UNITED STATES DEPARTMENT OF LABOR 
FRANCES PERKINS, Secretary 
CHILDREN’S BUREAU 
GRACE ABBOTT, Chief 
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 
MARTHA M. ELIOT, M. D. 
By 
Bureau Publication No. 217 
UNITED STATES 
GOVERNMENT PRINTING OFFICE 
WASHINGTON : 1933 
For sale by the Superintendent of Documents, Washington, D. C. 
Price 10 cents CONTENTS 
Page 
Letter of transmittal v 
Purpose of investigation 1 
General conditions affecting health of young children in Puerto Rico 3 
Climate and sunlight 3 
Hours of sunlight 4 
Intensity of sunlight 4 
Exposure to sunlight 5 
Density of population 6 
Race 6 
Economic conditions and diet 6 
Indexes of child health 8 
Infant mortality 8 
Communicable diseases 9 
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 17 
Sex and age 17 
Skin pigmentation 18 
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 
Onset of dentition 31 
Color of mucous membranes 31 
Other physical findings 32 
Incidence of rickets in Puerto Rico 35 
Number and age distribution of children examined for rickets 37 
Methods of examination for rickets 37 
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 48 
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 66 
Family income in relation to diet 67 
Per capita income 67 IV 
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 
Milk 78 
Butter 78 
Eggs 78 
Meat and fish 79 
Fruit 79 
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 
Value of breast feeding in relation to mother’s diet 94 
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 LETTER OF TRANSMITTAL 
United States Department of Labor, 
Children’s Bureau, 
Washington, May 81, 1933. 
Madam: 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 w~as 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. 
,Grace Abbott, Chief. 
Hon. Frances Perkins, 
Secretary of Labor. 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 
PURPOSE OF INVESTIGATION 
The study of Puerto Rican children 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 living in a temperate 
climate. During a previous investigation made in New Haven, 
Conn., by 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 interpreted 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 rick- 
ets, 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. Study of a group of infants and young children who had 
lived continuously in 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 
investigation, 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 in 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 rickets (unquestioned) in Puerto Rico and also 
of those minor 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 
1 Eliot, Martha M.: The Control of Rickets; preliminary discussion of the demonstration in New Haven. 
Journal of the American Medical Association, vol. 85, no. 9 (Aug. 29,1925), pp. 656-663. 2 
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. GENERAL CONDITIONS AFFECTING HEALTH OF YOUNG 
CHILDREN IN PUERTO RICO 
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 
1 Fassig, Oliver L.: The Climate of Puerto Rico. Puerto Rico Review of Public Health and Tropical 
Medicine, vol. 4, no. 5 (November, 1928), p. 203. 4 
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. 
HOURS OF SUNLIGHT 
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 
Month 
Daily average of possi- 
ble hours of sunlight 
Daily average of aetual 
hours of sunlight 
San Juan 
New Haven 
San Juan 
New Haven 
January 
11: 2 
9. 6 
6. 9 
4. 7 
February 
11. 6 
10. 6 
7. 8 
6. 4 
March . - _ _ . 
12. 0 
12. 0 
8. 2 
7. 1 
April _ _ 
12. 6 
13. 3 
7. 8 
7. 5 
May __ 
13. 0 
14. 5 
7. 6 
8. 4 
June 
13. 2 
15. 1 
7. 7 
9. 2 
July 
13. 1 
14. 8 
8. 1 
9. 2 
August 
12. 7 
13. 8 
8. 4 
8. 4 
September 
12. 2 
12. 5 
7. 4 
7. 7 
6. 6 
October 
11. 7 
11. 1 
7. 3 
N ovember 
11. 3 
9. 9 
6. 9 
5. 3 
4. 9 
December _ 
11. 1 
9. 3 
6. 8 
Total for year 
4, 416 
12. 1 
4, 456 
12. 2 
2, 774 
7. 6 
2, 592 
Daily average for year 
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 hoars 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 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. Tis- 
INTENSITY OF SUNLIGHT 
3 Figures supplied by 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. GENERAL HEALTH CONDITIONS 
5 
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 
Rico, 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 Rico 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. 
EXPOSURE TO SUNLIGHT 
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 
light, though apparently no conscious effort is made to insure their get- 
ting any. After this early period the amount of exposure to sunlight 
increases 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 them 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 
year 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. 
4 Tisdall, Frederick F., and Alan Brown: Relation of the Altitude of the Sun to Its Antirachitic Effect. 
Journal of the American Medical Association, vol. 92, no. 11 (Mar. 16, 1929), pp. 860-864. 6 
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 full- 
blooded 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. Boas 7 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 mixing 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 
role 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 
6 Fifteenth Census of the United States, 1930, vol. 1, Population. 
« Ibid, pp. 1251, 1263. 
7 Boas, Franz: The Anthropometry of Puerto Rico. American Journal of Physical Anthropology, vol. 
3, No. 2 (April-June, 1920), pp. 247-253. GENERAL HEALTH CONDITIONS 
7 
proportion of the men 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 tune 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 lactating 
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 dining the same year the total production of milk was 
estimated to be 7,805,143,792 gallons, or 26 ounces per capita daily.8 
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 of 
health; but this amount probably was small. 
8 Fourteenth Census of the United States, 1920, vol. 5, Agriculture, pp. 23, 654; vol. 6, pt. 3, Agriculture, 
pp. 383, 388. _ „ 
9 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1926, p. 83; 1927, p. 113. San 
Juan. 8 
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. 
INFANT MORTALITY 
That the infant mortality rate in a community varies with social 
and 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, 
10 During the fiscal year ended June 30, 1926, 3,743,803 pounds of canned milk (about 2}4 pounds per cap- 
ita), 4,666,931 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 Rico, pp. 19, 22. 
San Juan. 
11 A Handbook of Dairy Statistics, by T. R. Pirtle. U. S. Department of Agriculture. Washington, 
1928. 
12 In December, 1929, birth registration was only 55 per cent complete, according to the Report of the 
Commissioner of Health of Puerto Rico 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 Rico 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 birth- 
registration 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 Rico 
was admitted to the death-registration area as of 1932. 
13 Causal Factors in Infant Mortality, by Robert Morse Woodbury. U. S, Children’s Bureau Publica- 
tion No. 142. Washington, 1925. GENERAL HEALTH CONDITIONS 
9 
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: 
Fiscal year 
Deaths of 
infants 
under 1 year 
per 1,000 
live births 14 
1919-20 
146 
1920-21 
162 
1921-22 
152 
1922-23 
143 
1923-24 
128 
1924-25 
148 
Fiscal year 
Deaths of 
infants 
under 1 year 
per 1,000 
live births 14 
1925-26 
150 
1926-27 
167 
1927-28 
146 
1928-29 
15179 
1929-30 
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 live births, for 1926 it was 9.7, and for 1927 it was 7.8.17 
COMMUNICABLE DISEASES 
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 
14 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1925, p. 121; 1927, p. 133; 1930, 
p. 29. San Juan. 
15 The hurricane of 1928 directly or indirectly caused many deaths, which are reflected in the increased 
infant mortality rate. 
16 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1925 (p. 100 and Appendix 
Table, not numbered); 1926 (p. 101 and Appendix Table 25); and 1927 (p. 131 and Appendix Table 26). San 
Juan. 
•7 Birth, Stillbirth, and Infant Mortality Statistics. U. S. Bureau of the Census, 1925, pp. 2 and 195; 1926, 
pp. 7 and 200; and 1927, pp. 2 and 194. Washington. 
18 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1926, p. 99. San Juan. 10 
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 cent22 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 
15 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1926, p. 41; 1927, p. 61. San 
Juan. 
20 Ibid., 1925, p. 33. 
21 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1925, pp. 77-84. San Juan. 
22 Ibid., 1926, p. 53. 
22 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. GENERAL HEALTH CONDITIONS 
11 
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 Rfo Piedras,24 where 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 death- 
registration area for the corresponding periods show striking differ- 
ences. The tuberculosis death rates for every 100,000 population were: 
Puerto Rico (fiscal years)2S 
1923-24 
205 
1924-25 
221 
1925-26 
240 
1926-27 
267 
1927-28 
.. 237 
1928-29 
__ 27 301 
1929-30 
283 
u. s. 
death-registration area (calendar years) 26 
1923- _ 
94 
1924_. 
90 
1925__ 
87 
1926-_ 
87 
1927-- 
81 
1928-_ 
79 
1929-_ 
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 a far 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 Rico for the Fiscal Year, 1927, p. 135. San Juan. 
28 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1925, p. 112; 1927, p. 135; 1930, 
p. 70; and Twenty-Eighth Annual Report of the Governor of Puerto Rico, Fiscal Year 1928, p. 46. 
26 Mortality 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 during the time immediately following it. 
28 Framingham Community Health and Tuberculosis Demonstration of the National Tuberculosis Asso- 
ciation. Final Summary Report, 1917-1923, inclusive. Framingham, Mass., 1924. 
24 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1927, p. 135. 
80 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1927, Appendix Table, Mor- 
tality from Tuberculosis, and Appendix Tables 26 and 27. 
160326°—33 2 12 
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. 
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: 
CAUSES OF DEATH AMONG CHILDREN 
Table 1.—Cause of death and age at death (as reported hy mother); brothers and 
sisters of children examined in Puerto Rico 
Cause of death 
Brothers and sisters of children examined 
Total 
Age at death 
Under 1 
month 
1 month, 
under 6 
6 months, 
under 
1 year 
1 year 
and over 
Not re- 
ported 
All causes 
356 
38 
60 
58 
167 
33 
Causes known 
293 
28 
53 
57 
151 
4 
Communicable diseases 
102 
8 
19 
24 
49 
2 
29 
1 
5 
5 
18 
50 
1 
9 
13 
27 
Malnutrition,.. 
67 
7 
10 
15 
35 
5 
3 
2 
9 
2 
3 
4 
6 
2 
4 
25 
4 
7 
14 
Causes unknown... _ 
63 
10 
7 
1 
16 
29 
The age distribution of the deaths that occurred among children 
under 1 year is strikingly different among this group of Puerto Rican GENERAL HEALTH CONDITIONS 
13 
children from the distribution among children in the United States birth- 
registration area in 1927,31 as is shown in the following comparison: 
U. S. birth-repis- 
Puerto Rico 
tration area32 
Under 1 month 
... 24 per cent 
56 per cent 
1 month, under 6 months. _ 
38 per cent 
27 per cent 
6 months, under 1 year 
_ 37 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 Itico 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 one- 
third 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 w as 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 by 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 Rico 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 Mortality Statistics, pt. 1, 1927, p. 186. U. S. Bureau of the Census. 
Washington, 1929. 
33 Estudio de la Mortalidad Infantii en Puerto Rico, p. 24. San Juan. METHOD OF INVESTIGATION 
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 fandlies of 28 of the 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 
and 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. METHOD OF INVESTIGATION 
15 
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 iived; the type of house or dwelling; 
the income on which the family was maintained and its source; the 16 
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 in 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 consisted 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. PHYSICAL CONDITION OF CHILDREN EXAMINED 
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 Wassermann 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. 
Table la shows liow 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 study- 
SEX AND AGE 18 
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. 
SKIN PIGMENTATION 
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. 
Table la.—Age at examination; boys and girls examined in Puerto Rico 
Children examined 
Age 1 at examination 
Total 
Boys 
Girls 
Number 
Per cent 
distribu- 
tion 
Number 
Per cent 
distribu- 
tion 
Number 
Per cent 
distribu- 
tion 
584 
308 
276 
564 
100 
296 
100 
268 
100 
73 
13 
37 
13 
36 
13 
98 
17 
53 
18 
45 
17 
75 
13 
43 
15 
32 
12 
10 to 12 months 
74 
13 
44 
15 
30 
11 
104 
18 
54 
18 
50 
19 
76 
13 
37 
13 
39 
15 
25 to 34 months 
64 
11 
28 
9 
36 
13 
20 
12 
8 
1 Age is given as of nearest month; that is, “under 4 months” is actually under 3 months and 16 days, 
“4 to 6 months” 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 light, 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 light 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 
light group. Undoubtedly the increasing depth of pigmentation that 
the older children showed was due, in some part at least, to the in- PHYSICAL CONDITION OF CHILDREN EXAMINED 
19 
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. 
SKELETAL GROWTH AND BODY WEIGHT 
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 re- 
31 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. 20 
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 
Cliildren’s Bureau hi 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 born 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 Now 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 N ew Haven children under 34 
m 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 movable footpiece that slides in two grooves, one on each side of the board, parallel to the meter meas- 
uring sticks. Measurements can be read on either of the two measuring sticks. 
36 Observed data, smoothed values, and equations of curves are given in Appendix A, p. 100. 
37 Statures and Weights of Children under Six Years of Age, by Robert M. Woodbury, pp. 85, 102, 104. 
U. S. Children’s Bureau Publication No. 87. Washington, 1921. 
3* Data collected during study made by the U. S. Children’s Bureau in cooperation with the department 
of pediatrics, Yale University School of Medicine (unpublished), 21 
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 
Puerto Rican boys (293) 
United States white boys (more than 47,000) 
United States white boys whose mothers were born in Italy (4,366) 
United States Ne$ro boys (1,406) J 
Weight 
Kilo^rama 
Chart 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 
Month, of a^e 
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 
fifteentli 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. 22 
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 
Puerto Rican girls (26+) 
United States white girls (more than45,000) 
United States white girls whose mothers were born in \ta.\y (.4,226) 
United States Negro girls (1,504) 
Weigk-t- 
Kilo^rame 
Month of age 
Chart II.—Average weight of girls 1 to 33 months of age examined in Puerto Rico, 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 twenty- 
third 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 PHYSICAL CONDITION OF CHILDREN EXAMINED 
23 
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 
Puerto Kican-boys (292 in height groups 295 in head-circumference group) 
UnitedStates white boys (more than47,000) 
United States white boys whose mothers were born in Italy (4,366) 
United States Negro boys (1,406) 
New Haven white boys (curve bused on 350 boys 
I to 60 months; section shown represents 
trendferboy3 lto33 months) 
Height 
Centimeters 
Head circumference 
Month ofage 
Chart III—Average height of boys 1 to 33 months of age examined in Puerto Rico, 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 New 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 for 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 stand- 
38s Statures and Weights of Children under Six Years of Age, by Robert M. Woodbury, p. 107. U. S. 
Children’s Bureau Publication No. 87. Washington, 1921. 24 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Puerto Ricangirls (265 in height group; 266 in head-circumfference group) 
United States white girls (more than 45,000) . 
United States white girls whose mothers were born in Italy (4;226) 
United States Negro girls (1,504) 
New Haven white girls (curve based on 334girls I to 60 months; 
section shown represents trend for girls I to 33 months) * • 
Height 
Ce»*t>imetero 
Head circumference 
Month of age 
Chart IV.—Average height of girls 1 to 33 months of age examined in Puerto Rico, 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 New Haven 
■Puerto Rioan boys, 48 to 3+ cm. in height (302) 
United States white boys, 5o.8to9l.4cm. In height (51,185) 
Puerto Rioan girls, 46 to 92 cm. in height (2T0) 
United States white girl a, 50.8 to 914 cm. in height (51,529) 
weigkt in kilognsms 
Height in centimeters 
Chart 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 same height PHYSICAL CONDITION OF CHILDREN EXAMINED 
25 
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 clearly 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 haves 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 j udgment 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 the 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 26 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
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 
PIGMENTATION 
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 under 7 months of age at the 
time of examination. That breast feeding is probably responsible for 
Per cent 
Puerto Rico 
563 children 
New Haven 
918 children 
■Subcutaneous fab 
Very good 
Good 
Fair 
Poor 
Very poor 
Chart Va.—Percentage distribution of grades of subcutaneous fat of children examined in Puerto 
Rico, compared with that of children examined in New 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. 
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. PHYSICAL CONDITION OF CHILDKEN EXAMINED 
27 
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 of 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. 
Muscle 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 
160326°—33 3 28 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
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 bo 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 
a** The New Haven group includes children of older ages than the Puerto Rican, hut the proportion 
completing the various stages of motor development at the specified ages is probably not influenced by 
the differences in the age distribution. For a comparison of the age at examination of the Puerto Rican 
and New Haven children see Table 12. PHYSICAL CONDITION OF CHILDREN EXAMINED 
29 
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. 
Table 2.—Age at holding up head; children examined in Puerto Rico and in New 
Haven, Conn. 
Age 1 at holding up head, as reported by mother 
Children examined 
in Puerto Rico 
Children examined 
in New Haven 
Number 
Per cent 
distri- 
bution 
Number 
Per cent 
distri- 
bution 
Total 
584 
1,186 
448 
100 
597 
100 
Under 2 months 
22 
5 
18 
3 
2 months _ _ _ _____ 
155 
35 
168 
28 
3 months 
152 
34 
262 
44 
4 months,.. ... . _ _ _. __ _ ___ __ _ 
63 
14 
110 
18 
5 months and over 
56 
13 
39 
7 
56 
481 
Not holding up head at date of examination. . __ __ 
52 
104 
Not reported whether holding up head. 
28 
4 
'Age is given as of nearest month; that is, “under 2 months” is actually under 1 month and 16 days, 
“2 months” is from 1 month and 10 days to 2 months and 15 days, inclusive, and so on. 
Table 3.—Age at sitting alone; children examined in Puerto Rico and in New 
Haven, Conn. 
Age 1 at sitting alone, as reported by mother 
Children examined 
in Puerto Rico 
Children examined 
in New Haven 
Number 
Per cent 
distri- 
bution 
Number 
Per cent 
distri- 
bution 
584 
1,186 
Age reported 
346 
100 
697 
100 
Under 5 months 
25 
7 
44 
6 
5 months 
59 
17 
92 
13 
6 months 
110 
32 
260 
37 
7 months... - 
55 
16 
146 
21 
8 months.-- 
50 
14 
78 
ii 
9 months and over 
47 
14 
77 
a 
Age not reported ___ 
20 
215 
Not sitting alone at date of examination. - 
190 
270 
Not reported whether sitting alone--- -_ 
28 
4 
'Age is given as of nearest month; that is, “under 5 months” is actually under 4 months and 16 days, 
“5 months” is from 4 months and 16 days to 5 months and 15 days, inclusive, and so on. 30 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table 4.—Age at standing alone; children examined in Puerto Rico and in New 
Haven, Conn. 
Children examined 
in Puerto Rico 
Children examined 
in New Haven 
Age 1 at standing alone, as reported by mother 
Number 
Per cent 
distri- 
bution 
Number 
Per cent 
distri- 
bution 
Total - - 
584 
1,186 
252 
100 
479 
100 
36 
14 
25 
5 
8 months ... - - _ 
56 
22 
50 
10 
9 months. 
48 
19 
82 
17 
48 
19 
104 
22 
29 
12 
76 
16 
25 
10 
68 
14 
10 
4 
74 
15 
31 
237 
Not standing alone at date of examination. . 
273 
466 
Not reported whether standing alone 
28 
4 
’Age is given as of nearest month; that is, “under 8 months” is actually under 7 months and 16 days, 
“8 months” is from 7 months and 16 days to 8 months and 15 days, inclusive, and so on. 
Table 5.—Age at walking alone; children examined in Puerto Rico and in New 
Haven, Conn. 
Age 1 at walking alone, as reported by mother 
Children examined 
in Puerto Rico 
Children examined 
in New Haven 
Number 
Per cent 
distri- 
bution 
Number 
Per cent 
distri- 
bution 
Total 
584 
1,186 
Age reported 
219 
100 
520 
100 
Under 10 months- - „ 
35 
16 
27 
5 
10 months 
38 
17 
38 
7 
11 months . 
18 
8 
56 
11 
71 
32 
104 
20 
13 months - _ _ _ _ 
26 
12 
75 
14 
14 months.. _ 
6 
3 
85 
16 
15 months and over... — ___ 
25 
11 
135 
26 
8 
48 
Not walking alone at date of examination _ 
329 
614 
Not reported whether walking alone 
28 
4 
’Age is given as of nearest month; that is, “under 10 months” is actually under 9 months and 16 days, 
“10 months” 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 PHYSICAL CONDITION OF CHILDREN EXAMINED 
31 
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. 
Table 6.—Age at onset of dentition; children examined in Puerto Rico and in 
New Haven, Conn. 
Children examined 
in Puerto Rico 
Children examined 
in New Haven 
Age 1 at onset of dentition, as reported by mother 
Number 
Per cent 
distri- 
bution 
Number 
Per cent 
distri- 
bution 
Total 
584 
1,186 
Age reported 
336 
100 
706 
100 
Under 5 months 
60 
18 
76 
11 
5 or 6 months 
108 
32 
231 
33 
7 or 8 months . 
96 
29 
199 
28 
9 or 10 months 
44 
13 
126 
18 
11 months and over 
28 
8 
74 
10 
Age not reported ... _ . 
15 
143 
No sign of teeth at date of examination 
205 
328 
No report on onset of dentition 
28 
9 
1 Age is given as of nearest month; that is, “under 5 months” is actually under 4 months and 16 days, 
“5 or 6 months” 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 32 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
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 condi- 
89 Physical Status of Preschool Children, Gary, Ind., by Anna E. Rude, M. D., pp. 39, 78. U. S. Chil- 
dren’s Bureau Publication No. 111. Washington, 1924. 
40 Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1925, p. 80. PHYSICAL CONDITION OF CHILDREN EXAMINED 
33 
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 Wasser- 
mann 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 
Malnutrition 54 
Gastroenteritis 9 
Communicable diseases 188 
Rhinopharyngitis 117 
Laryngitis 2 
Bronchitis 27 
Broncho-pneumonia 1 
Tuberculosis 2 
Tuberculosis, suspected.. 16 
Fever, unexplained (prob- 
ably due to communi- 
cable disease) 7 
Erysipelas 1 
Mumps 1 
Malaria 1 
Malaria, suspected 2 
Nasal diphtheria, suspec- 
ted 1 
Congenital syphilis 3 
Congenital syphilis, sus- 
pected 5 
Dysentery 2 
Diseases of ears 3 
Chronic otitis media 3 
Diseases of nose and throat 138 
Hypertrophied tonsils or 
hypertrophied tonsils 
and adenoids 132 
Hypertrophied adenoids 
only 6 
Diseases of the eyes 18 
Corneal scar and ectro- 
pion 1 
Conjunctivitis 11 
Gonorrheal conjunctivi- 
tis 1 
Stye 2 
Internal strabismus 3 
Diseases of central nervous 
system 9 
Poliomyelitis, old, with 
paralysis 2 
Birth injury, probable 3 
Hydrocephalus 3 
Facial paralysis 1 
Genito-urinary diseases 9 
Vaginitis 2 
Inguinal hernia 5 
Hydrocele 2 
Skin diseases 62 
Impetigo 45 
Hemangioma 1 
Other skin diseases 16 
Congenital defects 52 
Congenital heart dis- 
ease 6 
Mongolian idiocy 1 
Mongolian idiocy, ques- 
tionable 1 
Supernumerary fingers-_ 2 
Tongue-tie 41 
Harelip 1 
The largest group of pathological conditions is that of communicable 
or infectious 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 oidy, 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 34 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
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 respiratory system 
Communicable diseases, positive 
148 
or suspected _ _ _ 
83 
Diseases of eyes. _ — — ._ 
68 
Earaches and otitis media. _ _ 
.. 57 
Wasting diseases 
__ 51 
Convulsions. 
._ 51 
Undefined fevers 
_. 23 
Diseases of nervous system 
9 
Others _ 
._ 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 raquitismo, or 
fatigue, 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). INCIDENCE OF RICKETS IN PUERTO RICO 
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 roentgenograpliic 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 hones. 
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 
41 Hess, Alfred F.: Rickets, Osteomalacia, and Tetany, pp, 54-55. Lea & Febiger, Philadelphia, 1929. 36 
DEVELOPMENT OP BONES OF CHILDREN IN PUERTO RICO 
bowing of tlie 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, lie 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 discontinued45 in the year following the 
Children’s Bureau study. 
42 Attention may be called to the fact that climatic conditions in the Isthmus of Panama are different from 
those in Puerto Rico and may well account for the rickets found there by 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 New Haven, Conn. (See p. 4.) During 
the rainy season, which lasts seven or eight months, the monthly 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 New Haven only in the late fall and winter months, and in Puerto Rico not at all. 
(See Climatological Data for Central America, by W. W. Reed, in Monthly Weather Review, vol. 51, no. 3, 
March, 1923, pp. 133-141, published by the Weather Bureau, U. S. Department of Agriculture, Washington.) 
43 Torroella, Mario A.: Raisons pour lesquelles le raehitisme n’existe pas au Mexique. Archives de 
Mfidecine des Enfants, vol. 32, no. 5 (May, 1929), pp. 262-269. 
44 Gonzales, Martin: Raisons pour lesquelles le raehitisme n’existe pas au Mexique. Bulletins de la 
Sod fit,e de Pfidiatrie de Paris, vol. 27, Jan. 15, 1929, pp. 42-54. 
48 See Report of the Commissioner of Health of Puerto Rico for the Fiscal Year 1927, p. 132, and later 
reports. INCIDENCE OF RICKETS 
37 
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. 
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 
shoidd 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, roentgeno- 
graphic 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. 
METHODS OF EXAMINATION FOR RICKETS 38 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
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 shoidd 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 borne 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 KICKETS 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 INCIDENCE OF RICKETS 
39 
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. 
Table 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 1 at examination 
Total 
No rickets 
Question- 
able 
Slight 
Moderate 
Marked 
diagnosis 
Num- 
Per 
Num- 
Per 
Num- 
Per 
N um- 
Per 
Num- 
Per 
her 
cent2 
her 
cent2 
her 
cent2 
her 
cent2 
her 
cent2 
Total 
584 
400 
68 
134 
23 
46 
g 
3 
1 
1 
(3) 
Under 7 months 
171 
143 
84 
25 
15 
3 
2 
7 to 12 months 
149 
94 
63 
39 
26 
14 
9 
i 
1 
1 
1 
13 to 18 months 
104 
65 
63 
26 
25 
12 
12 
i 
1 
19 to 34 months _ 
140 
83 
59 
39 
28 
17 
12 
i 
1 
Not reported 
20 
15 
5 
1 Age is given as of nearest month; that is, “under 7 months” is actually under 6 months and 16 days, “7 
to 12 months” is from 6 months and 16 days to 12 months and 15 days, inclusive, and so on. 
2 Per cent not shown where number of children was less than 50. 
3 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- 40 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
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 of 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 Bl, 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 incidence, 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 
Rico 
New 
Haven 
Enlarged costochondral junctions ____ 
16 
58 
Enlarged epiphyses at wrist _ _ _ 
10 
55 
Bowlegs __ _ 
30 
50 
Knock-knees _ ____ 
28 
23 
Harrison’s groove. 
6 
25 
Pigeon-breast _ _ _ 
(«) 
2 
Moderate enlargement of— 
Either frontal or parietal bosses 
7 
27 
Both frontal and parietal bosses 
1 
15 
Asymmetry of head 
6 
(47) 
Craniotabes _ _ __ 
1 
4 
« Less than 1 per cent. 
17 Not reported. 41 
INCIDENCE OF RICKETS 
Table 8.—Incidence of skeletal signs usually considered evidence of rickets; children 
examined in Puerto Rico and in New Haven, Conn. 
Skeletal signs 
584 children examined in Puerto 
Rico (1 to 34 months of age) 
956 children examined in New 
Haven (1 to 36 months of age) 
Total 
re- 
ported 
Signs 
present 
Signs 
absent 
Not re- 
ported 
Total 
re- 
ported 
Signs 
present 
Signs 
absent 
Not re- 
ported 
Enlarged costochondral junctions 
581 
92 
489 
3 
954 
549 
405 
2 
Enlarged epiphyses at wrist 
583 
60 
523 
1 
952 
519 
433 
4 
Bowlegs ... 
582 
176 
406 
2 
i 547 
276 
271 
2 409 
Knock-knees.. ... 
i 303 
85 
218 
2 281 
i 555 
127 
428 
2 401 
Harrison’s groove 
583 
37 
546 
1 
951 
233 
718 
5 
Pigeon-breast 
577 
1 
576 
7 
954 
15 
939 
2 
Moderate enlargement of— 
Either frontal or parietal bosses . 
f 43 
1 ro/i 
f 259 
\ KAK 
Both frontal and parietal bosses. _ 
} i)8o 
l 4 
{ 146 
> 545 
6 
Asymmetry of head 
551 
33 
518 
33 
(3) 
(3) 
(3) 
<3) 
Craniotabes 
581 
5 
576 
3 
945 
37 
908 
ii 
1 Includes only children who were standing at date of examination. 
2 Children not yet standing at date of examination. 
3 Not 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 showm by the roentgen- 
ray 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 wdio have not had 
rickets is made evident by the occurrence of these signs in Puerto Rico, 
wdiere, 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 wdien they did 
so occur, more weight, as would be expected, wras given to each one in 
making the diagnosis. 
Table 9 show's the degree in wdiich 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 wras, 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 one- 
eighth 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 42 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
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. 
Table 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- 
Rickets 
able 
diag- 
nosis 
Total 
Slight 
Moder- 
ate 
Marked 
92 
9 
50 
33 
30 
2 
1 
89 
9 
50 
30 
28 
2 
3 
3 
2 
1 
Epiphyses of long bones at wrists (slight en- 
60 
9 
30 
21 
19 
1 
1 
176 
90 
57 
29 
28 
1 
Slight 
153 
89 
45 
19 
19 
23 
1 
12 
10 
9 
1 
85 
' 51 
20 
14 
13 
1 
68 
45 
14 
9 
9 
17 
6 
6 
5 
4 
1 
37 
3 
15 
19 
15 
3 
1 
34 
3 
14 
17 
14 
2 
1 
3 
1 
2 
1 
1 
1 
1 
1 
47 
13 
24 
10 
8 
1 
1 
Parietal or frontal bosses moderately en- 
43 
13 
23 
7 
Parietal and frontal bosses moderately 
4 
1 
3 
1 
1 
1 
33 
20 
9 
4 
3 
1 
31 
18 
9 
4 
3 
1 
2 
2 
5 
1 
2 
2 
1 
1 
Slight 
4 
1 
2 
1 
1 
1 
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 43 
INCIDENCE OF RICKETS 
points of interest in interpreting their occurrence in this tropical 
region as compared with their occurrence in New Haven. 
Bowlegs.—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. 
Knock-knees.—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. 
Enlarged costochondral junctions and epiphyses of long bones at wrist.— 
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. 
Harrison's groove.—Harrison’s groove was found in only one-fourtli 
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, 
160326°—33 1 44 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
were also found, and in 14 additional children there was enlargement 
of the cervical lymph nodes, probably indicating infection in the nose 
or throat. Tlds 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. 
Head signs.—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 in 
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 liistories 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. 
RELIABILITY 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. INCIDENCE OF RICKETS 
45 
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. 46 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table 10.—Roentgen-ray diagnosis of rickets and age at date of examination; chil- 
dren examined in Puerto Rico 
Age 1 at examination 
Children examined in Puerto Rico 
Total 
Roentgen-ray diagnosis of rickets 
No rickets 
Doubtful 
diagnosis 
Rickets 
Total 
Active, 
marked 
Slight, 
heal- 
ing 
Slight. 
healed 
Num- 
ber 
Per 
cent 
Num- 
ber 
Per 
cent 
Num- 
ber 
Per 
cent 
Total -- 
584 
509 
87 
70 
12 
5 
1 
1 
3 
i 
Under 4 months . 
73 
60 
82 
12 
16 
1 
1 
1 
98 
67 
68 
29 
30 
2 
2 
2 
7 to 9 months . - 
75 
62 
83 
12 
16 
1 
1 
1 
10 to 12 months 
74 
62 
84 
12 
16 
13 to 18 months .. - 
104 
99 
95 
5 
5 
19 to 24 months 
76 
76 
100 
64 
63 
98 
1 
2 
1 
Not reported ... 
20 
20 
•Age is given as of nearest month; that is, “under 4 months” is actually under 3 months and 16 days, 
“4 to 6 months” 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 mdk 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 Jh—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 Figure 3.—Complete healing in bones of same 
infant after 3 months’ exposure to sunlight 
Figure 2.—Fresh lime-salt deposits in bones of 
same infant after 18 days’ exposure to sunlight 
Figure 1.—Severe active rickets and osteo- 
porosis in an infant (S. M., 7 months old) 
who had lived in a cellar in San Juan and 
had never been exposed to sunlight INCIDENCE OF RICKETS 
47 
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 complete 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- 
mal” 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 
in 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 48 
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 roentgeno- 
graphic 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 roeritgen-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 tliis 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 craniota- 
bes49, 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 
49 Though a slight enlargement of the costochondral junctions theoretically may be the earliest sign of 
rickets and may 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 
by 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 diagnosis, the roentgenogram also shows evidence of rickets in the radius and ulna. INCIDENCE OF RICKETS 
49 
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, roentgeno- 
grapliic 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 then 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 life 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 dckets 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 
during 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 clinical 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 50 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Chart VI.—Clinical and roentgenographic evidence of rickets at specified age periods in children examined in Puerto Rico and in children examined in New Haven 
who had been given no specific treatment to prevent rickets 
25 to 36 
months 
14 to 24 
months 
New Haven Children (956) 
13 to 18 
months 
10 to 12 
months 
7 to 9 
months 
4to 6- 
months 
3 months 
andunder 
25 to 34 
months 
19 to 24 
months 
Puerto Rican children (564) 
13 to 13 
months 
Clinical diagnosis 
Roentgen-ray diagnosis 
lOto 12 
months 
7to 9 
months 
4to6 
months 
3 months 
andunder 
Per Cent INCIDENCE OF RICKETS 
51 
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 VI 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. 
Table 11—Clinical diagnosis of rickets in relation to roentgen-ray diagnosis; 
children examined in Puerto Rico 
Roentgen-ray 
diagnosis 
Children examined 
Total 
Clinical diagnosis of rickets 
No rickets 
Question- 
able 
diagnosis 
Rickets 
Total 
Very 
slight1 
Slight* 
Mod- 
er- 
ate 1 
Marked 
(!) 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Total 
584 
100 
400 
100 
134 
100 
50 
100 
13 
33 
3 
1 
No rickets 
509 
87 
354 
89 
114 
85 
41 
82 
10 
29 
2 
Doubtful diagnosis 
70 
12 
44 
11 
19 
14 
7 
14 
2 
4 
1 
Slight or very slight 
3 
1 
1 
(2) 
1 
1 
1 
2 
1 
1 
(J) 
1 
(J) 
1 
(J) 
1 
2 
1 
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 52 
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 roentgeno- 
graphic 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 residt 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 61 in 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 
6« Kramer, Benjamin, and Frederick F. Tisdall: A Simple Technique for the Determination of 
Calcium and Magnesium in Small Amounts of Serum. Journal of Biological Chemistry, vol. 47, 
no. 3, (August, 1921), p. 475. 
si Briggs, A. P.: A Modification of the Bell-Doisy Phosphate Method. Journal of Biological 
Chemistry, vol. 53, no. 1 (July, 1922), p. 13. 53 
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. 
52 Howland, John, M. D., and Beniamin Kramer, M. D.: Factors Concerned in the Calcification of Bone. 
Transactions of the American Pediatric Society, vol. 34, p. 204. Washington, 1922. 54 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Case 
No. 
Initials 
Age in 
months 
Calcium 
in blood 
(mg in 100 
c c serum) 
Phos- 
phorus in 
blood 
(mg in 100 
c c serum) 
Product 
of Ca 
content 
and Ph 
content 
Roentgen-ray 
diagnosis 
Clinical 
diagnosis 
Clinical signs of rickets present 
General 
nutri- 
tional 
condi- 
tion of 
child 
Feeding 
Calcium 
(gms) in 
mother’s 
daily 
diet i 
15 
P. B 
3 
10. 2 
5. 5 
56.10 
Normal 
0. 497 
104 
G. F 
4 
9.9 
5.6 
55. 44 
.328 
50 
D. F 
4 
10.6 
5.2 
55. 12 
Good.. 
Breast milk; cow’s 
milk, 1 pint a 
.760 
day. 
11 
R. D 
5 
11.1 
6.2 
68.82 
Fair... 
.168 
mula. 
175 
L. V 
5 
8. 6 
5.8 
49.88 
. 178 
69 
B. L„ 
6 
14.1 
6.4 
90.24 
Questionable—. 
Slight costochondral junctions; moderate 
frontal bosses; head asymmetrical. 
Good.. 
Breast milk; 
malted milk. 
. 771 
5 
C. L 
6 
12.3 
5.6 
68.88 
do 
No rickets 
Slight Harrison’s groove; slight flaring ribs.. 
Very 
good. 
.482 
72 
R. N 
6 
10.0 
4.4 
44.00 
Poor.. 
Condensed milk 
and rice water. 
.453 
48 
J. S - 
6 
11.1 
6.4 
71.04 
Breast milk; cow’s 
milk, dried or 
.241 
fresh. 
180 
S, M 
6 
10.0 
5.5 
55.00 
Good- 
milk, 1 quart a 
day. 
63 
M. B 
6 
10.1 
5.2 
52.52 
do 
.213 
88 
G, R 
8 
11.3 
5.2 
58.76 
.096 
54 
J, A 
8 
11.0 
5.9 
64.90 
do 
Questionable... 
Slight pot-belly; moderate knock-knees 
. 168 
mula. 
133 
R. N 
9 
10.8 
5.0 
54.00 
No rickets 
.226 
53 
0. F . 
• 9 
10.9 
4.8 
52.32 
Questionable... 
Slight Harrison’s groove; flaring ribs; slight 
pot-belly; muscle tone fair, relaxation. 
___do___ 
Breast milk; cow’s 
milk, Yi pint a 
day. 
62 
M. B____ 
10 
11.1 
6.4 
71.04 
do_ 
do _ 
Moderate parietal bosses; flaring ribs; pot- 
belly. 
___do___ 
Breast milk; cow’s 
.214 
milk, 1)4 pints a 
day. 
102 
0. G..._ 
13 
10.5 
5.6 
58.80 
Slight Harrison’s groove; pot-belly; flaring 
ribs; muscle tone poor, relaxation. 
__.do.__ 
.287 
ly; cow’s milk 
formula and bar- 
ley water. 
331 
C. R.... 
15 
8.3 
5.2 
43.16 
Slight epiphyses at wrist; moderate pot-belly. 
Good- 
Breast milk; cow’s 
milk, Y Pint a 
.147 
day in coffee. 
563 
R. R 
15 
11.0 
5. 6 
61. 60 
No rickets 
245 
G. A 
2 
11.7 
6.9 
69.03 
Doubtful 
Questionable... 
.446 
259 
H. M 
3 
11.2 
5.4 
60.48 
Very 
good. 
Breast milk; con- 
densed milk. 
.372 
6 
M. E 
4 
11.8 
5.6 
66.08 
do 
Questionable... 
Slight epiphyses at wrist 
-..do..- 
Breast milk 
.431 INCIDENCE OF RICKETS 
55 
99 
M. P.... 
4 
10.2 
5.1 
52.02 
No rickets 
Good.. 
Cow’s milk for- 
.417 
8 
H. R..__ 
5 
11.6 
4.5 
52.20 
do 
mula. 
Breast milk; cow’s 
milk. 
Breast milk 
.283 
46 
R. G 
6 
(») 
10.1 
4.7 
(J) 
41.41 
do 
.789 
59 
C. R 
6 
4.1 
.214 
13 
E. L 
7 
(*) 
11.4 
4.7 
« 
55.86 
. ..do 
Muscle tone fair, relaxation 
___do___ 
do 
.314 
43 
F. D 
5 
4.9 
Doubtful; osteopo- 
rosis +. 
Questionable... 
Slight costochondral junctions; slight pot- 
belly; muscle tone fair, relaxation; flaring 
ribs. 
Fair.. 
do 
.405 
234 
S. M 
7 
7.1 
4.3 
30.53 
Marked active 
rickets; osteopo- 
rosis +++. 
Very marked 
rickets; tet- 
any; pneumo- 
nia. 
Moderate costochondral junctions: moderate 
epiphyses: moderate Harrison’s groove; 
moderate bosses, frontal and parietal; slight 
craniotabes; moderate pot-belly; flaring 
ribs; muscle tone poor, marked relaxation. 
...do... 
Cow’s milk formu- 
la. 
.306 
144 
R. P 
16 
8.7 
3.6 
31.32 
No rickets; osteo- 
porosis +++. 
No rickets 
Flaring ribs; moderate pot-belly; muscle tone 
fair, marked relaxation. 
Poor.. 
Cow’s milk, 1 pint 
a day. 
.173 
89 
A. F 
22 
11.2 
5.7 
63.84 
do 
Questionable..- 
Moderate parietal bosses; moderate bowlegs; 
flaring ribs; moderate pot-belly; muscle 
tone poor, marked relaxation. 
...do... 
do. 
.218 
279 
N. P..... 
7 
10.2 
3.7 
37.74 
No rickets; osteo- 
porosis -f+++. 
No rickets 
Marked relaxation; muscle tone very poor 
Very 
poor. 
Cow’s milk formu- 
la. 
.158 
568 
M. R 
24 
9.45 
5.6 
52.92 
do 
Moderate parietal bosses; flaring ribs; mod- 
erate pot-belly; muscle tone poor, marked 
relaxation; slight funnel chest. 
...do... 
Breast milk; cow’s 
milk. 
184 
M. G._._ 
32 
11.6 
4.7 
54. 52 
do - 
do 
Moderate knock-knees; muscle tone poor, 
marked relaxation; slight pot-belly. 
Fair... 
Cow’s milk in cof- 
fee only. 
.069 
1 Foods recorded for sample day. See p. 82 for method of estimation of amounts of calcium in mothers’ diets and discussion of findings. 
a Not reported ADDITIONAL ROENTGENOGRAPHIC STUDIES OF THE 
RADIUS AND 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 iq Early Scurvy, American Journal of Diseases of Children, vol.30, 
no. 2 (August, 1925), pp. 174-188. A. H. 
N. P. 
Figure 4.—Osteoporosis in two children: A. II., 6 months old; and N. P., 7 months old Figure 5.—Osteoporosis in two children: A. P., 34 months old; and M. G., 32 months old 
A. P. 
M. Q. ROENTGENOGRAPHIC STUDIES OF RADIUS AND ULNA 
57 
spongiosa, however, of rarefied trabeculae or of poorly calcified trans- 
verse trabeculae showing in the roentgenograms as transverse fines 
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 
centers 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 fines 
were seen around the periphery of the centers of ossification in these 
atrophic bones, which, though suggestive of rings, were too delicate to 
Puerto Rican boys (293) 
Puerto Rican jfirla (264) 
Individual Puerto Rican children having osteoporosis (51) 
Kilograms 
We'igMr 
Morrth of a£e 
Chart VII.—Weights of individual Puerto Kican children having osteoporosis, compared with aver- 
age weights of all boys and girls examined in Puerto Rico 
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 S&uglingsskorbuts. Zeitschrift fur Kinderheiikunde, vol. 30 
(1923), pp. 279-285. 58 
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 VII and VIII, 
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 
Puerto Rican boys (292) 
Puerto Rican girls (265) 
Individual Puerto Rican children having osteoporosis (51) 
Centimeters 
Hei^kfc— 
Month, of age 
Chart VIII.—Heights of individual Puerto Rican children having osteoporosis, compared with aver- 
age heights of all boys and girls examined in Puerto Rico 
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 osteoporosis described. Experimentally it has been shown that 
diets deficient only in calcium result in osteoporosis of the bones.55 
Osteoporosis or atrophy of the bones was described by Alwens 56 as 
55 Korenchevsky, V.: The Aetiology and Pathology of Rickets from an Experimental Point of View, pp. 
63-73. Medical Research Council, Special Report Series No. 71. London, 1922. 
66 Alwens, Dr. [W.]: tiber die Beziehungen der Unterernahrung zur Osteoporose und Osteomalazie. 
Munchener Medizinische Wochenschrift, vol. 66, no. 38 (Sept. 19,1919), pp. 1071-1075. ROENTGENOGRAPHIC STUDIES OF RADIUS AND ULNA 
59 
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 found 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. 
INCIDENCE OF TRANSVERSE LINES IN LONG BONES 
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.58 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 
hues 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 
57 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. 
58 Eliot, Martha M., Susan P. Souther, and E. A. Park: Transverse Lines in X-Ray Plates of the Long 
Bones of Children. Bulletin of the Johns Hopkins Hospital, vol. 41, no. 6 (December, 1927), pp. 364-388. 
59 Asada, Tameyoshi: Uber die Entstehung und pathologische Bedeutung der im Rontgenbild dea 
Rohrenknochens am Diaphysenende zum Vorschein Kommenden parallelen Querlinienhildung. Mit- 
teilungen aus der Medizinischen Fakultat der Kaiserlichen Kyushu-Universit&t, vol. 9. no. 1 (1924), pp, 
43-95. 
160326°—33 6 60 
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 lires in their 
bones; of the 332 children who seemed to be well nourished, only one- 
fourth showed them. It is also of interest that of the 59 children who 
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. 
Table 12.— Transverse lines in radius and ulna of children examined at specified 
age periods in Puerto Rico and in New Haven, Conn. 
Age 1 at examination 
Children examined in 
Puerto Rico 
Children examined in New 
Haven 
Total 
Lines present 
Lines 
absent 
Total 
Lines present 
Lines 
absent 
Number 
Per cent 
Number 
Per cent 
Total 
584 
200 
34 
384 
1,186 
339 
29 
847 
Under 8 months .. 
196 
34 
17 
162 
288 
58 
20 
230 
8 to 13 months 
146 
51 
35 
95 
296 
54 
18 
242 
14 to 34 months __ 
222 
105 
47 
117 
348 
105 
30 
243 
254 
122 
48 
132 
Not reported... ... ... 
20 
10 
(2) 
10 
'Age is given as of nearest month; that is, “under 8 months” is actually under 7 months and 16 days, 
“8 to 13 months” is from 7 months and 16 days to 13 months and 15 days, inclusive, and so on. 
2 Per cent not shown because number of children was less than 50. SOCIO-ECONOMIC CONDITIONS AS FACTORS IN THE 
HEALTH OF PUERTO RICAN CHILDREN 
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 62 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
households were visited that included various persons not members 
of the family60—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 “family,” 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 half- 
brothers and half-sisters hnd other relatives are not considered ruembers of the family even though they 
were living in the same household. SOCIO-ECONOMIC CONDITIONS 
63 
Table 13.— Number in child’s own family and presence in household of others than 
members of family; families visited in Puerto Rico 
Number in family 
Families visited 
Total 
Group in household 
Number 
Per cent 
distribu- 
tion 
Family only 
Family and others 
Group 
not 
reported 
Number 
Per cent 
distribu- 
tion 
Number 
Per cent 
distribu- 
tion 
Total 
506 
262 
241 
3 
Number reported 
503 
100 
262 
100 
241 
100 
1 
4 
1 
4 
2 
2 
55 
11 
11 
4 
44 
18 
3 
178 
35 
98 
37 
80 
33 
4 
105 
21 
56 
21 
49 
20 
5 
71 
14 
40 
15 
31 
13 
6 
35 
7 
19 
7 
16 
7 
7 
29 
6 
19 
7 
10 
4 
8 
14 
3 
11 
4 
3 
1 
9 
6 
1 
4 
2 
2 
1 
10 or more. _ 
6 
1 
4 
2 
2 
1 
Number not reported 
3 
3 
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 w'as 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 
FAMILY INCOME 
81 The “income group” consisted of the persons who were supported by the total family income, whether 
they were living in the household as members of the family or boarders or whether they were living away 
from home. 
98 For 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. 64 
PEVELOPMENT OE 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. 
Table 14.— Annual income of families visited in San Juan and in Ponce, P. R. 
Families visited 
Annual income 1 
Total 
San Juan 
Ponce 
Number 
Per cent 
distribu- 
tion 
Number 
Per cent 
distribu- 
tion 
Number 
Per cent 
distribu- 
tion 
Total 
506 
328 
178 
378 
100 
247 
100 
131 
100 
14 
4 
7 
3 
5 
36 
10 
18 
7 
18 
14 
134 
35 
77 
31 
57 
44 
$400, less than $600 
85 
22 
61 
25 
24 
18 
47 
12 
34 
14 
13 
10 
23 
6 
16 
6 
7 
5 
$1,000 or more..- 
39 
10 
34 
14 
5 
4 
128 
81 
47 
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. 
Table 15.—Annual income of family and number of persons maintained on this 
income; families visited in Puerto Rico 
Annual income1 
Families visited 
Total 
Number maintained on family income 
3 or fewer 
4 
5 
6 or 7 
8 or more 
Not re- 
ported 
Total 
506 
146 
95 
76 
77 
37 
75 
50 
23 
12 
6 
8 
1 
134 
46 
30 
27 
22 
9 
85 
23 
23 
18 
13 
8 
47 
10 
9 
12 
12 
4 
23 
7 
5 
4 
5 
2 
39 
15 
7 
3 
5 
9 
Amount not reported. 
128 
22 
9 
6 
12 
5 
74 
1 Aggregate amount contributed to family support. SOCIO-ECONOMIC CONDITIONS 
65 
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. 
Table 16.—Persons contributing to family support and person furnishing chief 
support; families visited in Puerto Rico 
Families visited 
Persons contributing to family support 
Person furnishing chief support 
Total 
Father 
Mother 
Other 1 
Not re- 
ported 
506 
399 
51 
24 
32 
258 
258 
116 
97 
9 
10 
37 
30 
3 
4 
20 
14 
2 
3 
1 
16 
16 
39 
24 
6 
9 
12 
12 
8 
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 
63 For average daily earnings in various industries see Puerto Rico; what it produces and what it buys, 
p. 11 (Trade Information Bulletin No. 785, Bureau of Foreign and Domestic Commerce, U. S. Department 
of Commerce, Washington, 1932). 66 
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 ambition 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 year—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 SOCIO-ECONOMIC CONDITIONS 
67 
families in which the father was unemployed for various lengths of 
time, 86 families (61 per cent) had incomes below $400. 
Table 17.—Annual income of family in relation to the father's unemployment; 
families visited in Puerto Rico 
Annual income* 
Families visited 
Total 
Unemployment of father 
Father 
dead, not 
living 
with fam- 
ily, or not 
contrib- 
uting 
None 
Less 
than 3 
months 
3 months, 
less 
than 6 
6 months, 
less 
than 9 
9 months 
or more 
Not re- 
ported 
Total 
506 
230 
46 
57 
50 
14 
54 
55 
Less than $200 
50 
6 
2 
6 
16 
1 
6 
13 
$200, less than $400 
134 
56 
11 
26 
17 
7 
13 
4 
$400, less than $000 
85 
45 
17 
15 
4 
i 
1 
2 
$600, less than $800 . _ 
47 
40 
5 
2 
$800, less than $1, 000_. 
23 
17 
3 
1 
1 
1 
$1,000 or more 
39 
30 
4 
1 
2 
2 
Amount not reported 
128 
36 
4 
6 
12 
5 
32 
33 
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 Rican 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. 68 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table 18.—Annual per capita income and number of persons maintained on annual 
income; families visited in Puerto Rico 
Annual per capita income 1 
Families visited 
Total 
Number maintained on annual income 1 
Num- 
ber- 
Per 
cent 
dis- 
tribu- 
tion 
3 or less 
4 
5 
6 or 7 
8 or 
more2 
Not 
re- 
port- 
ed 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Num- 
ber 
Per 
cent 
dis- 
tribu- 
tion 
Total.-- 
506 
146 
95 
76 
77 
37 
75 
Income reported- 
377 
100 
124 
100 
86 
100 
70 
100 
65 
100 
32 
Less than $50 
66 
18 
10 
8 
12 
14 
13 
19 
21 
32 
10 
$50, less than $100 
130 
34 
33 
27 
30 
35 
28 
40 
26 
40 
13 
$100, less than $150 
91 
24 
27 
22 
23 
27 
20 
29 
14 
22 
7 
$150, less than $200 
38 
10 
21 
17 
9 
10 
6 
9 
1 
2 
1 
$200 or more _ 
52 
14 
33 
27 
12 
14 
3 
4 
3 
5 
1 
Income not reported 
129 
22 
9 
6 
12 
5 
75 
1 Aggregate amount contributed to family support. 
2 Per cent distribution not shown because number of families was less than 50. 
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 wrell- 
nourished 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 
writli 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 wdiich 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 Pico, 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. 69 
SOCIO-ECONOMIC CONDITIONS 
HOUSING 
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. 
Table 19.—Number of persons in household and number of rooms (exclusive of 
kitchen) occupied by household; families visited in Puerto Rico 
Families visited 
Number in household 
Number of rooms occupied by household 
Total 
1 
2 
3 
4 or more 
Not re- 
ported 
.506 
165 
172 
86 
76 
7 
2 
11 
7 
3 
1 
3 ... . 
110 
61 
35 
8 
6 
4 _ ... 
92 
37 
41 
9 
5 
5 . . 
89 
21 
40 
16 
11 
1 
6 
65 
18 
19 
12 
5 
1 
60 
8 
17 
16 
17 
2 
8 
35 
4 
11 
7 
13 
9 
19 
3 
4 
4 
8 
10 
17 
6 
1 
7 
3 
11 
5 
1 
2 
2 
10 
4 
6 
Not reported 
3 
3 70 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table 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 
Number of rooms 
Families visited 
Total 
Oroup in household 
Oroup re- 
ported 
Family only 
Family and others 
Group 
not re- 
ported 
Number 
Per cent 
Number 
Per cent 
Total 
506 
503 
262 
52 
241 
48 
3 
1 
165 
165 
116 
70 
49 
30 
2 
172 
172 
101 
59 
71 
41 
3 
86 
86 
26 
30 
60 
70 
76 
76 
19 
25 
57 
75 
Not reported 
7 
4 
4 
3 
Not 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 majoiity 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. 
FURNISHINGS 
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 barrios, it was frequently a small 
plot of ground surrounded by several frame houses or huts. Of the 
dwellings reported to have no patio or yard, a majority were either 
huts crowded together or city tenements. SOCIO-ECONOMIC CONDITIONS 
71 
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 ranchons, or row houses, predominated. In 
the poorer districts on the outskirts of San Juan, such as La Perla, 
Puerta do Tierra, Tras Tallieres, 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 percent) lived in 
frame houses or ranchons 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. 
San Juan proper.—The 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 72 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
there were dark apartments that opened only on a hall; hut, 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. 
Outskirts oj San Juan.—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, ranchons, 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 from 
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 they 
were 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 ranchons, or rows of one-story houses, were found in San- 
turce. 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 ranchons 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. 73 
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 built 
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, Puerta 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 hill 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 all 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 
thrown 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 lower,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. 
Ponce.—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 74 
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 J)f 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 
G5 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 ranchons, 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. 
Table 21.—Children living in houses having estimated number of hours of sunlight 
a day; families visited in Puerto Rico 
Estimated number of hours a day of sun- 
light in house 
Children in families 
visited 
Number 
Per cent 
distri- 
bution 
556 
Hours reported .. 
553 
100 
360 
126 
26 
27 
14 
3 
65 
23 
5 
5 
3 
Less than 2 hours... ... . .. 
None in house, but sun easily acces- 
None... . SOCIO-ECONOMIC CONDITIONS 
75 
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. 
160326°—33 6 DIETS OF PUERTO RICANS64 
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. 
64 For further information on Puerto Rican diets see the series of bulletins entitled Tropical Foods, by 
Elsie Mae Willsey, published by the University of Puerto Rico, Rio Piedras. (First bulletin of series dated 
1925.) DIETS 
77 
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 78 
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 Ilican family. 
FAMILY DIETS 
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 three- 
fourths of those with the larger incomes bought a liter or more a day. 
MILK 
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-tliirds 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. 
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 
EGGS 
65 A. liter is approximately the equivalent of 1 quart. DIETS 
79 
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. 
Meat, or more often dried codfish, wTas 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 wdth incomes of less than $400 
were reported to use meat somewhat less frequently than those with 
larger incomes. 
MEAT AND FISH 
FRUIT 
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 wdth 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. Twenty- 
one 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 wdth 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 80 
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. 
DIETS OF MOTHERS 
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 day 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 
for lactation periods completed at varying lengths of time previous to 
the interview. PIETS 
81 
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. 
ANALYSIS OF 112 SAMPLE DIETS 
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 wrere 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, 82 
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. 
Table 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 bones 
Roentgen-ray diagnosis of con- 
dition of child’s bones 
Total 
moth- 
ers 
Calories 
Protein (gms) 
Calcium (gms) 
Aver- 
age 
Maxi- 
mum 
Mini- 
mum 
Aver- 
age 
Maxi- 
mum 
Mini- 
mum 
Aver- 
age 
Maxi- 
mum 
Mini- 
mum 
Total 
112 
870 
1,441 
220 
34.11 
66.08 
12.02 
0.283 
0.789 
0.050 
Normal 
56 
880 
1,441 
390 
34. 67 
66.08 
15.38 
.286 
.760 
.072 
Doubtful 
28 
890 
1,318 
357 
34. 62 
51.51 
12.18 
.344 
.789 
.064 
Osteoporotic 
28 
831 
1,075 
220 
32.48 
46. 57 
12.02 
.217 
.524 
.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 estimates 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 best 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,67 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— 
66 Sherman, Henry C.: Chemistry of Food and Nutrition, pp. 221-222. New York, 1932. 
67 Ibid., p. 511. DIETS 
83 
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,68 is 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, 
88 Sherman, Henry C.: Chemistry of Food and Nutrition, p. 287. 
89 Id. 
70 Ibid., p. 512. 
»Id. 84 
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 one- 
quarter 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 
72 Calculated from figures in Chemistry of Food and Nutrition, by Henry C. Sherman, (New York, 1932), 
pp. 195-19G. ' 
73 Ibid., p. 184. DIETS 
85 
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.) 
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: 
CLASSIFICATION OF DIETS OF ALL MOTHERS INTERVIEWED 
Diet grade 
Calcium (grams) 
Protein (grams) 
Combinations of food falling within range of cal- 
cium and protein 
Aver- 
age 
Maxi- 
mum 
Mini- 
mum 
Aver- 
age 
Maxi- 
mum 
Mini- 
mum 
1. Good 
0.66 
0.87 
0. 46 
50 
65 
36 
Milk, 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,7< 2 or more times a 
a day. 
Rice, potato, or bread 2 or more times a day. 
Meat, 5 or more times a week. 
.Oranges, 3 or more times a week. 
2. Fair 
0.37 
0.55 
0.18 
41 
53 
28 
'Milk, 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. 
3. Poor 
0.20 
0.28 
0.12 
37 
50 
24 
Milk, none.75 
Eggs, or leafy vegetables, 1 or more times a week. 
Beans or tuberous vegetables 7 or more times a 
week. 
1 Rice, potatoes, or bread, 2 or more times a day. 
Meat, 3 to 7 times a week. 
(Oranges, 1 to 4 times a week. 
4. Very poor. 
0.15 
0.24 
0.06 
18 
28 
7 
Milk, 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 
week. 
Rice, potatoes, or bread, 1 or 2 times a day. 
Meat, once a week. 
.Oranges 1 to 4 times a week. 
5. Exceeding- 
ly poor. 
Jo. 08 
0.14 
0.02 
15 
26 
4 
Milk, none. 
Eggs or leafy vegetables, none, or once a week. 
Beans or tuberous vegetables 3 or more times a 
week. 
Rice, potatoes, or bread 1 or 2 times a day. 
Meat, none, or once a week. 
.Oranges, none, or 1 to 2 times a week. 
74 Tuberous vegetables include, as well as potatoes, such native vegetables as yams and yautias. At the 
time when the calcium content of those 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 tho 
potato and the turnip, both of which the yautia resembles. Through a subsequent analysis the yautia has 
been found to be actually much lower in calcium than the estimate used in the present study. (Signifi- 
cance of Mineral Metabolism, I. Preliminary Report on Calcium and Phosphorus Content of Some Puerto 
Rican Food Materials, by D. H. Cook and Trinita Rivera, in the Puerto Rican Review of Public Health 
and Tropical Medicine, vol. 4, no. 2, pp. 65-69, August, 1928.) The amounts of calcium in these diets, 
therefore, are as a whole somewhat overestimated. 
75 Though the diets in the third grade contain no milk, they are slightly better than those in either the 
fourth or the fifth, because they include larger amounts of calcium and protein containing foods other than 
milk. 86 
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 small 
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.) 
Table 23.—Calcium content of mother’s diet on day 'previous to interview, and 
diet grade; families visited in Puerto Rico 
Families visited 
Calcium content (gms) of 
mother’s diet1 
Diet grade 
Total 
Per cent 
distri- 
bution 
Average 
Maximum 
Minimum 
506 
484 
100 
0.287 
1.345 
0.048 
Good 
48 
10 
.614 
1.345 
.456 
Fair 
194 
40 
. 338 
. 638 
. 183 
93 
19 
. 211 
. 328 
. 127 
2 107 
22 
. 180 
. 238 
.067 
2 42 
9 
. 104 
. 145 
.048 
Grade not reported. _ ________ . _ 
22 
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 DIETS 
87 
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. 
Table 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 lactat- 
ing on day previ- 
ous to interview 
Diet grade 
Number 
Per cent 
distribu- 
tion 
Number 
Per cent 
distribu- 
tion 
Number 
Per cent 
distribu- 
tion 
i 556 
2 526 
3 213 
531 
100 
505 
100 
197 
100 
Good. . 
98 
19 
60 
12 
15 
8 
205 
39 
208 
41 
76 
39 
82 
15 
89 
18 
44 
22 
109 
21 
109 
22 
45 
23 
37 
7 
39 
8 
17 
9 
Grade not reported 
25 
21 
16 
1 Of the 506 mothers in families visited, 50 had two children examined, and information was sought with 
regard to these mothers’ diets during both pregnancies. 
2 Of the 506 mothers in families visited, 50 had two children examined; of the 556 children examined, 30 
were never breast fed. 
s Of the 506 mothers in families visited, 293 were lactating on the day previous to interview. 88 
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. 
Table 25.—Mother's diet grade for day previous to interview and annual per capita 
income of family; families visited in Puerto Rico 
Families visited 
Diet grade 
Annual per capita income 
Total 
Less than 
$50 
$50,less 
than $100 
$100,less 
than $150 
$150, less 
than $200 
$200 or 
more 
Not 
reported 
506 
66 
130 
91 
38 
52 
129 
48 
1 
7 
6 
7 
19 
8 
194 
14 
50 
51 
18 
20 
41 
93 
19 
27 
14 
4 
4 
25 
107 
18 
32 
16 
8 
6 
27 
42 
11 
10 
3 
1 
2 
15 
Not reported 
22 
3 
4 
1 
1 
13 
METHOD OF COLLECTING INFORMATION ON DIETS 
DIETS OF CHILDREN 
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 lias, therefore, 
been considered as only approximate. DIETS 
89 
The exact quantities of food eaten by the child daily wrere 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) wrere used and the time when they wTere 
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. 
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 wTould receive four grades—one for each of the 3-month periods 
of the first year. 
The second year, whether complete or incomplete, w*as treated as 
a single period and given one grade; the third year wTas treated in 
the same wray. 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 
2}a 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 tune 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. 
METHOD OF GRADING DIETS 
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 milk 76 
BASIS FOB GRADES OF DIETS AT DIFFERENT AGES 
n Infant Care, p. 78, U. 8. Children’s Bureau Publication No. 8. Washington, 1933. 90 
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 Dl 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,77 was 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 (lj4 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 
77 When a child was partly breast fed and partly artificially fed, it was not possible to tell exactly whut 
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 milk 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 milk feeding in that period. 91 
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 modern 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 y2 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: 
DIETS 
Grade 
Cereal 
Eggs and vegetables 
Satisfactory 
Once or twice a day 
Each food 2 to 6 times a week. 
Either food 2 to 6 times a week 
or 
One food once a week, the other 
1 to 6 times a week. 
Either food once a week or 
neither given. 
Borderline.- 
Once or twice a day 
Poor 
Once or twice a day 
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 92 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
The grades for the use of solid foods during the second year were 
given as follows: 
Grade 
Cereal 
Eggs and vegetables 
How often eaten 
How often eaten 
At what age begun 
.2 . 
1 to 2 times a day 
Each food 5 to 7 times a week 
Both foods at or before 12 months, 
jry sat 
factory 
or 
One food at 12 months, the other at 13 
to 15 months. 
> 
1 to 2 times a day 
Each food 3 to 7 times a week 
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. 
o 
1 
11 to 2 times a day 
Each food 3 to 7 times a week 
One food at 12 months, the other 13 to 
15 months. 
03 
GO 
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. 
© 
I 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, 
or 
Both foods at 16 months. 
© 
T3 
lx 
O 
« 
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 
One food at or after 16 months, the other 
lx 
o 
© 
other 0 to 2 times. 
before 16 months, 
or 
Both foods at 16 months. 
te 
0 to 2 times a day 
Neither food in diet 
Both foods at or before 12 months, 
a 
or 
or 
. 
Either food 1 to 2 times a week. 
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 
One food at or after 16 months, the other 
>• 
other 0 to 2 times. 
before 16 months. 
c 
0 to 2 times a day 
Neither food in diet 
One food at or after 16 months, the other 
cs 
or 
before 16 months, 
fl 
d 
»- 
Either food 1 to 2 times a week. 
or 
Both foods at 16 months, 
or 
One food at or after 16 months, the other 
before 16 months. 
> 
0 to 2 times a day 
(7S) 
(n). 
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 IX79 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 
7» For data upon which Chart IX is based, see Appendix Tables Dl, D2, D3, D4, and D5, DIETS 
93 
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. 
cDict Grades 
Per 
cent 
VeC/5ood 
Good 
Fair' 
Poor 
Very poor 
354 children 
Period I 
(under 4 months) 
469 children 
Period 2 
(4-6 months) 
3fi>7 children 
Period 3 
(7-9 months) 
293 children 
Period4- 
(10-12 months) 
203 children. 
Period 5 
(13-24 months) 
Chart IX.—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 
first 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 
continued for about two-fifths of the children. In the period between 94 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
10 and 12 months, only about half the children received what w~as 
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 uncommon. 
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 DIETS 
95 
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 D1 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 of 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 McCollum, Elmer V., and Nina Simmonds: The Newer Knowledge of Nutrition, pp. 410-432. New 
York, 1929. 96 
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 
Breast feeding 
only 
141 children 
Per cent 
Mixed feeding 
(breast milk and 
cow’s milk) 
105 children 
Cow’s mi IK 
•feeding' only 
85 children 
Subcutaneous fat 
Chart X.—Subcutaneous fat in relation to type of milk feeding at time of examination; Puerto Rican 
children 13 months of age and under 
Good 
Fair 
Poor and very poor 
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 
and 
<£ood diet 
164 children 
Per cent 
Fair diet 
63 children 
Poor and very 
poor diet 
104 children 
Subcutaneous Fat 
Chart XI.—Subcutaneous fat in relation to diet grades at time of examination; Puerto Rican children 
13 months of age and under. 
Good 
Fair 
Poor and very poor 
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 XI. From Chart X there would seem 
to be little doubt that during the first year of life children who were DIETS 
97 
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 XI. 
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. 
ECONOMIC CONDITION OF FAMILY IN RELATION TO CHILD'S DIET 
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. 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 percent) 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 SUMMARY 
99 
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 oHntermittent 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. DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Appendix A.—HEIGHT, WEIGHT, AND HEAD CIRCUMFERENCE FOR AGE: CHILDREN EXAMINED IN 
PUERTO RICO 
Age 1 at examination 
Boys 1 to 33 months of age 
Total 
Height (centimeters) reported 
Height not reported 
Total reported 
i 
47, less than 49 
49, less than 51 
51, less than 53 
53, less than 55 
55, less than 57 
57, less than 59 
59, less than 61 
61, less than 63 
63, less than 65 
65, less than 67 
67, less than 69 
69, less than 71 
71, less than 73 
73, less than 75 
75, less than 77 
77, less than 79 
79, less than 81 
81, less than 83 
83, less than 85 
85, less than 87 
87, less than 89 
89, less than 91 
91, less than 93 
93, less than 95 
Total.. 
307 
304 
1 
2 
11 
10 
7 
15 
26 
24 
20 
28 
28 
27 
21 
23 
15 
14 
11 
7 
4 
5 
3 
2 
3 
Age reported 
295 
292 
1 
2 
11 
10 
7 
14 
25 
23 
20 
26 
28 
26 
18 
23 
14 
14 
11 
7 
4 
3 
3 
2 
3 
1 month 
9 
8 
1 
1 
5 
1 
1 
2 months ... 
11 
11 
1 
2 
5 
1 
2 
3 months 
16 
16 
4 
3 
1 
2 
4 
2 
4 months . 
19 
19 
2 
1 
8 
4 
3 
1 
5 months 
18 
18 
2 
3 
5 
5 
1 
1 
1 
6 months 
16 
16 
1 
2 
3 
4 
3 
1 
2 
7 months 
13 
13 
1 
1 
3 
1 
4 
2 
1 
8 months 
17 
17 
1 
1 
2 
3 
8 
2 
9 months 
13 
13 
1 
2 
5 
4 
1 
10 months 
15 
15 
1 
2 
5 
4 
2 
1 
11 months .. 
17 
17 
1 
1 
1 
1 
3 
3 
12 months 
12 
12 
1 
1 
1 
4 
4 
1 
13 months 
9 
9 
3 
3 
1 
2 
14 months. 
8 
8 
1 
2 
2 
1 
1 
1 
15 months 
8 
8 
2 
1 
1 
3 
1 
16 months 
11 
11 
4 
2 
2 
2 
1 
17 months 
12 
11 
1 
2 
1 
5 
2 
1 
18 months 
6 
6 
1 
3 
2 
19 months ... 
10 
10 
1 
4 
2 
1 
1 
1 
20 months 
7 
7 
3 
2 
1 
1 
21 months ... 
4 
4 
1 
2 
1 
22 months 
5 
5 
1 
1 
2 
1 
23 months 
7 
7 
1 
2 
1 
1 
1 
1 
24 months 
4 
3 
1 
1 
1 
1 
25 to 27 months 
14 
14 
1 
4 
4 
1 
1 
1 
1 
1 
28 to 30 months . 
7 
7 
2 
2 
1 
1 
1 
31 to 33 months 
7 
7 
1 
1 
1 
1 
1 
2 
Age not reported 
12 
12 
1 
1 
1 
2 
1 
3 
1 
2 
Table Al.—Height and age; hoys 1 to S3 months of age examined in Puerto Rico 
1 Age is given as of nearest month; that is, “1 month” is from 16 days to 1 month and 15 days, inclusive, and so on. APPENDIX A 
101 
Age1 at examination 
Girls 1 to 33 months of age 
3 
o 
b 
Height (centimeters) reported 
"O 
<V 
Total reported 
45, less than 47 
47, less than 49 
49, less than 51 
51, less than 53 
53, less than 55 
55, less than 57 
57, less than 59 
59, less than 61 
61, less than 63 
63, less than 65 
65, less than 67 
67, less than 69 
69, less than 71 
71, less than 73 
73, less than 75 
75, less than 77 
77, less than 79 
79, less than 81 
81, less than 83 
83, less than 85 
85, less than 87 
87, less than 89 
89, less than 91 
91, less than 93 
Oi 
<X> 
Lt 
o 
a 
-*-9 
-a 
'53 
M 
Total 
Age reported 
1 month.. 
2 months 
3 months... 
4 months 
274 
266 
272 
1 
1 
2 
1 
13 
9 
13 
17 
20 
14 
26 
13 
18 
19 
25 
23 
17 
10 
14 
3 
9 
2 
1 
1 
2 
265 
1 
1 
2 
1 
13 
9 
13 
17 
20 
14 
26 
13 
17 
17 
25 
22 
16 
8 
14 
3 
9 
2 
1 
1 
1 
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 
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 
7 
1 
1 
1 
1 
1 
4 
6 
1 
3 
2 
2 
4 
3 
2 
2 
2 
4 
3 
5 
2 
1 
1 
4 
6 
3 
3 
1 
1 
1 
1 
2 
1 
3 
5 
4 
2 
3 
2 
2 
1 
5 months.. 
6 months.. 
7 months 
8 months.. 
1 
1 
1 
i 
1 
1 
2 
3 
5 
1 
2 
1 
3 
1 
1 
1 
1 
2 
1 
9 months 
2 
2 
1 
3 
2 
1 
1 
1 
10 months 
2 
1 
1 
2 
3 
5 
11 months 
1 
2 
12 months 
13 months. 
2 
2 
1 
1 
2 
1 
1 
1 
1 
1 
3 
4 
2 
2 
1 
3 
1 
14 months 
15 months.. 
4 
1 
1 
1 
16 months 
17 months.. 
l 
1 
18 months 
1 
1 
1 
1 
1 
19 months 
20 months 
2 
6 
1 
21 months 
1 
1 
1 
1 
1 
1 
2 
22 months 
2 
1 
3 
1 
6 
3 
23 months. 
3 
3 
4 
1 
1 
1 
1 
1 
5 
1 
1 
24 months 
25 to 27 months 
3 
1 
1 
1 
l 
28 to 30 months 
31 to 33 months 
1 
1 
1 
2 
1 
Age not reported 
i 
2 
— 
I 
Table A2.—Height and age; girls 1 to 33 months of age examined in Puerto Rico 
i Age is given as of nearest month; that is, “1 month” is from 16 days to 1 month and 15 days, inclusive, and so on. 102 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table A3.—Average height at each month of age; hoys and girls 1 to S3 months of 
age examined in Puerto Rico 
Age 1 at examination 
Average height (centimeters) 
Boys 
Girls 
Observed 
Smoothed 
Observed 
Smoothed 
1 month __ _ 
53.3 
52. 98 
52.8 
51.94 
2 months 
56. 2 
56.83 
55.4 
55.91 
3 months _ _ ... 
58. 6 
59. 31 
.58.9 
58. 44 
4 months 
62.8 
61. 23 
61. 2 
60.38 
5 months - - 
62.9 
62. 84 
60.7 
62.00 
(i months 
64.0 
64. 26 
61. 5 
63.41 
7 months 
64. 5 
65. 54 
64.0 
64. 69 
8 months. _ _ 
66. 6 
66. 73 
64.5 
65.86 
9 months 
67.8 
67.84 
68.0 
66.96 
10 months 
68.7 
68. 89 
67.5 
67.99 
11 months ■ 
69.8 
69. 89 
68.8 
68.98 
12 months 
72.0 
70. 85 
69.6 
69.92 
13 months 
69.8 
71.78 
72.0 
70.82 
14months_. _ 
75.3 
72. 69 
73.4 
71.70 
1,5 months. 
75.5 
73. 56 
73.1 
72.55 
16 months ... ... ... _ _ 
75. 5 
74. 42 
75. 2 
73.38 
17 months 
74.9 
75. 25 
76.0 
74. 19 
18 months 
76.0 
76. 08 
73.7 
74.99 
19 months . 
77.0 
76.88 
74.4 
75.76 
20 months 
74.9 
77. 67 
79.0 
76.53 
21 months _ 
75.5 
78.46 
77.6 
77.28 
22 months __ 
78.4 
79.22 
75.6 
78.02 
23 months 
80.6 
79. 98 
78.4 
78. 75 
24 months _ 
84.0 
80. 74 
78.3 
79. 47 
25 months. . . 
I 
f 81.48 
] 
1 80.18 
26 months 
1 80.4 
{ 82.21 
} 81.1 
\ 80.88 
27 months 
1 
82.94 
1 
81.58 
28 months 
1 
( 83.66 
1 
82.27 
29 months .. __ _ 
\ 85.1 
-j 84.38 
\ 84.0 
\ 82.95 
30 months ... _ 
1 
85.09 
1 
83. 63 
31 months 
I 
85. 79 
1 
( 84.30 
32months... .... 
1 86.9 
\ 86.49 
84.0 
-1 84.96 
33 months ... 
l 87.19 
1 
l 85.63 
Equations of curves used in smoothing: 
Boys, y=52.429051+0.548181:r+10.976648 log. x. 
Girls, y = 51.434797+0.507865z+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 month” is from 16 days to 1 month and 15 days, inclusive. 
Note that from 26 months onward each observed average includes a 3-month interval. APPENDIX A 
103 
Table A4.—Weight and age; boys 1 to 33 months of age examined in Puerto Rico 
Boys 1 to 33 months of age 
Weight (kilograms) reported 2 
Age 1 at examination 
Total 
0 
Total 
re- 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
bn a 
ported 
© £ 
£ 
Total ... 
307 
305 
5 
16 
24 
30 
55 
49 
48 
33 
26 
9 
7 
3 
2 
Age reported 
295 
293 
5 
16 
23 
28 
53 
47 
46 
32 
26 
9 
6 
2 
2 
1 month 
9 
9 
2 
5 
2 
2 months ... 
11 
11 
2 
7 
2 
3 months . 
16 
16 
1 
7 
4 
4 
4 months ... 
19 
19 
1 
3 
5 
5 
4 
1 
5 months 
18 
18 
2 
1 
4 
9 
1 
1 
6 months ... 
16 
16 
1 
2 
3 
7 
1 
2 
7 months.. 
13 
13 
1 
4 
4 
3 
1 
8 months.. ... .... 
17 
16 
1 
1 
2 
6 
1 
3 
2 
1 
9 months 
13 
13 
1 
1 
1 
4 
5 
1 
10 months 
15 
15 
1 
5 
6 
2 
1 
11 months 
17 
17 
1 
1 
2 
9 
3 
1 
12 months 
12 
12 
1 
4 
5 
1 
1 
13 months 
9 
9 
1 
3 
3 
1 
1 
14 months 
8 
8 
2 
4 
1 
15 months 
8 
8 
1 
1 
1 
9 
3 
16 months _ ... . . 
11 
11 
1 
1 
3 
3 
3 
17 months 
12 
12 
1 
6 
5 
18 months . . . __ __ 
6 
6 
1 
3 
1 
1 
19 months... 
10 
10 
1 
4 
3 
1 
1 
20 months 
7 
7 
1 
2 
2 
2 
21 months 
4 
4 
1 
3 
22 months 
5 
5 
1 
2 
2 
23 months 
7 
1 
3 
0 
1 
24 months 
4 
4 
2 
2 
25 to 27 months 
14 
14 
1 
2 
1 
2 
1 
1 
28 to 30 months 
7 
7 
3 
3 
31 to 33 months 
7 
6 
1 
2 
Age not reported 
12 
12 
1 
2 
2 
2 
2 
1 
1 
1 
1 Age is given as of nearest month; that is, “1 month” 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. 104 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table A5.—Weight and age; girls 1 to S3 months of age examined in Puerto Rico 
Girls 1 to 33 months of age 
Age 1 at examination 
Weight (kilograms) reported 1 
o 
Total 
Total 
re- 
ported 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
Weight 
reporte 
274 
272 
1 
2 
21 
17 
42 
45 
36 
38 
40 
19 
6 
1 
2 
266 
264 
1 
2 
21 
17 
41 
43 
36 
36 
37 
19 
6 
4 
1 
2 
1 month... _. 
8 
8 
7 
1 
2 months _ 
17 
17 
1 
8 
4 
4 
3 months 
11 
11 
4 
7 
4 months 
12 
12 
3 
5 
3 
1 
5 months.- __ 
18 
18 
2 
2 
10 
4 
6 months-- - _ 
15 
15 
4 
1 
2 
6 
2 
7 months 
12 
12 
1 
2 
6 
2 
1 
8 months 
13 
13 
1 
1 
4 
3 
2 
2 
9 months 
7 
7 
1 
1 
6 
10 months 
11 
11 
2 
4 
3 
1 
1 
11 months 
10 
9 
2 
4 
3 
i 
12 months 
9 
9 
4 
2 
2 
1 
13 months 
13 
13 
1 
1 
6 
4 
1 
14 months 
7 
7 
2 
3 
2 
15 months-.- 
13 
13 
1 
4 
2 
6 
16 months 
5 
5 
1 
4 
17 months 
5 
5 
2 
2 
1 
18 months 
7 
7 
1 
1 
1 
4 
19 months..- _ 
5 
5 
1 
3 
1 
20 months 
4 
4 
2 
1 
1 
21 months _ 
9 
9 
1 
1 
2 
4 
1 
22 months 
5 
5 
1 
2 
i 
1 
23 months _ 
9 
9 
2 
1 
3 
1 
1 
i 
24 months 
7 
6 
1 
1 
2 
2 
i 
25 to 27 months.-. 
24 
24 
1 
i 
7 
3 
5 
4 
3 
28 to 30 months 
5 
5 
1 
3 
i 
31 to 33 months ... 
5 
5 
i 
1 
2 
1 
Age not reported 
8 
8 
1 
2 
2 
3 
1 Age is given as of nearest month; that is, “1 month” 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. APPENDIX A 
105 
Table A6.—Average weight at each month of age; boys and girls 1 to S3 months of 
age examined in Puerto Rico 
Age 1 at examination 
Average weight (kilograms) 
Boys 
Girls 
Observed 
Smoothed 
Observed 
Smoothed 
1 month 
4.0 
4.08 
4.1 
3.79 
2 months 
5.0 
5. 01 
4.6 
4.81 
3 months 
5.6 
5.69 
5.6 
5.44 
4 months . . 
6.6 
6.06 
6.2 
5.91 
5 months 
6.5 
6. 44 
5.9 
6.30 
6 months . 
6.5 
6. 77 
6.1 
6. 62 
7 months _ __ 
6.8 
7.07 
6.8 
6.91 
8 months 
7.3 
7. 34 
6.7 
7.18 
9 months 
7.2 
7.59 
7.4 
7. 42 
10 months 
7.8 
7.83 
7.5 
7. 64 
11 months _ 
7.8 
8.06 
7.4 
7.85 
12 months _ 
8.7 
8.28 
8.0 
8.05 
13 months 
7.9 
8.49 
8.2 
8.24 
14 months-- ... _ 
9.5 
8.69 
9.4 
8. 42 
15 months 
9.6 
8.89 
9.0 
8.59 
16 months 
9.5 
9.08 
9.8 
8. 76 
17 months 
9.3 
9.27 
9.8 
8.92 
18 months 
9.3 
9.46 
9. 1 
9.08 
19 months . 
9.5 
9.64 
8.8 
9.24 
20 months __ _ 
9.7 
9.81 
9.8 
9. 39 
21 months 
9.3 
9.99 
9.3 
9.53 
22 months- - 
10.2 
10.16 
8.6 
9.68 
23 months 
10.3 
10.33 
10.1 
9.82 
24 months - 
10.0 
10.49 
9.0 
9.96 
25 months 
( 10.66 
) 
( 10.09 
26 months _ 
10.9 
{ 10.82 
\ 10.4 
| 10.23 
27 months 
10.98 
1 
10.36 
28 months 
11.14 
1 
10.49 
29 months 
11.7 
\ 11.30 
\ 11.0 
{ 10.62 
30 months 
11.45 
1 
10. 75 
31 months . - - 
11.61 
] 
10.87 
32 months 
11.8 
\ 11.76 
\ 10.8 
\ 10.99 
33 months 
l 11.91 
1 
1 11.12 
Equations of curves used in smoothing: 
Boys, j/=3.966361+0. 116496z+2.701817 log. x. 
Girls, y=3.713483+0.080924r+3.116864 log. x. 
x representing age in months, y representing weight in kilograms. 
> Age is given as of nearest month; that is, “1 month” is from 16 days to 1 month 15 days, inclusive. 
Note that from 26 months onward each observed average includes a 3-month interval. 106 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table A7.—Weight for height of boys 1 to 38 months of age examined in Puerto 
Rico 
Boys 1 to 33 months of age 
Weight (kilograms) reported 
O 
Height (centimeters) 
Total 
Total 
2 
re- 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
ported 
£ 
Total 
307 
305 
5 
16 
24 
30 
55 
49 
48 
33 
26 
9 
7 
3 
2 
Height reported.. 
304 
302 
4 
16 
24 
30 
55 
49 
46 
33 
26 
9 
7 
3 
2 
1 
1 
1 
2 
2 
2 
11 
11 
1 
4 
6 
55, under 57... _ ...... 
10 
10 
1 
9 
57, under 59.... . . 
7 
7 
5 
2 
15 
15 
i 
3 
2 
6 
3 
26 
26 
4 
13 
7 
2 
24 
24 
2 
1 
14 
1 
1 
20 
19 
4 
11 
4 
1 
28 
28 
2 
10 
11 
28 
28 
6 
12 
3 
7l! under 73 
27 
27 
i 
14 
7 
4 
1 
21 
21 
i 
2 
11 
4 
3 
23 
23 
i 
2 
8 
10 
1 
i 
15 
15 
i 
1 
3 
8 
2 
79| under 81 
14 
14 
2 
2 
8 
2 
11 
11 
1 
1 
6 
2 
i 
7 
7 
1 
4 
1 
i 
4 
4 
1 
2 
1 
87’ under 89 
5 
5 
1 
1 
2 
1 
3 
3 
1 
1 
1 
Ql, under 93 
2 
1 
1 
1 
3 
3 
1 
2 
Table AS.—Weight for height; girls 1 to 83 months of age examined in Puerto Rico 
Girls 1 to 33 months of age 
Weight (kilograms) reported 
O 
Height (centimeters) 
Total 
Total 
re- 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
13 
14 
bL c, 
ported 
k* *" 
Total 
274 
272 
1 
2 
21 
17 
42 
45 
36 
38 
40 
19 
6 
4 
1 
2 
Height reported 
272 
270 
1 
2 
20 
17 
41 
45 
36 
38 
40 
19 
6 
4 
1 
2 
1 
1 
1 
1 
1 
1 
2 
2 
2 
1 
1 
1 
S3, under 55 
13 
13 
1 
10 
2 
9 
9 
1 
3 
3 
2 
13 
13 
2 
6 
5 
59, under 
17 
17 
2 
13 
2 
20 
20 
1 
3 
9 
7 
14 
14 
1 
7 
5 
1 
26 
26 
2 
17 
6 
1 
13 
13 
1 
5 
6 
1 
18 
18 
1 
5 
9 
2 
1 
19 
19 
1 
4 
11 
3 
25 
24 
1 
2 
5 
8 
7 
1 
i 
23 
23 
1 
2 
7 
12 
1 
77, under 79 
17 
17 
3 
6 
7 
1 
10 
10 
7 
3 
14 
14 
2 
2 
2 
1 
3 
3 
2 
1 
9 
8 
1 
3 
2 
2 
i 
2 
2 
1 
1 
1 
1 
1 
1 
1 
1 
Height not reported 
2 
2 
1 
1 APPENDIX A 
107 
Table A9.—Average weight for height; Puerto Rican boys and girls 
Average height 
(centimeters) 
Average weight (kilograms) 
Average height 
(centimeters) 
Avreage weight (kilograms) 
Boys 
Girls 
Boys 
Girls 
Observed 
Smoothed 
Observed 
Smoothed 
Observed 
Smoothed 
Observed 
Smoothed 
46 
2.0 
2.65 
72. 
8.6 
8.53 
8.8 
8. 47 
48 
3.0 
2. 89 
4.0 
3.10 
74 _ 
9.3 
9.00 
8.9 
8. 92 
50 
3. 36 
4.0 
76 .. 
9.5 
9.47 
9.4 
9. 36 
52 
4.0 
3.83 
4.0 
3.99 
78 
9.6 
9. 94 
9. 4 
9.81 
51 
4.5 
4.30 
4.1 
4.44 
80 
10.7 
10.41 
10.3 
10. 26 
50 
4. 8 
4. 77 
4.7 
4.89 
82 
11.1 
10.88 
10. 9 
10. 71 
4.3 
5. 24 
5. 2 
5. 33 
84 
11. 3 
11. 35 
11. 3 
11 15 
fit) ... 
5.5 
5. 71 
6.0 
5. 78 
86 
11.8 
11.82 
11.6 
11. 60 
02. 
6.3 
6.18 
6.1 
6.23 
88. 
12.6 
12.29 
12.5 
12.05 
64. 
6.6 
6. 65 
6.4 
6.68 
90 
13.0 
12.76 
11.0 
12. 50 
06 
7.0 
7.12 
7.2 
7.12 
92 
13.23 
14.0 
12.94 
68 . 
7.7 
7. 59 
7.5 
7.57 
94. 
13.0 
13.71 
70 
8.3 
8.06 
7.8 
8.02 
Equations of curves used in smoothing: 
Boys, y 
= 2.415280+0.235207X. 
Girls, v 
= 2.199033 +0.22384Sr. 
x representing centimeters of height above 46; y representing weight in kilograms. 
Table A10.—Head circumference and age; boys 1 to S3 months of age examined in 
Puerto Rico 
Boys 1 to 33 months of age 
Age 1 at examination 
Head circumference (centimeters) 2 
To- 
tal 
34 
35 
36 
37 
38 
39 
40 
41 
42 
43 
44 
45 
46 
47 
48 
49 
50 
51 
Total 
307 
1 
1 
3 
6 
9 
10 
23 
17 
28 
37 
25 
26 
44 
41 
23 
11 
1 
1 
Age reported 
295 
1 
1 
3 
6 
9 
10 
22 
16 
27 
35 
24 
24 
43 
40 
21 
11 
1 
1 
1 month... ... 
9 
1 
1 
2 
2 
2 
1 
11 
1 
1 
5 
3 
1 
16 
2 
2 
2 
6 
1 
3 
19 
4 
5 
5 
2 
3 
IS 
1 
4 
3 
6 
3 
1 
16 
1 
1 
2 
2 
3 
5 
1 
1 
13 
1 
4 
5 
3 
8 months. _ __ 
17 
1 
3 
7 
2 
3 
1 
13 
1 
1 
2 
5 
2 
2 
15 
2 
6 
2 
1 
3 
1 
11 months... 
17 
1 
2 
1 
3 
3 
4 
2 
1 
12 months .. __ 
12 
1 
1 
2 
7 
1 
13 months.. .. 
9 
1 
1 
3 
1 
2 
1 
14 months... . 
8 
1 
1 
2 
2 
2 
IS months.. 
8 
1 
4 
3 
11 
1 
1 
2 
3 
4 
17 months 
12 
1 
5 
5 
1 
6 
1 
1 
2 
2 
19 months 
10 
1 
1 
6 
1 
1 
20 months 
7 
4 
1 
2 
4 
1 
2 
1 
22 months 
5 
1 
3 
1 
23 months.. . 
7 
1 
1 
3 
1 
1 
4 
1 
1 
2 
14 
1 
2 
4 
3 
3 
1 
7 
2 
2 
1 
2 
7 
1 
1 
4 
1 
Age not reported 
12 
1 
1 
1 
2 
1 
2 
1 
1 
2 
1 Age is given as of nearest month; that is, “ 1 month” 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 cen- 
timeters, inclusive, and so on. 
160326°—33 8 108 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table All.—Head circumference and age; girls 1 to 88 months of age examined 
in Puerto Rico 
Girls 1 to 33 months of age 
Age 1 at examination 
Head circumference (centimeters) 2 
Total 
34 
35 
36 
37 
38 
39 
40 
41 
42 
43 
44 
45 
46 
47 
48 
49 
Total 
274 
2 
2 
8 
10 
10 
11 
18 
15 
24 
25 
45 
37 
32 
17 
9 
9 
Age reported 
266 
2 
2 
8 
10 
10 
11 
18 
15 
23 
25 
41 
35 
31 
17 
9 
9 
8 
1 
2 
1 
4 
17 
1 
5 
4 
3 
1 
3 
11 
2 
2 
2 
4 
1 
12 
2 
2 
3 
3 
1 
1 
18 
2 
3 
3 
5 
3 
1 
1 
15 
1 
3 
3 
1 
12 
1 
2 
2 
3 
2 
1 
1 
13 
1 
1 
5 
3 
2 
1 
7 
1 
1 
5 
11 
1 
4 
4 
1 
1 
10 
1 
1 
3 
3 
2 
9 
3 
4 
2 
13 
4 
5 
2 
2 
n 
1 
4 
1 
1 
13 
1 
5 
4 
3 
5 
2 
1 
2 
5 
1 
2 
2 
7 
1 
3 
3 
5 
1 
1 
3 
4 
1 
2 
1 
9 
2 
2 
2 
3 
5 
1 
2 
1 
1 
9 
1 
1 
2 
1 
1 
3 
7 
2 
1 
4 
24 
1 
6 
3 
5 
6 
3 
5 
1 
1 
2 
1 
5 
1 
2 
1 
1 
8 
1 
4 
2 
1 
1 Age is given as of nearest month; that is, “1 month” 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. APPENDIX A 
109 
Table A12.—Average head circumference at each month of age; hoys and girls 
1 to S3 months of age examined in Puerto Rico 
Average head circumference (centimeters) 
Age1 at examination 
Boys 
Girls 
Observed 
Smoothed 
Observed 
Smoothed 
1 month - 
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 
47.6 
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 
f 47.37 
( 47.51 
47.64 
f 47.77 
( 47.90 
48.02 
48.14 
\ 48.26 
1 48.37 
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 
t 46.8 
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 
2 months. 
3 months 
4 months.. 
5 months . 
6 months.. _ 
7 months . _ _ 
8 months . 
9 months 
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 . 
28 months 
29 months... 
30 months 
31 months 
32 months 
33 months . _ . 
47.4 
48.7 
| 46.8 
| 46.2 
46. 63 
1 46.73 
46.84 
46.93 
\ 47.03 
l 47.11 
Equations of curves used in smoothing: 
Boys, y = 36.241792+0.006690.r+7.839603 log. x. 
Girls, y=35.346584 — 0.019527X+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 month” is from 16 days to 1 month and 15 days, inclusive. 
Note that from 26 months onward each observed average includes a 3-month interval. Appendix B.—SKELETAL SIGNS OF RICKETS 
Table Bl.—Clinical diagnosis of rickets in relation to number of skeletal signs 
usually considered evidence of rickets; children examined in Puerto Rico 
Number of skeletal 
signs 
Children, examined 
Total 
Clinical diagnosis of rickets 
No rickets 
Questionable 
diagnosis 
Rickets 
Per cent 
Per cent 
Per cent 
Per cent 
Number 
distribu- 
Number 
distribu- 
Number 
distribu- 
Number 
distribu- 
tion 
tion 
tion 
tion 
Total 
584 
too 
400 
100 
134 
100 
50 
100 
None 
238 
41 
236 
59 
2 
1 
1 
208 
30 
136 
34 
09 
51 
3 
6 
2 
97 
17 
24 
6 
51 
38 
22 
44 
3 
30 
5 
4 
1 
11 
8 
15 
30 
4 
9 
2 
1 
] 
8 
16 
6 
2 
(>) 
2 
4 
i Less than 1 per cent. APPENDIX B 
111 
Table B2.—Clinical diagnosis of rickets in relation to combinations of skeletal signs 
usually considered evidence of rickets; children examined in Puerto Rico 
Children examined 
Clinical diagnosis of rickets 
Skeletal signs 
T 
No 
rickets 
Ques- 
tionable 
diag- 
nosis 
Rickets 
Total 
Slight 
Moder- 
ate 
Mark- 
ed 
Total 
584 
400 
134 
50 
46 
3 
i 
238 
236 
2 
One skeletal sign only 
208 
136 
69 
3 
3 
Costochondral junctions - - _ 
25 
2 
23 
Epiphyses . _ ._ 
15 
4 
11 
Bowlegs __ . 
87 
72 
12 
3 
3 
Knock-knees - - 
37 
32 
5 
Harrison’s groove. 
9 
2 
Parietal or frontal bosses moderately en- 
larged _ ___ . 
14 
7 
7 
Craniotabes _ _ .. ___ 
3 
1 
2 
Asymmetry of head 
18 
16 
2 
Two skeletal signs_ . _ 
97 
24 
51 
22 
21 
1 
Costochondral junctions and epiphyses 
5 
1 
4 
4 
17 
3 
9 
5 
5 
Costochondral junctions and knock-knees. 
Costochondral junctions and Harrison’s 
7 
2 
5 
1 
1 
Costochondral junctions and parietal or 
frontal bosses moderately enlarged 
3 
1 
1 
1 
1 
Costochondral junctions and craniotabes.. 
Costochondral junctions and asymmetry 
of head . 
1 
1 
1 
2 
1 
1 
1 
17 
1 
13 
3 
3 
Epiphyses and knock-knees .. . .. 
1 
1 
Epiphyses and Harrison’s groove. 
1 
1 
1 
1 
13 
10 
3 
Bowlegs and Harrison’s groove.. ... 
5 
3 
2 
2 
Bowlegs and parietal or frontal bosses 
9 
1 
8 
Knock-knees and Harrison’s groove .. 
5 
1 
1 
3 
3 
Knock-knees and parietal or frontal bosses 
5 
3 
2 
1 
1 
1 
Asymmetry of head and parietal or frontal 
2 
2 
Parietal and frontal bosses moderately 
1 
1 
1 
Three skeletal signs _ .... 
30 
4 
11 
15 
15 
Costochondral junctions, epiphyses, and 
parietal or frontal bosses moderately en- 
1 
1 
1 
Costochondral junctions, epiphyses, and 
6 
3 
3 
3 
Costochondral junctions, epiphyses, and 
2 
2 
2 
Costochondral junctions, bowlegs, and 
2 
1 
1 
Costochondral junctions, bowlegs, and 
parietal or frontal bosses moderately 
1 
1 
Costochondral junctions, bowlegs, and 
3 
3 
3 
Costochondral junctions, bowlegs, and 
knock-knees 
3 
2 
1 
1 112 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table B2.—Clinical diagnosis of rickets in relation to combinations of skeletal signs 
usually considered evidence of rickets; children examined in Puerto Rico.—Con. 
Children examined 
Clinical diagnosis of rickets 
Skeletal signs 
Total 
No 
rickets 
Ques- 
tionable 
diag- 
nosis 
Rickets 
Total 
Slight 
Moder- 
ate 
Mark- 
ed 
Three skeletal signs—Continued 
Costochondral junctions, parietal or fron- 
tal bosses moderately enlarged, and 
2 
1 
1 
1 
Costochondral junctions, Harrison’s 
1 
1 
1 
Costochondral junctions, Harrison’s 
1 
1 
Epiphyses, bowlegs, and asymmetry of 
1 
1 
1 
1 
Bowlegs, knock-knees, and parietal or 
2 
1 
1 
1 
Bowlegs, knock-knees, and Harrison’s 
2 
2 
2 
Knock-knees, parietal or frontal bosses 
moderately enlarged, and asymmetry 
1 
1 
Asymmetry of head and parietal and fron- 
1 
1 
Four skeletal signs.-- 
9 
1 
8 
7 
1 
Costochondral junctions, epiphyses, bow- 
1 
1 
1 
Costochondral junctions, epiphyses, bow- 
legs, and parietal or frontal bosses mod- 
1 
1 
1 
Cost.ochondral junctions, epiphyses, bow- 
2 
2 
1 
1 
Costochondral junctions, epiphyses, 
knock-knees, and parietal or frontal 
1 
1 
1 
Costochondral junctions, epiphyses, 
1 
1 
1 
Costochondral junctions, bowlegs, Harri- 
son’s groove, and parietal or frontal 
1 
1 
1 
Epiphyses, bowlegs, Harrison’s groove, 
1 
1 
1 
Epiphyses, bowlegs, Harrison’s groove, 
1 
1 
Six skeletal signs-_- _ 
2 
2 
1 
i 
Parietal and frontal bosses moderately 
enlarged, costochondral junctions, epi- 
physes, Harrison’s groove, and cranio- 
1 
1 
i 
Parietal and frontal bosses moderately 
enlarged, costochondral junctions, Har- 
rison’s groove, knock-knees, and asym- 
metry of head 
1 
1 
1 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 milliamperes 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 one- 
tenth of a second, if a current of 100 milliamperes 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 milliamperes may be passed 
through a 30-milliampere radiator tube without damage to the tube. 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 lour 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. 
Table Dl.—Diet grade and type of feeding of children during first 3-month period 
{birth to 3 months); children in families visited in Puerto Rico 
Diet grade 
Type of feeding 
Children in families visited 
Breast milk 
Cow’s 
milk 
(average 
amount 
daily) 
Total 
Age at d 
Under 4 
months 
ite of home vis 
4 months and 
over 
it 
Not 
re- 
ported 
Num- 
ber 
Per 
cent 
distri- 
bution 
Num- 
ber 
Per 
cent 
distri- 
bution 
Num- 
ber 
Per 
cent 
distri- 
bution 
556 
70 
473 
13 
Grade reported _ 
554 
385 
100 
69 
70 
54 
100 
77 
472 
322 
100 
68 
12 
9 
Good 
140 
25 
14 
20 
124 
26 
2 
Mainly 
1 CUP-- 
no 
20 
12 
17 
96 
20 
2 
Small amount— 
1 pint.-. 
30 
5 
2 
3 
28 
6 
9 
2 
9 
2 
1 cup 
20 
4 
2 
3 
17 
4 
1 
2 
1 
1 115 
APPENDIX D 
Table D2.—Diet grade and type of feeding of children during second 3-month 
period (fourth to sixth month); children 4 months of age and over in families 
visited in Puerto Rico 
Children 4 months of age and over 
in families visited 
Type of feeding 
Age at date of home visit 
Total 
Diet grade 
Milk feeding 
Some 
4 to 6 months 
7 months and 
over 
form 
Cow’s milk 
of 
solid 
food1 
Per- 
Per- 
Per- 
Breast milk 
(average 
Num- 
cent 
Num- 
cent 
Num- 
cent 
amount 
her 
distri- 
ber 
distri- 
ber 
distri- 
daily) 
bution 
bution 
bution 
473 
100 
373 
469 
100 
98 
too 
371 
100 
34 
7 
11 
11 
23 
6 
+ 
+ 
14 
3 
7 
7 
2 
20 
4 
4 
4 
16 
4 
Good 
300 
64 
68 
69 
232 
63 
0 
• 212 
45 
37 
38 
175 
47 
0 
61 
13 
23 
23 
38 
;o 
Small amount- 
1 pint 
+ 
21 
4 
6 
6 
15 
4 
None 
More than 
+ 
6 
1 
2 
2 
4 
l 
1 pint. 
Fair 
78 
17 
13 
13 
65 
18 
Small amount 
1 pint 
0 
33 
7 
4 
4 
29 
8 
M ore than 
0 
26 
6 
5 
5 
21 
6 
1 pint. 
None - . 
1 pint 
+ 
19 
4 
4 
4 
15 
4 
1 pint. 
0 
27 
6 
4 
4 
23 
6 
30 
6 
2 
2 
28 
8 
1 cup 
0 
9 
2 
1 
i 
8 
2 
do 
+ 
21 
4 
1 
i 
20 
5 
4 
2 
2 
1 “ + ” indicates that some solid food was given; “0” that no solid food was given. The kind of solid 
food was not considered. 116 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table 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 
Type of feeding 
Children 7 months of age and over in 
families visited 
Milk feeding 
Age at date of home visit 
Total 
Diet grade 
Some 
form 
7 to 9 months 
10 months 
and over 
Cow’s milk 
of 
Breast milk 
(average 
amount 
solid 
food 1 
Per 
Per 
Per 
daily) 
Num- 
cent 
Num- 
cent 
Num- 
cent 
her 
distri- 
her 
distri- 
ber 
distri- 
button 
button 
button 
Total 
373 
74 
299 
307 
100 
73 
100 
294 
100 
49 
13 
5 
7 
44 
15 
+ 
+ 
22 
6 
1 
1 
21 
7 
27 
7 
4 
5 
23 
8 
58 
16 
13 
18 
45 
15 
Small amount.. 
1 pint 
+ 
30 
8 
6 
8 
24 
8 
More than 
+ 
28 
8 
7 
10 
21 
7 
1 pint. 
Fair 
214 
58 
46 
63 
168 
57 
0 
105 
29 
21 
29 
84 
29 
0 
30 
8 
6 
8 
24 
8 
Small "amount— 
1 pint 
0 
23 
6 
6 
8 
17 
6 
More than 
0 
30 
8 
9 
12 
21 
7 
1 pint. 
+ 
22 
26 
7 
4 
5 
7 
0 
21 
6 
5 
7 
16 
5 
25 
7 
4 
5 
21 
7 
+ 
0 
11 
3 
1 
1 
10 
3 
14 
4 
3 
4 
11 
4 
Grade not reported 
6 
1 
5 
1 indicates that some form of solid food was given; “0” that no solid food was given. The kind of 
solid food was not considered. 117 
Table D4.—Diet grade and type of feeding of children during fourth 3-month period 
{tenth to twelfth month); children 10 months of age and over in families visited in 
Puerto Rico 
APPENDIX D 
Type of feeding 
Children 10 months of age and 
over in families visited 
' 
Age at date of home 
visit 
Total 
Diet grade 
Solid food J 
10 to 12 
months 
13 months 
and over 
Breast milk 
Cow’s milk 
(average 
amount 
daily) 
Number 
Per cent dis- 
tribution 
Number 
Per cent dis- 
tribution 
Number 
Per cent dis- 
tribution 
299 
66 
233 
Grade reported 
293 
100 
66 
100 
227 
100 
Very good 
12 
4 
2 
3 
10 
4 
None.. 
7 
2 
1 
2 
6 
3 
pint. 
1 pint 
Small amount.. 
5 
2 
1 
2 
4 
2 
Good 
2 
1 
2 
1 
Small amount.- 
1 pint 
1 
0) 
(*) 
2 
1 
3 
1 
1 
Fair.. 
7 
1 
2 
6 
3 
1 
1 
2 
2 
1 
2 
1 
2 
1 
2 
1 
2 
1 
Poor 
200 
68 
46 
70 
154 
68 
Unsatisfactory 
29 
10 
9 
14 
20 
9 
None. 
pint. 
16 
5 
16 
7 
1 pint . 
Unsatisfactory. 
24 
8 
9 
14 
15 
7 
Do 
do 
17 
6 
7 
11 
10 
4 
Unsatisfactory 
22 
8 
9 
14 
13 
6 
Do 
do 
14 
5 
2 
3 
12 
5 
Unsatisfactory 
28 
10 
4 
6 
24 
11 
Do... 
50 
17 
6 
9 
44 
19 
Very poor 
72 
25 
17 
26 
55 
24 
Unsatisfactory 
32 
11 
10 
15 
22 
10 
15 
5 
3 
5 
12 
5 
None - - 
1 cup 
Unsatisfactory... 
16 
5 
2 
3 
14 
6 
9 
3 
2 
3 
7 
3 
Grade not reported.. 
6 
6 
1 For method of grading solid food see p. 91. 
2 Less than 1 per cent. 118 
DEVELOPMENT OF BONES OF CHILDREN IN PUERTO RICO 
Table D5.—Diet grade and type of feeding of children during second year; children 
13 months of age and over in families visited in Puerto Rico 
Type of feeding 
Children 13 months of age and 
over in families visited 
Age at date of home 
visit 
Total 
Diet grade 
Solid food 1 
13 to 24 
months 
25 months 
and over 
Breast milk 
Cow’s milk 
(average 
amount 
daily) 
Number 
Per cent dis- 
tribution 
Number 
Per cent dis- 
tribution 
Number 
Per cent dis- 
tribution 
233 
176 
57 
203 
100 
147 
100 
56 
100 
1}4 to 2 pints 
Very satisfactory. 
2 
1 
1 
1 
1 
2 
6 
3 
2 
1 
4 
7 
1H to 2 pints 
1 
(2) 
2 
1 
2 
5 
2 
1 
3 
5 
23 
11 
21 
14 
2 
4 
to 2 pints 
7 
3 
7 
5 
9 
4 
7 
5 
2 
4 
4 
2 
4 
3 
3 
1 
3 
2 
71 
35 
48 
33 
23 
41 
1 Yi to 2 pints 
1 pint . 
Unsatisfactory 
10 
5 
8 
5 
2 
4 
28 
14 
18 
12 
10 
18 
4 
2 
3 
2 
1 
2 
Unsatisfactory 
26 
13 
17 
12 
9 
16 
2 
1 
2 
1 
1 
(’) 
50 
1 
2 
101 
75 
51 
26 
46 
1J4 to 2 pints 
Very unsatisfac- 
tory. 
2 
1 
1 
1 
1 
2 
3 
1 
2 
1 
1 
2 
Very unsatisfac- 
tory. 
Unsatisfactory or 
very unsatis- 
tory. 
3 
1 
2 
1 
1 
2 
42 
21 
28 
19 
14 
25 
3 
1 
3 
2 
Very unsatisfac- 
tory. 
Unsatisfactory or 
very unsatis- 
factory. 
3 
1 
2 
1 
1 
2 
32 
16 
27 
18 
5 
9 
Do - 
1 
(J) 
4 
1 
2 
Unsatisfactory or 
very unsatis- 
factory. 
9 
7 
5 
2 
4 
Do 
3 
1 
3 
2 
4 
Grade not reported- 
3 30 
3 29 
1 
1 For method of grading solid food, see p. 92 
2 Less than 1 per cent. 
3 Includes 26 children for whom diet grade was not reported as they had lived only one month into this 
period. APPENDIX D 
119 
Table D6.—Diet grade and type of feeding of children during third year; children 
25 months of age and over in families visited in Puerto Rico 
Diet grade 
Type of feeding 
Children 25 
months of age 
and over at 
date of home 
visit 
Milt 
Breast milk 
feeding 
Cow’s milk (av- 
erage amount 
daily) 
Solid food 1 
Total _ _ 
57 
Grade reported.-. .. .. 
40 
1 
Good 
2 
Fair . _ _ 
8 
None 
3 
2 
None _ 
3 
Poor 
9 
1 
None _ 
1 pint- 
6 
2 
Very poor. 
20 
18 
unsatisfactory. 
2 
Grade not reported 
2 17 
1 For method of grading solid food, see p. 92. 
2 Includes 16 children for whom grade was not reported as they had lived only 1 month into this period. 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 docl or 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 cranio- 
tabes 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 ot 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 
*i Wimberger, Hans: Klinisch-radiolische Diagnostik von Rachitis, Skorbut, und Lues inn Kindesalter- 
[Clinical and Radiological Diagnosis of Rickets, Scurvy, and Congenital Syphilis in Childhood.] Ergeb- 
nisse der inneren Medizin und Kinderheilkunde [Berlin], vol, 28 (1925), pp, 269-288, 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, 64y2 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 bowdng 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 
w as made. 
Case 8.—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 wras 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 wras 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. Breast fed. 
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 Avere well 
calcified, there was no osteoporosis, and the cortex was clearly defined and of 
about average width. The diaphyseal-epiphyseal junctions were well calcified 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. 
In the substance of the spongiosa, about 2 centimeters from the distal end of 
both the right and left ulme, 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 
of the medullary space. In the radii, also, similar zones, though even less we1! 
defined, could be seen at approximately 2 centimeters from the distal ends of the 
bones. In both radii at the region of this zone there was a slight bulging of the 
shaft. The 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.