JAITOABT 13, 1911] 



SCIENCE 



43 



as an exceedingly special and restricted one 

 and perhaps even question its right to be 

 classed among the physical sciences. And 

 yet this subject of the "earth's magnet- 

 ism," which may consider itself fortimate 

 if a brief section is devoted to it in the 

 average text-book of physics, is itself so 

 broad and so extensive that I know of no 

 single investigator who could to-day be re- 

 garded as equally eminent in all its various 

 branches and siib-branches. To be a 

 master in terrestrial magnetic research re- 

 quires the most intimate knowledge of sev- 

 eral of the so-called fundamental sciences, 

 especially of mathematics, physics and 

 geology, with some knowledge of astron- 

 omy, meteorology, chemistry and geography 

 besides. "The field of investigation," says 

 Maxwell, ' ' into which we are introduced by 

 the study of terrestrial magnetism, is as 

 profound as it is extensive. ' ' 



MAGNETIC DISTURBANCES 



Instead of beginning with the phenom- 

 ena usually chosen to illustrate the earth's 

 magnetism, let us begin with one of the ir- 

 regular, more or less spasmodic manifesta- 

 tions — one of the so-called "abnormal 

 features" — the earth's "magnetic storms" 

 as they were termed by Humboldt. Be- 

 cause of the troublesome nature of mag- 

 netic disturbances, when one is dealing 

 with a phenomenon like the diurnal varia- 

 tion, for example, various magneticians for 

 nearly three quarters of a century have 

 been seeking some logical method of decid- 

 ing just at what point a disturbance begins 

 — in brief, how large must be the fluctua- 

 tion in a magnetic element to be classed 

 and eliminated as a "disturbance." 

 Various rules have been set up, but none 

 has found general acceptance. 



There was a time when it was thought 

 that, by mere inspection of the photo- 

 graphic record of the variations of a mag- 



netic element during the day, it was pos- 

 sible to say whether that particular day 

 was disturbed or not ; and so arose in Great 

 Britain, for example, what are called the 

 ' ' astronomer royal 's five quiet days during 

 the month." Through Airy's initiative, 

 the first English magnetic observatory was 

 established at the Greenwich Observatory 

 during that period of intense magnetic ac- 

 tivity which prevailed in the first half of 

 the last century — during the days of Hum- 

 boldt and Gauss. The honor of deciding 

 on the five quietest days in any particular 

 month from which the "normal" diurnal 

 variation may be deduced, has therefore 

 been accorded to the head of the Greenwich 

 Observatory; these days are selected by 

 him, or more likely by his magnetic assist- 

 ant, from a mere inspection of the photo- 

 magnetograms. However, it has been 

 found that these supposedly quiet days 

 may themselves be subject to a more or less 

 constant disturbance which, prevailing 

 throughout the day, serves to elevate or 

 depress the entire curve and hence does 

 not reveal itself in superposed serrations 

 or "eruptions." Or the disturbances may 

 follow somewhat the same course as the 

 diurnal variation itself, and hence again 

 not be revealed by mere inspection, but 

 serve mainly to increase or decrease the 

 diurnal range. 



Van Bemmelen and Ghree have also 

 found that these supposedly quiet days are 

 subject to a non-cyclic effect similar to that 

 shown by other days. I recall that during 

 my first magnetic survey — that of Mary- 

 land, 1896-99 — there were months when a 

 searching examination of the photographic 

 records of the magnetic observatories 

 showed that there was hardly a day out of 

 a whole month without some kind of irregu- 

 larity. 



It is then evident that a phenomenon 

 which occurs so frequently and which in 



