The circumgalactic medium (CGM) houses large-scale gaseous inflows and outflows as well as a reservoir of material that can be used for future star formation. Theoretical gas accretion model predictions depend strongly on galaxy mass, morphology, and orientation. I present a two-pronged approach to unravel the origins of the gas flows in the CGM of galaxies at z < 1. First, I assess the prediction that accretion preferentially occurs along the major axis of galaxies by determining the orientation and morphological properties of galaxies associated with CGM absorbers of such low metallicity that they strongly suggest an intergalactic origin. For this I use HST imaging and IFU spectroscopy of fields with absorbers drawn from an HI-selected sample of 35 partial and Lyman limit systems (pLLSs/LLSs, 16.0 ≤ logNHI < 19.0 [cm-2]) — half of which have metallicities so low that an intergalactic medium origin is strongly suggested. Second, I test the prediction that cold, low-metallicity CGM gas accretion should be suppressed in the most massive galaxies by their strong virial shocks. For this I use CGM observations of the very massive luminous red galaxies (LRGs). This study uses two samples: 21 LRGs with QSO sightlines probing the outer CGM to 500 kpc (~Rvir) and 50 LRGs with QSOs probing the inner CGM (~0.3 Rvir), both using HST/COS spectroscopy complemented by ground-based MgII observations. Among the results of my work are that galaxies show no strong dependence of the HI covering factor on mass when comparing a sample of galaxies over the range 9 ≤ logM★ ≤ 12 Msun, and that the metallicity distributions of the CGM in low- and high-mass galaxies also appear indistinguishable, with some of the lowest-metallicity gas found in the halos of LRGs.