Terry’s Take: The Mattoon/Charleston tornado of 1917
The 3rd deadliest tornado disaster in Illinois history occurred on May 26, 1917. On that date 96 years ago, 101 people were killed across central Illinois, most of them in the Mattoon/Charleston area. Injuries were totaled in the hundreds with an approximate number of 638.
Originally, this was believed to be a single tornado with a path 293 miles long, extending from the Mississippi River near Louisiana, MO, all the way across central portions of Illinois and Indiana, ending near Mount Vernon, IN. The storm was on the ground for a total span of 7 hours 20 minutes. It was later determined that the event consisted of 4 to 8 separate tornadoes with the Illinois tornado approximately covering 155 miles (including times it was aloft), over a time of approximately 4 hours. The strongest part of the tornado, through Mattoon/Charleston, was later determined to be F4 intensity on the original Fujita scale.
In the days before areal photography and aircraft surveillance, it was not unusual for tornadoes to be reported to have long tracks, on the assumption that the tornado remained on the ground constantly. Minor directional changes were also not well documented, with many tornadoes assumed to have traveled in a straight line.
The first touchdown was shortly before noon on May 26 near Pleasant Hill, located in far southern Pike County. Three people were injured in Nebo, a few miles east, around noon. It then continued eastward.
Originally, it was believed this same tornado curved southeast and went through Clark County before moving into Indiana south of Terre Haute. However, it was later determined the original tornado curved northeast and ended near the Embarras River; a different tornado touched down 5 miles east-southeast of Charleston and moved southeast to Marshall, causing F3 intensity damage just north of Westfield.
The Mattoon/Charleston portion of the tornado track was surveyed by Clarence J. Root, the meteorologist in charge of the Weather Bureau office in Springfield. According to his report from the May 1917 edition of Climatological Data, the appearance of the tornado(es) changed over the course of the track:
Across the State from the Mississippi River almost to Mattoon, all eye witnesses agree that the storm had the typical funnel-shaped tornado cloud with the swinging tail, and east of Charleston the same type of cloud was reported, but the writer, who visited Mattoon and Charleston, failed to find anyone in those cities who saw a funnel-shaped cloud. Eye witnesses who were near the edge of the city, and had an unobstructed view, agree that the approaching storm appeared as a low, boiling mass of clouds, one part a little to the north and the other a little to the south. The parts seemed to roll toward one another, coming together and downward like the meshing of a pair of cog-wheels. There was an abundance of evidence of true tornadic action, … it might be suggested that the cloud was so low that there was no room for the usual pendant portion.
The tornado passed a mile north of the State Normal School (now Eastern Illinois University) in Charleston. There, barograph traces indicated the barometric pressure fell 0.27 inches, then immediately rose 0.40 inches, as the tornado passed. Winds were estimated to reach 80 mph, and hail of 2-1/4 inches in diameter was measured.
In early days, it was believed that tornado damage was caused by buildings “exploding” from the contrast in pressure. J.P. Carey, of the Department of Geography at the school, wrote the following in the June 15, 1917 edition of the journal Science:
The reason for the location of the area of complete devastation being to the right of the center seems to be plausibly explained when the agents of destruction are considered. On the right of the center there is the explosive action due to the reduced pressure on the outside of the buildings, the eastward component of the counter-clockwise wind of the tornado (probably over 400 miles per hour), the forward movement of the storm, and the west wind which was prevalent at that time all working in conjunction as agents of destruction…
Of course, later research determined that the wind speed in tornadoes does not reach as high as 400 mph, but much was unknown about tornadoes and meteorology in general at the time.
The destruction in Mattoon and Charleston was overwhelming. In Mattoon, nearly 500 houses were demolished. The survey results indicated that this particular part of town, where the worst of the damage occurred, consisted of cottages or small dwellings with no basements. To the north of the worst damage area, two manufacturing plants, a school, and a couple stores were badly damaged. In the areas closer to the business district, damage was confined to broken windows and tree limbs. In Charleston, the path was closer to the business district, with 15 businesses badly damaged, and two lumber yards and 2 railway stations completely destroyed; the high school, power station, and gas reservoir were badly damaged. Over 220 homes were destroyed as well. J.P. Carey reported that in both cities, the parts of town with the worst damage “were more completely demolished than if a gigantic roller had passed over them, for the buildings were broken into short sticks, split into narrow pieces, and some parts carried rods and even miles eastward.”