done. In many places the water may be deepened, the earth
276 PREVENTION [Sect.
taken from the bottom being used to build up a steep bank.
In other places large stones may be arranged along the margins.
Everywhere the streams should be so dealt with as to allow
the water to run freely, and above all, weeds should be dragged
out of the water, while the banks are rendered as " clean " and
permanent as possible. The Forest Department in Mauritius
" canalised " a stream in this way, at the cost of only 0*37 rupees
a running foot for both banks.^ While the rough training is
in progress, holes in neighbouring rocks may be filled in with
cement. Cattle often cause much trouble by leaving holes
in muddy banks in which Anophelines breed with readiness.
This problem is attacked in Panama by a rule that cattle be
allowed to water only at fixed spots which are properly pre-
pared by cobble-stones to prevent such breeding.
Exactly similar measures are required for rivers and lakes,
but the works involved for them must often be taken under
the heading of major works. Irrigation canals and irrigation
works in general must be attended to on the same principles.
It is often stated that irrigation is not compatible with
mosquito reduction. This is quite untrue, because in Ismailia
at the present day we have an example of the opposite.
Ismailia, though built in the desert, is now a well-watered
garden without mosquitos. Nothing but attention to the small
details which I have mentioned is required, and the expense
of the work is very small. Idleness always finds an excuse.
Small marshes can frequently be dealt with quite easily by
these simple means, that is, merely by clearing the natural lines
of drainage. Our mosquito gangs in Mauritius, consisting
merely of Indian workmen, made short work of considerable
areas of marsh in this manner. Only common sense is required ;
but, of course, in other instances major works by drainage have
to be performed in order to give a final outlet to the water.
Temporary marshes (section 30 (i)) can be attacked on the
same principles.
* See pp. 166 and 450.
35] MARSHES 277
(8). Major works. — I divide all works for the reduction of
mosquitos into two classes, namely {a) such as we have
hitherto dealt with, which can be performed by any intelligent
persons, and should be called minor works ; and {b) those
which require the advice of an engineer, and which should
be called major zvorks. Naturally, major works generally
require a greater expenditure of capital than minor works,
though this does not imply that they are always more ex-
pensive in the long run. As a general rule, I think, major
works need not be undertaken until minor works have failed,
unless expert advice to the contrary is given.
Marshes are of two kinds, those in which the natural outlets
for the water are sufficient but have become choked by vegeta-
tion, cultivation, or other small obstructions, which can usually
be removed by minor works ; and those in which the natural
outlets are insufficient and require deepening. For adequate
dealing with the latter, levels must be taken and extensive
works, directed by an engineer, are generally demanded.
Marshes may be caused not only by insufficient outlet,
but by excessive inlet. The water in a marsh may not come
merely from rainfall over the marshy area, but from streams
or drainage from beyond. Thus marshes are very apt to
occur at the base of hills, small or large, the rainfall upon
which runs out and stagnates on the plain. We must there-
fore consider whether we should increase the outlet or decrease
the inlet ; whether we should cut channels across a marsh in
order to free the water in it, or carry intercepting trenches
round it in order to remove the incoming waters.
In many cases, especially in marshes due to periodical
inundations by streams and lakes, the problem of how best
to deal with them can be solved only by an expert. The
marsh may be («) drained away, {b) filled up, or {c) converted
into an open lake too deep to breed mosquitos. Estimates
must be prepared, and the engineers have to select the best plan.
Marshes difficult to deal with are often formed by roads,
278 PREVENTION [Sect.
railways, houses, irrigation canals, ill-managed water conduits
and standpipes, and even by badly-made drains.
The form of drain to be employed often requires careful
thought. Drains may be (a) open channels cut in the soil ;
(d) open channels flanked with stone or concrete ; (c) channels
filled with large stones at bottom and gravel above (rubble
drains) ; or (d) various kinds of subsoil piping.
All major works require to be constantly kept in repair
by minor works, that is, after the major work is done, there
must be a constant expenditure for what engineers call
" maintenance."
Though it may be quite possible to treat a given breeding
surface by minor works only, yet the expense of this may
be so great that it will exceed the interest on the money
laid out for clearing the area by a major work plus the
expense of maintenance.
In all these matters the advice of the local engineers should
be sought. But this does not mean that the advice of the
director of the anti-malarial campaign should be disregarded.
They must work together. For instance, M. Watson notes
(section 57), that open channels are quite effective in the
lowlands of his district, because the local carrier will not
breed in them ; but piped drains are required in the uplands
where the carriers breed in any open water-course.^ Obviously
the habits of the local carriers should always be carefully ascer-
tained and considered, because this is likely to diminish largely
the cost of their reduction.
Engineering works made in ignorance or in defiance of
this knowledge may do more harm than good. For example,
I think that most of the malaria in Freetown, Sierra Leone,
was due to the badly-made drains ; and I have seen much
malaria caused by other works constructed without regard to
their possible evil sanitary consequences.
A whole book could be written on drainage of the soil as a
^ Compare sections 42 and 43.
35] VEGETATION 279
preventive of malaria — the history of the subject alone would
fill many pages. The reader will find further details in
the next chapter.
(9). Trees. — Vegetation has long been known to affect
malaria. In some places it seems to have a bad effect, and in
others a good effect. Probably this depends largely upon the
habits of the local carrier. We quite understand that certain
mosquitos prefer wooded country, while others like the open.
Thus M. Watson finds that in his district one of the most
important measures consists simply in clearing the jungle
to a considerable distance round habitations. On the other
hand, it has been sometimes claimed that trees round habita-
tions exclude mosquitos breeding in marshes beyond.
Suggestions were made some time ago that eucalyptus trees
would have the effect of drying up the soil, and that their odour
is inimical to mosquitos. Consequently many plantations of
eucalyptus were made in Italy, Mauritius, India and elsewhere,
but the results have been quite indefinite, as might be supposed.
I certainly think that eucalyptus tends to dry up the soil, but
whether it does so sufficiently to check mosquito-breeding is
another matter. The cost of making plantations will generally
exceed that of minor works.
It is well known that trees exhale a great amount of damp,
which is always favourable to mosquitos and unfavourable to
men. On the whole, then, I am strongly of opinion that many
trees should not be allowed in the proximity of houses or in
the middle of towns in the tropics. In addition to giving out
moisture they exclude the breeze and increase the heat, because
the amount of shade thrown by them scarcely compensates
for the cooling effect of wind. In the case of houses surrounded
by considerable grounds, trees are pleasant and beneficial at a
distance from the house. As already mentioned, moreover,
trees are almost sure to breed Culicines in holes in their trunks
and principal branches, I therefore think that every tropical
country should possess strict rules empowering the health
28o PREVENTION [Sect.
department to control the growth of trees within, say, lOO
metres of houses. In Port Louis, Mauritius, the Mayor, Dr
Laurent, very wisely made a great reduction in the forestation.
In the same place much discussion has occurred regarding
what are called the " river reserves." By an old law the forest
department insists upon maintaining a belt of trees in the
neighbourhood of each of the streams for the purpose, partly
of preserving the streams, and partly of maintaining the rainfall,
forests being found to increase the number of rainy days, a
thing which is very useful for various kinds of cultivation. On
consideration I was very doubtful as to the effect of such river
reserves, but suggested that in the neighbourhood of towns
and villages they should be kept scrupulously free from
undergrowth.
As a general rule undergrowth is favourable to subdomestic
and wild species of Anophelines, which, after leaving houses
where they find their food, like to take refuge in thick vegeta-
tion. Personally, I prefer houses surrounded by open lawns
rather than by thick shrubberies. In a house occupied by
Major Fowler and myself in Mauritius, and surrounded by
trees, the Scutomyia notoscripta was a pest by day, while C.
fatigans was equally troublesome by night. In the barracks
of the European troops, however, situated in an open grassy
maidan a few hundred yards away, these insects were not
nearly so numerous. Many similar experiences might be
cited.
(lo.) Houses. — Without making any general attack on the
numbers of mosquitos in a locality, they can be reduced in
dwellings simply by the manner in which the house is built.
As a general rule Anophelines much prefer dark damp houses,
especially with wooden or mud walls and thatched roofs, and
surrounded by thick vegetation. Lofty houses built of stone,
with large windows and plenty of ventilation, are not so
haunted by Anophelines, though they may sometimes be full
of Culicines.
36] HOUSES 281
I am sure that the colour of the walls is a very important
detail. No mosquitos seem to like white walls and ceiling,
especially if there is plenty of light (Nuttall [1901]).
Europeans called upon to live in the tropics are very apt to
bring with them their ideas of house decoration, and English
ladies especially like to furnish their drawing - rooms with
numbers of beautiful curtains, pictures and stuffed chairs, all
of which harbour the insects. This method of decoration is
rather barbarous at the best, and is quite inappropriate for
warm climates. Curtains especially check the breeze which is
so cooling to the inmates and so unpleasant for mosquitos.
The whole subject of houses in the tropics requires very
careful consideration. The housing of the poor is often simply
disgraceful. The building laws, if any, are generally evaded,
with the result that the mass of the people have to live in
the most miserable shanties made of wood, palm leaves, and
often of old kerosene oil tins. In every malarious country a
committee should be appointed to consider the whole of this
housing question^ especially with a view to excluding mosquitos
and malaria. Good houses suitable for the poor could easily
be designed, and building laws could be made in order to
enforce their construction by degrees. The same remarks
apply to tropical cities, where the native quarters are generally
in a dreadful condition. There is no reason which I can see
why houses in the tropics should not be as well built as those
in temperate climates. This, however, is a question of scientific
administration in opposition to those forms of administration
which we meet with too often in the world.
36. Prevention by Treatment. — It used to be said that
prevention is better than cure, but we are now finding that
cure is one of the best methods of prevention. In all parasitic
diseases, one way to remedy the disease is to destroy the
parasites in the patients, who then cease to spread them to
other persons. This has been undoubtedly the case with
282 PREVENTION [Sect.
ankylostomiasis, and will prove to be the case with other
parasitic diseases.
The history of prevention by treatment in malaria is a long
one : Laveran [1907] gives many interesting details. After
the mosquito theory was established, Koch was the first to
call marked attention to this method. He suggested that all
patients, especially children, should be carefully treated with
quinine. His principles were immediately tried in Italy and
in various German possessions (sections 49, 51); and in Pro-
fessor Celli's contribution to this book, an admirable account
will be found of the use of the method with more recent
extensions given by the Italians themselves. Equally good
results are quoted by other contributors to Chapter VII, where
many practical details are given.
As stated in section 33, treatment affects the malaria
equation partly by increasing the recovery factor r, and also
by reducing the proportion of infected persons with gametids
in their blood i. It thus modifies two of the factors. Treat-
ment of the sick only should be called case reduction ; and
treatment of those who have not yet shown symptoms, but
who may have become infected unknown to themselves, should
be called quinine prophylaxis.
We should always understand that quinine does not really
prevent infection. I presume that the protospores may be
inoculated into a person whether he is taking quinine or not,
but that they will not thrive in the former case — see negative
cases 5 and 6 in section 1 5.^ We should also understand from
section 23 that quinine, whether used for case reduction or for
prophylaxis, should be continued for a long time.
The drug does not have a very lethal effect on the gametids,
at least those of the malignant parasites, when these are once
formed ; but it certainly tends toward destroying the other
forms. The Italians insist that it does not have much effect
upon relapses, but, for reasons given in section 23, I doubt
^ And section 65 (5).
36] PREVENTION BY TREATMENT 283
whether they are right on the point, and am inch'ned to think
that the inadequacy of the treatment has been to blame for
the recurrences of fever. This point is made clearer by some
of our cases recently studied in Liverpool by enumerative
methods. On the whole, there is no doubt whatever that a
continuous and sufficient use of quinine will tend largely to
reduce the malaria ratio.
Another point must not be forgotten. For public health
work cases cannot be treated with the elaboration of detail
employed in hospital practice. All we can do is to recommend
a general line of treatment for the public ; and, in fact, the
people must become their own physicians. It is idle, therefore,
under these circumstances, to suggest complicated scales of
dosage and the use of various preparations.
I have always taught, in entire agreement with Professor
Celli,^ that daily dosage is much the best for public health work.
He insists strongly that physiological immunity of the patients
against the drug is much more quickly arrived at by daily
dosage than by intermittent dosage — and I have always found
the same. But there is a still stronger reason for daily dosage.
If we order quinine every few days, the patient is apt to put
off the evil hour until to-morrow — he complains that he has
much work to do to-day, or that he has indigestion, or is not
feeling very well. When to-morrow comes other excuses may
be found. Then the patient is attacked with fever and finally
decides that the quinine treatment is of no use. Personally,
I recommend that quinine should be given invariably once a
day, either just before the morning meal or at bedtime.
With regard to dosage, I do not advise quantities which
will disgust the people with the drug. It is perhaps better to
use amounts which will prevent the disease only in a percentage
of people, than to cure all of them at the expense of making
them all ill. CelH recommends 40 centigrams of the bisulphate,
hydrochlorate, or bihydrochlorate for adults and young persons ;
^ See section 49, p. 417.
284 PREVENTION [Sect.
20 centigrams of the same salts or 30 centigrams of the tannate
for children ; and in every case administered in the agreeable
form of comfits or of chocolates. But he increases the dose to
50 or 60 centigrams for adults in the presence of very severe
malaria. I have generally recommended 5 grains (33 centi-
grams) for adults, and much similar doses for children. It has,
of course, been shown that children can take quinine in larger
doses than adults ; but, on the other hand, I doubt whether
such larger doses are really necessary for them, and the vomit-
ing which is frequently caused in young children by the larger
doses is apt to alarm mothers. In fact we have to take the
greatest care not to impose too great a burden on the people
whom we wish to persuade into anti-malarial courses.
In India quinine is generally administered for public use in
the form of powders wrapped in papers. I think that this is
a bad method, partly because the taste is so very apparent in
powder, and also because much of it is wasted. I recommend
freshly-made pills or tablets, coated in a suitable manner. If
these are swallowed in milk or in water just before a meal, the
taste of them does not remain in the mouth for more than
a few seconds. Personally, I always take the drug thus in
preference to chocolates, etc., in which the bitterness seems to
me to be rather increased than diminished by contrast to the
sweetness of the preparation. Small children can generally be
bribed to take a pill by giving them some small sweet afterwards.
The experiments on the absorption and elimination of
quinine (section 23) cannot be accepted as having answered
the question as to which is the best salt for use in treatment
or prevention. For this purpose accurate observations by
enumerative methods, such as we are now adopting, must be
carried out at length. The only answer to the question can
be obtained by finding exactly how many parasites are killed
by a given dose. I have usually recommended the more soluble
salts, because a greater quantity is likely to be found in the
blood at a given moment ; the insoluble salts are certainly
36] SEGREGATION 285
absorbed very well, but probably not so suddenly as to produce
a maximum poisoning effect on the parasites. On the other
hand, the insoluble salts are likely to keep the blood more
continuously impregnated with the drug, and may therefore be
more useful for general prevention. Pending such researches,
it cannot be said that one salt is much more efficacious than
another ; and the salts selected must depend upon the local
price, and the local facilities for making up any forms useful for
administration. The sulphate is certainly the most generally
used in British possessions, but the Italians have lately insisted
very strongly upon the tannate, which certainly has great
advantages, being nearly tasteless and comparatively cheap.
Euquinine is more expensive, and is generally prescribed in
doses of 50 centigrams for adults and 25 centigrams for children.
Arsenic was used long ago for prophylaxis by Tommasi-
Crudeli, but has now been abandoned in Italy. We are making
further experiments with atoxyl and methylene blue.
It would be most useful if some drug could be found which
affects crescents more definitely than quinine does.
Various other agents, including serum of cattle, have
been tried, but the experiments will have to be repeated by
enumerative methods.
Segregation. — On going to West Africa in 1899, after service
in India, I was much struck by the fact that Europeans in the
former country are not segregated as they are in the latter. In
India the British officials and troops are nearly always housed
in separate locations commonly called cantonments, so that
they are frequently spared infection derived from neighbouring
crowded native quarters. But in Freetown, Sierra Leone, the
ofificials were oblisred to live almost in the midst of the native
population, even ladies being housed in rooms under which
native shops existed. I called attention to this immediately,
and attributed much of the large death-rate amongst Europeans
to it. The idea was elaborated by Stephens and Christophers,
who added the important consideration that native children are
286 PREVENTION [SECT.
the principal sources of the parasites, and J. E, Button strongly
recommended segregation for Bathurst. Others opposed the
idea, because they thought it would be wicked for the Europeans
to live apart from their coloured brethren.
At this time the advisibility of reducing mosquitos was
much disputed on the ground that the insects diffuse themselves
in all directions from a breeding centre, just as particles of gas
diffuse themselves from a generating bottle — which is, of course,
not the case. At the same time segregation was much urged
as an alternative measure ; but I pointed out [1904] that both
the measures depend upon the same principle — namely, that
mosquitos do not diffuse themselves without limit from a centre.
That is, I suggested the principles of random scatter given
in section 29 (14). Obviously, if mosquitos spread to great
distances, local drainage will not reduce their numbers, and,
equally certainly, segregation will not enable us to avoid them.
If this were the case, mosquitos infected in native villages would
be able to travel to such distances that segregation would be
useless. Happily, however, this is not the case, and therefore
both segregation and mosquito reduction remain valid.
We may segregate ourselves from the proximity either of
many infected persons or of breeding - places. S. P. James
[1903] removed a number of people from a locality in which
Anophelines abounded to one in which they were more scarce.
In October 1901, seventy-eight children belonging to these
people showed malaria parasites in 56%. In August 1902 they
were all moved into tents about 600 metres from the nearest
pool — no Anophelines now being found in the tents. The
result was that not a single case of fever occurred among
the adults, and that on 24th October the height of the fever
season, parasites were found only in one out of twenty-five
children examined — and this without treatment of any kind.
Their spleen rate had also decreased from 75% in April 1902
to 60% in October 1902 — the splenomegaly evidently dis-
appeared much more slowly than the parasite index.
37] MEASURES FOR PERSONAL PROPHYLAXIS 287
As proof of the good effect of segregation from infected
persons, I think that I may safely quote the case of the Indian
cantonments, where malaria is certainly much less rife than
in the crowded native quarters in the neighbourhood. Similar
segregation has recently been carried out in Sierra Leone, where
a special quarter has been provided for officials at some distance
from Freetown — owing, I believe, largely to my advice given
previously.
I look upon segregation as a most important measure for
preserving the health of European officials, not only against
malaria, but against many other diseases. It should always be
adopted until sanitation in general arrives at a much higher degree
of development in the tropics than it has hitherto attained.
37. Selection of Measures for Personal and Domestic
Prophylaxis. — The various preventive measures have been
described in the previous sections as briefly as possible, because
much more information is given in the important contributions
to this book contained in the following chapter. But we must
now discuss exactly an important matter, namely, the relative
values of these measures for use under various circumstances.
The preventive measures themselves are now well known and
are described in all the text-books ; but mere description is not
enough, as it is always difficult to employ every one of the
measures, and both the individual and the health officer may
be called upon to select the most feasible ones. We have
to examine, first, the measures most appropriate for the indi-
vidual — that is, for private persons, heads of houses, factories,
institutions, ships, and so on ; and, secondly, the much more
complicated problem of the selection of methods for public
prophylaxis.
(i). Self -protection of Europeans in the tropics. — I have had