i6o 



NA TURE 



[December i6, 1897 



to the coordinates in Hansen's tables, corrected from New- 

 comb's numbers. 



No. 



of 



Obs. 



Paris Mean 



Time 1897 D it ^a AS 



d. h. m. , „ „ s. s. „ // 



July 23 13 9 15 32'87io'i4 indeterminate + o"3oio'oi. —o'iio'2 36 



Oct. 13 14 5 15 32"86io"24 57'3""2i:i"'o + o"3iioo2.-t-o'5±o"4 29 



A New Form of Mirror for a Reflecting Telescope. 

 — During the dedication exercises held in connection with the 

 Yerkes Observatory, Dr. C. L. Poor advocated and exhibited a 

 reflecting telescope in which the mirror is a portion of a para- 

 boloid of revolution cut from the surface near the extremity of 

 the latus rectum. The reflected rays then being at right angles 

 to the incident rays, no dome would be required for such a 

 telescope, and there would be no secondary mirror. This form 

 of telescope was, however, recommended by Prof. Pickering 

 more than sixteen years ago (Nature, 1881, August 25) ; more- 

 over, Prof. Schaeberle shows in the Astronomical Journal, No. 

 419, the inefficiency of such an instrument, from the following 

 considerations. Let L denote the distance from the focus to the 

 centre of the mirror, which is evidently inclined about 45° to the 

 line of sight. If D denotes the minimum diameter of this 

 elliptical mirror, the maximum diameter must be D sec. 45° if a 

 circular cone of rays is to be used. The linear distance from 

 the focus to the nearest and most distant points of the mirror 

 will then be approximately — 



Least distance = L - ^ D sec 45°. 

 Greatest distance = L -f ^ D sec 45°. 



If we assume ^r = 1, the greatest distance divided by the 



least distance becomes i'"22. 



This quantity is approximately the blurring factor for the 

 given ratio of focal length to aperture for this form of instru- 

 ment. For a star which is only 5' from the optical axis of the 

 telescope, and in a place containing the longer axis of the 

 mirror, the image will, therefore, be a line no less than 66", or 

 more than a minute of arc. 



Exactly at the focal point this star image will be a point, but 

 for all other positions of the image the definition will be 

 unworkable. 



RECENT RESEARCHES ON TERRESTRIAL 

 MAGNETISM} 



'X'HE science of terrestrial magnetism has on one previous 

 ^ occasion formed the topic of a Rede Lecture. Twenty- 

 five years ago Sir E. Sabine delivered a discourse on this subject, 

 with which his name will always be honourably connected. 

 The length of time which has elapsed may perhaps justify a 

 return to the same theme, though it must be admitted that now, 

 as then, the study of the magnetic properties of the earth is 

 in an early stage of development. It is true that considerable 

 advances have been made in the theory of the nature of 

 magnetism itself, and of its connection with electricity ; but 

 when we attempt to apply theory lo explain the actual con- 

 dition of the earth progress is at once checked by difficulties, 

 many of which have hitherto proved insuperable. We have no 

 real knowledge of why the earth is a magnet, no real knowledge 

 as to why its magnetic state is continually changing, and thus 

 we are compelled to spend long periods of time in collecting 

 facts, which, though their number and complication oppress us, 

 are still insufficient to answer some of the simplest questions 

 that an inquirer, approaching the subject for the first time, 

 would be sure to ask. Terrestrial magnetism is in this respect 

 in the same stage as that occupied by astronomy during the 

 centuries in which the data were accumulated on which Kepler 

 and Newton worked. We have a certain grasp of the facts, but 

 have not yet found the thread of theory which binds them 

 together. 



And in one respect the magnetician is less favourably 

 situated than was the astronomer. The rapid repetition of 

 the principal astronomical events made it comparatively easy 

 to discover the laws which those events obey ; but, though 

 some magnetic phenomena run through their courses in a day, 

 a year, or a short period of years, the greatest change of all, 

 that which causes the magnet to point now to the east and now 



1 The " Rede Lecture " delivered in the Senate House, Cambridge, on 

 June 9, by Prof. A. W. Rucker, F.R.S. 



NO. 1468, VOL. 57] 



to the west of the geographical north, has been studied for three 

 hundred years and is still unfinished. It is a secular variation, 

 of which the period, if definite period there be, must be 

 measured by ages, and centuries may yet elapse before the first 

 cycle which man has watched will be complete. 



In spite of these difficulties attempts are continually being 

 made to draw from the facts at our disposal some more definite 

 information as to the causes of terrestrial magnetism ; to foretell 

 the future from the present ; to trace the connection between 

 the magnetic state of the earth and the constitution of the sun 

 or of the earth itself ; and I propose, therefore, to bring before 

 you some of the theories and speculations which are now 

 attracting the attention of those who take special interest in 

 this science. 



The fundamental fact, or rather series of facts, from which 

 we have to begin our investigation is a knowledge of the 

 magnetic state of the surface of the earth. To determine this, 

 observations have for many years past been made at many 

 different places, at sea and on land. The general result is a 

 matter of common knowledge. The compass needle points ap- 

 proximately north and south, and dips from the horizontal 

 towards the magnetic poles of the earth. 



The first and simplest hypothesis that will serve as a rough 

 approximate explanation of these facts, is that the earth itself 

 is uniformly magnetised, or that there is at the centre of the 

 earth a small but very powerful magnet by which the compass 

 and the dipping needle are controlled. 



If this suggestion were adequate, we should be compelled to 

 assume that the axis of the magnet was inclined to the axis of 

 the earth, for the magnetic and geographical poles do not 

 coincide. It would further follow that at the magnetic poles, 

 where the dipping needle is vertical, the magnetic force, which 

 determines the position of the needle, would be of maximum 

 intensity. 



But here the simple hypothesis breaks down. The distribu- 

 tion of terrestrial magnetism is more complex than that which 

 can be thus explained. It is true that there are two magnetic 

 poles, but the directive force is not greatest where the needle is 

 vertical. On the contrary there are in each hemisphere two 

 other points, generally called magnetic foci, at which the force 

 is a maximum. 



It is thus evident that the magnetic system of the earth might 

 be better represented by supposing that there are within it two 

 magnets inclined both to each other and to the geographical 

 axis, that the foci indicate the directions of these axes, and that 

 the magnetic pole or point where the needle stands vertical is 

 determined by their joint action. Mr. IL Wilde attempted to 

 imitate the magnetic state of the earth by the aid of a duplex 

 arrangement of this kind, but even this was insufficient. He 

 was compelled to supplement it by covering with thin sheets of 

 iron those portions of the globe which correspond to the oceans, 

 and with this modification he succeeded in making a capital 

 magnetic model of the earth. 



For the moment, however, I will not follow up the line of 

 inquiry thus suggested, but will only draw attention to the fact 

 that, in spite of all these complications, mathematical analysis 

 supplies us with the means of answering certain questions as to 

 the magnetic constitution of the earth, without the aid of a clear 

 mental picture of the causes to which that magnetic state is due. 

 Whether there be one or more independent magnetic systems 

 within the globe, whether some portions are more magnetic 

 than others, are points upon which at present we have but little 

 information, but there are a few facts from which we can argue 

 with the knowledge that the foundations of our investigation 

 are secure. 



Magnetic forces can be produced only by magnetised matter, 

 or by electric currents, and these may either exist within the 

 globe or be external to its surface. Some of the currents, how- 

 ever, may be both internal and external in the sense that their cir- 

 cuits pass partly through the rocks and partly through the air, and 

 that at certain points they traverse the surface from earth to air 

 or from air to earth. Thus the first important question with which 

 the investigator is confronted is : Are the forces which act upon 

 the compass produced within or without the globe ? and, if the 

 magnetic forces are in part due to electric currents, are all these 

 currents wholly internal or wholly external, or do some of them 

 flow in part within and in part without the earth ? 



With regard to the first inquiry, the great mathematician 

 Gauss furnished us with a method by which, if our knowledge 

 of the magnetic state of the surface of the earth is sufficiently 



