"If, furthermore, we suppose as the more likely theory that the projections follow a rectilinear direction, then the unknown luminary from which the apparitions emanate will have to satisfy two conditions: first, it must be in such a position that photographs can be taken from it; secondly, it must be in such a position that the images obtained can be transmitted to us. Let us take as an instance a case in which it is possible to fix the place and date. The first Montgolfier balloon, filled with hot air, was sent up from Annonay at four o'clock in the afternoon on the 5th of June, 1783. It is easy, by referring to the contemporary calendars, to learn which celestial bodies were at that moment above the horizon and at what height. We thus find that Mars, Jupiter and the moon were invisible, whereas the sun and Venus were at 50 and 23 degrees respectively above the horizon of Annonay and, of course, towards the west. These two luminaries alone then were in a position to witness the experiment of the brothers Montgolfier. But they did not witness it from the same altitude: a view taken from the sun would have shown things as seen from above, whereas, at the same hour, Venus would have shown then from an angle very nearly approaching the horizontal. "This is a first clue. Are we able to check it? Yes, by turning up the date on which the projections of the view then secured as observed by Victorien Beaugrand and by determining whether, on that date, the projecting luminary was able to light up the screen at Meudon. Well, on that day, at the hour which Victorien Beaugrand has given us, Mars and the moon were invisible, Jupiter was in the east, the sun close to the horizon and Venus a little way above it. Projections emanating from the last-named planet could therefore have fallen upon the screen, which as we know faced westwards. "This example shows us that, however frail my theory may appear, we are now able and shall be even better able in the future to subject it to a strict control. I did not fail to resort to this method in respect of the other pictures, and I will give in a special table, appended to this essay, a list of the data which I have verified, a list necessarily drawn up, in some haste. Well, in all the cases which I examined, the views were taken and projected under such conditions that they can logically be referred to the planet Venus _and to that planet alone_.