In this paper, we study the strong non-stationary stochastic processes that take place in the phase space of self-gravitating systems at the initial non-stationary stage of their evolution. The numerical calculations of the compulsive phase mixing process were carried out based on the model of chaotic impacts, according to which the initially selected phase volume experiences random impacts of a different and complex nature.

We study brightness variations in the double lensed quasar UM673 (Q0142-100) with the aim of measuring the time delay between its two images. In the paper we combine our previously published observational data of UM673 obtained during the 2003 - 2005 seasons at the Maidanak Observatory with archival and recently observed Maidanak and CTIO UM673 data. We analyze the V, R and I-band light curves of the A and B images of UM673, which cover ten observational seasons from August 2001 to November 2010. We also analyze the time evolution of the difference in magnitudes between images A and B of UM673 over more than ten years. We find that the quasar exhibits both short-term (with amplitude of \sim 0.1 mag in the R band) and high-amplitude (\sim 0.3 mag) long-term variability on timescales of about several months and several years, respectively. These brightness variations are used to constrain the time delay between the images of UM673. From cross-correlation analysis of the A and B quasar light curves and error analysis we measure the mean time delay and its error of 89 \pm11 days. Given the input time delay of 88 days, the most probable value of the delay that can be recovered from light curves with the same statistical properties as the observed R-band light curves of UM673 is 95{+5/-16}{+14/-29} days (68 and 95 % confidence intervals). Analysis of the V - I color variations and V, R and I-band magnitude differences of the quasar images does not show clear evidence of the microlensing variations between 1998 and 2010.

We report the results of our multicolor observations of PG 1115+080 with the 1.5-m telescope of the Maidanak Observatory (Uzbekistan, Central Asia) in 2001-2006. Monitoring data in filter R spanning the 2004, 2005 and 2006 seasons (76 data points) demonstrate distinct brightness variations of the source quasar with the total amplitude of almost 0.4 mag. Our R light curves have shown image C leading B by 16.4d and image (A1+A2) by 12d that is inconsistent with the previous estimates obtained by Schechter et al. in 1997 - 24.7d between B and C and 9.4d between (A1+A2) and C. The new values of time delays in PG 1115+080 must result in larger values for the Hubble constant, thus reducing difference between its estimates taken from the gravitational lenses and with other methods. Also, we analyzed variability of the A2/A1 flux ratio, as well as color changes in the archetypal "fold" lens PG 1115+080. We found the A1/A2 flux ratio to grow during 2001-2006 and to be larger at longer wavelengths. In particular, the A2/A1 flux ratio reached 0.85 in filter I in 2006. We also present evidence that both the A1 and A2 images might have undergone microlensing during 2001-2006, with the descending phase for A1 and initial phase for A2. We find that the A2/A1 flux ratio anomaly in PG 1115 can be well explained both by microlensing and by finite distance of the source quasar from the caustic fold.

We report new estimates of the time delays in the quadruple gravitationally lensed quasar PG1115+080, obtained from the monitoring data in filter R with the 1.5-m telescope at the Maidanak Mountain (Uzbekistan, Central Asia) in 2004-2006. The time delays are 16.4 days between images C and B, and 12 days between C and A1+A2, with image C being leading for both pairs. The only known estimates of the time delays in PG1115 are those based on observations by Schechter et al. (1997) -- 23.7 and 9.4 days between images C and B, C and A1+A2, respectively, as calculated by Schechter et al., and 25 and 13.3 days as revised by Barkana (1997) for the same image components with the use of another method. The new values of time delays in PG 1115+080 may be expected to provide larger estimates of the Hubble constant thus decreasing a diversity between the H_0 estimates taken from gravitationally lensed quasars and with other methods.

We present and analyse new R-band frames of the gravitationally lensed double quasar FBQ 0951+2635. These images were obtained with the 1.5m AZT-22 Telescope at Maidanak (Uzbekistan) in the 2001-2006 period. Previous results in the R band (1999-2001 period) and the new data allow us to discuss the dominant kind of microlensing variability in FBQ 0951+2635. The time evolution of the flux ratio A/B does not favour the continuous production of short-timescale (months) flares in the faintest quasar component B (crossing the central region of the lensing galaxy). Instead of a rapid variability scenario, the observations are consistent with the existence of a long-timescale fluctuation. The flux ratio shows a bump in the 2003-2004 period and a quasi-flat trend in more recent epochs. Apart from the global behaviour of A/B, we study the intra-year variability over the first semester of 2004, which is reasonably well sampled. Short-timescale microlensing is not detected in that period. Additional data in the i band (from new i-band images taken in 2007 with the 2m Liverpool Robotic Telescope at La Palma, Canary Islands) also indicate the absence of short-timescale events in 2007.

The paper has been withdrawn because double checking and comparison with other data sets after the original submission showed that a broken R-band filter at the Maidanak telescope had affected our quasar monitoring observations in the years 2004 and 2005. They had led to partially spurious measurements, hence our original analysis and conclusions are not reliable.

We present new brightness monitoring observations of the 4 components of gravitationally lensed system Q2237+0305, which show detection of an intrinsic quasar brightness fluctuation at a time of subdued microlensing activity, between June 27 and October 12, 2003. These data were used to determine the time delays between the arrivals of the four images. The measured delays are -6, 35, and 2 hours for images B, C and D relative to A, respectively, so they confirm that the long history of brightness monitoring has produced significant detection of microlensing. However the error bars associated with the delays, of order 2 days, are too large to discriminate between competing macro-imaging models. Moreover, our simulations show that for the amplitude of this intrinsic fluctuation and for photometric errors intrinsic to optical monitoring from our 1.5-m telescope or from the OGLE monitoring, a daily sampled brightness record cannot produce reliable lags for model discrimination. We use our simulations to devise a strategy for future delay determination with optical data. Nevertheless, we regard these first estimates to be significant, since they are the first direct measurements of time delays made for this system from ground-based observations in the visual wavelengths. Our results provide the most convincing confirmation of the gravitational-lens nature of Q2237+0305, and give observational justification to the extensive literature which attributes the quasar's previously observed brightness fluctuations to microlensing.

We present the results of a monitoring campaign of the double quasar SBS1520+530 at Maidanak observatory from April 2003 to August 2004. We obtained light curves in V and R filters that show small-amplitude \Delta m~0.1 mag intrinsic variations of the quasar on time scales of about 100 days. The data set is consistent with the previously determined time delay of \Delta t=(130+-3) days by Burud et al. (2002). We find that the time delay corrected magnitude difference between the quasar images is now larger by (0.14+-0.03) mag than during the observations by Burud et al. (2002). This confirms the presence of gravitational microlensing variations in this system.

Photometry of the Q2237+0305gravitational lens in VRI spectral bands with the 1.5-m telescope of the high-altitude Maidanak observatory in 1995-2000 is presented. Monitoring of Q2237+0305 in July-October 2000, made at nearly daily basis, did not reveal rapid (night-to-night and intranight) variations of brightness of the components during this time period. Rather slow changes of magnitudes of the components were observed, such as 0.08 mag fading of B and C components and 0.05 mag brightening of D in R band during July 23 - October 7, 2000. By good luck three nights in 1999 were almost at the time of the strong brightness peak of image C, and approximately in the middle of the ascending slope of the image A brightness peak. The C component was the most blue one in the system in 1998 and 1999, having changed its (V-I) color from 0.56 mag to 0.12 mag since August 1997, while its brightness increased almost 1.2 mag during this time period. The A component behaved similarly between August 1998 and August 2000, having become 0.47 mag brighter in R, and at the same time, 0.15 mag bluer. A correlation between the color variations and variations of magnitudes of the components is demonstrated to be significant and reaches 0.75, with a regression line slope of 0.33. A color (V-I) vrs color (V-R) plot shows the components settled in a cluster, stretched along a line with a slope of 1.31. Both slopes are noticeably smaller than those expected if a standard galactic interstellar reddening law were responsible for the differences between the colors of images and their variations over time. We attribute the brightness and color changes to microlensing of the quasar's structure, which we conclude is more compact at shorter wavelengths, as predicted by most quasar models featuring an energizing central source.

The problem of determination of the velocity anisotropy parameter (A) from star density profiles in globular clusters is considered. The calculations are performed for following clusters: M3, M12, M13, M15, M55, M56, Palomar 3, Palomar 14, NGC5139 (Omega Centauri) and NGC6535. We found that for most clusters is A=0. For M56 the parameter is calculated at limiting magnitudes B=21.5 and B=19.0 and we found A=0.8 and A=0.3 respectively. For Omega Centauri the anisotropy parameter is determined in three separate regions and we found: A=0 for central region, A=0.1 for middle region and A=0.3 for halo.