Sun-spot Frequencies Contrasted. 151 



sun-spot frequency — except as regards data for recent years — 

 practically the only source available has been the data pub- 

 lished by Wolf and Wolfer, which extend back to 1719. It 

 is improbable that the unit in Wolfer's latest table * repre- 

 sents an absolutely unchanging value throughout the whole 

 period, but great care has been taken to make the table as 

 homogeneous as possible, and the epochs it gives for the 

 occurrence of sun-spot maximum and minimum are pre- 

 sumably, in at least the great majority of cases, very 

 approximately correct. 



Auroral data are exposed to many more uncertainties. 

 The observed frequency varies enormously at different parts 

 of the Earth, and the number of auroras recorded in any 

 specified area is largely dependent on the provision made 

 for observing and recording them. Conspicuous auroras are 

 unlikely to escape notice in populous countries where they 

 are rare occurrences, but in Arctic latitudes where auroras 

 are common many doubtless fail to be recorded. With 

 the increase of population and the development of means oi 

 communication characteristic of the last 100 years, there has 

 no doubt been a tendency to an increase in the proportion 

 of auroras which come to be recorded. 



Thus auroral frequency is a quantity which is certainly 

 not expressed in terms of an invariable unit ; and the various 

 tables which have been published show irregularities due to 

 temporary and local causes, whose disturbing influence it is 

 practically impossible to assess. In the following investi- 

 gations the methods adopted aim at reducing to a minimum 

 the effect of the various uncertainties. 



§ 4. The sun-spot frequencies made use of are derived 

 exclusively from Wolfer's table, which gives data for each 

 individual month during the 153 years 1749 to 1901. The 

 auroral frequencies are from two sources, viz. : " Catalog 

 der in Norwegen bis Juni 1878 beobachteten Nordlichter 

 zusammengestellt von Sophus Tiomholt, herausgegeben von 

 J. JFr. Schroetei"" (Kristiania, 1902), and Joseph Lovering's 

 "On the Periodicity of the Aurora Borealis^ (Mem. 

 American Academy, New Series, vol. x. 1808). 



Of the several tables in the former work, that employed 

 is Table E, pp. 414-417, which gives auroral frequencies 

 derived from the whole of Scandinavia from July 1761 to 

 June 1878. In this table. Schroeter has combined Tromholt's 

 results for Norway with those of Rubenson for Sweden. In 

 the original the yearly totals are for years commencing in 

 July. In order, however, to obtain results more strictly 

 * Met. Zeit. 1902, vol. xix. p. 19-5. 



