75.0} %\special{psfile=/home/arana/julvesta/astronomy/aaslatex/submit/m348/U348NEW-RS.PS %hoffset=0 voffset=-100 hscale=75.0 vscale=75.0} \special{psfile=/home/arana/julvesta/astronomy/aaslatex/submit/m348/U348-JOINT.PS hoffset=0 voffset=-100 hscale=75.0 vscale=75.0} \caption{15-GHz VLBA images of Mrk 348 at epochs 1997.10 and 1998.75. The images have been aligned at the southern radio component, rotated by 15\arcdeg, then offset from each other horizontally. North and East are labeled, as are the two radio components. Both images are contoured logarithmically, with contours starting at 4~mJy~beam$^{-1}$ and increasing by factors of 2 to 512~mJy~beam$^{-1}$. The restoring beam at 1997.10 is $0.88\times 0.49$~mas in PA $-30^\circ$, while that at 1998.75 is $0.69\times 0.42$~mas in PA $-2^\circ$.} \label{fig:348-im} \end{figure} \clearpage \begin{figure}[htp] \vspace{14cm} \special{psfile=/home/arana/julvesta/astronomy/aaslatex/submit/m348/U231-JOINT.PS hoffset=0 voffset=-100 hscale=75.0 vscale=75.0} %\special{psfile=/home/arana/julvesta/astronomy/aaslatex/submit/m348/U231OLD-SRS.PS %hoffset=0 voffset=-100 hscale=75.0 vscale=75.0} %\special{psfile=/home/arana/julvesta/astronomy/aaslatex/submit/m348/U231NEW-SRS.PS %hoffset=0 voffset=-100 hscale=75.0 vscale=75.0} \caption{15-GHz VLBA images of Mrk 231 at epochs 1996.94 and 1998.71. The images have been aligned at the eastern radio component, rotated by 25\arcdeg, then offset from each other vertically. North and East are labeled, as are the two radio components. Both images are contoured logarithmically, with contours starting at 2~mJy~beam$^{-1}$ and increasing by factors of 2 to 32~mJy~beam$^{-1}$. The restoring beam at 1996.94 is $0.57\times 0.42$~mas in PA $0^\circ$, while that at 1998.71 is $0.54\times 0.42$~mas in PA $-2^\circ$.} \label{fig:231-im} \end{figure} \clearpage \end{document}