188 H.. A. Hazen— Tornadoes. 
ded thunder-storm of this year in the eastern United States. 
This occurred February 14, 1884, and was first noted at 
Augusta, Ga., early inthe morning. It appeared at hae it 
Va., at 4 A. M., and at Variety Mills, Va., about the same hour, 
where it seemed to come from southwest and passed “off to east- 
northeast. ioe ore Washington, . from west-south- 
west at 6.25 A. M. toward east- northeast. It was noted at 
Haverford Gcllene, ot and at New Castle, Del., at 8.30 a. M. 
ese times (about some of which there is a little uncertainty) 
give the following velocity: Lynchburg to Washington, about 
sixty-two miles per hour; Washington to Haverford College, 
counting the seconds between the flash of lightning and the 
thunder, on the west as it came up and on the east as it passed 
off, was a little less than a mile a minute. 
may not be that one and the same storm fete across 
thout a storm-center or disturbed region lg mewhat 
ties it. This center may not. make itself kno ag etek 
: 
thunder-storms into two classes—first, those caused in the 
presence of a low center of pressure, and, second, those due 
to the sun’s heat. It may be admitted that the sun’s direct 
heat is an important factor in the formation of summer thunder- 
storms, but this only acts in 22 ages with the disturbed 
region which is invariably prese 
If, now, ape a thunder-storm ee is general, gon is dene 
ent upon the presence of a storm-center, may we not hope by 
careful observation at short intervals of time and space, in the 
region over which the tornado action is expected to occur, to 
obtain a sounder basis of fact to reason upon than we now 
mere 
tep further we may venture. If we can correlate 
thandergirienn and storm-centers, may we not expect to obtain 
by a critical study of the former some. evidence of the forces 
acting in producing the latter? By such critical study we 
ought to be able to obtain —— to ihe following, among 
the questions ae to tornadoe 
t what distances and in what Se do they lie from the 
low center? How are they distributed with respect to thun- 
re 
