supply of water, and had been growing in the pipe for years. As the conduit was located very near the outlet and from 4 to 6 feet below the creek level, work was difficult; but thru the kindness of the Rochester Water-Works, Superintendent William H. Clark was able to bor row a trench pump which took care of the large amount of seepage water which flowed in from the creek. The root was removed from the conduit by sawing only one bevel piece, about 8 by 10 inches, from the top of the pipe, and by fastening a small rope around the head of the root, which was then pulled out by three men. What Cities Earn and Spend In 147 of the 227 American cities of more than 30,000 population the excess of expenditures for governmental costs, including interest and outlays for permanent improvements, over revenues during the fiscal year 1918 amounted to $70,923,990, or $3.48 per capita. In the remaining 80 cities the excess of revenues over expenditures was $22,323,060, or $1.60 per capita. Taking the entire 227 cities as a group, the excess of expenditures over revenues amounted to $48,600,930, or $1.42 per capita. Their net indebtedness averaged $77.53 per capita. These are among the significant facts presented in a report entitled "Financial Statistics of Cities Having a Population of over 30,000: 1918," just issued by Director Sam L. Rogers, of the Bureau of the Census, Department of Commerce. The total revenues of all the cities during the year were $1,124,094,899; the aggregate expenditures for current expenses and interest, $894,332,392; and the aggregate outlays, $278,363,437. Of the total revenues, $790,577,487, or 70 per cent, represents receipts from the various kinds of taxes. The bulk of this amount, $705,723,158, was derived from the "general property tax," made up of taxes on real and personal property. Of the remaining receipts from taxes, the largest item, $35,576,383, was derived from taxes on the liquor traffic. Public Service Enterprises Earned Nearly Double Total of Liquor Taxes Next to taxes, the source of the largest item of revenue is found in the earnings of public service enterprises, which amounted to $116,494,645. This sum is considerably more than double the amount of payments for expenses of public service enterprises $55,174,480, thus leaving a net revenue from this source of $61,320,165, which is almost double the amount derived from liquor taxes. The bulk of the earnings of public service enterprises came from water-supply systems, from which the receipts aggregated $90,139,705. Another important source of revenue consisted of special assessments and special charges the bulk of which were for outlays aggregating $72,673,785. For all the cities taken as a whole the per capita receipts from property taxes amounted to $21.03; from other taxes, $2.00; from earnings of public service enterprises $3.39; from special assessments and special charges for outlays, $2.12; and from all other sources combined, $4.20. The expenditures during the year for governmental costs in these 227 cities, which aggregated $1,172,695,829, were, in the order of their importance: For expenses of general depart- For outlays For interest on indebtedness. $690,160,283 278,363,437 148,997,629 55,174,480 For the 227 cities taken together, the payments for governmental costs, including interest and outlays, averaged $34.16 per capita and exceeded the revenues by $48,600,930, or $1.42 per capita; but the revenues exceeded the payments for current expenses and interest by $229,762,507, an amount equal to four-fifths of the total outlays, which aggregated $278,363,437. In other words, the cities over 30,000, taken as a group, are paying from their revenues all their current departmental expenses and interest and four-fifths of their outlays. In all but four cities the revenue receipts exceeded the payments for current governmental expenses and interest; and in 80, or 35 per cent of the total number covered by the report - including the larger cities of New York, St. Louis, Pittsburgh, Los Angeles, Washington, Portland, and Denver- the revenues exceeded the entire expenditures for governmental costs, including interest and outlays. The Relative Sanitary Values of Different Part I - The Results of Tests of Vertical-Nozzle Types Bacteriologist, Health Department, District of Columbia About ten or twelve years ago the first "bubble" drinking fountain was placed upon the market. Its advent was regarded as eliminating all the evils of the common drinking cup. No more would we be menaced by the invisible enemies of mankind— the pathogenic bacteria dislodged from the lips of the tuberculous, the syphilitic, the diphtheritic, awaiting their opportunity to take up their abode in the system of the next unsuspecting drinker. Without further recommendation than its newness, its more or less artistic beauty, the glowing statements as to its sanitary value, probably without one single serious test as to its real virtues, it was accepted on its face value by nearly all municipalities as the solution of the drinking cup problem. It is true that the literature on the subject does not show that it has ever been definitely recommended by any sanitarian or public health official, nevertheless its use has been almost universal. No town with a water-supply is too small or isolated to have its bubblefountain. This type of fountain has found its way to our railroad stations, our schools and parks, and even to our trains. From an inverted water spigot it has grown to the hundred and one types now on the market, all differing in one or more minor characteristics as to shape or construction, but, as will be shown, all having about the same value from a sanitary standpoint. An Awakening After nine or ten years of uninterrupted bubble fountain popularity, it suddenly dawns upon us, following innumerable epidemics of septic sore throat, diphtheria, influenza, etc., that maybe these fountains are not as sanitary as we thought, and that in all probability they have been an important factor in the spread of disease. With this germ of doubt in our minds, we proceed to make some simple tests, that should have been made at the very beginning. To our amazement we discover that instead of being a protection, our supposedly sanitary substitute for the common drinking cup is distinctly a menace to public health fully as great as the cup itself. The literature on the bacteriology of these drinking fountains is very sparse and of comparatively recent date.* All the findings have been condemnatory to the verticalnozzle type fountain, and with just cause. Unfortunately, as Professor Dunlap, of the Iowa State University, has pointed out in his extremely valuable article of THE AMERICAN CITY of May, 1919, little has been reported bearing upon practical and laboratory tests demonstrating clearly the causes underlying the sanitary inefficiency of the present-type fountains. The reports have so far dealt only with tests made on comparatively few fountains, which would leave open the question of faulty design as playing an important part. Then again, the literature is almost wholly deficient in regard to the three problems of paramount importance that have been brought to light during the later investigations of drinking fountains, namely: the methods used by drinkers and their bearing upon the sanitary qualities of any fountain; the danger from finger contamination; and a consideration of all types of fountains on the market. The Tests The preliminary examination of several types of drinking fountains in use in the public schools, which was started somewhat over four years ago by the late Dr. Joseph J. Kinyoun, at that time Bacteriologist of the District of Columbia Health Department, and the author, showed at once that a more extensive inquiry was necessary in order to determine whether any of the types on the market really served its purpose as a sanitary fixture. When we began this examination, we believed that there were but few types and that our task would be a simple one. We were very soon disillusioned, however, as shortly after beginning the study we found about thirty separate * Pettibone, Bogart and Clark. Jour. Bacteriology, Sept., 1916. H. A. Whittaker. Public Health Reports, May 11, 1917. VERTICAL BUBBLER TYPE FOUNTAIN THAT IS PARTICULARLY UNSANITARY Water from Supply Pipe Bubbles up Thru Waste Water and distinct types assembled for examination. To this original thirty have been added somewhat over sixty other types from time to time, including about eight or ten so-called side-stream types, until at the present time I am of the opinion that if any fountains have escaped this searching inquiry, it is by no fault of mine but rather my misfortune in not having been able to procure one for test purposes. All the drinking devices came under the following types: First, those in which the outlet of the water was in a small metal or porcelain bowl, which was always filled, and a stream of water was made to pass up thru the water in the bowl much after the manner of a spring. Second, those which had a catch-basin or bowl surrounding the nozzle and were arranged so that the surplus water was allowed to drain away either rapidly or slowly. Third, those which emitted a stream of water much after the manner of a faucet, the stream being continuous or intermittent, and the nozzle not supposed to come in contact with the mouth of the drinker. In some instances these were supplied with a wire guard. Fourth, those which projected three more streams of water toward a common center, the point of contact forming a more or less "drinkable" bubble. Fifth, those projecting one or more streams of water at an angle of 15 degrees or more. or The drinking devices were attached to the permanent fixtures in the laboratory so that all parts of these could be thoroly sterilized before the tests were made on each. The device was sterilized and the water turned on and allowed to pass thru for several minutes before the experiment was begun. Check analyses of the watersupply were made in each instance. Our first tests were made by using a solution of uranin, a powerful aniline dye, detectable in quantities smaller than one part to five million parts of water, but as this method depended too much upon the personal element, that is, the eyesight of the one making the test, twenty-four hour cultures of B. prodigiosus, an organism forming a brilliant red pigment in its growth, were substituted. The samples were then collected in bottles containing sterile glucose bouillon and allowed to incubate at about 25 degrees C. for the development of the organisms and their characteristic pigment. These cultures were employed to ascertain whether or not bacteria when placed on the device would be washed away within a certain length of time. The first test was made by using a bouillor culture of the organism, with and without sputum, pouring these cultures upon the stream of water being emitted from the nozzle or nozzles of the device, and in such a manner as to simulate as nearly as possible just what happens when a person drinking allows the washing from the mouth to fall back upon the supply pipe. The second test was to mix the culture with mucus, saliva and a solution of egg albumin. This mixture closely resembles food and the mucous saliva that might come from a diseased person and is often carried on the fingers and hands of children. The nozzle and such parts of the device as would come in contact with the mouth and fingers of drinkers, were smeared over with this material by means of a sterile cotton swab, the effect being practically what takes place when the mouth or dirty fingers of drinkers come in contact with these parts. Immediately thereafter the water was turned on and examination began. Each device was tested by both methods. with the stream running continuously and with the stream running intermittently. The tests of the water were made at short intervals, beginning at two seconds after contamination, then five seconds, ten seconds, twenty seconds, thirty seconds, one minute, two minutes, three minutes, four minutes, five minutes, one hour, and in some cases after twenty-four and forty-eight hours. This was done in order to determine the earliest possible time when the test germs would disappear. All tests were repeated for confirmation. As stated previously, the individual members of each group differed somewhat, from the point of mechanical construction, in certain minor characteristics, such as shape and size of nozzle; distance of top of nozzle from top of bowl; number and size of holes in nozzle thru which water entered; number and size of holes in bowl thru which water drained; presence or absence of any bowl or catch-basin; manner in which water entered nozzle, some being constructed in such manner that the water had a swirling or sidewise motion, the theory evidently being that this whirling spray would have a tendency to wash away any contaminating organisms; the presence or absence of a so-called guard, usually consisting of a metal ring; the projection of streams of water thru holes toward a common center, the number of holes varying from three to a dozen or so; the use of copper bronze or porcelain spheres of different sizes, the water bubbling from the center. The size of some of these spheres was such that I have observed children take the entire thing in their mouths while drinking. Data Resulting From Tests The following data show the results of tests made on the first four groups of types. These four groups include all the verticalstream types. Group V, the slanting stream types, will be considered in an article in the December issue of THE AMERICAN CITY, as they present a newer and somewhat different problem both in peculiarity of construction and results obtained. The methods of testing necessarily had to be modified to meet these changed conditions. Where several types, differing only in construction, under each group gave practically the same results when tested, they are reported collectively. Group I was made up of fountains in which the outlet of the water was in a small metal or porcelain bowl which was always filled while in use, and a stream or jet of water was made to pass up thru the waste water in the bowl much after the manner of a spring. Ten types in this group having no exhaust holes or holes leading back into supply pipes showed organisms present after both tests, lasting from 2 seconds to 48 hours. Twelve types with one or more exhaust holes and bowls emptying in from 20 to 60 seconds when the water was cut off, gave results as follows: on the first test (culture poured on bubble), with stream running continuously positive results from 2 seconds to 1 hour and negative results at 24 and 48 hours, and with the stream running intermittently positive results thru 24 hours and negative results at 48 hours; on the second test (culture with sputum, etc., smeared on parts), with both continuous and intermittent streams positive results thru 24 hours and negative at 48 hours. The six remaining types in this group were those having one or more exhaust holes and a bowl that empties itself in from 2 to 15 seconds when the water is cut off. On both tests and with both continuous and intermittent flow they all gave positive results thru 1 hour and negative results at 24 and 48 hours. The fountains in Group II had a catchbasin or bowl surrounding the nozzle so arranged that the surplus water was allowed to drain away either slowly or rapidly. The water from the nozzle did not run thru the waste water as in Group I, but the waste water and mucus from the mouth might run back over the nozzle. On the first test twelve types in this group gave with continuous flow positive results thru 4 minutes and negative at 5 minutes, 1 hour, 24 and 48 hours, and WATER RUNS CONTINUOSLY AT A HEIGHT OF 3 1-2 INCHES IN THIS FOUNTAIN with intermittent flow positive results only thru 3 minutes. On the second test with both continuous and intermittent flow they gave positive results thru 1 hour and negative at 24 and 48 hours. Fifteen other types in this group showed organisms present on the first test with continuous flow thru 5 minutes and absent at 10 minutes and thereafter; with intermittent flow the organisms first disappeared at 5 minutes. On the second test of these fifteen types organisms were present thru 1 hour and absent at 24 and 48 hours. Eight types in this group on the first test first showed negative results at 3 minutes; on the second test with continuous flow the results were positive thru 10 minutes and negative at 1 hour and thereafter, and with intermittent flow positive thru 1 hour and negative at 24 and 48 hours. Two types showed negative at 1 minute with continuous flow and at 20 seconds with intermittent flow on the first test; on the second test they first showed negative at 10 minutes with continuous flow, and with intermi.tent flow they showed positive at 10 minutes and negative at 1 hour. Group III was made up of fountains which emitted a stream of water much after the manner of a faucet. The members of this group varied in design from a simply inverted quarter-inch or half-inch pipe to devices made of metal or porcelain and varying considerably in shape and internal construction; some were so arranged that the water entered the nozzle in several different ways, or thru two or more holes, thus giving a swirling motion to the stream as it left the end of the nozzle. A large number of these could be and were, taken either entirely or partly in the mouth during the process of drinking, and even when they were supplied with a so-called guard in the shape of a ring or other device, some drinkers protruded the lips far enough to grasp the nozzle before attempting to drink. The results of the tests, all of which were run in duplicate, were as follows, giving the time limits of positive and negative results: Eight types: first test with continuous flow positive at 5 minutes, negative at 10 minutes, with intermittent flow positive at 3 minutes, negative at 4 minutes; second test with continuous flow positive at 10 minutes, negative at 15 minutes, with intermittent flow positive at 15 minutes, negative at 1 hour. Six types: first test - with continuous flow positive at 30 seconds, negative at 1 minute, with intermittent flow positive at 15 seconds, negative at 20 seconds; second test - with continuous flow positive at 10 minutes, negative at 15 minutes, with intermittent flow positive at 15 minutes, negative at 1 hour. Two types: first test with continuous flow positive at 1 minute, negative at 2 minutes, with intermittent flow positive at 30 seconds, negative at 1 minute; second test with continuous flow positive at 5 minutes, negative at 10 minutes, with intermittent flow positive at 10 minutes, negative at 15 minutes. |