The numbers of three types of compact binaries containing accreting NSs and BHs, which are most interesting from the point of view of their observational appearances, have been calculated as a function of time: i) X-ray transient source containing a NS in an eccentric orbit around a main sequence Be-star (the observed prototype: A0535+26), ii) BH accreting at a highly supercritical rate from the Roche lobe filling component (the assumed prototype: SS 433), and iii) BH accreting from the stellar wind of an OB-supergiant (the observed prototype: Cyg X-1). The evolution of the selected types of X-ray binaries during the first 10 million years after the starburst onset is presented in Fig. 1.
Fig. 1a shows the number of X-ray transient sources consisting of a NS in an eccentric orbit around a massive secondary that acquired enough angular momentum during the first mass exchange to become a rapidly rotating Be-star. This occurs for the binaries whose components have comparable initial masses and are not too distantly separated to avoid the common envelope stage formation during the first mass exchange (the last condition is always satisfied for the massive binaries at the Galactic center). To become an X-ray transient, the NS must accrete matter from the secondary Be-star, at least during the periastron passages. Since the duration of the accretion depends upon the orbital eccentricity, rotational period, and magnetic field of the NS, not all the transients are in the accretion stage at the same time, therefore the observed number of these sources can be a few times less compared to Fig. 1a. Fig. 1b shows the evolution of X-ray binaries with a BH, like Cyg X-1. The number of such sources depends strongly on time, giving source by t=7 Myr. The evolution of a superaccreting BH of SS 433 type is presented in Fig. 1c.
It is tempting to speculate that the presence of a radio jet in the well known X-ray source 1E 1740-2942 (Churasov et al. 1994) may indicate that it belongs to superaccreting BH binaries. It was shown that this source can hardly be a system like Cyg X-1 (Heindl et al. 1993), and there are arguments (Mirabel 1991) that it can be an isolated BH, accreting matter from the surrounding molecular cloud. At the moment, however, there are no direct observations which could prove that this object is an isolated BH.
The ratio of numbers of BH containing binaries (of both SS 433 and Cyg X-1 type) to X-ray transients with Be-stars is plotted in Fig. 1d. This ratio is remarkably sensitive to the time elapsed after the starburst and therefore it can be used for an independent estimation of the age of the X-ray binaries at the Galactic center. Absolute numbers of different systems seem to be worse age-indicators as they are subjected to different poorly known selection effects.