Context: Outflows from low-mass star-forming galaxies are a fundamentalingredient for models of galaxy evolution and cosmology. Aims: The onset of kpc-scale ionised filaments in the halo of the metal-poorcompact dwarf SBS 0335-052E was previously not linked to an outflow. We here weinvestigate whether these filaments provide evidence for an outflow. Methods: We obtained new VLT/MUSE WFM and deep NRAO/VLA B-configuration 21cmdata of the galaxy. The MUSE data provide morphology, kinematics, and emissionline ratios H$\beta$/H$\alpha$ and [\ion{O}{iii}]$\lambda5007$/H$\alpha$ of thelow surface-brightness filaments, while the VLA data deliver morphology andkinematics of the neutral gas in and around the system. Both datasets are usedin concert for comparisons between the ionised and the neutral phase. Results: We report the prolongation of a lacy filamentary ionised structureup to a projected distance of 16 kpc at $\mathrm{SB}_\mathrm{H\alpha} =1.5\times10^{-18}$erg s$^{-1}$ cm$^{-2}$arcsec$^{-2}$. The filaments exhibitunusual low H$\alpha$/H$\beta \approx 2.4$ and low [\ion{O}{iii}]/H$\alpha \sim0.4 - 0.6$ typical of diffuse ionised gas. They are spectrally narrow ($\sim20$ km s$^{-1}$) and exhibit no velocity sub-structure. The filaments extendoutwards of the elongated \ion{H}{I} halo. On small scales the $N_\mathrm{HI}$peak is offset from the main star-forming sites. Morphology and kinematics of\ion{H}{I} and \ion{H}{II} reveal how star-formation driven feedback interactsdifferently with the ionised and the neutral phase. Conclusions: We reason that the filaments are a large scale manifestation ofstar-formation driven feedback, namely limb-brightened edges of a giant outflowcone that protrudes through the halo of this gas-rich system. A simple toymodel of such a conical-structure is found to be commensurable with theobservations.

New results on the radio-quiet type 2 quasar known as the Teacup galaxy(SDSSJ1430+1339) based on the long-slit and 3D spectroscopic data obtained atthe Russian 6-m telescope are presented. The ionized gas giant nebula extendingup to r=56 kpc in the [O III] emission line was mapped with the scanningFabry-Perot interferometer. The direct estimation of the emission line ratiosconfirmed that the giant nebula is ionized by the AGN. Stars in the inner r<5kpc are significantly younger than the outer host galaxy and have a solarmetallicity. The central starburst age (~1 Gyr) agrees with possible ages forthe galactic merger events and the previous episode of the quasar outflowproduced two symmetric arcs visible in the [O III[ emission at the distancesr=50-55 kpc. The ionized gas velocity field can be fitted by the model of acircular rotating disk significantly inclined or even polar to the stellar hostgalaxy.

Lenticular (S0) galaxies are galaxies that exhibit a bulge and diskcomponent, yet lack any clear spiral features. With features consideredintermediary between spirals and ellipticals, S0s have been proposed to be atransitional morphology, however their exact origin and nature is stilldebated. In this work, we study the redshift evolution of the S0 fraction outto $z \sim 1$ using deep learning to classify F814W ($i$-band) HST-ACS imagesof 85,378 galaxies in the Cosmological Evolution Survey (COSMOS). We classifygalaxies into four morphological categories: elliptical (E), S0, spiral (Sp),and irregular/miscellaneous (IrrM). Our deep learning models, initially trainedto classify SDSS images with known morphologies, have been successfully adaptedto classify high-redshift COSMOS images via transfer learning and dataaugmentation, enabling us to classify S0s with superior accuracy. We find thatthere is an increase in the fraction of S0 galaxies with decreasing redshift,along with a corresponding reduction in the fraction of spirals. We find abimodality in the mass distribution of our classified S0s, from which we findtwo separate S0s populations: high-mass S0s, which are mostly red andquiescent; and low-mass S0s, which are generally bluer and include both passiveand star-forming S0s, the latter of which cannot solely be explained via thefaded spiral formation pathway. We also find that the S0 fraction in high-massgalaxies begins rising at higher $z$ than in low-mass galaxies, implying thathigh-mass S0s evolved earlier.