The mechanism of the measurement of spin-lattice relaxation time (T1) treated by BLOEMBERGEN, PURCELL and POUND (B. P. P.) has been modified arguing that DEBYE's value of dielectric relaxation time, used in the expression, is valid only for spherical molecules. The magnetic relaxation times calculated from B. P. P. equation show the values ranging from 1/2 to 1/10 times of the experimentally observed values. The principal source of the shortcoming of B. P. P. formula lies in the evaluation of rotational contribution to the spin-lattice relaxation time. A modified expression for T1 has been derived which shows a good agreement between estimated and observed values for a number of rigid spheriodal hydrocarbons.