260 Canadian Record of Science. 



at intervals and saw Aurora from 6.20 p.m. up to 5.05 a.m. (dawn) 

 morning of the 20th. At 4.40 a.m. 20th I had the good fortune to 

 observe, under very favourable circumstances, the formation of a field 

 of cirrus clouds almost overhead. It showed the wind action I ex- 

 pected — high W. to E., under N. to S., with a probable stratum 

 almost calm in between — but what is of exceptional interest, the 

 coincident behaviour of the Aurora showed a decrease while the clouds 

 grew, and an increase while the douds again melted away. At the 

 same time the high cirrus mist which makes the stars look dim, and 

 which was evident as a large halo round the moon, diminished in 

 density and decreased in size. These changes in the cloud formation, 

 the halo, and the Aurora were coincident and apparently in reverse 

 proportion. 



A somewhat similar cloud phenomenon was seen between 8 and 

 9 a.m. Of course the Aurora was invisible, but certain heavy low- 

 lying, cumulus clouds were distinctly exhibiting electrical influences, 

 so 1 presume if the Aurora had been still visible we would have seen 

 changes in it also. I consider the observation of these clouds important, 

 so the running notes I kept will be transcribed and forwarded to the 

 Society for anyone who may be interested. It may be remarked that 

 the period between these two displays is twenty five days, roughly one 

 revolution of the sun. On the 20th of September there was no sun 

 spot visible, although a group observed two weeks later must have 

 been forming on the reverse surface. On October 14th the old group 

 of sun spots that had been exciting so much attention could be seen on 

 the Western limb of the sun's disc, while a single new spot was just 

 appearing on the Eastern side. The meredian field, however, was 

 clear. In this respect, however, I may mention that the connection of 

 sun spots and terrestial Aurora Borealis is not direct, but probably 

 through the "protuberances " which accompany the outburst of spots, 

 or at least the earlier stages of their development, the solar pro- 

 tuberances in high latitudes are said to synchronize with terrestrial 

 magnetic disturbance, or magnetic storms (Lockyer), and the latter, 

 according to weather conditions and other modifying circumstances, 

 finds a consequent expression either in Aurora or thunderstorms, or 

 simply the unsettled state of the magnetic needle, indicating disturbed 

 earth currents. 



A great magnetic storm with displays of Aurora took place on the 

 two nights, October 30th and 31st and November 1st. It was supposed 

 to have originated in the sun, or at least to have had some connection 

 with two large groups of sun spots then attracting attention. Tele- 

 graphic communication was generally interrupted, but in some instances 

 lines were worked without battery power, as was the case between Mon- 

 treal and Fort William Friday night. James Kent, Esq., of the C.P.R. 

 Telegraph Company, gives me the information that trouble was first 

 experienced about midnight on the Winnipeg circuits, rapidly extended 



