Copy of Letters Sent to "Nature" on Partial Impact.

By Professor Bickerton.

The following fifteen letters were sent to "Nature," with the intention of their being published one each week, in order that the essential idea of each letter should stand alone, without being prejudiced by any inaccuracy or supposed inaccuracy of thought or fact in the other letters; also that the class of ideas contained in "partial impact" should be gradually appreciated. It was believed that this method would elicit discussion, and the weakness or merits of the hypothesis might be ascertained before publishing the whole in a book.

This was the more desirable, as, of course, but little discussion could occur in a young colony like New Zealand.

The letters are here printed in the form, and with the dates, on which they were sent to "Nature." All the letters have the fault of being too condensed, to comply with the regulations of "Nature "; in fact most of the easier reasoning is taken for granted.

So many interesting facts have been elicited since they were written, that the letters might advisedly be written again; but, as it would be many months before I could spare the time to do so, and for other reasons, I have left them as they were. Some of these new facts appear to actually demonstrate the hypothesis in its outlines.

A rough outline of the theory, is that there are stupendous numbers of bodies distributed throughout space, almost all in a state of motion, and that probably most of these bodies are not moving in regular orbits. Consequently they occasionally come into collision, and this probability is also increased by their mutual attractions. Secondly, that, in case of a collision occurring, it is far more probable that the impact should be partial than that they should meet fair, centre to centre. Thirdly, that, in the event of such a partial collision, generally, a piece common to both will be struck off, which will coalese, and the two wounded stars would pass on in space. Fourthly, that these two will be hotter on the wounded side, and, as they revolve, will form variable stars. The middle part forming a temporary star, if a very small ratio of the whole, otherwise a nebulas or a system according to the varying circumstances. The two variable stars may escape each others attraction, and that of the middle body, and may travel into space; but often they may be attracted back by the increased attraction, which must act upon them after collision, and the two will then become a double star, orbitally connected.

It will be at once seen that if this theory represents the truth, that it is reasonable to suppose that sometimes variable stars will be in pairs; also double stars, which have been recently connected, will have one on both of their constituents, still exhibiting traces of their former variability.

We should also expect to find that variable stars remain where temporary stars have appeared, and expect that temporary stars should be occasionally, though rarely, recurrent. That variable stars should gradually lose their variability, and become ordinary stars, and, lastly, for certain reasons variable stars would have great irregularities of motion, which would alter the time of their rotation, and would also alter the intensity or apparent intensity of their periods of maximum and minimum, and generally give a want of uniformity to their action.

To ascertain if any variable were in pairs, Herschel's list was searched, and a few pairs found; but Chamber's Astronomy was found to contain a few more variables than Herschel's, and a ohart has been made from this list. Eleven pairs—that is, page 3 22 stars—are extremely close together, and ten more pairs appear to be close enough together that, assuming a moderate proper motion, they may have acquired their positions in a few thousand years. Probably, in the larger lists to be found in Europe, a greater ratio would be found. Mr, Arthur Beverly, of Dunedin, has kindly made an approximate calculation of the probability of such an arrangement being the result of chance. To eliminate much of the extreme complexity, the problem was taken in the following form Given 50,000 spaces and 120 particles, what is the probability of the 120 only occupying 100 spaces; it is approximately one divided by 162 Sextillians. I need not say these associated variables must be so associated by law. It behoves any one, who is not prepared to accept the conclusion of partial impact, to show that it is in error, or to give a better hypothesis.

Of Variable Doubles—After I had searched all the astronomies that could be obtained for variable doubles, and could find only a few casually mentioned, a gentleman told me that he had seen a long account of them in the "Intellectual Observer" for 1862. I requested him to look and ascertain if any were variable, and he found the remarkable fact that Struve had actually satisfied himself that twenty-three doubles were variable, and suspected forty-two more. I dare say some of my readers, with Antipodean advantages, may ascertain if any of these doubtful ones have been confirmed. Mr. Townsend, who has a magnificent 6 in, equatorial, and who is devoting himself largely to double stars, is distinctly of opinion that Alpha Centaura is variable. All the other points mentioned above have each of them also many representatives.

Many nebulæ are accompanied by two stars. It would be particularly interesting should any of these be found to be variable, or orbitally connected, or both receding from each other. Although I need not say, that without such further evidence, the hundreds of coincidences already found, render any idea of mere chance, an absurdity. In addition to all these coincidenes the theory appears to give an absolutely satisfactory explanation of that permanent stumbling block, the origin of cosmical rotation. It accounts for the forms and changes of form of the nebulas; for the high velocity of meteors and comets. It gives a reasonable suggestion for the origin of the whole of the motion of the solar system, and possibly accounts for the structure of the whole visible universe. It gives a logical reason for an extension of the mathematicians age of the sun's heat, which naturalists so much demand, and not improbably furnishes a means of escape from the melancholly conclusion of the theory of dissipation of energy.

Of course it is not supposed that the whole of the vast number of inferences made are right, or that great modifications may not be required. It is impossible that an isolated worker can think of all the conditions, and it is the especial work of criticism to show such oversights. It is simply claimed that such a large series of coincidences could not occur, unless the outlines of the theory had some basis in truth, and consequently it is worth some investigation.