Abstract. Spectroscopic observations of the peculiar variable V838 Mon during the period from the second light outburst until the fast dimming are presented. We describe high resolution (R ≈ 60 000) high S /N spectra obtained a day before the second light maximum and low resolution (R ≈ 6000) spectra covering the whole period. The temporal run of intensities and radial velocities of various lines is presented. Using Na  D IS lines we determine the reddening distance of V838 Mon d > 3.1 kpc, and kinematic distance d > 4 kpc. We estimate that V838 Mon is slightly metal deficient but otherwise has a quite solar-like chemical composition except for enhanced abundances of Li, Ba and La.

Aims. We study the alignment of galaxies relative to their local environment in SDSS-DR8 and, using these data, we discuss evolution scenarios for different types of galaxies. Methods. We defined a vector field of the direction of anisotropy of the local environment of galaxies. We summed the unit direction vectors of all close neighbours of a given galaxy in a particular way to estimate this field. We found the alignment angles between the spin axes of disc galaxies, or the minor axes of elliptical galaxies, and the direction of anisotropy. The distributions of cosines of these angles are compared to the random distributions to analyse the alignment of galaxies. Results. Sab galaxies show perpendicular alignment relative to the direction of anisotropy in a sparse environment, for single galaxies and galaxies of low luminosity. Most of the parallel alignment of Scd galaxies comes from dense regions, from 2 . . . 3 member groups and from galaxies with low luminosity. The perpendicular alignment of S0 galaxies does not depend strongly on environmental density nor luminosity; it is detected for single and 2 . . . 3 member group galaxies, and for main galaxies of 4 . . . 10 member groups. The perpendicular alignment of elliptical galaxies is clearly detected for single galaxies and for members of ≤10 member groups; the alignment increases with environmental density and luminosity. Conclusions. We confirm the existence of fossil tidally induced alignment of Sab galaxies at low z. The alignment of Scd galaxies can be explained via the infall of matter to filaments. S0 galaxies may have encountered relatively massive mergers along the direction of anisotropy. Major mergers along this direction can explain the alignment of elliptical galaxies. Less massive, but repeated mergers are possibly responsible for the formation of elliptical galaxies in sparser areas and for less luminous elliptical galaxies.

Context. Longtime monitoring of gravitational lens systems is often done using telescopes and recording equipment with modest resolution. Still, it would be interesting to get as much information as possible from the measured lightcurves. From high resolution images we know that the recorded quasar images are often blends and that the corresponding time series are not pure shifted replicas of the source variability. Aims. In this paper we will develop an algorithm to unscramble this kind of blended data. Methods. The proposed method is based on a simple idea. We use one of the photometric curves, which is supposedly a simple shifted replica of the source curve, to build different artificial combined curves. Then we compare these artificial curves with the blended curves. Proper solutions for a full set of time delays are then obtained by varying free input parameters and estimating statistical distances between the artificial and blended curves.Results. We performed a check of feasibility and applicability of the new algorithm. For numerically generated data sets the time delay systems were recovered for a wide range of setups. Application of the new algorithm to the classical double quasar QSO 0957+561 A, B lightcurves shows a clear splitting of one of the images. This is an unexpected result and extremely interesting, especially in the context of the recent controversy about the exact time delay value for the system. Conclusions. The proposed method allows us to properly analyse the data from low resolution observations that have long time coverages. There are a number of gravitational lens monitoring programmes that can make use of the new algorithm.

Estimation of time delays from a noisy and gapped data is one of the simplest data analysis problems in astronomy by its formulation. But as history of real experiments show, the work with observed data sets can be quite complex and evolved. By analysing in detail previous attempts to build delay estimation algorithms we try to develop an automatic and robust procedure to perform the task. To evaluate and compare different variants of the algorithms we use real observed data sets which have been objects of past controversies. In this way we hope to select the methods and procedures which have highest probability to succeed in complex situations. As a result of our investigations we propose an estimation procedure which can be used as a method of choice in large photometric experiments. We can not claim that proposed methodology works with any reasonably well sampled input data set. But we hope that the steps taken are in correct direction and developed software is truly useful for practising astronomers.