IT is well known that there are some circumstances, connected with photometric observations, calculated to make us doubt whether, theoretically or observationally, we have, determined correctly the amount of light that is extinguished in its passage through our atmosphere. Foremost amongst these considerations may be mentioned the fact, pointed out some time since by Prof. Seeliger, that the very accurate and trustworthy observations made by Dr. Müller, at Potsdam, with a view to determine this quantity, are not rigorously represented by the theoretical expressions derived by Laplace. The deviations may not be large in amount, but they exhibit a systematic character which is suspicious. In the same connection may be mentioned the initial objection, urged by Prof. Langley, that the fundamental expressions used in those investigations are not equally applicable to light of all wave-lengths. There are, further, in use different numerical values of the coefficient of transmission, pointing either to various degrees of transparency in the atmosphere, or to peculiarities in the instruments themselves, or the methods employed in the reduction of the observations.