- Email: [email protected]
Optical colour maps of Seyfert galaxies
Vistas in Astronomy
Vol. 40, No. I, pp. 155-160, 1996 Copyright @ 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0083-6656/96 $32.00 + 0.00
0083-6656(95)00119-O
OPTICALCOLOURMAPSOFSEYFERT GALAXIES JAR1 K. KOTILAINEN
I** and
MARTIN
J. WARD
2
1European Southern Observatory, Karl Schwarzschild Str. 2, D-85748 Garching, Germany
2 Astrophysics, Nuclear Physics Building, Keble Road, Oxford OX1 3RH, U.K.
Abstract- We discuss optical broad band B-I maps of Seyfert galaxies, which show double or elongated blue morphological features. In NGC 1068, NGC 3227, Mrk 3 and Mrk 573 these features are roughly parallel to their extended radio emission and probably arise from scattering of nuclear light by extranuclear “mirrors”. In Mrk 78 and Mrk 348, there is a bluer region set against the red nucleus and host galaxy. In these cases the scatters may be either less efficient and/or further away from the nucleus than in the other galaxies. Our results demonstrate the usefulness of colour maps in enhancing the visibility of low contrast features. Copyright @Elsevier Science Ltd.
1. INTRODUCTION There is now compelling evidence that in Seyfert 1s we see the compact nucleus and BLR directly, whereas in Seyfert 2s optically thick material, possibly in the form of a torus, blocks our direct view. The EELR which lies along the minor axis of the torus is illuminated by the soft X-ray and UV continuum radiation emitted from the nucleus in two oppositely directed cones. Because the putative dusty torus will be redder than the surrounding stellar population, the geometry of the obscuring region should be apparent in colour maps. Here we discuss our high resolution colour maps of the Seyfert galaxies NGC 1068, NGC 3227, Mrk 3, Mrk 78, Mrk 348 and Mrk 573. The maps resolve the circumnuclear regions and reveal the existence of knots and elongations not discernible in single filter images. A full discussion of our results will appear as Kotilainen and Ward (1994).
2. OBSERVATIONS AND DATA REDUCTION CCD images of the galaxies were obtained through broad band B, V, R and I filters using the 2.5 m Nordic Optical Telescope (NOT) on La Palma, in November 1990 and * Tuorla Observatory, University of Turku, FIN-21 500 Piikkio, Finland. 155
156
1 K. Kotilainen and hi. J Ward
December 1992. The pixel scale employed was 0.2” px-‘. Conditions were photometric, with generally subarcsec seeing. The images were reduced using the STARLINK KAPPA and the ES0 MIDAS reduction packages, with standard procedures, i.e. bias and dark subtraction, flatfielding, subtraction of residual sky, removal of cosmic rays and bad pixels, and finally coadding of the multiple exposures.
3. CONSTRUCTION OF THE COLOUR MAPS The colour maps were produced from the broad band images, after alignment of the single colour images to within a small fraction of a pixel. This was achieved either using a field star or the central pixel with highest intensity as the reference point. The former method is prefered, because of the possibility of colour dependence in the position of the nucleus. The frames were also corrected for differences in the magnification factor between the different filters. To avoid undersampling, the flux calibrated images were first smoothed by a Gaussian filter with a width corresponding to the seeing profile appropriate for each frame. Then the frame corresponding to the longer wavelength was divided by the shorter wavelength frame. In order to check that the structures apparent in the colour maps are not artefacts, we constructed similar maps for the quasar 3C 273, which was observed during the November 1990 run. Reassuringly, neither 3C 273 nor the stars in that field show rings or other artefacts. Another potential problem is contamination from emission lines whose wavelength lies within the range covered by the broad filters. This effect can be much reduced by using the B-band (which includes only a weak contribution from Hfi 4861) and the I-band (which covers a region devoid of strong forbidden lines).
4. RESULTS NGC 1068: In the B-I map of the nuclear region, there is an elongated blue structure across the nucleus, extending along PA= 40 deg for about 6” (660 PC). Its colours are much bluer than a typical B-I index for a spiral bulge of T = 2(- 2.4) but redder than the average B-I index for Seyfert 1 galaxies (- 1.2). It is likely that the blue elongation is caused by scattered light from the nucleus of NGC 1068, which is superimposed on the stellar population of the nuclear region, which is redder. Recently, Pogge and De Robertis (1993) presented narrow band emission line and continuum imaging observations of NGC 1068. In their “UV/Red” (3600 A/6100 A) colour map, they find a number of blue features. The bluest region is offset from the nucleus by - 0.8” in PA - 33 deg. It is along the radio axis and roughly coincident with the peak of the [0111] cone structure (Evans et af., 1991) and the radio continuum NE peak (Wilson and Ulvestad, 1983). Additionally, there is an extended, diffuse UV excess continuum component with a conical morphology. The morphology in our B-I map is in good agreement with the structure observed by Pogge and De Robertis, and with the structure seen in the imaging polarimetry of NGC 1068 by Miller et al. (1991). NGC 3227: The B-I image of the nuclear region shows two blue maxima on either side of the nucleus (NW and SE) with a separation - 2”. There is a good correspondence with the maxima in the CO map (Meixner et al., 1990), especially for the SE maximum. Although a
Optical Colour Maps
157
correlation between the CO distribution (molecular clouds) and the colour maps would be expected if the blue maxima are caused by recent star formation, there is no direct evidence of ongoing star formation in NGC 3227. On the other hand, the nucleus of NGC 3227 is polarised, with values rising smoothly from - 1% in the red to nearly 3% in the blue (Schmidt and Miller, 1985), for both the continuum and the broad emission lines. This polarisation can be explained by interstellar dust scattering, where the dust is intermixed with the NLR gas (Schmidt and Miller, 1985). The double maxima might then represent scattering of the nuclear light in the unobstructed directions. Mrk 3: The B-I image of the central region shows the nuclear region to be redder than the surrounding bulge at PA - 100 deg with extent - 11”. Within this red elongation, there is an arclike blue extension, with two main blue maxima, roughly elongated at PA = 95 deg. The W maximum is bluer at - 1.3” from the nucleus, while the E maximum lies at 0.8” from the nucleus. In the E, the elongation curves towards larger PA, fading away - 4.6” from the nucleus. The double structure is roughly aligned with the radio axis $5 deg; Ulvestad and Wilson, 1984) and the curvature of the blue continuum is similar (S-shaped) to that seen in the emission lines (Capetti et al. 1993) and in the radio (Kukula et al., 1993). Probably the blue structure is scattered light from the nucleus and follows the biconical structure. Outside of the region of blue emission, the colour is typical of galactic bulges, showing no sign of star forming regions in its SO type host galaxy. Pogge and De Robertis (1993) presented a “W/Red” continuum colour map of Mrk 3, which revealed two distinct structures: a pair of blue knots on either side of the nucleus along PA = 85 deg, separated by l-5”, and a diffuse blue component with a distinct bi-conical morphology extending out to 5” on either side of the nucleus along PA = 114 deg. Our B-I map is in good agreement with their map. Mrk 78: The B-I map shows the nucleus to be redder than the bulge of the galaxy. The reddest region close to the nucleus is slightly extended (1.1 x 1.7”) along PA - 330 deg, almost perpendicular to the E-W [0111] morphology (Capetti et al., 1994). This may indicate the location of the obscuring dust. There is also an arc-like blue region forming a broken ring from E through N to SW at a distance - 3” from the nucleus. Although this could be a region of star formation around the nucleus, there is no supporting evidence for this. Another possibility is that it is caused by a scattering mirror which is less efficient or further away from the nucleus than in the cases of NGC 1068, Mrk 3 and Mrk 573. Mrk 348: The B-I map of the central region shows the nucleus to be redder than the circum-nuclear region. At larger distances from the nucleus the host galaxy colours become redder. Between these there is an intermediate blue ring-like region, about 3” in diameter, elongated - 15 deg. Again this blue structure may be associated with scattering of the nuclear light. Mrk 573: The B-I map of the central region (Fig. 1) shows two blue maxima on either side of the nucleus with a separation of - 7” at PA - 135 deg and a red “bridge” perpendicular to them crossing the nucleus. This double structure is roughly parallel to the radio axis (PA = 124 deg; Ulvestad and Wilson, 1984) but the blue component is more extended than the radio continuum emission. The regions of blue maxima lie within the radiation cones, with the red bridge indicating the presence of large scale obscuring material across the nucleus. Alternatively the blue maxima could arise from star formation triggered by the radio jet,
158
J K. Kotilainen and M. J Ward
Fig.
I. B-1 map of Mrk 573
however there is no sign of young blue stars in the continuum images. We note that the 10 pm luminosity of Mrk 573 is relatively larger than is typical for Seyfert 2s, suggesting the possibility of hot dust within the NLR (cf. NGC 1068) or the presence of a starburst component within the nucleus (Maiolino et al., 1994). Pogge and De Robertis (1993) presented a “UV/Red” map of Mrk 573, which shows a biconical structure oriented at PA = 124 deg with an opening angle - 60 deg. They also find a curved feature 3.5” SE of the nucleus and a similar feature 4.5” NW Both are coincident with [0111] emission line features, but lie outside the radio continuum lobes. There is good general correspondence between their map and our B-I map; except that we find regions of blue emission that remained unresolved in their map. These maxima are more diffuse and situated further away from the nucleus than in NGC 1068 and Mrk 3.
Optical Colour Maps
159
5. DISCUSSION In the simple unified model the optical, UV and soft X-ray radiation from the nucleus is collimated by the torus and escapes anisotropically along its axis, resulting in two oppositely directed cones. In several cases [OIII]IHa images directly show the conical or bi-conical geometry of the high excitation gas (e.g. Tadhunter and Tsvetanov, 1989; Haniff et al., 1991). How is this related to the structures seen in our colour continuum maps? They are unlikely to be regions of star formation because there is no direct evidence for this in the continuum images, and because the morphology of the regions is diffuse, unlike the sharp boundaries seen in the OB associations further out in the disk of NGC 1068 or the compact, knotty appearance of the star forming regions in the spiral arms of Mrk 573. More appealing alternative is scattering of light from the Seyfert nucleus by off-nuclear dust or electrons, which would bluen the continuum. For NGC 1068 electron scattering close to the nucleus is preferred, because of the wavelength-independent polarisation in the UV (Antonucci, 1993) but in the circumnuclear region dust particles also contribute to the scattering (Miller et al., 1991). The blue knots in NGC 1068 and Mrk 3 are significantly offset from their nuclei, as are their [0111] peaks. In Mrk 573, the blue maxima lie beyond the inner radio continuum lobes. NGC 1068 and Mrk 3 are both examples of Seyfert 2 nuclei with polarised broad lines indicating a hidden Seyfert 1 nucleus (Antonucci and Miller, 1985; Miller and Goodrich, 1991), whereas Mrk 573 has only weakly polarised broad Hfl. It may be that Mrk 573 lacks a compact scattering region, with most of the scattering being distributed over a larger region, producing correspondingly weaker polarisation. In NGC 1068 and Mrk 3 the scatters are close enough to the nucleus to produce strong polarisation. This is confirmed for NGC 1068, where there is a strongly polarised region NE of the nucleus (Miller et al., 1991) with the PA of polarisation perpendicular to the line of sight to the nucleus, and coinciding with the extended blue feature seen in our colour map.
6. SUMMARY Colour maps provide an alternative method to study the morphology of the extended scattering region around Seyfert nuclei. Imaging polarimetry is of course desirable but the levels of polarisation may be low due to dilution by starlight. We have detected clumpy and elongated blue structures in the B-I maps of several Seyfert galaxies. For NGC 1068, NGC 3227, Mrk 3 and Mrk 573, we interpret the colour structure as arising from scattering of the nuclear light from extranuclear mirrors (electrons or dust). In Mrk 78 and Mrk 348 the structure is less clear, which may indicate that the scattering mirrors are less efficient, or they could be situated further away from the nucleus.
References Antonucci R.R.J. (1993) Ann. Rev. Astr. Ap. 31, 473. Capetti A. et af. (1993)BAAS 25, 1472. Capetti A. et al. (1994) Ap. J 421, 87. Evans IN. et al. (1991) Ap. J Lett. 369, L27. Haniff C.A., Ward M.J. and Wilson AS. (1991) Ap. J 368, 167.
160
J. K. Kotilainen and M. J Ward
Kotilainen J.K. and Ward M.J. (1994) Astr. Ap. Submitted. Krolik L. (1989) Ap. J 347, 179. Kukula M.J. et al. (1993) M.NR.A.S. 264, 893. Maiolino ef al. (1994) Ap. L In press. Meixner M. et al. (1990) Ap. J 354, 158. Miller J.S., Goodrich R.W. and Mathews W.G. (1991) Ap. J 378,47 Pogge R.W. and De Robertis M.M. (1993) Ap. J 404,563. Schmidt G.D. and Miller J.S. (1985) Ap J 290, 517. Tadhunter C. and Tsvetanov Z. (1989) Nature 341,422. Tran H.D., Miller J.S. and Kay L.E. (1992) Ap. J. 397,452. Ulvestad J.S. and Wilson A.S. (1984) Ap. J 278, 544. Wilson AS. and Ulvestad J.S. (1983) Ap. J. 275, 8.
Vol. 40, No. I, pp. 155-160, 1996 Copyright @ 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0083-6656/96 $32.00 + 0.00
0083-6656(95)00119-O
OPTICALCOLOURMAPSOFSEYFERT GALAXIES JAR1 K. KOTILAINEN
I** and
MARTIN
J. WARD
2
1European Southern Observatory, Karl Schwarzschild Str. 2, D-85748 Garching, Germany
2 Astrophysics, Nuclear Physics Building, Keble Road, Oxford OX1 3RH, U.K.
Abstract- We discuss optical broad band B-I maps of Seyfert galaxies, which show double or elongated blue morphological features. In NGC 1068, NGC 3227, Mrk 3 and Mrk 573 these features are roughly parallel to their extended radio emission and probably arise from scattering of nuclear light by extranuclear “mirrors”. In Mrk 78 and Mrk 348, there is a bluer region set against the red nucleus and host galaxy. In these cases the scatters may be either less efficient and/or further away from the nucleus than in the other galaxies. Our results demonstrate the usefulness of colour maps in enhancing the visibility of low contrast features. Copyright @Elsevier Science Ltd.
1. INTRODUCTION There is now compelling evidence that in Seyfert 1s we see the compact nucleus and BLR directly, whereas in Seyfert 2s optically thick material, possibly in the form of a torus, blocks our direct view. The EELR which lies along the minor axis of the torus is illuminated by the soft X-ray and UV continuum radiation emitted from the nucleus in two oppositely directed cones. Because the putative dusty torus will be redder than the surrounding stellar population, the geometry of the obscuring region should be apparent in colour maps. Here we discuss our high resolution colour maps of the Seyfert galaxies NGC 1068, NGC 3227, Mrk 3, Mrk 78, Mrk 348 and Mrk 573. The maps resolve the circumnuclear regions and reveal the existence of knots and elongations not discernible in single filter images. A full discussion of our results will appear as Kotilainen and Ward (1994).
2. OBSERVATIONS AND DATA REDUCTION CCD images of the galaxies were obtained through broad band B, V, R and I filters using the 2.5 m Nordic Optical Telescope (NOT) on La Palma, in November 1990 and * Tuorla Observatory, University of Turku, FIN-21 500 Piikkio, Finland. 155
156
1 K. Kotilainen and hi. J Ward
December 1992. The pixel scale employed was 0.2” px-‘. Conditions were photometric, with generally subarcsec seeing. The images were reduced using the STARLINK KAPPA and the ES0 MIDAS reduction packages, with standard procedures, i.e. bias and dark subtraction, flatfielding, subtraction of residual sky, removal of cosmic rays and bad pixels, and finally coadding of the multiple exposures.
3. CONSTRUCTION OF THE COLOUR MAPS The colour maps were produced from the broad band images, after alignment of the single colour images to within a small fraction of a pixel. This was achieved either using a field star or the central pixel with highest intensity as the reference point. The former method is prefered, because of the possibility of colour dependence in the position of the nucleus. The frames were also corrected for differences in the magnification factor between the different filters. To avoid undersampling, the flux calibrated images were first smoothed by a Gaussian filter with a width corresponding to the seeing profile appropriate for each frame. Then the frame corresponding to the longer wavelength was divided by the shorter wavelength frame. In order to check that the structures apparent in the colour maps are not artefacts, we constructed similar maps for the quasar 3C 273, which was observed during the November 1990 run. Reassuringly, neither 3C 273 nor the stars in that field show rings or other artefacts. Another potential problem is contamination from emission lines whose wavelength lies within the range covered by the broad filters. This effect can be much reduced by using the B-band (which includes only a weak contribution from Hfi 4861) and the I-band (which covers a region devoid of strong forbidden lines).
4. RESULTS NGC 1068: In the B-I map of the nuclear region, there is an elongated blue structure across the nucleus, extending along PA= 40 deg for about 6” (660 PC). Its colours are much bluer than a typical B-I index for a spiral bulge of T = 2(- 2.4) but redder than the average B-I index for Seyfert 1 galaxies (- 1.2). It is likely that the blue elongation is caused by scattered light from the nucleus of NGC 1068, which is superimposed on the stellar population of the nuclear region, which is redder. Recently, Pogge and De Robertis (1993) presented narrow band emission line and continuum imaging observations of NGC 1068. In their “UV/Red” (3600 A/6100 A) colour map, they find a number of blue features. The bluest region is offset from the nucleus by - 0.8” in PA - 33 deg. It is along the radio axis and roughly coincident with the peak of the [0111] cone structure (Evans et af., 1991) and the radio continuum NE peak (Wilson and Ulvestad, 1983). Additionally, there is an extended, diffuse UV excess continuum component with a conical morphology. The morphology in our B-I map is in good agreement with the structure observed by Pogge and De Robertis, and with the structure seen in the imaging polarimetry of NGC 1068 by Miller et al. (1991). NGC 3227: The B-I image of the nuclear region shows two blue maxima on either side of the nucleus (NW and SE) with a separation - 2”. There is a good correspondence with the maxima in the CO map (Meixner et al., 1990), especially for the SE maximum. Although a
Optical Colour Maps
157
correlation between the CO distribution (molecular clouds) and the colour maps would be expected if the blue maxima are caused by recent star formation, there is no direct evidence of ongoing star formation in NGC 3227. On the other hand, the nucleus of NGC 3227 is polarised, with values rising smoothly from - 1% in the red to nearly 3% in the blue (Schmidt and Miller, 1985), for both the continuum and the broad emission lines. This polarisation can be explained by interstellar dust scattering, where the dust is intermixed with the NLR gas (Schmidt and Miller, 1985). The double maxima might then represent scattering of the nuclear light in the unobstructed directions. Mrk 3: The B-I image of the central region shows the nuclear region to be redder than the surrounding bulge at PA - 100 deg with extent - 11”. Within this red elongation, there is an arclike blue extension, with two main blue maxima, roughly elongated at PA = 95 deg. The W maximum is bluer at - 1.3” from the nucleus, while the E maximum lies at 0.8” from the nucleus. In the E, the elongation curves towards larger PA, fading away - 4.6” from the nucleus. The double structure is roughly aligned with the radio axis $5 deg; Ulvestad and Wilson, 1984) and the curvature of the blue continuum is similar (S-shaped) to that seen in the emission lines (Capetti et al. 1993) and in the radio (Kukula et al., 1993). Probably the blue structure is scattered light from the nucleus and follows the biconical structure. Outside of the region of blue emission, the colour is typical of galactic bulges, showing no sign of star forming regions in its SO type host galaxy. Pogge and De Robertis (1993) presented a “W/Red” continuum colour map of Mrk 3, which revealed two distinct structures: a pair of blue knots on either side of the nucleus along PA = 85 deg, separated by l-5”, and a diffuse blue component with a distinct bi-conical morphology extending out to 5” on either side of the nucleus along PA = 114 deg. Our B-I map is in good agreement with their map. Mrk 78: The B-I map shows the nucleus to be redder than the bulge of the galaxy. The reddest region close to the nucleus is slightly extended (1.1 x 1.7”) along PA - 330 deg, almost perpendicular to the E-W [0111] morphology (Capetti et al., 1994). This may indicate the location of the obscuring dust. There is also an arc-like blue region forming a broken ring from E through N to SW at a distance - 3” from the nucleus. Although this could be a region of star formation around the nucleus, there is no supporting evidence for this. Another possibility is that it is caused by a scattering mirror which is less efficient or further away from the nucleus than in the cases of NGC 1068, Mrk 3 and Mrk 573. Mrk 348: The B-I map of the central region shows the nucleus to be redder than the circum-nuclear region. At larger distances from the nucleus the host galaxy colours become redder. Between these there is an intermediate blue ring-like region, about 3” in diameter, elongated - 15 deg. Again this blue structure may be associated with scattering of the nuclear light. Mrk 573: The B-I map of the central region (Fig. 1) shows two blue maxima on either side of the nucleus with a separation of - 7” at PA - 135 deg and a red “bridge” perpendicular to them crossing the nucleus. This double structure is roughly parallel to the radio axis (PA = 124 deg; Ulvestad and Wilson, 1984) but the blue component is more extended than the radio continuum emission. The regions of blue maxima lie within the radiation cones, with the red bridge indicating the presence of large scale obscuring material across the nucleus. Alternatively the blue maxima could arise from star formation triggered by the radio jet,
158
J K. Kotilainen and M. J Ward
Fig.
I. B-1 map of Mrk 573
however there is no sign of young blue stars in the continuum images. We note that the 10 pm luminosity of Mrk 573 is relatively larger than is typical for Seyfert 2s, suggesting the possibility of hot dust within the NLR (cf. NGC 1068) or the presence of a starburst component within the nucleus (Maiolino et al., 1994). Pogge and De Robertis (1993) presented a “UV/Red” map of Mrk 573, which shows a biconical structure oriented at PA = 124 deg with an opening angle - 60 deg. They also find a curved feature 3.5” SE of the nucleus and a similar feature 4.5” NW Both are coincident with [0111] emission line features, but lie outside the radio continuum lobes. There is good general correspondence between their map and our B-I map; except that we find regions of blue emission that remained unresolved in their map. These maxima are more diffuse and situated further away from the nucleus than in NGC 1068 and Mrk 3.
Optical Colour Maps
159
5. DISCUSSION In the simple unified model the optical, UV and soft X-ray radiation from the nucleus is collimated by the torus and escapes anisotropically along its axis, resulting in two oppositely directed cones. In several cases [OIII]IHa images directly show the conical or bi-conical geometry of the high excitation gas (e.g. Tadhunter and Tsvetanov, 1989; Haniff et al., 1991). How is this related to the structures seen in our colour continuum maps? They are unlikely to be regions of star formation because there is no direct evidence for this in the continuum images, and because the morphology of the regions is diffuse, unlike the sharp boundaries seen in the OB associations further out in the disk of NGC 1068 or the compact, knotty appearance of the star forming regions in the spiral arms of Mrk 573. More appealing alternative is scattering of light from the Seyfert nucleus by off-nuclear dust or electrons, which would bluen the continuum. For NGC 1068 electron scattering close to the nucleus is preferred, because of the wavelength-independent polarisation in the UV (Antonucci, 1993) but in the circumnuclear region dust particles also contribute to the scattering (Miller et al., 1991). The blue knots in NGC 1068 and Mrk 3 are significantly offset from their nuclei, as are their [0111] peaks. In Mrk 573, the blue maxima lie beyond the inner radio continuum lobes. NGC 1068 and Mrk 3 are both examples of Seyfert 2 nuclei with polarised broad lines indicating a hidden Seyfert 1 nucleus (Antonucci and Miller, 1985; Miller and Goodrich, 1991), whereas Mrk 573 has only weakly polarised broad Hfl. It may be that Mrk 573 lacks a compact scattering region, with most of the scattering being distributed over a larger region, producing correspondingly weaker polarisation. In NGC 1068 and Mrk 3 the scatters are close enough to the nucleus to produce strong polarisation. This is confirmed for NGC 1068, where there is a strongly polarised region NE of the nucleus (Miller et al., 1991) with the PA of polarisation perpendicular to the line of sight to the nucleus, and coinciding with the extended blue feature seen in our colour map.
6. SUMMARY Colour maps provide an alternative method to study the morphology of the extended scattering region around Seyfert nuclei. Imaging polarimetry is of course desirable but the levels of polarisation may be low due to dilution by starlight. We have detected clumpy and elongated blue structures in the B-I maps of several Seyfert galaxies. For NGC 1068, NGC 3227, Mrk 3 and Mrk 573, we interpret the colour structure as arising from scattering of the nuclear light from extranuclear mirrors (electrons or dust). In Mrk 78 and Mrk 348 the structure is less clear, which may indicate that the scattering mirrors are less efficient, or they could be situated further away from the nucleus.
References Antonucci R.R.J. (1993) Ann. Rev. Astr. Ap. 31, 473. Capetti A. et af. (1993)BAAS 25, 1472. Capetti A. et al. (1994) Ap. J 421, 87. Evans IN. et al. (1991) Ap. J Lett. 369, L27. Haniff C.A., Ward M.J. and Wilson AS. (1991) Ap. J 368, 167.
160
J. K. Kotilainen and M. J Ward
Kotilainen J.K. and Ward M.J. (1994) Astr. Ap. Submitted. Krolik L. (1989) Ap. J 347, 179. Kukula M.J. et al. (1993) M.NR.A.S. 264, 893. Maiolino ef al. (1994) Ap. L In press. Meixner M. et al. (1990) Ap. J 354, 158. Miller J.S., Goodrich R.W. and Mathews W.G. (1991) Ap. J 378,47 Pogge R.W. and De Robertis M.M. (1993) Ap. J 404,563. Schmidt G.D. and Miller J.S. (1985) Ap J 290, 517. Tadhunter C. and Tsvetanov Z. (1989) Nature 341,422. Tran H.D., Miller J.S. and Kay L.E. (1992) Ap. J. 397,452. Ulvestad J.S. and Wilson A.S. (1984) Ap. J 278, 544. Wilson AS. and Ulvestad J.S. (1983) Ap. J. 275, 8.