We report the discovery of the most metal-poor dwarf star-forming galaxy (SFG) known to date, J0811+4730. This galaxy, at a redshift z=0.04444, has a Sloan Digital Sky Survey (SDSS) g-band absolute magnitude M g = −15.41 mag. It was selected by inspecting the spectroscopic data base in the Data Release 13 (DR13) of the SDSS. LBT/MODS spectroscopic observations reveal its oxygen abundance to be 12 + log O/H = 6.98 ± 0.02, the lowest ever observed for a SFG. J0811+4730 strongly deviates from the main-sequence defined by SFGs in the emission-line diagnostic diagrams and the metallicity -luminosity diagram. These differences are caused mainly by the extremely low oxygen abundance in J0811+4730, which is ∼ 10 times lower than that in main-sequence SFGs with similar luminosities. By fitting the spectral energy distributions of the SDSS and LBT spectra, we derive a stellar mass of M ⋆ = 10 6.24 − 10 6.29 M ⊙ (statistical uncertainties only), and we find that a considerable fraction of the galaxy stellar mass was formed during the most recent burst of star formation.

The demographics of dwarf galaxy populations have long been in tension with predictions from the Λ Cold Dark Matter (ΛCDM) paradigm 1-4 . If primordial density fluctuations were scale-free as predicted, dwarf galaxies should themselves host dark matter subhaloes 5 , the most massive of which may have undergone star formation resulting in dwarf galaxy groups. Ensembles of dwarf galaxies are observed as satellites of more massive galaxies 6-9 , and there is observational 10 and theoretical 11 evidence to suggest that these satellites at z=0 were captured by the massive host halo as a group. However, the evolution of dwarf galaxies is highly susceptible to environment 12-14 making these satellite groups imperfect probes of ΛCDM in the low mass regime. We have identified one of the clearest examples to date of hierarchical structure formation at low masses: seven isolated, spectroscopically confirmed groups with only dwarf galaxies as members. Each group hosts 3-5 known members, has a baryonic mass of ~4.4×10 9 to 2×10 10

Many hot subdwarf stars show composite spectral energy distributions indicative of cool main-sequence (MS) companions. Binary population synthesis (BPS) models demonstrate such systems can be formed via Roche lobe overflow or common envelope evolution but disagree on whether the resulting orbital periods will be long (years) or short (days). Few studies have been carried out to assess the orbital parameters of these spectroscopic composite binaries; current observations suggest the periods are long. To help address this problem, we selected 15 moderately bright (V ∼ 13) hot subdwarfs with F-K dwarf companions and monitored their radial velocities from 2005 January to 2008 July using the bench-mounted Medium Resolution Spectrograph on the Hobby-Eberly Telescope (HET). Here we describe the details of our observing, reduction, and analysis techniques, and present preliminary results for all targets. By combining the HET data with recent observations from the Mercator Telescope, we are able to calculate precise orbital solutions for three systems using more than six years of observations. We also present an up-to-date period histogram for all known hot subdwarf binaries, which suggests those with F-K MS companions tend to have orbital periods on the order of several years. Such long periods challenge the predictions of conventional BPS models, although a larger sample is needed for a thorough assessment of the models' predictive success. Lastly, one of our targets has an eccentric orbit, implying some composite-spectrum systems might have formerly been hierarchical triple systems, in which the inner binary merged to create the hot subdwarf.

Hot subdwarf stars with F-K main sequence binary companions have been known for decades, but the first orbital periods for such systems were published just recently. Current observations suggest that most have long periods, on the order of years, and that some are or once were hierarchical triple systems. As part of a survey with the Hobby-Eberly Telescope, we have been monitoring the radial velocities of several composite-spectra binaries since 2005 in order to determine their periods, velocities, and eccentricities. Here we present observations and orbital solutions for two of these systems, PG 1449+653 and PG 1701+359. Similar to the other sdB+F/G/K binaries with solved orbits, their periods are long, 909 d and 734 d, respectively, and pose a challenge to current binary population synthesis models of hot subdwarf stars. Intrigued by their relatively large systemic velocities, we also present a kinematical analysis of both targets and find that neither is likely a member of the Galactic thin disk.

The eclipsing binary system 2M 1938+4603 consists of a pulsating hot subdwarf B star and a cool M dwarf companion in an effectively circular three-hour orbit. The light curve shows both primary and secondary eclipses, along with a strong reflection effect from the cool companion. Here we present constraints on the component masses and eccentricity derived from the Rømer delay of the secondary eclipse. Using six months of publicly-available Kepler photometry obtained in Short Cadence mode, we fit model profiles to the primary and secondary eclipses to measure their centroid values. We find that the secondary eclipse arrives on average 2.06 ± 0.12 s after the midpoint between primary eclipses. Under the assumption of a circular orbit, we calculate from this time delay a mass ratio of q = 0.2691 ± 0.0018 and individual masses of M sd = 0.372 ± 0.024 M ⊙ and M c = 0.1002 ± 0.0065 M ⊙ for the sdB and M dwarf, respectively. These results differ slightly from those of a previously-published light curve modeling solution; this difference, however, may be reconciled with a very small eccentricity, e cos ω ≈ 0.00004. We also report an orbital period decrease ofṖ = (-1.23 ± 0.07) × 10 −10 .