140 ROCHESTER ACADEMY OF SCIENCE. [Jan. 26, 
interval which does not differ very much, if at all, from that finally 
adopted as above stated. This periodicity is, however, subject to 
modifications which require further investigation. The chief of these 
are certain annual interruptions which are plainly apparent in the 
accompanying plate. At first it was supposed that these were due 
to the interference of moonlight, or summer twilight, as the case 
might be, preventing auroras from being seen. But the records 
of magnetic perturbation which are not affected by moonlight show the 
same thing, so that further explanation is required. 
By the identification of the disturbed portions of the sun, couceniea 
in the production of particular series of outbreaks of magnetic 
phenomena, we may learn whether cessation of solar activity attends the 
annual interruption in the recurrence of auroras to which reference has 
been made. For the purpose of such identification the surface of the 
sun was considered to have been divided meridionally into as many 
sections as there are days, and fractions of a day, required for the com- 
pletion of a single synodic revolution. Lists were then made of all 
sunspots observed on each section, together with their sizes, as deter- 
mined by the measurements made at Greenwich Observatory upon the 
photographs, taken under the auspices of the Solar Physics Committee 
of the Royal Society. The sums of the numbers, indicating the sizes 
in each of these lists for the entire period selected, show the corre- 
sponding amount of sunspot formation on each section. The numbers 
of stations reporting auroras each day during the period selected, were 
also arranged in accordance with the time of a synodic revolv “on, and 
the sums for corresponding days of all these synodic pemrds were 
obtained. By this means the portion of the sun associated with a given 
series of outbreaks of magnetic phenomena at once becomes apparent. 
Thus it appears that the areas most frequented by sunspots are most 
actively concerned in the production of auroras, having this power 
occasionally, even when spots are temporarily absent, and in any case 
manifesting it chiefly, if not exclusively, when at the eastern limb, 
appearing by rotation. As regards explanation of the interruptions at 
regular intervals of series of recurrences of auroras, it appears from the 
tabulation here described that they cannot be accounted for by cessa- 
tion of solar activity. 
This tabulation, by increasing our knowledge of the nature and 
consequence of solar activities, may contribute positively to the expla- 
nation of the interruptions of series of auroras in question. 
The association of phenomena indicated and their periodicity, 
afford positive proof that the body of the sun is a coherent mass prob- 
