We characterise the conditional distributions of the HI gas-to-stellar massratio, $R_{HI}\equiv M_{HI}/M_{\ast}$, given the stellar mass, $M_{\ast}$, oflocal galaxies from $M_{\ast}\sim 10^7$ to $10^{12}$ $M_{\odot}$ separated intocentrals and satellites as well as into late- and early-type galaxies (LTGs andETGs, respectively). To do so, we use 1) the homogeneous "eXtended GALEXArecibo SDSS Survey", xGASS (Catinella et al. 2018), by re-estimating theirupper limits and taking into account them in our statistical analysis; and 2)the results from a large compilation of HI data reported in Calette et al.(2018). We use the $R_{HI}$ conditional distributions combined with the GalaxyStellar Mass Function to infer the bivariate $M_{HI}$ and $M_{\ast}$distribution of all galaxies as well of the late/early-type andcentral/satellite subsamples and their combinations. Satellites are on averageless HI gas-rich than centrals at low and intermediate masses, with differencesbeing larger for ETGs than LTGs; at $M_{\ast}>3-5\times 10^{10}$ $M_{\odot}$the differences are negligible. The differences in the HI gas content are muchlarger between LTGs and ETGs than between centrals and satellites. Ourempirical HI Mass Function is strongly dominated by central galaxies at allmasses. The empirically constrained bivariate $M_{HI}$ and $M_{\ast}$distributions presented here can be used to compare and constrain theoreticalpredictions as well as to generate galaxy mock catalogs.

Over the past decades, the kinematics of galaxies in the local Universe andat intermediate redshift (i.e., z~1-3) have been characterized in great detail,but only a handful of galaxies at high redshift (z>4) have been examined insuch a way. The Atacama Large Millimeter/submillimeter Array (ALMA) LargeProgram to INvestigate [CII] at Early times (ALPINE) survey observed 118star-forming main sequence galaxies at z=4.4-5.9 in [CII]158um emission,increasing the number of such observations by nearly an order of magnitude. Tocharacterize the morpho-kinematics of this sample, we apply a well-testedtilted ring model fitting code (3DBarolo), a quantitative morphologicalclassification (Gini-M20), and a set of disk identification criteria to theALPINE data. By exploring the G-M20 of z>4 rest-frame FIR and [CII] data forthe first time, we find that our 1"~6kpc resolution is insufficient to separategalaxy types based solely on these data. Of the 75 [CII]-detected ALPINEgalaxies, 29 are detected at high enough significance and with sufficientspatial resolution to allow for tilted ring model fitting and the derivation ofmorpho-kinematic parameters. By combining these results with diskidentification criteria, we are able to robustly classify 14 of the 29 fitsources (six rotators, five mergers, and three dispersion-dominated systems),with the remaining sources showing complex behaviour. We then compare therotation curves and dynamical mass profiles of the six ALPINE rotators to thetwo previously detected z~4-6 unlensed main sequence rotators, finding highrotational velocities (~50-250km/s) and a range of rotation curve shapes.