and the pools filled with gravel or rubble. It is seldom necessary in such cases to make formal stone, brick, or cement surface-drains, or to lay in subsoil pipes. We must remember that puddles can be spoilt for larvæ, even if they are not entirely drained away; either by running a current of water through them, or by breaking up their surface with stones and rubble. Pools formed in the soft earth at the bottom of water courses require filling in with large stones and gravel. Pools in the hollows of rocks, as those found in the bed of drying streams, require to be filled up by concrete (one part of Portland cement to two of sand) mixed with stones; or they can be often released by cutting the rims of rock. Water lying on flat fields or other surfaces can generally be released by small crow's-foot channels the smaller the better. Boggy ground on the margin of ponds, tanks, and streams, can be dealt with similarly. Made surface drains should be kept free of silt.

Deeper Pools, Pits, and Ponds.-These should be filled up where possible by earth, or by rubbish collected from the houses. Where they cannot be filled up they can often be spoilt for larvæ, by clearing away the vegetation on their banks in which the larvæ love to hide. Larvæ do not like open water of any depth, especially when it is free from vegetation. It is astonishing how soon even large pits and ponds can be filled up with rubbish collected from houses by the Culex gang (paragraph 12).

Disused Wells, Pits, and Tanks.-These should be filled up if possible; or cleared of vegetation and oiled regularly. Owners of premises often give great trouble about such places; in which case the municipal authorities ought to be approached on the matter. Often, however, no larvæ exist in them.

Used Wells, Washing Pits, and Drinking Pools.-When these contain larvæ, which is rare, recourse must be had to clearing vegetation and to periodical oiling. Wells can always be protected by a mosquito-proof cover with a small wire gauze grating for the bucket-as done at Bathurst, Gambia, and other places.

Streams.-Sedges and grass on the banks of sluggish streams sometimes shelter larvæ, and should be lifted. Torrents, which in the rains cannot harbour larvæ, may become prolific sources of mosquitoes when they begin to dry up. Cementing the pools in the rocks, "training" the banks, and a generous use of the broom at the end of the rainy season are the things required.

Marshes.-These must be gradually filled or drained; but this work ought to be undertaken by government. Gaol prisoners can be most usefully employed on the task, as at Lagos. It is probable that in most cases as much good will be done by clearing the rank vegetation, and by deepening the marsh and converting it into a lake, as by filling it up or draining it. A small Anopheles gang can generally do effective work by draining and dressing the boggy ground at the proximate margin of the swamp, or by oiling the marsh in the dry season, as is often done in America.

All this looks very formidable on paper. It is not so in reality. A very few men working day after day will do wonders in the course of a few months. The great thing is to make a beginning: not to form counsels of perfection, not to measure means with ends, but simply to set to work with whatever force there is available, however small it may be. A single private citizen can eradicate malaria from a whole town. In an enter

prise of this nature, the means grow as the work proceeds.

As already mentioned, the number and nature of the breeding pools depend greatly on local conditions. Freetown, Sierra Leone, contains perhaps an excessive number of pools. It was indeed partly for this reason that it was selected for the first formal campaign against mosquitoes, because success in its case would demonstrate the possibility of success in more favoured places. In Freetown there is a great rainfall (160 inches annually), and the ground, though hilly, is such that water collects in many spots. Even here, however, the Anopheles pools are not without limit, and a gang of twenty to forty men has been able to produce marked effects in so short a period as two months (vide Appendix). In my experience, most other towns will be easier to deal with than Freetown is, because they possess either a less humid climate or a more absorbent soil. In many localities, indeed, the breeding places will be found to be very few.

Much effect will be often produced in the dry season, when the pools are necessarily small in number and the larvæ are concentrated. In the rains they spread themselves more widely of course.

Perhaps the principal difficulty in dealing with Anopheles lies in the fact that they often breed in waters which, being required for drinking, washing, and irrigation purposes, cannot be drained away. Here the superintendent must be guided by his own common sense. It is generally, perhaps always, possible greatly to limit the number of such waters by closing those which are not really needed. The rest must be dealt with by cleansing from weeds, deepening, banking in, covering, or by persistent oiling. In all tropi

cal towns, natives are very fond of digging holes and banking up surface drains. The local government ought to set its face against such doings. With a pipe water-supply, and specially constructed cisterns for washing clothes, most wells and washing-pits may be filled up.

It will sometimes happen that the breeding places cannot easily be found owing to a quantity of rank vegetation. It is very wrong to allow such in the middle of towns, and the presence of ill-kempt areas full of weeds is a proof of incompetent municipal management. Efforts should be made to clear up such places. See YOUNG'S Paper in the Appendix.

A large proportion of the sickness prevalent in tropical towns is directly due to mismanagement on the part of the local authorities. These nearly always show a tendency to pay the least possible attention to the cleanliness and tidiness of their towns. Much of the money available for sanitation is often spent, not on getting the work done, but in paying far more than the market value for supervising officers who draw their salaries and write reports. It is high time that a radical change be made in these matters. Leading citizens who cannot get improvements made by quieter means will generally obtain attention if they write to the principal home papers on the subject. It is monstrous that large towns should be allowed to remain in the disgusting condition in which, as a matter of fact, they often are.

14. Destruction of Larvæ.-As argued in paragraph 3 it is better to obliterate the breeding places of mosquitoes than to spend much time over merely destroying the larvæ. We may continue to destroy the larvæ with oil for months, leaving the pools themselves alone, and yet

find that after every operation fresh eggs are laid by the adult insects living in the houses, and that fresh larvæ develop. It is indeed probable, if not certain, that if the work of destruction be carried out over a considerable area, long, thoroughly, and persistently, the mosquitoes will end by being vastly reduced in that area (unfortunately the experiment has never been thoroughly made); but the same result can be obtained more permanently, and in the end probably more economically, by obliterating the breeding places themselves.

Nevertheless, while the other work is in progress, it will do no harm to destroy larvæ wherever they are met with-which can be done either by the workmen in the course of their work, or by special men employed for the purpose. Indeed in some places, such as those just referred to, namely, collections of water in use for drinking, washing, and irrigation purposes, it is often impossible to drain at all, and the superintendent must fall back upon destruction of the larvæ.

Numerous means of destruction have been suggested -such as the introduction of fish, and of many chemical poisons or substances inimical to the larvæ. Fish, tadpoles, some kinds of beetles, and other animals do eat larvæ; but, unfortunately, they cannot always be introduced; and even if they are introduced, it is evident that their efforts will not always be successful, since we often find living larvæ in the same pools with them. It is much the same with chemical poisons. Waters which can be poisoned are those which are not required, and which had, therefore, better be drained away at once. Moreover, few poisons are effective against larvæ unless concentrated; and then most of them become dangerous, inasmuch as children and