The observations performed at the Pic du Midi Observatory with a K-coronameter at different altitudes above the solar limb allow us to compute the vertical density gradient of the corona, therefore giving the value of the hydrostatic temperature T _h computed under the assumption of hydrostatic equilibrium in the lower corona. We compare T _h with the ratio r = I ₅₃₀₃/I ₆₃₇₄ of the intensities of the green and the red coronal lines observed at the same position angle and on the same day: r is expected to yield the proportion of cold regions and hot regions along the line of sight under study. As a matter of fact no relation appears between individual values of T _h and r nor between averaged values of T _h and r (Figure 1). In a second step we compute the 5303 Å emission which should be observed, for selected position angles where 6374 Å emissions are faint, making use of the electron density values which can be deduced from the K corona observations. The ratio q = (I ₅₃₀₃) obs/(I ₅₃₀₃) calc of observed to computed 5303 Å intensities should depend upon the corona temperature according to the ionization theory. But again, we do not find the expected relation since Figure 2 displays only a large scatter of points. We conclude that large deviations to hydrostatic equilibrium probably occur in the corona but it is still possible that the presence of arch structures in the lower corona makes it unrealistic to interpret a ratio of intensities integrated along different lines of sight in term of a regular density gradient, although this procedure has been universally used up to now.