This paper presents a theory of collisions between small drops in a turbulent fluid, focusing on collisions between equal drops. The drops considered are much smaller than the small eddies of the turbulence, so collision rates depend on drop sizes, energy dissipation rate ε, and kinematic viscosity ν. The collision efficiency for nearly equal drops is assumed to be unity, and the collision rate is given by N = 1.30(r₁ + r₂)³n₁n₂(ε/ν)¹/², valid for r₁/r₂ between 1 and 2. A numerical integration shows how an initially uniform distribution changes due to collisions. An approximate calculation also considers collisions between drops of different inertia due to gravity and turbulence. The results are applied to atmospheric clouds to assess turbulence's role in rainfall initiation. Estimates of ε are made for typical conditions, and initial collision rates, mean properties, and large drop production are calculated. Turbulence effects in layer clouds are small, but moderate turbulence in cumulus clouds can broaden drop size distributions. In heterogeneous clouds, collision rates increase, and drop inertia becomes significant. Turbulence-induced collisions between unequal drops become comparable to gravity when ε ≈ 2000 cm² sec⁻³. The paper discusses two collision mechanisms: collisions due to motion with the air and collisions due to motion relative to the air. The former is due to spatial velocity variations, while the latter is due to differences in drop inertia. The collision rate is calculated using the similarity hypothesis, and the results are applied to estimate the time for mean drop mass to increase by 50%. The results show that turbulence can slowly affect cloud properties, but in cumulus clouds, it can initiate rainfall. The theory is validated by comparing with experimental data and shows that turbulence can lead to significant drop size distribution broadening in uniform clouds. The effects of gravity and turbulence become comparable when ε ≈ 2100 cm² sec⁻³. The paper concludes that turbulence plays a significant role in cloud dynamics, particularly in cumulus clouds, but not in stratiform clouds.This paper presents a theory of collisions between small drops in a turbulent fluid, focusing on collisions between equal drops. The drops considered are much smaller than the small eddies of the turbulence, so collision rates depend on drop sizes, energy dissipation rate ε, and kinematic viscosity ν. The collision efficiency for nearly equal drops is assumed to be unity, and the collision rate is given by N = 1.30(r₁ + r₂)³n₁n₂(ε/ν)¹/², valid for r₁/r₂ between 1 and 2. A numerical integration shows how an initially uniform distribution changes due to collisions. An approximate calculation also considers collisions between drops of different inertia due to gravity and turbulence. The results are applied to atmospheric clouds to assess turbulence's role in rainfall initiation. Estimates of ε are made for typical conditions, and initial collision rates, mean properties, and large drop production are calculated. Turbulence effects in layer clouds are small, but moderate turbulence in cumulus clouds can broaden drop size distributions. In heterogeneous clouds, collision rates increase, and drop inertia becomes significant. Turbulence-induced collisions between unequal drops become comparable to gravity when ε ≈ 2000 cm² sec⁻³. The paper discusses two collision mechanisms: collisions due to motion with the air and collisions due to motion relative to the air. The former is due to spatial velocity variations, while the latter is due to differences in drop inertia. The collision rate is calculated using the similarity hypothesis, and the results are applied to estimate the time for mean drop mass to increase by 50%. The results show that turbulence can slowly affect cloud properties, but in cumulus clouds, it can initiate rainfall. The theory is validated by comparing with experimental data and shows that turbulence can lead to significant drop size distribution broadening in uniform clouds. The effects of gravity and turbulence become comparable when ε ≈ 2100 cm² sec⁻³. The paper concludes that turbulence plays a significant role in cloud dynamics, particularly in cumulus clouds, but not in stratiform clouds.