The inelasticity of hadron-carbon collisions for energies exceeding 100 TeV
is estimated from the carbon-emulsion chamber data at Pamirs to be (K-c) =
0.65 +/- 0.08. When combined with data on hadron-lead collisions taken in
the same energy range it results in the K similar to A(0.086) mass number d
ependence of inelasticity. The evaluated partial inelasticity for secondary
(nu > 1) interactions, K-nu>1 similar or equal to 0.2, suggests that most
of the energy is lost in the first interaction.