From ZADEH@OSSENU.ASTRO.NWU.EDU Fri May 24 14:32:46 1996 From: ZADEH@OSSENU.ASTRO.NWU.EDU Date: Fri, 24 May 1996 13:32:40 -0500 (CDT) To: gcnews@astro.umd.edu Subject: eso paper \documentstyle[11pt,paspconf,epsf]{article} \nofiles \def\etal{{\it et al.}} \def\kms{km s$^{-1}$} % Some definitions I use in these instructions. \def\lsol{\hbox{$\hbox{L}_\odot$} } \def\emphasize#1{{\sl#1\/}} \def\arg#1{{\it#1\/}} \let\prog=\arg \def\edcomment#1{\iffalse\marginpar{\raggedright\sl#1\/}\else\relax\fi} \marginparwidth 1.25in \marginparsep .125in \marginparpush .25in \reversemarginpar \begin{document} \title{The 1720 MHz OH Emission from G359.1--0.5 and the CND} \author{F. Yusef-Zadeh and B.T. Robinson} \affil{Northwestern University, Dept Physics and Astronomy, Evanston, IL 60208, US} \author{D.A. Roberts} \affil{NCSA, 405 N. Mathews Ave, Urbana, IL 61801} \author{W.M. Goss and D.A. Frail} \affil{NRAO, P.O. Box 0, Socorro, NM 87801} \author{A. Green} \affil{University of Sydney, School of Physics, Sydney, NSW 2006} \begin{abstract} VLA observations of the CND, which surrounds Sgr A West, and G359.1--0.5 have been carried out at the 1720 MHz transition of the OH molecule. 1720 MHz OH maser emission has been detected from the CND and G359.1-0.5 at velocities near V$_{LSR}$ = +132 and --5 \kms, respectively. The OH maser features toward G359.1--0.5 have been detected along the interface between a large-scale non-thermal continuum shell G359.1--0.5 and its surrounding ring of high velocity molecular CO gas. The V spectra of a number of bright maser sources associated with the CND and G359.1--0.5 suggest strong magnetic fields between 0.4 and 4 mG. We argue that these masers are produced at the boundaries of G359.1--0.5 and the CND and that the maser features signify regions of shocked gas resulting from the supernova expansion and from cloud-cloud collisions at the Galactic center. \end{abstract} \section{Introduction} A new and powerful probe to find evidence of shock activity was recently discussed by Frail, Goss \& Slysh (1994). They observed numerous distinct 1720 MHz OH maser spots along the interface between the supernova remnant W28 and an adjacent molecular cloud. They suggest that the masers are being pumped collisionally behind the shock where H$_2$ molecules with densities and temperatures limited to 10$^3-10^5$ cm$^{-3}$ and 25-200 $^\circ$K cause population inversion in the OH molecules (Elitzur 1976). More recent cross section calculations (Offer \etal\ 1994; Lockett 1995) show a significant difference between para and ortho-H$_2$ rates, suggesting that the ortho/para H$_2$ ratio, as well as the temperature and density of H$_2$, can be important in collisional pumping of the 1720 MHz OH transition. Motivated by the 1720 MHz maser observations, a search was conducted to find examples such as the W28 SNR and its associated molecular complex in the Galactic center region. In this paper, we review recent detections of shocked maser emission from G359.1--0.5 and the circumnuclear disk (CND) in Sgr A West surrounding the inner 1$'$ of the Galactic center (Yusef-Zadeh, Uchida and Roberts 1995; Yusef-Zadeh \etal\ 1996). We argue that the expansion of a supernova into an adjacent molecular cloud and cloud-cloud collisions are responsible for producing shocked OH maser emission. In the accompanying paper by Green \etal, the results of maser activity associated with Sgr A East are described (see also Yusef-Zadeh \etal\ 1996). \subsection{G359.1--0.5} The shell-type SNR G359.1--0.5 was first discovered as a complete shell of polarized emission at $\lambda$6cm with a spectral index of $\approx$--0.4 (Reich \& F\"urst 1984). A $^{12}$CO survey of the Galactic center has revealed a nearly continuous ring of molecular gas concentric with this prominent non-thermal radio continuum source %complete shell makes it similar to Tycho's remnant (Uchida \etal\ 1992a). The cloud has a radius of about 12$'$, a radial velocity which ranges between $-$60 and $-$190 \kms\, and is characterized as a Galactic center cloud because of its large linewidth. This morphological correlation of the non-thermal and the CO shells (Uchida \etal\ 1992a) as well as an HI absorption line study (Uchida \etal\ 1992b) suggest that both the non-thermal continuum and CO shells are physically associated with each other and that they both lie near the Galactic center. % However, an alternative view is %described below based on the maser study. \begin{figure} \plotfiddle{cnd_fig1.ps}{6in}{0}{90}{90}{-290}{-100} \caption{The black spots show the positions of the 1720 MHz OH masers A, B, C1 and C2 integrated in velocity between $-$7.7 and $-$3.7 \kms at a spatial resolution of $21''\times12''$. They are superimposed on the gray scale $\lambda$20cm continuum image of the western half of G359.1--0.5 SNR shell at a resolution of 33$''\times31''$.} \end{figure} The detection of a number of compact 1720 MHz OH maser features along the interface between the SNR shell G359.1--0.5 and its surrounding ring of high-velocity molecular gas was recently reported by Yusef-Zadeh, Uchida \& Roberts (1995). An extended and weakly emitting OH feature was also noted along the brightest side of the non-thermal shell. The extended feature is considered to be a weak maser because of its small linewidth and its similarity to the velocity structure of the bright and compact maser sources. The morphological correlation between the neutral gas, the non-thermal shell and the maser features provide strong support for the hypothesis that the 1720 MHz maser line of OH arises from shocked gas by the impact of the expanding SNR into the molecular material. The boundary of the SNR and molecular clouds should be a natural place for population inversion because of the energy flow between the two components of the ISM with very different temperatures. Follow-up VLA observations with a spatial and spectral resolution of 20.7$''\times12''$ and 0.27 \kms\ were carried out in early 1996 in the 1720 MHz line and in the main lines of the OH molecule at 1665 and 1667 MHz. No main line emission from G359.1--0.5 was detected above $5\sigma=75$ mJy level, as theoretical considerations had predicted. In the collisional pump model, only the maser amplification of the 1720 MHz transition of OH is expected to be observed (Elitzur 1976). Figure 1 shows the maser spots based on these new observations superimposed on the grayscale continuum image of the western edge of the G359.1--0.5 shell. The brightest of the OH masers appears at the location where a non-thermal filament crosses the western edge of the SNR shell. This filament is known as the ``Snake'' and its morphology is distinguished somewhat from the other Galactic center filaments by its long (20$'$) and narrow (10$''$) extent, and by two uncharacteristic kinks along its length (Gray \etal\ 1991). If there is an interaction between the non-thermal filament and the non-thermal continuum shell, the ``Snake'' may be responsible for an enhanced shock activity at the location where the bright maser A is noted in Figure 1. We note, however, that Zeeman measurements of the V spectrum of this maser are not unusual as they show similar polarization characteristics to other maser components in G359.1--0.5. Estimates of the line of sight component of the magnetic field based on Zeeman measurements range between 0.4 and 0.6 mG (Robinson \etal\ 1996). The extended 1720 MHz OH feature that had been detected in the low-resolution (71$''\times33''$) observations (Yusef-Zadeh \etal\ 1995) was resolved out in the new high-resolution observations as shown in Figure 1. This weak and extended maser may be the site where population inversion is achieved at low OH densities by either radiative or collisional pumps (Elitzur 1976). Similar extended and weak OH features at 1720 MHz have been found at low galactic latitudes based on single-dish surveys of the Galactic plane (Haynes and Caswell 1977; Turner 1982). These extended maser features are argued to be associated with giant molecular clouds confined to the spiral arms of the Galaxy and their excitation is considered to be due to collisional pumping at low temperatures resulting in an enhanced T$_{ex}$ at 1720 MHz (Turner 1982). \begin{figure} %\plotone{cnd_fig2_new.ps} \plotfiddle{cnd_fig2.ps}{2.8in}{90}{55}{55}{210}{-70} %\plotfiddle{cnd_fig1.ps}{3.2in}{90}{55}{55}{210}{-20} \caption{The CS(2--1) spectrum toward the 1720 MHz OH maser source B of G359.1-0.5 showing a peak near -6 \kms. This peak velocity is close to the velocity of the OH masers toward G359.1-0.5. This spectrum is based on observations made with the 12m Kitt Peak NRAO telescope.} \end{figure} A strong argument in favor of the physical association of 1720 MHz OH masers and the surrounding molecular cloud near SNR is the similarity of the velocity of the maser and the systemic velocity of nearby thermal gas (Frail \etal\ 1994; Frail \etal\ 1996). However, in the case of G359.1-0.5, a puzzling aspect of the nature of the association between the masers, the molecular CO, and the non-thermal shell is the strong discrepancy that is noted between the ring of molecular gas with velocities in the range between --60 and --90 \kms\ and the velocity of 1720 MHz masers near --5 \kms. The possibility that all three features of G359.1-0.5 are associated with each other and lie near the Galactic center has been discussed by Yusef-Zadeh \etal\ (1995). Here, we explore the possibilities that these rare masers arise in low-velocity foreground gas which happens to be inverted. In this picture, the source of population inversion is unclear. %responsible for inverting the strong 1720 MHz line of OH molecule % which is thought to be collisionally pumped. Furthermore, it is difficult to imagine that a ring of molecular cloud lying in the foreground matches exactly the size of the supernova shell; there is no evidence for maser emission toward other Galactic center continuum sources that lie in the field centered on G359.1--0.5. Finally, the detection of CS (2--1) emission at low negative velocities near the maser positions %argues in favor of the gas cloud to be is evidence that the gas cloud can be associated with the SNR and with the population of Galactic center molecular clouds. Figure 2 shows the CS spectrum at V$_{LSR}=-6$ \kms\ centered on one of the bright masers (position B of Yusef-Zadeh, Uchida \& Roberts 1995). This spectrum, which is based on recent observation of G359.1-0.5 made with the NRAO 12m telescope (Robinson \etal\ 1996), shows clearly two velocity components with similar peak antenna temperature. The --6 \kms\ velocity feature of the CS gas toward G359.1-0.5 has a similar velocity to the radial velocity of the shocked OH masers. Alternatively, the alignment of the CO ring and the SNR is a coincidence. %the high-velocity CO ring is only coincidentally %superimposed on the non-thermal SNR shell. In this picture, it is conceivable that the molecular ring is not associated with the shell but still lies near the Galactic center because of its large linewidth. One argument against this model is that the largest concentration of the CO ring is concentrated on the brightest side of the SNR. This side of G359.1--0.5 faces the Galactic plane and most of the masers, including the extended masing feature, are found here. This morphology suggests that the CO ring and the shell may be interacting. \subsection{The Circumnuclear Disk} Near IR observations of shocked gas in the CND surrounding Sgr A West have been carried out for more than a decade by imaging the distribution of H$_2$ emitting gas (Gatley \etal\ 1984; Pak \etal\ 1996). The ratio of the V=2--1 and 1--0 S(1) lines of H$_2$ have been used to probe the shocked region. However, because of the dense environment of the Galactic center surrounded by an intense UV radiation field, it has not been possible to distinguish between shocked and UV heated gas (Gatley \etal\ 1984; Burton \& Allen 1992; Pak \etal\ 1996). In addition, because of the large linewidths of molecular and atomic clouds near the Galactic center, the broad CO lines accompanying shocked sites (e.g. HH objects, W28) are not good diagnostics of shocked gas in this region. 1720 MHz OH maser emission from Sgr A West has been detected at the velocity of +134 \kms. Figure 3 shows the position of the 1720 MHz line emission drawn as a circle. The maser source observed at a resolution of $\approx15''$ is superimposed on a gray-scale radio continuum image of Sgr A West at $\lambda$6cm with a resolution of 3.8$''\times3.2''$ (PA=$-$65$^\circ$). The high-resolution radio continuum image shows the peak +132 \kms\ maser line feature arising from a region of weakly emitting ionized gas sandwiched between the Eastern and Northern Arms of Sgr A West. In fact, a radio continuum radiograph of this region, as seen in Figure 4 of Yusef-Zadeh \& Morris (1987), reveals a limb-brightened hole in the distribution of continuum emission at the location of the northwest Streamers. This morphology is consistent with the presence of neutral gas surrounded by ionized gas (Yusef-Zadeh, Zhao \& Goss 1994). An HI absorption study of this region confirms the presence of the 130 \kms\ HI gas at the location of the peak maser emission (Plante, Lo \& Crutcher 1995). The position of the HI Zeeman splitting measurements at +130 \kms\ is shown as a triangle. Remarkably, the strength and the orientation of the magnetic field based on Zeeman measurements of HI gas and the 1720 MHz OH masers agree (see Plante \etal\ 1995; Yusef-Zadeh \etal\ 1996). However, if we apply the new method describing the polarization of masers (Elitzur 1996a), the line of sight magnetic field estimates will be lower by a factor of about five; Elitzur (1996b) suggests that the 1720 MHz OH masers in the Galactic center are saturated. \begin{figure} %\plotone{cnd_fig2.ps} \plotfiddle{cnd_fig3.ps}{6in}{-90}{80}{80}{-310}{450} %\plotfiddle{cnd_fig2.eps}{6in}{0}{80}{80}{-280}{-70} \caption{A $\lambda$6cm continuum image of Sgr A West with a spatial resolution of 3.7$''\times3.2''$ showing the location of the 135 \kms\ OH(1720 MHz) maser feature as a circle. The position of the HI Zeeman splitting measurements at +130 \kms\ by Plante \etal\ (1995) is also shown as a triangle.} \end{figure} Figure 4 displays contours of the integrated 1720 MHz emission feature based on low-resolution data integrated between +128 and +138 \kms which are superposed on the distribution of HCN (1--0) and [OI] line emission from the CND taken from Jackson \etal\ (1993). The maser feature is located where there is a gap in the distribution of the HCN emission molecular gas (dark-shaded areas) as traced by the HCN emission. The CND is known to orbit the Galactic center with a circular velocity of about 110 \kms; at this location the kinematics of the HCN and [OI] gas deviate from circular geometry (Jackson et al. 1993). H110$\alpha$ observations also indicate a feature at a velocity up to +144 \kms\ near this location. The kinematics of the ionized gas at the region of the gap are also known to be inconsistent with circular motion of ionized gas around the Galactic center (Yusef-Zadeh, Zhao \& Goss 1994). %More recent high-resolution observations of OH masers in the CND region %reveal that the spectrum of the peak 135 \kms\ feature consists of at %least three velocity components and that the spatially extended %feature is resolved out (Yusef-Zadeh \etal\ 1996). We believe that %the extended nature of the maser emission from the CND is due to an %enhanced T$_{ex}$, producing weak masers. %Morphological and kinematic comparisons %between thermal molecular gas at the gap of %the CND and the 1720 MHz OH maser gas %suggest that these features are related to each other. Thus, we consider %that the maser feature is produced as %a result of an interaction between the gas in the CND and a distinct % peculiar-moving cloud in the Galactic center. % Observational evidence for the existence of %peculiar-moving ionized and neutral %gas at the Galactic center raises the question of whether %this peculiar cloud is infalling toward or outflowing from %the Galactic center. A number of arguments have been made to account for %both possibilities, as schematically drawn in Figure 5, but the %direction of the gas flow is still ambiguous. We hope that the new maser %sources described above can be used as a probe of gas flows by %making high-resolution proper motion measurements in the future. \begin{figure} %\plotone{cnd_fig3.ps} \plotfiddle{cnd_fig4.ps}{6in}{0}{65}{65}{-210}{10} \caption{ Black, thin, solid contours of 1720 MHz OH emission integrated between +128 and +138 \kms\ are displayed at levels (0.75, 1, 1.5, 2, 3, 4, 5, 7, 9, 12, 15, 20)$\times$0.1 Jy \kms\ and are superimposed on the [OI] (broken contours) and HCN (dark and shaded areas) distribution of the CND (Jackson \etal\ 1992). The peak maser emission coincides with a gap in the distribution of the highest density HCN gas displayed as dark areas.} \end{figure} %The region where the maser spot at V$_{LSR}$=+132 \kms\ is detected %is unusual because of its peculiar velocity and its position in the %CND, as described earlier. A ``tongue'' of molecular gas at the location where the 135 \kms\ OH(1720) maser is located is distinguished from the gas in the CND by its magnetic field properties. %HI observations reveal Zeeman splitting of %the absorption line at +130 \kms\ line and The line of sight magnetic field based on HI measurements is estimated to be about --2 mG at the location of the 1720 MHz OH maser (Plante \etal\ 1995). The far-IR polarization observations of this region are also noted for the unusual geometry of the magnetic fields (Hildebrand \& Davidson 1994), showing a distribution which is quite similar to mid-IR polarization measurements along the Northern arm (Aitken \etal\ 1991). This geometry of the field differs from that predicted by an axially symmetric model of the CND which is dominated by circular motion (Hildebrand \& Davidson 1994), but argues for the coupling of the ionized gas of the Northern arm to the tongue of molecular gas observed in the gap of the CND. The kinematics as well as morphological and magnetic characteristics suggest that the +130 \kms\ feature is associated with the neutral gas that is observed between the Northern and Eastern arms (Jackson \etal\ 1993; Yusef-Zadeh \etal\ 1994). In this picture, the highly blue shifted molecular and atomic gas clouds noted in HCO$^+$, HI, and OH studies (Marr \etal\ 1992; Pauls \etal\ 1993; Yusef-Zadeh, Zhao \& Goss 1994; Yusef-Zadeh 1994; Zhao, Goss \& Ho 1995), the red shifted [OI] (Jackson \etal\ 1993), and the 1720 MHz OH gas cloud are likely to be part of a single feature. The molecular species closer to the Galactic center have velocities ranging between --150 and --180 \kms\ whereas the clouds closer to CND have velocities ranging between +70 to +140~\kms. This feature is considered to be interacting with the circular-moving gas in the CND. This cloud, with an assumed initial velocity of $\approx+$150 \kms, could be colliding with the CND gas with a relative velocity of $\approx+$50 \kms. The resultant collision would disrupt a segment of the CND near the gap and produce the shocked gas at $\approx$+130 \kms. Assuming that the area from which the shocked emission arises is $\approx30''$ (1.2 pc), the total luminosity that the shock generates is $\approx10^3$ \lsol assuming that the gas number density is about 2$\times10^3$ cm$^{-3}$. % This value exceeds the total luminosity of %the 1720 MHz OH maser by seven orders of magnitude. Is the intruding cloud infalling or outflowing from the Galactic center? If the cloud is infalling, a consequence of the interaction is that the red shifted motion of the intruding cloud should become blue shifted as it nears the Galactic center. At the location where the cloud collides with CND, the 130 \kms\ OH maser delineates the shocked region. The schematic diagram in Figure 5 presents a picture in which the intruding cloud is stretched along its trajectory as the magnetic field is expected to become uniform, as suggested by mid and far-IR polarization data. The Zeeman splitting of HI and OH suggest a line of sight magnetic field between --4 and --2 mG field at the position of the shocked gas. The large strength of the magnetic field can be explained as a result of the compression of the field as the intruding cloud collides with the the CND. In this picture the Northern and Eastern arms delineate the edges of the intruding cloud photoionized by the UV radiation field at the Galactic center (Jackson \etal\ 1992). The weakly-emitting ionized Streamers beyond the CND may then be due to the low-density ionized rims of the intruding cloud blown by the powerful winds associated with the IRS 16 cluster (Krabbe \etal\ 1991). \begin{figure} %\plottwo{cnd1.eps}{cnd2.eps} \plotone{cnd_fig5.ps} \caption{A schematic diagram showing the infall model of gas flow toward the Galactic center. This model represents a tongue of gas infalling toward the center. The cross represents the position of Sgr A$^*$.} \end{figure} The above infall picture does not explain the origin of the intruding high velocity cloud. The expansion of the Sgr A East shell into its molecular counterpart (M--0.02--0.07) can account naturally for the origin of the high velocity neutral clouds toward the Galactic center. However, the red-shifted radial velocity of the infalling cloud is inconsistent with the relative location of Sgr A East with respect to the CND. Low-frequency radio continuum observations have revealed that Sgr A East must lie behind Sgr A West (Yusef-Zadeh and Morris 1987; Pedlar \etal\ 1989). Thus, we consider an alternative schematic model, as presented in Figure 6, in which a high negative velocity cloud is accelerated toward the Galactic center as it follows a parabolic orbit around Sgr A$^*$ colliding with the northern part of the CND. This model implies that Sgr A East is behind but in the vicinity of Sgr A West and the CND. There is no evidence favoring the infall vs the outflow models. In the outflow picture, however, the expansion of Sgr A East into the +50 \kms\ molecular cloud can naturally explain the origin of peculiar-moving neutral and ionized clouds (e.g. Yusef-Zadeh, Lasenby \& Marshall 1993). \begin{figure} %\plottwo{cnd1.eps}{cnd2.eps} \plotone{cnd_fig6.ps} \caption{A schematic diagram showing the outflow model of gas flow from the Galactic center. In the outflow picture, the origin of the cloud is associated with the expansion of the Sgr A East SNR which lies behind the Northern and Eastern arms of Sgr A West.} \end{figure} Acknowledgements: We thank Jennifer Reddy for drawing Figures 5 and 6 and Mark Morris for useful discussions. Yusef-Zadeh's work was supported by NASA grant NAGW-2518. The National Radio Astronomy Observatory is a facility of the National Science Foundation, operated under a cooperative agreement by Associated Universities, Inc. \section{References} \begin{references} \reference Aitken, D.K., Gezari, D., Smith, C.H., McCaughrean, M. \& Roche, P.F. 1991, ApJ, 419, 419 \reference Burton, M. \& Allen, D.A. 1992, in {Proc.Astr.Soc.Australia}, 10, 55 \reference Elitzur, M. 1976, ApJ, 203, 124 \reference Elitzur, M. 1996a, ApJ, 457, 415 \reference Elitzur, M. 1996b, ApJ, submitted. \reference Frail, D.A., Goss, W.M. \& Slysh, V.I. 1994, ApJ, 424, L111 \reference Frail, D.A., Goss, W.M., Reynoso, E.M., Giacani, E.B., Green, A.J. \& Otrupcek, R. 1996, A.J., 111, 1651 \reference Gatley, I., Jones, T.J., Hyland, A.R., Wade, R., Geballe, T.R. \& Krisciunas, K. 1986, MNRAS, 222, 299 \reference Gray, A.D., Cram, L.E., Ekers, R.D. \& Goss, W.M. 1991, Nature, 353, 237 \reference Haynes, R.F. \& Caswell, J.L. 1977, MNRAS, 178, 219 \reference Hildebrand, R.H. \& Davidson, J.A. 1994, in {\it The Nuclei of Normal Galaxies: Lessons from the Galactic Center}, eds. R. Genzel and A. Harris, Dordrecht: Kluwer, p199 \reference Jackson, J. M. et al. 1993, ApJ, 402, 173 % Geis, N., Genzel, R., Harris, A. I., Madden, S., % Poglitsch, A., Stacey, G. J. \& Townes, C. H. 1993, \reference Krabbe, A., Genzel, R., Drapatz, S. \& Rotaciuc, V. 1991, ApJ, 382, L19 \reference Lockett, P. 1995, BAAS, 27, 34.01 \reference Marr, J.M., Rudolph, A.L., Pauls, T.A., Wright, M.C., Backer, D.C. ApJ, 1992, 400, L29 \reference Offer, A.R., van Hemert, M.C. \& van Dishoeck, E.F. 1994, J.Chem.Phys, 100, 362 \reference Pak, S., Jaffe, D.T. \& Keller, L.D. 1996, ApJ, 457, L43 \reference Pedlar et al. 1989, ApJ, 342, 769 %, A., Anantharamaiah, K.R., Ekers. R.D., Goss, W.M., van %Gorkom, J.H., Schwarz, U.J., \& Zhao, J.-H. \reference Plante, R.L., Lo, K.Y., \& Crutcher, R.M. 1995, ApJ, 445, L113 \reference Reich, W. \& F\"urst, E. 1984, A.A.Suppl. 57, 165 \reference Robinson, B.T. et al. 1996, in preparation. \reference Turner, B.E. 1982, ApJ. 255, L33 \reference Uchida, K.I., Morris, M., Bally, J., Pound, M. \& Yusef-Zadeh, F. 1992a, ApJ, 398, 128 \reference Uchida, K.I., Morris, M. \& Yusef-Zadeh, M. 1992b, AJ, 104, 1533 \reference Yusef-Zadeh, F., Lasenby, A. \& Marshall, J. 1993, ApJ, 410, L27 \reference Yusef-Zadeh, F. \& Morris, M. 1987, ApJ, 322, 721 \reference Yusef-Zadeh, F., Roberts, D.A., Goss, W.M., Frail, D. \& Green, A. 1996, ApJ, submitted. \reference Yusef-Zadeh, F., Uchida, K.I. \& Roberts. D.A. 1992, Science, 283, 601 \reference Yusef-Zadeh, F., Zhao, J.-H. \& Goss, W.M. 1995, ApJ, 442, 646 \end{references} \end{document} \begin{figure} \plotone{cnd2.eps} \caption{Schematic diagram showing two possible accounts of the} \end{figure} %\begin{figure} %\plottwo{cnd1.eps}{cnd2.eps} %\caption{Schematic diagram showing two possible accounts of the} %\end{figure} \end{figure} \verb"\plottwo{"\arg{file}\verb"}{"\arg{file}\verb"}" \end{quote} The \arg{file} argument is used to name the file(s) to be included. The \verb"\plotone" command includes one figure that is scaled to the width of the current text column; \verb"\plottwo" inserts two figures side by side, and the pair is scaled to fit the text width. If one uses these macros, the necessary vertical space is provided automatically. %\begin{figure} %\plottwo{cnd1.eps}{cnd2.eps} %\caption{Schematic diagram showing two possible accounts of the} %\end{figure} or \begin{quote} \verb"\begin{figure}"\\ \verb"\plottwo{"\arg{mygraph.eps}\verb"}{"\arg{another.eps}\verb"}"\\ \verb"\caption{"\arg{Two related graphics.}\verb"}"\\ \verb"\end{figure}" \end{quote} Please note that the caption will be centered under the {\tt pair} of graphics when \verb"\plottwo" is used. It is not possible to caption the two plots individually with this package at this time. As with tables, figures will be identified with arabic numerals, e.g., ``Figure 1.''. If you must fuss with the positioning and scaling of the EPS plot on the printed page, you can try using this command: \begin{quote} \verb"\plotfiddle{"\arg{file}\verb"}{"\arg{vsize}\verb"}{"\arg{rot}\verb"}{"\arg{hsf}\verb"}{"\arg{vsf}\verb"}{"\arg{htrans}\verb"}{"\arg{vtrans}\verb"}" \end{quote} \begin{quote} \begin{tabular}{lp{3in}} \tt vsize & vertical white space to allow for plot, any valid \LaTeX\ dimension\\ \tt rot & rotation angle, in degrees, counter-clockwise\\ \tt hsf & horiz scale factor, percent\\ \tt vsf & vert scale factor, percent\\ \tt htrans & horiz translation, in PS points 72/in\\ \tt vtrans & vert translation, in PS points 72/in\\ \end{tabular} \end{quote} If you {\em can} produce EPS but you do {\em not} have \prog{dvips}, you can still put the \verb"\plotone" or \verb"\plottwo" commands in the the appropriate places, but you will have to comment them out and put in a \verb"\vspace{"\arg{dimen}\verb"}" command to open up the text. The \prog{dvips} program is in the public domain and is available from labrea.stanford.edu. A special note to authors: Color EPS files should be avoided if possible. And since it is sometimes necessary to edit EPS files to make them printable, authors should try to avoid EPS files with lines longer than 1024 characters. \subsection{Sections} \subsection{Tables} Tables should appear in {\tt table} environments. \begin{quote} \verb"\begin{table}"\\ \verb"\caption{"\arg{text}\verb"}"\\ \verb"\begin{tabular}{"\arg{cols}\verb"}"\\ \verb"\end{tabular}"\\ \verb"\end{table}" \end{quote} There should be only one table per environment. The {\tt table} environment encloses not only the tabular material but also any title (caption) or footnote information associated with the table. Tabular information is typeset within \LaTeX's {\tt tabular} environment; the \arg{cols} argument specifies the formatting for each column. Tables and figures will be identified with arabic numerals, e.g., ``Table 2.''; the identifying text, including the number, is generated automatically by the \verb"\caption" command. There is a \verb"\tableline" command for use in {\tt tabular} environments. \begin{quote} \verb"\tableline" \end{quote} This command produces a single horizontal rule. There should be a \verb"\tableline" above and below between the column headings, and two at the end of the table. Authors should not use additional \verb"\tablelines" themselves, and are discouraged from using vertical rules unless essential. \subsection{EPS Files} ----- End Included Message -----