In spite of the ambiguity in analyzing the eclipse flash spectrum, pointed out by Giovanelli (1975), the general conclusion of Kanno et al. (1971) is shown to be still valid. Giovanelli (1975) has criticized the conclusion of Kanno et al. (1971), hereafter referred to as Paper I, that coronal emission originates actually in the interspicular region below 104 km above the Sun's limb. (Here the interspicular region corresponds to the atmospheric medium except the spicules which may be regarded as a spherically symmetric atmosphere, because the fractional volume occupied by the spicules is negligibly small in heights greater than some 103 kin.) His principal argument is based on the fact that the observed integrated intensity distribution of the 5303 A line at Point A in Paper I is compatible with even a simple model with zero emissivity up to 5.5 x 103 km, if data fitting is allowed to be within the standard deviation of the observed values. So far as one is concerned with only integrated intensity, it is true in the above sense that the emissivity of 5303 N up to 5.5 x 103 km is to be between zero and twice the value given in Paper I. Furthermore, since the scatter of the data is greater for other 5303 A observations, it is impossible to say anything about the structure of the 5303/~ corona below 10 4 km observed in Paper I.
However, we have additional information on the peak intensity, Ip, as shown in * Add 8.75 to the values in the figures to get logI~ in units of erg s -1 cm -1 ~--1 sterad-1.