hospital or hospitals as may be necessary for that purpose, and to provide therein a room or rooms, or ward or wards, for the care of confinement cases, and a room or rooms, or ward or wards, for the temporary care of persons suffering from mental or nervous disease, and also to make provision in separate buildings for patients suffering from tuberculosis and other communicable diseases, and from time to time to add thereto accomodations sufficient to take care of the patients of the county. This time may be extended by the State board of health for good cause shown. Unless adequate funds for the building of said hospital can be derived from current funds of the county, available for such purpose, issuance of county warrants and script, it shall be the duty of the commissioners' court to submit, either at a special election called for the purpose or at a regular election, the proposition of the issuance of county bonds for the purpose of building such hospital. If the proposition shall fail to receive a majority vote at such election said commissioners' court may be required thereafter at intervals of not less than 12 months, upon petition of 10 per cent of the qualified voters of said county, to submit said proposition until same shall receive the requisite vote authorizing the issuance of the bonds.

SEC. 16. Where found to be more practicable and when approved by the State board of health, two or more adjacent counties having each a population of less than 15,000 persons may join for the purpose of this act, and erect one or more hospitals for their joint use, under the terms and conditions above set forth for a single county. In such cases such combined counties shall have the same powers, and be subject to the same liabilities as a single county, herein provided for; and the district court in either county shall in such case have the same powers for the purpose of enforcing this act, as are herein provided for in case of single counties.

PUBLIC HEALTH REPORTS

VOL. XXVIII.

OCTOBER 31, 1913.

No. 44.

RAT PROOFING A MUNICIPAL SEWER SYSTEM.

A REPORT OF AN INVESTIGATION TO FIND A PRACTICAL METHOD OF RAT PROOFING THE SEWER SYSTEM OF SAN FRANCISCO.

By FRIENCH SIMPSON, Passed Assistant Surgeon, United States Public Health Service.

The city sewer system may be considered a permanent harboring place for rats, and it provides, in addition; a convenient and extensive highway for their rapid travel from one point of the city to another.

To a considerable extent rats enter and follow this highway in their migrations from house to house, and the number present in sewers bears a somewhat definite relation to the number infesting bordering premises. The rat proofing of the city sewers would not only reduce the danger of spread of disease by rats, but would reduce the number of rats infesting such premises and prove an important element for their general control and final elimination.

An examination of the rat-trapping records of this office indicates that about 13 per cent of all rats trapped are caught in the sewers. These rats were trapped through the use of 252 cage traps, representing 2.9 per cent of the total traps in use in the city.

Rats may enter sewers in four different ways:

First. Through the catch basins at present in use at the corners of city streets.

Second. Through house sewer systems, where the premises have remained vacant for a period of time, which has resulted in the evaporation of the water contained in the various traps.

Third. Through laterals from the main sewer system which extend to vacant lots and there find openings on the surface. Such laterals represent abandoned house sewer systems following the destruction of the buildings by the fire of 1906.

Fourth. Entrance is effected wherever the discharge pipes of the system are above low tide.

Without doubt the majority of rats enter and leave the sewers by means of the numerous and convenient corner catch basins, and the report contained herein will be confined to the discussion of the rat

proofing of catch basins alone. The rat proofing of house sewer systems in vacant premises, the closure of open laterals in vacant lots, and the question of lowering the outfalls which remain uncovered at low tide will not be considered at this time.

Catch Basins.

It is stated by the city engineer's office that there are at present in use in this city approximately 20,000 catch basins, consisting of three general types:

First. The old type, comprising 40 per cent, 8,000.

Second. The modern type, comprising 40 per cent, 8,000.

Third. The reconstructed type, comprising 20 per cent, 4,000.

The first, or old type, consists of an elliptical basin measuring 3 feet by 5 feet and having an average depth of 7 feet. This basin is divided by a brick wall, a portion of the wall being omitted below to provide a water trap. The water from the street enters through a vertical opening in the curb. This type of trap was found to be in very poor condition, and in every instance where examined, dry or semidry sediment was found present up to the sewer outlet. However, a vigorous clean-up of catch basins has put them in fair condition in certain sections. In nearly every instance this sediment contained rat burrows, which extended under the dividing wall and afforded easy passage to rats from the streets. The walls were not plastered, and, being rough, afforded sufficient foothold for rats to climb to the curb inlet. Many of these basins were badly in need of repair.

Type No. 2 is the form constructed under present contract work. It consists of a half-spherical and half-square pit, with a horizontal opening in the street near the curb. The cover of this type consists of two series of rectangular openings about 1 inches in width. The trap used is a mechanical device, consisting of a flap door of cast iron that opens when the water reaches a certain point in the basin and closes automatically by force of gravity. From actual examination, however, few of these flap doors were found operating properly. Some were rusted in an open position, but the majority were held open by sticks or some form of débris that had fallen into the catch basin. Sediment was found up to the level of the sewer outlet in most cases examined. The interior walls were not plastered, and rats could come and go at will.

The third type represents a reconstructed form. The pit is the same as in the second type. The traps are of the water type, constructed of quarter bends of vitrified pipe. The walls of such basins, where found smoothly plastered on the inside, and where the outlet provided was not less than 3 feet from the top of the street inlet,

were considered rat proof. It is believed that this remodeled form of trap, when properly reconstructed, is rat proof, but frequent cleaning will be required to remove accumulated sediment.

It is stated that a gang of three men with a wagon and driver can clean about 12 basins per day. With six gangs operating it would require 277 days to get around and clean each basin once. The following estimate covers approximately the cost of such cleaning:

Eighteen men clean 72 basins per day.

Cost for labor, 18 men, at $3 per day....

Cost for teams and drivers, 6 men, at $5 per day..

Cost of 6 gangs per day..

Six gangs clean 20,000 basins in 277 days at a cost of $84 per day. Total cost for cleaning each basin once in 277 days ($84 by 277)...

$54

30

84

23, 268

Since these gangs are continuously employed throughout the year of 300 working days, the total cost per year for the cleaning of catch basins would be 300 times 84, or $25,200.

The various types here outlined were all carefully examined in order to determine why they are not rat proof. This was found to be due to several causes:

First. The openings in basin inlet gratings, necessary for the proper admission of storm waters or other water accumulating in the street, were in every case sufficient in size to permit the easy passage of rats.

Second. The traps of very few of the basins examined contained water, and in a city with a long dry season, unless mechanically filled with water, such traps are ineffective as such, and in the absence of water offer no obstruction to the passage of rats to and from the

sewer.

Third. Practically all catch basins contained dry or semidry sediment up to the opening of the trap outlet to the sewer. This sediment raises the bottom of the basin to within from 20 to 27 inches of the street inlet. Rats very readily jump 2 feet vertically, and coming from the sewer easily leap upward the distance required to reach the openings in the street gratings.

An attempt was then made to devise some form of catch-basin cover or some practical mechanical attachment thereto, which would prevent the entrance or exit of rats to or from the sewer, and at the same time offer no obstruction to the flow of storm or other water from the street.

A series of 11 experiments, using 20 live rats in actual catch basins, was carried out to prove the inefficiency of the present form of basin and to test the various schemes suggested for improvement, but the ingenuity of the entire force was exhausted without finding a rat-proof cover which was both efficient and practical.

In the series where the distance from the bottom of the basin to the street inlet grating did not exceed 27 inches rats in every instance were able to reach the grating by jumping and escape through its openings.

A series of experiments as follows was then conducted in order to determine the minimum distance rats would be unable to jump in their efforts to escape through openings in inlet gratings:

First. A catch basin was cleaned to a depth of exactly 3 feet and the sewer outlet closed. Six rats were placed therein over night. No rats escaped.

Second. A piece of vitrified sewer pipe 24 inches in diameter was arranged verti cally on end. The distance inside from the bottom to the top was made exactly 3 feet 6 inches. Three rats were left therein over night. None escaped.

Third. Eight pieces of vitrified sewer pipe-four 18, two 21, and two 24 inches in diameter, respectively-were arranged vertically on end. The distance inside from bottom to top was made exactly 3 feet. On the top of four pieces of pipe, one for each size, was placed the ordinary catch basin cover. Twenty-four rats were distributed throughout and allowed to remain over night. No rats escaped.

From these experiments it is assumed that a vertical distance of 3 feet between the inlet grating and the bottom of the basin will prevent the escape of rats to the street, provided the walls of such basins are smooth.

It is understood that recently during the dry season an attempt has been made to render the traps in basins effective by cleaning, flushing, and filling the traps with water. Unfortunately the water in such traps has proved an ideal breeding place for mosquitoes. In the investigation of numerous mosquito complaints it has been found that in about 98 per cent of all cases examined the breeding of such mosquitoes was confined to corner catch basins alone. In many instances, on lifting the street grating, the basin was found to be swarming with mosquitoes, while the water in the bottom (sealing the trap) contained thousands of wrigglers. A force of men, constituting a "flying squad," is now engaged, as expeditiously as possi ble, in cleaning, flushing, and oiling all such breeding places.

From the above it will be seen that catch basins in present use present many objections, which may be stated briefly, as follows: First. Practically all are nonrat proof.

Second. When such contain water they form ideal breeding places for mosquitoes.

Third. They require an expenditure of about $25,000 per year for cleaning.

Before taking up the question of rat proofing it might be well to consider the need for catch basins.

All examined three months after cleaning were found to be full of sediment up to the trap outlet, which permitted further additions of sediment to enter directly into the sewer. From this it is estimated that 68 per cent of the time the catch basins are not performing their