Search

MANY of the compact near-infrared sources observed in the central parsec of our Galaxy have been shown to be very massive, luminous blue stars with strong helium lines and fast winds, perhaps similar to Wolf–Rayet stars1–3; and the complex near-infrared source IRS16, which appears to be the central concentration of such sources, has now been shown3 to consist of about 20 luminous blue stars, not 100 main-sequence stars, as had been suggested earlier4. These young blue stars, rather than any object associated with the nearby compact radio source SgrA* (ref. 8), may be the pre-dominant source of ionizing radiation1 and hydrodynamic activity5–7 in the inner two parsecs of the Galaxy. An earlier argument4 that the distribution of IRS16 components precludes the existence of a million-solar-mass black hole at or near SgrA* remains valid even if there are fewer than two dozen stars in IRS16. Here I argue further that the presence of young stars in this region is incompatible with the presence of a giant black hole or any other centrally condensed mass, because tidal forces would have inhibited star formation.

A 2.2-µm image of IRS 16, the relatively blue infrared source near the galactic centre, is presented at the highest spatial resolution yet reported. It comprises several discrete, extended sources. By intercomparison of radio and infrared images of the region, the non-thermal radio source Sgr A* can be located relative to IRS 16 without the uncertainty of intermediate astrometric frames. No component of IRS 16 is coincident with Sgr A*. These data preclude the presence of a massive (>106 M⊙) black hole at the galactic centre. Any black hole present has a mass nearer to 100 M⊙.

A typical extragalactic radio source is a simple double or elongated structure with a well-defined orientation on the sky. In 1972, Willson1 investigated the relative orientation of 74 extragalactic 3C radio sources mapped at the Cambridge 1-mile telescope, and noted that extended extragalactic radio sources tend to be parallel if they are separated by less than 10° on the sky. Since 1972, many more radio sources have been mapped with resolutions 10–20 times greater than that of the observations available to Willson. I have therefore reinvestigated this problem in two separate and largely independent samples of some 300 sources each, and report here that the alignment is present to a high significance level, provided that only the more distant sources are considered. The effect could be due to gravitational imaging if most of the matter in a closed universe is in thin filaments.

Hills1 has pointed out that in some conditions black holes in galactic nuclei would rapidly grow by disruption and accretion of stars thus producing Seyfert luminosities (>1043 erg s−1). In this hypothesis only those galactic nuclei with a high central stellar density are susceptible to Seyfert pathology, and they evolve through only one Seyfert phase. Here, I suggest an alternative model for recurring activity in normal galactic nuclei based on the assumed presence of a massive nuclear black hole (∼107 M⊙) and the observation of a very clumpy distribution of interstellar gas in the inner 200 pc of our Galaxy. The system of massive molecular clouds with low net angular momentum could provide an ‘accretion event’ every 107 yr with a duration of ∼105 yr. In such a picture, most spiral-galaxy nuclei evolve through recurring Seyfert episodes of rather short duration; the 1% accretion duty cycle is roughly consistent with the fact that a few per cent of all spiral galaxies are Seyferts2.