It is evident that torsional resistance of a reinforced concrete (RC) member is attributed to both concrete and steel reinforcement. However, recent structural design codes neglect the contribution of concrete because of cracking. This paper reports on the results of an experimental and numerical investigation into the torsional capacity of concrete beams reinforced only by longitudinal rebars without transverse reinforcement. The experimental investigation involves six specimens tested under pure torsion. Each specimen was made using a cast-in-place concrete with different amounts of longitudinal reinforcements. To create the torsional moment, an eccentric load was applied at the end of the beam whereas the other end was fixed against twist, vertical, and transverse displacement. The experimental results were also compared with the results obtained from the nonlinear finite element analysis performed in ANSYS. The outcomes showed a good agreement between experimental and numerical investigation, indicating the capability of numerical analysis in predicting the torsional capacity of RC beams. Both experimental and numerical results showed a considerable torsional post-cracking resistance in high twist angle in test specimen. This post-cracking resistance is neglected in torsional design of RC members. This strength could be considered in the design of RC members subjected to torsion forces, leading to a more economical and precise design.