Droplet spectra observed in adiabatic and close-to-adiabatic volumes of convective clouds are typically significantly wider than those predicted by an adiabatic parcel rising from the cloud base, 1-2 microns versus a couple tenths of a micron. Presence of cloud turbulence and its impact on the diffusional growth of cloud droplets has been often used as a possible explanation. The idea is that droplets arriving at a given location follow different trajectories through a turbulent cloud and they feature different growth histories. This is what we refer to as the “eddy hoping”. This mechanism has been investigated over the last two decades applying theory as well as numerical simulations using idealized frameworks of stochastic models, DNS, and scaled-up DNS. These studies suggest that eddy hopping can explain large spectral width of the adiabatic droplet spectra. However, as I discussed in my lecture in May, all those computational studies feature a fundamental flaw of droplets dispersing in the vertical direction, with the large spectral width because of the correlation between droplet radius and its vertical position. In this lecture, I will discuss a new extremely-high-resolution (grid length of 7.5 m) numerical simulation of a turbulent cumulus cloud applying Lagrangian particle-based microphysics. The simulation indeed shows spectral width up to 1 micron in adiabatic volumes not far from the cloud base. To understand these results, I will discuss an idealized framework of a kinematic one-dimensional vertical air plume that crosses the cloud base and forms a cloud. Without turbulence, the results are consistent with the adiabatic parcel model: the spectral width above the cloud base is close to 0.1 micron. However, with turbulence, the kinematic model predicts spectral widths between 1 and 2 microns. Physical mechanisms involved will be discussed.