We have obtained high resolution (2.9″∼260 pc) aperture synthesis imaging of NGC 1068 in the CO J = 1 → 0 line made with the Owens Valley mm Interferometer. The major features seen in CO are: (1) the inner spiral arms of molecular gas at ∼15″ radius (1.5 kpc) which originate from the ends of the central stellar bar, and (2) a compact source (∼3″) coincident with the Seyfert 2 nucleus. The components within 2 kpc of the galactic nucleus account for approximately 30% of the total molecular gas content in NGC 1068. The spiral arm CO emission is resolved into 38 discrete complexes which closely correlate with regions of strong Hα and 10 μm emission. The sizes of these structures range up to 500 pc and their masses derived from the CO line flux are 2×107−7×10 M⊙. Somewhat smaller mass estimates are obtained from the virial theorem using the observed CO linewidths and sizes, suggesting that these complexes may be self-gravitating.

Optical and near infrared imaging of the luminous infrared galaxies show significant asymmetries, distortions, double nuclei, bars, and tidal tails indicating that the galaxies have undergone significant dynamical perturbations. Single dish CO data indicate extremely high molecular gas contents (1-20 times that of the Milky Way) in these galaxies. Approximately 25 of them have been mapped at high resolution using the millimeter interferometers and in virtually all cases, extremely high gas concentrations are found in their nuclei. In most instances, a substantial fraction of the total molecular gas content is seen within radii ≲ 1 kpc; in a few cases, the gas is concentrated in a central bar or massive spiral arms (eg. NGC 1068). The dense molecular gas probably plays a pivotal role in the evolution of such dynamically disturbed systems: being dissipative, the gas can readily sink to the center of the system where it may fuel a nuclear starburst and/or build up and fuel a central active nucleus.

The presence in the central few kpc of a disk galaxy of (1) a bar or a triaxial bulge, (2) a central mass concentration, and (3) a weak dissipation, make up the dynamical ingredients able to speed up considerably, (1) the thickening of the disk by vertical instabilities, and (2) the angular momentum losses, necessary to further accretion of mass. Some consequences compatible with observations are that bulges can be formed in large part after the disk formation phase, and that bars can be destroyed by a central mass concentration of about 109M⊙.

We present first results of a survey for CO emission from a sample of southern Seyfert galaxies using the SEST. The sample consists of all galaxies from the Revised Shapley-Ames Catalog which are south of the celestial equator and are listed in Table 3 of the compilation of active galaxies by Veron-Cetty and Veron (A Catalogue of Quasars and Active Nuclei - ESO Scientific Report No. 7, 1989). During Dec 21- 31, 1989, we observed 30 galaxies and detected 6/8 Type 1 Seyferts, 7/14 Type 2 Seyferts, 0/2 LINERs, 4/4 unclassified Seyferts, and 2/2 galaxies with active nuclei, but not classified as Seyferts.

We have made a search for circumnuclear 12CO (J=1−0) emission from 28 starburst and 29 Seyfert galaxies with recession velocities less than 5000 km s−1, using the Nobeyama Radio Observatory 45-m telescope. The full half-power beam width of 17 arcsec covers less than about 5 kpc in diameter for the sample galaxies. The mean beam size-to-optical diameter ratios are 0.19±0.13 and 0.16±0.11 for the starburst and Seyfert galaxies, respectively. Circumnuclear CO emission has been detected from 12 starburst and 14 Seyfert (7 type 1, 6 type 2, and one peculiar AGN) galaxies.

The interacting edge-on galaxy NGC 4631 has been observed in the 12CO(J = 2 − 1) line emission using the IRAM 30-m telescope with a resolution of 13″ (330 pc). The molecular gas is strongly concentrated in a ring-like disk of 1 kpc radius, which is rotating rigidly. Line proriles show several velocity components, which are attributable to spiral arms. Although the J=2−1 to 1−0 transition line ratio indicates that the gas is generally opaque against the lines, we find some optically thinner regions, as well.

We have mapped the edge-on galaxy NGC3628 in the 12CO(2−1) and 13CO(2−1) lines with the IRAM 30m MRT and in radio continuum at λ20 cm with the VLA. The 12CO(2−1) spectra were obtained out to a distance of 3 kpc east and 2 kpc west of the edge-on galaxy (resolution 0.4 kpc) with supplementary spectra ≃ 9 kpc east and west (1 arcmin ≃ 2 kpc at a distance of 6.7 Mpc). The resulting map shows a central peak of CO emission but also a multiple peaked structure from which we deduce a ringlike enhancement of molecular gas with a diameter of ≃ 400 pc and two outer maxima likely indicating spiral arms of the galaxy. The view of a molecular “ring” is supported by the fact that the maxima of the CO intensity are located close to the turnover radius where rigid rotation passes into differential rotation. Lesch et al. (1990) found for a number of galaxies with molecular rings that these rings can be formed by viscous accretion of molecular gas even at these radii. We find for the rigid rotating disk a molecular mass of m ≃ 108M⊙ assuming a conventional conversion factor. From our channelmaps we were able to distinguish between six unresolved clumps in the central region with comparable CO emissivity and masses of ≃ 106M⊙.

The detected 13CO J = 3 → 2 line in the center of the Sc galaxy NGC 253, along with lower frequency line data, provide further evidence that the starburst nucleus contains warm (TK ∼ 100 K), dense (n(H2) ∼ 104−105 cm−3) molecular gas.

39 pairs of galaxies have been observed (Ref. 1–4 and references herein). They are listed in Table 1. They can be classed in two categories. For 23 of them, one component has a compact stellar-like nucleus (Byurakan classes 4 and 5) and the other component has a faint nucleus of undefined form (Class 2) or no nucleus (Class 1); for one pair in this category, one component has a compact nucleus and the other belongs to Class 3. The components of the 16 other pairs have similar nuclei, compact or not. Out of the 39 pairs 6 have one component with a complex nucleus (Subclass 2s), the other component having a compact or non-compact nucleus.

A review of recent developments in three problems is made. (1) Galaxy formation. Quite different patterns are found if the galaxy is mainly gaseous or mainly stellar. In the first case the motions are highly organized, while in the second case they are mostly irregular. (2) Barred galaxies. One can construct roughly self-consistent stellar models for the bar and for the outer spiral, beyond the −4/1 resonance. The stellar orbits near corotation are stochastic, but gas can join smoothly the bar with the outer stellar spiral. (3) Normal Spirals. We found that strong and open spirals (Sb, Sc) terminate near the 4/1 resonance, while weak tight spirals (Sa) can extend all the way to corotation.

Numerical simulations show that repetitive starbursts will occur in a merger of two gas-rich disc galaxies. These bursts are caused by the orbital motion of two dense gaseous cores which have been formed at the centres of the original galaxies during the early phase of the merger. Such repetitive starbursts are able to explain the observational result that merging galaxies show a wide range of star formation activity.

Some regular 2-armed spiral galaxies (e.g. M81, M100) reveal also local 4-armed features, especially close to the radius of UHR (ultraharmonic, or 4:1 resonance). They may at least partially originate in gasous disk component. We performed some SPH (smoothed particle hydrodynamics) simulations of a model gas disk perturbed by the harmonic (2-armed) perturbation potential representing the gravity of the stellar arms. SPH calculations produce the inter-arms near the UHR due to non-linear dynamical effects. They can be interpreted as the first wavelength of a viscously damped density wave originating at UHR. We present elswhere (Artymowicz, 1990; Artymowicz and Lubow, 1990) an analytical slightly non-linear theory of harmonic and ultraharmonic wave launching, propagation and dissipation in gaseous disks with arbitrary soundspeed, self-gravity and viscosity, in the perturbing gravitational potential of arbitrary form. The main conclusions are as follows:

(i) The physics of wave generation at UHR and the Lindblad resonances is similar. The total torque exerted on the disk by the external potential is practically independent of the amount of viscosity and self-gravitation in both cases.

(ii) The total torque depends sensitively on the disk's rotation curve and the tightness of the imposed spiral structure.

(iii) Applied to M81, the theory including linear and quadratic terms predicts a large and hence observable amplitude of the 4-armed waves at UHR, consistent with the recent observational work of Elmegreen et al. (1989).

(iv) Transfer of negative angular momentum from the stellar to the gaseous disk results in a significant damping of the stellar wave. Direct conversion of the harmonic to ultraharmonic stellar wave at UHR may be even more efficient.

(v) We propose that UHR plays a major role in limiting the global instabilities (e.g. of SWING or WASER-type) in M81 and presumably many other spirals.

M81 is clearly interacting with its two neighbouring galaxies M82 and NGC3077. First we simulated the deformation of the atomic gas in the triplet, using a straightforward model (restricted to the 3-body problem), in order to determine the parameters of the trajectories for the three galaxies. The resulting external tidal features satisfactorily resemble the observed HI bridges and tails. We then used a model of cloud collisions to simulate the response of the M81 molecular content. However the 3-body model parameters are not compatible with such a model. Stronger perturbations are needed in a self-consistent gas and stars simulation to produce and amplify a spiral density wave in the inner disk of M81. Preliminary tests show that a strong wave can be amplified both in the stars and gas components, the global structure bearing some resemblance with the observations. The model parameters must now be improved in the light of the most recent observational data.

Traditionally, the potential used for the study of the orbital structure of a bar is derived from a more or less realistic analytical mass distribution. Since the light distribution of real barred galaxies does not look like the mass distribution of the models (see [1]), we have directly computed the potential from CCD frames. We have applied our method on a sample of 17 early-type barred galaxies and, for one of them, we have computed orbits in its calculated potential.

We present simulations of the gas behaviour in the M51 spiral potential, perturbed by the tidal field of NGC5195. These 3D computations are very helpful to interprete the molecular observations (arm-interarm contrast, Z-behaviour, map of velocity dispersion, kinematics, streaming motions) and in particular to guess the pattern speed of the spiral.

Computer simulations of the encounter of a satellite galaxy with a host galaxy are described. These are being carried out using tree-code gravity on an NCUBE parallel processor with 128 processors. The particles of the host - representing both stars and gas - are distributed over the processors, and the potential of the whole ensemble is calculated using a fast parallel communications algorithm to pass information between the processors, with the result that the amount of processing and communications scales as log N where N is the number of particles.

A backend processor has been constracted which is a pipeline for calculating the gravitational force exerted on one particle from the rest in an astrophysical many-body system (Sugimoto et al. '90). The first version of the processor, GRAPE-1 (GRAvity PipE), was designed with 8 bits floating point format for use in collision-free systems (Ito et al. '90). It is in operation at the speed of 120M flops equivalent.

We demonstrate some peculiar features that can be taken as signs of interaction between galaxies. The features are sharp edged one armed spirals and two-component density and/or velocity structure. We give a kinematical explanation of the first arm and show typical velocities in this arm.

When a small galaxy of mass M2 passes in the vicinity of a triaxial or oblate primary galaxy of mass M1M2, some gas could be stripped from the secondary falling down in the gravitational field of the companion and/or lifted from the less bound regions of the last one. These gaseous particles will point out the presence of a torque induced by the disturbing object on the primary. Assuming a homogeneous ellipsoid for the main body, analytical evaluations are made as function of the geometry on the possible co/counter-rotation or polar rings for the gas involved. An application is considered to the well studied galaxy A0136-0801 (Schweizer et al 1983), testing the possibility to form the observed polar ring. We assume the primary is a normal S0 disk and the ring's gas is coming from the disk plane. Due to the axial symmetry, for two symmetric elements of mass ΔM/2 lying on the x-axis at a distance l/2 from the center of mass, the quadrupole torque due to M2 passing at distance r with colatitude θ is: where are the frame vectors and G the gravitational constant. Comparing it with the restoring torque due to a homogeneous spheroid of major semi-axis a, miming the main body of the galaxy, we obtain for the ratio: where ∊ is the semi-axis ratio, z the height on the equatorial plane and k' the positive solution of the equation for the cofocal spheroid touching the ΔM/2 particle positions. We obtain that the static configuration considered is consistent with the possibility to send mass to the pole assuming: M1 = 4.5 1010 MO, M1/M2 = 2.25, l/2 = 10kpc, a=4kpc, θ=45° and r=15kpc, if (i) Γqlasts for about a revolution period of the ΔM/2 particle and (ii) the mass of the main body is enclosed in a not too much small or large scale length, this last being one of the more strict conditions. The weight of the higher harmonic of multipole has been considered.

The stellar system and the ISM are coupled to each other by their common gravitational field. Since the point-like structure of the individual, randomly moving stars represents a permanent perturbation of the gravitational potential, collective reactive modes are generated in the stellar as well as in the gaseous component [2, 4]. Thus the evolution of the distribution function of the stars is determined by pure stellar encounters as well as by interactions with collective modes. Using fluctuation theory in quasilinear approximation we derive a Fokker-Planck-equation including both effects [1, 3]: where Dpoint describes pure stellar encounters and Dcollec describes the interaction with collective modes. For a Gaussian velocity distribution, the expression in the curved brackets vanishes. Hence collective modes - induced by the point-like structure of the stars - give rise to an enhanced relaxation towards a Gaussian velocity distribution. The individual stars suffer an additional dynamical friction with the ambient gas represented by R(v). R is very small for subsonic velocities but exhibits a sharp increase near the sound speed cs. For v > cs it is R ∝ v−3. The contribution to Dcollec arising from collective modes of the gas has the same v-dependence as R(v), the respective amplitudes, however, depend in different ways on the large-scale parameters of the system. In the case that the friction time scale is smaller than the relaxation time scale, the velocity distribution may considerably deviate from a Gaussian shape. The general consequence of the additional friction leads to a ‘cooling’ of the stellar component [Just et al., this volume]. For example, newly formed stellar clusters may ‘cool down’ within their parental molecular cloud till σ< cs. We find the respective time scale to be of the order of some 106yr.