standing of the water supply of the city is essential to an understanding of the outbreak, the following brief description is given.

The water supply of Watervliet is derived from a public supply pumped from the Hudson and Mohawk rivers, from 23 public wells, and from a large number of wells located on private property. Three of the latter are used so extensively by the public that they may be classed as public wells.

The public water supply is derived in largest part from the Mohawk river at Dunsbach Ferry, about sixteen miles distant, from which about two and one-half million gallons per diem are pumped into a reservoir located on a height west of the city. The investigation by the Engineering Division disclosed the fact that owing to a breakdown in the pumping station at this point it was necessary to install an auxiliary supply from the Hudson, the latter being put into operation July 24. This pumping station was located between Broadway and the river bank about midway between Fifteenth and Sixteenth streets. A sewer empties into an obsolete ferry slip about one hundred and fifty feet above the intake. Other sewers empty at varying distances above and below. A similar condition had led to installing such an auxiliary supply from the Hudson in the spring of 1916, at which time the plant was located four or five hundred feet south of the city limits a fact of some significance as the typhoid incidence at that time was in Port Schuyler or that part of the city south of the Arsenal.

The public wells are all located on the streets or sidewalks, the waste water running into the gutters or standing in puddles about the bases of the pumps. The pumps themselves are in many

instances of wooden construction and venerable aspect. The more recent installations have been iron pumps. The tops are usually fairly well protected by flags, the iron pumps being bolted to boards laid across the flags.

In most instances the city sewers, constructed of vitrified tile, follow the centers of the streets. These sewers carry both domestic sewage and storm water and the joints are supposed to be laid with cement, but in a number of instances where the sewers have been dug up, the ground about the sewer has shown evidence of leakage. Usually, too, the sewer has been found to be laid in

shale, offering an excellent opportunity for underground pollution of wells even at a considerable distance.

The wells, with the exception of the Beattie well and the well at Seventh avenue between Seventh and Eighth streets, are simply dug wells with no protection from ground water. Some of them have required some blasting, but the depth and the particulars as to the strata penetrated in their fabrication are unrecorded. The danger of pollution of such wells from leaking sewers within fifteen to fifty feet is an imminent one, and despite the fact that not all of them can be held responsible for the typhoid infection in the past, there is no reason why they may not become such a source in the future with the two exceptions noted above. The analyses of the waters from these wells show present or past pollution and their proximity to sewers renders it reasonable that it is from them they receive their contamination.

A notable exception to the proximity of sewers is found in the well at Second avenue and Thirteenth street. This well, which is held responsible for a large group of cases, is at the edge of the sidewalk about fifty feet from the canal bank and one hundred feet from Thirteenth street. There is no sewerage either on Second. avenue or Thirteenth street at this point. The nearest sewer is on Third avenue about three hundred feet distant, up a rather steep incline. A number of privies exist in this block, which is in reality located on a hillside. The storm sewer from the northern and western part of the city carrying storm water only, courses under the Erie canal at Thirteenth street about one hundred and seventy-five feet distant from the well and at a depth probably greater than the depth of the well itself. Even had this sewer carried domestic sewage, it seems impossible that it could have been the source from which this well was infected.

As to the Hudson and Mohawk supplies, both are of course from highly polluted sources and were treated with liquid chlorine, the rate of application prior to September 17, 1917, being supposedly about .33 parts per million. On this date samples taken by the Engineering Division showed a bacterial count of 66,000 in the raw water and 55,000 in the chlorinated water, with colon bacilli in all dilutions, clearly indicating that the chlorine was insufficient to produce any appreciable results in the reduction

of the bacterial content. The quantity of chlorine was increased from day to day on the basis of an examination for free chlorine, and the samples taken on the twenty-second of September gave the first satisfactory bacterial analysis.

Watervliet's death rate from typhoid fever has been consistently high as shown by the following table. The upstate rates for the corresponding years are given for comparison.

The above rates are prima facie evidence of a polluted water supply. As a matter of fact, the disease is endemic in Watervliet and almost every year rises at some time or other to epidemic proportions. The death rate is in reality much higher as a considerable number of the severe cases are taken to hospitals in Troy or Albany for treatment. The records of the registrar of vital statistics at Troy show that since 1913 the following number of deaths from typhoid have occurred in Troy among residents of Watervliet:

1914, 4; 1915, 2; 1916, 2.

Adding these, the typhoid death rates for these years are:

1914, 60.00; 1915, 53.3; 1916, 46.7

Prior to 1913, the certificates of death failed to show the usual residence of decedents and one can but surmise the actual loss of life from typhoid among residents of this community. Even this does not tell the entire story, for many residents of Troy, Albany and smaller places in the vicinity work in Watervliet and contract typhoid there.

The unusually high death rate of 1913, was due to a flood of the Hudson river which inundated a large part of the city lying east of the Erie canal and infected the wells lying in that locality. This flood also probably contributed to an increase in the infection in the general water supply itself.

Although the death rates above given are quite high and would be higher if the typhoid deaths attributable to Watervliet but occurring elsewhere were added, it is probable that even this corrected ” rate would be less than the truth, through failure to accurately diagnose atypical cases of the disease. Seemingly but few of the physicians of the city used a Widal reaction for the purpose of identifying their continued fevers and most of these failed to send more than one specimen of blood to the laboratory for this purpose. Instances were encountered of cases diagnosed as malaria, grippe, gastric ulcer, etc., which were in the immediate families of cases of typhoid and recounted histories indicating that a Widal would have been of extreme interest.

The preliminary investigation by Doctor Duryee above referred to excluded milk and any common food product and the explosive character of the outbreak eliminated flies, contact, et cetera, from consideration except as minor sources.

the

Although few cases of typhoid were reported during early September from the northern and southern parts of the city latter known as Port Schuyler- in the latter part of the month there was a considerable increase in cases occurring in the district north of Nineteenth street and as far west as Fourth avenue. The public wells were blamed in the minds of a number of people for this increased incidence but it did not seem reasonable to believe that such a considerable number of wells could be simultaneously infected with typhoid organisms. It appeared much more plausible on the face of things to attribute this increase of infection to the city water supply.

In all one hundred and sixty-nine discovered cases were investigated, not including several cases reported from Troy as having received their infection in Watervliet. An analysis of the data collected by personal visits of representatives of the State Department of Health very clearly showed:

1 The presence of some unusual condition in the neighborhood of Thirteenth street, Second and Third avenues.

2 That milk, food, flies and contact were not responsible for the outbreak.

3 That but three explanations for the outbreak could be considered as solving the problem, namely: