Total places supplied. Forty states and the District of Columbia reported a total of 7651 places having water supplies, most of the figures apparently being for some date in 1924 but some as early as the close of 1922. These states had 2698 places supplied with water at or about the close of 1896. The gain in places supplied from 1896 to 1924 (approximately) was 4953, or over 180 per cent. The states not reporting in 1924, with the number of places in each supplied with water in 1896 were as follows: Arizona, 5; Nebraska, 92; Nevada, 10; North Dakota, 10; Ohio, 154; Pennsylvania, 445; Washington, 51; Wyoming, 15; total, 782. (The Water-Works Resources Inventory of Pennsylvania compiled by the State Water Supply Commission from returns collected in 1914, showed 676 municipalities having public water supplies.) Assuming the same percentage (180) of growth from 1896 to 1924 in places supplied for the eight states as is shown for the 40 states there would be for the whole United States about 9850 places supplied in 1924. The separate table for possessions of the United States, shows 250 places supplied (with no report for Alaska), thus bringing the estimated total places supplied in the United States and its possessions to 10, 100, in round numbers. For Canada an estimate in the number of works (not places supplied) in 1924 follows: The three provinces for which 1924 data are available reported 282 works in 1924, against 219 in 1915; an increase of 63 works or about 30 per cent. Allowing the same percentage gain for the other six provinces would bring their combined number of works to 402, the total for all Canada to 684, against 528 in 1915. Presumably 700 to 750 places are supplied by these works. Numbers of works. The number of water-works in the United States in 1924 is unknown; many privately-owned works supply two or more places, with a maximum of 55 places supplied by one New Jersey company, and some cities supply outside territory. Estimates for Canada are given just above. Ownership. Thirty-eight states and the District of Columbia report ownership, embracing 7347 places supplied. Of these 5077 or 70 per cent, are supplied from municipally-owned works, 2194 from privately owned, and 76 have supplies of mixed or unknown ownership. If the 70 per cent of municipal ownership holds good for the whole United States, then about 7000 places are supplied from municipal and about 3000 from private works. Of the 250 supplied, reported in the possessions of the United States, 80 per cent are publicly owned. Canadian water supplies are nearly all publicly owned. Sources of supply. For 36 states and the District of Columbia, having a total of 6937 places supplied, the source of supply is reported, at least in part. The division is 2331 from surface, 4148 from underground sources, and 458 either mixed or unknown. Dropping the latter, of 6479 places, 2331 use surface water and 4148 underground water, or 36 and 64 per cent respectively. If data for populations supplied and water consumed were available probably the percentages would be much more than reversed. Canada has more surface than underground supplies. In the earlier years surface water supplies were relatively more common than now. "The Manual of American Water-Works" for 1888 gave the sources of supply for 1591 works in the United States (not places; but there was little difference between the two then). Of these works 907 or 57 per cent supplied surface and 684 or 43 per cent ground water. Although many private water companies supply more than one community (up to 55 in one case) there are but few "parent companies," or corporations owning and controlling many local water companies, as in the gas and electric utility fields. The largest parent water company known controls some 30 companies, located in 15 states. TABLE 2 Growth in number and changes in ownership of United States water-works since 1800 * Since this table was originally compiled one additional works, privately owned, in existence before 1800, has come to light, but as the figures up to 1896 have been before the public for many years and the change is so slight, with percentages not affected after the first few lines, it has not seemed worth while to remake the table. † Estimated. The investment in water works runs into billions and is rapidly increasing. The total volume of reported water works "contracts let" in the Construction News Section of Engineering News-Record totaled $61,445,000 for 1924 and was slightly higher for 1923, the figures including no jobs under $15,000 each, but taking in work reported as to be done by force account. Water bonds sold in 1923 totaled $86,831,000 and in 1922, $93,693,000. These figures, combined with the facts that many water works jobs run less than $15,000, that some are not reported, and, turning to the water bond issues, that the cost of many water works extensions is met from revenue, warrant the conclusion that the water works investments of the United States are now increasing at the rate of $100,000,000 to $125,000,000 or more a year. SIGNIFICANCE OF WATER SUPPLY IN DISEASE INCIDENCE Typhoid fever organisms, so far as known, breed only in the bodies of infected persons, whence they are discharged in the feces and urine. Unless these discharged organisms get back into the human body the line is broken. Water is one of the carriers from the infected person back to the second individual. Milk, other foodstuffs and direct contact are the other media of infection, but are not the concern of the water works official except to confuse the question as to cause and its elimination. The term "infected persons" comprises all persons ill with typhoid fever, a considerable proportion of recent convalescents, and about 2 per cent of all persons who have previously had typhoid fever; perhaps also a small proportion of persons who have never shown recognizable symptoms of infection. By a careful survey it is possible to enumerate all the visible sources of water-borne typhoid infection. The concealed sources, that is convalescent and "carriers," cannot be individually located; but their probable numbers in a well-studied population can be estimated with reasonable accuracy. All water, whether taken from surface or underground sources, has at some time since its precipitation been in contact with the surface of the earth and has consequently been more or less exposed to pollution with the intestinal discharges of persons and lower animals upon its catchment area. Between the sources of pollution and the ultimate destination of the water are numerous agencies operating to reduce the numbers of typhoid bacilli and other intestinal bacteria which may reach the consumer. These safeguards comprise all the agencies which tend to destroy typhoid bacilli and other intestinal bacteria in the natural and artificial processes of sewage disposal and water purification. In order, then, to arrive at an estimate of the numbers of typhoid bacilli which will reach the consumer of a water supply from known sources of pollution, it would be necessary to calculate the extent of the action of all these agencies which tend to diminish pollution. Tests for bacteria of the B. coli group afford a direct measure of the numbers of intestinal bacteria present. Since typhoid bacilli are found only in association with intestinal discharges, these tests have the most sanitary significance of any that can be made in a laboratory. Nevertheless, such tests do not distinguish between typhoid and non-typhoid pollution. A more or less accurate estimate of the probable numbers of typhoid bacilli present in a water supply represents the maximum of information which we may hope to obtain regarding the nature and extent of the significant pollution. Even this complete knowledge, however, would fail to give an accurate measure of the sanitary quality of the water, unless it could be correlated with a knowledge of the effect which the known numbers of typhoid bacilli will produce in the causation of typhoid fever among persons who drink the water. Other things being equal, we may well suppose that dosage, that is, the number of typhoid bacilli introduced, is the main factor which determines whether or not infection will result. It is highly probable that the resistance of the human body against typhoid infection is relative, not absolute; that an individual capable of resisting the invasion of a given number of typhoid bacilli would not be able to resist invasion of say ten times that number. Or, to state it differently, the more numerous the typhoid bacilli introduced, the greater is the probability that some of them will succeed in passing the defenses of the body. In sufficiently large groups it may be concluded that the dosage or exposure to infection has been greatest in that group exhibiting the highest proportion of cases. This principle of reasoning is commonly applied in undertaking to locate the source of typhoid infection. Lacking experimental observations, since it is not possible to undertake human experiments, we have only indirect means of determining the probable dosage of typhoid bacilli necessary for infection. A sharply defined epidemic of typhoid fever affords the best opportunity to determine definitely the effect of a water supply in the causation of this infection. This is so primarily because the very distinctness and sudden development of an epidemic argue the operation of a single, distinct and unusual cause, the spread of infection from a single source, since it is essentially improbable that two or more unusual sources of infection should have developed independently and simultaneously. A great number of well studied typhoid epidemics, conclusively traced to infection of water supplies, have afforded opportunities to observe and measure definitely the extent of infections caused by these water supplies among their consumers. Unfortunately, the water supplies responsible for such definite epidemics have seldom been closely examined bacteriologically during the period of their known infectivity, so that the extent of pollution which has produced the definitely determined effect is not known. The study of epidemics has, therefore, failed to give a definite idea of the probable dosage of typhoid bacilli responsible for the causation of known percentages of infection among those exposed. The effect of water supplies in the causation of endemic typhoid fever is a matter of greater public health importance and at the same time more difficult to determine, because the endemic typhoid in a community represents the sum total of infections from all sources and through all routes. Our most definite knowledge and nearest measures of the effects of known water supplies in contributing to the endemic prevalence of typhoid fever are derived from the study of communities in which a distinctly marked change in the extent of pollution of the water supply has been followed by a correspondingly definite change in the rate of typhoid prevalence. Even here, however, the observed reduction in the amount of waterborne typhoid, eliminating a number of local sources of infection, may have resulted in a reduction of cases previously caused by secondary infection from these sources, thus magnifying the apparent effect of the water supply. On the other hand, it may be that the water-borne typhoid, though obviously reduced, has not been completely eliminated. The effect of this would be to reduce. the apparent previous effect of the water supply. JURISDICTION OF STATE AND FEDERAL AUTHORITIES OVER Control and supervision of water supplies by health departments have been a gradual development of the past thirty years. In earlier times cities found little difficulty in securing supplies from |