576 REPORT— 1901. 



resolving power, or purity of the best spectrum that can be obtained when the 

 specti-ometer is employed in its usual way, i.e., with the image of the sun thrown 

 on the slit. No doubt, the light being now derived from a large extent of the 

 sun's disc, sharp lines must have been fringed with faint and narrow wings owing 

 to the rotation of the sun ; but the wings were too faint and too narrow to be 

 visible in the second spectrum.] 



On the whole, the verification of the efi^ect predicted by the new analysis 

 appears to be satisfactory. 



A modification of the experiment can be made in the absence of sunshine by 

 throwing the image of a flat sodium flame upon the grating, when the D lines will 

 be seen beautifully defined, and may be reversed if suitable arrangements are made in 

 the flame. But a sodium flame cannot be made truly flat or truly steady so as to 

 furnish an image the purity of which may be relied on like that of the sun. The 

 solar arrangement for making the experiment is therefore to be preferred when 

 sunshine and sufficiently good apparatus are available. 



8. A Long Period Solar Variation} By William J. S. Lockyer, 



This paper consists of a discussion of the observations of the measurement of 

 sunspot areas made since the year 1833, this year being the epoch when Schwabe 

 commenced his series of sunspot observations on a systematic basis. The actual 

 dates of the epochs of maxima and minima of sunspot area used in this investi- 

 gation were those given by Dr. Wolf and Dr. Wolfer. As a check on the work 

 the important results of Mr. William Ellis' discussion of the Greenwich 

 Observations of the Magnetic Elements were utilised, as he has shown that the 

 curves representing the magnetic elements are in almost exact accord with that 

 representing the solar spotted area. 



, In dealing with the sunspot curve the first result of the investigation was 

 to indicate that the intervals between a minimum and a following maximum 

 varied regularly, the length of this period of variation amounting to a little more 

 than three eleven-year periods, or about thirty-five years. The magnetic curves 

 examined in the same way indicated precisely a similar variation. 



An inquiry into the amount of spotted area included in each interval between 

 consecutive sunspot minima indicated also a regular variation, the period being 

 similar to that mentioned above — namely, about thirty-five years. 



Further, it was found that the interval in time between consecutive minim a 

 was not constant but varied, as far as could be judged, regularly, the length of 

 the period increasing and decreasing in alternate eleven-year periods from a mean 

 value. 



The paper then indicated that as the sun may be considered as a ' variable ' 

 star, it may be likened to the well-known variable r; Aquilte, the light of which 

 changes rather similarly — /,e., the interval between a minimum and a following 

 maximum has a short-period variability, and the period from minimum to minimum 

 alters. 



In conclusion the author referred to the important work of Professor Ed. 

 Bruckner, who had indicated that the changes of climate were periodical, and 

 that the mean length of the period was about thirty-five years ; to Mr. Charles 

 Egeson's investigations on territorial meteorology for South Australia ; and to 

 Professor Ed. Richter's results on his researches on the movements of glaciers. 



All the.«e investigations indicated clearly a periodical change in the meteorology 

 of the earth's atmosphere, which were the result of this thirty-five yearly solar 

 period, as shown by the correspondence of the respective epochs. 



The paper then indicated that the next 'great ' maximum of sunspots, similar 

 to that of 1870 and 1835, should occur at the approaching maximum, and it 

 would be interesting to see whether all the solar, meteorological, and magnetic 

 phenomena of those two periods were repeated. 



' See Proc. Royal Soe. vol. Ixviii. p. 285. 



