July 17, 1896.] 



SCIENCE. 



n 



fact, the few points Pi'of. Franklin specifies are 

 faulty to an unnecessary degree. 



1. In determining the 26.68-day period of 

 the solar rotation, Prof. Franklin insists (p. 

 808, cols. 1 and 2) that I used the ' aspects of 

 the solar corona as photographed during several 

 eclipses ' for that purpose. It is true that in 

 my earlier papers an attempt was made in this 

 direction, but it was not employed in my final 

 computation. On p. 712 Astron. and Astro- 

 Physics, No. 118, Oct., 1893, it is said, " the re- 

 sults there published have been entirely super- 

 seded by the method which will now be de- 

 scribed." There is no possible excuse for hav- 

 ing thus confused my work, even spreading out 

 the erroneous conception over an entire page 

 of Science. 



2. Prof. Franklin does not seem to know 

 whether I used the ordinary Gaussian method 

 of least squares in dealing with the residuals, 

 but is inclined to think I did. As a matter of 

 fact, I did not use that method, but avoided it 

 as incapable of reaching the desired results ; it 

 having been demonstrated to be so by many 

 previous investigations. My method is de- 

 scribed on pp. 713, 714, A. and A. P., No. 

 118, and bears no resemblance to Gauss's pi*o- 

 cess, uses the variations of the earth's magnetic 

 field and not the visible coronal lines. It in- 

 volved a large amount of labor in computation, 

 and the details have not yet been published. 

 In this instance as well, a confusion of mind, to 

 such an extent, is inexcusable in a critic. 



3. On p. 808 Prof. Franklin says : "At this 

 place we interpose the remark that the position 

 is regarded as proven that the sun and the 

 moon do not continuously influence the terres- 

 trial field by direct action as magnets?" This 

 is an old view, held by students generally, I ad- 

 mit, but it was the very essence of my research 

 to inquire whether it is sound or not. A critic 

 may throw the entire case out of court by 

 such a remark, and then proceed to abuse 

 the author of the investigatiod, but it is not 

 criticism and it is not science. The old view is 

 based upon a very oflf-hand computation of 

 magnets (Mascart & Joubert, Elect. & Mag. , p. 

 417, Vol. I.), adduces no study of observations 

 to test it, and in general lets the case go by 

 default. Yet the experimental evidence is 



abundant that the sun does in some ivay send 

 its energy to the earth, in peculiar periodic 

 variations, aside from the ordinary light field. 

 To go no deeper now, we know that the sun 

 spots and corona, as well as the faculte and 

 prominences, vary in an eleven-year period; 

 synchronously on the earth there are variations 

 of the magnetic field, the aurora, the pressure, 

 temperature, rainfall, the position of the storm 

 tracks and the velocity of eastward movement; 

 also there are two barometric periods, in the 

 polar and equatorial belts, as yet entirely with- 

 out explanation. How is this sympathetic con- 

 dition to be accounted for except by solar ac- 

 tion ? I have given my explanation of what 

 the nature of the energy, is, and the experi- 

 mental evidence is decidedly on the side of my 

 statements. 



My treatment of the magnetic observations 

 has been entirely simple, following the lines of 

 Bulletin No. 2, and the process is too obvious 

 to need any defence. The result is to show 

 that the. earth is surrounded by two great sys- 

 tems of magnetic vectors {sQe Astron. and Astro- 

 Phys., No. 118, and ^mer. Journ. Science, Vol. 

 L. , August, 1895). Both indicate that the earth 

 acts like a conducting spherical shell, with its 

 magnetization at right angles to one field and 

 nearly parallel to the other field, the former 

 therefore parallel to the plane of the ecliptic, 

 and the latter at right angles to the same plane. 

 In the Meteorol. Zeitschrift, September, 1895, Dr. 

 W. van Bemmelen, in an interesting paper, 

 finds that the large magnetic disturbances enter 

 the earth along lines nearly orthogonal to the 

 auroral ovals, and gives their relation to the 

 magnetic meridians more in detail than I had 

 done. His work confirms my result published 

 two or three years earlier : ' Die Theorie und 

 Ergebnisse Bigelow^s scheinen aber ein sehr 

 beachtenswerther Versuch, dieses ganze System 

 zu erklaren,' u. s. w., p. 327. The scientific 

 character of my residuals, and my use of them 

 are similar to those commonly employed in de- 

 termining the diurnal, annual and secular vari- 

 ations of the terrestrial magnetic field, and if 

 Prof. Franklin proposes to throw mine out he 

 must needs accompany them with the entire 

 subject of terrestrial magnetism. I have merely 

 pursued my analysis into deeper conditions,. 



