D. The premises outside the houses must be made clear of trashchips, leaves, pieces of board, etc., rotting wood being believed to be an especially bad nidus of infection. The mere wetting of these things as they lay with bichloride solution is not thorough disinfection, the underside seldom being reached by the solution.

It is an injury to wet the leaves of living plants with bichloride of mercury. It kills the leaves, etc., and after a rain has washed the bichloride off, these dead leaves are a good nidus (culture nudium) for any infection not destroyed, as on the underside of the leaf. It is not reasonable to believe that the living leaf would serve such an end. E. The thorough wetting of the cleaned ground, ditches, etc., with bichloride solution or covering it with chloride of lime is doubtless efficient, but unquestionably the disinfection of the outside premises by fire is the method of election.

This is best done by the Barber asphalt furnace as used by Farrar in New Orleans in 1897, which is fairly manageable. If this machine is not available, ordinary fires built and continued for a considerable time, as done by the writer at Conquests Camp and at Brunswick, Ga., in 1893, is absolutely efficient but far less manageable and more apt to set neighboring structures afire.

The introduction of the asphalt furnace by Farrar is a distinct and valuable addition to our disinfecting armamentarium, and indeed is equivalent to giving a new and most efficient "method," for it renders quite generally applicable a method which was but rarely used by the crude means used in 1893.

By watching the houses to the leeward of the disinfected house one can sometimes form a fairly good idea of the efficiency of the work done about the premises, even if the nonimmunes do not return to those disinfected.

F. Disinfection, when we hope to suppress the fever, must be thorough. Everything must yield to this. The evil we seek to avert is too serious to weigh expense, or convenience, or hardship against it.

G. When we no longer hope to suppress the fever and use this measure simply to limit or lessen the rapidity of its spread the extent to which it should be carried out depends, as so many other quarantine measures, on the balance between the good to be attained and the cost, including hardship and inconvenience. In general, the fabrics of the sick room and nurses' room and these rooms themselves should at least be disinfected. This, if done by some method not injurious and little annoying-formaldehyde and steam, and scrubbing the floor with bichloride solution-will not make the second factor large, and will, I believe, in a considerable number of instances prevent the further infection of the household. No one who has examined the lists of the sick by houses, as in post-epidemic disinfection, can fail to be impressed with the fact that yellow fever is to a considerable extent a house disease, even during an epidemic.

MEASURES TO BE TAKEN IN A DISTRICT THREATENED BY

YELLOW FEVER.

By Surg. H. R. CARTER.

Perfect cordon

impossible.

The measures to be taken in the district threatened by the yellow fever epidemic depends to a great degree on what is done in the infected district, including its immediate environment.

If a cordon absolutely perfect were around the infected district and the arrangements for mail, freight, express, and passengers were also perfect, nothing need be done by places in the threatened district any more than by the country at large with a good maritime system on account of an epidemic in Havana. It seldom or never happens that these conditions exist, and precautions to supplement imperfections in our first guard are requisite. It is well at first to determine on the limits (next to the infected town) of certainly clean territory. The district between this and the territory adjudged infected, or probably infected (the Neutral zone. so-called "neutral zone"), requires special attention.

Before a cordon is established there is almost invariably an exodus into the country adjacent to an infected town, and after it is established there is especially at first a certain amount of seepage, so to speak, through the cordon. Also the district adjacent to a large town will probably elect to continue some direct communication with it, choosing rather to run a certain risk of infection than to interrupt their trade relations entirely. Unless this action unduly risks other communities, I think it is to be allowed. This communication, too, may be so guarded that the risk is minimal ("daylight communications," etc.), and yet these districts should not be considered for quarantine purposes as "certainly clean."

This region (neutral zone), of possible contamination, yet which we try to keep clean, is to be held as infected so far as communication with clean places is concerned and as clean with regard to that infected district. It should be

subjected to frequent inspection, and it frequently happens that a district rendered suspicious by an early exodus of refugees may subsequently be pronounced clean.

tion from South.

The local passenger trains running from the direction of Train inspec the infected town should be inspected. This is because a certain number of people may come out even to considerable distance from the infected town or district in private conveyance and take the train in clean territory at way stations. A proper system of inspection and certificates of residence materially lessens this risk.

Inspection must begin at the starting point and go as far as these people are likely to travel by private conveyance. More than this is unnecessary. At some suitable place on each road a place for the temporary detention of suspects is to be established. They should be put off the train at this place and as soon as practicable sent back to the general camp of detention. It may be advisable, however, to make other disposition of these suspects. There is no need of inspectors meeting the train before coming to each city. A main object of this inspection (of local passenger traffic) is as far as possible to prevent suspects getting on the train-that of the through passenger traffic to prevent them getting off in clean territory. The station agents of the railroads may be our most efficient assistants.

Trains from North.

Whether it be necessary to inspect trains from the North, to prevent the return of those who have gone North and not stayed long enough, may be a question. I think it is generally unnecessary and generally inefficient. On such trains, however, as have inspectors on them going North the inspection on the return trip should be done as an added precaution. There is a section of highland, the southern end of the Uninfectible Appalachian mountain system, which projects into the South, into which refugees may safely be allowed to go, but from which, owing to its proximity and other causes, a train inspection should be maintained. In this district, especially in railroad centers like Atlanta, it is also well to keep as much supervision of the movements of refugees as possible, keeping their addresses, etc., for ten days.

The country adjacent to the infected district, having due regard to means of communication as well as distance, should be inspected thoroughly and often, and kept inspected. This is necessary both for its own safety and that of the country beyond. The inability to do this on account of unwise quarantine restrictions was, I think, one cause of the spread of the recent epidemic. It is also of

region South.

the greatest commercial advantage in quieting rumors and allowing traffic to go on unimpeded, and if this work be done as a matter of routine the presence of the inspector Inspection of will cause no alarm or excitement. This inspection of healthy towns. healthy places may be widely extended in territory having even well-guarded communication with the infected district with the greatest advantage.

Freight.

In towns of considerable size it is well to have an inInspection of spector of freight and express to see that none is landed without proper certificate. It is a check on the work in the infected district.

Towns considered especially threatened by yellow fever Provisional should establish in some safe place a small isolation camp

isolation camps. of both tents and buildings, where an important case of fever can be removed, if it be right to move him, and where those presumably exposed to his infection, or rather suspects, may be isolated. Time and risk of infection will be saved by having this ready.

A place may be so isolated as to be in extreme danger of infection, generally by proximity, from an infected town. In this case all ingress must be subject to sanitary superCordon arounp vision and a cordon around the clean town may be required, healthy towns.

and practically the same system be adopted as for a be sieged city. By this method Kenner, La., 11 miles from New Orleans, received no infection in 1897,

MEASURES TO BE TAKEN IN AN INFECTED AND NONINFECTED

TOWN.

By P. A. Surg. A. H. GLENNAN.

The experience of this year has tended to confirm my observations. made at Key West, Fla., in 1887, where I gained some practical knowledge in the early stages of a yellow-fever outbreak. The first case was discovered in a large boarding house in a populous section of the city. Immediately upon the announcement of the disease these unacclimated boarders scattered to different parts of the city, and owing to defective sanitary and police powers they were kept under a desultory surveil lance only. An immediate exodus of other unacclimated people took place, by every available means, to the neighboring keys and the mainland. This is the inevitable tendency in any community upon the announcement of the first cases of yellow fever, and the immediate steps to be taken are, first, the isolation of the sick, guarding the infected premises and contiguous blocks as well, and second, a systematic supervision over all persons and effects leaving such a city or town, with the least amount of hindrance compatible with safety.

ISOLATION OF INFECTED CASES.

In incorporated municipalities, infected premises are generally posted with a yellow flag, and a guard stationed at the front door. This is not sufficient. A cordon of guards-immunes if obtainable-should be established around the infected block and contiguous squares as well. These guards, when relieved from duty, should not be allowed to go to and return from different parts of the city, but should be maintained at some near central point. A captain of these guards should inspect the day and night watches at irregular times, prevent persons, vehicles, or street cars entering or departing through the lines, see that supplies and necessary articles are delivered at stated times and places, and also to arrest and return persons escaping through the lines. In addition to this, a house-to-house inspection of the infected area should be made once or twice daily by competent physicians, twice daily being preferable, as it tends to reassure the persons in the restricted district.