which according to Walter F. Wilcox* of Cornell University, means a saving of something like 46,000 lives each year in that city alone.

The mortality in the registration area in the United States has been reduced since 1890 from 19.6 to 16.2 per 1,000 in 1905, which means a saving of over 290,000 lives a year.

It has long since been known that rivers are always purer near their source; the amount of impurities increases as we descend the stream, since the water courses are the natural drainage channels of the country and the wastes of human life and occupations find their way into the streams. It is also well known that our large American rivers are the sewers and at the same time the source of water supply for nearly all the cities located on their banks. These cities show, moreover, a marked prevalence of typhoid fever, thus confirming what has been observed over and over again, that this disease, as also cholera,.dysentery and diarrhoeal discases can be carried from one town or city to another by means of inland waterways. Indeed the question is one of extreme interest even to the residents along the Great Lakes; we know that large cities like Buffalo, Erie, Cleveland, Detroit and Milwaukee discharge their sewage into the Lakes, and we also know how Chicago and Cleveland suffered from typhoid-fever visitations by contaminating their own water supplies. It is also a well-known fact that many of the river cities were obliged to resort to purification of their water supplies in order to arrest the ever increasing typhoid-fever wave.

INFLUENCE OF WATER SUPPLIES UPON TYPHOID-FEVER DEATH RATES.

For the purpose of determining the influence of public water supplies on the typhoid-fever death-rates in general, Mr. M. O. Leighton, chief of the Water Resources Branch of the United States Geological Survey, very courteously complied with my request for a list of the principal American cities with a population of over 30,000, classified according to their water supply. Dr. Cressy L. Wilbur, chief Statistician of Vital Statistics, Bureau of the Census, with equal promptness and accuracy has computed the mean rate (not the average annual rate, which, however, differs only slightly for the five years 1902 to 1906), and has arranged them in the diagram, Chart I. The statistics in spite of the many factors concèrned in the dissemination of typhoid fever conclusively show that the water supply plays the most important rôle in the spread of the disease. A summary of the typhoid-fever death-rate is here given:

MEAN TYPHOID FEVER DEATH RATE FROM 1902 тo 1906 PER 100,000 OF POPULATION.

4 cities using ground water from large wells..

18 cities using impounded and conserved rivers or streams.

8 cities using water from the small lakes..

7 cities using water from the Great Lakes.

18.1

18.5

19.3

32.8

45.7

61.1

*Monthly Bulletin, New York State Department of Health, March, 1908.

5 cities using both surface and underground water.

19 cities using polluted river water.....

The rates for cities using unpurified river water fluctuate from 33.1 at Minneapolis, to 122.1 at Allegheny and even 133.1 at Pittsburg.

GENERAL MOVEMENT OF TYPHOID FEVER.

Chart II. illustrates the general movement of typhoid fever in different countries and cities, showing percentage of decrease from first to last period shown. The period covered by Dr. Wilbur is (as nearly as convenient) the last quarter of a century, and the rates are usually given for the successive five-year periods, beginning with 1881. The table shows that during the last twenty-five years the death-rate from typhoid fever has fallen in those fourteen countries and cities from an average of 42.3 to 18.1 per 100,000, a reduction of 54.3 per cent. A more striking reduction could have been made, if statistics going back as far as 1870 had been included. The typhoid rate in Berlin in 1872, at a time when that city was riddled with cesspools and supplied with polluted water, was as high as 142 per 100,000. On account of the incomplete mortality returns everywhere prior to 1881, we have deemed it best to exclude all older foreign statistics, and for similar reasons Dr. Wilbur begins his statistics for the United States with 1890.* We have likewise excluded Mr. Whipple's statistics, which tend to show that the death-rate from typhoid fever in twelve states, including all of the New England States, New York, New Jersey, Maryland, California, Minnesota and Michigan, has fallen from 55 in 1880 to 21 per 100,000 in 1905.

THE HYGIENIC VALUE OF PURE WATER: ANNUAL COST OF TYPHOID FEVER

IN THE UNITED STATES.

According to the census of 1900 there were 35,379 deaths from typhoid fever during the census year throughout the United States; and based on an estimated mortality of ten per cent it is within reason to assume a yearly prevalence of 353,790 cases of this disease. If we calculate the average cost for care, treatment and loss of work to be $300 and the average value of a human life at $5,000, we have a total loss in the United States of $283,032,000 from one of the so-called preventable diseases. Mr. George C. Whipplet presents some striking evidence to indicate that a

*Dr. Wilbur obtained the foreign statistics from data compiled from the International figures given in the last report of the Registrar General of England and Wales, from which report the rates for London are also taken. The rates for the cities of Paris and Berlin are given in the "Annuaire Statistique" of the city of Paris for the year 1904, and the rate of Berlin for the period 1901-1905 was supplied by Dr. Wilbur from data in his office. Dr. Wilbur laments the fact that "it is even now difficult to obtain a satisfactory statement of the number of deaths from such an important disease as typhoid fever in certain foreign countries and the difficulty of securing comparative data increases as we go back. The disease was first accurately compiled by the Registrar of England in 1869."

†The Value of Pure Water, New York, 1905, p. 5.

CHANGE IN WATER SUPPLY

I-FROM UNFILTERED RIVER SUPPLY TO FILTERED RIVER SUPPLY
2-FROM UNFILTERED RIVER SUPPLY TO WELLS

3-FROM POLLUTED RIVER SUPPLY TO CONSERVED RIVER SUPPLY

THE NORRIS PETERS CO, WASHINGTON, &

loss of $10,000 for every death from typhoid fever is a conservative estimate, in which case the decrease in the "vital assets" during the census year of 1900, would amount to $353,790,000. Reduce the prevalence of the disease one half (which has been accomplished in Europe and our own country), and the question of the hygienic value of pure water will be answered from an economic point of view.

THE EFFECT OF IMPROVED WATER SUPPLY ON TYPHOID-FEVER DEATH RATES.

Chart III. shows clearly the effect of change in water supply on typhoidfever death-rates in seven American cities. Dr. Wilbur of the Bureau of the Census has given the death-rate for a considerable time before and after the date of change and also the average annual death-rate before and after purification, and the percentage of reduction. From this table we learn that the combined average annual death-rate from typhoid fever in cities with a contaminated supply was 69.4 and after the substitution of a pure supply it fell to 19.8 per 100,000; a reduction of 70.5 per cent.

The Bulletin for the month of April, 1908, of the New York State Department of Health contains an interesting article showing that the deathrate from typhoid fever in ten cities of that state has been reduced 53.4 per cent by improved water supplies.

It may be urged that improved methods of medical treatment are responsible for a considerable reduction in the death-rates from typhoid fever, but when we see such a striking change immediately after the installation of filtration plants as in the case of the American cities shown in Chart III. and also more recently in Cincinnati and Philadelphia, we are forced to the conclusion that water purification plays the most important rôle by diminishing primarily the number of cases. It should be stated, however, that the effects are still more marked when combined with a good system of sewerage. The history of every sewered town shows a lessening of the typhoid death-rate and that the typhoid rate is always higher in sections of the same city supplied with makeshifts. The writer in 1895 pointed out that typhoid fever prevailed in Washington in one of 81 houses supplied with privies, and only one in 149 of those connected with sewers, and offered as the only reasonable explanation, that the sewers carry away the filth and germs that otherwise would contaminate the soil and ground water, but even if there were no wells these makeshifts are still a source of danger in so far as they favor the transmission of the infection by means of flies,* nor can the possibility be ignored that the germs in leaky or overflowing boxes may reach the upper layer of the soil and with pulverized dust gain access to the system.

The writer believes that about 80 per cent of the cases of typhoid fever

The agency of flies in the transmission of typhoid-fever germs was first pointed out by Dr. Kober in the Report of the Health Officer of the District of Columbia in 1895.