Full text of The Effect of Tropical Sunlight on the Development of Bones of Children in Puerto Rico : A Roentgenographic and Clinical Study of Infants and Young Children with Special Reference to Rickets and Related Factors, Bureau Publication No. 217

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

UNITED STATES DEPARTMENT OF LABOR
CHILDREN’S BUREAU

PUBLICATION No. 217

THE EFFECT OF TROPICAL SUNLIGHT
ON THE DEVELOPMENT OF BONES
OF CHILDREN IN PUERTO RICO

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

UNITED STATES DEPARTMENT OF LABOR
FRANCES PERK INS, Secretary

C H IL D R E N ’S BUREAU
G RACE ABBO TT, Chief

THE EFFECT OF TROPICAL SUNLIGHT
ON THE DEVELOPMENT OF BONES
OF CHILDREN IN PUERTO RICO
tn
A Roentgenographic and Clinical Study of Infants
and Young Children with Special Reference to
Rickets and Related Factors

By
MARTHA M . ELIOT, M. D.

Bureau Publication No« 217

U N ITE D STATES
GOVERNM ENT PR IN T IN G OFFICE
W ASHINGTON : 1933

For sale by the Superintendent of Documents, Washington, D . C.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Price 10 cents

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

CON TENTS

Letter of transmittal________________________________________________ _
v
Purpose of investigation______ _____________________ ________________ ~~
1
General conditions affecting health of young children in Puerto R ico____
3
Climate and sunlight___________________ i _________________ _______
3
Hours of sunlight_______________________________ ______________
4
Intensity of sunlights___________________________________4
Exposure to sunlight______________________________________
~
5
Density of population_________________ ____ _______ _______________
g
Race________________________________ _________ ______ ______ ________
6
6
Economic conditions and diet__ ______ ________________ _____ ______
Indexes of child health______________ ___________________________
I
8
Infant mortality_________________________ ______________________
g
9
Communicable diseases__________ _______________________ ______
Causes of death among children_______________________________
12
Method of investigation____ ____ ___________________________________14
Source of material and selection of cases________________ ____ ’______
14
Collection of social and economic data_____________________________
15
Physical condition of children examined______________ £_________________
17
Physical examination_________________________________j____________
17
Sex and age____________
17
lg
Skin pigmentation_______________________________
Skeletal growth and body weight_______ ____________ ___________
19
Amount of subcutaneous fat_____________
25
Relation of amount of subcutaneous fat to sex, age, and skin
pigmentation___________________________________
26
Muscular development_________________________________
27
31
Onset of dentition_________________________________
Color of mucous membranes________ 1____________________ _____
31
Other physical findings________________________________________
32
Incidence of rickets in Puerto R ico__________________________________________ 35
Number and age distribution of children examined for rickets_______
37
37
Methods of examination for rickets________________________________
Diagnosis of rickets by clinical examination_____________________ _
38
Incidence of rickets at clinical examination and its relation to
age.
--------- ----------- --------- ___----------------------------------------38
Physical signs used as basis for clinical diagnosis of rickets_____
39
Reliability of clinical diagnosis________________________________
44
Diagnosis of rickets by roentgenographic examination______________
45
Incidence of rickets at roentgenographic examination and its
relation to age_____ - _______________________________________
45
Interpretation of clinical diagnosis in the light of roentgen-ray
diagnosis_____________________________________________|_____
4g
Determination of amount of calcium and phosphorus in the blood of
34 selected infants___________
51
Additional roentgenographic studies of the radius and ulna______________
56
Incidence of osteoporosis___________________________________________
56
Incidence of transverse lines in long bones______________________ _
59
Socio-economic conditions as factors in the health of Puerto Rican children.
61
Size and composition of family and of household_______________ 61
Family income______________________________
63
Source of income___________________
65
Employment and wages_______________________ _______ ___ ~~~~
65
Unemployment of father_________________________________~_I__
66
Family income in relation to diet______________________________
67
Per capita income__________________________________________________ 67

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

CONTENTS

Socio-economic conditions as factors in the health of Puerto Rican child
ren— Continued.
Page
Housing___________________________________________________________
69
Crowding__________________________________________ ____________
69
Furnishings_______________________________________ ^---------------70
Exposure to sunlight----------------70
Diets of Puerto Ricans_________________________________________________
76
Family diets____________________________________________________________78
78
M ilk_________
Butter--------------------------78
Eggs-------------------------------------------------------------------------------78
Meat and fish___________________
79
79
Fruit________________________ :------------------------------------ ------------Vegetables--------------------- . . . . --------------- ---------------------------------79
Diets of mothers------------------- .--------- ---------------------------------------------80
Analysis of 112 sample diets______________________
81
Classification of diets of all mothers interviewed_______________
85
Diets of children________________________________
88
Method of collecting information on diets______________________
88
Method of grading diets_______________________________________
89
Basis for grades of diets at different ages------------------- --------------89
Grades given to diets____________________________
92
94
Value of breast feeding in relation to mother’s diet_________ —
Child’s physical condition in relation to diet at time of interview.
95
Economic condition of family in relation to child’s diet________
97
Summary______________________________________________________________
98
Appendix A.— Height, weight, and head circumference for age; children
examined in Puerto Rico (12 tables)___________________
100
B. — Skeletal signs of rickets (2 tables)______________________
110
C. — Technique of roentgen-ray examination________________
113
D . — Grading of children’s diets (6 tables)___________________
114
E. — Case histories of children showing roentgen-ray evidence of
rickets_________________________________________________
120

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

L E T T E R O F T R A N S M IT T A L

U nited States D epartment of L abor,
C hildren’ s B ureau ,

W ashington M a y 3 1 , 1933.

M adam : Herewith is transmitted a bulletin entitled, The Effect of

Tropical Sunlight on the Development of Bones of Children in Puerto
Rico.
The investigation upon which this report was based was planned
and carried out under the direct supervision of Dr. Martha M . Eliot,
director of the child-hygiene division of the Children’s Bureau, who
is also the author of the report. The medical examinations were
made with the assistance of Dr. Edith B. Jackson. The study of
economic and social factors was made by staff members of the Bureau
experienced in social investigation.
The Bureau is indebted to the staff members of the Department of
Health of Puerto Rico and of the Presbyterian Hospital in San Juan,
of the Hospital Asilo des Damas in Ponce, and of the School of
Tropical Medicine, University of Puerto Rico, in San Juan. Their
interest and cooperation made the investigation possible.
Thanks are especially due to Dr. E. A. Park, of the Johns Hopkins
University School of Medicine, formerly of the Yale University School
of Medicine, for his assistance and interest in the planning of the
investigation and the interpretation of the results.
Respectfully submitted.
G race A bbott, Chief.

Hon. F rances P erkins,
Secretary o f Labor.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

THE EFFECT OF TROPICAL SUNLIGHT
ON THE DEVELOPMENT OF BONES OF CHILDREN
IN PUERTO RICO
A Roentgenographic and C linical Study o f Infants and Young Children with S pecial
R eference to R ickets and Related Factors

P U R P O S E OF IN V E S T IG A T IO N

The study of Puerto Rican children here reported was undertaken in
order to observe the roentgenographic appearance of the bones of
mfants living under the influence of tropical sunlight and to make
comparisons between the roentgenographic appearance of the bones
of such mfants and that of the bones of infants living in a temperate
climate JJurmg a previous investigation made in New Haven
*
n \’- b
/ . the Children’s Bureau in cooperation with the department
of pediatrics, Yale University School of Medicine, it had been found
that the bones of a large proportion of infants living in this temperate
climate showed sooner or later .by roentgenographic examination cer
tain minor changes that were mterpreted as evidence of slight rickets
Because such a large proportion of infants in New Haven showed these
slight changes, regardless of the fact that they had been given what
was thought to be an amount of cod-liver oil sufficient to prevent rickets, the question had arisen whether the changes should not be re
garded as physiological variations of normally growing bone rather
than as the evidences of beginning rickets.1
Questions had arisen also in New Haven as to the correct interpre
tation of certain skeletal signs used in the clinical diagnosis of mild
rickets, btudy of a group of infants and young children who had
lived continuously m a tropical climate, exposed the year around to
intense sunlight, would, it was hoped, answer a number of these
questions.
Though it was assumed, when Puerto Rico was selected for the
mvestigation, that the intensity of the sunlight there and the possi
bilities of exposure were such as to insure the prevention of rickets and
allow for normal growth of bone, it was nevertheless essential to the
main purpose of the study that this assumption be confirmed before
the data collected m Puerto Rico could be used with certainty as a
normal control for the data collected in New Haven. In this report
then, the incidence of nckets (unquestioned) in Puerto Rico and also
ATSe Ifilnor deviations from the normal commonly interpreted in
the New Haven study as evidence of slighter degrees of the disease
will be discussed as a basis for conclusions regarding the roentgeno
graphic appearance of the bones of normal infants. From time to

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

time, also, comparison will be made between these Puerto Rican
children and an unselected group of New Haven children who, during
the previous study, had been examined by the same physicians and
whose mothers had had no specific advice with regard to the preven
tion of rickets.
The investigation was made in the two principal cities of the island,
San Juan and Ponce, because of the facilities in these two places for
making roentgenographic examinations and for observing a large num
ber of infants within a short period of time. During a period of six
weeks (from the last week of January, 1927, to the second week of
March) 584 children were examined, and the homes of 556 of these
children were visited in order to obtain information regarding diet and
living conditions, especially exposure to sunlight.
In the course of the investigation the examinations of the Puerto
Rican children and the visits to their homes brought out so many other
findings of interest in connection with the health of these children
that it seemed desirable that the report should deal with other aspects
of health besides those relating to the presence or absence of rickets.
Indeed, the interpretation of the roentgenographic and clinical find
ings with regard to rickets is itself so dependent on an understanding
of the growth and development of children that a discussion of the
general health of children in the island and the conditions under
which they live is given before the discussion of the special examina
tions for rickets.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

GENERAL

C O N D IT IO N S A F F E C T IN G H E A L T H
C H IL D R E N IN P U E R T O R IC O

OF YOUNG

The health of young children in any community depends to some
extent on underlying racial and climatic factors; to a larger extent on
the general health conditions prevailing in the community, including
the facilities for health education and medical care; and to an even
larger extent on the economic and. social status of the community as a
whole. The health of the children in Puerto Rico is no exception to
this general rule. Intermixing of races, a tropical climate, ignorance,
poverty, overpopulation, irregularity of employment, probably all
contribute to the high incidence of malnutrition as well as to the high
incidence of disease and death. A report on the health of the children
studied in the present investigation would be incomplete without some
picture of these underlying factors. A detailed discussion of certain
social and economic factors such as income, housing, family grouping,
and diet will be given in later sections. (See pp. 61-97.) As a back
ground for the more detailed discussion a general discussion of climatic,
racial, and economic factors, as well as of the general health condi
tions, is given in the following sections.
CLIMATE AND SUNLIGHT

The island of Puerto Rico, the fourth largest of the islands of the
West Indies, lies between the eighteenth and nineteenth parallels of
latitude north of the equator, about 5° south of the Tropic of Cancer.
It is located, therefore, well within the Tropics.
The temperature2 varies comparatively little from season to season,
averaging 75° to 76° F. in the winter months and 80° to 81° F. in the
hottest summer months. Throughout the year the sunlight is intense,
and even in the so-called winter months it is so hot at noon that in
order to avoid the great heat and bright light the people remain, if
possible, indoors or in the shade. The windows of the houses or huts
do not have glass, but many have wooden shutters to keep out heavy
rain and intense sunshine.
Though the island is comparatively small— a little more than 100
miles long from east to west and about 40 miles wide— it has several
varieties of climate. Heavy tropical showers are frequent in the
northern part of the island and comparatively infrequent in the south
ern. The sunlight is even more continuous in the south than in the
north and for long periods of time is unrelieved by cloud or shower.
The city of Ponce, situated on the southern coast of the island, has
less rain and more sunlight than has the city of San Juan, on the
northern coast.
Because of the well-recognized association between normal calci
fication of bone and adequate exposure to sunlight, it is obviously
of importance in any community where the growth of bone or the
* Fassig, Oliver L .: T he Climate of Puerto R ico.
Medicine, vol. 4, no. 5 (Novem ber, 1928), p. 203.

Puerto R ico Review of Public Health and Tropical

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

incidence of rickets is being studied to consider the amount of sunlight
to which children, especially young infants, are ordinarily exposed.
H O U R S O F S U N L IG H T

The following comparison shows the daily average of possible
hours and of actual hours of sunlight during each month in the year
for San Juan, P. R., and for New Haven, Conn. It also shows the
total number of possible and actual hours of sunlight a year in both
San Juan and New Haven, and the daily averages for the year.3
D aily average of possi
ble hours of sunlight

D aily average of actual
hours of suniight

M onth
San Juan

January.
February
_ _ _ _ _
__
March.
__ _
A pril..
_ _ _
__ _
M ay_______ __
_ _ _ _
__ _____ ______
June___
July_____
_
_ ______ _____
August______ __
September.
____
_ _
October
November.
_ __
_
December ___________
Total for year_________
Daily average for year.

______ __

11.
11.
12.
12.
13.
13.
13.
12.
12.
11.
11.
11.

2
6
0
6
0
2
1
7
2
7
3
1

4, 416
12. 1

New Haven

9.
10.
12.
13.
14.
15.
14.
13.
12.
11.
9.
9.

6
6
0
3
5
1
8
8
5
1
9
3

4, 456
12. 2

San Juan New Haven

6.
7.
8.
7.
7.
7.
8.
8.
7.
7.
6.
6.

9
8
2
8
6
7
1
4
4
3
9
8

2, 774
7. 6

4.
6.
7.
7.
8.
9.
9.
8.
7.
6.
5.
4.

7
4
1
5
4
2
2
4
7
6
3
9

2, 592
' 7. 1

Although the daily average of possible hours of sunlight is about the
same in the two cities, the average of actual hours is somewhat greater
in San Juan. Much less seasonal variation occurs in San Juan, how
ever, in both possible and actual hours of sunlight than in New Haven.
In San Juan, moreover, the uniformly warm climate allows the actual
hours of sunlight to be continuously available throughout the year to
children living there, whereas in New Haven the long cold season oc
curs coincidentally with the decrease in actual hours of sunlight and
limits still further the availability of what sunlight there is. Though
during the summer months the daily average of actual hours of sun
light in New Haven exceeds that in Puerto Rico, this period is short
and does not bring the total number of actual hours of sunlight a year
up to the number available to children in Puerto Rico.
IN T E N S IT Y O F S U N L IG H T

In addition to the advantage that Puerto Rico has over New Haven
in actual hours of sunlight, it has the advantage also in intensity of
sunlight. Though accurate measurements are not available, it is prob
able that at the latitude of Puerto Rico (about 18° N.) the intensity of
the sunlight is relatively high and that it varies little from month to
month, whereas at the latitude of New Haven (about 41° N.) without
much doubt the intensity is lowered during the winter months. Tis3Figures supplied b y Weather Bureau, U. S. Department of Agriculture. Averages of actual hours are
based on a 25-year period for San Juan and on a 20-year period for New Haven.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

GENERAL HEALTH CONDITIONS

dall and Brown,4 who studied the relation of the altitude of the sun to
its antirachitic effect, regard 35° as the altitude below which there is
definite decrease in the ultra-violet content of the sunlight. They
state that in Jamaica, which is at about the same latitude as Puerto
Kico, the minimum altitude of the sun for the year is 50°, whereas in
Boston, which is at about the same latitude as New Haven, it is below
35° for four months of the year. When the uniformly high tempera
ture of Puerto Kico is taken into consideration there is little doubt that
the possibility of long daily exposure to sunlight, which is effective in
the prevention of rickets, is very great for infants in San Juan— far
greater than for infants in New Haven.
E X P O S U R E T O S U N L IG H T

Out of the equable climate of Puerto Rico have grown customs and
habits of living which insure an out-of-door life for children as well as
adults. The construction of the houses, the type of clothing, the
customs with regard to housework, all make exposure to sunlight
inevitable.
In Puerto Rico it is customary to keep babies indoors for the first 40
days of life, but because of the absence of window glass and the almost
universal accessibility of sunlight inside the houses (see pp. 70-75), it
is probable that even these youngest babies receive some direct sun
li&ht, though apparently no conscious effort is made to insure their getting any. After this early period the amount of exposure to sunlight
mcreases rapidly, since it is then considered safe to take the baby out.
Baby carriages are not used in Puerto Rico except by the well-to-do,
and it seemed to be the universal custom for the mother to take the
baby outdoors in her arms, or, when he was older, astride her hip, sup
ported by her arm or by a sling hung from her neck. The mother usu
ally took the baby with her when she did her daily errands, held him
in her lap as she sat in the sun, or let him sit in a box used as a make
shift play pen while she worked in the patio or yard. A few mothers
made special efforts to protect their young babies (usually those with
the fairest skins) from exposure to the sun in order to prevent tanning,
but most of these mothers admitted that they were not successful in
keeping the baby out of the sun all the time. As soon as a baby
learned to creep or to walk, he would begin to get more sunlight.
Again and again a baby would be seen playing in an open sunny door
way, or hanging over a bar placed across the door to keep him from
falling down the steps, or playing about the patio where his mother
could watch him while she worked.
. After a baby has learned to walk exposure to sunlight is almost
inevitable, as his playground is the yard, the patio, or the street, all
of which are sunny most of the time.
The fact that little children in Puerto Rico need no clothing to pro
tect hhem from cold makes it possible for them, to get plenty of direct
sunlight. During the early months of the baby’s life one or two gar
ments are customarily put on him, but many a child 6 months or a
old was seen playing about the house or in the patio without
clothes, or wearing but one scanty garment such as a shirt or short
dress. Even children 2 or 3 years of age wore little or no clothing.
Brown: Relation of the Altitude of the Sun to Its Antirachitic Effect.
Journal of tne American Medical Association, vol. 92, no. 11 (M ar. 16, 1929), pp. 860-864.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

DENSITY OF POPULATION

The population of Puerto Rico is very dense, the territory being fifth
in order of density of all the States, Territories, and possessions of the
United States; only the District of Columbia, Rhode Island, New Jer
sey, and Massachusetts have denser populations.5 In 1930 the popu
lation of Puerto Rico was 1,543,913. Of these persons 1,116,692 (72
per cent) lived in the rural parts of the island; the remaining 427,221
(28 per cent) lived in the 40 cities and towns of more than 2,500 popu
lation: 114,715 in San Juan, 53,430 in Ponce, and 259,076 in the other
38 cities and towns.6 The wide distribution of the people throughout
the hills and coastal plains with consequent isolation is responsible for
many of the health problems that are most difficult to solve.
RACE

That the Puerto Ricans are a mixture of highly pigmented peoples,
being partly Indian, partly Negro, and partly Spanish, is of interest
in this study in view of the well-known high incidence of the severer
degrees of rickets among the children of highly pigmented peoples liv
ing in temperate climates. In the early sixteenth century, at the time
of the settlement of Puerto Rico by Spaniards, the island was inhabited
by Indians. During the years of settlement the number of fullblooded Indians was greatly reduced by slavery, war, epidemics, and
intermarriage, until, according to the early Spanish records, the Indi
ans as a distinct race had disappeared 50 years after the coming of the
Spaniards. During these years there was undoubtedly a mixture of
Indian and Spanish blood, but it is difficult to determine how much
Indian blood still exists in the island. B oa s7 calculates on theoretic
grounds that possibly 14 per cent of the natives of the island have
Indian blood in their veins and thinks that the amount of Indian blood
in the rural districts is greater than is ordinarily assumed. Early in
the period of settlement the first African negroes were brought to
Puerto Rico as slaves; from then on the number of negroes in the island
increased gradually. After the first importation of slaves a m ixing of
the two races began, which has continued down to the present day.
The result is that in Puerto Rico to-day is found a race of people, who,
though predominantly Spanish in type, show many evidences of negro
blood and, in districts where less mixture with the negroes has taken
place, show some remaining evidences of Indian blood.
How much influence this mixing of races has had upon the growth
and stamina of the people can not be estimated. The apparent
retardation in the physical growth and development of the children is
probably a result not only of this mixture of races but also of various
environmental factors.
ECONOMIC CONDITIONS AND DIET

The economic condition of the great majority of the native Puerto
Ricans is extremely poor, and this fact unquestionably plays a major
rôle in the generally poor physical condition of the people, and of the
children in particular, as will be shown later. (See p. 61.) A large
8 Fifteenth Census of the United States, 1930, vol. 1, Population.
8 Ibid, pp. 1251, 1263.
^Boas, Fmnz:T The Anthropom etry of Puerto R ico. American Journal of Physical Anthropology, vol.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

GENERAL HEALTH CONDITIONS

proportion of the moil are employed, in the various agricultural pur
suits of the island, but they own no land and move about from plan
tation to plantation as the various crops need, planting or harvesting.
Wages for such work are very low, and the supply of laborers is very
large. The standard of living, at the same time, is so low that it is
possible for a family to exist on an unbelievably small income. In the
rural districts the farm laborer usually lives in a hut on the plantation
where he works. He may or may not be allowed to cultivate a small
piece of land about his hut. If he does have the opportunity, he may
raise a few banana trees or have a small truck garden. The fact, how
ever, that these farm laborers do not own the land on which they build
their houses and that they move frequently from plantation to plan
tation with the rotation of crops prevents them from having either the
desire or the time to cultivate the land around their huts to any extent.
The economic situation is such that many families do not have com
mon necessities, such as an adequate diet and. a house equipped with
beds, chairs, tables, cookstove, cooking utensils, and proper sanitary
arrangements. Living is, on the whole, so simple that it requires
little effort. There is no heating problem, the children need little or
no clothing, and the adults need only enough to cover them; few per
sons in the rural districts wear shoes. A few cents a day suffices to
provide the rice and beans that are the basis of the diet. The wide
scattering of the rural population through the hills of the island makes
the distribution of perishable foods difficult. Milk, eggs, and fresh
meat are not available to the majority of people in the rural districts.
In the cities the conditions are little better, for wages are low, and
though milk, eggs, and meat are more easily obtained there, these
foods are comparatively little used because of their high price.
The poverty of the Puerto Ricans affects the health of the children
primarily through the resulting inadequacy of their diet, especially
with respect to milk. The importance of milk as an indispensable
part of the diet of all growing children and of pregnant and lactatmg
mothers is well recognized everywhere to-day, but in Puerto Rico not
nearly enough milk is either produced or imported to supply the needs
of the children or of the mothers. The United States Census of 1920
reported only 50,311 dairy cows in Puerto Rico, or 1 cow for every 26
persons of the total population. At the time of that census there were
in the United States 19,675,297 dairy cows, or 1 for every 5.37 persons.
In Puerto Rico at that time the total production of milk was estimated
to be 7,613,071 gallons, or 2 ounces per person daily; in continental
United States during the same year the total production of milk was
estimated to be 7,805,143,792 gallons, or 26 ounces per capita daity“
It is difficult to estimate what proportion of the milk produced in
Puerto Rico is actually consumed by the people. From data reported
to the commissioner of health of the island it is estimated that the
consumption of milk by the total population in the fiscal year 1925—26
was approximately 1 ounce per capita daily and that in 1926—27 it was
less than 1 ounce per capita daily.9 Undoubtedly some milk, both
cow’s milk and goat’s milk, was consumed in the rural districts that
was not sold and was therefore not reported to the commissioner ot
health; but this amount probably was small.
s Fourteenth Census of the United States, 1920, vol. 5, Agriculture, pp. 23, 654; vol. 6, pt. 3, Agriculture,
PP R ep ortof the Commissioner of Health of Puerto R ico for the Fiscal Year 1926, p. 83; 1927, p. 113.
Juan.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

San

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

The importation of canned milk, butter, and cheese from the United
States and other countries in 1926 10 added but little to the total sup
ply of milk and milk products used by the Puerto Ricans.
When all forms of milk are considered, it is probable that the actual
consumption of milk and milk products in 1926 was less than 2 ounces
of milk per capita daily. It is estimated that in the United States the
consumption of milk alone in 1925 amounted to 19.2 ounces per capita
daily and that in addition there were consumed per capita annually
17.4 pounds of butter, 4.3 pounds of cheese, 14.87 pounds of condensed
and evaporated milk, and 2.8 gallons of ice cream.11
The need for more milk in the diet of the Puerto Rican children and
mothers, as_well as other needs of the families, will be discussed in
more detail in the section on social and economic conditions in Puerto
Rico (p. 61). The importance of the economic needs of the Puerto Rican
people in the problem of child health can not be overemphasized.
INDEXES OF CHILD HEALTH

The best available indexes of the health of a community are the
relation of birth rates to death rates, the incidence of communicable
diseases, and the mortality rates from certain causes for special age
periods. The health of the children in a community is especially re
flected in the trend of infant mortality, in the incidence of communi
cable diseases and nutritional diseases, and in the mortality from such
diseases^ as gastroenteritis and tuberculosis. Since, however, nutri
tional diseases and general malnutrition are not reportable to health
authorities, the extent of these conditions is rarely known on a com
munity-wide basis. Where nutritional disturbances are widespread
and medical care inadequate, as in Puerto Rico, study of vital statis
tics gives an inadequate estimate of the true health conditions. In
Puerto Rico, moreover, reporting of vital statistics is very incom
plete,12 and the information that can be assembled from a study of
these statistics can be regarded only as roughly indicative of the
health conditions in the island.
IN F A N T M O R T A L IT Y

That the infant mortality rate in a community varies with social
ami economic factors such as the father’s earnings, the family’s per
capita income, the mother’s employment, housing congestion, and
feeding has been demonstrated by the Children’s Bureau.13 The high
infant mortality rate reported in Puerto Rico probably reflects
(besides incomplete registration of births) the poverty of the people,
14 During the fiscal year ended June 30,1926, 3,743,803 pounds of canned milk (about
pounds per cap
ita), 4,665,931 pounds of cheese, and 789,448 pounds of butter were imported, according to the records of the
customs offices of Puerto Rico. Twenty-sixth Annual Report of the Governor of Puerto R ico, pp. 19, 22.
San Juan.
11 A H andbook of Dairy Statistics, b y T . R . Pirtle. U. S. Department of Agriculture. Washington,
12 In December, 1929, birth registration was only 55 per cent complete, according to the Report of the
Commissioner of Health of Puerto R ico for the Fiscal Year 1930, p. 8 (San Juan). From October, 1930, to
February, 1931, the United States Bureau of the Census made tests of the completeness of birth and death
registration in Puerto R ico to determine whether the island was eligible for admission to the United States
birth and death registration areas. Birth registration was soon found to be extremely poor, and the birthregistration tests were discontinued. The death-registration tests showed that 90 per cent or more of the
deaths were registered (the percentage of completeness required for admission to the death-registration
area). The island authorities took the requisite administrative steps to gain admission, and Puerto R ico
was admitted to the death-registration area as of 1932.
13 Causal Factors in Infant M ortality, b y Robert Morse W oodbury. U. S. Children’s Bureau Publica
tion N o. 142. Washington, 1925.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

GENERAL HEALTH CONDITIONS

their illiteracy, their ignorance in matters of child health, and the
prevalence of disease in the island.
On account of incomplete birth registration infant mortality rates
in Puerto Rico are probably far from accurate. The rates are given,
however, to indicate the general trend, as follows:
Deaths of
infants
under 1 year
per 1,000
live births 14

Fiscal year

19191920192119221923- 24.
1924-

20.
21.
22.
23.
25.

146
162
152
143
128
148

Deaths of
infants
under 1 year
per 1,000
live births 14

Fiscal year

1925192619271928- 291929-

26.
27.
28.
30.

150
167
146
«179
133

The infant mortality from gastrointestinal diseases is perhaps an
even better index of the unfavorable social and economic conditions
affecting the health of young children than the infant mortality as a
whole, since infant deaths from these diseases are known to result in
even larger measure from poverty and from ignorance of methods of
proper feeding and care. In Puerto Rico the mortality rate from
diarrhea and enteritis for children under 1 year was 41.2 deaths per
1.000 live births in the fiscal year 1924-25, 45.1 in 1925-26, and 55.8 in
1926-27.16 The incompleteness of birth registration affects these
rates as it does the total rate. Though the situation in Puerto Rico
is not comparable with that in the United States, it is of interest to
note that the rate from diarrhea and enteritis in the United States
birth-registration area for the calendar year 1925 was 11.2 deaths per
1.000 five births, for 1926 it was 9.7, and for 1927 it was 7.8.17
C O M M U N I C A B L E D IS E A S E S

The communicable diseases that constitute the greatest menace to
the health of the people of Puerto Rico are hookworm, malaria, tuber
culosis, and syphilis. The actual incidence of these diseases in the
island can not be stated with any degree of accuracy. Because of the
inadequacy of medical care and of facilities for diagnosis and treatment
in the rural districts, the reporting of disease is far from complete. In
1926 there was 1 physician for every 4,500 persons in Puerto Rico; but
as nearly half the physicians in the island were in San Juan and Ponce,
the rural population had only 1 physician for every 6,800 persons.18
Some of the physicians have very wide areas to cover; and since it
is often impossible for them to see sick persons before they die, they
can only guess the cause of death. Moreover, there are undoubtedly
many persons with communicable diseases who are never seen by a
physician.
Hookworm disease.

For a long time hookworm disease has infected the people of Puerto
Rico, stunting the children and incapacitating the adults. That
44 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1926, p. 121; 1927, p. 133; 1930,
p. 29. San Juan.
i® The hurricane of 1928 directly or indirectly caused m any deaths, which are reflected in the increased
infant m ortality rate.
16 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1925 (p. 100 and Appendix
Table, not num bered); 1926 (p. 101 and Appendix Table 25); and 1927 (p. 131 and Appendix Table 26). San
Juan.
I? Birth, Stillbirth, and Infant M ortality Statistics. U. S. Bureau of the Census, 1925, pp. 2 and 195; 1926,
pp. 7 and 200; and 1927, pp. 2 and 194. Washington.
48 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1926, p. 99. San Juan.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

hookworm disease was prevalent in 1926 and 1927 is shown by the
very large number of cases that were found wherever special investi
gation was made. Of the 67,727 rural inhabitants who were examined
in 1926 the proportion found to be infected was 85 per cent; of 86,029
examined in 1927 the proportion infected was 67 per cent.19 The pro
portion of infected persons living in the large cities, such as San Juan
and Ponce, was not reported, but it is believed that at the time of the
study 50 per cent of the people in the small towns were infected.20
Intensive work is being carried on by the bureau of rural sanitation of
the insular department of health to eradicate the disease, and each
year the effect of the treatments given and of the education of the
people in modern methods of sanitation and in prevention of the
disease is becoming more widespread throughout the island.
The incidence of hookworm disease in the group of children included
in this study is not known because appropriate laboratory tests could
not be undertaken. As the group was primarily an urban one and as
a large proportion of the children were little exposed to the disease
because they were too young to walk, it is probable that the incidence
was not high.
Malaria.

Malaria was reported most frequently from the lower lands of the
coastal plain, especially in the regions of the sugar plantations, where
irrigation is extensive. Since malaria is not as a rule fatal, mortality
statistics give little idea of the prevalence of the disease or of the
economic waste that results from it. The incidence of the disease
varies greatly even within a single municipality, depending on the
nearness of the dwellings to the irrigated districts. The proportion of
infected persons in the lowland districts has been found to vary from
25 per cent to as high as 75 per cent in certain small colonies.21 The
economic loss due to incapacity for work during an attack of malaria
is great. In the malaria districts many children were infected; in
Fajardo, a town on the east coast of the island, 45 per cen t22 of the
cases treated by the department of health in 1926 occurred in children
under 15 years of age. No attempt was made to determine the
incidence of malaria in the group of children included in this study,
since blood examinations were not possible. As, however, only 6 per
cent of the group were found to have enlarged spleens, it is probable
that the incidence of malaria was not high.
Tuberculosis.

The incidence of tuberculosis was reported from all but two of the
municipalities during the fiscal year 1926-27, the largest number of
cases being reported from the large cities and towns where special
diagnostic and therapeutic clinics were held. The tuberculosis death
rates were very high, and it is probable that if the causes of death had
been reported accurately they would have been still higher. It has
been estimated 23_ that the number of deaths from tuberculosis in
Puerto Rico is twice as great as is actually reported. In 1926-27 the
reported rate of deaths from tuberculosis for the island as a whole was
267 for every 100,000 population, whereas in the cities of San Juan and
19 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1926, p. 41; 1927, p. 61. San
Juan.
so Ib id ., 1925, p. 33.
» Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1925, pp. 77-84. San Juan
22 Ibid., 1926, p. 53.
28 Personal communication (Feb. 1,1930) from Dr. G. S. Pesquera, who, under the auspices of the National
Tuberculosis Association, has studied the morbidity and mortality from tuberculosis in Puerto Rico.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

GENERAL HEALTH CONDITIONS

Ponce it was 455 and 389, respectively.24 The maximum rate of 662
deaths for every 100,000 population for this same year was reported
from the city of Rio Piedras,24where an active campaign for the control
of tuberculosis had been begun. It should be remembered that at the
time of the study death registration in Puerto Rico was inaccurate.
The rates of death from tuberculosis for Puerto Rico for the fiscal
years 1923-24 to 1929-30 and the rates for the United States deathregistration area for the corresponding periods show striking differ
ences. The tuberculosis death rates for every 100,000 population were:
Puerto R ico (fiscal y ea rs)25

1923-24________________________ 205
1924^25________________________ 221
240
192526____________________
192627____________________
267
192728______________
237
1928-29________________________ 27301
192930.........
283

U. S. death-registration area (calendar y e a rs)26

1923
1924
1925
1926
1927
1928
1929

________________________
________________________
________________________
_________________
_____________________
________________________
________________________

94
90
87
87
81
79
76

The reported rates of death from tuberculosis are approximately
three times as high for Puerto Rico as for continental United States,
and the rates for the United States decreased during the period from
1923 to 1929, while those for Puerto Rico were increasing. The
increase in rate for Puerto Rico may indicate that conditions have
been growing worse or that better diagnostic work is being done and
more complete reports of deaths are being made. Public-health
officers generally believe, as a result of a demonstration made in
Framingham, Mass.,28 that for every death from tuberculosis reported
in a community nine active cases exist. On this basis, in 1926-27
when 3,842 deaths from tuberculosis were reported in Puerto Rico,29
there would have been at least 34,578 active cases.
Such a high incidence of tuberculosis can not fail to have a very
grave effect on the rates of morbidity and mortality in infancy and
early childhood. What proportion of infants and young children in
any community become infected is not known; but under the Puerto
Rican conditions of crowding and inadequate diet, it would not be
surprising to find that tuberculosis is playing afar more important
part in the infant morbidity and mortality rates than is now suspected.
In 1926-27, of the 3,842 deaths reported as caused by tuberculosis,
270 (7 per cent) were deaths of children under 15 years of age; of
these 270, 27 were children under 1 year of age and 36 were children
1 to 2 years of age.30 It is recognized now that infants and young
children become infected with tuberculosis very easily and that the
death rate among young infants so infected is very high, and it is
known that the younger the infant the less likely is the diagnosis of
tuberculosis to be made. It seems probable, therefore, that many of
the deaths attributed to other causes, such as congenital debility,
broncho-pneumonia, acute bronchitis, meningitis, or even enteritis,
may have been due to tuberculosis.
24 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year, 1927, p. 135. San Juan.
28 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1925, p. 112; 1927, p. 135; 1930,
p. 70; and Tw enty-Eighth Annual Report of the Governor of Puerto R ico, Fiscal Year 1928, p. 46.
26 M ortality Statistics (annual), 1923-1929. U. S. Bureau of the Census, Washington.
27 Conditions due to the hurricane of 1928 undoubtedly were responsible for the unusually high tubercu
losis death rates dining the time immediately following it.
28 Framingham Com m unity Health and Tuberculosis Demonstration of the National Tuberculosis Asso
ciation. Final Summary Report, 1917-1923, inclusive. Framingham, Mass., 1924.
29 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1927, p. 135.
80 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1927, Appendix Table, M or
tality from Tuberculosis, and Appendix Tables 26 and 27.
160326°—33----- 2

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

In the present study it was found that of 449 families for whom
a report was obtained concerning tuberculosis, 182 (41 per cent) gave
a history of this disease; that is, some member of the family was
reported either to have the disease at the time of the investigation
or to have died of it. The history of tuberculosis was given by some
what more families living in Ponce and in the crowded tenements of
San Juan proper than in the less crowded districts on the outskirts
of San Juan.
Syphilis.

At the time of the study it was the belief of the staff of the insular
department of health that syphilis was very common in Puerto Rico,
but no figures were available that gave an accurate picture of the
incidence of the disease in the island as a whole. In the course of
this investigation approximately 300 of the 506 families visited re
ported that Wassermann tests had been made and that in 66 of these
(22 per cent) one or both parents of the children examined had posi
tive reactions. It is probable, however, that this figure does not
represent the true incidence of syphilis in the community, since the
group studied were families known for one reason or another to the
staff members of the department of health, and a certain proportion
may have come to their attention because syphilis had been diagnosed
or suspected. As will be pointed out in a later section (p. 33), the
incidence of syphilis found at clinical examination in the children
was relatively small compared with the reported incidence in the
parents.
C A U S E S O F D E A T H A M O N G C H IL D R E N

As part of the general inquiry regarding the health conditions under
which the children were living, information was sought on the num
ber of deaths that had occurred among the brothers and sisters of
the children examined previous to the date of the interview. The
total number of these deaths, the cause of death, and the age at death,
as reported by the mothers, are shown in Table 1:
T a b l e 1.— Cause of death and age at death (as reported by mother); brothers and

sisters of children examined in Puerto Rico
Brothers and sisters of children examined
Age at death
Cause of death
Total

6 months,
Under 1 1 month,
1 year
under
under 6
and over
month
1 year

N ot re
ported

A ll causes__________________________

356

167

Causes kn ow n . _________________________

293

151

102
29
50
67
5
9
6
25

8
1
1

19
5
9
10

24
5
13
15

49
18
27
35

C om m unicable diseases_________

Causes unknown________ ________________

3
2
2
4

2
3

4
4
14

7
7

The age distribution of the deaths that occurred among children
under 1 year is strikingly different among this group of Puerto Rican

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

GENERAL HEALTH CONDITIONS

children from the distribution among children in the United States birthregistration area in 1927,31 as is shown in the following comparison:
Under 1 month_________
1 month, under 6 months
6 months, under 1 year. _

Puerto R ico

U . S. birth-registration area32

24 per cent
38 per cent
37 per cent

56 per cent
27 per cent
17 per cent

A proportion similar to that found in the present study is reported
in a bulletin of the Department of Health of Puerto Rico by A. Fernos
Isern and J. Rodriguez Pastor 33 who studied the ages at death of
children dying in the first year of life.
Though the large proportion of deaths of Puerto Rican children
over 1 month of age, and especially of those over 6 months, may be
accounted for to a great extent by the high incidence of gastrointes
tinal diseases as a cause of death, it will be seen from the tabulation
on page 12 that communicable diseases and diseases of the respiratory
tract are also important as causes of death in this age period. Con
versely, the relatively small proportion of deaths of infants over 1
month of age in continental United States may be accounted for by
the reduction in the number of deaths from gastrointestinal diseases
and, to some extent, communicable and respiratory diseases.
The large proportion of deaths from communicable disease among
the children in the families studied in Puerto Rico— more than onethird of the total— is striking. Meningitis was said to have caused 40
deaths; tetanus, 18; measles, 14; influenza, 6; malaria, typhoid fever,
and whooping cough, 5 each; tuberculosis of glands or of bones, 4;
syphilis and diphtheria, 2 each; and scarlet fever, 1. The deaths from
meningitis are of special interest. The reports of the commissioner
of health for the years 1925 to 1927 include, each year, approximately
250 cases of simple meningitis, largely occurring in children under 2
years of age, but no cases of epidemic cerebrospinal meningitis. In
the absence of this latter form of meningitis and in view of the known
high rate of deaths from tuberculosis in the population as a whole,
it would seem likely that the simple meningitis was in most instances
of tuberculous origin. Of the 40 deaths from meningitis reported,
35 were said to have occurred in children under 2 years. That tuber
culosis was not generally recognized in Puerto Rico as a cause of death
in infancy and early childhood is brought out by the fact that no case
of pulmonary tuberculosis was reported in this group, and only four
cases of bone or gland tuberculosis.
Diseases of the digestive system and malnutrition together account
ed for two-fifths of all the deaths reported; respiratory diseases for
only one-tenth, a smaller proportion, perhaps, than might have
been expected.
The causes of death here reported would indicate that a large pro
portion of the deaths probably are preventable. Control of tubercu
losis and other communicable diseases, better food, and better eco
nomic conditions would undoubtedly do much toward reducing the
mortality rate among children.
31 The age of death for Puerto Rican infants is that remembered and reported b y the mother; the age of
death in the United States birth-registration area is that given on the death certificate. The age reports are
in all probability less accurate in Puerto R ico than in the United States, but the differences are so striking
that they are given in spite of probable inaccuracies in the Puerto Rican percentages.
32 Birth, Stillbirth, and Infant M ortality Statistics, pt. 1, 1927, p. 186. U. S. Bureau of the Census.
Washington, 1929.
S3 Estudio de la Mortalidad Infantil en Puerto R ico, p. 24. San Juan.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

M E T H O D O F IN V E S T IG A T IO N

During the 6 weeks of the study 584 children, in 534 families,
ranging in age from 1 to 34 months, were given both physical and
roentgenographic examinations. In most of the families 1 child was
examined, the youngest. In 50 families 2 children were examined, the
youngest and the next to youngest. Data on the socio-economic con
ditions under which these children were living were obtained at home
visits to 506 families (556 children). The families of 28 of thè children
examined could not be visited.
The families of these children lived in the largest cities of the island,
San Juan and Ponce. Some, including a few from the rural districts,
lived on the outskirts of San Juan.
In order that the results of the examinations might be comparable
with the material collected in the New Haven study, the method of
investigation was duplicated as far as possible in every detail. The
personnel—two physicians, two social investigators, a roentgenologist,
and a secretary—was the same as that conducting the New Haven
study ; the technique of examinations, both physical and roentgeno
graphic, was in all respects identical with that of the previous study;
the same record forms were used by physicians and social investigators.
The most variable element was the clinical judgment of the physicians.'
Though every attempt was made to maintain the same basis for
judging the physical condition and development of the children as
had been used in New Haven, it is probable that the great prevalence
of poorly nourished children so influenced the physicians that the
whole scale of clinical estimate was lowered from that used by them
in New Haven. With regard to the clinical evidences of rickets, it is
possible that the examiners, in their efforts not to overlook or discount
signs that might be considered clinical evidences of the disease,
counted signs as positive, which under other conditions would have
been disregarded. It is possible that errors have been made, but if so,
they have been made in the direction of reporting the nutritional con
dition of the children to be better than it actually was and of making
diagnoses of clinical rickets when perhaps not wholly justified.
SOURCE OF MATERIAL AND SELECTION OF CASES

Examinations were made only in San Juan and Ponce, where
roentgen-ray equipment was available. In San Juan they were made
at the roentgen-ray laboratory of the insular department of health
ana at the Presbyterian Hospital; in Ponce, at the Hospital Asilo
des Damas. At each of these institutions every consideration was
given to the needs of the investigation. The division of the examina
tions between the two cities proved to be of considerable interest
because of the difference in exposure to sunlight already referred to
(p. 3) and in the economic condition of the people, which will be
discussed in later sections.
14

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

METHOD OF INVESTIGATION

Most of the children examined were brought from the well-baby
conferences conducted in San Juan and Ponce by the bureau of social
medicine of the insular department of health. In San Juan a few of
the children examined belonged in families living in the immediate
neighborhood of the roentgen-ray laboratory of the department of
health, and a few were brought to the Presbyterian Hospital clinic
from the rural districts. Practically all were children already known
to the visiting nurses of the department of health. The mothers with
their babies and little children gathered at their local well-baby
station and were taken to the place of examination in ambulances
belonging to the department of health. An average of 20 children a
day were seen.
The nurses who arranged for the children to be brought for examina
tion were asked to select children who were under 2 years of age,
principally infants under 1 year. Special effort was made to examine
infants under 1 year, as the earliest roentgen-ray evidence of rickets
had been found in children at this age in New Haven. The nurses
were requested also to select supposedly well children, such as would
be brought to a well-baby station, and to eliminate children who were
known to have an infectious disease. (It was impossible to eliminate
entirely children who were suffering from nutritional disorders, espe
cially in the group over 1 year of age, and also a few children obviously
having one infectious disease or another were seen.) As a result of
this selection, the group of children examined probably did not repre
sent a true sample of the children of Puerto Rican cities; they were
probably somewhat better nourished and less likely to have an infec
tious disease than the average.
COLLECTION OF SOCIAL AND ECONOMIC DATA

The social and economic data were collected by two investigators
who had gathered the same type of information in New Haven. Some
adaptations in the method of collecting data were necessary because of
the many differences in the customs of the people, their language, and
their diet. Fortunately for the success of the study, nurses and social
workers from the insular department of health were assigned to assist
the Children’s Bureau investigators and to act as interpreters. The
familiarity of these local workers with the customs of the people made
their help invaluable. A large proportion of this interpreting was
done by nurses who already knew the families of the children.
Information was also sought regarding any illnesses that the child
under observation had had, the occurrence in the family, so far as was
known, of tuberculosis and syphilis, and the number of children in the
family who had died, and the causes of these deaths, if known.
At the home visits information was obtained with regard to the
social and economic situation of the family, their diet, and the availa
bility of sunlight in the house or yard, as well as the family custom
with respect to exposing infants to sunlight. A detailed discussion of
each of these general subjects will be found in later sections, but it may
be said here that the social and economic data covered in general the
size and membership of the family and of the household; the number
of rooms in which the household lived; the type of house or dwelling;
the income on which the family was maintained and its source; the

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

employment of father and mother and of others in the household; and,
if the father was unemployed, the period of unemployment.
The accuracy of the data regarding diets was at first seriously ques
tioned by the investigators. By careful questioning, however, and by
observation of the meals being prepared m the homes and the provi
sions being purchased in the markets, the investigators became con
vinced that the data were essentially accurate and gave a true picture
of the food eaten by the families whose children were studied. The
uniformity of the diets in the great majority of households was notice
able. The data recorded cpnsisted of an estimate of the total quantity
of certain foods purchased by the family for a week and the frequency
with which they were used; a detailed statement of the food eaten by
the mother on the day previous to the investigator's visit; estimates
of variations in her diet during any recent periods of pregnancy and
lactation; and as detailed a statement as possible of the food eaten by
the child throughout his life.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

P H Y SIC A L C O N D IT IO N OF C H IL D R E N E X A M IN E D

The general physical condition of the children was found to be far
from satisfactory. Even from casual observation of the children in
the streets, the examining physicians received the impression that
many children were very poorly nourished, an impression that was
fully corroborated by careful physical examinations. As has been
pointed out, it is probable that the children who were examined belong
to a group of families somewhat more intelligent than the average, who
had taken advantage of the medical and nursing service offered by the
health department. It is possible that the group was weighted with
somewhat better nourished children than the average, especially
among the younger children.
PHYSICAL EXAMINATION

Though a complete clinical estimate of the physical condition of a
child should take into consideration all factors contributing to growth,
development, and health, it was found essential in this investigation
to limit the study of the physical condition to the factors that had a
bearing on the development of rickets. Since rickets is primarily
a disturbance of the nutritional processes, manifesting itself most
strikingly in poor skeletal and muscular development, and since it is
intimately associated with growth, as detailed a study as possible was
made of those clinical evidences of satisfactory or unsatisfactory
nutrition, which, when taken together, usually lead a physician to
describe a child as “ well nourished,” “ undernourished,” or “ poorly
nourished.” In addition to a clinical examination of the bones for the
signs usually considered to be indicative of rickets, certain measure
ments were taken to show the amount of skeletal growth and of body
weight, and estimates made of the amount of subcutaneous fat, the
development of the muscles, the color of the mucous membranes, the
degree of natural pigmentation of the skin, and the extent of tanning.
Routine examinations of the skin, eyes, ears, nose, throat, heart, lungs,
and abdomen were made, and any evidences of disease or other infec
tion were recorded. Because of lack of time no systematic attempt
was made to verify clinical impressions by laboratory procedures such
as tuberculin or W assermann tests or examination of blood or stools
for parasites. Suggestion for follow-up of cases that needed treatment
were made to the nurses in attendance at the clinic.
SEX AND AGE

Table la shows how the 584 children included in the study were dis
tributed according to sex and according to age at time of examination.
It will be seen that 320 (more than one-half the children whose ages
were reported) were 1 year old or less at time of examination and that
171 (nearly one-third) were 6 months old or less. From the point of
view of studying roentgenograms of the arm bones of young infants,
this age distribution was satisfactory; from the point of view of study17

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

ing growth and nutrition it would have been desirable to include also
more older children.
There is no significant difference in the age distribution of the boys
and the girls.
S K IN P IG M E N T A T IO N

Many degrees of skin pigmentation may be seen in the Puerto
Ricans, varying from that characteristic of the south European to that
of the full-blooded negro. Since, in temperate climates, the degree of
skin pigmentation associated with race seems to influence in some way
the susceptibility of children to rickets, and since, if rickets was found
in Puerto Rico, it was desirable to know whether it occurred more fre
quently in the lighter or the darker children, the children were classi
fied in four groups— according to whether their skin was light, medium,
dark, or very dark.
T a b l e la . — Age at examination; boys and girls examined in Puerto Rico
Children examined

Age 1 at examination

Total

Number

T otal.................................... ............

Boys

Per cent
distribu N um ber
tion

584

Girls

Per cent
distribu Num ber
tion

308

Per cent
distribu
tion

276

Age reported....................................................

564

100

296

100

268

100

Under 4 months____________________
4 to 6 m onths____ ___________
7 to 9 m onths___________ ____________
10 to 12 m onths_____ _____________
13 to 18 m onths________________
19 to 24 m onths......................
25 to 34 m onths.......... .......................

73
98
75
74
104
76
64

13
17
13
13
18
13
11

37
53
43
44
54
37
28

13
18
15
15
18
13
9

36
45
32
30
60
39
36

13
17
12
11
19
15
13

Age not re p o rte d ...____ _________________

1 Age is given as of nearest month; that is, "und er 4 m onths” is actually under 3 months and 16 days,
4 to 6 m onths” is from 3 months and 16 days to 6 months and 15 days, inclusive, and so on.

Of the 584 children examined, 239 were classified as fight, 191 as
medium, 122 as dark, and 29 as very dark; for 3 children no report
was made. It is obvious that a large proportion of the children
examined were but lightly or moderately pigmented, representing a
preponderance of Spanish stock. For most purposes of this study
the dark and very dark groups will be combined, since they represent,
on the whole, children largely of negro stock.
It became clear at the beginning of the study that it would be
impossible in many children to distinguish between pigmentation that
was racial in origin and pigmentation that had been acquired by ex
posure to the sun. That an appreciable degree of the pigmentation
was due to exposure is suggested by the definite preponderance in
the fight group of infants under 7 months of age, the age at which
exposure to sun, with consequent tanning, was likely to be least com
mon (see p. 19), and by the fact that the children in the medium
and in the two darker groups were in general older than those in the
fight group. Undoubtedly the increasing depth of pigmentation that
the older children showed was due, in some part at least, to the in
https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

PHYSICAL CONDITION OF CHILDREN EXAMINED

creased exposure to the sun that took place as the children grew
older and were able to get outdoors by themselves.
Though it was assumed at the beginning of the study that Puerto
Rican children not only had access to sunlight that was adequate
to prevent rickets, but were probably exposed to it constantly, never
theless it seemed worth while to record the presence or absence of
tanning of the skin, since this could be used as objective evidence
that such exposure had or had not taken place. It was, of course,
realized that the absence of tanning did not necessarily mean
that no exposure to the sun had taken place, since slight degrees
of tanning are difficult to recognize and may easily be overlooked in
persons whose skin is already somewhat pigmented. The presence
of tanning (distinguished from racial pigmentation whenever possible
by examination of parts of the skin that were usually covered with
clothing) was, however, taken as positive evidence of exposure to
sunlight, and has been used as such in connection with the study of
the roentgenograms of the bones and of certain aspects of the physical
examination.
Of the 584 children examined in Puerto Rico, 386 were reported
as tanned and 188 as not tanned; for 10 children no report was made.
Of the tanned children 79 were tanned on face, neck, and hands only,
245 on arms and legs in addition to face, neck, and hands, and 53
over the whole body; for 9 the extent of tanning was not reported.
Though many children under 7 months of age are customarily taken
out of doors almost daily, a large proportion of them do not receive
enough direct sunlight for tanning to show. At the examination of
children of this age group only one-third were found to be tanned.
It should perhaps be added, however, that even though in many chil
dren the amount of exposure possible at this age is not sufficient to
produce tanning, nevertheless, as will be shown later, the exposure
is enough to prevent the development of rickets in practically all.34
During the second six months of life a considerable increase in expo
sure evidently took place, since approximately three-quarters of the
children in this age group showed tanning. After this the proportion
showing tanning increased still more, until in the oldest group (19 to
34 months) it was found in more than 90 per cent.
That geographical conditions even within such a relatively small
territory as Puerto Rico may also make a difference in a child’s expo
sure to sunlight is shown by the fact that more of the children were
tanned in Ponce (76 per cent) than in San Juan, 35 miles away (63
per cent). This difference may be due partly to the difference in
structure of the houses (see p. 71), but it is probably due mostly to
the more continuous sunlight and drier climate of Ponce.
SK E L E T A L G R O W T H A N D B O D Y W E IG H T

Certain body measurements, namely, weight, height, and head cir
cumference, were taken for the purpose of studying the general trend
of growth of this group of Puerto Rican children in comparison with
the trend of growth of children in continental United States as a whole
and in New Haven.
The weighing was done on balance scales by one of the public-health
nurses. The weight was taken with the child stripped and was rem The large proportion of children that are breast fed during their first eight months (see p. 93) and the
long daily exposure of the mothers to the intense sunlight should be mentioned in this connection.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

corded in pounds and ounces, later converted into kilograms for the
sake of having the unit of measurement conform to that of other
series of measurements with which they were to be compared.
The measurements of height and head circumference were made
by one of the physicians, assisted by a nurse, and recorded to the
nearest millimeter.
The crown-heel length or height was taken with a measuring board 35
that had been used in similar studies in New Haven. The child was
laid flat on his back in the concavity of the measuring board, with
the crown of his head touching the headpiece of the board. The
nurse held the child’s head in this position and helped to keep his
body straight and flat on the board. The crown-heel length was
measured by bringing the footpiece into contact with the soles of
both feet. Care was taken to read and record only measurements
made when the footpiece was in contact with' the child’s heels, and
not merely in contact with the balls of his feet. The knees were
kept fully extended.
The occipito-frontal circumference of the head was measured with
a steel tape drawn snugly around the head and the measurement
recorded to the nearest millimeter. Care was taken to obtain the
circumference of the largest part of the head.
In order to show the trends of growth among the Puerto Rican
children, curves have been fitted 36 to the average weight, height, and
head circumference of children of each month of age from 1 to 33
months— boys and girls—and these curves are shown in Charts I to
IV. The trend of the weight-for-height relation is shown in Chart V
by straight lines fitted to the average weight for height. In studying
these trends the reader should bear in mind that the Puerto Rican
children are a mixed group of Spanish and Negro stock, with possibly
a slight admixture of Indian— a fact that among others may influence
their growth in height.
For the sake of comparison trends of growth in weight and height
for groups of children in continental United States, reported by the
Children’s Bureau in 1921, are shown also on the charts. The curves
show the smoothed averages of weight for age and height for age for
United States white children and the observed averages for weight
and for height of United States negro children and for United States
white children whose mothers were bom in Italy.37 Figures for children
of Spanish stock are not available. Those for children of Italian
mothers represent a south European group and are given as the next
best basis for comparison.
The trend of growth in head circumference of the Puerto Rican boys
and girls examined has been compared with that for white boys and
girls measured in New Haven, excluding children of south European
stock.38 The curves are based on children 1 to 60 months of age; but
as the Puerto Rican children included were all under 34 months, only
the sections of the curves representing New Haven children under 34
36 The measuring board is 1 meter long, solidly constructed of well-seasoned, matched wood, with a con
cavity for the child to lie in and a meter measuring stick inlaid on each side. It has a fixed headpiece and
also a m ovable footpiece that slides in tw o grooves, one on each side of the board, parallel to the meter meas
uring sticks. Measurements can be read on either of the tw o measuring sticks.
39Observed data, smoothed values, and equations of curves are given in Appendix A , p. 100.
3? Statures and Weights of Children under Six Years of Age, b y Robert M . W oodbury, pp. 85, 102,104.
U. S. Children’s Bureau Publication N o. 87. Washington, 1921.
38 Data collected during study made b y the U. S. Children’s Bureau in cooperation with the department
of pediatrics, Yale University School of M edicine (unpublished).

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

PHYSICAL CONDITION OF CHILDREN EXAMINED

months are presented. The trends of growth in head circumference
for negro children and for white children of south European stock
(largely Italian) in New Haven so closely approximate the trends for
New Haven white children (exclusive of south Europeans) that they
are not reproduced.
Charts I to IV show that these Puerto Rican boys and girls were
lighter in weight and shorter in height and had slightly smaller heads
than the reported groups of white children of continental United
States of the same age and sex, and that they were lighter in weight
and shorter than the reported group of negro children and the reported

M onth o f a i c

C h art I.—Average weight of boys 1 to 33 months of age examined in Puerto Rico, compared with
averages for certain United States boys of the same age period

group of white children of Italian mothers. The difference in height
between the Puerto Rican children and the white children of conti
nental United States is roughly 2 centimeters or less during the period
from 1 to 6 or 7 months, 3 centimeters from then to the thirteenth or
fifteenth month, and 4 centimeters thereafter. The difference for
both boys and girls from the sixth month onward is 4 to 5 per cent.
The trends of growth in height for the Puerto Rican children after the
first few months of life parallel fairly closely those for the white
children of continental United States, and the monthly percentage
increment in the two groups is fairly similar. Skeletal growth in
height proceeds for both groups at approximately the same rate,
though the average height of the Puerto Ricans is somewhat less than
that of the white children of continental United States.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

The difference in head circumference between the Puerto Rican
children and the New Haven white children is practically constant
throughout the age period represented by the curves.
The differences in weight, however, are striking. During the early
months of the first year the differences between the Puerto Rican
children and the white children of continental United States of the
same sex are at a minimum— 3 to 5 per cent for girls in the second,
third, and fourth months, and 6 per cent for boys in the second and
third months. With increasing age there is increasing difference in
the average weights of the Puerto Rican girls and the white girls of
continental United States, the maximum, 14 per cent, being reached

C hart II.—Average weight of girls 1 to 33 months of age examined in Puert 9 R ico, compared with
averages for certain United States girls of the same age period

in the thirty-second and thirty-third months. In the averages for
boys the maximum percentage difference (14 per cent) occurs between
the ninth and eleventh months. From the twelfth to the twentythird months the average Puerto Rican boy weighs 13 per cent less
than the average white boy of the same age in continental United
States, and from the twenty-fourth month onward 12 per cent less.
The averages for negro children and for white children of Italian
mothers (both in continental United States) are much closer to the
averages for white children in continental United States than are the
averages for the Puerto Rican children.
Differences in growth in height, weight, and head circumference
for age are evident between Puerto Rican children and white chil
dren in continental United States. These differences appear in the
skeletal framework and are indicated by the curves showing height

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

PHYSICAL CONDITION OF CHILDREN EXAMINED

for age and head circumference for age. The differences in average
weight are proportionately greater than the differences in either stat
ure or head circumference and are probably too great to be accounted
for only by the Puerto Rican child’s smaller skeletal framework. In
the early months of life the differences in height and weight are in
general less than in later months. Doubtless ethnic stock, climate,
disease, and general economic and social conditions, such as family in
come and diet, affect the Puerto Rican children so as to lead to these
differences.
Chart V shows the straight lines that have been fitted to the average
weight for height of Puerto Rican boys and girls 1 to 33 months of age
1

i------- r------- 1------- 1------- 1------- 1------- 1------- 1------- 1------- 1------- —
—
11 P u erto 'Rican ■boy» (£92 in h eigh t g rou p j 295 in head-cirou m ftror ic e g r o up)
____ m United S t a t e s w hite b oys (m ore than 47000 )
ii

•

United. S t a t e s w hite boya w hose m oth ers w e re b o rn ir Italy 4,366)
United S t a t e s Negro b o y s (1,406)
sea on 3SO DO
I t o 60 m on th s; s e c t io n s h o w n re prese I t s ^
t r e n d f o r b o y s lt o 3 3 m on th s)

” x>

v *

—

a
Hei^iit

x * .
0 65
K

1
| 60

✓

/
8

S'
50
44 W 4M*

* 44 **

» ♦ *’

Head c ir c u mffere

W
nee

40
y

.35
■

Month oPage
C hart III.—Average height of boys 1 to 33 months of age examined in Puerto R ico, compared with aver
ages for certain United States boys of the same age period; average head circumference of same Puerto
Rican boys compared with averages for white boys examined in N ew Haven

and to the average weight for height of white boys and girls in conti
nental United States of approximately the same height as the Puerto
Rican children.38a Similar averages lor negro and Italian children of
this age group are not available. The oldest children of the group
of white children in continental United States were probably younger
than the oldest of the Puerto Rican children studied. The chart
shows that the relation between height and weight among both boys
and girls in Puerto Rico is very similar to that in continental United
States. The similarity of these weight-for-height curves suggests
that these measures are not sufficient indexes of growth and develop
ment. Certainly the general physical appearance of the children
indicated that they were much below par according to clinical stand38» Stefures and Weights of Children under Six Years of Age, b y Robert M . W oodbury, p. 107.
Children’s Bureau Publication N o. 87. Washington, 1921.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

U. S.

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO
t-----------r

-------- 1----------- 1--------- -r— — i----------- 1----------- 1— -------1-----------,----------------------,

—— P u © r t° R ican g i r l s (265 in
- - U n ited s t a t e s w h it e g ir ls
U n ited S t a t e s w h ite g ir ls
U nited S t a t e s N egro g ir ls

X
o
##

height groups 266 in h ea d-circum ference group)
(rtiore t h a n 4 5 .Q0 0 )
w h o se m o t h e r s w ere b orn in Italy (4 226;
(1,504.)
^

—

«4 " o
*

m onths;

s ection shown rep rej;e n ts îren d For g ir ls 1t o 33 mon t h Q)

+ ~

\ x o

Heig kfc

■ o r '

✓

I
O 55

J
f

1Mi

*H »

4+ * + +

44« H H

Head c i r c u m fere i c e

s
%

M o n t h of* a g e

IV .—Average height of girls 1 to 33 months of age examined in Puerto R ico, compared with aver
ages for certain United States girls of the same age period; average head circumference of same Puerto
Rican girls compared with averages for white girls examined in N ew Haven

C hart

V .—Average weight for height of boys and girls 1 to 33 months of age examined in PuertoRico
compared with averages for United States white boys and girls of about the sam« height

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

P H Y S IC A L

C O N D IT IO N

C H IL D R E N

E X A M IN E D

ards used by physicians for judging the nutritional state of children in
continental United States.
AMOUNT OF SUBCUTANEOUS FAT

There is no doubt that the general nutritional condition of the great
majority of Puerto Rican children, as seen from a clinical point of view,
was far from being as satisfactory as that of children in continental
United States. Many of the most poorly nourished or atrophic chil
dren were in a class entirely outside the public-health experience of the
physicians, and could be compared only with the children suffering
from severe marasmus or starvation who are seen elsewhere in hospital
wards. Excellently nourished children, moreover, were much less
commonly found and then usually among the younger infants. The
great mass of children belonged in a group whose nutritional condition
would have been called fair or poor in continental United States.
That the usual standard of gauging the physical condition of the
children was not adhered to (because of the preponderance of poorly
nourished children), but that a standard based on the range within the
group itself was unintentionally substituted, will be shown later.
The amount of subcutaneous fat was estimated for each Puerto
Rican child and a grade given, as had been done for the New Haven
children, on a scale of five grades: Very good,good, fair, poor, and very
poor. It is realized that such clinical estimates are subjective and
may vary to a considerable degree according to the judgment of the
physician making the examination. That they may also be influenced
greatly by the variations and extremes within the group under obser
vation at the time, is in general well recognized, and this has been
illustrated dearly by the findings in the present study.
It was without question the intention of the physicians who made the
examinations in Puerto Rico to use the same standards for estimating
subcutaneous fat as they had used in similar studies in New Haven,
and so to have comparable data from the two places. However, in the
face of the preponderance of poorly nourished children and the scar
city of really well-nourished ones, the judgment of the physicians with
regard to estimating amounts of subcutaneous fat rapidly became
warped, and unintentionally there occurred, in conformity with the
variations within the group, a definite readjustment in their whole
scale of values, as has been pointed out. Children who in New Haven
would have been considered to have a ‘ ‘ fair” amount of subcutaneous
fat, were, because of this unconscious readjustment of standards,
reported as having a “ good” amount, and those who in New Haven
would have been considered to have a “ poor’ * amount were reported as
having a “ fair” amount. There is little doubt that the ratings of the
fat of these Puerto Rican children are high as compared with the
ratings given in New Haven by thè same physicians.
Even .though the results of the estimates of subcutaneous fat for the
two groups are not, therefore, comparable, it may be assumed that the
distribution of grades within each group is descriptive of that particu
lar group. Chart Va shows the percentage distribution of the grades of
subcutaneous fat for children under 3 years of age— 563 Puerto Rican
children and 918 New Haven children, graphically recorded according
to the scale descriptive of each group. It will be seen that, in spite of
the differences in the descriptive scales, the Puerto Rican children
have less satisfactory amounts of subcutaneous fat than the New

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELO PM EN T

OF BON ES

C H IL D R E N

PUERTO

R IC O

Haven children and also that there is a proportion of very poorly
nourished. Puerto Rican children that has no counterpart in the New
Haven group. If, moreover, a readjustment of Puerto Rican grades
were made by dropping each grade one point in order to bring the whole
scale more nearly into correspondence with that used in New Haven,
a better idea would be given of the actual contrast between the rea
sonably good condition of the New Haven children and the markedly
unsatisfactory condition of the Puerto Ricans.
RELATION OF AMOUNT OF SUBCUTANEOUS FAT TO SEX, AGE, AND SKIN
p ig m e n t a t io n

The observers found no differences between boys and girls with
regard to amount of subcutaneous fat. Study of the children by age
groups, however, showed that the amounts of fat were on the whole
more satisfactory for those who were mider 7 months of age at the
time of examination. That breast feeding is probably responsible for
Per cent

100

P u erto Rico
563 children

New Haven
918 children
•Subcutaneous f a t

3 Very ^ood
E 2 S 3 Poor

3 Good

Fair

Very poor

C hart Va —Percentage distribution of grades of subcutaneous fat of children examined in Puerto
R ico, compared with that of children examined m N ew Haven

this, and poor artificial feeding for the less satisfactory nutritional
condition found in the later months, is a reasonable assumption; this
will be discussed later.
.
■■ > '
EkB
It was the impression of the examiners that the children with. the
darkest pigmentation of the skin, that is, those who were predomi
nantly Negro, had more satisfactory amounts of subcutaneous fat than
those with lighter skin pigmentation, that is, those who were predomi
nantly Spanish. This was shown to be true of the very dark group,
since 26 out of 29 children in this group belonged in the “ very good’
or “ good” grade for subcutaneous fat, and none in either the “ poor’
or the “ very poor.” The light and medium groups, on the other hand,
each had fewer children in the “ very good” and “ good” grades than
did the dark and very dark groups combined. It may be that the
darker-skinned children tend to thrive better in the tropical climate
than do the lighter-skinned ones. Whether or not this is because the
darker groups are indigenous to the Tropics and can therefore thrive,
whereas the lighter groups whose forebears came from temperate
regions can not, is not known. It is, of course, possible that racial
factors account for the better development of the darker children.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

P H Y S IC A L

C O N D IT IO N

O F C H IL D R E N

E X A M IN E D

That exposure to sunlight may affect the nutritional processes
directly, as in the prevention of rickets, is, of course, an accepted fact;
that it may also affect other aspects of the nutritional process in some
less direct way would seem to be possible. In an effort to find out
whether recent exposure to sunlight had affected the child’s general
nutritional condition, as evidenced by the amount of subcutaneous fat
found at examination, the amount of subcutaneous fat found in chil
dren who were tanned was compared with the amount found in those
who were not tanned. A larger proportion of the children who were
tanned than of those who were not tanned had relatively good amounts
of subcutaneous fat. Whether in this group of children living in the
Tropics there is any real relation between exposure to sunlight and
general nutritional condition as shown by the amount of subcutaneous
fat, or whether the racial factors represented by the natural degree of
pigmentation of the skin are of greater significance than the superim
posed tanning, can not be determined from this small group of cases;
but it seems clear that there is some association between the deeper
degrees of pigmentation of the skin and the better amounts of subcu
taneous fat.
MUSCULAR DEVELOPMENT

Since a child’s muscular development is one of the indexes of his
general physical condition and since it is commonly believed to be
disturbed when rickets develops, an attempt was made to observe and
estimate in a general way the muscular condition of each child at the
physical examination, and to record the stage that he had attained in
motor development.
It was the impression of the physicians that many more children in
Puerto Rico than in New Haven showed weak, flabby muscles and
relaxation of the joints and ligaments. The lack of good muscular
development shown by the Puerto Rican children was perhaps most
striking in their lack o f resistance when being handled during the
course of the examination. Many lay completely relaxed through the
various procedures of the examination, without showing the ordinary
child’s resistance to being measured or to having a roentgenogram
taken. The most marked cases of muscular relaxation were found in
the children who were most malnourished. Poor muscle tone and
relaxation of joints and ligaments seemed in general to accompany the
less satisfactory amounts of subcutaneous fat; but, as will be shown
later, there was no evidence that rickets had anything to do with the
muscular condition.
M uscle tone and relaxation of joints and ligaments.

Nearly half (45 per cent) of the 553 children for whom a report was
obtained showed either poor muscle tone or relaxation of joints and
ligaments, or both. Of those who showed poor muscle tone, more
than three-quarters, as would be expected, also showed relaxation of
joints and ligaments. Poor muscle tone and relaxation both occurred
more often in children who were over 6 months of age than in those
who were under 6 months.
The association between unsatisfactory amounts of subcutaneous
fat and poor muscular development, which seemed to exist at the time
of examination, was borne out by the later analysis. Three-fourths of
the children with good muscle tone and nearly three-fourths of those
with no relaxation of joints and ligaments were in the groups having
160320°—33------3

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF BON ES

O F C H IL D R E N

PUERTO

R IC O

the more satisfactory amounts of subcutaneous fat. In all age groups
the child with poor muscular development was, in general, the one with
the less satisfactory amount of subcutaneous fat.
Pot-belly.

One of the usual manifestations of poor muscular development is
pot-belly. It is common knowledge that this frequently occurs in the
Tropics. Of the 578 children for whom a report was obtained on this
item, 278 (48 per cent) had pot-belly, 119 of the 578 (21 per cent) to a
moderate or marked degree. As will be shown later in the discussion
of the clinical diagnosis of rickets, the presence of pot-belly, especially
one of a moderate or marked degree, was occasionally the clinical evi
dence that swung a diagnosis from negative to questionable or from
questionable to positive. It is true that in northern climates pot-belly
frequently accompanies rickets, but that it may accompany other less
specific disturbances of nutrition is often overlooked. When the
child’s diet consists largely of carbohydrates, as it does in the Tropics,
the size of the pot-belly is undoubtedly increased by the distension of
the intestines. Practically half of the Puerto Rican children showed
evidence of this poor musculature of the abdominal wall. That pot
belly in these children was associated with poor nutrition and not with
rickets is made clear by the negligible amount of rickets found at
roentgen-ray examination.
Motor development: Age of holding up head, sitting up, standing, and walking.

The progress in the muscular development of a child may be judged,
though perhaps somewhat roughly, by the age at which he is able to
perform certain acts requiring muscular strength and coordination of
different muscle groups, such as holding up the head steadily when the
trunk is supported, sitting up unsupported on a firm surface, standing
alone, or walking without support. Since the first performance of
these various acts takes place usually at rather definite stages during
the first 15 or 18 months of a child’s life, the ages at which such first
performances occur may be taken in a general way as an indication of
the progress of an individual child’s muscular development, and these
ages may be used to compare one group of children with another.
Muscular development necessary to perform these acts at the usual
age is dependent upon good health, proper nutrition, normal innerva
tion of the groups of muscles used, and normal mental development.
Data concerning the ages at which these Puerto Rican children first
held up their heads, sat up alone, stood alone, or walked alone were
obtained from the mothers at the time of the home visit. Many of
the mothers had not observed carefully or could not remember exactly
when their children were able to do these various things; therefore the
ages reported should be considered as approximate.
Practically no sex differences were found in the ages at which each
stage of motor development was accomplished, and therefore, the num
ber of boys and girls have been combined for the sake of uniformity in
presentation of the data.
Tables 2, 3, 4, and 5 show the ages at which the Puerto Rican chil
dren were reported to have held up their heads, sat up, stood, and
walked, in comparison with the ages at which a group of unselected
New Haven children were reported to have done these things.38b The
38k The N ew Haven group includes children of older ages than the Puerto Rican, but the proportion
completing the various stages of motor development at the specified ages is probably not influenced b y
the differences in the age distribution. For a comparison of the age at examination of the Puerto Rican
and N ew Haven children see Table 12.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

P H Y S IC A L

C O N D IT IO N

C H IL D R E N

E X A M IN E D

Puerto Rican children tended to stand and walk somewhat earlier than
did the New Haven children. The ages at which Puerto Rican chil
dren first held up their heads and first sat up are somewhat more scat
tered in their percentage distribution than the ages at which the New
Haven children reached the same stages of development.
T a b l e 2.— Age at holding up head; children examined in Puerto Rico and in New

Haven, Conn.
Children examined
in Puerto R ico

Children examined
in N ew Haven

Age 1 at holding up head, as reported by'm other
Num ber

T otal_______________________

Per cent
distri
bution

584

Age reported___________________
Under 2 m onths________________
2 months______ ____________
3 months_________________
4 months_________________
5 months and over___________
Age not rep orted ..____ _____________
N ot holding up head at date of examination. . .
N ot reported whether holding up h e a d _____

N um ber

Per cent
distri
bution

1,186

448

100

597

100

22
155
152
63
56

5
35
34
14
13

18
168
262
110
39

3
28
44
18
7

56
52
28

481
104
4

,J A g e is given as of nearest month; that is, “ under 2 m onths” is actually under 1 month and 16 davs
2 months is from 1 month and 16 days to 2 months and 15 days, inclusive, and so on.

T a b l e 3. — Age at sitting alone; children examined in Puerto Rico and in New

Haven, Conn.
Children examined
in Puerto Rico

Children examined
in N ew Haven

Age 1at sitting alone, as reported b y mother
Number

T o ta l........................ .
Age reported_________________
Under 5 m onths____________
5 m onths___________
6 m onths____ __________
7 months________ ____
8 m onths___________
9 months and over______
Age not reported___________
N ot sitting alone at date of examination.
N ot reported whether sitting alone____
<
5 months

Per cent
distri
bution

584

Number

Per cent
distri
bution

1,186

346

100

697

100

25
59
110
55
50
47

7
17
32
16
14
14

44
92
260
146
78
77

6
13
37
21
11
11

20
190
28

215
270
4

®f? eare! t moni ^ L ^ *?» “ under 5 m onths” is actually under 4 months and 16 days,
is from 4 months and 16 days to 6 m onths and 15 days, inclusive, and so on.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF BON ES

C H IL D R E N

PUERTO

R IC O

T a b l e 4.— Age at standing alone; children examined in Puerto Rico and in New

Haven, Conn.
Children examined
in Puerto R ico
A g e 1 at standing alone, as reported b y mother
Number

Children examined
in N ew Haven

Per cent
distri- - Num ber
bution

Per cent
distri
bution

1,186

Total________________________ ______ - ------- ---------------------

584

Age reported__________________________________________________

252

100

479

100

36
56
48
48
29
25
10

14
22
19
19
12
10
4

25
50
82
104
76
68
74

5
10
17
22
16
14
15

237
466
4

31
273
28

iAge is given as of nearest m onth; that is, “ under 8 m onths” is actually under 7 months and 16 days,
“ 8 m onths” is from 7 months and 16 days to 8 months and 15 days, inclusive, and so on.

T a b l e 5. — Age at walking alone; children examined in Puerto Rico and in New

Haven, Conn.
Children examined
in Puerto R ico
A g e 1 at walking alone, as reported b y mother
Num ber

Per cent
distri
bution

Children examined
in N ew Haven

N um ber

Per cent
distri
bution

1,186

584
Age reported.................... - .........- .............................................r.........

219

100

520

100

Under 10 months__________________________________________
10 months------------ ------- -----------------------------------------------------11 months_________________________________________________
12 months------------------------------- ------- -----------------------------------13 months--------------------------------------------------------------------------14 months_____________________ ___________________ _______
15 months and over______ *------------------------- -----------------------

35
38
18
71
26
6
25

16
17
8
32
12
3
11

27
38
56
104
75
85
135

5
7
11
20
14
16
26

8
329
28

48
614
4

iAge is given as of nearest m onth; that is, “ under 10 m onths” is actually under 9 months and 16 days,
“ 10 m onths” is from 9 months and 16 days to 10 months and 15 days, inclusive, and so on.

Further comparison of the average age at which each stage of motor
development was reached by the children in Puerto Rico and by those
in New Haven shows differences which, though slight, indicate the
earlier development of the Puerto Rican children. The average age
of holding up the head for Puerto Rican children was 3 months; of
sitting up, 6.7 months; of standing, 9.5 months; and of walking, 11.9
months. The average age of holding up the head for New Haven
children was 3 months; of sitting up, 6.6 months; of standing, 10.6
months; and of walking, 13.5 months. There is no difference between
the averages of the two groups with respect to holding up the head and
sitting, but there is a significant difference in favor of the Puerto

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

P H Y S IC A L

C O N D IT IO N

C H IL D R E N

E X A M IN E D

Rican children of about a month in the average age of standing and
of walking.
It is probable that various factors, many of which can not be eval
uated, have to do with this tendency to slight precocity on the part
of the Puerto Rican children. It would seem possible, however, that
the slower growth in height and weight and the almost complete
absence of rickets in the Puerto Rican group may have some direct
bearing on the situation, since it is pretty generally recognized that
children of the small, slender, wiry type tend to develop in motor
skill earlier than those of the large, heavy type, and also that motor
development may be delayed when rickets is in its active stages. The
poor muscle tone and the greater degree of relaxation of the joints
and ligaments observed at examination of these Puerto Rican chil
dren apparently did not affect their motor development. To what
extent tropical sunlight (apart from its antirachitic effect), or dif
ferences in race, or habitual diet may influence muscular develop
ment can not, of course, be evaluated in such a small group of cases.
ONSET OF DENTITION

Closely associated with the growth and development of the skeleton
in infancy is the development of the teeth. The age of eruption of the
first deciduous teeth is often used as one gauge of physical development,
and delay in eruption is thought to be one of the evidences of rickets.
The age of eruption of the first teeth was reported by the mothers of
336 Puerto Rican children and is shown in Table 6. This table shows
also the age of eruption of the first teeth for 706 New Haven children.
Comparison shows that the teeth of Puerto Rican children tend to
erupt earlier than those of New Haven children. The average age
of first eruption of teeth reported for Puerto Rican children was 7
months, and that for New Haven children 7.3 months.
T a b l e 6 . — Age

at onset of dentition; children examined in Puerto Rico and in
New Haven, Conn.
Children examined
in Puerto R ico

Age 1 at onset of dentition, as reported b y mother
Number

Per cent
distri
bution

Children examined
in N ew Haven
N um ber

Per cent
distri
bution

T otal................................................. .

584

reported..................................................

336

100

706

100

Under 5 m onths_____________________
5 or 6 m onths..................... ............
7 or 8 m onths_______________________
9 or 10 m onths................................
11 months and over_________

60
108
96
44
28

18
32
29
13
8

76
231
199
126
74

11
33
28
18
10

Age not reported________ _________________
N o sign of teeth at date of exam ination..........
N o report on onset of den tition ...

15
205
28

Age

143
328
9

I Age is given as of nearest month; that is, “ under 5 m onths” is actually under 4 months and 16 days.
6 or 6 m onths” is from 4 months and 16 days to 6 months and 15 days, inclusive, and so on.

COLOR OF MUCOUS MEMBRANES

Because of the generally poor physical condition of the children
and because of the high incidence in Puerto Rico of malaria, hookworm

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF BONES

C H IL D R E N

PUERTO

R IC O

disease, syphilis, and tuberculosis, any of which may be accompanied
by anemia, it was thought worth while to make a rough estimate of the
presence or absence of anemia as indicated by the pallor of mucous
membranes. Though it would have been desirable to make accurate
diagnoses of anemia by examination of the number of red blood cells
and the amount of hemoglobin in the blood, such procedure was
obviously impossible because of the limited time of the study. It was
believed, however, that pallor of the mucous membranes of the eyelids
and lips might be regarded as fair evidence of anemia, though not
evidence from which the degree of anemia could be judged accurately.
Pallor of the mucous membranes of the eyelids and lips was found
in 122 children (22 per cent of the 552 children for whom the condition
was reported upon), a proportion strikingly higher than that found in
a study of children in a city in continental United States.39 In the
latter study about 8 per cent of the children from 2 to 7 years of age
showed pallor, and only 2.2 per cent of the children under 2 years.
Among the Puerto Rican children examined, pallor was found in 27 per
cent of the children 12 months of age and under and in 16 per cent of
those 13 to 34 months of age. Pallor of the mucous membranes in
these infants and young children was found to be associated with the
less satisfactory amounts of subcutaneous fat, with poor muscular
development, and with lack of evidence of exposure to sunlight.
Only 14 per cent of the 315 children with “ very good” or “ good”
amounts of subcutaneous fat showed pallor of the mucous membranes,
whereas 49 per cent of the 71 children with the “ poor” or “ very poor”
amounts showed this pallor.
It seems likely that such diseases as syphilis and tuberculosis and
nutritional disturbances of infancy were the more important causes of
this pallor. Most of the children examined lived in cities where the
incidence of hookworm disease was relatively low. Moreover, only 6
per cent of the children for whom a report was obtained were found to
have enlarged spleens— a low “ spleen index” for a country in which
malaria is more or less prevalent. Malaria, therefore, was probably a
minor factor in the production of this pallor, since the spleen index
of malaria is usually considerably higher in Puerto Rico, sometimes as
high as 75 per cent in a community where malaria is prevalent.40
The association between pallor of the mucous membranes and lack
of tanning of the skin is in all probability a significant one. The
presence or absence of pallor and of tanning was reported for 545
children. Seventy-three per cent of the children whose mucous
membranes were of good color were tanned, whereas only 48 per cent
of those with pallor of mucous membranes were tanned. Apparently
both exposure to sunlight and the grade of the child’s nutritional
condition as shown by the amount of subcutaneous fat are related to
the color of the mucous membranes. The largest proportion of chil
dren with mucous membranes of good color were those with tanned
skins and well-nourished bodies; the largest proportion of pale chil
dren were those with no tanning and less well-nourished bodies.
OTHER PHYSICAL FINDINGS

During the course of physical examination certain findings other
than those which bore a direct relation to the child’s nutritional condi8» Physical Status of Preschool Children, Gary, Ind., b y Anna E . Rude, M . D ., pp. 39, 78. U . S. Chil
dren’s Bureau Publication N o. 111. Washington, 1924.
40 Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1925, p. 80.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

P H Y S IC A L

C O N D IT IO N

O F C H IL D R E N

E X A M IN E D

tion were noted. Evidences of infection were specially sought, though
as has been noted, the time at the disposal of the examiners did not
permit the use of special tests, such as the tuberculin or the Wassermann tests.
The diagnoses, other than that of rickets, which were made at the
time of the physical examination have been grouped as follows:
Nutritional disturbances_________ 63
54
Malnutrition___________
Gastroenteritis_________
9
Communicable diseases__________ 188
Rhinopharyngitis_________117
Laryngitis______________
2
Bronchitis______________ 27
Broncho-pneumonia____
i
Tuberculosis____________
2
Tuberculosis, suspected.. 16
Fever, unexplained (prob
ably due to communi
cable disease)________
7
Erysipelas______________
i
Mumps_________________
1
Malaria________________
1
Malaria, suspected______
2
Nasal diphtheria, suspec
ted___________________
1
Congenital syphilis_____
3 C»\\V
Congenital syphilis, suspected________________ «**>5
Dysentery______________
2
Diseases of ears_________________
3
Chronic otitis media____
3
Diseases of nose and throat______ 138
Hypertrophied tonsils or
hypertrophied to n s ils
and adenoids_________ 132
Hypertrophied adenoids
only-------- -----------------6

Diseases of the eyes_____________
Corneal scar and ectro
pion_________________
1

Conjunctivitis_________
Gonorrheal conjunctivi
tis___________________

Stye_________________

11
1

3
Internal strabismus_____
Diseases of central nervous
s y s t e m .... ________ ____ _
9
Poliomyelitis, old, with
paralysis_____ _______
2
3
Birth injury, probable__
H ydrocephalos-c. . A ____i l f f
Fapiajtea$öysäs. . *t
y * *1
Gemt^-^rma*ry
_____
9
\^. W ^ in itis^ d ^ ilM ____
2
IngöQ&Plrernia5
(a\ ü f l y a r o c e l e .
2
iS*nn diseasfijkaUPr’ V__________
62

----------------451
\jUHemangioma__
________
Other skin diseases_____
16
Congenital defects_______________ 52
Congenital heart dis
ease__________________
6
Mongolian idiocy_______
1
Mongolian idiocy, ques
tionable______________
1
Supernumerary fingers. _

Tongue-tie_____________
Harelip_________________

41
1

The largest group of pathological conditions is that of communicable
or mfectious diseases because of the inclusion under this heading of
the 117 cases of simple rhinopharyngitis (common cold). In a number
of cases diagnoses of tuberculosis, diphtheria, or congenital syphilis
were suspected, but could not be confirmed. Only 3 definite clinical
diagnoses of syphilis were made at examination, but roentgenograms
of the bones of the forearms showed syphilis to be present in 6 cases.
The incidence of positive Wassermann tests in the parents has already
been discussed. (See p. 12.)
Of 565 children whose tonsils, adenoids, and cervical lymph nodes
were examined, 427 (76 per cent) had no obvious defect of tonsils
or adenoids, 100 (18 per cent) had hypertrophied tonsils only, 31
(5 per cent) had both hypertrophied tonsils and hypertrophied ade
noids, 6 children had hypertrophied adenoids only, and 1 child had
hypertrophied and diseased tonsils, with no involvement of the ade
noids reported. ^ Comparison of these findings with those made by
the same physicians in New Haven shows a considerably smaller pro
portion of children in Puerto Rico having hypertrophied or diseased
tonsils or adenoids. In New Haven 39 per cent of the children of
an age comparable to the Puerto Rico group had tonsils that were
either enlarged or diseased, and 14 per cent had adenoids that either
were definitely enlarged or seemed to be so. Whether the differences

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF B O N ES

O F C H IL D R E N

PUERTO

R IC O

in climate between New Haven and Puerto Rico can account for the
difference in the proportions of hypertrophied or diseased tonsils found
in these two communities can only be surmised.
Among the children examined were a relatively large number with
skin infections, chiefly impetigenous in nature, and with conjunctivitis.
An unusual number of children were found to have tongue-tie; no
explanation of the great frequency of this defect was apparent. _
In connection with these diagnoses made at the examination, it is
of interest to note the diseases that these same children were reported
to have had previous to the time of the study. Five hundred and
thirty-two children were reported to have had one or more diseases;
only 16 were reported to have had none. The diseases reported have
been grouped as follows:
Colds___________________________ 499
Diseases of digestive system-------- 356
Diseases of skin---------------------------- 257
Diseases of respiratorysystem .___ 148
Communicable diseases, positive
or suspected---------------------------- 83
Diseases of eyes-------------------------- 68 i

Earaches and otitis media---------Wasting diseases------------------------Convulsions_____________________
Undefined fevers_________________
Diseases of nervous system--------Others___________________________

57
51
51
23
9
37

Colds were extremely prevalent among these children. Some form
of disturbance of the digestive system was reported for two-thirds of
them, of skin disease for about one-half, and of respiratory disease for
more than one-fourth. The group of “ wasting diseases” included
various poorly defined conditions, locally known as r a q u i t i s m o , or
f a t i g u e , that were probably due to frequent digestive disturbances or
starvation or to tuberculosis, undiagnosed. It was noticeable that
the digestive disturbances in children under 1 year of age were less
frequent among breast-fed infants than among those artificially fed.
That communicable diseases other than colds were reported by rela
tively few mothers probably may be accounted for by the fact that
the children were young and therefore had been comparatively little
exposed to such diseases or by the fact that in many cases of illness
they were not seen by a physician.
.
Because of the well-recognized association of the convulsions of tet
any with rickets, it is of interest to note the high proportion (9 per
cent) of children reported to have had convulsions in a group known
to show, on the whole, almost no evidence of active rickets at the
time of examination. The occurrence of a convulsion instead of a
chill at the onset of an acute infectious disease is common in infancy
and early childhood and may account partly for the relatively large
number of convulsions reported.
Some comment may be made on the large number of cases of skin
infections and of diseases of the eyes, and the question raised as to a
possible relation between them and the inadequate diets that many
of the. children were receiving. The association between vitamin-A
deficiency and xerophthalmia among children whose diets are low (or
entirely lacking) in milk, butter, eggs, and green vegetables is well
known; the association between vitamin-A deficiency and skin and
respiratory infections is less well defined. That the children examined
were receiving diets deficient in these foods will be pointed out later
(p. 94),

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

IN C ID E N C E O F R IC K E T S IN P U E R T O R IC O

♦

It has already been pointed out that the study here reported was
undertaken in order to observe the roentgenographic appearance of the
bones of infants living under the influence of tropical sunlight and to
make comparisons between the roentgenographic appearance of the
bones of such infants and that of the bones of infants examined pre
viously in New Haven, Conn. Though the selection of Puerto Rico
as the place in which to undertake this study of normal infants’ bones
was based on the presumption that the intensity of its sunlight would
be adequate to prevent the development of rickets, nevertheless it was
necessary to confirm this presumption by a careful study of the actual
incidence of rickets in the island before the data there collected could
be used with certainty as a normal control for the data collected in
New Haven.
In the New Haven study it had been found that a relatively large
proportion of infants, even though given what was thought to be a
satisfactory amount of antirachitic treatment, showed sooner or later,
if examined repeatedly by roentgen ray, certain deviations from the
normal, which were interpreted as evidences of slight rickets. In the
course of that study, moreover, the question was continually being
raised whether these slight bone changes were in reality those of
rickets or whether they were variations of normally growing bones.
The study in Puerto Rico was undertaken in the belief that the
presence of similar slight deviations from the preconceived normal
in a tropical region, where presumably the sunlight was adequate to
prevent rickets, would indicate that such deviations are not rickets
but are variations within the normal, whereas the absence of these
changes in Puerto Rico would tend strongly to confirm the opinion
formed during the New Haven study that they are in reality the
roentgenographic evidences of early rickets.
From a clinical point of view, also, questions had arisen with regard
to the correct interpretation of certain skeletal findings. Was it
possible that some of the slight skeletal signs, interpreted as those of
rickets, were only variations in normal skeletal growth? It was
hoped, when the study in Puerto Rico was undertaken, that the
roentgenograms would establish clearly the differentiation between the
roentgenographic appearance of the bones of normal infants and that
of the bones of infants with slight or early rickets. It was hoped
also that a better understanding of the clinical diagnosis of rickets
could be reached. It was, of course, essential that the study should
be carried out along the lines followed in New Haven.
A review of current literature shows that there has been some
difference of opinion on the question of the prevalence of rickets in
tropical regions. Hess41 has reported his observations on rickets,
made in 1928 in the West Indies, in Panama, and in Costa Rica. In
Kingston, Jamaica, he found among the children at the General Hos
pital “ many suffering from mild rickets, according to clinical criteria,
although they did not present a rachitic appearance or pronounced
‘ i Hess, Alfred F .: Rickets, Osteomalacia, and Tetany, pp. 54-55.

Lea & Febiger, Philadelphia, 1929.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

V
36

DEVELOPM ENT

OF BO N ES

O F C H IL D R E N

PUERTO

R IC O

bowing of the legs. This held true also for the children of the babies’
welfare clinic.” In Trinidad, at the General Hospital, he found that
“ mild rickets was of common occurrence, but no case of severe rickets
was seen.” Here, also, he found rickets in fully half the babies (all
colored) that he examined at the babies’ welfare clinic.
In hospital wards in San Jose, Costa Rica, Hess found “ considerable
rickets, some even of moderate degree,” and on the Isthmus of Panama
he found the same in two hospitals. In the hospital at Ancon, Canal
Zone, conducted by the United States Government, he saw “ roent
genograms of rickets of mild and even marked degree which had
developed in the Canal Zone.” He later studied roentgenograms of
the epiphyses of about 100 children living in the region of the Canal
Zone and found that eight showed definite signs of rickets, most of
these children being between the ages of 4 and 6 months; six other
children under 7 months showed questionable signs at the epiphyses.42
Furthermore, with regard to the protection of infants in these
regions, Hess says: “ But even under favorable circumstances, unless
the babies are taken out of doors, they do not receive sufficient ultra
violet rays to protect them against rickets.” With the exception of
the conclusions based on roentgenograms of children in the Canal
Zone, Hess apparently based his conclusions on clinical observation
only.
Other observers quoted by Hess apparently had never seen rickets
in the West Indies or in Trinidad. The differences in opinion were
probably due to differences in the interpretation of clinical findings,
a subject to which this report will refer later. An article by Torroella43
and one by Gonzales44 state that no case of rickets was found during
the examination of 6,000 children in Mexico.
In Puerto Rico, prior to the time of the investigation here reported,
a large number of infants were reported every year to the insular
department of health as having died of rickets. It has long been the
belief of the physicians of the department of health that these deaths
should not have been so classified and that the confusion had resulted
from the anglicization of the Spanish term “ raquitismo” — meaning
malnutrition— to rachitis or rickets. In actual practice in Puerto
Rico any severe wasting disease in infancy is called “ raquitismo,”
just as in continental United States such a disease is called marasmus.
The present investigation was, therefore, specially welcomed by the
Department of Health of Puerto Rico, since it would help to clear up
the question of whether or not rickets actually occurs in the island
and since it might lead to greater accuracy in vital statistics. That
the term “ rickets” as used on death certificates in Puerto Rico is a
misnomer was shown clearly by the present investigation. Its use in
the vital-statistics reports was discontinued45in the year following the
Children’s Bureau study.
42 Attention m ay be called to the fact that climatic conditions in the Isthmus of Panama are different from
those in Puerto R ico and m ay well account for the rickets found there b y Hess. In Colon, Republic of
Panama, the daily average of actual hours of sunlight is 6.6, and in Ancon, Canal Zone, it is 5.9; both of these
figures are lower than the average in either San Juan, P. R ., or N ew Haven, Conn.
(See p. 4.) During
the rainy season, which lasts seven or eight months, the m onthly averages of actual hours of sunshine in
Colon range from 4.6 to 6.7 daily. In Ancon, corresponding averages range from 4.2 to 4.8. Similar low
figures are reached in N ew Haven only in the late fall and winter months, and in Puerto R ico not at all.
(See Climatological Data for Central America, b y W . W . Reed, in M on th ly Weather Review, vol. 51, no. 3,
M arch, 1923, pp. 133-141, published b y the Weather Bureau, U. S. Department of Agriculture, Washington.)
43 Torroella, Mario A .: Raisons pour lesquelles le rachitisme n ’existe pas au Mexique. Archives de
M édecine des Enfants, vol. 32, no. 5 (M ay, 1929), pp. 262-269.
44 Gonzales, Martin: Raisons pour lesquelles le rachitisme n ’existe pas au Mexique. Bulletins de la
Société de Pédiatrie de Paris, vol. 27, Jan. 15,1929, pp. 42-54.
43 See Report of the Commissioner of Health of Puerto R ico for the Fiscal Year 1927, p. 132, and later
reports.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

IN C I D E N C E

O F R IC K E T S

NUMBER AND AGE DISTRIBUTION OF CHILDREN EXAMINED FOR
RICKETS

Examination for rickets was made of 584 Puerto Rican children
ranging in age from 1 to 34 months. Of these, 171 were under 7
months of age, 149 were 7 and under 13 months, 180 were 13 and
under 25 months, and 64 were 25 months or over; the ages of 20
were not known. The preponderance of children under 13 months
of age was a matter of selection, since the first year of life, and espe
cially the first six months, is the period when the changes in the bones
interpreted as the earliest evidence of rickets had been found in New
Haven children, and therefore the period for which examinations were
most desired in Puerto Rican children who were presumably free from
rickets. The number of children examined in the second and third
years was relatively small, but the results of the examinations indicate
to some extent the incidence of rickets at this age.
M ETHODS OF EXAMINATION FOR RICKETS

When as accurate a diagnosis as possible is desired of the presence of
rickets, not only should a physical examination be made to detect any
clinical evidence of the disease, but roentgenograms of the long bones
should be taken to determine whether any evidence of the disease is
present and, if so, to determine the activity and severity of the process,
and also to obtain a graphic record of deformities such as bowing of
the bones of the legs. Supplementary chemical examination of the
blood will aid the examiner in differentiating between an active process
and one that is subsiding or healing, and also in making a very early
diagnosis when rickets is suspected because of craniotabes or enlarged
costochondral junctions but is not yet demonstrable in the roentgeno
grams of the radius and ulna.
In the investigation in Puerto Rico physical examinations were made
of all children, the investigators laying special emphasis on examina
tion for the signs generally accepted as manifestations of rickets.
Roentgenograms were taken of the bones of the forearm only, since,
as will be explained later, roentgenograms of these bones are the most
satisfactory ones in making an early diagnosis of rickets. Opinions
with regard to the rachitic origin of deformities such as bowlegs and
knock-knees were formed from physical examination alone. In a few
cases, as an aid to diagnosis, chemical tests were made to determine
the calcium and inorganic phosphorus content of the blood serum.
Many elements enter into a clinical examination for rickets, mak
ing it a less reliable method of determining the real incidence of
the disease in a community than a roentgen-ray examination. The
fact that the child’s skeleton is constantly changing with growth, the
fact that many of the evidences of mild rickets are but slight varia
tions from the normal, and the fact that the clinical judgments of
different observers vary because these limits of normal can not be
arbitrarily defined and because the observers differ in their training
and experience make unreliable any study of the incidence of rickets
based on clinical examination alone. On the other hand, roentgenographic examination has the advantage of presenting permanent
records of certain attributes of the bones, which can be studied and
reappraised by several observers and for which the limits of normal
may be more clearly defined.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF BON ES

O F C H IL D R E N

PUERTO

R IC O

If a true picture of the incidence of rickets in a community is to
be obtained, however, as many children as possible should be examined
by both methods during the period of life when the disease is usually
active—that is, the first two years of life—and also during the third
and possibly the fourth year, when the disease is as a rule no longer
active, but when the clinical evidences of previous disease should be
found approximately in proportion to the roentgen-ray evidence of
activity of the disease in earlier age periods. The incidence of rickets
found by roentgenographic examination during the first year or two of
life should form, it is believed, some sort of background for the inci
dence to be found by clinical examination coincidentally or in later
years. It should be remembered, however, that the incidence of rickets
found at clinical examination at any given age is the cumulative
record of the more or less permanent effect on children’s skeletons of a
transitory disease which may have come and gone before that age or
which may still be active at the time of examination. Moreover, as
will be pointed out later, the total number of clinical diagnoses of
rickets probably includes a certain proportion of diagnoses of mild
rickets that are incorrectly made, since they are based on skeletal signs
not easily differentiated from those of normal growth and development.
DIAGNOSIS OF RICKETS BY CLINICAL EXAMINATION

Since it seemed desirable to compare the results of the physical
examinations in Puerto Rico with those in New Haven, every effort
was made to interpret the skeletal findings in Puerto Rico just as they
had been interpreted in New Haven. In retrospect it is believed that
in the effort not to overlook any of the clinical evidences of possible
rickets, too much emphasis may have been placed on slight varia
tions, and a clinical diagnosis of rickets may have been made in some
cases in which it was not warranted. Since, however, similar con
clusions had undoubtedly been drawn in New Haven, the clinical
diagnoses have been retained as made, and have been analyzed in
some detail to show how varied and inconsistent were the observations
upon which the diagnoses were based. It should be borné in mind,
however, that the phrase “ clinical diagnosis of rickets” as used in
this study does not mean that rickets was necessarily present as an
active disease; indeed, in a majority of cases the evidence was only
such as might have been produced by previous rickets of a mild
degree, and few cases of active rickets were even suspected at
physical examination.
INCIDENCE OF RICKETS AT CLINICAL EXAMINATION AND ITS RELATION TO AGE

Examinations for clinical evidences of rickets were made on 584
children. The diagnoses made and the ages of the children at the
time of examination are shown in Table 7. It will be seen that in only
50 children (9 per cent) was the evidence sufficient to lead to a definite
diagnosis of rickets, in 134 it was sufficient to lead to a questionable
diagnosis, and in 400 there was no evidence of rickets. Of the 50 cases
in which the clinical diagnosis of rickets was definite, 46 were consid
ered to be of slight degree, 3 of moderate degree, and 1 of marked
degree. (See case 1, Appendix E, p. 120.)
It will be seen also from Table 7 that the proportion of children
showing signs interpreted as evidences of rickets or of questionable
rickets was considerably smaller in the age group under 7 months than

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

IN C ID E N C E

O F R IC K E T S

in any other, and that the proportion showing no evidence of rickets
was largest in the age group under 7 months. Though this is not
incompatible with the usual increase in the incidence of clinical evi
dences of rickets during the later months of infancy and the early
months of childhood, the fact that it occurred in these Puerto Rican
children who were exposed constantly to intense sunlight suggests
that in some cases the process of growth and development of the bones
may alone account for the slight signs commonly used as a basis for
the clinical diagnosis of mild rickets.
T a b l e 7.— Clinical diagnosis of rickets and age of child at examination; children

examined in Puerto Rico
Children examined in Puerto Rico
Clinical diagnosis of rickets

Age 1at examination

Total N o rickets

Question
able
diagnosis

Slight

Moderate

Marked

Num - Per Num - Per Num- Per N um - Per Num - Per
ber c e n t 2 ber c e n t 2 ber c e n t 2 ber c e n t2 ber c e n t 2
584

400

134

171
149
104
140
20

143
94
65
83
15

84
63
63
59

25
39
26
39
5

15
26
25
28

3
14
12
17

2
9
12
12

1
1
1

(3)
1

1 Age is given as of nearest month; that is, "u n d er 7 m onths” is actually under 6 months and 16 days, “ 7
to 12 m onths” is from 6 months and 16 days to 12 months and 16 days, inclusive, and so on.
1 Per cent not shown where number of children was less than 50.
* Less than 1 per cent.

PHYSICAL SIGNS USED AS BASIS FOR CLINICAL DIAGNOSIS OF RICKETS

The clinical diagnoses of rickets made in Puerto Rico were based in
large part upon certain skeletal signs usually considered to be evi
dences of the disease, but also to some extent upon certain signs of
muscular weakness, which were regarded as secondary rather than
primary diagnostic evidence. Analysis of the data shows that both
the number of these signs and the degree of deformity presented by
each were taken into consideration when the diagnosis was made. It
is evident, moreover, from study of the records, that some skeletal
signs appeared much more frequently than others and that certain
signs or combinations of signs clearly were given more weight by the
examiners in the diagnosis of rickets than others. When the diagno
sis of rickets was questioned, it was done either because the skeletal
signs were few, usually only one or two having been found, or because
they were present in such slight degree that their significance was
questioned. The difference in the degree in which the signs were
present probably accounted most frequently for the differences in the
diagnoses made when the same group of skeletal signs were present.
Because of the great variety of factors entering into a clinical diagnosis
of rickets it is to be expected that many inconsistencies in diagno
sis would appear. Lastly, it should not be forgotten that a diagnosis
often depends on the observer’s impression of the child as a whole, a
very subjective and unreliable factor, but one that enters into nearly
every clinical diagnosis and may account for many inconsistencies.
The skeletal signs that were considered as clinical evidence of rickets
were: Enlargement of the costochondral junctions of the ribs, enlarge

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF BONES

C H IL D R E N

PUERTO

R IC O

ment of the epiphyses of the long bones (especially those at the wrist),
bowlegs, knock-knees, Harrison’s groove, pigeon-breast, moderate
enlargement of the parietal or the frontal bosses or both, cranio tabes,
and asymmetry of the head. The evidences of muscular weakness
that were considered as giving contributory evidence of rickets were:
Decreased muscle tone, increased relaxation of the joints and liga
ments, and pot-belly. A scale ranging from what was considered
normal through two or three degrees of enlargement or deformity was
adopted for each sign. That different physicians should use a scale con
sistently is scarcely to be expected. In this study the examinations were
made by two physicians who had made many of the examinations in
New Haven, and, though, undoubtedly, individual judgment varied
to some extent, the conclusions drawn were, on the whole, fairly
consistent.
Relation of number of skeletal and muscular signs to clinical diagnosis o f rickets.

It became apparent from study of the records that the observers
when making the clinical diagnoses took into consideration not only
which skeletal signs were present but also how many. Though analy
sis showed that the observers were influenced more by certain skeletal
signs than by others, it also shows that, regardless of what the signs
were, the greater the number of signs found the more likely was it that
a diagnosis of rickets would be made. (See Appendix Table B 1, p. 110.)
A similar relation was found between the number of signs of muscular
weakness and the clinical diagnosis, indicating probably that these
signs influenced the diagnosis when it was uncertain on the basis of
skeletal signs alone.
Incidence and degree of each skeletal sign.

More important in attempting to understand the basis for the clini
cal diagnosis of rickets than the number of signs is the incidence of
certain skeletal signs and the degree in which they were observed.
The inconsistencies of clinical judgment in making such diagnoses may
be demonstrated also by analyzing the diagnoses made when various
signs or combinations of signs were present. The following detailed
study of the skeletal signs found in Puerto Rican children is given
in an attempt to throw some light on their value in diagnosis, and in
order that comparisons may be made by other observers as to the
relative frequency and severity of these signs in communities having
differing degrees of intensity of sunlight.
The mcidence, in the total group of children examined in Puerto
Rico, of each of the skeletal signs usually considered to be evidence of
rickets compared with the incidence in a group of children of about the
same age examined in New Haven is shown in Table 8. The percent
age incidence of each sign in the two groups of children is as follows:
Puerto New
R ico Haven

Enlarged costochondral junctions___________________________________ 16
Enlarged epiphyses at wrist-----------------------------------------------------------10
Bowlegs______________________________ __________________________—
30
Knock-knees______________________________________________________
28
Harrison’s groove__________________________________________________
6
Pigeon-breast______________________________________________________ (46)
Moderate enlargement of—
Either frontal or parietal bosses_______________________________
7
Both frontal and parietal bosses_______________________________
1
Asymmetry of head________________________________________________
6

Craniotabes------------------

« Less than 1 per cent.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

a N ot reported.

58
55
50
23
25
2
27
15
(47)

IN C ID E N C E
T

able

OF R IC K E T S

8.— Incidence

of skeletal signs usually considered evidence of rickets; children
examined in Puerto Rico and in New Haven, Conn.
584 children examined in Puerto
R ico (1 to 34 months of age)

Skeletal signs

956 children examined in New
Haven (1 to 36 months of age)

Total
Total
Signs N ot re
Signs
Signs N ot re
Signs
re
re
present absent ported ported present absent ported
ported

581
Enlarged costochondral junctions___
583
Enlarged epiphyses at wrist-------------582
B ow legs..______ __________ __________
1303
Knock-knees________________________
583
Harrison’s groove___________________
577
Pigeon-breast__________ ____________
Moderate enlargement of—
Either frontal or parietal bosses.. }
583
Both frontal and parietal bosses..
551
Asym m etry of head-------- '----------------581
Craniotabes_________________________

92
60
176
85
37
1

{

33
5

}

489
523
406
218
546
576

3
1
2
1 281
1
7

536
518
576

33
3

954
952
1 547
i 555
951
954
f
950 \
(3)
945

2
4
9 409
2 401
5
2

549
519
276
127
233
15

405
433
271
428
718
939

259 }
146
(3)
37

545

(3)
908

(3)
11

1 Includes only children who were standing at date of examination.
2 Children not yet standing at date of examination,
s N ot reported.

It will be seen that the incidence of each of the signs, with the excep
tion of knock-knees, is considerably lower in Puerto Rico than in New
Haven. That the high incidence of rickets as shown by the roentgenray examination in the New Haven group (approximately 30 per cent)
plays an important part in this difference is undoubted, but that these
signs may occur also in slight degree in children who have not had
rickets is made evident by the occurrence of these signs in Puerto Rico,
where, as will be shown later, rickets is in reality a rare disease.
In a very large proportion of cases in which the skeletal signs were
found in Puerto Rican children, they occurred only in a very slight or
slight degree. It is probable that most of the signs recorded as occur
ring to a very slight degree were within the range of normal and were
given little weight by the observers when making the diagnoses. The
weight given to signs occurring in a slight degree varied, apparently
depending on the presence of other skeletal signs or on evidences of
muscular weakness. Skeletal signs were reported to have occurred
in a moderate degree in relatively few instances; but when they did
so occur, more weight, as would be expected, was given to each one in
making the diagnosis.
Table 9 shows the degree in which each skeletal sign was present
among the Puerto Rican children and indicates the frequency with
which a clinical diagnosis of rickets was made in the presence of each
sign. The ultimate diagnosis was, of course, made by consideration
of all the signs, both skeletal and muscular, that were found in each
case. A detailed tabulation of the combinations of skeletal signs
found in each of the 584 children will be found in Appendix Table B2.
Table 9 shows that a positive diagnosis was made in approximately
half the children with a Harrison’s groove, in one-third of those with
either enlarged costochondral junctions or enlarged epiphyses at the
wrist, in one-fifth of those with moderately enlarged parietal or frontal
bosses, in one-sixth of those with bowlegs or knock-knees, and in oneeighth of those with asymmetrical heads. It is apparent that the
observers must have felt that the presence of enlarged costochondral
junctions, enlarged epiphyses at the wrist, and a Harrison’s groove

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF BONES

C H IL D R E N

PUERTO

R IC O

was more definite evidence of rickets than was the presence of either
slight bowing of the legs or slight knock-knees. As would be expected
when either bowlegs or knock-knees occurred in a moderate degree, a
clinical diagnosis of rickets was more likely to be made than when it
occurred in a slight degree, but the diagnosis was not necessarily con
sidered positive just because one of these signs was present to a
moderate degree.
T a b l e 9. — Skeletal signs usually considered evidence of rickets and degree in

which these signs were present in relation to clinical diagnosis of rickets; children
having these signs who were examined in Puerto Rico
Children having skeletal signs
Clinical diagnosis of rickets
Skeletal signs
Total
No
rickets

Ques
tion
able
diag
nosis

Rickets

Total

Slight

Costochondral junctions......................................

Slight enlargement__________ ____________
Moderate enlargement.-............... ................

89
3

30
3

28
2

M oder
Marked
ate
2

1
1

Epiphyses of long bones at wrists (slight enlargement)_____________ ____ ______________

Bowlegs_______ ____ _________________________

176

Slight................................ .............................
Moderate____ ___________________________

153
23

89
1

45
12

19
10

19
9

Knock-knees___________________ ____________

S lig h t............................... .......................... .
Moderate___________ ____________________

68
17

45
6

14
6

9
5

9
4

34
3

14
1

17
2

14
1

2
1

Harrison’s groove............................................... .
Slight enlargement____ _ ______ _________
Moderate enlargement_________ _____ ___
Pigeon-breast, slight.............................................

Parietal or frontal bosses_________ ____________

Parietal or frontal bosses moderately enlarged_________________________________
Parietal and frontal bosses moderately
enlarged______________ _________

Asymm etry of head...............................................

31
2

18
2

S lig h t................................................... .........
M o d e ra te _____________________ ____ ____

1
1

Craniotabes. „ ............................... ......................

Slight..................... ............ ................ ..............
M oderate________________________________

4
1

1
i

1
1

The reports were analyzed to find out whether there was any sig
nificant difference in the incidence of these skeletal signs as found by
the two physicians. With the exception of a more frequent report of
Harrison's groove by one observer and of moderately enlarged parietal
or frontal bosses by the other, there were no significant differences.
The following discussion of the incidence and degree of each of the
skeletal signs usually thought of as indicating rickets brings out some

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

IN C I D E N C E

O F R IC K E T S

points of interest in interpreting their occurrence in this tropical
region as compared with their occurrence in New Haven.
B o w l e g s . — Bowlegs were found less often in Puerto Rico than in
New Haven. Though in 30 per cent of the Puerto Rican children
examined bowlegs were found to be present in a degree that was
considered outside the limits of normal, they were also found in a
less degree in an additional 49 per cent not shown in the table. It is
probable that most of these cases of bowing, whether considered
within the limits of normal or not, were in reality only exaggerations
of the normal curves due, in many cases, not to rickets but to the
mother’s habit of carrying the infant astride her hips. In this position
the infant, though supported by the mother’s arm or a sling, soon
learned to help support himself by clinging with his legs to his
mother’s body. The tendency was, therefore, for the natural bowing
of the child’s legs to be increased.
No case was found of the bowing of the lower third of the legs
that is so characteristic of rickets.
K n o c k - k n e e s . — Knock-knees (estimated only for children who could
stand at the time of examination) was found in Puerto Rico in nearly
as large a proportion of cases as was an abnormal degree of bowing;
but unlike bowing, it occurred even more often in Puerto Rico than in
New Haven. Since in many cases the knock-knees disappeared when
the child was placed in a prone position and the knees brought together
it seems probable that the knock-knees were not due to any bony de
formity such as might be produced by rickets.
Furthermore, in no case were the knock-knees asymmetrical— the
type of knock-knees characteristic of severe rickets.
E n l a r g e d c o s t o c h o n d r a l j u n c t i o n s a n d e p i p h y s e s o j l o n g b o n e s a t w r i s t .—
The incidence of enlarged costochondral junctions and of enlarged
epiphyses of the long bones at the wrist was strikingly lower in Puerto
Rico than in New Haven. The difference may be attributed, in all
probability, to the difference in climate and the consequent difference
in the incidence of rickets in the two localities. The fact, however,
that even 16 per cent of the Puerto Rican children examined were
thought to have had slight enlargement of the costochondral junc
tions, and 10 per cent had slight enlargement of the epiphyses at the
wrist (whereas, as will be shown later, in only 1 per cent was evidence
of rickets found by roentgen ray) suggests either that the presence of
such enlargement does not always indicate the presence of rickets or
that the observer’s impression of enlargement is sometimes incorrect.
Moreover, it may be pointed out that in New Haven a slight degree
of enlargement of the epiphyses at the wrist had apparently been
present in a number of children who showed no evidence of rickets
even on repeated roentgenographic examination of the bones of the
forearm. There is little doubt, therefore, that in Puerto Rico, as in
New Haven, a number of cases of slight enlargement of the epiphyses
of the wrist and probably also of the costochondral junctions are the
result of physiological growth and development, and are not necessa
rily pathological. Probably it is impossible to distinguish with cer
tainty between the slight enlargement due to rickets and that due
to normal growth.
H a r r i s o n ’ s g r o o v e . — Harrison’s groove was found in only one-fourth
as many children in Puerto Rico as in New Haven. In 15 of the 37
children showing this sign hypertrophied tonsils or adenoids, or both,

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPM ENT

OF BON ES

O F C H IL D R E N

PUERTO

R IC O

were also found, and in 14 additional children there was enlargement
of the cervical lymph nodes, probably indicating infection in the nose
or throat. This association in Puerto Rican children of Harrison’s
groove with hypertrophy of the lymphoid tissues of the nose and
throat suggests that obstruction of the respiratory passages due to
simple hypertrophy or to infection may be one of the underlying causes
of this deformity. The greater incidence of Harrison’s groove among
New Haven children indicates without much doubt, however, that
rickets increases the frequency with which this sign occurs.
H e a d s i g n s .— The shape of the children’s heads in Puerto Rico,
probably a racial characteristic, was very different from that most
frequently seen in New Haven, although in the latter city heads sim
ilar in shape to those found commonly in Puerto Rico were occasion
ally found among both negroes and whites. A large proportion of
the Puerto Rican children had heads that were narrow and low m
the frontal region, broad and high in the parietal, and more or less
flat in the occipital; the New Haven children’s heads were character
istically squarer or rounder in shape, with prominent or high fore
heads. In Puerto Rico this usual breadth of head in the parietal
region commonly gave the impression that the parietal bosses were
somewhat enlarged; but unless either the parietal or the frontal bosses,
or both, appeared to be at least moderately enlarged, the child’s head
was not considered abnormal. Moderately enlarged frontal bosses
were comparatively rare in Puerto Rico, the great majority of diag
noses of abnormal heads having been based on moderate enlargement
of the parietal bosses only. It is doubtful, however, whether the
heads should have been considered abnormal unless the frontal bosses
were enlarged also, since even moderately enlarged parietal bosses
were probably only exaggerations of the shape characteristic of the
Puerto Rican heads in general. On the whole, however, neither mod
erately enlarged parietal nor moderately enlarged frontal bosses were
common in Puerto Rico as compared with New Haven.
Craniotabes was found in five Puerto Rican children. For two the
roentgenograms showed rickets (see case histories 1 and 4 in Appendix
E, p. 120), and for these two clinical diagnoses of rickets were also made.
Three children showed no rickets by roentgenogram. They were 2,
3, and 4 months of age and showed slight craniotabes as the only
clinical sign of the disease. Positive clinical diagnoses were not made
in these three cases. None of the five children showed evidence of
syphilis either at physical examination or by roentgenogram.
RELIABH.i t Y OF CLINICAL DIAGNOSIS

That the clinical diagnoses of rickets made on these Puerto Rican
children were based in large part on relatively slight evidence and
were not made consistently has already been indicated. This may
be seen even more clearly by studying the various combinations of
skeletal signs and the diagnoses made from each combination, as
shown in Appendix Table B2 (pp. 111-112).48 Though in retrospect the
basis for diagnosis seems inadequate in a number of cases, it should
be remembered that a similar basis for diagnosis is often used in tem
perate climates and is assumed to be justified because a relatively large
<s It should be pointed out, however, that this table does not include signs of muscular weakness, such
as pot-belly or poor muscle tone, which no doubt influenced the diagnosis in many cases, nor does it show
the degree of the skeletal signs, as does Table 9.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INCIDENCE OF RICKETS

proportion of such diagnoses can be corroborated by other methods
of examination. The fact (as will be pointed out later) that there is
no basis for belief that rickets exists among the Puerto Rican children
examined except as a rare disease, together with the fact that most
of the clinical diagnoses of rickets made in Puerto Rico could not be
corroborated by roentgenograms, throws doubt not only on the diag
noses made in Puerto Rico but also on a certain number of the diag
noses made on similarly slight evidence in temperate climates. Prob
ably it is impossible to distinguish clinically between the skeletal
changes in the long bones that are within the limits of normal growth
and development and those that are the result of a slight degree of
rickets.
That the clinical diagnoses of rickets in Puerto Rico, based on
skeletal signs that deviated but little from the preconceived standard
of normal, were probably unreliable will be shown in the following
analysis of the roentgenographic findings.
DIAGNOSIS OF RICKETS BY ROENTGENOGRAPHIC EXAMINATION

Roentgenograms were taken of the radius and ulna of the 584 chil
dren who were given physical examinations by the same technique
as that previously used in New Haven. (See Appendix C, p. 113.) Of
all the long bones the radius and ulna are the most satisfactory to
use for the early diagnosis of rickets, since the examination of these
presents fewer technical difficulties than examinations of other bones.
Though rickets probably always develops in the ribs before it
develops in the ulna, the difficulty of obtaining good roentgenograms
of the costochondral junctions is so great that study of the develop
ment of rickets in the rib by this method is practically out of the
question. Besides, the thickness of the femur or tibia and the
irregularity at the diaphyseal-epiphyseal junction makes the use of
these impractical. Because of its smaller and more uniformly regular
epiphyseal surface, the ulna is even more satisfactory than the radius
for the early diagnosis of rickets.
INCIDENCE OF RICKETS AT ROENTGENOGRAPHIC EXAMINATION AND ITS RELATION
TO AGE

The diagnoses of rickets made from the roentgenograms are shown
in Table 10. That rickets is actually very rare in Puerto Rico as
compared with New Haven is shown by the fact that for only 5 chil
dren was there definite evidence in the roentgenograms that the child
had rickets or had had it previously. Three of these children (all
less than 6 months of age) showed changes in their bones comparable
to those found commonly in young infants in New Haven and inter
preted there as very slight or slight rickets; 1, who had lived all his
life (until 2 weeks before examination) in a stone cellar apartment
lighted only by electric light, showed severe, active rickets; and 1
showed the scars of an old process (probably one of moderate degree
at the time of its activity) which had occurred when she lived in New
York City and which had healed some months before the time of the
study. The total incidence of rickets in the 584 children examined
was less than 1 per cent.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

T a b l e 10.— Roentgen-ray diagnosis of rickets and age at date of examination; chil

dren examined in Puerto Rico
Children examined in Puerto R ico
Roentgen-ray diagnosis of rickets
Rickets

Age 1 at examination
Total

N o rickets

Total
N um
ber

T otal......................... .

584
73
98
75
74
104
76
64
20

Doubtful
diagnosis

509
60
67
62
62
99
76
63
20

Per
cent
87
82
68
83
84
95
100
98

N um
ber
70
12
29
12
12
6

Per
cent

N um
ber

Active,
Per marked
cent

16
30
16
16
5

1
2
1

Slight, Slight.
heal healed
ing

1
2
1

iAge is given as of nearest month; that is, “ under 4 m onths” is actually under 3 months and 16 days,
“ 4 to 6 m onths” is from 3 months and 16 days to 6 months and 15 days, inclusive, and so on.

Because of the rarity of rickets in Puerto Rico, the case histories of
the five children showing roentgen-ray evidence of the disease are
given in detail in Appendix E (p. 120) and are summarized as follows:
Case 1.— S. M. was a 7-month-old, full-term male infant who was born in a
cellar of one of the large stone tenements of San Juan and had lived there until
two weeks before examination. The apartment in which he lived had no outside
window. At no time had he been taken out of doors. He had been fed practi
cally from birth on a cow’ s milk mixture and had been given no cod-liver oil.
There was a history of frequent attacks of bronchitis and daily convulsions for
three or four months.
Physical examination showed broncho-pneumonia and evidences of severe
rickets. Chvostek’s sign was not elicited. Roentgenograms of the bones of the
forearm (fig. 1) showed advanced rickets with a marked degree of osteoporosis.
Blood studies showed a calcium content of 7.1 milligrams per 100 cubic centi
meters of serum and a phosphorus content of 4.1 milligrams. A diagnosis of
severe active rickets was made from the clinical and roentgen-ray examinations,
and a diagnosis of tetany from the chemical examination of the blood. Figures 2
and 3 show the result of treatment with sunlight alone— Figure 2 at the end of
18 days, Figure 3 at the end of 3 months.
Case 2.— W. V. was a 5-month-old, full-term male infant who was taken out of
doors very little— about half an hour a day— in the shade. He was on a mixed
feeding of breast milk and condensed milk.
Physical examination showed slightly enlarged costochondral junctions, on the
basis of which a clinical diagnosis of questionable rickets was made. The roent
genogram showed evidence of slight rickets with a slight degree of osteoporosis.
Case 3.— J. L. was a 5-month-old, full-term female infant, entirely breast fed.
Physical examination showed none of the signs of rickets. The roentgenogram
showed evidence of very slight rickets.
Case 4-— M. S. was a 3-month-old, full-term male infant, wholly breast fed.
Physical examination showed a widely open fontanelle, a moderate degree of
craniotabes, and slightly enlarged costochondral junctions. A clinical diagnosis
of slight rickets was made. The roentgenogram showed very slight rickets.
Case 5.— M. G. was a 25-month-old, full-term female infant who was born in
New York City, where the first 21 months of her life were spent. Her diet was
mixed. No cod-liver oil had ever been given. Physical examination showed
slightly enlarged costochondral junctions and slight knock-knees, but the evidence

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1
#

F igure 1.— Severe active rickets and osteoporosis in an infant (S. M ., 7 months old)
who had lived in a cellar in San Juan and
had never been exposed to sunlight

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

F igure 2.— Fresh lime-salt deposits in bones of
same infant after 18 days’ exposure to sunlight

F igure 3.— Complete healing in bones of same
infant after 3 m onths’ exposure to sunlight

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INCIDENCE OF RICKETS

was not sufficient to make a clinical diagnosis of rickets. The roentgenogram
showed evidence of old, healed rickets. Though this case is included in Table 10,
it can not legitimately be counted as a case of rickets developing in the Tropics.

W. V., J. L., and M . S. (cases 2, 3, and 4) were all young infants
who showed evidence of slight or very slight rickets by roentgen ray.
The changes in their bones were comparable to those found in many
infants in New Haven. It may be emphasized here that neither S. M .
nor M . G. (cases 1 and 5) developed rickets under the influence of the
usual environmental conditions of Puerto Rico— S. M ., because the
disease followed c o m p l e t e isolation of the infant from sunlight; M . G.,
because the disease undoubtedly developed while the child lived in
New York City. J. L. and M . S. (cases 3 and 4) had apparently led
the usual existence of young infants in Puerto Rico; W. V. (case 2)
probably had been taken out of doors less than was customary.
Of the remaining 579 roentgenograms, 509 were classed as“ nor
m al” according to the New Haven criteria, and 70 for one reason or
another were classed as doubtful. Detailed study of each of the
doubtful roentgenograms has led to the belief that the findings that
raised some question as to their interpretation were nearly all due to
movement of the child’s arm while the roentgenogram was being
taken, to a position of the bones that allowed the epiphyseal surfaces
to show more than was usual, or to improper roentgen-ray technique.
The roentgenograms classed as doubtful, though tabulated separately
in Table 10, in all probability should be thought of as showing normal
bones.
The almost complete absence in Puerto Rico of the slight deviations
from the nprmal commonly found in New Haven and diagnosed there
as rickets lends great weight to the opinion that these deviations are
in reality the earliest evidences of that disease. Though it is believed
that their absence in Puerto Rico may be attributed primarily to the
protective influence of the sunlight and their presence in New Haven
to inadequate protection, nevertheless, the fact that the relatively
slower growth of the Puerto Rican children may play some part must
not be overlooked. Though the Puerto Rican children as a group
were shorter and lighter in weight than white children of continental
United States, as has been shown in Charts I and II, it has also been
pointed out (pp. 21, 22) that the differences were comparatively slight
m the first six months of life, the period when the bone changes most
commonly made their appearance in New Haven children. It may be
supposed, therefore, that many more than three Puerto Rican chil
dren would have shown changes similar to those found in New Haven
children, had these changes been due to physiological growth alone.
Slower growth was probably a minor factor in the prevention of the
changes interpreted as rickets, as compared with the intense sunlight.
Age incidence.

The ages of the children at the time of examination are shown in
Table 10. It will be seen that 3 of the 5 children showing rickets were
under 7 months of age, and that 1, the child with severe rickets, was
between 7 and 9 months. The child with a moderate healed process
was over 2 years of age; and since the active stage of the disease
must have occurred when the child was in New York, the case, as has
been pointed out, should not be classed as rickets developing in Puerto

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

Rico. Of the 246 Puerto Rican children 9 months of age or under at
the time of examination, only 4 (less than 2 per cent) showed roentgenographic evidence of rickets, whereas in a group of 407 New Haven chil
dren of the same age who had been given no treatment to prevent
rickets 142 (35 per cent) showed such evidence. Of the Puerto Rican
children 10 to 15 months old, none showed roentgen-ray evidence of
rickets, whereas 79 of 249 New Haven children of the same age group
(32 per cent) showed such evidence of the disease.
INTERPRETATION OF CLINICAL DIAGNOSIS IN THE LIGHT OF ROENTGEN-RAY
DIAGNOSIS

The proper interpretation of the clinical signs usually considered to
be manifestations of rickets is of practical importance, since in most
communities it is impossible to provide roentgen-ray or chemical
examination for all children suspected at physical examination of
having the disease. A clinical diagnosis of rickets when the disease is
moderate or severe is relatively easy, and few mistakes are made;
but the diagnosis of mild rickets is often difficult and is probably
made on many children who have not had rickets. In the study
made in Puerto Rico considerable doubt may be thrown on the correct
ness of the clinical diagnoses because of the discrepancy between
the incidence of rickets as shown by roentgen-ray examination and
that shown by clinical examination.
When attempting to interpret the results of the clinical examina
tions in the light of the roentgenographic examination of the radius
and ulna alone it is necessary to bear in mind some of the facts regard
ing the development and course of the disease and the limitations of
both types of examinations. It should be remembered that rickets
makes its appearance first in the ribs and soon afterwards affects the
other long bones at their more rapidly growing ends. Theoretically,
then, roentgenograms of the ribs should provide one of the best
means of making a very early diagnosis of rickets, but the technical
difficulties of obtaining satisfactory roentgenograms of the costochon
dral junctions make the use of this method virtually impossible.
The roentgenographic method of diagnosis is, moreover, not delicate
enough to reveal the truly incipient changes due to rickets, and it is
not until the process has progressed to the stage at which the
changes in the ulna are sufficient to cast a shadow that a diagnosis
can be made by roentgenographic examination.
There is a brief period, then, of one or possibly two months at the
beginning of the disease when roentgenography does not assist in
the diagnosis of rickets. At this stage, too, the clinical evidences of
the disease are usually indefinite and clinical diagnosis is unreliable.
In certain cases of incipient rickets definite clinical evidence may,
however, precede roentgen-ray evidence, as in the case of craniotabes49, but more often clinical evidence accompanies the roentgen-ray
evidence or follows it closely.
From the time that rickets is first seen in the roentgenogram of the
ulna— usually at about the third or fourth month— and throughout
a Though a slight enlargement of the costochondral junctions theoretically m ay be the earliest sign of
rickets and m ay precede roentgen-ray evidence of the disease in the radius and ulna, the diagnosis of rick
ets on this basis in the absence of any other clinical signs of the disease has not been considered safe because
of the great uncertainty of clinical judgment and the frequent errors made in the attempt to differentiate
b y palpation alone the costochondral junction that is normal from the one that is just over the border of
the pathological. In a majority of cases in which enlargement of the costochondral junctions is sufficient
for certain diagnosi s, the roentgenogram also shows evidence of rickets in the radius and ulna.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INCIDENCE OF RICKETS

the period of activity and healing of the disease, roentgenograms of
the bones of the forearm have their greatest value in confirming the
clinical diagnosis. But when the disease has subsided and the process
of repair has wiped out the scars in the radius and ulna, roentgenographic examinations of these bones are no longer useful in judging
whether the clinical signs commonly thought of as signs of mild rickets
are, in individual cases, sequelae of the disease or normal variations
associated with growth. Some light can be thrown on these later
clinical diagnoses by comparing their incidence with that of active or
healing rickets as shown by roentgenograms in earlier age periods.
As has been pointed out, the incidence of rickets found by roentgen
ray in the first year or two of fife should form a background for the
incidence of rickets found by clinical examination either coinciden
tally or later. Since, moreover, the incidence of rickets found by
clinical examination at any given age level represents in general an
accumulation of current cases and cases that have occurred at one
time or another previous to the age in question, it is to be expected
that after the early months of infancy are passed the clinical incidence
of rickets would be higher than the maximum incidence found by
roentgen-ray examination at the same time or in any previous age
period. This maximum incidence found by roentgen-ray examination
in a cross-section study must fall short of the true total incidence,
first, by the number of cases of incipient rickets not yet advanced
enough to show in the roentgenogram, and secondly, by the number
of cases of rickets that have already receded and hence no longer show
in the roentgenogram. How much allowance should be made in
order to arrive at a true estimate of the total incidence of rickets is
not known. ^If, however, a great disproportion exists between the
amount of rickets^ demonstrable by roentgen ray during the period
when the disease is usually active and the amount of rickets found
concurrently or later by clinical examination it is believed that this
disproportion probably represents a certain amount of error in clinical
diagnosis due to the difficulty of differentiating the clinical signs of
mild rickets from the variations of normal growth and development.
A roentgenogram may be expected, then, to help in confirming an
individual clinical diagnosis of rickets most often during the age of
greatest activity of the disease—roughly from 3 to 18 months. After
this period is passed, and in many cases of mild rickets during the
latter half of it, roentgenograms may not help in the confirmation of
individual clinical diagnoses because the rachitic process has subsided,
leaving only the bony deformities on which the clinical diagnoses were
based. ^ When the incidence of rickets is being studied on a commun
ity basis, however, the roentgenographic method of diagnosis gives
the most satisfactory picture of the total incidence of the disease
durmg the period when it is in its active stages and provides a basis
for interpretation of the clinical diagnoses made on the total group.
In an attempt to understand the great preponderance in Puerto
Rico of clinical diagnoses of rickets over those made by roentgen ray,
two methods of study have been used: First, a direct correlation of
the clmical with the roentgen-ray diagnosis made at the same time,
and, second, a comparison of the results of the two types of examina
tion at different age levels.
Table 11 gives the clinical diagnoses made on 584 children and
shows whether or not they were corroborated by the roentgen-ray
diagnosis made coincidentally. It will be seen that only two (4 per

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

7to9
I0t0l2
months months

!3tOt8 l9to24 25to34
months months months

Puerto Rican children (564)

3month3 4 to 6 and under months

7t0 9
months

I0t0l2
13to 18
|4to24 25to36
months months months months

New Haven Children (956)

C h art V I.— Clinical and roentgenographic evidence of rickets at specified age periods in children examined in Puerto R ico and in children examined in N ew Haven
who had been given no specific treatment to prevent rickets

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

3months 4 to 6
and under months

INCIDENCE OF RICKETS

cent) of the clinical diagnoses of rickets were corroborated by the
roentgenograms. One of these diagnoses was of a case of marked
clinical rickets in a child 7 months of age, the other a case of very
slight rickets in a child 3 months of age. Both cases (cases 1 and 4)
will be found discussed in detail in the case reports in Appendix E
(p. 120). A similar correlation of clinical and roentgen-ray diagnoses
was made for a group of 956 New Haven children of the same age
group who had been given no specific antirachitic treatment, and it
was found that 37 per cent of the clinical diagnoses of rickets were
corroborated by roentgen-ray examination, a striking contrast to
the Puerto Rico findings.
The relation between the incidence of rickets found by roentgen-ray
examination and that found by clinical examination in different age
periods is shown in Chart V I for 564 children examined in Puerto Rico
and 956 children of a similar age group examined in New Haven.
The contrast in the incidence of rickets (both clinical and roentgen
ray) in the two communities is marked.
T a b l e 11— Clinical diagnosis of rickets in relation to roentgen-ray diagnosis;

children examined in Puerto Rico
Children examined
Clinical diagnosis of rickets

Total
Roentgen-ray
diagnosis

N o rickets

Per
cent
N um dis N um
ber tribu ber
tion
T ota l______ _____
N o rickets-------------------Doubtful diagnosis------Slight or very slight
rick ets........................Moderate healed rickets
Marked active rickets..

584

100

400

509
70

87
12

354
44

3
1
1

1
(*)
«

1
1

Question
able
diagnosis

Per
cent N um
dis
ber
tribu
tion
100
89
11
(*)
(a)

134

Rickets
Total

Per
cent N um
dis
ber
tribu
tion
100

M od
Per Very Slight1 er Marked
cent slight1
(>)
ate 1
dis
tribu
tion
100

29
4

2
1

114
19

85
14

41
7

82
14

10
2

1 Per cent distribution not shown as number of children was less than 50
2 Less than 1 per cent.

If, however, the relation of one type of examination to the other is
studied for the two groups considered separately and the results com
pared, it will be seen that in the Puerto Rican group the incidence of
rickets found by roentgen-ray examination in the early age periods
would seem to be far too small to account for the incidence found at
clinical examination in the later age periods; whereas in the New
Haven group, the incidence of rickets found by roentgen-ray exami
nation in corresponding early age periods is so large as to account very
well for the proportion of the cases showing clinical evidences of
rickets during the following years. The very small proportion of
clinical diagnoses of rickets in Puerto Rico that were corroborated by

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

roentgen-ray examination, the discrepancy between the incidence of
roentgen-ray manifestations in the early age periods and clinical mani
festations in the later, and the almost complete absence of roentgenographic evidence of rickets at any age period strengthen the opinion
that many of the clinical diagnoses were incorrectly made.
As has already been pointed out in the discussion of the clinical
diagnoses, errors in the diagnosis of mild clinical rickets are probably
made because of the difficulty of differentiating the skeletal changes
due to mild rickets from those which are the result of normal growth
and development. How much such error in diagnosis may account for
the excess of clinical diagnoses over roentgen-ray diagnoses in New
Haven is not known; in Puerto Rico, however, in the face of the
almost complete absence of roentgen-ray evidence of rickets, it
would seem probable that a great majority of the clinical diagnoses
of rickets were incorrect.
DETERMINATION OF AMOUNT OF CALCIUM AND PHOSPHORUS IN
THE BLOOD OF 34 SELECTED INFANTS

Chemical examination to determine the amount of calcium 60 and
of inorganic phosphorus 51in the blood serum was made for 34 infants,
most of them 8 months of age or younger. Nineteen of the samples
came from the group of infants for whom roentgenograms showed
unquestionably normal bones, 9 from the group for whom roentgeno
grams showed bones which were probably normal but which for one
reason or another were classified as doubtful, 5 from the group for whom
roentgenograms showed severe osteoporosis or atrophy of the bones,
and 1 from the infant with marked rickets. Unfortunately it was not
possible to obtain samples of blood from any one of the three infants
who showed slight or very slight roentgenographic evidence of rickets.
The findings on these samples of blood, as well as some clinical and
roentgenographic data regarding the infants from whom they were
taken are given on pages 54-55.
There are not enough cases to warrant drawing any definite con
clusions with regard to the relation between the blood findings and
the child’s diet or his nutritional condition. It is of interest to observe,
however, that of the 32 samples of blood for which the calcium content
was reported 15 contained 11 or more milligrams per 100 cubic centi
meters of serum, 11 contained 10 to 10.9 milligrams, 5 contained 8 to
9.9 milligrams, and 1 (the blood of the infant with active rickets and
tetany) contained only 7.1 milligrams. The great majority of the
calcium findings fell in or above the upper range of what is usually
considered normal— 9 to 11 milligrams per 100 cubic centimeters of
serum, the average amount being 10.6 milligrams.
Of the 34 samples of blood examined for inorganic phosphorus, 23
contained 5 or more milligrams per 100 cubic centimeters of serum
(each of 3 samples containing 6.4 milligrams, the highest reading), 9
contained 4 to 4.9 milligrams, and 2 contained just less than 4 milli
grams. Both of the last-mentioned samples of blood were from
infants the roentgenograms of whose bones showed severe osteoporosis
50 Kramer, Benjamin, and Frederick F. Tisdall: A Simple Technique for the Determination of
Calcium and Magnesium in Small Am ounts of Serum. Journal of Biological Chemistry, vol. 47,
no. 3, (August, 1921), p. 475.
m Briggs, A . P .: A M odification of the B ell-D oisy Phosphate M ethod.
Journal of Biological
Chemistry, vol. 53, no. 1 (July, 1922), p. 13.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

INCIDENCE OF RICKETS

but no evidence of rickets. The average amount of inorganic phos
phorus was 5.2 milligrams per 100 cubic centimeters of serum.
The product obtained by multiplying the calcium content (in milli
grams per 100 cubic centimeters of serum) by the phosphorus content,
the criterion suggested by Howland and Kramer62 for the differentia
tion of rachitic from nonrachitic blood, is also shown on pages 54—55.
In the opinion of these authors any product falling below 40 indi
cates rickets. In 25 of the samples of blood examined, this product
was more than 50; in 4, between 40 and 50; in 3, between 30 and 40;
for the other 2 samples the product was not determined, as the cal
cium content was not reported. Of the 3 with product below 40,1 was
blood from the infant with active rickets and the other 2 were from
infants with severe osteoporosis, but no evidence of rickets. The
average product of the calcium content and the phosphorus content
in these 34 samples of the blood of infants living in the Tropics was
56.2.
u Howland, John, M . D ., and Benjamin Kramer, M . D .: Factors Concerned in the Calcification of Bone.
Transactions of the American Pediatric Society, vol. 34, p. 204. Washington, 1922.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

General
nutri
tional
condi
tion of
child

Initials

15
104
50

P. B ____
O. F ____
D . F ____

3
4
4

10.2
9.9
10.6

R . D ___

11.1

6.2

68.82 ........d o ...................... ____ d o ......... .......

175
69

L. V .......
B . L ____

5
6

8.6
14.1

5.8
6.4

49.88 ____ d o____________ ........d o_________ Flaring ribs; muscle tone fair......
90.24
Questionable.— Slight costochondral junctions; moderate G o o d - B r e a s t
m ilk ;
frontal bosses; head asymmetrical.
malted milk.
68.88 ........d o...................... N o rickets
Slight Harrison’s groove; slight haring r ib s .. V ery
good.
44.00 ........do...................... ........ d o................. Head asymmetrical; muscle tone poor...........
and rice water.
71.04 ____ d o____________ _ .A._ d o _________
Breast m ilk; cow ’s
milk, dried or
fresh.
55.00 -------d o ...................... ____ d o.................
Good— Breast milk; cow ’s
m ilk, 1 quart a
day.
52.52 ........d o....... .............. ____ d o............. .
P o o r .. Condensed m ilk ...
58.76 ........d o...................... ____ d o _________
G ood—
64.90 ........ d o...................... Questionable... Slight pot-belly; moderate knock-knees____ . —d o — C o w ’ s m ilk fo r
mula.
54.00 ........do____ ____ . . . N o rickets
Muscle tone fair; relaxation____
52.32 ____ do...................... Questionable.. . Slight Harrison’s groove; flaring ribs; slight -__do__ _ Breast m ilk; cow ’s
pot-belly; muscle tone fair, relaxation.
milk, Y pint a
day.
71.04 ____ do......................
Moderate parietal bosses; flaring ribs; pot
d o ... Breast m ilk; cow ’s
belly.
m ilk, 1J4 pints a
day.
58.80 ........ do...................... ____ do............... . Slight Harrison’s groove; pot-belly; flaring — do__ Breast m ilk chief
ribs; muscle tone poor, relaxation.
ly; cow ’s milk
formula and bar
ley water.
43.16 ........ do...................... ........do...... .......... Slight epiphyses at wrist; moderate pot-belly. G o o d - Breast milk; cow ’s
m ilk, Yi pint a
day in coflee.
61.60 ____ do..................... N o rickets
Slight pot-belly........ .................................. ........ Fair. __
69.03 D o u b tfu l............... Questionable...
Good
60.48 ........ do...................... N o rickets_____
V ery
Breast milk; con
good.
densed milk.
66.08 ____ d o____________ Questionable... Slight epiphyses at wrist.................................. . —d o .— Breast m ilk______

5.5
5.6
5.2

C . L .......

12.3

R. N - - -

10.0

4.4

J. S.........

11.1

6.4

180

S, M ___

10.0

5.5

63
88
54

M . B ___
G, R ___
J, A ........

6
8
8

10.1
11.3
11.0

5.2
5.2
5.9

133
53

R . N ___
0 . F ........

9
9

10.8
10.9

5.0
4.8

M. B—

11.1

6.4

102

O. G . . . _

10.5

5.6

331

C. R _ . . .

8.3

5.2

563
245
259

R . R ___
G. A . . . .
H. M ___

15
2
3

11.0
11.7
11.2

5.6
5.9
5.4

11.8

5.6

M. E—

https://fraser.stlouisfed.org
Federal Reserve Bank #of St. Louis

Roentgen-ray
diagnosis

Clinical
diagnosis

Clinical signs of rickets present

56.10 N orm al.......... ........ N o rickets_____ Slight pot-belly...............
55.44 ........d o .. .................. ____ d o................. Muscle tone fair; relaxation_______
55.12 ........ do...................... ____ d o_________

Feeding

Fair__
Good— Breast m ilk; cow ’s
milk, 1 pint a
day.
Fair__ C o w ’ s m ilk for
mula.

Calcium
(gms) in
m other’s
daily
diet 1

0.497
.328
.760
.168
.178
.771
.482
.453
.241

.213
! 168
.226

.214
.287

.147

.446
.372
.431

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

Product
of Ca
Content
and Ph
content

Phos
Calcium
Age in in blood phorus in
months (m g in 100 blood
c c serum) (m g in 100
c c serum)

Case
No.

M. P _ „.

10.2

5.1

H . R ___

11.6

4.5

52.20

O ___
R ___
L ____
D ___

6
6
7
5

(!)
10.1
(2)
11.4

4.7
4.1
4.7
4.9

S. M ___

7.1

4.3

____ do................
(3)
41.41
____ d o____ _______ ____ d o_________
(s)
55.86 Doubtful; osteopo Questionable... Slight costochondral junctions; slight pot
rosis - f .
belly; muscle tone fair, relaxation; flaring
ribs.
30.53 Marked
active V ery marked Moderate costochondral junctions: moderate
rickets; tet
rickets; osteopo
epiphyses: moderate Harrison’s groove;
any; pneumo
moderate bosses, frontal and parietal; slight
rosis + - H - .
nia.
craniotabes; moderate pot-belly; flaring
ribs; muscle tone poor, marked relaxation.
31.32 N o rickets; osteo N o rickets_____ Flaring ribs; moderate pot-belly; muscle tone
porosis + -(—(-.
fair, marked relaxation.
63.84 ........d o...................... Questionable.. . Moderate parietal bosses; moderate bowlegs;
flaring ribs; moderate pot-belly; muscle
tone poor, marked relaxation.
37.74 N o rickets; osteo N o rickets......... Marked relaxation; muscle tone very poor___
porosis + + + + •
52.92 ........d o...................... ____ d o................. Moderate parietal bosses; flaring ribs; m od
erate pot-belly; muscle tone poor, marked
relaxation; slight funnel chest.
54.52 ------ d o...................... ........do_............. . Moderate knock-knees; muscle tone poor,
marked relaxation; slight pot-belly.

46
59
13
43
234

R.
C.
E.
F.

144

R . P ____

8.7

3.6

A . F .......

11.2

5.7

279

N . P ___

10.2

3.7

568

M . R ___

9.-45

5.6

184

M . G ___

11.6

4.7

1 Foods recorded for sample day.
* N ot reported

52.02 ____ d o____________

N o rickets_____
____ d o ................

See p. 82 for method of estimation of amounts of calcium in mothers’ diets and discussion of findings.

.417

G ood..
mula.
Breast milk; cow ’s
milk.

F a ir .. ____ do____________

.283
.789
.214
.314
.405

_-_do___ C ow 's milk formu
la.

.306

P o o r .. C ow ’s milk, 1 pint
a day.
___do___ ____ do......................

.173

C ow ’s milk formu
Very
poor.
la.
— do__ Breast milk; cow ’s
milk.

.158

Fair__ C ow ’s milk in cof
fee only.

.069

.218

INCIDENCE OF RICKETS

CTt
Oi

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

A D D IT IO N A L R O E N T G E N O G R A P H IC S T U D IE S O F T H E
R A D IU S A N D ULNA

In addition to the study of the incidence of rickets, special study
was made of the roentgenograms to show the incidence of osteoporosis
and of transverse lines in the radius and ulna. The frequent associa
tion in temperate climates of osteoporosis with rickets adds interest
to the fact that osteoporosis was found to be present in a climate where
rickets was found to be practically nonexistent. That an inadequate
amount of calcium in the diet is the most probable etiological factor in
the production of osteoporosis in these Puerto Rican children and that
poor nutrition and irregular growth account for the high incidence of
transverse lines will be brought out in the following discussion.
INCIDENCE OF OSTEOPOROSIS

The most striking pathological finding in the roentgenograms of the
bones of the children examined in Puerto Rico was the presence of
osteoporosis. Osteoporosis is a rarefaction of the bone tissues, which
may be secondary either to decreased formation and increased resorp
tion of bone tissue or, as is more common in temperate climates, to
lack of calcification of newly formed bone tissue, as in rickets. It is
believed that in this group of Puerto Rican children a great majority of
cases of osteoporosis were due to decreased formation of bone tissue
rather than to a lack of calcification of newly formed bone tissue, since
in only two cases (S. M., case 1, and W. V., case 3) was the osteo
porosis associated with rickets.
Roentgenograms of the bones of 59 children— approximately 10 per
cent of the total group examined— showed osteoporosis— 19 to a mild
degree, 40 to a more severe degree. In many of the latter the osteo
porosis was so marked that it was described as atrophy. Border-line
cases were found in large numbers, but unless the rarefaction was
definite a diagnosis of osteoporosis was not made.
Detailed study of the roentgenograms showed a general rarefaction
or thinning throughout the cancellous bone tissue. (See figs. 4 and 5.)
Though this rarefaction varied considerably in degree in the cases
studied, in the majority of cases it was of moderate degree; in a few,
very marked. With the exception of the two children with rickets,
in every child studied the cortex of the shaft was thin and delicate,
being in some instances not more than a millimeter in thickness and
tapering off toward the ends of the shaft to the thinness of paper.
The cortex, as well as being thin, usually appeared very dense and
solidly calcified. In the children with rickets the cortex was rare
fied but thicker than normal, appearing layered, as is usual with ra
chitic bones in which osteoid tissue laid down by the periosteum is
not being calcified normally.
In nearly all the cases of atrophy the marrow cavity was relatively
wide and long, encroaching upon the spongiosa. In some, the mar
row cavity was nearly free from trabeculae and the spongiosa so rare
fied as to suggest the “ ground-glass” appearance described by Pelkan
as characteristic of the early stages of scurvy.53 The presence in the
“ Pelkan, K .F .: The Roentgenogram ia Early Scurvy,
no. 2 (August, 1925), pp. 174-188.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

American Journal of Diseases of Children, vol.30,

A. H.

N . P.

F igure 4.— Osteoporosis in two children: A . H ., 6 m onths old; and N . P ., 7 months old

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

A. P.

M . G.

F igu re 5 — Osteoporosis in tw o children: A . P ., 34 months old; and M . G., 32 m onths old

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

ROENTGENOGRAPHIC STUDIES OF RADIUS AND ULNA

spongiosa, however, of rarefied trabeculae or of poorly calcified trans
verse trabeculae showing in the roentgenograms as transverse lines
differentiates this extreme rarefaction of atrophy from that of the
early stage of scurvy, in which usually no bony trabeculae are visible
in the marrow cavity. The absence of definite rings 54 around the
oenters of ossification of the epiphyses, and the absence of any broad
ening of the zone of calcification are further evidence that the rare
faction was not of scorbutic origin. Occasionally fine penciled lines
were seen around the periphery of the centers of ossification in these
atrophic bones, which, though suggestive of rings, were too delicate to

M onth o f a^e

C hart V II.—Weights of individual Puerto Rican children having osteoporosis, compared with aver
age weights of all boys and girls examined in Puerto R ico

be so interpreted. In some cases the spongiosa appeared to consist
of many very delicate trabeculae which cast but poor shadows; in the
most severe cases, described as atrophy, the trabeculae were relatively
few and very slender, forming a delicate tracery of lines parallel to the
long axis of the bone, or sometimes a coarse network, with the prin
cipal lines running parallel to the long axis. In many of these atro
phied bones transverse lines (figs. 4 and 5) were seen crossing the spon
giosa at right angles to the long axis of the bone, indicating, as will be
pointed out, the frequent interruptions of growth that had taken
place.
The general physical condition of the children who showed this
osteoporosis of the bone was for the most part exceedingly poor, as
m Wimberger, Hans.: Zur Diagnose des Säuglingsskorbuts.
(1923), pp. 279-285.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Zeitschrift für Kinderheilkunde, vol. 30

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

judged by the amount of subcutaneous fat found and by their weight
and height. More than half were markedly undernourished and some
were emaciated to an extreme degree. The weight and height of each
child with osteoporosis have been plotted on Charts V II and V III,
according to the child’s age at time of examination, to show the re
lation of his weight and height to the average weight and height of
all the Puerto Rican boys or girls that were examined. The retar
dation in growth of these children is striking.
The generally poor nutritional condition of these children and the
almost complete absence of rickets suggest that the rarefaction of the
bone tissue was due not to inability of the organism to calcify newly

formed bone tissue, as in rickets, but to some profound nutritional
disturbance resulting in slow growth of bone with decreased deposi
tion and increased resorption of lime salts. Whether the nutritional
disturbance bringing about this slow growth was due primarily to a
generally inadequate diet or to a major deficiency in calcium is not
known, but it is probable that a low calcium intake in the diets of the
Puerto Ricans (see p. 82) played a definite part in the production of
the osteopQrosis described. Experimentally it has been shown that
diets deficient only in calcium result in osteoporosis of the bones.
Osteoporosis or atrophy of the bones was described by Alwens 56 as
m Korenchevsky, V .: The Aetiology and Pathology of Rickets from an Experimental Point of View, pp.
63-73. M edical Research Council, Special Report Series N o. 71. London, 1922.
m Alwens, Dr. [W .j: Über die Beziehungen der Unterernährung zur Osteoporose und Osteomalazie,
Münchener Medizinische Wochenschrift, vol. 66, no. 38 (Sept. 19,1919), pp. 1071-1076.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

ROENTGENOGRAPHIC STUDIES OF RADIUS AND ULNA

occurring in adults as a result of starvation following the World War.
The diets that he described were very similar to the Puerto Rican
diets— they were lacking chiefly in calcium and protein containing
foods, such as milk and milk products, green vegetables, fruit, and
eggs. The roentgenographic appearance of the bones described by
Alwens is apparently very like that foimd in this study.
So, too, the atrophy of the bones found in coeliac disease is appar
ently very similar to that here discussed. Parsons describes the
roentgenographic appearance of the long bones in coeliac disease as
follows:
If the child be under 7 years of age there probably will not be
any evidence of rickets, but the bones look unduly small, even
allowing for the small stature of the child. The trabeculae may be
well formed, but should the disease have been present for any
length of time the trabeculae appear to be thinner and the whole
bone more osteoporotic and fragile than normal. The cortex also is
thinner than would be expected, but the epiphyseal line is quite
sharp and distinct.57

Since in coeliac disease the growth of bone is greatly retarded, oste
oporosis or atrophy results rather than rickets; but when improvement
takes place and growth starts, rickets may be superimposed on the
osteoporosis. In Puerto Rico slow growth of bone and adequate ex
posure to sunlight combine to prevent the development of rickets in
these atrophic bones.
IN C ID E N C E OF T R A N SV E R SE LIN ES IN L O N G B O N E S

The narrow transverse lines of dense bony tissue frequently seen
crossing the spongiosa in roentgenograms of long bones are believed
to be evidence of periodic or intermittent growth. Each line is the
shadow cast by a thin plate or latticework of well-calcified trabeculae
which has formed at the end of the bone at a time when growth of the
bone in length has temporarily ceased. The anatomical condition re
sponsible for these transverse lines in the roentgenograms has been
described in the literature.5®59 It is further believed that lines occur
more frequently in older children and in those who have been chroni
cally ill or have had some wasting disease. Though transverse lines
may be found in roentgenograms of children with rickets, it is not
believed that rickets is responsible for the formation of these lines
except as it interferes with the growth of the bones.
Study of the roentgenograms made in Puerto Rico seemed to offer
an interesting opportunity to observe the incidence of these transverse
lines in a group of young children living under tropical conditions,
a group in which many were retarded in physical development but
practically none had rickets. The incidence of these lines in children
examined at specified ages is shown in Table 12 both for the Puerto
Rican group and the New Haven group. It is apparent that the pro
portion of children under 3 years of age showing transverse lines in
their long bones is greater in Puerto Rico than in New Haven and that
«7 Parsons, Leonard G.: The Bone Changes Occurring in Renal and Coeliac Infantilism, and Their
Relationship to Rickets. Part II. Coeliac Rickets. Archives of Disease in Childhood, vol. 2, no. 10
(August, 1927), p. 200.
.
. __ „
„ ,
, , _
«8 Eliot, Martha M ., Susan P. Souther, and E. A . Park: Transverse Lines in X -R a y Plates of the Long
Bones of Children. Bulletin of the Johns Hopkins Hospital, vol. 41, no. 6 (December, 1927), pp. 364-388.
m Asada, Tameyoshi: Über die Entstehung und pathologische Bedeutung der im Röntgenbild des
Röhrenknochens am Diaphysenende zum Vorschein Kom m enden parallelen Querlimenbildung. M it
teilungen aus der Medizinischen Fakultät der Kaiserlichen Kyushu-Universität, vol. 9.. no. 1 (1924), pp.
43-95.
160326°—33----- 5

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

the difference occurs chiefly after the early months of infancy are
passed. Nearly half the Puerto Rican children more than 1 year of age
showed lines. In New Haven in no group of children under 3 years did
so large a proportion show lines; but when children between 3 and 6
years of age were studied, it was found that 48 per cent showed lines.
These findings are to be expected if interruptions in growth are the
pathogenic factors.
Among the Puerto Rican children transverse lines occurred more
frequently in children whose nutritional condition was poor. In the
group of 72 Puerto Rican children reported as being greatly under
nourished, 39 (more than one-half) showed transverse lines in their
bones; of the 332 children who seemed to be well nourished, only onefourth showed them. It is also of interest that of the 59 children wTho
had osteoporosis or atrophy of the bones, 42 (nearly three-fourths)
showed transverse lines in their long bones, and of the 525 who were
not classified as having osteoporosis less than one-third showed them.
When malnutrition was so severe that osteoporosis developed in the
bones, either interruptions in growth became more frequent or, be
cause of the thin texture of the bone, the scars were more easily seen.
T

able

12.—

Transverse lines in radius and ulna of children examined at specified
age periods in Puerto Rico and in New Haven, Conn.
Children examined in
Puerto R ico

Age 1 at examination

Children examined in New
Haven

Lines present
Total
Number Per cent

Lines
absent

Lines present
Total
N um ber Per cent

■Lines
absent

T otal............ . . . .....................

584

200

384

1,186

339

847

Under 8 m onths_________________
8 to 13 m onths__________________
14 to 34 m onths..................... ..........
35 to 60 m onths_________________
N ot reported____________________

196
146
222

34
51
105

17
35
47

162
95
117

288
296
348
254

58
54
105
122

20
18
30
48

230
242
243
132

(*>

‘ Age is given as of nearest month; that is, “ under 8 m onths” is actually under 7 mont.hs'and 16 days,
“ 8 to 13 m onths” is from 7 months and 16 days to 13 months and 15 days, inclusive, and so on.
J Per cent not shown because number of children was less than 50.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

S O C IO -E C O N O M IC C O N D IT IO N S AS F A C T O R S IN T H E
H E A L T H O F P U E R T O R IC A N C H IL D R E N

Because of the influence that certain social and economic conditions
have in temperate climates on the incidence and severity of rickets,
special study was made of the environment of these children living in
a tropical climate. That poverty and ignorance and the accompany
ing poor hygiene and inadequacy of diet contribute in temperate
climates to the development of rickets in its severest forms is generally
conceded, but that these factors are not primary in the causation of
the disease is shown by the essentially complete absence of the disease
in Puerto Rico, where, according to the data presented in the following
sections, the people are very poor and the diets of the lowest quality.
Information on family groups, income, housing, and diet was sought
in Puerto Rico, as in New Haven, primarily with the idea of demon
strating whether any relation exists between the socio-economic con
ditions and the presence of rickets, and secondarily as of interest in
relation to the general health of the children studied. Since, as has
been pointed out, no evidence of rickets was found in the vast majority
of the Puerto Rican children examined, the absence of the disease under
conditions that, in a colder and less sunny climate, contribute mate
rially to its development in the severest forms, is extremely striking.
The great amount of severe rickets reported among Puerto Rican
children living in crowded tenements in the city of New York makes
even more striking the absence of rickets in the children studied in
Puerto Rico.
As has been pointed out earlier in this report, the children included
in the study were brought for examination through the efforts of
public-health nurses, and most of them had been brought previously
by their mothers to the child-welfare conferences of the insular depart
ment of health. The data may be regarded, therefore, as character
istic of families that were intelligent enough to seek such health advice,
probably representing the middle and lower economic groups of the
people of Ponce and of San Juan and its outskirts. Of these two
cities, San Juan not only is the larger but is the seat of the insular
government and the port at which nearly all ships dock. In San Juan,
therefore, there are greater opportunities for laborers and skilled
artisans to find work, and, in general, as will be shown by the data
on income, the people of San Juan are not so poor as the people of
Ponce.
Home visits were made to 506 of the 534 families having children
included in the study; 28 families could not be visited. Fifty of the
families visited each had 2 children included in the study; the infor
mation collected through home visits, therefore, has to do with 556
of the 584 children examined.
SIZE AND COMPOSITION OF FAMILY AND OF HOUSEHOLD

The households in which these Puerto Rican children were living
varied considerably in membership. Besides households consisting
of an ordinary family—mother, father, and their children— many
61

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

households were visited that included various persons not members
of the fam ily60— children and adults, some relatives, some not. In
many houses, because of economic pressure, two families had moved
in together to form a single household. The houses were small, and
the crowding was therefore great, as will be shown in the section on
Housing (p. 69).
Households were visited also from which one or more members of
the family were absent. A number of children were being cared for
away from home because of the poverty of their own families, and a
number of fathers were living away from home. Some of these
customarily remained away for months at a time as the seasonal
demand for labor arose in different parts of the island, and some
because by local custom family ties are loose. Many of the marriages
were consensual or common-law marriages; but though it is possible
that some of the fathers did not feel full responsibility for supporting
the family, in general such families were relatively stable units of
society. The children of such marriages, though reported as illegit
imate in the vital statistics, were considered legitimate in the social
group to which they belonged and were accepted without question as
members of the family.
Table 13 shows the size of the family of which the child under
observation was a member and also shows whether or not the house
hold in which he lived included persons other than members of this
family. It will be seen that nearly one-half the families were com
posed of 3 or fewer persons living at home, about one-fifth of 4, and
one-third of 5 or more. The size of the families ranged from 1
person (the child under observation, cared for away from home) to
15, the majority being composed of 3, 4, or 5 persons. The average
number of persons in these families was 4. There was no difference
on the whole in the size of the families in Ponce and in San Juan.
Of the 503 families for which information was obtained on this
point Table 13 shows that 241 were living in the same household
with another family or with some person or persons not members of
their own families. As would be expected, the smaller families were
more likely to take other persons to live with them in the same house
hold than were the larger families. The number of persons in the
household who were not members of the family of the child under
observation varied considerably. Ninety-one households had but
1 member in addition to the family of the child under observation;
45 had 2; 41 had 3; and 64 had 4 or more.
60 The term “ fam ily,” as used in this study, includes the child under observation, his mother, his full
brothers and sisters and his maternal half-brothers and half-sisters living with him and his mother, and
the man with whom the mother was living, whether they were legally married or not. Paternal halfbrothers and half-sisters and other relatives are not considered members of the family even though they
were living in the same household.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SOCIO-ECONOMIC CONDITIONS
T

able

13.— Number in child’s own family and presence in household of others than
members of family; families visited in Puerto Rico
Families visited
Total

Group in household

N um ber in family

Family only
Number

Per cent
distribu
tion

T otal__________

506

Num ber reported____

503

100

9 . . ..................
10 or m ore________

4
55
178
105
71
35
29
14
6
6

1
11
35
21
14
7
6
3
1
1

N um ber not reported.

..................

1
2 . . ................................
3.............................

4 ..........................

5..............................
6 . . ........ ................
7 . . ..................

8..................

Num ber

Fam ily and others

Per cent
distribu Number
tion

262
262
11
98
56
40
19
19
11
4
4

Group
not
Per cent
reported
distribu
tion

241
100

241

100

4
37
21
15
7
7
4
2
2

4
44
80
49
31
16
10
3
2
2

2
18
33
20
13
7
4
1
1
1
3

F A M IL Y IN C O M E

Information was sought as to the total family income for the year
preceding the study for each family visited, the source of this income,
the number of persons maintained on it,61 the occupation of both
father 62 and mother, and any period of unemployment of the father
during the year. Effort was made to determine the relation of either
the total or the per capita income to various socio-economic factors,
such as housing, unemployment of the father, necessity for the
mother’s working, and adequacy of diet.
As it was not possible for the investigators to interview the fathers,
the information on income and unemployment was obtained from the
mothers at the home visits.
The total family incomes for the year preceding the study are shown
in Table 14. It will be seen that almost half the families studied
were existing on incomes of less than $400 a year and that only 10
per cent had incomes of $1,000 a year or more. Of the 39 families
that had an income of $1,000 or more, only 7 had $2,000 or more.
Many families with very low incomes might be cited; as, for example,
a mother and child who had a total income of less than $50 for the
year and two families of 3 persons each, existing on $75 a year for
each family. Each of 14 families reported a total annual income
of less than $100. These figures represent estimates of the income
in money only; any assistance in the form of food, clothing, and shelter
that may have been given by friends or relatives is not recorded. A
m The “ income group ” consisted of the persons who were supported b y the total family income, whether
tney were living m the household as members of the family or boarders or whether they were living awav
from home.
B
J
u F °r the purpose of this study the term “ father” is used to designate the man with whom the mother
was living, whether or not she was married to him and whether or not he was the father of the child under
observation.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

great majority of the families studied were exceedingly poor. The
relative smallness of the incomes reported may be judged by the fact
that food prices in Puerto Rico for staple supplies such as rice and
beans were about the same as in continental United States, but it
should be noted that rents were a small item in the budget, that the
problem of heat for the house did not exist, and that little was spent
for clothing.
Table 14 also brings out the fact that incomes in Ponce were signifi
cantly smaller than those in San Juan. In Ponce 63 per cent of the
families had incomes of less than $400 for the preceding year, whereas
in San Juan only 41 per cent had similarly low incomes.
T a b l e 14.— Annual income of families visited in San Juan and in Ponce, P. R.
Families visited
Total

San Juan

Annual in com e1
Num ber
T otal___________ _______________

Per cent
distribu Num ber
tion

506

Ponce

Per cent
distribu N um ber
tion

328

Per cent
distribu
tion

178

Amount reported______________ _________

378

100

247

100

131

100

Less than $100- ........................................
$100, less than $200___________________
$200, less than $400___________________
$400, less than $600___________________
$600, less than $800_____ _____________
$800, less than $1,000_________________
$1,000 or more—...................... . ........ .

14
36
134
85
47
23
39

4
10
35
22
12
6
10

7
18
77
61
34
16
34

3
7
31
25
14
6
14

7
18
57
24
13
7
5

5
14
44
18
10
5
4

Amount not reported____________________

128

1 Aggregate amount contributed to family support.

Table 15 shows the number of persons maintained on the incomes
reported. Comparison of the incomes of the smaller families (four
persons or fewer maintained) with those of the larger families (five per
sons or more maintained) shows that, though the smaller families
tended to have slightly smaller incomes, on the whole incomes did not
increase significantly with increase in number maintained. In other
words, as would be expected, the larger the group maintained the
smaller was the income per person and the less satisfactory the eco
nomic conditions in general.
T

able

15.— Annual income of family and number of persons maintained on this

income; families visited in Puerto Rico
Families visited
N um ber maintained on family income

Annual incom e1
Total
3 or fewer
T otal_____________ ............

506

Less than $200_____ ____ ______
$200, less than $400_____________
$400, less than $600_____________
$600, less than $800_____________
$800, less than $1,000___________
$1,000 or m ore__________________
Am ount not reported__________

50
134
85
47
23
39
128

6 or 7

8 or more

146

23
46
23
10
7
15
22

12
30
23
9
5
7
9

6
27
18
12
4
3
6

8
22
13
12
5
5
12

9
8
4
2
9
5

1 Aggregate amount contributed to family support.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

N ot re
ported
75
1

SOCIO-ECONOMIC CONDITIONS
SOURCE OF INCOME

The source of income— that is, the persons who furnished it— was
known for 498 of the 506 families visited. For 474 families it was also
known whether the father or the mother or some other person was the
chief support. In all but 24 of these families the mother or the father
was the sole or the chief support. As will be seen in Table 16, in 399
families the father was the sole or chief support, and in 17 he contrib
uted a smaller proportion to the income. In 51 families the mother
was the sole or chief support, and in 120 she contributed a smaller
proportion to the income.
Of the 506 families that were visited only about three-fourths (378)
reported their total annual incomes. Of these all but 1 reported also
the source of income. Among the 377 families whose incomes were
reported and for whom source of income was known, it was plain that
as a whole those whose sole or chief support was the father had greater
incomes than those whose sole or chief support was the mother or some
other person. Of the 332 families whose income was known and who
were supported solely or chiefly by the father, more than one-half had
total annual incomes of $400 or more; of the 45 families supported
solely or chiefly by the mother or some other person less than one-third
had incomes of $400 or more.
T

able

1 6 .—

Persons contributing to family support and person furnishing chief
support; families visited in Puerto Rico
Families visited
Person furnishing chief support

Persons contributing to family support
Total

Father
Total__________________
Father only_______________
Father and m other__________
Father and other__________
Father, mother, and other____
M other only____________
M other and oth er......... .........
Other only i _________
N ot reported_____________

N ot re
ported

Mother

Other i
24

3
3

10
4
1

506

399

258
116
37
20
16
39
12
8

258
97
30
14

9
2
16
24

6
12

9
8

1 Includes relatives and outside agencies.

EMPLOYMENT AND WAGES

Because of the fact that most of the children examined in this study
lived in cities, it is not surprising to find that very few of the fathers
were employed in the leading industries of the island, namely, the
raising of sugar, tobacco, coffee, and fruit.* A large proportion of the
fathers were engaged as skilled or semiskilled workers or as laborers in
manufacturing and mechanical industries or in transportation. Others
were employed in various branches of trade or in public or domestic
service; a few were employed in clerical positions.
Wages were low, there was an oversupply of laborers, and they had
long periods of unemployment. Wages for men working in the fields
varied from 25 cents a day-to $1.25.63 In the cities wages were a little
higher. Many jobs, however, because of seasonal or other periodic
6i , % a^er? g? daily earnings in various industries see Puerto R ico; what it produces and what it buys,
p. 11 (Trade Information Bulletin N o. 785, Bureau of Foreign apd Domestic Commerce, U. S. Department
pf Commerce, Washington, 1932),
'
’
i

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

variations, provided a relatively small amount of work during the
year; and many, though lasting the year around, provided only two,
three, or four days’ work a week.
In addition, because of the long-continued habit of subsisting on
an extremely small income, many of the men voluntarily worked only
a few days a week. The oversupply of laborers, their general lack of
physical vigor— for which hookworm, malaria, and inadequate diet
were probably largely responsible— and their lack of knowledge of
any mode of life other than that providing the barest subsistence were
probably the underlying causes of this intermittency of work. The
poor physical condition of the average laborer was alone enough to
account for his lack of ainbition for a steady job. He was usually
satisfied with just enough to provide the traditional rice, beans, and
coffee for his family.
The women of the family frequently attempted to supplement the
income derived from the men’s wages, but they were even more under
paid than the men. In the fields and factories women might receive
as little as 40 to 50 cents a day, and for home needlework the pay
was often less. A minimum-wage law attempted to meet the problem
of low wages for factory workers. It did not, however, meet the
problem of handwork done in the home. Many women who had to
stay at home to care for children spent long hours at embroidery or
other forms of needlework, earning $1.25 to $2 a week. Consequently,
even the mothers who worked steadily could add but little, to the
family income. As a rule, the mother who worked probably earned
$25 to $100 a year. Some earned as little as $10 a year, others as
much as $200 or $300. Since it often happened that the mother
worked when the father was unemployed, the income of a family to
which the mother as well as the father was reported to be con
tributing might not be very different from the income to which the
father alone contributed. Other members of the family sometimes
added a little to the family income; but since they had to meet the
same labor conditions as the father and the mother, the amount they
were able to contribute was usually small. The majority of the women
who were employed were engaged in domestic labor or in sewing.
UNEMPLOYMENT OF FATHER

The income was low in many families, partly because the father was
unemployed much of the time. The duration of the father’s unem
ployment, the total annual income, and the source of the income are
known for 334 families. For these families it was reported that 140
of the fathers (42 per cent) had been unemployed for varying periods
of time during the past ye.ar—42 fathers for less than 3 months; 51
for 3 to 5 months; 38 for 6 to 8 months; and 9 for 9 months or more.
The figures for duration of unemployment include both long periods
of unemployment at a single stretch and the estimated aggregate of
unemployment of men employed regularly but for part of the time
only.
Table 17 shows the duration of unemployment of the father and
also the annual income for the 334 families for which the data with
regard to unemployment and income were known. Of the 194
families in which the father was regularly employed, the total income
fell below $400 in only 62 families (32 per cent); whereas of the 140

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SOCIO-ECONOMIC CONDITIONS

families in which the father was unemployed for various lengths of
time, 86 families (61 per cent) had incomes below $400.
T

able

17.— Annual income of family in relation to the father’s unemployment;
families visited in Puerto Rico
Families visited

Total
None

T otal.....................
Less than $200______
$200, less than $400___
$400, less than $600...
$600, less than $800 .
$800, less than $1,000.. . .
$1,000 or m ore_________
Am ount not reported_____

Father
dead, not
living
with fam
Less
3 months, 6 months,
9
months
N
ot
re
ily,
or not
than 3
less
less
or more ported
contrib
months
than 6
than 9
uting
Unemployment of father

Annual incom e1

506

230

50
134
85
47
23
39
128

6
56
45
40
17
30
36

2
11
17
5
3
4
4

6
26
15
2
1
1
6

16
17
4

1
7

i
12

1 Aggregate amount contributed to family support.

FAMILY INCOME IN RELATION TO DIET

The relation of the total family income to the use of various essen
tial articles of diet, such as milk, butter, eggs, and green vegetables,
will be discussed in the section on Family diets (p. 78). Suffice it to
say here that, as would be expected, a definite association was found
between the size of the family income and the frequency with which
the essential foods were eaten by the family. There would seem to be
little doubt that if the families were to receive an adequate income
and some further education, the quality of the Puerto Kican diets
would rapidly improve.
PER CAPITA INCOME

As the per capita income for the year takes into consideration not
only the family’s annual income but also the number of persons main
tained by it, a better picture of the amount of money available to
support an individual over a year’s time is given by per capita income
than by family income. In the study of such factors as the mother’s
diet and the child’s physical condition, therefore, the per capita
income is used instead of the family income. Since it has already
been shown that the total annual family incomes did not vary signifi
cantly with the number of persons maintained, it is obvious that the
per capita income was considerably smaller in the larger families and
larger in the smaller families. Table 18 shows this relation.
Some of the per capita incomes were found to be extraordinarily
small. In 12 families, varying in size from 3 to 14 members, the per
capita income was less than $25 a year. There is no reason to believe
that these figures are not approximately correct. Persistent effort
was made to get all the information possible regarding incomes,
the incredible smallness of the figures acting as a stimulus to the
investigators.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

18.— Annual per capita income and number of persons maintained on annual
income; families visited in Puerto Rico

able

Families visited
N um ber maintained on annual in co m e 1

Total

Annual per capita in com e1

Per
cent N um
dis
ber
tribu
tion

N ot
re
Per 8 or
cent m ore2 port
ed
dis
tribu
tion
37

146

6 or 7

3 or less
Per
cent
N um dis
ber- tribu
N um
tion
ber

T otal_______________

506

Incom e reported----------------

377

100

124

100

$50, less than $100----$100, less than $150------$150, less than $200____
$200 or m o r e .-................

66
130
91
38
52

18
34
24
10
14

10
33
27
21
33

8
27
22
17
27

12
30
23
9
12

14
35
27
10
14

13
28
20
6
3

19
40
29
9
4

21
26
14
1
8

32
40
22
2
5

10
13
7

+ _ . _ n r fn J

129

1 Aggregate amount contributed to family support.
distribution not shown because number of families was less than 50.

2 Per cent

Per capita income and diet of mother.

The relation of the per capita income to the quality of the mother’s
diet will be discussed in a later section (p. 88). Similar study of the
relation of the per capita income to the child’s diet proved to be not
feasible, as at the time of the study the number of children who had
been weaned and were, therefore, dependent on the family budget for
their food was too small.
Per capita income and physical condition of children.

The physical status of the children, as judged clinically by the
amount of subcutaneous fat, varied with the per capita income of the
families to which they belonged, when the group of relatively wellnourished children was compared with the group of poorly nourished
ones. Of 247 children with amounts of subcutaneous fat estimated
as “ very good” or “ good,” 122 (49 per cent) were in families with a
per capita income of less than $100 a year, whereas of 51 children
with subcutaneous fat estimated as “ poor” or “ very poor,” 35 (69 per
cent) came from families with similarly low incomes. As the group
of children with “ poor ” amounts of subcutaneous fat is small, the
difference in the proportions is probably not great enough to be con
sidered definitely significant, but this finding certainly suggests that
low incomes had a direct bearing on the poor physical condition of
many of these children. That such a large proportion of children
(49 per cent) who showed what was estimated to be a “ good” grade
of subcutaneous fat should be found in families in which the per capita
income fell in the lower group is surprising until it is remembered
that the standards for estimating amounts of subcutaneous fat were
in all probability lowered to fit the group in Puerto Rico, and that
the “ good ” grades of subcutaneous fat should be thought of as “ fair ”
according to usual standards in continental United States, and the
“ fair” grade as poor.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SOCIO-ECONOMIC CONDITIONS

H O U S IN G

Interest in the housing conditions of the families included in this
investigation was centered in the amount of crowding in the house
and in the possibility of exposure of the children to sunlight, both in
the houses and out of doors in the yard or patio.
CROWDING

The degree of crowding in these Puerto Rican households is shown
by the relation between the number of rooms in the house or apart
ment and the total number of persons occupying them. (Since the
kitchen in some houses was part of the house and in others was in a
shed or in the patio, it has not been counted in the number of rooms
used by the household.) From Table 19 it is seen that in 98 house
holds of 5 to 6 members each and in 54 of 7 to 10 members each the
members of the household were crowded together in one or two rooms;
one household of 11 members had only two rooms. This crowding
is especially serious in consideration of the size of the rooms, usually
about 8 to 10 feet by 10 to 14 feet. In connection with this crowding
it should be remembered that the risk of transmission of communicable
disease is very greatly increased by such close contact. It is not
surprising that tuberculosis, when once introduced into such a house
hold, spreads rapidly.
Table 20 shows the relation between the number of rooms occupied
by the household and the presence in the household of persons other
than members of the family. That nearly one-third of the house
holds living in one room and two-fifths of those living in two rooms
included some person or persons other than the members of the child’s
family is striking evidence of the economic pressure under which
these families were living.
T a b l e 19.— Number of persons in household and number of rooms (exclusive of

kitchen) occupied by household; families visited in Puerto Rico

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILD KEN IN PUERTO RICO

T a b l e 20. — Number of rooms (exclusive of kitchen) occupied by household and

presence in household of others than members of child’s own family; families
visited in Puerto Rico
Families visited
Group in household
Number of rooms
Family only

Total
Group re
ported

Family and others

Number Per cent Number Per cent

Total__________________

506

503

262

241

1 ........................... ...................
2_______________ ____________
3_____ _____ _____ ___________

165
172
86
76
7

165
172
86
76
4

116
101
26
19

70
59
30
25

49
71
60
57
4

30
41
70
75

Group
not re
ported

FURNISHINGS

N ot only did the families lack the household water supply and sani
tary conveniences usual in cities and towns in continental United
States, but their homes were bare of the household furnishings and
utensils that are ordinarily considered essential. A few families had
beds, but the majority of the houses that were visited had only canvas
or rope hammocks that could be taken down and put aside during the
day. In the city dwellings and in some of the larger frame cottages
chairs were found occasionally. Nearly every house had a small table
or a built-in shelf that could be used as a table, and most of the houses
had a small charcoal stove on which all the family cooking was done
and on which the water for washing clothes and the flatirons were
heated. In the poorer families tin cans were used as cooking utensils
or even as a substitute for the stove itself.
EXPOSURE TO SUNLIGHT

The majority of the families included in the study lived in the largest
cities on the island, San Juan and Ponce, and some lived on the out
skirts of San Juan. The children in Ponce had much better opportu
nities to get sunlight than did the children of San Juan and its out
skirts. The types of houses varied considerably in the different sec
tions of the cities, but they had one characteristic in common, namely,
a complete absence of window glass. The great majority of the houses
were provided with latticed shutters for windows and doors; and as
these doors and windows were almost invariably open except in stormy
weather, sunlight entered most of the houses or apartments at some time
during the day. A few apartments, mostly in the tenements of San
Juan, admitted no sunlight at any time. These will be discussed later.
More than four-fifths of the dwellings investigated had direct
access to a yard or patio where much of the family housework
was done and where the children could play. Such yard space was
sometimes inclosed within a large building as a court, or between two
buildings, or, as in the suburbs or b a r r i o s , it was frequently a small
plot of ground surrounded by several frame houses of huts. Of the
dwellings reported to have no patio or yard, a majority were either
huts crowded together or city tenements.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SOCIO-ECONOMIC CONDITIONS

Types of houses in relation to sunlight

The houses in which the families lived varied in type considerably
with their location. In the city of San Juan proper, the most charac
teristic dwellings were the tenements; some of these had been con
verted from old Spanish buildings and some had been built recently.
In the better suburbs of San Juan, such as Santurce and Barrio Obrero,
small frame houses and r a n c h o n s , or row houses, predominated. In
the poorer districts on the outskirts of San Juan, such as La Perla,
Puerta de Tierra, Tras TaUieres, and the swamp district, the houses
were largely huts or shacks, hastily and poorly constructed. In Ponce
the majority of the houses were either small frame houses or huts.
These are more characteristic of the usual Puerto Rican dwelling than
are the tenements of San Juan.
Of the 556 children whose homes were visited, 113 (20 per cent)
lived in the tenements of San Juan proper, 144 (26 per cent) lived in
frame houses or r a n c h o n s in the suburbs of San Juan, 104 (19 per cent)
lived in huts in the poorest districts on the outskirts of the city, and
190 (34 per cent) lived in Ponce, where the characteristic dwelling was
a small frame house or hut. For five children living in San Juan the
district in which they lived was not reported.
S a n J u a n p r o p e r . — T h e tenements of San Juan proper were of four
kinds. One kind consisted of tenements converted from old Spanish
private dwellings. Most of these were two or three stories high; and,
as they were ordinarily surrounded by similar high buildings, the pa
tios, which were either in the center of the house or next to the wall of
an adjoining building, were small and in a few cases completely shaded.
Frequently there was an overhanging balcony around the patio on two or
three sides, so that the rooms below were much darker than those above.
In some of these tenements the patio was sunny for three to four hours
during the day; in others the patio as well as the lower apartments re
ceived only reflected sunlight. There were many apartments in each
of these houses. A few of these apartments did not open to the out
side and were dark, with the exception of those on the top floor, which
had skylights. This type of house had less sunshine than any other.
A second kind consisted of converted public buildings, such as mon
asteries or palaces. These were usually two stories high, built of
stone, with stone floors in the rooms and patios. The original large
rooms had been divided by low, thin partitions into apartments, the
larger rooms being turned into three or four apartments, the smaller
ones into one or two. These apartments usually had one to three
small rooms, each room being about 8 to 10 by 10 to 14 feet in size.
Frequently one room opened on a patio or balcony, and another on
the street. Those apartments having three rooms usually had a cen
ter room with no opening to the outside except through the other two
rooms. Since the two end rooms were small and usually sunny during
some part of each day, these center rooms were, as a rule, not totally
dark. Some of the buildings, however, had basement apartments that
were totally dark, having no window nor door opening to the outside.
The third kind of tenement, which, on the whole, had more sun
than either of the kinds previously described, was built on the side of
a hill, with a series of patios on different levels or with a patio on the
ground floor and receding balconies on the upper floors. The apart
ments in the buildings had one to four rooms, usually with at least one
opening on a sunny patio or balcony, In some of these buildings

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

there were dark apartments that opened only on a hall; but, even if
the sun did not actually enter the house, there was opportunity for the
occupants to have easy access to sunshine for several hours each day.
The fourth kind, recently built and few in number, was constructed
so that sunlight entered all apartments and was easily available in the
patios. Such buildings were one to two stories high, and each sur
rounded a good-sized, sunny patio. In many cases apartments of
three or four rooms extended from the street to the patio, with an
opening at each end. As the partitions between the rooms were usu
ally low, considerable light reached the inner rooms from the door or
windows that opened into the street or patio. The tenements that
were placed on the three sides of the patio away from the street were
divided into one and two room apartments, the front room of each
opening on the patio. Usually the patio was large and the building
low enough so that it was sunny nearly all day long. Moreover, most
of these apartments had sun shining into one or more rooms for part
of the day, so that on the whole they constituted the sunniest type of
house in San Juan proper.
It should be pointed out in conclusion that a great majority of the
apartments in all these tenements had access to a sunny patio or to
the street and that only a few of the apartments themselves were
totally dark.
O u t s k i r t s o f S a n J u a n . — Santurce, Barrio Obrero, Sunoco, and Rio
Piedras were places on the outskirts of San Juan from which children
were brought to be examined. The houses there were smaller than in
the tenement districts and were in general of four types: Frame cot
tages, huts, r a n c h o n s , and 2-story houses.
The majority of the people in these places lived in little frame cot
tages that were constantly exposed to sunlight. These houses had
one, two, or four small rooms averaging about 8 feet by 10 feet in size.
Usually there was a door opening from each room to the outside, and
each house had two to four windows. The houses were raised f r o m
the ground on stilts, some only a few inches, others several feet. In
Barrio Obrero and. Rio Piedras the houses were arranged along streets,
each house having its own yard. Some of the houses were shaded by
trees, but the great majority were very sunny. In Sunoco and San
turce many of the houses were like those in Barrio Obrero. In some
sections, however, they were not built along the streets but were
grouped together irregularly with openings here and there resembling
patios or yards. Occasionally the houses were fairly close together
with little or no yard space, but since they were low and small theywere nearly all very sunny. A few small huts, similar to those in the
swampy districts, to be described later, were found scattered among
the other types of dwellings in Santurce.
Numerous r a n c h o n s , or rows of one-story houses, were found in Santurce. They varied somewhat in size and shape and were divided
usually into apartments that were four rooms deep with a door at
each end. The rooms were very small, so that light passed easily into
the inner rooms. Despite the fact that these r a n c h o n s were often
built close together, it was found that usually one room of each apart
ment was sunny for a part of each day. Most of them opened into a
sunny yard, so that sun was easily accessible to anyone living in these
houses.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SOCIO-ECONOMIC CONDITIONS

In addition, there were a few small 2-story houses in Santurce.
Some were occupied by one family, more by two families, and many
by four or more families. As a rule, each family occupied one room.
Most of these houses were sunny, but some of them that were buüt
close together along the main streets were not sunny nor had they
available sunny patios.
In the poorest districts on the outskirts of San Juan (La Perla and
San Miguel, Pqerta de Tierra, and Tras Tallieres), as well as along
the edges of Santurce in the swampy region, the houses were, as a
rule, the poorest type of hut or shack.
La Perla and San Miguel are two districts that extend down the
cliff from the city wall to the edge of the sea. The bill is very steep,
and many of the houses stand on high stilts on one side and are
crowded together on the face of the cliff, without streets or sidewalks.
The shacks usually had but one room about 10 feet square; they were
constructed of ail sorts of material, such as pieces of boxes or sheets
of iron. Sometimes there was a window, with or without a shutter,
but ordinarily the only opening was the door. If there was a second
room it had, as a rule, no opening to the outside. In these huts one
or more families lived, crowded together. There was no yard in
which the children might play. The sanitation of these districts was
very poor, few latrines were provided, and garbage and refuse were
tlmown out into the passageways between the shacks. However, both
districts were sunny; and though there was little space where the
children could play in the sun, they must have had the benefit of a
considerable amount of the sunlight that entered the huts through
doorways or windows.
Puerta de Tierra was divided into two parts. The upper part was
dry and sunny, with little frame houses and a few 2-story tenements.
In the lower swampy part, an open sewer ran down the middle of the
street and overflowed under the houses when the tide was high. The
huts resembled those in La Perla, except that they were perhaps even
more dilapidated. They were all on stilts, and the ground under
many of them was very wet. The huts were exposed constantly to
sunlight, but there was no play space for children around them.
In Tras Tallieres and the swampy regions of Santurce the sanitary
conditions were somewhat better, as there were no open sewers here.
On the higher ground stood little frame houses, and on the lo wer, huts sim
ilar to those already described. Some of the houses stood on stilts over
water and others were on little mounds surrounded by a ditch and
marsh; these had to be approached by means of narrow plank walks
over the swampy land. Many of the houses had no sunny play space
about them, but the houses themselves were almost invariably sunny ,
P on ce.
The types of houses found in Ponce were very similar to
those found in the outlying districts of San Juan. There were a few
blocks of 2-story buildings in the downtown section; but most of the
people lived in small frame houses, in huts, or, occasionally, in 2-story
houses, which were scattered over the plain and upon the hills with
nothing to shade them. The ground upon which these houses were
built was dry, and the sanitation was much better than in the poorer
districts of San Juan. Nearly all the houses of Ponce had plenty of
sunshine available within the house.
It is readily seen from this discussion of types of houses that with
the exception of some of the worst of the tenement houses in San

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

Juan, practically all the Puerto Rican houses that were visited had
some sunshine easily available either in the patio or in the house itself.
In the tenements of San Juan proper, where the apartments were
dark, sunlight was usually available in the neighboring patios; in the
huts of the swampy districts, where sunny play space was more or less
lacking, sunlight was usually available inside the house or in the door
way or on a porch.
Availability of sunlight in houses and diagnosis'of rickets.

Each of the houses studied was graded according to the approximate
length of time that sunlight entered by either door or windows. At
the time of year when the study was made, every house had both win
dows and doors open all day long, and most of the mothers stated that
they closed them only when it rained hard or when it was very hot.
Table 21 shows the length of time that the houses were reported to be
sunny. A large proportion (512, or 92 per cent) of the children stud
ied lived in houses that were sunny part of the day at least. About
65 per cent of the whole group lived in houses into which the sun shone
through the open door or through unglazed windows for five or more
hours a day. These were chiefly the frame houses or huts common in
the suburbs and swamp districts of San Juan and in Ponce. One
hundred and twenty-six children (23 per cent) lived in houses or apart
ments that had sun two to four hours a day. Most of the apartments
were in r a n c h o n s , tenements, or 2-story houses, with the sunny doors
or windows on one side only, so that the sun shone into the apartment
only part of the day.
The 26 apartments that were sunny for less than two hours daily
were nearly all in the tenements of San Juan. They had doors or
windows opening under balconies or on patios surrounded by high
buildings. In some of these houses the sunlight entered only for a few
feet. Twenty-seven children lived in houses into which the sun did
not shine at all but which were near a sunny patio or open space. A
few of these houses were fairly dark, but as the sunlight was easily
accessible, they were graded as slightly better than the 14 houses that
not only were dark but also had no sunny yard nor patio. Three of
these 14 latter houses opened into small, dark patios, 6 opened into
narrow passageways, and 5 opened only into hallways.
T a b l e 21. — Children living in houses having estimated number of hours of sunlight

a day; families visited in Puerto Rico
Children in families
visited
Estimated number of hours a day of sun
light in house
Number

Per cent
distri
bution

556
Hours reported....... ......................... .......... •5 hours or m ore.................
...............
2 to 4 hours_________________________
Less than 2 hours___________________
None in house, but sun easily accessible____ _________________________
N one_____________________ _______

558

100

360
126
26

65
23
5

27
14

5
3

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SOCIO-ECONOMIC CONDITIONS

The almost complete absence of roentgen-ray evidence of rickets
goes to show that the amount of sunlight received was adequate to
prevent the disease in nearly all the children. Moreover, it is obvious
that there were too few cases of rickets to make it possible to draw any
conclusions regarding the effect of the different amounts of sunlight
that were available indoors. The fact that sunlight was available in
doors as well as out is, however, probably of greater significance in the
prevention of rickets among the infants, who spent a large proportion
of their time in the house, than among the older children, who could
go out by themselves. As has already been pointed out (see p. 46),
only one infant showed severe rickets, and it is of interest to note here
that this child lived in a house that received no sun at any time and
that he was not taken out of doors at any time. The group of children
for whom a clinical diagnosis of rickets was made is also too small to
show what part was played in the diagnosis by the amount of sunlight
available in the house.
The influence of the amount of sunlight available in the house on the
prevention of rickets may be shown indirectly, however, through its
relation to the presence of tanning of the skin, which is evidence of
direct exposure to sunlight. A larger proportion of children who were
tanned than of children who were not tanned lived in houses that were
sunny for more than four hours each day or had a sunny yard available
as play space.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

D IE T S OF P U E R T O R IC A N S 64

Poverty, local customs, and climatic conditions in Puerto Rico tend
to limit the diet of the people. The basis of the diet was rice and
dried beans, which, though largely imported, were the cheapest foods
available. Tuberous vegetables and small amounts of meat or salt
codfish were also used, and fruit in small quantities. There was not
an adequate local supply of such foods as milk, butter, eggs, and fresh
vegetables, which in continental United States are considered essen
tial for the health of adults and for the health and normal growth of
children, nor could the people as a rule afford to buy them. Besides,
most of these families had no way to keep these perishable foods from
day to day, and they had to buy them in small quantities if at all.
The diets of the families visited were unbelievably monotonous;
there was very little variation from day to day and from family to
family. The diets were poor in minerals, vitamins, and protein;
many were low in caloric value also. The daily diet for the average
family can be summarized as rice and beans twice a day— usually in
large amounts, especially of rice; coffee twice or three times a day,
with unrefined sugar and with or without a small amount of milk;
bread in very small quantities, usually without butter; tuberous vege
tables; and fruit in small quantities, the fruit chiefly for the children.
Some families had nothing but rice and beans, and a few were found
that did not even have both rice and beans every day.
The traditional dish of rice and beans was kept from being too
monotonous by varying the method of cooking the rice, by using dif
ferent types of beans (such as black beans, kidney beans, and navy
beans), and by using peppers and spices for flavoring sauces made
from the beans or by using olive oil on the rice in place of bean sauces.
Nearly all the families visited during the study— all except the very
poorest— used some vegetables other than beans several times a week
at least, as well as small amounts of meat or dried codfish. By use of
these additional foods they added some vitamins and minerals to their
diets, but not enough to bring the diets up to recognized standards.
The additional foods were often served as stew or a cold dish similar
to stew that was called “ salad.” A small amount of meat or salt cod
fish, rarely more than a quarter of a pound and often less, was custom
arily cooked with tuberous vegetables, such as white potatoes, sweet
potatoes, or yautias, or with green platano (plantain), and sometimes
with a small amount of tomato, cabbage, or beans. When such a
stew or salad was served it was likely to take the place of rice and
beans at one meal or to be served with rice in place of beans.
In many households enough rice and beans were prepared at noon
to serve at both the noon and evening meals. The lack of cooking
facilities made it impossible to have much variety of food at any meal.
« For further information on Puerto Rican diets see the series of bulletins entitled Tropical Foods, by
Elsie Mae Wlllsey, published b y the University of Puerto Rico, R io Piedras. (First bulletin of series dated

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

As there were no facilities for storage of food or for refrigeration,
only the staple supply of rice and beans could be kept on hand; all
other foods— vegetables, meats, fruits, and milk— had to be bought,
prepared, and eaten on the same day. In the market women were
seen buying a part of a tomato, one onion, a few tuberous vegetables,
or a few peppers as the daily supply for the family. Milk was often
bought in very small quantities— even a cupful at a time— from street
vendors. Milk for the baby may have been bought from a dairy.
The families who were better off than the average and who lived
near enough to a market so that they could go there often were able
to add more variety to their diets in the way of fruits and fresh green
vegetables. Among the vegetables added were string beans, water
cress, lettuce, cabbage, celery root, as well as gandules (pigeon peas),
chayote (a vegetable resembling summer squash), and other vegetables
grown on the island.
Fruits— oranges, bananas, mangoes, pineapples, and cherries—
were used more or less regularly by a large proportion of the families,
although the adults did not eat fruit so often as did the children. As
there were no facilities for baking in the average home and as bakery
food was both expensive and not easily available, bread and pastry
were little used. Cereals, except rice, were used only occasionally.
Raw sugar-cane was eaten as a sweet by both adults and children.
Sweets other than sugar were used hardly at all. Granulated sugar,
unrefined, was used freely in coffee, but not for other purposes.
Coffee was used customarily in all the households one or more times
a day by the adults, and frequently by any or all of the children who
were no longer breast fed. When a child was weaned, he was all too
often given coffee, sometimes with milk, sometimes black. Most of
the adults drank their coffee black, although some added a little milk.
Practically all the milk taken by the adults was the small amount
that they put in their coffee. Though some milk was provided by
many families for the children, relatively little was bought for the
adults. Some families, who bought no fresh cow’s milk, used small
amounts of condensed, evaporated, or dried cow’s milk, or of goat’s
milk.
The price of butter was so high as to be practically prohibitive.
Eggs were used very little.
Detailed information was sought from the 506 families visited
regarding the use of certain foods that in continental United States
are considered necessary for an adequate diet. Estimates were ob
tained of the amount of milk used daily, the amount of butter and the
number of eggs used weekly, and the number of times a week that
meat or fish, fruit, leafy vegetables, and other vegetables were used.
Effort was made to ascertain that the amounts reported were approxi
mately those that were used customarily.
The data on these foods will be taken up under the following gen
eral divisions: (1) The diet of the family as a whole; (2) the diet
of the mother of the family, with special emphasis on the food taken
during periods of pregnancy and lactation; and (3) the diet of the
child, as reported by the mother.
It should be borne in mind that nearly all the families studied
lived in cities or on the outskirts of cities and therefore had easier
access to markets than did many rural families, and that they were
practically all families that were being taught by public-health nurses

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

to use such essential foods as milk and leafy vegetables. The diets of
these families were probably better, therefore, than those of the
average Puerto Rican family.
F A M IL Y D IE T S
M IL K

Information with regard to the purchase of milk was obtained from
487 of the 506 families visited. Only 10 of these families bought no
milk at all; 15 bought fresh milk for the use of one special member of
the family only, usually the youngest child or the mother; 393 bought
fresh milk for the use of the whole family (both these groups were
able to report with fair accuracy the quantity purchased daily); 69
bought canned milk, but reported the quantities so roughly that no
attempt was made to record them.
Of the families who reported the quantity of fresh milk purchased,
it will be seen that nearly two-fifths bought less than 1 liter 65 a day
(as a rule about half a liter, or 1 pint). Nearly one-half the families
bought 1 liter a day or a little more. Only about one-sixth bought
as much as 2 liters a day. Considering the fact that each of these
families averaged four members and that each included at least one
child under 3 years of age, who should have had milk as the basis of
his diet, or a child who was still breast fed and whose mother there
fore should have been drinking milk freely, these quantities are small.
Compared, however, with the per capita consumption of milk by the
population of the island as a whole—namely, 1 to 2 ounces daily—
these quantities seem relatively good and illustrate what could be
accomplished educationally by the nursing staff of the department of
health throughout the island if the supply of milk were greater and
more easily available.
Families with incomes smaller than $400 yearly purchased consid
erably less milk than those with incomes of $400 or more. Less than
one-half the families with the smaller incomes and more than threefourths of those with the larger incomes bought a liter or more a day.
BUTTER

Butter was used in very small amounts by the majority of the fam
ilies interviewed; 31 per cent of those reporting used no butter at
all, 23 per cent used less than a quarter of a pound a week; 35 per
cent, a quarter of a pound to a pound; and 12 per cent, 1 pound or
more. No family used more than 3 pounds a week; and though in
general the larger families used somewhat more butter, some of the
largest, even those consisting of 10 or more members, used less than
a quarter of a pound a week. On the whole, the families that used
very little milk were the ones that used very little butter also.
As with milk, the poorer the family, the less butter was used.
About two-thirds of the families whose total annual incomes were less
than $400 each used no butter, or less than a quarter of a pound a week,
and about one-third of the families whose incomes were $400 or more
used similarly small amounts.
EGGS

Very few eggs were used by the families interviewed. Thirty-eight
per cent of the familes reporting had used no eggs in the week before
« A liter is approximately the equivalent of 1 quart.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

the interview; 35 per cent had used not more than half a dozen; 18
per cent, half a dozen to 2 dozen; and 9 per cent, more than 2 dozen.
As with milk and butter, the poorer the family the fewer eggs were
used.
M E A T A N D F IS H

Meat, or more often dried codfish, was reported to have been used
several times a week by all but four of the families interviewed;
nearly half the families reported using it less than five times a week.
In this connection it should remembered that meat and codfish were
usually used in very small amounts, chiefly as flavoring in stews
or salads. Rarely was more than a quarter of a pound used daily
by the average family. The families with incomes of less than $400
were reported to use meat somewhat less frequently than those with
larger incomes.
F R U IT

Fruit, chiefly citrus fruit, was used by the children in a large pro
portion of the families interviewed, only 9 per cent reporting that
no fruit at all was used and 52 per cent reporting that fruit was used
at least once a day. The families with incomes of less than $400 used
fruit less frequently than did those with incomes of $400 or more.
VEGETABLES

Green vegetables, such as lettuce, water cress, spinach, cabbage,
and string beans were little used by the families interviewed. Twentyone per cent reported that they never used such vegetables; 36 per
cent, that they used them once or twice a week; 24 per cent, three or
four times a week; and 18 per cent, five or more times a week. Like
the other articles of diet discussed, green vegetables were used more
often by families whose incomes were $400 a year or more.
In addition to the inquiry on green vegetables, inquiry was made
with regard to the use of other types of vegetables, including dried
beans and root vegetables. Such vegetables were used very frequently
and with rice formed the basis of the great majority of the diets. Only
7 per cent of the families interviewed reported using these vegetables
less often than 7 times a week, 44 per cent used them from 7 to 10
times a week, and 49 per cent used them 11 or more times a week. As
would be expected, these vegetables, which, together with rice, are the
cheapest foods obtainable and have become by custom the basis of the
diet of all Puerto Rican families, were used about as often by the fam
ilies with incomes smaller than $400 a year as by those with incomes
of $400 or more.
In general the family diets were found to be deficient in many of the
foods considered essential to a diet for growing children and for preg
nant or lactating mothers. Poverty seemed to be the chief cause of
the inadequacy of these family diets. Larger amounts of milk, butter,
and eggs were bought by the families with incomes of $400 or more a
year than by those with incomes of less than $400 a year. Meat or
fish, leafy vegetables, and fruit were used more frequently by the group
with the better incomes; beans and tuberous vegetables about equally
by the two groups.
The families visited in Ponce used less milk and butter, but consid
erably more beans and tuberous vegetables than did the families visited

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

in San Juan, who on the whole, had larger incomes. Apparently leafy
vegetables, fruit, and small amounts of meat were used about equally
in the two cities. Fewer of the families visited in Ponce than of those
visited in San Juan used eggs.
D IE T S OF M O T H E R S

Information was sought regarding the diets of the mothers during
periods of pregnancy and lactation, with the idea that study of the
diets of those periods might throw some light on the general health of
the mothers and children. As a basis for comparison with these diets
information was sought also regarding the diets of the mothers when not
pregnant or lactating. That the diets of the mothers were very poor
at all the times studied will be shown in the discussion on pages 82-88.
Each mother was questioned as to the amounts and kinds of food
she had eaten on the day before the interview and also whether this
food differed from the food eaten during her recent pregnancy and, if
so, in what respects. If she was not lactating, she was asked whether
the food eaten on the dajr before the interview differed from the food
eaten when she was lactating and in what respects it differed. Inquiry
was made with regard to foods considered essential for good nutrition,
emphasis being placed on those containing calcium, protein, and vita
mins in largest amounts— milk, butter, eggs, meat, fruit, and leafy
vegetables. Inquiry was also made with regard to the use of dried
beans, tuberous vegetables, and rice and other cereals. The amounts
of sugar and other sweets and of breadstuffs were not recorded, as at
the time the data were gathered it was not intended to compute the
caloric content of the diets.
Exact amounts of food taken by the mothers were not measured,
but approximate amounts of foods eaten by the mothers were esti
mated from the total amount cooked for the family and the portions
served to the members, the mother in particular. The information
obtained from the mother regarding her diet was compared with that
obtained for the diet of the family as a whole.
When inquiry was made concerning the foods eaten during periods
of pregnancy and lactation, special emphasis was placed on changes
in the amounts of milk, butter, eggs, fruits, and leafy vegetables.
The statements made by the mothers as to the amounts and types of
food used were so much alike that the single day’s diets, when considered
together, probably give fairly accurate information on the usual diet of
the Puerto Rican mothers with respect to the special items listed.
Diet histories for the day before the interview were obtained for 484
of the 506 mothers, 287 of whom were still nursing at the time of the
interview, and 197 who were no longer nursing. As 50 of the 506
mothers each brought two children to be examined, and as many of
them reported diets for the two pregnancies or the two periods of lac
tation, or both, the diets were recorded for each period of pregnancy
and of lactation. Diets, therefore, were reported for 531 periods of
pregnancy and for 505 periods of lactation. Of the 505 diets for pe
riods of lactation, 287 were for periods not yet completed (that is, the
mother was still nursing her child at the time of the interview), and in
these cases the diet reported for the period of lactation is the same as
the diet for the day before the interview; the remaining 218 diets were
for lactation periods completed at varying lengths of time previous to
the interview,

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

Before the discussion of the mothers’ diet, a word should be said
with regard to the general health of these women. No data were col
lected with regard to weight, but it was the impression of the examin
ing physician that the women were in general poorly nourished. Most
of them also were small in stature. As with the children, this may
have been due partly to race and partly to prolonged underfeeding.
The women of 25 or 30 years of age looked old and worn, and many
of them were listless. Though most of them were well tanned, many
showed pallor of the mucous membranes of the lips. It was esti
mated that the average mother probably weighed not more than 50
kilograms (110 pounds), and a weight of 50 kilograms has been used
as the basis of computations concerning the diets.
In order that the diets might be compared with one another and
with recognized standards and that relationship between the diets
and such factors as family income might be studied, each diet had
to be graded by a definite plan, according to its adequacy in certain
food elements. In the course of two previous studies, one in Wash
ington and one in New Haven, that had to do with the incidence of
rickets, a plan for grading diets had been worked out that was based
upon the foods usually eaten in continental United States. These
diet grades were based primarily on the calcium and protein content
of foods rather than on their caloric value.
Though the vitamin content of the diets was taken into consider
ation to some extent in planning the diet grades, no estimates on the
vitamin content of individual diets have been made. It was soon
found that the gradations planned for judging diets in continental
United States were not suitable for judging the diets in Puerto Rico.
The extreme poverty of the Puerto Rican diets necessitated an exten
sion downward of the usual scale, so as to provide for the inclusion of
diets that were poorer in calcium and protein than any found in conti
nental United States.
A N A L Y S IS O F 112 S A M P L E D I E T S

In order to determine the variation in certain constituents in the
Puerto Rican diets, to be used in the preparation of the grading plan, a
detailed study was made of a sample group of 112 mothers’ diets. The
cases included in this special study were taken at random from three
groups of cases differentiated from one another only by the diagnosis
made at roentgen-ray examination of the children’s bones. Half the
diets were those eaten by mothers of children showing entirely normal
bones, a quarter, by mothers of children whose roentgenograms were
considered somewhat doubtful (see p. 47), and a quarter, by mothers
of children whose roentgenograms showed rarefaction or osteoporosis
of such a degree that it was described as atrophy. The diets to be
studied were selected at random within the groups described, those
eaten by mothers of children with normal bones being selected in ap
proximately equal numbers from the children in San Juan and Ponce.
Of the diets so selected, 59 were those of mothers who were still lactating
at the time of the interview and 53 of mothers no longer doing so.
It was believed that such a sampling would represent the variation
of all the mothers’ diets in calcium content and probably in other
constituents, and that possibly some interesting facts might be brought
out with regard to the relation between the mothers’ diets and the de
velopment of the children’s bones,

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

The approximate amounts of protein and of calcium in each of these
112 diets and also the number of calories were calculated from the data
recorded with regard to certain foods eaten on the day before the
interview. These will be found in Table 22.
T

22.— Caloric, protein, and calcium content of diets of 112 mothers 1 on day
previous to interview and roentgen-ray diagnosis of condition of child's hones

able

Calories
Protein (gms)
Calcium (gms)
Roentgen-ray diagnosis of con Total
moth
dition of child’s bones
A
ver
M
axi
M
ini
A
ver
M
axi
M
ini
A
ver
Maxi M ini
ers
mum mum
age
age
mum mum
mum mum
age
Total__________________
N o rm a l.. ___________________
D oubtful________________
Osteoporotic___ _________

112
56
28
28

870
880
890
831

1,441
1,441
1,318
1,075

220

34.11

66.08

12.02

0.283

390
357

34.67
34.62
32.48

66.08
51. 51
46.57

15.38
12.18

.286
.344
.217

220

12.02

0.789
.760
.789
.524

0.050
.072
.064
.050

1 For foods recorded and for method of selecting these 112 mothers see pp. 80, 81.

Protein content.

The number of grams of protein in the food eaten by each mother
on the day before the interview was estimated, and, as the amounts of
practically all the protein-containing foods, except bread, were re
ported, it is likely that the estimate^ give a reasonably accurate
picture of the protein content of these diets. The maximum number
of grams of protein estimated for a single diet was 66, the minimum
12, and the average 34. According to Sherman,66 who summarized
data from 25 independent investigations, the average protein require
ment is 44.4 grams per 70 kilograms of body weight per day, or 32
grams a day for a woman weighing 50 kilograms. Sherman further
concluded from the studies that if only the b e s t data were used (those
which allowed a “ reasonable period for adjustment to such a low
protein diet” ), the average requirement would probably be lower, or
about 0.5 gram daily per kilogram of body weight. Using this latter
figure, then, as more suitable for estimating the requirement for Puerto
Rican women who are customarily on a low protein diet, it is found that
for a person weighing 50 kilograms, the daily maintenance require
ment would be 25 grams. The “ standard” allowance recommended
by Sherman— one that permits a reasonable margin of safety— is,
however, 1 gram daily per kilogram of body weight,6? or approxi
mately 50 grams daily for a person weighing 50 kilograms. Such a
standard is somewhat more than 50 per cent above the average re
quirement of 32 grams for a woman weighing 50 kilograms and is
twice as much as the estimated requirement of 25 grams a day. The
average number of grams of protein taken by the mothers in this
sample group, then, was probably adequate for maintenance but was
considerably below the standard and obviously did not provide the
margin of safety permitted by such a standard allowance as that
suggested by Sherman.
Calcium content.

The amount of calcium contained in the foods eaten by each mother
the day before the interview was calculated; and since the quantity
of milk, the chief calcium-containing food, had been obtained with
the greatest care, and careful estimates had been made of other cal
cium-containing foods— vegetables, especially dried beans, and fruit—
86Sherman, H enry C .:
87Ibid., p. 511.

Chemistry of Food and N utrition, pp. 221-222. N ew York 1932

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

8 3

it is believed that the information regarding the calcium content of
these diets is reasonably complete.
The maximum amount of calcium in any of these diets was found
to be 0.79 gram, the minimum, 0.05 gram, and the average, 0.28 gram.
This average falls slightly below the amount arrived at when the main
tenance requirement as estimated by Sherman for a person weighing
70 kilograms, namely, 0.45 gram,68is adjusted to the needs of a person
weighing 50 kilograms, namely, 0.32 gram. It should be borne in
mind, moreover, that Sherman states that this maintenance require
ment “ approximates the minimum of actual need rather than a nor
mal allowance,” and that “ the margin for safety should probably
be at least as large for calcium as for protein.” 69
Sherman suggests that a 50 per cent increase above the average
minimum requirement be allowed.70 If this is done, so as to arrive
at a “ standard ” allowance of calcium, it will be found that the average
woman of this Puerto Rican group should have been receiving approx
imately 0.48 gram of calcium daily instead of what she actually
received, namely, 0.28 gram. To provide for the increased demand
for calcium during pregnancy or lactation, moreover, this standard
allowance should again be liberally increased, probably until it ap
proaches the amount suggested by Sherman as the daily requirement
of a growing child, 1 gram per day.71 No such increase in the calcium
content was evident in the diets of the 59 women who were lactating
at the time the diet was taken. The average in the diet of these 59
women was 0.30 gram of calcium as against an average of 0.26 gram
for the 53 women who were no longer lactating.
Caloric content.

Though at the time when the dietary data were gathered it was
not the plan to calculate the caloric content of the mother’s diets, it
was later decided upon because the diets were found to be so inade
quate in other respects.
The caloric content of each daily diet was estimated as closely as
possible from average helpings of reported foods. The maximum
estimated number of calories per day in these 112 diets was 1,441, the
minimum 220, and the average 870. That this average is so very
low is at least partly accounted for by the fact that the amounts of
sugar and other sweets and of bread, articles of diet that are obviously
high in calories, were not recorded. Though sugar was probably used
very freely in coffee, bread and such sweets as pastry and cakes were
used relatively little. If the amounts of these foods used were added
to the total, this addition would, of course, raise the average number
of calories considerably. It is possible, moreover, that the mothers
whose diets were low in the recorded foods increased the caloric con
tent of their diets by using sugar, and possibly bread, proportionally
even more than those whose diets were high in the recorded foods.
If this is true, the number of calories would be increased more in the
diets that were low in the recorded foods than in those higher in the
recorded foods.
There is no way of estimating exactly the caloric value of the sugar
and small amounts of bread that were eaten; but if the general average
of 870 calories were increased 50 or 75 per cent to allow for sugar,
68Sherman, Henry
«»Id.

C.: Chemistry of Food and Nutrition, p. 287,
1

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

bread, and sweets, or even if it were doubled, it would still fall below
the minimum energy requirement of approximately 1,800 calories 72
for a woman of 50 kilograms in body weight who is in normal health
and who spends eight hours in sleep, eight hours in sewing, ironing,
sweeping, or other types of light housework, and eight hours in sitting
or standing at rest.
That the Puerto Rican women were able to maintain their body
weight even at a poor level on such a low energy intake may perhaps
be accounted for partly by the fact that on the whole their work was
light and the climate is warm. It is probably true also, as suggested
by Sherman,73 that ‘ ‘ chronic undernutrition in adults, or even simple
restriction of food consumption in health, if continued sufficiently,
may bring the organism to a lower level of energy metabolism than
would be indicated by the weight or surface. ”
It should be remembered, however, that 59 of these 112 women
not only had to maintain their own weight but had to provide food
for a nursing infant. Moreover, the unsatisfactory nutritional con
dition of the women indicated that the caloric content of their diet
was probably inadequate.
It has already been pointed out that Table 22 shows for each diag
nostic group the average and the range in calcium, protein, and cal
ories. It will be seen that on the average the diets of the mothers of
the children showing atrophy of the bones were but slightly lower in
calories and protein than either of the other groups. They were, how
ever, approximately one-third lower in calcium than the diets of the
mothers of the children with “ doubtful” roentgenograms, and onequarter lower than those of the mothers of children with normal
roentgenograms. The numbers in each group are too small to allow
any conclusions to be drawn. It should be pointed out, moreover, in
this connection that most of the children in the group showing osteo
porosis (18 out of 28) had been weaned and that the calcium in the
mother’s diet can not, therefore, be regarded as indicative of the cal
cium in the children’s diet. It may, however, indicate the amount of
calcium in her diet before weaning took place and thus have some
bearing on the child’s condition.
No explanation of the greater amount of calcium in the diets of
mothers whose children had doubtful roentgenograms is suggested.
On the whole this group of children were younger (mostly under 6
months of age), and most of them were breast fed (21 out of 28). In
the New Haven study doubtful diagnoses were made more often in
young infants, and it had been suggested at the time of this study that
it was perhaps less easy to be certain of the condition of the bones
when growth was taking place rapidly. It is possible that this factor
of rapid growth entered into the doubtful diagnoses in the Puerto
Rican children. The higher calcium content of the diets of the
mothers in this group might, under such circumstances, be associated
with more rapid growth of the child. There is not enough evidence,
however, to substantiate such a hypothesis.
The diets of this sample of 112 mothers were deficient in calcium,
protein, and calories, but the deficiency in calcium is in all probability
the most serious, the calcium content falling approximately 40 per
78Calculated from figures in Chemistry of Food and Nutrition, b y Henry C. Sherman, (N ew York, 1932),
pp. 195-196.
78Ibid., p. 184.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

3DÎETS

cent below the optimum for mothers no longer lactating and even
more below the optimum for mothers who are lactating. The defi
ciency of calcium in the mothers' diets probably indicates a similar
deficiency in the diets of the other members of the family, with the
exception, perhaps, of children who were breast fed or who were re
ceiving more than a cup of cow's milk a day. Though it can not be
demonstrated, it is suggested that the osteoporosis of the bones found
in so many of the children may be a reflection of this deficiency of
calcium in the diet. (See pp. 56-59.)
CLASSIFICATION OF DIETS OF ALL MOTHERS INTERVIEWED

Using the 112 sample diets to show the variation in foods eaten and
in the calcium and protein content of these foods, five grades were
established, into which the diets of all the mothers would fall. These
grades were based primarily on the calcium and protein content of the
diets. The average and range of the calcium and protein content of
the diets in each of these grades and the combinations of foods that
fell within each range are as follows:
Calcium (grams)
D iet grade

Protein (grams)

A ver M axi M ini A ver M axi M in i
age mum mum age mum mum

1. G ood _____

0.66

0.87

0.46

2. Fair............

0.37

0.55

0.18

3. P oor______

0.20

0.28

0.12

4. V ery poor.

0.15

0.24

0.06

5. Exceeding
jo. 08
ly poor.

0.14

0.02

Combinations of food falling within range of cal
cium and protein
M ilk, 1 cup or more a day.
Eggs, 3 or more times a week.
Leafy vegetables, 3 or more times a week.
Beans or tuberous vegetables,74 2 or more times a
a day.
Rice, potato, or bread 2 or more times a day.
M eat, 5 or more times a week.
,Oranges, 3 or more times a week.
M ilk, few teaspoonfuls (in coffee) to 1 cup.
Eggs, 1 or 2 times a week.
Leafy vegetables, 1 or 2 times a week.
Beans or tuberous vegetables, 7 or more times a
week.
Rice, potatoes, or bread, 2 or more times a day.
Meat, 3 to 7 times a week.
.Oranges, 1 to 4 times a week.

'M ilk, none.75
Eggs, or leafy vegetables, 1 or more times a week.
Beans or tuberous vegetables 7 or more times a
week.
24
Rice, potatoes, or bread, 2 or more times a day.
M eat, 3 to 7 times a week.
.Oranges, 1 to 4 times a week.
M ilk, few teaspoonfuls (in coffee).
Eggs, or leafy vegetables, none or 1 to 2 times a
week.
Beans or tuberous vegetables, 3 or more times a
7
week.
Rice, potatoes, or bread, 1 or 2 times a day.
M eat, once a week.
.Oranges 1 to 4 times a week.
M ilk, none.
Eggs or leafy vegetables, none, or once a week.
Beans or tuberous vegetables 3 or more times a
week.
4
Rice, potatoes, or bread 1 or 2 times a day.
M eat, none, or once a week.
(Oranges, none, or 1 to 2 times a week.

7i Tuberous vegetables include, as well as potatoes, such native vegetables as yams and yautias. A t the
iime when the calcium content of these diets was calculated, no figures as to the amount of calcium in the
yautia could be found. It was assumed that in calcium content the yautia was somewhere between the
potato and the turnip, both of which the yautia resembles. Through a subsequent analysis the yautia has
been found to be actually m uch lower in calcium than the estimate used m the present study. (Si,,mhjance of Mineral M etabolism, I. Preliminary Report on Calcium and Phosphorus Content of Some Puerto
Rican Food Materials, b y D . H . Cook and Trinità Rivera, in the Puerto Rican Review of Public Health
and Tropical M edicine, voi. 4, no. 2, pp. 65-69} August, 1928.) The amounts of calcium in these diets,
therefore, are as a whole somewhat overestimated.
„ , „ v ^
,,
.
... ' , ,
75 Though the diets in the third grade contain no milk, they are slightly better than those m either tbe
fourth or the fifth, because they include larger amounts of calcium and protein containing foods other than
milk.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

On the basis of calcium and protein content the diets that would fit
into the first grade could, on the average, be called “ good” and those
near the top of the range within this grade “ very good.”
Whether
those near the bottom of the range should be called “ good” on the
basis of but one cup of milk a day is very doubtful. The second
grade, called “ fair,” provides for the diets that on the average con
tained at least the daily maintenance requirement of calcium and
protein for a woman of 50 kilograms; the diets in the third, fourth,
and fifth grades are all three poor, and fall progressively lower.
The diets in the fourth and fifth grade are both very poor, but may
be distinguished from each other, as the fourth contains a very s m s . l l
amount of milk and the fifth contains none.
Diets reported for day before interview.

Table 23 shows the grades given to the diets eaten by 484 mothers
on the day before the interview and also the calcium content of each
diet group (calcium content not calculated for 2 mothers' diets). The
poverty of the diets is very striking. That only 10 per cent of the
total group of mothers had eaten foods that could be graded as “ good”
is no less worthy of note than that nearly 9 per cent were so. exceed
ingly poor as to fall in a group below the usual “ very poor” one.
The average amounts of calcium and the range calculated for the
different diet groups indicate the quality of each grade from the point
of view of calcium content. As was done in studying the sample of
112 mothers’ diets, the average amount of calcium in the day’s food
was calculated for the total group of 482 mothers. This was found to
be 0.29 gram, an amount almost identical with the average for the
sample and therefore also slightly lower than the amount calculated as
a daily maintenance allowance for a woman of 50 kilograms, 0.32 gram.
(See pp. 82-83.)
T

able

23.— Calcium content of mother’s diet on day previous to interview, and
diet grade; families visited in Puerto Rico
Calcium content (gms) of
m other’s d ie t 1

Families visited
Diet grade
Total

Per cent
distribution

Average

M axim um M inim um

Total__________

506

Grade reported...........

484

100

0.287

1.345

0.048

G ood ____________
Fair_______ ____ _
P oor____________
V ery poor_______
Exceedingly poor.

48
194
93
2 107
2 42

10
40
19
22
9

.614
.338
.211
.180
.104

1.345
.638
.328
.238
.145

.456
.183
.127
.067
.048

Grade not reported...

1 For foods recorded see p. 85.
2 Includes 1 diet for which calcium content was not reported.

Diets reported for periods of pregnancy and lactation.

With a view to determining whether any improvement was made by
Puerto Rican women in their diet during periods of lactation, the
foods recorded as eaten on the day before the interview by the mothers

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

4 k

who were still lactating were compared with the foods recorded for
those no longer lactating, and almost no difference was found.
Apparently, as a general thing, mothers in Puerto Rico do not or can
not get a better diet during lactation than that which they take
usually. Moreover, no significant difference was found between the
lactating and the nonlactating women with regard to the inclusion of
milk in their diets, and the average calcium content of the diets was
found to be about the same in the two groups, the diets of the mothers
who were still lactating averaging 0.29 gram and those of mothers no
longer lactating averaging 0.28 gram. As was true of the sample
diets already discussed, the calcium content of the diets of the
mothers who were carrying the extra burden of lactation was, there
fore, not only far below the “ standard” allowance (see p. 83), but
even farther below the allowance desirable for lactating women.
Table 24 shows a comparison of the grades given to diets eaten by
mothers during 531 periods of pregnancy and during 505 periods of
lactation (combining complete and incomplete periods) with the grades
given to diets eaten on the day previous to interview by 197 mothers
who were no longer lactating. The latter group probably represents
the usual grade of diet eaten by Puerto Rican women.
Contrary to what might be expected, however, it can be seen that
the diets taken by the mothers during pregnancy were slightly, but
definitely, better than those taken on the day previous to the inter
view and also better than the diets taken during the lactation periods.
The greatest difference was found in the proportion of mothers who
during pregnancy took enough milk to have their diets classified as
good. Though the differences found are slight, they are large enough,
in all probability, to be considered significant and to indicate a tend
ency on the part of some Puerto Rican women to improve their diets
during pregnancy, though the same effort apparently was not made
during lactation.
T a b l e 24. — Diet grade of mothers during pregnancy, during lactation, and on day

previous to interview for mothers not lactating on that day

Periods of
pregnancy

Periods of lactation

Mothers not laetating on day previ
ous to interview

D iet grade
Number

Per cent
Per cent
Per cent
distribu Number distribu Num ber distribu
tion
tion
tion

T otal________

i 556

Grade reported....... .........

531

100

505

100

197

100

98
205
82
109
37

19
39
15
21
7

60
208
89
109
39

12
41
18
22
8

15
76
44
45
17

8
39
22
23
9

G ood_______
Fair_______
Poor__________
Very poor_______
Exceedingly p o o r...........
Grade not reported. . .

«526

2 213

1 Of t]16 506 mothers in families visited, 50 had two children examined, and information was sought with
regard to these mothers diets during both pregnancies.
s Of the 506 mothers in families visited, 50 had two children examined: of the 556 children examined 30
were never breast fed.
’
3 Of the 506 mothers in families visited, 293 were lactating on the day previous to interview.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

Grade of mother’s diet in relation to income.

That the quality of the mother’s diet is closely related to income is
shown in Table 25, and has already been referred to in the section on
Income (p. 68). Since individual diets are under consideration, the
income per capita has been used rather than the total family income.
As would be expected, income influenced the quality of the mother’s
diet to a great extent. Of the mothers reported as receiving good or
fair diets, nearly two-thirds were in families with per capita incomes
of $100 or more; of those who had poor diets, only one-third had
incomes as high. The high cost of milk in Puerto Rico and its
scarcity are probably responsible for this, since the amount of milk
was given great weight in grading the diets, but the high cost of eggs,
butter, meat, and fresh green vegetables is also without doubt a con
tributory factor.
T a b l e 25. — Mother’s diet grade for day previous to interview and annual per capita

income of family; families visited in Puerto Rico
Families visited
Annual per capita income

D iet grade
Total

506
F a ir._______ _______________

48
194
93
107
42
22

$200 or
more

N ot
reported

Less than $50, less
than $100
$50

$100, less
than $150

$150, less
than $200

130

129

7
18
4
8
1

19
20
4
6
2
1

8
41
25
27
15
13

66
1
14
19
18
11
3

7
50
27
32
10
4

6
51
14
16
3
1

DIETS OF CHILDREN
METHOD OF COLLECTING INFORMATION ON DIETS

When the information on Puerto Rican children’s diets was being
collected, an effort was made to collect it in such form that a general
description of the diets from the child’s birth to the time of the exam
ination could be given and that the quality of the diet at the time of
examination could be studied in relation to the physical condition of
the child. The diet information for the child was obtained from the
mother at the time the home was visited, which was, as a rule, a few
days after the physical examination of the child. The food eaten by
the child on the day before the home visit was recorded and also a
statement from the mother giving in as great detail as possible an
account of the child’s diet from the time of his birth to the date of the
visit. As the majority of the children were 1 year of age or under at
the time of the visit, and the rest in the second or third year, it was
possible for the mothers to remember with a fair degree of accuracy
the time at which certain events took place, such as weaning and the
addition to the diet of cow’s milk and of various solid foods. The age
when certain foods were first used was recorded in months. It was
realized, however, that there could easily be an error of one or two
months either way; the time of such changes in diet has, therefore,
been considered as only approximate.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

The exact quantities of food eaten by the child daily were not
recorded, with the exception of milk, but the frequency with which
such articles of diet as fruit, eggs, leafy vegetables or other vegetables,
and cereals (including rice) were used and the time when they were
first used were recorded. The daily quantity of cow’s milk, whether
in a formula or otherwise, and whether supplementary to breast feed
ing or not, was recorded, as well as variations in amount from period
to period.
METHOD OF GRADING DIETS

As with the mothers’ diets, each child’s diet was graded so that it
might be compared with that of the other children and might be
studied in relation to such factors as income and physical condition.
As changes in diet occur at shorter intervals during the first year
than in later ones, the first year of the child’s life was divided for the
purpose of description of his diet into four periods of three months
each, whereas the second and the third years were not divided;
instead, a whole year, or that part through which the child had lived,
was used as the unit for grading. A diet grade was given for each
period, even if it was incomplete. Thus a child 2 months old would
receive one grade— a grade for the first 3-month period, even though
incomplete. A child 5 months old would receive two grades— a grade
for his first 3 months of life and a grade for the second 3 months,
even though the second period was incomplete. A child just a year
old would receive four grades—one for each of the 3-month periods
of the first year.
The second year, whether complete or incomplete, was treated as
a single period and given one grade; the third year was treated in
the same way. Thus a child 16 months old would receive 5 grades—
a grade for each of the 4 complete 3-month periods of the first year
and a grade for the second year, even though incomplete. A child
2K years old would receive 6 grades— 4 for the first year, 1 for the
second, and 1 for the third.
By this method of grading the children’s diets at various age levels,
it was possible to study the adequacy of diets at different ages, to
compare different groups of children on an age basis, and, in addition,
to study the relation of the diet at the time just before the examina
tion (the most recent diet period, even if incomplete) to the physical
condition of the child at the time of examination. It may be said
here that essentially no difference was found in the quality of the
children’s diets taken during a given period with respect to whether
the period was complete or incomplete.
BASIS FOR GRADES OF DIETS AT DIFFERENT AGES

The grades for diets in the different age periods were based in gen
eral upon standards for feeding infants and young children generally
used in continental United States, with some adaptation to fit the
customs in Puerto Rico. Five grades were given: “ Very good,”
“ good,” “ fair,” “ poor,” and “ very poor.” For each age period the
“ very good ” grade represents a relatively good type of feeding but
not always one equal to what would be described as “ very good”
in continental United States. This is especially true of the second and
third periods, in which the standards for foods in addition to m ilk76
76 Infant Care, p. 78, U. S. Children's Bureau Publication No. 8.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Washington, 1933.

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

have been kept low in an attempt to adjust the grading to local
customs. The “ good,” “ fair,” and “ poor” grades also are based on
standards lower than those accepted in continental United States.
For example, the diets in the very good grade for periods two and three
(especially in period three), are not up to the usual standards, which
require that vegetables and eggs, as well as cereal, be begun during
these months. The poor and very poor grades include many diets
that are exceedingly low in both quality and quantity. For grades
given to diets at different age periods see Appendix Tables D l to D6
(pp. 114-119).
The diet grades were based first on the type of milk feeding—
breast feeding or artificial feeding— and, if the child was artificially
fed, the quantity of cow’s milk and secondly, upon solid foods— the
kinds of solid foods, the number of times a week they were used, and
the age at which they were begun. In general, as the child grew older,
breast milk was given decreasing weight in the grading, and cow’s
milk and solid foods, increasing weight. Since nearly every child was
found to receive some orange juice or other form of fresh fruit almost
daily, fruit was not taken into consideration in grading.
In the grading of milk feeding, breast milk was considered the most
satisfactory type for a child during the first 9 months of life. This
was decided upon with full recognition of the fact that in many
instances the mother’s diet was so inadequate as probably to affect
the quality and quantity of her milk. For the first 3-month period of
the first year breast milk, unsupplemented by cow’s milk, was con
sidered the most satisfactory type of milk feeding; for the second
3-month period (4 to 6 months) and the third 3-month period (7 to 9
months) breast feeding, either alone or in major part,77was considered
most satisfactory. For the third 3-month period (7 to 9 months),
if breast feeding was found to be the minor part of the baby’s milk
feeding and 1 pint or more of cow’s milk a day was given to supple
ment it, or if an average of 1% pints of cow’s milk alone was given,
the milk feeding was considered satisfactory.
After the first 9 months of life, breast feeding, unsupplemented,
was no longer considered the most satisfactory type of milk feeding,
but instead, an adequate amount of cow’s milk (1% pints to 2 pints)
or a small amount of breast milk with 1 pint or more a day of cow’s
milk in addition was so considered. During the second and third
years the grading of milk feeding was based on cow’s milk alone, 1%
pints or more a day being considered a very satisfactory amount of
milk, 1 pint satisfactory.
For children under 3 months of age no solid food was required to
have the diet considered very good, a very satisfactory milk feeding
alone being enough.
In the grading of the diets for the age periods after the first three
months, however, increasing weight was given to the addition of solid
foods to the child’s diet, namely, cereal (usually rice), vegetables
(usually some form of dried beans, but occasionally fresh vegetables
such as string beans or lettuce), and eggs. The use of rice water in
ii W hen a child was partly breast fed and partly artificially fed, it was not possible to tell exactly what
proportion of his milk feeding was from the breast. If a child was breast fed for more than half of a 3-month
period or if he received an average of only 1 cup of cow ’s milk a day (less than 12 ounces) in addition, the
breast milk was empirically considered to be the major part of his total m ilk feeding in that period. If he
was breast fed less than half of a 3-month period or received an average of 1 pint of cow ’s milk a day (12 to
20 ounces) or more, the breast milk was considered to be the minor part of his m ilk feeding in that period.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

feeding formulas during the first 9 months of life was counted as the
use of cereal, though the investigators realized that relatively little
value was added to the diet in this way.
For children 4 to 9 months of age only one solid food had to be given
daily in addition to a satisfactory milk feeding to have the diet con
sidered very good or good. That this standard for solid food during
the period from 4 to 9 months was low in the light of modem standards
for infant feeding was recognized; but it proved to be relatively high
for the usual Puerto Rican infant’s diet.
In the age period 10 to 12 months the very good diet grade included
what was considered an adequate amount of all three solid foods;
that is, eggs and vegetables used 2 to 6 times a week each, as well as
cereal once or twice daily. If such foods were eaten in combination
with 1% pints of cow’s milk a day or with a pint of cow’s milk a day
supplementing breast feeding, the whole diet was considered very
good; if 1 pint of cow’s milk a day was given, or only an average of 1
cup supplementing breast feeding, or, if the child’s milk feeding was
entirely from the breast, and cereal, eggs, and vegetables were all
taken in adequate amounts, the diet as a whole was considered good.
The fair, poor, and very poor grades represented less and less adequate
use of solid foods combined with less and less adequate milk feedings.
The grades for the use of solid foods during the period from 10 to 12
months were given as follows:
Grade

Cereal

Eggs and vegetables

♦
Satisfactory_____

Once or twice a day

Each food 2 to 6 times a week.

Borderline_______

Once or twice a day

Either food 2 to 6 times a week
or
One food once a week, the other
1 to 6 times a week.

Poor.

Once or twice a day

Either food once a week or
neither given.

During the second year also the grade of the child’s diet depended
both on the kind and quantity of milk and on the use of cereal, eggs,
and vegetables. No diet was considered either very good or good
unless both cow’s milk and these specified solid foods were well repre
sented. To be graded as very good, a diet had to contain 1 pint or
more of cow’s milk a day, and all three solid foods had to be used to an
extent considered very satisfactory. To be graded as good, a diet
had to contain 1 pint or more of cow’s milk and all three types of solid
foods in amounts considered satisfactory. Whether the use of solid
foods was satisfactory depended not only on the number of times a
week that each was given but on how early in life the use of these foods
was begun. Throughout the grading of the second-year diets, the
time when eggs and vegetables were started was given considerable
weight in determining the grade of the diet. When, as frequently
happened, eggs or vegetables were given in fair or poor amounts or
were not started until the child was 13 months or older the grade of
the diet as a whole was low.
160326°— 33------ 7

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

The grades for the use of solid foods during the second year were
given as follows:
Cereal

Eggs and vegetables

Grade
H ow often eaten

H ow often eaten

1 to 2 times a day—

Each food 5 to 7 times a w eek____

1 to 2 times a d a y ..

Each food 3 to 7 times a w eek____

Both foods at or before 12 months.

1 to 2 times a day___

Each food 5 to 7 times a week____

Both foods at 13 to 15 months.

1 to 2 times a day___

Each food 3 to 7 times a week____

One food at 12 months, the other 13 to
15 months.

1 to 2 times a day___

One food 3 to 7 times a week, the
other 0 to 2 times.

Both foods at or before 12 months.

1 to 2 times a day___

Each food 3 to 7 times a w eek____

One food at or after 16 months, the other
before 16 months,
or
Both foods at 16 months.

1 to 2 times a day.

Each food 3 to 7 times a week,
or
One food 3 to 7 times, the other 0
to 2 times.

Both foods at 13 to 15 months.

1 to 2 times a day___

One food 3 to 7 times a week, the
other 0 to 2 times.

One food at 12 months, the other at 13
to 15 months.

’0 to 2 times a day.

Each food 3 to 7 times a week.

One food at or after 16 months, the other
before 16 months.

0 to 2 times a day.

One food 3 to 7 times a week, the
other 0 to 2 times.

One food at or after 16 months, the other
before 16 months,
or
Both foods at 16 months.

0 *o 2 times a day___

Neither food in diet
or
Either food 1 to 2 times a week.

Both foods at or before 12 months,
or
One food at 12 months, the other at 13
to 15 months,
or
Both foods at 13 to 15 months.

'0 to 2 times a day.

One food 3 to 7 times a week, the
other 0 to 2 times.

One food at or after 16 months, the other
before 16 months.

0 to 2 times a day.

Neither food in diet
or
Either food 1 to 2 times a week.

One food at or after 16 months, the other
before 16 months,
or
.
B oth foods at 16 months,
or
One food at or after 16 months, the other
before 16 months.

0 to 2 times a day___

(78)-

A t what age begun
Both foods at or before 12 months,
or
One food at 12 months, the other at 13
to 15 months.

78 Neither eggs nor vegetables in diet.

The diet grades for the third year also were based on the use of
milk and solid foods, the same standards in general being used as for
the second year.
GRADES GIVEN TO DIETS

Chart I X 79 shows, for successive periods of the first two years of
the child’s life, the percentage distribution of the grades given to the
diets taken by the children for whom diets were reported. It will be
seen that in these two years, according to the grading plan used, the
diets deteriorated rapidly in adequacy. In the early periods, as has
78Foy data upon which Chart I X is based, see Appendix Tables D l, D2, D3, D4, and D5,

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

already been pointed out, the most important element that affected
the grades given to the diets was breast milk; in later periods it was
cow s milk and solid foods— especially solid foods.
D ie t Grades
V e ry ^ o o d

G ood

Fair*

P oor

Very p o or

t
K
.554 ch ild re n

Period I
50
(under 4 months)

P eriod 5
(13-24 m onths)

C hart I X .—Diet grades during specified age periods in the first 2 years of life; Puerto Rican children
1 to 34 months of age

Breast feeding, either alone or supplemented with feedings of cow’s
milk, was continued for more than half the children to the end of the
krst year. During the second year, or that part of it through which
the child had lived up to the time of interview, breast feeding was
contmued for about two-fifths of the children. In the period between

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

10 and 12 months, only about half the children received what was
considered a relatively satisfactory amount of cow’s milk (1 pint or
more), either alone or as supplementary to breast feeding. Nearly
all the rest of the children were still breast fed, either entirely or with
supplementary feedings of less than 1 pint of cow’s milk a day; a
very few received no breast milk and an inadequate amount of cow’s
milk. During the second year (or that part of it through which the
child had lived) half the children received 1 pint or more of milk daily,
the majority of the rest receiving only an average of 1 cup of cow’s
milk daily, either alone or supplementary to breast milk. A very few
children were still entirely breast fed in the second year.
For infants during their first three months of life solid food did not
enter into the grading, the diet grade being based on the type of milk
feeding only. In the diets of children 4 to 6 months of age the use of
any sort of solid food in addition to milk was relatively u n com m on.
Of the 469 children graded for this second 3-month period only a small
proportion (19 per cent) had received cereal in any form. Two chil
dren only were reported as having had vegetables in this second period
and 1 as having had eggs. Since, however, the diet grades in this
period were planned so as not to depend greatly upon the addition of
solid food, but more upon breast feeding, a large proportion of the
diets of children of this age were graded as “ good.”
In the third period, from 7 to 9 months of age, only 144 of the 367
children whose diets were graded (39 per cent) received any solid food
in addition to milk. Many of these received cereal only; only 32
received vegetables, with or without cereal; and only 3 received eggs.
In the fourth period, from 10 to 12 months, 172 of the 293 children
whose diets were graded (59 per cent) received some solid food in addi
tion to milk. Of these, 98 received cereal only (usually rice); 69,
vegetables, (usually dried beans, but occasionally a green vegetable)
either alone or in combination with other foods; and 9, eggs, either as
the only solid food or in combination with others. In spite of the
fact, then, that by the end of the first year of life about half the children
were receiving 1 pint or more of cow’s milk daily, very few could be
graded as having very good, good, or even fair, diets because of the
extreme lack of supplementary foods.
In the second year only 2 of the 203 children received all three solid
foods in “ very satisfactory” amounts; 9 received all three foods in
satisfactory amounts; 27 received vegetables or eggs or both in bor
derline amounts, as well as enough cereal; and 165 in unsatisfactory
amounts, or none at all. Ten received no solid food at all, not even
cereal; these diets were in the very poor grade.
The diets taken by these Puerto Rican children were, then, far below
the standards usually accepted in continental United States with
regard to the quantity of milk and also with regard to the use of
other foods usually considered necessary. It is not surprising that
during the period when a full diet of cow’s milk and solid foods should
have been established the nutritional condition of the children became
increasingly unsatisfactory.
VALUE OF BREAST FEEDING IN RELATION TO MOTHER'S DIET

The classification of the children’s diets has been based upon the
empirical assumption that breast feeding, regardless of the adequacy
of the mother’s diet, is, when supplemented at appropriate ages by

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

certain additional foods, the best method of feeding children under
10 months of age, and that after 10 months an adequate amount of
cow’s milk, with liberal use of certain solid foods, constitutes the best
diet. As the mothers’ diets during both pregnancy and lactation
were on the whole poor (see p. 86), it may well be questioned whether
breast feeding should have been considered very satisfactory for these
children. Recent experimental work has shown that the quality of
the mother’s diet influences the nutritive value of her milk.80 As 95
per cent of the children were breast fed either entirely or partly during
the first 3 months of life, and 77 per cent during the period from 4 to 6
months, there would seem to be ample opportunity to show the effect
of inadequacy of the mothers’ diets on the ability of the mothers to
nurse their children.
Comparison, however, of the diets eaten during lactation by women
who were able to nurse their children through six months with those
who did not nurse them after the first three months shows little dif
ference in the quality of the diets. Indeed, a larger proportion of
mothers who nursed their children through six months than of those
who nursed their children for less time than that reported diets that
fell in the two poorest groups. The quality of the mother’s diets
during lactation apparently did not influence the length of the nursing
period.
So, too, the diets eaten during pregnancy could not be shown to have
influenced the length of the nursing period. The fact that there were
so few women whose diets could be classified as really good may ac
count for the lack of contrast between diets of good quality and of poor
quality in their effect on the length of the nursing period.
Considering the poverty of these mother’s diets, the question may
very well be raised with regard to the desirability of grading breast
feeding as very good, but since the real quality of the breast milk was
not known, and since many of the artificial formulas substituted would
probably have been even more inadequate, as can be judged from the
few reported during these first 6 months (see Appendix Tables D l and
D2), it seemed best to consider breast milk as the most satisfactory
food on the whole for these children in their early months.
CHILD'S PHYSICAL CONDITION IN RELATION TO DIET AT TIME OF INTERVIEW

The assumption that breast milk was on the whole the best food
may be shown to be reasonably correct by comparing the nutritional
condition of children who were examined during periods when breast
milk was the chief part of the diet with that of children who were
examined in later age periods, when breast feeding was less common,
and also by studying the relation of the diets as a whole that the
children were receiving at the time o f examination to their physical
condition.
It has already been pointed out in the section on physical condition
of the children (see p. 26) that the children who were examined during
the first 6 months of life showed on the whole more satisfactory
amounts of subcutaneous fat than those examined during later months,
and also (see p. 22) that the children under 6 months more nearly ap
proached the average weight for age of white children in continental
80 M cCollum , Elmer V ., and Nina Simmonds: The Newer Knowledge of Nutrition, pp. 410-432.
York, 1929.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

N ew

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

United States than did those over 6 months. This alone would point
to the fact that the breast milk that these children were receiving,
regardless of its quality, was in all probability a more satisfactory food
20

Per cent
60
50

Breast feeding
only
141 children

Mixed feeding
(breast mi IK and
cow’s miih)
105 children

Cow’s milK

feeding only
B5 children

Subcutaneous fat

CZ) Good
C

hart

Poor and very poor

Fair

—Subcutaneous fat in relation to type of milk feeding at time of examination; Puerto Rican
children 13 months of age and under

for children of this age than any food that the older children were
receiving. Moreover, when children 1 year of age or under were
studied, a definite relation was found between the food that the child
Per cent
Very ¿ood and
¿bod diet
164 children

0
10
--------- L-

20
1

30
1

•‘»■U
i 1

OU
- 1

vy
1

100

U lli

Fair diet
63 c h ild re n

Poor arid very
poor diet
104 children
Subcutaneous fa t

Good
C hart

Fair

Poor and very poor

X I .—Subcutaneous fat in relation to diet grades at time of examination; Puerto Rican children
13 months of age and under.

was receiving at the time of examination and the child’s physical con
dition as shown by the amount of subcutaneous fat. This relation
may be seen in Charts X and X I. From Chart X there would seem
to be little doubt that during the first year of life children who were

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

DIETS

entirely breast fed up to the time of examination, that is, received no
cow’s milk at all, were the most satisfactorily nourished, whereas those
on cow’s milk alone were the least satisfactorily nourished. In this
connection, however, it should be pointed out that 70 per cent of the
141 children who were entirely breast fed at the time of examination
were at that time under 7 months of age and that only 22 per cent of
those receiving no breast milk were equally young.
When the grade for the entire diet (solid food as well as milk) that
the child was receiving at the time of the home visit was considered
in relation to the amount of subcutaneous fat, a similar though some
what less striking relation was found; this is shown in Chart X I.
From the study of the nutritional condition of the children and of
their diet it would appear that breast feeding, regardless of the quality
of the mother’s diet, was the diet that brought relatively the best re
sults; but it is also apparent that after the children became 6 months
of age even these results were not satisfactory according to general
standards in use in continental United States. If the mother’s diet
had been better, and if adequate solid foods had been included in the
child’s diet in addition to the breast milk, the nutritional condition
of the children would, without much doubt, have been better.
E C O N O M IC C O N D IT IO N OF F A M ILY IN R E LA TIO N T O C H IL D 'S D IE T

The grade of the child’s diet as a whole and the type of milk feeding
were studied separately in relation to several economic conditions—
the per capita income of the family, the number of persons who were
sharing the family diet, and, if the father had been unemployed, the
duration of his unemployment— but no relation was found. The ab
sence of any such relation is probably due to the fact that the great
majority of incomes were so low that contrasts could not be drawn.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SUMMARY

The main part of this investigation consisted in obtaining roent
genograms of the forearm bones of nearly 600 children under 3 years
of age living in Puerto Rico, where children are exposed to sunlight
all the year round. These were obtained so that they might be com
pared with roentgenograms of children of the same age group living
in New Haven, where much less sunlight is available. Besides the
roentgenographic examinations, physical examinations were given to
the children.
As part of the physical examinations, the weight and height of each
child were taken. In weight the group of Puerto Rican children
examined fell considerably below a group of children examined some
years before in continental United States, and in height the Puerto
Rican group fell somewhat below it. The differences for children
under 6 months of age were, however, less than those for children
over 6 months.
Many of the children examined seemed to be poorly nourished.
From study of the final tabulations it is apparent that the point of
view of the physicians became warped as they proceeded with the
examinations and that their standards for the various grades of
nutritional condition were unintentionally lowered by the prevalence
of poorly nourished children. Grades for subcutaneous fat are given
in the report in order to show the variation within the Puerto Rican
group, but these estimates should not be compared grade for grade
with estimates for children in continental United States, since the
grade “ good” given to the group in Puerto Rico is probably com
parable with the grade “ fair” as used in continental United States,
and so on.
Special clinical and roentgenographic examinations were made in
order to determine whether rickets was as infrequent in this region as
would be expected on account of the sunny climate. At clinical
examination 50 children (about 9 per cent of those examined) were
thought to show (usually in slight degree) the physical signs that are
commonly considered evidence of rickets. At roentgenographic
examination, only 5 (less than 1 per cent) showed evidence of rickets—
3 in a slight or very slight degree and 1 in an advanced degree; 1
showed a healed process of many months’ standing. Study of the
roentgenographic incidence of rickets and of the relation of the roent
genographic incidence to the clinical has led to the belief that many
of the clinical diagnoses of rickets made during the present study were
wrong and that the physical signs upon which they were based were
in all probability within the limits of variation of normal growth and
development.
The slight deviations from the preconceived normal which had been
considered signs of slight rickets in a previous study made in New
Haven by the United States Children’s Bureau in cooperation with
the department of pediatrics, Yale University School of Medicine,
were almost totally absent (3 children with such deviations out of
584 children examined). It is therefore concluded that the deviations
98

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

SU M M ARY

found in the New Haven study were properly considered signs of
rickets and that the standards of normal used in that study were
not too limited.
A report is given of 59 cases of marked osteoporosis occurring in
children who, with the exception of 2, had no rickets, and it is sug
gested that the osteoporosis was due to gross inadequacies in the diet,
especially inadequacy of calcium.
The incidence of transverse lines appearing in the roentgenograms
of the bones of the forearm was studied. These lines are thought to
be evidence of intermittent growth and, as was expected, were found
more frequently in these Puerto Rican children than in children of
the same age in New Haven.
The socio-economic conditions under which the families were living
at the time of the investigation were studied— the size of the families,
housing facilities, income, and diet. The diets of the mothers and
children were considered of especial interest because of the relation
of the content of these diets to the growth of the children and to the
roentgenographic appearance of their bones. The low calcium con
tent of the mothers’ diets was most striking and was due primarily to
the fact that almost negligible amounts of milk were taken by a
majority of the women. The calcium content of the children’s diets
could not be calculated, but it is probable that the inadequacy of
calcium in the mothers’ diets is characteristic of the diets of the
children also. The quality of the children’s diets deteriorated rapidly
after the breast-feeding period was over. Unsatisfactory amounts of
cow’s milk were given to at least half the older children. The solid
food that was given was added to the diet later than is usual in con
tinental United States and was insufficient in amount and variety;
the use of green vegetables and of eggs was relatively rare. The
basis of the average child’s diet after weaning was rice, beans, and
coffee, with a little milk and, occasionally, green vegetables. The
diets of both mothers and children were grossly inadequate, especially
as regards milk and green vegetables.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Appendix A.— HEIGHT, WEIGHT, AND HEAD CIRCUMFERENCE FOR AGE: CHILDREN EXAMINED IN
PUERTO RICO

100

A l .— Height and age; hoys 1 to 88 months of age examined in Puerto Rico__________________________ _

DEVELOPM ENT

able

B oys 1 to 33 months of age

05
d
c3
d

T5
A g e 1 at examination
Total

O
a
M
Pc3
O

T ota l...................... .......
Age reported_______________

307
295

304
292

14
7
7
12

C
O
tO

»o
to

2
2

11
11
2
4

10
10
1
5
3
i

d
03
rd
co

d
c8
d
4-»

aT
to

7
T

15
14

1
1
2
2

2
2
1
3
2
1
i

1
i

4
8
3
3
1
1

d
03
dl
CO

d
c3
d
CO

d
c3
d

d
03
d
CO

05
CO

CO
!>»

d-t
t>-

28
28

23
23

CO
CO

*o
CO

CO
26
25

d
cS
dl

d
03
d|

2
4
5
4
1
2
21
1
1

3
1
3
4
3
2
1

8
28
26

1
1
l 1
2
8
5
5
1
1

27
26

15
14

1
2
4
1
2
4
2

1
3
1
1

1
3
4
2
2
1
2
1

1
2
5
3
4

2
2

1
3

d
03
d

2
1
1
1
2

11
11

1
1
2

2
1
2
1
4

1
2

14
14

7
7

to
00

oT
00

4
4

8
©

o
d

C
O
05

5
3

2 __ 3
2
3
i

3
3

1
1

1
1
1
1
2
1

1
1
1
2
1

1
1
1
1

1
1
1
2

1
1
1

Age is given as of nearest month; that is, “ 1 m onth” is from 16 days to 1 m onth and 15 days, inclusive, and so on.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

8
09
CO

8
CD

05
d
c3
dl

05
00
d
CÖ
dl
-4-a

1
2
1
2
4
4
7
1
3
1

d
C5
dl

D'
GO
d
CÖ
dl
4-»

R IC O

d
c3
d

CO
00
d
03
5
co

PUERTO

11
16
19
18
16
13
17
13
15
17
12
9
8
8
11
11
6
10

d
03
d

05
CO
d
03
d

to
CO
d
CS
d
4-S

P'.
CO
d
a

CO
CO
d
3
d

1
1

O F C H IL D R E N

8
11
12
6
10
7

Age not r e p o r te d ----------------

jS
oT

1
11
16
19
18
16
13
17
13
15
17
12

d
0
o
d

t''»
to
d
C3
rC

OF BO N ES

CO
05
d
C3
5
8

Height (centimeters) reported

A2.— Height and age; girls 1 to 33 months of age examined in Puerto Rico

able

Height not reported

Girls 1 to 33 months of age

„ 1

2
4
3
2
2

2
4
3
5
2
1

1
1
1

67, less than 69

1
4
6
3
3
1
1
Ï

2
3
5
2
1

2
1
3

1
2
ï
3
1
1
1
1
2

4
2
3
2
2
1

1
2
2
1
3
2
1
1
1
1

1
1
2
3
5
4
1
1

2
2
2
i
1
2
1
1
1
1

1
3
4
2
2
1
3
1
1
1

3
3
4 v 3
S

1
1
1
2
6
1
1
1

■■

1
1
1
1

1
3

1 Age is given as of nearest m onth; that is, “ 1 m onth ” is from 16 days to 1 month and 15 days, inclusive, and so on.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

91, less than 93

89, less than 91

87, less than 89

85, less than 87

9
1
3
2

83, less than 85

79, less than 81

13
4
6

81, less than 83

77, less than 79

1
1

75, less than 77

2
1
1

73, less than 75

63, less than 65

1
1

71, less than 73

61, less than 63

69, less than 71

59, less than 61

65, less than 67

57, less than 59

55, less than 57

53, less than 55

51, less than 53

49, less than 51

47, less than 49

45, less than 47

272
265
8
16
11
12
18
15
12
13
7
11
10
9
13
7
13
5
5
7
5
4
9
5
9
7
24
5
5

1
2

1
1

1
. . . . .

APPENDIX

274
266
8
17
11
12
18
15
12
13
7
11
10
9
13
7
13
5
5
7
5
4
9
5
9
7
24
5
5
8

Total
Total___ _____________
Age reported.............................
1 m o n th .. ............. .........
2 months____________
3 m on th s.. ____________
4 m onths......................
5 m onths_______________
6 m onths______________
7 m onths___________
8 m onths_________ ____
9 m onths_____ _____ _
10 months...... ..........
..
11 m onths______________
12 months____________
13 months____ ____ ____
14 m on th s.____ _______
15 m onths____________
16 m onths______________
17 months___________
18 m onths___ __________
19 m onths______________
20 months______________
21 m onths_______ ____ _
22 m onths..........................
23 m onths________ ____ _
24 m onths______ ____ _
25 to 27 months. ...............
28 to 30 months_________
31 to 33 m o n t h s _______
Age not reported____________

Total reported

Height (centimeters) reported
A g e 1 at examination

102

DEVELOPM ENT

A3.— Average height at each month of age; boys and girls 1 to S3 months of
age examined in Puerto Rico

able

OF BONES

C H IL D R E N

PUERTO

R IC O

Average height (centimeters)
A g e 1 at examination

Boys
Observed

2 months___________ ___________ __________

18 m onths________ ____
___ __ ____________
19 m onths_____________________________________

1
26 months_______________ ____________________ |
1
|
32 months_______________________ ___________

1
1

Smoothed

53.3
56.2
58.6
62.8
62.9
64.0
64.5
66.6
67.8
68.7
69.8
72.0
69.8
75.3
75.5
75.5
74.9
76.0
77.0
74.9
75.5
78.4
80.6
84.0
80.4
85.1
86.9

Girls

52. 98
56.83
59. 31
61. 23
62.84
64.26
65.54
66.73
67.84
68.89
69.89
70.85
71.78
72.69
73.56
74.42
75.25
76.08
76.88
77.67
78.46
79. 22
79.98
80.74
81.48
82.21
82.94
83.66
84.38
85.09
85.79
86.49
87.19

Observed
52.8
55.4
58.9
61. 2
60.7
61. 5
64.0
64.5
68.0
67.5
68.8
69.6
72.0
73.4
73.1
75.2
76.0
73.7
79.0
77.6
75.6
78.4
78.3
1
f

f
81.1 \

1
1

84.0 -!

1
f

i
84.0 l
1

Smoothed
51.94
55.91
58. 44
60. 38
62.00
63. 41
64.69
65.86
66.96
67.99
68.98
69.92
70.82
71.70
72.55
73.38
74.19
74.99
75.76
76.53
77.28
78.02
78.75
79.47
80.18
80.88
81.58
82.27
82.95
83.63
84.30
84.96
85.63

Equations of curves used in smoothing:
Boys, g = 52.429051+0.548181X+10.976648 log. X.
Girls, y =51.434797+0.5078652:+11.478762 log. X.
x representing age in months, y representing height in centimeters.
1 Age is given as of nearest month; that is, “ 1 m on th ” is from 16 days to 1 month and 15 days, inclusive.
N ote that from 26 months onward each observed average includes a 3-month interval.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

103

APPENDIX A
T

able

A4.— Weight and age; boys 1 to S3 months of age examined in Puerto Rico
Boys 1 to 33 months of age
Weight (kilograms) reported 2

Weight
reportée

Age 1 at examination
Total
Total
re
ported

T o t a l..----------------------------

307

305

Age reported______________ ____

295

293

9
11
16
19
18
16
13
17
13
15
17
12
9
8
8
11
12
6
10
7
4
5
7
4
14
7
7

9
11
16
19
18
16
13
16
13
15
17
12
9
8
8
11
12
6
10
7
4
5
7
4
14
‘ 7
6

5
2

2
7
7
3
1

4
1
1
3
1
5
6
9
4
3
2
1
1
1
1

1
1
2
1
3

2 m onths__ _______________
3 months____ _____________
4 months__________________
5 m o n t h s ................ ..............
6 m onths________________ .
7 months__________________
9 m on th s.. ___________ 11 m onths_______ . . . _____

1
1
1

1
2
2
1
1
1
1

1
1
1

2
4
5
4
3
4
2
1
1
1

4
5
9
7
4
6
4
5
2
3
1
1

1
1

2
3
5
1
4
1
3
6
3
4
2

1
1
1
1

2
1

1
1
1
2
3
5
1
3
2
3
2
3
2
1

2
1

1
1
1
3
3
1
1
2
2
2
2
1
3

1
4
3

1
1
3

1 Age is given as of nearest month; that is, “ 1 m onth” is from 16 days to 1 month and 15 days, inclusive,
and so on.
2 Weight is to nearest kilogram; that is, “ 3 kilograms ” is from 2.50 to 3.49 kilograms, inclusive, and so on.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

104

DEVELOPM ENT

A 5 .—

able

OF BON ES

O F C H IL D R E N

PUERTO

R IC O

Weight and age; girls 1 to 38 months of age examined in Puerto Rico
Girls 1 to 33 months of age
W eight (kilograms) reported a

o
Ö'O

A g e 1 at examination
Total

T otal_______________
Age reported..........................

Age not reported----------------

Total
reported

fd o
bßQ,
14 &
®

274

272

266

264

8
17
11
12
18
15
12
13
7
11
10
9
13
7
13
5
5
7
5
4
9
5
9
7
24
5
5

8
17
11
12
18
15
12
13
7
ii
9
9
13
7
13
5
5
7
5
4
9
5
9
6
24
5
6

7
8

1
4
4
3
2
1

1
1

2
4

1
1

4
7
5
10
2
2
4
2
2
1

3
4
6
6
3
1
4
4
4
1
2
1

1
2
2
2
5
3
2
6
4

1
1

1
2
2
1
1

1
1
1

1
2
1
3
2
4
2
1
2
1
3
2
2
1
2
7

1
1

1
1
1
3
6
4
2
4
1
1
4
1
3
3
1
1

1
2
1
1
1
1
1
2
5
3
2

4
1

1
1

1 Age is given as of nearest month; that is, “ 1 m onth” is from 16 days to 1 month and 15 days, inclusive,
id so on.
2 Weight is to nearest kilogram; that is, “ 3 kilograms” is from 2.50 to 3.49 kilograms, inclusive, and so on.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

A P P E N D IX

105

T able A 6.— Average weight at each month of age; hoys and girls 1 to 33 months of
age examined in Puerto Rico

Average weight (kilograms)
Boys

Age 1 at examination

Smoothed

Observed
1 m onth_________________ ____________________ _
2 m onths________________________ _____________
3 m onths______________________________________
4 months______________________________________
5 months______________________________________
6 months______________________________________
7 months______________________________________
8 months_________________________________ ____
9 m onths______________________________________
10 m onths___ ___ __ - ________ ______________
11 months_____________________________________
12 months_______________ __________ _______
13 months_____________________________________
14 months___ _ ______________________________
15 m onths_________ __________ ________________
16 months______________________ _____________
17 months_____________________________________
18 months__________ ____ _____________ ________
19 months_____________________________________
20 months____________________________ ____ ___
21 months_____________ ______________________
22 months_____________________________________
23 m onths_____________________________________
24 months_____________________________________
25 months______________ __________________ ___ )
26 months_____________________________________
27 months___________________ _________________ f
28 months___________________________________ _ 1
29 months_________ ______ ________________ ___ }•
30 months_____ _______________________________ 1
31 months_____________________________________ ]
32 months_____________________________________
33 months________________ ____ ____ ____ ______ I

}

4.0
5.0
5.6
6.6
6.5
6.5
6.8
7.3
7.2
7.8
7.8
8.7
7.9
9.5
9.6
9.5
9.3
9.3
9.5
9.7
9.3
10.2
10.3
10.0
(
10.9 ■1
11.7

Girls

11.8 f
l

4.08
5.01
5.69
6.06
6.44
6.77
7.07
7.34
7.59
7.83
8.06
8.28
8.49
8.69
8.89
9.08
9.27
9.46
9.64
9.81
9.99
10.16
10.33
10.49
10.66
10.82
10.98
11.14
11.30
11.45
11. 61
11.76
11.91

Observed

Smoothed

4.1
4.6
5.6
6.2
5.9
6.1
6.8
6.7
7.4
7.5
7.4
8.0
8.2
9.4
9.0
9.8
9.8
9.1
8.8
9.8
9.3
8.6
10.1
9.0
i

1
>
1

10.4

}

11.0

[

f
10.8 •J
l

)
1

3.79
4.81
5.44
5.91
6.30
6.62
6.91
7.18
7.42
7.64
7.85
8.05
8.24
8.42
8.59
8.76
8.92
9.08
9.24
9.39
9.53
9.68
9.82
9.96
10.09
10.23
10.36
10.49
10.62
10.75
10.87
10.99
11.12

Equations of curves used in smoothing:
Boys, ¡/=3.966361+0.11649&T+2.701817 log. x.
Girls, y =3.713483+0.080924a:+3.116864 log. x.
x representing age in months, y representing weight in kilograms.
1 Age is given as of nearest month; that is, “ 1 m onth” is from 16 days to 1 month 15 days, inclusive.
N ote that from 26 months onward each observed average includes a 3-month interval.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

106

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

A7.— Weight for height of boys 1 to S3 months of age examined in Puerto
Rico

able

not
id
1

Boys 1 to 33 months of age
Weight (kilograms) reported

Height (centimeters)
Total

Total
reported

£ R
13

14
£

Total....................................

307

305

Height reported................. ..........

304

302

47, under 49________________
49, under 51________________
51, under 53________________
53, under 55________________
55, under 57________________
57, under 59________________
59, under 61________________
61, under 63____
______
63, under 65________________
65, under 67________________
67, under 69____________ . . .
69, under 71_________ _____
71, under 73________________
73, under 75________________
75, under 77_______________
77, under 79_______________
79, under 81________________
81, under 83_______________
83, under 85________________
85, under 87________________
87, under 89_________ ____
89, under 91_______ _______
91, under 93....... .....................
93, under 95________________

2
11
10
7
15
26
24
20
28
28
27,
21
23
15
14
11
7
4
5
3

2
11
10
7
15
26
24
19
28
28
27
21
23
15
14
11
7
4
5
3

6
13
5
4
2

3
7
14
11
10
6
1
1
1
1

Height not reported___________

able

1
1
1

2
4
5
3
2

6
9
2
2
4
1

2
1
4
h

12
14
2
2
1

1
1
5
7
7
11
8
3
2
1

3
4
4
10
8
2
1
1

1
3
]
2
8
6
4

1
1

1
2
2
1
2
1
1

1
1
2
1

1
1
1

1
1

A8.— Weight for height; girls 1 to S3 months of age examined in Puerto Rico

Weight (kilograms) reported
Height (centimeters)
Total

T ota l________

Total
re
ported

Weight not
reported

Girls 1 to 33 months of age

_____

274

272

Height reported______ ____

272

270

1
1
2
1
13
9
13
17
20
14
26
13
18
19
25
23
17
10
14
3
9
2
1
1

1
1
2
1
13
9
13
17
20
14
26
13
18
19
24
23
17
10
14
3
8
2
1
1

1
6
6
9
4
5
2
3

1
1
2
11
8
7
6

45, under 47.....................
47, under 4 9 ...................
49, under 51___ _______
51, under 53___________
53, under 55...... .......... .
55, under 57___________
57, under 59___________
59, under 61____ ______
61, under 63___________
63, under 65___________
65, under 67___________
67, under 69........ ............
69, under 71.. ________
71, under 73_____ ______
73, under 75................ .
75, under 77___________
77, under 79_____ .•.........
79, under 81....................
81, under 83...... .......... .
83, under 85___________
85, under 87___________
87, under 8 9 ................ .
89, under 91___________
91, under 9 3 . ...............
Height not reported....... .

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1
1

1
2
1
10
3
2
1

2
3
6
2
3
1

2
5
13
9
7
2
1
1
1

2
7
5
17
5
5
1
2
1

1
3
7
12
7
7
2
1

1
1
1
3
7
2
3

2
1
2
1

1
1

2
1

1
1
1

APPENDIX A
T

able

107

A9.— Average weight for height; Puerto Rican boys and girls
Average weight (kilograms)
Boys

Average height
(centimeters)

3.0

2.0

2.89
3.36
3.83
4.30
4.77
5.24
5. 71
6.18
6.65
7.12
7.59
8.06

4.0
4.5
4.8
4.3
5.5
6.3

6.6

7.0
7.7
8.3

68.
70-

Girls

Boys

Average height
(centimeters)

46485052.
54.
56.
58.
6062.
64.

66_

Avreage weight (kilograms)

2.65
3.10
3.54
3.99
4.44
4.89
5.33
5.78
6.23

4.0
4.0
4.0
4.1
4.7
5.2

6.0
6.1

6.4
7.2
7.5
7.8

8.53
9.00
9.47
9.94
10.41
10.88
11.35
11.82
12.29
12.76
13.23
13.71

8.6

9.3
9.5
9.6
10.7
11.1
11.3
11.8

12.6

6.68

13.0

7.12
7.57

13.0

8.02

Girls

8.47
8.92
9.36
9.81
10.26
10.71
11.15
11.60
12.05
12. 50
12.94

8.9
9.4
9.4
10.3
10.9
11.3
11.6
12.5

11.0
14.0

Equations of curves used in smoothing:
Boys, y=2.415280+0.235207z.
Girls, g=2.199633+0.223848j-.
x representing centimeters of height above 46; v representing weight in kilograms.
T

able

A10.— Head circumference and age; boys 1 to S3 months of age examined in
Puerto Rico
Boys 1 to 33 months of age

A g e 1at examination

Head circumference (centim eters)2

To
tal
34
Total_______

307

295

1 m onth____
9
11
2 m o n th s ...
3 m onths___
16
4 m o n th s ...
19
5 m o n th s ...
18
6 m on th s.. .
16
7 months___
13
8 m on th s.. .
17
9 m onths____
13
10 m onths__
15
11 months___
17
12 months__
12
13 m o n th s ...
9
14 m onths..
8
15 m on th s.. .
8
16 m onths__
11
17 m onths........
12
18 m onths. _.
6
19 m on th s..
10
20 m o n th s ...
7
21 m onths__
4
22 m o n th s ...
5
23 m on th s..
7
24 m o n th s ...
4
25 to 27 months. 14
28 to 30 months. 7
31 to 33 months. 7

Age reported__

—

1
1

1
1
1
1

and so on.

1™ nm

X T

160326°—33----- 8

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

1
3
2
1
4
2
1
1
1
1
1
1
1

Age not rep orted -.. 12

tta ftS

88 ° ' “earest

1
1
2
3
2
7
2
2
4
2
5
2
T
2
I
1
2
2
1

1
1
1
i
2
3
3
5

2
4
1
2
1
2

6
1
1
3
3
2
4
2
1

1
1

3
1
1

3
?
4

....

—

is irom io days to 1 month and 15 days, inclusive,
«*“

centimeters” is from 33.50 to 34.49 een-

108

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

A l l .— Head circumference and age; girls 1 to S3 months of age examined
in Puerto Rico

able

Girls 1 to 33 months of age
Head circumference (centim eters)2

A g e 1 at examination
Total
34

Total_______________

274

Age reported.............. ..........

266

8
17
11
12
18
15
12
Î3
7
11
10
9
13
n
13
5
5
7
5
4
9
5
9
7
24
5
5

1
1

1
5

4
4
2

3
2
2
2
1

1
2
2
3
3

3
4
3
3
3
2

1
3
5

1
3

]
1

2
i
i

3
5
1
1
1

1
1
1
2
3
1
4
3
3
4

1
1

1
1
2
5
4
3
4
5
5
2
1

1
1
1
2
1
1

1
2
1

1
1
1
2
2
4
4
1
2
3
1
2
2
1
6
1

1
1

1
2
2

1
3
2
2
3
3
2
2
1
2
3
1
2

1
3
Î
4
5
2

1
1

1 Age is given as of nearest month; that is, “ 1 m onth” is from 16 days to 1 month and 15 days, inclusive,
and so on.
2 Head circumference is given as of nearest centimeter; that is, “ 34 centimeters” is from 33.50 to 34.49
centimeters, inclusive, and so on.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

109

APPENDIX A
T

able

A12.— Average head circumference at each month of age; boys and girls
1 to S3 months of age examined in Puerto Rico
Average head circumference (centimeters)
A g e 1 at examination

Boys
Observed

1 m onth—
2 months—
3 m onths..
4 m onths..
5 months. .
6 m onths..
7 months. .
8 m onths..
9 m onths..
10 months .
11 months.
12 months.
13 months.
14 months.
15 months.
16 months.
17 months.
18 months.
19 months.
20 months.
21 months.
22 months.
23 months.
24 months.
25 months.
26 months.
27 months.
28 months.
29 months.
30 months.
31 months.
32 months.
33 months.

36.8
38.5
39.7
40.7
41.5
42.0
42.7
43.3
43.9
44.0
43.7
45.5
44.6
46.4
46.3
46.7
46.5
46.2
46.5
46.7
46.0
46.8
47.0
46.3
f
47.6 {
f
47.4 {
1
[
48.7 \
l

Girls
Smoothed
36.25
38.62
40.00
40.99
41.76
42.38
42.91
43.38
43.78
44.15
44.48
44.78
45.06
45.32
45.56
45.79
46.00
46.20
46.39
46.58
46.75
46.91
47.07
47.22
47.37
47.51
47.64
47.77
47.90
48.02
48.14
48.26
48.37

Observed
36.0
37.4
39.0
40.3
40.7
40.7
41.6
42.2
43.6
43.7
43.6
43.9
44.2
45.4
44.6
45.0
45.2
45.1
45.2
46.3
45.7
45.2
46.8
45.9
1
>
1
)
}
1

]
}
1

(
46.8 <
l
f
46.8 \
f
46.2 1
l

Smoothed
35.33
37.77
39.19
40.19
40.96
41.59
42.12
42.57
42.97
43.33
43.64
43.93
44.20
44.44
44.67
44.88
45.07
45.26
45.43
45.59
45.75
45.89
46.03
46.16
46.29
46.41
46.52
46.63
46.73
46.84
46.93
47.03
47.11

Equations of curves used in smoothing:
Boys, y=36.241792+0.006690x+7.839603 log. x.
Girls, y=35.346584—0.019527Z+8.173817 log. x.
x representing age in months; y representing head circumference in centimeters.
1 Age is given as of nearest month; that is, “ 1 m onth” is from 16 days to 1 month and 15 days, inclusive.
N ote that from 26 months onward each observed average includes a 3-month interval.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Appendix B.— SKELETAL SIGNS OF RICKETS
B l.— Clinical diagnosis of rickets in relation to number of skeletal signs
usually considered evidence of rickets; children examined in Puerto Rico

T able

Children, examined
Clinical diagnosis of rickets
Total

N um ber of skeletal
signs

Questionable
diagnosis

N o rickets

Number

Per cent
Per cent
distribu Number distribu Num ber
tion
tion

Rickets

Per cent
distribu Number
tion

T otal_________

584

100

400

100

134

100

N one____________. . .
1____ _______________
2_____ ______________
3____ _______________
4____ _______________
6____________________

238
208
97
30
9
2

41
36
17
5
2

236
136
24
4

59
34
6
1

2
69
51
11
1

1
51
38
8
1

1 Less than 1 per cent.

110

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Per cent
distribu
tion

100

3
22
15
g
2

6
44
30
16
4

111

APPENDIX B

B2.— Clinical diagnosis of rickets in relation to combinations of skeletal signs
usually considered evidence of rickets; children examined in Puerto Rico

T able

Children examined
Clinical diagnosis of rickets
Skeletal signs
T

QuesNo
tioudble
rickets diag
nosis
Total

584

400

134

N o n e .______ _______ ______ ____________ ____

238

236

One skeletal sign only________________ ______

208

136

25
15
87
37
9

2
4
72
32
2

14
3
18

7
1
16

7
2
2

24
3
2

Costochondral junctions.................. ............
Epiphyses_______________________________
Bowlegs___ _____________________________
Knock-knees__________________ _________
Harrison’s groove
_____ _ __________
Parietal or frontal bosses moderately en
larged_________________________________
Craniotabes____________ ______ __________
Asym m etry of head.......................................
T w o skeletal sig n s.._________ ___ ___________
Costochondral junctions and epiphyses___
Costochondral junctions and bowlegs____
Costochondral junctions and knock-knees.
Costochondral junctions and Harrison's
groove______ ______ _________ _________
Costochondral junctions and parietal or
frontal bosses moderately enlarged______
Costochondral junctions and craniotabes..
Costochondral junctions and asymmetry
of h ea d .. ______ _______ . . ___________
Epiphyses and bowlegs_________ ________
Epiphyses and knock-knees................... .
Epiphyses and Harrison’s g r o o v e . . . ____
Epiphyses and asymmetry of head. ____
Bowlegs and knock-knees_______________
Bowlegs and Harrison’s groove_______
Bowlegs and parietal or frontal bosses
moderately enlarged___________________
Knock-knees and Harrison’s groove__
Knock-knees and parietal or frontal bosses
moderately enlarged___________ ______
Harrison’s groove and pigeon-breast.. . .
Asym m etry of head and parietal or frontal
bosses moderately enlarged__________
Parietal and frontal bosses moderately
enlarged___________________________
Three skeletal signs_________ _______ ______
Costochondral junctions, epiphyses, and
parietal or frontal bosses moderately en
larged_____________________________
Costochondral junctions, epiphyses, and
bowlegs- ____ ______________________
Costochondral junctions, epiphyses, and
knock-knees...... .............. .............. ...
Costochondral junctions, bowlegs, and
asymmetry of head________ _________
Costochondral junctions, bowlegs, and
parietal or frontal bosses moderately
enlarged____ _____________________
Costochondral junctions, bowlegs, and
Harrison’s groove.._____ ________ ______
Costochondral junctions, bowlegs, and
knock-knees______ ____________________

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

5
17
7
1
3
1
2
17
1
1
1
13
5

23
11
12
5
7

1
9
5

4
5

1
1

1
13

1
3

3
3

1
1

5
1

1
4

1
6
2
1

1
1

3
3

1
1
10

8
1

M oder M ark
ate
ed

1
1

Slight

9
5

Rickets

112
T

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

B2.— Clinical diagnosis of rickets in relation to combinations of skeletal signs
usually considered evidence of rickets; children examined in Puerto Rico.— Con.

able

Children examined
Clinical diagnosis of rickets
Skeletal signs
Total

Three skeletal signs—Continued
Costochondral junctions, parietal or fron
tal bosses moderately enlarged, and
asymmetry of head............... ...................
Costochondral junctions,
Harrison’s
groove, and asymmetry of head............
Costochondral
junctions,
Harrison’s
groove, and knock-knees_______________
Epiphyses, bowlegs, and asymmetry of
head__________________________________
Epiphyses, bowlegs, and knock-knees___
Bowlegs, knock-knees, and parietal or
frontal bosses moderately enlarged_____
Bowlegs, knock-knees, and Harrison’s
groove_________________________________
Knock-knees, parietal or frontal bosses
moderately enlarged, and asymmetry
of head________________________ ______ _
Asym m etry of head and parietal and fron
tal bosses moderately enlarged_________
Four skeletal signs__________________________
Costochondral junctions, epiphyses, bow
legs, and knock-knees_________________
Costochondral junctions, epiphyses, bow
legs, and parietal or frontal bosses m od
erately enlarged....................................... .
Costochondral junctions, epiphyses, bow
legs, and Harrison’s groove_____ ______
Costochondral junctions,
epiphyses,
knock-knees, and parietal or frontal
bosses moderately enlarged____________
Costochondral junctions,
epiphyses,
knock-knees, and Harrison’s groove----Costochondral junctions, bowlegs, Harri
son’s groove, and parietal or frontal
bosses moderately enlarged.....................
Epiphyses, bowlegs, Harrison’s groove,
and knock-knees______________________
Epiphyses, bowlegs, Harrison’s groove,
and asymmetry of h e a d ...____________
Six skeletal signs...................................... - ..........
Parietal and frontal bosses moderately
enlarged, costochondral junctions, epi
physes, Harrison’s groove, and craniotab es......................... - ................... - .........Parietal and frontal bosses moderately
enlarged, costochondral junctions, Har
rison’s groove, knock-knees, and asym
metry of head_________________________

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Ques
No
tionable
rickets diag
nosis
Total

Rickets

Slight

M oder M ark
ate
ed

Appendix C.— TECHNIQUE OF ROENTGEN-RAY
EXAMINATION

The technique of the roentgen-ray examination of the bones of the
forearm followed in every detail that used in New Haven. The roent
genograms were taken on 8 by 10 inch superspeed films, placed in flat
film holders without screens. It had been found that the detail of the
bone structure was partly lost if screens were used, and though the
length of exposure was necessarily longer without screens, the resulting
roentgenogram was very much more satisfactory. In order to keep
the time of exposure as short as possible, filters were not used. A
fine-focus 30-milliampere radiator tube was used with a target-film
distance of 20 inches. The target-film distance was kept unchanged
throughout the examinations. A current of 40 milliampères with a
spark gap of 3 % inches was used. The time of exposure varied from
one-fourth to three-fourths of a second, depending on the thickness
of a child’s wrist, the smallest wrists requiring a bare one-fourth
second and the largest, those of well-nourished children of 18 months
or over, three-fourths of a second. Experience in New Haven had
shown that it was of utmost importance that the child’s arm should be
held absolutely still during the exposure. If any movement took
place, even the very slightest, the detail of the bone structure was
blurred and the interpretation of the film made more difficult or im
possible. To avoid this as far as possible, one of the physicians held
each child for the roentgen-ray examination. The hand was placed
palm up with the arm extending at right angles to the body and in
extreme external rotation, so that the two bones of the forearm would
lie parallel to each other and not crossed. As often as possible the
roentgenograms were developed before the child left the clinic, and,
if they were not satisfactory, further films were taken. In this way
roentgenograms were obtained which were on the whole satisfactory
for diagnosis.
Further experience has shown, however, that the detail of the bone
structure will be clearer and the chance lessened that movement will
spoil the roentgenograms if the time of exposure is shortened to onetenth of a second, if a current of 100 milliampères and a kilovolt
peak of 75 are used, and if the target-film distance is lengthened to 30
inches. If the exposure is accurately timed to one-tenth of a second
and the kilovolt peak is 75 or less, 100 milliampères may be passed
through a 30-milliampere radiator tube without damage to the tube.
113

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Appendix D.— GRADING OF CHILDREN’S DIETS

Appendix Tables D - l to D -6 show the grades given to the children’s
diets for each of the four 3-month periods of the first year and for the
whole second and third years and show also the different combinations
of milk and solid foods that fall within the range of each grade. The
number of children whose diets fell in each grade is shown and in
addition the number of children in each grade for whom the period
was complete and the number for whom it was incomplete. (See
p. 89.)
It should be remembered in studying these tables that together
they represent diet histories for 554 children (2 diet histories were
not reported). The diet histories of 293 of these children were reported
for all four 3-month periods of the first year; of 367, for three periods;
of 469, for two periods; and for 554, for 1 period. More than half
the children appear, therefore, in all four tables representing the first
year.
It will be seen in each table that the number of children that had
not yet completed the diet period (those whose ages at the time of
the interview still fell within the age range of the period) is compara
tively small and that the differences in the distribution of diet grades
of the two groups are insufficient to be considered significant. The
total number of diets in each grade, therefore, has been used in the
comparisons, regardless of whether they are for complete or incomplete
periods.
T

able

D l .— Diet grade and ty-pe of feeding of children during first 3-month period
(birth to 3 months); children in families visited in Puerto Rico
T yp e of feeding

Children in families visited
Age at date of home visit
Total

C ow ’s
milk
(average
amount
daily)

D iet grade
Breast milk

Under 4
months

4 months and
over

N ot
re
Per
Per
Per
ported
N um cent N um cent N um cent
ber
distri ber
distri ber distri
bution
bution
bution

T otal___________

556

Grade reported_______

554

100

385

140

1 cup___
1 p in t ...

110
30

20
5

1 cup___

V ery good________

N one___

G ood_____________
M ainly ________
Small amount___
Fair_________ ____

None ...................

P o o r .-.:__________
Grade not reported___

___________
1

114

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

473

100

472

100

322

12
2

17
3

96
28

20
6

9
4

1
1

115

A P P E N D IX D

T a b l e D 2 .— Diet grade and type of feeding o f children during second 3-month
period (fourth to sixth m onth); children 4 months o f age and over in fam ilies
visited in Puerto Rico

Children 4 months of age and over
in families visited
T yp e of feeding
Age at date of home visit
Total

Diet grade

4 to 6 months 7 months and
over
Some
form
of
C ow ’s milk
Per
Per
solid N um Per
(average
cent N um cent N um cent
food1
amount
ber
distri ber
distri ber distri
daily)
bution
bution
bution

M ilk feeding

Breast milk

Total____

373

Grade reported.

100

371

Very g o o d ..
Entirely.
M ainly. .

None.
1 cup.

Entirely_____
M ainly______
Small amount.
None_________

None_____
1 cu p ........
1 pint____
More than
1 pint.

Small amount.
None......... .......

Good.

Fair.

Poor...........

212
61

21
6

175
38
15
4

None.

1 pint____
M ore than
1 pint,
lp in t ........

N one.

1 pint.

None.
None.

V ery poor.
cup...
do.

Grade not reported.
1 “ + ” indicates that some solid food was given; “ 0 ” that no solid food was given.
food was not considered.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

The kind of solid

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

116
T

D3.— Diet grade and type of feeding of children during third 8-month period
{seventh to ninth month); children 7 months of age and over in families visited in
Puerto Rico

able

Children 7 months of age and oyer in
families visited

Type of feeding
M ilk feeding

Age at date of home visit
Total

D iet grade

Breast milk

Some
10 months
7 to 9 months
form
and over
of
C ow ’s milk
solid
(average
food 1
amount
Per
Per
Per
daily)
Num- cent Num - cent Num- cent
ber
distriber
distri- ber distribution
bution
bution
373

pint____
M ore than
1 pint.

100

294

100

+
+

22
27

6
7

1
4

1
5

21
23

7
8

+
+

30
28

8
8

6
7

8
10

24
21

8
7

Fair

Small amount-. 1 pin t___
M ore than
1 pint.

0
0
0
0

214

168

105
30
23
30

29
8
6
8

21
6
6
9

29
8
8
12

84
24
17
21

29
8
6
7

11
14

3
4

1
3

1
4

10
11

3
4

Very poor________
+
0
Grade not reported___

299

100

G ood........................
Small amount—

367

1 “ + ” indicates that some form of solid food was given; “ 0 ” that no solid food was given.
solid food was not considered.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

5
The kind of

APPENDIX D
T

117

D 4 . — D iet grade a n d t y p e o f fe e d in g o f children d u rin g fo u r th 8 -m o n th period
(tenth to twelfth m onth) ; children 1 0 m on th s o f age and over i n fa m ilie s visited in
P u e rto R ico

able

Children 10 months of age and
over in families visited

T yp e of feeding

Age at date of home
visit
Milk feeding

Total____

Per cent dis
tribution

13 months
and over

Number

Per cent dis
tribution

Cow’s milk
(average
amount
daily)

Num ber

Breast milk

Per cent dis
tribution

Solid food1

N um ber

10to 12
months

Diet grade

66
233
100 66 100 227 100
12 4 2 3 10 4
2 1 2 6 3
7
2 1 2 4
5
2
2 1
2
1
1 (2)
1 (2)
1 (2)
1 (2)
2 1 2 6 3
7
1 1 2 2
3
1
2 1
2 1
2 1
2 1
200 68 46 70 154 68
29
10 9 14 20 9
299

Grade reported.

293

V ery g o o d ..

Small amount.

M ore than 1 Satisfactory.
pint.
___ do_____
1pint______

Small amount.
Mainly_______

1pint.
1cup..

Borderline...
Satisfactory.

None___
Mainly..
Entirely.

1p int.
1cu p..
N one.

Borderline.
-----do........
-----do____

N one.

M ore than 1
pint.
. . d o _______
1pint______
___ d o ______
1cu p______
----- d o ______
N one______
N one______

None_________
Unsatisfactory.
None_________
Unsatisfactory.
None_________
Unsatisfactory.
None_________

N one_________

Good.

Fair.

Poor.

None_________
Small amount.

Do__...........

Mainly_______
D o.........

Entirely______
D o..............

Unsatisfactory__

V ery poor.
N one.
NoneNone .
N one.

1p in t..
----- d o .
1c u p ...

____ d o.

Grade not reported.

1For m ethod of grading solid food see p. 91.
2Less than 1 per cent.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Unsatisfactory.
None____ ____
Unsatisfactory.
None............. .

16
24
17

8
6
22 8
14
5
28
10

7
7
4

15
5
3
3

22
12

10
5
6

11 10
14
13
2 3 12
4
6 24
6 9 44

32
15
16
9

11 10
5
3
2
5
3
2

16
15

9
7
9

14
7

6
5
11
19

118

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

D5 .—

able

D iet grade and ty p e o f fee d in g o f children d u rin g secon d y e a r ; children
1 3 m onths o f age and over i n fa m ilie s visited i n P u e rto R ic o

Children 13 months of age and
over in families visited

T yp e of feeding

Age at date of home
visit
M ilk feeding

Total
13 to 24
months

D iet grade
Solid food 1

Breast milk

S-i
©
rÛ

C ow ’s milk
(average
amount
daily)

'S g
S i
»A c

25 months
and over

'S g-

ÖH

u £
©.£5

’S g
O rO

3
&

öS

Ph
57

Total........

233

Grade reported.

203

100

147

100

V ery g o o d ..

N one.

IK to 2 pints Very satisfactory.

G ood ______
N one.
N one.

I K to 2 pints Satisfactory.
1pint........... ____d o ______

Fair.

Poor.

None________
None________
None________
Small amount _
Mainly______
Entirely_____

IK to 2 pints
1 pint______
1 cup______
1 pint______
1 cup______
None______

Unsatisfactory.
____d o _________
Borderline____
Unsatisfactory.
Borderline____
____d o _________

V ery poor.

2
1

5
5
3

10
1

4
18

7
9
4
3

28
4
26
2
1
101

N one.

IK to 2 pints

None.
None.

____do..
1pint..

None.

1cu p .

N one_________
Small amount.

____d o .
1pint..

M ainly_______

1 cu p ..

Do...
Entirely .

Do.
Grade not reported.

____d o.

V ery unsatisfac
tory.
N one___________
Very unsatisfac
tory.
U nsatisfactory or
very unsatistory.
N one___________
Very unsatisfac
tory.
Unsatisfactory or
very unsatis
factory.
N one___________..
Unsatisfactory of
very unsatis
factory.
N one.....................

' 100

7
7
4
3

23
N one_________ I K to 2 pints B orderline..
1pint______ ____ d o ______
None_________
Small amount _ ____d o______ ____ d o ______
M ainly_______ 1cup_______ Satisfactory.

176

3
4

14
13

18
3
17

2
12
1

0
50

1
1
1

3
3
42

3
3

1
9
3

330

9
1

1
2
2
28
3

1
1
1
19

2
1

1
1
1
14

16
2

2
2
2
25

1
2

329

1For method of grading solid food, see p. 92
2Less than 1 per cent.
3Includes 26 children for whom diet grade was not reported as they had lived only one month into this

period.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

APPENDIX D
T

able

119

D 6 .—

D iet grade and ty p e o f fe e d in g o f children d u rin g third y e a r ; children
2 5 m onths o f age and over in fa m ilie s visited in P u erto R ico

T yp e of feeding
M ilk feeding
Diet grade
C ow ’s milk (av
Breast milk erage amount
daily)

Solid food 1

Children 25
months of age
and ov8t at
date of home
visit

57
40

1V>. to 2 p ints..

V ery satisfactory_______
Satisfactory____________

F air..

_________ __________

1
2
8
3

2
3

9
N one_____

W i to 2 pints.. Unsatisfactory_________

Unsatisfactory or verv
unsatisfactory.
........do._______ _________

1
6
2
20
18

2
2 17

1 For method of grading solid food, see p. 92.
2Includes 16 children for whom grade was not reported as they had lived only 1 m onth into this period.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Appendix E.— CASE HISTORIES OF CHILDREN SHOWING
ROENTGEN-RAY EVIDENCE OF RICKETS

Because of the rarity of rickets in Puerto Rico the case histories of
the five children showing roentgen-ray evidence of the disease are
given in detail.
Case 1.— S. M., male, age, 6 months, 16 days. Full term, fourth pregnancy,
normal birth. Three older children were dead; the first at 10 days, of hemorrhage,
the second, at 9 months, of “ meningitis,” the third, at 3 months, of colic.
S. M. was born in a cellar of one of the large stone tenements of San Juan, in a
1-room apartment that had no windows opening to the outside and was lighted
only by electricity. He had been sickly from birth, was breast fed only a few
days, and was never taken out of the cellar for fear he would “ catch cold.” Two
weeks before examination he was moved to a new and lighter tenement. Until
this time he had literally never been exposed to daylight; nor, because he was
artificially fed, had he received the benefit of any antirachitic vitamin through
breast milk. His diet consisted of a cow’s milk formula, without additional food
of any sort. No cod-liver oil had ever been given. There was a history of fre
quent attacks of bronchitis, and, for the past three or four months, daily convul
sions associated with what was thought by the local doctor to be meningitis.
Physical examination showed a fairly well-nourished infant; weight, 7 kilo
grams; height, 64 centimeters. He was pale, cyanotic, breathing rapidly, and
obviously sick. There was no evidence of tanning, and his musculature was
flabby and weak. ^He could not sit up. His head was large and hydrocephalic
in type; the anterior fontanelle was full, and measured approximately 7 centi
meters in its anteroposterior diameter and 7 centimeters in lateral diameter; the
sagittal, coronal, and lambdoidal sutures were open. There was slight craniotabes in the parietal bone just behind and above the left mastoid process. The
frontal and parietal bosses were moderately enlarged. Examination of the chest
showed moderately enlarged costochondral junctions and a Harrison’s groove
with accompanying flaring of the costal margins of the ribs. The epiphyses of
the long bones at the wrist were moderately enlarged. There was a moderate
degree of pot-belly, and an enlarged spleen. The legs showed no rachitic deform
ities. Chvostek’s sign for tetany was not elicited. Examination of the lungs
revealed a definite broncho-pneumonia, with signs of consolidation at the right
upper lobe. Roentgenogram of the chest corroborated the diagnosis of broncho
pneumonia. The heart was negative.
Roentgenograms of the bones of the forearm showed advanced rickets with a
marked degree of osteoporosis accompanying it (fig. 1). The cortex was poorly
calcified; the periosteum, faintly visible, was elevated on both radius and ulna.
There was no evidence of deposit of calcium in the zone of primary calcification of
the lower ends of radius and ulna, and there was a small amount of fraying or
fringing of the distal end of both radius and ulna, but the rachitic intermediary
zone, sometimes called the rachitic metaphysis, was not defined by lime-salt
deposit at its periphery. There was, therefore, essentially no cupping, and only
slight spreading of the ends of both bones. The trabeculae toward the end of
the shaft were irregularly placed, and, at the extreme end of the shaft, lay at
various angles, forming a dense irregular line, such as is commonly seen in severe
rickets. The picture was that of the type of rickets described by Wimberger81
as occurring in inactive children, a type which though very severe, shows com
paratively little evidence of cupping and fraying, but marked osteoporosis of
the bones. Such a picture occurs when the disease is so severe that there is
inability on the part of the organism to reinforce the weakened bone by laying
down lime salts in the periphery of the rachitic intermediary zone.
Blood studies were made in the chemical laboratory of the School of Tropical
Medicine and showed a calcium content of 7.1 milligrams per cubic centimeter of
serum, and a phosphorus content of 4.1 milligrams. A diagnosis of severe active
rickets was made from the clinical and roentgen-ray examinations, and of tetany
si Wimberger, Hans: Klinisch-radiolische Diagnostik von Rachitis, Skorbut, und Lues inn Kindesalter[Clinieal and Radiological Diagnosis of Rickets, Scurvy, and Congenital Syphilis in Childhood.] Ergeb
nisse der inneren M edizin und Kinderheilkunde [Berlin], vol. 28 (1925), pp, 269-288,

120

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

APPENDIX E

121

from the chemical examination of the blood. Treatment with sun baths was insti
tuted in the Presbyterian Hospital. The diet was kept the same. No cod-liver
oil was given. At the end of 18 days healing had begun (fig. 2), as was shown
by the deposit of lime salt both in the subperiosteal tissues and in and around the
intermediary rachitic zone at both ends of the radius and ulna, and by the increase
in the blood calcium to 8.3 milligrams and in the phosphorus to 5.2 milligrams.
Three months later advanced healing had taken place (fig. 3).
Case 2.— W. V., male, age 5 months, 12 days. Full term. He was breast
fed at night with supplementary feeding of condensed milk and rice water during
the day. He was taken out of doors very little, about half an hour, daily in the
shade. His mother went out of doors very little.
Physical examination: Weight, 6 kilograms; height, 64)4 centimeters. The
amount of his subcutaneous fat and his muscular development were estimated
as fair; the color of his mucous membranes was good; his skin was lightly pig
mented, but there was no tanning. The costochondral junctions were slightly
enlarged; there was slight bowing of the legs, which was probably not abnormal
at his age. A clinical diagnosis of questionable rickets was made on the basis
of the slightly enlarged costochondral junctions.
The roentgenogram showed a fairly thick type of bone with a slight degree of
osteoporosis. The cortex was poorly calcified and thick, apparently having been
laid down in layers, a condition frequently seen at this age when growth is very
rapid; the periosteal surface of the shaft was not everywhere well defined; the
zones of primary calcification of both radii and ulnae were poorly defined and
irregularly calcified, there was slight cupping of both ulnae, and irregular calcifica
tion which suggested fringing. A roentgenographic diagnosis of slight rickets
was made.
Case 3.— J. L., female, age 5 months 13 days. Full term. Breast fed entirely.
Physical examination: Weight, 6.24 kilograms; height, 61 centimeters. A
well-nourished, plump infant, with good muscular development and good color
of mucous membranes. Her skin was lightly pigmented and was tanned only
on the knees. There was no clinical evidence of rickets.
Roentgenographic examination showed a less heavy type of bone than that of
the previous case, and one which was fairly well calcified throughout. The cortex
was of average thickness. There was definite though slight cupping at the distal
ends of both ulnae. The zone of primary calcification was not quite clearly defined
nor everywhere complete. There were certain breaks in the contour which sug
gested veiy early rickets. The ulna side of the distal end of the radius showed
a slight decrease in density; otherwise the distal ends of the radii appeared normal.
A roentgenographic diagnosis of very slight rickets was made.
Case 4-— M. S., male, age 2 months 26 days. Full term. Breastfed.
Physical examination: Weight, 4.88 kilograms; height, 56 centimeters. A
fairly well-nourished infant with good muscular development, pale mucous mem
branes, lightly pigmented skin, and no tanning. The anterior fontanelle was
4 centimeters long and 5 centimeters wide. There was a moderate degree of
craniotabes, which was bilateral. The costochondral junctions were slightly en
larged. A clinical diagnosis of slight rickets was made.
Roentgenographic examination showed bones of average thickness. The shaft
was well calcified, with cortex of normal density and thickness. There was slight
cupping of the distal ends of both ulnae. The zones cf primary calcification of the
ulnae showed slight irregularities and what appeared to be breaks in calcification.
The distal ends of the radii were also not clearly defined. A roentgenographic
diagnosis of very slight rickets was made.
Case 5.— M. G., female, age 25 months. Full term. Born in New York City,
where the first 21 months of life were spent, and taken to Puerto Rico 4 months
before examination. She was breast fed for 3 months, then given a diet at
first of fresh cow’s milk and later of condensed milk, supplemented by soup, eggs,
potatoes, yautias, cereal, and fruit. No cod-liver oil had ever been given.
Physical examination showed a well-nourished child— weight, 12.02 kilograms;
height, 85 centimeters. Her skin and mucous membranes were of good color,
and she was tanned on face, neck, arms, and legs. There was very little clinical
evidence that would lead to a diagnosis of rickets, slightly enlarged costochondral
junctions and slight knock-knees being the only signs present.
The roentgenogram of the bones of the forearm taken on the day of examina
tion showed evidence of old healed rickets. The shafts of the bones were well
calcified, there was no osteoporosis, and the cortex was clearly defined and of
£bout average width. The diaphyseal-epiphyseal junctions were well ealgified

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

122

DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO

and clearly defined. The zones of primary calcification were unbroken. The
trabeculae at the distal ends of both radii and ulnae were regular, well calcified
and clearly visible up to the zone of primary calcification.
’
,
substance of the spongiosa, about 2 centimeters from the distal end of
both the right and left ulnae, there was an irregular, rather poorly defined zone
of increased density, approximately 3 millimeters in width, made up of irregularly
and rather closely placed trabeculae. There was a slight thickening of the inner
surface of the cortex in this region and just above, with some irregular narrowing
« ^medUi j ai y sPace- In the radii, also, similar zones, though even less wed
detmed, could be seen at approximately 2 centimeters from the distal ends of the
b° n®8* tn both radii at the region of this zone there was a slight bulging of the
shatt. lh e zones of increased density probably represent an old healed rachitic
process, which was of moderate severity when active. The amount of bone that
has grown since the zone formed would indicate that the rachitic process
had probably occurred in the winter and healed in the summer previous to the
examination.
Though this case is included in Table 10 it can not legitimately be counted as
a case of rickets developing in the Tropics.

https://fraser.stlouisfed.org
Federal Reserve Bank of St. Louis

Full text of The Effect of Tropical Sunlight on the Development of Bones of Children in Puerto Rico : A Roentgenographic and Clinical Study of Infants and Young Children with Special Reference to Rickets and Related Factors, Bureau Publication No. 217

FRASER