The “Plague” at the Chicago 1933 Century of Progress

The 1893 the World’s Columbian Exposition in Chicago had its serial killer, H.H. Holmes, made famous as “The Devil in the White City.” The actual extent of Holmes’s crime spree is probably far smaller than the legend and the book suggest, but popular imagination keeps the legend alive. Most people don’t know that 40 years later, the 1933 Century of Progress, the next World’s Fair in Chicago, had its own actual killer, far more deadly than Holmes.

In the summer and fall of 1933 an outbreak of illness at the Century of Progress sickened, at a minimum, 1,400 people, with 98 resultant known deaths. Most of the cases had a common thread: people who got sick had stayed at or visited one or the other of two hotels, the Auditorium Hotel or the Congress Plaza Hotel, across the street from one another.

The Century of Progress was a bit unusual in that it had two “seasons,” one in 1933 and the other in 1934. Its first season, from May 27 to November 1, 1933, had 22,565,859 visitors. An estimated 8,500,000 of those visitors were from out of town. About 160,000 of them had contact with the two hotels in question, over half of them as registered guests at the hotels.

By mid-August, 1933, health authorities had identified the disease as amoebic dysentery. Not only that, suspicion arose that something in one or both of the hotels was causing the outbreak. But getting to where it was coming from was made all the more difficult by the nature of the population at the fair, a mix of locals and visitors from elsewhere.

The story of solving the mystery of this epidemic is a remarkable one. In fact, it’s really two stories: that of medical investigators determining the cause of the disease, and that of engineers tracking down the infrastructure problem that allowed the disease to spread.

THE MEDICAL CASE

In the period from June 1, 1933 to June 30, 1934, 1,409 cases of illness that were ultimately related to the Century of Progress had been recorded. According to a report by the U.S. Public Health Service, “slightly more than two-thirds of those infected were out-of-town visitors to Chicago. Approximately 75 percent had had contact with one or both of the hotels, but less than one-half of the infected Chicago residents reported such contacts.” (1)

That this was a single epidemic disease was the first thing that had to be agreed upon. That it was, as it turned out, amoebic (referred to in reports as “amebic”) dysentery was quite surprising. As the very first point in the U.S. Health Service’s report summary points out, “An epidemic of amebic dysentery had its origin in Chicago during the summer and fall of 1933. It was the first recognized water-borne outbreak, and the only known extensive epidemic of this disease in a civilian population” (italics mine). This meant civilian populations anywhere. (2)

The out-of-town visitors usually became ill after returning home, a fact that masked recognition of the epidemic, at least at first. It’s also highly likely that a significant number of illnesses associated with the epidemic never were reported, making the final estimate of 1,409 cases quite low. In any event, the amoebic dysentery attributed to the fair was reported “from over 400 cities scattered in 43 states, the Territory of Hawaii, and three Canadian Provinces” according to the U.S Health Service’s report, which examines the history, epidemiology, and physical causation of the outbreak in great detail. (3)

It is a record of a medical scientific mystery solved by persistent and excellent detective work. Investigators were very careful to query health departments in other cities and states to determine if identifiable cases of amoebic dysentery had been found and to link those cases with visits to the Century of Progress. To confirm possible cases, they sought histories in which stool samples in affected people showed the amoebic cysts typical of infection by entamoeba histolytica, the organism that causes the disease. Only a few cases were included in the report data in which such clinical evidence was not presented. They were accepted because in all other respects the disease the victims had was consistent with amoebic dysentery.

Where was it coming from?

The possibility that one of the hotels was a hot spot for the growing number of cases of illness arose in August 1933 at about the same time the disease was first identified. Simultaneous reports of two cases involving registered guests at that hotel raised suspicions. “By the end of the month 7 others had been reported. In September 2 more involving guests, 1 from each of the hotels, came to light. In addition, an intensive study of employees of one hotel revealed several (11 officially reported) infections in that group.” (4)

Apparently, substantial effort was expended looking for the “focus” of the disease in the hope that a single location could be identified. As it turned out, the focal point was both the Congress Plaza Hotel and the Auditorium Hotel. “Only one focus was discovered which accounted for any considerable number of cases, namely the two hotels. The similarity of the epidemic period to that of A Century for Progress Exposition suggested that the epidemic was related to conditions at the exposition. . . . The general incidence of infection was higher in one of the hotels than in the other and markedly so for the early weeks of the epidemic.” (5)

Chicago’s Water Supply in 1933

By the time of the Century of Progress, Chicago had come a long way from the contaminated water supply of most of the 19th century. The source of water for the city, Lake Michigan, was (and had been for a long time) no longer also its sewer. However, there were, at the time of the exposition, some differences in the sanitary quality of water coming from the different pumping stations. As the engineering section of the U.S. Health Service’s report on the epidemic says:

While normally the supply of water to hotels X [Congress Plaza] and Z [Auditorium] was from the Fourteenth Street pumping station, there is a possibility that at times these hotels might have been supplied with water from the Chicago Avenue pumping station.

The water pumped from this station comes from Lake Michigan, most of it through the Two Mile intakes. Because of the shorter distance from shore, the lake water at these intakes is generally not as satisfactory from a sanitary standpoint as that at the Four Mile intakes. However the daily bacterial examinations of the samples of chlorinated water indicate that this water throughout the year of 1933 was consistently of excellent quality. (6)

We in our day might think that water chlorination was a later development. However, it was practiced in Chicago starting in 1916, when a modern liquid anhydrous chlorine system was adopted for use at all of the city’s water pumping stations. (7)

THE QUEST OF THE ENGINEERS

The search for the source of the disease affecting so many people was thorough, and the engineers and public health authorities concluded that the source was not the city water supply. They then had to look elsewhere.

The engineers looked at all of the internal water systems of both hotels. As described in their report, in the course of this part of the investigation, the engineers identified a separate system in the Congress Plaza “for distributing cooled drinking water” in the hotel. “When guests called for ice water it was usually obtained from this system, either from one of the faucets on various floors of the buildings or from outlets in the room service department in the basement.” (8) The system included a cooling tank, from which all cooled water for guests, and for table service in the restaurants and pantries was drawn.

An electric motor drove the pump that distributed the water from the tank.

Directly over the motor there was found, on February 8, 1834, a wooden plug which had been driven into a hole in the cleanout ‘Y’ at the foot of a soil stack. The condition of the plug indicated that there had been leakage from the sewer [line] at this point. If this leakage occurred in appreciable quantities, it could have entered the cooling tank and polluted the drinking water in [the Congress Plaza Hotel]. It is our opinion that this did occur during the period of the 1933 amebic [sic] dysentery outbreak. (9)

The report goes on to say that the tank which held the cooled water was

made of riveted steel . . . supported by a wooden base a few inches from the floor in a small room in the basement of the [Congress Plaza Hotel]. . . . There was an opening of about 2 feet square in the top of the tank for inspection purposes. The cover over this opening was not tight-fitting and leakage into the top of the tank near this opening could drain into the tank. The wood of the cover was deteriorated considerably. (10)

According to the report,

In endeavoring to rod out the sewer through the cleanout “Y” a workman [quite some time before the fair] probably was unable to remove the cap and therefore made a hole in the center of its face. Instead of properly sealing the hole by inserting a threaded metal plug, he closed the opening by driving in the wooden one. How long this plug had been in this sewer is unknown. . . . The plug when found was so loose that it was easily removed by the finger tips. It was badly deteriorated by decay which extended both inside and outside the pipe on the lower side. . . .When removed, there was a noticeable black crust around the plug at the rust stain which marked the ring of contact between the plug and the iron. This crust appeared to be dried sewage solids which had accumulated on the plug as the liquid escaped through the leak. (11)

While this part of the system belonged to the Congress Plaza Hotel, something clearly was affecting both hotels. As the engineering report notes, “The epidemiological evidence shows a marked parallelism as to time of case infections among registered guests in both hotels . . . especially after July 17 [1933]; hence one would suspect the same or closely related sources of infection for both hotels.” (12)

The culprit, it turned out, was the house water system, which included the cold water distribution from the tank below the sewer pipe. “The house water system was common to both hotels and had the same sources of potential pollution at the condenser cross-connections. There is much presumptive evidence that one or both of these cross-connections was used; also that the valves in the two cross-connections were not tight.” (13)

Up until July 16, 1933, the very large preponderance of infections occurred among guests at the Congress Plaza Hotel. The first days when simultaneous infections occurred appear to have been July 17-19, but the cases were few. However, August changed things. “The next sharp rise in infections among [Congress Plaza Hotel] guests, and the one which marked the beginning of the major period of infection in both hotels for the entire epidemic, began on August 13.” The periods of rising infection in most instances coincided with the loss of use of one of three major water condensers for the hotels. Between August 17 and August 23, the affected condenser was shut down entirely, and “infections in both hotels rose rapidly, attaining, on August 22, the largest number of reported infections on any one day of the epidemic.” (14) Once the condenser came on line again, infection frequency decreased. Each time it stopped, infection rates increased.

The strong likelihood is that when the errant condenser was running, it helped flush water through the waste system fast enough and in enough quantity that the leaking waste pipe was not contaminated.

But later surges in infection suggested that in fact the sewer pipe was often overloaded when the need for the condenser was less, especially as weather cooled.

Peak periods of infection occurred around the 17th of September and the 3d [sic] and 11th of October. It should be noted that the latter two peak period of infection bear no special relationship to the quantity of water used in the condensers, for both of these October peaks occurred during periods when the water consumption for condensing purpose[s] was low. It seems to us that the extent of overloading sewer no. 2, which in turn is influenced by the number of registered guests in the hotel, became a more important influence in the pollution of the condenser water, and the operation of condenser C less important, as cooler weather came. Thus the fluctuations in infection rates in the hotels during October bear a resemblance to those in the numbers of guests. (15)

Infections continued to occur after that, but at a much reduced rate. “After the peak of October 11 the guest infections at [Congress Plaza Hotel] declined rapidly. Cases were reported among registered guests at both hotels until November 6, after which the outbreak in epidemic proportions subsided.” (16)

Hindsights: Explaining the June and July 1933 Outbreaks

In late June and early July, 1933, several things conspired to create the environment for infection, particularly at the Congress Plaza, mostly influenced by a substantial increase in the amount of leakage from the sewer over the drinking water: a rapid increase in the number of registered guests at the hotel; extremely warm weather; and excessive rainfalls on June 29 and July 2.

On June 1, the number of registered guests at the Congress Plaza was 361; by the 28th of the month the number was 1,044. This high occupancy came with very hot weather during the latter part of June. The engineering report states that on June 29, “conditions at [the Congress Plaza] and in Chicago were such as to permit gross pollution of the cooled drinking water. The hot weather, which had persisted for several preceding days, was featured, on this day, by a rainstorm of unusual character. Heavy rainfall began at 5:10 p.m. and continued until 8:20 p.m. central standard time, the total precipitation amounting to 1.56 inches.” Much of the rain fell at dinner time at the hotels and “caused serious flooding of sewers in the downtown section of Chicago.” (17)

Because of the time of day, “heavy discharge from the gravity sewer systems of buildings” and from fixtures in the Congress Plaza would be expected. “This would place sewer no 2 of [the Congress Plaza] under pressure, even without a storm. This, coupled with the simultaneous flooding of the street sewer by the storm, probably caused serious leakage from the branch sewer over the drinking-water tank around the wooden plug.” Later tests proved that the likely leakage from the pipe over the drinking water tank during and after the storm was “about two-tenths of a gallon per minute.” The report states that during the very heavy rainfall of 47 minutes, “the amount of freshly infected sewage water which could have drained into the cooled-water tank at [the Congress Plaza] . . . would have been 9.4 gallons.” Further, “owing to the probable heavy use of this cooled water at mealtime a large percentage of this contaminated water might have been consumed during that evening. Lumbermen constituted a special group of registered guests at [the Congress Plaza] when this heavy rainfall occurred. Among them, 20 cases of amebic dysentery were reported, including 5 deaths. Most of them checked out of the hotel on July 1.” (18)

Interestingly, another powerful storm, with very heavy rain, occurred in the early morning of July 2. The occupancy of the hotel was quite similar, and the 20 rooms in the Congress Plaza served by the sewer over the cooled drinking-water tank in question were occupied, as before. But only one case of amebic dysentery was reported for hotel guests from this period. The rainfall was “practically the same as that of June 29,” and caused worse flooding. However, “service activities in the hotel would be at a low point at this hour of the morning, when almost all registered guests would be sleeping. . . .The rate of discharge from fixtures in the hotel was undoubtedly low.” (19)

A “Providential” Breakdown

A disaster occurred during the July 2 storm that, contrary to what one might expect, kept the drinking water in the hotels from become even more polluted. A major street sewer line broke. It overloaded a very old, fragile house sewer in the Congress Plaza, breaking it in turn and flooding the ice storage room and the laundry, bakery, pantries, ice boxes, and other sections of food storage and preparation with sewage to a depth of 6 inches. These areas were, of course, taken off line and totally disinfected.

The report tells us that “the break in the sewer, while causing serious contamination of the hotel basement, seems providential in preventing a second gross contamination of the cooled drinking water, under conditions similar to those which existed on June 29. With the rupture of the two hotel sewers and the flow into the hotel basement, the city sewer was relieved of pressure. This . . permitted better flow and little or no pressure in the hotel sewers.” (20)

The Little Wooden Plug

That workman of the past who years before had hammered the wooden plug into the hole in the lateral sewer over the cooling tank had, of course, no idea that his makeshift repair would have such dire consequences. Yet, in a certain sense, for this World’s Fair, he was (unintentionally, of course) just as deadly as the notorious H. H. Holmes, Erik Larson’s “Devil in the White City,” who killed dozens of people during the 1893 Columbian Exposition.

Reports of amoebic dysentery continued to come in until the end of June, 1934. It’s not clear from the report if infection continued until the end of the Century of Progress in October, 1934, but the number of cases declined substantially after the plumbing defects in the cold water system were discovered in February 1934 and repaired. This is not to say that all sanitary issues were resolved by the end of the Fair, but the issues that occasioned an epidemic of such proportions were addressed such that the disease abated considerably.

Thanks for your patience in reading this account, if you got this far. I searched extensively for a narrative on this extraordinary story but did not find one, other than the extensive report I’ve cited, which is not very likely read by anyone (except me) in our time. I felt that the story deserved the detail that tells and completes it. The saga is, to me, an exemplar of cooperation and persistence by a remarkable team of investigators from the Board of Health of the City of Chicago, the city’s Division of Water Purification engineers, the U.S. Public Health Service, and the Department of Preventive Medicine of the State University of Iowa.

_____________________________________________

Special thanks to Roberta Molyneux, who has a strong interest in the history of the Century of Progress. She alerted me to the Fair’s disease outbreak.

Thanks also to Judith McDevitt, the professional copy editor in my life (before she became a medical professional). She carefully and quite thoroughly edited and critiqued this piece. Twice. And thanks to Burton Andersen, MD, for his thoughtful comments on the article. Burt, an Infectious Disease expert, cautioned me about several things in the article. He noted, for one thing, that calling it a “plague” might not be appropriate, since Plague is an entirely different disease. I concede the point, but keep the title because that’s how what little popular press there has been on the epidemic has referred to it. I hope the quotation marks around the term sufficiently qualify it.

NOTES

(1) “Epidemic Amebic Dysentery: The Chicago Outbreak of 1933.” U.S. Treasury Department/Public Health Service, National Institute of Health Bulletin No. 166, March 1936, p. vii.
Viewed at http://www.idaillinois.org/cdm/ref/collection/isl/id/24653
(2) Ibid.
(3) Ibid, p. vii
(4) Ibid, pp. vii-viii
(5) Ibid, p. viii
(6) Ibid, p. 54
(7) John Ericson, City Engineer, City of Chicago, “Chlorination of Chicago’s Water Supply,” read before the Sanitary Engineering Section, American Public Health Association, December 9-12, 1918, at Chicago, Ill.
https://ajph.aphapublications.org/doi/pdf/10.2105/AJPH.8.10.772
(8) U.S. Public Health Service report cited, p. 56
(9) Ibid, p. 57.
(10) Ibid, p. 57
(11) Ibid. p. 73
(12) Ibid, p. 78
(13) Ibid, p. 78
(14) Ibid, pp. 80-81
(15) Ibid, p. 81
(16) Ibid, p. 81
(17) Ibid, p. 83
(18) Ibid, p. 84
(19) Ibid, p, 85
(20) Ibid, p. 87

© Edmund J. McDevitt
May, 2019