BERI-BERI, THE RESULT OF HIGHLY MILLED CEREALS
mainly amongst the rice-eating people of the Malay States, Dutch Indies,
Philippine Islands, China and Japan. The disease has been known in the Far East
since the earliest times of which any record exists (2697 B.C.), but its
prevalence has greatly increased since the introduction of machine-milling.
Generally considered to be a tropical disease, beri-beri is by no means so
limited in its distribution; it has been reported as far south as Sydney in
Australia and as far north as Saghalien Island. The disease has originated in
the Port of London in the crews of ships which have been in dock for several
months; there have been cases in lunatic asylums in the United States and in
Europe; in 1894 at the Richmond Asylum, Dublin, there were 150 cases and 23
deaths from beri-beri. In 1914 Surgeon-General Blue reported the existence of
beri-beri in the county gaol at Elizabeth, New Jersey, U.S.A., and at one time
all the inmates serving sentences of over sixty days had contracted the
disease. In Newfoundland and Labrador there have been several seasonal
epidemics of beri-beri. During the last war British troops in the Dardanelles
and Mesopotamia and Annamite and Chinese troops in France suffered from
The disease is characterised by severe nervous
and heart symptoms and digestive troubles. In the
" dry " form there is a paralysis
and great wasting of the limbs; in the V wet" form
there is a dropsical condition (cedema1) of the lower
1 In oedema the tissues are distended and
waterlogged with colourless fluid (lymph). (Edema of the limbs and trunk
produces visible swelling, in which a " pit " or little hollow is
temporarily produced by pressure.
14 VITAMINS AND THE
CHOICE OF FOOD
in severe cases the trunk and arms are also affected. The departure from normal
health is usually very gradual, beginning with such vague symptoms as loss of
appetite and debility. As a practical test for the detection of incipient cases
of beri-beri among the troops in the Dardanelles, Dr. (now Sir W. H.) Willcox
made the men squat down, and those in whom the disease was beginning were
unable to rise.
The symptoms do not appear in any definite sequence. Slight oedema of
the feet and ankles is usually noticeable at an early stage, but may disappear
again. The knee-jerk and other reflexes are increased at first but absent
later. The early symptoms of nervous disorder are feelings of " pins and
needles," or numbness of the feet and legs; the muscles, especially of the
calves, are very tender to touch. The gait is unsteady, then becomes
high-stepping, and finally the legs may be completely paralysed. The pulse is
quick, palpitation is complained of; the heart is enlarged, especially on the
right side, and may be displaced towards the right. Shortness of breath
(dyspnoea) is sometimes very distressing. In a considerable number of cases the
voice is lost.
From time to time stray cases exhibiting these symptoms are observed in
Europe, and it is very likely that they are cases of beri-beri which have
escaped diagnosis owing to the mistaken belief that it is a tropical disease
peculiar to people living chiefly on rice.
During the period 1878 to 1885 it was recognised by certain medical men
that this disease in the East was common among the poorer people living almost
entirely upon a rice diet, and rare among the richer people whose food was much
more varied. Wernich (1878) thought that the bulkiness of a rice diet
interfered with the assimilation of food. Van Leent (1880) suggested that it
might result from food containing too little fat and protein; the diet of the
natives was usually very poor in fat and in nitrogenous foods such as meat,
milk and eggs. It has since been shown that beri-beri can develop on a diet
containing plenty of fat, so that an insufficiency of fat cannot be a
Takaki (1880-90), Medical Inspector-General of the Japanese Navy, believed
that the prevalence of beri-beri in the Japanese Fleet was caused by nitrogen (i.
starvation. He made a comparative study of the general hygienic conditions of
the Japanese and European navies and could find no essential difference except
in the food; he concluded that the freedom from beri-beri in the European
navies was connected with their higher protein diet. After a great deal of
opposition he persuaded the Naval Authorities to make a change in the men's
food. The quantity of rice was reduced, and in its place there was issued more
meat, fish, flour, milk and vegetables, including beans; x the cost
of the new ration was double that of the old. On the old ration 32 per cent. of
the men in the Navy had suffered from beri-beri, and on one occasion some
warships sent on active service to Chemulpo (1882) were obliged to return to
port because the crews were unfit for battle, 195 out of 330 men being ill with
beri-beri. The new ration was introduced in 1884, and in the first year the
number of cases of beriberi fell from 32 to 0-6 per cent. A minute
investigation into the cause of these few cases showed that the disease was
limited to a few prejudiced individuals who had persisted in eating rice only
and refused the new food. Three years after the changed diet there was not a
single case of beri-beri, although the personnel of the Navy had meanwhile been
Takaki's wonderful success in eliminating beri-beri from the Japanese
Navy attracted much attention to his theory that nitrogen starvation was the
cause. Numerous instances were now brought forward from other parts of the
world of the development of this disease in people living on,a good protein diet, and also of its non-appearance amongst
people who had a ration even lower in nitrogen than the old Japanese Navy
ration. The cause of the disease was therefore generally attributed to some
infection. Nevertheless a better ration was continued in the Japanese Navy,
but the Army ration still consisted chiefly
1 The term beans will be used to
denote the dried seeds of haricot and other varieties of beans, and does not
include the green pods of French or runner beans.
16 VITAMINS AND THE CHOICE OF
rice. The difference in the nutritive value of the two diets was remarkably
demonstrated during the Russo-Japanese War in 1905; there was no beri-beri in
the Navy, but 200,000 cases in the Army, which suffered from no other serious
epidemic. The rations issued to the Navy and Army at the siege of Port Arthur
may be considered as typical of the difference in the food of the two
Japanese Navy. Japanese Army.
1 lb. of meat. 5 oz. meat.
10 oz. of barley. 30 oz. of rice.
20 oz. of rice.
It is now known that it was not the meat, but some constituent of the
barley, which prevented beri-beri in the Navy. Experience in Japanese prisons
proved the value of a mixed barley and rice ration without meat.
The prevention of beri-beri is easily effected in an army, navy, or
prison by a change of diet, but it is impossible suddenly to reform the diet of
the large populations in the East.
A new conception of the origin of beri-beri was given by Eijkman (1897),
who noticed that some fowls kept at a prison in Java, of which he was medical
officer, fell ill and suffered from a peculiar form of paralysis, which he
called Polyneuritis gallinarum. No micro-organisms could be found which
might be the cause of the disease. Eijkman discovered that these birds were fed
upon the remains of the boiled rice from the prison. He therefore carried out
feeding experiments upon birds with different kinds of rice. Some were fed upon
the whole rice grain, some upon half-peeled rice, and some upon polished white
rice. Only those birds which had been fed upon the polished rice for three or
four weeks developed the disease; they were cured by being given either the
whole grain or the half-peeled rice, or the polishings (= rice bran). The
disease was produced by feeding with any sort of polished rice, whether fresh
or old, cooked or raw, home-
or imported. None of the rice was infected with organisms of any sort. It was
suggested that the monotony of the diet and consequent starvation might be the
cause of the disease. This was disproved by the fact that birds on an equally
monotonous diet of the half-peeled rice, or of the whole grain or potatoes, did
not suffer; moreover, the symptoms of starvation were quite distinct from those
of polyneuritis. Not only rice, but also sago, tapioca, or peeled barley as the
exclusive diet were found by Eijkman to cause the disease.
Eijkman, with the help of his colleague Vorderman, put the results of
these experiments to a practical test in 101 gaols in Java, containing
altogether about a quarter of a million prisoners. The inmates of some prisons
were given polished rice, of others half-peeled rice, and of still others a
mixture of the two.
In the prisons receiving polished rice, I in 39 of the inmates had
beri-beri; in the prisons receiving mixed rice, 1 in 416 had beri-beri; while
in those receiving half-peeled rice, 1 in 10,000 had beri-beri.
The results with both birds and prisoners were thus very conclusive, but
it could not definitely be proved that Polyneuritis gallinarum was the
same as beri-beri. The evidence was not at that time considered sufficient to
prove that a disease could be caused by the absence of some unknown substance
in the food. With our present knowledge this evidence would be accepted. Grijns
(1901) confirmed and extended Eijkman's observations, and showed that the
disease could also be cured if " katjang idjo " 1 beans
were added to the polished rice.
These Dutch observers also found that wheat germ, rye, barley, whole
rice and potatoes lost their power of protecting from beri-beri if they were
heated to 1150 to 1200 G. to sterilise them in order to
eliminate microorganisms as the cause of the disease.
A common criticism of Eijkman's experiments with birds was that they
were carried out in an area in which beri-
Phaseolus radiatus. a variety of haricot bean. C
18 VITAMINS AND THE CHOICE OF FOOD
was prevalent, so that the birds might have become infected. Eijkman repeated
his work in Holland, where beri-beri was unknown, on ducks and fowls, and
confirmed his former results, so that the disease could not be due to a place
Neither Eijkman nor other medical men understood at that time the real
meaning of these discoveries, for it was difficult and almost impossible to
dissociate microorganisms from epidemic disease, either directly or indirectly
through the action of toxins (poisons) produced by them. Much more work had to
be done to establish the fact that such heavy mortality could be produced by a
diet deficient in some unknown substance.
Fletcher and Gilmore Ellis, who had both believed firmly in the
infectious origin of the disease, were convinced by their independent
experiences with outbreaks of beri-beri in Malay and Singapore of the
association of beri-beri with the use of white rice. A complete survey of the
field was published by Braddon (1907) in his book Causes and Prevention of Beri-beri, and although he was wrong in attributing the cause to a poison in
rice and other farinaceous foods, yet his book served a useful purpose by
focussing attention on the rice question.
A thorough revision and extension of the existing state of knowledge of
the cause of the disease was carried out by Fraser and Stanton in the Malay
States (1911). They tested and examined every detail and really established the
cause of the disease as the absence of some unknown substance in the food. They
examined every sort of rice used in Malay, home-grown and imported, and the
various methods of its preparation for food.
The whole rice grain or padi consists of four parts: the external husk;
the pericarp, or thin covering skin of the seed; the germ or embryo plant; the
endosperm or store of food for the young plant. The endosperm forms the greater
part of the grain and consists chiefly of starch. The outer endosperm layers
contain the so-called aleurone
Fig. i.-Transverse section of
whole rice grain (husk removed) showing intact pericarp.
section of white (polished) rice
grain. Pericarp and most of sub-pericarpal layers
have been removed. It consists
almost entirely of
Reproduced by kind permission of Drs. Henry
Fraser and A. T.
Stanton from " Etiology of
Beri-Beri," Study No. 12,
for Medical Research, Federated Malay States,
composed of protein, fat, and other substances; this part of the grain is of
special significance and is known as the subpericarpal layer. The germ lies at
one end of the endosperm, from which it easily becomes detached if the pericarp
be removed. The removal or retention of the pericarp, subpericarpal layer and
germ is the main difference in the various processes of preparing the grain for
food. The structure of the grain is shown in Figs. I and 2.
Rice as prepared by the Malays, and by the natives of the Philippine
Islands, is pounded at home by hand until the husks are detached and can be
blown away. Some of the pericarp is removed by this process, but if it is dark
in colour the rest is often picked off by hand, as white rice is preferred. The
hand-milling does not interfere with the germ and subpericarpal layer, the
essential parts of the grain.
Polished (or uncured) white rice is prepared in machine mills and is the
rice commonly sold in England. The rice bran, or polishings, contains the pericarp,
germ and most of the outer layers of the endosperm; it is used as pig and
Parboiled (or cured) rice is prepared by soaking in water for
forty-eight hours and then steaming for five minutes (= curing). This process
causes the pericarp and germ to adhere to the endosperm, so that they are not
completely removed by the subsequent machine-milling. The final product is
yellowish and translucent, and often has an unpleasant smell; it is therefore
avoided by the Malays and Chinese settlers, who suffer in consequence from
beri-beri; it is only used by the Indian settlers (Tamils), who are thereby
protected from the disease.
These two kinds of rice were tested by Fraser and Stanton on 500
Javanese coolies employed on road construction in Malay and living in the
virgin jungle far from any village or town. The men were examined carefully and
all found to be free from beri-beri.
After an interval
20 VITAMINS AND THE CHOICE OF FOOD
for any latent cases to develop, so that the possibility of infection was
excluded, they were separated into two groups, one party on the favourite white
polished rice and one party on the distasteful parboiled rice. The daily ration
also contained small quantities of dried salt-fish, onions, potatoes, coconut
oil, tea and salt; the diet satisfied the standards for energy value and
quantity of protein. There were no cases of beri-beri among 273 men on
parboiled rice, but there were 20 well-marked and numerous slight cases among
220 men on white polished rice. No case of beri-beri occurred in anyone who had
been less than 87 days on white rice. No evidence of any infective agent could
be found; the beri-beri patients were in contact with the healthy men fed on
parboiled rice and the disease was not communicated. Outbreaks of beri-beri
amongst porters fed on white rice were stopped by substituting parboiled rice.
All the evidence thus pointed to white rice as the sole causative agent.
No important chemical difference in the two kinds of rice could be
detected, but Fraser and Stanton brought forward a most final and conclusive
piece of evidence by showing that a small amount of an acid or alcohol extract
of rice polishings added to a white rice diet prevented or cured birds suffering
from the disease described by Eijkman. They thus proved that an unknown
chemical substance is a necessary constituent of the diet; if it is absent,
The investigations of Vedder and his colleagues in the Philippine
Islands, in which there was a large mortality from beri-beri, confirmed those
of Fraser and Stanton. Vedder found that the poorer grades of white rice, which
were not milled and polished with such care as the better grades, were
equivalent to parboiled rice as a preventive of beri-beri. The native troops,
or Philippine Scouts, who had suffered much from the disease, were completely
freed from it in 1910 by the substitution of under-milled rice for white rice
at the instigation of Major Chamberlain.
main difficulty was to secure sufficiently under-milled rice, as the factories
generally carried the milling process too far, and each fresh lot of rice had
to be tested before issue. It is interesting to know that this is easily done
by staining a sample with a solution of iodine; the highly milled grains stain
almost black; grains still protected by some pericarp stain a light grey
In spite of all the evidence that polished white rice was the cause of
the disease, scepticism still existed. Strong and Crowell in 1912 therefore
carried out another experiment. They were able to use 29 prisoners under
sentence of death at Billibid prison. The prisoners were isolated and every
precaution taken to exclude the possibility of infection. Beri-beri developed
in 13 out of 17 prisoners fed on white rice, but there was only one doubtful
case amongst prisoners fed on whole unmilled rice. Strong and Crowell were thus
convinced that the disease was of dietetic origin.
Beri-beri is more prevalent at certain seasons. The seasonal epidemic
appears at different times in different countries and bears a direct relation
to the available food supply. Vedder found that in the Philippines the incidence
of the disease was dependent upon the rice crop; if last season's supply of
home-grown, hand-milled rice were used up before the new one was ready,
imported highly-milled rice had to be used until the next harvest, and the
cases of beri-beri were more numerous the longer this interval.
The Occurrence of Beri-beri on a Diet containing White
Beri-beri follows from a diet consisting largely of white wheaten flour.
Several epidemics have been described by Little in Labrador and Newfoundland,
where white bread is the chief food during the winter months; beriberi appears
in the spring. In the old days when brown flour was used there was no beri-beri
in these countries.
22 VITAMINS AND THE
CHOICE OF FOOD
remarkable incident is related by Little.
In 1910 a ship laden with a cargo of wholemeal wheat flour ran ashore on the
coast of Newfoundland. In order to refloat the ship a considerable portion of
flour was removed and used by the population in the neighbourhood. No case of
beri-beri was reported in that district for more than a year after the wreck.
became a common disease in Norwegian ships after the use of white wheaten flour
was made compulsory through the mistaken idea that it would help to ameliorate
the men's hard life at sea. Previously they had eaten biscuits made from rye
flour and had not suffered from beri-beri. The explanation is that in the
milling of rye there is no separation of the germ which contains the antiberi-beri
substance, but the germ is lost in the milling of white wheaten flour. Hoist
tells an amusing story of an old sea captain. In 1894 the masters of Norwegian
sailing ships were compelled to supply their men with white wheaten flour;
objecting to the new regulation, this old captain continued to use rye flour
for his own food. Beri-beri appeared amongst his crew and the sick men were
cured by rye biscuits from his private supply.
The occurrence of beri-beri amongst the British troops in the
Dardanelles and Mesopotamia during the War is easily explained. The deficiency
of the antiberi-beri substance in highly-milled wheaten flour is made good,
under ordinary circumstances, by other articles in the diet, such as eggs,
fruit and vegetables. The " mixed " diet, on active service far from
a base, consists chiefly of non-perishable or sterilised foods, such as white
flour, tinned meats, white rice, jam, sugar, cheese, margarine and salt fish.
The deficiency in the white flour is thus not compensated for by the other
articles in the diet, the antiberi-beri factor having been destroyed by heat
during the preparation of the " tinned " ration of meat and
vegetables (see table, p. 57). The Indian troops were protected from beri-beri
by the use of dahl (= dried pulse) and by atta (= wholemeal
flour). This is particu-
well illustrated in the report of Major-General Hehir on the health of the troops
during the siege of Kut. The Indian troops never suffered from beri-beri, but
there was beri-beri amongst the British troops during the early days of the
siege; it disappeared later. Beri-beri followed upon the use of white flour,
and when this was exhausted the British had to fall back on the coarse wholemeal
wheat and barley flour used by the Indians, and the disease was cured. The
reason for the disappearance of beri-beri was not appreciated at the time.
Ashmead, in 1901, mentioned the prevalence of beriberi amongst Scottish
crews who were large consumers of oatmeal; in Brazil the disease follows upon
the too exclusive use of manioc (tapioca).
Beri-beri cannot therefore be dismissed from consideration as a
tropical disease peculiar to rice-eating peoples ; large eaters of white bread
and other prepared farinaceous foods also suffer from it if other protective
food-stuffs are not taken.
Although a few medical men still cling to the belief that beri-beri is
caused by an infection, or by a toxin, the majority agree that the disease is
of dietary origin; this is shown by the resolutions passed by the section of
Tropical Medicine at the International Medical Congress in London, in 1913 :-
" (1) Beri-beri amongst natives living principally on rice is
brought about by the continued and too exclusive use of rice, submitted to a
too complete milling, which removes cortical and sub-cortical layers of the
" (2) All authorities charged with the health of native communities
are urged by every means in their power to restrain the use of this rice in the
dietary of coolies.
" (3) In view of the proved non-infectiousness of beriberi, the
section suggests that all port and sanitary authorities should abolish foreign
quarantine and other restrictive measures against this disease."
24 VITAMINS AND THE CHOICE OF FOOD
Distribution of the Antiberi-beri
The work of the several investigators has shown that foods can be
divided into two groups :-
Those containing Antiberi- Those not containing Anti-
beri Substance. beri-beri Substance.
(= rice Polished white rice.
Under-milled rice (= un- Sago.
Potato, whole. Tapioca.
Beans, dried. Peeled barley.
Barley, whole. Sterilised foods.
Rye flour. Milk. Yeast.
For many years beri-beri was considered to be a disease to which only
adults were subject. Hirota (1888) was the first to describe in infants a
condition closely resembling adult beri-beri, and his observations were
confirmed later by other Japanese physicians. The disease known as taon or
suba in the Philippines, which causes a very high infant mortality, has
been found to be identical with infantile beri-beri described by Hirota in
Japan. A special investigation, organised by the Philippine Bureau of Health
(1909), showed that of all the infants under one year of age 56-6 per cent,
died from this disease. Amongst the poorer classes many mothers lost five or
six children from it. Nine-tenths of the deaths from this disease occurred
between the ages of one and three months, and less frequently up to ten months
It is a peculiarity of the Philippine Islands that 75 per cent, of the
infant deaths are amongst the breast-fed, chiefly from infantile beri-beri; in
Europe, and practically everywhere else, this proportion is reversed and the
greater mortality is amongst the artificially fed infants.
This disease in infants was first attributed to some poison in the
mother's milk, but later it was proved to be the result of a deficiency of the
antiberi-beri substance in her milk, because her diet is poor in this
substance. The natives believe that food is eaten merely to satisfy hunger and
not to supply material for flesh and blood; if they have no sensation of hunger
they think they are well-fed. and are therefore perfectly content with their
filling diet of white rice. Women with well-developed beri-beri seldom become
pregnant; some whose diet is just on the border-line of deficiency conceive and
produce plump and apparently well-nourished babies. The diet, just sufficient
to maintain the woman before pregnancy, is inadequate for the needs of both
mother and child, and beri-beri may develop in the mother during pregnancy or
lactation. Her nursling becomes restless and sleepless, its face is usually
very full, and there is a blue look round the mouth and nose; the plump limbs
pit on pressure (oedema). The child appears out-of-breath; the voice alters and
may be lost. Finally the child dies from convulsions or heart-failure. The
right side of the heart is much dilated and displaced to the right as in adult
The child's condition can be relieved only by a change of food.
Artificial feeding is not practicable amongst these people because of their poverty
and ignorance. Formerly the only chance of saving the child's life was to
transfer it to a healthy wet nurse, but, since the recognition that the
disease is caused by something lacking in the mother's milk, it has been
possible to cure the child while still suckled by the mother, by giving the
infant an extract of rice polishings. This fact proves once again that the
disease is not caused by a poison in the milk.
Andrews quotes the case of a young woman and her infant both suffering
from beri-beri, the mother so paralysed that she could not walk. In hospital
the mother continued to suckle her baby, who received no other food, but the
woman was given under-milled rice instead of
26 VITAMINS AND
THE CHOICE OF FOOD
rice and also some mongos, a kind of bean similar to the cow-pea of the
United States. Both mother and child made a rapid recovery, and after only
twenty days were discharged cured.
The Philippine doctors now recognise the value of changing the mother's
diet, but as under-milled rice cannot be bought "in the open market, they
advise the use of mongos in its place.
The Philippine Government passed a Bill in 1914 to provide money for the
free distribution of extract of rice polishings to the suffering children. The
extract, if given in time, produces a disappearance of all the symptoms as if
by magic. At the end of three days' treatment the child is quite well except
for loss of voice, which may not return for some months.
POLYNEURITIS IN BlRDS AND ANIMALS.
The actual identity of polyneuritis in birds and of beri-beri in man
cannot be indisputably proved, but there is much evidence to suggest that both
diseases have a common cause, that is the absence of the same unknown chemical
substance from the food. Absolutely identical symptoms cannot be expected in
different species; even in cases in which the same cause is known to operate,
as in the injection of a certain poison or of a particular type of bacterium,
the symptoms produced in different species are not identical.
Experiments on birds with polyneuritis have certainly justified the
application of similar methods for the prevention and treatment of beri-beri
in man. Birds are apparently much more susceptible than man to a shortage of
the antiberi-beri substance, and show signs of the disease on a diet which does
not produce beri-beri in man; that this is a difference of degree rather than
of kind is proved by the fact that foodstuffs which will prevent and cure
polyneuritis in birds will also prevent and cure beri-beri in man.
disease in birds
resembles " dry" beri-beri;
Fig. 3.-Polyneuritis. First day of disease.
Fig. 4*.-Polyneuritis; severe case. Second day of disease.
Reproduced by kind permission of
Drs. Henry Fraser and A. T. Stanton from "Etiology of Beri-Beri,"
Study No. 12, Institute for Medical Research, Federated Malay States, 1911.
Fig. 5.-Pigeon suffering from
Fig. 6.-Same pigeon cured by
treatment with B-Vitamin. (From photographs kindly supplied by Dr. Drummond.)
is the most marked feature, and there is no oedema. At first the bird shows a
peculiar high-stepping gait not observed in any other disease of fowls. The
bird " teters " forward on its toes and stumbles if hurried; the
voice is often lost. Paralysis of the legs increases rapidly till at length the
bird can only lie on its side-in farmyard phraseology " off its legs
" (Figs. 3 and 4). Wing-drop is seen in some birds in the early stages,
and later the head may be drawn back and the whole body convulsed. Paralysis of
the swallowing muscles may cause death from starvation.
The disease develops at different rates, depending partly on individual
susceptibility, but chiefly on the degree of deficiency of the diet. Definite
symptoms do not appear on unpolished rice alone before twenty to thirty days.
On a diet affording considerable protection, Eijkman recorded a case which took
more than a year to develop.
Recovery is extraordinarily rapid if the bird is treated with some
special food, such as extract of rice polishings, when it is first off its
legs; chronic cases of long standing are generally impossible to cure (Figs. 5
The hypersensitiveness of birds to a shortage in their food of the
protective substance against beri-beri, needs to be better appreciated by
poultry-keepers. Delicacy in young chicks, as shown by so-called leg-weakness,
or by increased susceptibility to infection, may be traced in some cases to a
diet deficient in the antiberi-beri substance, such as cereals free from germ
which are frequently used for birds.
A characteristic neuritis has been produced experimentally in dogs,
goats, pigs and monkeys; the latter also showed the oedema and heart symptoms
of beri-beri. An experiment was devised by McLaughlin and Andrews in which
puppies were suckled by Philippine women whose infants had just died of
beri-beri. The puppies developed unmistakable signs of the disease.
Beri-beri has been observed in animals on board ship in which the crew