The difficulty is so great that many efforts—some bold and daring, others positively wild in the unscientific absurdity of their nature—have been made to overcome it.
Among the most ingenious of these is (or rather was, for I think it is no longer maintained even by its eminent author), Prof. Tyndall’s theory of a comet’s tail as an actinic cloud, generated by the passage of the solar rays through exceedingly tenuous matter after those rays had been in part deprived of their heating power, during their passage through the comet’s head. According to this theory the actinic cloud can not be formed under the heating rays, but so soon as the actinic rays fall on the tenuous matter alone, the cloud is formed,—so that all round the region in which would be the comet’s shadow, there is no luminous cloud, while along that region the cloud exists. The rapidity with which light travels would of course make this explanation absolutely perfect in explaining cometic tails lying always exactly in a straight line directed from the sun, or with their axis so situated. But unfortunately this exceedingly rapid formation of the tail (a tail of ninety million miles in length would be formed in about eight minutes) is more than observation requires or can explain. Prof. Tyndall made a slight oversight in dealing with this part of his theory. Noticing that the actinic cloud, as he called it, is not formed instantly, but after a delay of a few seconds, in his experiments, he reasoned as though it would follow from this that the formation of the actinic cloud behind a comet’s head in space might be a process extending its action in distance from the head at a rate considerably less than that at which light travels, yet still fast enough to account for the exceedingly rapid formation of the tail of Newton’s comet, and of other similar tails. But a little consideration will show that the few seconds following the fall of light on the vapors dealt with by Tyndall, before the luminous cloud appeared, would produce no such effect as he imagined. The rate of formation of the tail would still be that at which light travels. Imagine the head at A, for the sake of argument, and the sun’s light after reaching A, passing on to B, C, D, E, etc., to Z, a distance say of one hundred million miles, in nine minutes:
A . . . B . . . C . . . D . . . E . . . . . . Z.
Suppose that, when the light has reached the vaporous matter lying at B, an interval of one full minute (much greater than any noticed in Tyndall’s experiments), occurs before the actinic cloud comes into view, a similar interval after the light has passed C before the cloud is seen there, and so on, up to the time of the arrival of the light at Z. Professor Tyndall’s reasoning implied that all the time intervals thus occurring at B, C, D, E, etc., up to Z, had to be added together, to give the total time of the formation of the tail from A to Z, and hence naturally a long time might elapse, and the head having at the end of this time reached a different position from that which it had occupied at the beginning, the divergence of the tail from the direction exactly opposite to the sun, and the curvature of the tail, would be alike readily accounted for. But what are the actual facts of the case. The part of the tail formed latest by the supposed solar actinic action, namely, the part at Z, would be formed just nine minutes after the light had left A, and ten minutes after the part nearest to A had been formed (by the same light waves), for, nine minutes after leaving A, the light would be at Z, and a minute after each epoch (according to our supposition) the actinic cloud would be formed respectively at A and at Z. We get just the same interval—nine minutes—whether the actinic cloud appears immediately after light has traversed the vapour which is to form the cloud, or a minute after, or an hour after. In every case the tail would be formed outwards from A, at the rate at which light travels. This does not accord with the phenomena—in fact, the supposition that a tail could be formed at the rate at which light travels, will be found, on examination, to lead to many most manifest absurdities, which Professor Tyndall doubtless recognized when he sought escape from the supposition of such rapid tail formation, through the effects he attributed to the delayed appearance of the actinic cloud.
Another theory in explanation of the rapid formation of such a tail as that of Newton’s comet is worthy of far less notice. Professor Tyndall’s theory was based on an interesting physical fact, which he had himself discovered, and which was also manifestly akin in character to the formation of a comet’s tail. The one to be now noticed was suggested to a mathematician by a rather familiar phenomenon, the effects of which on his imagination he seems to have been never able to entirely overcome—at any rate no amount of evidence against the theory seems to counterbalance in his mind the notion once conceived that the theory might be true. (It is a way some theorists have.)
Professor Tait was once looking at a part of the sky which seemed clear. As he looked, a long streak rapidly formed, which presently disappeared (if I remember his original description aright) almost as rapidly as it had formed. At any rate, the appearance of the streak was rapid enough to remind him of what astronomers said about the rapid (apparent) development of comets’ tails. The phenomenon itself was easily explained. There had been a flight of seabirds, traveling after their wont in a widely extended layer, which when he began his observations had been looked at somewhat aslant, so that—the distance being too great for the birds to be seen individually—nothing of the flight could be discerned at all. But it is evident that in such a case a very slight movement on the part of each bird would suffice so to shift the position of the layer in which they were traveling, that it would be seen edgewise, and then the birds, being so situated that the range of sight toward any part of the layer passed athwart a great number of them, would of course be seen, not individually but as a cloud, or long straight streak, a side view in fact of the layer in which they were traveling. Eureka! shouted Professor Tait; and presently announced to the world the marvelous theory that the rapid formation of comets’ tails may be accounted for on the same general principle. Astronomers have found that along the tracks of some comets (where the tails never lie, by the way, but that is a detail) are countless millions of meteoric bodies separately undiscernable (and never yet discerned as a cloud—another detail); therefore it follows that the tails of all comets are formed by movements of “brickbats and paving stones” in them (Professor Tait’s own description of meteors), after the manner of the seabirds he saw from Arthur’s Seat. Professor Thomson at the Edinburgh meeting of the British Association endorsed this theory with special reference to the value of the “seabird analogy” in explaining the phenomena of Newton’s comet. Dr. Huggins, who, as he does not claim to be a mathematician (or to speak more correctly, as his labors in physical research have not given him time for profound mathematical research), may be more readily excused, also speaks of the seabird theory as if it had some legitimate standing. “The tail, he conceives,” he says, referring to Dr. Tait, “to be a portion of the less dense part of the train illuminated by sunlight, and visible or invisible to us, according, not only to circumstances of density, illumination, and nearness, but also of tactic arrangement, as of a flock of birds under different conditions of perspective.” Of course, the theory is utterly untenable—by astronomers who know something of the actual facts, and have enough mathematics to consider simple geometrical relations. Bodies moving in a plane surface like birds, if they individually travel in the same plane, keep its position unchanged. But if they move individually at an angle to that plane (as they occasionally do), they change its position—the surface, however, in which they collectively are at any moment, still remaining plane. In such a case only could such a phenomenon as was observed by Professor Tait be seen. But in such a case the visibility of the streak formed by the flight of birds would last but a few minutes, for the same motion which had in a few minutes brought the streak into view would in the next few minutes take it out of view. During the short time that a flight is visible in this way, it has an unchanging position, or a scarcely changing one. If the tail of Newton’s comet had rapidly formed and as rapidly vanished, remaining, while visible, in an almost unchanging position, the “seabird analogy” might explain that particular phenomenon, however inadequate to explain multitudes of others. But the phenomena to be explained are entirely different. Leaving out of the question the varying position and length of the tail as it approached the sun, and after it left the sun’s neighborhood, all of which were entirely inconsistent with the seabird analogy, what we are called upon to explain is that a visible tail ninety millions of miles in length, seen in position 1A on one day, was seen three days later in position 3A (having manifestly in the meanwhile passed through all the intermediate positions, including 2A). If Professor Tait, profound mathematician though he be, though he may “differentiate and integrate like Harlequin,” can show how any flight of bodies, like or unlike seabirds, can accomplish such a feat as the above, appearing first to form a thin streak A1, and in less than four days a thin streak A3, each ninety millions of miles long, without some of them having had to travel a distance nearly equal to the line 1 to 3—or some one hundred and fifty millions of miles long, instead of the trifling journeys he assigned them, he should take a rank above Newton and Laplace as a mathematician. But there is another feat, apparently equally difficult to him, which he might achieve very readily with great advantage to those non-mathematicians among astronomers whom his name—well deserved, too—as a mathematician has hitherto misled, and with not less advantage to his own reputation: he might frankly admit that the idea which occurred to him while watching those unfortunate seabirds, had not quite the value which at the moment he mistakenly attached to it, and has since seemed to do.
But apart from the consideration of theories such as those, either demonstrably untenable, though ingenious, like Professor Tyndall’s, or altogether and obviously untenable like Professor Tait’s, there are certain phenomena of comets’ tails which force upon us the belief that they are phenomena of repulsion, though the repulsive action is of a kind not yet known to physicists.
1. The curvature of all the cometic tails when not seen from a point in or near the place of their motion.
2. The existence of more tails than one to the same comet, the different tails being differently curved.