In a second memoir Herschel followed up this inquiry with an acuteness worthy of his genius. Unfortunately he allowed himself to be carried away with the idea that the Sun was inhabited in order to sustain this theory. He needed a solid kernel upon which his imaginary inhabitants could dwell; and also a means whereby he could protect them from the radiations of the photosphere. With this idea in view he conjectured the existence above the Sun’s solid body of a layer of clouds always contiguous to the photosphere which enveloped it, and which always being rent when the photosphere was rent, thus enabled us to see the solid body of the Sun lying behind. These notions can only be described as very arbitrary, as unsupported by observation, and as involving explanations quite out of harmony with the principles of modern physics. However, the labours of Herschel resulted in so many positive discoveries of visible facts, and in so many just conclusions, that they contributed greatly to the growth of our present knowledge of the true constitution of the Sun. Since Wilson’s time, as Secchi pointedly remarks, astronomers generally have verified his observations with good instruments, and by an investigation of a great number of spots. De La Rue, discussing the Kew observations, found that of 89 regular spots 72 gave results which conformed to Wilson’s ideas, whilst the remaining 17 were opposed thereto. There is nothing surprising in the existence of a contrarient minority when we consider the great changes which in reality often occur in the forms of the spots. De La Rue suggested a very simple expedient for showing that the spots are cavities. Take two photographs of the Sun made at an interval of one day: during that time every point on the Sun’s surface will have been displaced, so far as the telescope is concerned, by about 15°. Place these photographs in a stereoscope, and we shall readily see the interior cavity, the edges of which will appear raised above the photosphere. It is impossible therefore to entertain the least doubt as to the truth of the theory that the spots are excavations in the luminous stratum which envelopes the whole of the solar globe. If it be true that a spot is a cavity, it follows that when it reaches the margin of the solar disc we ought to detect a hollow place; and this will be so much the more easy to observe according as the cavity is larger and deeper. As a matter of fact, numerous observations of this sort have been recorded from the time of Cassini down to the present time under the designation of “notches” on the Sun’s limb. On July 8, 1873, Secchi observed such a notch 8″, or 3600 miles deep.