Skip to main content

Full text of "The effects of inanition and malnutrition upon growth and structure"

See other formats

HI 1 ■ II mm 













C. M. JACKSON, M.S., M.D., LL.D., 






Copyright, 1925, By P. Blakiston's Son & Co. 


To My Teachers 




The widespread occurrence of human famine during and since the world 
war has raised serious questions concerning both the immediate and the remote 
effects upon the human species. Even in the more fortunately situated coun- 
tries, recent investigations have revealed, especially among children, a large 
amount of malnutrition, with possible consequences of great importance to 

Inanition in animals and plants is likewise a subject of much interest, and 
presents a method of the utmost value in the study of the living organism. 
By withholding or decreasing the normal diet (total inanition) or merely one or 
more of the essential nutritional elements (partial inanition), we may observe 
effects which throw much light upon the process of nutrition from the standpoint 
of normal morphology, of physiology, or of pathology. 

Inanition is therefore a subject of both theoretical and practical importance 
to scientific workers in various fields — to biologists (both zoologists and botan- 
ists), who are concerned with the fundamental characters of living organisms; 
to anatomists, who are interested in the problems of morphogenesis; to physiolo- 
gists and biochemists, working in the various fields of human and animal 
nutrition; to pathologists, since inanition is one of the primary factors in patho- 
genesis; and to physicians, who recognize inanition, not merely as a therapeutic 
measure of occasional utility, but especially as a complication in most of the 
disorders with which they have to deal. Inanition and malnutrition have long 
been considered of importance in connection with the diseases of infancy, but 
only recently has their significance become more fully apparent likewise for 
adults, and above all in the so-called "deficiency diseases." The nature of 
these deficiency disorders as forms of partial inanition, and their relationship 
to each other and to inanition in general, constitute an outstanding problem 
in medicine. 

On account of these varied and important relations, a systematic review of 
the subject of inanition seems desirable, especially from the viewpoint of mor- 
phology. The present work will therefore include not only the results published 
by the author and his coworkers during a decade of research in this field, 
together with a considerable amount of unpublished material, but also a com- 
prehensive review of the widely scattered data in the literature concerning 
the morphological effects of inanition in all living organisms. While an ade- 
quate critique in so vast a field is impossible, it is hoped that even a brief 
survey of the literature will be of interest and service. Especial care has been 
taken to make the bibliography as complete and accurate as possible, although 
errors and omissions are unavoidable. 


Grateful acknowledgment is due to several of my friends and colleagues, 
especially to Professors Donaldson, Scammon, Downey, Sigerfoos, McKnight, 
Bell and Litzenberg, who have read and criticised certain chapters; to the 
artist, Miss Jean Hirsch, and to various authors for the illustrations used; and to 
the Graduate School of the University of Minnesota for special support in the 
original investigations which are utilized in the present work. 

C. M. Jackson. 




Preface vii 

Introduction xi 



I. Effects of Inanition on Plants . i 

II. Effects on the Protozoa 15 

III. Effects on the Higher Invertebrates 26 


IV. Effects on the Body as a Whole (Total Inanition) 67 

V. Effects on the Body as a Whole (Partial Inanition) 98 

VI. Effects on the Integument, Adipose Tissue and Mammary Gland . . . 117 

VII. Effects on the Skeleton 133 

VIII., Effects on the Teeth 156 

IX. Effects on the Musculature 162 

X. Effects on the Brain 173 

XI. Effects on the Spinal Cord 19 1 

XII. Effects on the Peripheral Nervous System 203 

XIII. Effects on the Visual Apparatus 210 

XIV. Effects on the Heart and Blood Vessels 222 

XV. Effects on the Blood 238 

XVI. Effects on the Lymph and Lymphatic Glands 261 

XVII. Effects on the Spleen 269 

XVIII. Effects on the Thymus 285 

XIX. Effects on the Alimentary Canal 3°° 

XX. Effects on the Liver 3 2 4 

XXI. Effects on the Pancreas and Salivary Glands 345 

XXII. Effects on the Respiratory Tract. Lungs, Trachea and Larynx ... 361 

XXIII. Effects on the Urinary Tract. Kidneys and Bladder 37° 

XXIV. Effects on the Female Reproductive Tract 389 

XXV. Effects on the Male Reproductive Tract 403 

XXVI. Effects on the Suprarenal Glands 417 

XXVII. Effects on the Thyroid and Parathyroid Glands 435 

XXVIII. Effects on the Hypophysis 448 


XXIX. Conclusions 457 

Tables 1 to 13 462 

Bibliography 479 

Index 587 

,■> a I ix 


It is desirable at the outset to define some of the terms used throughout the 
present work. 

Inanition is defined by the Standard Dictionary as the state of being void 
or empty; specificially, in pathology, exhaustion from lack of nourishment, 
whether by absence of food or disorder of the nutritive system (from Latin 
inanis, empty). Accordingly the term inanition as used in the present work 
indicates in a broad sense the lack of food or of any foodstuff (including water) 
which is essential to the living organism. Malnutrition or dystrophy are often 
used as synonymous with inanition in general, but are wider in scope, including 
disturbances of nutrition from causes other than mere lack of nutriment. 
Starvation, in the broad meaning of the term, is synonymous with inanition; but 
is more frequently used to indicate the extreme stages of inanition, leading to 
death. Famine denotes inanition or starvation on a large scale, especially in 
the human species, with widespread scarcity of food, usually resulting from war, 
drought, floods, insects, etc. Hunger and thirst indicate the sensations arising 
from the lack of food or water, respectively; and to famish means to undergo 
extreme hunger or thirst. 

Inanition as above defined includes many different conditions, which may 
be classified in various ways. As to the character of the inanition, we may 
distinguish (^4) total inanition, with absence or insufficiency of all nutriment; or 
(B) partial inanition, with absence or insufficiency of merely one or more of the 
essential elements of food. As synonymous with "total" inanition, the terms 
"absolute," "general," or "quantitative" inanition are frequently used in the 
literature. As synonymous with partial inanition, the terms "relative," "spe- 
cial" or "qualitative" inanition are often employed. Some authors use the 
term "complete" inanition as synonymous with "total," and "incomplete" 
as synonymous with "partial;" but they are defined differently in the present 
work, as noted below. 

As to the degree of either total or partial inanition, we may distinguish either 
(i) complete inanition, with entire absence of all food (in total inanition) or of the 
deficient elements (in partial inanition); or (2) incomplete inanition, with merely 
an insufficient amount of all food, or of the deficient elements. The terms 
"underfeeding," "subnutrition," or "caloric insufficiency" are frequently used 
to indicate incomplete total inanition. 

As to the duration and severity of the inanition, we may distinguish (1) 
acute inanition, which is severe and of shorter duration; and (2) chronic inanition, 
which is milder and of longer duration. 

Inanition may also be classified according to the mode of occurrence which 
results in the malnutrition of the cells of the living organism. Thus inanition 
may arise from extrinsic causes, which in some way prevent the necessary nutri- 



ment from reaching the cells; or from intrinsic causes, affecting protoplasmic 
metabolism. The intrinsic causes may interfere with the proper assimilation 
(anabolism) of the food, even though it may reach the cells in adequate quantity 
and quality; or they may occasion an abnormally rapid consumption (katabo- 
lism) of the food, thereby creating a condition of relative inanition. 

Thus among the extrinsic causes of inanition in the higher organisms, in 
addition to (a) lack of adequate food, there may exist (b) faulty ingestion or 
mastication of the food, due to oral defects, stenosis of the alimentary canal, etc.; 
(c) faulty digestion, due to glandular deficiency; (d) faulty absorption through the 
alimentary mucosa; or (e) faulty transportation, through defects in the blood or 
vascular system. 

Similarly, various intrinsic conditions may prevent the normal metabolism 
of the food, even though it is brought to the cells in normal quantity and quality. 
Thus the faulty cell metabolism may arise from (a) hereditary, inherent defects 
in cell structure or composition; (b) toxic influences which prevent normal metab- 
olism; or (c) lack or excess of the hormones normally concerned with cell 
metabolism. For example, a condition of inanition may arise either from lack 
of the pancreatic hormone (insulin) in diabetes mellitus, or from excess of the 
thyroid hormone (thyroxin) in hyperthyroidism (cf. Lubarsch '03; Watson '99; 
Barker '16). 

The various types of inanition are summarized in the following table: 

I 1. Complete (no food whatever) 
2. Incomplete (insufficient nutriment; general under- 

A. Total 


Inanition \ 

B. Partial (qualitative) 



of one or more 


(entire absence) 

of the neces- 



sary food- 











While the present work is concerned chiefly with the effects of inanition 
upon animals, a brief (and necessarily incomplete) review of the effects upon 
plants also will be found of interest and value. The general observation that 
plants, as well as animals, thrive according to the quantity and quality of 
their nutriment was doubtless made even in prehistoric times. More exact 
knowledge has slowly accumulated, but apparently the process of starvation 
has been studied less extensively in plants than in animals. The metabolic 
processes are fundamentally similar in plants and animals, and some aspects of 
inanition (especially of partial inanition) are more clearly apparent in the 
simpler plant organism. The chief effects of inanition on plants will first be 
summarized briefly, followed by a more detailed account of the results of total or 
partial inanition upon the various species. 

Summary of the Effects on Plants 

Plants in general, much more than animals, appear susceptible to modification 
by various external factors, including the food supply. It is difficult to sum- 
marize briefly the principal morphological effects of inanition upon plants, on 
account of the wide range in the character of these organisms and the great 
differences in their mode of nutrition. In general, however, it will appear that 
the effect of inanition is to restrict or inhibit their growth during the develop- 
mental period, often resulting in premature development with the production of 
marked abnormalities of form and structure. In the poppy (Papaver) an inheri- 
tance of some of the experimentally produced variations is claimed. In later 
stages of growth the plants are usually less susceptible, but sooner or later a 
deprivation of nutriment will usually produce protoplasmic atrophy, with pro- 
gressively degenerative changes in the cells and tissues, finally resulting in 
the death of the organism. Of the cell constituents, the formed storage prod- 
ucts (starch, oil, etc.) are usually consumed first; then the cytoplasmic struc- 
tures are attacked; lastly the nucleus, which is the most resistant. 

These effects are produced not only by general or total inanition (either com- 
plete or incomplete) but also often in a strikingly characteristic manner by par- 
tial inanition, when there is a marked deficiency of only one (or a few) of the 


numerous essential factors in plant nutriment. Lack of any one of these factors 
water, calcium, potassium, magnesium, iron, phosphorus, sulphur, manganese, 
nitrogen, carbohydrates and possibly vitamins— will cause stunting of growth 
with variable, degenerative cell-changes. These are expressed by morphological 
and physiological derangements, often resulting ultimately in the death of the 
organism. The effects are apparently most severe in the case of phosphorus or 
nitrogen deficiency. 

In addition to the deleterious effects more or less common to all these 
partial deficiencies, there are in each case certain peculiarities due to the special 
functions which each of the food-elements normally performs. These peculiari- 
ties appear also to vary considerably in different classes of plants. 

Thus water deficiency (aqueous inanition) is usually expressed promptly by 
characteristic changes in form and structure of plants, on account of its funda- 
mental importance in morphogenesis and adult structure, as well as in transporta- 
tion. Of the various salts, those of calcium, potassium and magnesium are 
especially essential for chlorophyll production and starch formation. Phos- 
phorus appears to be more concerned with the transformation than with the 
origin of carbohydrates. Cellulose formation proceeds in the absence of 
phosphorus, but is impossible without calcium. Mitosis may occur in the 
absence of calcium or of magnesium, but not without potassium or phosphorus. 
The process of inflorescence and the development of the sexual organs in general 
appear to be unusually susceptible to the effects of malnutrition, and in some 
cases (especially in fern prothallia) a deficiency of calcium or nitrogen may 
influence sex by inhibiting the development of the female organs (archegonia) . 
Numerous other instances of characteristic effects of the various types of partial 
inanition in plants will be cited in the following pages. 

Some of these effects of inanition upon plants, especially with reference to 
cellulose, chlorophyll and starch production, are peculiar to the plant kingdom. 
Most of the effects, however, resemble those found in animals. A study of these 
effects is therefore of interest to the biologist, since the phenomena common to 
plants and animals indicate characters of fundamental importance. A knowl- 
edge of these fundamental characters is likewise useful in the analysis of 
conditions found in the higher animals, and is essential to the comprehension 
of many of the complex problems of human medicine, as will appear in later 

The effects of inanition on plants will be considered in detail under (A) total 
inanition (or on water only), and (B) partial inanition. Although subsistence on 
water alone is, according to definition, a form of partial inanition, the results 
are similar to those of total inanition, and are grouped with them for convenience. 

(A) Effects of Total Inanition, or on Water Only 

According to Winkler ('13), the theory has recently been emphasized (by v. 
Gobel, Klebs and Sachs) that the morphogenesis of plants depends chiefly upon 
t heir metabolic activities, in which case the production of abnormal forms through 
inanition is to be expected. Palladin ('18) concluded that "each external condi- 


tion — such as heat, light, atmospheric pressure, humidity, gravitation, and the 
supply of nutrient material — exerts an influence upon plant growth, and conse- 
quently upon both external form and internal structure." 

As to the duration of starvation, Coupin ('21) observed that seedlings, which 
after germination were kept in distilled water in the dark, lived the following 
number of days before death from inanition: nut-bearing pine, 60; pumpkin, 46; 
winter vetches, 44; lentils, 40; marvel of Peru, 39; peas, 33; beans, 32; sunflower, 
30; buckwheat, 25; radish, 24; nasturtium, 23; spinach, 22; tomato, 21; beet, 20; 
common cress, 18; mustard, 18. Thus considerable variation occurs, which is 
ascribed by Coupin to variations in the resistance of the plant itself, and espe- 
cially to the quantity and quality of the reserve material available. London 
('97) found that active bacteria may survive without food for 49 to 88 

As to the effect on size, it is well known that a decrease in size (or a stunted 
growth) is the general effect of insufficient nutriment in plants; but the result is 
usually less obvious than in animals on account of the rigidity of the cell walls 
in plants. Quantitative studies on this topic in plants appear scarce. London 
('97) used a centrifuge in measuring the volume of bacteria {Bacillus anthracis, 
B. subtilis and Streptococcus pyogenes), and found during inanition a progressive 
loss amounting to an average total of 51 per cent (range 27-72 per cent). Some 
measurements of retardation in weight of plants in the absence of certain growth- 
promoting substances (vitamins) were made by Bottomley ('14). 

According to Winkler ('13), the dwarfing (nanism) due to inanition does not 
always reduce the plant organism proportionately in all parts, but morpho- 
logical peculiarities may occur (Gauchery, Kraus). Such modifications are 
generally not hereditary, although de Vries ('00) obtained inheritance of 
experimentally produced variations in Papaver sommiferum polycephalum. 
According to Thomson ('88) and Rignano ('11), Hoffmann's ('87) researches 
prove the inheritance of variations (such as relatively large number of atypical 
flowers) produced by insufficient nourishment in Papaver, Migella and 
Argemone. Inflorescence may also occur prematurely (Winkler) or in great 
profusion (Gagnespan '19) as a result of starvation. 

Another effect of inanition upon the reproductive mechanism is possibly 
the modification of sex (extensive review of earlier literature by Strassburger 
'00, and 0. Schultze '03). Various botanists have claimed that the prothallia 
of ferns grown under unfavorable nutritive conditions produce only antheridia 
(male organs) , and no archegonia (female organs) . A suppression of the develop- 
ment of female organs by malnutrition was observed in maize (Zea mays) 
by K. Muller ('64) and Cugini ('80); and in Equisetum by Buchtien ('87) and 
others. The relation of inanition to maleness is discussed by Hoffmann ('85). 
Heyer ('84) was skeptical as to the modification of sex in Mercurialis by environ- 
mental factors, though Klebs ('95, '96, '03) obtained positive results in some 
Algae and Fungi. According to Geddes and Thomson ('01) : " The experiments 
of Klebs may perhaps be regarded without unfairness as marking the real begin- 
ning of a physiology of reproduction in plants. For he has set himself to show how 
definite environmental conditions of nutrition, temperature, etc., are definitely 


associated with the occurrence of particular modes of reproduction in Algae and 

DeVries ('oo) noted that the extent to which the stamens are metamorphosed 
into carpels in Papaver somniferum polycephalum is a highly variable character, 
and is determined by external (especially nutritional) factors at a certain critical 
period, about the seventh week of development. Influences at earlier or later 
periods are ineffective. This may perhaps explain some of the conflicting results 
which have been obtained in other forms. 

There is also clearly a difference between the monecious and diecious 
forms, in respect to their susceptibility to sex-modification by environmental 
changes, as has been emphasized by Strassburger ('oo) and O. Schultze 
('03), who concluded that the sex (in diecious forms) is predetermined 
in the ovum, and apparently unmodified by environmental changes. 
Thus Strassburger could obtain no modification of sex in Melandrium. Noll, 
while getting positive results in the usually monecious Equisetum Telmateja 
(only male prothallia developing in cultures without phosphates), had negative 
results with the diecious Marchantia polymorpha. 

Coulter, Barnes and Cowles ('11) cited numerous examples of the reproduc- 
tive process as influenced by various unfavorable conditions, including dessica- 
tion and decreased food supply. The effects vary greatly in different species. 
In moulds generally the formation of asexual spores is favored by dessication 
and starvation (and of zygospores by the opposite conditions), but the sexual 
mode of reproduction is induced by food scarcity in Saprolegnia. According to 
Morini ('85), a reduction of nutrition appears necessary for zygosporulation in 
the Ustilagineae. As to sex determination, Coulter, Barnes and Cowles stated 
that in diecious plants undernourishment and xerophytic (dry) conditions 
apparently facititate the development of male (staminate) plants, but other 
factors must be considered. Recent evidence indicates that in diecious plants, 
as in animals, sex is determined at a much earlier period than was formerly 
supposed, being usually predetermined in the gametes, independently of external 
factors. Apparent change of sex seems best explained by assuming that such 
forms are at least potentially bisexual, and that external factors may either 
cause suppression of one of the sexes (as in Zea) or stimulate development of the 
sex commonly suppressed (in Carica). 

Love ('09) found that in peas a decrease in the food supply decreases the 
coefficient of variability in yield and number of internodes, but increases the 
variability in height. 

Data upon the cytological effects of starvation in plants appear relatively 
few, and have been described chiefly in connection with various forms of partial 
inanition (to be considered later). Cunningham ('80) found protoplasmic 
atrophy and fatty degeneration in the mycelium and fruiting organs of certain 
moulds (Choanephora and Pilobolus crystallinus) kept in distilled water. Bok- 
orny ('92) likewise found cytoplasmic atrophy, as well as changes in fat content, 
cell-sap, etc. of Spirogyra cells (see also later under potassium deficiency). 
Kosinski ('02) found in fungus cells {Aspergillus niger) during starvation dim- 
inished respiration and a gradual consumption of the plastic formed material to 


yield energy. Growth ceases, but returns on refeeding. Hottes observed that 
in developing beans, upon the removal of the cotyledons with their food supply, 
the meristematic tissue which would normally produce the lateral roots is 
transferred to the tip of the root and there used for growth. In 3 or 4 weeks 
all the cells in the upper part of the root have lost most of their cytoplasm, and 
the only actively functioning cells are those at the tip. Decrease in the size 
of the root is due to decrease in the number of cells, rather than to decrease in 
their size. Winkler ('13) also concluded that cell size is not greatly affected as 
a rule by environmental factors, the size of a plant in general being due to the 
number rather than to the size of the constituent cells. Maige ('23), however, 
found a relation between the size of the nucleus and nucleolus and the amount 
of nutrition in the bean seedling. 

The effects of inanition upon the growth and differentiation (as noted above 
for sex) of plants also vary according to the age or stage of development of the 
organism. Thus Davidson and Le Clerc ('18) found that the greatest increase 
in the yield of wheat occurs when the soil fertilizer (sodium nitrate) is added 
during the first stage of growth,' with a slighter effect in the second stage and 
none in the third. Hoagland ('19) similarly noted that the yield of barley is 
largely conditioned upon favorable supply and concentration of nutrients for the 
plant during the first eight or ten weeks of the cycle. Urbain ('20) observed that 
embryos deprived of their normal food (endosperm) at an early stage and placed 
in nutrient solutions are inhibited in their later development, being dwarfed 
and presenting various abnormalities (to be described later). 

Mode of Action. — Reed ('07) pointed out that the essential elements in 
plant nutrition appear to act in two ways: (a) as component parts of the cell 
structures or fluids; and (b) as indirect agents in causing less understood physi- 
cal or chemical conditions necessary for the proper functioning of the cell, 
whether as carriers of other ions, or as specific antidoting agents. Thus v. 
Liebig ('76) found that ammonia stimulates plant growth, provided the other 
necessary nutrients are present. Reed ('07) noted that algae thrive best in 
neutral or slightly acid cultures, while phanerogams thrive best in neutral or 
slightly alkaline solutions. Moore, Roaf, and Knowles ('08), in experiments on 
the hyacinth and onion, found that alkalinity in general apparently stimulates 
growth, producing increased nuclear division, changes in the chromosomes and 
obscure cell-outlines; while acidity causes decrease or absence of nuclear division, 
and thickening of cell-walls. Steinberg ('19) concluded that " Increased acidity 
of the Pfeffer nutrient solution within a certain range results in the exhibition in 
Aspergillus niger cultures of growth 'stimulation' like, but less in amount than, 
that observed by addition of salts of certain heavy metals." The mode of 
action in such cases is obscure and it is difficult to draw the line between growth 
stimulants (including the vitamins) and nutrients. 

(B) Effects of Partial Inanition 

Heretofore we have been considering chiefly the effects of inanition in general 
— of total inanition (excepting water), the nutriment being either entirely absent 
(complete inanition) or reduced in amount (incomplete inanition). We have 


now to consider the effects of partial inanition, only one or more of the essential 
food factors being deficient. Aside from the carbon dioxide used for starch 
formation in the green plants, the necessary factors include water and certain 
mineral salts, especially nitrates and phosphates of calcium and potassium. 
Magnesium, iron and sulphur also appear essential, at least in small amounts. 
Some plants (fungi and many algae) require also organic (protein) nutri- 
ment. The various factors are of unequal importance, however. Thus in the oat 
plant, Dickson ('i8) found that a deficit of phosphorus or nitrogen is much more 
injurious than a deficit of calcium, potassium or magnesium, which is in general 
agreement with the earlier results of Reed ('07) in various plants. The effects 
of the various substances upon plant morphogenesis and structure may be 
designated as " Chemomorphoses " (Winkler). 


Water, as is well known, is a fundamental necessity for both plants and ani- 
mals. Hygromorphosis is the primary factor in determining the form in many 
plants, both higher and lower. With reduction of the normal supply of water, 
many of the higher plants wilt through loss of turgescence due to osmotic pres- 
sure. Sagot noted that dryness has a marked effect upon the structure of 
plants. When growing plants are deprived of water, dwarfing and various 
abnormalities of form may occur. Some amphibious plants have a water form 
and a land form, apparently conditioned by the environment. 

Coulter, Barnes and Cowles ('11) classified plants, with reference to their 
water requirements, as xerophytes, mesophytes and hydrophytes. "It would 
seem that the chief determining factors of leaf size and proportion are those that 
control the water supply." In the xerophytes (associated with water defi- 
ciency), there is high transpiration through the leaves and low absorption of 
water through the roots, resulting in leaves of small size and great thickness. 
The thickness is due to a relatively large amount of cell-division in planes 
parallel with the surface. They cite many variations experimentally produced 
in the leaves and stems, the latter being shortened and thickened by dessication. 
Upon the reproductive structures, dessication usually has an effect similar to 
that of other unfavorable external conditions, frequently stimulating reproduc- 
tion, but often inducing modifications of the normal process. 

Palladin ('18) likewise emphasized the marked differences in the dandelion 
{Taraxacum officinale) and broom (Genistica anglica) grown in moist and in dry 
air. "The difference is so great that they might be taken for distinct species." 
As Palladin pointed out, however, water deficiency may arise in either of two 
ways, diminished intake (through roots) or increased exit (transpiration through 
leaves). The two modes may result in different morphological effects, as shown 
by the experiments of Kohl ('86) on Tropaeolum majus (see accompanying 


Effects of Various Degrees of Moisture upon the Structure of Tropaeohim majus 

(Kohl '86) 

External conditions 


Anatomical characters 


size of 

Kind of 












Cells tangentially 
elongated; thin 
outer walls 







Cells radially elon- 
gated; thick outer 

Two adjacent 
layers well 






Cells almost cubical 

Poorly developed 






Cells very much 
elongated radially 

Less developed 
than in 2 

The general effects of the second type of water-deficiency (increased exit 
through transpiration) were summarized by Winkler ('13) as follows: 

"Werden Pflanzen in trockener Luft erzogen, so ist im allgemeinen das 
Langenwachstum verzogert, auch werden die Internodien weniger lang, dafiir 
aber dicker und ihre Zahl wird erhoht; die Festigkeit der Achsenorgane ist 
grosser; die Haarbildung an Blattern und Stengeln erscheint gefordert; die 
Wurzelbildung erfolgt reichlicher, der Blattfall, die Blute- und Fruchtbildung 
eher; die Epidermis-, Rinden- und Markzellen bleiben kleiner; die Bildung von 
Sekretionsorganen und Kristalzellen wird begiinstigt, die Entstehung von Kork 
und Sclerenchym beschleunigt, die Gefassbildung gefordert. Kultur in feuchter 
Luft hat gerade den umgekehrten Erfolg (Kohl, Eberhardt)." 

Dickson ('18) found that the water requirement in the oat plant varies 
considerably according to the proportions of the various salts found in the 
nutrient medium. The requirement is decreased by a deficiency of magnesium, 
slightly increased by a deficiency of calcium, and greatly decreased by a defi- 
ciency in potassium, phosphorus or nitrogen. 


That calcium is essential for the normal growth of the higher plants has long 
been recognized (e.g. by Stohmann '62, in the maize plant), but its function is 
not yet entirely clear. Loew ('92) considered calcium as one of the important 
mineral bases entering into the constitution of the proteins in the cell-nucleus 
and chlorophyll-bodies. Reed ('07) pointed out that calcium differs from 
potassium and phosphorus in forming but a small proportion of the actual living 
protoplasm; but it has varied and important functions, influencing processes and 
products into which it does not itself seem to enter. True ('22) recently 


" It appears that a certain quantity of Ca ions must be present in the medium 
for the maintenance of the chemical and functional integrity of the deeper 
King living parts of the cells of absorbing roots of higher green plants . . . 
When this necessary minimal supply of Ca ions is lacking ... the function of 
absorption is upset and a more or less marked leaching of ions from the plant 
follows. In the absence of this necessary minimum of Ca, the soil solution or 
culture solution may be rich in all other required ions, but these are useless 
to the plant. They are unabsorbable." 

Von Liebig ('76) stated that calcium is doubtless necessary for cellulose 
formation. Boehm ('75) in Phaseolus vulgaris, v. Raumer ('83) in Phaseolus 
multiflorus, and Molisch ('95) in Spirogyra noted that a calcium deficit inter- 
feres with the formation of new cellulose walls after cell-division. Reed ('07) 
likewise noted that in Spirogyra and roots of Zea mais kept in solutions without 
calcium salts, although nuclei divide mitotically, new cell septa are formed 
imperfectly or not at all. After two months in Ca-free solutions, most of the 
Spirogyra cells were found dead and those alive markedly degenerated. He 
states, however, that some investigators (Bruch '02) have found calcium unnec- 
essary for nutriment in fungi and lower algae, and even toxic in some cases; but 
all agree that seedlings in calcium-deficient solutions usually develop small 
leaves. Reed also found that in Spirogyra, Zygnema and Vaucheria, calcium 
appears necessary for the growth and activity of chlorophyll and chloroplasts 
(in agreement with Bokorny '95 and Loew '92); perhaps serving also as an 
antidote to the bad effects of magnesium, oxalic acid, etc. He further observed 
that spores of Gymnogramme sulphur ea, cultivated in calcium-free solutions, 
developed prothallia in which antheridia were produced in large numbers, but 
archegonia were absent. Controls were normal. In certain mosses (Atri- 
chum), he noted that the spores fail to germinate in calcium-free solutions, 
although a retarded development (with peculiar moniliform cells) occurs when 
the calcium is replaced by sodium. Palladin ('18) emphasized the importance 
of calcium for normal leaf development of plants. In the development of the 
oat plant, Dickson ('18) found that a deficiency of calcium increased the general 
vigor of growth, lengthened the period of development and increased the total 
dry weight of the plant. The grain production was lowered, however, the ratio 
of grain to straw and the weight of the individual kernels being decreased. 


According to Reed ('07), de Saussure, in 1804, established the necessity of 
potassium for the growth of terrestrial plants, which has been confirmed by 
Nobbe ('71) and many later investigators. Dickson ('18) found that in the oat 
plant a deficiency of potassium in the nutrient solution causes a decrease in the 
general vigor of growth, a shortened period of development and a decrease in 
the total dry weight of the plant. The grain production is also lowered, 
although the ratio of grain to straw is increased and the weight of the individual 
kernels greater. 

Cytological effects of potassium deficiency upon algae (chiefly Spirogyra) 
were noted by Bokorny ('92). He found the chains of cells broken up into small 


segments or individual cells; cells shorter, and more turgid, with bulging ends. 
Atrophy of the cytoplasm and chlorophyll bands occurs, and changes in the fat 
content, cell-sap, etc. may appear. Molliard and Coupin ('03) found that when 
Sterigmatocystis (Aspergillus) niger is grown in the absence of potassium, the 
normal form of the conidial apparatus disappears, variable outgrowths replacing 
the normal conidiophores. Prothallia of Gymnogramme sulphurea are able to 
grow and form chloroplastids with only traces of potassium present, but no 
starch is formed. In cells of Hydrodictyon and Basidiobolus ranarum on K-free 
media, the cytoplasm becomes abnormally vacuolated and the nuclei indistinct. 
In Zygnema filaments the pyrenoids rapidly lose starch, the chloroplasts becom- 
ing vacuolated and the radiating strands of protoplasm disappearing. In 
Spirogyra cells kept in K-free solution, after 35 days no mitosis occurs, although 
both cells and nuclei undergo preliminary elongation. The inhibition of mitosis 
cannot be attributed to the lack of potassium in the nucleus, however, since 
Macallum ('05) has shown that potassium is normally absent from the nucleus 
in both plant and animal cells. 

Reed, in agreement with previous investigators (see literature reviewed 
by him), found that in some cases the function of potassium can be partially 
accomplished by the substitution of sodium. For example, when a subminimal 
amount of potassium is present, some evidence indicates that certain moss-spores 
can germinate and utilize sodium during the embryonic stages of development. 
Dassonville ('98) found that the substitution of sodium for potassium in the 
wheat plant produces less growth but more lignification; in the secondary roots 
of the tomato certain characteristic histological changes result. 


Although von Liebig ('76) and earlier investigators recognized that magne- 
sium is essential for the growth of the higher plants, its function long remained 
somewhat obscure. Loew ('92) thought it served as a carrier for phosphoric acid 
in the formation of nucleoproteins. Willstaetter ('06) has proved that magne- 
sium is a constituent of the chlorophyll molecule. Numerous investigators 
have noted the antagonistic action of magnesium and calcium (literature 
reviewed by Reed), including possibilities of their substitution for each other 
in some cases. 

Von Raumer ('83) noted in Phaseolus multiflorus deprived of magnesium a 
stunted growth of stem and leaves, with unhealthy chloroplasts. He con- 
cluded that magnesium is essential for starch transportation. Bokorny 
('95) with magnesium deficiency in various algae likewise found modifications 
of the chloroplasts, with markedly decreased size of nuclei. Reed ('07) observed 
that the ratio of magnesium and phosphorus is an important factor in spore 
formation in Aspergillus niger, and in Vaucheria the oil droplets disappear in 
Mg-free cultures. In Spirogyra, the chloroplasts become, variably disarranged 
and retracted into an irregular cytoplasmic mass near the center of the cell, 
although the pyrenoids and nuclei appear unaffected. Mitosis is still possible- 
though greatly retarded. According to Dickson ('18), a deficiency of magne, 


sium in the medium apparently increases the general vigor of growth, lengthens 
the period of development and increases the total dry weight of the oat plant. 
The grain production is lowered, however, the ratio of grain to straw and the 
weight of the individual kernels being decreased. Thus the results are closely 
similar to those of a calcium deficit. In general, the results of magnesium 
deficiency also resemble those of etiolation, when plants are grown in darkness 
(to be discussed later). This is probably because of the interference with 
chlorophyll formation or function in both cases, a conclusion which is supported 
by recent studies on the effects of magnesium deficiency in tobacco and maize 
by Garner, McMurtrey and Moss ('22). 


Von Liebig ('76) recognized the necessity for a certain amount of sulphur 
in the formation of proteins in plants, and of iron for chlorophyll-formation in 
green plants. He stated that although only a very slight amount of iron is 
necessary, in its complete absence the plants become yellow and stunted in growth, 
a condition known as chlorosis. Knop ('64) had found iron necessary for the 
growth of buckwheat and grasses in general, though apparently not for maize 
and peas. Gile and Carrero ('16) concluded that in the absence of special 
precautions it is often possible that the rice plant may be inhibited in its growth 
by an insufficient supply of available iron; a view also in accordance with the 
experience of Hoagland ('19) for the barley plant. The exact role of iron in 
plant growth is not clear, for it does not enter into the chemical composition of 
chlorophyll (Willstaetter). McHargue ('22) found that in sand cultures man- 
ganese is necessary for plant growth and development to maturity. In cul- 
tures without manganese, growth ceased in about six weeks; the plants became 
chlorotic, with death of the young leaves and buds. 


Ville ('61) was one of the first to demonstrate (in sand cultures) the indis- 
pensability of phosphates for all plant growth. This view has been universally 
adopted, the possible substitution of arsenic (Bouilhac '94) having been refuted. 
Loew ('91) found that when phosphorus is withheld the cells of Spirogyra cease 
to grow and the chloroplasts turn yellow. Starch formation continues for some 
time, however, and fats and proteins accumulate in the cells. Becquerel 
('04) and Schoene ('06) found it possible to germinate moss spores in solutions 
lacking phosphorus; but the development of the protonema is inhibited, and 
the rhizoid system is abnormally extensive. 

Reed ('07) found that of all the elements the lack of phosphorus seems to be 
the most injurious, its influence predominating in normal cell metabolism. 
In the spores of moss (Atrichum) no germination occurs without phosphorus. 
The thickness of the cellulose cell walls is increased in Basidiobolus, and in 
Spirogyra may even be doubled. In the latter, a few cells elongate, as in the 
K-f ree cultures, but no cell division occurs. (On adding a few drops of potassium 


phosphate, mitosis appeared in one hour.) The cells show progressively degen- 
erative changes in three stages: (i) the cells become cloudy, due to the accumula- 
tion of fat droplets (ascribed to lack of transformation into lecithin in the 
absence of P); (2) the chlorophyll bands become disarranged; (3) chlorophyll 
disappears from the chloroplasts and the cell contents become completely dis- 
organized. Reed believed that phosphorus is more closely connected with the 
transformation than with the origin of carbohydrates; and that in its absence 
abnormal transformations occur. No especial susceptibility of the nucleus to 
phosphorus deficiency has been observed, which is surprising since this element 
is an essential component of the nucleoproteins (chromatin). 

Moore, Roaf and Knowles ('08) observed that in hyacinth and onion the 
phosphatic ions have a peculiar effect on inflorescence. In the oat plant, Dickson 
('18) found the general development of the plant severely affected by deficiency 
of phosphates. The vigor of growth is greatly decreased, the developmental 
period shortened, and stooling prevented; the total dry weight of the plant is 
diminished, and grain production lowered, although the weight of the individual 
kernels and the ratio of grain to straw are increased. 


Nitrogen ranks with phosphorus as of primary importance, being an essen- 
tial constituent of protoplasm in both plants and animals. Aside from those 
cases. where atmospheric nitrogen is fixed by the aid of the nitrifying bacteria, 
plants require nitrates or higher organic compounds of nitrogen. In spite of 
the fundamental importance of nitrogen in plant nutrition, however, compara- 
tively few data appear as to the morphological effects of its absence. In ger- 
minating moss spores, Schoene ('06) found: "Bei Stickstoffhunger schreitet 
Funaria zu einer machtigen Ueberlangung des Rhizoidsystems unter voll- 
standiger Unterdriickung des Chloronemas, die tibrigen Moossporen entwickeln 
sich zu chlorophyllosen Hemmungsbildungen." Winkler ('13) cited several 
examples to illustrate the dependence of form upon the quality of food in plants 
which require organic nutriment (fungi and many algae). No descriptions of 
cytological changes as a result of nitrogen deficiencies in plants have been found 
in the literature. Prantl ('81) noted that fern prothallia cultivated in nitrogen- 
free solutions produce antheridia only (archegonia also appearing in controls on 
complete nutrient solutions). Dickson ('18) found that in the oat plant the 
effects of a nitrogen deficit resemble those of a phosphorus deficit, either one 
severely crippling the development and preventing stooling. The develop- 
mental period is thereby lengthened and the total dry weight of the plant as 
well as the grain production is lowered, although the weight of the individual 
kernels and the ratio of grain to straw appear increased. 

Urbain ('20) obtained some interesting results in numerous plants, including 
wheat, oats, barley, Mirabilis jalapa, Spinacea cleracea and Pinus pinea. The 
grains were soaked in water, and at successive intervals of time the embryos 
were removed from the endosperm (thereby depriving them of their normal 
organic nutriment, vitamins, etc.), and placed in artificial nutrient solutions. 


It was found that the endosperm is not absolutely essential, since without it 
the embryos develop, though at a retarded rate and with marked changes appear- 
ing later. The roots are less branched; stems simpler; leaves fewer, simpler and 
smaller; inflorescence precocious and less well developed, showing various 
abnormalities; fruits often aborted; all parts dwarfed. Comparison of sections 
with those of normal controls in Nigella, Papaver, Solarium, Torilis, and Zea 
showed that the internal structure also is much simpler in plants deprived of 
their endosperm. In the stem, the cells are fewer and simpler; the cortex 
shows fewer layers; the tissues of the central cylinder are less differentiated, the 
vascular bundles being relatively undeveloped; the pith is relatively large. 
Similar modifications occur in the root and petiole. In the leaf the epidermis 
appears nearly normal, but the other tissues show a reduction in the number and 
size of their elements. The greater the dwarfing, the more pronounced are 
these modifications. The modification in the growth of beans observed by 
Hottes upon the removal of the cotyledons was mentioned previously. These 
are most striking examples of the profound morphological changes resulting 
from an early partial inanition in the higher plants. 


It. has long been known that a green plant deprived of light undergoes a 
series of striking changes, designated as " etiolation." These changes are 
associated with, and at least partly dependent upon, the lack of carbohydrate 
food normally produced through the chlorophyll. The plant is thus thrown 
upon its stored food material for nutrition, much like an animal during starva- 
tion. In producing etiolation, Sachs ('87) believed that this interference with 
nutrition is the primary factor. MacDougal ('03) while recognizing this factor, 
was inclined to lay greater emphasis upon other factors, such as the withdrawal 
of the direct effect of sunlight upon the living organism. Palladin ('18) similarly 
considered etiolation as due only partly to deficiency of the organic assimilation 
products, partly to lack of direct effects of sunlight, and partly to results of 
diminished transpiration, with disturbance of the water balance. 

While the relative importance of the carbohydrate inanition factor in pro- 
ducing etiolation is thus somewhat uncertain, the changes in the plant as a result 
of etiolation are very marked. In addition to the yellow leaves and white stems 
(green chlorophyll lacking in the absence of sunlight), von Sachs ('87) noted 
that phanerogam seedlings develop longer shoot-axes and smaller leaves; and 
that after a variable length of time growth ceases and death results in the mal- 
formed, diseased plant. Further details in the process have been added by later 
investigators, notably MacDougal ('03), who made extensive observations 
upon the effects of etiolation in a large number of plants. He found that the 
form and structure of the plant are usually much altered, but the results are 
widely divergent in various types. There is an abnormal differentiation of 
tissues, " some being suppressed, others accentuated and in some instances new 
tissues arising. Variations occur in the form, size and number of the elements, 
the structure and characteristics of the walls being materially different from the 


normal, while the protoplasts differ chiefly in the character of the inclusions and 
the composition of the vacuolar fluids." For further details, MacDougal's 
monograph may be consulted; also Palladin ('18). The results of etiolation, 
where there is a carbohydrate deficiency, may be contrasted with the converse 
experiments of Urbain (previously mentioned), in which the reserve food material 
(protein, fats, etc. of the endosperm) was removed and other nutrients fully 


The vitamins, or accessory food factors (of unknown composition) so essen- 
tial in animal nutrition are, as is well known, derived directly or indirectly from 
plants. Funk ('22) and Sherman and Smith ('22) have recently reviewed at 
length the literature covering numerous researches which indicate the necessity, 
or at least the advantage, of the vitamins or apparently similar growth-promoting 
substances in the development of plants themselves, including yeast, bacteria, 
fungi and higher plants. It is quite possible, for instance, that the remarkable 
changes which Urbain ('20) effected by removal of the endosperm are at least 
partly due to the loss of vitamins stored to supply the needs of the plant during 
development. The fully-grown plants are apparently able to synthesize these 
vitamins (whatever they may be), possibly in some cases by the symbiotic aid 
of nitrifying bacteria. Possibly some of the growth-promoting effects of minute 
quantities of certain substances may belong in a similar category, as shown, for 
example, in the experiments of Bottomley ('14) on wheat seedlings. It remains 
for the future to reveal the chemical nature and mode of action of these obscure 
substances. Thjotta and Avery ('21) concluded that two distinct vitamin-like 
substances are required for the growth of the hemophilic bacteria. Ellis and 
Macleod ('22) have recently reviewed the literature on the relation of vitamins 
to the growth of yeast. 

Liebig's Law of the Minimum 

In connection with the inanition of plants, it seems desirable to consider 
what is generally referred to as Liebig's "law of the minimum" or the "limiting 
factor," because of its fundamental importance in animal as well as plant 
nutrition under conditions of partial inanition. Since this doctrine has occa- 
sioned much controversy, it is advisable to examine somewhat carefully Liebig's 
original ideas, which have been frequently misunderstood and greatly modified 
by subsequent investigators. 

In the edition which I have consulted (Die Chemie in ihrer Anwendung auf 
Agricultur and Physiologie, 9. Aufl. Braunschweig, 1876), v. Liebig approaches 
the question from the point of view of fertilizers and their application to obtain 
the maximum harvest of crops. Under "Lehre (Gesetz) des Minimums" 
(p. 330 ff .) he explains how soils vary in composition and how successive crops 
may exhaust the soil by removing certain of its constituents which form the 
essential nutrients for plants. In the application of manures or fertilizers to 
enrich an impoverished soil, he emphasizes the importance of a knowledge of 
its content in the various essential plant nutrients, all of which are indispensable 


(in variable amounts) and none of which can be substituted for each other. As 
v. Liebig puts it: 

"Ein jedes Feld enthalt ein Maximum von einem oder mehreren und ein 
Minimum von einem oder mehreren Nahrstoffen. Mit diesem Minimum, sei 
es Kalk, Kali, Stickstoff , Phosphorsaure, Bittererde, oder ein anderer Nahrstoff , 
stehen die Ertrage im Verhaltniss, es regelt und bestimmt die Hohe oder Dauer 
der Ertrage. 1st dieses Minimum z.B. Kalk oder Bittererde, so werden die 
Ernten an Korn and Stroh, an Ruben, Kartoffeln oder Klee dieselben bleiben 
und nicht hoher ausfallen, auch wenn man die bereits in Boden vorhandenen 
Kalis, der Kieselsaure, Phosphorsaure, etc., urn das hundertfache vermehrt." 

Furthermore, field experiments are cited to prove that when enough of the 
minimum constituent is added to bring up its proportion to that of the next 
lowest constituent, further addition of the first constituent has no further favor- 
able effect upon the yield. 

While it is impracticable to follow in detail the controversy which has fol- 
lowed as to the validity of this law in plants ( and later as extended to animals), 
the result may be summed up briefly in the statement that, as formulated and 
applied by v. Liebig, the law in general remains well established. This does 
not mean, however (as some have claimed), that no growth whatever will occur 
in the absence of any factor essential for normal nutrition, or that growth under 
restricted nutriment must always result in normally proportioned dwarfs, 
growth remaining normal until the limit is reached. Numerous examples of 
pathological development under such conditions have been cited in the foregoing 
pages, and some instances of such were recognized by v. Liebig. 

Furthermore, this law (as strictly interpreted) must be modified by the recog- 
nition that growth may be limited or affected by many factors other than the 
amount of the minimum nutrient. The proportions of the other nutrients 
present, conditions of acidity or alkalinity, and other factors (some of which 
were elsewhere cited by v. Liebig) certainly have more or less influence in deter- 
mining the amount, character and outcome of growth. 

Thus the question becomes largely a matter of definition. In the sense 
intended by v. Liebig, the law still holds good in its general application to 
fertilizers and farm crop yields. If more strictly interpreted and extended, 
however, it certainly requires modifications. From this point of view, one must 
agree with the critics, such as Mitscherlich ('20), to take a recent example, who 
formulates a law of physiological relation of growth factors. He properly 
insists that, strictly speaking, there can be no such thing as any one minimum 
factor alone determining the amount of growth production, since the influence 
of all growth factors together must be considered. The significance of this 
question for animal growth will be considered in later chapters. 



The effects of inanition on the protozoa are of especial importance, since 
the changes can be directly observed under the microscope in these unicellular 
forms. These changes are of extreme importance in interpreting the effects of 
inanition in the higher organisms, including man. In the present chapter, 
some of the more important results of inanition among the protozoa will first be 
indicated briefly in relation to duration of starvation, changes in cell size and 
form, endoplasm and ectoplasm, nucleus, reproduction, and recovery upon 
refeeding. The changes will then be explained more in detail in the various 
species of protozoa. 

Summary of the Effects on the Protozoa 

Period of Survival. — The period of survival during inanition among protozoa 
in general is usually quite limited, being only a few (four or five) days in Stentor 
and Rhizopoda without symbiotic algae. Amoeba terricola may survive ten 
to twenty days; Amoeba proteus and Trichosphaerium, two or three weeks. 
Paramecia may survive from three to twenty-one days. The minimum repre- 
sents Lipska's observation for the (exhausted?) individuals immediately after 
conjugation. Her research indicates that five to seven days more nearly repre- 
sents the average for complete inanition. The prolonged period is for excep- 
tional individuals or (usually) represents incomplete inanition, where some food 
is available. Allescher's observations on Dileptus indicate that temperature is 
an important factor, the duration period increasing from four or five days at 
6°C. to a maximum of seven to twelve days at 15 , then decreasing to five or six 
days at 25-30 . Death in starving Paramecium sometimes occurs suddenly, 
by rupture of the cell-membrane, but as a rule slowly, by gradual atrophy and 
degeneration. Death from starvation may be escaped by encystment (in Didin- 
ium) or by conjugation or encystment (in Dileptus). 

Change in Size and Form of the Cell. — All the evidence shows a marked 
reduction in the size of protozoa during inanition. The maintenance of the 
original size of Noctiluca is merely apparent, the increase in peripheral vacuoles 
masking the decrease in cytoplasm. The recorded data indicate a decrease in 
length to about one-half of the original in Paramecium, one-fifth in Stentor and 
one-tenth in Dileptus and Pleurotrichia. Thus in these extreme cases the volume 
may be reduced to less than 1 per cent of the original. There are also more 
or less marked and variable changes in form accompanying the diminution in 
the size of the body. 

Changes in the Endoplasm. — The cytoplasmic changes at first appear chiefly 
in the endoplasm, which becomes progressively reduced in amount and 



transparent, on account of the disappearance of the food- vacuoles and similar 
inclusions. In the later stages of inanition, a progressive vacuolation of the 
endoplasm has been described in Trichosphaerium, Didinium and Colpidium; and 
in Paramecium by all recent observers except Lipska. Such vacuolation of the 
endoplasm apparently does not occur in Noctiluca and Pleurotrichia, however, 
and Lipska's results indicate that even in Paramecium it is probably only an 
indirect effect of inanition, being due primarily to other environmental factors. 

Changes in the Ectoplasm. — In all cases the cell-membrane and associated 
ectoplasmic structures (cilia, trichocysts, cytopharynx, etc.) appear more 
resistant than the endoplasm, but in the later stages of inanition they also may 
be attacked and partially resorbed (in Noctiluca, Didinium and Paramecium). 
Associated with these regressive changes, there is a progressive decrease in 
motility and in other vital phenomena. 

Nuclear Changes. — The nucleus in general is much more resistant than the 
cytoplasm, but may show changes in form with loss of chromatin content (in 
A moeba, Gregarina, Didinium, Stentor, Dileptus) . In forms such as Paramecium, 
with macronucleus and micronucleus, the former shows more distinct changes. 
In the earlier stages of inanition it frequently elongates and enlarges. Later it 
usually divides, and may show degenerative changes, with granulation, vacuola- 
tion, fragmentation and variable extent of resorption. The micronucleus usu- 
ally persists practically unchanged, although rarely it may divide with reunion 
of the daughter nuclei. This persistence of the nucleus is a factor of great 
importance for the survival of the organism during periods of inanition. The 
less important constituents of the organism are usually consumed first, the most 
essential apparently survive longest. The persistence of the nucleus during 
starvation recalls a similar behavior when living cells are engulfed and digested 
(e.g., by Trichosphaerium, according to Schaudinn) and in trypsin digestion 
experiments. This may be of significance as indicating the presence of similar 
enzymes during the autolysis of cells in starvation. 

Effects on Reproduction. — The relations of inanition to reproduction in 
Protozoa have attracted much attention. Many observers have noted an 
apparent stimulus to division resulting from a brief period of starvation (in 
Stentor, Didinium and Paramecium). R. Hertwig ('99, '03, '03a), however, 
observed that while under certain conditions fasting protozoa may divide more 
readily than well-nourished, as claimed by Jickili ('02) and others, this does not 
occur as a general rule. In Paramecium, which has been most extensively stud- 
ied, divisions occur to a variable degree in the early stages of inanition, but 
rarely or never in the later stages. According to Hertwig's theory, inanition 
upsets the cell-equilibrium as expressed by the nucleus-plasma ratio; but the 
relative increase in the nucleus may be equalized later by a reduction in chro- 
matin, associated with the process of conjugation. The literature on the nucleus- 
plasma ratio is reviewed by Erdmann ('12). 

Rolph ('84), who proposed a nutritive theory of sex, claimed that conjugation 
occurs in protozoa when conditions result in an interference with their nutrition. 
Thus sex is considered as primitively a form of hunger, which drives the organism 
to engulf its neighbor ("isophagy"). A similar theory of sex was elaborated 


by Geddes and Thompson ('01). The studies of Maupas ('88, '89) and of 
numerous more recent investigators, however, have shown that the life cycle of 
the protozoa, with its phases of sexual and asexual reproduction, is apparently a 
very complicated process, involving both internal and external factors. Among 
the latter, inanition is doubtless an important factor, which under certain condi- 
tions may induce conjugation. This was observed by Maupas and others in 
numerous species of Infusoria. Hertwig discovered that when low temperature 
cultures (which have a relatively large nucleus) are placed at a high temperature 
and kept without food, an artificial "depression" is produced, which leads to 
conjugation. This was confirmed by Prandtl ('06) for Didinium, and by Popoff 
('07) for Epistylis. Conjugation has been observed in the earlier stages of 
ordinary inanition in Paramecium by Kasanzeff ('01), Wallengren ('02) and 
Chainsky ('06); but Calkins ('02) andLipska ('10) found inanition very unfavor- 
able to conjugation. It is evident that further research is necessary in order to 
clear up this fundamental problem. 

Recovery upon Refeeding. — The protozoa show an astonishing capacity for 
recuperation upon careful refeeding after inanition. Maupas ('88) found that 
an atrophic Stentor regained its original size in two days of abundant realimenta- 
tion, which evidently involved an increase of over 100 times its reduced volume. 
Joukowsky ('98) observed a similar remarkable capacity for recuperation in 
Pleurotrichia. In Paramecium, Wallengren ('02) found a normal recovery 
possible, even in the greatly vacuolated condition after fifteen days of inanition. 
Lipska ('10), however, found recovery in general possible only up to four or five 
days of inanition. The process of recovery is the inverse of that during inani- 
tion, and cell-divisions begin after three to five days of refeeding. Nirenstein 
('10) observed regeneration of endoplasmic fatty granules in Paramecium 
within a few hours of refeeding. Jennings ('08) noted that variability in size is 
increased during refeeding, some individuals recuperating more rapidly than 
others; but the normal condition is eventually restored. 

Effects on the Various Species of Protozoa 

Protozoa Other Than Infusoria. — Among the unicellular animals constitut- 
ing the phylum Protozoa, only a few observations as to the effects of inanition 
have been made upon forms outside of the class Infusoria. Brass ('83) noted 
that in starving Amebae and Gregarinae, as well as in Infusoria, the chromatic 
nuclear substance becomes resorbed, serving (according to his interpretation) 
as a reserve nutritive material. Schaudinn ('99) observed that in Trichosphaer- 
ium Sieboldii (a species of Foraminifera) during starvation the first change is a 
retraction of the pseudopodia in two or three days, and a resorption of the nutri- 
tive granules. "Zugleich mit diesen Vorgangen beginnen die Zellkerne sich an 
einzelnen Stellen zu kleinen Gruppen zusammenzulagern. Nachdem das 
Plasma rein geworden ist, wird dasselbe grob vakuolisiert, und zwar scheint 
diese Vakuolisierung von der Peripherie gegen das Zentrum vorzuschreiten. 
Im weiteren Verlaufe vereinigen sich die einzelnen Kerngruppen zu einer einzigen 
grossen Gruppe, und die Zelle rundet sich kugelig ab. Die Vakuolisierung 


nimmt immer merhr zu, unci zwar werdern jetzt umgekehrt wie zu Anfang die 
zentralen Vakuolen immer grosser . . . Schliesslich zerfallt das Plasma (nach 
etwa 3 Wochen) in eigenthiimlicher Weise, indem es sich zunachst in wenige 
grosse Kugeln zerteilt, die wieder in kleinere sich auflosen, welche dann ganz 
verschwinden. Der Kernhaufen bleibt schliesslich allein in der zusammenge- 
falteten Gallerthiille iibrig, und leistet noch lange Widerstand." 

In Actinosphacriutn (a Heliozoan), and also in certain Infusoria, Hertwig 
('03a, '04) obtained by prolonged inanition a marked change in the nucleus- 
plasma ratio, the nucleus becoming relatively enlarged. The significance of 
this change will be mentioned later. Pearl ('06) found that poorly-nourished 
individuals of the flagellate Chilomonas are reduced in size and slenderer in form. 
Borowsky ('10) observed that Actinosphaerium eichhorni resists inanition 
14-18 days. The nuclei become fewer (apparently by fusion), and in advanced 
stages the cytoplasm undergoes marked vacuolation. 

Penard ('05) noted that Amoeba terricola may endure starvation for ten to 
twenty days; but that other Amebae and Rhizopoda, free from "zoochlorelles," 
perish in four or five days. Stole ('06) observed that in Amoeba proteus by 
withdrawal of food for four days (as well as by overfeeding and other environ- 
mental changes) binuclear cells are obtained, which return to the usual mononu- 
clear form upon restoring normal conditions. Gruber ('n) found that Amoeba 
proteus after six or seven days of starvation becomes sluggish, the plasma becom- 
ing denser and darker. The cell becomes decreased in size, although the nucleus 
gradually increases. After two or three weeks, the animal becomes rounded up, 
degenerates and dies. During starvation in Amoeba diploidea, Erdmann ('10) 
noted vacuolation and pigmentary degeneration. 

As an example of the effects of partial (calcium) inanition, Maas ('12) noted 
a reduction of the calcareous skeleton in the Foraminifera Bilocularia and 
Qiiinquelocularia in Ca-free sea water. Thompson ('17) stated (p. 415) that 
" when Foraminifera are kept for generations in water from which they gradually 
exhaust the lime, their shells grow hyaline and transparent, and seem to consist 
only of chitinous material." 

Pratje ('21) has recently made an extensive and careful study of the changes 
found in Noctiluca miliaris kept without food and observed alive. The normal 
fat-droplets disappear gradually (in eight to fourteen days), so that the cyto- 
plasm becomes transparent. The central cytoplasmic mass is greatly reduced 
in amount, with fewer and thinner radial extensions to the periphery. There is a 
corresponding increase in the peripheral vacuolar fluid, however, so there is no 
appreciable decrease in the size of the whole cell. The nucleus becomes more 
distinctly visible and more transparent. The nucleoli often become visible. 
Thus far only the more fluid endoplasm has been affected, but as inanition pro- 
gresses, the firmer, peripheral protoplasmic structures (cell membrane and 
organs) are attacked and consumed. In about three weeks, all available sources 
of energy are exhausted. Recovery by refeeding is now impossible, and death 
soon occurs. " Fassen wir unsere Ergebnisse der Hunger versuche bei Noctiluca 
zusammen, so finden wir, das im Hungerzustande zuerst die vorhandenen 
Nahrungsstoffe und Reservesubstanzen aufgebraucht werden und erst dann 


das Protoplasma im Innern der Zelle in Angriff genommen wird. Audi der 
Kern zeigt geringe Veranderungen. Nach dem fliissigeren Innenplasma, wird 
das festere periphere Plasma als Energiequelle fur den Stoffwechsel benutzt, 
sowohl Bestandteile der Zellmembran wie auch der Organellen, Staborgan, 
Tentakel, Zahn und Lippe." 

Infusoria. — Among the Infusoria, the Paramecium has been studied most 
extensively, but some observations have also been made upon other species. 
Gruber ('86) noted cell-divisions with decreased size in malnourished Stentor 
coeruleus. The phenomena were more carefully described by Maupas ('88), 
who experimented at temperatures of 23-26°C. with the following results: 
"Lorsqu'on prend un Stentor bien nourri et arrive a son maximum d'accroisse- 
ment, et qu'on l'isole en le privant completement d'aliments, il n'en continue 
pas moins a se multiplier. Mais chaque bipartition n'etant plus suivis d'accro- 
issement, la taille diminue rapidement et apres trois ou quatre divisions, on 
arrive a. ne plus avoir que de petits avortons mesurant 235/z en longeur et 105^ 
en largeur (PL XII, Fig. 14). Si on continue a laisser ces avortons sans nourri- 
ture, ils s'etiolent de plus en plus et finissent par perir apres deux ou trois jours. 
Mais en leur donnant une abondante nourriture, je les ai vus en deux jours, 
par une temperature de 24 a 25 degres, s'accroitre rapidement et reprendre leur 
taille normale de 1176M en longeur et 270^ en largeur (PL XII, Fig. 15)." 

Further observations were made by Sosnowsky ('99), who noted, in Stentor 
coeruleus fasting in distilled water, the appearance of nuclear vacuoles, often 
collapsing at the periphery so as to give an irregular nuclear contour. Allescher 
('12) in the same species found the rate of loss during starvation to vary greatly 
according to temperature. Thus the loss in volume after three days at 6°C. 
averages 18.53 per cent; at 15 , 41.21 per cent; at 25 , 60.03 percent. Frequency 
curves of distribution in size were made. ''Pure line" cultures show no reduc- 
tion in size variations. The nuclear substance shows no decrease in the 
cold (6°) cultures, but at higher temperatures there is a reduction in the 
size and number of the nuclear masses, and sometimes the nuclear chain 
is broken. 

In Colpidium, Miyoshi ('96) and Jensen ('99) observed during inanition a 
diminution in size and a disappearance of the cytoplasmic granules, with 
resultant increase of transparency. Wallengren ('02) also noted vacuolation 
of the endoplasm during the later stages of starvation in Colpidium, together 
with other changes similar to those found in Paramecium (to be mentioned later). 
Peters ('20, '21) observed that in cultures of Colpidium colpoda in synthetic 
nutrient mixtures the structure is not easily modified by chemical changes in 
the medium. The size of the organism may be reduced by inadequate nutrition, 
but is regained by new subcultures on adequate diet. Experiments indicate 
that ammonium phosphate and chloride, and also magnesium salts, cannot be 
omitted from the medium without stopping growth. Phosphate deficiency 
causes apparent disintegration, and potassium deficiency results in loss of 
movement, with final death. However, sodium, calcium or sulphates may 
apparently be separately omitted (excepting traces) without injurious 


In Pleurotrichia, Joukowsky ('98) described a remarkable reduction in 
length from 200^1 to 30^, or even 15/x, during inanition, with no visible degenera- 
tive changes; the body becoming enlarged again upon careful refeeding. In 
DUeptus gigas, Hertwig ('03) likewise observed a great decrease in size during 
starvation. Death may be escaped by conjugation or encystment, a result 
confirmed by Prandtl ('06). Hertwig's pupil, Allescher ('12) studied the changes 
in DUeptus in more detail. In "hunger cultures" at 6°C, the animals lived 
four or five days with estimated average loss of 56 per cent in volume; at 15 , 
seven to twelve days with loss of 97 per cent; at 25 , five days with loss of 98 
per cent; at 30 , six days with loss of 99 per cent in volume (length reduced to 
less than one- tenth, and breadth to one-third of the initial). The DUeptus 
has normally a very large number of minute nuclear particles, probably over 
1,000, which become reduced to 50 or 60 somewhat larger particles during 
starvation. The plasma is consumed more rapidly than the nuclear material, 
and becomes darker in color, due to formation of brownish pigment. 

In Didinium nasutum, Thon ('05) noted two stages of inanition similar to 
those found in Paramecium by Wallengren ('02). "Die Haupterscheinung 
der ersten Periode ist das Verdauen der Einschlusse und ^Condensation des 
Plasmas, in der zweiten wird das Plasma vakuolisiert und die Tiere encystieren 
sich." In the first stage nearly all the individuals divide, and the contractile 
vacuoles become enlarged (especially also in Spirostomum ambiguum) . In the 
later stages the organs associated with the cytopharynx may become disinte- 
grated and absorbed. The changes in the nucleus appear early and are quite 
variable. Its horseshoe form may be preserved or distorted. The nuclear 
membrane is persistent, but wrinkled; and the nucleoli lose their chromatin 
content. Cell-division becomes impossible and encystment follows. 

Encystment following inanition was likewise found by Root ('14) in Podo- 
pliyrae, and Mast ('17) usually obtained similar results in Didinium nasutum. 
Recently, however, Mast and Ibara ('23) conclude that inanition does not 
facilitate encystment in Didinium. 

Paramecium. — The effects of inanition upon Paramecium have been observed 
by several investigators. Verworn ('00) noted remarkable changes in starving 
paramecia. "Ihr Zellkorper war durch ein Anzahl grosser kugeliger Vakuolen 
nicht nur vollkommen deformiert, sondern auch in seiner Struktur in tief- 
gehendster Weise verandert." 

The first extensive study of starving paramecia was by Kasanzeff ('01), 
a pupil of R. Hertwig. He found that Paramecium caudatum survives six to 
nine days, with marked decrease in size. In five days, the average length 
decreases from 0.232 mm. to 0.147 mm., and the breadth from 0.084 mm - to 
0.042 mm. The endoplasm becomes transparent and the food-vacuoles dis- 
appear; but abnormal vacuolation occurs later. The macronucleus at first 
elongates, increasing in volume and in chromatin content; but in later stages it 
decreases greatly in size, extruding small yellowish masses into the endoplasm, 
and may finally become fragmented. Normal cell-division and conjugation 
usually occur only in the earlier stages of inanition. The micronucleus shows 
increase rather than decrease in size; it may divide with the later reunion of the 


daughter nuclei (as in conjugation), a process interpreted as "self-fertilization." 
Kasanzeff distinguishes two phases of inanition, the second phase beginning 
with the disintegration of the macronucleus. 

Wallengren ('02), a pupil of Verworn, studied "hunger cultures" of Parame- 
cium caudatum (also of Colpidium colopoda). He also divided the inanition 
period of about 15 days into two phases. In the first phase of 8 or 9 days, the 
reserve food material becomes exhausted. Through the disappearance of the 
food-granules and food-vacuoles, the endoplasm becomes transparent and is 
greatly reduced in volume. In the second phase, the endoplasm is attacked 
and numerous vacuoles of variable size arise, "die wahrscheinlich durch osmot- 
isch wirkende Zellprodukte entstanden sind" (see Fig. 6). The contractile 
vacuoles are reduced, and the ectoplasmic organs (trichocysts, cilia) become 
smaller and sometimes partially consumed. In the macronucleus appear 
granules which fuse to form a mulberry-like mass (nucleolus), which persists 
while the remainder of the macronucleus becomes progressively vacuolated, 
deformed, and often finally fragmented and largely resorbed. Of all the cell- 
organs, the micronucleus alone resists the destructive effects of inanition, 
although it may show changes preliminary to a normal division. Normal 
conjugation may occur during the early stages of inanition. "So schreiten 
also bei der einzelnen Zelle die Inanitionserscheinungen von den unwichtigeren 
Teilen zu der wichtigeren fort, die unentbehrlichsten halten am langsten stand." 

In size, Wallengren's paramecia were reduced in ten days from 0.25-0.3 mm. 
to 0.16-0.17 mm. in length, and from 0.07-0.08 mm. to 0.028-0.042 mm. in 
breadth. (The loss in Colpidium was relatively even greater.) In the later 
stages, the breadth may be increased by enormous vacuolation. Death may 
occur either suddenly, by rupture of the cell-membrane; or more slowly, with 
gradual decrease of the cell until death from exhaustion. On refeeding, the 
paramecia may gradually regain their normal size and structure, even when 
markedly vacuolated by fifteen days of inanition. 

Most of the observations of Wallengren upon Paramecium caudatum were 
verified by Calkins ('02, '04), who found that inanition is unfavorable to con- 
jugation. He also studied the somewhat similar periodic " depression " phenom- 
ena, and found them to occur independent of nutrition. 

Another extensive investigation was made by Chajinski (Chainsky) ('03, 
'06), who made over 15,000 observations on starving Paramecium caudatum. 
His results in general are very similar to those of Kasanzeff and Wallengren. 
In the earlier period, he found disappearance of the food-vacuoles, with succes- 
sive vacuolation of the endoplasm, and vacuolation (sometimes division) of the 
macronucleus. In the later stages, the ectoplasm also becomes vacuolated, the 
macronucleus greatly vacuolated and fragmented, and the entire cell markedly 
deformed. The micronucleus persists unchanged. Conjugations may occur 
in the earlier stage of inanition. 

Jennings ('08) made a careful and extensive statistical study of size, includ- 
ing the effects of varied nutrition, in Paramecium caudatum and Paramecium 
aurelia. "Pure line" cultures were utilized. A large number of P. caudatum 
were taken from a 24 hour hay-infusion culture, and were kept in a small 



quantity of the same fluid. At the end of eleven days, "They had evidently 
begun to starve; they were small and thin, and almost half of them had died." 
In a similar test with P. aurelia, at the end of twenty-one days only 37 specimens 
out of many hundreds survived, and these were in the extremes of starvation. 
The numerical data (in microns) for 100 individuals in each group (except the 
starved P. aurelia) are summarized in the accompanying table. 

Effects of Inanition on Paramecium. (Jennings '0? 


of variation 

Normal Paramecium caudatum 

1 84. 680 ±0.848 
64.880 + 0.580 


12 . 596 ± 0.600 
8.624 + 0.412 

6.821 +0.327 
13 -292 ±0.645 

Paramecium caudatum starved n days 

Length. . 

149. 360 ±0.736 

10.896 ±0.520 


7. 296 ±0.350 

Normal Paramecium aurelia 

Length . . 

144.880 + 1.097 
54. i6o±o.765 

11.346 ±0.541 


20.948 ± 1 .042 

Paramecium aurelia starved 21 days 

Length . . 

I02.594± 1 . 161 
23. 892 ±0.644 

lo.467±o.82i io.202±o.8o8 

5.8o4±o.455 24.29i±2.oi4 

Thus during inanition the breadth of Paramecium is decreased more than 
the length, and is more variable. On refeeding, the variability is increased, 
since some individuals recover more rapidly than others; but eventually a nor- 
mal equilibrium is restored. 

Nirenstein ('10) proved that fatty granules are normally abundant in the 
endoplasm of Paramecium, but that they disappear more or less completely 
during the earlier days of inanition. Emaciated fasting individuals may be 
completely deprived of fat without the appearance of vacuoles or other degener- 
ative phenomena. The characteristic fatty granules reappear within a few 
hours after refeeding with fatty emulsions, carbohydrates (starch) or even 
protein (Merck's egg-albumin). The possibility of fat formation from ingested 
bacteria was excluded. 

The most recent extensive study of Paramecium caudatum during inanition 
is that by Lipska ('10), a pupil of Yung in Geneva. Lipska claimed that the 
"hunger cultures" of previous observers do not represent complete inanition. 
In such cultures, it is impossible to avoid bacterial contamination. Moreover, 
in the hundreds or thousands of individuals hitherto used in such cultures, many 



die during the experiment and their disintegrated corpses, or more likely the 
bacteria which thrive thereupon, serve as food to prolong the life of the sur- 
vivors. Lipska sought to avoid these difficulties by improvements in the experi- 
mental technique: (1) by repeated washings of the paramecia in boiled water at 
the beginning; and (2) by isolation of individuals, each in sterile water in a 
capillary glass tube two or three cm. in length, kept horizontally in a moist 
chamber. About 3,500 individuals were so observed, and in 350 cases the para- 



f.v. Is 

I — cr. 

Figs, i to 6. — Paramecia in various stages of inanition. Figs, i to 5 from Lipska ('10); 
Fig. 6 from Wallengren ('02). Magnification about X400. b.g., buccal groove; c, cilia; cr, 
excretion crystals; d.v., degeneration vacuoles; ect., ectoplasm; end., endoplasm;/., fold in body; 
f.v., food vacuoles; TV, macronucleus; n, micronucleus. 

Fig. 1. — : Normal Paramecium caudalum of average dimensions. 

Fig. 2. — Paramecium after six days of inanition. The macronucleus (TV) is greatly 
enlarged, occupying the posterior half of the pear-shaped body. 

Fig. 3. — Paramecium after seven days of inanition. The macronucleus (TV) is elongated; 
micronucleus (w) emigrated to posterior end of body. Two depressions are visible on the 
body, separated by a thick ridge or fold (/). 

mecia were finally fixed, stained and mounted for more detailed microscopic 
study. The results, in comparison with those of her predecessors, will be sum- 
marized briefly (Figs. 1 to 5). 

The average duration of life under such conditions was found by Lipska to be 
only five to seven days. The time varies according to circumstances, however. 
The (exhausted?) individuals starved immediately after conjugation survive 
only about three days. A few individuals, especially those containing symbiotic 
green algae, may survive ten or twelve days; and those previously hypernour- 
ished may survive even fifteen to twenty days. The higher averages of pre- 
vious observers represent merely the few survivors of an incomplete inanition. 



The body of Paramecium (as observed by Lipska) becomes progressively 
smaller, with no change of form until the fourth day of inanition. In later 
stages, it may become variously deformed — flattened dorsoventrally (Fig. 3), 
later often crescent-shaped (Fig. 4) or pyriform (Fig. 2). The average data for 
length and breadth in microns are as follows: normal Paramecium, 238 (length) 
— 54 (breadth); starved and flattened, 124-28.5; starved and crescent-shaped, 
136.8-28.8; starved and pyriform, 118-46. Thus there is a decrease of nearly 

Fig. 4. — Paramecium after eight days of inanition. The characteristic crescent-shaped 
curvature of the body appears in this case. The macronucleus (N) and micronucleus (n) have 
emigrated to the anterior end of the body. 

Fig. 5. — Paramecium after nine days of inanition. The macronucleus has divided into 
two spheroidal bodies (N, N), the micronucleus (n) remaining undivided. 

Fig. 6. — Paramecium after twelve days of inanition, showing the degeneration vacuoles 
(d.v.), considered characteristic by Wallengren ('02) but not by Lipska ('10). 

one-half in the various dimensions, which is somewhat greater than the average 
decrease obtained by Jennings, but not very different from that of Kasanzeff and 
Wallengren. A decrease of one-half in dimensions would correspond to a loss 
of seven-eighths (87^ per cent) in volume. By a curious error, Lipska confuses 
dimensions and volume, concluding that "II meurt lorsqu'il a perdu a peu 
pres la moitie de son volume initial." 

In the endoplasm of the starving Paramecium, Lipska found, in agreement 
with all previous investigators, a progessive disappearance of the food-vacuoles, 
with corresponding increasing transparency of the cytoplasm. The "excretion 
crystals" become more clearly visible, but decreased in size and number (as 
previously noted in P. caudatum by Schewiakoff '94). While apparently all 
previous observers (including Kolsch '02) have described a marked and progres- 
sive vacuolation of the endoplasm (cf. Fig. 6), Lipska could not find even a single 
case of such vacuolation during inanition. She found the vacuolation well 
marked in old cultures with the ordinary technique, however, and concluded 


that it is due primarily, not to inanition, but to the toxic action of the bacterial 
products in such cultures. 

In the ectoplasm, Lipska found no vacuolation (versus Chainsky) and rarely 
any involvement of the trichocysts and cilia (versus Wallengren). The move- 
ments become sluggish, however, and there is "un ralentissement general des 
phenomenes vitaux." 

The macronucleus in the early stages of inanition becomes elongated and 
slightly enlarged (Fig. 2) confirming Kasanzeff and Chainsky, staining more 
deeply and often moving toward one end of the body. Toward the third or 
fourth day it frequently divides into two nearly equal spheroidal masses (Fig. 
5), which separate later, but do not undergo the granular degeneration and 
fragmentation described by Kasanzeff, Wallengren and Chainsky (cf. Fig. 6). 

The micronucleus may leave the vicinity of the macronucleus, but (in agree- 
ment with previous observers) usually undergoes no appreciable change in size, 
form or structure. Although the Paramecium in normal cultures averaged 
one division per day, only eleven cases of division were observed by Lipska in the 
thousands of individuals during inanition. Even these few divisions may have 
begun before inanition, as they occurred shortly after isolation. No cases of 
conjugation were found by Lipska during inanition, special tests for this purpose 
being made by placing five to ten paramecia together in a capillary tube. These 
tests are admittedly inconclusive, however. 

Death of the paramecia from inanition was found to occur (as described by 
Wallengren) either (1) slowly, by a process of gradual granular degeneration and 
ultimate disintegration or (2) rapidly, an infrequent form due usually to a 
mechanical injury of the (probably weakened) cell-membrane. Granular 
masses of chromatic (nuclear) origin are frequently long recognizable in the dis- 
integrating dead cells. 

Recovery of the starving paramecia was found possible by careful refeeding, 
if begun not later than the third to fifth day of inanition. Division recom- 
mences three to five days later. The process of recovery is exactly inverse to 
the process of degeneration during inanition (confirming Wallengren). 



The striking effects of inanition on the invertebrates are very illuminating 
in comparison with the more complicated phenomena met in starvation and in 
the various deficiency disorders of the higher vertebrates, including man. It is 
difficult to summarize satisfactorily the effects of inanition upon the large and 
varied group of organisms included under the metazoan invertebrates. Never- 
theless it will be profitable to review briefly some of the outstanding results as 
to duration of inanition, effects on weight, size and form of the body, changes 
in various organs and tissues, effects on the gonads and sex, effects on cell 
structure, effects on development, regeneration during inanition, and recovery 
upon refeeding. The data for the various groups of invertebrates will then be 
given in detail. 

Summary of Effects on the Higher Invertebrates 

Duration of Inanition. — The period of endurance naturally varies not only 
with the species, age, and individual, but also with the type of inanition, tem- 
perature and other external conditions. Hibernation or similar dormant peri- 
ods are not comparable with ordinary inanition, because in the former case 
special stores of food are accumulated in the body for this purpose, and also 
because the metabolism is decreased so as to lessen greatly the need for nutri- 
ment. Some of the longer periods (years) of starvation reported for the leech, 
snail, spider and various insects are probably to be accounted for by unusually 
extensive periods of dormancy. 

For complete starvation (total inanition) of adults, the periods reported for 
different species range from one or two days up to several years. In the Coelen- 
terata, the period ranges from six weeks or less in the jellyfish Cassiopea to four 
months for Hydra; but newly hatched Hydra may survive only four or five days. 
The Platyhelminthes (fiatworms) may endure six to fourteen months. The 
parasitic roundworm Ascaris can endure only seven to ten days, whereas the 
blood-leech Hirudo after engorgement may last one to three years. Among 
Echinoderms, the adult starfish can endure for months, while the larvae usually 
perish within a few weeks (maximum 60-70 days). Of the Mollusca, the mussel 
Mya endures only 18 days, the snail Helix four months (or in some cases possibly 
for years). Of the Arthropoda, the waterflea Daphnia endures only n days, 
the crayfish Astacics and spider Epeira a few months, the crab Cancer and 
the scorpion 18 months. The insects show enormous variation in different 
species, varying from only one or two days in the flies, bees and ants, dragon-flies, 
etc. to possibly several years for the bedbug. Among the beetles (Coleoptera) 
alone, the endurance in different species varies from two days to over six years. 



The Orthoptera (grasshoppers) and Lepidoptera (butterflies and moths) endure 
starvation generally only a few days. The time varies inversely with the 
temperature and (in general) directly with the age; but in some cases (certain 
flies and beetles) the normal larval period may be prolonged by underfeeding. 
In the grasshoppers, the period of endurance is doubled if water is allowed. 
In all species there are also great individual variations. 

Effects on Body Weight and Size. — As pointed out by Putter ('n) the possi- 
ble decrease in dimensions during inanition is much less in those forms having 
a firm skeleton (e.g. Arthropoda). The maximum possible reduction in body 
weight and size varies greatly among the higher invertebrates, ranging from 15 
or 20 per cent in some Arthropoda to over 90 per cent in some coelenterates and 
planarians. Among coelenterates, Hydra is reduced to ^7 or less in length; 
the jellyfish Amelia to ^ and Cassiopea to ^5 of the original volume. Plan- 
arians (flatworms) may be reduced to ^{2 m length and to 3^300 i n volume. In 
the snail Helix, the loss in weight apparently varies from n to 50 per cent or 
more, according to species and conditions, with much individual variation. 
Among the Arthropods, a loss of 75 per cent in weight is recorded for the water- 
flea Daphnia, and of 15 per cent for the crayfish Astacus. Great variation has 
been found among the insects. 

There are relatively few data on the loss during inanition in larval stages. 
The larvae of the sea-urchin Strongylocentrotus Hindus may decrease to half the 
diameter of the original ovum. Chortophaga nymphs lose 20-25 per cent, and 
the tent-caterpillar (Clisiacampa) 35 per cent. The remarkable reduction (to 
3^o 0) m the larva of the beetle Trogoderma is apparently exceptional. The loss 
during the pupal stage in general appears relatively slight. 

Hall ('22) found marked differences in the limits of exsiccation in various 
invertebrates and vertebrates (see table on p. 116). 

Effects on the Form of the Body. — During inanition the body is not only 
reduced in size, but also frequently more or less changed in form. Hydra for 
example, becomes at first abnormally elongated, later greatly contracted and 
relatively broader. The water-flea Daphnia presents a peculiar modification 
of body form attributed to malnutrition, and said to be hereditary (Woltereck). 
In the sponges, the coelenterates (Hydra and Cassiopea) and the planarians the 
changes are remarkably great, not only in external form, but also in correspond- 
ing internal structure. The so-called "reduction" process in these forms more 
or less resembles a reversal of the developmental process, which will be discussed 
later. In planarians, the involution frequently involves the posterior portion 
of the body to a greater extent, resulting in a relative enlargement of the head 
region. This is also true of the nemertin worm Lineus lacteus, but not of 
Lineus ruber, which illustrates the differences which may occur between species. 

Changes in Various Organs and Tissues. — It is a remarkable fact that the 
various organs and tissues of the body differ very greatly in their resistance to 
inanition. Some undergo changes very quickly, others only after longer periods, 
and still others show a remarkable resistance. There are also differences in the 
extent of atrophy, as well as the order of sequence. The gonads will be dis- 
cussed separately later, and also the changes in the developing organism. 


In the medusa Cassiopea, the loss is almost entirely at the expense of the 
gelatinous ground substance; while in sponges, on the contrary, this substance is 
increased in amount. The ectodermal structures are more resistant, but they 
atrophy to a variable extent in different regions. In planarians, the intestinal 
epithelium, eyes, and pigment tissue and sexual duct system are affected rela- 
tively early; the muscles atrophy later, while the gonads and nervous system 
persist with great tenacity. In Lineus, the relative resistance of the various 
tissues is somewhat similar to that in planarians. Although among the higher 
invertebrates the skeleton is usually unaffected by starvation, in the snail 
Helix the calcareous shell is attacked as well as the soft tissues; while in sponges 
the results appear variable. Calcium deficiency usually causes involution of 
the invertebrate calcareous skeleton, both in larval and adult stages. In the 
reproductive passages of Helix, the non-glandular parts appear more resistant 
than the accessory albuminous gland, which undergoes an enormous involution. 
While the gland cells are greatly reduced, however, the associated " parenchyma" 
(stroma) nuclei persist and proliferate by amitosis. 

Effects on the Gonads and Sex. — The effects of inanition upon the repro- 
ductive system and process have aroused much interest and have been carefully 
observed in various invertebrates as well as in other organisms. In forms where 
the process of reproduction may be either asexual by budding, fission or parthe- 
nogenesis) or sexual, the tendency of abundant nutrition is to favor asexual 
reproduction; while inanition or other unfavorable environment usually occa- 
sions a change to the sexual form, in many cases especially favoring the develop- 
ment of males. The gonads themselves are usually very resistant to starvation, 
being (like the nervous system) as a rule among the last of the organs to undergo 
involution. There are, however, evident variations in different species and 

Among sponges, the process of involution during inanition frequently results 
in the formation of numerous small bodies ("gemmules"), from which new 
individuals may be reproduced later. In Hydra, budding is inhibited but the 
gonads (especially the testes) may mature in spite of the atrophy of the body 
as a whole. The predominating development of males in Hydra during inani- 
tion was emphasized by Nussbaum ('93), Schultz ('06) and Berninger ('10), 
but was denied by Hertwig (06), Whitney ('07), Hanel ('08), Krapfenbauer 
('08) and Frischholz ('09). 

In planarians (flatworms), asexual reproduction is inhibited by inanition 
and there is a return to sexual reproduction, but the "cocoons" are reduced in 
size and the number of enclosed young markedly diminished. In the nemertin 
worm Lineus, the gonads undergo partial involution, but some areas (likewise 
the ducts) are very persistent. 

In the rotifers (wheel-animalcules), Leydig noted that inanition causes 
prompt atrophy of the ovary and a tendency to production of males. Nuss- 
baum O97) concluded that in the rotifer Hydatina underfeeding in the phase 
preceding the ripening of the ovum tends to produce male offspring. This was 
confirmed by Lennsen ('98) but opposed by Whitney ('08). Shull ('10, '11) 
concluded that sex in Hydatina is determined by both internal and external 


factors, but that the production of males during inanition is probably only an 
indirect effect. 

The relation of nutrition to sex has also been studied extensively in the 
crustacean water-fleas by Leydig and others. The change from parthenogenetic 
to sexual mode of reproduction under unfavorable conditions (especially inani- 
tion) in this group was noted by Kerherve ('92), Cuenot ('94) and Issakowitsch 
('05). More recently, the internal factors have been emphasized, although it is 
admitted that the hereditary tendency may be influenced by external factors, 
such as malnutrition, especially during labile period of the ovum (McClendon 
'10; Woltereck 'n). Green ('19) recently concluded that in Simocephalus 
the sex is probably predetermined in the ovary; but is also subject to environ- 
mental influence, though probably not through starvation. 

Among insects the relations of nutrition to sex have been noted especially 
in the aphids (plant-lice), the Lepidoptera and the Hymenoptera. The appear- 
ance of winged male forms in aphids as a result of underfeeding was noted by 
Kyber (1813), Leydig, and others. The same phenomenon in the grape-louse 
Phylloxera was found by Keller ('87) and Behr ('92). Thus apparently all 
observers admit the effect of unfavorable environment, especially underfeeding, 
in causing the cessation of parthenogenesis and the appearance of males and 
sexual reproduction among the aphids. The conditions in this group therefore 
form strong evidence supporting the theory of nutritional sex-determination. 

In the Lepidoptera, Treat ('73) observed a preponderance of males as a 
result of underfeeding the larvae of certain butterflies. This was not confirmed 
by Poulton ('93), although he admitted that a lesser resistance to inanition in 
the female larvae might result in a selective mortality with survival of a relatively 
larger number of males. This, of course, is not strictly a process of sex-deter- 
mination, but rather of sex-survival. In the underfeeding experiments by 
Kellogg and Bell ('03, '04) on the silkworm Bombyx, and by Holmes ('10), 
Guyenot C13) and Loeb and Northrop ('16) on the fruitfly Drosophila, no 
effect on the sex-ratio was observed. 

Among the termites and Hymenoptera, the sex is known to be determined by 
fertilization (the unfertilized eggs producing the males). The diet, however, 
determines the development of the female reproductive tract, which in the few 
larvae well-fed (with "royal diet") becomes functionally developed, producing 
the queens;" while in the less richly nourished larvae the reproductive tract 
remains rudimentary, producing the "workers," of which there may be different 
varieties, according to the amount of food available. There is some evidence 
indicating, that in wasps (and possibly other Hymenoptera) the sex-ratio may be 
affected by nutrition, but this is somewhat uncertain. 

Among the invertebrates in general, we may therefore conclude from the available 
evidence that malnutrition tends to favor the sexual, rather than the asexual, mode 
of reproduction. Furthermore, especially in Hydra, rotifers, daphnids and 
aphids, the sex-ratio is at least to some extent subject to environmental influence, 
inanition tending to produce a preponderance of males. Wilson ('00) concluded 
that nutrition is one factor which may determine sex. Schultze ('03) reached 
this conclusion for Hydra and Hydatina, but felt less certain regarding the 


daphnids and aphids. How this influence becomes effective, however, is still 
a matter of uncertainty. In view of the conclusion by some investigators that 
temperature, rather than nutrition, is the decisive factor in sex-determination, 
we may recall the dictum of Cuenot ('94) that "la plupart des influences de 
milieu se ramenent en somme a des differences de nutrition." The whole 
question requires further investigation, especially in the light of the recent 
theory of sex-determination by the accessory chromosome. 

Effects on Cell Structure. — We have noted that the effect of inanition is to 
produce a variable decrease in the size of the body as a whole, which is found to 
involve a variable reduction in the various parts, organs and tissues. This, of 
course, depends ultimately upon the changes in the underlying units, the com- 
ponent cells of the organism. Some cells are destroyed and absorbed, others 
persist more or less changed. From his study of the histological changes in 
Hydra and planarians during the involution process of inanition, Schultz 
inclined to attribute the marked decrease in size of the body to a decrease in 
the number of cells, those persisting being practically unchanged in size. The 
preponderance of evidence is against this view, however. Especially in the 
adult organism, all the cells of the body during inanition tend to undergo more or 
less atrophy. The extent and character of this atrophy vary in the different 
tissues, and in different cells of the same tissue. The decrease in the size of the 
body is therefore due partly to the complete disappearance of cells and partly 
to an atrophy of those persisting. 

During the process of atrophy, the cytoplasm of the starving cells undergoes 
in general a series of characteristic changes, first losing its stored food material, 
pigment, mitochondria, and various inclusions. Later the cytoplasm of the 
cells often fuses into a syncytium. Vacuolation usually appears, with progres- 
sive decrease in amount of the cytoplasm, and terminal disintegration and com- 
plete absorption in the case of some of the cells. There is usually an earlier 
stage of reduction in size (simple atrophy) and a later stage of degeneration. 

The nucleus is usually more resistant than the cytoplasm, giving a higher 
nucleus-plasma ratio. At first the nucleus may even enlarge, but later it tends 
to shrink (pycnosis) with perhaps ultimate fragmentation, karyolysis, and final 
absorption. The reduction in cytoplasm and relative increase in nuclear size 
frequently gives the atrophic tissue an embryonal appearance. 

Some further special features in the atrophic cell changes may be reviewed 
briefly. As noted by Terroine ('20) from the chemical viewpoint, the fat changes 
show great variation among species and individuals. While fat in general is 
quickly absorbed during inanition, in the vitelline gland cells of planarians it is 
tenaciously retained, disappearing only when the cells undergeo final necrosis. 
In planarians and molluscs, the nerve cells may show certain degenerative 
changes, although the nervous tissue as a whole is remarkably resistant. It 
may be noted that apparently the final absorption of the degenerated cells in 
the various tissues is usually accomplished by simple solution in the interstitial 
tissue fluids; although phagocytosis has been noted in sponges, sea-urchin larvae, 
and especially in the case of the nemertin worm Linens. The process of phag- 
ocytosis is said to occur also in the metamorphosis of insects. 


Some specific changes in cell-structure are noted during various forms of 
partial inanition. Thus calcium deficiency may loosen the intercellular attach- 
ments, but it does not affect the ciliary mechanism, and apparently permits 
mitosis to continue in the embryonic tissues of various invertebrates. Phos- 
phorus and potassium, however, are evidently necessary for mitosis, as in plants. 
In addition to the mineral salts, certain proteins, fats or carbohydrates, water, 
etc. are doubtless essential to life, but we have as yet few data upon the morpho- 
logical effects of their deficiencies among the higher invertebrates. 

Effects on the Developing Organism. — The effects of inanition upon the 
developing invertebrate organism, especially during the earlier embryonal 
stages, in many respects often differ markedly from those previously described 
for the adult. The embryonal cells have a characteristic tendency to growth 
and differentiation, which may enable certain organs and tissues not only to 
persist but even to develop at the expense of the remainder of the starving organ- 
ism. This is true for general inanition, and is also especially evident during 
various forms of partial inanition; for example, in the case of specific salt defi- 
ciencies in the sponges, coelenterates, etc. Thus in many cases it is evident 
that deficiency in one limiting factor does not necessarily altogether inhibit the 
development of the invertebrate organism when that factor is exhausted, but 
occasions instead a disproportionate, abnormal growth, the extent and char- 
acter of which will vary according to circumstances. Liebig's law does not 
apply in these forms. 

On 'the other hand, in many cases the developing embryo or larva (in some 
species even the adult organism) tends to undergo during inanition a series of 
retrogressive stages more or less exactly reversing the normal order of develop- 
ment. This process (technically called "reduction") is often well-marked in 
sponges, coelenterates and planarians, though apparently rare in the more 
highly organized invertebrates. Even in the lower invertebrates, however, there 
is some question as to whether the (often remarkable) resemblance of the atrophic 
organism to the earlier embryonic stages may not be more apparent than real. 
But in some cases these atrophic remnants are actually able, under proper condi- 
tions of nutrition, to regenerate the normal structure of the organism. Child 
concludes that starvation results in morphological rejuvenation, followed by 
physiological regeneration upon refeeding. 

The relatively great resistance to inanition usually offered by the gonads has 
already been mentioned for adults. This applies also in some cases to the 
developing organism, so that sexual maturity may occur in undersized bodies 
as a result of underfeeding (e.g., silkworm and certain bees). But in others 
(starfish) sexual maturity is reached only upon the attainment of a certain body 
size, or the development of the gonads may even be entirely inhibited by inani- 
tion (fruitfly). 

The sex of the offspring (as already mentioned) may also be influenced through 
the effect of inanition upon the germ cells in the larval stages, particularly at 
certain critical periods (rotifers, daphnids, aphids, etc.). In some insects, sexual 
maturity appears to be determined entirely by larval nutrition, in others chiefly 


by the adult nutrition. Cuenot ('99) doubts whether the sex-ratio is notably 
affected by nutrition in species reproducing by fertilization only. 

Underfeeding during the larval period may also result in undersized adults 
(various insects), and also sometimes in marked structural modifications (pig- 
mentation, etc.). In some cases, these acquired characters appear hereditary, 
at least for a few generations; although ultimately upon adequate diet there is 
an evident tendency to return to the original condition. 

The experiments with media or diets variously deficient have shown that 
compounds of P, K, Na, Ca, Mg, Fe, S and CI appear necessary for development, 
growth being inhibited or variously perverted when the amount stored in the 
ovum is exhausted, unless a supply from without is available. The remarkable 
results (especially of Herbst, Loeb and Maas) in this field have been summarized 
by Driesch ('06). The marvelous effect of fat upon the development of the 
reproductive tract in the female Hymenoptera illustrates the morphogenetic 
potency of a single dietary factor. As to other forms of partial inanition, 
Baumberger holds that protein is in general the limiting factor in the growth of 
insects. The fruitfly Drosophila apparently requires yeast as well as sugar for 
growth, and other insects living on fermenting substrata of low protein content 
usually feed on the microorganisms present. Little or nothing is known con- 
cerning the possible requirement of vitamins by the invertebrates in general. 

Regeneration during Inanition. — In many of the invertebrates, indications 
of regenerative activity may be observed in certain cells or regions during the 
general degenerative process of involution. Driesch ('01) observed that the 
reserve materials of the body which are consumed during starvation may be 
used to build up other parts of the organism. Thus in daphnids, Crustacea and 
insects, repeated ecdysis (moulting), with regeneration of a new exoskeleton 
and appendages, may occur during starvation. This regenerative process occurs 
notably in the lower invertebrates, such as the sponges and coelenterates, and 
has been studied especially in planarians. Here the starving organism exhibits 
the remarkable capacity to regenerate large portions of the body, such as the 
head, with all its parts, pharynx, nerve ganglia, musculature and excretory 
system. Thus even during starvation, regeneration may restore a complete 
normal individual of much smaller size. In this process there is an extensive 
involution of the older tissues to furnish material for the regenerated parts. 

Recovery upon Refeeding. — In general, it is possible for the starving inver- 
tebrates to recover their normal size and structure upon appropriate refeeding, 
if the process of involution has not proceeded too far in degeneration. This 
applies also to partial inanition. In sponges, for example, the skeleton is 
reformed upon restoring calcium carbonate to the medium. In planarians, 
regeneration of the gonads is possible even when they have nearly disappeared 
after three or four months of inanition. In the gland cells of the snail Helix 
starved five months, evidences of recuperation were found even after only two 
days of refeeding. Drosophila (fruitfly) larvae are capable of normal develop- 
ment after long periods of retardation on inadequate diet; and in this form/as 
likewise in the larvae of the beetles Trogoderma and Tribolium, the normal per- 
iod of life may be greatly extended by retarding the developmental process 


through alternate periods of fasting and refeeding. Underfed larvae of the 
moths Acronyeta and Bombyx produce smaller pupae and adults, the effect in 
the case of Bombyx being often carried over to the second and third generations. 

Effects on the Various Phyla 

Among the metazoan invertebrates, the effects of inanition upon growth and 
structure have been studied most extensively in the phyla Coelenterata, 
Platyhelminthes and Arthropoda. Some investigations, however, have been 
made also upon the Porifera, Nemathelminthes, Trochelminthes, Echinodermata, 
Annulata, and Mollusca. A few observations upon the Urochorda (Tunicata) 
are also included. 


The effects of inanition upon sponges have been studied chiefly by Maas. 
His first work ('04, '04a) was on the effects of a partial inanition upon the 
development of Sycandra setosa in CaC0 3 -free sea- water. "Die 
kiinstlichen, karbonatfreiem Seewasser machten ebenfalls ihre Metamorphose 
durch, indem sie ungefahr in der gleichen Zeit festsetzten und ihre dermalen 
Zellen nach aussen, die gastralen nach innen kehrten; sie zeigten aber nach 24 
Stunden und noch spater keine Spur von Nadeln oder sonstigen Kalkkonkre- 
menten." The skeleton is normally formed if the CaC0 3 is merely reduced in 
amount, but CaSCu cannot be substituted. In the CaC03-free sea-water, the 
body form is abnormally flattened, and various developmental irregularities 
occur. Usually the cells finally spread out into a flattened mass, which ulti- 
mately disintegrates. 

In 1906, similar experiments by Maas on Sycandra raphamis were extended 
to include various larval and adult stages. When placed in artificial sea-water 
without CaC0 3 , larval metamorphosis occurs. "Es kommt zum Ansetzen, 
zur sog. Umkehr der Schichten, Kalknadeln werden nicht gebildet; trotzdem 
zeigt sich ein Hohlraum, ja mitunter auch ein Osculum." These organisms 
without skeleton soon collapse unless CaC0 3 is supplied, in which case they may 
recover normal structure if the cells have not been too greatly injured. 

When place in entirely Ca-free solutions (both carbonate and sulphate absent), 
the development is arrested earlier, in the amphiblastula stage. The ciliary 
mechanism is unaffected. On the fifth day, the granular ectoderm cells of the 
posterior half of the body become loosened and detached. The anterior mass 
of ciliated entoderm cells may persist for a week. Upon replacing the larvae 
in normal sea-water, recovery of more or less normal structure is possible, unless 
the granular ectoderm cells have been lost previously. 

When the larvae, after normal metamorphosis, are transferred to CaC0 3 - 
free solution, the previously formed calcareous spicules are gradually dissolved. 
The gastral cells form a rudimentary cavity and the dermal cells continue differ- 
entiation, but development gradually ceases in a variable time. In totally Ca- 
free solution (containing only Na and Mg salts) the skeletal spicules begin to 
disappear in a few hours. In one day, the soft parts also are degenerating ; dermal 



cells markedly loosened and dissociated; inner mass of cells usually compact, 
may show rudimentary gastrula cavity. "Die Zwischensubstanz ist auffallend 
schwach entwickelt." 

When adult sponges {Sycandra) are placed in the CaC03-free solution, the 
skeleton shows little or no obvious change, although the soft parts are affected. 
The histological changes are somewhat similar to those seen in the experi- 
ments with larvae. "Im Innern des Tubar-Hohlraums zeigt sich Detritus mit 
Porenzellen; die Gastralzellen geben ihre histologische Auspragung auf, ballen 
sich zusammen.'" In the gastral cells, karyokinesis may continue long in the 
CaCCVfree solution. In completely Ca-free solution, the skeleton of the adult 
sponge Sycandra is but slightly affected. The soft parts, however, degenerate 
in a manner resembling the normal process of gemmulation. 

Maas later ('07, '10) studied the effects of ordinary inanition upon sponges 
(and other organisms with calcareous skeleton). In Ascandra Lieberkiihnii 
and Sycon raphanus, while in Ca-free medium the skeletal spicules disappear 
and the soft parts persist, during ordinary inanition (deprivation of food) 
the converse occurs. During hunger involution the organism is reduced in 
size, with involution of the old osculum, degeneration of part of the sponge and 
formation of a new osculum (sometimes two). The body becomes irregularly 
lobulated externally, with syncytial structure and excessive gelatinous ground 
substance internally. A similar reduction of the calcareous skeleton with per- 
sistence of the soft parts was observed in certain molluscs, the tube-worm 
Spirorbis, and Foraminifera (Bilociilaria and Quinquelociilaria) . 

The involution changes produced in sponges by unfavorable conditions 
(Ca-deficit, general inanition, etc.), according to Maas ('10), all involve loss of 
cellular differentiation (dedifferentiation). Through phagocytosis, syncytia 
arise which are considered equivalent to "gemmules." The body is reduced 
to a 2-layered condition, but these layers are not comparable to the normal germ 
layers. Wilson ('07), Urban ('10), and Miiller ('n), likewise obtained experi- 
mentally a reduction or degeneration of sponges, with the ultimate formation 
of gemmule-like bodies. It is not clear, however, to what extent inanition was 
a factor in these cases. 


Among the Coelenterata, observations upon the effects of inanition, aside 
from a few studies on Scyphozoa, and a single observation on Ctenophora, have 
apparently been confined to the class Hydrozoa. Of the Hydrozoa by far the 
greater number of investigations have concerned the fresh-water polyp, Hydra, 
which will be considered first. 

Hydrozoa. — The earliest recorded inanition experiments on Hydra were 
those of Trembley (1744), who noted that both Hydra viridis and Hydra fusca 
required four months for death from starvation. Retrogression and premature 
detachments of buds during inanition were also observed. "Quand la nourri- 
ture manque, les jeunes polypes se separent plutot. II est apparent que, presses 
par la faim, ils se detachent pour aller chercher ailleurs de quoi la satisfaire" 
(I.e., p. 159). This was confirmed by Marshall ('82) and Berninger ('10). 


Baudelot ('69), Kleinenberg ('72), Marshall ('82), and Schultz (06) found 
that in underfed Hydra, well-formed buds, instead of becoming detached, may 
undergo reduction, retrogression and final disappearance. 

Greenwood ('88) described the changes in the entoderm cells of Hydra during 
the earlier stages of inanition. The nutritive vacuoles (containing protein and 
fat) persist for two or three days, but become fewer and smaller or disappear 
entirely in more protracted fasting. The smaller "gland cells" of the entoderm 
become progressively more conspicuous, but in prolonged starvation their 
secretory granules may become smaller and partly dissolved. The pigment, 
though abundant during inanition, may also be slowly discharged. 




n s r 

Fig. 7, a to d. — To illustrate the changes in size and external form of the fresh-water polyp. 
Hydra fusca, during starvation. All Xio. (After Schultz '06.) a, Normal Hydra; b, Hydra 
starved 1J-2 weeks; size reduced, but form still nearly normal; c, later stage of reduction; 
tentacles have become rudimentary; the testis has matured, in spite of the atrophy of the body 
as a whole; d, terminal stage of reduction (cf. Figs. 8-1 1). 

Nussbaum ('87, '93) was the first to study the effect of inanition upon sexual 
development in Hydra. He concluded that the (normally hermaphroditic) 
Hydra when abundantly fed will produce ovaries only; when moderately fed, 
both ovaries and testes; while those scantily fed produce testes only. This 
question is of fundamental importance in the theory of sex-determination, and 
has occasioned much controversy. 

Hertwig ('06) opposed Nussbaum's view, and maintained that the condi- 
tions for sexual differentiation in Hydra are more complicated, temperature 
being a more important factor than nutrition. In his colony, sex-organs 
(always male) appeared only at lower temperature, whether starved or well-fed. 
The testes were better developed in the well-fed. 


Schultz ('06) noted that although the process of asexual budding in Hydra 
fusca is inhibited by inanition, the differentiation of the testes proceeds, they 
being among the organs most resistant to starvation. Even in extreme stages 
the sex-cells (spermatogonia) persist and ripen into spermatozoa (Figs. 7c, 10). 

Whitney ('07), on the basis of very careful and extensive experiments with 
Hydra viridis, concluded that temperature rather than food supply is the pri- 
mary factor in sex-production . Low temperature followed by higher temperature 
causes budding and also formation of sex-organs, irrespective of food conditions. 

Oral area 





Entoderm^ ^Jf&^Eclbderro 

Fig. 8. — A longitudinal section of Hydra fusca after eight weeks of starvation. Body 
concontracted; mouth aperture still open. (After Berninger '10.) 

Fig. 9. — A longitudinal section of Hydra fusca after twelve weeks of starvation. Body 
oval in form; mouth aperture has closed. (After Berninger '10.) 

Hanel ('08) found that neither inanition nor low temperature causes sex- 
production in Hydra grisea. Krapfenbauer ('08) obtained positive results by 
lowered temperature, but not by inanition. Frischholz ('09) found that within 
certain limits of temperature, sex-production in Hydra tends to appear in definite 
cycles of 20 to 40 days, depending upon the nutritional conditions. Even in 
extended inanition, however, sexual forms appear, either male or female (the 
strains used being always monosexual, never hermaphroditic). A change of 
temperature is unnecessary, but may accelerate sex-production. However, 
permanent exposure to high temperatures in Hydra grisea or to low tempera- 
tures in Hydra fusca, permanently inhibits sex-production, irrespective of the 
degree of nutrition. Finally, Berninger ('10) again found that in Hydra fusca 
(also in Dendrocoelum lacteum) inanition stimulates the development of the 
testes, with abundant ripening of the spermatozoa during starvation. 

In summary, it is evident that Hertwig was correct in concluding that 
sexual differentiation in Hydra is a complicated phenomenon, various factors 
being involved. Besides temperature and nutrition, there is probably a cyclic 
or seasonal variation, and other hereditary characteristics, apparently varying 
in different species or strains of Hydra. 



In addition to the effect upon the reproductive system, the general process 
of reduction in Hydra fusca during inanition has been described in detail by 
Schultz ('06). During the first week there is noted (except at low temperatures) 
a remarkable elongation, both body and tentacles forming thread-like extensions 
often reaching ten times the normal length. In the following weeks (Fig. 7, 
a-d) the animals gradually retract and become smaller, resulting in very small 
hydras (reduced to l - normal length or less), but of normal body form. The 
tentacles become shortened, with swollen extremities; finally they disappear. 
The body, deprived of tentacles, gradually becomes pear-shaped and later 
spherical, the oral aperture closed and obliterated. Finally there results a 
spherical planula of ectoderm and endoderm, resembling the embryonal form 
(Figs. 8 to 11). Thus involution apparently reverses the processes of embryonic 
development in Hydra. 

The histological changes in the tentacles of Hydra during reduction are not 
very striking. Apparently the loss of cells is chiefly at the tips, where degener- 
ating cells are frequently seen. 




^1 w>xVacuole 

Syncytium x^T^^Hf 


Fig. 10. — Ectoderm cells from starving Hydra fusca. Highly magnified. The cells 
are separated by large interstitial spaces. The cytoplasm is scanty; the nuclei relatively 
large. A group of the persistent sex-cells (spermatogonia) is shown. (After Schultz '06.) 

Fig. 11. — Entoderm cells from starving Hydra fusca. (After Schultz '06.) Highly 
magnified. The cells have fused into an irregular syncytium, containing vacuoles, pigment 
granules and inclusions. The free surfaces of the cells show irregular processes. 

In the body ectoderm (Fig. 10), there is great variation in the resistance 
of different cells. The epithelium-muscle cells remain flatly extended, with 
progressive loss of plasma until the nucleus is barely covered. The cytoplasm 
usually becomes vacuolated. The mematoblasts disappear entirely. Other 
cells, however, may retain much cytoplasm, sometimes with gigantic nuclei. 
The gland cells of the foot remain apparently unchanged. The development of 
the testes (as above mentioned, Fig. 10) proceeds in spite of inanition, "Oder 
richtiger gerade infolge des Hungers." " Je mehr das betreffende Tier reduziert 
war, desto weiter war meistens die Reife der Testikel geschritten. Auch 
wimmelten meine Aquarien, wo ich die hungernde Tiere hielt, bald von Sperma- 
tozoen. Doch wachsen die Testikel nie zu jenen mammaformigen Gebilden, 
wie sie gewohnlich bei Hydra fusca erscheinen. Sie blieben weit kleiner und 
bildeten nur geringere Anhaufung unter dem Ectoderm. Eine Reifung der 
Ovarien beobachtete ich nie." 

Of the entoderm in Hydra, the so-called gland cells show no morphological 
changes, although they decrease in number. The intestinal epithelium, how- 


ever undergoes marked changes, the earlier stages of which were described by 
Greenwood ('88), as above mentioned. In contrast with the ectoderm, these 
entodermal cells become less vacuolated by disappearance of the food- vacuoles. 
Later the cells fuse into a syncytium, containing pigment, vacuoles and other 
inclusions (Fig. n). The entoderm cells of the foot become indistinguishable 
from the intestinal epithelium. The entodermal nuclei at first become some- 
what swollen, with indistinct nuclear membranes, later frequently undergoing 
chromatolysis. The intestinal cavity contains extruded masses of cytoplasm, 
intermingled with desquamated epithelial cells containing nuclei in various 
stages of degeneration. "Der Tod des Tieres wurde durch die voile Degener- 
ation des Entoderms eingeleitet, wahrend Ectoderm und Genitalanlage noch am 
Leben bleiben." 

Schultz claimed that the tremendous reduction in the size of Hydra (as in 
Planarians) during inanition is accomplished chiefly through decrease in the 
number of cells, the size of those persisting being not very different from the 
normal. He opposed Roux's theory that those cells persist which require less 
food in the struggle for existence, stating that in the disappearance of organs 
during inanition the sequence is in general the opposite to that by which the 
organism developed, ontogenetically and phylogenetically. This order of loss 
is not always that most advantageous to the individual organism, however. 

Berninger ('10) obtained results very similar to those of Schultz. He con- 
sidered Hydra viridis unsuitable for inanition experiments, on account of the 
symbiotic algae, and studied chiefly Hydra fusca. In ordinary filtered tap water 
they all died within three weeks, but in spring water they lived twelve to four- 
teen weeks. As an age difference, Frischholz ('09) noted that newly hatched 
Hydra fusca die of starvation in four or five days. There is notable elongation 
in the first week, as observed by Schultz and by Krapfenbauer. The involution 
changes are not marked until the fifth or sixth week, when the body is reduced to 
about half its original length (Fig. 7, a to d). In fourteen weeks of starvation 
the Hydra is reduced from 7-8 mm. length and i}4,-2 mm. width to 0.2 mm. 
length and 0.13 mm. width. This is estimated to be about }£o of the original 
size, corresponding to a Hydra embryo a few hours old. Structurally the 
reduced Hydra differs from the embryo only in the higher differentiation of the 
cells, and the absence of yolk granules. "Also kann man wohl sagen, dass die 
Hunger bewirkte Reduktion die Hydra ungefahr auf ein embryonales Stadium 
zuriickbrachte, wobei der umgekehrte Weg eingeschlagen \vurde, welchen die 
Entwicklung durchlief." 

Child and Hyman ('19) observed that in Hydra "The differences in diameter, 
general appearance and opacity between body and stalk become less marked 
with lack of food and in advanced starvation may almost or entirely disappear." 
In starved animals, the entoderm becomes much less susceptible to the disinte- 
grative effect of cyanide, dyes, etc. 

A few observations have been recorded as to the effects of inanition upon 
other Hydrozoa. Semper ('81) states that "The observations made on Hydroid 
Polyps by Hincks, Allman and Schneider are highly interesting. According 
to these, in the first place a Medusa of the group of the Hydroidea can be 


induced by lack of nourishment to assume the polyp-form, i.e., the larva form 
of the species." Linko ('00) noted a disappearance of pigment in the ocelli of 
the medusa Margellium retro punctatum which had been in an aquarium without 
food for a long time. Citron ('02) found that in Syncoryne Sarsii the ectodermal 
cells become flattened, the cytoplasm greatly reduced in amount, and the cells 
fusing into a syncytial condition. 

In connection with his pioneer work on physiological morphology, J. Loeb 
('92) found that potassium must be present in small quantity in the surrounding 
water to permit regeneration of polyps in Tubularia mesembryantheum, and 
magnesium in addition to permit normal growth. Although at this time he 
considered the salts of these two elements, in addition to NaCl, as sufficient for 
regeneration and growth in Tubularia, he later (1905) admitted that traces of 
calcium salts were also present in the water used. Herbst ('97) observed that 
decapitated Tubularia mesembryantheum are able to regenerate their heads in 
media free from calcium phosphate, perhaps because a sufficient supply was 
already stored in the body. The further work of Loeb and Herbst will be 
considered later in connection with the Echinodermata. 

Scyphozoa. — Among the Scyphozoa, deVarigny ('87) observed three 
medusae of the jellyfish Aurelia aurita, weighing 98, 82 and 57 g., respec- 
tively. After 150 days in the laboratory, two survivors weighed 82 and 75 
g., representing a loss of %$ or ^ of the original weight. Since protozoa 
and bacteria were not excluded from the sea-water, however, this probably 
represents an incomplete inanition. 

Hadzi ('09) studied the effects of inanition upon larvae, probably of the Scy- 
phomedusa Chrysaora mediterranea, in various advanced stages of development. 
Upon placing the hydriform scyphulae in sea-water without food (excepting 
the Ciliata present), the process of strobilization begins at once and many free- 
swimming ephyrulae appear. On account of lack of food, the ephyrulae are 
unable to develop further into Scyphomedusae, but undergo instead a series of 
characteristic retrogressive changes. The umbrella becomes smaller, later 
spherical, the marginal lobules and sense-organs (tentaculocysts) becoming 
detached. The gut lumen becomes constricted; the gastric diverticula retract 
and disappear. The body surface becomes ciliated. The reduced body later 
assumes an ellipsoidal form (Gastrea type) with a simple mouth opening, which 
finally becomes closed, resulting ultimately in a planula form, with distinct bound- 
aries between ectoderm and endoderm. "Von einem iiber 1 mm. grossen, 
schon ziemlich hoch differenzierten Tiere, der Medusenlarve Ephyra, ist unter 
allmahlicher Riickbildung eine moglichst einfach gebaute, oft nur 80^ grossen, 
Planula entstanden." This reversal of the developmental process is completed 
within three weeks. The planulae may live fourteen days longer, but finally 
lose their cilia and disintegrate. Certain modifications of the process may 
occur, and the parent scyphula also undergoes similar regressive changes. 

Stockard ('10) noted that vigorous regeneration occurs in experiments upon 
starving jellyfish, Cassiopea xamachana, but the corresponding decrease in body 
size is greater than usual in starvation, since regeneration proceeds at the 
expense of the older tissues. 



Mayer ('14) made a more extensive and detailed study of the inanition 
changes in the same species of Cassiopea. The body weight decreases according 
to the formula: 

Y = W(i- a)" 

W representing the initial body weight, Y the body weight after X days of 
starvation, and a (the "index of katabolism") being the fairly constant fraction 
of the existing body weight which is lost in any single day. 

The rate of starvation varies according to circumstances. If the rate for 
the normal medusa in large aquaria of filtered stagnant sea-water is taken as 
1.0, the rate in small aquaria (400 cc.) is 1.7; in running water, 2.4. For 
starving medusae regenerating their bell-rims, the rate is 0.96; if starving with 
stomachs removed, 1.27. In general the starving medusae regenerate as 
rapidly as the well-fed. 

Bell rim 

Bell rim 






Lateral view, 


Fig. 12. — Normal medusa of the jellyfish, Cassiopea xamachana. 
size. (After Mayer '14.) 

Fig. 13. — Medusa of the jellyfish, Cassiopea xamachana, starved 41 days in darkness, with 
loss of about 96 per cent in weight. Lateral view, natural size. (After Mayer '14.) 

Fig. 14. — Same as Fig. 13, but magnified to original size, for comparison with Fig. 12, 
to show more clearly the change in form undergone by the jellyfish Cassiopea during inanition. 
Note the relatively small bell (umbrella) with upturned rim, and the relatively large arms with 
rudimentary tentacles. 

In six weeks of inanition, with final loss of over 96 per cent in body weight, 
there are likewise progressive changes in body form (Figs. 12, 13, 14). The 
bell-rim becomes shrunken and bent upward, and the arms atrophic. The 
mouths become closed by coalescence in about three weeks, so that subsequent 
recovery by refeeding is impossible. The cells are reduced in size; many become 
fused into a syncytial condition or degenerate and disappear. The gelatinous 
substance, which forms about 95 per cent of the organism, is greatly reduced in 
amount and becomes vacuolated. The commensal green algae become crowded 
in the diminutive starving Cassiopea, and ordinarily most of them escape from 
the body; but if the experiment is conducted in darkness most of the algae 
degenerate, and die, although a few may persist and regenerate a new supply 
upon refeeding. 

Hatai ('17) confirmed Mayer's formula for the loss in weight of starving 
Cassiopea (after the first day), but found little change in the relative weights of 
the mouth organs, umbrella and velar lobes. 

In connection with his experiments on partial inanition with various salt 
deficiencies, Herbst ('97) made a few incidental observations on Cotylorhiza 
tuberculata. Phosphorus was found necessary to enable the planulae to develop 


into normal scyphostomata, the larvae disintegrating in P-free solutions. Simi- 
larly in potassium-free mixtures, the planulae developed no further, but gradu- 
ally died off within seven days; while controls developed normally. 

Ctenophora. — In connection with his experiments on partial inanition with 
various salt deficiencies, Herbst ('97) incidentally noted that when segmenting 
ova of Beroe ovata are placed in mixtures free from calcium phosphate, they soon 
perish; while the controls continue development. 


Of the Platyhelminthes or flatworms, the class Turbellaria or planarians 
have been studied most extensively during inanition. Some observations 
have also been made upon the Nemerteans, a related group of somewhat doubt- 
ful classification. 

Turbellaria. — F. F. Schultze ('36) was apparently the pioneer in observing 
the marked reduction in the size of planarians during protracted inanition. The 
first recorded measurements, however, are those of Voigt ('94), who noted that 
in ten or eleven months of fasting the length of Planaria alpina is reduced from 
12 mm. to 1 3^ or 2 mm. The sexual reproduction is also affected, the "cocoons" 
being reduced to less than half their normal length, and the number of young 
from each being reduced from 55 in the well-fed to four in the starving individ- 
uals. Cuenot mentions that asexual division in some Turbellaria {Microstoma, 
Planaria subtentaculatd) " tres prospere dans les moments d'abondance, s'arrete 
lorsque la nourriture devient rare." According to Rywosch and Zacharias 
{'86),' however, the advent of unfavorable conditions may lead to the develop- 
ment of sex-organs, with return to sexual reproduction. 

Lillie ('00) observed that in 43 days of starvation a Planaria maculata was 
reduced from 9 mm. (length) and 0.75 mm. (breadth) to 0.6 mm. and 0.25 mm. 
Assuming a decrease of one-half in the third (dorsoventral) dimension, this 
involves a reduction to V90 of the initial volume. The sex-organs appear 
immature. The pigment cells near the surface seem greatly reduced in number, 
but not in size. Although no detailed histological examination was made, the 
reduction process apparently reverses the steps of normal development. " Cer- 
tain it is that specimens reduced by starvation to a smaller size than just hatched 
specimens of the same species resemble these in their general proportions, the 
relatively greater breadth in proportion to their length, as compared with mature 
specimens, the smallness of the cephalic lobes, and in the small number of intesti- 
nal diverticula and branches of the longitudinal nerves." During regeneration, 
Lillie also noted considerable loss, owing to the destructive metabolism in work- 
ing over the old tissues into new form. 

Morgan ('01) observed that if regenerating Planaria lugubris are fed, the 
old tissue loses but little and the new tissue grows faster; if the worms are unfed, 
the old tissue loses more and the new part grows less, forming a smaller worm. 
The decrease in the old part appears to be due, not to cell-migration, but to 
loss of substance, which is transported to the regions of active regeneration. 

Stevens ('01) found a remarkable resistance of the nervous tissues to inani- 
tion in Planaria lugubris. "The nerve fibres are more easily traced in speci- 


mens that have not been fed for several weeks before cutting: the other tissues 
degenerate somewhat while the nervous tissue does not to nearly the same extent. 
A specimen starved for 18 weeks showed the head ganglion nearly as large as 
those in corresponding specimens fed during the same time, while other tissues 
were much reduced; everything in the shape of fat or yolk material having 
entirely disappeared, and the actual size of the animal having been reduced from 
n X 3 mm. to 3 X M mm." 

More detailed investigation of the histological changes in planarians during 
inanition was made by Schultz ('02, '04, '04a, '08, '08a). In his first paper, 
Schultz ('02) noted a marked decrease in the body-size of Dendrocoelum lacteum. 
"Bei hungernden Dendrocoelum lacteum konnte ich beobachten, dass die Seiten- 
verzweigungen des Darmes allmahlich immer armer wurden. Auf Schnitten 
erwies es sich, dass das Darmepithel in den Seitenzweigen von den feinsten 
Endverzweigungen angefangen und weiter zum Centralstamme fortschreitend 
sich aus seinem Verbande lost. So sieht man auf Durchschnitten oft das Darm- 
lumen schwinden und die einzelnen Epithelzellen freiim Mesenchym liegen." 

The histological changes during inanition in Planaria lactea were described 
by Schultz ('04) in greater detail. In six months the body length is reduced to 
}{o or yi.2, and the worms begin to die after seven months. Some cells undergo 
degeneration, terminating in necrosis; others undergo reduction, i.e., a"dediffer- 
entiation" with return to embryonal condition; while some cells (body epithe- 
lium) remain unchanged in size and structure. The nuclei are very resistant 
and never decrease in size. 

The alimentary canal in general decreases in proportion to the entire body, 
the form of the gut and branches being usually well preserved. The lining epi- 
thelial cells early begin to show changes, with loss of granules and fat, followed 
by progressive cytoplasmic atrophy. In some cases the cells degenerate, the 
nuclei often undergoing hypertrophy and karyolysis. In other cases the cells 
fuse into a syncytium with scanty cytoplasm, which may obliterate the gut 
lumen. The eyes likewise degenerate relatively early, by the fourth or fifth 
month, the optic cup breaking up. The pigment cells become disintegrated, 
leaving intercellular masses of pigment granules which are later resorbed. The 
parenchyma 1 (interstitial " Grundgewebe ") is largely resorbed by the sixth 
month of fasting, although some cells remain unchanged. The muscles persist,, 
even in extreme stages of inanition, but are reduced in size proportional to the 
whole body. The muscle-cells become shorter, but retain the same thickness. 
The nervous system is very resistant, although degeneration of nerve-cells 
occurs. "Das Gehirn und die Nervenstamme mit ihren Quercommissuren 
sind noch zuletzt, wenn das Tier schon dem Hungertode ganz nahe ist, gut 
entwickelt." The reproductive system varies in its different parts. Reduc- 
tion begins relatively early in the penis, which ultimately disappears. The fate 
of the ovaries is uncertain. The vasa deferentia, oviducts and associated genital 

1 It should be noted that many writers on the histology of invertebrates unfortunately use 
the term " Parenchyma " in a sense quite opposite to that generally used for vertebrates, where 
the corresponding interstitial tissue is designated as the "stroma." 


passages later undergo progressive reduction. The testes are quite resistant, 
but in six months of fasting are reduced greatly in size and consist of small 
groups of three to five cells each, scattered through the "parenchyma." Even 
the spermatogonia may finally undergo necrosis and chromatolysis. Schultz 
(like Lillie) compared the reduction process to a reversal of the normal 
development, so that inanition may lead to a rejuvenation of the organism. 
He also considered the order of sequence in the loss of organs as a useful adapta- 
tion, the less important organs being sacrificed first, the most essential (and least 
differentiated) cells persisting longest. 

Stoppenbrink ('05) made a most careful and thorough study of the changes 
during inanition, chiefly in Planaria gonocephala, but also in Planaria alpina, 
Dendrocoelum lacteum and Polycelis nigra. The extreme limit of duration found 
was fourteen months in Planaria gonocephala and ten months in Planaria alpina. 
The marked reduction in body size corresponds in general to that noted by pre- 
vious investigators. There is also a marked change in the body form, the 
relatively large head and short postpharyngeal portion of the body resulting 
from unequal reduction in the different regions. 

The reduction process in the various organs was found in general somewhat 
similar to that described by Schultz in Planaria lactea, but certain differences or 
additional features were noted. In the reduction of the intestine, no scattering 
of the epithelium in the mesenchyme was found (P. gonocephala, D. lacteum). 
The vitelline glands ("Dotterstocke"), in which the fat deposits of the body are 
concentrated, do not lose this fat quickly during starvation, as might be expected, 
but retain it tenaciously, until the organ has undergone extensive retrogressive 
changes, with cellular necrobiosis and syncytial degeneration. The sexual duct 
system degenerates much later than the vitelline glands. The ovary and testes 
are the most resistant of all, although the sex-cells undergo regressive changes. 
Stoppenbrink summarizes his results as follows: 

"Wahrend die Grossenreduktion in einem gleichmassigen Kleinerwerden 
samtlicher Zellen eine ausreichende und einfache Erklarung finden wiirde, 
deutet die Veranderung der Korperform auf anderweitige, gleichzeitig mitwirk- 
ende Ursachen hin. Diese Ursachen sind darin zu erblicken, dass eine 
ungleiche Beeinflussung der verschiedenen Gewebe stattgefunden hat, indem 
die entbehrlicheren Organe zugrunde gingen, um mit ihrem S toff material die 
Organe vor dem Untergang zu bewahren, die fur das Tier unumganglich not- 
wendig sind. Eine stattfindende Nekrobiose lasst sich nur dort feststellen, 
wo untergehende Zellen in grosserer Menge beieinander angetroffen werden. 

"Im Nervensystem, Darm, Exkretionsgefasssystem, Parenchym, Hauttnuskel- 
schlauch und Korperepithel trat ein gleichzeitig stattfindender Zerfall von 
Zellen in grosserem Umfange nicht ein. Dagegen liessen sich Degenerations- 
prozesse deutlich im Bereiche der Geschlechtsorgane beobachten, die zu einer 
totalen Riickbildung dieses Organsystems fiihrten. 

"Dieses Prozess erfolgte in der Weise, dass zuerst die Dotterstocke 
angegriffen wurden, im spateren Verlauf der Begattungsapparat und zuletzt die 
Hoden und Ovarien. Dabei trat eine Phagocytose nicht ein, die Elemente 
zerfielen an Ort und S telle und wurden resorbiert. 


"Beachtet man, dass die postembryonale Entwicklung der Geschlechts- 
organe in der Reihenfolge vor sich geht, dass zuerst die Bildung der Ovarien 
und Hoden, viel spater erst die Entwicklung des Begattungsapparates und am 
Schluss die Anlage der Dotterstocke erfolgt, so findet man, dass die Involution 
der Geschlechtsorgane in der umgekehrte Reihenfolge stattjlndet, wie ihre Entstehung." 

Berninger ('n) likewise studied the effects of inanition in various species of 
planarians (Planaria alpina, gonocephala, torva; Polycelis nigra; Dendrocoelum 
lacteum), with results in general agreement with those of Stoppenbrink, Schultz 
and other earlier investigators. Death from starvation occurs in six to twelve 
months, with length reduced to about 3-12 5 volume to ^300- The nervous and 
muscular systems suffer no actual degeneration; the gut and "parenchyma" 
(stroma) but little. In darkness, the eyes degenerate and are entirely resorbed 
in seven or eight months. Of the reproductive system, the vitelline glands and 
penis are resorbed first, then the genital passages, later the ovaries and lastly 
(just before death) the testes. The "cocoons" are dwarfed and the enclosed 
embryos reduced in size and number. Regeneration of the sex-organs is possible 
upon refeeding, even when they have almost disappeared after three or four 
months of fasting. 

Lang ('12) in connection with a study of regeneration in planarians, was 
able to confirm in general the results of previous investigators as to the effects 
of inanition. "Exkretionsgefasssystem, Muskulatur und Nervensystem blei- 
ben nicht nur vor dem Zerfall verschont, sondern regenerieren auch noch 
abgeschnittene Teile. Insbesondere regeneriert sich im Verlauf der Langs- 
nerven stamme an Querschnitten ein neues Gehirn. Bei den gesamten 
Reduktionen und Regenerationen werden diejenigen Organe verschont bezw. 
gefordert die entweder zum Leben des Individuums unbedingt notig sind oder 
die eine Vorbedingung fur Beseitigung des Hungerzustandes bedeuten, 
insbesondere Pharynx und Nervensystem." 

In connection with a series of physiological studies, based upon experiments 
with planarians (chiefly Planaria dorotocephala), Child ('11, '15, '19, '20) has 
made incidental observations as to the morphological effects of inanition. Accord- 
ing to his conception, age is characterized physiologically by decreased metabo- 
lism, expressed morphologically by accumulation of structures which hinder 
cell-metabolism. "Starvation removes the structural obstacles to a greater or 
less extent, but without increasing the rate of metabolism, except perhaps at 
first; by means of food the rate of metabolism is increased and rejuvenation is 
accomplished. Starvation brings about morphological rejuvenation, the 
following feeding, physiological regeneration." 

In connection with his experiments on partial inanition with various salt 
deficiencies, Herbst ('97) made some incidental observations upon the marine 
polyclad worms, Stylochus neapolitanus, Discocoelis tigrina and Thysanozoon 
Brochii. In solutions without calcium phosphate, they died and disintegrated 
within about eighteen hours (controls unaffected), thus indicating the necessity 
for this salt in the medium for adult polyclads. 

Nemertinea. — Aside from the observations by Giard ('05) onLineus bilineatus, 
the effects of inanition upon the nemertin worms have been studied by Nus- 



baum and Oxner. Oxner ('n) reported briefly the results in Lineus ruber and 
Linens lacteus. On decapitation of young fasting individuals (or of older ones 
just after extrusion of the sexual products), the body undergoes a process of 
reduction and involution, without sexual maturation. If the sexual system has 
reached a certain stage of maturity, however, it will continue independent develop- 
ment to maturity in spite of the decapitation with consequent inanition, which 
prevents further growth of the body in general. 

The changes in Lineus ruber and Lineus lacteus were studied in detail by 
Nusbaum and Oxner ('12). The starvation was in sea- water for various periods, 
six to thirteen and one-half months. The external dimensions of Lineus ruber 
reduced to }■■■§ or x %, all parts being nearly proportionately reduced (thus differ- 


Fi G I5 — a magnified portion of a cross section of the nemertin worm, Lineus ruber, after 
starvation for I2 1 A months. (From Nusbaum and Oxner '12.) Surface epithelial cells (ep) 
are markedly reduced in size. Stroma ("parenchyma") cells (St) of the underlying connective 
tissue are more closely packed, due to loss of the interstitial gelatinous substance. Muscle 
fibers (Mf) are greatly atrophied. Pigmented tissue (originally abundant) has largely dis- 
appeared from the body wall, being transported by phagocytic wandering cells (Pc) to the 
intestinal wall (7m) serving as food for the starving organism. Areas- of intestinal epithelium 
also undergo pigmentary degeneration into syncytial areas (So) which are finally absorbed. 

ing from the planarians). In Lineus lacteus, however, the posterior portion is 
more reduced. The body wall is normally 0.06 to 0.09 mm. thick; including 
0.02 mm. for cutaneous epithelium, 0.025 mm. for the "parenchyma" (stroma) 
and outer muscle layer, the remainder being the middle and inner muscle with 


corresponding "parenchyma." In starvation, the body wall is reduced to about 
0.04 mm. in thickness; including 0.02 mm. for the epithelium, 0.01 for the 
"parenchyma" and outer muscle, and 0.01 mm. for the remainder. 

The cutaneous epithelial cells become lower, but the gland-cells may retain 
their original height. The reduction in size of the muscular and "parenchyma" 
layers is chiefly at the expense of the interstitial gelatinous substance, but most 
of the " Parenchymzellen und Bindegewebeselemente" also disappear. The 
muscles are also greatly atrophied; the individual fibers may be reduced to one- 
half in length and diameter. The depigmentation of the body is striking, the pig- 
ment cells undergoing degeneration and resorption in various parts of the body 
(including intestine, brain and eyes). Resorption is accomplished through 
phagocytosis by the wandering cells, which transport the material chiefly 
in pigment form to regions where most needed (Fig. 15). 

The intestine in Linens is decreased about one-third in thickness and two- 
thirds in number of epithelial cells counted in a cross section. In many places, 
especially in the hindgut, a degenerative involution occurs, particularly involving 
the folds projecting into the lumen (Fig. 15). The epithelial cells form extensive 
degenerative syncytial areas, within which occur occasional islets with cells of 
embryonal appearance. These may be able to regenerate the epithelium upon 
refeeding. The eyes degenerate and disintegrate, as described for the planarians. 
The reproductive system undergoes a partial reduction. "Sowohl in den 
Hoden wie auch in den Ovarien unterliegen gewisse Abschnitte des Keimepithels 
einem vollstandigen Zerfalle under Mitwirkung von Wanderzellen . . . 
Gewisse Abschnitte der Gonaden bleiben aber bestehen und hier bilden sich die 
Geschlechtsprodukte aus . . . Die Gonodukte unterlogen keiner Reduction, 
sogar in sehr spaten Inanitionsstadien." The nervous system is very resistant; 
an apparent skrinkage in volume is due to atrophy of the connective tissue, the 
nervous elements remaining intact. There is likewise little or no change in the 
proboscis, nephridia, vessels, etc. 

In general, therefore, there is in Linens a remarkable difference among the 
various tissues and organs as to the time and extent of their reduction. A 
decrease in both number and size of cells is involved. In some cases the nucleus 
is more resistant than the cytoplasm, but in general the "Kernplasmarelation" 
is not much changed. In certain regions progressive, regenerative changes may 
occur among regressive, degenerative changes. The changes in Linens during 
inanition and in regeneration of the body are naturally similar, since the regen- 
erating organism takes no food and structures are regenerated at the expense of 
the remainder of the organism. 


Among the roundworms, but few observations on the effects of inanition are 
available. Maupas ('00) noted that in the hermaphroditic nematodes mal- 
nutrition reduces the number of ova, but does not affect the sex. In Ascaris, 
Weinland observed that the duration of starvation is greatest (seven to nine 
days) in fluid saturated with carbon dioxide. There is a marked decrease in the 
(normally very high) glycogen content. Ono ('20) studied the effect of starva- 


tion upon the mitochondria in Ascaris megalocephala. The normal muscle cells 
of the body contain mitochondria, chiefly of filamentous form, but granular in 
the perinuclear region. "When starvation begins, however, the filamentous 
forms become granular ones, some by diminution, others seemingly by segmen- 
tation, while still others by the fusion of two or more of them may form globules, 
clumps, etc., usually connected by supporting fibrils; in short the whole number 
and quantity of mitochondria gradually diminish to a very small minimum at 
the end of about 10 days starvation." Similar results were observed also in 
the epithelial cells of the intestine. 


Relatively little attention has been paid to the effects of inanition upon 
the annelid worms. Apparently only the Hirudinea (leeches) have been studied 
in this respect, aside from a single reference to the Chaetopoda. 

Chaetopoda. — In connection with the previously mentioned study of the 
effects of calcium-inanition on sponges, Maas ('12) noted that in calcium-free 
water a reduction of the calcareous substance without injury to the soft parts 
may likewise occur in the tubeworm, Spirorbis. 

Hirudinea. — Only a few data, chiefly physiological, are available concerning 
the effects of inanition upon the leeches. Valisnieri is cited as authority for 
the statement that Hiriido medicinalis requires three years for death from starva- 
tion. Some observations by Cajal ('04a) and Dustin ('06) on the nerve cells of 
fasting leeches will be stated in Chapter X. Bialaszewicz ('19) found but slight 
decrease in the fat of leeches during starvation. 

Putter ('11) states that in the blood-leech one meal of blood may last six or 
seven months, and that an additional six or seven months or more of fasting may 
be endured without the slightest injury. Weber ('14) says the blood-leech has 
an enormous gastric capacity (five or six times the volume of the entire empty 
body) and may not feed for months. 

Smallwood and Rogers ('10) noted that the leech Semiscolex kept without 
feeding for long periods gradually decreased in size. When the nerve cells from 
such starved animals are examined, either fresh or in stained sections, great 
changes are noted. "The whole cytoplasm of the cells has the appearance 
of a coarse foam structure. Here and there may be found the remnants of 
previously existing solid particles of stored up material." These particles were 
interpreted as food material, which is consumed during starvation. 

Similarly in the fish parasite, Pisciola, Erhard ('n) observed that droplets of 
glycogen occur normally in the glia tissue around the large ganglion cells. After 
three days of starvation, however, the droplets of glycogen begin to decrease 
around the nerve cells, but now appear in the cells. Glycogen is said to behave 
similarly also in other parts of the body. 


The investigations of inanition in the Echinodermata have concerned 
chiefly the effects of partial inanition (various salt deficiencies) upon the develop- 


ment of the sea-urchin and the starfish. A few observations have also been 
made upon general (total ) inanition in these forms. 

Echinoidea. — The effects of calcium deficiency upon the development of the 
sea-urchin were studied first by Pouchet andChabry ('89, '89a, '89b), who reared 
the larvae in sea- water from which a part or all of the calcium had been removed 
by precipitation with sodium or potassium oxalate. In media with about nine- 
tenths of the calcium remaining, the development appears normal up to about 40 
hours. "Mais, a la 60 heure, elles sont encore a l'etat de gastrula, tandis que 
les temoins ont des spicules ramifies et un intestin complet. Apres 90 heures 
ces larves, sans prendre des spicules, entrent dans une veritable phase pluteus, 
caracterisee pour elles par le differenciation de l'intestin en trois regions: 
oesophage, estomac et rectum. Mais la forme generate reste spherique, sans 
prolongements, et la mort survient apres quelques jours d'existence en cet etat. 

"En poursuivant une elimination plus complete de la chaux, les larves ne 
depassent plus le stade gastrula et meme le nombre de cellules qui l'atteignent 
devient de moins en moins grand. Lorsqu'on reste, au contraire, en deca de la 
quantite que nous avons indiquee, le developpement des spicules est simple- 
ment retarde, et ils subissent en outre, une deformation variable. Sur quelques 
larves, on observe la formation d'un appendice probosciforme, median, et 
qui semble tenir la place des deux prolongements anterieurs frontaux, sans 
spicule interne." Thus calcium deficiency results in retarded and abnormal 
development, especially in the skeletal system. 

The necessity and significance of the various salts in development was 
further revealed by the brilliant investigations of Herbst ('97). In artificial 
sea-water of varied composition he studied the effects of various deficiencies at 
various stages in the development of numerous organisms. He used chiefly 
sea-urchin and starfish, but made incidental observations upon several other 
invertebrates (also the fish Labrax lupus). The salt mixtures used contained 
various combinations and proportions of NaCl, KC1, MgSo 4 , MgCl 2 , CaS0 4 , 
CaCo 3 , Ca 3 P20 8 and FeCl 3 . Chemically pure salts were found desirable, since 
in some cases {e.g., iron), mere traces appear notably to affect the results. 

In the sea-urchin {Sphaer echinus granulans and Echinus microtuberculatus) , 
the results may be summarized as follows: In phosphorus-free solutions, the 
segmentation of the fertilized ova is abnormal and soon arrested (Fig. 17). 
Experiments beginning with later stages (blastula, gastrula or pluteus) like- 
wise resulted in prompt arrest of development and rapid death of the organisms. 
In sulphur -free solutions, the fertilized ova undergo retarded development, 
which proceeds only up to the formation of abnormal gastrulae (Fig. 18). 
Experiments beginning with later stages (blastula, gastrula or pluteus) resulted 
in death without further development. Chlorine and potassium (Fig. 24) were 
similarly found essential for segmentation, and also for continued development 
beginning at the later stages. In magnesium -free solutions the fertilized ova 
show no apparent difference from controls up to the gastrula stage (second day), 
but thereafter become retarded in development, not exceeding an abnormal 
pluteus stage with imperfect skeleton (Fig. 20). Magnesium was found essen- 
tial also for development in experiments beginning with later stages. Calcium 



was likewise found necessary for development of the fertilized ova» Even with 
CaS0 4 and CaCl 2 present in the solution, the development proceeds no further 
than an abnormal pluteus, with rudimentary skeleton (Fig. 22). The presence 
of CaC0 3 was found necessary, not only for the development of larvae with a 
normal skeleton, but also for the preservation of the already formed skeleton in 


Fig. 17. Fig. 16. Fig. 18. 

Figs. 16 to 25 illustrate the effects of various salt deficiencies upon the development 
of the sea-urchin, causing inhibition or distortion of the normal growth process. After 
fertilization in normal sea-water, the ova were placed in artificial media containing various salt 
mixtures. The normal controls in the complete salt mixtures pass through the segmentation 
stages, blastula (Fig. 25) and gastrula stages, reaching the normal pluteus stage shown in 
Fig. 16. All these figures (16 to 25) are taken from the monograph by Herbst ('97). 

Figs. 17 to 24 represent the most advanced stages reached in the solutions variously 
deficient. (After Herbst '97.) 

A, arms; a, a, a — segments of alimentary canal; Be, blastocele; Cc, (abnormal) ciliated 
crown; em, egg membrane; m, mesoderm cells; Sk, skeletal rods; Sp, "snout-like" process; 
Vac, vacuoles in blastula cells. 

Fig. 16. — Normal pluteus stage of the sea-urchin Echinus, reared in a complete salt mixture 
containing NaCl, KC1, MgS0 4 , CaSCu, CaCOs, FeCOs, and CaHP0 4 . Fourth day. 

Fig. 17. — This stage of (abnormal) segmentation represents the maximum development 
reached in the same salt mixture as the foregoing (Fig. 16), with omission of the phosphate. 
Most of the ova failed to segment at all, or were arrested in still earlier stages of abnormal 
segmentation. This illustrates the necessity for phosphorus in normal development of the 
sea-urchin, Echinus. 

Fig. 18. — Abnormal gastrula of Echinus, representing the maximum stage of development 
reached in an S-free mixture of NaCl, K'Cl, MgCl 2 , Ca 3 P 2 8 , CaCOs and FeC0 3 . Third day. 
Shows that sulphur is necessary, for normal development. 

experiments beginning with the pluteus stage. Finally, the fertilized ova and 
later stages were observed to undergo retarded and abnormal development 
unless iron was present, at least in traces (Fig. 23). 

The abnormal form and structure of the larvae appeared variable, yet to 
some extent characteristic according to the type of deficiency, as shown by 
Figs. 16 to 25. Herbst reached the general conclusion that "Das wichtigste 



Fig. 19. Fig. 20. 

Fig. 19. — Larva (abnormal pluteus) of the sea-urchin Echinus reared in a solution the 
same as the preceding, but with addition of MgS04. Fourth day. Except for the absence 
of the skeleton, the internal structure corresponds somewhat to that of the normal pluteus 
stage (cf. Fig. 16); but the external form is very different, arms undeveloped, etc. The 
addition of CaS04 to the medium permits normal development, indicating that the organism 
requires this salt as a source of sulphur, that in MgS04 being inadequate. 

Fig. 20. — Abnormal pluteus representing the maximum stage of development reached by 
the sea-urchin Echinus reared in an Mg-free mixture. Third day. No arm-processes or oral 
invagination. Gut and skeletal development somewhat retarded. Indicates the necessity 
of magnesium for normal development. 

Fig. 2i. Fig. 22. 

Fig. 21. — Abnormal pluteus of sea-urchin Sphaerechinus reared in a mixture containing 
CaHP04, but not CaC03. Third day, showing the maximum stage of development reached. 
The form is abnormal; the "snout-like" process and absence of skeleton being especially 
characteristic in the absence of calcium carbonate. 

Fig. 22. — Abnormal pluteus of the sea-urchin Echinus, reared in mixture containing NaCl, 
KC1, MgS0 4 , CaS0 4 , Ca 3 P20s and FeC03. Deficiency in CaC03 results in complete absence 
of skeletal development, and in abnormal external form differing from that shown for Sphaere- 
chinus in the preceding figure. Third day, representing the most advanced stage of develop- 
ment reached. 



meiner Versuchsresultate besteht in dem Nachweise, dass die zum Aufbau des 
Embryo nothwendigen Baustoffe im Ei nicht in solchen Quantitaten vorhanden 
sind, dass sie bis zu dem Stadium, wo die Vermehrung des Bildungsmaterials 
durch Nahrunsgsaufnahme moglich ist, also zum Pluteusstadium reichen, son- 
dern dass sie dem Meerwasser zum Theil bereits bei der Furchung entzogen 
werden. Die nor male Entwicklung der Seeigellarven hdngt also nicht nur von 
einer bestimmten physikalischen, sondern vor alien Dingen von einer bestimmten 
chemischen Beschajjenheit des umgebenden Mediums ab." Herbst's results upon 
the starfish and other animals (Coelenterata, Platyhelminthes, Tunicata and 
fish) are mentioned under the corresponding sections. 

Fig. 23. Fig. 24. Fig. 25. 

Fig. 23. — Abnormal pluteus of the sea-urchin Echinus, reared in mixture of NaCl, KC1, 
MgS04, CaSO-i, CaaP20s and CaC03. Third day, representing the most advanced stage of 
development (rarely reached) in Fe-free media. Indicates that iron is essential for normal 

Fig. 24. — Abnormal blastula of sea-urchin Sphaerechinus, reared in same mixture as the 
specimen shown in Fig. 25, excepting absence of potassium salt. Note small size, thick wall, 
absence of cilia and vacuoles. Demonstrates that K is necessary for normal development. 

Fig. 25. — Normal blastula of sea-urchin Sphaerechinus, reared in complete salt mixture. 
Control to the preceding Fig. 24. Note size of blastula, cilia and cellular vacuolation. 

In a later investigation, Herbst ('00) studied in greater detail a peculiar 
histological change found in the developing sea-urchin and other forms placed 
in Ca-free solutions. " Durch das Fehlen von Calcium im umgebenden Medium 
wird der Verband der Furchungszellen membranloser Eier der Seeigel derartig 
aufgelockert — und zwar bei Echinus radikaler als bei Sphaerechinus — dass die 
einzelnen Zellen zum Teil sogar durch grossere Zwischenraume von einander 
getrennt werden. Trotz dieser ganzlichen Isolation oder Auflockerung ver- 
lauf t aber die Furchung bis zu Ende, ja es trifft sogar Differenzirung in Wimper- 
zellen ein, die, auch wenn sie ganzlich isolirt sind, doch einige Zeit am Leben 
bleiben und sich munter bewegen konnen. Der Calciummangel wirkt also 
zunachst nur specifisch auf den Zusammenhalt der Zellen, nicht aber auf die 
Lebensenergie ein, deren endlichen Erloschen vielleicht iiberhaupt nicht an dem 
Fehlen des Kalkes, sondern vielmehr an der Isolation, an dem Fferausreissen 
aus dem Gesamtorganismus liegt." Experiments on later stages gave similar 


results. When the egg membrane is left around the segmenting ovum in the 
Ca-free water, complete separation of the cells is prevented, and upon restora- 
tion to normal sea- water they may reunite and continue normal development. 
These results of Herbst have been applied in the theory of calcium deficiency in 
scurvy of vertebrates. 

We have already noted the pioneer work of J. Loeb ('92) in demonstrating 
the necessity for potassium, sodium and magnesium salts for the normal growth, 
development and regeneration of the Coelenterate Tubularia. This work was 
continued by Loeb ('05, 'n), extending his results in general to Echinodermata 
{Arbacia), Crustacea (Gammarus) and Vertebrates (Fundulus). With the fer- 
tilized Arbacia eggs, placed in various mixtures of NaCl, KC1, and MgCU, 
Loeb found that any one salt permitted segmentation (often abnormal) only 
up to a maximum of 64 cells (in MgCl 2 ). A mixture of two chlorides (MgCl 2 
and CaCL) may permit reaching the blastula stage; while three chlorides 
(NaCl, CaClg and KC1) made it possible to reach the gastrula, or even the plu- 
teus stage, without skeleton. The addition of NaoCOs gave plutei with normal 

While Loeb's results are thus to a certain extent in general agreement with 
those of Herbst, his interpretation of them is quite different. On account 
especially of his experiments with Fundulus eggs, Loeb believed that the injuri- 
ous effect of various salt deficiencies is due, not to the direct need for the 
deficient salt in the developing organism, but rather to the toxic effect of the 
other salts remaining in solution. Thus Loeb ('05) concludes: "It seems to me 
that my experiments necessitate the introduction of a new conception, namely, 
that of physiologically balanced salt solutions. By this I mean salt solutions 
which contain such ions and in such proportions as completely to annihilate the 
poisonous effects which each constituent would have if it were alone in solution." 
The principle that certain salts may function in neutralizing the toxic action of 
others has been referred to previously in the chapter on plants. 

As to the mechanism of this protective action, Loeb ('11) states: "These 
observations on the sea-urchin egg, therefore, suggest the possibility that the 
combination of the three salts in their definite proportion and concentration has 
the function of forming a surface film of a definite structure or texture, around 
the protoplasm of each cell, by which the protoplasm is kept together, protected 
against and separated from the surrounding media." 

The effects of total inanition upon the larvae of the sea-urchin, Strongylo- 
centrotus lividus, were carefully studied by Runnstrom ('12, '12a). The rate of 
involution varies directly with the temperature, but the larvae may survive 
starvation for 60 or 70 days at i8-iq°C, decreasing to one-half the diameter 
of the original ovum, or less. Two types of involution occur: (1) the skeleton is 
less affected and the arms persist, although the alimentary canal shows marked 
changes; or (2) the skeleton is markedly resorbed and the arms greatly shortened, 
although the other structures may be less changed. In general the hindgut 
(rectum) undergoes the most marked reduction. The lining epithelial cells 
becomes first cylindrical, later shortened and finally detached and migratory. 
The mesenchyme cells are actively phagocytic and migratory, transporting 


materials (including disintegrating pigment cells ) from the regions undergoing 
atrophy to those where further growth occurs. Thus, by a process of "auto- 
differentiation," further development (e.g., anlages of the pedicellariae) may 
occur at the expense of the remainder of the body. The inanition involution is 
therefore not purely a reversal of the normal developmental process. Runn- 
strom refers the morphological changes directly to physico-chemical conditions, 
varying with the food supply and resulting differences in the acidity, permeabil- 
ity, etc. of the various cells and tissues. 

Asteroidea. — The experiments of Herbst ('97) upon the effects of various 
salt deficiencies, which have already been described for the sea-urchin larvae, 
were also in some cases extended to the developing starfish (Asterias glacialis) 
with very similar results. In phosphorus-free salt mixtures the fertilized ova 
fail to segment normally and development never goes beyond the blastula stage. 
Bipinnaria stages in phosphorus-free solutions die within a day. Sulphur 
(sulphates) and potassium salts were likewise shown to be necessary for the 
normal early development of starfish. Tests were not made for the other sub- 
stances found necessary in the development of the sea-urchin. 

Mead ('oo) obtained great differences in size between fed and unfed starfish. 
He noted that sexual maturity is correlated with the attainment of a certain 
size (50 mm.). "When food is accessible, the starfish eats voraciously and 
grows with great rapidity, but, on the other hand, it will live for months almost 
without food and apparently remain healthy, though it does not grow." 

Schultz ('08b) kept recently metamorphosed larvae of the starfish Asterias 
rubens in filtered water to see whether they undergo a "reduction" to embryonal 
form as he found in Hydra and Planaria. Growth ceases, but no developmental 
reversion occurs. At the end of three weeks, most of the starfish show no change 
in size or structure, excepting the intestinal gland-cells or granule-cells, which 
become fewer and may disappear entirely. Even in more prolonged starvation, 
no typical "reduction" occurs, but degeneration gradually supervenes. The 
cells of the epithelial band soon become detached. Proliferated and degen- 
erated cells fill all the cavities of the body, the gut-lumen first, and the stone- 
canal last. The lumina of the blood-vessels and water-vascular system become 
obliterated, and the body cavity filled with connective tissue. The muscle 
and nervous system persist unchanged, however, and the size of the nuclei in 
general is unaffected. 


Of the phylum Mollusca, the effects of inanition have been studied chiefly 
in the class Gastropoda, with a few observations upon the Pelecypoda. 

Gastropoda. — The remarkable resistance to inanition by the snail Helix 
was noted by several earlier observers cited by Lucas (1826). Helix (sp.?) 
was said by Macbride (1774) to endure starvation for a period of 15 years! For 
Helix nemoralis, Miiller noted a period of one year. For Helix pomatia, 
Sorg (1805) observed endurance for six months, and Wiesmann for one year. 
Semper ('81) stated: "I myself kept various species of land-snails for years 


wrapped in paper and quite dry in wooden boxes, and thus wholly without 
food, and many of them are at this day alive and active." 

In various species of (edible) Helix during starvation, Sabrazes ('02) found: 
"La perte de poids peut se chiffrer par la moitie du poids primitif, et cela 
dans un laps de temps de quelques jours au bout dequels les escargots s'accolent 
les uns aux autres et sont proteges contre la dessication par une mince membrane, 
d'une aspect parchemine, parfois creusee d'un petit opercule, tendue a orifice 
de la coquille. Les animaux peuvent rester ainsi, a l'etat de vie latente, plus- 
ieurs mois . . . M. Devaux a observe aussi des faits analogues." 

More recently Krahelska ('13) found that Helix pomatia endures starvation 
at i7°C. for six months to a year, but that Helix arbustorum is less resistant. 

Slowtzoff ('03a) found in Helix pomatia subjected to total inanition a gradual 
loss of body weight up to 25.74 per cent, the loss affecting the shell as well as 
the soft parts. The change in chemical composition was determined. 

From their experimental studies on the cytology of invertebrate nerve 
cells, especially in the molluscs Planorbis, Limax agrestris, and Limax maximus, 
Smallwood and Rogers ('08, '09, '10,) conclude that the lipochrome pigment 
granules contained in these cells vary according to nutritive conditions. They 
slowly decrease in size and number during hibernation and prolonged starvation, 
with corresponding increase in cytoplasmic vacuolation. The pigment granules 
in the nerve cells of Limax maximus disappear entirely in advanced starvation, 
apparently with no shrinkage of the cell body or nucleus. The granules repre- 
sent stored nutriment, probably comparable to the Nissl bodies. Legendre 
('09) described in detail the various forms of degeneration observed in Helix 
pomatia after prolonged inanition and other nutritional disturbances. The 
cell changes are variable, including chromatolysis and cytoplasmic vacuolation. 

Although hibernation represents a special condition not comparable to 
ordinary starvation, it may be noted that Cattaneo ('92) observed decreased 
ameboid activity in the blood cells of Helix during hibernation; and Erhard 
('11) found a decrease in the glycogen droplets in the nerve cells and surrounding 
glia tissue. I have not been able to obtain the thesis by Bellion ('09), but 
Moglia ('10) and Legendre ('13) noted an apparent increase in the pigment 
granules of the nerve cells during hibernation. 

The most extensive study of inanition in molluscs was made by Krahelska 
('10, '12, '13). In her first ('10) paper, the losses in body weight were found 
subject to variation according to individuals and species {Helix pomatia, 
Helix arbustorum, Helix jruticum and Leucochroa candidissima), the total loss 
varying from 10.76 — 47 per cent or more. In general the loss during hiber- 
nation is much less than that during a corresponding period of inanition while 

After two months of starvation, the kidneys of Helix arbustorum show 
marked histological changes. The cytoplasm of the epithelial cells becomes 
greatly reduced in amount and often syncytial in character, being homogeneous, 
finely fibrillated or vacuolated (beginning degeneration) . It also shows enlarged 
concretions in vacuoles. The nuclei are sometimes enlarged and vesicular, 
sometimes almost pycnotic. After four months, these changes are more pro- 



nounced; all nuclei are now small and variably pycnotic. The changes during 
hibernation are somewhat similar, but slighter, and there is no cytoplasmic 
atrophy, probably because of the preparatory period preceding hibernation. 
Krahelska later ('12, '13) studied thoroughly the histological changes in 
the albuminous gland (accessory gland of the hermaphroditic sexual duct) 
of Helix pomatia and Helix arbustorum during inanition and hibernation. In 
Helix pomatia, during five or six months of total inanition, with loss of 40 










V \ 



>. N. 



— _ 


. >< --- 

— — - 



■^rrfeio " ■ 




10 12 14 

Weeks of Inanition 





Fig. 26. — Chart showing the changes in the volumes of entire gland-cell, cytoplasm and 
nucleus, in the albuminous (accessory sexual) gland of the snail. Helix pomatia during inani- 
tion up to a period of twenty-two weeks. (From Krahelska '13.) The volumes were esti- 
mated by projecting the magnified cells in sections upon paper, and measuring the correspond- 
ing average areas of cell, cytoplasm and nucleus. There is a marked and progressive decrease 
in the entire cell and the cytoplasm. The nucleus decreases but slightly, however, which 
causes a marked rise in the nucleus-plasma ratio. 

per cent (or more) in body weight, the non-glandular sexual ducts are not 
greatly atrophied, but the albuminous gland undergoes reduction to about one- 
half in length. The color of the gland also changes from milk-white to yellow, 
later orange-brown. The individual gland-tubules show marked atrophy 
(Fig. 27). The corresponding progressive decrease in the size of the cells, 
together with changes in the relative amount of nucleus and cytoplasm, are 
shown in Fig. 26. It is evident that the cytoplasm decreases much more rapidly 
than the nuclei, resulting in a marked increase in the nucleus-plasma ratio. 
The "parenchyma" (stroma) nuclei, which at first are only half as numerous as 
the glandular epithelial nuclei (in a given field) proliferate by amitosis and become 
more abundant, finally becoming almost equal to the glandular nuclei in number. 
The cytoplasmic changes during inanition (Figs. 27, 28, 29, 30) may be 
roughly classified in two stages. (1) During the phase of "reduction," the cells 
become undifferentiated, and return to a somewhat embryonal condition (Fig. 



;t mm 

g^i— f 

Fig. 27. Fig. 28. 

Fig. 27. — Cross sections of two gland-tubules from the albuminous (accessory sexual) 
gland of the snail, Helix pomalia; to show the effect of inanition. (From Krahelska '13.) 
a, section of a normal gland tubule; bases of the cells external, with nuclei and granular cyto- 
plasm; the vacuoles in the cells correspond to the secretion granules; lumen lined by a flattened 
synctial layer, b, corresponding section of a gland-tubule after about five months of starva- 
tion. Tubule and constituent cells greatly reduced in size; cells fused into a syncytial mass, 
enclosing the deeply-staining, pycnotic nuclei. 

Fig. 28. — Normal gland cell from the albuminous (accessory sexual) gland of the snail, 
Helix arbustornm. (From Krahelska '13.) This corresponds to a single cell of those shown in 
Fig. 27a, but more highly magnified ( X800). The nucleus, N, is at the base of the cell; the 
cell body is largely composed of vacuoles enclosing secretory granules and their associated 
" chromatoplasts " (chromidial apparatus). 

- ... rn ^.J& 

Fig. 29. Fig. 30. 

Fig. 29. — Three gland cells from the albuminous (accessory sexual) gland of the snail, 
Helix pomalia, after eight weeks of starvation. (From Krahelska '13.) X800. The nuclei 
are hyperchromatic but slightly changed in size; while the cytoplasm is greatly reduced in 
volume. Two of the cells each contain a large vacuole, enclosing granular remnants of the 
secretion-granules and associated structures. 

Fig. 30. — A portion of a gland-tubule from the albuminous gland of the snail, Helix 
pomatia, after twenty weeks of starvation. X800. (From Krahelska '13.) Degenerative 
stage, the nuclei of gland cells and stroma being intermingled in a scanty vacuolated degener- 
ated mass of cytoplasm. Many cells show necrosis, with disintegrating nuclei. 


29). The secretory granules undergo a progressive absorption, and the "chro- 
matoplasts" (chromidial apparatus, etc.) undergo regressive changes. This 
phase occupies about two months. Then follows (2) the phase of degeneration, 
with syncytial cell-fusion, karyorrhexis and pycnosis. These are destructive 
changes leading to necrosis (Fig. 30). 

If the ordinary room temperature (i7°C.) is increased to 32°C, the inanition 
changes usually requiring four months are attained in three weeks. Changes 
very similar to those during inanition appear also as a result of functional 
exhaustion at the end of the normal egg-laying period. No structural changes 
occur in the gland during hibernation, unless artificially prolonged. In a snail 
richly fed for two days after five months of starvation, evidences of recuperation 
were found already beginning in the gland cells. 

Pelecypoda. — Sorg (1805) noted death from starvation in 18 days in My a 
ftictorum. Mead Coo) made a few incidental observations indicating that the 
relations of growth and nutrition in the clam and oyster are similar to those 
already metioned for the starfish. Maas ('07, '12) noted that in Ca-free water 
the calcareous substance in the shells of young mussels (species not stated) 
may be resorbed without injury to the soft parts. Schultz ('08b) found that 
during inanition, in filtered water, the young mussels Mytilus (like the starfish) 
cease to grow, but undergo no "reduction" to embryonal type and die in about 
three weeks. Sections show no important changes. "Die Verdauungsorgane 
erleiden natiirlich wahrend des Hungers die ersten Veranderungen, und die 
Zellen der Leber, z.B., verschmelzen und bilden ein Syncytium." 


The effects of inanition in the rotifers (wheel animalcules) have been studied 
chiefly in their relation to reproduction and sex. Leydig noted that when 
rotifers are kept a few days in water without food, the ovary shrinks, and the 
granular yolk-mass almost entirely disappears. All such individuals produce 
winter eggs. Regarding the effects of nutrition upon sex, he concludes: " Wenn 
wir sehen, dass bei Aphiden, Daphniden, Rotatorien, Mannchen unter dem 
Einfluss allgemeiner Ursachen, als da sind Nahrung, Warme und Kalte, zum 
Vorschein kommen, so haben wir einstweilen einen Anhaltspunkt zu der Ver- 
muthung dass die Differencirung des Geschlechts auch in anderen Gruppen 
ahnlichen allgemeinen Einwerkungen unterworfen sein konne." 

Nussbaum ('97) concluded: "Bei Hydatina senta bestimmt wahrend einer 
gewissen Entwicklungsphase die Ernahrung das Geschlecht des ganzen Geleges 
eines jungfraulichen Weibchen. Wird das auskriechende Weibchen bis zur 
Reifung seines ersten Eies gut ernahrt, so legt es nur weibliche Eier; wird es bis 
zur Geschlechtsreife mangelhaft ernahrt, so legt es nur mannliche Eier. Vor 
und nach dieser Periode hat die Ernahrung auf das Geschlecht keinen Einfluss." 
Nussbaum's results were comfirmed by Lenssen ('98), but opposed by Punnett 
('06), who concluded that the sex is due to internal factors, unaffected directly 
by food or temperature. 


Whitney ('08) likewise obtained negative results with Hydatina senta. 
Temperature has no influence upon sex-determination (versus Maupas), and 
"starving the young females for the first few hours after they hatch does not 
cause them to produce a higher percentage of male eggs." 

Shull ('10, 'n) after careful and extensive experiments upon Hydatina 
senta concluded that while sex-production is dependent upon both internal and 
external factors, the quantity of food probably has no influence in this respect. 
Starvation may be accompanied by an increased proportion of parthenogenetic 
male-producers, but this is probably only an indirect effect, due to a decrease in 
certain substances incidentally introduced by the food. Thus the question as 
to the effect of nutrition upon sex-production in rotifers still remains somewhat 


In the phylum Arthropoda, the effects of inanition have been studied most 
extensively in the class Insecta. There have also been numerous observations 
upon the Crustacea, and a few on the Myriapoda and Arachnida. 

Crustacea. — The investigations in this class have included both the sub- 
classes, Malacostraca and Entomostraca, and those in the latter group will be 
considered first. The Entomostraca, like the rotifers, have attracted attention 
on account of the apparent effect of inanition upon sexual reproduction. Leydig 
included the water flea Daphnia among the forms in which sex is determined by 
external factors, including nutrition. Kerherve ('92) claimed that Daphnia 
magna is parthenogenetic during abundant nutrition, but is quickly transformed 
into sexual reproduction by unfavorable conditions, especially by inanition. 
Cuenot ('94) noted a similar condition in Moina rectirostris. Issakowitsch 
('05) in Simocephalus vetulus found that sexual reproduction is induced during 
the asexual stages by either low temperature or inanition. Woltereck ('08, 
'09, '11) admits the hereditary cycle in daphnids, but claims that in Daphnia 
longispina a peculiar variety of different body form is produced by unfavorable 
environment, chiefly by malnutrition. The effects of environmental factors 
apparently may be hereditary in Daphnia longispina and in Hyalodaphnia 
cucullaia. The external factors affect sex only in the "labile period" and then 
only indirectly, through influence upon the internal mechanism of sex-production. 
Finally McClendon ('10) concludes that "The life cycle of a Daphnid is therefore 
an hereditary tendency but can be influenced by nutrition and probably by 
temperature and the accumulation of excretions." 

On the other hand, Green ('19) has recently made careful and extensive 
experiments upon Simocephalus vetulus, concluding that: "The sexual state is 
probably determined in the ovary of the preceding generation. There are 
almost certainly predisposing factors in the environment but it is not certainly 
known what they are. Food or lack of food does not offer a sufficient explana- 
tion." It would therefore appear that among the Entomostraca, as already 
noted for the rotifers, the effect of inanition upon sex-production is still an open 
and uncertain question. 


Kerb ('10) found that Daphnia during starvation may undergo repeated 
ecdysis, moulting in spite of rapid decrease in body weight (from 105-111 mg. 
to 24-28 mg. in n days). Lipschiitz ('13) cites similar observations by Knor- 
rich ('01) and Wolff ('10). 

Among the Malacostraca, Haller (cited by Lucas, 1826) stated that the crab, 
Cancer maurinus, can endure a starvation period of eighteen months. Mead 
('00) noted incidentally that the relations of nutrition and growth in the lobster 
{Homarus) are similar to those already stated for the starfish. Przibram ('07) 
described the regeneration of appendages in certain Crustacea (Trypton spongi- 
cola, etc.) and concluded: "But not only may the means of regeneration and 
compensation be clearly shown to occur in this case but also reduction is involved 
to an appreciable degree, especially if the crayfish is starved during the experi- 
ment. Then each moult shows the shedding of a smaller skin and the animal is 
at the end of the transposition in all dimensions smaller than at the time of the 

Irvine and Woodhead ('88, '89) experimented with common edible shore 
crabs {Cancer) which, after ecdysis, were placed in artificial sea-water. They 
were found able to obtain the calcium carbonate necessary for the calcareous 
exoskeleton when CaCl2 is the only calcium salt present in the sea-water, but 
cannot utilize CaSo4 for this purpose. J. Loeb ('11) found that the marine 
Gammarus dies quickly in distilled water, even when made isotonic by addition 
of sugar solution. The addition of NaCl alone, or of KC1 and CaC^, is insuffi- 
cient ;• but if all three salts are added in proper strength, life is made possible. 
Loeb's theory of the action of the salts in such cases has been stated in connection 
with the Echinoderms. 

Brunow ('11) in a biochemical study of the crayfish Astacus fluviatilis found 
the loss in body weight proportional to the length of inanition, decreasing from 
19.16 g. to 17.47 g. (loss of 8.8 per cent) in 70 days, and to 16.26 g. (loss of 15. 1 
per cent) in 140 days. Morgulis ('23) found but 3 per cent decrease in the body 
weight of the lobster (Homarus) after 56 days of starvation, and cites observa- 
tions (by Moore and Herdman) showing no loss in weight after eight months 
of fasting, the loss in dry substance being masked by absorption of water. 

Arachnida. — Lucas (1826) observed that the spider Epeira phalangoides 
lived one and one-half months without food, while Haller noted a period of 
several months for Epeira Walk. According to Lucas, De Geer found an 
apparent increase in the weight of Epeira diodema kept one month without food 
in 78 cubic inches of air. Treviranus gave 18 months as the starvation period 
for Scorpio europaeus. Morgulis ('23) cites reports by Blackwell of 17 months 
of starvation for spiders, and by Jacquet of 368 days for scorpions. 

Myriapoda. — Plateau ('78) observed circular constrictions of the intestine in 
the Myriapod Julius starved 15 days. Childs ('21) starved Parajulus for 13- 
42 days and obtained results somewhat resembling those of Needham ('97) on 
dragon-fly nymphs. "Prolonged absence of food matter from the midgut of 
Parajulus causes more or less irregular thickenings of the epithelium which give 
the margin of the lumen a wavy contour and are attended by a reduction in the 
thickness of the brush border. These thickenings are caused by the inhibition 


of the tendency toward senescence and discharge on the part of the mature cells, 
coupled with continued reproduction and growth on the part of the regenerative 

Cattaneo ('92) observed that in the Myriapod Glomeris, as in the snail 
Helix, the ameboid activity of the blood cells appears decreased during 


Baumberger ('19) concluded that growth of insects in general may be 
limited by lack of protein. More or less extensive observations upon the effects 
of inanition of various types upon insects have been made upon all of the princi- 
pal orders, excepting the Aptera. 

Orthoptera. — According to Lucas (1826), Vaillant noted a starvation period 
of five months in the grasshopper, "eine grosse Heuschrecke" (Sp.?). Sanford 
('18) noted that the cockroach Periplaneta orientalis may endure starvation for 
three weeks; or even two months or more if the crop is distended with food. The 
epithelium of the crop may absorb and store large globules of fat which are 
slowly absorbed during inanition. Bodine ('21) found that the grasshopper 
Melanoplus femur rubrum endures total starvation without water 73 hours, or 
with water, 144 hours; the loss in body weight being 30-35 per cent. Melano- 
plus differ entialis, a large species, endures 96 hours without water, or 172 hours 
with water; loss in body weight, 20-25 per cent. Quiescent nymphs of Chorto- 
phaga live without food for about two weeks at o°-9°C, one week at 23 , but 
only three or four days at 3 8°; the maximum loss in body weight at death being 
20-25 per cent. 

Neuroptera. — Lucas (1826) observed death from starvation in Hcmerobius 
(Perla) after two days; in Perla bicaudata after two and one-half days; in 
Agrion (Fabr.) Virgo after four days; and in Nemoura (Latr.) nebulosa after 
seven days. Needham ('97) studied the midgut epithelium of dragon-fly 
nymphs, and found that after two months of fasting the columnar cells increase 
three-fold in height. The striated border has mostly disappeared, and the 
cells are filled with secretory granules. The small cells in the depressed cell- 
nests increase in number, and appear to be centers of regeneration for the sur- 
face epithelium. Slowtzoff ('04) found that the Libellulidae (dragon-flies) 
die after only 60-84 hours of total inanition, with excessive loss of water, 
which perhaps causes the rapid death. The loss in body weight averages 22.55 
per cent. 

According to Grassi, among the Termites ("white ants") the development 
of workers and soldiers is regulated by the character of the nutrition. 

Hemiptera. — Dufour (1833) found that the bedbug (Cimex lectularius) 
may live for a year in a closed vial without taking food. In a Brazilian 
species, Conorhinus megistus, Neiva ('10) observed a female specimen which 
had been alive 57 days in a tightly closed box. According to Weber ('14): 
"Von der Bettwanze wird angegeben, dass sie in den Bettvorhangen eines 6 
Jahre lang unbewohnten Zimmers angetroffen wurde, blattdiinn und fast 
durchsichtig." Riley and Johannsen ('15) noted that the ability of the bedbug 


to endure starvation is one factor explaining the long periods in which deserted 
houses and camps may remain infected. 

Among the plant-lice (Aphidiidae), as in the case of the rotifers and daphnids, 
the effects of inanition have been studied chiefly in their relation to sexual repro- 
duction. Kyber (1813) noted that male forms result from underfeeding, a 
result apparently confirmed by Leydig and others. Goldi ('85) found that with- 
drawal of food results in the appearance of the winged forms in Pemphigus 
xylostei, Pemphigus bumeliae and Lachnus. Thus even in June the winged 
forms of Schizoneura lanigera are obtainable, leading directly to the sexual 
generation. Keller ('87) and Behr ('92) observed that also in the grape-louse, 
Phylloxera vastatrix, deficiency of food causes cessation of parthenogenesis and 
the appearance of the sexual, winged forms, containing males as well as females. 
According to Keller ('87), this confirms the theory of Landois and Dusing as 
to the relation of nutrition to sex, the males appearing only under unfavorable 

Diptera. — Cuenot ('99) found that in the maggots of flies the sex-ratio is 
not materially affected by the quality or quantity of food. Tangl ('09) 
found that in fasting larvae of Ophyra cadaverium, the metabolism is chiefly 
at the expense of the fat. Starvation of the larvae retards the process of 
metamorphosis. In connection with a study of the chromidia, Popoff ('10) 
noted changes in the fat cells, "oenocytes" and pericardial cells of the housefly 
during feeding and inanition. Guyenot ('13, '13a, '13b, '13c) made a series of 
studies upon the relations of nutrition and reproduction in the fruitfly, Droso- 
phila ampelophila. If the larvae are reared on sterilized potato, instead of 
yeast, they undergo metamorphosis but with atrophic gonads, sexual maturity 
being markedly retarded. The eggs laid by such abnormal forms are few in 
number and give rise to weak and short-lived larvae. In the second article 
('13a), it was shown that the fecundity depends upon the environment, not 
only of the larvae, but also of the pupae and the adult fly. Drosophila is thus 
intermediate between these insects (certain butterflies) whose sexual maturity 
is determined entirely by larval nutrition, and those (Calliphora) where the 
sexual maturity is not so affected and depends almost entirely upon adult nutri- 
tion. Later ('13b) it was shown that when adult females are placed on poor 
nutriment (potato or carrots) the eggs become abnormal and greatly reduced in 
number. " Apres avoir pondus quelques oeufs, donnant des larves, les femelles 
pondent quelques oeufs dans lesquels se forme un embryon, generalement 
anormal, qui meurt a un stade plus ou moins avance." Later the malnourished 
females lay unfertilized ova, even when conjugation with the male occurs, the 
spermatozoa apparently being resorbed in the seminal receptacle of the female. 
Finally, Guyenot ('13c) demonstrated that the process of egg deposition is, 
in part, determined by the nutritional conditions. 

Further studies on the nutritional relations of the fruitfly (banana-fly) 
Drosophila were made by J. Loeb ('15, '15a) who grew five successive generations 
on a solution of glucose, cane sugar, ammonium tartrate, citric acid, dipotas- 
sium phosphate and magnesium sulphate. Bacteria and yeast were not 
excluded, however. Loeb and Northrop ('16), by special precautions, reared 


twelve completely sterilized generations on a nutrient solution containing baker's 
yeast and citric acid. In many experiments on other media (e.g. filter paper 
plus cane sugar and salts; likewise with addition of casein, edestin, egg albumin, 
milk or a mixture of aminoacids), the larvae attained normal size, but did not 
undergo metamorphosis. Sterile flies, grown on sterile bananas or potatoes, show 
no sexual development. Yeast is apparently the only adequate food for these 
flies. Loeb and Northrop ('17) and Northrop ('17) found that, unless yeast is 
added, growth of Drosophila ceases on aseptic cultures and that by inadequate 
feeding during the larval period the total normal duration of life of nineteen 
days may be prolonged up to twenty-nine days. The length of life of the imago 
is not affected by the earlier period of malnutrition. 

Baumberger ('19) concluded that "Drosophila living in fermenting fruit 
are dependent for their food supply on the synthetic and absorptive powers of 
yeast cells. In a similar manner, my study of the relation of Musca domestica 
to manure, of Desmometopa to decaying meat, and of Sciara and Tyroglyphus 
to decaying wood shows clearly that these Arthropods also feed on microorgan- 
isms." Both adults and larvae of Drosophila require sugar as food, but cannot 
live on sugars or nucleoproteins alone. Since Drosophila can be reared normally 
on yeast nucleoprotein, sugar and salts, any "special substance" required must 
be present in this mixture. The larvae can be maintained for a long time on a 
minimum of protein (banana diet) at constant or slowly increasing size, and may 
later develop to normal size on adequate (yeast) diet. The total span of life 
may therefore be increased eleven days to forty days of more. There is a 
tendency for the larva to pupate after a certain length of time, whether it reaches 
the maximum size before this period or is still undersized from malnutrition. 

Pearl and Parker ('24) made an extensive experimental study of the duration 
of life in Drosophilia melanogaster during complete (total) inanition. The 
mean duration is slightly less than two days. The duration in the females is 
more variable than in the males, both absolutely and relatively, but in both 
sexes the variability is relatively much less than during full feeding. The females 
are longer lived than the males, during inanition as well as when full fed. 
Although the normal wild-type flies during full feeding live about three times 
as long as the vestigial type, during total inanition the mean duration of life is 
nearly the same in both types. The genetic significance of these results is 

Vinokuroff ('22) found the mean duration of life in the common housefly 
{Musca domestica) to be 1.3 days when starved without water, or 1.8 days on 
water alone. Similar observations were made by Glaser ('23). 

Lepidoptera. — Lucas (1826) noted death from starvation in the moth Bombyx 
in three days; in another case (Bombyx cerura Schrank) in 15 days, in June. In 
the saturniid moths, which normally take no food in the imaginal stage and live 
only about eight days at ordinary temperature, Rau and Rau ('12) found the 
duration of life extended to about 18 days at low temperatures. 

Semper ('81) stated that T. Gentry of Philadelphia, found underfed larvae of 
Acronyeta (sp.?) produce smaller pupae and moths. Von Linden ('07) observed 
that one specimen of adult Hylophila prasinana lost 43 percent of its body weight 


in 11 days of starvation; another lost 70 per cent in 17 days; while a larger 
Papilio podalirius lived only ten days with loss of 29 per cent. The loss 
of the corresponding pupae is much less, being only 7.8 per cent for the Hylo- 
pkila in 30 days. The loss is relatively greatest in the latter part of the fasting 
period for the pupae, but in the earlier part for the imago. Kellner ('87) noted a 
slight prolongation of the larval stage in underfed silkworms (Bombyx mori). 
Kellogg and Bell ('03) found that underfeeding the larvae in this species causes 
dwarfing of the moths, the effect being decreasingly evident in the second and 
third generations. They further ('04) noted that starvation of the larva one to 
four days does not affect the time of metamorphosis, or the size or fertility of 
the moth. Starvation of four to seven days, however, reduces the last inter- 
moulting stage, resulting in a normal (though smaller) cocoon and moth. Death 
occurs if the larva is starved eight days or more. The cocoon loses about four 
per cent of its weight on the first day. The pupa loses slightly but steadily 
during the pupal period (2-10 days), with total loss of about 14 per cent. 
In the Tent caterpillar (Clisiacampa) and the Mourning-cloak butterfly 
(Euvanessa antiopa) the loss is steady but greater, amounting to 35 and 65 
per cent, respectively. Kopec ('24) during intermittent starvation of the larvae 
of the moth, Lymnantria dispar L., found a considerable prolongation of the 
larval life with a slight abbreviation of the pupal period. The body weight 
is decreased and the adipose tissue exhausted. Kopec ('24a) further noted that 
when the female moths derived from starved caterpillars are mated with normal 
males, the eggs laid are reduced in number but apparently develop normally. 
The spermatozoa of the male moths from starved caterpillars appear normal, 
and are capable of fertilizing normal eggs, but the resultant larvae show a 
higher mortality and the pupae appear dwarfed. Thus inanition is more injuri- 
ous to the male than to the female. 

As to the effects of inanition upon the sex of butterflies, Treat ('73) found 
a larger proportion of males resulting from underfeeding the larvae of Papilio 
asterias, Vanessa antiopa, and Dryocampa rubicunda. Poulton ('93) found no 
evidence to indicate that the sex in Smerinthus populi can be determined by 
external conditions. "It may be admitted that the larger female larvae 
require more food, chiefly to prepare for the amount of material to be stored up 
in the ova. It would not therefore be at all surprising if the female larvae were 
starved before the males when a minimum food was supplied." Cuenot ('99) 
cited the negative results of Riley, Bessels, Briggs, Andrews and Fletcher in 
starvation experiments on butterfly larvae. Kellogg and Bell ('04) likewise 
obtained no evidence that underfeeding the silkworm larvae produces an excess 
of males, although they ('03) found fertility greatly reduced. Kopec ('24a) 
obtained similar results. 

In the underfed silkworm larvae, Kellogg and Bell ('03) also found great 
individual variation in the resistance to inanition. The time required for 
metamorphosis is abnormally prolonged, usually with five moults instead of 
four, and the silk-production is greatly reduced. Pictet ('05, '05a) also found 
in malnourished butterflies a prolongation of the larval period, with shortening of 
the pupa period. There is also an imperfect pigmentation of the imago, which 


increases with each generation and may persist for a time even after abundant 

Coleoptera. — Great variations have been observed in the time required for 
death from starvation in various beetles. Sorg (1805) noted six days for 
Chrysomela populi, 13 days for Dermestes lardarius and Cerambyx fuliginator, 
and 36 days for Lampyris noctiluca. Fingerhuth (cited by Lucas, 1826) found 
two days in Coccionella 14-guttata, three days in Calandra granaria, five days 
in Curculio scrophularia, six days in Melolontha horticola, nine days in Carabus 
auratus and Cicindela campestris , 14 days in Geotrnpes Latr. (Scarabaeus 
stercorarius) , 27 days in Lucanus cervus, and one month in Cetonia aurata. 

Wodsedalek ('17, '21) observed a remarkable endurance in the larvae of 
Trogoderma tar sale kept without food at constant room temperature. Newly- 
hatched specimens, about one millimeter in length, lived four months; older and 
larger larvae lived progressively longer, the maximum for full-grown (8 mm. in 
length) being over six years. The larvae gradually decreased in size during 
inanition. They usually dwindled to the hatching size of one millimeter before 
death, representing for the larger larvae a decrease to 3^oo 0I the original mass. 
In the latter paper ('21) it is stated: "In another experiment groups of speci- 
mens varying in size from 2-8 mm. in length are undergoing periods of feasting 
and fasting. The larvae in various stages of starvation when given plenty of 
food again begin to grow in size. For example, some of the large specimens are 
on their way to their fourth "childhood" after having attained the maximum 
larval size four times; while specimens originally 4 mm. in length are on their 
way to their ninth "childhood" after reaching 4 mm. eight times." 

According to Chapman ('20), "Tribolium conjusum has its egg stage short- 
ened from ten to five days by a rise from 24 to 34 , and it will develop one 
generation after the other throughout the year. On the other hand, the life 
cycle may be prolonged by a reduction of the amount of moisture and also 
by a limitation of the quantity or quality of the food. Thus the length of 
life and the number of broods may be altered by changing any one or all of 
these three factors. A larva now under observation has had its life prolonged 
from thirty to ninety days due to food conditions, and during this time it has 
moulted twelve times rather than the normal six times." 

Biedermann ('98) found that certain intranuclear particles ("Kernkrys- 
talloide" of Frenzel) in the midgut epithelial cells of the meal-worm, Tenebrio 
molitor, become smaller and finally disappear during starvation. Similar 
cytoplasmic granules, crystalline or irregular in form, decrease but never 
disappear, even after prolonged starvation. Kriznecky ('14) and Szwajsowna 
('16) observed that in this species starved larvae undergo metamorphosis 

Slowtzoff ('03) noted in May-beetles on absolute inanition a loss of about 
24 per cent in body weight. The daily loss is greater at first (2.39 per cent), 
then sinks to a minimum (0.66), with a premortal rise. Geotrupes stercoralis 
loses 21.73 P er cen t in body weight during starvation of five to eleven days. 
The studies of Fatta and Mundula ('08) and Manca and Fatta ('o3-'o4, '05) 
were inaccessible. 


Hymenoptera.— Lucas (1826) cites Trembley ("Biologie") as authority 
for the statement that "Die Insecten-geschichte gibt uns Beispiele von Bienen, 
Ameisen, verschiedenen Raupenarten von Wurmern, Schmetterlingen und 
Fliegen, welche ganze Monate ohne die geringste Nahrung hinbringen." This 
probably refers to the period of dormancy or hibernation, however, as perhaps 
likewise the period of one year noted by Reaumur for the wasp, Vespa vulgaris. 
Lucas himself found the following periods required for death from starvation: 
ant, Formica fusca, two days (in May); bee, Apis terrestris, three days; Apis 
mellifica, four to six days or more (in June) ; Vespa vulgaris, eight days. Slowt- 
zoff ('04a) observed that the bumble-bee, Bombus terrestris, dies after only 24-48 
hours of absolute inanition, with loss of about 24 per cent in body weight. 
The loss is chiefly in water content, which is considered the cause of death. 

The effect of the nutrition upon sex-development in ) the Hymenoptera 
has attracted much attention. Von Siebold observed in the wasp, Nematus 
ventricosus, a progressive increase in the percentage of females during the 
spring and summer, probably due to the effects of increased warmth and 
food. In the honey-bee {Apis mellifica), however, it is well known that sex is 
determined by fertilization, the unfertilized eggs producing the males (drones). 
If larvae from the fertilized eggs are well-fed with the "royal diet," they become 
queens, with functional reproductive tract; but if fed with the less rich, ordinary 
diet, they become workers, with rudimentary reproductive tract. According 
to Von Planta the "royal diet" is relatively twice as rich in fats as 
the ordinary diet, though slightly poorer in glucose and protein. The drones are 
males resulting from unfertilized eggs and are not due to difference in the diet. 

Eimer ('88) concluded that " Geschlechter giebt es in der Hummelfamilie 
nur zwei: Mannchen und Weibchen, denn die kleinen und grossen Arbeiter 
sind nichts als in Folge von schlechteren Ernahrung, vor Allem der mangelnden 
Honigfutterung wahrend des Larvenlebens, geschlechtlich unvollkommen 
entwickelte Weibchen." Attempts to determine sex by nutrition of the honey- 
bee larvae have been unsuccessful (Dalla Torre, '10). 

Popovici-Baznosanu ('10) observed that among the bees Osmia rufa and 
Osmia cornuta a greater amount of food is deposited in the cells of the larvae 
producing females than in those producing males. Experimental removal 
of a portion of this food resulted in a marked reduction in the size of the adult 
bees, both male and female. 

Among the ants (Formicidae), according to Emery ('94), the females, like 
those of termites and bees, exhibit a remarkable "Nahrungspolymorphismus." 
In this case, the quality of the food apparently determines whether the larva 
shall become queen or worker. There are, moreover, two different kinds of 
workers, small and large, probably determined by the quantity of food supplied. 
In some species, there are intermediate stages between the large and small 
workers. O. Hertwig ('20) adopts Spencer's theory that these intermediate 
forms may depend upon the developmental stage at which the larvae are sub- 
jected to inanition. Apparently no experimental evidence upon this question 
is available. 



A few observations have been made upon inanition among the Tunicates, 
a sub-phylum of the Chordata, closely related to the Vertebrata. Herbst 
C97), in connection with his study of the effects of various salt deficiencies 
upon the development of Echinoderms and other organisms, noted that in 
Phallusia mammillata and Ciona intestinalis, "Die Furchung geht also bei 
Ascidien auch in phosphorfreiem Medium vor sich, fur die spatere Organbildung 
und deren Ablauf ist aber die Phosphorzufuhr von aussen unentbehrlich." 
The Ascidian ova thus appear to have an unusually large initial supply of 
phosphorus. The tendency to loosening of the intercellular substance, resulting 
in the detachment of the cells as a result of calcium deficiency was also noted 
(among other forms) in the larvae of Ciona intestinalis by Herbst ('oo). . Schultz 
('07) found that the "reduction" phenomena during regeneration in Clavellina 
lepadiformis resemble those produced by inanition, since the regeneration process 
withdraws nutriment from the remainder of the organism. 



The second part of the present work concerns the effects of inanition upon 
vertebrates, including man. The general effects upon the body as a whole 
will be considered first, followed by chapters upon the various systems and 
organs of the vertebrate body. In general, each organ or part will be considered 
in relation to the effects of (A) total inanition, or on water alone; or (B) the 
various forms of partial inanition (deficiencies in protein, salts, vitamins, etc.). 
As previously noted, subsistence on water alone, although by definition a 
form of partial inanition, is for convenience considered with total inanition, 
on account of their similarity in effects. 



The effects of inanition upon the vertebrate body are of great interest and 
importance in relation to the corresponding conditions met in human famine and 
disease. The general effects will first be summarized briefly, followed by a 
more detailed consideration of the data. 

Summary of the Effects of Total Inanition on the Body as a Whole 

The duration of inanition among vertebrates (as in invertebrates) varies 
exceedingly, ranging from a few days in small birds and mammals to possibly 
years in some reptiles. In general, the period of endurance is greater if water 
is allowed; in larger and older than in smaller and younger animals; in carnivora 
than in herbivora; and in cold-blooded than in warm-blooded animals. An 
increased amount of stored food reserves, especially of fat, increases the endur- 
ance, while exercise and cold are unfavorable. In general, all factors are 
effective in proportion to their influence upon metabolism. Also the duration 
of human disease is often limited by the capacity to withstand the inanition 

The relative loss in body weight during inanition is subject to less variation, 
the maximum averaging about 40 per cent (range 30-50 per cent), as shown by 
Chossat ('43) and many later observers. The loss in general is less in younger 
and greater in fat animals, and is also somewhat variable according to the type 
of inanition, species and environment. The curve of loss in body weight is 
logarithmic in type, resembling an inverted growth curve. The head loses 



relatively less than the remainder of the body, and hence becomes relatively 
larger. These rules in general apply likewise to conditions of human 

The embryonic vertebrate is usually protected from inanition through 
storage of nutriment in the egg of oviparous forms and through sacrifice of 
the maternal organism in mammals. Starvation of the pregnant mother must 
be very severe in order to reduce the birth weight of the offspring. Attempts 
to reduce the size of the human fetus by restriction of the diet during pregnancy 
therefore appear generally impracticable. 

Even under relatively normal conditions, there is a large amount of malnutri- 
tion among children, and this is greatly increased by the conditions of war and 
famine. The best simple physical index of nutrition (with the possible exception 
of Bornhardt's index) is probably the weight: height 3 ratio, but this is normally 
variable, even in individuals of the same age and race. Such indices may serve 
a useful purpose in directing attention to suspicious cases, but require confirma- 
tion by clinical evidence. 

During "physiological inanition," certain growth changes may proceed 
in the body of young vertebrates, as illustrated by the salmon during migration, 
by amphibia during metamorphosis and by the human infant during the post- 
natal loss in weight. In various mammals (including man), despite under- 
feeding or malnutrition sufficient to prevent increase in weight, there is a 
persistent tendency to growth in certain parts (especially skeletal) at the expense 
of others, resulting in dystrophic growth with characteristic abnormal propor- 
tions, the body being elongated and the head enlarged. These dystrophic 
growth changes are contrary to Liebig's "law of the minimum," if strictly 

Recovery from inanition is generally possible, unless extreme stages have 
been reached. Recuperation under proper nutritional conditions is especially 
rapid in the young, but permanent stunting or dwarfing with failure to attain 
normal adult size may occur when the inanition has been severe or prolonged, 
and especially when occurring at a very early age. The conditions limiting 
the possibility of recovery from inanition and malnutrition are of practical 
importance in human medicine. 

Effects of Total Inanition, or on Water Alone 

The topics considered under this heading include the duration of inanition, 
the effects on body weight in adult and young, nutritional indices, dystrophic 
growth, changes in adult proportion (head, trunk and limbs), and recovery 
upon refeeding. The effects of partial inanition will be discussed in the next 

Duration of Inanition. — That different animals exhibit marked variations in 
their resistance to inanition has long been known and the earliest experiments 
were concerned chiefly with this feature. Many of the earlier observations 
(by Redi, Spallanzani, Haller, Blumenbach, and others) were compiled by 
Lucas (1826), whose data for invertebrates have been noted in the previous 


chapter. The periods of endurance for vertebrates (usually for inanition with 
water) are stated by him as follows: 

Fishes: Esox lucius, 2 months; Cyprinis carpio, 50 days. 

Amphibians: Rana (sp.?), 1 year. Bufo (sp.?) 4 months to "several years;" 
Salamandra (sp.?), 6 months to 1 year; Proteus (sp.?) over 2 years. 

Reptiles: Testudo terrestris, 18 months; Lacerta africana, 8 months; Lacerta 
crocodil., 8 months; Lacerta muralis, i}<2 months; Lacena lacustr., 2-3 months; 
Chamaeleon (sp.?), 1 year; Coluber vipera, 6—10 months; Chamaeleon cerastes, 
5 years; Vipera vulgaris, 10-12 months; Vipera caudisona, 5-6 months. 

Birds: Vultures, 11-21 days; Aquila regia, 21-28 days; Falco milvus, 
18 days; Cuculus canorous (young), 3 days; Fringilla canaria, 3 days; Fringilla 
caclebs, 3 days; Fringilla domestica (10 days old), 16 hours; same (14 days old), 
27 hours; same (adult) 3 days; Columba oenas 2-13 days; Phasianus capo 
(total inanition), 5-9 days; same (on water only), 20-24 days. 

Mammals: Antilope (sp.?) 20 days; Viverra Zibetha, 10 days; Canisfamiliaris 
(nursing puppy), 3 days; same (adult), 25-39 days, Canis lupus, 5 days; 
Felis catus, 25-39 days; Sciurus vulgaris, 2^-3 days; Scavia porcellus, 4-6 
days; Lepus cimiculus, 8^-9 days; Manis (sp.?), 2 days. 

For the human species, Lucas cites a long series of fasting cases of variable 
credibility, including the following: newborn child, 8 days; a boy, 5 days; 
girls, 11 days to 7 years (!); adult men, 5-71 days; adult women, 34 days to 
10 years (!). Some of the longer periods are merely impostures or incredible 
legends of religious fanatics; some, however, refer to cases of insanity or other 
neural disturbances (cataleptic and similar states somewhat comparable to 
hibernation) in which a lowered rate of metabolism may render possible a 
period of inanition far beyond that possible in normal individuals. Reviews 
of the literature on long fasts are given by Hammond ('79) and Rochas ('02). 

Since the publication by Lucas (1826), a large number of observations has 
accumulated in the literature concerning the endurance of various types of 
inanition by various species, some of which are mentioned later (Table 1). 
The weight of evidence indicates that for normal human adults the extreme 
limit of endurance, even under favorable circumstances and with water ad 
libitum, will rarely exceed 2 months. Several human fasts of 30 days or more 
will be mentioned later. The variations observed in fatal cases range from 
24 days (Voelkel '86) to 72 days (Lussana '68). For total inanition the time 
is much shorter, ranging from 1-2 weeks (Birch-Hirschfeld '92), rarely more. 
In congenital atresia of the esophagus, death from inanition occurs in 3 or 4 
days (Hirschsprung '61); and in complete duodenal atresia in 4-12 days 
(Theremin '77). Further data concerning periods of endurance of inanition 
in various species are cited especially by Bardier ('13), Beeli ('08), Chossat 
('43), Colin C73), Falck ('81), Fowler ('71), Rosenstern ('11), and Schaeffer 

Factors Involved. — Some of the various factors influencing the duration of 
inanition may be mentioned briefly. As to the type of inanition, total inanition 
is usually fatal in a much shorter time than partial inanition, with water. 
In birds, however, there is usually but little difference. Other forms of partial 


inanition, with deficiencies in proteins, salts, etc. may be endured for various 
(usually much longer) periods of time, to be mentioned later. As to age, the 
young are much less resistant than adults, as will be discussed later. Among 
animals, the carnivora endure inanition longer than the herbivora, the cold- 
blooded longer than the warm-blooded, and, in general large animals longer than 
small animals. Individuals with large amounts of stored food reserves, espe- 
cially of fat, can endure starvation for longer periods. Exercise, exposure to 
extremes of temperature, infections, etc. are unfavorable factors. 

In general, all factors are effective in proportion to their influence upon 
metabolism. An increased rate of metabolism will more rapidly exhaust the 
stored food reserves, or reduce them below the necessary minimum, with resultant 
death from starvation. On the other hand, a lowered rate of metabolism dur- 
ing inanition will prolong the period of endurance. This, together with the special 
provision of food reserves, makes possible the extensive periods of inanition 
during hibernation and some allied conditions of "physiological inanition" 
(cf. Alexandre '88, '89). 

Loss of Body Weight in Adults. — For adult vertebrates a large amount of 
data is available upon various species for the loss in body weight during both 
total and partial inanition. Chossat ('43) established 40 per cent as the usual 
average of maximum loss in adult vertebrates, but recognized some exceptions 
and variations, as are apparent in Table 1. The range is usually between 30 and 
50 per cent. On the whole, however, there is in the various classes of vertebrates 
comparatively little difference in the percentage of body weight which may be 
lost in extreme inanition. The most important factor is the amount of stored 
food reserves (especially fat). Many of the observed variations (for example, 
sex differences) are probably due to this factor; although Ott ('24) finds a lesser 
resistance in the female frog, in spite of the large ovarian mass available for 
resorption (Table 6). Larger species or individuals will usually endure a greater 
loss, while unfavorable environment (temperature, etc.) may produce death 
with a smaller loss of body weight. Kahan ('85) observed that in pigeons the 
daily average loss in weight increased with repeated intermittent fasts, but this 
was not confirmed by Seeland ('88) in rabbits or by Stewart ('16) in albino rats. 

As to the course of the loss in body weight of animals during inanition, 
Chossat ('43) concluded that in birds the loss is greatest in the first third of the 
inanition period, least in the second third and intermediate in the last third. 
Bourgeois ('70) comfirmed this for mammals. Moleschott ('59), however, found 
that the final acceleration of loss is variable or absent. Lazareff ('95) for 60 
guinea pigs on total inanition found a progressive decrease in the daily loss rate, 
the average percentage loss for the eight successive days being roughly 9, 7, 6, 
5, 4, 3, 3, 1.5. According to Rosenstern ('11) similar results were obtained by 
Finkler (pigeons), Kuckein (chicks), Richet (ducks), Rubner (rabbits), Petten- 
kofer, Luciani, and Laborde (dogs). For loss of body weight in various animals 
during hibernation, see Polimanti ('05, '13), Valentin ('57); Rulot ('01). 

The data of Falck ('75) for dogs on total inanition may be taken as typical, 
and the curves for two adults are shown in Fig. 31. Morgulis ('23) shows similar 
curves for fasting dogs, based on Avrovov's data. 



In the human adult, no accurate data are available as to loss of weight during 
total inanition. On water alone, a loss of about 40 per cent may occur before 
death, according to Bright ('77), Duflocq (cf. Fernet '01) and Meyer ('17). 
In about 400 autopsies on victims of the Madras famine of 1877-8, Porter ('89) 
estimated that about 86 per cent of the men and 83 per cent of the women had 
lost one-third or more in body weight, emaciation being the most outstanding 

The loss in body weight in many of Porter's cases was partially masked by 
dropsy or "hunger edema," a condition which has been noted frequently by 






1 : / : 


M > 



1 I 1 r^r 


/ ' 

/M ' 1 

1 1 1 


1 ' 

'. ' fonnn 

1 \ i 



L M 


'/ ' 

1 Ml 

. 1 

' ' , A \ I 


M. I 1 V 


i |i | 


ja \ 


1 1 1 1 

1 i : 

Mi M 




# ' | . 

\J 1 y 


/. j J_ 


' \ ' 




; i M 

am p 

r/ \ -!/ 

! 1 



i 1 


1 1 

1 M 


' I ! I\! / 

1 1 

1 19000 





+-t i 


1 1 1 i 1 

1 ' ' s 

I 1 

, i> 

/ X 



i i s 


1 / 1^ 

i 1, v-^ 


1 i 


■v ! 

X 1 



I I ! 



r-i+r +S 

1 1 | 1 

-+H4- -r 

1 / 


' ~'rC~/i 

; i 

1 ' 

11 1 c 


^ §7 

' ! 

"°° irionn 

MXf - 


1 I \% l0UUU 


j/*^ N. 


1 5 



/ X. 

1 | 





TV* 1 / 





■ j 

8 i^onf) 


. - N?fcU 

j M '(§ ouuu 


/ / 


iS?0 r 

' J ^ ' 




r^- " 


"Sr ' — / 


s 3§<- 

. J_l 

1 ^SO 1 ' 

i 1 

1 1 14UUU 






^t— - 


: j_ie: 

7/ / 


ii 1 . 

i 1 1 


/ "" 


1 1*5000 


v _j L 

I / / ' 

1 I 

i > i ! 

! i i 

1 1 / 

M ! 

< 1 





1 1 1 

1 •■ - 

< IrTMY) 



: \ ' 

Mil i i 



i 1 J ' ' 

h-T" - 



- ■ LJ-H4- 


d-ci To&l-I'n, 



"S 1 1(W) 

J 5 

10 15 

£0 £5 

50 3 

5 4 


15 50 55 

>&0 65 

Fig. 31. — Chart showing curves of body weight in dogs on (complete) total inanition 
The curves showing the percentage of loss of the initial weight demonstrate that the loss 
becomes progressively slower with age. The descending curve shows the loss in absolute 
body weight by the adult dog IV. (Falck '75.) 

other observers in cases of chronic malnutrition and especially during famine or 
allied conditions. Edema under such conditions has been noted by Abel ('23), 
Aron ('20), Bigland ('20), Beyermann ('19), Burger ('19, '20), Croftan ('17), 
Digby ('76), Enright ('20), Fliigge ('22), Gaspard (1821), Hecker ('44), Kraus 
('19), Landa ('17), Lange ('17), Leys ('14), Lubarsch ('21), Maase and Zondek 
('17, '20), Mann, Helm and Brown ('20), Matthias ('19), Maver ('20), McLeod 
('81), Menzies ('20), Oberndorfer ('18, '19), Paltauf ('17), Park ('18), Prince 
('21), Prinzing ('16), Rossle ('19), Schiff ('17), Schittenhelm and Schlecht ('18), 
Strauss ('15), Tallquist ('22), Vacker ('71), Vandervelde and Cantineau ('19), 
Wells ('18) and Nicolaeff ('23). While many of the authors ascribe "famine 
edema" and allied conditions to a general total (incomplete) inanition, others 

7 2 


believe it to be caused primarily by one or more forms of partial (especially pro- 
tein) inanition, as will be mentioned in the next chapter. 

Some data upon the loss in weight of the human body in various fasting 
periods (on water only) were cited by S. Weber ('02) (see table on p. 73). 

The course of the body weight on water alone has been carefully studied in 
voluntary fasts of 30 days or more by Luciani ('89, '90), Paton and Stockman 
('89), Penny ('09) and Benedict ('15). The 40 day fasts of Succi and Dr. 
Tanner and the 50 day fast of Merlatti were less rigidly controlled. The 
Tanner case is discussed by Taylor ('20). The Italian "hunger artist," 

lb 10 X 

Fig. 32. — Chart showing the individual curves of body weight in three men (Jacques, 
Succi and Levanzin) fasting i month, on water only. The curves are nearly parallel, 
although there are evident irregularities and individual variations. The fourth curve (con- 
tinuous line) represents the theoretical curve (according to Mayer's formula) for Levanzin, 
assuming that each daily loss represents the same percentage of the body weight at the begin- 
ning of the corresponding day. It is clear that the actual loss is relatively greater in the earlier 
stages, and relatively less in the latter part of the period. 

Succi, underwent four fasts of 30 days each at different times, the percentage 
losses in body weight for the successive fasts being 22.7, 21.7, 19.2, and 17.2 
(Schondorff '13). 

Individual variations are evident in the three curves of body weight shown 
in Fig. 32. That of Levanzin (Benedict '15), with loss of 21.6 per cent in 31 
days, is the most regular, that of Jacques (Paton and Stockman '89), with loss 
of 16.6 per cent in 30 days, the most irregular, and that of Succi (Luciani '89, 
'90), with loss of 19.2 per cent in 30 days, intermediate in character. Falck 
('81) stated that the decreasing body weights form a straight line; but it is clear 



that the rate of loss in body weight in general is greatest in the earlier days, 
gradually decreasing later. Luciani attempted to devise a formula to 
represent the course of the decrease in body weight, but without success. 

Subject (and observer) 

Length of 
fast, days 







per cent 




per cent 

Schwede (Johannson) 

Breithaupt (Senator) 

Cetti (Senator) 





62. 79 




1 1 . 16 


1 .01 

Succi (Luciani) 


As shown by the continuous line in Fig. 32, the formula by Mayer ('14) 
for starving medusae does not fit the present case, as the relative loss in 
the human body is too great in the earlier stages. The course of the body weight 
during inanition in man and mammals more nearly resembles a reversed growth 
curve of the logarithmic type, as found by Morgulis ('23) in curves based on 
Avrovov's data for the fasting dog. 

Kohlschutter ('87) presented some curves showing loss of body weight in 
adults with typhoid fever. Although he recognized that these curves are loga- 
rithmic in form, he rejected the idea that the loss might be due to starvation, 
since the only data available to him indicated a uniform loss (straight line) 
during inanition. He attributed the loss in weight during typhoid and tuber- 
culosis to the increased oxidation of body substance due to the fever. 

Incomplete Inanition. — The types of inanition heretofore considered have 
been complete, either total or on water alone. With incomplete diets the food 
given is inadequate in amount, the effect varying with the extent and character 
of the deficit. Thus Chossat ('43) found that birds on one-third normal rations 
lived twice as long as those without food, but the total body loss in weight was 
about the same. Petroff ('83) noted that rabbits on one-fourth of the normal 
ration (water ad lib.) lived only 36 days, or slightly longer than on water alone. 
On one-third ration the time was extended to 47 days. Ochotin ('86), however, 
found that a rabbit on one-fourth normal ration lived only 15 days with loss of 
48.75 per cent. 

Loss (or Retardation) in Body Weight during Inanition in the Young.— 
It has already been noted that age is an essential factor, the resistance to inani- 
tion in general increasing progressively from birth to maturity. Resistance in 
the young is lessened by the apparently smaller storage of reserve food materials 
(especially fat). The food requirement is also relatively greater in the young 
because metabolism is more intense, and the needs for growth as well as for 
maintenance must be supplied. Thus it is quite possible for a growing organism 
to starve to death on a diet sufficient to maintain the body without loss of weight. 
According to the degree of inanition, we may find an actual loss in body weight, 
maintenance, or merely a retardation in the normal growth rate. 



Some of the data indicating the relative effects of inanition upon body weight 
and length of life in young vertebrates may be mentioned briefly, for comparison 
with the effects in adults. The effects upon the prenatal as well as the postnatal 
stages must also be considered. 

Although the lesser resistance of the young was noted even by Hippocrates 
and Galen, the first accurate, quantitative observations were made by Chossat 
('43). He found that on total inanition young turtle doves (initial weight 
no g.) survived only 3 days with loss of 25 per cent in body weight; adolescent 
(143 g.) survived 6 days with loss of 36 per cent; while adults (189 g.) survived 
13 days with loss of 46 per cent. 

The more extensive experiments on chicks by Petroff ('86) are summarized in 
the accompanying table. The exceptional endurance of the very young chicks 
may be due to the presence of unabsorbed yolk material, and also to additional 
protection from loss of heat under the wings of the mother. 

Effect of Total Inanition upon Chicks at Various Ages (Petroff '86) 

Age of chicks 

(from hatching) 


No. of obser- 
vations at each 

Average time of 
survival hours 

Average loss of 

body weight, 

per cent 

Average loss in 

weight per hour, 

per cent 

































6 ' 
































Among mammals, Falck ('75) observed that 3 puppies placed on total 
inanition at 18 hours of age survived 2.8-3.4 days, with loss of 19.3-26.3 
per cent in body weight; 2 puppies at 14-16 days of age survived 13-15 days 
with loss of 46-48 per cent; young adult dog (1 year) survived 23 days with 
loss of 48 per cent; an old adult (fat) survived 60 days with loss of 49 per 
cent (see Figs. 31 and ^ ) t ) ). 

Von Bechterew ('95) found that a newborn puppy on total inanition lived 
6 days with loss of 37 per cent in body weight; one at 3 days lived 8 days with 
loss of 34 per cent; two at n days lived 15-17 days with loss of 38-41 per cent. 
Similarly a kitten aged 2 days lived 4 days with loss of 18 per cent; one at 4 
days lived 6 days with loss of 22 per cent; while one at 6 days lived 6 days with 
loss of 26 per cent. 

On water alone,- Dehon ('05) found that 7 kittens aged 3-17 weeks lived 
4-8 days with loss of 10-45 P er cent m body weight. Similarly, Howe, 



Mattill and Hawk ('09) noted that a puppy aged 1 month lived 6 days with 
loss of 22 per cent, while adult dogs survived 48-117 days with loss of 53-63 
per cent. 

In reptiles (serpents), Pellegrin ('01) observed that 10 young Tropidonotus 
natrix on total inanition survived an average of 36 days with loss of 38 per cent, 
while 10 on water alone lived 116 davs with loss of 43 per cent. (Adult 
Pelophilus survived 3-4 years.) 

In amphibia, Swingle ('18) starved yearling tadpoles of Rana catesbiana 
for 5 months with marked skrinkage of the body (weight undetermined). 

Fig. 33. — From a photograph of a female dog after 60 days of complete total inanition with 
loss of 49 per cent in body weight. (Falck '75.) 

On incomplete inanition (diet merely restricted in amount), Aron ('n) 
discovered that growing puppies can be held at maintenance (constant body 
weight) for several months; but ultimately the amount of food has to be increased 
so as to permit some increase in body weight, otherwise death from inanition 
results. Jackson ('15a) similarly observed that albino rats held at maintenance 
by underfeeding live only about 2 months unless a slight increase in body weight 
is permitted, and Stewart ('18, '19) found that newborn albino rats can be held 
at maintenance by underfeeding for only 2-3 weeks. 

Thus the period of survival in growing animals on maintenance diet varies 
directly with the age. A diet below the maintenance requirement will pro- 
duce death more rapidly, with actual loss of body weight, while a diet above the 

7 6 


maintenance requirement, but still subnormal in amount, will cause a retardation 
in growth proportional to the degree of deficiency. Thus Evans and Bishop 
('22) found that on optimum standard diet the albino rat at one year reached 
a body weight of about 330 g. ; slightly underfed, 220 g.; on two-thirds normal 
ration, 140 g.; and on half ration, 60-85 g. (see Fig. 34). As will be shown 
later, however, the dystrophic growth under such circumstances is not only 
decreased in rate but also often abnormal in character. 



Fig. 34. — Chart showing curves of average growth in female albino rats on various planes 
of nutrition. The upper curve (light line) represents the average normal growth of littermate 
controls on an abundance of "Standard Diet I." The 3 lower curves in heavier line represent 
averages for the groups in which this ration was reduced slightly (in the upper curve), about 
one-third (in the middle curve), and about one-half (in the lower curve), respectively. Circles 
mark the average times of occurrence of the first estrus, which did not occur at all in the most 
underfed group. (Evans and Bishop '22.) 

Prenatal and Larval Inanition. — Since the resistance to inanition in general 
varies directly with age, it may be inferred that during prenatal or embryonic 
stages, the organism is particularly susceptible to nutritional deficiencies. This 
is probably true, but it is somewhat difficult to prove. Especially in mammals 
the embryo and fetus are carefully protected against inanition by the food supply 
through the placenta from the mother, even when the latter is severely underfed. 
In oviparous forms the eggs are usually provided with an abundance of nutrient 
yolk material, which in part may be carried over into the embryonic body as a 
reserve supply and may for some time aid in the resistance to inanition. 

Among fishes, Fabre-Domergue and Bietrix observed that the young may 
perish from inanition before complete absorption of the yolk material. In 
amphibia, Brehm ('12) cites a remarkable case in which Nussbaum observed 
that an adult Proteus anguinus after 13 months of starvation gave birth to a 


defective young proteus 12.6 cm. long. This had possibly developed at the 
expense of other eggs in the oviduct. Swingle ('18) observed that larvae oiRana 
pipiens after emergence from the egg capsule may survive inanition for over 
100 days (doubtless due to the large amount of yolk material present). Podh- 
radsky ('23) noted shorter periods (up to 32 days) in young tadpoles of Rana 
fusca. Yung ('78, '83) found that in the later developmental stages malnutrition 
retards the growth in size and prevents metamorphosis of frog tadpoles until 
they become sufficiently large. In tadpoles of Rana fusca, however, Barf urth 
('86, '86a, '87) discovered that at the time of metamorphosis into frogs starva- 
tion actually accelerates this process, and thus hastens the development. 
Wolterstorff ('96) concluded that the result varies according to the stage of the 
tadpoles used. Although Barfurth's results were denied by Bataillon ('91), 
Bohn ('04, '04a) noted that while in the earlier larval stages of Rana temporaria 
the removal of the albuminous capsule (which serves as nutriment) retards 
growth, it causes metamorphosis into tadpoles at a subnormal size. Kopec 
('22, '22a) found that starvation of tadpoles before they are 50 days old retards 
their development, but after 65 days of age inanition accelerates their 

Even normally, according to Barfurth, the frog tadpole eats but little during 
metamorphosis, so fasting merely accelerates the normal process. This was 
confirmed by Duesberg ('06), who described the histological process of absorp- 
tion in the tail during metamorphosis (c/. also Morse '18 and Morgulis '23). 
Pfliiger states that larvae of Alytes obstetricans cease to eat when 8.1 cm. long, 
and live 5 weeks during which time the extremities are formed at the expense 
of the tail. According to Chauvin('76) and Kaufman ('18) a similar "physiolog- 
ical inanition" occurs during the metamorphosis of the Axolotl; and, according 
to Powers ('03), the metamorphosis of Amblystoma tigrinum is accelerated by 
starvation. The "physiological inanition" during amphibian metamorphosis 
recalls the similar conditions found in insects; and also the well-known fasting 
period of the migrating salmon, during which the sexual products are matured at 
the expense of the musculature (Valenciennes '48; Siebold '63; Miescher '97; 
Stone '97; Paton '98; Greene '10, '12; Heitz '18). A similar phenomenon 
apparently occurs in the male fur-seal (Parker '17) and the gander (Stieve '22), 
which fast during the reproductive season. 

In mammals, the results on this point have been conflicting. Ver Eecke 
(01) and Jagerroos ('02) found that the offspring of underfed pregnant animals 
(rabbits and dogs) develop normally, at the expense of the maternal organism. 
Similar conclusions were reached for the calf fetus by Tapke ('10) and by 
Eckles ('16), the latter stating that "All the data indicate that the weight of a 
calf at birth is not ordinarily influenced by the ration received by the mother 
during gestation." Zuntz ('19) found that the rat fetus is but slightly modi- 
fied in weight, even when the fasting mother loses greatly in weight. 

On the other hand, Roloff ('66) stated that underfeeding pregnant animals 
usually produces fetal maldevelopment. Rudolski ('93) starved pregnant 
rabbits and a dog, obtaining apparent reduction in the size of the offspring. 
Similarly positive results were obtained by Paton ('03) for guinea pigs, and by 



Reeb ('05) for rabbits and dogs. King ('15, '21) observed subnormal average 
weights of young albino rats born (sometimes stillborn) from mothers in ill- 
health, and ascribed the result to prenatal malnutrition. 

The extensive experiments of Barry ('20, '21) indicate that underfeeding of 
the pregnant albino rat must be severe in order to obtain positive results. 
When starvation is instituted shortly after copulation, pregnancy is interrupted 
by resorption of the ovum. If inanition is instituted in the second half of preg- 
nancy, death and degeneration of the fetus occur in a few cases. No abortions 
were observed (contrary to Diatschenko ('97, '99) and Reeb ('05) in rabbits) 
the pregnancies proceeding to full term, as was likewise observed by Paton 
('03) and Rudolski ('05). Forty-one of 120 offspring were apparently stillborn, 

original body wciuHt 
~~ \ ~~~~£- 

TE5TI5\ 2~ 






lil h N H H U 

nitiL ulii L U 1.1 U U l\ IwliLili I 

I H M M M h'l M M M M M M M M W 

HVP0PH\3Ij> — 



Fig. 35. — Diagram showing the relative degree of atrophy (change in weight) of the body 
and of various organs in pigeons during (i) total inanition and (2) partial inanition (milled 
rice diet, deficient in various respects). The loss in body weight is about 40 per cent in each 
case. There is a marked increase in weight of the suprarenals; the brain and hypophysis show 
little or no change; while the other organs show variable losses in weight. (From McCarrison 
'21; "Studies in deficiency disease," Oxford Med. Publ.; body weight corrected.) 

however, and the average weight of the newborn rats from these severely starved 
mothers was only about 3 g. or 40 per cent below the normal birth weight of 
5 g. (see Table 4). 

Human Prenatal Inanition. — In the human species, the possibility of reduc- 
ing the fetus by underfeeding the pregnant mother has received much attention 
on account of its clinical importance, especially in cases of maternal contracted 
pelvis. The conflicting views found in the literature during the past century 
have been summarized by Hoffmann ('92), Florschutz ('95), Reijenge ('96), 
Schaeffer ('02) and Reeb ('05). A low maternal diet during pregnancy in order 
to reduce the size of the fetus was proposed by Brunninghausen (1804) and was 
advocated by Ackermann (1804), Reisinger (1820), Bandelocque (1820), 


Depaul ('49) and others. The plan was opposed, however, by Jorg (1806), v. 
Siebold (1806), Osiander (1820) and others (cf. Cyr '69), as ineffective or as 
dangerous to the mother. 

Prochownick ('89, '01) modified the plan by restricting, not the quantity, 
but the quality, of the diet, making it rich in protein, but poor in carbohydrates 
and water. This modified diet has been approved by Fraenkel ('96) and other 
clinicians, although Bondi ('13) stated that it is unsupported by either clinical 
experience or animal experiments. Prochownick ('17) still maintained his 
position, although admitting that the restricted war diet produced no evident 
decrease in average birth weight. 

The evidence afforded by the birth weights in Europe during the war period 
is of much interest. In Germany the general shortage of food resulted in an 
average loss in adult body weight of about 15 per cent, as estimated by Rubner 
('19). Kettner ('16) of Charlottenburg claimed the appearance of an atrophic 
type of newborn, but later ('16a) admitted that it could hardly be due to mater- 
nal underfeeding. A statistical comparison with the prewar records by Tschirch 
('16) in Jena, Mossmer ('16) in Posen, Momm ('16) in Freiburg, Ruge ('16), 
Bendix ('16), Rabnow ('16), Briining ('18) and Linde ('19) in Berlin, Hofmann 
('19) in Rostock, Soergel ('18) in Halle, Loenne ('18) in Bonn, Schmidt ('18) 
in Tubingen, Jahreiss ('19) in Augsburg, and Linke ('21) in Heidelberg (?) 
reveal no significant change in the German birth weight during the war period. 
Beninde ('19), however claimed that by 191 8 the conditions were worse, result- 
ing in a decreased birth weight in some regions. David ('22) found an average 
apparent decrease of 3-3.75 per cent in weight of the newborn, and of 2.5-3.2 per 
cent in length, but did not establish a relation to the skrinkage in diet during 
the war period. 

In Austria (Vienna), Peller ('17), Richter ('17) and Schauta ('17) found no 
appreciable reduction in average birth weight. Similarly, Murray ('24) found 
no significant decrease of the average birth weight in London during the war 
period. In Belgium, Demoor and Slosse (/20) claim that the prewar birth 
weight of about 3,000 g. was reduced to an average of 2,500 g. during the 
war famine. It seems probable that in most of the countries involved in 
the war, the food shortage, although markedly affecting the general population, 
was not severe enough to reduce the weight of the newborn, excepting districts 
of actual famine. Ivanovsky ('23) states that in Russia, although the number of 
births decreased enormously, there was a marked increase in the premature or 
stillborn, also in the occurrence of monsters and various anomalies. Troizky 
found no decrease of birth weight in 22,000 Russian cases (up to 1917). 

Effect on Sex Ratio. — In the first three chapters, the evidence was reviewed 
as to the effect of nutrition upon sex determination among plants and inverte- 
brates. It appears that in many species the effect of malnutrition is to produce 
a preponderance of males. While in some cases this may be due merely to a 
selective mortality, in others it appears probable that nutrition may, under 
certain conditions, affect the sex from the beginning. 

Among vertebrates the evidence as to the influence of nutrition upon sex 
determination is more conflicting and less conclusive. In the human species, 
there is a widespread tradition that relatively more males are born following 


periods of war or famine (Ploss; Ruge '16), but convincing statistical evidence is 
lacking. Richter ('17), Loenne ('18) and Linke ('19) found no evident change 
in sex ratio during the world war. 

Among domestic animals, Girou de Buzaieingues (1828) found that the per- 
centage of male lambs from 150 abundantly nourished ewes was about 40; 
while in 150 poorly fed ewes the percentage of male lambs was increased to 60. 

Girou's conclusion was supported by the experiments of Duesing ('85) and 
by Wilckens ('86), the latter presenting statistical data from about 30,000 
domestic mammals (including 16,000 colts, 4,900 calves, 6,750 lambs and 2,300 
pigs). Wilckens concluded that good nutrition in utero tends to produce 
relatively more females; malnutrition, more males. Slonaker and Card ('23), 
however, found an increased ratio of females in albino rats on a restricted 
vegetable diet. 

Among amphibians, the earlier experiments of Born ('81, '94) upon Rana 
fusca indicated that relatively more females are obtained from larvae placed on a 
nourishing diet; and Yung ('83, '85) in Rana esculenta apparently produced 
more females on richly protein diets. Cuenot ('99), however, obtained negative 
results with Rana temporaria. King ('07), in careful and extensive experiments 
upon larvae of the toad (Bufo lentiginosus) , likewise failed to find any significant 
differences in sex ratio on various diets (meat, wheat, mixed, and egg-yolk) ; 
but her underfeeding experiments failed on account of the high mortality. 
The problem of sex determination in amphibians is greatly complicated by 
the difficulty in distinguishing the sexes during the early larval stages, and the 
question is still unsettled. 

In reviewing the evidence, Geddes and Thomson ('01) concluded that in 
vertebrates, as well as invertebrates, sex may be determined by various external 
factors, especially nutrition. Mitchell ('n), on the other hand, concluded 
that "In nearly every case, however, other observers have obtained conflicting 
results, or placed another interpretation on similar results, whilst in none of 
the cases has the factor of selective mortality been sufficiently excluded." 
In the recent biological literature, the views concerning sex determination 
have been dominated largely by the "accessory chromosome" theory, according 
to which the sex is supposed to be determined by the relative amount of nuclear 
chromatin present in the gametes at the time of fertilization. It may be pointed 
out, however, that even under these conditions the sex ratio might still be 
affected by previous malnutrition of the gonads in the parent organism during 
the process of oogenesis and spermatogenesis. Further investigation is there- 
fore necessary before this important question can be settled. 

Postnatal Loss in Birth Weight. — The normal postnatal decrease in human 
body weight was discovered by Chaussier about a century ago, according to 
Benestad ('14). A similar decrease in newborn puppies, kittens and rabbits 
was denied by Kehrer ('70) but was found in the guinea pig by Minot ('91) 
and by Bessesen and Carlson ('23). According to Ostwald ('08), such a loss 
is very general in animals. Although chiefly due to other factors (loss of 
meconium, urine, etc.), it is generally admitted that a part of this postnatal 
decrease in body weight is due to inanition, on account either of insufficient 


food intake or of inability of the digestive system to adapt itself suddenly to 
the new conditions. 

The postnatal decrease in the human infant usually reaches a maximum 
of about 200 g. on the third day (cf. data cited by Vierordt '06), and the birth 
weight is normally recovered by the tenth day (Winckel '62). Further details 
are given by Kehrer (70), Monti ('89), Schaeffer ('96), Gundobin ('12), Robert- 
son ('14, '15, '23), Benestad ('14), Bailey and Murlin ('15), and Ramsey and 
Alley ('18). Schick ('15) was able to prevent the initial loss in 12 cases by 
abundant feeding of breast milk. On the other hand, in cases of deficient 
breast milk, artificial feeding or digestive inability (especially in prematures), 
the loss may be greater and continue longer, with corresponding retardation 
in recovery. Thus Briining ('18), Hofmann ('19) and Kutting ('21) found 
that although the German "war-babies" were normal in birth weight, they 
failed to thrive normally in the first ten days, probably on account of deficient 
quality in the maternal milk. 

Inanition during Infancy and Childhood. — We come now to the considera- 
tion of the effects of inanition upon the body as a whole during infancy and 
childhood. A case of actual starvation in twins one month old was described 
by Jones ('89). Infantile malnutrition is very common and varied in character 
and in degree of severity. When pronounced, it leads to a marked condition 
of infantile atrophy, which, according to Albarel ('05), was first described by 
Soriano, a Spanish physician of the 16th century. It was described under the 
name "athrepsie" by Parrot ('77). Other terms which have been applied 
to the condition are "inanition," "cachexia," "marasmus," "hypotrophie," 
"pedatrophy," "decomposition," "denutrition," etc. These terms have been 
used by various authors with different meanings, which are discussed in the 
works of Thiercelin ('04), Rosenstern ('n), Vigor ('11), Lesage ('n), Czerny 
('12), Nobecourt ('16), Raimondi ('17), Marfan ('20), Talbot ('21), Variot 
('21), Utheim ('22), and others. 

For present purposes, it suffices to note that infantile atrophy is usually 
considered, not a definite clinical entity, but a condition of malnutrition which 
may result from many different causes. Anything which interferes with the 
normal nutrition of the tissue cells is a cause of inanition. As mentioned in 
the introduction, this may result from a food-intake deficient in quantity or 
quality; from varied lesions of the alimentary tract preventing proper ingestion, 
digestion or absorption of the food; or from imperfections in the blood or 
vascular transporting system. Even when all of these extrinsic factors are 
absent, however, and adequate nutriment is brought to the cells, they may still 
be unable to absorb and assimilate it, due to intrinsic defects in their protoplasm, 
as emphasized by Czerny (08). Such intrinsic defects may be congenital or 
even hereditary in origin (Variot and Guyarder '04, Lesage 'n) or they may be 
caused by the action of toxins produced in the system (cf. Stransky '22). 
Intestinal infections may cause inanition both as an extrinsic factor, through 
interference with digestion and absorption, and as an intrinsic factor, through 
toxic action directly upon the cells of the body. 


Utheim ('22) has recently reviewed in detail the question as to the etiology 
of infantile atrophy, as studied in Marriott's clinic, and reaches the following 

" Based on the material in this clinic, it is believed, then, that in the etiology 
of athrepsia, feeding is the main factor, a quantitative and especially a qualita- 
tive starvation being responsible for the development of most cases, that the 
constitutional factor is of less importance and that the parenteral infections 
will often contribute in developing the picture." 

In agreement with the observations upon the young of lower animals, 
the human infant perishes quickly from total inanition, with a relatively slight 
loss in body weight. Chauvin ('40) reported an incredible case of a premature 
(7 months) infant which remained quiescent and at constant body weight 
without food or drink for 7 weeks after birth, with ultimate recovery. Cantala- 
massa ('92) observed a case of premature (7 months) twins, one of which died 
without nursing in 11 days with loss of 24 per cent in weight; the other nursed 
slightly and died in 23 days with loss of 22 per cent. Bouchaud ('64) noted 34 
Paris cases, mostly prematures with incomplete inanition, all of which lost 
over 25 per cent, and 18 over 30 per cent in weight. The extreme case was 
one of a premature (6 months) of 1,400 g. which lost 570 g. (40.7 per cent) 
in 17 days. 

Thiercelin ('04) states that among athreptics, "il n'est pas rare de voir 
des enfants de trois semaines qui ne pesent que la moitie de leur poids de nais- 
sance," but apparently no other author has noted such an extreme loss in weight. 
According to Richter, a child may die of starvation in 5 days with a loss of 
less than 25 per cent in body weight. 

Of 38 Breslau infants dying with loss of over 25 per cent in body weight, 
Quest ('05) found the maximum loss at death to be 38.0 per cent, and 34.8 
per cent in those which recovered. His results were confirmed by Rosenstern 
('11), who observed only 3 nurslings surviving a loss of over 32 per cent (maxi- 
mum 35 per cent). Four fatal cases reached 35 to 38 per cent. Czerny ('n) 
states that nurslings during chronic inanition may survive a loss of one- third 
in weight, but the danger point is reached before this during acute inanition. 

In 12 cases of deaths from (chiefly chronic) inanition in infants 5-255 
days of age, final body weights 1,695-3,972 g., and body lengths from 46-64 
cm., the losses in body weight were estimated by Jackson ('21) in different 
ways as follows: loss in body weight, the final weight being compared with the 
maximum observed during life, average loss 19.2 per cent (range 13-7-25.5); 
loss in body weight, the final being compared with the normal for final body 
length, average loss 28.5 per cent (range 7.4-52.3); retardation in body weight, 
the final weight compared with the normal for corresponding age, average 
deficit 56.8 per cent (range 41. 6-7 1.6) (see also Table 3). 

In cases of incomplete inanition (underfeeding) or of partial (qualitative) 
inanition, the weight of infants may be stationary or merely retarded to a 
variable degree, depending upon the length and severity of the malnutrition. 
Camerer noted that artificially fed infants lag considerably behind the normally 



As to the effects of inanition during childhood, a large amount of data has 
accumulated from observations upon school-children. One of the first estab- 
lished facts is that the children of the poorer classes average in height and weight 
below those of the well-to-do of the same age (Pagliani '79; Landsberger '87; 
Geissler and Uhlitzsch '88; Geissler '92; Boas '97; Pfaundler '16; and others). 
This is ascribed to underfeeding and malnutrition, along with other unfavorable 
hygienic conditions. Even among the wealthier classes, many children are 
malnourished, due to improper feeding (including various types of partial 
inanition) rather than to underfeeding. 

Fig. 36. — Field graph showing the body lengths (circles) and weights (dots) for atrophic 
infants of the first year, plotted according to age. The larger circles and dots represent original 
Minnesota data; the others are from various sources. The curves for normal body weight and 
length are from data compiled by Prof. R. E. Scammon. Note that in the malnourished cases 
the weight is subnormal to a much greater degree than the height. 

Medwedjew ('82) observed the growth in length of 50 individuals during 
the great Russian famine, but his original publication was inaccessible to me. 
Nicolaeff ('23) found the body weight 20-40 (sometimes 50) per cent sub- 
normal for age among children 1-16 years old at Kharkow during the recent 
Russian famine. The conditions in Russia are described also by Morgulis ('23). 
Stefko ('23a) found that the girls, having more body fat, showed greater loss 
in weight, but lower mortality. 

Even in countries like the United States, where extreme poverty is relatively 
infrequent, numerous investigations have revealed a surprisingly large number 
of apparently malnourished school-children, at least in the large cities. Thus 
Sill ('09) found 40 per cent of 1,000 primary school-children, in the Jewish 
quarters of East Side New York, malnourished and subnormal in weight. 
He cites evidence of a similar prevalence of malnutrition in London and Edin- 


burgh. A somewhat smaller, but still alarming, amount of malnutrition among 
New York school-children was found by Chapin, Baker, Mitchell, and others. 
In a recent investigation by the Bureau of Child Hygiene (Baker '18a, '18b), 
171,691 school-children in the Borough of Manhattan were classified according 
to the Dumferline scale as follows: No. 1 (normal) 17.3 per cent; No. 2 (passable 
61. 1 per cent; No. 3 (distinctly undernourished) 18.5 per cent; No. 4 (pro- 
nounced malnutrition or marasmus) 3.1 per cent. As to age, of the children 
up to 6 years, 22.5 per cent were undernourished. The percentage increased 
to a maximum of 25.2 per cent at 9 years; then decreased to 12.1 per cent at 
16 years. Similar conditions probably exist in one million school-children in 
New York City. Nationality was not found to be an important factor, although 
the percentage of undernourished varied from 19.8 in children of Russian and 
Polish origin to 28.7 per cent among those of Italian parentage. Among 
894 New York school-children, Mitchell ('19) found 69 "special" and "open 
air" cases; of the remaining 800 children 16.8 per cent were 7 per cent or more 
subnormal in weight according to height, in comparison with norms based 
upon data of Holt, Burk and Boas. As an example of conditions in a smaller 
city of the Middle West, Brown ('20) found that 41 per cent of the children 
in the Lowell School of Kansas City were more than 10 per cent underweight 
(compared with the standard table of Dr. Wood). 

From a large series (172,000) measurements of American children in 1918-19, 
Woodbury ('21) concluded that 4.4 per cent were notably deficient in stature, 
and 15.7 per cent were deficient in body weight. The largest proportion of 
deficient cases occurred in infants below one year of age, the deficient group 
at this age averaging almost 25 per cent below normal in weight. These data, 
from material secured in the "Children's Year" measurements, indicate a lesser 
extent of malnutrition than those above mentioned. 

It seems probable that the percentage of undernourished children was 
increased by the scarcity and high price of food during the recent war period, 
(Chapin, et al. '18) even in countries not directly involved; although this was 
not found evident in Amsterdam school-children by Lubsen ('17). In Belgium 
Demoor and Slosse ('20) state that children of 6-14 years at the end of the war 
were retarded at least one year in height and weight. Calmette ('19) found 
even greater apparent retardation in Lille, amounting to 4 or 5 years in the 
older children. Newman estimates that fully 10 per cent of the English school- 
children are seriously malnourished. 

In Vienna, evidence of marked undernourishment of the children is given by 
Gribbon and Paton ('21) and Gribbon and Ferguson ('21). Among children 
from 1-14 years of age, the worst effects were found in those from 2-3 
years of age, which averaged 13.6 per cent below normal in height for that age, 
and 26.5 per cent subnormal in weight. 

In Germany, numerous observers have described severe malnutrition and 
retarded growth in childern as a result of the war conditions (Fuhge '18; Addams 
and Hamilton '19). Beninde ('19) found conditions progressively worse since 
191 7. Schlesinger ('20) presents extensive data and concludes that malnutri- 
tional effects of the war appeared later in children than in adults, consisting in 
(1) retardation (average of 2 cm.) in growth in length; ^2) loss in weight; (3) 


constitutional disturbances. In Munich, even before the war, Oppenheimer and 
Landauer ('n) found much malnutrition; and at the close of the war, Pfaundler 
('19) concluded that the Munich children of 6 years had been retarded by 3 cm. 
in height and 1 kg. or more in weight. In Berlin, Czerny ('21) states that no bad 
results among children were visible in the earlier years of the war, but retarded 
growth became apparent later. Goldstein ('22) among the institutional children 
of Berlin found 89 per cent subnormal according to age and 50 per cent were 
10—20 per cent subnormal in weight for height. They were also retarded 1—3 
years in height. Fuhge ('18) made a study of the metabolism in such retarded 
children, finding the growth retarded by insufficient food, with relatively small 
proportion of protein and fats. 

Indices of Nutrition. — In connection with the study of malnutrition, espe- 
cially in childhood, the desirability of some exact method of determining the degree 
of undernourishment has been increasingly evident in recent years. A favorite 
method has been to compare the height and weight with those given in standard 
tables as normal for the corresponding age, but this is unsatisfactory on account 
of the great variability even among healthy children of the same age. Since 
emaciation of the body is the most outstanding symptom of inanition, a com- 
parison of the ratio between length and weight (independent of age) should 
be useful. In order to obtain a normal index which is approximately constant 
for age, however, it is obvious that the simple height: weight ratio (as used by 
Dreyfuss '06 and earlier authors) is inadequate, since the weight (or volume) 
of the body in general increases as the cube of the height, assuming the body pro- 
portions to remain constant. But even the ratio height 3 : weight does not 
remain constant, since Quetelet (Anthropometric, 1835) demonstrated that 
between birth and maturity the human body normally becomes relatively 
elongated, so that a constant ratio is expressed more nearly by height 5 : 
weight 2 . Not even this ratio remains constant, however, and no completely 
satisfactory formula or index has yet been discovered. 

Various indices or modifications have been proposed, which are discussed in 
the papers of Fleischner ('06), Rohrer ('08, '21), Pirquet ('16), Matusiewiez 
('14), Pfaundler (/16), Manny ('16, '18), Holt ('18), Davenport ('20), Dreyer 
and Hanson ('20), Retan ('20), Emerson and Manny ('20) /Gerber (' 21), Wagner 
('21), Huth ('21), Carter ('21), Bardeen ('20, '21, '23), Guttman ('22), Clark 
('22), Van derLoo ('22) and Davenport ('23). 

It is impossible to discuss the various proposed indices in detail, but a few 
more important points may be mentioned. The original "Index ponderalis" 

of Livi ('86) was in the form: 100 v ^eig ) gramS ' ) - This was modified bv 

length (cm.) 

weight _. ., . . 

Rohrer ('08) as the " Korperfiillenindex " = 100 -7^. Pirquet ('16) prefers 

the sitting height instead of the total body length in his index "Pelidisi" = 

\/io. weight (grams) . p irquet > s index has been used by Helmreich and 
sitting height (cm.) 
Kassowitz ('23) and others. 



Guttman ('22), using Bornhardt's ('86) index, and Van der Loo ('22) claim 
that by using the chest measurement as one factor in the formula, a better 
correlation with the clinical findings is obtained. The thoracic circumference is 
utilized also in the index of Pignet (cited by Ivanovsky '23), and by Dreyer and 
Hanson ('20); while Davenport ('23) uses the ratio weight: chest-girth 2 as the 
"index of build." 

Bardeen ('20) uses a weight-height formula with the English units of inches 
and pounds. For the sake of uniformity, however, the metric system is pref- 
erable, and in the present work I have used Rohrer's weight-height index = 
body weight (grams) 

height (cm.) 3 

Assuming the density as one, this expresses what 

3.61, lj KM. 4 

42 1 iii 1 — 1 , i ; 1 ■ 1 1 — i-p — n— — 1 ; ■ ; . ' — rm — 1 — 1 — n 1 


i j | • 

±tt tu-t~4 ! ^4-r- ! L ~t~ _L 1 -+44-- -- -1- -1-4- 

t*~ H ft ^T i n r I ' f- + h 1 ^ -4-P1 -m- -t p 4 

T> r 

IT T ' ' '- ! I ' j ! ' ■ 

JJ ' — 1- -H- 


1 ", . .lI j 

+ 111 .11, 1 : ; j : ... i_ -4-L.4- 

_t J- ; | 1 1 

■ * 111 


1 1 • II II 1 ' : 1 

~£T ~ f ~ 

i I 1 1 1 

^er— -*-* 

[ »[« 1 

?0 E " 

<± • !• 1 1 1 1 j ' 1 

C.J v ' 

• 1 ■ » • III I ' ' ' 


p" tl " ! ! M ' M ' 

• III J_ ... ..... . 1 M .' 1 i j 


• •• . a 

->r > . 1 L 

-ftp v- 1 i 1 1 -,- 1 71 n ■ ! 

2.6 • 

' » Mi 'u»i a 

. ' 

1 •»• "^ 1 ' . ■ 1 1 1 1 II 1 1 


* • ?L« -X 

• PI • V • 

"• I .' « •••: :♦ : ♦ ?\ " TT ^T ' f M 

c.3 •; 

.• • • ; •••! . 1 rsj it 


•• la ? \ I I 1 1 I 1 

* a • r \ 

• •• J t - T !•! p* 1 i X ' 11 1 


i* 1 ***a* ••» . 1 \ Ar • 

'"■ "al •?' " ' ^ ! 1 M 

"a* a •»•*>.. • 1 I ,,] XgU, - 

► ■ •«• 

• • a!j . *a f Jt'1 1 . X &M, 


1 - . „. aa Li •*. • 1 1 i Tsfeir/J . M ' - U- 

» J» h 1 

1 a*a' *f • .•• • NS4. 

17 T "*" 

„ •. a -aa-f,;a..-a • 1 >q&i 1 1 Mj 

" • 1 lj iS » a L,,% t ! v vJ'PX.. 1 I :*i. 

• J 1 • 1 ** i i » | 1 | a V^< /„ 


. ,!• • l * ' L M - M ' i • >^H i ' 

*ll IJ» a' • J • * ^<- 

\A """ " 

' * • *• »,, ! j i H "ipM J Ps. > 

II J Ml' 1 " 11 ; >L - - 

■ T •!> « f p • 1 , ^"^ 

.a 1 •! „ ^— 1 

• r »n T n it ^"^ — u — . 

. a 1 | li iZ 

""- "" 

,r * • • 

lat . 1 

a Ml*,!! 1, 'III | 1 


1 i r -j,, r j'i 'il -i ■ 1 4 1 1 i 1 Ml 


35 41 47 

53 59 65 71 77 63 69 95 101 107 113 119 125 131 137 


weight (g.) 


leight (cm.)' XN """ / "* 334 
consecutive necropsies of children at the Johns Hopkins Hospital. Data obtained through 
courtesy of Professor John Howland. Note that in only 43, or 12.9 per cent, of these cases is 
the weight-height index above Bardeen's norm, and the extremely emaciated cases are more 
than 50 per cent below. 

percentage of the volume of the cube of the body length would be occupied by 
the body. According to the data compiled by Bardeen ('20), this weight- 
height index normally remains at about 2.54 in the fetus and infant up to about 
six months of age, then decreases gradually to about 1.20 at maturity. 

Since the weight is relatively depressed during emaciation (see Fig. 36), 
this index is correspondingly lowered during inanition. The relative degree of 
emaciation may therefore be accurately expressed by the weight-height index, 
as shown by Fig. 37, indicating graphically the index for 334 consecutive cases 
of children autopsied in the Johns Hopkins Hospital. It will be noted that in 


only 43 cases, or 12.9 per cent, was the index above the normal average. The 
remaining 87 per cent are below the normal average to a variable extent. 

This brings us to the difficult problem of determining, if possible, the point 
below which the weight-height index should be considered abnormal, denoting 
pathological emaciation. It must be remembered that in general at any age, 
and in each sex of a given race, the individual heights and weights are normally 
variable, the frequencies at various intervals above and below the average being 
distributed roughly according to the "probability curve." Robertson ('23) 
estimates that in children of a given age the normal variability in weight varies 
(at different ages) from 11-20 per cent, and that in stature from 4-6 per cent. 
This necessarily affects the normal weight-height index, which Matusiewicz 
('14), Pfaundler ('16), and Gerber ('21) have shown to be subject to marked 
individual variation. A low index therefore does not necessarily indicate 
malnutrition in the individual case, as Pfaundler ('21) and others have repeatedly 
emphasized. On the other hand, malnutrition may exist in some cases without 
marked depression of the weight-height index. Nobecourt ('16) describes two 
types of infantile denutrition,one ("cachexie") with emaciation and one ("hypo- 
trophie") without. With reference to the racial factor, Dublin and Gebhart 
('23) have shown that the Wood -Woodbury weight-height tables fail as an index 
of malnutrition in Italian children of New York. Davenport ('23) concludes 
that body-build is determined largely by heredity, probably acting through the 
endocrine glands. 

Of the various indices proposed, aside from Bornhardt's index (or others 
involving the thorax), the weight-height index is perhaps the most convenient 
and reliable single physical index of the nutritional condition of the body, but 
no single index can be conclusive for every individual case. A low weight-height 
index is presumptive evidence of malnutrition, but it needs confirmation by 
clinical evidence. The various physical signs and symptoms of malnutrition in 
children have recently been summarized by Roberts ('23) and Goldberger ('23). 
As a result of clinical experience, Holt ('18) proposed an arbitrary zone of 10 
{or 12) per cent or more below the normal weight-height relation as an indication 
of malnutrition, while Emerson and Manny ('20) and Emerson ('22) draw the 
line at 7 per cent (below the Boas-Burk norm). Baldwin ('24) finds in healthy 
school-children an average variability of 6-9 per cent in weight, according to 
height, age and sex. Any such boundary line is necessarily arbitrary and mis- 
leading, however, unless the limitations above mentioned are kept clearly in 
mind. Thus in a series of 506 healthy children, Clark, Sydenstricker and Collins 
('23) found 13 per cent of the individuals more than 10 per cent underweight 
according to Dreyer's standard (stem length and chest circumference tables) ; 
20 per cent were more than 10 per cent underweight according to the Wood 
standard (height-weight-age tables); and 17 per cent were subnormal accord- 
ing to Pirquet's standard, having a " pelidisi " of 94 or less. None of the indices 
so far proposed appears to be very closely correlated with clinical findings 
(Huth '21; Van der Loo '22; Clark, Sydenstricker and Collins '23; Helmreich 
and Kassowitz '23 and others). The subject needs further investigation {cf. 
Taylor '22). 


In infantile atrophy and allied conditions of inanition, the weight-height 
index may drop very low. Thus in 12 fatal cases studied by Jackson ('22) 
the index averaged 1.65 (range 1.20-2.05), or about 35 per cent below Bardeen's 
norm (2.54) (see Table 3). A similar degree of emaciation is indicated in a 
series of 82 observations by Porter ('89) on older children who perished in the 
Indian famine (Fig. 38). Only two (on account of dropsy) of these were above 
normal in weight according to height. The data shown in Figs. 37 and 38 
indicate that a weight-height index 50 per cent subnormal is rarely reached, 
even in cases of fatal inanition. In many cases of chronic inanition the index is 

^yu 1 *. — r- . 1 1 1 

1 [ 

| . Mi 


1 | ; 1 

M , 


■ 1 

i ! 


?q*5-t * 




~I ' ~ ' " . 

C, "j^lt - «_ \v-1- 

IS VfcK 

ii^ - ' -&&£ 


M \<^ : 

_ _ . .. . _ ^ 

•r ■ ir« i>- ~"~ ' S s|j$< '- 


■ST • -1 -P^vi 

IJD izz :::::::::::::: iz ::~- :2l±::i: 

* -~^ 

- .,,-■' x • s> - - 


• 1 \ 

-« i -\ 

• _i> 

?•=> ' 3? ■» X' 

i • » ' 

i * ' • ,, . 

T -t- *~ 

** ! * . ! <• v 

50 ~r | -Doay Lcn 

40 50 60 TO RO 90 100 IK 

) 1£0 150 140 150 160 170 ISO 

Fig. 38. — Field graph showing the weight-height index ( T — . 


weight (g.) 


00) in 


ight (cm.) 1 

victims of the Madras famine of 1877-78. (Data from Porter '89.) Of the 82 eases, only 2 
(probably dropsical) appear above Bardeen's normal weight-height index. The average varies 
from about 33 per cent below in the younger to 40 per cent below in the older. 

lowered not only by the resultant loss in body weight, but also by an actual 
coincident increasing body length, due to a persistent skeletal growth, which 
will next be considered. 

Dystrophic Growth. — That inanition may result in deformity of the body 
in growing organisms has long been known, but the effect in general was usually 
attributed merely to variation in the relative reduction of the various organs or 
parts, as in adults. It is true that (as previously mentioned) cases of "phys- 
iological inanition" had been recognized in amphibia at the time of metamor- 
phosis and in the migration of the salmon, during which certain growth changes 
proceed at the expense of the remainder of the body, with no intake of food from 
the exterior. These were considered exceptional cases, however, without parallel 
in the phenomena of inanition in other vertebrates. The possibility of a con- 


tinued growth in some parts at the expense of other during inanition has but 
recently been recognized as a general phenomenon. Although apparently (in 
vertebrates) first observed in the human species, we may first consider the 
phenomena as found in lower animals. 

Waters ('08, '09), at the Missouri Agricultural Experiment Station, held 
15 young (yearling) steers at maintenance by underfeeding for several months 
and noted some curious changes. The skeleton continued to grow, resulting in 
progressive increase in height and length, apparently at the expense of the adi- 
pose and muscular tissues. Although remaining constant in weight, the body 
underwent various changes in proportions, the head becoming larger and the 
thorax elongated dorsoventrally. About the same time, Lassabliere noted in 3 
malnourished puppies a dissociation of growth in length and weight, similar to 
that found by Variot in human infants (to be mentioned later). Aron ('10, 
'n) likewise noted that puppies underfed so as to maintain nearly constant body 
weight continued to increase in length and height on account of persistent 
skeletal growth. Other changes noted will be mentioned later. A similar 
persistent growth of the skeleton at the expense of the remainder of the body 
was also noted by Falke ('10) in underfed calves, by Aron ('13, '13a) in underfed 
rats, and by Mendel and Judson ('16) (judging by ash-content) in underfed mice; 
although Tschirwinsky ('10) had obtained negative results in undernourished 

The first complete and systematic analysis of the relative weight changes in 
the various parts of the growing body as a result of underfeeding was made 
by Jackson ('15a) in the albino rat, beginning at three weeks of age or later. 
Although there was an increase in body length, and especially in tail length, 
as a result of persistent skeletal growth, the changes in general body proportions 
otherwise were not marked. There was apparently a slight increase in the head 
weight, counterbalanced by a similar decrease in the trunk and extremities 
(Fig. 39). Judson ('16), Thompson and Mendel ('18) likewise noted changes 
in the proportions of underfed white mice, the head becoming relatively enlarged 
and the tail elongated. The various abnormalities of growth on insufficient 
or inadequate diet were reviewed by Mendel ('17). 

The results of Jackson ('15a) upon albino rats underfed beginning at 3 
weeks (age of weaning) or later were confirmed by Stewart ('16). In experi- 
ments beginning on the newborn rats, however, Stewart ('18, '19) found some- 
what different results. In this case, in addition to the increase in body and tail 
length, there was a marked increase (45 per cent) in the head weight, while the 
body weight decreased slightly (4 per cent) . The differences in other organs were 
even more strongly marked, as will appear later. 

In albino rat fetuses retarded in growth by underfeeding the mother during 
the latter half of pregnancy, Barry ('20, '21) found a slight increase in the weight 
of the head and limbs, with a corresponding decrease in the trunk. The body 
length and tail length were also nearly normal, the skeletal growth being much 
less pronounced than that found by Jackson and Stewart during postnatal 
inanition. The results therefore appear to differ materially according to the 
age at which the inanition occurs. j 



The investigations of Waters upon the growth of steers at various planes of 
nutrition were continued and extended by Van Ewing and Wells ('14), by 
Trowbridge, Moulton and Haigh ('18, '19) and by Moulton, Trowbridge and 
Haigh ('21, '22). The results confirmed those of Waters as to changes in the 
form and proportions of the body, with additional observations upon the various 
organs (to be mentioned later). 

A persistent growth in length of very young fasting tadpoles of Ranafusca was 
recently found by Podhradsky ('23). 

£0.6 percent. 

££.T per cent. 

10. 1 percent. 

fore -limbs 
6.9 per cent. 

15.6 percent. 

9.6 percent. 

8.5 per cent. 

15.7 per cent. 

15.4 per cent. 

6T.4 percent. 

54.1 percent. 

55.4 per cent. 

Controls at 3 weeks Constant 5-10 weeks Controls at 10 westo 

Fig. 39. — Graph showing the changes in the relative weights of the head, trunk and limbs 
in young albino rats held at constant body weight by underfeeding from 3 weeks to 10 weeks 
of age, in comparison with normal initial and final controls. (Jackson '15a.) 

In children, Auboyer ('81) ascribed the abnormal growth in length of the 
extremities sometimes observed during and after certain fevers to the stimula- 
tion of the epiphyseal cartilages by the febrile toxins. Camerer ('93) noted an 
apparent increase in length during illness, but in the absence of exact measure- 
ments concluded that this was an illusion, the length appearing exaggerated by 
the emaciated condition. In 1905, however, Camerer, Jr. observed a case of 
chronic gastrointestinal trouble in a child beginning at 9 months and extending 
U P to 5 years of age. There was marked retardation in weight, but normal 


increase in length, which Camerer attributed to dissociation of growth in weight 
and length during undernourishment. He concluded: 

" Der Umstand, dass dasLangenwachstum trotz chronischer Unterernahrung 
keineswegs zu klein ist, beweist die Unabhangigkeit des Langen- vom Gewichts- 
wachstum bei Unterernahrung and liefert einen schlagenden Beweis fur die 
Selbstandigkeit der assimilierenden Kraft wachsender Organe." 

Fleischner ('06) observed that in malnourished children of the same weight 
but at different ages the height shows a regular increase with age. He failed to 
demonstrate the significance of this apparent increase, however. 

Variot (.'07) from cranial measurements on living hypotrophic infants, as 
well as from brain weights at autopsy, concluded that the growth of the brain 
proceeds independently during undernourishment. In a remarkable series of 
publications during the succeeding year, Variot ('07a, '07b, '07c, 'o7e, '08, '08a) 
demonstrated beyond question the frequent dissociation of statural and ponderal 
growth during chronic infantile inanition. The growth in height is retarded 
relatively less than the growth in weight, and in severe cases the height may 
increase while weight is stationary or even slightly decreasing. He found the 
dissociation more distinct in prematures, twins, and weaklings, especially in 
atrophy of gastrointestinal origin. 

Variot ('08b) also demonstrated that a "physiological" dissociation of 
statural and ponderal growth occurs during the normal postnatal loss in body 
weight. In a series observed, the average birth weight at the end of 10 days had 
increased but slightly (from 3,000 to 3,100 g.) while the body length simultane- 
ously increased from 49.5-51.8 cm. Lascoux ('08) confirmed in general the 
results of Variot, but found in some exceptional cases an inversion of the disso- 
ciation, increase in weight occurring during stationary body length. 

Freund ('09) claimed that short, acute infections depress weight, with no 
marked influence upon growth in height, while malnutrition produced by chronic 
infections almost always causes complete cessation of growth in both height and 
weight. Numerous cases illustrating the dissociation of growth in weight and 
height during infantile malnutrition are cited by Vigor ('11), Birk ('11), Lust 
('13), Opitz ('13), Stolte ('13), and Aron ('14). Birk ('n) found that in very 
young undernourished infants (receiving too little breast milk) growth in 
both length and weight were nearly stationary; while in older infants the growth 
in length was more persistent. Lust ('13), on the contrary, found the growth 
in length inhibited less in atrophic nurslings than in cases arising later. The 
whole question of the abnormal (dissociated or uncorrelated) growth during 
infantile malnutrition is reviewed and discussed fully in the treatises of Baud- 
rand ('11), Lesage ('n), Schloss ('n) and Tobler and Bessau ('14) (cf. also 
Jackson '23). 

More recently, Waser ('20) concludes that the variable relations of statural 
and ponderal growth during malnutrition in nurslings may be classified as follows : 
(1) body weight and length stationary (during inanition in pyloric stenosis, 
underfeeding, dyspepsia, and "decomposition" with transient inhibition of 
the growth impulse); (2) decrease in weight, length stationary (in dyspepsia, 
"decomposition" and pyloric stenosis); (3) increase in weight, length stationary 


(in rapid recovery from malnutrition); (4) increase in length, weight stationary 
(in "decomposition" and metabolic disturbances); (5) increase in length, weight 
decreased (in dyspepsia and "decomposition" during the first 3 months of 

Jackson ('22) in 12 fatal cases of atrophic infants (Table 2) found that the 
actual loss of body weight, the final being compared with the maximum recorded 
during life, averaged 19.2 per cent; when the final body weight is compared with 
that normal for the final body length, the apparent loss is greater, averaging 
28.5 per cent. The average difference of 9.3 per cent in the latter case is evi- 
dently due to the average increase in length due to skeletal growth during the 
period of malnutrition.- 

Summarizing briefly the dystrophic growth changes in the young of man 
and other vertebrates, it is clear that growth does not necessarily cease during 
insufficiency or even total absence of food intake. Under such conditions of 
inanition, certain tissues appear able to appropriate nutriment and grow at the 
expense of other parts of the body. Cases of "physiological inanition" in which 
such developmental or growth changes occur are found in the migrating salmon, 
the metamorphosis of amphibia, and the human infant during the postnatal 
loss of birth weight. The most frequent change in the form of the body during 
dystrophic growth is an abnormal elongation, due to persistent skeletal growth, 
which accentuates the emaciation caused by the atrophy of the softer tissues. 
In some cases there is also an enlargement of the head, and disproportionate 
growth of the extremities. The extent and character of this disproportionate 
(or uncorrelated) growth vaiies according to age. As will appear later, it also 
varies remarkably according to the type of inanition (various forms of partial 
inanition) and in the different parts and organs of the body. All of these 
dystrophic growth changes apparently contradict Liebig's "Law of the minimum 
or limiting factor," as strictly interpreted by some authors (cf. von Bunge '01). 

Changes in Adult Proportions (Head , Trunk and Extremities) . — It has already 
been shown that inanition may produce certain changes in the form of the body 
in the young, with dystrophic skeletal growth resulting in elongation of the body 
and accentuating the emaciation. A tendency to abnormal growth of the head 
and changes in the proportions of the extremities were also noted. It remains to 
consider what changes in the form of the body and the proportional size of the 
parts may result from inanition in adult vertebrates. 

Among the lower vertebrates, Harms ('09) observed a definite shortening 
of the vertebral column in Triton taeniatus and Triton cristatus in 2 or 3 months 
of inanition. The tail, however, remained unchanged and thus became relatively 
longer. In the salamander Diemyctylus, however, Morgulis ('n) found a greater 
shrinkage of the tail than of the body during starvation. According to 
Kammerer (' 1 2) , in Proteus anguinus fasting in a dark, cool place, the tail becomes 
relatively shorter, the extremities longer, and the head larger; whereas in a 
warm, light place the body proportions remain normal. 

In adult albino rats, the proportions were studied by Jackson ('15), 15 rats 
being subjected to acute inanition (water only) with average loss of ^ per cent 
in body weight in 9 clays; while 6 were subjected to incomplete inanition (insuffi- 



cient normal diet) with loss of 36 per cent in 5 weeks. In comparison with 
normal controls, the tail appeared relatively elongated, probably due to an 
actual decrease in the trunk length during the inanition period. The relative 
changes in the weights of the head, limbs and trunk are represented graphically 
in Fig. 40. It is evident that the head and fore-limbs become relatively heavier, 
since during inanition they lose in absolute weight less than the body as a whole. 
The hind-limbs likewise increase in relative size during acute inanition, but 
remain nearly unchanged in relative (percentage) weight during chronic inani- 

9 (10) per cent. 

1 1 Z per cent. 

11.4 percent. 

Tore -limbs 
5.0 percent. 

IX. percent. 

6.9 percent. 

15.0 percent. 

Hind- limbs 
IT. 5 per cent. 

15.3 per cent. 

T1.0 percent. 

66.4 percent. 

64 1 per cent. 

Normal initial Acute inanition Chronic inanition 

Fig. 40. — Graph showing the changes in the relative weights of the head, trunk and limbs 
in adult albino rats subjected to acute inanition (water only) with loss of 33 per cent in body- 
weight and to chronic inanition (underfeeding) with loss of 36 per cent in body weight. (Jack- 
son '15.) 

tion. In both cases the trunk decreases in relative weight, compensating for 
the relative increase in head and limbs. Certain changes in the proportions of 
adult steers on submaintenance rations were noted by Benedict and Ritzman 


Changes in body weight and circumferences of thorax and abdomen in 
Germany during the war famine are discussed by Fischer ('23). In 2,114 
adults 20-55 years of age, examined before, during and after the Russian 
famine, Ivanovsky ('23) found marked atrophic effects resulting in emaciation 


and senile appearance. The loss in body weight often exceeded 30 per cent. 
The trunk became bent and shortened, with a decrease in stature of 3.8-6.6 
cm. in the males and 3.6-4.8 cm. in the females. There was apparently both 
absolute and relative decrease in the size of the head, also a tendency to increased 
dolichocephaly and minor changes in nasal measurements. The thoracic cir- 
cumference was markedly decreased; but the extremities appeared elongated, 
in relation to stature. 

Recovery from Inanition. — It is well known that recovery from inanition is 
generally possible, if adequate nutriment is provided before the extreme stage 
is reached. After a certain point, varying according to species, individual, 
type of inanition. and environmental conditions, recovery becomes impossible 
and death is inevitable. The process of recovery upon proper refeeding after 
inanition will be considered briefly, first among vertebrates in general and second 
in man. 

As examples of the extreme extent of inanition from which recuperation is 
possible, Kahan ('85, '86, '04) obtained recovery after loss of 30-45 per cent 
in the body weight of adult pigeons; and after loss of 50 per cent in a hen and 
30.9 per cent in rabbits. In the dog, Laborde ('86) noted recovery after loss 
of 48 per cent in body weight; and Howe, Mattill and Hawk ('09) after a loss 
of 63 per cent in a fast (water only) of 117 days. Liberge reports the 
recovery of a cow after total inanition for 40 days (weights not stated). 
Kammerer ('12) found that in advanced stages of fasting, Proteus refuses food 
and is therefore incapable of recovery, though remaining alive for a long period. 

The recovery after repeated periods of alternating inanition and refeeding 
has also been studied. Kahan ('85, '86) noted that adult pigeons upon refeed- 
ing after repeated inanition acquired a body weight greater than the initial 
weight; and the same tendency was observed in rabbits ('86a). Similar results 
were obtained by v. Seeland ('87) in pigeons and chicks; and by Noe ('00) in 
guinea pigs and rabbits, but not in rats. Thus intermittent fasting in adult 
animals may result in increased body weight. Morgulis ('13) found that in 
salamanders a single protracted fast is less injurious than intermittent fasting. 

Rowntree ('22) states that one of Poletaeff's dogs survived 22 days (total 
inanition) with 47 per cent loss in body weight. In a subsequent starvation it 
succumbed only after a loss of 60 per cent. The necropsy showed some fat 
still present. A rabbit survived 8 days of total inanition, with loss of 32 per 
cent in body weight. After 7 days, it was again starved and died on the 10th 
day, with loss of 40 per cent. 

The effects of inanition upon the subsequent growth in young animals have 
also been frequently studied. In the first place numerous investigators (Minot 
'91 and Lepine '75a in guinea pig; Hatai '07, Ferry '13, Stewart '16, Jackson 
and Stewart '18, '19, '20, in the albino rat; Robertson '15 in mice; Springer '09 
and Morgulis 'n, in the salamander) have found unusually rapid growth upon 
refeeding after inanition, so that (unless the inanition is prolonged) the organism 
tends to recover the body weight normal for its age. According to Tobler and 
Bessau ('14), the abnormally rapid growth following retardation is probably 
due to the fact that cell-division has proceeded meanwhile, making conditions 



unusually favorable for growth on refeeding. Robertson ('23) and others have 
thought that inanition may facilitate subsequent growth by the removal of 
accumulated inhibitory products. 

Ultimate Effects. — As to the question of the ultimate effects of the inanition 
and the possibility of a permanent stunting in the later growth of the body, the 
results of animal experiments appear somewhat conflicting. Thus Hatai 
('07) and Stewart ('16) and Jackson and Stewart ('18, '20) obtained complete 

11 41 62. 62 [01 ill [41 162. 162 

1 I — I — I — r I 1 T 
Day.5 of refeeding 

£0 40 60 60 100 1C0 140 160 160 COO 220 240 260 280 300 3Z0 340 

Fig. 41. — Chart showing curves of growth in albino rats amply refed after underfeeding 
from 3 to 20 weeks of age. (Jackson and Stewart '20.) L indicates the birth of a litter in the 
control female. 

recovery upon refeeding young albino rats which had been underfed for various 
periods beginning at three or four weeks of age. Aron ('14) noted no permanent 
stunting of rats underfed less than 150 days. Similar results were obtained by 
Osborne and Mendel ('14a, '15a), who found no suppression of growth capacity 
in young rats whose growth had been retarded for long periods by various inade- 
quate diets (to be mentioned later). On the other hand, Aron ('io, 'n, '14) and 
B riming ('14) found that severe underfeeding of young dogs and rats apparently 
prevents them from reaching normal adult size upon later full feeding. Jackson 

9 6 


and Stewart ('19, '20) obtained the same result if the inanition of the rats is 
begun at a very early age or is protracted over a long period. They therefore 
conclude that "The ultimate effect varies according to the length of the under- 
feeding period, the age at which the inanition occurred, the sex (body weight 
more affected in males), the severity and the character of the inanition" (see 
Figs. 41 and 42). In steers held at maintenance by underfeeding for nearly a 
year, Moulton, Trowbridge and Haigh ('21) failed to obtain full recovery of 
the noimal body weight, length, width and circumference, even after 3 years of 
full refeeding. 

As to recovery of normal proportions, the results of Stewart ('16) and of 
Jackson and Stewart ('19) indicate that upon refeeding young albino rats after 

18 38 58 Ya 96 118 138 158 178 

— i i i i i r 
Days of refeeding 

30 40 60 80 100 IK) 140 160 180 £00 HO C4Q ffiO d60 300 3Z0 340 360 380 400 4£0 440 460 460 500 5Z0 

Fig. 42. — Chart showing curves of growth in albino rats amply refed after underfeeding 
from 3 weeks to 1 year of age. (Jackson and Stewart '20.) The test rats fail to recover fully, 
remaining permanently stunted. 

underfeeding for various periods, the head, limbs, trunk and tail rapidly regain 
their normal conditions (Table 7). Even in rats permanently dwarfed by long 
underfeeding, Jackson and Stewart ('20) found that the parts of the body regain 
the weights nearly normal for corresponding body weight, although body length 
and tail length become slightly subnormal relative to body weight (Table 8). 
In the human species, the available evidence indicates that the process of 
recovery from inanition is very similar to that in other mammals (cf. Carrington 
'08). As previously mentioned, numerous voluntary adult fasts of 30 days or 
more are recorded, with subsequent recovery from loss of 20-25 per cent in body 
weight. Recovery from still greater losses is not unusual after periods of mal- 
nutrition due to chronic illness. Durlocq (Fernet '01) noted complete recovery 
in a hysterical girl, aged 15, who had lost 38 per cent in weight through total 
inanition. In the Russian famine, Ivanovsky ('23) observed recovery from 
extreme emaciation. On the other hand, there are definite limits beyond which 
recovery is impossible, as in the case reported by Meyer ('17), where recovery 


failed in a man after loss of about 40 per cent in body weight from a fast of 60 
days on water alone. 

Some data concerning the recovery from inanition in infants and children 
have already been mentioned. Rapid recuperation after temporary retardation 
of growth has been noted by Coudereau ('69), Pagliani ('79), Camerer ('93), 
Filliozat ('09), Holt ('18), Czerny ('21), Goldstein ('22), and others. Quest 
('05) found that in infants a loss of more than 34 per cent in weight is usually 
fatal, while Rosenstern ('n) has observed only 3 cases of nurslings surviving a 
loss of over 32 per cent (the maximum being 35 per cent). Baudrand ('n) 
distinguishes primary atrophy (constitutional, due to defect in the germ cell) 
from secondary atrophy (due to inanition from extrinsic causes), only the latter 
type being capable of recuperation. Thus in 73 underweight (not premature) 
newborn, Opitz found that 28.7 per cent made quick recovery of normal weight; 
27.4 per cent continued growth parallel to the normal curve; while 43.9 showed 
continued retardation, the growth curve becoming progressively subnormal. 
It is generally believed that children poorly nourished over long periods will 
not attain normal adult size (Burk '98, and others) although this is difficult to 
prove. Tobler ('13) observed a child weighing 4,500 g. at birth which upon 
weaning at 7 months suffered a chronic malnutrition from dyspepsia, which 
retarded growth until well into the second year. Although the digestive trouble 
was later fully overcome, the child was permanently stunted in growth, with a 
length of only 89 cm. and a weight of 10,500 g. at 8 years of age. 

Recent data published by the German public health office on the height and 
weight of 69,000 Leipzig children of the Volksckulen (Jour. A.M. A., Nov. 4, 
1922, 79:1623) indicate an increase in their average height in the period 1919- 
192 1, especially for the ages 7 to 10. The increase in weight was proportionately 
less, and was greater in 1919-20 than in 1920-21. The present weights are still 
below those just previous to 1914 (prewar period). These children may yet 
become normal, but it is quite possible that permanent dwarfing may occur as 
a result of severe or protracted inanition during infancy or childhood, in accor- 
dance with the results obtained by experimental inanition upon animals. 


Effects of Partial Inanition 

Under this heading will be considered the changes produced by deficiencies in 
the various essential food factors — protein, fats, carbohydrates, inorganic salts, 
vitamins and water. Some of these have already been referred to incidentally 
under total inanition. In many cases, indeed, it is quite probable that the 
effects of total inanition, especially when incomplete (general underfeeding), 
are in reality due not to caloric insufficiency, but to the inadequacy of some one 
of the special dietetic factors, in accordance with Liebig's" law of the minimum." 
During protracted underfeeding, the limiting factor will be that essential which 
is the first to become exhausted in the body so that the deficiency becomes 
manifest. The time required for exhaustion of the various essential substances 
will depend upon (i) the amount of each substance which is stored in the body 
as available reserve; (2) the rapidity with which each substance is consumed in 
the body (which in turn depends upon various factors influencing metabolism) ; 
and (3) the intake of each substance in the food, in case of incomplete inanition. 
The exact character of the deficiency in total inanition, either complete or incom- 
plete, is therefore exceedingly variable according to the species, age, individual, 
environment, etc. 

Exhaustion of all the essential food substances in the body at the same time 
is practically impossible; and the depletion of but a single substance is rare under 
ordinary circumstances. What usually happens during under-nutrition is that 
more than one substance becomes exhausted, or falls below the minimum 
required, so the symptoms are likely to be those of a mixed deficiency. The 
variation in the factors which become exhausted may explain why the " defi- 
ciency diseases" are so variable in their character, with more or less resemblance 
to total inanition on the one hand, and to the various individual forms of partial 
inanition on the other. Of interest in this connection is the claim of F. Mtiller 
{'97) that several emaciating conditions (infectious fevers, malignant growths, 
severe diabetes, etc.) exhaust the protein content of the body more than does 
ordinary inanition. Opitz (Zentralbl. f. Gyn., 1924, 48:2) holds that the so- 
called toxicoses of pregnancy are essentially deficiency manifestations of inani- 
tion in the maternal organism. The general pathology of partial inanition is 
discussed by Weill and Mouriquand ('19). 

The cause of death from inanition, either total or partial, has often been dis- 
cussed but is still uncertain (Howe '12). Various theories have been proposed. 
It was early recognized that death is not due simply to exhaustion of stored food 
materials, since fat may persist in appreciable amount. Collard de Martigny 
(1828) ascribed death from starvation to impoverishment of the blood, resulting 


from loss of solids. Chossat ('43) concluded that death is caused by the fall of 
temperature. Although accepted by Bourgeois ('70), this cannot be the 
primary cause, since it is obviously inapplicable to cold-blooded animals; and 
even in the warm-blooded, death cannot be prevented by artificial maintenance 
of heat. Lukianov believed that death is due, not to exhaustion of reserves, 
but to inability of the organs to utilize them; but he failed to explain this 
inability. Beeli ('08) held that death is due to asphyxia, the accumulation of 
toxic materials causing paralysis of the respiratory center in the medulla. 
Lipschiitz ('18) concluded that death from inanition is produced by auto-intoxi- 
cation, due to toxins resulting from disordered metabolism of the malnourished 

Whether it be a direct effect of the inanition, or indirectly caused by toxins 
in circulation, a lowered resistance to infection is a well-known result of various 
types of inanition, both total and partial. Thus the immediate cause of death 
following inanition is frequently an infectious complication, such as terminal 
bronchopneumonia in the human species. The effects of inanition upon the 
ductless glands may also form an important complication as will be mentioned 
later under the various organs. It is therefore evident that the immediate cause 
of death from inanition may vary according to circumstances. 

On the other hand, since most diseases, especially the chronic disorders, 
interfere more or less with the process of nutrition, inanition is usually present 
as a complication. As expressed by Chossat: "L'inanition, on peut done le 
dire, est la cause de mort qui marche de front et en silence avec toute maladie, 
dans laquelle l'aliment n'est pas a l'etat normal. Elle arrive a son terme naturel, 
quelquefois plus tot, quelquefois plus tard que la maladie qu'elle accompagne 
sourdement et peut devenir ainsi maladie principale la ou elle n'avait ete 

The general effects of partial inanition upon the body as a whole will now be 
summarized briefly, following which the various individual types of partial 
inanition will be considered. 

Summary of Effects of Partial Inanition on the Body as a Whole 

Various essential proteins (amino-acids), salts (of P, Ca, Na, Fe, I), vitamins 
(A, B, C and probably others) and water cannot be synthesized in the vertebrate 
organism and must therefore be present in the food-intake. If they are absent 
from the food-intake, or inadequate in amount, decline in body weight with 
various other characteristic symptoms of malnutrition supervene, as soon as 
the available supply of reserves of these substances stored within the body is 
exhausted. The period required for such depletion varies according to the 
substance in question, the species, age and previous nutritive condition of 
the individual, as also according to various environmental factors influencing 
the process of metabolism whereby the essential substances are consumed. 

Each of the essential food factors has its special function in nutrition, and 
the deficiency in each case results in the characteristic symptoms of the corre- 
sponding "deficiency diseases." Thus malnutritional edema and pellagra are 


probably due chiefly to protein insufficiency. Phosphorus and calcium are 
especially important in the formation of bone, and their deficiency (with lack 
of a special vitamin) may result in rickets or similar disorders with skeletal 
lesions. Deficiencies in vitamins A and B give rise to characteristic syndromes 
(the former to xerophthalmia; the latter to beriberi and polyneuritis), and defi- 
ciency in vitamin C causes scurvy. Water is also essential, but fats and carbo- 
hydrates may be replaced by an excess of protein in the diet. In most cases of 
partial inanition, more than one essential factor is simultaneously depleted, the 
resulting malnutrition therefore being due to a mixed deficiency. 

In a strict sense, a total inanition is non-existent. Even in the complete 
absence of food-intake, the living cells of the body are not entirely deprived of 
nutriment, being still nourished by the blood-stream and lymph, which are more 
or less imperfectly replenished by absorption of such stored or reserve materials 
as may be available in the organism. When this nutriment becomes inadequate 
with respect to any essential factor, the cells of the tissues most concerned with 
that factor will undergo disordered metabolism, although in young animals an 
abnormal, dystrophic growth may continue for some time. Ultimately, and 
especially when the cell-nutriment becomes inadequate to supply the energy 
required for the vital functions, cell atrophy and degeneration occur (as will 
be shown later), the body weight falls and death follows. 

On account of this fundamental similarity between total and partial types of 
inanition, it is not surprising to find that they resemble each other in many of 
the more general effects. Thus in both cases there is usually a marked and pro- 
gressive loss of body weight with resultant emaciation in the adult, the limits 
being about the same in both types of inanition. In the young there is in both 
cases a retardation or inhibition of growth, frequently accompanied by dystro- 
phic growth phenomena which vary according to circumstances. Finally there 
is, after partial as well as total inanition, the possibility of recuperation upon 
adequate refeeding, depending upon the character and extent of the injury 
produced by the inanition. In both cases, there is an upper (somewhat 
variable) limit beyond which death is inevitable. There is also a lower limit, 
below which perfect recovery is possible. Between these upper and lower limits 
of inanition, there is probably in all cases a degree of injury possible which 
permits of only partial recovery, resulting in a variable degree of dwarfing 
and deformity of the body. The available evidence indicates that in some cases 
the long-continued suppression of growth during partial (inadequate protein) 
inanition injures the capacity of subsequent growth much less than does total 
(incomplete) inanition. The more specific effects of the various types of inani- 
tion will be considered in the subsequent chapters upon the individual organs. 

Effects of Various Types of Partial Inanition on the Body as a Whole 

Protein Deficiency. — It has long been known that protein forms an essential 
factor in human and animal nutrition. The earlier experiments (cf. Munk 
'91; Rosenheim '91) upon protein deficiency, however, were unsatisfactory in 
two important respects. In the first place, they neglected other factors, such 


as inorganic salts and the recently discovered vitamins, so that they often repre- 
sent mixed deficiencies of uncertain character from which no definite conclusions 
can be drawn. In the second place, only within the past decade has it been 
generally recognized that the quality of the dietary protein is even more impor- 
tant than the quantity (Hart, McCollum, Steenbock and Humphrey 'n). 
Chiefly through the investigations of Osborne and Mendel ('n, 'ua, '12a, 
'12b, '14, '15, '15a, '16b, etc.) and their co-workers it is now recognized that 
among about 18 amino-acids which, in varying proportions, constitute the ordi- 
nary proteins of foods, several, though essential for nutrition, cannot be synthe- 
sized in the animal body and must therefore be present in adequate amount in 
the food-intake. Among these essential amino-acids, tryptophan, lysin, tyrosin, 
and cystin are especially important for normal growth, which is prevented or 
retarded in the absence or insufficiency of any one of them. Liebig's "law of 
the minimum" is therefore considered applicable to the essential amino-acids 
in the diet (Osborne and Mendel '15a, '16b). Thus on account of their varying 
amino-acid content, diets with some isolated proteins (gliadin, edestin, glutenin, 
casein) may permit maintenance of young rats without growth for long periods, 
while others (zein, gelatin) do not even suffice for maintenance, but occasion 
more or less rapid decline in body weight. 

Contrary to the previously mentioned results of (incomplete) total inanition 
in man and animals, Osborne- and Mendel ('n, 'ua), in young rats held at 
maintenance for long periods (up to a year or more) by incomplete protein diets, 
found no changes in the body proportions, excepting the possibility of continued 
growth in the nervous system. Mendel and Judson ('16), however, described 
persistent skeletal growth in mice retarded by diets inadequate in protein or 
salts, as well as by simple underfeeding. Mendel later ('17) described abnor- 
malities of growth on various insufficient or inadequate diets. 

Recovery upon Refeeding. — Osborne and Mendel also ('n, 'na, '12a, 
'12b, '14a, '15a; Mendel '14, '15) found in the retarded rats a remarkable capacity 
for recuperation upon adequate refeeding, and concluded that it seems impos- 
sible by this type of inanition, no matter how long continued, to suppress the 
capacity for growth, or to produce permanently dwarfed individuals. It also 
appears possible in this way to increase the total span of life in rats, and refed 
females have borne apparently vigorous young after the normal age of meno- 
pause (Osborne and Mendel '17). Wheeler ('13) likewise obtained normal 
recovery in young mice retarded in growth for long periods on gliadin or casein 

This failure to produce permanent dwarfing by long continued suppression 
of growth in rats and mice on incomplete protein diets is in striking contrast 
with the (previously mentioned) results of Jackson and Stewart and others 
in animals dwarfed by underfeeding. It should be noted, however, that the 
incomplete protein experiments were not begun upon very young animals, 
and it is possible that a similar retardation of growth at earlier stages (between 
birth and the weaning period) would result in permanent dwarfing, as found in 
rats by the underfeeding experiments, and in certain types of partial inanition 
(to be mentioned later). 


Among experiments representing protein deficiency, complicated by absence 
of other essential factors, may be mentioned those of Hatai ('04, '07), who 
obtained loss of about 30 per cent in body weight in young rats on starch-suet 
diets, with rapid recovery on normal refeeding. Schulz ('12) found no increase 
in length or weight in puppies fed farinaceous gruels. On refeeding adequate 
diet after 3^ months, full recovery was obtained if the experiment began with 
puppies at 2 to 4 weeks of age, but not with those at 4 days of age. Briining 
(' 14) also found that young rats soon cease to grow on an unbalanced carbohydrate 
diet, recalling the " Mehlnahrschaden " of human infants. Albrecht ('13) 
stated that protein-poor diet affects pregnant mares, though not the develop- 
ment of the fetus; but Evvard, Cox and Guernsey ('14) observed that feeding 
maize diet (mixed deficiency) to pregnant sows causes reduction in the size and 
vigor of the newborn pigs. Funk and Macallum ('14) held the body weight of a 
young chicken nearly constant (at 150-160 g.) for about 7 months on a diet of 
rice and cod liver oil (mixed deficiency). 

In the growth of the frog tadpoles, Emmett and Allen ('19) concluded that 
the quality of the protein in the diet is more important than the quantity. 
Evans and Bishop ('22) demonstrated the effect of variable quantities of protein 
upon the growth curve of albino rats. Slonaker and Card ('23, '23a, '23b, 
'23c, '23d) have shown that in general the growth and reproduction of albino 
rats are much less upon a protein-poor mixed vegetable diet than upon the same 
with the addition of animal protein (omnivorous diet). 

Malnutritional Edema. — That general or localized edema (with or without 
ascites) may be produced in rats by deficiency in protein (or fats) , with abundance 
of water in the diet, is the conclusion reached by Denton and Kohman ('18) 
and Kohman ('19, '20). This may also be responsible for the dropsy occurring 
in malnourished sheep, cattle and horses. (Friedberger and Frohner '08; 
Hoare '15; Frohner and Zwick '15; Hutyra and Marek '16). Many authors 
believe that this deficiency is likewise the primary (or at least an important) 
factor in the edema commonly observed in conditions of human famine (Cornish; 
Maase and Zondek '17, '17a; Schiff 'i7;Lange '17; Park '18; Wells '18; Schitten- 
helm and Schlecht '19; Maver '20; Prince '21; McCollum '22). This condition 
may resemble the dropsical type of beriberi (Budzynski and Chelchowski '16); 
and McCollum ('22) holds that "wet beriberi" is due to a double deficiency of 
protein and vitamin B. Edema may also be associated with pellagra (Enright 
'20). Harden and Zilva produced edema in a monkey on a diet without vitamin 
A, and Fracassi ('22) considers vitamin deficiency an important factor in 
"hunger edema." Many writers, however, as already mentioned, have con- 
sidered that "famine edema" is due to general quantitative (incomplete total) 
inanition, rather than to qualitative, specific or partial inanition. Excess of 
carbohydrates and water is also frequently considered an accessory factor 
(McCarrison '21; McCollum '22). Curschmann ('22) believes the effect 
is due primarily to endocrine disturbance. 

In describing the "famine edema" in German cities, Kraus ('19) states that 
"Die bleichen, bis 40 pet. ihres urspriinglichen Gewichtes abgemagerten, 
hydropisch geschwollenen, durch Muskelschwache unbeweglich gewordenen 


Menschen boten einen nicht weniger schrecklichen Anblick als die Facies pestica. 
Matthias ('19) describes two types of edema, one with ascites but without 
adipose atrophy, ascribed to partial inanition; the other involving general 
edema and brown atrophy of adipose tissue, ascribed to general inanition. The 
immediate cause of the edema in most cases is probably an injury of the capil- 
lary walls (to be considered in a later chapter), without renal or cardiac lesions. 
Croftan ('17) has emphasized the importance of edema as a danger signal in 
the starvation treatment of diabetes. 

Pellagra. — This malady was observed in Spain by Casal about 1730 and in 
Italy by Frappoli (1771), who introduced the word pellagra ("skin lesion"). 
Various theories as to its etiology have been proposed, which have recently 
been reviewed extensively by Raubitschek ('15), Harris C19), Snyder ('23) and 
Vaughan ('23). The characteristic condition of malnutrition associated with 
the disease was observed early, and the theory that pellagra is caused by gen- 
eral inanition was advocated by Soler (1791), Cerri (i8o4-'o5), Robolotti ('65) 
and others. 

That maize diet may be a factor in pellagra was suspected even by Casal, 
and was noted by many of the earlier writers. More specifically, the deficiency 
of this diet in protein as a causal factor was advanced by Marzari (1810), 
Morelli ('55),Lussana and Frua ('56) and Calmarza ('70) although other environ- 
mental factors were recognized. This theory of protein deficiency, in more or 
less modified form, has recently been supported by Boyd ('20), Roberts ('20), 
and by. Goldberger and his associates in several papers, but is opposed by 
Hindhede ('23). The theory that pellagra is caused by toxins produced by 
changes in the maize was also supported by various early writers, and more 
recently by Lombroso and others, even down to the present (Marie '08, '10; 
Nichols '12, '13; Centani '14; Niles '16, '17). The theory of infection as a cause 
of pellagra has also had many adherents (cf. MacNeal '21), and recently the 
possibility of avitaminosis has been suggested by Rondoni ('15, '19) and by 
Funk ('22). McCollum and Simmonds ('17) state that typical pellagra- 
producing diets are deficient in the fat soluble A vitamin, inorganic salts, and 

Animal experiments to determine the cause of pellagra have been somewhat 
inconclusive. Maize or other presumably pellagra-producing diets were fed by 
Nicholls ('12, '13) to rats; by Rondoni ('15, '19) and Rondoni and Montagnani 
('15) to guinea pigs; by Chittenden and Underhill ('17) to dogs ; by Sundwall 
('17) to rats, monkeys and pigs; by Sullivan ('20, '20a) to pigeons, and by Chick 
and Hume '('20) to monkeys. Varied symptoms of malnutrition were thereby 
produced, but apparently in no case did these resemble very closely those of 
typical human pellagra. 

On account of the immense literature on pellagra (Raubitschek '15 gives a 
bibliography of 1,472 titles including "nur bemerkenswerte und wissenschaft- 
liche Publikationen"), it will be possible to mention only a few of the papers, 
more especially those dealing with various phases of its pathology. As to the 
effects on the human body as a whole, there is noted a condition of general 
malnutrition, with variable loss in body weight, up to extreme emaciation 


(Tuczek '93; Nichols '15, '19; Sundwall '17; Goldberger and Wheeler '20; and 
others). In some cases, there is a tendency to edema ("wet form" of pellagra), 
ascribed by Fraenkel ('69-'7o) to renal and cardiac lesions in "pellagra typhosus." 

Marie ('08, '10) states that loss in body weight is not a constant symptom, 
but occurred in 84 per cent of the females and 74 per cent of the males. Calder- 
ini ('47) found 514 out of 1,005 cases notably under weight. The decrease in 
weight is said to correspond to progressive repugnance for food, although some 
cases (especially of the typhoid type) may appear well nourished. In late 
stages there is usually extreme emaciation. 

Deficiency in Fats and Carbohydrates. — It has generally been held that 
carbohydrates are desirable in the diet, in order to avoid the ill effects of the 
large amounts of protein or fat otherwise necessary to supply sufficient calories. 
It is well known, however, that dogs and other carnivorous animals may thrive 
upon a meat diet nearly free from both carbohydrate and fat. Prochownick 
('89, '01, '17), advocated a diet rich in protein and poor in carbohydrates and 
water, in order to reduce fetal size. Osborne and Mendel ('12, '20a) obtained 
normal growth in rats on nearly fat-free rations. They ('21b, '21c, '2id) 
have recently shown that it is quite possible to obtain growth of rats to adult 
size on a diet practically free from digestible carbohydrates; and the body weight 
may be at least trebled without either fat or carbohydrate. Evans and Bishop 
('22) have also obtained excellent growth in rats on diets nearly free from fats or 

Most of the earlier experiments on fat deficiency were complicated by the 
fact that the (then unrecognized) " fat-soluble A" vitamin was also eliminated, 
which is now known to be essential for growth. This probably explains the 
results of McCollum and Davis (,'13), Hatai ('15) and Stepp ('17), who found 
growth retarded in rats on lipoid-free diets. Drummond ('20) observed good 
health but subnormal growth (possibly due to other factors) on fat-free rations; 
but Drummond and Coward ('21) obtained normal growth from weaning to 
maturity in rats on a diet nearly free from neutral fat. 

It is well known that both fats and carbohydrates may be synthesized in 
the animal body, and McCollum, Halpin and Drescher ('12) have shown that 
this is true also for lecithin (phosphatized fat). Normal nutrition is therefore 
possible in the absence of these substances from the diet, although the length of 
life during total inanition is materially affected by the amount of body fat 
present (Voit '01 a). 

Deficiency of Inorganic Salts. — The necessity for an adequate supply of 
salts in the food-intake has long been recognized, and was emphasized by v. 
Liebig. Their relative abundance in foods and their storage in the body are 
discussed by Forbes ('19). Forster ('73) reviewed the earlier literature on this 
question and experimented with low salt diets (also probably somewhat deficient 
in vitamins). On such diets, pigeons perish in 13-29 days, and dogs survive 
26-36 days, with progressive weakness and paralysis. Gaube ('97) found that 
mineral hunger in pregnant rabbits causes abortion, with stillborn or poorly 
developed young. A diet poor in minerals (Na, Ca, Mg, K and Fe) likewise 
resulted in less vigorous chicks. Hart and McCollum ('14) concluded that the 


restricted growth of herbivora, rat and swine on wheat or maize diets is not due 
to protein deficiency alone, since the addition of certain salt mixtures resulted in 
improved (though still subnormal) growth. McCollum and Davis ('15a) 
found that the addition of salts to wheat diets gives great improvement, but 
still less than half normal growth in rats, on account of other deficiencies. 
Czerny and Finkelstein emphasize demineralization as a factor in athrepsia. 
Grabley ('19) believes that mineral deficiency in the diet is a cause of imperfect 
growth and nutrition in man. Only slightly subnormal growth in rats on low 
salt diets was found by Evans and Bishop ('22), but H. G. Miller ('23) noted 
marked retardation in the growth of young rats with dietary deficiency of 

Babcock ('05) noted that cattle ordinarily obtain from their rations sufficient 
sodium chloride for maintenance, but that during protracted lactation they 
become progressively malnourished and will perish unless additional salt is 

Among experiments resulting in malnutrition or growth failure on diets 
considered deficient in calcium (in some cases also deficient in other essentials) 
are those by Chossat ('42) on pigeons with calcium-poor wheat diet; by Weiske 
('74) on rabbits with "calcium-free barley;" by Hart and Steenbock ('19) 
on pigs with maize and oats; by Elliot, Crichton and Orr ('22) on pigs with oat- 
meal, etc.; by Russell and Morrison ('19) on cattle with oats; and by Haigh, 
Moulton and Trowbridge ('20) on a calf with silage and maize diet. 

The maize diet fed by Evvard ('12) to pregnant sows, resulting in weak 
and underweight offspring, was deficient in calcium as well as in protein. Dib- 
belt ('10, 'n) maintained that there is normally a physiological "calcium 
hunger" in newborn mammals (including human infants), the shortage of cal- 
cium in the maternal milk (denied by Wieland '13) being supplemented by 
absorption of calcium stored up in the bones of the offspring during the fetal 
period. McCrudden ('13) concluded that human dwarfing may be due to 
metabolic disturbance associated with calcium deficiency during the growth 
period. The constitutional effects of calcium deficiency in children have been 
reviewed recently by Stheeman ('21). This question will be discussed further 
in connection with rachitis, and in the chapter on the skeletal system. Some 
effects observed by Korenchevsky ('23) in the offspring of rats on diets deficient 
in calcium or vitamin A will be mentioned later under the vitamins. 

Voit ('80) noted that in spite of the skeletal lesions the general growth of the 
body is not inhibited in dogs on calcium-poor diet. Stoltzner ('09a) claimed that 
this, as well as the continued growth of the body during experimental anemia 
with iron-poor diets, is contrary to the "law of the minimum" (as advocated by 
von Bunge). Similar results with diets deficient in phosphorus will be men- 
tioned later. The calcium and phosphorus necessary for growth under these 
conditions are provided by absorption from the supply already stored in the 
skeleton. Osborne and Mendel ('18a), however, maintain that the "law of the 
minimum" holds for all essential salts in the diet, failure of growth in the whole 
body resulting where the limiting factors are deficiencies in the salts of chlorine, 
sodium, magnesium, potassium, calcium and phosphorus. 


McCollum and Simmonds (McCollum '22) found that rats, kept at an early- 
age upon diets in which the inorganic content is unsatisfactory, develop abnormal 
forms and become permanently stunted. They become stocky, owing to failure 
to grow in length. This permanent suppression of growth is contrasted with the 
successful recuperation after retardation by vitamin deficiency, likewise after 
inadequate protein diet (Osborne and Mendel) and resembles the results of 
general underfeeding (Jackson and Stewart). 

The effects of phosphorus deficiency upon growth have likewise been demon- 
strated. Hart, McCollum and Fuller ('09, '09a) found that pigs at 40-50 
pounds made normal gain in body weight up to 75-100 pounds on low phosphorus 
grain diets. Then followed loss of weight, weakness and collapse (presumably 
due to exhaustion of the available phosphates stored in the skeleton). Good 
results followed upon the addition of calcium phosphate to the diet. 

Lipschutz ('10, '11) obtained continued though subnormal growth in puppies 
on a phosphorus-poor diet of rice and egg albumin, supplemented by salt mix- 
tures. The skeletal lesions resembled those of scurvy (vitamin deficiency). 
Heubner ('11) similarly observed that puppies continued to grow for 7 weeks on 
rice diet, which is poor in phosphorus (also otherwise deficient), with subsequent 
decline. On a tapioca diet, growth ceased in 3 or 4 weeks, and marked emacia- 
tion followed. On refeeding with rich mixed diet, normal appearance and 
health were rapidly restored, but in body size and weight the dog remained 
permanently dwarfed, corresponding to a puppy of 3 months. Masslow ('13) 
also found that puppies on phosphorus-poor diet continue nearly normal 
growth for about one month, but later lose weight, become emaciated and die. 
Sherman ('n) concluded from a study of typical American dietaries that 
human malnutrition is frequently due to phosphorus deficiency. Further data 
on calcium and phosphorus deficiency will be mentioned later in connection 
with the discussion of rickets. 

Emmett, Allen and Sturtevant ('20) and Swingle ('22) cite evidence that 
iodin is effective in causing the metamorphosis of amphibia, which recalls the 
well-known experiments of Gudernatsch, showing acceleration of metamorpho- 
sis by thyroid feeding. The question of iodin deficiency will be considered later 
in the chapter on the thyroid gland. Smith ('17) concluded that iodin deficiency 
in the diet during gestation in sows results in the birth of weak, hairless pigs of 
full size but with edematous skin and other abnormalities, ascribed to disturbance 
of function in the fetal thyroid gland. There may be a similar occurrence in 
sheep and other domestic animals. 

Fetzer ('13) found that, on diets deficient in iron pregnant rabbits can supply 
from their own bodies the iron necessary for fetal growth only up to a certain 
limit, beyond which fetal death occurs. Hess (,'22) cites evidence indicating 
that in some cases a lack of iron in the diet of infants may lead not only to anemia 
but also to marked retardation in growth, with prompt recovery upon addition of 

Rachitis. — While the lesions in rickets appear most prominent in the skeleton 
(to be considered in a later chapter), it is nevertheless a general metabolic dis- 
order with certain broader aspects which may be briefly mentioned here. The 


etiology of rickets has been much disputed. It has generally been held to be due 
to dietary insufficiency of some kind, although infection, toxins, or bad hygienic 
conditions in general have been frequently considered primary factors, even 
down to the present (Looser '20; Paton and Watson '21). Esser ('07) claimed 
the production of rickets by overfeeding in young rats; and it has even been 
observed to appear spontaneously in this species by Erdheim ('14) and Pappen- 
heimer ('14). Some consider that rickets is produced indirectly through effects 
on the endocrine system (Stoeltzner, '21). 

Among those who consider rickets to be caused by dietary deficiency, 
there have been varied opinions as to what factor is deficient. For human 
rickets, a deficiency in fats was sometimes held responsible (Vincent '04, 
Cheadle and Poynton '07; opposed by Hutchinson '20 and others); although 
mineral deficiency has frequently been suspected, especially in rachitoid dis- 
orders in animals, such as that in cattle described by Lotsch andLange ('12). 
Lehnerdt ('10) and Rohmann ('16) hold that rickets or "pseudorickets" may 
be caused by (1) deficiency of calcium salts in the diet; (2) deficient calcium 
absorption in the intestine; (3) excessive calcium excretion; or (4) defective 
calcium assimilation by the osteoblasts. The disputes as to whether the dis- 
orders produced by calcium and phosphorus deficiencies in animal diets are 
really identical with human rickets will be discussed in the chapter on the 
skeletal system. 

Recently the controversy has shifted to the question concerning the causa- 
tion of rickets by deficiency in vitamin A, which was advocated by Mellanby 
('19), Nathan ('20), Higier ('22) and others; but was opposed by Hess and Unger 
('19), Hess, McCann and Pappenheimer ('21), Mackay ('21), Paton and Watson 
('21, '21a) and others. It may be stated here, however, that the most recent 
work on experimental rickets by E. Mellanby ('21), Sherman and Pappenheimer 
('21) and by McCollum and his associates indicates strongly that the cause of 
rickets is not a single deficiency, but it is complicated in character and dependent 
upon various factors. For puppies, Mellanby ('21) considers the following 
as important: (1) a deficiency of calcium and phosphorus in the diet; (2) 
a deficiency of fat containing the antirachitic vitamin; (3) excess of bread, other 
cereals and carbohydrates; (4) absence of meat; (5) excess of the protein portion 
of caseinogen, free from calcium; (6) confinement. McCollum and his coworkers 
have recently shown that in young rats the experimental production of rickets 
(or closely allied conditions) depends largely upon the calcium-phosphate ratio 
in the diet, either low calcium or low phosphorus being effective, in the absence 
of a "fourth vitamin" (distinct from, though closely associated with vitamin A) 
which promotes calcium deposition in ossifying tissues (McCollum, Becker and 
Simmonds '22, '22a; McCollum '23). Byfield and Daniels ('23) found it impos- 
sible to produce typical rickets in rats on diets low in calcium, phosphorus or 
butter fat, except when the experiment was extended to the second generation. 
Chick et al. ('23) conclude from clinical observations and animal experiments 
that the three main factors in the etiology of rickets are (1) an organic dietary 
factor concerned with the calcification of bone; (2) light; and (3) the amount and 
relative proportions of calcium and phosphorus in the diet. An extensive 







historical survey of the questions concerning the etiology of rickets has recently 

been made by Park ('23) and by Vaughan ('23). 

Human Rickets. — For the present, we are concerned merely with the more 

general effects of rickets upon the growth of the body as a whole, leaving a 

consideration of the effects upon the skeleton and other organs for later chapters. 

The occurrence of fetal rickets has often 
been claimed, but according to Wieland 
('10) it is very doubtful. Huenekens 
('17) found premature infants especially 
susceptible to rickets, and J. H. Hess 
('23) gives the characteristics in this class 
of cases. Rickets in the human species 
does not ordinarily appear before the 
latter half of the first year, and is most 
frequent during the rapid growth period 
of the first two years. It may occur at 
any later age, however, and in the adult 
I it is usually designated as osteomalacia. 
According to Ruffer ('21), rickets 
has existed in Egypt since 2,000 B.C., 
§ and probably much earlier. Deformity 
of the spine and legs (probably rachitic 
in origin) was described by Soranus as 
prevalent among Roman children of the 
first' century, A.D. In most civilized 
countries today rickets is still a very 
potent factor in stunting growth, produc- 
ing deformity and lowering resistance to 
infection (Findlay and Ferguson '18). 
Kissel ('97) noted a variable degree of 
rickets in 80 per cent of 2,530 children 
examined at Moscow ; and Schmorl ('09a) 
found evidence of either active or healed 
rickets in 345 (89.4 per cent) of 385 
children under 5 years of age autopsied at 
Dresden. Increased frequency of rickets 

Fig. 43. — Ventral view of a child, 8 years , . , . ., , , , • , 

old; died from rickets. Note the enlarged during and since the world war has been 

head, deformed limbs, distended abdomen reported by numerous observers. 

and deformed thorax (marked groove in the m-, , , ., , , 

costochrondrai region, with enlarged costo- The external appearance ot the body 

chondral joints, forming a "rickety rosary"), in rickets is shown in Fig. 43. The 

well-known characteristic features were 
recognized by Whistler (1645) an d Glisson (1650), and described by many 
later observers. They include an enlarged cranium, deformed thorax (with 
enlarged costochondral joints often forming a "rickety rosary"), distended 
abdomen ("pot belly"), enlarged wrist, knee and ankle joints, with variable 
curvature of the lower limbs. Most of the deformities are the mechanical result 


of the softening of the skeleton, as will be discussed later. Seibold ('27) described 
three typical stages in the development of the disorder. 

In more recent descriptions of human rickets, Engel ('20, '20a) found that in 
German children of 2-5 years, rickets has become increasingly prevalent since 
191 7. Growth is much retarded in severe cases. According to data cited from 
Baginsky by Wohlauer ('n), body length is not much affected by rickets in the 
first year, but retardation becomes progressively evident in the second and third 
years. The body, though dwarfed, may be well proportioned and nearly normal 
in form; but the musculature is scanty and there is marked deformity of the 
limbs and thorax in severe cases, resulting in a prodigious number of crippled 
dwarfs. Looser ('20) stated that late rickets (or osteomalacia) is characterized 
by a general inhibition of body growth, with retardation in the development of 
the sexual organs and secondary sex characters. Well marked deformities of 
the trunk and limbs are characteristic, as described by Jenner ('95), Comby 
('01), Vincent ('04), Wohlauer ('n) and others. 

In experimental rickets of animals, the reported effects on the growth of the 
body as a whole are somewhat contradictory. Thus in puppies E. Voit ('80), 
Miwa and Stoltzner ('98) and Quest ('06) found continued growth without 
emaciation in rickets caused by calcium-poor meat diet, and Lipschutz ('10) 
noted that the general growth is not much retarded in rickets caused by phos- 
phorus deficiency. This is confirmed by the more recent experiments of 
Mellanby ('21), with deficiency in vitamin A, etc. On the other hand, Sherman 
and Pappenheimer ('21) and McCollum, Simmonds, Kinney, Shipley and Park 
('22) found growth retarded or suppressed in young rats with experimental 
rickets. Jackson and Carleton ('23) noted that in such rats the weight appears 
normal for body length, but that loss in body weight may be masked by increase 
in intestinal contents. Nevertheless, in animals, as in man, rapid growth appears 
most favorable for the development of rickets (at least during the latent period). 
General malnutrition (total inanition) appears distinctly unfavorable to the 
development of rickets, so that starvation, like sunlight, may even serve as 
a preventive or healing factor (Sweet '21; Jundell '22; McCollum, Simmonds, 
Shipley and Park '22). 

Elliot, Crichton and Orr ('22) state that during rickets in pigs the growth 
rate appears to be retarded less in the head than in the rest of the body, so the 
head often appears unusually large in the later stages. 

Stoeltzner ('09) and Mellanby ('21) have pointed out that rickets presents an 
apparent exception to the "law of the minimum," since the absence of an essen- 
tial factor in the diet may result in distortion of skeletal growth, without sup- 
pression of growth in the body as a whole. 

Vitamin Deficiencies. — As early as 1881, Lunin found that mice are unable 
to live long on an apparently adequate synthetic diet of proteins, fats, carbohy- 
drates, salts and water. Since the addition of milk gave good results, he con- 
cluded that other (unidentified) substances indispensable for nutrition must 
be present in the milk. Similar experiments with better results were made by 
Rohmann ('03, '08, ' 16), probably because his artificial diets were not sufficiently 
purified. Hopkins ('06, '12) experimented upon rats with artificial purified 



diets, confirming and extending the results of Lunin. His curves of body weight 
(see Fig. 44) indicate clearly the presence in milk of elusive "accessory factors," 
of which minute quantities are essential for growth. Osborne and Mendel 
('n, '12) and McCollum and Davis ('13) similarly found that although mainten- 
ance of adult rats and growth in the young may be obtained for a time upon 
isolated and purified food substances, nutritive decline and failure inevitably 
follow, unless certain essential factors (designated as "vitamines" by Funk) 
are added to the diet. Billard ('22) noted a dropsical condition and subnormal 
growth in frog tadpoles on vitamin-free diet. 

It is impossible here to review the results of the numerous investigations 
upon this phase of nutrition, which especially during the past decade have clari- 





1 p— 

i J 

in m 

£ / 

O / 
c f 

-£■ / 

3 / ! 

^ / 


(Q f | 

¥ 1 

l Age in 




Fig. 44. — Chart showing the effects of vitamin deficiency upon growth in body weight of 
rats. The lower curve (up to the 18th day) shows the average retardation of growth (in 8 
male rats) upon a vitamin-free dietary; the upper curve, for 8 similar rats, with addition of 3 
c.c. of milk daily to the diet. On the 1 8th day, the milk was transferred from one set to the 
other, with marked effect upon the growth in each case. (Hopkins '12; Med. Res. Comm. 

fied the relations of partial inanition to growth and have placed the etiology of 
several "deficiency diseases " upon a fairly substantial basis. The present status 
of the vitamin question was discussed in the symposium by McCollum, Mendel, 
Sherman, Shipley, Holt and Hess ('22), and the literature is reviewed in detail 
by the report of the Medical Research Council (by Hopkins et al., iQio)and in 
the works by Langstein and Edelstein ('17), Stepp ('17), Aron ('20), McCarrison 
('21), Funk ('22), Ellis and Macleod ('22), Sherman and Smith ('22), McCollum 
('22), Mellanby ('22) and others. The following vitamins are now generally 
recognized: "fat-soluble A" or vitamin A, with which is usually closely asso- 
ciated the antirachitic factor or "fourth vitamin;" "water-soluble B" or 


vitamin B (" antineuritic ") ; and the antiscorbutic "water-soluble C " or vitamin 
C. A fifth, or "vitamin X," has recently been discovered by Evans. 

The chemical nature and the exact physiological functions of the vitamins 
are still uncertain. Hopkins concluded that they are catalytic, stimulative 
. agents, possibly affecting both external and internal secretions. Similar views 
have been advanced by various workers. Chick ('20) emphasizes the greater 
need for vitamins when metabolism is accelerated (by work, low temperature, 
growth, pregnancy, lactation, etc.), which is very suggestive. McCarrison 
('21) believes that in the absence or inadequacy of vitamins, there results a 
disturbance of metabolism, so that the other dietary constituents cannot be 
properly utilized in the various organs. Abderhalden ('22) concludes that the 
vitamins promote the oxidative processes in the cells of the body. 

As many investigators have observed, there are marked differences among 
species as to the effects of vitamin deficiency, which will be mentioned later. 
Thus Hess ('22) states that: 

"In relation to the antiscorbutic vitamin, man reacts as does the guinea 
pig; in respect to vitamin B, he reacts like the pigeon or fowl; and in respect 
to vitamin A (fat soluble factor), he resembles the rat. The rabbit, for reasons 
entirely unexplained, withstands deprivation of any vitamin with comparative 
impunity, and therefore is not employed in any biologic test for these factors. 
Furthermore, a diet which leads to a definite avitaminosis in one animal, leads 
to a quite different one in another animal. For example, a diet of polished rice 
brings about polyneuritis in the fowl or in the pigeon or in the rat, but induces 
scurvy in the guinea pig." Sugiura and Benedict ('23) have raised pigeons 
from hatching to maturity on diets deficient in both vitamins A and C. 

The effects of vitamin deficiency upon the various individual organs of the 
body will be considered in the appropriate later chapters, but some more general 
effects upon the body as a whole may now be considered. 

McCollum and Simmonds ('17, '18) observed that, since vitamins (A and B) 
are not synthesized in the body, they must be present in the diet or (after exhaus- 
tion of the amounts previously stored in the body) they will be absent from the 
milk of nursing rats, with consequent failure of growth in the young. It seems 
inappropriate to speak of any single "growth vitamin," since all appear (in 
most species) to be necessary for optimum growth. Funk, however, claims that 
the growth-promoting factor in vitamin B may be separated as a distinct 
"vitamin D" (cf. Funk and Paton, '22). Evans and Bishop ('22a, '23a, '23b) 
distinguish a "fertility conferring factor X," in the absence of which the pla- 
centae of rats are abnormal and the embryos invariably resorbed. 

Vitamin A. — We have previously noted that the earlier failures to obtain 
growth upon diets free from fats and lipoids were probably due chiefly to the 
elimination of the closely associated "fat-soluble A" (cf. Stepp '22). The 
term was introduced by McCollum and Davis ('15, '17), who had previously 
('13, '14) noted failure of continued growth in rats upon diets free from this 
factor. Osborne and Mendel ('13, '17a, '21) and Hess, McCann and Pappen- 
heimer ('21) obtained similar results in rats, and Mackay ('21) likewise found 
that kittens on a diet deficient in vitamin A cease growth and become emaciated. 



The effects of deficiency in vitamins A and B upon the growth curves of young 
rats are shown in Fig. 45. 

When young animals are placed on diets deficient in vitamin A, there is a 
variable latent period during which growth continues (though at a retarded 
rate), probably by the aid of the vitamin stored in the fats and lipoids of the 
body. Evans and Bishop ('22) found that the body storage of vitamin A in 
young rats is apparently exhausted in 4-100 days. Osborne and Mendel 
('21) hold that for growth of rats vitamin A is much less important than vitamin 
B; and especially for nutrition in adults (Mendel '20). Shipley, Park, 
McCollum and Simmonds ('21) found that in young rats on diets deficient in 
phosphorus and vitamin A, the arrested growth was resumed upon the addition 
of vitamin A (see also Fig. 45). Korenchevsky ('23) found that diets deficient 
in vitamin A, or calcium, or both, when fed to the male rat only, have no 


"•••••# , 

*1 month-* 




Fig. 45. — Chart showing growth curves of young rats fed on diet deficient in both vitamins 
A and B. The curves represented by dots (....) show the inability of the rats to grow when 
both vitamins, A and B, are absent from the diet. When A alone is added, e.g., in butter fat, 

(curves marked ), there is no improvement; but upon adding both A and B (curves marked 

), excellent growth follows. When vitamin B alone is added, there is sometimes 

slight growth (curves marked xxxx), probably due to unexhausted reserve stores of A, but 
growth failure eventually follows. (Drummond; Med. Res. Comm. '19.) 

apparent effect on the offspring. When such diets are fed to the pregnant 
female, still-births are more frequent. Even though the newborn may appear 
normal, they are unusually susceptible to rachitoid disorders later, especially 
when the deficient diet of the mother is continued during lactation. 

The relation of vitamin A to the cause of rickets was mentioned above in 
connection with mineral deficiencies. The characteristic ocular lesions (xeroph- 
thalmia) produced by deficiency of vitamin A will be discussed in Chapter XIII. 

Vitamin B. Polyneuritis and Beriberi. — The importance of vitamin B in the 
production of polyneuritis in animals (especially birds) and of human beriberi 
became evident through the work of Eijkman ('97) and numerous later investi- 
gators. For review of various other theories of beriberi, see Vedder ('13, '23) 
and Vaughan ('23). Nagayo ('23) claims that human beriberi, although it 
may involve a deficiency of vitamin B as one factor, is a disorder distinct from 
experimental polyneuritis and more closely resembles the infantile "Mehlnahr- 
schaden." The marked atrophy of the body in experimental polyneuritis 
does not occur in human beriberi. The failure of growth in young animals on 
diets deficient in vitamin B was demonstrated by Funk and Macallum ('15), 
Osborne and Mendel ('17a), Abderhalden ('19) and Shipley, McCollum and 


Simmonds ('21) in rats; by Drummond ('16) in chicks; and by Emmett and 
Allen ('19) and Emmett, Allen and Sturtevant ('20) in frog tadpoles (see 
also Fig. 45). 

Although Eijkman ('97) observed that in some cases there was no emaciation 
in the body of adult chickens, even at death from polyneuritis, this appears to be 
exceptional. Marked loss in body weight during polyneuritis in chicks and 
pigeons was noted by Tasawa ('15) and others. The loss in body weight on diets 
deficient in vitamin B is so marked (frequently 40 per cent) and constant that 
many investigators believe it is due to the associated deficient food-intake 
(incomplete total inanition). This view is shared by Karr ('20), Simonnet ('20), 
Lumiere ('20a, '20b), McCarrison ('21), Hoffman ('22), and others. For exam- 
ple," No varo ('20), in substantial agreement with Findley ('20), found that in 
pigeons on polished rice diet the body weight, temperature, heat loss and food- 
intake remain constant for 7-13 days. Then they decline in the following 
order: heat loss; food-intake; body weight; temperature. The body weight 
was found to decrease in polyneuritis even more rapidly than in fasting (total 

Hoffman ('22) found that polyneuritic pigeons on polished rice diet, modified 
so as to be deficient only in vitamin B, nevertheless lose weight as during total 
inanition. In both cases, recovery was made upon refeeding with adequate 
diet. Gotta ('23), however, modified the diet so as to produce polyneuritis with 
but slight loss in body weight. 

Vitamin C. Scorbutus. — The historical development of our knowledge of 
scurvy has recently been reviewed by Hess ('20) and Vaughan ('23). It was 
perhaps the first human disorder to be recognized as a definite deficiency disease, 
due to the lack of fresh vegetables or antiscorbitic fruit juices in the diet 
(Ronsseus, 1564; Kramer, 1720). More exact knowledge dates from the recent 
experimental production of scurvy in the guinea pig by Hoist and Frolich ('07, 
'12). The essential factor was designated as the "water-soluble C " vitamin by 
Drummond ('19). Emmett and Peacock ('22) find that the requirement of 
chicks for vitamin C is much less than for A and B. The rat even thrives upon 
diets devoid of vitamin C, possibly because it may be synthesized in the body of 
this animal (Parsons '20). 

As to the changes in body weight, Hoist and Frolich found, during experi- 
mental scurvy in young guinea pigs, usually stationary weight for 1-2 weeks, 
followed by rapid loss, averaging 30-40 per cent, rarely reaching 50-60 per 
cent. Dogs were found susceptible, but negative results were obtained on mice, 
rats and cats. Similar results were noted by more recent investigators. 
Findlay ('21b) found that rabbits deprived of vitamin C gradually lose weight 
and die without signs of scurvy, although the offspring born during this period 
may show hemorrhages in the joints and viscera. The weights of Bessesen 
('23) for scorbutic guinea pigs are shown in Table 12. For curves by Chick and 
Hume ('17) showing loss of body weight in the guinea pig during scurvy, see 
Fig. 46. Growth may continue during the first 2 or 3 weeks, but a later marked 
decline is constant. Cohen and Mendel ('18) likewise found that scorbutic 
symptoms (tender and swollen joints) often appear while the guinea pigs still 



have a good appetite and are growing rapidly, so that (general) inanition plays 
little or no part at this stage. During the last 10 days, however, the loss in 
body weight corresponds to that in starvation. Bessesen ('23) found an average 
loss of 16 per cent in beginning scurvy, and 37 per cent at death (Table 12). 
Recovery is possible, even in extreme stages, upon the addition of antiscorbutics 
to the diet (Fig. 46). 

Hess ('15, '16, '20, '23) has studied especially the effects of scurvy upon 
children. The body weight usually becomes stationary, and growth in length 


11 id 14 30 36 41 4b 54 

Fig. 46. — Chart showing growth curves of guinea pigs as affected by vitamin C. Curve A 
shows growth on typical scurvy diet of oats, bran and water (deficient in vitamin C). Curve B 
shows better growth for a time on oats and bran plus autoclaved milk, but eventual decline 
and death from scurvy. Curve C shows growth curve on diet of oats, bran and water, result- 
ing in scurvy; but cured by adding vitamin C (orange juice and autoclaved milk) to the diet. 
Curve D shows the growth on the scurvy diet (oats, bran and water) plus 5 c.c. of orange juice 
daily; autoclaved milk added on the 56th day. Curve E shows optimum growth on diet of 
oats, bran and cabbage leaves; and Curve F on diet of oats, bran, autoclaved milk and 3 c.c. 
of orange juice daily. (Med. Res. Comm. Report '19; from Chick & Hume '17.) 

is also greatly retarded (in contrast with the effects of chronic general inanition). 
The symptoms may be obscure in latent cases. Rapid improvement in growth 
is obtained by addition of vitamin C in the form of orange juice, a result also 
noted by Chick and Dalyell ('21) and others. 

Water Deficiency (Aqueous Inanition). — The period of toleration of thirst 
varies greatly in man, as noted by Rowntree ('22), the recorded duration ranging 


from 36-72 hours for travelers lost in a desert to 18 days in the case of Viterbi 
(an Italian political prisoner) on total inanition. As previously stated, the 
period of duration during inanition is greatly extended when water is available. 
Marriott ('23) states that during the development of anhydremia the loss in body 
weight is more rapid than that observed in any other condition, reaching 10-25 
per cent in one or two days. He also cites the observations by King and 
McGee as to the effects of desert thirst on man. 

Rosenfeld ('86) held that Oertel's obesity cure, which involves a restriction 
of the liquids in the diet, is dangerous on account of producing lesions in the 
kidneys, heart and nervous system. 

Among lower animals, Falck and Scheffer ('54) observed a duration of 4 
weeks in the dog on dry biscuit; while Pernice and Scagliosi ('95a) noted death 
after n days in a dog on dry bread, and after 8-10 days in young chicks on dry 
maize. Bowin ('80) found that both dogs and rabbits die in about 23 days 
on dry diet. Nothwang ('91) observed death from thirst in pigeons at an aver- 
age of 4}4 days. Kudo ('21) kept adult albino rats on dry diet 6-7 days, 
while on total inanition one survived n days. With variable amounts of 
liquid (milk) added to the diet, the duration period was correspondingly 

The loss in body weight observed during thirst is also variable, but is usu- 
ally marked, with great emaciation, as in total inanition. According to Loren- 
zen ('87) a relatively dry diet is very effective in reducing the amount of fat in 
man, a principle used in the "reducing" diet of Oertel and others. As pre- 
viously mentioned, Prochownick's diet (for reducing fetal size) is low in water 
content as well as in carbohydrates. 

Chossat ('43) noted a loss of 35 per cent in the body weight of frogs subjected 
to evaporation. The losses recorded before death on dry diet in other animals 
are as follows: Schuchardt ('47), 44 per cent in pigeons; Falck and Scheffer- 
('54), 20 per cent in a dog; Bowin ('80), 50 per cent in rabbits, somewhat less in 
dogs; Skoritschenko (JSt,), very irregular loss in rabbits; Pernice and Scagliosi 
('95a), 24 per cent in a dog, 34-41 per cent in young chicks; Maurel ('04, '04a), 
30 per cent in adult guinea pigs; Kudo ('21), 36 to 51 per cent average in adult 
albino rats (Table 9). 

The importance of water in growth has been emphasized by Davenport 
('97, '99). It is frequently stated that the water content of living organisms 
can be modified experimentally to only a very limited extent; but this is not 
supported by the experiments of Hall ('22), who subjected various animals to 
exsiccation in a dry chamber, without food (excepting the mice, which were 
fed dry corn and oats). Subsequent recovery was obtained in all cases by 
giving water. The periods of exsiccation and the losses in body weight and in 
water content are shown in the accompanying table (p. 116). 

In human infants, O. and W. Heubner ('10) stated that lack of water in the 
diet may cause inhibition of growth, at least in weight. Similarly, Meyer 
('13) found that upon a diet of concentrated "Eiweissmilch," the growth of 
healthy infants was retarded, with prompt recovery upon the addition of merely 
aqua destiUata. Meier ('21) noted that the lack of water in breast-fed infants 



may cause a disturbance resembling alimentary toxicosis, and recalls the related 
"dessication fever" and "inanition fever" of the newborn. Utheim ('22) 
has recently reviewed the evidence that the weight of the body fluctuates accord- 
ing to its water content; and that the latter is greatly influenced by the diet. 
The dehydration produced by diarrhea is well known. 

Effects of Exsiccation (Hall '22) 


Per cent exsiccated of 

Body weight 



Earthworm (Allobophora chloroticus) 

Leech (Placobdella parasitica) 

Meal worm (Tenebrio molitor) 

Newt (Amby stoma punclatum) 

Frog (Rana pipiens) 

Turtle (Chrysemys marginata) 

Chameleon (Anolis carolinensis) 

Horned toad {Phrynosoma cornutiim) 

Lizard (Sceloporus spinosits) 

Wood mouse (Peromyscus leucopus) 

Meadow mouse {Microtus pennsylvanicus) 
House mouse (Mus musculus) 


41 .0 
33 -i 

3°- 7 
24. 2 





105 min. 

450 min. 

1 ,084 hrs. 

116 hrs. 

33 hrs. 
288 hrs. 
186 hrs. 
119 hrs. 

86 hrs. 

77 hrs. 

68 hrs. 
270 hrs. 

Spiegler ('01) found that young puppies are very sensitive to a dietary water 
deficit, which causes inanition and general retardation in growth. Kudo 
('21a) held albino rats one month old at constant body weight for several weeks 
by a restricted amount of liquid (milk) in a diet otherwise adequate for growth. 
The rats show a progressive tolerance of thirst, so that less liquid milk is required 
daily for maintenance as the experiment proceeds. The tail becomes elongated, 
while the body length remains constant (thus differing from the results of Jack- 
son and Stewart by underfeeding). 

The results of Kudo's thirst experiments upon the various organ weights 
(see Tables 9 and 10) resemble somewhat those of underfeeding experiments or 
total inanition on the same species, though certain differences occur. The 
general resemblance may be due in part to insufficient food-intake during thirst, 
which has also been observed by Straub ('99) in dogs, and by Maurel (04, '04a) 
in guinea pigs. On the whole, we may conclude that with aqueous inanition 
(dry diet) the length of life and loss in body weight usually do not differ much 
from those previously noted during total inanition (cf. Table 1). 

Although oxygen is not strictly a food-stuff, the effects of its deficiency are 
of interest for comparison. J. Loeb ('96) observed that the resistance of develop- 
ing ova to lack of oxygen varies greatly in different species. Eggs of the fish 
Ctenolabrus are injured almost immediately when deprived of oxygen, and the 
segmenting cells tend to fuse into a syncytial mass. The eggs of Fimdulus, 
on the contrary, are very resistant to lack of oxygen. 




The present chapter includes the effects of inanition upon the skin and appen- 
dages, including the mammary gland. In connection with the tela subcutanea, 
the effects upon adipose tissue in general are also noted. 

The effects of inanition on the skin and appendages are of interest in der- 
matology, and especially in relation to the diagnosis of the various deficiency 
diseases (pellagra, malnutritional edema, scurvy, etc.). Atrophy of the skin is 
likewise characteristic in many other disorders involving general malnutrition. 
The effects on the mammary gland are of obvious importance in pediatrics. 
After a summary of the more important effects the changes will be considered in 
detail under {A) total inanition, or on water alone, and (B) partial inanition. 

Summary of Effects on the Skin and Appendages 

These effects will be summarized for both total inanition and partial inanition 
including the deficiency disorders. 

Changes in Weight. — In adults, aside from the tela subcutanea, the loss in 
weight of the integument during inanition is usually relatively less than in the 
body as a whole, although nearly equal in the rat and frog. In the newborn rat, 
the skin may increase in weight while the body is held stationary; but the growth 
impulse soon decreases to a minimum, later increasing in experiments begun 
toward the adult stage. The loss in weight of the skin is promptly regained 
upon refeeding. 

The general appearance of the skin during total inanition is variable, thicken- 
ing and roughening of the hair coat being frequent. The epidermis becomes 
somewhat atrophic, but mitoses persist in reduced number in the deeper epi- 
thelial cells. In the corium, the pigment is generally reduced in amount. The 
healing of skin-wounds is slow and imperfect during inanition and hibernation. 
The tela subcutanea during inanition loses heavily in weight (up to 90 per 
cent or more), chiefly through atrophy of the adipose tissue. The rate of loss 
varies in. different regions, however, and in some cases fat may persist in con- 
siderable quantities, even at death from starvation. Flemming established 
three histological types of adipose atrophy; simple, serous and proliferative; 
to which may be added gelatinous or mucoid atrophy, occurring chiefly in adi- 
pose bone marrow. Hibernating animals subsist chiefly upon fat stored in the 
so-called hibernating gland. In general, the ordinary neutral fats of the body 
appear to be easily mobilized, while the lipoidal fats contain phospholipid 
which are relatively resistant to starvation. The atrophic adipose tissue is 
easily restored to normal upon adequate refeeding. 



The mammary gland undergoes marked atrophy during inanition in the 
adult female, and retarded development of the gland occurs in the malnourished 
young. Histologically the atrophy of the adult gland cells is confined to the 
cytoplasm, with prompt recovery upon adequate refeeding. During total 
(incomplete) inanition, lactation is persistent, although reduced in amount, with 
variable changes in the chemical composition of the milk. During partial 
inanition, the effects vary according to the type of deficiency. 

During partial inanition, the changes in the skin also are variable, according 
to the type of the deficiency. Protein deficiency occasions marked cutaneous 
disturbances, as, for example, in the edema so frequently characteristic of famine 
and similar chronic malnutiitional conditions. In pellagra (primarily due to 
protein deficiency), the skin lesions are very marked and characteristic, with 
variable inflammatory changes in the acute stage, and atrophic changes in the 
chronic stage. Iodin deficiency causes a myxedema, probably secondary to 
thyroid lesions. During rickets (due to calcium-phosphorus and vitamin defi- 
ciency), the skin is variably atrophic. In other vitamin deficiences, retarded 
growth or abnormal structure frequently occurs in the skin and appendages. 
Cutaneous hemorrhages, and occasionally edema, are found in scurvy. Aque- 
ous inanition (water deficiency) causes a cutaneous loss of weight somewhat 
similar to that during total inanition, accompanied by variable structural 

(^4) Effects of Total Inanition, or on Water Alone 

Changes in Weight of the Integument. — If the tela subcutanea is excluded, 
the loss in the weight of the skin proper, including hair, nails, etc., in adult 
animals during inanition is usually relatively less than that in the body as a 
whole. Thus in pigeons with average loss of 40.4 per cent in body weight, 
Chossat ('43) found a loss of only 33.3 per cent in the skin, not including the 
feathers (in which there was no loss in weight). Pfeiffer ('87) found but little 
loss in the weight of the skin in the rabbit, aside from subcutaneous fat. In 
3 rabbits with loss of 35-41 per cent in body weight, Weiske ('97) noted a loss 
only 22-25 P er cent m tne s kin. Voit ('94, '05, '05a) in the dog noted a loss of 
32 per cent in the body and of 20 per cent in the skin. In the fat-free skin, 
the loss is relatively less than in the fat-free body. Sedlmair ('99) in starved 
cats found the loss in weight of the skin relatively somewhat less than that of 
the body. In adult albino rats, Jackson ('15) observed that in acute inanition 
the body weight lost ^ per cent in weight, the skin 31 per cent; in chronic inani- 
tion, the body lost 36 per cent, the skin 39 per cent. 

The course of the loss in weight of the skin in the adult guinea pig at succes- 
sive stages of total inanition was noted by Lazareff ('95). With losses in body 
weight of 10, 20, 30 and 36 per cent, the corresponding average losses in weight 
of the skin were 1.97, 8.17, 12.71 and 17.94 per cent. Thus it appears that the 
loss is relatively slight in the earlier stages, becoming progressively greater but 
attaining a maximum percentage only half that in the body as a whole. On 
the other hand, in frogs (Rana pipiens) with progressive loss in body weight up 
to 5c or 60 per cent, Ott ('24) found relatively a still greater loss in weight of 


the integument in the male, although somewhat less in the female (Table 6). 
The percentage of dry substance remained nearly constant. 

In the young, the changes in the weight of the integument differ from those 
in the adult. Manassein ('68, '69) in rabbits on total inanition at age of about 
3 weeks obtained an average loss of about 35 per cent in body weight and of 40 
per cent in skin weight; at the age of 3 months, body loss was 33 per cent, skin 
19 per cent; in adults, body loss was 39 per cent, skin 27 per cent. Aron 
('10, '11) concluded that in underfed puppies the skin increases slightly in rela- 
tive weight. Trowbridge et al. ('18) likewise observed a slight increase in the 
weight of the integument of yearling steers held nearly at maintenance (con- 
stant body weight) by underfeeding; while in those losing about 17 per cent in 
body weight the hide lost only 3-6 per cent. 

In albino rats held at maintenance by underfeeding for various periods 
beginning at 3 weeks of age or later, Jackson ('15a) found a marked loss (36 per 
cent) in the weight of the integument in the rats beginning at 3 weeks of age 
(Table 4), with smaller losses in those beginning later. Stewart ('18, '19) in 
still younger rats observed the greatest loss (48 per cent) in those underfed 
from birth to 10 weeks; in those underfed from birth to 3 weeks there was no 
loss of weight in the integument; and in those held at birth weight by underfeed- 
ing 16 days there was an actual increase of 25 per cent in the weight of the integu- 
ment. This appeared to be normal developmental growth, and not merely an 
increase in weight due to edema. A smaller relative increase (10 per cent) was 
found by Barry ('20, '21) in the skin of newborn rats which had been stunted 
prenatally by severe underfeeding of the pregnant mother. 

These results (shown in Table 4) demonstrate clearly that the dystrophic 
growth of the integument varies greatly according to age. In the prenatal and 
newborn rats, the tendency to increase in weight of the skin during subnutrition 
appears greatest, but the growth impulse (m underfed animals) rapidly decreases 
to a minimum, the greatest losses occurring at body weights of 15-25 g. 
Later the growth impulse again increases, so that the losses become relatively 
less, up to the adult stage, where the loss is nearly proportional to that of the body. 
There are also differences according to the intensity of the underfeeding and 
probably according to the type of inanition (to be considered later). 

For the human species, data as to changes in the weight of the integument 
during inanition are scarce. Ohlmuller ('82) noted that in a normal infant of 
56 days, the body weight was 4,195.5 g., the skin weight 1,291.67 g., or 31.16 per 
cent; while in an atrophic infant of 56 days, the body weight was 2,381.2 g., the 
skin weight 290.55 g., or 12.21 per cent. It is evident that the tela subcutanea 
was here included with the skin proper, however, for Ohlmuller remarks that the 
enormous loss was due to the adipose content. A few data upon the weight of 
the integument, without tela subcutanea, as observed by me in autopsies of 
atrophic infants are given in Table 3. Lack of norms for comparison makes it 
difficult to draw any conclusions, however. 

Recovery in Weight of Skin upon Refeeding.— In albino rats held at 
maintenance by underfeeding from 3-12 weeks of age, Stewart ('16) 
found that upon full refeeding the integument rapidly recovers the marked 


loss in weight, reaching its normal proportionate weight within two weeks. 
In albino rats underfed from birth to 3 weeks (with retarded body weight of 
10 g.) or to 10 weeks (15 g.), and then fully refed to body weight of 25, 50 and 
75 g., Jackson and Stewart ('19) obtained nearly normal weight (or above) for 
the integument in all except one group refed to 25 g., which was sub- 
normal (Table 7). In rats similarly underfed for longer periods (from 3 
weeks to nearly a year of age), and then refed until permanent dwarfed size was 
reached, although the body weight was markedly below normal adult, the 
weight of the skin was but slightly below that normal for corresponding body 
weight (Table 8). 

General Appearance.— The integument varies in appearance during inani- 
tion. Tiedemann ('36) and Falck ('81) noted that in starvation the human skin 
is atrophic, lax, dry, shrunken and pale. A somewhat similar description for 
atrophic infants is given by Bourgeois ('55), Vincent ('04), Birk ('n), Lesage 
Cn), Nobecourt ('16), Nicolaeff ('23) and others. According to Thiercelin 
('04) and Rosenstern ('11) the skin of such infants may be at first eczematous, 
later pale or cyanotic. In the adult man studied by Meyer ('17) the skin was 
dry, rough, and discolored for several days before death. In this case there 
were no folds, excepting the volar and plantar regions. As a result of war 
famine, there is a notable increase in the occurrence of skin infections (Rubner 
'19, Richet and Mignard '19, Vandervelde and Cantineau '19, Ivanovsky, '23), 
probably due to lessened immunity. The frequent occurrence of edema during 
famine has already been mentioned, and this, of course, may alter greatly the 
appearance of the skin. 

The hairs in dogs during starvation remain firmly attached, according to 
Falck C75), or become easily detached, according to Bich ('95). An increased 
growth in thickness of the hair coat in underfed cattle was observed by Weiske 
('75) and Van Ewing and Wells ('14). Thickening and roughening of the hair 
coat were also found in underfed mice by Judson ('16) and Thompson and 
Mendel ('18); and in underfed horses by Moehl ('22). Trowbridge, Moulton 
and Haigh ('18) observed that poorly nourished steers shed their hair very 
late in the season. Irregularities in the hair coat also occur during various 
forms of partial inanition, to be mentioned later. Porter ('89) noted that absence 
of pigment is characteristic of the famine-stricken, and that in women the black 
hair may become yellowish, devoid of pigment. In the emaciated victims of the 
Russian famine, Ivanovsky ('23) stated that " the hair grew more slowly, fell out 
prematurely and tended to rapidly become gray. Growth of the nails on hands 
and feet was retarded, and the teeth readily decayed. The eyes became limpid 
as with aged people; the skin lost its elasticity and became wrinkled." 

Epidermis. — Cunningham ('80) found fatty degeneration of the deeper epi- 
dermal cells, and also atrophic changes in the dermis, in starved larvae of B-nfo 
melanosticus and Rana tigrina. Among the victims of the Indian famine, Por- 
ter ('89) observed that fatty degeneration of the cuticle may give rise to the 
"famine skin," a harsh, dry, patchy scurf also described by Donovan in the 
Irish famine of 1847. The vitality of the skin appears too low to throw off 
the dead epithelium. 


Rabl ('85) noted many mitoses in the epidermis of Salamandra atra starved 
5-7 months. In the epidermis cells (chiefly in the deeper layer) of both young 
and adult rabbits, Morpurgo ('88, '89, '89a) likewise found mitoses persistent 
even in extreme starvation, though reduced in number. Mitoses were 
also found in the peripheral portion of a sebaceous gland in an adult rabbit 
after death from starvation. In the epidermis cells of starved larvae of Triton 
taeniatus and 'Triton alpestris, Schultze ('88) observed irregular and lobulated 
nuclei, with reduced amount of chromatin. "Die Masse des Chromatins der 
Mitosen bei den Hungerlarven — ich sah immer noch sparliche — ist, wie ich in 
Uebereinstimmung mit den von Rabl fur Salamandra atra gemachten Angaben 
finde, dieselbe wie bei den gut genahrten Thieren." 

In a lizard starved 6 days, Konstantinovitsch ('03) found a thinning not only 
in the basal, Malpighian layer, but also in the layer of pigmented cells. In 
Diemyctylus viridescens starved for various periods, Morgulis ('n) described a 
marked atrophy of the epidermis, the decrease in the nucleus being much less 
than that in the cytoplasm of the cells. Persistent mitosis was also noted. 

Ruzicka ('17) found that in adult Triton cristatus and Triton taeniatus ecdy- 
sis during fasting occurs twice as frequently as in full-fed controls. Although 
fasting apparently causes an acceleration of metabolism (in agreement with 
Child's theory), it does not result in rejuvenation of skin structure. The skin 
in the fasting adults becomes thinner, but differs in structure from that in normal 
larvae; the nuclei become chromatin poor, Leydig cells are absent, and keratin- 
isation is accelerated (though absent in the larval skin, even during fasting). 
"Es sind also auch die Hungerreduktion als eine Alterserscheinung anzufassen; 
der absolute Hunger bewirkt schliesslich einen den kiinstlichen Seneszenz 
analogen Zustand." 

Corium. — Some of the observations above mentioned included the dermis as 
well as the epidermis. Samuel ('85) observed that in fasting pigeons growth 
continued four days at a retarded rate in the papillae which produce the large 
wing feathers. This retarded growth was correlated with a diminished blood 
supply to the papillae. Harms ('09a) described a degeneration produced in the 
digital glands of Rana fusca and Rana esculenta during inanition. There is no 
phagocytosis. The duct resists degeneration longest. Regeneration of the 
gland occurs upon refeeding. Tornier (07) noted a disappearance of pigment 
in the skin of underfed salamander tadpoles. Kammerer ('13) in similar experi- 
ments on Salamandra maculosa found that the richly-fed became yellow, the 
underfed variably spotted, chiefly black. He admits, however, that in all cases 
the pigment is reduced in amount by inanition, the black pigment of the chroma- 
tophores being more resistant than the yellow. Weber ('14) explains the 
" Hungermelanismus " of frogs as due, not to pigment formation, but merely to 
expansion of the chromatophores. The integument becomes green again upon 
refeeding. Ruzicka ('17) found the skin pigment greatly increased in adult 
fasting Triton cristatus and Triton taeniatus. 

Healing of Skin Wounds. — Collard de Martigny (1828) noted that skin 
wounds in animals during starvation heal poorly and form an imperfect cicatrix. 
He explained this as due to the anemia. The process was studied in detail by 


Chudnovski ('90) in skin wounds on starved rabbits. He found that the regen- 
eration of the epidermis cells occurs as in well-nourished animals, but more 
slowly. The number of mitotic figures is diminished and their form may be 
abnormal. The epithelial cell nuclei are poor in chromatin and may undergo 
chromatolysis. Granulation tissue is absent 01 diminished in amount, and the 
infiltration of multinucleated cells (found in wounds of well-nourished animals) 
is feeble or absent. Rous and McMaster ('24) state that in albino rats on 
water alone following laparotomy the abdominal wound heals slowly but by first 

According to Valentin ('58) in the hibernating marmot no appreciable 
amount of regeneration occurs in skin wounds. Hansemann ('98) states that no 
mitoses ordinarily occur in the skin of the hibernating marmot or hedgehog; 
but in a few hibernating hedgehogs, which were killed 1-14 days after incisions 
were made on the snout, immigration of leucocytes occurred, and cell-divisions 
appeared in both epithelium and connective tissue. 

Tela Subcutanea. — -The most conspicuous changes during inanition are not 
in the skin proper, but in the underlying tela subcutanea. This is because in 
well-nourished individuals it is very largely composed of adipose tissue, which is 
known to be rapidly depleted by inanition. The atrophy of this layer produces 
the characteristic looseness and folds of the skin during starvation. On account 
of the progressive character of this atrophy, the thickness of the skin folds may 
be used as a convenient index of the stage of inanition, especially in infants, 
according to Batkin ('15), Marfan ('21), Peiser ('21), Gerber ('21), Kading 
('22), Nicolaeff ('23) and Hille ('23). In a starved girl of 19 years, Schultzen 
('62, '63) noted that the adipose panniculus had disappeared from the trunk, 
but was still evident in the extremities. Likewise in infantile atrophy, the 
subcutaneous fat does not disappear at a uniform rate over the whole body. 
Marfan ('21) describes the panniculus adiposus as becoming thin and finally 
disappearing altogether in the various regions in the following manner: 

" Ce processus de destruction de la graisse commence par le ventre et finit 
par la face. II atteint successivement les regions suivantes: (1) l'abdomen; (2) 
le tronc (d'abord la poitrine, puis le dos, puis les lombes); (3) les membres 
(d'abord les superieurs, puis les inferieurs, puis les fesses); (4) la face (d'abord le 
front, puis les joues et le menton). On constate parfois quelques derogations 
a cet ordre, mais elles sont tres rares." 

The histological changes in the adipose tissue of the tela subcutanea 
of atrophic infants are shown in Figs. 47-49. 

In extreme stages of inanition, the ordinary adipose tissue of the tela subcu- 
tanea (and elsewhere) is usually almost completely depleted, so as to lose 90 
per cent or more in weight. Observations were made by Chossat ('43) in 
pigeons; Bidder and Schmidt ('52) in the cat; Falck ('75) in dogs; Ohlmuller 
('82), DeTommasi ('94), Klose ('13) and others in the human body. Neverthe- 
less, small areas of subcutaneous fat may be macroscopically visible, even after 
death from starvation, as noted by Falck ('75) in the dog; Meyer ('17) and others 
in man. In some cases the fat may persist in considerable amount (Fowler 
'70; Jewett'75; Voelkel '86; Ffartman '00), indicating thatdeath from starvation 



is not due to exhaustion of nutritive material. The persistence of the human 
"sucking pad" (corpus adiposum buccae) during general emaciation has been 
noted, not only during infancy (Ranke '84; Lehndorff '07; Scammon '19) but 

Fig. 47. — Normal adipose tissue, with large, polyhedral fat cells, from the tela subcutanea of 
a newborn infant. X400. (Parrot '77.) 

Fig. 48. — Adipose tissue from the tela sub- 
cutanea of an emaciated, athreptic infant. Adipose 
cells unequally atrophied; some still contain a 
moderate amount of fat in droplets of variable size. 
Cell nuclei and granular cytoplasm evident. X400. 
(Parrot '77.) 

Fig. 49. — Adipose tissue ' from the 
tela subcutanea of an extremely emaci- 
ated, athreptic infant. The atrophic 
adipose cells have been almost completely 
depleted of fat, and are closely packed so 
as to resemble the Malpighian layer of 
the epidermis. X400. (Parrot '77.) 

even in the adult (Gehewe '52). Beneke (05) stated that the cells of a lipoma 
differ from ordinary fat cells in that they do not atrophy in general emaciation 
of the body. In cases of malnutritional edema (due chiefly to partial inanition, 


however) the tela subcutanea may fail to show the usual decrease in weight, on 
account of the replacement of the fat by water. 

Adipose Tissue in General. — In considering the behavior of fat in the body 
during inanition, we must distinguish the ordinary fat of adipose tissue from the 
lipoidal fat which occurs especially in epithelial cells of glands, and elsewhere. 
As pointed out by Nikolaides ('99), Carini ('01), Traina ('04), and Dietrich 
('io), the ordinary fat is easily mobilized during inanition, hence called "wander- 
ing" or "usable" fat; while the lipoidal fat is usually "sessile" or "permanent" 
in character. The ordinary neutral fats appear to be readily resorbed, while 
the lipoidal fats contain lecithin and other phospholipins, which are more 
resistant to inanition. Aschoff ('09) described the relations of the various types 
of fat in various forms of fatty metamorphosis ("degeneration"). 

Bichat (1801, 181 2) long ago described the metamorphosis of adipose tissue, 
which assumes a gelatinous aspect and consistency in the epicardial region and 
and in the adipose bone marrow during phthisis and similar chronic diseases 
involving emaciation. 

Virchow ('59) added further histological details of the process: 

"Bei allgemeiner Abmagerung ist nichts gewohnlicher als dass das Fettge- 
webe unter dem Pericardium, im Nierenhilus sich wieder in deutliches Schleim- 
gewebe umbildet. Das Fett schwindet aus den Zellen, diese verkleinern sich, 
in die Zwischensubstanz tritt eine schliipfrige, gallertige Flussigkeit, welche die 
schonsten Mucin-Reactionen gibt." 

Gurlt (cited by Schwann, 1839) discovered " dass bei abgemagerten Personen 
die gewohnlichen Fettzellen mit Serum gefiillt sind." Similarly, Czajecwiz 
('66) described the process in young and adult rabbits as follows: 

"Bei Nahrungsentziehung wird der Fetttropfen in der Zelle resorbiert, 
seine Stelle grosstenteils durch eine sehr feinkornige Flussigkeit ersetzt; bei 
langerem Hungern schwindet dass Fett ganzlich und es bleiben die Formele- 
mente des Bindegewebes in Form von grossen, schonen, runden, mit seroser 
Flussigkeit gefullten und mit deutlicher Membran und mit einem oder mehreren 
Kernen versehenen Zellen zuriick." 

On refeeding, the fat cells were found resume their normal structure. 

Toldt ('70) described the fat cells in emaciated animals as decreasing in 
size and approaching the primitive " Protoblasten " in appearance. 

Flemming ('71, '71a, '76) studied the inanition atrophy of adipose tissue in 
young and old animals, including the fish, frog, rabbit, guinea pig, rat, cat, dog 
and man. He concluded that the " serumhaltige Fettzelle " of previous observers 
is not the final form of the atrophic fat cell, but merely an (inconstant) inter- 
mediate stage. This may explain the contradictory observations by previous 
investigators. Flemming described 3 types of fat cell atrophy as follows: 

1. Simple or Normal Atrophy. — The fat droplet gradually decreases, with 
formation of secondary droplets and granules, the cell finally reverting to the 
ordinary connective tissue cell of variable form. 

2. Serous Atrophy. — In this case the fat droplet is replaced by a serous drop- 
let or droplets, sometimes with small fatty granules also. Ultimately the serous 
content disappears, and the cells assume the same final form as in (1). 



3. Proliferative Atrophy. — In this case the nuclei of the fat cells proliferate, 
which may occur with either simple or serous atrophy. 

Flemming considered the simple atrophy more characteristic of chronic 
malnutrition and old age, serous atrophy arising in acute inanition and prolif- 
erative atrophy in either acute or chronic inanition. His classification, with 
more or less modification, has since been generally followed. Hammar ('95) 
has shown a variation according to the type of adipose tissue, to be considered 
later (under hibernation). Further details in the process of adipose atrophy 
are also given by Poljakoff ('88, '95), Metzner C90), Lindemann ('99), Pasini 
('03), Bell (cited by Waters '08), Schidachi ('08) (with extensive review of the 
literature), Monckeberg ('12), Matsuoka ('15), and Lubarsch ('21a). Metzner 

Fig. 50. — Normal bone marrow from the 
tibia of a young adult rabbit, showing the 
fibrous reticulum, large fat cells (containing 
fat droplets of variable size), and a few mar- 
row cells and erythrocytes. X500. (Jack- 
son '04.) 

Fig. 51. — Gelatinous marrow from the 
tibia of a starved rabbit. The fat cells 
have assumed a shrunken stellate form. The 
fat has disappeared, excepting a few small 
granules in two of the cells. The stroma 
presents a gelatinous, amorphous mass with 
lighter staining areas around each cell. The 
reticulum fibers are present in reduced num- 
ber, and the blood vessels are evident. X500. 
(Jackson '04.) 

('90) described fuchsinophile (Altmann) granules in atrophic fat cells, which 
probably correspond to the lipoidal fat granules mentioned by subsequent 
observers (cf. Cramer '20). 

Herter ('97) described the replacement of the subcutaneous fat and adipose 
bone marrow by a gelatinous substance in pigs during fat starvation. 

The mucoid (or "gelatinous") atrophy of Bichat and Virchow is indeed 
especially characteristic in the adipose bone marrow during inanition. Neumann 
('68), Feigel ('72), and Hoyer ('73) described in starved animals the transforma- 
tion of the marrow fat cells into a network of branching cells lying in a hyalin, 
mucin-containing ground substance. Fenger ('73), Ricklin ('79), Geelmuyden 
('86), and Helly ('06) described a similar involution of the adipose marrow in 
various human cachexias. Further details were added by Bizzozero ('69, 


'89, '89a), Bizzozero and Torre ('81) and Denys ('87) in birds, and by Herter 
('98) in the pig. According to L. Neumann ('82), this description applies only 
to chronic inanition; in acute inanition the cells do not become stellate. Geel- 
muyden ('86) found gelatinous marrow in the bones of frogs after the winter 
fast, and cited numerous observations on the same in human long bones in 
diseases with emaciation (cf. Dickson '08). A review of the literature, with 
description of the histological changes in the bone marrow of rabbits and pigeons 
during inanition and refeeding was given by Jackson ('04). He concluded that: 

"Beim hungernden Thier ensteht Gallertmark, indem das Fett verschwindet, 
und die Zellen ihre urspriingliche Reticulumform wieder annehmen. Die 
Reticulumfasern liegen dann meistentheils zwischen den Zellen in der reich- 
lichen gallertigen Grundsubstanz, thielweise aber auch innerhalb der Zellen oder 
unmittelbar neben ihnen" (cf. Figs. 50 and 51). 

During hibernation the changes in adipose tissue have received much atten- 
tion, since it has been shown that during hibernation the organism subsists 
chiefly at the expense of its stored fat. Valentin ('57) found that in a marmot 
hibernating 44 days, with loss of 8 per cent in body weight, the ordinary adi- 
pose tissue had lost 19 per cent and the "hibernating gland" (" Winterschlaf- 
driise") 27 per cent. In 3 other marmots at the end of hibernation (average 
166 days), with loss of 35 per cent in body weight, the ordinary adipose tissue 
had lost 99 per cent and the "hibernating gland" 68 per cent. Valentin ('58), 
Afanassiew ('77), Ehrmann ('83), Carlier ('03), Hansemann ('02) and others 
have shown that the so-called "hibernating gland" is composed essentially of 
adipose tissue, whose cells during hibernation undergo simple or serous atrophy 
similar to that found in adipose tissue in general during inanition. 

The subject of adipose involution during inanition was exhaustively investi- 
gated by Hammar ('95) who described in the albino rat two types of fat, white 
and brown. The "white fat" corresponds to the usual description of adipose 
tissue; during inanition it is reduced to remnants of fibrous connective tissue 
(never gelatinous) in the subcutaneous, subserous and intermuscular localities. 
The "brown fat," which corresponds in general to what has been described as the 
"hibernating gland" in many animals, has a different structure, each cell con- 
taining a spheroidal nucleus and numerous fat droplets of variable size. Dur- 
ing inanition the decrease in the weight of the "brown fat" is much less marked 
than that of the "white fat," and its structural involution is also different. 
It remains lobular in form, the color becoming dark reddish brown from blood- 
vascular congestion. The fat cells are reduced in size, and form a syncytium 
filling the intervascular spaces. The fat droplets are decreased in size and 
number, or completely absent. The cytoplasm usually becomes coarsely 
ganular, sometimes vacuolated, the nuclei unchanged. Apparently the cells 
may undergo either simple or serous atrophy. Hammar's results were in general 
confirmed by Auerbach ('02) for various species of rodents and insectivora, 
including hibernating and non-hibernating forms, and by Rasmussen ('22, '23) 
in the American woodchuck (Marmota monax rufescens). Changes in the 
hibernating gland of the woodchuck appear slower than in other species; but 
subsequent to hibernation and before food is plentiful it decreases rapidly 



with ultimate loss of about 75 per cent of the initial weight (body weight 
loss about 40 per cent). Although the fat cells of the gland decrease from 30/i 
to io/jl in average diameter, the spheroidal nucleus undergoes a slight increase 
in size. 

Mammary Gland. — The only quantitative data available concerning the 
effects of inanition upon the weight of the mammary gland during inanition 
are those of Manassein ('69) which are quoted in the accompanying table. 

Effects of Inanition on Weight of Mammary Glands in the Rabbit 
(from Manassein '69, Table XI). 



Length of 

Body weight 

Mammary glands 

Initial, grams 


Weight, Per cent 
grams : of body 

Normal controls 

1 1 mo. . . 
5 mo.... 

5 mo 

14^2 mo. 

5 mo. 
5 mo. 
5 mo. 

had borne 
had borne 


1,361 .6 















Adult had borne 1 

16 mo. 
16 mo. 
16 mo. 

had borne 
had borne 
had borne 

313 hrs. 

390 hrs. 
454 hrs. 
156 hrs. 
378 hrs. 
423 hrs. 
, 105 hrs. 












1. 81 




J -3° 







o. 29 


After total inanition 

14^ mo 

had borne 

460 hrs. ' 1 ,471 .3 




After inanition on water only 

Refed for 1-4 months after inanition 

14% mo 

had borne 

463 hrs. 

i, 5 i8.7(-43-2%) 




1 1 mo 

had borne 

411 hrs. 

i,5 I 7-9(~36.9%) 



1 .06 

4% mo 


266 hrs. 





3 hrs. 

post partum 

324 hrs. 




2. 11 

1 Nursed I rabbit during inanition period. 


From the data in this table it appears that there is marked variation in the 
normal weight of the mammary glands in the rabbit, the average being 0.76 
per cent of the body weight (range 0.29-1.29). After inanition, the range is 
equally great (0.186-1.30 per cent), but the average is only 0.58 per cent, 
indicating that the weights of the mammary gland have decreased relatively 
more than the body weights. Those refed show a prompt recovery in the weight 
of the mammary gland, with hypertrophy (as might be expected) in those 
becoming pregnant. Unfortunately it is uncertain to what extent the weight 
changes in the gland during inanition are due to loss in the associated fat, 
rather than to atrophy of the glandular parenchyma. 

In an adult female dog subjected to total inanition for 60 days (see Fig. 33), 
Falck ('75) noted that the mammary glands were entirely atrophied, the 
nipples elongated and inelastic. A marked reduction in the amount of milk 
secretion, with chemical changes including an increased fat content, were 
observed during starvation in sheep by Barbera ('00) and Barbera and Bicei 
('00a); likewise in goats by Lusk ('01, '17). The effects of underfeeding on 
milk production in cattle have been studied by Eckles and Palmer ('16) and 
Moehl ('22), who found that normal secretion may continue for a time in spite 
of a considerable degree of underfeeding. Ultimately there is a decreased flow, 
with variable changes in chemical composition. Data for human milk secretion, 
and for changes during partial inanition, are mentioned later. 

The histological changes during total inanition in the active mammary 
gland of 11 rabbits and 13 guinea pigs were investigated by Meynier ('06, 
'08). He found that in spite of the marked atrophy of the gland, secretion 
might still continue, within certain limits, sufficient to maintain the nursing 
young alive. The atrophy of the gland cells is confined almost entirely to the 
cytoplasm, and in prolonged inanition there is a marked fatty infiltration. On 
proper refeeding, this fat decreases in amount, and the gland tends to resume its 
normal functional structure. Cell division was not observed in the epithelial 
cells during either inanition or refeeding. 

The effects of inanition upon the mammary gland in young albino rats 
underfed for various periods were studied by Myers ('19), who found that the 
development of the mammary glands is thereby retarded, roughly in proportion 
to the retardation of body weight. When such stunted rats are refed, however, 
the development of the mammary glands for some time lags behind that in 
normal rats of corresponding body weight, although later they recover the 
normal condition. 

Human Mammary Gland. — The atrophy of the mammary gland in the 
adult human female during starvation has often been observed (Bright '77; 
Falck '81), the girl of 19 years described by Schultzen ('62, '63) apparently 
forming the only exception. In the autopsies during the Indian famine, Porter 
('89) states that the mammary glands "had shrivelled to such a degree in the 
emaciated women that their position was only ascertainable by the presence of 
the nipple. No gland could be seen or felt."' 

The accounts concerning the effects of war famine upon the milk secretion 
of nursing mothers are somewhat variable. Thus during the siege of Paris 


in 1870, according to Decaisne ('71), in some cases (12 out of 43 studied) young 
and vigorous women, in spite of malnutrition, were able to continue milk secre- 
tion sufficient to maintain their nursing infants. Similarly, during the recent 
war, Tschirch ('16), Steinhardt ('17) and Lande ('19) found the ability to nurse 
their offspring was not reduced in German women. Opitz ('18) and Momm 
('20) found the milk reduced in amount, but unchanged in chemical composition. 
As Lusk ('21) remarks: "All this confirms the biological principle of the sacrifice 
of the mother for the welfare of the offspring." This principle has its limita- 
tions, however, and in some localities the results were less favorable, doubtless 
depending upon the character and degree of the inanition. Ruge ('16), for 
example, found that malnutrition of the mother causes marked decrease in 
both quantity and quality of the milk secreted. Loenne ('18) likewise noted a 
decreased capacity for lactation, which, according to Klitting ('21) resulted in a 
greater postnatal loss of weight in the newborn. Briining ('18) ascribed the pro- 
longation of the time required to recover the postnatal loss of weight to a deter- 
ioration in the quality, rather than in the quantity, of the maternal milk. In 16 
of Decaisne's cases (above mentioned) there was practically no milk secreted, 
and three-fourths of the infants died of inanition. 

It is well known that an extreme degree of atrophy of the mammary glands is 
also commonly found in various diseases involving marked emaciation, such as 
tuberculosis and cancer, as well as in certain types of partial inanition to be 
mentioned later. 

(B) Changes in the Integument during Partial Inanition 

Under this heading will be considered the structural changes in the integu- 
ment upon diets deficient in protein (including malnutritional edema and pel- 
lagra), salts (including rickets,) vitamins (including scurvy), and water. 

Protein Deficiency. — Osborne and Mendel ('n) noted certain changes in 
the hair coat of young albino rats whose growth was retarded by diets incomplete 
in proteins. Similar changes were observed by Wheeler (/13) in young mice on 
inadequate protein diets. Evvard ('12) found that maize-fed pregnant sows 
give birth to many weakling pigs, with the skin lighter, anemic and deficient in 
hair development. He ascribed this to calcium deficiency; but experiments 
later (Evvard, Cox and Guernsey '14) indicated that the protein deficiency is 
the most important factor in the maize diet. (Osborne has shown that zein is 
deficient especially in tryptophan and lysin.) Skin lesions were observed also 
in rats on the maize diet by Abderhalden ('19), and by several investigators 
attempting to produce experimental pellagra in animals by maize diets (cited by 
Marie '08, '10). Osborne andMendel ('16a) observed that with corn-gluten food 
(deficient in trytophan and lysin) young chicks were stunted in growth and the 
body remained covered with down, as at the beginning of the experiment. 
However, a few feathers continued growth which had already begun. In 
another experiment ('16b), in a young rat held at maintenance by a diet lacking 
lysin, a patch of hair on the animal's back was dyed red at the beginning of the 
experiment. This color remained unchanged for 6 months; then lysin was 


added to the diet and the color soon disappeared, on account of the resumption 
of growth in the hairs as in other parts of the body. Drummond ('16), however, 
noted continued growth in beak and feathers of chickens held at constant body 
weight (100-150 g.) by rice diet (mixed deficiency) for periods of 20 — 80 days. 
Zuntz ('20) concluded that the growth of the epidermal structures (including 
hair and nails) may be restricted by lack of cystin, and he obtained increased 
growth of hair in sheep and man by addition of hydrolized horn to the diet. 
The literature on alopecia in animals on various artificial diets is reviewed by 
Bruning ('14a). 

Malnutritional Edema. — As already mentioned, cutaneous edema (with or 
without general anasarca and ascites) may be produced experimentally in rats 
by diets deficient in protein or fats, and similar deficiencies may be responsible 
for the edemic cachexia found in domestic animals and in conditions of human 
famine. Many authors, however, consider that "famine edema" and allied 
conditions are due to general inanition rather than to any specific deficiency. 
Cutaneous edema is usually associated with atrophy of the adipose tissue. 
Cramer ('23) found that in young rats on a diet deficient in tryptophan: " The 
hair falls out in patches after 6 or 7 weeks of this diet, and after another 8 or 10 
days there is an extensive oedematous condition extending over parts of the 
trunk and stretching the skin." This cutaneous myxedema was ascribed to 
hypothyroidism, due to extensive lesions in the thyroid gland. 

Cutaneous Lesions in Pellagra. — As previously mentioned, the exact etiology 
of pellagra is yet somewhat uncertain, although it seems probable that protein 
deficiency in the diet is at least an important factor. The cutaneous lesions are 
variable according to the stage and severity of the disease, and are sometimes 
absent ("pellagra sine pellagra"). While space does not permit the review in 
detail of numerous observations on the cutaneous pathology in pellagra, the 
more recent papers by Griffini ('70), Raymond ('89), Babes and Sion ('01), 
v. Veress ('06), Gurd ('n), Roberts ('12), Fiocco ('12), Raubitschek ('15) and 
MacNeal ('21) may be cited. It is of interest to note that in some preliminary 
therapeutic trials by Goldberger and Tanner ('22), the dermal lesions in 2 cases 
reacted favorably to cystin, and in a third case to cystin and tryptophan. 

In general, the human skin in pellagra shows an acute stage of erythema 
(similar to that in sunburn), followed by a chronic stage of atrophy and pig- 
mentation. The pigmentation usually affects the areas exposed to the sun, and 
appears in patches symmetrically disposed on both sides of the body. 

The histological changes in the acute (erythematous) stage usually involve 
inflammatory phenomena, with more or less edema in both epidermis and dermis. 
The epidermis undergoes proliferation and desquamation, with degeneration 
changes frequently forming blebs in the stratum germinativum (Malpighian 
layer). The dermis shows congestion, with marked fibrosis and pigmen- 
tation. Later the acute symptoms subside and a chronic condition of general 
atrophy in all layers of the skin gradually supervenes. This atrophy resembles 
that characteristic of old age, with hyalin degeneration and sclerosis in the 
blood vessels. Marked cutaneous edema may be present, but this "wet form" 
appears less frequently. 


According to Marie ('08 '10), there is found in pellagra a precocious tendency 
to alopecia, with sclerosis and pigmentation of the ectodermic structures. 

Deficiency in Salts. — Von Hoesslin ('82) gives some weights of the skin in 
dogs on nearly iron-free diet, but unfortunately normal controls are lacking. 
Roughening of the hair coat was observed in puppies by Quest ('06) with calcium- 
poor diet, and by Masslow ('13) with phosphorus-poor diet. Babcock ('05) 
found in cows on low salt rations during lactation no decrease in the milk yield 
for a considerable period of time, and a slight increase in fat content of the milk. 
Meigs ('22) reviewed the effects of partial inanition upon lactation, concluding 
that a marked deficit of calcium or phosphorus may not affect the amount of 
secretion for a long time, whereas a serious shortage of protein, fat or carbohy- 
drate causes an immediate reduction. 

In rickets the skin may be affected in children (Cheadle and Poynton '07). 
In rats subjected to experimental rickets, by deficiency in phosphorus or calcium 
and antirachitic vitamin, the skin becomes emaciated, and the hair coat fre- 
quently appears rough and uneven (McCollum, Simmonds, Shipley and Park 
'21; Shipley, Park, McCollum and Simmonds '21; McCollum, Simmonds, 
Kinney, Shipley and Park '22). Jackson and Carleton ('23) found a markedly 
subnormal weight of the integument in rachitic rats, even when the body weight 
appeared normal (Table n). 

Smith ('17) and Hart and Steenbock ('18a) found that a deficiency of iodin 
in the diet of pregnant sows results in the birth of weak, hairless pigs, with 
edematous skin and undeveloped hoofs. This condition, which is ascribed to 
malfunction of the enlarged fetal thyroid gland, may also occur in sheep, and 
occasionally in cattle and horses. 

Vitamin Deficiency. — Funk and Macallum ('14) found the feathers and beak 
apparently normal in a chick whose growth was suppressed for several months 
by a diet of unpolished rice and cod liver oil. In chicks and pigeons on polished 
rice diet with experimental polyneuritis, Tasawa ('15) found the skin atrophic, 
thin and dry; the fat in the subcutaneous tissue and elsewhere greatly reduced in 
amount. In the so-called "wet" beriberi of man (due to deficiency of vitamin 
B) edema and anemia of the skin are characteristic. 

According to Emmett and Allen ('20) and Funk ('22), lack of vitamin A in 
the diet of young rats produces a coarse and sparse hair coat. Cramer ('20) 
claims that if rats are deprived of vitamins for 25 days, the lipoids disappear 
from the "brown fat" (as well as from the suprarenal cortex); hence he proposes 
to call this type of adipose tissue the "lipoid gland" or "cholesterin gland." 

The experiments of McCollum and Simmonds ('18) indicate that vitamins 
(A and B) are not synthesized in the mammary gland by the nursing mother 
rat; therefore if vitamins are absent from the diet they are lacking also in 
the milk, and the young fail to grow properly. 

Scurvy. — The skin undergoes characteristic changes in scurvy, which is 
caused by a deficiency in vitamin C. Hoist and Frolich ('07, '12) discovered 
that scurvy can be produced in young guinea pigs by a diet of oats or other 
cereals with water. The characteristic cutaneous hemorrhages occur; likewise 
subcutaneous edema, but only occasionally. Petechiae in the follicles of the 


vibrissae appear constant, however. Bessesen ('23) found the integument 
usually subnormal in weight (Table 12). 

The extensive literature on human scurvy has recently been reviewed by 
Hess ('20) who notes that the skin is usually pale or livid and dotted with 
numerous petechiae. These are variable in size and most frequently located on 
the lower extremities. There may also be larger superficial hemorrhages, with 
color varying from reddish in the more recent to blue, green or brown in the 
older lesions. Emaciation and edema may occur, the latter most frequently 
localized in the regions of the ankles or eyes. According to Sato and Nambu 
('08) and Comrie ('20), cutaneous hemorrhages, often associated with sclerosis 
and marked edema, occur especially in the lower extremities. 

That the regions of the sweat glands and especially of the hair follicles are 
particularly susceptible to the petechial hemorrhages was observed by Lasegue 
and Legroux ('71), and confirmed by Aschoff and Koch ('19), Bierich ('19) and 
Wiltshire ('19). Wiltshire also described a peculiar "hyperkeratosis" of the 
affected follicles, each presenting a hard swelling, of pin-head size, due to the 
accumulation of epithelial debris at the mouth of the follicle. The hairs become 
atrophic, and may be broken off and regenerated. A similar hyperkeratosis is 
said to occur in other malnutritional states. 

Water Deficiency. — Tiedemann ('36) stated that in man during extreme 
thirst the skin becomes dry and hot. Schuchardt ('47) found no apparent loss 
in the skin of pigeons with loss of 44 per cent in body weight on a dry barley diet. 
Scheffer ('52) and Falck and Scheffer ('54) in a dog which lost 20.7 per cent in 
body weight during 28 days on dry biscuit, noted an apparent loss of 28 per cent 
in the weight of the skin. Bowin ('80) observed that in dogs on a dry diet the 
skin becomes roughened and the hairs are easily detached. Pernice and Scag- 
liosi ('95a), in chicks fed dry maize only, noted that the comb, at first red, 
becomes pale and later cyanotic. At autopsy the skin is dark reddish or nearly 
black (from passive congestion), and the fat has almost disappeared. As shown 
in Table 9, Kudo ('21) found that in adult albino rats on dry diets (acute or 
chronic thirst), the relative loss in the weight of the integument is slightly less 
than that of the whole body, as found in total inanition. As shown in Table 10, 
Kudo ('21a) noted that in young albino rats (about 4 weeks old) which were 
held at nearly constant body weight by a dry diet for various periods (1-13 
weeks), the integument usually loses slightly in weight (7.5-14) per cent. This 
slight loss appears early and is not progressive in the longer experiments. The 
skin becomes somewhat roughened, but the hair is not easily detached (as it is 
in adult rats). Dryness and desquamation were observed on the plantar sur- 
faces. The claws become much elongated, especially in the longer test periods. 

According to Tobler and Bessau ('14) the loss of turgor in the skin of mal- 
nourished infants with diarrhea is due to the withdrawal of water from the skin. 
Marriott ('23) states that during anhydremia the skin becomes gray, wrinkled 
and dry, and loses its elasticity. There is marked stagnation in the peripheral 



It has long been recognized that the skeleton loses little or nothing during 
general inanition, the wasting of other parts of the emaciated body giving rise to 
the characteristic appearance expressed by the phrase "reduced to a skeleton" 
(see Fig. 33). During certain types of partial inanition (rickets and scurvy), 
however, the skeleton is markedly affected. After a brief summary, the effects 
of inanition upon the skeleton will be considered under (A) total inanition, or on 
water only, and (B) partial inanition. 

Summary of Effects on the Skeleton 

During total inanition (or on water alone) there is but slight loss in the weight 
of the skeleton in adults, since the solids lost are largely replaced by water. 
In the young, during incomplete inanition there is even a persistent skeletal 
growth, variable in amount according to circumstances. 

Histologically there is but little change in the adult bony tissue during total 
inanition, aside from resorption tending to osteoporosis in chronic conditions. 
Degenerative changes may occur in the cartilage cells, and especially in the adi- 
pose bone marrow, which undergoes mucoid atrophy (as described in Chapter 
VI). There is also an atrophy of the red marrow, with degenerative changes in 
the hemopoietic cells. Bone fractures heal poorly during inanition; but regen- 
eration and restitution of normal skeletal structure occur upon appropriate 
refeeding, if the preceding inanition is not too prolonged or severe. 

Protein Deficiency, including pellagra, tends to produce in adults a skeletal 
condition of osteoporosis, somewhat similar to that in chronic total inanition. 
In young animals on incomplete protein diets there appears to be less tendency 
to persistent dystrophic skeletal growth, but more perfect recovery upon refeed- 
ing after long periods of suppressed growth. 

Calcium and Phosphorus. — The skeleton serves as a storehouse of reserve 
mineral supply, and is especially sensitive to a dietary deficiency in salts of 
calcium or phosphorus. A deficiency in calcium or phosphorus alone apparently 
produces skeletal softening grossly resembling rickets, but histologically 
presenting a "pseudorachitic osteoporosis" of variable extent and character. 

True rickets apparently involves an accessory factor ("fourth vitamin") 
and also a relative deficiency in calcium or phosphorus. The fresh weight of 
the skeleton is not much affected, but there is marked softening (with conse- 
quent deformities) and the dry weight is greatly reduced. Histologically 
rickets is characterized, in the region of enchondral ossification, by an increased 
width and irregularity of the proliferative cartilage zone, a failure of the pro- 



visional calcification, and an intensified vascular invasion from the marrow 
and perichondrium. A characteristic, irregular "metaphysis" zone is thereby- 
produced, composed of excessive osteoid substance (uncalcified bone), which 
replaces the cartilage, apparently by metaplasia. Subperiosteal osteoid is 
also formed. The bone marrow becomes atrophic and variably fibroid in 
character. Cessation of osteogenesis, together with continued absorption 
in the spongiosa and cortical bone, may produce a variable degree of osteo- 
porosis. Recovery is possible upon appropriate diet, but severe rickets causes 
permanent dwarfing and deformity. In late or adult rickets (osteomalacia) 
there is much decalcification of bone (halisteresis), and the enchondral ossi- 
fication changes are absent. 

Scurvy, caused by lack of vitamin C, is characterized by a general hemor- 
rhagic tendency, inhibited osteogenesis and marked atrophy of the osseous tissue. 
The skeleton is fragile and brittle, with a histological structure quite different 
from that in rickets. In the young (infantile scurvy or Barlow's disease), 
the proliferative cartilage zone is widened, as in rickets, but the provisional 
calcification takes place. The normal invasion and replacement of this carti- 
lage by osteogenic marrow tissue fails to occur, however, and the accumulation 
of calcified trabeculae forms a wide, weakened zone, which easily fractures, 
causing hemorrhages and an irregular swollen area ("Triimmerfeld"). The 
adjacent marrow undergoes a fibroid degeneration, related to that in rickets and 
other forms of inanition, but with characteristic multiple hemorrhages. Atrophy 
of the osteogenic tissue, together with continued absorption in both spongiosa 
and cortex, results in osteoporosis which is more constant and pronounced than 
in rickets. Except in the most extreme stages, recovery is possible upon appro- 
priate diet. 

Fetal scurvy is produced when pregnant guinea pigs are placed upon scor- 
butic diet, but the occurrence of fetal rickets is doubtful. There are marked 
differences among species in the susceptibility to scurvy and rickets, as found 
also in other dietary deficiencies. 

Aqueous inanition, with dry diets, occasions skeletal changes apparently 
similar to those found in total inanition. This is probably due in part to the 
lessened food-intake on dry diets. 

(A) Effects of Total Inanition, or on Water Only 

Weight Changes in Adults. — Actual weights of the skeleton in mammals 
subjected to various degrees of inanition, in comparison with those of normal 
controls, indicate little or no loss, according to observations on pigeons by 
Chossat ('43) and Lukjanow ('89); on the cat by Bidder and Schmidt ('52); 
on dogs by Falck ('54, '75) and Schondorff ('97); on hibernating marmots by 
Valentin ('57); on various mammals by Bourgeois ('70), C. Voit ('66, '94), 
and E. Voit ('05a); on rabbits by Gusmitta ('93), Pfeiffer ('87), Weiske ('97) 
and Sedlmair ('99); on the femur of guinea pigs by Lazareff ('95); on albino 
rats by Jackson ('15); and on man according to Rokitansky ('54) and Cohnheim 
('89). Where apparent loss in weight of the skeleton occurs, it is almost always 


less than 10 per cent, usually below 5 per cent, which is very slight in comparison 
with the corresponding loss in body weight. 

The nearly stationary weight of the skeleton during starvation, despite the 
loss of calcium phosphate and marrow fat, is possible because the substances 
lost are largely replaced by water, the specific gravity of which is higher than 
that of the fat (cf. Wellman '08, Jackson '15, Lusk '17). 

In the skeleton of amphibia, however, the destructive effect of starvation may 
be much greater. Harms ('09), Kammerer ('12) and Nussbaum ('14) found in 
fasting Triton and Proteus a marked shrinkage in the length of the vertebral 
column, but no weights of the skeleton were recorded. Ott ('24) found that 
during hibernation and subsequent inanition in the leopard frog (Rana pipiens), 
the ligamentous skeleton in general showed no loss in weight, but rather a 
slight increase (Table 6). In the group with loss of 60 per cent in body weight, 
however, there was an apparent decrease of 12 per cent in the weight of the 
skeleton. There was a marked and progressive decrease in dry substance. 

Weight Changes in the Young. — The earlier observations upon the skeleton 
in malnourished young children by Bouchaud ('64), Ohlmuller ('82) and Herter 
('08) indicated merely that the skeleton, though retarded in growth, fails to 
lose in weight and (as in adults) becomes relatively heavier in proportion to 
body weight. Later it became apparent, as already stated in Chapter IV, that 
in young individuals subjected to underfeeding, certain dystrophic growth 
changes may occur, chief among which is a persistent growth of the skeleton. 
This, was observed by Camerer ('05), Variot ('07a, etc.), Stolte ('13), Jackson 
('22), and others in the human infant; by Waters ('08), Falke ('10), Trowbridge, 
Moulton and Haigh ('18, '19) and Moulton, Trowbridge and Haigh ('21) in 
calves; by Aron ('10, 'n, '13) in puppies and rats; by Jackson ('15a), Jackson 
and Stewart ('18), Barry ('20, '21) and Stewart ('16a, '18, '19) in the albino rat 
(see p. 89; Table 4); by Thompson and Mendel ('18) in the mouse; and by 
Podhradsky ('23) in young fasting tadpoles of Rana jusca. Further data upon 
the weight of the ligamentous skeleton in atrophic infants are given in Table 3. 

That such dystrophic skeletal growth does not invariably occur in mal- 
nourished young, however, is indicated by the observations of Tschirwinsky 
('10) on lambs, and by Lascoux ('08), Freund ('09), Lesage ('11), Waser ('20) 
and others on atrophic children, as mentioned in Chapter IV. The skeletal 
growth is doubtless affected by the amount and character of the insufficient diet, 
and apparently varies also according to age (Birk 'n; Lust '13). The age 
factor appears most clearly in the work of Jackson, Stewart and Barry (Table 4), 
which indicates that in the albino rat the persistent growth tendency of the 
skeleton in undernourished rats is relatively strongest after the weaning period 
(3 weeks of age), being less in older as well as in younger stages; and not evident 
in the fetus. 

Structural Changes in the Adult Bone. — Various observers have noted that 
the bones tend to become weakened and brittle during conditions of prolonged 
inanition (cf. v. Recklinghausen '10), but in most cases it appears probable that 
these changes are primarily due to certain specific (especially mineral) defi- 
ciencies, which will be mentioned later. This applies especially to the various 


forms of "Hungerosteopathie" (osteoporosis and osteomalacia) frequently 
observed during periods of famine (by Chelmonski '21; Dalyell and Chick '21; 
Koepchen '19; Schlesinger '19; Richet and Mignard '19; Sauer '20; Simon '21; 
Szenes '21; Seeliger '23; Nicolaeff '23) and to similar skeletal disorders in 
malnourished animals (by Theiler et al., '12, '20; Gans '15; Hedinger '20). It 
is also possible, as will appear later, that some of the effects of inanition upon the 
skeleton may be indirect, through injury to some of the endocrine glands (para- 
thyroid, etc.) which are concerned in calcium metabolism. (Cf. Stefko '23a.) 

Gusmitta ('93) found apparent dilation of the Haversian canals and enlarge- 
ment of the lacunae in the bones of a starved dog; aside from this no data appear 
as to the effect of total inanition upon osseous tissue. 

Changes in Adult Marrow. — The changes in the adipose bone marrow during 
inanition were described in the preceding chapter, p. 125. It undergoes mucoid 
atrophy, with the absorption of the fat and metamorphosis of the fat cells, 
which return to their primitive stellate form and become embedded in an 
abundant gelatinous or mucoid intercellular substance. Hyperemia usually 
occurs. The reticulum fibers, which are observed in the adipose marrow, appear 
still quite distinctly in the atrophic mucoid marrow (Jackson '04) (Figs. 
50, 51). Kolliker ('89) described a transformation of yellow (adipose) marrow 
into red (lymphoid) marrow through inanition. The formation of gelatinous 
marrow in the long bones of various mammals was reviewed by Ricklin ('79) 
and Ackerknecht ('12). 

The red or lymphoid marrow may also undergo mucoid atrophy, depending 
upon the relative abundance of adipose cells present (Denys '87; Roger and 
Josue '00; Traina '04; Jackson '04; Dickson '08; Dantschakoff '09; Meyer '17; 
Jolly '20; Stefko '23). The various stages in the marrow changes during 
inanition were described by Solts ('94), in fasting dogs with loss of 13-52 per 
cent in body weight. There is hyperemia in the early stages, followed by mucoid 
degeneration in the middle stages. Some of the fat cells become stellate; some 
are destroyed. Other marrow cells, including the giant cells, undergo vacuolar 
degeneration and final necrobiosis. Capillary thromboses and hemorrhages 
may occur. Traina ('04) likewise found the bone marrow nearly normal in 
fasting rabbits up to about 10 per cent loss in body weight. At 10-20 per cent 
loss, the fat disappears from the adipose marrow cells (in the long bones, first 
at the ends). The fat may be replaced by serous fluid, with undiminished 
cell volume; or the fat cells may form a branched reticulum, enclosing marrow 
cells of normal or diminished size. Interstitial gelatinous substance appears, 
and the cells, including the megakaryocytes, undergo degenerative changes. 
At death from starvation, small fat droplets may still be found in the fat cells, 
also in the leukocytes and even free in the blood vessels. Altmann's granules 
become more numerous and distinct in the marrow cells. The observations of 
Opie ('04) on the eosinophiles in the bone marrow of fasting guinea pigs are 
stated in Chapter XV. The changes in the marrow cells during inanition have 
been reviewed by Helly ('06). 

Although the lymphoid tissue is generally found to undergo atrophy with 
disappearance of the lymphocytes in the bone marrow as elsewhere during 


inanition (Jolly '20), a proliferation of leukocytes was found by Sanfelice ('89) 
and of marrow cells in fasting rabbits by Roger and Josue ('00). Dantschakoff 
('09), however, described in the marrow of fasting birds (chick and duck) 
a general rarefaction of cells with progressive decrease in hemopoiesis. Foa 
('99) described the degenerative changes in the megakaryocytes of fasting 
rabbits. A decrease in the blood-forming elements in human red marrow dur- 
ing malnutrition was described by Dickson ('08). Lossen ('10) found that in 
the red bone marrow of fasting rabbits there is a marked decrease in the number 
of erythroblasts but a relative increase in the number of lymphocytes and 
myelocytes. Similar changes were observed in human cachexias. Ikeda ('22) 
observed in the bone marrow of fasting rabbits at first a transient proliferation 
of the myeloid and (slightly) the erythroblastic elements; later the process 
becomes normal, but in protracted inanition atrophy of the myeloid tissue 
occurs. Stefko ('23) concludes from an extensive study of material from 50 
necropsies that inanition stimulates the formation of myelocytes in the bone 
marrow, thereby affecting also the blood picture. 

Hibernation. — Some observations have also been made on the- bone marrow 
during hibernation. According to Pappenheim ('01), in hibernating spermo- 
philes, the red marrow of the ribs shows only slight changes; but the adipose 
marrow of the long bones undergoes atrophy proportional to the emaciation of 
the body. On awakening, this marrow also becomes red. Beretta ('02) 
noted frequent mitoses in the erythroblasts and leucocytes of the femur marrow 
in the hibernating hedgehog (Erinaceus europaeus). 

Cartilage. — The cartilage during inanition has been studied most frequently 
with reference to its fat content. Manassein ('68, '69) observed in the costal, 
articular and laryngeal cartilage cells of rabbits a variable number of highly 
refractive (fatty?) granules, which persisted during starvation. Similarly a 
persistence of the fat in the cartilage cells during inanition was found by Sacer- 
dotti ('98, '98-'99, '00) in starving rabbits; and by Bell ('09) in underfed cattle. 
Rabe ('10) noted an increase in the size of the fat droplets of the cartilage cells in 
the rabbit's ear during starvation, with a gradual decrease in glycogen content. 

Structural changes in the vertebral column of Triton taeniatus after several 
months of starvation were described by Harms ('09). Degenerative changes 
affect first the marrow, with complete disappearance of the fat, enlargement and 
disintegration of the marrow cell nuclei. Later the cartilage cells are similarly 
affected, with vacuolar degeneration, nuclear enlargement and karyorrhexis. 
Ultimately the intervertebral disks become completely fibrous. Podhradsky 
('23) noted ultimate resorption of the caudal chorda dorsalis in fasting tadpoles 
of Rana fusca. 

Degenerative changes were found also by Meyer ('17) in the cartilage cells 
of the epiglottis and trachea in a man who had starved to death. The greatest 
degeneration appears in the interior of the cartilage, although surface cells are 
also affected. The nuclei are pycnotic or absent; and often entire cells have 
disappeared, leaving empty lacunae. 

Skeletal Changes in the Young. — The dystrophic growth changes in under- 
nourished young children and animals, with dissociation of growth in height and 


weight due to persistent growth of the skeleton, have already been considered. 
Certain further details as to structural changes will now be mentioned. 

A thorough histological study of the zone of enchondral ossification (upper 
femur, etc.) in the full term fetus of rabbits subjected to total inanition for 2-8 
days was made by Diatschenko ('97, '99). No data as to the effects of the 
maternal inanition upon the fetal body are given, but marked and progres- 
sive changes were found in the zone of ossification. The cartilage showed no 
changes except in this zone, although some increase in the number of osteo- 
blasts beneath the neighboring periosteum was evident. The hyperplastic 
zone (columns of flattened cartilage cells) increases from 0.6 to 0.8 mm. in 
width. The hypertrophic zone becomes indistinct, but the atrophic zone (of 
regressive cartilage cells) increases from 0.12 to 0.23 mm. in thickness. The 
atrophic cartilage cells are markedly shrunken, in rounded, strongly calcified 
capsules, separated by thickened septa. 

The bony trabeculae in the adjacent marrow zone become slightly thickened 
and covered by osteoblasts 2 or 3 layers deep, instead of the normal single 
layer. The nuclei of the cartilage cells and the osteoblasts in the region stain 
feebly, giving a characteristic light zone in stained preparations, which increases 
progressively in width during the period of inanition. 

The newly formed bone laid down upon the trabeculae and walls of the 
medullary spaces may present a somewhat thickened osteoid (uncalcified) 
layer, but never the irregular osteoid masses characteristic of rickets. The 
progressive changes produced in the fetal ossification zone are designated by 
Diatschenko as " chondrodystrophia foetalis ex inanitione" and resemble some- 
what those of syphilitic osteochondritis. They differ from those in rickets 
especially in the increased width of the atrophic cartilage zone and in the inten- 
sity of calcification. 

In a radiographic study of the skeleton in the hands and feet of children 
retarded in growth by malnutrition, Variot ('05a, '06, '06a, 'oyd) found the 
centers of ossification appearing nearly normal according to the height. Thus 
the process of ossification is somewhat in advance of that in normal infants 
of the same body weight, but is behind the normal for corresponding age. A 
similiar condition was found by Jackson ('15a) in the skeleton of young albino 
rats held at maintenance by underfeeding from 3 weeks to 10 weeks of age. The 
persistent growth of the skeleton involves the normal appearance and union 
of the various epiphyses, corresponding to the size or weight of the skeleton. 
Unpublished observations upon sections of the lower end of the femur in these 
rats by Dr. F. P. Silvernale, in the Institute of Anatomy, University of Minne- 
sota, show an apparently normal process of ossification, with no indication of 
the dystrophic changes observed by Diatschenko ('97, '99) in the rabbit fetus. 

In the cranium of athreptic infants, a depression of the fontanelles with a 
tendency of the bones to overlap, so as to produce ridges at the sutural borders, 
was noted by Tardieu ^80), Thiercelin ('04) and others. This may be due to 
cranial overgrowth exceeding that of the brain, or to decreased intracranial 
fluid. Dystrophic structural changes in the cranium of athreptic infants, 
according to Lesage ('14), may involve a "soft atrophy," with retarded ossifi- 


cation of the fontanelles and sutures, similar to the craniotabes of syphilis or 
rickets. In "congenital spasmodic atrophy," on the contrary, Lesage and 
Cleret ('14) described the head as small and the cranium hard, an intense 
osseous hyperplasia sometimes filling the fontanelles and sutures with a dense 
bony tissue. Zuntz ('19) noted that the hardness of the skull is not modified 
in the fetus of rats during maternal starvation. Nicolaeff ('23) found osteo- 
porosis, especially of the vertebrae and cranial vault, in famine-stricken children. 
Recovery upon Refeeding. — As previously noted, a recovery of normal 
body weight is possible, if the inanition has not exceeded certain limits in 
length or severity. If the normal size as well as the weight of the body is to 
be recovered in the young, the subsequent growth of the skeleton is necessary. 
Aron ('11, '13) found that puppies and young rats underfed for long periods 
do not regain normal size upon full refeeding. This was confirmed by Jackson 
and Stewart ('18, '20) in underfed young rats. If the underfeeding period 
is not too severe or prolonged, however, and if not begun at too early an age, 
perfect recovery was found possible (Stewart '16). The weight of the 
skeleton (ligamentous, cartilaginous and dry) at various stages of recovery 
was studied by Jackson and Stewart ('19). 

In children refed after prolonged inanition, Goldstein ('22) observed that 
the increase in weight begins quickly while increase in length (skeletal growth) 
does not appear until after a stationary period. 

As to histological structure, Solts ('94) refed 2 dogs and 3 puppies after 
inanition, finding that the bone marrow regains normal structure. The mucoid 
substance gradually disappears and the necrotic cells are absorbed. Many 
mitoses occur in the marrow cells, and nucleated red cells become abundant. 
Jackson ('04) similarly found that the bone marrow in a pigeon, richly fed for 
3 weeks after a severe underfeeding period of 16 days, showed disappearance 
of the gelatinous bone marrow with complete recovery of normal structure. 
Bourgeois ('70) found that during inanition in animals the formation of a 
callus and the consolidation of fractures is difficult or impossible. The effect 
of inanition upon regeneration of bone was studied also by Trifiliev ('01) in 
fractures of the radius in starved and normal rabbits. The callus was found 
less voluminous and the process of repair delayed several days in the test 
rabbits in comparison with the controls. Hammer ('20) noted a similar effect 
as a result of human malnourishment during the war famine. "Wir haben 
namlich seit langerer Zeit beobachten konnen, dass einzelne Frakturen eine 
auffallend geringe Neigung zur Heilung zeigen die in einer verzogerten Kallus- 
bildung ihre unmittelbare Ursache hat, eine Beobachtung die auch anderwarts 
gemacht worden ist, worauf neben anderen (Eisler, Rupp, Tietzke) vor alien 
Melchior mehrfach hingewiesen hat." Hammer concludes that this inhibition 
of the regenerative capacity of the bone, which may exist in the absence of 
clinical symptoms, is not due to dietary deficiency of calcium or phosphorus, 
and is probably an indirect result of the effects of inanition upon the endocrine 
glands. A delay up to 8 or 10 months in the healing of fractures during the 
Russian famine was recorded by Abel ('23). 


(B) Effects of Partial Inanition 

The effects of partial inanition on the skeleton will include deficiencies of 
protein (with pellagra), of salts (calcium, phosphorus and rickets), of vitamins 
(A, B and C) and of water. The effects on skeletal growth were summarized 
by Jackson ('21). 

Protein Deficiency. — As already mentioned in Chapter V, Osborne and 
Mendel ('11, etc.) found that on various incomplete protein diets young albino 
rats remain unchanged in size for long periods, after which they are capable 
of complete recovery upon adequate refeeding. In both respects these results 
indicate that the skeleton reacts differently in protein deficiency in comparison 
with general underfeeding. Schulz ('12) found no increase of length in puppies 
fed farinaceous gruels (mixed deficiency), with full recovery upon later refeed- 
ing, excepting the very young puppies, which remained permanently stunted. 
Briining ('14, '14a) noted no persistent skeletal growth in young rats on one- 
sided carbohydrate diet with stationary body weight. Mendel and Judson 
('16), however, found persistent growth of the skeleton in mice on diets with 
protein or salt deficiency, as well as in simple underfeeding. 

Evvard, Cox and Guernsey ('14) noted defective bone development in the 
offspring of pregnant swine fed on maize (deficient in both protein and calcium). 
Osteoporosis and osteomalacia have often been observed in cases of human 
malnutrition, involving deficiency in protein as well as in calcium and phos- 
phorus (Alwens '19). On milled rice diet (deficient in protein, salts and 
vitamins), fragility of the bones and atrophy of the bone marrow were found 
in animals by McCarrison ('21) and others. Regenerative activity in the 
bone marrow of dogs on rice diet was noted by Brucco ('20). 

While the chemical data are too numerous to be mentioned here, it may be 
noted that Klose ('13), in an infant with alimentary edema (mixed deficiency), 
found that the skeleton shows a marked increase in water content, with decrease 
in fat, protein and ash. Similar data are given by Aron ('10, 'n, '13, '13a) 
for underfed puppies. 

According to Findlay and Mackenzie ('22), dietary deficiency of protein 
in rats causes patchy hemorrhagic areas in the bone marrow of the femur. 

In pellagra (assumed to be due chiefly to protein deficiency), fragilitas 
ossium was noted frequently, especially in the ribs, by Fraenkel ('69-' 70), 
Lombroso ('92), Tuczek ('93), Marie (08, '10), Raubitschek ('15) and Harris 
('19). Roberts ('12) presented X-ray figures showing rarefaction of the spon- 
giosa and cortical layer in the bones of the hands. 

In rats on lipoid -free ration with retarded growth of body, Hatai ('15) found 
the weight and length of the long bones normal for the body length, but the 
tail length slightly above normal. 

Salt Deficiency. — In human experiments with complete dietary deficiency 
of inorganic salts, Munk ('93) concluded from urinary analysis that there is 
a loss of calcium phosphate from the skeleton (cf. Forster '76; Wellman '08; 
Lusk '17). Lotsch ('12) described in cattle on salt-poor diet a skeletal disorder 
resembling human rickets and osteomalacia. Forbes ('19) has studied the 


relation of minerals to the growth and structure of the body, and especially 
of the skeleton. "The readiness with which minerals may be deposited in the 
bones, the lack of a definite upper limit of such a deposit, and the readiness 
with which these minerals may be withdrawn constitute the skeleton a true 
store of mineral nutriment." McClendon ('22b) states that in a normal white 
rat weighing about 70 g. the skeleton contains 82 per cent of the phosphorus 
and 96 per cent of the calcium in the body. 

In most of the earlier experiments, the mineral deficiencies were complicated 
by other (especially vitamin) deficiencies, so their interpretation is often 

Calcium Deficiency. — It is difficult to consider the effects of calcium 
deficiency aside from the question of rickets. Some of the more general effects 
of calcium deficiency upon the skeleton will be mentioned now, leaving those 
papers dealing more specifically with rickets for later consideration. 

Chossat ('42) observed that on diet of wheat (calcium-poor) and water, 
adult pigeons die before 10 months with a diarrhea and skeletal lesions ascribed 
to the calcium deficiency and prevented by addition of calcium carbonate to 
the diet. The bones, especially the femur and sternum, become very porous, 
softened and fragile, so as to fracture easily. Similar results were obtained 
by Friedleben ('60), Milne-Edwards ('6i) (with loss of ^ in weight of the 
skeleton), C. Voit ('81) and numerous more recent workers. E. Voit ('77) 
found that in pigeons on calcium-poor diet the bones more actively used in 
movements and support retain their calcium content longer, while the inactive 
(e.g., skull and sternum) become porous and thin. The negative results of 
Weiske ('71, '74) and Weiske and Wildt ('73) with goats, rabbits and lambs 
on calcium-poor diets were ascribed by later investigators to starvation (total 
inanition) since the animals refused the food. Numerous more recent workers 
have investigated the histological changes in the skeleton of various animals 
on different calcium-poor diets, and have described, in addition to the resorp- 
tive changes associated with osteoporosis, variable other lesions more or less 
closely related to those of rickets (to be mentioned later). 

According to Kellner ('16) fragility of the bones occurs also in the large 
domestic animals on calcium-poor diets. The relation between malnutritional 
and senile osteoporosis has been emphasized by Alwens ('19) and others. 
Rubner ('20a), however, states that the bone lesions occurring during the war 
famine were not curable by calcium alone. McClendon ('22a) found that low 
calcium diet produced osteoporosis in young rats, with marked reduction in 
the thickness of the wall in the shafts of the long bones. 

Dibbelt ('n) cites evidence indicating that there is a "physiological osteo- 
porosis" in the skeleton of the human infant and in the young of other animals. 
He ascribed this to the relative poverty of calcium in the milk, a condition which 
favors bone resorption. Wieland ('13) also reviewed the evidence for "physi- 
ological osteoporosis," but found no proof of a universal calcium-deficit in the 
nursling's milk. In the case of the human infant, it is difficult to exclude the 
possibility of latent rickets, scurvy or similar disorders due to dietary 


Albrecht ('13) found that in pregnant cows on calcium-poor diet the skeleton 
of the fetus develops normally, the necessary calcium being withdrawn from 
the maternal skeleton. Evvard, Cox and Guernsey ('14) noted Imperfect 
development of the skeleton in pigs from sows fed maize diet during pregnancy, 
but ascribed this to deficiency of protein rather than of calcium. 

Phosphorus Deficiency. — Not many experiments have been made with diets 
deficient in phosphates, aside from those to be mentioned later in connection 
with rickets. Weiske ('71) and Weiske and Wildt ('73) fed phosphorus-poor 
diets to adult goats and young lambs, but found no apparent effect on the skele- 
ton. Later investigators (Roloff, E. Voit) explained these negative results as 
due chiefly to the failure of the animals to eat the diet, with resultant general 
starvation. Hart, McCollum and Fuller ('09 /09a) found that low-phosphorus 
diet causes osteoporosis in pigs, but made no histological examination. 
Heubner ('10), Lipschutz ('10, '11a) and Schmorl ('13) fed puppies diets poor in 
phosphorus, and found the skeletal changes more like those of scurvy than of 
rickets. It is probable that these diets involved also deficiency in vitamin C 
(to be considered later). 

Rachitis. — Some of the gross deformities resulting from human rickets as 
described by Whistler (1645) an d Glisson (1650) were mentioned in Chapter V. 
Vincent ('04) states that the enlargement of the costochondral joints ("rickety 
rosary") is one of the earliest and most striking changes, found by Holt in 
142 out of 144 cases. Vincent claims this is absent in all other diseases, but it 
occurs also in scurvy. "Pigeon breast" is frequent, with lateral compression of 
the thorax, and a deep longitudinal groove on each side of the protuberant 
sternum and costal cartilages. A deep transverse furrow ("Harrison's sulcus") 
may occur at the lower end of the sternum, or there may be a funnel-shaped 
depression in the ensiform region. Spinal deformity is common — usually 
kyphosis, more rarely scoliosis or lordosis. Deformities of the clavicle, pelvis 
and limbs (especially the lower) may also occur, and the epiphyses, especially 
of the wrists and ankles, are enlarged. The skull is also especially affected, with 
arrested growth of the facial region but enlargement of the cranium (Figs. 43 
and 52). Regnault ('99) claimed that the enlargement of the rachitic cranium 
is apparent, rather than real. The forehead is square and projects forward, with 
frontal and parietal eminences thickened, vertex and occipital region flattened. 
Thin spots ("craniotabes" of Elsasser) may occur, perhaps due partly to pres- 
sure atrophy. The cranial sutures are usually open and the closure of the 
fontanelles delayed. The changes in the individual bones were described fully 
by Seibold (1827), Beylard ('52), Comby ('01), Jenner C95), and more recently 
with excellent illustrations (including X-ray) and review of the literature by 
Wohlauer ('n). Kauffman ('22) and Kauffman, Creekmur and Schultz ('23) 
claim that the changes in the temporal bones in otosclerosis may represent a 
form of late or adult rickets. 

These gross changes, which are chiefly the mechanical result of the softened 
skeleton, have recently been described for human rickets with further details 
by Cheadle and Poynton ('o7),Lesage ('n), Peckham ('20), Engel ('20), Maass 
C21), Schmidt ('21), Brusa ('21), Park and Howland ('21), Marfan ('22) and 



Pfaundler ('22). Some of these changes are shown in Figs. 43 and 52. Similar 
lesions have been described by numerous investigators in various animals, espe- 
cially puppies and rats, during experimental rickets. 

Although it is often stated that the growth of the skeleton is retarded during 
rickets, but few data are available upon the fresh weight of the skeleton to sup- 
port this view. Trousseau and Lasegue ('50) noted a dry weight of only 1 
kilogram in the skeleton of a rachitic dwarf of 8 years, the normal weight being 

Fig. 52. — The skeleton of a female dwarf, aged 42 years, 4 feet and 2 inches in height; 
showing extreme deformities as a result of rickets during early life. The lateral curvature 
of the humeri is probably due to the use of crutches. (Seibold, 1827.) 

7 or 8 kilograms. Friedleben ('60) confirmed Virchow's ('54) finding of con- 
tinued growth (especially subperiosteal) of bone during human rickets and con- 
cluded: "Von Bedeutung ist es, dass der Gesammtknochen betrachtlich massen- 
hafter und dicker erscheint, als der normale, was einestheils von der lockeren 
porosen Beschaffenheit der frischem Auflagerungsschichten, anderntheils von 
einer betrachtlichen Zunahme des Wassergehalts herriihrt." Especially in late 
rickets, however, the skeleton may share in the general retardation in growth of 
the body as a whole, as stated in Chapter V. Wohlauer ('n) stated that in 


comparison with the norms of Wilms and Sick, the centers of ossification in rachi- 
tic children usually appear normal in time, but are delayed in severe cases. 
Jenner ('95) emphasized the arrested growth of the bones in rickets. Lehnerdt 
('10) concluded that in infantile rickets the skeletal mass is usually normal or 
even increased. Voit ('80) and Dibbelt ('09) found the weight of the fresh 
bones and cartilages normal or above in puppies with rickets produced by cal- 
cium-poor diet (extracted meat plus fat). Jackson and Carleton ('23) likewise 
found nearly normal weights, in proportion to body weight, for both the liga- 
mentous and cartilaginous skeleton in rats with experimental rickets (Table 11). 
In nearly all cases, however, there was found a marked, but variable, decrease in 
the dry weight of the skeleton — a conclusion supported by a large number of 
observations in the literature on rickets in man and other mammals. Owing 
to this loss (chiefly of calcium phosphate) the dry content of the bones during 
rickets may decrease from about 65 per cent down to 30 per cent or less, depend- 
ing upon the stage and severity of the lesion. 

As previously mentioned, the experiments on diets deficient in calcium and 
in phosphorus have been made chiefly to determine the cause of rickets. As 
early as 1839, Guerin noted that the bones in puppies on a meat diet become 
softened and deformed, and he made the diagnosis of rickets produced by lack of 
calcium phosphate. Roloff ('66, '75) confirmed these results, finding softening 
of the long bones, scapulae, pelvic bones, vertebrae, etc. The epiphyseal joints 
become enlarged, forming a " rickety rosary" along the costochondral junctions. 
The bones become lighter (due to osteoporosis) and the thorax elongated dorso- 
ventrally. The condition develops most readily in young, rapidly growing 
puppies. Although no histological examination was made, it was diagnosed as 
rickets, due to lack of calcium. The addition of calcium salts to the diet pre- 
vented the disease, but non-calcium salts (including phosphates) did not. 
Friedleben ('60) recognized clearly that the osteoporosis produced in pigeons 
by mineral-poor diets differs fundamentally from true rickets. 

The first production of experimental rickets in which the histological struc- 
ture was carefully studied was apparently that of Wegner ('72). He found that 
the addition of small doses of phosphorus to the ordinary diet stimulates osteo- 
genesis in young animals (calf, rabbit, chicken). A salt-extracted grain diet in 
young chickens makes the bones soft, thin and brittle (confirming Chossat '42). 
The addition of phosphorus to this diet was found to produce rickets. 

"Aeussert interessant ist es, dass unter dem gleichzeitigen Einfluss der 
Phosphorfiitterung und der Entziehung anorganischer Substanzen, namentlich 
des Kalkes, der Wachstumsmodus der Knochen eine Aenderung erfahrt, die 
auf das vollkommenste dem entspricht, was wir beim Menschen als Rachitis 
zu bezeichnen gewohnt sind (Taf. 1, Fig. 34). Man sieht bei einem Ver- 
gleich mit dem daneben stehenden normalen Knochen an dem Beispiel 
gewahlten oberen Ende der Tibia eines jungen Huhnes die ausserordentliche 
hohe, von zahlreichen weiten Markraumen durchzogene gallertig, durch- 
scheinende gewucherte Knorpelmasse; in sie greift sehr unregelmassig in wellig- 
hiigeligen Linien ein die Zone der Kalkinfiltration, die tibrigens an sich sehr 
unvollkommen ist. An der Stelle, wo sich ausgebildete, weitmaschige spongiose 


Knochensubstanz bilden sollte, existirt ein ganz ungewohnlich dichtes osteoides 
Gewebe; die mikroskopische Untersuchung weist dann noch des Genaueren 
nach, wie alle diese Vorgange auf das bunteste durch einander gehen, kurz wir 
haben die Rachitis, wie sie im Buche steht." 

Wegner stated that his work confirms the theory that rickets is due to two 
factors: (1) insufficient supply of inorganic salts (due to defective intake or 
excessive excretion); and (2) a constitutional stimulus upon the osteogenic tissue. 

Lehmann ('78) and E. Voit ('77, '80) produced experimental rickets 
in puppies. Voit stated: "Ich will es unternehmen, jeden jungen Hund 
grosser Rasse in 3-4 Wochen durch Futterung mit kalkarmen Muskelfleisch 
und reinem Fett ohne Abmagerung hochgradig rachitisch zu machen." Voit 
('80) based his diagnosis upon the characteristic softening and deformity of the 
various bones, with broadening of the epiphyseal cartilage and marked 
histological irregularities in the process of ossification. 

While it was early recognized that rickets involves an abnormal widening of 
the epiphyseal cartilage and a disturbance of the normal process of enchondral 
ossification (Rufz '34; Guerin '39; Beylard '52; Friedleben '60) the exact 
histology of this disorder was somewhat neglected until comparatively recent 
times. Meyer ('49) described the structure of rachitic bone as similar to that 
decalcified by acids. A detailed and accurate histological description was given 
by Broca ('52), whose excellent work has usually been overlooked by later inves- 
tigators. He recognized that the "spongoid tissue" described by Guerin in 
human rickets is essentially an uncalcified osteoid layer, a transition between the 
epiphyseal cartilage and the calcified diaphyseal bone, normally present in small 
amount, but accumulating in large quantity when the ossification process is 
interrupted in rickets. 

The existence of osteoid substance (uncalcified bone) in rickets was also 
recognized by Friedleben ('60), Wegner ('72), Kassowitz ('82-'85) and others, 
but its significance was not fully appreciated until later. Pommer ('85) empha- 
sized the formation of excessive osteoid substance as a cardinal point in true 
rickets. This distinguishes it from osteoporosis, which is due merely to exces- 
sive absorption of already formed bone. Halisteresis, or decalcification of 
formed bone, is less frequent, but may occur (as was noted by Broca), especially 
in adults. The gross appearance and consistency of bone may be very similar 
in rickets and osteoporosis, and the two conditions may be coincident. 
Osteomalacia is now generally considered as a form of late rickets arising after 
growth has ceased, and hence not involving the phenomena in the zone of 
enchondral ossification (Pommer '85; Wild '01; Schmorl '05; Looser '05, '08, '09, 
'20; Schmidt '09; Ribbert '09; Stoeltzner '09; v. Recklinghausen '10; Boehme 
'19; Higier '22; Korenchevsky '22; Maxwell '23; et al.). This view was held 
long ago by Beylard ('52) and others. 

It is impracticable to mention here the large number of papers (chiefly 
German) dealing with the histology of rickets, which appeared in the pre-vitamin 
epoch, chiefly between 1885 and 1910. Detailed reviews of this literature will 
be found in the works of Rievel ('07), Schmorl ('09b), Lehnerdt ('10), and 
v. Recklinghausen ('10). The chemical phases of rickets are reviewed by 


Schabad ('10). The essential structural features which were established for 
rickets may be summarized briefly as follows (Schmidt '21, p. 204) : 

"Die anatomische Grundlage dieser Veranderungen ist sehr kompliziert. 
Sie besteht erstens im Auftreten osteoider Substanz in einer das normale Mass 
in Dicken- und Flachenausdehnung weit uberschreitender Menge, und oft in 
einer mit starker Hyperamie verbundenen, iibermassigen knochenbildenden 
Tatigkeit des Periostes und Endostes, welche zu Verdickung und Verdichtung 
der Knochen fdhrt; zweitens in einer Storung der enchondralen Ossifikation, 
namlich in iibermiissiger Proliferation, mangelnder Verkalkung und unegelmas- 
siger Vaskularisierung und Ossifikation des Knorpels." 

Associated with the endosteal proliferation, especially in the long bones, 
the marrow may be extensively replaced by a fibrous connective tissue. Pri- 
mary lesions in the marrow have been described by Marfan and Baudouin 
('09), Marfan, Baudouin and Feuille ('09), confirmed by Hutinel and Tixier 
('09). There is apparently an early stage of proliferation in the various types of 
marrow cells, followed later by their regression and replacement by fibrous 
marrow. Networks of osteoid trabeculae may be formed in this fibrous marrow, 
or may arise by decalcification of pre-existing bone (halisteresis). The decalci- 
fied, osteoid structures may later be resorbed through the activity of the osteo- 
clasts (Morpurgo '09), giving rise to a variable degree of osteoporosis. 

In the region of enchondral ossification, the proliferative zone of the epiphy- 
seal cartilage becomes abnormally wide. According to Pommer ('85), Schmorl 
('06) and Heubner ('06), this is due not to increased proliferation of the cartilage, 
but to lack of its removal as occurs in the normally succeeding stages in the 
ossification process. Provisional calcification fails to occur in the adjacent zone. 
Schmorl ('09) considers this defective calcification of the cartilage primarily 
responsible for the further irregularities in the process of ossification. The 
vascularization of the cartilage is excessive, with vessels coming not only from 
the marrow, but also (largely) from the adjacent perichondrium or periosteum. 
Kassowitz ('78, '82-'85, '12) has long maintained that this hyperemia is inflam- 
matory in character and of primary importance in rachitis. Owing to the irregu- 
lar invasion of the cartilage, the normally even plane of ossification is replaced 
by a wide irregular "spongoid zone" or "metaphysis," composed of cartilagi- 
nous masses of variable size, intermingled with osteoid substance (uncalcified 
bone) and vascular marrow. This occasions the widening of the zone between 
the epiphysis and diaphysis, as seen by the Roentgen-rays or in gross 

According to Strelzoff ('73), Kassowitz ('82-'85), Schmidt ('09), and 
Wohlauer ('n), the osteoid substance (at least in part) is formed by direct meta- 
plasia of the persistent cartilage. Although this metaplasia theory is doubted 
by Schmorl ('06, '09), it has recently been supported by McCollum and his 
co-workers. Ribbert ('13) described necrosis of the cartilage cells, which he 
considered evidence of a toxic agent in rickets. Simultaneous with the forma- 
tion of the osteoid metaphysis, osteoid substance is also deposited under the 
periosteum of the bone. The osteoblasts become surrounded by matrix, as 
normally, but the latter fails to become calcified. 


The various stages in the process of healing in the rachitic skeleton have been 
described by various authors, including Schmorl ('06, '09), Marfan and Baudouin 
('09), Wohlauer ('n) and Schmidt ('21) for man; by Mellanby ('21) in puppies; 
and by Pappenheimer ('22) for rats. Calcification appears in the metaphysis 
near the zone of proliferative cartilage, and the normal process of enchondral 
ossification ensues. Several layers may appear, indicating alternating stages of 
recrudescence and healing. The abnormal osteoid structures become calcified, 
but later may be largely resorbed. A correction of minor deformities is possible. 
Extensive deformities cannot be corrected, however, and permanent dwarfing 
frequently results. Entire destruction of the epiphyseal cartilage (a rare occur- 
rence) necessarily precludes the possibility of further growth in length of the long 

Whether the histological features above mentioned as characteristic for 
human rickets are to be found in the somewhat similar disorders experimentally 
produced in the lower animals is a question which has been much disputed. 
Mouriquand ('23) claims that in human rickets the hyperemia and proliferation 
of the bone marrow entail a decalcifying chondromyelitis which is not apparent 
in experimental rickets in rats. There is no doubt, as Heubner ('06) and others 
have shown, that many of the typical changes in the zone of enchondral ossifica- 
tion in rickets can be produced in puppies by feeding calcium-poor or phos- 
phorus-poor food. Hypertrophy of the proliferating cartilage, vascular invasion 
of the cartilage and absence of calcification in this zone can be thus produced. As 
to the crucial point, the production of osteoid tissue, but few investigators have 
been successful until quite recently. Absence of excess osteoid tissue, with 
increased absorption producing merely "pseudorachitic osteoporosis" were 
found in experiments with calcium-poor diets (chiefly in puppies fed horse-meat 
plus lard) by Korssakow ('92), Reimers andBoye ('05), Miwaand Stoeltzner 
('98), Stoeltzner ('99/08), Aron and Sebauer ('08), Dibbelt ('09), Gotting 
('09), and Schabad do). 

Stilling and von Mehring ('89) found that the puppies from a bitch fed 
calcium-poor diet (horseflesh plus fat) showed no bone changes, but the 
mother after continuing on the diet 126 days showed softening of the vertebral 
column and pelvic skeleton. Histological examination of the affected bones 
revealed active resorption, with bony trabeculae, covered by layers of osteoid, 
as in human puerperal osteomalacia. 

Dibbelt ('10) fed a calcium-poor diet (rice and horseflesh, plus sodium and 
potassium chloride) to an adult dog during pregnancy. A resected rib showed 
considerable decalcification (halisteresis) of bone, with many osteoclasts, 
Howship's lacunae, etc., indicating an active resorption comparable to that in 
puerperal osteomalacia. Her 6 puppies appeared normal at birth; but 2 of the 
puppies continued nursing the calcium-poor milk of the mother (still on the 
calcium-poor diet). Although they increased normally in body weight, they de- 
veloped marked skeletal lesions, so that they were scarcely able to crawl within 
a few weeks. One of the puppies was then fed horseflesh plus calcium 
phosphate, and improved rapidly. The other was fed horseflesh only and 
became worse. A resected rib (on the third day of the sixth week) showed 



retarded ossification and deficient bony tissue, but no osteoid substance. The 
proliferative zone of the cartilage was greatly increased in breadth. The 
primary marrow spaces were filled with vascular fibrous tissue and few marrow 
cells. Osteoclasts were present, but not abnormal in abundance. Korenchev- 
sky and Carr ('23) likewise found young rats much more susceptible to rickets 
when the mother was fed during pregnancy or lactation on diets deficient in 
calcium or vitamin A. 

Stoeltzner and Salge ('01) confirmed Wegner's ('72) experiments resulting 
in the production of osteoid tissue of puppies by adding phosphorus to the 
calcium-poor diet, but they still designated this as "pseudorachitische Osteopo- 
rose" rather than true rickets. 

Stoeltzner ('08), Lehnerdt ('09, '10) and Shipley, Park, McCollum, 
Simmonds and Kinney ('22) found that strontium cannot successfully replace 
calcium in preventing rachitoid lesions in puppies fed a calcium-poor diet. It 

Fig. 53. — From a photograph of a portion of a section of the upper extremity of the tibia 
in a normal young albino rat about i month old, body weight 36 g. C, epiphyseal cartilage; 
to the left of which is a vertical black band, representing calcified bone of the epiphysis. 
To the left of this, a small part of the epiphyseal marrow (M) is visible. To the right of the 
epiphyseal cartilage is a wide zone of enchondral ossification, with a network of ossified 
trabeculae, finer next to the cartilage, and becoming progressively coarser toward the marrow 
cavity (M') of the diaphysis. Prepared by von Kossa's silver method (calcified tissue black). 
X50. (Preparation by O. J. Morehead.) 

appears that strontium, like calcium and phosphorus, may stimulate the forma- 
tion of osteoid substance, which (in the absence or malassimilation of the salts 
necessary for calcification) may persist in excessive amounts. The identity of 
the lesions with those of human rickets is questionable, however. In genuine 
rickets, the osteoid tissue appears incapable of calcification, even in the presence 
of calcium salts, whereas in "pseudorickets" the osteoid tissue (if present) 
becomes calcified as soon as the necessary salts are supplied. 

Lipschutz ('10, 'n, 'na) observed that puppies on a phosphorus-poor diet 
develop bone lesions resembling those of scurvy, in connection with which they 
will be discussed later. 


As noted in the discussion of the etiology of rickets in Chapter V, the more 
recent work on experimental rickets has recognized a vitamin factor. Elliot, 
Crichton, and Orr ('22), however, produced rickets (with excess osteoid) in 
pigs on diets of oatmeal and bran, in spite of abundance of vitamins A, B and C, 
but preventable by the addition of calcium salts. Mellanby ('19, 21) has espe- 
cially emphasized the importance of vitamin A, or an allied antirachitic factor, 
though recognizing also other factors. By various deficient diets, he has pro- 
duced in puppies skeletal lesions which appear in all respects essentially identical 
with thosejof human rickets. Similar success has been obtained with experi- 
mental rickets in rats by Korenchevsky ('21, '22, '22a), Sherman and Pappen- 

p IG 24. — From a photograph of a portion of a section of the upper extremity of the tibia. 
Albino rat (McCl 14.2) had been placed on a phosphorus-poor diet (white flour, 93 per cent; 
spinach, 1 per cent; NaCl, 2 per cent; lime, 2 per cent; yeast 2 per cent) for 1 month, beginning 
at 3 weeks of age, resulting in severe rickets. Final body weight, 31 grams. Compare with 
Fig. 53, and note the great hypertrophy of the epiphyseal cartilage (C), to the left of which is a 
thin black band, representing the remnant of calcified bone layer of the epiphysis, surrounded 
by a light band of uncalcified osteoid tissue. (The epiphyseal marrow cavity appears black.) 
To the right of the epiphyseal cartilage, the zone of enchondral ossification is replaced by 
a wide irregular zone (Z), representing the "metaphysis," composed of osteoid tissue, invading 
marrow, and remnants of cartilage. The calcified bony trabeculae (shown in Fig. 53) have 
nearly disappeared, but some remnants (T) are still visible. Around and between the slender 
calcified trabeculae are relatively wide bands of uncalcified, osteoid tissue (0). The dark area 
below and to the right represents marrow of the diaphysis. Von Kossa's silver method (calcified 
tissue black). X50. (Preparation by O. J. Morehead.) 

heimer ('21), Pappenheimer, McCann, Tucker and Hess ('21), McCollum 
Simmonds, Shipley, and Park ('21, '21a, '22), McCollum, Simmonds, Parsons, 
Shipley and Park ('21), Shipley, Park, McCollum and Simmonds ('21, '21a, '22), 
McCollum, Simmonds, Kinney, Shipley and Park ('22), Park, Shipley, McCollum 
and Simmonds ('22), McCollum ('22), Shipley ('22), Hess ('22), Jobling, 
Pappenheimer and Hess, ('22), and Pappenheimer, McCann and Tucker ('22). 

The characteristic histological changes produced in the bones of young 
albino rats by experimental rickets are shown in Figs. 53 and 54. 

As previously noted, the work of McCollum and his co-workers strongly 
indicates that at least two factors are concerned in the production of rickets: 
(1) a "fourth vitamin" or organic factor which is closely associated with 
vitamin A and promotes calcium deposition; and (2) a dietary deficiency of 


either calcium or phosphorus. Thus there is apparently a low-calcium rickets 
and a low-phosphorus rickets, each of which may produce the essential lesions, 
including the osteoid substance. There are minor differences in the histological 
details, which vary much according to the stage and severity of the disease, but 
the low-phosphorus rickets appears morphologically to resemble more closely 
the ordinary human rickets. As noted above, however, most investigators 
have obtained osteoporosis, rather than rickets, on low calcium diets. 

Vitamin Deficiency. — The probability of a "fourth vitamin" factor in rickets 
was mentioned above. Skeletal changes in other vitamin deficiencies will now 
be considered. In young rats on a vitamin-free diet (polished rice with salt), 
Ishido ('23) found in the bone marrow of the femur and tibia numerous fat cells, 
which did not occur in rats exposed to ultraviolet light, or in full-fed controls. 

Vitamin A. — Herter ('97) noted mucoid degeneration of the bone marrow, 
and also bloody synovial fluid in the knee-joints, in pigs during fat starvation, 
involving deficiency in vitamin A. Tozer ('18, '20, '21) found that the changes 
in the costochondral junctions of guinea pigs on a diet deficient in vitamin A, 
but otherwise adequate, closely resemble those of mild experimental scurvy in 
these animals. Mackay ('21) and Tozer ('21a) obtained similar results in 
kittens on a diet deficient in vitamin A, but otherwise adequate. Hess, McCann 
and Pappe'nheimer ('21) likewise obtained no rickets in young rats on diets 
deficient merely in vitamin A, although histological examination showed 
retarded osteogenesis. Emmett and Peacock ('22) noted enlarged knee-joints 
and beading of the ribs in chicks on a diet deficient in vitamin A, but give no 
histological data. Findlay and Mackenzie ('22) found that diets deficient in 
vitamin A cause gelatinous degeneration of the marrow in the femur of the rat. 
Shipley, Park, McCollum and Simmonds ('21) observed that diets deficient in 
vitamin A, but otherwise complete, produce merely osteoporosis in the bones of 

Vitamin B. — But few changes have been observed in the skeleton during 
beriberi or polyneuritis, due to deficiency in vitamin B. Shipley, McCollum 
and Simmonds ('21) found that under these conditions rats show no gross 
deformity of the skeleton, but histologically present lesions essentially identical 
with those seen in guinea pigs with scurvy. These changes include osteoporosis; 
thin epiphyseal cartilage with strongly calcified zone of provisional calcification; 
no zone of osteoid; marrow congested and hemorrhagic, sometimes showing 
reticulum only. Findlay and Mackenzie ('22) likewise found that diets deficient 
in vitamin B produce hemorrhagic congestion in the bone marrow of the femur 
in the rat. Findlay ('21) noted atrophy of the skeleton in avian beriberi. 

Vitamin C. Scorbutus. — Although scurvy has doubtless occurred in the 
human race at various intervals since ancient times, it was not clearly recognized 
and differentiated as a distinct deficiency disease until about the 17th century. 
Fragility of the bones in scurvy was noted by Gideon Harvey (1675). The 
classical treatise by Lind (1772) mentioned briefly the gross skeletal lesions, 
including the occasional separation of the epiphyses in young patients. Since 
that time, a voluminous literature on scurvy has accumulated, which has 
recently been well summarized by Hess ('20) and Hojer ('24). 



The skeletal changes in human adult scurvy have been described recently by 
Aschoff and Koch ('19), Bierich ('19) and Comrie ('20). In children, scurvy was 
long confused with rickets. Infantile scurvy was first clearly demonstrated by 
Barlow ('83, '94) and is therefore commonly known as Barlow's disease. At 
that time but little was known of the essential histological changes in the 
skeleton during infantile scurvy, the details of which have since been thoroughly 
investigated by numerous workers. Naegeli ('97) was the first to give a detailed 
description of these changes, which have been confirmed and extended by 
Schmorl ('99, '01, '07), Schodel and Nauwerk ('00), Jacobsthal ('00), Looser 
('05), Erdheim ('18), and others. Fraenkel ('04, '06, 08,) has studied especially 
the skeletal changes as shown by the Rontgen-rays, including the so-called 
"white line." The pathology and pathogenesis of scurvy have also been 


■ Jfe* 


Pig. 55. — From a photograph of a portion of a section through the costochondral joint of a 
normal guinea pig; body weight 236 g. C, costal cartilage; Z, zone of enchondral ossification 
(bony and calcified cartilaginous trabeculae, and invading marrow) ; M, costal marrow; D, 
bone of costal diaphysis; P, costal periosteum and adjacent intercostal muscle. Zenker 
fixation; hematoxylin-eosin stain. X40. (Preparation by Everett Rowles.) 

reviewed by Vincent ('04), Lesage I'n), Schmidt ('21) and especially by 
Hess ('20). Hojer's ('24) recent monograph is excellent. 

The gross skeletal lesions include osteoporosis, with fragility and thinning of 
the cortex in the shaft of the long bones, and frequent occurrence of fractures in 
severe cases. Enlargement of the costochondral joints occurs, and has frequently 
been mistaken for rickets (cf. Hess '20; Hess and Unger '20). The general 
hemorrhagic condition in scurvy is manifested in the skeleton by frequent sub- 
periosteal and marrow hemorrhages. The bone marrow undergoes changes, 
becoming more fibrous or gelatinous in appearance, especially at the ends of the 
long bones. Hemorrhages and fractures with enlarged calluses are most 
frequent in pre-adult cases at the junction of the diaphysis with the epiphyseal 

The microscopic changes in the skeleton have been studied in detail, espe- 
cially during infantile scurvy. Although there is much variation in different 



bones, and in different individuals according to age and to the stage and inten- 
sity of the disorder, the structural changes in general involve a fibroid and 

Fig. 56. — From a photograph of a portion of a section through the costochondral joint of a 
young guinea pig on a diet of oats and milk-powder for 10 days; body weight reduced from 
273 g. to 212 g. Incipient scurvy. C, costal cartilage; Z, zone of enchondral ossification, 
largely replaced by fibroid marrow; M, more nearly normal costal marrow; D, bone of costal 
diaphysis; P, adjacent periosteum and (atrophic) muscle. Zenker fixation, hematoxylin- 
eosin stain. X40. (Preparation by Everett Rowles.) 

Fig. 57. — From a photograph of a portion of the enlarged costochondral junction of a 
young guinea pig; body weight reduced from 254 g. to 169 g. in 21 days on a diet of oats and 
milk-powder, with death from scurvy. C, costal cartilage; Z, irregular zone of echondral 
ossification; H, zone of fibroid marrow with distended blood vessels and hemorrhages; F, zone 
of non-hemorrhagic fibroid marrow; M, more nearly normal marrow; D, bone of costal dia- 
physis; P, adjacent periosteum, hemorrhagic, with markedly atrophic muscle. X40. (Prepa- 
ration by Everett Rowles.) 

hemorrhagic degeneration of the marrow, with associated inhibition of osteogenesis 
and atrophy of the preexisting bone. 


In a typical section through an area of enchondral ossification (e.g., a 
costochondral joint), as shown by Figs. 55-57, the structure of the cartilage 
appears abnormal. The zone of proliferation is usually increased in thickness, 
and the normally columnar arrangement of the cells becomes very irregular. 
The provisional calcification of the cartilage matrix occurs. But since this 
zone is not (as normally) invaded and replaced by the osteogenic tissue from the 
marrow, the calcified trabeculae do not become covered with bone, but form a 
widened and weak layer which easily fractures. The consequent hemorrhages, 
with partial organization and resorption from the adjacent marrow and 
periosteum, result in the very irregular and variable structure of this swollen 
area ("Trummerfeld" of Fraenkel). 

Adjacent to this "Trummerfeld," the bone marrow presents a fibroreticular 
structure with a gelatinous ground substance and few marrow cells 
("Gerustmark" of Schodel and Nauwerk). Scattered through the marrow 
of this region multiple hemorrhages frequently occur. The marrow in the 
remainder of the shaft is more nearly normal. The interspersed bony trabeculae 
appear variably thinned, due to failure of bony apposition, combined with con- 
tinued resorption. Schmorl ('07) opposed Looser's theory that the rachitic 
bone changes are secondary to the marrow hemorrhages. 

Osteogenesis is also retarded or entirely inhibited in the periosteum, with 
primary atrophy of the osteoblasts. Resorption continues, however, although 
the osteoclasts are not abnormal in number. The entire bone, both cortex and 
spongiosa, therefore becomes progressively osteoporotic. The bony tissue is 
normally calcified, but great reduced in amount. Bahrdt and Edelstein C13) 
and others have found that the fat content of the bones in infantile scurvy may 
be nearly unchanged, although the calcium and phosphorus content is greatly 

The histological changes in the skeleton of scorbutic adults appear very 
similar to those in children, excepting the intensive changes in the cartilage at 
the junction with the marrow. It may be noted that in children a coincidence 
of rickets and scurvy is not infrequent, as in the case described by Ingier ('13). 
Such complications, as also the changes during recovery from scurvy, may pro- 
duce very puzzling structural conditions. 

Experimental Scurvy. — Bartenstein ('05), Frolich ('12), and others found 
that young guinea pigs fed on raw or sterilized milk diet develop skeletal lesions 
(including microscopic) which have a remarkable similarity to those of infantile 
scurvy, excepting the absence of hemorrhagic tendency. Lipschiitz ('10) 
obtained similar results in puppies fed a phosphorus-poor diet of rice and egg 
albumin, supplemented by salt mixtures. The addition of casein, lecithin, 
nucleins and phosphates gave normal structure in controls. Schmorl ('13) 
obtained in puppies on phosphorus-poor diet skeletal lesions even more closely 
resembling human scurvy, with occasional small hemorrhages in the subchondral 
zone. The more typical hemorrhagic condition was absent, however, and 
Schmorl concluded that the disorder is not identical with human scurvy. 

A new epoch in experimental scurvy began with the work of Hoist and 
Frolich ('07, '12). They found that by means of various diets (cereals or bread) 


it is easy to produce in young guinea pigs a disorder which in every essential 
respect corresponds to that found in human infantile scurvy. They worked 
out the gross and microscopic changes (including the skeletal) with great care. 
They also compared the lesions with those found during ordinary starvation 
(water only), in which the mucoid degeneration of the marrow is similar, but 
the other changes (hemorrhage, etc.) different. For details, see Hojer ('24). 

The results of Hoist and Frolich in guinea pig scurvy have been confirmed 
and extended by Jackson and Moore ('16), Chick, Hume and Skelton ('18), 
Tozer ('18), and others. Howe ('21) obtained softened skull bones and enlarged 
joints in guinea pigs on vitamin-poor diets,but the relation to scurvy is uncertain. 

Ingier ('13, '15) made an extensive study of the effect of a scorbutic diet 
upon the fetus in pregnant guinea pigs. There aje marked individual varia- 
tions, but the skeletal changes appear greatest in the earlier fetal stages. The 
earlier fetuses usually die showing marked evidence of inhibited growth. 
Fetuses from later periods of pregnancy are born alive with comparatively 
slight skeletal changes. The pregnant mothers also suffer severely, especially 
in the earlier stages of gestation. 

Experimental scurvy, in all essential respects apparently identical with 
human scurvy, was also produced by K. Hart ('12) in young monkeys fed on 
condensed milk. The entire skeleton is affected in typical fashion, although 
there are marked individual variations. These results on the monkey were 
fully confirmed by C. Hart and Lessing ('13) and by Talbot, Todd and Peterson 
('13). In both cases, they confirmed Fraenkel as to the constancy of the "white 
line," which appears on the X-ray negative in the area of increased density at 
the junction of the epiphysis and diaphysis. (Hess doubts the diagnostic value 
of this sign.) More recently, Harden and Zilva ('19a) have also produced 
typical scurvy in monkeys, and noted the histological changes in the costochon- 
dral joints. 

In contrast with the striking success in producing experimental scurvy in 
the guinea pig and monkey, the results in other species of animals have been 
largely negative. Hoist and Frolich failed with the rat, mouse and cat. Mc- 
Clendon, Cole, Engstrand and Middlekauff ('19) fed oats to a rabbit for 9 
months, resulting merely in fragility of the bones. Findlay ('21b) likewise 
obtained merely loss in weight, excepting the offspring (above mentioned). 
As noted by Hess ('20), other evidence indicates that birds, pigs and cattle 
likewise show little or no susceptibility to scurvy. These are striking examples 
of the marked nutritional differences between species. 

Aqueous Inanition. — Schuchardt ('47) found an apparent loss of 7 per cent 
in the bones of pigeons with loss of 44 per cent in body weight on a dry barley 
diet. Falck and Scheffer ('54) noted an apparent loss of 5.3 per cent in the 
ligamentous skeleton of a dog fed dry biscuit 4 weeks with loss of 20 per cent in 
body weight. In adult albino rats, Kudo ('21) observed in the acute thirst 
series (body loss 36.1 per cent) an average loss of 4.3 per cent in the ligamentous 
skeleton and of 11.8 per cent in the cartilaginous skeleton. In the chronic 
thirst series (body loss 52.4 per cent) the relations were reversed, the ligamentous 
skeleton losing 10.3 per cent and the cartilaginous 5.0 per cent (Table 9). In 


experiments with young rats held at nearly constant body weight by dry diet 
for periods of 1-13 weeks, Kudo ('21a) found in general a progressive increase 
in skeletal weight, reaching a maximum average of about 40 per cent increase 
in the ligamentous skeleton, 58 per cent in the cartilaginous skeleton, and 32 
per cent in the humerus and femur alone (Table 10). 

The only data for structural changes in the skeleton during aqueous inanition 
are apparently those by Pernice and Scagliosi ('95a). In a dog which died after 
11 days on dry bread, with body loss of 24 per cent, the bones were noted as 
showing a moderate stasis hyperemia. Three young chicks were fed dry maize 
and lost 34-41 per cent in body weight in 8-10 days. At autopsy, no 
change was noted in the periosteum and osseous tissue, but the marrow appeared 
dark red in color. Microscopically cells of the cartilage in the lingual region 
appeared markedly atrophic, with zigzag borders and poorly stained nuclei. 
In some cases, the entire cell had degenerated into an amorphic granular mass. 

So far as known, the skeletal changes, both in weight and structure, during 
aqueous inanition are thus similar to those found during total inanition or on 
water alone (Table 4). This is perhaps due, at least in part, to the invariable 
lessening of the food intake on a dry diet, which would naturally produce under- 



Like the skeleton, the teeth appear very resistant to inanition in general, 
though especially susceptible to rickets and scurvy. After a brief summary, the 
effects of inanition upon the teeth will be considered under (.4) total inanition, 
or on water only ; and (B) partial inanition. 

Summary of Effects on the Teeth 

In total inanition, or on water alone, the teeth in adults show no appreciable 
change in weight or structure, but there are slight changes in chemical composi- 
tion, especially in chronic (incomplete) inanition. In the young, such inanition 
may delay the process of dentition, but persistent growth and development of 
the teeth (as of the skeleton) occur in young rats held at constant body weight 
by underfeeding. 

The effects of partial inanition have been studied chiefly in rickets and 
scurvy. In both human and animal rickets there is delayed and abnormal 
dentition. Both enamel and dentine may be defective and imperfectly calci- 
fied. Caries and lesions of the peridental membrane are frequent; but the dental 
defects are not closely correlated with the skeletal lesions, and are exceedingly 
variable in both human and animal rickets. 

In scurvy, the gums are markedly congested and swollen in about 80 per cent 
of the adult human cases, but apparently in a much smaller proportion of guinea 
pigs. The alveolar bone and peridental membrane undergo necrosis, with 
consequent loosening of the teeth, and ulcerations or pyorrhea may occur 
(more rarely in animals). Congestion and hemorrhage appear very early in 
the dental pulp (guinea pig), with consequent pulpar degeneration and fibrosis, 
and possibly osteodentine formation. Scorbutic changes may affect the teeth 
before eruption, even in the fetus. Recovery of normal structure in the teeth 
is possible upon antiscorbutic diet, unless extreme degeneration has occurred. 

(A) Effects of Total Inanition or on Water Alone 

Comparatively few observations have been made upon the teeth during 
simple inanition. Weiske ('97) noted that in 4 adult rabbits 7-1 1 days on 
distilled water alone, with loss of 35-41 per cent in body weight, there is a slight 
decrease in the organic content of the teeth, with corresponding relative increase 
in the inorganic content. Trowbridge, Moulton and Haigh ('18), in connection 
with an elaborate study of the changes in seven yearling steers subjected to 
various planes of prolonged underfeeding, found an apparent increase in the 
weight of the teeth, even with stationary or decreasing body weight. The 



chemical composition of the teeth is also changed, with an increase in water 
content and a decrease in nitrogen and ash. 

Talbot ('09) and others have pointed out the importance of malnutrition 
in connection with defective teeth of children. Sill ('09), for example, in 1,000 
school children 6-12 years of age in the Jewish quarters of East Side, New 
York City, found 40 per cent malnourished, more or less anemic and under 
weight; while 86 per cent had dental caries. He believes that dental caries is a 
causative factor in malnutrition. It might, however, be an effect, rather than a 
cause; or possibly a "vicious circle," malnutrition producing defective teeth, 
which in turn tend to prevent an adequate food-intake. Emerson ('22), how- 
ever, found no increase in caries among malnourished children. 

Tschkwinsky ('10) observed that in underfed lambs there is a delay in the 
replacement of the temporary with the permanent incisors. Jackson ('15a) 
found that in albino rats held at constant body weight by underfeeding from 
3-10 weeks of age, there is a progressive development of the teeth, the formation 
and eruption of the third molars proceeding in spite of the stationary body 
weight. This was confirmed by Stewart ('18). Similarly Boas ('23) finds that 
among poor children the eruption of the permanent teeth is not retarded, as is 
the general body development. It therefore appears that the teeth share with 
the skeleton the persistent tendency to growth during incomplete total inanition 
(general underfeeding). 

Bean '(14) and others have suggested the relative development of the teeth 
(appearance and replacement of the deciduous teeth) as a method of determining 
the "physiological age" of children. This would appear to be a more conven- 
ient index than the height or skeletal epiphyses, but infortunately there has not 
been as yet a sufficient investigation of the correlation of dentition with skeletal 
and other changes during retardation of growth in malnourished children. This 
is necessary before it can be determined how reliable is the stage of dentition 
as an index of "physiological " (or better "anatomical") age, in comparison 
with "chronological" age. 

(B) Effects of Partial Inanition 

King ('18) incidentally noted the frequent occurrence of defective teeth in 
rats suffering from malnutrition on diets which apparently contained relatively 
too much starch and too little protein. 

Mineral Deficiency. — Aside from observations in connection with rickets, 
not much is known as to the effects of dietary mineral deficiencies upon the 
structure of the teeth. Miller ('87) found a very slight apparent decrease in 
calcium content in the teeth of adult dogs after 13 weeks on a calcium-poor 
diet. The deficit is restored by refeeding on a diet rich in calcium. Leonard 
('20) observed imperfect formation of enamel in many infants 6 months to 3 
years of age. This was ascribed to malnutrition from diets lacking in essential 
salts and vitamins, but did not appear in the permanent teeth of the older children. 

Rachitis. — Irregularity of the teeth in rickets was noted even by Whistler 
(1645) and Glisson (1650). An abnormal delay in the eruption of the teeth in 


rachitic children has frequently been observed (Seibold '27; Woronichin '76; 
Jenner '95) and is one of the well-known clinical symptoms of rickets. 

Comby ('01) described the rachitic changes observed by himself and others in 
the jaws and teeth. Eruption is retarded and pronounced deformities occur, 
though fortunately rarely. "Ces dents sont malades dans leur germe; elles 
sortent noiratres, fendillees, insumsamment revetues d'email; au lieu de se 
developper normalement, elles tombent en poussiere, et les racines seules per- 
sistent au milieu des gencives tumefiees, fongueuses et saignantes." The per- 
manent teeth are usually good, though delayed in appearance. 

The work of Veve ('02) was inaccessible. 

In his comprehensive work on rickets, Wohlauer ('n) gave a detailed 
account of the changes in the jaws and teeth, which are frequent and important. 
The mandible becomes deformed, with the alveolar process slanting obliquely 
inward (Fleischmann). The maxilla is bent inward at the attachment of the 
zygomatic process, while the alveolar process is pressed outward. These 
deformities of the jaws, which are caused primarily by tension of the 
attached muscles upon the softened bones, naturally disturb the normal position 
of the teeth and give rise to various degrees of malocclusion. 

The dentition is delayed, and the teeth appear at extended intervals and in 
abnormal sequence (Baginsky) . Moreover the teeth show various abnormalities 
in form. They may or may not be attacked by caries, depending upon the 
time at which the rickets appears. If dentition occurs at the florid stage of 
rickets, the teeth are markedly affected; but not if the dentition is completed 
before the onset of rickets. The same principle applies to the permanent teeth, 
according to Wohlauer. 

According to Burchard and Inglis (/08), Hopewell-Smith has described an 
imperfect development of enamel during rickets, the first formed enamel con- 
taining numerous spaces, probably filled with soft tissue. Wells ('19) found 
delayed dentition in 32 out of 42 consecutive cases. There are frequently no 
teeth erupted at the end of the first year; sometimes none up to 18 months. 
The developmental process may be arrested, and there is a striking tendency to 
early caries. The enamel may be completely destroyed even before the tooth is 
fully erupted. 

According to Pfaundler ('22), in addition to the delay in dentition, the fol- 
lowing peculiarities of the teeth may occur in rachitic children: "The individ- 
ual teeth appear at unusually long intervals; erupt asymmetrically and 
in atypical order. Particularly in the upper jaw, they are frequently small, 
soft, easily broken and discolored by caries, to which they are peculiarly liable. 
They are often frightfully misshaped and foreshortened. There is occasionally 
an excessive formation of enamel. The temporary teeth show striped or circu- 
lar erosions at neck and root. The permanent teeth, the germs of which are 
also affected, show these erosions at the crown." 

Marfan ('22) and Ruden ('22) have recently likewise described delayed denti- 
tion, abnormal development of the teeth, and malformation of the jaws as a 
result of rickets in children. Park ('23a) holds that with proper diet during 
pregnancy and by the use of sunlight and cod liver oil during infancy "more 


could be accomplished in regard to the eradication of caries of the teeth than in 
all other ways put together, and that rickets would be abolished from the earth." 

In experimental rickets in puppies, Voit ('80) noted that the teeth are small 
and poorly developed. In spontaneous rickets in white rats, Weichselbaum and 
Erdheim ('09) found imperfect calcification of the dentine, especially of the 
youngest layer (adjacent to the odontoblasts), which was penetrated by vascular 
loops. The rickety teeth are very transparent to the X-rays. They cited similar 
observations by Fleischmann ('07) in human teeth during rickets. 

M. Mellanby ('18, '20, '21) has described a general hypoplasia of the teeth in 
puppies with rickets caused by diets deficient especially in vitamin A. The 
following defects are noted: (1) delayed shedding of the deciduous teeth; (2) 
delayed eruption of the permanent teeth ; (3) irregular position and overlapping, 
especially of the incisors; (4) enamel defective or partially absent; (5) general 
softening of the teeth, due to low calcium content. The dental defects 
appear independent of oral sepsis or other complications. 

In experimental rickets of young rats, Shipley, Park, McCollum and 
Simmonds ('21) observed that the incisor teeth are frequently loose, fragile and 
sometimes fractured. The conditions were studied more in detail by McCollum , 
Simmonds, Kinney and Grieves ('22), who found the greatest percentage 
of oral defects in the rats fed diets deficient in protein, calcium and vitamin 
A. The next highest incidence occurred in rats on diets low in calcium; and a 
still lower percentage occurred on diets low in both calcium and vitamin A. No 
caries-like lesions, pulp exposure, osteodentine, or defects in the attaching tissue 
or maxilla occurred in the stock rats on complete diet. They conclude that 
severe oral disease may result from relatively defective diets, where the distur- 
bance appears out of all proportion to the cause. The diet is thus of primary im- 
portance in determining the quality of the teeth and their vitality in resisting 
invasion by microorganisms. 

A further report by Grieves ('22) places the rats studied in 3 groups. In 
group 1 (on low calcium diet causing a pseudorachitic condition with excess 
osteoid), 22 per cent of the rats show caries-like defects and 3.65 per cent of their 
molars are involved. In 41 per cent of these rats, attaching- tissue defects 
occur, involving also the molars. Diets affecting the bones do not always affect 
the tooth attachments, however. 

In group 2 (diets low in calcium and vitamin A), 31 per cent of the rats show 
caries-like defects and 5.21 per cent of their molars are involved. Twenty per 
cent of these rats show attaching-tissue defects. But many definitely rachitic 
rats show little or no dental defects; rachitis and caries-like lesions are rarely 

In group 3 (diets high in calcium and butter fat), 17 per. cent of the rats 
show caries-like defects and 1.71 per cent of their molars are involved. In 28 
per cent of these rats, attaching-tissue defects occur, disturbing 33 per cent of 
their molar attachments. 

The caries-like and attaching-tissue lesions are described by Grieves in detail. 
The variability of the lesions in the test rats resembles that found in human 
rickets, indicating differences in individual resistance. Endocrine and other 


factors may be involved. " Until further facts are available, one can think only 
of the necessity for a proper Ca-P-organic factor balance in any diet as the most 
important in the formation and maintenance of normal bones and teeth and 
healthy attaching-tissues." 

Marshall ('23) in puppies on diets "insufficient or improperly proportioned" 
in calcium and phosphorus found a marked delay in dentition and relative 
absence of dentine, with normal amount of enamel. 

Scorbutus. — In human adult scurvy, the involvement of the gums and teeth 
usually forms one of the first and most conspicuous symptoms. The lesions 
have recently been described and illustrated in detail by Aschoff and Koch 
('19). Comrie ('20) noted swollen and bleeding gums in 80 per cent and severe 
gingivitis or pyorrhea in about half of 600 cases. Bierich ('19) found the 
gingival swellings most pronounced in those with carious teeth. The lesions 
have recently been summarized by Hess ('20). The gums become congested 
and hemorrhagic. Later the teeth become loose and may fall out, and the 
alveolar bone undergoes necrosis. Pyorrhea may be present. The gums 
may become so swollen as to hide the teeth, forming foul, fungoid growths. 

In infantile scurvy (Barlow's disease) the gingival lesions are similar but some- 
what less pronounced if teeth are present; and are slight or absent before the 
eruption of teeth. Talbot ('19) emphasizes the susceptibility of the alveoli and 
peridental membrane, rather than the teeth, to the changes produced by scurvy. 

In experimental scurvy the dental and gingival lesions have been frequently 
studied. Hoist and Frolich ('07, '12) in their pioneer work on experimental 
scurvy in the guinea pig noted loosening of the teeth, some gingival hyperemia 
and hemorrhages, and rarely gingival ulcerations. Only 20 per cent of the scor- 
butic guinea pigs show marked congestion of the gums. These results have 
been confirmed by Cohen and Mendel (/18) and many others. Jackson and 
Moore ('16) found congestion, hemorrhages and necrotic degeneration 
in the pulp of both incisors and molars. 

The first detailed histological study of the teeth in scurvy was made by Zilva 
and Wells ('19). They find that in guinea pigs on scorbutic diet profound 
changes occur in the teeth very early (at 10 days), when only slight lesions are 
seen elsewhere. In the teeth the pulp undergoes a fibroid degeneration. "In 
complete pulpar fibrosis no cellular elements of any description occur . . . 
Nerves, cells, blood vessels and odontoblasts have all shared the process of 
fibrification and are no longer recognisable. The fine cellular connective tissue, 
which is but a loose mass of network in the normal state, has either become 
grossly hypertrophied or quite obliterated, and its place taken by a new firm 
fibrous structure, devoid of cells, nuclei or any regular arrangement of constituted 
parts . . . The irregular osteoid condition is well marked . . . In a scuivy 
tooth the condition persists right up to the apex of the root; the change appears 
to start first in the odontoblastic cells at the top of the pulp, working down to- 
ward the apex, followed by distended blood vessels and hemorrhage; then com- 
plete fibroid degeneration follows. . . In advanced cases of scurvy the teeth 
were apparently sound but useless, inasmuch as .they had been loosened by the 
gradual absorption of the cement membrane of the alveolar sockets, which had 


left exposed that portion below the neck." Nearly normal structure of the 
teeth had been recovered in a scorbutic guinea pig cured by one month on nor- 
mal mixed diet. Scorbutic changes were also noted (but not in detail) in the 
teeth of monkeys. 

Wells ('19 ,'21) added further data to the results of Zilva and Wells ('19). 
Pregnant guinea pigs were fed scorbutic diet and six in advanced stages of preg- 
nancy aborted after 11-15 days on the diet. "Microscopical sections were 
made of the teeth of the mother and offspring and in every case an advanced 
state of scurvy could be seen." Wells ('21) states that the irregularly osteoid 
condition of the dentine is "probably due to the hemorrhagic condition of the 
dentinal fibrils," an explanation which appears unintelligible. 

Howe ('20) confirmed the results of Zilva and Wells, finding a loosening of the 
teeth and a condition resembling pyorrhea in scorbutic guinea pigs. Unless 
the conditions are extreme, recovery follows the use of antiscorbutics. A similar 
disorder was produced by Howe ('21) in guinea pigs on a diet deficient in all 
three vitamins (A, B and C). Howe ('22) also reports loosening of teeth, exces- 
sive tartar formation, etc., in monkeys on scorbutic diet. 

Robb, Medes, McClendon, Graham and Murphy ('21) in guinea pigs on scor- 
butic diet (equal parts of white flour and alfalfa meal) report that "The teeth of 
our scorbutic animals become very loose. The dropping-out-of-the-teeth 
indicates a loss of cementum and possibly of material from the alveolar process. 
The changes in the teeth proper were surprising. There was marked hyperemia 
of the pulp with some hemorrhage. The odontoblasts lost their tall columnar 
form and secreted osteodentine very rapidly. The osteodentine nearly filled the 
pulp cavity in some cases." Unfortunately the normal process of development 
in the teeth of the guinea pig is unknown; but osteodentine occurs normally in 
the pulp of the incisor of the rat, which suggests a possible source of error in 
interpretation. However, Hojer ('24) obtained similar results. 

In young rats fed McCollum's rachitic diet No. 291 1 (calcium-deficient), 
Bracco ('23) obtained no gross appearance of rickets, but microscopically the 
teeth showed intense congestion of the pulp and marked irregularities in the 
formation of dentine. 

In the case of the pulpar degeneration observed by Zilva and Wells, it should 
be noted that the histological changes closely resemble those described by Bur- 
chard and Inglis ('08) following traumatic thrombosis or " jugulation," which 
results in death of the pulp or extravasations leading to fibroid degeneration. 
"Inflammation of a low grade may persist in the pulp for long periods, giving 
rise to an increase of its fibrous tissue with atrophy of the cellular elements, 
producing a condition found in chronic interstitial inflammation in some 
other tissues — a sclerosis." Fibroid degeneration of the pulp may also occur 
normally in old age, and is ascribed by Hopewell-Smith to capillary thrombosis. 
It therefore seems probable that the scorbutic pulpar fibrosis described by Zilva 
and Wells is not a unique condition, but is probably secondary to the hemor- 
rhagic condition resulting in interference with the normal blood supply to the 
bone and is thus related to the fibroid degeneration which occurs also in the 
pulp marrow. 



The present chapter deals with the effects of inanition upon the skeletal 
musculature only. The cardiac muscle will be considered in connection with 
the heart, and smooth muscle in connection with the various organs in which it 
occurs. (See Index.) The marked atrophy of the musculature explains the 
characteristic weakness generally appearing in both total and partial inanition, 
as well as in various chronic diseases involving malnutrition. Under such 
conditions the musculature appears to serve as a storehouse of protein, fat and 
glycogen reserves for the exhausted organism. After a brief summary, the 
effects of inanition upon the musculature will be dismissed under (A) total 
inanition and (B) partial inanition. 

Summary of Effects on the Musculature 

During total inanition, there is in adults, both human and animal, a marked 
loss in the weight of the skeletal musculature, which in general is roughly pro- 
portional to that of the entire body. The loss in the musculature is relatively 
greater in some species (frogs), and in general is somewhat greater in chronic 
(incomplete) than in acute inanition. The degree of atrophy apparently varies 
in different regions of the body. 

In young animals (rats) held at constant body weight by underfeeding, the 
musculature tends to increase slightly in weight. In malnourished human 
infants, the musculature appears atrophic, but it is questionable whether there 
is much actual loss of weight except in extreme cases. The appearance of ema- 
ciation may be increased by the relative growth of the skeleton. Recovery in 
the musculature after inanition usually appears promptly upon adequate 

The histological changes in the skeletal muscle fibers involve first a simple 
atrophy — a decrease in size with no evident changes in structure. The ordinary 
(neutral) fat, both interfibrous and intrafibrous, undergoes progressive resorp- 
tion, but the phosphorized lipoidal granules are very resistant to inanition. 
Later certain of the muscle fibers begin to show degenerative changes, with 
progressive loss of the characteristic striations in the myofibrillae, granular 
(fatty, albuminous or pigmentary) degeneration in the sarcoplasm. Waxy 
degeneration is rare. The nuclei are more resistant, and often undergo prolifera- 
tion. A variable degree of hyperplasia (fibrosis) occurs in the interstitial 
connective tissue. The extent of the degeneration varies greatly, not only in 
different fibers of the same muscle, but also in the muscles in different regions of 
the body. 



In hibernating animals, during the feeding period, there is a marked storage 
of fat in the muscle, which serves as an important storehouse of fat and protein for 
the organism during the subsequent fasting period. In the frog and the salmon 
the musculature thus contributes largely to the materials for growth of the sex 
glands during the fasting period. Similarly in all species the musculature during 
inanition is apparently consumed to supply the needs of the more essential vital 
organs of the body. 

In the various forms of partial inanition, atrophy and degeneration in the 
skeletal musculature are frequently evident, especially in those conditions 
involving general emaciation of the body, such as famine edema, pellagra, etc. 

In rickets, there is apparently a regressive dystrophy of the musculature, with 
a slight progressive loss of its weight (in rats). In beriberi, there are found the 
usual atrophic degenerative changes in the muscle fibers, with nuclear prolifera- 
tion and interstitial fibrosis — a condition frequently resembling that of chronic 
myositis. In scurvy, the muscles share in the general hemorrhagic condition, 
but the fibers also independently undergo the typical atrophic degenerative 
changes with interstitial fibrosis. 

In aqueous inanition (on dry diets), the skeletal musculature undergoes 
atrophy with loss of weight and degenerative changes in the muscle fibers 
resembling those in other forms of both total and partial inanition. The 
intermuscular connective tissue may present a round cell infiltration, as in 
rickets and scurvy, which does not ordinarily appear in total inanition. 

(^4) Effects of Total Inanition or on Water Only 

The effects of total inanition upon the skeletal musculature and the very 
similar effects on water alone will be discussed (i) as to the gross changes, 
especially in weight, and (2) as to the histological changes involved. 

Changes in Weight of the Musculature. — These changes may conveniently 
be considered separately in the adult and in the young organism, human and 

Adult Human. — The atrophy of the skeletal musculature during starvation 
has often been observed, but quantitative data are lacking. Tiedemann ('36) 
stated that: "Die Leichname Verhungerter fand man in hohem Grade abge- 
magert, besonders waren die Muskeln sehr dunn, welk und leicht zerreissbar. " 
Willien ('36) noted that especially the muscles of the trunk become atrophied 
during inanition. Rokitansky ('54) concluded that in general the atrophy of 
the musculature during malnutrition is relatively less than that of the blood, 
adipose and areolar tissues, but greater than that of the viscera, nervous system 
and skeleton. Falck ('81) noted that the muscles at autopsy appear "braun, 
klebrig, atrophirt. " In a case of starvation with loss of about 40 per cent in 
body weight, Bright ('77) observed that the musculature throughout appeared 
wasted, especially that of the trunk, and entirely devoid of fat. Theile ('84) 
recorded the weights of the various groups of muscles in both adults and children 
in different conditions of nutrition. In a greatly emaciated man 31 years old, 
it appears that the upper extremity muscles have lost relatively less than the 


lower, and the diaphragm less than either. In an emaciated (tuberculous) 
man of 39 years, the perineal musculature was atrophied, apparently more than 
that of the extremities. (The children will be mentioned later.) 

Adult Animal. — In adult pigeons on total inanition with average loss of 40.4 
per cent in body weight, Chossat ('43) found the loss in the skeletal musculature 
to be very slightly greater, averaging 42.3 per cent. Bidder and Schmidt ('52) 
in a cat with loss of 50 per cent in body weight found an apparent loss of about 
67 per cent in the musculature. In various mammals, Bourgeois ('70) noted 
that the musculature during inanition loses slightly more (relatively) than the 
whole body, the loss in dry weight averaging about 45 per cent. The trunk 
muscles appear to lose relatively more than those of the neck and limbs, confirm- 
ing Collard de Martigny (1828) and Chossat ('43). 

C. Voit ('66) found during starvation in the cat a loss in the musculature 
relatively slightly less than that in the whole body; while Sedlmair ('99) found it 
slightly greater. In the dog, a relative loss in the musculature slightly greater 
than that in the body as a whole was found by C. Voit ('94), Kumagawa ('94), 
and E. Voit ('05, '05a). In the rabbit, Pfeiffer ('87) found the musculature to 
lose relatively somewhat less than the whole body; Weiske ('97) found the rela- 
tive loss slightly greater than in the whole body; while Voit ('05) found it nearly 
unchanged in relative weight. Jackson ('15) in albino rats on acute inanition 
found an average loss of ^t, per cent in body weight and of 31 per cent in the 
musculature; while in chronic inanition, with body loss of 36 per cent the muscu- 
lature lost 41 per cent in weight (Table 4). 

Gaglio ('84) noted a loss of 85 per cent in the musculature of a frog starved 
with loss of 56 per cent in body weight. In leopard frogs (Rana pipiens) with 
previous losses in body weight up to 50-60 per cent, Ott ('24) found that the loss 
in the musculature always relatively exceeds that of the body in the male. In 
the female, the loss in the musculature is even greater in the earlier stages of 
inanition, but later it more nearly corresponds to that in the body as a whole 
(Table 6). There is a progressive decrease in the percentage of dry substance 
in the musculature. 

Certain special conditions of total inanition require attention. In 3 mar- 
mots hibernating an average of 166 days, with loss of 35.5 per cent in body 
weight, Valentin ('57) observed an average evident loss of 30.3 per cent in the 
musculature, the loss being apparently more rapid in the earlier part of the 
period. Miescher ('80, '97) found that in the fasting Rhine salmon, the sex 
glands develop at the expense of the musculature, which may lose over 50 
per cent in weight. The superficial lateral trunk musculature is attacked (his- 
tology mentioned later), while the remaining muscles appear relatively unaf- 
fected. Changes in the weight of the fasting salmon are also given by Gillespie 
(Paton '98). In the Pacific salmon, Greene ('i3-'i9) likewise found a loss of 
40-50 per cent in weight of the musculature during the fasting period of 

An apparently comparable condition exists in the frog, as observed by Gaule 
('01), in which the musculature reaches its maximum weight during the summer 
feeding period (July-August). It declines to a minimum during the winter 


fasting period, and is apparently sacrificed in part to serve as material in the 
development of the sex glands. These are conspicuous examples of dystrophic 
growth changes during adult inanition, certain portions of the body growing at 
the expense of others, as occurs so generally during chronic inanition in young 

Human Infants. — In malnourished, athreptic infants, marked atrophy of the 
musculature has been observed by DeTommasi ('94), Thiercelin ('04) and many 
others, so that in extreme cases the emaciated body appears reduced to "skin 
and bones." Ohlmuller ('82) found that in an atrophic infant of 56 days the 
musculature formed 23.6 per cent of the body, whereas in a "normal" infant of 
the same age it formed 25.8 per cent. This would indicate that the muscula- 
ture had relatively lost slightly more than the body as a whole, or had been more 
retarded in growth. Theile ('84) likewise observed the musculature of an emaci- 
ated infant apparently forming only about 17 or 18 per cent of the body, 
the normal at birth being 20-22 per cent. Nicolaeff ('23) found variation in the 
amount of atrophy in the individual muscles of famine-stricken children. The 
functionally active muscles of mastication (temporal, masseter) lost relatively 
less than the biceps brachii, which sometimes lost 50 per cent in weight. 

Young Animals. — In nursing puppies subjected to chronic inanition by insuf- 
ficient or improper diet, Quattrochi ('01) observed marked emaciation, with 
atrophy of the skeletal musculature. Aron ('10, '11) noted that in an underfed 
puppy the musculature formed only 29.3 per cent of the body weight, while 
in a .full-fed litter-mate control it formed 50.2 per cent. Since the control 
was much larger, however, the comparison indicates not the loss, but merely 
the amount of retardation in growth. An initial control of the corresponding 
body weight is required to determine whether an actual loss in the musculature 
has occurred. 

In young albino rats held at nearly constant weight by underfeeding for 
various periods, Jackson ('15a) found a slight increase in the weight of the 
musculature (Fig. 39; Table 4), a result confirmed by Stewart ('16, '18) and by 
Jackson and Stewart ('20). In the fetuses of rats subjected to inanition by 
underfeeding the pregnant mother, however, Barry ('20, '21) obtained a 
slight loss (or retarded growth) in the weight of the musculature. 

In a thin yearling steer held at maintenance by underfeeding for several 
months, Trowbridge, Moulton and Haigh ('18) found that the musculature 
formed 44.5 per cent of the body weight, whereas in a fat control of the same age 
it formed only 33.1 per cent. This would apparently (though not necessarily) 
indicate a persistent growth of the musculature during inanition. On the other 
hand, Thompson and Mendel ('18) believe that in underfed albino mice the 
characteristic curvature of the spinal column is caused by the arrested growth of 
the skin and muscle, together with the persistent growth of the vertebral 

It maybe noted that although there may be no actual loss in the weight of 
the musculature, or even a slight gain, in young organisms on chronic inanition, 
an appearance of emaciation may be produced by the concurrent persistent 
growth of the skeleton. 



Recovery upon Refeeding. — Stewart ('16) found that in young albino rats 
held at maintenance for several weeks, and in which the musculature was pre- 
sumably above normal in weight, approximately normal conditions are restored 
after the first week or two of refeeding upon adequate diet. In younger rats, 
underfed from birth to 6 or 10 weeks of age, Jackson and Stewart ('19) observed, 
upon refeeding to a body weight of 25-75 g. that the musculature lags 
behind and appears slightly below normal weight. Similarly in young rats 
refed fully after long periods of inanition, Jackson and Stewart ('20) found in 
most cases a slight deficit in the weight of the musculature. However, the 
differences are so slight and variable that their significance is somewhat doubtful. 

Histological Changes in the Musculature. — The histological changes also 
may be grouped according to those in the adult and the young, both human and 

Fig. 58. — -Cross section showing the his- 
tological structure of skeletal muscle in a man 
who died from starvation. The muscle fibers 
appear extremely atrophied and separated 
from the endomysium by extensive spaces. 
CMeyer '17.) 

Fig. 59. — Cross section showing the his- 
tological structure of cardiac muscle in a man 
who died from starvation. The muscle fibers 
appear variably atrophic and shrunken, in 
places separated by extensive spaces inter- 
mingled with the connective tissue stroma. 
(Meyer '17.) 

Adult Human. — Schultzen ('62, '63) described fat droplets and indistinct 
cross stria tion in the skeletal muscle fibers of a 19 year old girl who had starved 
to death. "Striationem transversam bene perspicuam reddere non licuit. 
Inter fibrillas magnae adipis guttae." Hayem ('77), in cases of starvation dur- 
ing and after the siege of Paris, found that the lesions in the muscle fibers appear 
more distinct in chronic than in acute inanition, and include: (1) simple atrophy, 
the cross striation remaining unaffected; (2) granular degeneration; (3) fatty 
degeneration; (4) pigmentary degeneration. The interstitial connective tissue 
tends to hyperplasia (fibrosis). 

Popow ('85a) studied the changes in muscle fibers (human and animal) 
during starvation, noting decreased diameter, also granular, fatty and some- 
times waxy degeneration. Landau ('10) noted fatty degeneration in the muscle 
fibers in various diseases involving general cachexia or nutritional disturbance 


of the muscle tissue through interference with the circulation. In a man who 
died from starvation, Meyer ('17) found a marked atrophy of the skeletal muscle 
fibers, which in cross sections appear to have shrunken away from the endo- 
mysial sheaths (or sarcolemma?) leaving empty spaces of variable width (Fig. 
58). The striations in most fibers are only faintly visible, though often well 
preserved. The sarcoplasm may form granular masses. The nuclei may be pro- 
liferated, sometimes forming a degenerated mass. 

Adult Animal. — Heuman ('50) observed a decrease in the size of the pec- 
toral muscle fibers in starved pigeons. Valentin ('58) could find no evident 
changes in the microscopic structure of the muscle fibers in hibernating marmots, 
in spite of loss in body weight up to 35 per cent. Manassein ('69, '69a) in the 
muscle fibers of starved rabbits found granular degeneration (albuminous or 
fatty), sometimes pigmented or waxy (Zenker's) degeneration. The granular 
degeneration was found oftenest in the recti muscles of the eye, the order of fre- 
quency in other muscles observed being: tongue, diaphragm, abdominal muscles, 
shoulder muscles, intercostals. Lepine ('74) noted brownish pigmentation and 
disappearance of fat in the muscles of starved animals. Carville and Boche- 
fontaine ('74, '75) found the muscles in starved dogs to be yellowish red. No 
decrease was noted in the diameter of the fibers, which are in most cases finely 
granular, many with loss of cross striation. Some fibers appear vitreous (waxy 
degeneration?) with spaces separating the sarcolemma. Eichhorst ('79) 
observed non-fatty granulation in the muscle fibers of starved pigeons and raven. 

In fasting summer frogs, Sokoloff ('76) studied the degeneration (granular 
or waxy) in the muscle fibers, also the regeneration upon refeeding. The nerve 
endings in muscle appear very resistant to starvation. In the atrophic and 
degenerated muscle fibers of starved frogs, Gaglio (,'84) found numerous fine 
albuminous (non-fatty) granules, with proliferation of the nuclei; also slight 
increase in the interstitial connective tissue. 

In rabbits subjected to total complete or incomplete inanition, Ochotin ('85, 
'86) described the usual granular or fatty (?), occasionally waxy, degeneration 
in the muscle fibers of the diaphragm. Coen ('90), in the starved rabbit and 
kitten, found the muscle fibers mostly well preserved, some showing finely 
granular degeneration with loss of striation. The interstitial stroma shows 
congestion and a variable degree of nuclear proliferation. Morpurgo ('90) 
observed occasional mitoses in the skeletal muscle of adult rabbits on ample 
refeeding after inanition. 

During inanition in the pigeon Knoll ('80) noted in the muscle fibers a reduc- 
tion in the fatty interfibrillar granules, but an increase in the non-fatty. Knoll 
('89) and Knoll and Hauer C92) observed during starvation a decrease in the 
granulation in the dark fibers of the pectoralis major, but the greatest degree of 
atrophy occurs in the non-granular, light fibers. Morpurgo ('89) found the 
average diameter of the skeletal muscle fibers in starved pigeons reduced from 
33fx to 18.6M, indicating a loss in volume of 68 per cent. 

The size of the skeletal muscle fibers under various degrees of nutrition was 
measured by Kunkel ('87), Schwalbe ('90), Schwalbe and Mayeda ('90), 
Mayeda ('90), and Halban ('94), who found the average diameter of the fibers 


decreased in both human and animal malnutrition. In the fasting frog, 
Kunkel ('87) found a marked decrease in the diameter of the sartorius fibers, 
but not in their number. The large range of normal variation in the size 
of the muscle fibers makes comparison somewhat difficult. Frankl and Freund 
('84) held that the diminution in the volume of muscle during emaciation is due 
only in part to decreased caliber of the atrophic fibers, the greater part of the 
decrease being ascribed to actual disappearance of disintegrated fibers, the 
interstitial connective persisting in increased amount. In the starved dog, 
Morpurgo ('89b) found no decrease in the number of muscle fibers, and the 
muscle nuclei undergo but slight atrophy. 

Statkewitsch ('94) studied the microscopic changes in the muscle and other 
tissues during inanition in numerous animals (mammals, birds, reptiles and 
amphibians), giving also a detailed review of the earlier literature. He found 
that in general the skeletal muscle is affected earlier and more intensively than 
smooth muscle. "Abgesehen von einer Abnahme der Grossenverhaltnisse 
und einer Triibung treten in den quergestreiften Muskelfasern je nach der 
Dauer des Hungerns Schwellung, kleinkornige und spaterhin auch grosskornige 
Degeneration auf, wobei die quere, wie auch Langsstreifung nicht mehr beo- 
bachtet werden kann." The degenerative changes appear first in the cervical 
musculature, then (in order) in the extremity muscles, pectoralis major, heart, 
rectus abdominis, and finally the smooth musculature. The granules are 
albuminous; fatty granules (by ether or osmic tests) not being found during 
inanition. Zenker's (waxy) degeneration is rare (found in 1 cat and 1 dog, 
6-24 hours post mortem), and pigmentary degeneration was never observed. 
Since the muscle fiber shrinks greatly, the nuclei appear more numerous, being 
nearly unchanged in size. In extreme stages of inanition, nuclear degeneration 
may occur. Very similar changes were found by Konstantinowitsch ('03) 
in the muscle fibers of starved rabbits, lizards and frogs; and by Beeli ('08) in 
cats, showing decrease in the average nuclear diameter. 

Athanasiu and Dragoin ('08) found no fat in the striated muscle fibers of 
the summer frog, but a storage of large amounts in rows of interfibrillar granules 
during the winter. Similarly Miescher ('80/97) observed fat droplets between 
the myofibrillae of the muscle fibers in the superficial lateral trunk muscles of 
the fasting salmon. Further details as to this fat storage were noted by Mahal- 
anobis (Paton '98) and especially by Greene ('12, '12a, '13, '19) and Greene and 
Greene ('14). Greene finds that in the Pacific salmon at the beginning of its 
migratory fast the fat is stored chiefly in the muscles: (1) in the dark superficial 
lateral trunk muscle, chiefly as large droplets within the fibers; (2) in the great 
mass of pink muscle, with large quantities of fat, at first entirely interfibrous, in 
droplets of variable size up to iooju; (3) in the smaller fin muscles, a slight amount 
of fat, chiefly interfibrous. The stored fat is gradually consumed on the river 
journey to the spawning grounds, but it is not completely exhausted even at 
death. Chemical analysis indicates that the muscle also decreases markedly 
in protein content, with slight loss in ash and increase in water content. 

Bell ('09, '10, '11, '12) made a careful study of the granules in muscle fibers 
during inanition, finding that in the ox the " liposomes " (fatty or lipoidal granules 


staining with scarlet red, etc.) do not appear to vary much under moderate 
fluctuations in the degree of nutrition. In the rat and cat, however, the 
liposomes during starvation decrease notably in size, number and staining 
capacity. For the rat, this was confirmed by Bullard ('12). Krause ('n) 
also states that the fat droplets in muscle fibers are dependent upon the nutritive 
condition of the animal. In starved dogs, Morgulis, Howe and Hawk ('15) 
found indistinctness of the striations, but no swelling or granular degeneration 
of the muscle fibers. 

In the muscle fibers, as previously mentioned in Chapter VI for adipose 
tissue, the fat is apparently of two kinds: (1) the ordinary (neutral) fat, which is 
easily removed by inanition ; and (2) the phosphorized, lipoidal fat, which strongly 
resists inanition and in extreme stages becomes increased in amount by fatty 
degeneration or infiltration (v. Gierke, '21). 

In Amia calva after 20 months of starvation, Smallwood ('16) found the 
skeletal muscle fibers in various degrees of degeneration and disintegration, 
which appeared to involve progressively: (1) the cross striations; (2) the sarco- 
plasm; and (3) the nucleus. 

Moulton ('20, '20a) observed that in the skeletal muscle of underfed steers 
there is a marked loss of nitrogen as well as of fat, with notable decrease in the 
size of the muscle fibers, but no obvious change in histological structure. Thus 
the muscles form an important storehouse for protein, as well as for fat and 

In the Young. — So far as they have been observed, the changes in the struc- 
ture of the muscle in young individuals appear in general similar to those in 
adults. Walbaum ('90) observed that in malnourished children there is a 
decreased content in the fatty granules of the skeletal muscles excepting the 
eye muscles. Moenckeberg ('12) described the atrophic and degenerative 
changes during malnutrition in the muscle and other tissues. Lesage and Cleret 
('14) found a marked interstitial sclerosis in the muscle tissue of infants with 
congenital spasmodic atrophy. According to Nobecourt ('16), Variot and 
Ferrand studied the diameter of crural muscle fibers in malnourished infants. 
In the " hypotrophic " infants (with moderately retarded growth) the muscle 
fibers show a variable degree of diminution in diameter; but the fibers in the more 
severely malnourished are said to show no decrease in diameter. "La fibre 
striee des enfants amaigris, meme dans le cas ou cet amaigrissement est consider- 
able et ou Tenfant n'a que la moitie du poids qu'il devrait avoir pour son age et 
pour sa taille, n'est presque pas diminuee de volume." This remarkable finding 
needs verification, although it is quite possible that in the young the muscle 
fibers tend to resist a decrease in volume during inanition. As was noted 
above, the total mass of the musculature in underfed young rats not only fails 
to decrease, but usually even increases slightly in amount, with nearly constant 
body weight. 

Morpurgo ('98a) concluded that the general law of post-embryonal develop- 
ment in the musculature is the same as in other tissues. There is an early 
period of cellular differentiation, governed by heredity and independent of nutri- 
tion and function (Roux). The later growth of the muscle fibers is not governed 


chiefly by heredity, but varies according to nutrition and function. During 
this later period, inanition causes a simple atrophy of the muscle fibers. 

Changes in the chemical composition of the musculature during inanition 
have been referred to incidentally in the foregoing account, and are further 
described in the papers of Aeby ('75), Pfeiffer ('87), Lukianoff ('88), Aldehoff 
('8.9), Tonninga ('93), Rubow ('05), Roger ('07), Maignon ('07), Tobler ('11), 
Terroine ('20), and Moulton, Trowbridge and Haigh ('22a). In general, they 
support the doctrine that the muscles serve as an important storehouse for 
reserve protein, fat and other materials, which are consumed by the organism 
during inanition. 

(JB) Effects of Partial Inanition 

We have here to consider the effects upon the musculature in certain forms 
of partial inanition, including deficiency of protein, minerals, vitamins and 

Protein Deficiency. — As mentioned in Chapter V, the edema found during 
human famine and allied conditions may be due to a mixed deficiency, but usu- 
ally a lack of protein appears to be of primary importance. In this malnutri- 
tional edema, Budzynski and Chelchowski ('16) and others have generally 
observed profound atrophy and weakness of the musculature. Maase and Zon- 
dek ('17), however, ascribed the muscular atrophy in the cases of "war edema" 
to general inanition, and especially to deficiency of fat. In the muscles' of an 
infant with alimentary edema, Klose ('13) found an increased water content, 
with reduction in fat, protein and ash. 

In the "cachexia aquosa" and allied conditions in sheep and cattle on inade- 
quate (especially low protein) diets, Hoare ('15) and Froehner and Zwick ('15) 
state that the muscles are emaciated, pale, flaccid and sometimes edematous. 
Kohman ('20) was able to produce a malnutritional edema with muscular 
atrophy in rats on a watery diet, low in protein and fat. 

Pellagra, as stated in Chapter V, may be considered as primarily due to 
protein deficiency. In this condition, especially in advanced stages, muscular 
atrophy is very common. Fraenkel ('69-'7o) found it in 21 out of 48 cases, the 
pectoralis major being especially involved. Muscular atrophy in pellagra is 
mentioned also by Tuczek ('93) and Marie ('08, '10); the literature on the 
subject is reviewed by Raubitschek ('15). 

Mineral Deficiency. — In a puppy on an iron-poor diet, von Hoesslin ('82) 
observed continued growth of the body; but the muscles of the extremities 
showed fatty degeneration, and microscopically fat droplets appeared among 
the myofibrillae. 

In human rickets, atrophy of the musculature is mentioned by Whistler 
(1645), Glisson (1650), Seibold (1827), Vincent (04), Cheadle and Poynton 
(07), Wohlauer ('n) and Engel ('20). Bing ('07), Stoeltzner ('09) and Banu 
('21) also support the theory of a specific regressive dystrophy of the muscu- 
lature in rickets, although Heubner and Comby consider the process a disuse 
atrophy. Jenner ('95) observed a transparent appearance and indistinct cross 
striation of the muscle fibers, but no fatty degeneration. Banu ('21) finds the 


muscle fibers uniformly atrophied, with disappearance of the cross striations, 
increased distinctness of the longitudinal striations, multiplication of the muscle 
nuclei and increased connective tissue. 

In experimental rickets (or pseudorickets ?) of puppies, E. Voit ('80) found no 
significant change in the weight of the musculature. Aron and Sebauer ('08) 
noted that in experimental rickets in a puppy the musculature appeared less 
well developed than in the control, perhaps on account of lack of exercise in the 
former. In rachitic rats, Jackson and Carleton ('23) found a slight but progres- 
sive loss in the weight of the skeletal musculature, amounting to 12 per cent in 
the severely rachitic group. 

Vitamin Deficiency. — The effects of vitamin deficiency on the muscles have 
been noted chiefly for vitamins B and C. 

Vitamin B. Beriberi and Polyneuritis. — In human beriberi, Balz ('82) found 
nuclear proliferation and sometimes atrophy of muscle fibers. Scheube ('94) 
observed parenchymatous-fatty (sometimes colloid) degeneration of the muscle 
fibers, some fibers being atrophic and some hypertrophic; with some increase in 
the number of muscle nuclei and in stroma. Rumpf and Luce ('00) stated that 
the muscle lesions are not those ordinarily found in simple neuritis, but indicate 
a myopathic process, designated as polymyositis acuta parenchymatosa et chronica 
inter stitialis. Duerck ('08, '08a), however, considered the changes found in 
the musculature (atrophic degeneration, nuclear proliferation, etc.) as non- 
specific and identical with those occurring in other diseases. 

Kato and Shizume ('19), McCarrison ('19) and others have described the 
changes in the musculature of the chick and pigeon during polyneuritis gal- 
linarum. Findlay ('21) found considerable muscular atrophy and loss of cross 
striation. Funk ('22) also has recently reviewed the changes found in birds 
with beriberi. The muscle fibers "exhibit atrophy and fatty degeneration, 
but the changes disappear rapidly on returning to normal nutrition." 

Vitamin C. Scorbutus. — In human scurvy, the hemorrhagic tendency is 
manifested in the musculature as elsewhere. According to Sato and Nambu 
('08), the muscles, especially those of the lower extremities, show hemorrhages 
to a variable extent. The muscles also present edema and marked atrophy, 
in connection with the general cachexia. Histologically, the muscle tissue 
exhibits a myositis, with interstitial hemorrhages, atrophy and degeneration of 
the muscle fibers. The changes are also described by Aschoff and Koch ('19), 
who found the hemorrhages frequent near the muscular attachments, and in the 
tendons, fascia, etc. The musculature of the legs is affected most, the arms less 
and the trunk least, exposure to trauma being an important factor. Comrie 
1/20) noted marked atrophy of the muscles, with deep-seated hemorrhages in 
over half of the cases. Hess ('20) has recently summarized the various changes 
in human and animal scurvy, including degeneration of the muscle fibers, hemor- 
rhages and variable pigment deposits secondary thereto, and interstitial fibrosis. 

In experimental scurvy of guinea pigs, Hoist and Frolich ('07, '12) found 
intramuscular hemorrhages, especially in the neighborhood of the bones and 
joints, with microscopic changes similar to those in human scurvy. "Die 
Muskelfasern waren in grosser Verbreitung abnorm schmal und zeigten ofters 


einige fettige Degeneration. Auch bestand oft ein Zerfall in unregelmassige, 
hyaline Klumpchen, die zum Teil nicht in derselben Weise wie in normalen 
Fasern gefarbt wurden. Zwischen diesen Klumpchen bestanden hier und da 
kleine Ansammlungen von Sarkolemmkernen. Sonst wurde aber keine Ver- 
mehrung der Zellen bezw. eine rundzellige Infiltration nachgewiesen." 

In experimental scurvy of monkeys, Hart ('12) described typical intra- 
muscular hemorrhages, and also some peculiar granules, the staining reactions 
of which indicated a calcium content. 

Jackson and Moore ('16) stated that the degenerative lesions found in the 
muscles of the guinea pig in scurvy are independent of the intramuscular hemor- 
rhages. Hojer ('24) found hemorrhages, atrophy of the muscle fibers, and 
necroses with calcification. 

Aqueous Inanition. — Longet ('68) stated that the autopsy of a man after 
death from thirst shows disappearance of fat and marked emaciation in the 
musculature. Schuchardt ('47) in pigeons on dry diet with loss of 44 per cent 
in body weight noted an apparent loss of 37 per cent in the pectoral musculature. 
Nothwang ('91) in pigeons after death from thirst found the muscles apparently 
well preserved and dark red in color. Scheffer ('52) and Falck and Scheffer 
('54) found in a dog on dry diet an apparent loss of 20 per cent in body weight 
and of 29 per cent in the musculature. 

In histological structure, Pernice and Scagliosi ('95a) observed that in a dog 
subjected to a dry diet the muscle fibers are pale, poorly stainable, somewhat 
homogeneous in appearance, with indistinct striations. The fibrillae appear 
attenuated; the nuclei are numerous and some show mitosis. In chickens under 
similar conditions, the skeletal muscle shows occasional slight hemorrhages, with 
partial loss of striation. The interstitial stroma presents a round cell infiltra- 
tion. As previously mentioned, this occurs also in the muscles during rickets 
and scurvy, but rarely in total inanition. It may represent an inflammatory 
reaction to toxic substances in the circulation. 

The foregoing investigators found also a marked drying of the musculature 
which in many cases was measured by chemical analysis. Straub ('99) observed 
a loss of 20 per cent in the water content of the musculature in dogs subjected 
to aqueous inanition on a dry diet. Durig ('01) in frogs deprived of water by 
exsiccation found that the organs vary much in their loss in weight and in water 
content, the musculature losing most heavily. Similarly in dogs subjected to 
experimental diarrhea, with loss of 25-30 per cent in body weight, Tobler 
('10) noted that the skin and musculature suffer most, losing up to 50 per cent 
of their water. The water content of the viscera is much less affected. 

The loss of weight in the skeletal musculature of rats on a dry diet was noted 
by Kudo ('21, '21a), as shown in Tables 9 and 10. In adult rats in the acute 
thirst series with average loss of 36 per cent in body weight, the musculature lost 
33 per cent; in the chronic thirst series, the body lost 52 per cent and the muscu- 
lature 61 per cent. In young rats held at nearly constant body weight by a 
dietary deficiency of water for 1-13 weeks, there was a loss of about 5-7 per 
cent in the musculature of the various groups. 



In general the brain appears relatively resistant to the effects of both total 
and partial inanition. Usually little or no loss in weight or changes in gross or 
microscopic structure are apparent. In advanced stages of starvation, however, 
and especially in types of partial inanition (beriberi, pellagra) involving neural 
or psychic disturbances, there are well-marked degenerative changes in the 
nerve cells. After a brief summary, the effects of inanition upon the brain will 
be considered under (A ) total inanition, and (B) partial inanition. 

Summary of Effects of Inanition on the Brain 

The brain in general is extremely resistant to loss in weight during total inani- 
tion or on water alone. In adults, both human and infrahuman, even with a 
loss of 40 or 50 per cent in body weight, the loss in brain weight is below 10 per 
cent, usually below 5 per cent, and often shows no appreciable change. This 
is also true in general during the various forms of partial inanition; but in some 
cases there may be an actual increase in brain weight (with edema in pellagra, 
etc.) or a definite atrophy, with decreased weight (during protein deficiency, 
pellagra, etc.). 

In atrophic infants, the brain is capable of continued growth with retarded, 
or even stationary, body weight. The same is true in young animals, especially 
in newborn subjected to prolonged underfeeding, although in some cases (acute 
inanition, various forms of partial inanition, and at later ages) the brain weight 
may remain unchanged, or even show a slight loss. In human rickets the brain 
appears enlarged (often hydrocephalic), but in experimental rickets in animals 
it is usually normal in weight. After severe underfeeding, the brain may fail to 
grow properly upon subsequent ample refeeding. 

In structure, the brain during total inanition (or on water only) grossly 
appears normal, excepting a variable degree of congestion, especially in the 
meninges. Microscopically the white substance (medullated fibers) usually 
appears normal. Aside from a variable degree of hyperemia, the gray substance 
usually likewise shows no marked change, except in the nerve cells. Even most 
of these cells frequently appear normal, but there are often atrophic and 
degenerative changes which are extremely variable in different individuals, in 
different cells, and in different regions of the brain. As a rule, the changes 
appear well marked only in advanced stages of inanition. 

These brain cell changes, which have been extensively studied in various 
animals, are especially evident in some of the large cells (of Betz and Purkinje, 
etc.), although frequently more intensive in the smaller cells. The changes 




involve cell atrophy with progressive chromatolysis (of the Nissl substance), 
neurofibrillar degeneration, cytoplasmic vacuolation, rarely nuclear degenera- 
tion. "Steatosis" is not characteristic. Only slight changes appear in the 
brain cells during hibernation. In the young during inanition the normal 
developmental changes in the brain are largely arrested, but the degenerative 
changes are less conspicuous than in the adult. 

In the various forms of partial inanition, the structural changes of the brain 
in general resemble those described for total inanition, including a variable 
degree of congestion and of atrophic degeneration in the nerve cells. The 
changes are especially marked in beriberi, with the associated phenomena of 
paralysis; likewise in insane pellagrins, frequently involving karyorrhexis and 
complete cell disintegration. Sclerosis due to glial proliferation is more frequent 
than in total inanition. Hemorrhages occur in scurvy and beriberi. In general, 
it may be noted that the lesions of the nerve cells during the various types of 
inanition are not specific in character, but closely resemble those produced by 
toxic and other injurious agents. 

It is a remarkable fact that although the brain appears relatively resistant to 
the effects of inanition in general, it is, as pointed out by Clark ('23), more sus- 
ceptible than any other tissue to the effects of oxygen deficiency. 

(A) Effects of Total Inanition, or on Water Only 

Under this heading will be included the effects upon the brain, human and 
infrahuman, adult and young, as to (1) weight and (2) structure. 

1 . Effects on Brain Weight. — In the human adult, it has long been known 
that the brain is very resistant to inanition in general (Rokitansky '54), but 
exact quantitative data appear relatively scanty. Von Bischoff ('64) stated 
that during emaciation the brain does not share the loss in body weight, at least 
not to the same degree. 

Porter ('85— '87, '89) published average weights of the adult brain found in 
autopsies upon victims of the Madras famine of 1877-78, as shown in the 
accompanying table. 

Brain Weights in Victims of the Madras Famine (Porter '89) 




Condition of body 

No. of cases 


No. of cases 


Plump (normal) 













39 - 2 



Atrophic (greatly emaciated) 


This table would indicate in the extreme cases an apparent loss of about 9.8 per 
cent in the brain weight of the men, and 8.1 per cent for the women. All of these 



had probably lost over 40 per cent in body weight. Porter concluded that the 
brain wastes like other tissues, but to a lesser extent, "and no doubt to this is 
due the effusion of serum into the subdural and subarachnoid cavities so fre- 
quently found in these cases." The relatively slight loss in the dropsical cases 
was ascribed to edema in the brain substance. 

Marchand ('02) found no significant difference in brain weight as the result 
of inanition. Matiegka ('04), however, in adults from 20 to 59 years of age, 
obtained average brain weights as shown in the accompanying table. 

Average Brain Weight in Different Conditions of Nutrition (Matiegka '04) 

Nutritional condition 

21 Men, grams 

38 Women, grams 





1261 .0 

Comparing the extremes, Matiegka's data would indicate a loss in brain weight 
of about 7.3 per cent for the men, and of 5.6 per cent for the women. 

Krieger ('20) similarly compared the brain weights in several groups of 
emaciated adults with various norms, as indicated in the accompanying table. 
The apparent loss in body weight, estimated by comparing with Gartner's 
norm for body length, ranged from 36-48 per cent in the various groups. 
The loss in brain weight was estimated by comparison with the norms of (1) 
Marshall, based on age and height; (2) Bischoff, based on body weight; and (3) 
Marchand, based on age and height. 

Average Brain Weight in Various Conditions of Emaciation in Adults. All Males, 
Except as Indicated in Group I. From Autopsies in the Pathological Institute, 

Jena (Krieger '20) 


No. of 



Percentage deviation of brain weight 
from the norm of 




I. Insane. No chronic organic dis- 








1 ,216 



— 0. 16 

— 1 .02 
+ 0.8 
+ 0.36 

+ 0.6 

— 0.52 
+ 6.2 

— 0.22 

— 1.2 


ease \fem 





f -6.2 
t -3- 1 1 

1 Making allowance for the normal decrease with age. 

From these data, Krieger concludes that it is doubtful whether the human brain 
loses appreciably in weight during inanition, although comparison with Mar- 
chand's norm indicates a small but fairly constant loss of 2.8-5.6 per cent. 


Weber ('21) compared the weights of the brain in 1,257 autopsies among 
civilians at Kiel during the separate years 1914-1918. Body lengths are 
available, but not body weights. The brain weight averages 1,357 g- m the 
males, and 1,246 g. in the females, with no apparent difference between the 
period of good nutrition (1914-1915) and the period of subnutrition (1916- 

Brain Weight in Atrophic Infants. — Von Buhl ('6i) found in 52 cases the 
average brain weight of the newborn about 352 g. (range 193.5-482). In 
infants dying in the second or third week after birth, the average brain weight 
found was 41 1.5 g. "So ist damit unumstosslich dargethan, dass mit der Ab- 
nahme des Korpergewichts das des Gehirnes nicht oder doch am wenigsten 
abnehme." Parrot ('82) published a table showing the brain weight averaging 
286.7 g- m IO infants 1-7 days of age, with average body weight of 1,994 g.; 
and 359.7 g. in 26 infants 8-36 days of age, with body weight of 1,969 g. 
Manouvrier deduced from these data the independent growth of the brain during 
inanition in infancy; but he failed to exclude the possibility that in the latter 
group the brain during an earlier period of normal growth might have attained 
its greater weight, which was not lost subsequently, in spite of a possible decrease 
in body weight. 

Similarly Ohlmiiller ('82) observed that the brain and spinal cord combined 
weighed 528.8 g. in a well nourished infant of 56 days (body weight 4149.5 g.); 
while in an atrophic infant of 56 days (body weight 2,381.2 g.), they weighed 
only 480.9 g. This does not justify the conclusion that there has been a decrease 
in the weight of the brain and cord in the atrophic infant, because its previous 
body weight is not stated. Ohlmiiller cited data showing that in the newborn 
the brain weight averages 13 per cent of the body weight, increasing to 15-24 
per cent in emaciated infants up to 42 days old. In the absence of exact data 
concerning the previous body weights, however, it is impossible to draw any 
conclusions as to what changes in absolute brain weight have occurred during 
the period of inanition. Ohlmiiller found no appreciable difference in the 
water content. There was an apparent increase in fat content of the brain in 
the atrophic infants, but no appreciable difference in lecithin in the brain and 
cord of a starved puppy. 

Cantalamassa ('92) at the autopsies of twin infants which had died of star- 
vation in 11 and 23 days after birth, respectively, observed an unusual degree of 
overlapping at the sutures between the parietal and frontal bones. This he 
attributed to reduction in the volume of the brain, but it might equally well be 
ascribed to increased growth in the cranial bones, or to decrease in the cerebro- 
spinal fluid. Overlapping at the cranial sutures was also noted in athreptic 
infants by Thiercelin ('04), who also concluded that: "Le cerveau s'est atro- 
phic, le liquide cephalorachidien s'est en partie tari, et ce dessechement du 
contenu de la boite cranienne a mene une depression considerable des fonta- 
nelles f ormant une veritable cavite dont la profondeur peut atteindre 3 ou meme 
4 millimetres." 

This brings us to the work of Variot ('07) who first demonstrated that, in 
addition to the persistent skeletal growth, there is in infants during inanition an 



actual increase in the brain weight, which often appears to correspond to the 
age, rather than the body weight. He concluded that: 

"Les mensurations faites sur le crane des enfants hypotrophiques vivants, 
aussi bien que les pesees du cerveau apres la mort, semblent concordantes 
pour nous montrer que les processus hypotrophique ne s'exerce pas sur les 
centres nerveux comme sur le reste de 1'organisme." 

This conclusion was confirmed by Variot and Lassabliere ('09), who give 
data on 12 hypotrophic infants autopsied at 3-21 (average 13%) months of 
age. Compared with the normal for corresponding age, the brain weight 
averages nearly normal, ranging from 18 per cent below to 20 per cent above; 
whereas the corresponding height of the body averaged 10 per cent below nor- 

koo ^n — | 1 1 n ' — 

, : '| 

! i 1 1 : 1 


i i, 

1100 ' ! 

. _*J!_LjlJ_ L 

Ik 1 " / 

,i j Ml ' I'll 

1 • 1 ^ 

* S 

tii \j^*„ 

M i 

• • !i^r 

it y 

. ' ! .i ' ' ! J/T\ " 

■■if > 1 1 ' 

1 ! 

" 11 j > X ! * 

1 i ' ' : 

* ^-i^>^* : 

800 p | ■ 

^W*^ .'ii k 


* oJ$r'- I'M 


v WH it III 

^>T IT 1 

TOO •■«] I ,, " 

'-■tfr * i • 

It ' 1 

^ -j- X-i- T 

T ' " 



^ < 

000 >, , >' 

u- -IL X 

DP i • . S 

— ' ,rn» ^ 

\ * 'y 

n (i 

• • ^^ H ;, 

_l_ m 

400 - — m ' ' ' " — L>*^ 


i , , ^^ i t 

• ^1 ■ 

T \s m 

! i 

300 -j- **\\ — °~ 1 \ — 


j j) | 

?00 \ i o cd y Lcngrr 

$»■ 1 

42. 44 46 43 50 5£ 54 5b 56 60 

6£ 64 66 66 TO 72 T4 76 

Fig. 6o. — A graph showing the individual brain weights in emaciated, atrophic infants. 
The larger dots represent original Minnesota cases; the others are from various sources. The 
curve represents the normal brain weight, according to body length, from data compiled by 
Prof. R. E. Scammon. It is evident that inanition had little or no effect upon the weight of the 

mal, and the body weight 37 per cent below normal. Thus during inanition 
in infants the brain appears capable of continued growth at the expense of 
the remainder of the body. 

Variot's conclusion was also confirmed by the extensive data of Fayolle 
('10). In a series of 128 infants, he found the average brain weight in the 
hypotrophic 0.2 per cent above the normal for corresponding age; the body 
height being 7.2 per cent below, and the body weight 36.4 per cent below. In 
a second series of hypotrophies, the head circumference averages 0.01 per cent 
above normal for age (indicating normal size for the brain); the body height 
being 5.14 per cent below, and the body weight 24.19 per cent below. 


Lesage ('11) agreed that the brain in atrophic infants is the last organ to 
be affected, but states that if the inanition is prolonged the brain also is finally 

From head measurements on 125 children in Czerny's clinic, Sawidowitsch 
('14) concluded that "Ernahrungsschadigungen, welcher Artsieauchseinmogen, 
bewirken eine Hemmung in der Gehirnentwicklung." The changes in body 
length, body weight and brain volume appear to be independent of each other. 

Lesage ('14) found a brain weight of 460 g. in an atrophic infant of 4 
months, at which age the normal weight is 620 g. Nobecourt ('16) and 
Marfan ('21), however, support the doctrine that the brain weight in infants is 
largely independent of the body weight. This is likewise confirmed by the 
observations of Nicolaeff in famine-stricken children. 

Jackson ('22) studied the weights of the body and of various organs (see 
Tables 2 and 3) in about 50 atrophic infants in comparison with the norms for 
(1) final body weight; (2) maximum body weight; (3) body' length; and (4) age. 
In this series, the brain averaged about 26 per cent above the norm for final 
body weight; 1.5 per cent above the norm for the maximum body weight 
observed during life; 7.7 per cent below the norm for body length; and 12.3 per 
cent below the norm for age. Thus the brain weight averages slightly higher 
than that corresponding to the maximum body weight, but lags slightly behind 
that corresponding to the body length (which has been shown to increase during 
inanition). That the brain weight in emaciated infants in general averages 
approximately normal according to body length is apparent from the field 
graph shown in Fig. 60. Of the individual data represented, 25 (larger dots) 
are Minnesota cases; the others are from various sources. 

Brain Weight in Adult Animals. — The resistance of the brain toward inan- 
ition was discovered by Collard de Martigny (1828), who noted that the brain 
appeared unchanged in size in several starved dogs and rabbits. Chossat ('43) 
first studied the weight of the brain during inanition. In 10 pigeons on total 
inanition with loss of about 40 per cent in body weight, the brain weight aver- 
aged 2.25 g., which exactly equals that in 10 controls of the same initial 
body weight. Since it is difficult to determine precisely the line of separation 
between the brain and spinal cord, their combined weight was found to be 
3.08 g. in the controls and 3.02 g. in the test pigeons, giving an apparent 
loss of 1.9 per cent. The (combined) decrease in dry weight appears relatively 
greater, from 0.64 to 0.58 g., a loss of 9.4 per cent. Findlay ('21) recently 
likewise found practically no loss in the brain weight of starved pigeons and 
fowls (Table 13). 

In a cat losing about 50 per cent in body weight on total inanition, Bidder 
and Schmidt ('52) compared the weight of brain and spinal cord with that in 
a normal control of similar initial body weight. This gave an apparent loss 
of 37.6 per cent, which was evidently due to a large individual variation. Von 
Bibra ('54) found no essential change in the brain weights (fresh or dry) in 
starved rabbits. Valentin ('57) noted weights indicating a loss of about 5 per 
cent in the brain of marmots with loss of 35.5 per cent in body weight after 166 
days of hibernation. 



Voit ('66) observed an apparent loss of 3.2 per cent in the brain and spinal 
cord (combined) of a cat on total inanition with loss of 33 per cent in body 
weight. Manassein ('68, '69) in 47 adult rabbits with average loss of about 
39 per cent in body weight found an apparent gain of 3 per cent in the weight 
of the brain. In 2 cats there was an apparent loss of 10 per cent in brain 
weight, while in 2 crows the brain weight remained unchanged (with loss of 36 
per cent in body weight). 

Bourgeois ('70) gave no weights, but stated that the brain often appears 
entirely normal in starved animals (guinea pigs, rabbits, cats, dogs), contrast- 
ing strongly with the loss of weight in other organs. 

In the guinea pig, Lazareff ('95) found in guinea pigs (10 in each group) on 
total inanition with losses in body weight of 10, 20, 30, and 36 per cent corre- 
sponding apparent losses in brain weight averaging 1.5 1, 3.02, 5.54 and 6.05 
per cent, respectively. Sedlmair ('99) obtained an apparent gain in the brain 
weight in 2 starved cats, but the difference in the initial weights in comparison 
with the control makes conclusions uncertain. 

Donaldson ('n) stated that the relative weight of the central nervous 
system in the leopard frog {Rana pipiens) remains nearly constant during 
hibernation. Ott ('24) found that in this species the weight of the brain remains 
nearly constant during hibernation and subsequent inanition, although in 
the male during the later stages of inanition (with body loss of 50-60 per cent) 
there is an apparent loss of 15-22 per cent in brain weight (Table 6). The per- 
centage of dry substance remains nearly constant. 

Jackson ('15) in adult albino rats on water only (acute inanition series), 
with loss of about 33 per cent in body weight, noted an apparent loss of 5.1 per 
cent in brain weight. In the chronic inanition series (underfeeding), with 
loss of about 36 per cent in body weight, there is an apparent loss of 6.6 per 
cent in the brain weight. 

Brain Weight in Young Animals. — In 8 rabbits 3 months and 20 days old, 
with average loss of about 33 P er cent m body weight from starvation, Manassein 
('69) noted that there was an apparent gain of about 3 per cent in the brain 
weight, as in adults. In 3 rabbits only 23-25 days old, with loss of 35 per 
cent in body weight, there was an apparent increase of 7 per cent in brain 
weight, indicating a persistent growth tendency (unrecognized as such by 
Manassein, however). 

Von Bechterew ('95) found apparently a slight loss, or at least a greatly 
retarded growth, in the brain weight of newborn kittens and puppies on water 
only. Schukow ('95) confirmed and extended these results on puppies and 
chicks, concluding that there is not an actual decrease in the brain weight 
during inanition, but merely a retardation in growth. 

Hatai ('08) stunted 5 litters of albino rats by moderate underfeeding begin- 
ning at 30 days of age, so that at 170 days they averaged only 91.5 g. in 
body weight (controls reaching 146.5 g). In the test rats the brain 
weight appears normal for corresponding body weight. Donaldson ('n), in 
a series of 22 litters of albino rats held at nearly constant body weight by more 
severe underfeeding from 30-51 days of age, found the average brain weight 


7.7 per cent less than that in full-fed controls of the same age, but 3.6 per cent 
greater than the (calculated) initial brain weight. Aron ('n) noted that the 
brain weight in puppies underfed for long periods is nearly equal to that in 
the full-fed controls of the same age, but the lack of initial controls makes it 
impossible to estimate the amount of growth in brain weight during the under- 
feeding period. 

The changes in the brain weight of albino rats of different ages during 
various degrees of underfeeding were studied by Jackson and coworkers, the 
results being epitomized in Table 4. Jackson ('15a) observed practically no 
change in the average brain weight of rats underfed for various periods begin- 
ning at the age of 3 weeks or later. Stewart ('18), however, found that if the 
underfeeding was begun at birth, allowing the body weight to increase slowly 
from about 5 g. to 15 g. at 10 weeks of age, the brain weight increases 
to 8 per cent above the normal for corresponding body weight. If the under- 
feeding is more severe, retarding the body weight to only 10 g. the brain 
weight increases to 60 per cent above the normal. In still severer inanition, 
holding the body nearly constant at birth weight for about 16 days, Stewart 
('18a) found the brain weight 125 per cent above normal! Sugita ('18), in 
nursing rats retarded 29-39 per cent in body weight by underfeeding 3-4 
days, similarly found the brain 24 per cent above the standard for correspond- 
ing body weight. Thus at this early age the brain shows a most remarkably 
persistent growth tendency during inanition, which enables it (with certain 
other organs) to grow at the expense of the remainder of the body. This 
tendency is apparently not so strong in the fetus, however, since Barry ('20, 
'21) found the brain only 12.5 per cent above normal weight in full term fetuses 
which had been retarded 40 per cent in body weight by maternal underfeeding 
during pregnancy. 

Stewart ('18a) also studied the weights of the parts of the brain in albino 
rats held at birth weight (about 5 g.) for 5-18 days. The brain weight 
in this series averages 114 per cent above normal. In another series, the body 
weight was allowed to reach 10 g. at 3 weeks, and in these the brain weight 
appears 33 per cent above normal for corresponding weight. The weights of 
the various parts (cerebrum, cerebellum and brain stem) in general preserve 
approximately the same relative weight as in normal individuals having the 
same brain weight. The olfactory bulbs, however, appear hypertrophied in 
the younger group. 

Data published by Trowbridge, Moulton and Haigh ('18) and by Moulton, 
Trowbridge and Haigh ('22) indicate that the brain weight in underfed young 
steers of various ages is both relatively and absolutely somewhat higher than 
in full-fed controls of the same body weight. This would indicate that in 
the bovine species, as in man and the rat, the growth of the brain during incom- 
plete inanition appears relatively independent of the body as a whole. 

Recovery of Weight upon Refeeding. — In young albino rats amply refed 
after underfeeding from 3 to 12 weeks of age, the weight of the brain was 
observed to be normal by Stewart ('16). In rats underfed from birth for 
arious periods and then refed, Jackson and Stewart ( '18, '19) found that the 


brain does not preserve its relatively high weight, but instead soon lags behind 
in growth and appears below normal for corresponding body weight. As 
mentioned in Chapter IV, such rats were found by Jackson and Stewart ('20) 
to be permanently stunted in body weight. The brain weight in such dwarfed 
adults was found to be still slightly below the normal for body weight. This 
would seem to indicate that even though the brain grows persistently in weight 
during periods of severe subnutrition, it is nevertheless injured in some way so 
that it may not be able to recuperate fully when subsequently placed under 
good conditions of nutrition. 

2. Effects on Brain Structure.— The effects of total inanition, or on water 
only, upon the structure of the brain in human and infrahuman adult and 
young will now be considered. 

Human Adult. — Aside from occasional references to meningeal congestion 
or brain softening, no data concerning the effects of inanition upon the structure 
of the human brain are given by the earlier observers. Rokitansky ('54) 
stated that the nervous system, "so far as relates to its constituent elements, 
remains exempt," even during extreme general atrophy of the body. Cyr ('69) 
adopted Parrot's doctrine of an encephalic "steatosis" (to be mentioned later). 
Bright ('77) found, in a case of human starvation, the brain apparently normal, 
excepting a small tubercular area. 

Popow ('82, '85, '85a) appears to have been the first to report cytological 
changes in the human brain cells, in a case of starvation from esophageal 
stricture. Hemorrhagic extravasations, proliferation of the neuroglia and 
connective tissue, atrophy and cloudy swelling of the ganglion cells were 
observed. These changes were held to be the possible cause of the mental 
disturbances observed in this case. As late as 1889, Filipi adhered to Parrot's 
doctrine of encephalic steatosis, while Cohnheim stated that the tissue of the 
central nervous system is but very slightly affected by inanition. 

Tarassewitsch ('98) described the histological changes in the brain of a 
religious fanatic, aged 30 years, who died after 35 days of inanition. The 
Purkinje cells of the cerebellum show slight chromatolysis. The cells of the 
cerebral cortex present cytoplasmic vacuolation and pigmentation. The nuclei 
are usually peripheral in position, and stain diffusely, but are rarely changed in 
shape. Complete breaking down of the cells was not observed. The neu- 
roglia appears loose in texture and the blood vessels congested. Dreyfuss ('06) 
considered the changes produced by inanition in the nerve cells during insanity 
as secondary in character. Agostini and Rossi (07) described vacuolization 
of the nerve cells and changes in the neurofibrillae during inanition in the 

In a man who died of starvation, Meyer ('17) found the brain large 
(1600 g.) and edematous. The cells of the cerebral cortex appear shrunken, 
with wide pericellular spaces, and atrophy especially of the cytoplasm. In a 
victim of the Russian famine, Hassin ('24) found no macroscopic abnormalities 
in the brain and meninges. Microscopically some degenerative changes 
appeared, especially an accumulation of lipoids in the ganglion cells, glia, and 
blood vessels. 


Human Infants. — The first detailed account of the changes in the brain of 
infants during inanition was that of Parrot ('68), who described a "steatosis" 
(fatty degeneration) as characteristic for various organs in athreptic infants. 
In the brain and spinal cord, the arachnoid presents irregular, yellowish, opal- 
escent spots, due to fatty degeneration of the connective tissue cells. In the 
substance of the brain and cord, the neuroglia cells appear similarly infiltrated, 
forming fatty granulations, either microscopic or visible to the naked eye. 
The corpus callosum is most frequently affected. The cerebral vessels rarely 
show changes. In his classic treatise, Parrot ('77) described three chief lesions 
in the brain of athreptic infants — steatosis, hemorrhage and softening, none of 
which has been found characteristic by most of the later observers. Thiercelin 
('04) concluded that the cerebral symptoms (coma, strabismus, convulsions) 
are due to cerebral congestion and toxemia, rather than to softening or 

Tardieu ('80) mentioned cerebral and meningeal congestion among the 
signs of death from inanition in the newborn. Talbot ('09) thought the arrested 
development of the facial region and teeth in malnourished children may be 
due to interference with the blood supply on account of malnutrition of the 
brain, "since the brain presides over the development of the tissues." Moncke- 
berg ('12) described the cell changes in "pedatrophy" as primarily an inanition- 
atrophy affecting all the organs and tissues in various degrees. Nicolaeff ('23) 
sometimes observed hyperemia and increase of ventricular and subarachnoid 
fluid in the brains of famine-stricken children. 

Adult Animals. — Carville and Bochefontaine ('74, '75) noted that the 
meninges, brain and spinal cord appear congested in starved dogs. Falck ('75), 
on the other hand, found the meninges and brain anemic and firm, with a slight 
amount of fluid at the base of the brain and in the ventricles. 

Mankowsky ('82), during inanition in rabbits and dogs, found the meninges 
and brain pale and edematous, with loss of weight. The cerebral and spinal 
ganglion cells show atrophic degeneration, with vacuolation, pigmentation, and 
fatty degeneration; some appearing shrunken and granular. The white sub- 
stance and neuroglia are unchanged. The changes are sometimes general, 
sometimes in localized, softened areas. Two dogs were refed to normal weight 
after a long fast. In one the nervous system appeared anemic; in the other, 
hyperemic. The ganglion cells showed fatty degeneration, but not the atrophy 
found in the starved animals. 

Rosenbach ('83, '84) described marked changes in the nerve cells of the 
brain and spinal cord in starved dogs, although the white substance and stroma 
are but slightly affected. The nerve cells react differently in different regions. 
The spinal ganglion and anterior horn cells suffer most, with marked atrophy, 
vacuolation and albuminous degeneration. The posterior horn cells show only 
cloudy swelling. The cerebellar cells rank next, with shrinkage and vacuolation 
of the Purkinje cells. The cells of the cerebral cortex show but slight traces of 
cloudy swelling. The pyramidal cells rarely present coagulation necrosis and 
vacuolation, but the round cortical cells are more strongly affected. The 
nuclei in these cells appear less resistant, and show granules, but the nuclear 


contour is unchanged. Rosenbach believed that the nerve cells which appear 
latest in their embryonic development show the greatest resistance to inanition. 
Further details in the nerve cell changes were described bv Rosenbach ('84a, 

Ochotin ('86) studied the changes in the central nervous system in human 
starvation, also especially in rabbits subjected to various degrees of incom- 
plete total inanition. The changes in the brain were found similar to those in 
the spinal cord, which will be described in the next chapter. 

Coen ('90) studied the changes in 3 rabbits and a kitten after death from 
total inanition or on water only. The cerebral cortex shows notable atrophy 
of the nerve cells, with extensive pericellular spaces. The cytoplasm is much 
reduced, but the nuclei remain unchanged. The layer of small pyramidal 
cells appears most affected; next come the large pyramidal cells. The nerve 
fibers show fine, glistening granules, and free droplets of myelin occur, but the 
neuroglia appears normal. No marked vascular changes occur in the cortex 
or meninges. 

Peri ('93) made a careful study of the nervous system in rabbits, cats and 
dogs after starvation with loss in body weight up to 45 per cent. He used vari- 
ous fixatives and stains, including those of Weigert, Marchi, Golgi, etc. The 
changes, both macroscopic and microscopic, in various regions of the brain and 
cord were found less marked than those described by previous investigators. 
Venous stasis, diapedesis and slight edema were noted, but no "steatosis." 
The. nerve cells usually appear unchanged, a few showing a slight degree of 
atrophy but never marked degeneration. The silver methods of Golgi reveal 
nothing abnormal. The changes found in the spinal cord and sciatic nerve 
will be mentioned later. Recovery of the brain cells in dogs refed after fasting 
was described by Lubimoff ('94). 

Monti ('95, '95a) obtained more definite changes in the brain cells of fasting 
rabbits, using the osmic-bichromate and Golgi methods. Some cells remain 
normal, but others show a "varicose atrophy" with degenerative changes begin- 
ning peripherally on the dendrites and extending toward the cell body. This 
may finally become involved, but the axone and nerve fiber are not injured. 
The number of cells affected and the extent of the degeneration vary greatly 
in different regions. In the cerebellum, the Purkinje cells are more resistant 
than the small cells of the molecular layer. 

Bich ('95), in a study of the retinal changes in starved dogs, noted that the 
meninges appear normal; the brain substance anemic and always edematous. 

Ganfini ('97), using the Nissl method, found no appreciable change in the 
brain cells of fasting rabbits, aside from slight decrease in staining ability. The 
changes in the spinal cord will be mentioned later. 

Lugaro and Chiozzi ('97) used both Nissl and Golgi methods on the brain and 
cord of fasting dogs and rabbits. In general, changes appear in the nerve cells 
only in the later stages of inanition, and vary greatly in the site and the intensity 
of the lesions. The spinal ganglion, Purkinje and cerebral cortical cells are 
among the first affected, the anterior horn cells being the most resistant. The 
Nissl substance in the affected cells gradually undergoes chromatolysis, but 


the achromatic parts and the nucleus are affected only in later stages. Degenera- 
tion of the dendrites was not observed. Since the cytological changes resemble 
toxic lesions, they concluded that the effects of inanition may be through 
autointoxication. Soukhanoff ('98, '98a), however, suggested that the toxic 
agents may interfere with the cellular nutrition, which might explain the simi- 
larity of toxic and inanitional lesions. 

Daddi ('98, '98a) also found that in fasting dogs the nerve cell changes 
are in general slight, appearing in but few cells and in late stages of inanition. 
The chromatic (Nissl) substance undergoes a variable degree of chromatolysis; 
later the achromatic portion becomes degenerated and vacuolated. In general, 
the lesions appear more pronounced in the cerebrum, cerebellum and spinal 
ganglia than in the brain stem and spinal cord; the spinal ganglion and Purkinje 
cells being affected more than the cerebral. Daddi opposes the autointoxication 
theory of Lugaro and Chiozzi. Changes similar to those described by Daddi were 
found by Puglisi-Allegra ('00) in fasting dogs and guinea pigs. Marinesco 
('00) and Muhlmann ('10) claimed that chronic nutritional disturbances of 
the nerve cells in the brain and cord result in pigment formation. 

Marchand and Vurpas ('01) in fasting rabbits and guinea pigs found no 
appreciable changes in the cerebellum. The lesions in the cerebral cortex appear 
similar to, but slighter than, those in the spinal cord, which will be given in the 
next chapter. Geeraerd ('01) found the chromatolysis in the cortical cells of 
the guinea pig slighter in prolonged inanition than in fatigue. Weygandt ('04) 
briefly describes cortical lesions produced in mice by starvation or insomnia. 
Panella ('06) found a decrease in the nuclein content of the brain in fasting dogs. 
Donaggio ('06, '07), using the silver method, found the neurofibrillae in the 
nerve cells of adult rabbits very resistant to either inanition or cold alone, but 
markedly affected by their combination. In general, the fibrolysis is less easily 
produced than chromatolysis. Coarser bands appear in the fibrillar network, 
probably by fusion of the fine neurofibrillae. The extent and character of the 
changes vary in different individuals, and also in different parts of the central 
nervous system. In the cerebral cortex, the lesions are less intense than in the 
cerebellum, brain stem and spinal cord. Vacuolation of the cells frequently 
occurs. Marinesco ('06, '09) described a hypertrophy of the neurofibrillae in 
nerve cells of kittens subjected to inanition, especially in combination with 
cold or intoxications. 

Riva ('05, '07) found the neurofibrillae of the nerve cells in fasting dogs and 
rabbits in general very resistant to inanition, but the appearance may be greatly 
changed by the cytoplasmic vacuoles, which may displace and modify the neuro- 
fibrillar network. If the vacuoles are small or absent, the network remains 
normal. Balli ('07) produced marked lesions in the neurofibrillar network by a 
combination of inanition and thyro-parathyroidectomy. 

In cats subjected to starvation, Beeli ('08) observed that in spite of the failure 
of the nervous system to lose in weight, degenerative changes occur in the nerve 
cells. In the cerebellum, the Purkinje cells undergo progressive shrinkage and 
vacuolation. The changes are less marked in the cerebral cortex. The gray 
and the white substance contrast sharply in gross appearance. 


Following the subcutaneous transplantation of the cerebellum in white mice, 
Laignel-Lavastine and Jonnesco ('12) described a lipoidal degeneration in the 
Purkinje cells, which may be in part due to imperfect nutrition. Sundwall 
('17) found congestion but slight cytological changes in the cerebral cortex of 
starved white rats. Asada ('19) made similar observations on fasting rabbits 
and Findlay ('21) in birds (pigeons and fowls). The papers of Damlevski 
C91) and Frankenberger ('17) were inaccessible. The effects of inanition 
and other abnormal conditions upon nerve cells in general were reviewed by 
Barbacci ('99), Robertson ('99), Marinesco ('09) and Bardier ('13). 

The changes in the nerve cells during hibernation have been studied by 
several investigators. Querton ('98) claimed that there is a partial retraction 
of the cortical neurones during hibernation, but this has not been confirmed. 
Chromatolysis with changes resembling those described in the nerve cells during 
starvation were found by Legge ('99) and by Barconcini and Beretta ('00) in 
the bat and other hibernating mammals. Cajal ('04) verified the observation 
of Tello that in reptiles coarse longitudinal neurofibrillae form during hiberna- 
tion, but split up into finer fibrillae upon the awakening. Rasmussen and 
Myers ('16) found no significant change in the chromophilous (Nissl) substance 
in the woodchuck (Marmota monax) during hibernation. According to their 
review of the literature, Legge, Baroncini and Beretta, and Marinesco ('05) 
found marked changes in the Nissl substance during hibernation, while Levi 
('98) and Zalla ('10) observed no chromatolysis. Rasmussen ('19) also found 
that " Complete inanition for 3 months during winter sleep and for 3 weeks after 
waking does not modify the morphology, number or distribution of mitochondria 
in nerve cells" (of cerebellum, spinal cord and spinal ganglia). 

Young Animals. — A few observations are available concerning the effects 
of inanition upon the structure of the brain in young animals. Von Bechterew 
('95) found the brain softened and hyperemic (especially the gray matter) in 
starved kittens and puppies. He concluded that: 

"Die mikroskopische Untersuchung des Gehirnes der verhungerten Neuge- 
borenen zeigte ausser den Gewebsveranderungen, welche dem Hungertode 
iiberhaupt eigenthiimlich sind und in ausgepragte Coagulationsnecrose und 
dem Myelinzerfall in den Markscheidenhaltigen Fasern bestehen, eine Verspat- 
ung in der Entwickelung und Markscheidenbekleidung derjenigen Systeme, 
welche bis zum Antritt der Hungerperiode noch unentwickelt waren." 

The results of von Bechterew were confirmed and extended by Schukow 
('95), who found that the retardation of growth in brain weight of fasting new- 
born puppies is due to failure of myelinization in tracts which had not already 
begun to myelinize. Since in dogs the brain fibers are the latest to myelinize, 
these parts are most retarded during inanition. In chicks 15-20 days old, the 
brain weight is much less retarded. Intensive coagulation necrosis occurs in 
the brain cells of fasting newborns. On refeeding, the lesions vanish and the 
brain recovers in weight and development. If the newborn puppies are merely 
undernourished by restricted nursing, the same pathological changes occur, but 
they develop much more slowly. 


Reinke ('06) found that larvae of Salamandra maculata after successive 
exposures to 4 per cent ether solution may live up to 152 days, but take no food. 
The medullary tube becomes distended and thin- walled in 10 days; many cells 
perish, others undergo mitosis (up to 80 days) . The final result is a partial regen- 
eration of the central nervous system, which is nourished at the expense of the 
remainder of the body, especially of the musculature, which becomes strongly 
atrophied. Reinke further states: "Lasst man Salamanderlarven zunachst 
hungern und futtert sie dann sehr stark mit lebenden Wurmern (Naiden), so 
treten etwa nach einer Woche eine enorme Menge von Mitosen in fast alien 
Organen auf, aber ganz regelmassig schubweise." 

Sugita ('18), in the underfed young albino rats previously referred to, 
concluded from a histological study " that by starvation in the early days the 
brain suffers much in its development in toto, but the cell division is going on 
quite normally according to age. The growth of the cells in size is retarded and 
the formation of myelin fibers somewhat diminished by inanition. S.o the smaller 
weight and size of the underfed brain is due to an arrest in the growth and 
development of the constituent neurons and not to a decrease in their number." 

Changes in Chemical Composition. — Although the chemistry of inanition is 
outside the scope of the present work, it may be noted incidentally that the 
brain undergoes relatively slight chemical changes during inanition. Its fat 
(lecithin) belongs to the phosphorized lipins, which in general are very resistant 
to starvation. In addition to those found in the papers above cited, data on 
the chemical changes in the brain during inanition are contained in the works of 
Aeby ('75), Pfeiffer ('87), Lukianov ('88, '89), Tonninga ('93), Voit ('94), 
Herter ('98), Durig ('01), Roger ('07), Donaldson ('n), and Donaldson, Hatai 
and King ('15). 

(B) Effects of Partial Inanition 

The effects upon the brain by various forms of partial inanition, including 
dietary deficiencies in protein, fat, salts, vitamins and water, will now be 

Protein Deficiency. — Paltauf ('17) found the brain weight unaffected in 
cases of human malnutritional edema and allied conditions. 

In a series of young albino rats with loss of about 30 per cent in body weight 
(initial weight 40-101 g.) on a nearly protein-free diet of starch, suet and 
water, Hatai ('04) found an apparent average decrease of about 5 per cent in the 
absolute weight of the brain. No marked alteration occurs in the Nissl 
substance. In a similar series of rats amply refed after the inanition period, 
Hatai ('07) obtained prompt recovery in the body weight and brain weight, 
although certain changes in the chemical composition of the brain (higher water 
content and lower percentage of ether-alcohol extracts) still persisted. 

In albino rats held at nearly constant body weight for prolonged periods on 
various incomplete protein diets, Osborne and Mendel ('n, '11a) held that the 
normal proportions in the various parts of the body were maintained, excepting 
the possibility of a continued growth of the central nervous system, for which 
some evidence was cited. 


Frohner and Zwick C15) noted that the brain (like the body in general) 
appears edematous in cattle on watery, low-protein diets. Koch and Voegtlin 
('16) found loss in the weight, and changes in chemical composition, of the body, 
brain and spinal cord in monkeys and rats on diets of corn-oil cake, corn meal 
and sweet potatoes, etc. (mixed deficiency). They conclude: "This is, we 
believe, the first recorded instance in which such an atrophy of the central nerv- 
ous system has been produced in animals by a change in diet." In monkeys and 
pigs on these diets, Sundwall ('17) found a variable degree of meningeal conges- 
tion, chromatolysis and other cell changes in the cortex, with occasional degen- 
eration of fibers in the internal capsule. In albino rats stunted in growth by an 
inadequate maize diet (mixed deficiency) after weaning, Holt ('17) noted a 
slight increase in the brain weight. The relative weight of the olfactory bulbs 
appears reduced (cf. opposite result by Stewart ('18a) on rats underfed from 
birth), but the number of nerve cells therein is not affected. 

Pellagra. — The enormous literature on pellagra (assumed to be due primarily 
to protein deficiency) contains many observations upon the associated changes 
in the brain, which are of especial interest on account of the frequency with which 
insanity follows pellagra. Only a few of the papers can be considered here, 
however. For more complete review of the pathology of pellagra, with exten- 
sive bibliographies, the works of Marie ('08, '10), Raubitschek ('15), Sundwall 
('17), and Harris ('19) may be consulted. 

The pioneer in this field wasLombroso ('69), who emphasized the importance 
of the brain lesions in pellagra, including thickening and opacity of the meninges, 
edema and softening of the cerebrum, atrophy of the cerebral cortex, abundant 
corpora amylacea, and fatty and pigmentary changes in the cortical brain cells. 
Fraenkel's ('69-'7o) extensive data were chiefly from Lombroso's clinic. Fraen- 
kel in 113 autopsies found meningeal thickening in 33, purulent exudate in 4, 
subarachnoid hemorrhages in 5, marked brain edema in 24, atrophy of cerebral 
cortex in 11. The brain weight was variable; in 28 cases, 7 appeared increased 
and 8 decreased. 

Tuczek ('93), however, in a careful study of 8 cases of pellagra, found nothing 
abnormal in the cerebral cortex, cerebellum, pons and medulla, although the 
nerve cells of the cranial nerve nuclei appear richly pigmented. Rossi ('98) noted 
cytoplasmic vacuolation, pigmentation and disintegration of the Nissl bodies, 
sometimes also nuclear displacement and degeneration, in the cerebral cortical 
cells of pellagrins. These findings in the brain cells were confirmed and extended 
by Babes and Sion ('00), and several other investigators cited by Harris ('19). 
Parhon and Papinian ('05) and others demonstrated lesions in the neurofibrillae 
of the pyramidal cells in the cerebral cortex, especially in the large Betz cells. 
Marinesco ('09) found that the brain cells undergoing chromatolysis also show 
pigmentary degeneration. The neuroglia proliferates. He found, as an excep- 
tion, that the cells of Purkinje remain normal; but Harris ('10) noted marked 
degenerative changes in the Purkinje cells in one case. Hamill ('12) observed 
a variable chromatolysis in the large pyramidal cells of the cortex. 

The changes in the brains of insane pellagrins were thoroughly studied by 
Kozowsky ('12), who found the meninges variably hyperemic and thickened. 


The brain is always markedly sclerosed in all parts, sometimes edemic. There 
is proliferation of the neuroglia, and the corpora amylacea probably represent 
altered glia cells. The nerve cells of the brain all show a variable degree of 
change. Pigmentation is the commonest (also found in the sympathetic 
ganglia). The tigroid (Nissl) substance undergoes granular disintegration, 
first in the circumnuclear zone; later the whole cell becomes homogeneous, 
sometimes greatly vacuolated; the nucleus may be displaced and finally dis- 
appears; ultimately the whole cell may disintegrate into several small masses. 
The degree of cell change is proportional to the length of the period of insanity. 

Mott ('13) also found degenerative changes with a variable degree of chroma- 
tolysis and lesions of the neurofibrillae in the Betz and Purkinje cells in a case of 
pellagra. The changes are less intensive than those found in the spinal cord (to 
be mentioned later). 

Fat Deficiency. — In several rats which had been fed by McCollum on lipoid- 
free rations, with retarded growth of the body, Hatai ('15) found the brain and 
spinal cord each apparently reduced about 2 per cent in weight, the gray sub- 
stance being affected more than the white. In 45 human autopsies in cases of 
edema apparently due to insufficient food, especially poor in fat and protein, 
Prince ('21) observed no macroscopic lesions in the nervous system. 

Salt Deficiency. Rachitis. — In human rickets, the occurrence of hydro- 
cephalus was noted by Whistler (1645) ar >d Glisson (1650). Seibold (1827) 
noted an increase in the fluid of the brain ventricles (also subdural), and an 
enlargement of the Pacchionian bodies. Comby ('oi ) found a variable tendency 
to hydrocephalus in rachitic infants, and Stoeltzner ('03) mentioned mild hydro- 
cephalus as a possibly significant condition. Cheadle and Poynton ('07) 
likewise believed that the brain may be involved in rickets. Pfaundler ('22) 
states that in human rickets the brain is enlarged, as though swollen; vascular 
congestion causes ventricular dropsy, and the softened cranium permits hydro- 
cephalic enlargement. Karger ('20) claims that the enlargement of the rachitic 
brain is not due to hydrocephalus. "In den bekannten grossen rachitischen 
Schadeln findet sich in der Regel kein Hydrozephalus, sondern ein abnorm 
grosses Gehirn; in diesem sind bisher mikroskopisch keine wesentlichen Abweich- 
ungen von normalen nachgewiesen worden und die Versuche, diese Frage auf 
chemischen zu beantworten, haben bis jetzt zu eindeutigen Ergebnissen nicht 

Neurath ('24) reviews the neurologic changes in rickets, especially the 
cerebral changes, which are associated with mental retardation in infants. 

In dogs and pigeons on low salt diets, Forster ('73) observed weakness, 
trembling and paralysis, but no macroscopic lesions of the central nervous 
system at autopsy. In puppies on calcium-poor meat and lard diet, producing 
a rachitic (or pseudorachitic) condition, Voit ('80) found the brain weight nearly 
normal or possibly (as he thought) above normal. Quest (,'06) demonstrated 
that the brain in such rachitic puppies shows an increased water content, but 
normal calcium content. Jackson and Carleton ('23) found the brain weight 
nearly normal in a large series of albino rats in various stages of experimental 


Vitamin Deficiencies. — Aside from rickets, which (as stated in Chapter V) 
probably involves a vitamin factor in addition to the mineral deficiency, the 
brain has been studied in beriberi (vitamin B deficiency) and scurvy (vitamin C 
deficiency). Meyerstein ('22) made a few observations on the brain in young 
white rats on diets deficient in vitamins A and B. Lopez-Lomba ('23) found 
the brain unchanged in weight in pigeons on a vitamin-free diet. 

Beriberi. — The conspicuous neural symptoms in beriberi clearly indicate 
lesions of the nervous system. These have been found chiefly in the spinal 
cord and peripheral nervous system, but Rumpf and Luce ('00) cited observa- 
tions by themselves and previous investigators indicating hyperemia and edema 
of the brain, and occasionally slight hydrocephalus internus, in human beriberi. 
In experimental beriberi of pigeons, chicks, cats, dogs and white mice, no definite 
change was found in the nervous system. 

Walshe ('18, '20) has recently reviewed the literature and finds that "Since 
the original investigations of Baelz, Scheube, Pekelharing and Winkler (1882- 
1887) there has been a striking unanimity among pathologists that the nervous 
lesion of beriberi is not specific for the disease, and is not to be distinguished from 
that of a toxic polyneuritis." Walshe agrees with Eijkman that even though 
beriberi be due to a vitamin deficiency ("Teilhungertheorie"), the ultimate 
cause may yet prove to be a nervous poison produced by a disordered metabo- 
lism arising from vitamin deprivation. 

McCarrison ('19, '19a, '21) believes that absence of the so-called "anti- 
neuritic" factor, vitamin B, leads to functional and degenerative changes, not 
only in the central nervous system, but also in every organ and tissue in the 
body. In pigeons on a diet of autoclaved rice with butter and onions (deficient 
in protein as well as in vitamin B), the brain shows an increase of 14 per cent in 
weight, although this increase does not appear in polyneuritic pigeons on rice 
diet alone. According to Findlay ('21) the brain in avian beriberi, though 
nearly constant in weight (Table 13), shows a loss of Nissl granules and a 
decrease in nucleic acid content, both of which are restored upon administration 
of vitamin B. Hofmeister ('22) found in severe beriberi of rats marked lesions 
with hemorrhages in the cerebellum and brain stem, which are proportional to 
the nervous symptoms, and lead to degeneration of the brain cells. As no 
evidence of a degeneration in the peripheral nerves was found, the condition in 
rats is considered not a polyneuritis but rather a cerebral purpura similar to the 
hemorrhagic encephalitis which occurs in chronic poisoning with alcohol, lead or 
arsenical compounds. 

Scorbutus. — In scurvy, there are in general no specific changes in the nervous 
system, aside from occasional hemorrhages, which Sato and Nambu found in the 
brain once in 6 cases. There are no definite changes in the nerve cells or fibers, 
according to the literature reviewed by Hess ('20). Bessesen ('23), however, 
found an apparent average increase of 10-12 per cent in the weight of the 
brain in scorbutic guinea pigs (Table 12). This is due to the fact that the loss 
in brain weight is relatively less than the loss in body weight. 

Aqueous Inanition. — Schuchardt ('47) noted an apparent loss of 6 per cent 
in the weight of the brain and cord in pigeons with loss of 44 per cent in body 


weight on a dry barley diet. In a dog with loss of 20 per cent in body weight 
after 4 weeks on dry food, Falck and Scheffer ('54) found an apparent gain of 7.2 
per cent in the weight of the brain, in comparison with a control; but the water 
content remained normal. Bowin ('80), in rabbits and dogs on a dry diet with 
marked loss in body weight, apparently obtained an increase in the relative 
weight and the water content of the brain, but the exact data are lacking in the 
abstract by Muhlmann. 

Pernice and Scagliosi ('95, '95a) studied the effects of a dry diet upon a dog 
and 3 half-grown chicks, which died with loss of 24-41 per cent in body weight. 
In the dog, the meninges and brain were strongly congested. The various 
layers of nerve cells and fibers in the cerebral cortex in general appear cloudy, 
poorly staining with carmine, and often distinctly atrophic. The fibers stain 
poorly by the Pal-Weigert method. The cerebellum also appears markedly 
hyperemic, with degenerative changes especially in the superficial molecular and 
the Purkinje cell layers. In the chicks, somewhat similar changes were found. 
Atrophy and cloudy degeneration are especially evident in the large pyramidal 
cells. Similar atrophic and degenerative changes appear in the cells of the 
cerebellar cortex and in the gray substance of the medulla oblongata. 

The effects of a dry diet upon the brain weight in albino rats were noted by 
Kudo ('21, '21a), as shown in Tables 9 and 10. The brain was often found 
markedly congested, especially in the adult. In the adult acute thirst series, 
with average loss of 33 per cent in body weight, the brain weight remains 
unchanged; while in the chronic thirst series, with loss of 52 per cent in body 
weight, the average loss in brain weight is 4.2 per cent. In young albino rats 
held at nearly constant weight on relatively dry diets from one month of age for 
various periods (up to 13 weeks), the brain weight shows no significant change. 



The spinal cord, like the brain, appears relatively resistant to the effects of 
inanition, although the nerve cells show marked degenerative changes in certain 
types of partial inanition (pellagra, etc.). After a brief summary, the effects of 
inanition upon the spinal cord will be considered under (A) total inanition, and 
(B) partial inanition. 

Summary of Effects on the Spinal Cord 

In general, as might be expected, the changes in the spinal cord during inani- 
tion resemble those in the brain, although differing in some details. In weight 
the adult spinal cord suffers little or no loss, the decrease usually being less than 
10 per cent, even at death from either total or partial inanition. In young 
animals during inanition the spinal cord, like the brain, exhibits a persistent 
growth with retarded or even declining body weight. This may occur during 
partial inanition (scurvy, aqueous inanition). Upon ample refeeding after a 
period of inanition, the spinal cord usually recovers its normal proportionate 
weight, although after severe or prolonged inanition in young animals the later 
growth of the spinal cord upon refeeding may be subnormal. 

The structural changes in the spinal cord are also in general comparatively 
slight, especially in the white substance, although congestion is frequent and 
variable changes in the gray substance appear, especially in the later stages of 
inanition. These changes resemble those found in the brain, sometimes being 
less intensive but often more so. During total inanition (complete or incomplete) 
the nerve cells, especially in the anterior horn, may undergo atrophic degener- 
ation with a variable degree of cytoplasmic vacuolation, chromatolysis, 
neurofibrillar derangement, and nuclear enlargement or pycnosis, rarely kary- 
olysis or karyorrhexis. In extreme cases, some cells may degenerate and 
disappear entirely; while others show comparatively slight changes. The 
medullated fibers may show diffuse degenerative changes, probably secondary 
to the disturbances in the nerve cells. Proliferation of the neuroglia may occur. 
The changes during hibernation are comparatively slight. 

In the various forms of partial inanition, the structural changes in the spinal 
cord in general resemble those during total inanition, with certain special 
features in addition. In protein deficiency (including pellagra) the congestion 
is greater and degenerative changes in the nerve cells are usually more intensive, 
involving pigmentary degeneration and frequently total necrosis and dis- 
integration, especially in some regions. There are also more extensive 
degenerative changes in the white substance, especially in the posterior and 



lateral columns, with associated sclerosis, due to proliferation of the neuroglia. 
Ependymal proliferation often obliterates the central canal. In beriberi and 
scurvy, the changes in the cord are slight, but in aqueous inanition (thirst) the 
degenerative changes, like those in pellagra, are markedly intensive. The 
structural changes found during the various types of inanition are not specific, 
but resemble those produced by various toxic agents and other abnormal 

(A) Effects of Total Inanition, or on Water Only 

These effects will be considered under (1) changes in weight and (2) changes 
in structure. 

1. Changes in Weight. — For the changes in weight of the human spinal 
cord during inanition, no data have been found in the literature. In 4 cases 
observed by me, in which the spinal cord was weighed in atrophic infants (see 
Table 3), there appears to have been little if any loss in weight in this organ; 
but conclusions are uncertain on account of the small number of cases and the 
lack of an adequate norm for comparison. 

In Adult Animals. — Chossat ('43) found in starved pigeons an apparent 
decrease in average weight of the spinal cord from 0.83 g. to 0.77 g. or a 
loss of 7.2 per cent (body loss about 40 per cent). In 3 marmots hibernating 
about 166 days, with average loss of 35.5 per cent in body weight, Valentin 
('57) noted an apparent decrease of only about 3 per cent in the weight of the 
spinal cord. Voit ('66) observed an apparent loss of about 3 per cent in the 
combined weight of brain and cord in a cat starved with loss of 33 per cent in 
body weight. Manassein ('68, '69) found the average weight of the spinal 
cord practically unchanged in 11 adult rabbits starved with loss of about 39 per 
cent in body weight. In a normal control dog of 15.4 kilograms body weight, 
studied by Voit ('94), the spinal cord weighed 22.6 g. while in a test dog 
reduced by starvation from 17.4 to 11.78 kilos the spinal cord weighed 23.5 g. 

In 4 groups of guinea pigs (10 in each group) on total inanition with average 
losses of 10, 20, 30 and 35.5 per cent in body weight, Lazareff ('95) found an 
apparent increase of 1.05 per cent in the average weight of the spinal cord in the 
first group, and losses of 6.82 per cent in each of the other 3 groups (Table 5). 
Sedlmair ('99) apparently obtained a slight increase in the weight of the spinal 
cord in 2 starved cats, but the difference of initial body weights in comparison 
with the control makes conclusions uncertain. Jackson ('15) found the 
average weight of the spinal cord practically unchanged in adult albino rats in 
the acute inanition series (on water only), with loss of 33 per cent in body weight; 
while in the chronic (incomplete total) inanition series, with loss of 36 per cent 
in body weight, there is an apparent average loss of 4 per cent in the weight of 
the spinal cord. Ott ('24) observed in frogs but slight loss in the weight of the 
spinal cord during hibernation and with losses in body weight up to 40 per cent. 
With losses of 50 and 60 per cent, however, there was an apparent loss of 14 and 
25 per cent, respectively, in the male frogs (Table 6). In general, therefore, as 
in the brain, there is little or no loss in the weight of the spinal cord during 
starvation, excepting extreme stages. 


In young animals, the spinal cord was observed by Bechterew ('95) to increase 
in weight in fasting newborn kittens and puppies, even with decrease in body 
weight. Hatai ('08) found the spinal cord weight apparently normal for the 
body weight in albino rats underfed from 30 days of age to 170 days, the body 
weight being retarded to 91.5 g. (controls of same age reaching 146.5 g.). 
Donaldson ('11a), however, noted an apparent increase in the spinal cord of 
albino rats held at about 34 g. in body weight from age of 30 days to 51 days. 

The changes in the weights of the spinal cord in young albino rats underfed 
at various ages by Jackson and his associates are shown in Table 4. In rats 
held at constant body weight by underfeeding from 3 to 10 weeks of age (or 
later), Jackson ('15a) found an increase of about 36 per cent in the weight of the 
spinal cord. In rats underfed from birth, Stewart ('18, '19) obtained still more 
intensive growth of the spinal cord, which averaged 70 to 83 per cent above the 
normal for corresponding body weight. In the newborn offspring retarded by 
maternal underfeeding during pregnancy, however, Barry ('20, '21) did not find 
this intensive growth, the spinal cord being nearly normal in weight. 

The weight of the spinal cord was found above the normal for corresponding 
body weight in young steers with growth retarded for long periods on subnormal 
rations by Trowbridge, Moulton and Haigh ('18), and Moulton, Trowbridge 
and Haigh ('22, '22a, '22b). This would indicate that in the bovine species, 
the spinal cord, like the brain and skeleton, shows a persistent growth on a low 
plane of nutrition. 

Effects of Refeeding. — In young rats amply refed after underfeeding from 3 
to 12 weeks of age, Stewart ('16) found that the spinal cord had returned to its 
normal weight in proportion to the body within two weeks. In rats underfed 
from birth for various periods (resulting in relative hypertrophy of the spinal 
cord), and then amply refed to body weights of 25 to 75 g., Jackson and Stewart 
('19) found the spinal cord 7.5 to 11.2 per cent below normal weight after the 
longer underfeeding periods, indicating an inhibitory after effect of inanition, 
similar to that previously mentioned for the brain. Likewise in young rats 
permanently stunted by early or long periods of underfeeding, Jackson 
and Stewart ('19a, '20) found that the spinal cord failed to reach a weight 
proportional to that normally found at corresponding body weight. 

2. Changes in Structure. Human Adults. — Ochotin ('86) described the 
nerve cells in the spinal cord -of a very emaciated young man (with mandibular 
necrosis, anemia, chancroids, etc.). The cells appeared normal in form and 
size, but mostly cloudy or granular; nuclei variable, sometimes normal or even 
enlarged, rarely appearing filled with lymphoid corpuscles (nuclear 

In a woman in whom death from chronic starvation was suspected, Placzek 
('98, '99) found the spinal cord macroscopically normal, but microscopically 
showing variable atrophic degenerative changes in the anterior horn cells. In 
the earlier stages of degeneration the Nissl bodies appear slightly irregular and 
the nucleus peripherally placed; in advanced stages the Nissl bodies are 
destroyed, the cytoplasm atrophic, and the nucleus shows pycnosis or karyor- 
rhexis. In the cervical and thoracic regions of the cord, many of the anterior 


horn cells are reduced to small structureless remnants or have totally 
disappeared. The Marchi method also shows diffuse degeneration in the nerve 
fibers of the white substance. 

In a man who died from starvation, atrophic changes were likewise found by 
Meyer ('17) in the anterior horn cells. "The nuclei are small and frag- 
mentation of some nuclei apparently has taken place in these cells, and consider- 
able loss of material has occurred. Some of the cells are mere remnants and all 
are surrounded by very wide clear zones. Clear areas are also scattered 
throughout the gray substance." The fiber tracts of the cervical cord and a 
dorsal nerve root "contain many more neuroglial cells than are normally present. 
Considerable shrinkage and vacuolation of the myelin are present, and the cross 
sections of the fibers are irregular in outline." 

In the human infant, Parrot ('68, '77) described as characteristic for athrep- 
sia a steatosis (fatty degeneration) in the meninges and neuroglia of the spinal 
cord, as previously mentioned for the brain. Thiemich ('00) found in the 
spinal cord of atrophic infants by the Marchi method variably degenerative 
changes in certain tracts, but no correlation of these lesions with clinical nervous 
symptoms was evident. 

In adult animals, certain effects of inanition upon the structure of the 
spinal cord were mentioned in the preceding chapter in connection with similar 
changes in the brain, as observed by Mankowsky ('82), Rosenbach ('83, '84), 
Lugaro and Chiozzi ('97), and Daddi ('98, '98a). Rosenbach found the degen- 
erative changes more apparent in the nerve cells of the spinal cord than in the 
brain cells; while Lugaro and Chiozzi and Daddi found the anterior horn cells 
more resistant, especially with reference to the Nissl granules. 

Carville and Bochefontaine ('74, '75) noted congestion of the meninges, 
brain and spinal cord in a starved dog. The fat around the spinal cord, like 
the orbital, appears transformed into a gelatinous, amorphic mass. Schulz 
('84) maintained that the vacuoles described by Rosenbach in the nerve cells 
during starvation are artefacts. 

Popow ('85) described marked changes in the spinal cord of starved rabbits 
(total inanition), including proliferation of the neuroglia, congestion and 
sometimes hemorrhage. The nerve cells show variable degenerative characters; 
some atrophic, with scanty granular or homogeneous cytoplasm (the granular 
sometimes more peripheral and the homogeneous circumnuclear), or even naked 
nuclei. In other cases the nucleus appeared more active, with " Kernfiguren," 
occasionally double or multiple (fragmented?) nuclei. 

Ochotin ('85, '86) studied rabbits subjected to various degrees of complete 
or incomplete total inanition. In both types the changes appear greatest in 
the lower and upper parts of the spinal cord. Congestion and hemorrhage are 
conspicuous in the incomplete (chronic) inanition, but not in the complete. 
In animals killed after loss of 10 to 13 per cent in body weight during complete 
inanition, or 14-30 per cent during incomplete inanition, the nerve cells show 
beginning degeneration with whitish appearance. In more advanced stages, 
the cells show variable degenerative changes. Some are greatly enlarged, 
with disintegrated or obliterated nuclei (sometimes multiple) and vacuolated, 


poorly-staining cytoplasm. Others show cloudy or granular cytoplasm, the 
nuclei staining well with carmine. The coagulation necrosis and plasmatic 
exudate described by Rosenbach were not found. 

Downarowitsch ('92) was apparently the first to make systematic nuclear 
measurements of the nerve cells, in 3 rabbits killed after 8 days of total inanition 
with loss of about 40 per cent in body weight. In these and 3 normal controls, 
the cervical and lumbar enlargements of the cord were fixed in mercuric chloride 
and the long and short axes of the nuclei of 600 anterior horn cells measured in 
stained sections. From the average measurements, the corresponding volume 
of the nucleus (assumed to be an ellipsoid) was calculated to be 11 15.97 cubic 
micra in the normal, and 832.47 cubic micra after inanition, indicating a decrease 
in volume of 25.4 per cent. The nucleolus was also measured, and its average 
volume (considered spherical) apparently loses 42.5 per cent during inanition. 

Peri ('92, '93), carefully studied various regions of the central and peripheral 
nervous system by various methods (Weigert, Marchi, Golgi, Cajal, etc.). 
Grossly only venous stasis and mild edema were found, never the "steatosis" 
described by Parrot ('68), Filipi ('89) and others. Histologically, but very 
slight changes appear in the rabbits, probably on account of the short duration 
of inanition (3-5 days, with water); in the cats (fasting 15 days, with loss of 43- 
45 per cent in body weight) the changes were somewhat greater; and in the dogs 
(fasting 27-34 days, with loss of 43-44 per cent) the changes were greatest. 
But even in the dogs most of the nerve cells in the brain and spinal cord appear 
normal, relatively few showing degenerative changes. In the anterior horn 
cells, hyalin degeneration occurs rarely, with nuclear disappearance; but the 
Marchi and silver methods show nothing abnormal. The neuroglia appears nor- 
mal, but diapedesis is somewhat frequent. Peri ascribed the degenerative 
appearances described by Mankowski, Rosenbach, Popow and Coen to the 
inadequacy of the older technique employed. 

Barrows ('98), using Hodge's method measured the nerve cells of the spinal 
cord, spinal ganglia and occipital cortex in 3 starved rats with normal controls, 
and found: 

"(1) A decided shrinkage in size of the cells and nuclei in the famished 
animals, averaging about 20 per cent, and a still greater shrinkage in the 

" (2) An evident exhaustion of the substance of the famished cells, as shown 
by their faint staining with osmic acid and the notable absence of nuclei and 
nucleoli. The protoplasm of these cells shows a very fine vacuolation, not so 
marked as that described by Rosenbach for starving animals and by Hodge for 
extreme fatigue. In the brains of famished rats the pericellular lymph spaces 
are considerably enlarged." 

About this time, the Nissl method was introduced and was applied by several 
investigators to the nerve cells of animals subjected to inanition. Tauszk 
('94), a pioneer in this field, found by this method that chromatolytic changes 
appear more distinct in chronic than in acute inanition, and were most apparent 
in the cervical region of the spinal cord in rabbits. Ganfini ('97), in rabbits 
killed after 5-7 days of inanition, found the most marked changes in the anterior 



group of the anterior horn cells (Fig. 61). In these cells the Nissl granules 
undergo chromatolysis, and the nucleus appears swollen. Schaffer ('97) noted 
that the chromatolytic changes in the anterior horn cells of fasting rabbits are 
more intensive, and accompanied by greater vacuolation of the cytoplasm, in 
total inanition, than with water only. The nucleus also becomes deeply stained 
throughout. Jacobsohn O97) failed to confirm Schaffer, however, and 
Placzek ('99) found no nuclear changes, but merely chromatolysis (also some 
degeneration in the posterior column by the Marchi method). The work on 
nerve cell changes during inanition up to 1898 was reviewed by Barbacci ('99), 
Robertson ('99), and more recently by Marinesco ('09). 

Further studies on the anterior horn cells of fasting rabbits and guinea pigs 
were made by Marchand and Vurpas ('01), using various stains (picrocarmine, 
haematoxylin, Nissl's, Pal-Weigert, Marchi and Golgi methods). The changes 

are described in 3 stages, with 
intermediate forms: (1) cell size 
unchanged; cytoplasm becomes 
decolorized, with pale Nissl gran- 
ules ; nucleus central and unstained , 
with distinct nucleolus; (2) cell 
somewhat shrunken, with fewer 
and thicker processes; Nissl bodies 
finely granular; nucleus eccentric 
and irregular in form, deeply stain- 
ing with nucleolus still distinct; (3) 
Fig. 61. — Nerve cells from the ventral horn of preceding changes more extreme, 

the spinal cord in the rabbit, stained by Nissl's •.■■ • , n .-,. ■. 

method, a. cell from normal control; b, c, cells show- Wlth "Tegular Cell Outline, shorter 

ing the progressive disintegration of the Nissl bodies and less numerous Cell processes; 

after 5 to 7 days of total inanition. (Ganfini '97.) 1 1 . • • 1 

deeply-staining, non-granular, vac- 
uolated cytoplasm; nucleus atrophied and deeply staining or disappeared; no 
cell pigment. The lesions appear by various stains, best by Nissl's method. 
The neuroglia and medullated nerve fibers appear normal. 

Holmes ('03) described chromatolysis, vacuolation and nuclear swelling 
in the nerve cells of the spinal cord in frogs subjected to exhaustion and 
inanition; and Mourre ('04) similarly found chromatolysis in the nerve cells of 
guinea pigs. Gurewitsch ('08) found chromatolysis to a slight degree in starved 
rabbits; but more pronounced in dogs, surviving for a longer period. The 
changes found by various investigators in the nerve cells during hibernation 
will be mentioned later 

We come now to the era of investigation of the intracellular neurofibrillae. 
By means of a modified silver method, Cajal ('04a) and Dustin ('06) demon- 
strated marked changes in the nerve cells of leeches starved 2 months or more. 
There is partial degeneration and resorption of the neurofibrillar network; the 
fibrillae thicken, the meshes become narrower, and finally the nucleus breaks up 
into irregularly scattered, deeply- staining granules. A similar thickening of the 
neurofibrillae probably occurs in the nerve cells of adult mammals (dog and rab- 
bit) subjected to cold, especially in combination with inanition. 



Donaggio ('06, '07) likewise obtained modifications of the endocellular 
network of neurofibrillae in the nerve cells of rabbits under the combined influ- 
ence of cold and inanition, neither factor alone appearing effective. In the 
anterior horn cells, the fine fibrillar network presents coarser bands, possibly 
formed by fusion of the neurofibrillae. Riva ('05, '06, '07) also found that inani- 
tion alone usually produces relatively slight changes in the neurofibrillae of the 
brain, spinal cord and spinal ganglia in fasting dogs and rabbits (Fig. 62). 
The endocellular network may be considerably modified, however, in cases where 
large vacuoles cause mechanical displacement of the neurofibrillae. The results 
of Donaggio and Riva were confirmed and extended by Gurewitsch ('08a) in the 
spinal cord of fasting rabbits and dogs, and by Mattioli ('io) in rabbits. The 

Fig. 62. — Nerve cells from the ventral horn of the spinal cord in the dog, showing the 
variable changes in the neurofibrillae by Donaggio's silver method after a loss of about 50 per 
cent in body weight from total inanition, a, a cell in which the neurofibrillae have retained 
nearly normal structure; b, cell showing some vacuolation in the perinuclear region; c, cell with 
disorganized neurofibrillae crowded into bundles between the numerous large cytoplasmic 
vacuoles. (Riva '06.) 

literature on these effects of inanition upon the nerve cells in general is fully 
reviewed by Marinesco ('09). 

In Necturus maculatiis starved 4 months the spinal cord was found nearly 
normal by Smallwood and Rogers ('n). After 16 months, however, the spinal 
cord appears greatly reduced in size, with atrophy especially of the gray sub- 
stance. Changes found in the spinal ganglia will be mentioned in the next 
chapter. In the nerve cells of the spinal cord in Triton cristatus fasting 1-6 
months, Frankenberger ('17) described nuclear changes resembling pycnosis. 

In starved albino rats (on water only), Sundwall ('17) noted marked conges- 
tion and edema of the spinal cord; anterior horn cells swollen, vacuolated and 
chroma toly tic, with vesicular or pycnotic nuclei. There is diffuse degeneration 
of the various nerve fiber tracts, especially in the posterior columns. 

Changes during Hibernation. — Since hibernation represents a special type 
of inanition, the corresponding changes in the nerve cells of the spinal cord may 
be separately considered. (The observations on the brain cells during hiberna- 
tion by Querton '98,Legge '99, BaronciniandBeretta'ooand Marinesco '05 were 
mentioned in Chapter X.) First as to the Nissl substance, no appreciable change 


in the anterior horn cells was found in the hibernating hedgehog (Erinaceus) by 
Jacobsohn ('97) and Levi ('98). The results of Zalla ('10) were negative for 
the dormouse (Myoxus glis), and inconstant for amphibia (Rami, Bufo, Bombi- 
nator). Rasmussen and Myers ('16), who reviewed the literature in detail, could 
find no change in the hibernating woodchuck (Marmota monax). On the other 
hand, definite seasonal changes in the Nissl substance were found by Biihler 
C98) in the frog; likewise by Levi ('98) in Rana,Bufo and Zamenis viridis (adder). 
Chromatolytic and other nerve cell changes during hibernation were observed 
by Legge C99) in hibernating bats {Vespertilio murinus, Rinolophus ferrum 
equinum, etc.) ; by Baroncini and Beretta ('00) in Myoxus, Vespertilio and Vesper- 
ugo; by Marinesco ('05) in Erinaceus; by Cutori ('07, '08) in Testudo graeca; 
and by Zalla ('10) in reptiles (Lacerta, Zamenis, Tropidonotus). 

The changes in the neurofibrillae during hibernation may be due to the com- 
bined effect of cold and inanition. Tello ('03) found in dormant lizards the 
appearance of unusually coarse neurofibrillae in the motor cells of the anterior 
horn, but not in the brain cells. Cajal ('04 '04a) and Dustin ('06) obtained 
similar results in leeches and in mammals exposed to cold, especially during 
fasting. The coarse fibrillae apparently split up again upon the awakening and 
resumption of activity. Marinesco ('05) noted similar results in young cats and 
dogs, but no change in the hibernating hedgehog (Erinaceus). Cutore ('07, 
'08), however, found somewhat different changes in the anterior horn cells of 
Testudo graeca during hibernation, the peripheral zone of cytoplasm becoming 
vacuolated and the neurofibrillae more attenuated. Rossi ('10, '10a) noted the 
thickening of the neurofibrillae in the hibernating adder (Zamenis) and dormouse 
{Myoxus). Zalla ('10) also obtained the characteristic changes of the neuro- 
fibrillae in reptiles, and also (less markedly) in mammals (Myoxus). These 
changes in the neurofibrillae apparently have no definite relation to the changes 
in the chromophile (Nissl) substance, however, and represent a non-specific 
reaction to various pathological conditions. The conflicting results obtained 
by various investigators doubtless depend chiefly upon differences in species, in 
the technique employed, and in the stage of hibernation studied. 

As mentioned in Chapter X, Rasmussen ('19) found no change in the mor- 
phology, number or distribution of the mitochondria in the nerve cells of the 
cerebellum, spinal cord or spinal ganglia in the woodchuck (Marmota monax) 
after 3 months of hibernation and even after 3 weeks of further inanition upon 

The experiments of Reinke ('06) on the regeneration of the central nervous 
system of etherized (fasting) Salamander larvae were mentioned in Chapter X. 
Rossi ('10, '10a) studied the phenomena of regeneration in the spinal cord of the 
adder (Zamenis viriflavus) and the dormouse (Myoxus glis) during hibernation. 
The regenerative process appears to be somewhat retarded, especially in the 
cold-blooded animals, as shown by the Cajal method. 

(B) Effects of Partial Inanition 

The effects of partial inanition upon the spinal cord have been observed 
chiefly in protein deficiency (pellagra), vitamin deficiency (beriberi and scurvy), 


and water deficiency. A slight relative atrophy of the brain and spinal cord of 
albino rats on a lipoid-free ration was observed by Hatai ('15), as mentioned in 
Chapter X. 

Protein Deficiency.— Hatai (07) found that in albino rats 1 month old, 
subjected to a starch-fat diet for 3 weeks (with marked loss in body weight) 
and then amply refed on mixed diet, the spinal cord and brain, as well as the 
body, recovered normal weight, though certain differences in chemical composi- 
tion appear. Koch and Voegtlin ('16) observed marked loss in weight of the 
spinal cord and brain, as well as of the body, in monkeys and rats on protein- 
poor diets of corn-oil cake, corn meal, sweet potatoes, etc. (mixed deficiency). 
In these monkeys, Sundwall ('17) found meningeal congestion, degeneration of 
Burdach's column in the cervical region and of Goll's column in the dorsal region, 
with swelling and chromatolysis of the cells in the anterior horn and spinal ganglia. 

Pellagra. — Although degenerative lesions of the spinal cord in pellagrins 
were noted by Lombroso ('69), the first detailed account was given by Tonnini 
('83, '84). In 51 cases, he found the meninges anemic in 8; opaque and thick- 
ened in about half; calcareous infiltration of the arachnoid in 27. These changes 
appear rare in the cervical region. The cord is usually asymmetrical and ane- 
mic ; hyperemia and softening were found in 20 cases. In 13 cases studied micro- 
scopically, great pigmentation of the anterior and the posterior horn cells 
appears in 8; and granulo-pigmentary degeneration is frequent. Cell atrophy 
was noted once in the cervico-dorsal, and twice in the lumbar region. Degenera- 
tion was found twice in the lateral column and once in the posterior. Bel- 
mondo ('89) examined 8 cases and likewise found in severe pellagra a sclerosis 
of the posterior and lateral columns, involving the crossed pyramidal tracts, 
together with changes in the pia and gray substance (chiefly pigmentary 
atrophy). Corpora amylacea occur and the central canal may be closed by 
ependymal proliferation. Similar changes were found in 8 cases by Tuczek 
('93) in the posterior and crossed pyramidal tracts, with variable atrophy of the 
nerve fibers and glial fibrosis. The intensity of the lesions, which are sym- 
metrical, decreases from below upward. The central canal is obliterated, but 
the meninges, nerve roots and gray substance appear normal. 

Marie ('94) concluded that the lesions in the spinal cord are due to a polio- 
myelitis posterior, with associated degeneration of the endogenous fibers of the 
posterior and lateral columns. 

Using the Nissl method, Rossi ('98) found a variable degree of chromatolysis 
in the spinal cord of pellagrins, affecting either the periphery or the entire nerve 
cell. The cells show variable deformity, sometimes pigmented, sometimes homo- 
geneous or vitreous in appearance. The nuclei may be peripheral or disappear. 
The dendrites may be greatly altered or absent; and the entire cell may disinte- 
grate. Rossi's results were confirmed and extended by Babes and Sion ('00), 
who concluded that the changes in the white substance are chieflv or entirely 
exogenous in origin, opposing the endogenous theory of Tuczek and Marie. The 
degenerative changes were traced from the dorsal roots into the cord, with result- 
ant changes resembling tabes. Changes also occur in the cells of the gray 
substance, especially in Clarke's column and the anterior horn, with a peculiar 


proliferation of neuroglia around the degenerating nerve cells. Further investi- 
gations on the results by Nissl's method are cited by Marie ('08, '10). Roberts 
('12) described the degenerative changes in the nerve cells as shown by Nissl's 
technique and by other methods. 

Parhon and Papinian ('05) applied the modified silver technique and found 
neurofibrillar lesions in the nerve cells of the spinal cord, most pronounced in 
the cervical region. In the lumbar and sacral regions the central neurofibrillae 
are most affected, those in the cell periphery and processes being more resistant. 
These results were confirmed by Valtorta ('12), Bravetta ('11), Rezza ('12) and 
others (cited by Harris ? io). 

Harris ('10) found sclerosis of the posterior and lateral columns in 4 out of 5 
cases of pellagra, with constant degenerative changes in the nerve cells, similar 
to those described by earlier investigators. Anderson and Spiller ('11) con- 
cluded that in pellagra "The degeneration is caused by some toxic or infectious 
substance affecting all parts of the cerebrospinal axis, producing cellular degen- 
eration and diffuse degeneration of nerve fibers in the posterior and anterolateral 

In the spinal cord of 16 insane pellagrins, Kozowsky ('12) found the typical 
changes in the nerve cells, especially of the anterior horn and Clarke's column, 
and mostly in the middle and lower segments of the cord. The lateral pyra- 
midal tracts are most frequently involved; the posterior columns next. There 
is increased fibrosis along the blood vessels, which may be obliterated; some- 
times passive congestion and small hemorrhages occur. Similar lesions were 
described by Hamill ('12), Mott ('13), and other investigators cited by Raubit- 
schek ('15) and Harris ('19). 

Vitamin Deficiencies. — The vitamin deficiencies in which the spinal cord has 
been studied concern chiefly vitamin B (in beriberi) and vitamin C (in scurvy). 

Beriberi. — Rumpf and Luce ('00) reviewed the literature indicating that 
previous investigators of human beriberi found no significant changes in the 
spinal cord, aside from occasional atrophy of the anterior horn cells. Rumpf 
and Luce found: "einen sparlichen Ausfall sowie eine unbedeutende Degenera- 
tion der Vorderhornganglienzellen in alien Ruckenmarkssegmenten." Duerck 
('08) also made an extensive review of the subject (bibliography of 245 titles) 
and described 11 original cases of human beriberi. The changes in the nervous 
system, including the spinal cord, are variable in degree and not specific in 
character, resembling the degenerations due to toxic causes. 

Experimental Beriberi. — In avian polyneuritis, Vedder and Clark ('12) by 
Nissl's method found absence of the tigroid bodies in the cells of both ventral 
and dorsal horns of the spinal cord (Figs. 63, 64). The stainable substance is 
massed at one side of the cell and the nucleus sometimes stains poorly. The 
mitochondria appear normal in these cells, however, even when the tigroid 
bodies are markedly altered. Schnyder ('14), however, could find no appreci- 
able change in the structure of the spinal cord in various animals (white mouse, 
pigeon, chick, cat, dog) dying from beriberi. He used various histological 
methods, including the Pal-Weigert. Findlay ('21) noted nearly complete 
disappearance of the Nissl granules in avian beriberi (pigeons and fowls). 



In pigs placed on a ration of wheat, etc., Hart, Miller and McCollum ('16) 
obtained nervous symptoms and lesions resembling those of beriberi. The 

Fig. 63. — Nerve cells from the ventral horn of the spinal cord in fowls (Giemsa blood stain). 
The nucleus, n, is indicated in each, a, cell from normal animal, showing Nissl bodies; b, c, 
cells from fowl with polyneuritis (beriberi) after 24 days on polished rice diet. The nucleus 
has degenerated; the Nissl bodies have disintegrated, and the stainable substance is collected 
in irregular masses at one side of the cell. (Vedder and Clark '12.) 

Fig. 64. — A portion of a cross section of the ventromarginal column of the spinal cord in a 
fowl with polyneuritis (beriberi) on a polished rice diet. Stained with hematoxylin and acid 
fuchsin. /, /, nerve fibers in advanced stages of degeneration. (Vedder and Clark '12.) 

Fig. 65. — Sciatic nerve fiber (teased preparation) from fowl with polyneuritis (beriberi) on 
polished rice diet. Shows advanced degeneration, with disintegration of the axone, a. Hema- 
toxylin and acid fuchsin stain. (Vedder and Clark '12.) 

anterior horn cells and nuclei appear shrunken and degenerated; the Nissl 
granules indistinct and the cytoplasm homogeneous. They considered the condi- 


tion probably due to some toxic substance in the wheat rather than to the 
absence of any dietary factor, however. 

Scorbutus. — In human scurvy, lesions of the spinal cord appear rare. 
Sato and Nambu ('08) found nothing abnormal by Marchi's method. Feigen- 
baum ('17) observed hemorrhages in the spinal cord, however, and Hess ('18a) 
described focal degeneration in the lumbar cord, involving mainly the anterior 
horn cells, in a case of infantile scurvy. For experimental scurvy, no data have 
been found in the literature, excepting those of Bessesen ('23) who found a 
marked apparent increase in the weight of the spinal cord in scorbutic young 
guinea pigs (Table 12). This is explained in part by the corresponding loss in 
body weight. 

Aqueous Inanition. — As to the effects of thirst (dry diet) on the weight of the 
spinal cord, Falck and Scheffer ('54) found that in a dog with loss of about 21 
per cent in body weight, the spinal cord apparently lost 7.1 per cent. Kudo 
('21, '21a) studied the effects of a dry diet in albino rats (Tables 9 and 10). 
In adults, the spinal cord in the acute thirst series shows an apparent average 
loss of 1.8 per cent (body weight loss 36 per cent); while in the chronic thirst 
series it lost 6.7 per cent (body weight loss 52 per cent). In young albino rats 
(1 month old) in which the body weight was retarded for various periods by a 
dry diet the spinal cord shows a marked and progressive increase in weight, rang- 
ing from 20.8 to 53.1 per cent. This resembles the increase found in young rats 
during general underfeeding (incomplete total inanition) as previously mentioned. 

The structural changes in the spinal cord produced by dietary deficiency of 
water were studied by Pernice and Scagliosi ('93, '95, '95a). In a clog, the 
cord presents capillary congestion. The nerve fibers appear variably atrophic, 
especially near the gray substance. There is a distinct decrease in the size 
and number of the nerve cells, especially in the anterior horns. Many cells 
have entirely disappeared; others have degenerated into an amorphic mass of 
fatty granules. The stroma is increased in amount and the central canal 
widened. In 3 chicks, the pia mater appears thickened, rich in nuclei, with a 
few sub-pial hemorrhages. There is congestion, glial proliferation and atrophy 
of the nerve fibers and cells (Fig. 66), as in the dog. In general, the cells are 
small, poorly stained, and often without processes. Irregular, vacuolated or 
granular masses, or empty cavities, are found replacing degenerated cells, 
especially in the anterior horn of the lumbar region. The gray substance 
shows whitish-yellow, granular areas of softening, especially near the central 
canal, which in the lumbar cord appears enlarged and contains blood, etc. 

In rabbits dehydrated by various methods, Brasch ('98) found variable 
changes by the Nissl method, especially in the nucleus of the nerve cells of 
the spinal cord and spinal ganglia. The nuclear changes present two types: 
(a) nucleus rather small and pycnotic (sometimes karyorrhexis); (b) nucleus 
large, with stellate masses of variable stainability around the nucleolus. Transi- 
tional forms also occur. Brasch concludes that the retraction of the nuclear 
contents from the nuclear membrane is not a necrobiotic, but a purely physical 
phenomenon, caused by the dehydration and capable of recovery when the 
normal water supply is restored. 



As might be expected, the peripheral nervous system resembles the central 
nervous system in its notable resistance to inanition, likewise in showing a 
marked susceptibility to certain types of partial inanition (pellagra, beriberi, 
thirst). The effects of inanition upon the peripheral nervous system will 
first be summarized briefly, and then considered in detail under (^4) total inani- 
tion, and (B) partial inanition. 

Summary or Effects on the Peripheral Nervous System 

In general, the peripheral nervous system, like the central, appears relatively 
resistant to inanition. The nerve cells, however, in both spinal and sympathet- 
ic ganglia, may show progressive degenerative changes, including cytoplasmic 
atrophy, vacuolation and chromatolysis as well as nuclear changes, similar to 
those found in the central nervous system. The spinal ganglion cells and 
stroma in amphibia contain small fat droplets, which are chiefly resorbed during 
inanition. The nerve cell changes in general appear variable in the different 
types of partial inanition, being most pronounced in pellagra, in which pigmen- 
tary atrophy is frequent. 

Although the medullated nerve fibers are also resistant, the myelin sheath 
may show a slight degree of atrophy, especially in extreme stages of inanition. 
Wallerian degeneration may also occur, probably secondary to the above 
mentioned degenerative changes in the nerve cells. The nerve endings in 
muscle appear but slightly affected. Atrophy and degeneration of the periph- 
eral nerves, including cranial, spinal and sympathetic, are also found variably 
developed in the different types of partial inanition. They occur in pellagra, 
and are especially characteristic of beriberi (excepting in mice and rats), 
although they may be slight or absent unless the disorder is somewhat prolonged. 
In aqueous inanition, a notable decrease in the weight of the sciatic nerve trunk 
has been demonstrated in adult rats, which may be due partly to the loss in the 
included adipose tissue. In the young rat, there is during thirst a persistent 
growth in weight of the sciatic trunk, similar to that occurring in the spinal cord 
during inanition. 

During inanition the perineurium and endoneurium may undergo prolif- 
eration, and the resulting fibrosis tends to replace the loss in substance due to 
atrophy) of the neurones. In certain types of partial inanition (pellagra, 
beriberi and thirst), the changes may resemble those of a chronic neuritis, with 
round cell infiltration. 

In this connection it may be recalled (as shown in Chapter III) that a 
relatively marked resistance of the nervous system to inanition was likewise 



observed among invertebrates; in Planaria by Schultz ('04), Stoppenbrink 
('05), Berninger ('n) and Lang ('12); in Lineus by Nusbaum and Oxner ('12); 
in leeches by Cajal ('04a) ; and in Limax by Smallwood and Rogers ('08, '09, '10). 
So far as known at present, we may therefore conclude that in general through- 
out the animal kingdom the nervous tissue appears relatively resistant to 
inanition although subject to degenerative changes, especially in the extreme 

(.4) Effects of Total Inanition, or on Water Only 

In man, but relatively few observations upon the peripheral nervous system 
during inanition are available. Rokitansky ('54) held that the elements of 
the nervous system are exempt from general atrophy, but Luciani ('89, '90) 
believed the rapid decline toward the end of starvation to be due to disorganiza- 
tion of the nervous system. The neurasthenia so frequently found during 
undernourishment, as noted by Blanton ('19) in German school-children and by 
Rubner ('19) in adults, is probably due primarily to central rather than periph- 
eral nervous lesions. Blaschko ('83) did not find a primary degeneration in 
the plexuses of Auerbach and Meissner during infantile intestinal atrophy, 
although such degeneration is mentioned by Baginsky ('84). Meyer ('17) 
found increased neuroglia with considereabl shrinkage and vacuolation of the 
myelin in the nerve fibers of a cervical nerve root in a man who died of starvation. 

The observations upon the lower animals are more numerous. Carville 
and Bochefontaine ('75) found the sciatic nerve fibers apparently normal in 2 
dogs starved 27 days (on water only). Degenerative changes in the spinal 
ganglion cells, more or less resembling those in the nerve cells of the brain and 
cord, were found by Mankowski ('82), Rosenbach ('83, '84), Lugaro and Chiozzi 
('97), and Daddi ('98, '98a), as mentioned in Chapter X on the brain. Changes 
observed by Barrows ('98) in spinal ganglion cells, and by Cajal ('04a) and 
Dustin ('06) in the neurofibrillae of nerve cells in the leech were mentioned in 
Chapter XI on the spinal cord. The observations by Riva ('05, '06, '07) on 
the neurofibrillae and by Rasmussen ('19) on the mitochondria of spinal 
ganglion cells during inanition were also mentioned in Chapter XL 

Microphotographs showing in general but slight changes in the spinal 
ganglion cells of starved rabbits were published by Martinotti and Tirelli 
('01). A few cells show marked cytoplasmic and nuclear degeneration. 

Morat ('01) and Bonne ('01) discovered that the spinal ganglia of the frog 
(species not stated) appear yellow in winter during hibernation, due to an 
abundance of fat droplets of variable size in and around the ganglion cells. The 
fat droplets appear to be derived from the capsule cells; they diminish when 
the frog revives from torpor in the spring, and disappear completely in the 

Smallwood and Rogers ('n), as previously mentioned, found but little 
change in the spinal cord of Necturus maculatus after 4 months of fasting, but 
marked changes after 16 months. The reddish masses of fat associated with 
the dorsal root ganglia disappear. The ganglion cells each contain an apparently 


normal nucleus, with finely granular cytoplasm enclosing a yellow oil droplet. 
The greater part of the cell body is occupied by a large vacuole, containing a 
lymph-like fluid. 

Changes found by Lodato ('98) in the retinal ganglion- cells of fasting dogs 
will be mentioned in Chapter XIII, in connection with the eyeball. 

In starved rabbits, cats and dogs, Peri ('93) made a careful study of the 
sciatic nerve and concluded: 

"Dans le systeme nerveux peripherique, les alterations sont generalement 
atrophiques. La myeline est diminuee, specialement chez les animaux qui 
resterent longtemps a jeun. Aucune alteration dans la constitution de cette 
substance. Le cylindreaxe ne differe en rien du cylindreaxe normal. Toutes 
les preparations, faites avec les differentes methodes, sur les nerfs peripheriques, 
servent uniquement a confirmer les diminutions de la myeline." 

Findlay ('21) found the sciatic nerve degeneration much less marked in starva- 
tion than in avian beriberi. 

Merzbacher ('03) found that section of the peripheral nerves in hibernating 
bats causes little or no degeneration, but the degenerative process is more 
rapid in the animals artificially warmed and awakened. Similar phenomena 
were observed in hibernating frogs. Hibernating mammals temporarily 
resemble the cold-blooded animals in their reactions, 

Sokoloff ('76) discovered that in fasting summer frogs the nerve endings in 
muscle (studied chiefly by the gold method) are apparently not much affected 
and, show up clearly among the degenerated muscle fibers. The "muscle 
corpuscles" and "nerve end buds" appear hypertrophic and hyperplastic. 

In the sympathetic system, Isaew ('87) studied the intestinal ganglia (plexuses 
of Auerbach and Meissner) in starved dogs. The ganglion cells were rarely 
found cloudy and swollen; oftener granular and vacuolated. The nuclei are 
sometimes well preserved, sometimes degenerated. The nerve fibers also appear 
cloudy, and. the interstitial tissue infiltrated with round cells. Statkewitsch 
('94) found a marked cytoplasmic vacuolation, less frequently a fatty degenera- 
tion, in the cardiac ganglia of a starved cat. 

Uspenski ('96) described the changes in various peripheral ganglia (nodosum, 
superior cervical sympathetic, celiac and cardiac) of rabbits fasting with loss 
in body weight of 15.3 to 45 per cent. After osmic fixation, two cell types are 
found, dark and light. The dark cells are smaller and most affected. The 
celiac ganglia show the greatest changes; the cardiac ganglia present the least. 
The cytoplasm undergoes a variable degree of vacuolation and hyalin degenera- 
tion. The nucleus may also degenerate, becoming hypochromatic or pycnotic. 
The nucleolus is often extruded. Some cells may be entirely destroyed; others 
nearly normal. The cytoplasm may be markedly degenerated, with nearly 
normal nucleus, or vice versa. The degenerative changes appear even in the 
earlier stages of inanition. If the rabbits are refed after inanition, the nuclei 
soon become normal but complete regeneration of the cytoplasm requires a 
long time. Eve ('96), however, found no apparent decrease in the Nissl sub- 
stance in the sympathetic nerve cells of the starved frog and rat. The article by 
Zuboff ('03) was inaccessible. 


(B) Effects of Partial Inanition 

The effects of partial inanition upon the peripheral nervous system have 
been studied chiefly in connection with protein deficiency (pellagra), vitamin 
deficiency (beriberi and scurvy) and water deficiency (thirst). 

Pellagra. — Fraenkel ('69-' 70) mentioned pigment formation in the sympa- 
thetic ganglion cells as frequent in pellagra; but Tuczek ('93) found the 
peripheral sympathetic nerves in general unchanged; likewise the spinal 
nerve roots. 

Rossi ('99) by Nissl's method found in the spinal ganglia of pellagrins pro- 
gressive chromatolysis, nuclear displacement and degeneration, with prolifera- 
tion of the neuroglia. These findings were confirmed by Amabilino ('03, cited 
by Harris '19). Babes and Sion ('00), aside from occasional fibrosis, noted but 
little change in the spinal ganglion cells. They found, however, degenerative 
changes in the spinal nerve roots and larger peripheral nerve trunks, indicating 
peripheral neuritis, and supported the theory of an exogenous origin of the 
degeneration in the white substance of the cord (opposing the endogenous 
theory of Tuczek and Marie, also advocated by Marinesco '09). Marie ('08, 
'10) mentioned the occurrence of degeneration in the anterior root fibers and of 
pigmentation in the peripheral spinal and sympathetic ganglia. Kozowski 
('12) noted degenerative changes in the peripheral nerves and nerve endings, 
also pigmentation of the sympathetic ganglion cells. Raubitschek ('15) made 
an extensive review of the changes during pellagra in the peripheral nervous 
system, including sclerosis in the spinal ganglia and posterior roots, degeneration 
in the peripheral nerve fibers, and fibrosis with simple or pigmentary atrophy in 
the sympathetic ganglion cells. 

Beriberi. — In human beriberi, Baelz ('82) described degenerative changes in 
most of the spinal nerves, and especially in the various sympathetic plexuses 
(cardiac, pulmonary, splanchnic, celiac and renal). Rumpf and Luce ('00) 
summarized the changes in the peripheral nerves as: "eine Neuritis chronica 
interstitialis mit ziemlich betrachtlichem Markfaserausfall und parenchymatoser 
Markfaserdegeneration." They cite earlier observations (by Baelz, Scheube, 
Pekelharing and Winkler) showing a chronic interstitial neuritis with atrophy 
and degenerative changes in the peripheral nerves (vagus, recurrent, phrenic and 
especially the spinal nerves). 

In 125 necropsies in cases of beriberi, Ellis ('98) similarly found marked 
degenerative changes constantly in the phrenic, and vagus nerves and cardiac 
plexuses (probably causing death) ; also in the splanchnic nerves, and the pul- 
monary, celiac, and renal plexuses and their branches. He concluded that the 
symptoms of beriberi are obviously caused by degeneration of the peripheral 
spinal nerves in the paralytic cases, and of the phrenic and vasomotor nerves in 
the "moist" cases. Duerck ('08) and Vedder ('13) gave excellent illustrations 
showing the degenerative changes in the peripheral nerves. Strong and 
Crowell ('12) found that all nerves (vagus, phrenic, femoral, popliteal, sciatic) 
show marked degeneration of the medullary sheaths by the Marchi method. 
There is no proliferation of neurilemma nuclei or leukocytic infiltration. 


In 18 necropsies in cases of infantile beriberi, Andrews ('12) sectioned and 
stained various nerves (vagus, phrenic, intercostal, anterior tibial) by the 
Marchi method. There is degeneration of some fibers, but not so extensively 
as in the adult. Nagayo ('23) states that the peripheral nerve lesions are similar 
in human beriberi and in experimental polyneuritis. 

In lower animals, the condition of polyneuritis was found by Eijkman ('97) 
in his classic work on experimental beriberi by a polished rice diet in chicks. He 
observed atrophic and degenerative changes in the peripheral nerves as well as 
in the spinal cord (especially in the anterior horn cells). Eijkman ('13) thought 
the lesions caused directly by a toxin, arising probably from a metabolic distur- 
bance involved in the dietary deficiency. Vedder and Clark ('12) studied poly- 
neuritis gallinarum in 56 fowls on polished rice diet. In marked cases, the vagus 
fibers usually all show degenerative changes, but none appear in the cervical 
sympathic ganglia or in their preganglionic or postganglionic fibers. All of 
the fowls on the diet 35 days or more showed degeneration in the sciatic nerve 
fibers, whether neuritic symptoms were present or not (Fig. 65). In advanced 
degeneration, the myelin sheath breaks up into globules and the axis cylinder 
disintegrates. Similar degenerative changes appear also in the nerve fibers of 
both dorsal and ventral spinal nerve roots, with chromatolysis and other 
changes in the large nerve cells of both ventral and dorsal horns. 

Schnyder ('14), however, found no changes in the sciatic nerve of white 
mice dying from experimental beriberi, and but slight degeneration in birds, 
cats, and all but one of 4 dogs (one showing marked degeneration in the sciatic). 
He suggests that the lesions in the peripheral nerves may appear only when the 
disorder is prolonged. Tasawa ('15) observed various degenerative changes in 
experimental polyneuritis (200 chicks and 150 pigeons), and also the regenerative 

Voegtlin and Lake ('19) described degeneration of the myelin sheaths in the 
spinal cord and peripheral nerves of cats, dogs and rats with polyneuritis pro- 
duced by dietary deficiency. Kimura ('19) claimed that in beriberi of birds 
the degeneration of the medullated nerve fibers begins in the axone, rather than 
in the medullary sheath. He also describes degenerative changes in the nerve 
fibers and cells of the spinal cord. According to Funk ('22), similar studies 
were made by Weill and Mouriquand ('17), Kato and Shizume ('19) and Paguchi 
('19). Findlay ('21) noted nearly constant myelin sheath degeneration in the 
sciatic, and also degenerative changes in the sympathetic ganglia of the intestine 
and suprarenal, in avian beriberi. 

Karr ('20) showed that in the dog on diets deficient in vitamin B, the nerve 
lesions appear only in those which continue to eat nearly up to the time of 
appearance of the nervous symptoms; otherwise they die of starvation without 
the nerve lesions. Hofmeister ('22), in conformity with the above mentioned 
observations of Schnyder ('14) on white mice, finds that in beriberi of rats there 
is no evidence of a degeneration of the peripheral nerves, although in severe 
cases lesions occur in the brain (as mentioned in Chapter X). 

McCarrison ('21) points out that typical Wallerian degeneration may occur 
in the sciatic and vagus nerves of apparently healthy pigeons, though less 


frequently than in those with polyneuritis columbarum. Degenerative changes 
were also found in the nerve cells of the intestinal and suprarenal sympathetic 
ganglia. Wallerian degeneration was likewise noted in the femoral nerve of 
monkeys on diets deficient in vitamin B, but not in controls. 

Scorbutus. — In human scurvy, no lesions in the peripheral nerves were 
found by Schodel and Nauwerk ('oo), Sato and Nambu ('08) or Aschoff and 
Koch ('19), the last named having examined the vagus in 22 cases. In experi- 
mental scurvy in the guinea pig, Hoist and Frolich ('12) observed frequent indi- 
cations of Wallerian degeneration in the nerves, often typical in isolated 
fiber-bundles, but attached no great significance to the findings. Ingier ('13) 
also found a variable degree of degeneration in the peripheral nerves (including 
sciatic, peroneal, vagus and phrenic) of scorbutic guinea pigs. Hess ('20) states 

1 \ 

•-. ■ ■ • ■ 

Fig. 66. — A portion of a cross section of the thoracic spinal cord in a young fowl subjected 
to aqueous inanition (dry diet). Stained by the Pal method, showing some nerve fibers of the 
white matter (above) and a portion of the adjacent grey matter (below). Several atrophic 
and shrunken ventral horn cells appear; also some vacuoles where cells have disappeared. 
The neuroglia shows proliferation of nuclei. X266. (Pernice and Scagliosi '95a.) 

that the sheaths of the large nerves as well as those of the vessels are very often 
invaded by hemorrhage in scurvy. The extravasated blood is found to lie 
around but rarely among the nerve fibers, which show no pathological changes. 
Aqueous Inanition. — In a dog which died after n days on dry bread, Pernice 
and Scagliosi ('95, '95a) found that the sciatic nerve fibers appear below normal 
in size and stain less intensely (by alum-carmine and Weigert's method). But 
few structural changes were observed, including a thickening of the perineurium 
and (partly) of the endoneurium, accompanied by a slight round cell infiltra- 
tion. In 3 chicks which died after 8-10 days on a dry maize diet, the sciatic, 
vagus and glossopharyngeus nerves were studied in cross section (Schultze- 
Ranvier method), showing atrophy of the medullated fibers, and granular 
degeneration of the myelin sheath . Irregular and poor staining of the medullated 
fibers was also found in teased preparations (Fig. 67). The axone is often 
atrophic, and may show irregular swellings and deformities (especially when 


fixed in Mliller's fluid and stained by Weigert-Pal method). There is also 
hyperemia and proliferation of the connective stroma, and the markedly 
atrophic nerve fibers may be separated by a granular substance apparently 
derived from the degenerated fibers. 

The effect of various methods of dehydration upon the nerve cells in the 
spinal ganglia and cord of rabbits was studied by Brasch ('98) who found finely 
granular degeneration of the Nissl substance in extreme cases only. The 
nuclear changes are more prominent and present two types: (a) nucleus rather 
small and pycnotic (sometimes karyorrhexis) ; (b) nucleus large, with stellate 
masses of variable stainability around the nucleolus. The former type appar- 
ently develops more slowly, the latter more rapidly; and transitional forms 
occur. The changes are apparently physical rather than necrobiotic, and are 
capable of recovery upon restoration of water. 

Fig. 67. — Two degenerated nerve fibers (Pal stain; teased preparation) from the sciatic 
nerve of a young fowl subjected to aqueous inanition (dry diet). The axone is visible in the 
upper fiber. X600. (Pernice and Scagliosi '95a.) 

In young albino rats (1 month old) held at nearly constant body weight by a 
relatively dry diet for various periods (see Table 10), Kudo ('21a) found a con- 
tinued growth in the weight of the sciatic nerves, similar in most cases to that 
in the spinal cord. With adult rats in the acute thirst series (body loss 36 per 
cent), he found an apparent average loss of 21.3 per cent in the weight of the 
sciatic nerves; and in the chronic thirst series (body loss 52 per cent) a loss of 
22.1 per cent. This would indicate that in adults the loss during thirst (aque- 
ous inanition) is relatively greater in the peripheral than in the central nervous 
system, according to the data cited in Chapters X and XL This may be due 
partly to the atrophy of adipose tissue which occurs normally intermingled in 
the trunk of the sciatic nerve; but it nevertheless is in harmony with the findings 
of Pernice and Scagliosi (above cited), who described an atrophy of the sciatic 
nerve fibers. 



The visual apparatus includes the eyeballs, together with the accessory 
organs, lacrimal glands, conjunctiva, etc. General impairment of the visual 
apparatus has frequently been observed in cases of famine or malnutrition. 
Recently a specific disorder (xerophthalmia) has been shown to result from 
deficiency of vitamin A. Apparently the deficiency lowers the resistance to 
bacterial infection, which is the direct cause of the ophthalmia. Thus dietetics 
becomes a factor of importance in ophthalmology, and particularly among 
malnourished children. Following a brief summary, the effects of inanition 
upon this apparatus will be considered under (A) total inanition, and (B) partial 

Summary of Effects on the Visual Apparatus 

During total inanition, or on water only, there is but little or no loss (some- 
times even an increase) in the weight of the eyeballs, although the remaining 
orbital contents (muscles, fat, etc.) undergo the usual atrophy. In the young, 
both human and infrahuman, there occurs during chronic inanition a persistent 
growth in weight of the eyeballs, similar to that found in the nervous system. 
The eyeballs usually resume their normal proportions after appropriate 

During total inanition (complete or incomplete) the eyes may remain normal 
in appearance; or conjunctivitis, corneal ulceration, etc. may occur (probably 
due chiefly to vitamin deficiency). Histologically, progressive atrophic and 
degenerative changes occur in the tissues of the eyeball. These changes are 
usually comparatively slight, affecting the iris, ciliary processes and muscle, 
choroid and especially the cornea and retina. The retina may present anemia, 
edema and cellular degeneration, notably in the ganglionic layer. In sala- 
mander larvae, mitosis is almost completely suppressed in the cornea during 
total inanition. 

In the various types of partial inanition, as in total inanition, there is a 
marked tendency to progressive increase in the weight of the eyeballs in the 
young, and little or no loss in adults. Dimness of vision in the form of nyctal- 
opia (night blindness) or hemeralopia (day blindness) has been observed during 
famine edema and scurvy, with retinal and conjunctival hemorrhages in the 
latter. Conjunctivitis may occur in these conditions, as well as in pellagra and 
especially during aqueous inanition (thirst). 

The visual apparatus in the young is remarkably sensitive to a deficit in 
vitamin A, which produces a typical ophthalmia in both human and lower species 
(mammals and birds). In the earlier stages, there is a mild conjunctivitis with 


xerophthalmia, involving cornification and degeneration of the corneal epi- 
thelium, edema of the substantia propria and progressive leukocytic infiltration. 
Later the conjunctivitis becomes more severe, with purulent panophthalmia and 
keratomalacia, leading in extreme stages to corneal perforation and consequent 
destruction of the eyeball. There are also characteristic lesions in the lacrimal 
glands, with a disturbance of secretion which may contribute in producing the 
changes in the eyeball. The conjunctivitis and keratomalacia are associated 
with bacterial infection. This appears to be secondary to the primary effect 
of the vitamin deficiency, which produces a specific lowered resistance in the 
tissues of the visual apparatus. Upon adding vitamin A to the diet, recovery is 
prompt, excepting in extreme stages with corneal destruction. 

(.4 ) Effects of Total Inanition, or on Water Only 

In the human adult, the earlier observation upon the visual apparatus 
during malnutrition included conjunctivitis and corneal lesions, evidently 
corresponding to the disorder now known as xerophthalmia and considered as 
due to a deficiency of vitamin A. The earlier literature, as reviewed by Cyr, 
('69) and Blegvad ('24), will therefore be presented later, in connection with this 
topic. Bourgeois ('70) noted that the human eyeball retains its volume during 

Luciani ('89, '90) observed the retina during the 30 day fast of Succi. He 
found nothing abnormal until the 28th day, when there was a slight narrowing 
of the visual field and also a slight constriction of the retinal vessels, of 
doubtful significance. 

In a man who died of starvation, Meyer ('17) noted that the eyes were deeply 
sunken, the periorbital fat " absolutely depleted" and the eyeballs very soft. 

Numerous observations indicate an increase in certain visual disorders 
attributed to malnutrition during the recent war period. Thus Seef elder ('19) 
noted an increase in acute glaucoma, conjunctivitis and keratomalacia (the latter, 
at least, probably due to partial inanition). Feilchenfeld ('20) concluded that 
the eye is very sensitive to malnutrition, which during war famine resulted in an 
increased number of lesions of the eyelids (extensions from facial eczema); 
retinal hemorrhages and thrombosis of the central vein; apoplexy and glaucoma. 
The general weakness also causes a marked decrease in the power of accommoda- 
tion. Pick ('20) likewise noted a marked increase in certain ocular disorders as 
a result of malnutrition during and since the war. 

Most of these ocular symptoms during famine are probably due chiefly to 
lack of vitamin A (xerophthalmia) and vitamin C (scurvy), to be considered 
later, as well as to the lowered resistance with increased susceptibility to infec- 
tions in general. 

In malnourished and emaciated infants, the corneal ulcerations (kerato- 
malacia) described by Mackensie ('57), Blessig ('66), von Graefe ('66), Gama 
Lobo ('66), Teuscher ('67), Tardieu ('80), de Gouva ('83), Thalberg ('83), Kubli 
('87), Schiele ('07) and Stolte ('22) doubtless likewise belong chiefly in the 
category of xerophthalmia, to be considered later. Ohlmuller ('82) recorded for 
the eyeballs a weight of 5.41 g. in an atrophic infant of 56 days (body weight 


2,381 g.) and of 7.90 g. in a well-nourished infant of the same age; but no con- 
clusion can be drawn, since the previous body weight of the emaciated infant is 
unknown. The weights of the eyeballs observed by me in atrophic infants 
(Table 3) are always above the normal weight at birth (3.2 g.), even in 
those infants which have never reached a body weight of 3,200 g. This would 
indicate a persistent growth of the eyeballs in atrophic infants, corresponding 
to that which will be shown later for underfed young animals. 

The regenerative process in corneal lesions of an atrophic infant was studied 
by Sachsalber ('03). He noted that the non-vascularity of the cornea is 
unfavorable to vitality, so that keratomalacia and xerosis conjunctivae often 
precede loss of body weight as signs of general infantile malnutrition. 

Schindler ('19), by a comparison of 288 healthy infants of the first year with 
172 malnourished infants, demonstrated a marked increase in the pigmentation 
of the iris in the latter group. Schindler thinks this increased pigment may be 
hematogenous in origin due to increased destruction of blood in the atrophic 

The changes in the visual apparatus during total inanition in the lower 
animals include (1) weight changes, in young and adult; and (2) structural 
changes. As to changes in weight in young animals, Manassein ('69) observed 
an apparent average increase of 12 per cent in the weight of the eyeballs during 
inanition in 3 young rabbits (23-25 days) and of 24 per cent in 8 somewhat 
older (3 months, 20 days) . The corresponding decrease in body weight was about 
30-35 per cent. Jackson ('15a) found that in albino rats held at constant 
body weight by underfeeding from 3 to 10 weeks of age, the eyeballs increase 
about 50 per cent in weight, showing under these conditions an intensity of 
growth greater than that in any other organ of the body (Table 4). A similar 
tendency was found also in somewhat older rats, and in those underfed for longer 
periods. Stewart ('18, '19) noted that if the underfeeding was begun at still 
earlier periods, the relative intensity of growth in the eyeballs appears even 
greater; a maximum increase of 143 per cent occurring in rats held at birth 
weight by underfeeding for an average of 16 days. In the newborn offspring 
retarded in growth by maternal underfeeding, however, Barry ('20, '21) found an 
increaseof only 31 per cent above normal in weight of the eyeballs. 

In young albino rats amply refed after underfeeding from 3 to 12 weeks of 
age, Stewart ('16) found that the relatively heavy eyeballs apparently return 
to normal proportions within four weeks of refeeding. In rats underfed from 
birth to 3, 6 or 10 weeks and then amply refed to a body weight of 25-75 g-> 
Jackson and Stewart ('19) observed that the eyeballs still tend to be slightly 
above normal in weight (Table 7). In rats permanently stunted in body weight 
by underfeeding from 3 to 20 weeks of age, and then refed to maximum size 
attainable, Jackson and Stewart ('20) found the eyeballs averaging 10.7 per cent 
subnormal in weight; while in those which had previously been underfed for 
nearly a year before refeeding, the eyeballs average 18 per cent overweight 
(Table 8). Apparently there is much irregularity in the recovery of normal 
proportionate size in the eyeballs after periods of underfeeding in young albino 


In adult animals, Collard de Martigny ('28) noted a depression of the cornea, 
which might indicate an atrophic collapse of the eyeball during starvation, but 
there is no evidence by weights to support this idea. Chossat ('43) observed 
even an apparent increase of about 5 per cent in the average weight of the eye- 
balls in starved pigeons. Schuchardt's ('47) data indicate a slight loss in the 
eyeballs of starved pigeons. Bidder and Schmidt ('52) found an apparent loss 
of 68 per cent in the orbital contents of a starved cat, but this included the 
orbital fat, muscles, etc., as well as the eyeball. Valentin ('57) published data 
indicating an apparent loss of 8.6 per cent in the average weight of the eyeballs 
in 3 hibernating marmots, with loss of 35.5 per cent in body weight. In 47 
adult rabbits, with average loss of about 39 per cent in body weight, Manassein 
('68, '69) found the weight of the eyeballs practically unchanged (increase of 
1 per cent). Bourgeois ('70) likewise found practically no loss in the weight of 
the eyeballs during starvation in rabbits, guinea pigs, cats and dogs, which was 
confirmed by Voit ('94) for the dog, by Sedlmair ('99) for the cat, and by 
Cattaneo^'oo) for the rabbit. Bich ('95) found the weight of the eyeballs 
usually increased (ascribed to edema) in dogs either on total inanition or on 
water alone. 

Jackson ('15) noted an average apparent loss of only 4 per cent in the weight 
of the eyeballs in adult albino rats during acute inanition (on water only), and 
a loss of 6 per cent during chronic (incomplete total) inanition (Table 4). 

Ott ('24) found that in frogs during hibernation and subsequent inanition 
with loss up to 60 per cent in body weight, the eyeballs remain nearly constant 
in weight up to the later stages. Then they present an increase in weight, 
reaching 12 per cent in the males and 22 per cent in the females (Table 6). This 
increase is not due merely to absorption of water. 

Manassein ('69) recorded the weight of the Harderian glands in fasting 
rabbits. In 47 adult rabbits with average loss of about 39 per cent in body 
weight, there was an apparent loss of 28 per cent in the glands. In 8 younger 
fasting rabbits (3% months old) with body loss of about ^3 per cent, the glands 
apparently lost only 2 per cent; and in 3 rabbits 23-25 days old, with loss of 
35 per cent in body weight, the glands apparently increased 22 per cent in 
weight. These glands were also found 22 per cent above normal weight in 5 
(adult) rabbits which had been fully refed after a period of inanition. The 
large weights (usually nearly 1 g., often more, for the normal) recorded by 
Manassein for these glands, however, raises the suspicion that they may not 
have been the Harderian glands. Krause (Anatomie des Kaninchens, Lpz., 
1868) gave the weight of the Harderian gland of the rabbit as 0.06 g.; of the 
(closely associated) infraorbital salivary gland as 0.15 g.; of the parotid as 
1.1 g. 

As to structural changes in the eyeballs of animals during inanition, Bour- 
geois ('70) observed that in starved mammals (guinea pigs, rabbits, cats and 
dogs) the cornea is flaccid and opaque, but does not present ulceration and 
perforation, such as has often been observed in human starvation. Healing 
of corneal wounds during inanition is imperfect. Carville and Bochefontaine 
('74, '75) stated that in a starved dog the orbital fat is replaced by a gelatinous 


mass, which under the microscope shows an amorphous structure with blood 
capillaries (usually empty). In fasting dogs, Falck ('75) found the conjunctiva 
inflammed; the cornea cloudy and opaque; the eyeball white, moist and soft; 
the eye-muscles greatly atrophied; and the orbital fat reduced or absent. 

Von Bechterew ('95) observed that the opening of the eyelids is delayed in 
newborn puppies and kittens subjected to inanition. Similarly Stewart 
('18), in albino rats underfed from birth, noted that the opening of the eyelids 
is somewhat delayed in time, but nevertheless appears at a lower body weight 
than in the normal controls. 

Bich ('95) made an extensive study of the visual apparatus, especially of the 
retina, in 24 dogs during inanition, with or without water (no difference noted). 
He described the conjunctiva as pale and frequently dry; the cornea transparent 
and shiny; the pupil usually dilated, sometimes contracted; the orbital fat 
almost disappeared. He figured and described a series of progressive retinal 
changes, beginning when the dogs have lost about 20 per cent in body weight. 
Histologically the ganglion cells show at first cloudy swelling and well-defined 
pericellular spaces; later a cytoplasmic vacuolation. The normal structure may 
be recovered upon refeeding, except in extreme stages, when recovery may be 
delayed, even when the body weight is restored to normal. Aside from the 
ganglion cells, all other elements of the retina during inanition appear abnor- 
mally separated by a condition of edema. 

Lodato ('98, '98a) likewise made a careful study of the ocular changes in 8 
dogs subjected to inanition, with or without water, for 11-30 days, with vari- 
ous methods of fixation and staining. No histological differences were found 
between those with and those without water; but the latter lived longer and, 
although vision is conserved, showed more clearly certain ophthalmoscopic 
changes, including retardation of pupillary reflex, constriction of retinal arteries, 
and dilation of veins. 

Histologically the sclera and cornea in these dogs show no change, excepting 
a partial loss of the epithelium behind Descemet's membrane. The iris appears 
thin and anemic, with constricted vessels; the anterior layer of epithelium very 
deficient; muscle cells and nuclei, also the ciliary muscle, stain faintly. The 
cells covering the ciliary processes also present cloudy swelling with poorly 
staining nuclei; numerous free pigment granules appear to have migrated from 
the pigment cells of the ciliary processes. The choroid and vessels appear 
atrophic, especially peripherally; less so toward the optic papilla. The optic 
nerve shows changes of doubtful significance, but the ciliary nerve presents 
marked atrophic degeneration by the Weigert-Pal method. 

The retinal vessels are constricted, with dilated perivascular spaces, espe- 
cially in the vessels of the papilla. The pigmented epithelial cells of the retina 
are swollen, with granular cytoplasm containing few pigment granules. No 
apparent change occurs in the rods and cones. By Nissl's method, the amacrine 
cells show chromatolysis. The ganglionic and nerve fiber layers appear moder- 
ately edemic. The ganglion cells appear swollen, with poorly staining nuclei 
and widened pericellular spaces. Nissl's stain shows the cytoplasm affected 
with a variable degree of chromatolysis, vacuolation and degeneration. The 


nucleus in many cells is irregularly shrunken or pycnotic. Lodato thinks it 
remarkable that vision is not more affected by these changes in the ganglion 

Cattaneo ('oo) studied the functional and chemical changes (which are 
slight) in the eyeballs of 3 rabbits. One eyeball (control) was enucleated before 
complete inanition of 6-1 1 days. Kammerer ('12) observed that in fasting 
Proteus anguinus the eye pigment is reduced and absorbed, which recalls a 
similar process in certain invertebrates (planarians and nemertine worms) as 
mentioned in Chapter III. 

Kornfeld ('22) studied the effect of the plane of nutrition upon the rate of 
mitosis in the corneal cells of the larvae of Salamandra maculosa, which were 
richly fed after various periods of total inanition. The total number of mitoses 
per cornea drops to about 4, after total inanition for 3 or 4 days. Upon refeed- 
ing, the number does not change for 4-5 days, then increases rapidly to a 
maximum of about 400 after 6-14 days of refeeding. While nutrition is a 
condition necessary for cell division, Kornfeld thinks that hormones or other 
stimuli may also be necessary factors. 

(B) Effects of Partial Inanition 

The types of partial inanition affecting the visual apparatus include chiefly 
deficiencies of protein (edema and pellagra), of salts (rickets) of vitamins (xero- 
phthalmia, scurvy) and of water. 

Protein Deficiency. — In Chapter V, evidence was cited to indicate that the 
edema frequently occurring in connection with famine and similar malnutri- 
tional conditions is in many cases probably due chiefly to protein deficiency, 
though often associated with other dietary defects. Lesions of the visual appa- 
ratus frequently occur in connection with the edema. Maynard ('09) observed 
a slight cloudiness of the cornea, with dimness of vision and evidences of increased 
intraocular tension in 20 cases of "epidemic dropsy." Budzynski and Chel- 
chowski ('i6)mentioned night-blindness as almost constant, generally preceding 
the edema. Schittenhelm and Schlecht ('19) likewise noted frequent hemer- 
alopia and xerosis in such cases. The literature is reviewed by Maver ('20). 

In pellagra (also probably due primarily to protein deficiency), ocular lesions 
may occur. According to Marie ('08, '10), these include conjunctivitis with 
pterygium and hemeralopia, sometimes pigmentary retinitis. 

In rickets (deficiency of phosphorus or calcium and of antirachitic vitamin), 
according to Juaristi ('19), the eyes are more round and show more of the sclera, 
associated with changes in the fundus. In experimental rickets in rats on diets 
deficient in vitamin A and phosphorus, Shipley, Park, McCollum and Simmonds 
produced both xerophthalmia and rickets. The addition of phosphates pre- 
vents the rickets, but not the xerophthalmia (which will be considered later). 
In albino rats with experimental rickets, Jackson and Carleton ('22, '23) found 
an increase in the weight of the eyeballs (Table 2), amounting to over 4c per 
cent in the group with severe rickets. Possibly this increase in weight may be 
related to that above mentioned as occurring in young rats with general growth 
retarded during incomplete total inanition. 


An interesting case was reported by Haigh, Moulton and Trowbridge ('20), 
in which a Jersey heifer, which had been on a calcium-deficient ration of silage 
and corn, gave birth to an undersized and maldeveloped calf, with no eyes and 
with hair growing from the "eye sockets." 

Deficiency of Vitamin A. Xerophthalmia. — In the human species, the sus- 
ceptibility of the eyes to the effects of malnutrition has long been known. In 
victims of the Irish famine, Donovan ('48) observed at autopsy blood-shot eyes, 
as noted also in death from other wasting diseases causing absorption of orbital 
fat. Redness and inflammation of the sclera are mentioned by Falck ('75) 
among the characteristic symptoms of death from starvation. In atrophic 
infants, Mackensie ('57) noted: "In particularly emaciated infants I have on 
many occasions seen the cornea of one or both eyes lose its substance, become 
prominent and perforate with almost no inflammation." The earliest symptom 
of xerophthalmia, xerosis conjunctivae, was described by Bitot ('63) in mal- 
nourished children. Von Graefe ('66) apparently gave the first minute descrip- 
tion of keratomalacia, which was also noted by Gama Lobo ('66),Teuscher ('67) 
and de Gouva ('83) in malnourished negro children in Brazil. A similar condi- 
tion as a result of long fasting during the Lent Quadragesima in Russia was 
described by Blessig ('66), Thalberg ('83), Kubli ('87) and others. Tardieu 
('80) mentioned corneal ulceration as one of the characteristic lesions in death 
from inanition in newborn infants, and keratomalacia in athreptic infants was 
also described by Koun ('03). According to Prugavin ('06), even long after the 
Russian famine of 1898, nearly all of the children suffered from purulent inflam- 
mation of the eyes, and much blindness resulted. The observations by Zak 
('17) on "chicken-blindness" are mentioned later under "Scorbutus." Schiele 
('07) found cod liver oil to be curative even when administered only to the 
mothers of the nursing infants affected by keratomalacia. The more recent 
work on human keratomalacia will be mentioned later. 

The nature and significance of these eye lesions during human malnutrition 
were not understood until quite recently, when the subject has been cleared up 
by animal experiments. Falta and Noeggerath ('06) and Knapp ('08) observed 
that young rats malnourished on diets deficient in various factors (including 
vitamins, then unrecognized) develop a marked tendency to conjunctivitis 
and corneal ulceration. Knapp found Staphylococcus present in the conjunc- 
tiva. Freise, Goldschmidt and Frank ('15) demonstrated that this experimental 
keratomalacia is not contagious, however. It does not occur in rats merely 
underfed; hence it is not due to ordinary inanition. They found that the 
histological changes present the typical picture of a keratomalacia, with an early 
cornification of the corneal epithelium, swelling and decreased stainability of 
the middle epithelial cells, and inflammatory infiltration of the lower epithelial 
cells. The substantia propria also presents edema, vascular invasion and local 
areas of cellular infiltration. Severe cases may develop a perforating ulcer. 
The condition may be prevented or cured by the addition of 2 c.c. of milk daily 
to the artificial diet. Goldschmidt ('15) concluded that the efficiency of the 
milk depends upon its content of "noch unbekannten, aber fur das Leben 
notwendiger Substanzen," and that experimental keratomalacia in animals 



corresponds to human keratomalacia, and belongs in the category of diseases 
due to partial inanition, such as scurvy and beriberi. 

Some of the changes in the eyes of rats with experimental xerophthalmia are 
shown in Figs. 68 and 69. 

Fig. 68. — Photograph of two albino rats of the same litter, placed at 3 weeks of age on a 
diet deficient in vitamin A (patent wheat flour, 66 per cent; "Crisco," 20 per cent; casein, 
5 per cent; yeast, 5 per cent; plaster of Paris, 2 per cent; sodium chloride, 2 per cent). In 
about a month, both developed xerophthalmia. This is shown in rat "A," with perforated 
cornea and protruding lens in the right eye. Rat "B" had a similar ophthalmia, without 
corneal perforation, but recovered perfectly (as shown in the photograph) in 6 days after the 
addition of dried spinach to the diet. (Courtesy of Professor McClendon and Miss Schuck.) 

Fig. 69. — Sections illustrating the corneal changes in xerophthalmia produced in the rat by 
a diet deficient in vitamin A. A, normal cornea. B, stage showing moderate changes: 
proliferation of surface epithelium, with numerous mitoses; substantia propria invaded by 
blood vessels and round cell infiltration, with occasional fibroblasts; Bowman's membrane 
absent. /, surface epithelium; II, Bowman's membrane; III, substantia propria; IV, Desce- 
met's membrane; V, posterior endothelium. (After Wason '21.) 

McCollum and Simmonds ('18) (also Jour. Biol. Chem., 1917, 3 2:iSl ) 
described the condition in young rats resulting from lack of vitamin A as a type 


of xerophthalmia. Bulley ('19) concluded that the xerophthalmia is due to 
infection, rather than to specific food deficiency, although admitting that the 
latter may cause a lack of resistance to the infection. Stephenson and Clark 
('20) found an invasion of leukocytes to be the earliest change in the cornea of the 
rat, followed by edema, vascularisation, etc., which may lead to complete 
corneal degeneration with protrusion of the lens. They obtained various 
bacteria in cultures from the conjunctival sac, and concluded that "the condi- 
tion directly attributable to dietetic deficiency is a predisposition to bacterial 
infection," leading to the characteristic lesions. 

On the other hand, observations and conclusions are presented supporting 
the theory that the absence of vitamin A is the primary factor in experimental 
ophthalmia in young rats by Mendel ('20), Emmett ('20), and Emmett and 
Sturtevant ('20), Wason ('21), Osborne and Mendel ('21 ; also observed by them 
as early as 1913), Hess, McCann and Pappenheimer ('21), Mori ('22), Walker 
('22), Yudkin and Lambert ('22, '22a) and Holm ('22); though not found 
(in adults ?) by Emmett and Allen ('20). Ophthalmia with similar lesions has 
been produced, by diets deficient in vitamin A, in the chick by Guerrero and 
Conception ('20) (on polished rice diet), and by Emmett and Peacock ('22) and 
Beach ('23); in the young rabbit by Nelson and Lamb ('20), and Nelson, Lamb 
and Heller ('22); in the dog by Steenbock, Nelson, and Hart ('21); in the duck 
by Rumbaur ('22); and in the pigeon by McCarrison ('23). There is found, 
however, much difference in the susceptibility to this disorder, varying accord- 
ing to species, individuals and diets used. 

The extent of recovery upon adequate refeeding varies according to the 
degree of the ocular lesions. In most cases in rats complete recovery is possible, 
according to Stephenson and Clark ('20), but where degeneration is advanced 
the cornea may remain opaque. "In some cases the cornea has so far degener- 
ated before cure is begun that the lens is forced through the aperture during 
life, and cure consists in the disappearance of pus and the healing over of the 
injured tissues." 

Wason ('21) found that in rats the anatomic lesions (Fig. 69) in experimental 
xerophthalmia include hyalinization or necrosis of the outer layer of corneal 
epithelium, exudation of serum and cells into epithelium and stroma, a prolifera- 
tion of blood vessels and fibroblasts, and, in advanced cases, an invasion of the 
anterior or (occasionally) the posterior chamber. The degree of restoration 
possible upon proper diet depends upon the extent of the secondary injury. The 
manner in which the deficiency of vitamin A renders the cornea susceptible to 
bacterial invasion is unknown. Walker ('22) also found a staphylococcus-like 
organism present, but he (like other investigators) was unable to prevent or cure 
the disorder by external antiseptics. He concludes that there may be some 
other (possibly hereditary) factor concerned. 

Mori ('22) concludes that the xerosis (dryness) of the conjunctiva and cornea 
is the essential change produced in the eyes of rats by the deficiency in vitamin 
A, and that the corneal ulcers (keratomalacia) are produced by the secondary 
infection. He finds that the two characteristic initial changes are (1) a cornifi- 
cation of the outer layer of epithelial cells of the conjunctiva bulbi and cornea; 


and (2) the formation of keratohyalin granules in the second layer of the epi- 
thelial cells in the conjunctiva, but not in the cornea, except at the limbus. 
These granules have also been described in xerosis conjunctivae in human 
lagophthalmos. Later stages in the rats, as in human keratitis, show alterations 
in the corneal epithelium, producing either an abnormal keratosis (due to the 
drying), or a necrosis (from interference with nutrition). The epithelial cells 
lose their nuclei and fuse, usually with infiltration by pus cells, and may disap- 
pear in small areas. The substantia propria shows marked edema and diffuse 
cell infiltration, which is denser in exposed areas. Cellular exudates also appear 
in the anterior chamber and iris, and perforating ulcers often develop. Mori 
('23) claims that xerophthalmia and keratomalacia may be produced in rats 
also by diets containing abundant vitamin A, but with certain unfavorable salt 

Yudkin and Lambert ('22) found the experimental xerophthalmia of young 
rats presenting "watery lacrimation with a serosanguinous conjunctival secre- 
tion, becoming after a short time somewhat viscid." Early focal lesions were 
found always beginning in the conjunctiva, and consisting of degeneration 
and cellular infiltration of the epidermis, sometimes extending into the sub- 
jacent' stroma. The cornea is involved later. They also find ('22a) in the 
lacrimal glands definite lesions, which appear degenerative or inflammatory 
in character. The lesions include variations in the size, form and staining 
properties of the lacrimal gland cells, which are probably correlated with 
functional derangement. The disturbance of lacrimal secretion may account 
for some of the phenomena of the xerophthalmia, particularly the characteristic 
drying of the cornea in the later stages. 

More recent histological study by Lambert and Yudkin ('23) indicates that 
the changes in the lacrimal and Meibomian glands of the rat during experimental 
xerophthalmia are of doubtful significance. More definite degenerative and 
inflammatory lesions occur in the Harderian gland. Disturbances in the 
(probably fatty) secretion of this gland may render the conjunctiva more 
susceptible to infection. 

Finally, Yudkin and Lambert ('23), from experiments on young white rats, 
concluded that: 

"The earliest lesions in ophthalmia of rats resulting from deficiency of 
vitamin A consist in focal inflammatory lesions in the conjunctivae of the lids 
and nictitating membrane. The involvement of the cornea, which constitutes 
the most conspicuous feature of the well developed ophthalmia, is a secondary 
phenomenon. The characteristic corneal plaque consists of keratinized epithe- 
lium beneath which the deeper layers of epithelium are generally found intact." 

"Pathologically the ocular manifestations of a deficiency of vitamin A 
are referable to a low grade inflammatory process, originating in the palpebral 
conjunctiva and spreading to the cornea. The rapidity of development and 
the degree of destruction probably depend in large part on the type of bacterial 

Certain recent observations on the effects of dietary deficiency in vitamin 
A upon the human eye remain to be considered. As previously stated, numer- 


ous observers, such as Seefelder ('19) and Feilchenfeld ('20), have noted various 
ocular lesions resulting from malnutrition during the war. Among these lesions 
are conjunctivitis and keratomalacia, which we are now justified in assuming 
to be due to the same vitamin deficiency causing xerophthalmia in the lower 
animals. Hess and linger ('19), however, found no eye trouble in 5 infants 
fed on a diet considered deficient in vitamin A for periods of 8 or 9 months. 

On the other hand, Mori ('04), Bloch ('18, '19, '21, '24) and others have 
found typical eye lesions following this dietary deficiency, especially in infants. 
Bloch ('21) states that: "Xerophthalmia is considered a rare disease. Only 
amongst the negro slaves of Brazil and amongst the poorest and most ignorant 
inhabitants of Russia is it said to have been observed to any extent. The 
disease is generally described as follows: the first symptoms are dryness of the 
ocular conjunctiva, which becomes wrinkled and shrunken. Later on, small 
yellowish white spots appear, as though the conjunctiva had been dotted with 
paraffin wax. At this stage the disease is termed Xerosis conjunctivae. The 
dryness rapidly spreads over the whole conjunctiva and over the cornea, which 
becomes dull, uniformly hazy and insensitive. Later the cornea turns greyish 
and still later yellowish, until at last a more or less extensive necrosis of the 
cornea sets in, followed by ulceration (keratomalacia). The necrosis and 
ulceration may appear in the course of a few hours . . . All authors agree 
that the keratomalacia is due to insufficient nutrition of the cornea, and that 
this again is the consequence of the disorganisation of the whole mechanism 
of nutrition. The disease therefore occurs only amongst children who have 
been ailing for a considerable time and is often met with amongst infants who 
have been insufficiently nourished." Bloch finds this disorder, which he terms 
Dystrophia alipogenica, due to a deficiency in vitamin A. It is fairly common 
in Denmark, and is probably an important factor in causing permanent blind- 
ness. Bloch ('24) has recently reviewed the subject in detail. Xerophthalmia 
may occur in children during the late stages of various malnutritional disorders. 
Death may be caused by secondary infections, especially bronchopneumonia. 
In case of recovery, if the cornea is only partly damaged, the child can usually 
be saved from complete blindness. The resultant scars are always hazy 
and opaque. 

Ross ('21) describes 4 cases of keratomalacia in infants with a dietary 
malnutrition corresponding closely to Czerny's "Mehlnahrschaden." Both 
clinically and histologically the eye lesions are very similar to those produced 
in animals by dietary deficiency in vitamin A. A similar keratomalacia is 
described by Stolte ('22). Wright ('22) observed numerous cases of human 

Blegvad ('24) has recently given the subject a thorough discussion, based 
on 434 cases of keratomalacia in children and 19 in adults (also 148 cases of 
xerosis conjunctivae without keratomalacia), all observed by Danish oculists 
between 1909 and 1920. He also gives a full review of the literature, with a 
bibliography of 200 titles. 

Scorbutus. — The ocular lesions in scurvy (due to deficiency in vitamin C) 
are chiefly connected with the general hemorrhagic condition associated with 


this disease. Thus retinal hemorrhages were reported by Jacobsthal ('oo). 
Kitamura ('io) found retinal hemorrhage and edema, with circumscribed 
hypertrophy of the nerve fiber and ganglionic layers. He also cited earlier 
observations by Grenet, Belowsky, and Sato and Nambu, indicating the 
occasional occurrence of conjunctival hemorrhage in scurvy. This was also 
noted by Blake ('21). 

A weakness of vision in the form of nyctalopia (night-blindness), or more 
rarely hemeralopia (day-blindness) has been noted during adult scurvy by 
Zak ('17), O'Shea ('18), Bierich ('19) and others. Zak stated that in Russia 
at Easter time, following the 7 weeks' fasting period, a visual disturbance arises, 
termed " chicken-blindness." There are no objective symptoms, aside from 
a conjunctival xerosis, and the disorder is easily curable by using fresh liver 
or cod liver oil. Therefore it is apparently related to the xerophthalmia above 
mentioned. O'Shea noted pallor of the optic disc in 3 out of 22 cases with 
scorbutic night-blindness. Blake ('21) describes a case of exophthalmos due 
to orbital hemorrhage in infantile scurvy, and cites similar observations by 
previous investigators (found in 49 out of 379 cases by the collective investi- 
gation of the American Pediatric Society). The ocular lesions in scurvy 
are included in the recent review by Hess ('20). Bessesen ('23) found the eye- 
balls to remain nearly constant in absolute weight, therefore appearing relatively 
above normal on account of the loss in body weight by scorbutic guinea pigs 
(Table 12). 

Aqueous Inanition. — A few data are available concerning the effect of 
water deficiency (thirst) upon the visual apparatus. Schuchardt ('47) noted 
an apparent loss of 4 per cent in the weight of the eyeballs in pigeons with loss 
of 44 per cent in body weight on a dry barley diet. In a dog (initial age 76 days) 
with loss of 20 per cent in body weight after 4 weeks on a diet of dry biscuit, 
Falck and Scheffer ('54) found, in comparison with a normal control, an apparent 
increase of 19.7 per cent in the weight of the eyeballs, with a very slight increase 
in their water content (from 89.8 to 90.9 per cent). 

In adult albino rats on a dry diet, Kudo ('21) found that in the acute thirst 
series, with body loss of 36 per cent, the eyeballs lose 10.2 per cent in weight; in 
the chronic thirst series, with body loss of 52 per cent, the eyeballs lose 13.3 
per cent; and in total inanition, with body loss of 47 per cent, the eyeballs lose 
13.0 per cent (Table 9). Opacity of the lens apparently caused visual dis- 
turbance in some cases; the conjunctiva sometimes appeared congested, and 
once hemorrhagic. In young albino rats held at constant body weight for 
various periods by a relatively dry diet, Kudo ('21a) found a progressive 
increase in weight of the eyeballs, amounting to about 71 per cent in those on 
the diet 9-13 weeks (Table 10). Thus during aqueous inanition the eyeballs 
in the adult rat lose slightly; but in the young rat they increase remarkably 
in weight, much as has been found during total inanition (Table 4). 

Pernice and Scagliosi ('95, '95a), in an adult dog which died after 9 days 
on diet of dry bread, noted that the white of the eyes became slightly yellow; 
and the right eye, later both eyes, developed a purluent conjunctivitis. 



The cardiac, musculature, like the skeletal, undergoes atrophy and degenera- 
tion during starvation, resulting ultimately in cardiac weakness and circulatory 
disturbances. During partial inanition, the human heart may be either 
atrophied (in malnutritional edema, pellagra or thirst) or hypertrophied (in 
rickets, beriberi or scurvy). The blood vessels are also affected, especially in 
edema and scurvy. After a brief summary, the effects of inanition upon the 
heart will be considered under (.4) total inanition and (B) partial inanition. 
Finally (C) the effects of inanition upon the blood vessels will be discussed. 

Summary of Effects on Heart and Blood Vessels 

In human adults the loss in the weight of the heart during total inanition 
appears variable, but is roughly proportional to that of the skeletal muscula- 
ture and the body as a whole. As a rule, this applies likewise to atrophic 
human infants. In adult animals, the same rule holds, with variations; but 
in the guinea pig, at least, the cardiac loss appears relatively greater toward 
the end of starvation than at earlier periods. In young animals, the heart 
appears relatively more resistant during underfeeding, and in the young rat 
it may even increase in weight while the body weight is stationary. 

The structural changes in the heart during total inanition (or on water only) 
involve a variable degree of atrophy and ultimate degeneration of the cardiac 
muscle. Among human adults during inanition, brown or pigmentary atrophy 
of the cardiac muscle fibers is characteristic, with more or less vacuolation and 
nuclear proliferation (sometimes degeneration). The myofibrillae become less 
distinct in cross striation, and fatty degeneration may occur to a variable extent. 
The pericardial and interstitial fat disappears, and the interstitial spaces 
between the cardiac muscle fibers become more extensive. In atrophic human 
infants, the changes appear somewhat similar, but less extensive and more 
variable. Pigmentation apparently does not occur as in adults. The myo- 
cardium may appear nearly normal, and in some cases the cardiac muscle fibers 
may even appear hypertrophied. There may also be a variable degree of 
interstitial fibrosis. 

Among lower animals, both young and adult, the cardiac muscle during 
total inanition likewise undergoes a variable degree of atrophy and degenera- 
tion. During abundant nutrition there may be a considerable amount of fat 
stored as small droplets within the cardiac muscle fibers, especially in hiber- 


nating animals. This fat is progressively consumed during inanition, although 
certain lipoidal granules (probably containing phospholipins) here as elsewhere in 
the body appear very resistant to starvation. Through fatty degeneration, 
on the other hand, there is apparently in some cases an increase in the amount of 
fat within the cardiac muscle fibers, although the question of fatty degeneration 
has been much disputed. 

The various types of partial inanition also occasion cardiac changes more or 
less resembling those during general inanition, but with some characteristic 
differences. In human malnutritional edema (protein deficiency) the heart 
appears atrophic and brown atrophy has been observed. In pellagra, the heart 
is variable in size, usually atrophic, and presents a pigmentary degeneration 
similar to that found in starvation. In rickets, the heart (especially the right side) 
appears variably hypertrophied, perhaps due to obstruction of the circulation 
from thoracic deformity. A similar cardiac hypertrophy occurs in human 
beriberi (due to deficiency in vitamin B); but in birds there is usually a marked 
cardiac atrophy. In both cases, there may be a variable degree of myocardial 
degeneration. In infantile scurvy there is a marked cardiac hypertrophy, which 
is slight or absent in guinea pigs. Slight degenerative changes may occur, 
occasionally hemorrhagic infiltration. During aqueous inanition (thirst) the 
heart weight undergoes changes nearly proportionate to the body weight 
(rat), and structural changes somewhat inflammatory in character have been 
observed (dog), especially in the endocardium. 

The blood vessels during inanition likewise undergo changes which are 
somewhat variable in extent and character. During total inanition (or on 
water only) there is an apparent atrophy of the larger arteries and veins, with 
variably degenerative changes in all the tunics. Atrophy of the capillaries 
has also been observed in various regions. A primary lesion of the capillary 
endothelium is probably responsible for the characteristic edema arising from 
protein deficiency, and for the hemorrhagic tendency in scurvy. Abnormal 
permeability of the vascular endothelium may be due also to calcium deficiency. 
In pellagra, the dermal vessels show degenerative and sclerotic changes of 
primary importance. During aqueous inanition (thirst), congestion and degen- 
eration likewise appear in both the capillaries and the larger blood vessels (dog 
and chick). 

(^4) Effects of Total Inanition, or on Water Alone 

The effects upon the weight of the heart will be considered first, in man and 
lower forms, followed by the effects upon cardiac structure. 

In human adults, Schultzen ('63) found the heart very small and devoid of fat 
in a starved girl of 19 years. Curran ('74) likewise found the heart very small 
(5 oz.) in an old woman who died from starvation; while Bright ('77) noted a 
weight of 7% oz. in the famous case of Harriet Staunton, with final body weight 
of 74 pounds. Miiller ('83) stated that during emaciation the pericardial fat, 
like the body fat in general, is nearly or entirely consumed; and that the heart 
weight is decreased, though proportionately less than the body weight. In an 


extensive series of 459 autopsies upon victims of the Indian famine, Porter 
('89) noted that in 45 per cent of the men the heart weight was under 6 oz. 
(average 5^ oz.) and in the remaining cases it averaged only 7 oz. In 37.4 
per cent of the women it was below 5 oz. (average 4.24 oz.) and in the remainder 
averaged barely 6 oz. For all the men autopsied, the heart weight averaged 
6.17 oz. (ratio to body weight 1 :i96); and for all the women 5.3 oz. (ratio 
1 :i8o). As a norm for comparison, he cited Quain's (European) ratio of 
1 : 158 for men and 1 : 149 for women; which would indicate a loss of heart weight 
in the famine victims relatively greater than the loss in body weight. This is 
somewhat doubtful, however. Most of the cases above childhood represented 
in the chart of Fig. 70 are from Porter's data. 

Fig. 70. — Graph showing the individual weights of the heart, according to body length, in 
atrophic human cases, newborn to adult, from various sources. The curve of normal heart 
weight is from data compiled by Prof. R. E. Scammon. It will be noted that, although there 
is much individual variation, in most cases the heart weight is below normal, the degree of 
atrophy apparently becoming greater with increasing age and body length. 

Askanazy ('13) claimed that during inanition relatively the least loss occurs 
in the heart, brain and bones, the heart losing relatively much less than the 
whole body. Hirsch ('99), however, stated that in cachexias the cardiac muscle 
is reduced in proportion to the skeletal muscle and body weight (edemas 
excepted). This is confirmed by the observation of Meyer ('17), and by the 
extensive data of Roessle ('19). 

Bean and Baker ('19) in data from autopsies at the Johns Hopkins Hospital 
and the Charity Hospital of New Orleans, found the average cardiac weight 
(excluding pathological hearts) in adults, classified according to their nutri- 
tional appearance (body weight unknown), as shown in the accompanying table. 


Heart Weight in Various Conditions of Nutrition (Bean and Baker '19) 

Color and sex 

No. of 









White, male 








White, female... . 








Negro, male 








Negro, female.. . . 








The reduction in heart weight during malnutrition is very evident from this 

Sison ('20) found in adults during voluntary fasting a diminished area of 
cardiac dulness on percussion, which he thought might be due partly to increased 
resonance of the lungs. 

The observations of Krieger ('20) indicate that the loss in adult heart weight 
during malnutrition from varied causes averages relatively slightly less than the 
loss in body weight, as shown in the accompanying table. 

Average Heart Weight in Various Conditions of Emaciation. All Males, Except 

as Indicated in Group I. From Autopsies in the Pathological Institute, Jena. 

In Estimating the Loss in Heart Weight, the Normal Was Assumed to be 0.5 

Per Cent of the (Initial) Body Weight, Which Was Calculated by 

Gartner's Formula from the Body Length (Krieger '20) 





est. loss, 
per cent 

Average heart weight 


Est. loss, 
per cent 

I. Insane. No chronic organic disease ? 

II. Chronic diarrhea 

III. Malignant growths 

IV. Chronic general infections 

V. Tuberculosis 

VI. Aged. Various conditions 












3 1 












271 .0 

33 o 
(29. 6) 1 

1 Making allowance for normal age change; heart assumed to be normally 0.58 per cent of the body 
weight in the aged. 

Weber ('21) found but slight decrease in the average cardiac weight of adults 
autopsied at Kiel between the years 1914 and 1919, the average decreasing from 
295 to 286 g. in the male and from 261 to 257 g. in the female. This is admittedly 
inconclusive, however, since the corresponding body weights are unavailable. 
Pearl and Bacon ('22), from a biometric analysis of 5,000 consecutive adult 
autopsies at the Johns Hopkins Hospital, conclude that in fatal tuberculosis 
cases there is a decrease in the absolute weight of the heart, which is probably 



due to inanition. My own observations indicate a marked decrease in the 
weight of the heart during tuberculosis in adults, but not in children. 

In atrophic infants, Ohlmuller ('82) found weights as shown in the accom- 
panying table. 

Heart Weight in Atrophic Infants (Ohlmuller '82) 



Body length, 

Body weight, 



weight, per 
cent of body 

56 da. 
56 da. 

4 mo. 

4 mo. 




27. 1 
21 . 1 






Thus apparently the heart is variable in relative weight in atrophic infants, 
but the lack of data concerning the previous body weight in these cases makes 
conclusions very uncertain. 

41 - 





• 80 

• 60 Ml .33 »'i 1 





1 Ml 


. < 1 

• • 

: I i 1 

1 i 


'1 1 

» y 


: i ' M 


. J 

1 1 n~ 

" " ' 


1 i 

! , 

X I 

1 ! i 


! i 


i 1 


1 1 1 




' I I 


1 j 

I : , 

1 : ' ■ ,, ! 

1 i 


| . • « 



i$X . 

L J I'll 1 1 


• ^}W* ' 


. • t 

£ ' 

trs j Mil 1 1 1 

^ . 



L 1 

X U- 

-Vy « 

• 11 1 



* . 

1 1 1 1 1 

! 1 1 

ttr 1 1 j , 

mi 1 1 


\s ' 

'3 ! ! 1 


\ <A 



< .. 

._ - 


| ,. _ 


i ♦» 



. n 

r ! ' 



1 . 


1 1 






♦ O 

_^-> m 

• • 

• • 


1 • 

! ' M 



I 1 1 

M Pi 






I | 







1 ' 1 T II 




V • 

1 1 


1 i 



i 1 

••■ : 



1 1 

; 1 r 

• u 7 

1 1 




! 1 l 

^< H " 1 


• ; 


-q l M « 

1 1 

1 1 

1 1 1 

' 1 1 ! 1 



1 i 

1 ■ 


1 ' p> ,p j . 1 


! t 


44 46 48 50 52. 54 56 58 60 62 64 66 

66 TO 

Fig. 71. — Graph showing the individual weights of the heart, according to body length, in 
atrophic infants. Data from various sources, the larger dots representing original Minnesota 
cases. The curve of normal heart weight is from data compiled by Prof. R. E. Scammon. 
In most cases, the heart weight appears clearly below normal, although there is much variation. 

DeTommasi ('94) and Thiercelin ('04) stated that the heart is small in 
infantile atrophy, while Bovaird and Nicoll in an extensive series (571 autopsies, 
birth to 5 years of age) found the heart weight in general reduced nearly in pro- 
portion to the body weight. Lange and Feldmann ('21) concluded from fluoro- 


scopic examination that in (living) emaciated infants the heart appears 
diminished in size; while Marfan ('21) claimed that in athreptic infants the 
heart, like the brain, is very resistant to inanition. Nicolaeff ('23) found the 
heart weight usually 20 to 40 (sometimes 50) per cent subnormal for age in 
famine-stricken children in Russia. Stefko ('24) noted greater cardiac atrophy 
in the female, especially about the age of puberty. 

My own data (Tables 2 and 3) confirm the principle that in malnourished 
infants the heart weight approximates the normal for corresponding body weight, 
although markedly below the normal for the previous maximum body weight, 
or the final body length (Fig. 71); and especially retarded in comparison with 
the normal for corresponding age (Table 2). The individual data from various 
sources for cardiac weights plotted in Fig. 70 seem to indicate a relatively 
greater cardiac resistance to inanition during infancy than in later years up to 
the adult. 

In various adult animals, Lucas (1826) found the heart apparently normal, 
but Collard de Martigny ('28) noted that it appeared small and atrophic in 
starved dogs and rabbits. Chossat ('43) observed that in starved pigeons the 
cardiac fat usually disappears, and the heart loses about 45 per cent in weight, 
while the loss in body weight is only about 40 per cent. 

In a starved cat, Bidder and Schmidt ('52) found that, in comparison with 
a normal control, the heart weight had increased from 10.85 g- to I2 -33 g-> which 
indicates perhaps an error or abnormality. Similarly Voit ('66) found in two 
cats an apparent loss of only 3 per cent in cardiac weight. These observations 
indicating little or no loss in the heart during inanition have since been fre- 
quently cited in the literature (e.g., Kitt '18), although a large number of later 
data lead to a different conclusion. The earlier observations are sometimes 
cited in support of the theory that the most active organs lose relatively least 
during inanition. 

Manassein ('68, '69) in 47 adult fasting rabbits with average loss of 39 per 
cent in body weight, found a loss of 24 per cent in heart weight; and in several 
cases the normal weight was recovered upon refeeding. In 2 crows with loss of 
36 per cent in body weight, the apparent loss in heart weight was 40 per cent. 
In various mammals (guinea pig, rabbit, cat and dog) and birds (fowl, pigeon), 
mostly on total inanition, but some with water or on incomplete inanition, 
Bourgeois ('70) observed an average loss of about 40 per cent in body weight. 
The heart was found atrophied nearly in proportion to the other musculature, 
with an average loss of 45 per cent, sometimes over 50 per cent. Luciani and 
Bufalini ('82) noted an atrophied heart in a starved dog; likewise Voit ('94), 
the loss in body weight being 32 per cent, and in heart weight 22 per cent. In 
a starved rabbit, Pfeiffer ('87) noted an apparent loss of 28 per cent in heart 
weight, and of 27 per cent in body weight; while in pigeons on total inanition 
Lukjanow ('89) found a loss of only 15 per cent in heart weight, with loss of 34 
per cent in body weight. 

Lazareff ('95) in a series of fasting guinea pigs, 10 in each group, with average 
losses of 10, 20, 30 and 36 per cent in body weight, noted corresponding losses 
of 4.84, 9.14, 20.97 and 33.33 per cent in heart weight. The loss in heart weight 
thus appears relatively greater toward the end of starvation (Table 5). 


A peculiar exception was found by Kusmin ('96) in rabbits and dogs, the 
heart during starvation with hyperthermia (with or without water) appearing 
even increased in weight, which was interpreted as a functional hypertrophy. 
This may perhaps explain other exceptions occasionally found. Thus Weiske 
('97) in 3 rabbits on water only, with body loss of 35 to 41 per cent, found an 
apparent decrease of about 29 per cent in heart weight in two, while in the third 
there was an apparent increase of 10 per cent. 

Sedlmaier ('99) concluded that in starved rabbits the heart loses relatively 
somewhat less than the whole body; while Beeli ('08) observed an apparent loss 
of 72 per cent in heart weight, with loss of only 51 per cent in body weight. 
With losses of 34 and 36 per cent in the body weight of guinea pig and rabbit, 
Heitz ('12) noted corresponding apparent losses of 20 and 24 per cent in the heart 

Jackson ('15) found in albino rats with acute inanition an average loss of 
28 per cent in heart weight (and of 33 per cent in body weight), and with 
chronic inanition a loss of t,t, per cent in heart weight (and of 36 per cent in body 
weight.) (Table 4.) In fasting pigeons with loss of about 40 per cent in body 
weight, McCarrison ('21) noted a comparable loss in heart weight (Fig. 35). 
In fasting frogs with body losses varying from 10 to 60 per cent, Ott ('24) 
found nearly corresponding losses in heart weight in the males, but usually 
somewhat less in the females (Table 6). 

In young animals during inanition, the heart usually appears somewhat more 
resistant. Bechterew ('95) found that in newborn kittens and puppies the loss 
in heart weight is relatively less than in body weight. In young albino rats 
held at constant body weight by underfeeding from 3 to 10 weeks of age, Jackson 
('15a) found practically no decrease in heart weight; and in those underfed from 
birth, Stewart ('18, '18a, '19) found even an apparent increase in heart weight, 
up to 25 per cent. In the atrophic offspring of underfed pregnant mother rats, 
Barry ('20, '21) observed an average cardiac weight 8 per cent above normal 
(Table 4). 

In underfed rats, after maintenance from age of 3 to 12 weeks, Stewart 
('16) found the heart 16 to 34 per cent underweight, with apparent overcompen- 
satory recovery (+17 per cent) after amply refeeding 4 weeks, but normal rela- 
tions after refeeding 16 weeks. Jackson and Stewart ('19) noted nearly normal 
heart weight in rats amply refed to 25-75 g-> after underfeeding from birth to 3 
or 6 weeks (Table 7). The heart appeared slightly above normal in weight in 
young rats refed by Jackson and Stewart (,'20) to adult stages after various 
periods of severe early underfeeding (Table 8). 

Trowbridge, Moulton and Haigh ('18, '19) and Moulton, Trowbridge and 
Haigh ('22a) found the heart weight approximately normal for body weight in 
young steers of various ages, irrespective of the plane of nutrition. 

Effects on Cardiac Structure. — The effects of total inanition (or on water 
only) upon cardiac structure will be considered first in man and later among 
infrahuman species. In human adults, the earlier observations (such as Dono- 
van '48 and Muller '83) on the structural changes during inanition included 
merely the gross features, including disappearance of the pericardial fat, and 


the "pale, soft and flabby" appearance. The first detailed microscopic study 
of the human heart was apparently by Hayem ('77), who observed that during 
inanition the heart is affected much like other (skeletal) muscles, sometimes with 
even more intense lesions. He found that in acute inanition, the heart muscle 
undergoes atrophy, with either simple granular or granulo-fatty degeneration 
(as found in rabbits by Manassein '69). Vitreous (waxy) degeneration is rare, 
but brown or pigmentary atrophy which does not occur in other muscles, may 
appear in the heart in all forms of cachexia. He stated that: 

"Cette lesion est generalisee. L'organe devient petit; ses parois sont amin- 
cies et les cavites sont en general retrecies, plus rarement un peu elargies (parti- 
culierement quand il existe des lesions d'orifice); souvent elles conservent leurs 
dimensions normales. Le tissu musculaire en s'atrophiant prend une teinte 
brune speciale, qu'on a comparee a celle d'une feuille morte; mais, quand les 
lesions sont tres-accentuees, les parois cardiaques ont la teinte de la terre de 
Sienne brulee. Le coloration feuille-morte indique une alteration mixte, 
d'ailleurs frequente, a la fois graisseuse ou pigmentaire. Au microscopique les 
fibres plus ou moins nettement atrophiees sont remplis d'amas de pigment accu- 
mules autour des noyaux. De plus, dans quelques points, particulierement 
dans ceux qui repondent a la teinte feuille-morte, on trouve des granulations 
graisseuses eparses, ne masquant pas habituellement la striation." 

Voelkel ('86) noted a small heart with fatty degeneration in a starved man. 
In victims of the Indian famine, Porter ('89) found edema frequently replacing 
the epicardial fat. In about one-sixth of the adults, there were surface patches 
of a white, detachable lymphy deposit, known as " soldier's spot." In a man who 
died from starvation, Stschastny C98) noted cardiac changes, with brown 
atrophy, disappearance of cross-striation, and vacuolation of muscle fibers and 
nuclei. In a similar case, Meyer ('17) found marked atrophy of the cardiac 
muscle fibers, which appeared separated by extensive spaces (Fig. 59). 

Krieger ('20), in various human cachexias, found brown atrophy character- 
istic in all except the infectious group, and most marked in tumor-cachexias. 
The pericardial fat usually underwent gelatinous degeneration or disappeared, 
but was found notably persistent in a few cases. 

According to Rubner ('20), various German observers have noted a displace- 
ment of the heart resulting from the resorption of pericardial and abdominal fat. 
Bradycardia occurs in chronic underfeeding, as well as during certain forms of par- 
tial inanition, to be mentioned later. Reiss ('21) found sclerotic changes. 

The question as to the cause and character of fatty degeneration of cardiac 
muscle during inanition has been much discussed, in connection with the prob- 
lem of fatty degeneration in general. Krehl ('93) concluded that "Mangel- 
hafte Versorgung der Gewebe mit O gait als haufigste Ursache der fettigen 
Degeneration." For human adults, Borchers ('14) and Eyselein ('14) found no 
constant relation between the degree of malnutrition and the amount of fat 
present in the cardiac muscle fibers. This question is discussed later, in connec- 
tion with the effects on lower animals. 

In human infants, Parrot ('68), in accordance with his doctrine of visceral 
steatosis during inanition, claimed the occurrence of a slight fatty degeneration in 


the heart in athreptic newborn. Thiercelin ('04) likewise held that the athrep- 
tic heart may undergo fatty degeneration, as in all chronic cachexias. Helm- 
holz ('09), however, reported negative cardiac findings, except in one case of 
ischemic necrosis. Monckeberg ('12) mentioned cardiac atrophy as character- 
istic in cases of pedatrophy, while Nobecourt ('16) stated that slight hyper- 
trophy occurs in some cases. Lesage and Cleret ('14) found the myocardium 
apparently normal in cases of congenital spasmodic atrophy. 

Variot and Cailliau ('12) made a careful study of the cardiac histological 
structure in atrophic infants, describing three stages corresponding to the degree 
of malnutrition: (1) In nurslings not very atrophic; heart large. The lesions 
are discrete, with no atrophic muscle fibers and some even hypertrophic (up to 
40-50/x). The muscle nuclei are slightly increased in size and number; peri- 
nuclear sarcoplasm abundant; vacuoles rare. (2) In more advanced cases, 
the heart still appears hypertrophied (although the body weight and stature are 
subnormal), but the cardiac muscle fibers are distinctly atrophic; none hyper- 
trophic. Some fibers appear moniliform. The nuclei are proliferated and 
deformed. The perinuclear sarcoplasm is increased in volume, and frequently 
contains numerous non-fatty vacuoles. (3) In extreme marasmic atrophy the 
myocardial lesions are very marked, with atrophy as well as generalized vacuola- 
tion, involving the entire contractile substance. The interstitial connective 
tissue of the myocardium shows multiplication of nuclei and sometimes 
increase of fibrous tissue or enlarged interstitial spaces. 

In famine-stricken children of various ages, Nicolaeff ('23) noted subepicardial 
edema with absence of fat. The cardiac muscle fibers appear atrophied, with 
loss of transverse striations in places. The sarcoplasm is sometimes vacuolated, 
especially near the nuclei, which appear more numerous. Nicolaeff made 61 
necropsies, with histological study in 19 cases. 

In animals, the changes in cardiac structure during inanition have been fre- 
quently and carefully studied. Manassein ('69) in fresh preparations from 
starved rabbits found marked histological changes in the heart muscle. The 
cross-striation disappears to a variable extent. Granules appear which are 
soluble in ether but not in acetic acid; and in marked cases large and confluent 
fat droplets may appear. Brown pigment granules were sometimes noted; 
also mentioned by Lepine ('74). In fresh cardiac muscle fibers of a starved dog, 
Falck ('7 5) noted distinct cross-striation, with diffusely scattered granules, 
resistant to acetic acid and caustic potash. 

Zander ('79) found that in pigeons which died from inanition following sec- 
tion of the vagi there is a fatty degeneration of the heart muscle fibers. Eich- 
horst ('79) doubted the fatty character of the granules thus produced, since 
they failed to blacken in osmic acid. Knoll ('80) concluded that in starved 
pigeons the amount of ordinary fat in cardiac muscle is decreased; but there is an 
increase in other granules of lipoidal nature, with intermediate stages between 
these and true fat droplets. Popov ('85) and Ochotin ('85, '86) noted cloudy 
swelling, granular and fatty (?) degeneration of the cardiac muscle fibers in 
starving rabbits; and similar changes, with disappearance of cross-striation, 
were found by Isaew ('87) in starved dogs. 


Morpurgo ('89b) noted that in starved pigeons the cardiac muscle fibers 
appear more or less degenerated. Measurements gave a decrease in average 
diameter from 9.22/z (in controls) to 6.50/x in starvation, indicating a decrease of 
about 50 per cent in volume. Heitz ('12) also found a marked reduction in 
both length and breadth of the myocardial muscle fibers in fasting rabbits and 
guinea pigs; likewise a decrease in nuclear size. 

Coen ('90) in fasting rabbits and kittens found the cardiac and other muscle 
mostly well preserved, although some fibers appeared atrophic, with cloudy 
swelling and loss of the characteristic cross-striation. Statkewitsch ('94) noted 
granular, non-fatty degeneration in the cardiac muscle fibers of starved animals 
(cats, dogs, rabbits, pigeons) ; and vacuolar (rarely fatty) degeneration of the 
cells in the cardiac ganglia in a starved cat. Uspensky ('96) likewise described 
vacuolar degeneration in the cells and nuclei of the cardiac ganglia in fasting 
rabbits, although less marked than in other sympathetic ganglia. Kusmin 
('96) noted extravasations of blood in the myocardium of fasting rabbits and 
guinea pigs during hyperthermia. 

In Myoxus, Vesperugo and Vespertilio, Baroncini and Beretta ('00, '00a) 
found at the beginning of hibernation an enormous accumulation of fat in the 
interfibrillar sarcoplasm of the cardiac muscle fibers. These fat droplets 
decrease irregularly until by the end of hibernation the fibers are nearly or quite 

Konstantinowitsch ('03) in rabbits and Beeli ('08) in cats described during 
starvation a decrease in the size of the cardiac muscle fibers, with a loss of the 
cross-striation; the nuclei become small, irregular, and deeply staining. 

Bell ('11), like Knoll ('80), recognized in cardiac (also in skeletal) muscle 
fine lipoidal granules, which are stainable with scarlet red, though not with osmic 
acid; and hence previously confused with albuminous granules in many cases. 
These "liposomes" he found markedly decreased in the cardiac muscle of 
starved rats (Mus decutnanus) . Wegelin ('13) and Bullard ('12, '16) likewise 
found a decided decrease or disappearance of fatty granules in the cardiac muscle 
fibers of fasting albino rats. Bullard ('16), however, concluded that the ordi- 
nary fatty granules are distinct from those containing phospholipins (lecithin, 
etc.), the latter being largely unaffected by inanition. 

While a review of the chemical literature upon this problem is beyond the 
scope of the present work, it may be noted that chemical analyses, such as 
those of Rubov ('05) on dogs and Terroine ('20) on rabbit, dog and pigeon, 
show in starvation no very marked decrease in the cardiac fats, either the phos- 
pholipins or the neutral fats and fatty acids. The question therefore appears to 
be still unsettled. 

(B) Effects of Partial Inanition 

The effects of partial inanition upon the heart will be considered under 
deficiencies of protein (including malnutritional edema and pellagra); salts 
(rickets) ; vitamins, including vitamin A, vitamin B (beriberi and polyneuritis) 
and vitamin C (scurvy) ; and water. 


Protein Deficiency. — Evidence was cited in Chapter V indicating that in both 
malnutritional edema and pellagra the chief factor is probably protein deficiency, 
although there may be other factors of importance. 

Malnutritional Edema. — The effects upon the heart in this or allied condi- 
tions among animals have not been extensively studied (cf. review by M aver '20). 
In "cachexia aquosa" of sheep, Frohner and Zwick ('15) stated that the heart 
is soft and atrophic. Hedinger ('15) mentioned not infrequent cardiac dilation 
in lamziekte of cattle. 

In "famine edema," the condition of the heart among human victims has 
frequently been noted. Practically all investigators of this condition agree 
that bradycardia (slow heart beat) is a cardinal symptom. This suggests a 
cardiac lesion, although it might, of course, be of nervous or other origin. 
Although it is generally agreed that this edema is not to be classed with those 
arising from cardiac insufficiency, an atrophy or decrease in cardiac weight has 
been noted at autopsy by Hiilse ('17, '18), Schittenhelm and Schlecht ('19), 
Prince ('21) and others. Paltauf ('17) reported the heart weight as nearly 
always below 300 g., occasionally as low as 200 g. Oberndorfer gives the weight 
at 180 g., and Enright C 2 °) states that: "The heart usually weighed only 3 to 5 
oz., but in other respects was apparently normal." Mann, Helm and Brown 
('20) reported the heart normal in size in 200 necropsies. 

In structure, v. Jaksch ('18) and Schittenhelm and Schlecht ('19) found brown 
atrophy of the cardiac muscle in human famine edema. Jansen ('19a.) noted 
pigment masses at the poles of the cardiac cell nuclei; otherwise no regressive 
changes. Oberndorfer ('18) reported complete absence of glycogen and fat 
droplets in the heart muscle. Mann, Helm and Brown ('20) noted edema or 
gelatinous appearance of the cardiac surface. The edema of the "auriculo- 
ventricular" junction, found almost constantly by Park ('18) and Menzies 
('20), probably represents merely the local gelatinous metamorphosis of the 
epicardial fat, which has frequently been observed also in other conditions of 
chronic inanition. 

Pellagra. — In general, the heart is usually found more or less atrophic in 
pellagra, though sometimes hypertrophied, according to the review of the litera- 
ture by Marie ('08, '10), Raubitschek ('15) and Harris ('19). Thus Fraenkel 
('69-'7o) reported the heart hypertrophied in only 12 out of 113 cases; while it 
appeared atrophic or emaciated in 49. Lombroso ('92) stated that the hyper- 
trophic appearance of the heart is often deceptive, on account of its flaccidity. 
Actual weights of 26 hearts of pellagrins showed 2 slightly above normal; 
5 slightly below; and 19 markedly below normal. Nicholls ('12) observed in 
8 cases an average cardiac weight of 7^ oz., the normal being 9 oz. 

As to histological changes in the cardiac muscle, pigmentary atrophy was 
noted by Fraenkel ('69-'7o) as characteristic. Lombroso ('92) found brown 
atrophy in 28 out of 35 cases examined, and fatty degeneration in 3. The 
cardiac fibers often appear abnormally separated. Brown atrophy and fatty 
degeneration were also found by Tuczek ('93) and Marie ('08, '10), Kozowsky 
('12), and Raubitschek ('15). Kozowsky also mentions cardiac fibrosis, espe- 
cially along the vessels, and pigmentary degeneration of the cardiac ganglia. 


Cardiac lesions somewhat resembling those of human pellagra were found by 
Nicholls ('12) and Sundwall ('17) in rats and other animals fed maize and similar 

Mineral Deficiency. Rachitis. — In a dog, with a rachitoid disorder pro- 
duced by a calcium-poor diet, Voit ('80) found an apparent slight hypertrophy 
of the heart. This was confirmed by Jackson and Carleton ('22, '23) who 
found a moderate hypertrophy of the heart in rats subjected to experimental 
rickets (Table 11). In human rickets, Cheadle and Poynton ('07) stated that 
the heart appears variable, but probably shares in the general malnutrition. 
According to Wohlauer ('n), the heart may be displaced (due to thoracic de- 
formity), and is usually hypertrophied, especially in the right ventricle, on 
account of the increased labor due to the impediment to pulmonary circulation. 

Vitamin Deficiency. Vitamin A. — Emmett and Allen ('20) and Davis and 
Outhouse ('21) found no definite changes in the heart of rats on diets deficient 
in vitamin A. The effects of this and other vitamin deficiencies have been 
reviewed recently by Funk ('22). Herter ('97) noted cardiac enlargement, 
hemorrhages, and slight "granular degeneration" in the cardiac muscle fibers 
of pigs during starvation, involving a deficiency in vitamin A. Meyerstein 
('22) made a few observations on the heart in young white rats on diets deficient 
in vitamins A and B. 

Vitamin B. Beriberi and Polyneuritis. — In human beriberi, Ellis ('98) found 
a marked cardiac hypertrophy, with the right side in every case much enlarged. 
In 125 necropsies of beriberi, the average heart weight was 13.37 oz -> while in 
204 controls it averaged slightly below 9 oz. Duerck ('08) noted in beriberi a 
variable heart weight though it was usually found hypertropied, even in emaci- 
ated individuals. He also observed atrophic degeneration and nuclear prolifera- 
tion in cardiac as well as in skeletal muscle. He stated ('08a) that the right 
heart is especially dilated and hypertrophied, and that the cardiac muscle shows 
yellowish spots and streaks of a fatty nature. In a case of acute beriberi, 
however, Strong and Crowell ('12) found the cardiac muscle fibers swollen, ede- 
matous, granular, and irregularly vacuolated; but not fatty. In infantile beri- 
beri, Andrews ('12) described the right heart as greatly hypertrophied, so as to 
equal or exceed the left ventricle. The cardiac muscle fibers were also found 
hypertrophied, with congestion but no degeneration. Nagayo ('23) claims that 
hypertrophy of the right ventricle distinguishes human beriberi from experi- 
mental polyneuritis, in which cardiac atrophy is common. 

In experimental avian beriberi, cardiac lesions have frequently been found, 
although Schnyder ('14) reported negative findings. In 200 fowls and 150 
pigeons, Tasawa ('15) found the heart usually markedly atrophic, although the 
right heart (sometimes also the left) is distinctly dilated. Vedder and Clark 
('12) likewise concluded that no cardiac hypertrophy, comparable to that in 
human beriberi, occurs in polyneuritis gallinarum (in 56 fowls). The heart 
may show no microscopic changes; or there may be slight edema, increased 
pigmentation, or incipient mucoid or parenchymatous degeneration. The 
cardiac lesions are described by Vedder ('13). In pigeons on diet deficient in 
vitamin B, McCarrison ('19, '21) observed marked decrease in cardiac weight 


(Fig. 35). Findlay ('21) in 43 fowls and 41 pigeons found the heart not greatly 
atrophied (Table 13), and the right heart (especially the atrium) often dilated. 
The heart muscle appears pale and soft, with a line of edema frequently visible 
at the atrioventricular junction. Microscopically the myocardial cells show 
cloudy swelling and often beginning fragmentation. Hydropericardium oc- 
curred in 20 to 26 per cent of the birds. Very similar cardiac changes were 
found in inanition with water only. Hoffman ('22) in polyneuritic pigeons 
reported the heart "enlarged and flabby," sometimes "jelly-like." Souba and 
Dutcher ('22) and Souba ('23) noted a significant loss in cardiac weight in 
several hundred chicks on diets deficient in vitamin B. Lopez-Lomba ('23) 
found the cardiac weight normal in pigeons on a vitamin-free diet, excepting an 
early slight transient increase. 

In mammals on diets deficient in vitamin B, the results upon the heart appear 
less striking. Schnyder ('14) found no changes in the heart muscle of mice, 
aside from congestion (stasis). Drummond ('18) noted enlargement of the 
right heart in a few black rats. Voegtlin and Lake ('19) observed slight degen- 
erative changes in the heart of cats, dogs and rats. Emmett and Allen ('20) 
noted some atrophy of the cardiac muscle fibers of albino rats. 

Vitamin C. Scorbutus. — The cardiac changes during scurvy have 
frequently been studied, and the literature is fully reviewed by Hess ('20). In 
5 necropsies on infantile scurvy, Schodel and Nauwe.rk ('00) discovered a hyper- 
trophy of the ventricles and dilation of the right heart. This was confirmed in 
21 out of 31 necropsies by Erdheim ('18), who concluded that a direct ratio 
exists between the degree of cardiac enlargement and the intensity of the disorder. 
Cardiac enlargement was sometimes found even coexistent with general maras- 
mus, and may be due to the thoracic deformity (as in rickets). Hess ('20) 
described and demonstrated by means of roentgenograms the enlarged heart 
in infantile scurvy. There is also almost always an increase in pericardial fluid. 

In human adult scurvy, Sato and Nambu ('08) found the heart not especially 
enlarged, but the cardiac musculature often anemic and brownish, sometimes 
cloudy, with fibrosis. Aschoff and Koch ('19) described the enlargement of 
both ventricles, with possible fatty degeneration of the cardiac musculature. 
Comrie ('20) found the heart feeble, showing brown atrophy. 

In experimental scurvy of guinea pigs, Hoist and Frolich ('07, '12) frequently 
observed a fatty degeneration of the cardiac muscle. Lesions of the cardiac 
musculature were also described by Rondoni and Montagnani ('15). Findlay 
('21a) found dilation but no hypertrophy of the heart in scorbutic guinea pigs. 
The cardiac muscle fibers show loss of striation, but no fatty degeneration. The 
interstitial stroma is edematous in places, with capillary congestion; and definite 
hemorrhagic infiltration of the heart wall was noted in 5 out of 12 cases. A 
slight increase in the weight of the heart in scorbutic guinea pigs was found by 
La Mer and Campbell ('20) and Bessesen ('23) (Table 12). A relative increase 
is to be expected, on account of the loss in body weight. Hojer ('24) noted 
atrophy of the cardiac fibers, with necrosis and tendency to calcification. 

Aqueous Inanition. — In pigeons perishing on a dry barley diet, Schuchardt 
('47) found a loss of 36 per cent in the average weight of the heart. Bowin ('80) 


in rabbits and dogs on dry food found that the heart undergoes relatively less 
loss in weight than does the body, and retains its original water content (not 
confirmed by Skoritschenko '83). Pernice and Scagliosi ('95a), in a dog 
which died on a dry bread diet, found inflammatory changes in the heart, espe- 
cially the endocardium. The perivascular connective tissue is thickened, rich 
in nuclei and spindle cells. The cardiac muscle fibers appear pale, less distinctly 
stainable and more homogeneous in appearance with indistinct striation. The 
myonbrillae seem thinner and less closely packed; the nuclei numerous and some 
showing mitosis. 

Durig ('01) observed that in frogs upon withdrawal of water from the body, 
the organs lose unequally in weight. Excepting the brain, the heart loses less 
than any of the other organs. Kudo ('21) found that in adult albino rats on 
acute thirst experiments the heart lost 30.6 per cent in weight (body loss 36.1 
per cent) ; while in chronic thirst experiments the heart lost 46.3 per cent (body 
loss 52.4 per cent). In total inanition, the results were very similar to those in 
chronic thirst (Table 9). In thirst experiments in which young albino rats 
were held at constant body weight for various periods, Kudo ('21a) found that 
the heart remains nearly constant in weight, with a slight increase (of doubtful 
significance) in the longer periods (Table 10). 

(C) Effects of Inanition upon the Blood Vessels 

Numerous scattered observations are available concerning the effects of the 
various types of inanition upon the blood vessels. 

Total Inanition (or on Water Only). — Chossat ('43) gave pigeons water only 
until death with loss of about 40 per cent in body weight. He found an apparent 
decrease of nearly 30 per cent in the weight of the great vessels, but considered 
the result inconclusive on account of variability in the extent of removal of the 
vessels. Bidder and Schmidt ('52) found an apparent loss of about 38 per 
cent in the weight of the aorta and vena cava in a starved cat, with body loss 
of about 50 per cent. Strelzoff ('64) observed an apparent atrophy of the capil- 
laries in fasting rabbits and guinea pigs, especially in the small intestine, stomach 
and pancreas; to a lesser extent in the liver, large intestine and muscles. The 
nuclei of the capillary wall apparently undergo a fatty metamorphosis, and the 
vessels finally disappear entirely. In starved rabbits and dogs, Mankowsky 
('82) found the vessels of the spinal cord unaffected, aside from endothelial 
proliferation. On refeeding, a swelling of the endothelial nuclei of the cerebral 
vessels was noted. Poljakoff ('88) noted that in the albino rat the blood capil- 
lary plexus of the fat lobules undergoes atrophy together with the adipose tissue 
during inanition. 

Trivus ('99) made a thorough study of the changes in the walls of ligated 
arteries (femoral and common carotid) in rabbits and dogs fasting various 
periods with losses in body weight up to 40 per cent. In general the vascular 
wall shows a weak inflammatory reaction. Near the ligature the leukocytes 
predominate over the cells of the granulation tissue; many eosinophile and 
pigmented cells appear. The endothelium of the intima frequently thickens, 
forming several layers of cells. In the tunica media, the cytoplasm and nuclei 


of the muscle cells undergo vacuolation and the elastic layers are more wrinkled 
(due to the shrinkage of the vessel). The stellate connective tissue cells of the 
adventitia undergo fatty degeneration, and extensive hemorrhages into the 
surrounding tissue are frequent. In animals amply refed after inanition, these 
changes in the vascular wall have largely disappeared, excepting slight necrosis 
and fatty degeneration which still persisted in the cells near the ligature. 

In dystrophic infants, according to Lesage ('11), the blood vessels usually 
appear normal in structure, although arteriosclerosis sometimes appears. 

Stefko ('23) calls attention to the frequent hemorrhagic diathesis (purpura) 
in cases of human starvation. The condition appears to be due to degeneration 
in the walls of the blood vessels, which results from the impoverishment of the 
blood. Possibly this hemorrhagic condition should be ascribed to the exhaustion 
of vitamin C, as occurs in scurvy. 

Protein Deficiency. — Many investigators of malnutritional edema ascribe 
the condition primarily to an injury of the capillary walls (Lange '17; Schitten- 
helm and Schlecht ('19); Maver '20; Burger '20; Maase and Zondek '20; etal.). 
Oberndorfer ('18) considered the universal capillary congestion of significance, 
and Maver, Maase and Zondek mention the possibility of direct toxic injury to 
the capillary walls. The question is closely related to that of the role of the 
capillaries in the production of edema in general, the literature upon this topic 
being reviewed by Lange ('17). 

Although, as previously stated, protein deficiency has usually been accepted 
as the primary cause of famine edema, the question has been raised as to whether 
this (as also other edemas) may not involve also a calcium deficiency. The 
work of Herbst upon invertebrates (see Chapter III) showing the effects of 
calcium deficiency in dissolving the intercellular cement substance has been cited 
in this connection; as well as in disorders such as scurvy, with characteristic 
capillary hemorrhages. Chiari ('10) has reviewed the evidence indicating 
that calcium deficiency in general tends to increase the permeability of the capil- 
lary walls. 

In pellagra (presumably due chiefly to protein deficiency), vascular changes 
have been described. Fraenkel ('6o-'7o) mentioned pigmentation of the 
capillaries and fatty degeneration of the adventitia in the brain vessels. 
Primary degeneration and sclerosis of the blood vessels was emphasized by Marie 
('08, 'jo), Nicholls ('12), Kozowsky ('12), Raubitschek ('15) and Harris ('19) 
as of importance in the pathogenesis of pellagra. 

In rickets, Kassowitz ('12), as previously mentioned, has long advocated the 
theory that the characteristic changes in the zone of enchondral ossification are 
primarily due to hyperemia and increased vascular proliferation in the region. 
This theory, however, has not met with general acceptance. 

In relation to vitamin deficiency, vascular changes have been noted especially 
in connection with scurvy (deficiency of vitamin C), in which the hemorrhages 
form a constant and conspicuous lesion, and petechial extravasations have 
been described as occurring in nearly every organ. The exact cause of the 
hemorrhagic tendency is uncertain. Direct toxic injury to the capillary wall 
has been suggested by Sato and Nambu ('08). The possibility of a calcium- 


deficit was discussed by Gerstenberger ('18), which recalls the fundamental 
work of Herbst (mentioned above). Bierrich ('19) thought the capillary 
damage might be due to the lack of some N-containing building stone in the 
capillary endothelium. Wallgren ('21) calls attention to the occurrence of 
scorbutic edema, likewise due to an abnormal permeability of the capillary walls. 
Endarteritis of the medium sized arteries has also been observed by Ide ('22), 
and Sato and Nambu state that the aorta sometimes shows sclerosis. The 
increased permeability of the capillary walls has been used by Hess ('14) in his 
"capillary resistance test" in the diagnosis of scurvy. The whole question is 
thoroughly reviewed by Hess ('20). 

Findlay ('21a) has recently discussed the vascular changes in guinea pig 
scurvy, which appear to involve a primary interference with the nutrition of 
the capillary endothelium. The endothelium becomes swollen and degenerated, 
producing congestion. This occasions increased transudation of fluid through 
the capillary wall, and, as the intercellular substance is weakened, leads to 
diapedesis and characteristic hemorrhages. Hojer ('24) ascribes the hemorrhages 
to a weakness of the vascular wall, caused by an atrophy of the collagenous 
connective tissues, which is considered characteristic of scurvy. 

Aqueous Inanition. — In a dog on dry diet, Pernice and Scagliosi ('95a) 
found the blood vessels generally congested (passive hyperemia) and showing 
degenerative changes, notably in the nervous system and viscera (especially the 
kidneys). Similarly in young chickens, chronic thirst involved vascular 
congestion in the capillaries and larger vessels. In the aorta and right carotid, 
marked hyperemia, small hemorrhages and round cell infiltration were noted in 
the tunica adventitia and tunica media, although the intima showed but slight 
changes. Kudo ('21a) made observations indicating an apparent increase in 
the weight of the aorta in young albino rats subjected to chronic thirst for 
various periods. 



The effects of inanition upon the blood are of unusual interest, not only 
because of its fundamental importance to the organism, but also on account of 
its practical use in diagnosis. Although relatively stable, the blood is 
found to undergo variable changes in nearly every type of inanition. The 
effects of inanition upon the blood will first be summarized according to its 
various components, and later considered in detail under (.4) effects of total 
inanition and (B) effects of partial inanition. 

Summary of Effects on the Blood 

While the blood exhibits in some respects a considerable degree of stability 
under various conditions of inanition, the available data appear so variable and 
conflicting that a summary is unusually difficult. The variability affects chiefly 
the blood cell counts, the structure of the cells rarely being appreciably modified. 
The blood counts are affected primarily by the water content of the plasma, 
which undergoes marked fluctuation under various conditions. In addition, 
there are also frequent and extensive changes in the differential leukocyte 
count, probably due chiefly to changes in the hemopoietic system. It is evident 
that the number of blood cells in general will vary according to the ratio between 
blood destruction and blood regeneration, both of which may be variably 
affected by inanition. In interpreting the differential leukocyte count, it must 
be remembered that any marked variation in the polymorphonuclears will 
affect the percentage count of the other varieties, independent of their absolute 
numbers. Other variations observed in both red and white cells may depend 
upon the species, age, individual and the type of inanition concerned, as well as 
(sometimes) imperfect technique and other unknown factors. The principal 
results will be summarized briefly for the separate elements of the blood under 
varied conditions of inanition. 

The total volume of the blood, so far as has been accurately determined, 
although somewhat variable, tends to maintain its normal ratio to the entire 
body. Changes in distribution, however, with diminished peripheral circulation, 
may give a deceptive appearance of anemia in various (especially chronic) types 
of inanition. 

The blood plasma is subject to various changes in chemical composition, 
although for a long time its losses may be restored through absorption from the 
various tissues. The water content undergoes marked fluctuations, which 
affect the concentration of the blood and consequently the cell counts. Hydre- 
mia frequently occurs, especially in later stages of inanition and in certain 



chronic forms, particularly in protein deficiencies (malnutritional edema). 
On the other hand, anhydremia, with concentration of the blood, often occurs, 
especially during thirst. 

The erythrocytes rarely show structural changes (anisocy tosis, poikilocy tosis) , 
although there is evidence of their increased destruction in severe inanition, and 
the frequent appearance of nucleated red cells in the circulating blood indicates 
intensive regeneration, especially upon refeeding after inanition. During 
human inanition, the erythrocyte count is often within normal limits, but 
sometimes increased (especially in total inanition and earlier stages), or decreased 
(especially in chronic and late stages). In animals, the red cell count appears 
more frequently increased in the earlier stages of total inanition, often decreasing 
later. In hibernation, the erythrocyte count is variable. 

Among the various types of partial inanition, in malnutritional edema (due 
chiefly to protein deficiency), hydremic anemia is very characteristic in both 
man and lower forms. The results of a dietary deficiency of iron are variable, 
but apparently anemia may be produced in young animals. In rickets the 
blood is sometimes normal, but there is usually anemia, somewhat proportional 
to the severity of the rickets. In vitamin B deficiency (beriberi) there is usu- 
ally well marked anemia. In vitamin C deficiency (scurvy), the erythrocytes 
are variable; but there is usually a secondary anemia, often of the chlorotic 
type, with hemoglobin disproportionately low. In experimental scurvy of 
guinea pigs, the blood changes are usually slight. Aqueous deficiency (thirst) 
on dry diets usually produces in human and animal experiments an increased 
red cell count (in extreme cases nearly doubled) through concentration of the 

The leukocytes likewise rarely present morphological changes, though 
cytoplasmic and nuclear degeneration has occasionally been noted. The 
total leukocyte count during inanition in human adults is variable, often show- 
ing an early increase, with a later decrease. In atrophic infants, it is usually 
increased, sometimes normal or decreased. In animals it is variable, but 
usually decreased. During hibernation, there is in all cases a remarkable 
decrease in the number of leukocytes, which apparently migrate out of the 
vessels. Also during malnutritional edema there is a marked leukopenia. 
In pellagra and rickets, leukocytosis is usually found. In human and animal 
beriberi, a leukocytosis has been observed in most cases, though sometimes a 
leukopenia. In infantile scurvy there is usually a leukocytosis, but in human 
adults and in guinea pigs the total leukocyte count is nearly normal. During 
thirst the leukocyte count is variable. 

The differential leukocyte count is in general quite variable during inanition. 
In fasting human adults, the polymorphonuclear percentage is variable, usually 
increased at first, with decrease in later stages. The lymphocytes are also vari- 
able, usually decreased, but the eosinophiles are generally increased. In 
atrophic infants the polymorphonuclears frequently show a relative increase, 
sometimes also the lymphocytes; the other varieties are normal or variable. 

In animals, starved with or without water, the polymorphonuclears are usu- 
ally decreased. The lymphocytes are variable, most frequently showing a 


decreased percentage, sometimes after a preliminary increase. The mononuclear 
and transitional forms are variable. The eosinophiles are usually normal or 
increased, though sometimes decreased. 

During malnutritional edema, a lymphocytosis is very constant and char- 
acteristic. In pellagra, the results are variable and inconclusive. In rats on 
vitamin B deficiency, a marked lymphopenia has been observed. In human 
scurvy (especially adult) a relative lymphocytosis is usual; and a decrease in 
polymorphonuclears has been observed. There are no marked changes in 
scurvy of guinea pigs. 

The blood platelets have been studied but little during inanition, but seem 
as a rule to show no marked changes. A progressive decrease in number has 
been found during deficiency in vitamin A. In scurvy, they are normal or 
increased, hence they cannot be responsible for the hemorrhagic tendency in 
this disorder. 

On refeeding after inanition, the blood as a rule quickly recovers its normal 
condition, although a transient hydremia may occur, due to the more rapid 
regeneration of the plasma. Following prolonged or severe types of inanition, 
however, there may be a considerable delay in the recovery, doubtless due to 
delayed regeneration in the hemopoietic system. Regeneration of the blood 
is much retarded on protein-poor diets. During convalescence in human scurvy 
the erythrocyte count is sometimes remarkably high. 

(.4) Effects of Total Inanition, or on Water Only 

After an introductory discussion, the results will be considered first in man, 
adult and infant, followed by a review of the results in lower animals. 

On account of the intense interest in the blood, and the ease with which it 
may be examined, even during life, an enormous literature has accumulated 
upon the subject, including the effects produced by various forms of inanition. 
Harvey (1651), who discovered the circulation of the blood, concluded that it is 
to be considered a tissue of the body, rather than merely a liquid food, since it 
remains in quantity during starvation in man and animals. 

Opinions have varied widely concerning the extent of loss in the blood during 
starvation. The earlier authors were impressed with the general appearance of 
anemia, both during life and at autopsy. Rokitansky ('54), for example, con- 
cluded that in starvation, or atrophy from other causes, the loss is relatively 
greatest in the blood, exceeding even that in the adipose tissue. A general 
condition of anemia at death from starvation was likewise claimed by David 
(1815), Collard de Martigny (1828), Tiedemann ('36), Taylor ('20), Voelkel 
('86), Porter ('89) and many others. Claude Bernard stated that the dimin- 
ished resistance of starved animals is doubtless due to the decreased mass of 
blood. On the other hand, microscopic examination revealed surprisingly 
little structural change, so that Carl v. Voit ('81) and others held that the blood 
is one of the tissues least affected by inanition. According to Heidenhain 
('57), Cyr ('69), Falck ('81), Cohnheim ('89), Grawitz ('95) and most recent 
authors, the atrophy in total mass of the blood during inanition is, in general, 
proportionate to that of the whole body. 


There is, however, much variation in the changes in mass and composition 
of the blood, varying according to species, age, individual and type of inanition, 
as will appear in the subsequent review of the literature. The changes in 
physico-chemical constitution, which in general lie outside the scope of the pres- 
ent work, are reviewed by Burckhardt ('93), Lackschewitz ('93), Weber ('02), 
Lewinski ('03), Tria ('11), Schulz ('12), Robertson ('13), Gerpott ('13), Ash 
('14, '15), Nobecourt ('16), Trowbridge, Moulton and Haigh ('15, '18, '19), 
Lusk ('17), Hatai ('18), Moulton ('20) and others. In general, excepting 
extreme stages and variations in water content, it may be stated that the chem- 
ical composition of the blood undergoes relatively little change during inanition, 
its losses being largely compensated through absorption of materials stored in 
the various tissues and organs. Accurate analyses, however, show certain 
definite and characteristic changes during various types of inanition. 

Special mention may be made of the fat content of the blood, since the fat 
granules (as hemokonia) are visible by the ultramicroscope. Schulz ('96, '97) 
found the fat content of the blood markedly increased (sometimes doubled) 
in fasting pigeons and rabbits. This was confirmed by Daddi ('98b) for dogs in 
short fasting periods, although a progressive decrease occurs in longer fasts. 
Further data are cited by Weber ('02), Rothschild ('15), and Greene and 
Summers ('16). Bloor ('14) demonstrated that in dogs the result varies accord- 
ing to the nutritional condition of the dog preceding the inanition. Aside from 
the observations of Reicher ('09), Nobecourt and Maillet ('14) and Gage ('20, 
'21), apparently no attempt has been made to correlate these chemical changes 
with the morphological structure shown by the ultramicroscope. Some possible 
relations between the lipemia during inanition and the deposition of lipoidal 
fat in certain organs will be mentioned later. 

Changes in Human Adults. — The effects of total inanition (or water alone) 
upon the blood have frequently been studied in man. Morgagni (1761) found 
very little blood in the larger vessels of two men who died from chronic starva- 
tion. This was confirmed by Haller (1771). Dutrochet (1816) stated that the 
number of red blood corpuscles is increased by rich nutriment, and decreased 
by fasting. Donovan ('48) observed marked anemia in the Irish famine vic- 
tims. Brouardel ('76) found a high red cell count and a syrupy consistency of 
the blood in a case of starvation from esophageal stricture. The blood changes 
during Dr. Tanner's fast are given by Van der Weyde ('79-'8o). The digestive 
leukocytosis was studied by Detoma ('80). Cadet ('81) found that a 24 hour 
fast causes a slight increase in the number of red cells (per cu. mm.), and a 
decrease in the white cells and platelets. Curtis ('81) observed during a 45 
day fast great variations in the number (2,370,000-6,770,000), size and form of 
the erythrocytes, probably due chiefly to imperfections of technique. 

Hayem ('82, '89) claimed that in chronic cachexias, the loss in the total mass 
of blood is relatively greater than that of the whole body. The decrease in the 
red cell count is at first slight; later it may be extreme. The leukocyte count 
varies according to the cause of the cachexia. The number is usually increased; 
but, as likewise for the platelets, it may decrease before death. The structure 
of the leukocytes may also change, with cytoplasmic vacuolation and less 


distinct granulation. The ameboid movement is preserved. The nuclei may- 
appear larger and more vesicular. The differential leukocyte count is often 

In fasting insane patients, Andreesen ('83) found that the red cell count at 
first increases, later decreases. Ingestion of water causes a transient decrease, 
through dilution of the blood plasma. Refeeding may likewise cause a tempo- 
rary decrease in the red cell count. 

Senator ('87) in Cetti's 10 day fast (with loss of 11. 14 per cent in body 
weight) found the initial red cell count 5,720,000, decreasing to 5,287,000 on 
the 4th day, and increasing to 6,830,000 on the 9th day. After 2 weeks of refeed- 
ing, the count decreased to 5,730,000. The leukocyte count on the 9th fast day 
(not observed earlier) was 4,200, the ratio to red cells being 1 : 1,619. On the 
2nd day of refeeding, it was 12,300 (1:533). The hemoglobin, by Fleischl's 
hemometer, decreased from (initial) 110-118 to 85-90 on the 9th day of fasting. 

Luciani ('89, '90), in the 30 day fast of Succi, noted slight fluctuations in 
the red cell count (usually between 4.5 and 5 millions), ascribed chiefly to varia- 
tions in dilution of the plasma due to water ingestion. The leukocytes de- 
creased from 14,536 (initial) to 861 on the 7th day; increased to 1,550 on the 
9th day; with slightly higher counts later. On the whole, the blood appeared 
relatively resistant to change. 

Von Limbeck ('92) claimed an increase in the red cell count and hemoglobin 
content during fasting. Lehman et al. ('93), in Breithaupt's fast, found an 
increase in the red cell count from 4,953,000 to 5,150,000 on the 2nd day; with 
4,801,000 on the 6th day, and 4,812,000 on the 2nd day of refeeding. 

Tauszk ('94a, '96), in a repeated 30 day fast by Succi, found (as in the 
previous fast) but slight variations in the red cell count (range 4,840,000-5,472,- 
000). The form of the erythrocytes remained normal throughout. The 
leukocyte count decreased progressively from 9,600 on the 3d day to 4,200 on 
the 30th day. Differential count showed a relative increase in the eosinophiles 
(2.7-4.7 per cent) and in the polymorphonuclears (64.1-79.2 per cent), but a 
decrease in the mononuclears, including lymphocytes (33.1-16.0 per cent). 

Grawitz ('95) reviewed the results of inanition upon the blood of man and 
animals. He concluded that there is in general a decrease in total quantity, 
proportional to that of the body; but relatively slight change in composition, 
aside from a variable degree of hydremia, which may also appear upon refeeding. 
Extensive reviews of the literature on the subject were likewise made by 
Schwinge ('98), Pashutin ('02), Benzancon and Labbe ('04), Bardier ('13), 
Ash ('15), and Morgulis ('23). 

Cabot ('04) stated that fasting causes a temporary increase in the red cell 
count by concentration of the blood. In chronic malnutrition the leukocytes 
may decrease to 3,000. Opie ('04) found that starvation may decrease the 
eosinophile leukocytes to less than 0.5 per cent; but Meyer ('05) observed an 
increase of eosinophiles from 3.4-3.7 to 6.5 per cent in a healthy man after a fast 
of 24 hours. 

In a 14 day fast (1 liter of water taken daily), the observations on the blood 
by Charteris ('07) are summarized in the accompanying table. 


Blood Counts Observed during a Fast of Fourteen Days (Charteris '07) 




Red cells 

Differential count — percentage of 




phil le 





1-17 1 

1-2 1 
1-3 1 

2-3 ! 










11 , 148 
















II. 7 




11. 6 
















11 .0 






1 .0 
1 .0 

1 .0 





1 Signifies dates preceding the fast. 2 Subsequent refeeding. 

The blood changes appear relatively slight, aside from a moderate leukocytosis 
with gradual increase in eosinophiles. 

Gordon observed the blood of Martin ('07), a medical student, who fasted 
9 days on 24 ounces of water daily. There was no significant variation in the 
erythrocytes or total leukocytes, excepting a rise in the latter of 10,000 on the 
2nd and 9th days. The differential count was somewhat irregular. 

Benedict ('07, '08), in short fasts of 2-7 days, found a progressive fall in 
the erythrocyte count (with corresponding decrease in hemoglobin); and a 
relative leukocytosis, with a high per cent of polymorphonuclears, but a pro- 
gressive fall in the total leukocyte count in the longer fasting period. 

Penny ('09) records a few observations on the blood of a physician (self- 
experiment) fasting on water only for 30 days, as follows: 

Day of fast 

Red cells 




per cent 

per cent 

per cent 

per cent 

1 2th 

7 , 000 , 000 
6 , 000 , 000 


1 1 , 000 









Lustig ('n) concluded that during human fasting there is but little change in 
the red cell count, with a decrease in the number of leukocytes. Turk ('12) 


held that the human blood is relatively resistant during fasting. There is a 
tendency to increase in the red cell count; and in long underfeeding a notable 
decrease in the number of leukocytes, especially the neutrophiles. Lazarus 
('13) similarly concluded from the available evidence that total inanition in 
man and lower animals does not produce anemia; but in chronic inanition the 
results are more uncertain. 

Howe and Hawk ('12) in 2 men fasting 7 days noted an initial rise in the 
number of polymorphonuclear leukocytes, with a decrease below normal at the 
end. The lymphocytes have an opposite tendency. One man showed a pro- 
gressive increase in eosinophiles. Mann and Gage ('12) found an increased 
staining capacity in the leukocytic nuclei of man and frog, upon refeeding after 
a fasting period. 

Gruner ^-14) concluded from a review of the literature on the differential 
leukocyte count during starvation that the number of lymphocytes, after a 
preliminary decrease, remains nearly constant; while the neutrophiles show a 
progressive increase. He attempted to explain the blood changes in relation to 
those in the hemopoietic system. Schwartz ('14), in an exhaustive review of the 
eosinophiles, concluded that, while variable, they usually show a tendency to 
increase during inanition, especially in human fasting. 

Ash ('14, '15) reviewed the literature on the blood changes during inanition, 
showing the variable and often discordant results. He also made careful daily 
observations on the blood (hemoglobin, erythrocyte and leukocyte count, total 
and differential), together with some observations on coagulation time, density, 
etc., in the case of Levanzin, who fasted 31 days on water only, with loss of 21.9 
per cent in body weight. In general, the blood appeared very stable in composi- 
tion, with fluctuations not exceeding the normal range, and no significant change 
in the size, form or structure of the blood cells. As shown in Fig. 72, the ery- 
throcyte count ranged between 6 and 7 million, and the hemoglobin between 85 
and 93 per cent. There was, however, a transient marked rise in the total 
leukocyte count, from 6,000 up to 12,400 on the 3d fast day, decreasing to nearly 
normal in the latter half of the fasting period (Fig. 74). The polymorphonu- 
clear neutrophile count closely parallels the total leukocyte count (Fig. 74), 
rising from 60 to 79 per cent on the 3d fast day, decreasing irregularly thereafter, 
but about normal in the latter half of the test. There is no significant change 
in the lymphocytes and large mononuclears. The transitional form shows 
transient rises (to 5.0-6.5 per cent) on four different days (Fig. 73), and the 
eosinophiles show a decrease in the second half of the fasting period. 

Gage ('20, '21), using the dark-field microscope, studied the fat particles 
(hemokonia or "chylomicrons") which appear in the blood after ingestion of fat, 
and gradually disappear during fasting. 

Reiss ('21) reviewed the effects of malnutrition (war diet) on the blood. 
Lubarsch ('21a) emphasized the extensive destruction of blood corpuscles as an 
evident effect of the malnutrition (chiefly due to mixed deficiencies in the diet) 
during the war period. Stefko ('23) concludes from an extensive study that the 
blood during starvation may be either thickened or thinned. The thickened 
blood may collect in the inner organs, leaving the periphery anemic. The 



April. Ma/ 

11 E D 14 15 16 17 18 19 gOO ZZ £3 24 £5 26 £7 28 29 30 1' I 3 4 5 6 T 6 91011 IE 13 14 15 16 fT 

Day 3 


















16 17 











































































. — 




































Inanition Period 


Fig. 72. — Chart showing the slight changes in the number of erythrocytes (red blood cells 
per cu. mm.) and percentage of hemoglobin in Levanzin, fasting 31 days on water only. (Ash 

April May 

11 12 13 14 15 16 17 18 19 20 Z\ Zl £3 Z4 Zb Zb U 282930 1 Z 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 

Day 3 






4 5 

6 7 







14 1 

5 16 1 


3 20 


























; ' 

















[ I 





\ 1 





















1 ; 





/' \ 



\ .. 

». ! 














/ s 

1 ,' 



\ / 






! \ 






V / 













n Pe 





Fig. 73. — Chart showing the fluctuations in the differential leukocyte count (eosinophile, 
basophile and transitional forms) in the blood of Levanzin during his fast of 31 days on water 
only. (Ash '15.) 



April M/v/ 

11 \l 13 14 15 16 IT B19Z0glgaZ4g5ae7ZBe930 1 £ 3 4 5 6 7 8 9 10 11 [I D 14 15 16 17 

Day 3 


1 1 1 

3 4 5 

6 ' 

r a 



Ll 12 13 




[ 18 









n m 

c9 30 












1 1 * ' Tnariirio 

n- Period 








f f 
































1 i 























f f 


























_ a 










V j 







v ' 






























1 y 




























\ / 
























7 _ 













/ ^ 





/ \ 














Small Lymph, 


1 / 














Large Mono. 








Large Lymph. 



Fig. 74. — Chart showing the changes in the total leukocyte count and in the various types 
of leukocytes (per cu. mm.) in the blood of Levanzin during his fast of 31 days on water only. 
There is an initial rapid rise in the total count, with irregular decrease to nearly normal in the 
latter part of the fasting period. This change in the total count is apparently due to the 
change in the number of polymorphonuclear neutrophiles, there being no significant change in 
the small and large lymphocytes and large mononuclear forms. (Ash '15.) 


thickening is due to an excessive consumption of the plasma, resulting in a high 
red cell count (" Pseudopolyglobulie ") . There is also a leukocytosis, with relative 
increase of mononuclear and transitional forms. Thinning of the blood is due 
to a hydremia associated with edema, and may be due chiefly to protein inani- 
tion. Further data upon this topic will be cited in connection with infantile 
malnutrition and also under the various types of partial inanition. 

Changes during Malnutrition in Infants. — In considering the blood changes 
in atrophic infants, it must be kept in mind that pedatrophy is usually the result 
of gastrointestinal or other chronic disorders, and in many cases represents toxic 
effects as well as those of chronic inanition. 

Parrot ('77), who described infantile athrepsia, considered progressive leu- 
kocytosis a characteristic symptom. D'Orlandi ('99), however, found the total 
and differential leukocyte counts unmodified in hypothrepsia (mild or moderate 
uncomplicated inanition). In acute and chronic athrepsia, Cuffer ('78) ob- 
served an increased red cell count (up to 9 millions) as well as increase in total 
leukocytes (10,000-40,000). Cantalamassa ('92), claimed that anemia results 
in children who are starved as well as in cachexia from chronic diseases. 

Schlesinger ('03) found that in moderate infantile atrophy there is a variable 
anemia, due to dilution of the plasma; but in more severe cases the anemia may 
be masked by loss of the plasma. In uncomplicated atrophy, the leukocyte 
count may be normal or below normal, with normal differential; but in gastro- 
enteritis the blood is increased in density, with increased red cell and leukocyte 
count, showing lymphatic and polynuclear hyperleukocytosis. Just before 
death, there is a marked and rapid fall in the density of the blood, with decreased 
red cell count. 

Thiercelin ('04) claimed marked and constant blood changes in athrepsia. 
At first, the blood appears concentrated; but later there is progressive anemia 
and the red cell count may decrease to 3 million, or even below 1 million (Luzet). 
The red cells are also malformed, and often nucleated. There is considerable 
leukocytosis. The blood is easily coagulable ; hence frequent venous thromboses 
may occur in the brain, lungs and kidneys. Rieber ('05) noted leukopenia in 2 
moderately severe cases of pedatrophy, with an increase of polymorphonuclears 
during severe complications. 

Arneth ('05) concluded that the characteristic leukocytosis found in newborn 
infants is not due to the deficient nutrition during that period. 

In chronic athreptics, Minet ('07), found 19 showing an increased erythro- 
cyte count (5-7 millions) due to dehydration; while only 5 showed a decrease 
(below 5 million). Nucleated red cells were found in 9 out of 19 cases. Pro- 
gressive leukocytosis (11,000-37,000) appeared in 15 cases; and a decrease 
(below 9,000) in 4. A relative increase in polymorphonuclears occurred in 18 
out of 19 cases; often with indistinct nuclei and neutrophile granules scarce or 
absent. The polynuclear eosinophiles appeared rare (0.5-1 per cent) or absent. 
The neutrophile myelocytes or basophiles were constant (0.2-3 per cent). 
Minet also noted that digestive leukocytosis occurs as in normal infants, and is 
sometimes very marked, especially in late stages of cachexia (confirmed by 
Villa '18). 


Herter ('08) found a variable degree of simple anemia in infantilism resulting 
from chronic intestinal infection. Benjamin ('08) stated that severe infantile 
malnutrition causes a polynuclear leukocytosis; but in atrophy (decomposition) 
the lymphocytes are decreased. The eosinophiles tend to disappear and the 
large mononuclears are mostly replaced by younger forms. 

Lenoble ('08) in a 7 months' infant convalescent from malnutrition found 
slight anemia. The differential count shows 77 per cent of lymphocytes and 2.3 
per cent of eosinophiles, the other varieties diminished. Lesage ('n) stated 
that atrophic infants in general show a variable degree of anemia, and often 
leukocytosis (especially lymphatic). 

Lust ('11) found that acute nutritional disturbances in infants usually cause 
a concentration of the blood, which may change to hydremia in terminal stages. 
In chronic malnutrition, however, the blood usually changes but slightly in water 

Rosenstern ('11), however, concluded that in (total) complete inanition there 
is no change in concentration of the blood, contrary to the earlier views. The 
total quantity of blood decreases in proportion to the body. There is a decreased 
leukocyte count. In chronic malnutrition the effects are variable. There may 
be hydremia upon refeeding. 

Nobecourt and Maillet ('14) found that the minute fat granules (hemokonia), 
visible in the blood through the ultramicroscope, usually decrease in number or 
disappear during infantile malnutrition, but reappear during recuperation when 
milk diet is resumed. These granules evidently correspond to the "chylomi- 
crons" of Gage ('20, '21) above mentioned. 

Nobecourt ('16) reviewed the literature of blood changes in hypotrophic and 
cachectic infants. The changes appear variable, with more or less anemia, but 
sometimes an increase in erythrocyte count. The leukocytes are also variable 
and may be normal, increased or decreased in number. The differential count 
may be nearly normal. 

Meyer and Japha ('19) described 3 types of infantile anemia, associated with 
hydremia. These may be caused by toxic influences, constitutional weakness 
(aplastic anemia) or alimentary deficiencies. 

Schindler ('10) thought that the increased pigmentation of the iris in mal- 
nourished infants may be hematogenous in origin, due to increased destruction 
of erythrocytes. The hemosiderosis observed by Helmholz ('09) and others 
in the liver and spleen may be similarly explained. 

Bergel ('19, '21) believes that the fats and lipoids exert a specific chemotac- 
tic effect upon the lymphocytes; and that the lymphocytosis in the blood during 
fasting is due to the characteristic lipemia associated with the mobilization of 
the reserve body fat. 

Marfan ('21) found the data upon blood changes in malnourished infants 
obscure and contradictory. In mild or moderate malnutrition (hypothrepsia) 
there is usually anemia with reduction in the erythrocyte count to 3 or 4 millions, 
the leukocytes being unchanged. But in severe stages (athrepsia) the blood is 
concentrated through dehydration, with erythrocyte count of 5 or 6 millions; 
also marked leukocytosis (up to 30,000), with relative increase of polynuclears 
and decrease in mononuclear forms. 


Utheim ('22) states that in Marriott's clinic the decrease in blood volume 
and the concentration (anhydremia) are considered important factors in the 
pathogenesis of athrepsia. 

Inanition in Animals. — The changes in the blood of animals (vertebrates) 
during total inanition, or on water alone, will now be considered. 

Morgagni (1761) observed that in a starved puppy the heart was distended 
with blood, but the large vessels nearly empty. Haller (1771) likewise noted 
a decreased amount of blood in the vessels of fasting frogs. Lucas (1826) 
confirmed Dutrochet (1816) as to the apparent decrease in the number of red 
blood corpuscles, but observed no change in the color and consistency of the 
blood in starved animals (various amphibians, birds and mammals). Collard 
de Martigny (1828) in starved dogs and rabbits always found "la presque 
entiere vacuite du systeme sanguin." In rabbits of the same litter he measured 
the amount of blood escaped and remaining in the large vessels, finding a 
progressive decrease up to n days of starvation. Thus the blood, though 
concentrated, decreased in volume so that he believed death results from its 
insufficiency to supply the tissues. Magendie (i85i-'52) likewise observed a 
concentration of the blood, with marked increase of solids and red corpuscles, 
in a stallion given water only. 

Chossat ('43) in starved pigeons (loss of 40 per cent in body weight) found an 
average decrease in collectable blood from 12.74 to 4.88 g., an apparent decrease 
of 61.7 per cent. He considered the decrease in solids still greater, since the 
water content of the blood increases, whether water is ingested or not. Starved 
chickens likewise became anemic. Bidder and Schmidt ('52) observed an even 
greater apparent decrease of 94 per cent in the escaped blood of a starved cat, 
with loss of about 50 per cent in body weight. The unreliability of this method 
is shown by Sedlmair's ('99) observation of 54.3 g. in the weight of the collect- 
able blood in a control cat, 1.5 g. in one starved cat, and 27.9 g. in another. 
Heidenhain ('57), however, by more accurate quantitative methods found the 
blood volume in starving mammals to decrease nearly in proportion to the body 
weight. His results were confirmed by Panum ('64) and Voit ('94) for the dog. 
See ('66) also concluded that during inanition the loss in blood volume is roughly 
proportional to the decrease in body weight, with a tendency to hydremic anemia. 
Bourgeois ('70), like many earlier authors, noted merely an apparent poverty 
of blood in the vessels of various mammals and birds, starved with or without 

C. H. Schultz (1843) ascribed death in starved proteus, cats and dogs to 
oxygen deficiency caused by the marked shrinkage in the red blood cells. A 
similar shrinkage described by Jones ('56) and other observers was probably 
due to imperfect technique in the examination of the blood. 

Malassez ('75) gave an extensive review of the earlier literature; and also 
made observations on fasting guinea pigs and frogs, indicating a decrease in the 
total mass of blood. Red cell counts indicate a preliminary increase, due to 
concentration of the blood, with a later decrease, due to hydremia. Duperie 
('78) found in a young guinea pig an increase of 500,000 per cu. mm. in the red 
cells in 1 day of fasting. In an adult in 4 days the red cells increased from 5,518,- 
oco to 6,696,000, but the leukocytes meanwhile decreased from 12,000 to 5,200. 


Buntzen ('79) found in fasting dogs an increased red cell count, which 
decreased on refeeding. Both consumption and regeneration thus appear more 
rapid in the plasma than in the red corpuscles. Reyne ('81) in a starving dog 
likewise found a progressive increase in the red cell count up to death at 25 days; 
but the leukocyte count showed great variation. In guinea pigs on absolute 
inanition up to 6 days, Cadet ('81) also found a progressive increase in the red 
cell count, but a decrease in leukocytes and platelets. 

Nasse ('50) stated that the white corpuscles become less numerous in the 
blood of emaciated frogs. In starving larvae of Rana and Bufo, Cunningham 
('80) found the red blood cells to undergo decolorization and fatty degeneration. 
Leonard ('87) noted seasonal changes in the blood cells of Rana temporaria. 
Mosso ('87) found the red blood cells of fasting frogs and tortoises more resis- 
tant to hemolysis (confirmed by Gallerani '92). Ehrlich ('91) found more eosino- 
philes in the frog's blood during the winter, and more mast cells upon refeeding 
in the spring. In starved tritons, Jolly ('01) noted that after refeeding there 
appear in the blood spherical erythroblasts which show mitoses. In Amia calva 
starved 20 months, Smallwood ('16) found a marked reduction in the number of 
red and white blood cells, but no apparent morphological changes. 

In pigeons, Lukianow ('88, '89) found no marked change in the water con- 
tent of the blood during total inanition. Hoffman ('22) and Palmer and Hoff- 
man ('22), however, found in starved pigeons a reduction in the red cell count 
from 3.1 millions to 2.8 millions, and an increase in leukocytes from 170,000 
to 280,000. 

In a dog, given water alone for 43 days, Luciani and Bufalini ('82) found the 
blood more concentrated, with increased hemoglobin, in the earlier days; but a 
decrease later. Hayem ('82, '89), in a dog fasting 25 days, found the red cell 
count increased from 4,200,000 to 5,550,000 on the 18th day, then decreasing to 
4,800,000 at the end. Kahan ('83, '84) in 6 starving dogs likewise found an 
increased concentration of the blood in the earlier days; later it decreased but 
may not fall below normal. He also noted in later stages that the red cells 
become smaller and crenated. 

Groll ('87) and Hermann ('88) found the ratio of hemoglobin to dry sub- 
stance during total inanition increasing in rabbits up to 16 days, in cats up to 22 
days, and in dogs up to 21 days; but the ratio decreases in longer experiments. 

Luibomudrow ('93), in 17 fasting dogs, found the red cell count stationary 
or increasing up to a loss of 10-15 P er cent m body weight, later decreasing up 
to death. There are great individual variations. In 15 of the dogs, the leu- 
kocyte count decreased up to a loss of 20 per cent in body weight; later increas- 
ing, sometimes above normal. The lymphocytes decrease, both relatively and 
absolutely, except in the early period. The mononuclears increase from 
10 to 25 per cent; and the polynuclears decrease relatively at the beginning. 
Eosinophiles sometimes appear although previously absent. 

Lackschewitz ('93) claimed that in fasting cats the water content of the 
erythrocytes may vary considerably, but that of the serum remains fairly 
constant, contrary to the general belief. 

Poletaew ('93, '94, '95) made daily blood counts in 10 dogs starved to death, 
8 on total inanition and 2 on water only. During total inanition, the red cell 



count increases progressively, with some fluctuations, to the end, excepting 
in some cases a slight decrease in the last few days. With water, the red cell 
count increases up to a loss of about 30 per cent in body weight, then decreases 
gradually, finally below the initial count. The changes in the leukocyte count, 
total and differential, are shown in the accompanying table (average of 8 dogs 
on total inanition). 

Leukocyte Counts in Starved Dogs (Poletaew) 

Stage of body weight 












Before inanition 

To loss of 10 per cent. . . 
20 per cent 

14, 196 

1 ,082 

7. 12 




4. 11 



40 per cent 


Okintschitz ('93, '93a) made differential leukocyte counts on 8 rabbits 
during total inanition and refeeding. During inanition the relative number of 
lymphocytes and polymorphonuclears is progressively decreased, while the 
eosinophiles and "round nucleated" forms correspondingly increase. On refeed- 
ing, the normal relations are gradually reached. His results, however, have 
been criticised by Kallmark ('n) and Schwartz ('14). 

In 12 rabbits on total inanition, London ('95, '96) found that the total blood 
volume decreases in proportion to the body weight. Popel ('96) found a con- 
stant increase in the density of the blood in 10 rabbits and 10 dogs during total 
inanition. In rabbits starved 4 days, or underfed 12 days, Kieseritzky ('02) 
noted a concentration of the blood, with increased red cell count, which gradu- 
ally decreased to normal upon refeeding. Ronsse and Van Wilder ('03) in 
8 rabbits on total inanition likewise noted increase in concentration and red 
cell counts, although there was some tendency to decrease in the longer fasts, 
as shown by the accompanying table. 

Daily Erythrocyte Count, by Ronsse and Van Wilder ('03), in Fasting Rabbits, 
Averaged for 3-day Periods (I-IX) 



wt. grams 

wt. grams 

Average erythrocyte count (millions) in successive periods 

















2 ,760 

3 ,022 

1 ,402 






















In guinea pigs fasting for various periods up to 8 days, Opie ('04) found a 
temporary increase in x the eosinophile leukocytes of the blood, followed by a 
decrease in relative and absolute numbers. This was ascribed to the effects on 
the bone marrow, although the marrow showed no marked decrease in eosin- 
ophiles. There was some diminution in the number of eosinophiles in the 
tissues of the lung and small intestine, and especially in the spleen. On 
refeeding, the number of eosinophiles appeared irregular. 

Bidault ('04, '04a) noted a slight increase of eosinophiles in the blood of a 
horse after 1 day without food. 

Cesaris-Demel ('06) observed a marked decrease in the red cells of rabbits 
with experimental marasmus produced by toxins. This would, perhaps, corre- 
spond to the condition frequently found in infantile athrepsia. Roger ('07), 
in 8 rabbits on absolute inanition, found the water content of the blood to under- 
go a slight increase, but dropping below normal on the 4th day. There was a 
marked increase on refeeding, with gradual return to normal. 

In a dog starved 8 days, Keuthe ('07) noted at first a decrease in the relative 
number of polymorphonuclears, with an increase in lymphocytes; later the 
converse, with recovery on refeeding. In cats, Beeli ('08) observed during 
the first third of the starvation period an average red cell count of 6,950,000, 
and total leukocyte count of 10,000. During the last third of starvation, 
the red cells averaged 8,000,000; the leukocytes 5,400. No morphological 
changes in the blood cells were found. Schelble ('10) also made some total 
and differential leukocyte counts in cats during various stages of digestion and 

Kallmark ('n) made a careful study of the leukocytes in young rabbits 
(2^-6^-2 months old). Some were subjected to acute inanition, on water 
alone, for 7-14 days; others were fed barely enough to maintain body weight, 
18-47 days. Considerable individual variation was found, but there usually 
appears, especially in acute inanition, a transient decrease in the lymphocytes 
and amphophiles (pseudoeosinophiles) at the beginning, and a similar increase 
on refeeding. These fluctuations are correlated with changes in the lymphoid 
organs. The acidophile and basophile leukocytes undergo no significant change 
in number, but the latter show rarefaction and peripheral displacement of the 
basophile granules. Nucleated cells occur rarely in the blood of fasting rabbits, 
but appear numerous upon refeeding. Argaud and Billard ('n) in 2 rabbits 
on total inanition noted on the fourth day a hypoleukocytosis, with inversion 
of the formula (3 mononuclears to 1 polynuclear). 

Howe and Hawk ('12) found that 3 out of 4 dogs fasting for various periods 
showed a decreased polymorphonuclear count, with an increase in small lympho- 
cytes; the fourth showed the reverse. The large lymphocytes were variable in 
2; nearly constant in the other 2. The transitional, basophile and eosinophile 
blood cells usually appeared nearly constant. The blood became normal on 
refeeding. One dog on a repeated fast showed a different effect, all leuko- 
cytes excepting the large lymphocytes being nearly constant. Brasch ('12) 
studied the effects of various diets upon digestion leukocytosis in the rabbit 
and dog after 3 days of fasting. 


Trowbridge, Moulton and Haigh ('15, '18, '19) and Moulton ('20) observed 
the composition and quantity of (escaped) blood in cattle on different planes of 

Whipple and Hooper ('18) and Whipple, Hooper and Robscheit ('20) found 
some regeneration of erythrocytes in dogs with anemia due to hemorrhage, 
even during total inanition. Protein for regeneration in this case is apparently 
derived from the breaking down of other tissues. 

Utheim ('21, '22) observed in fasting rabbits a decrease in the blood volume 
(through loss of water); but in young rabbits held at maintenance by under- 
feeding, the blood volume rapidly returns to normal. Ikeda ('22) studied the 
blood of rabbits in various stages of total complete and incomplete inanition, 
finding a transient increase in the leukocyte (especially the lymphocyte) count, 
followed by a progressive decrease. The number depends upon the rate of 
regeneration in the lymphoid organs. There is in the earlier stages of fasting a 
lipemia, which is expressed by a transient fatty infiltration of the liver and 
kidney. Okuneff ('22) similarly explained the deposition of cholesterin lipoids 
in the suprarenal cortex, liver and spleen as a result of the cholesterinemia in 
the blood of fasting rabbits. 

Moehl ('22) found anemia (frequently pernicious anemia) among underfed 

Fisch and Emmel ('24) observed no significant change in the erythrocyte 
count of albino rats during acute inanition, but polychromasia and reticulation 
(normally present in a small percentage of erythrocytes) disappeared. At 108 
hours of acute inanition, Streicher and Emmel ('24) found an average decrease 
of 31 per cent in the total leukocyte count, associated with a relative neutro- 
phile increase, a pronounced lymphocytic decrease, and a 75 per cent decrease 
in the azurophile granulation in the lymphoid cells. These results differ 
markedly from those during lactation leukopenia, which is apparently an 
independent phenomenon. 

Some observations upon the blood during hibernation have been recorded. 
Prunelle (181 1) noted that in hibernating mammals the peripheral blood tends 
to collect in the central portion of the body (confirmed by Baroncini and Beretta 
'01). Valentin ('38) found the total blood volume relatively unchanged in the 
hibernating marmot. He also discovered ('65, '8i) that the white corpuscles 
have nearly disappeared from the blood, which he ascribed to stasis 
in the lymph stream. He also noted a decrease in the size and number of 
the red blood cells. Quincke ('82) observed a decrease of about 30 per cent 
(to 3 1/2 millions) in the red cell count; these cells being normal in form, but of 
various sizes, some containing yellow droplets. 

In the hibernating hedgehog, Carlier ('92) found the red cell count unchanged 
or increased. The leukocytes, however, undergo a remarkable decrease from 
18 or 20 thousand down to 1-3 thousand. They emigrate out into the connec- 
tive tissue, being found abundant in the gastric mucosa and submucosa. 

Pappenheim ('01) observed no anemia in the blood of hibernating spermo- 
philes, and no degenerative changes in the red bone marrow. Argaud and Billard 
(,'n) induced dormice to hibernate in summer by starvation. After 10 days 


only a few leukocytes, and these of the mononuclear type, were found in the 

Polimanti ('13) found in the hibernating marmot an increased red cell 
count, due to concentration of the blood; but a decrease in leukocytes. 
Rasmussen ('16) and Rasmussen and Rasmussen ('17) in hibernating marmots 
found but little change in the relative blood volume, or in the red cell and plate- 
let counts. The number of circulating leukocytes is reduced about one-half. 
On feeding and drinking after awakening, the normal blood count is restored, 
but no digestion leukocytosis is found. 

(B) Effects of Partial Inanition 

The effects of partial inanition upon the blood will be considered under 
deficiencies of protein (including malnutritional edema and pellagra), salts 
(including rachitis), vitamins (including beriberi and scurvy), and water. The 
malnutrition during infantile atrophy is doubtless frequently a mixed deficiency, 
involving one or more forms of partial inanition, although it was classified under 
total inanition in the previous section. 

Protein Deficiency. — For reasons previously given, malnutritional edema 
and pellagra are classified under protein deficiency, although their etiology is 
still somewhat uncertain. 

Various earlier experiments (Verdeil '49, Subbotin '71, et al.) upon the dog 
with bread or similar protein-poor diets indicated a production of anemia. Voit 
and his school held that the hemoglobin content of the blood varies according 
to the protein content of the diet. Von Hosslin ('82), however, found but 
little difference in the blood of dogs on protein-rich or protein-poor diet. 

Morozoff ('97) concluded that in man a meatless diet causes an increase in 
red cell count, but a decrease in the leukocytes. 

Boycott and Chisholm ('n) observed marked variation in the normal red 
cell count for rats, with no significant change on various (especially protein- 
poor) diets causing marked loss in body weight. 

Whipple and Hooper ('18) and Whipple, Hooper and Robscheit ('20) found 
that the regeneration of blood in dogs following hemorrhage is much less with 
a diet of sugar, gliadin or casein, than with a diet containing hemoglobin, 
gelatin or meat. No single amino-acid appears to determine the blood regenera- 
tion. Geiling and Green ('21) likewise found that in rats blood regeneration 
after hemorrhage is markedly retarded on diets poor in protein, vitamins or 
salts. Rubner ('19) stated that the lowered birth rate in Germany during the 
war was probably due to the lack of protein in the diet which prevented normal 
regeneration of the red blood cells, thereby interfering with normal menstrua- 
tion and fertilization. 

During malnutritional edema, anemia is almost constantly observed. This 
was noted by Vacker ('71) in the malnourished children during the siege of 
Paris. During the recent world war, hydremic anemia associated with edema 
was observed by Woltmann ('16), Budzynski and Chelchowski ('16), Maase 
and Zondek ('i7),Lange ('i7),Landa('i7), Knack and Neumann ('17), Gerhartz 


('17), Jansen ('18), Burger ('20), Enright ('20), Mann, Helm and Brown ('20), 
Lubarsch ('21a) and Tallquist ('22). 

As to the detailed blood changes associated with this edema, Woltmann 
described anisocytosis, polychromatophilia, marked leukopenia and lympho- 
cytosis. Budzynski and Chelchowski found the hemoglobin reduced 50 per 
cent, but only slightly decreased red cell count; also a definite leukocytosis, 
mainly due to increase in lymphocytes, which equalled or surpassed the polymor- 
phonuclears. Eosinophilia was almost constant. Maase and Zondek found 
decreased hemoglobin (50-70 per cent), red cell count (3-4 millions) and leu- 
kocytes (4-5 thousand) ; with marked increase of lymphocytes (up to 45 per 
cent) and large mononuclears and transitionals (n-25 per cent). Gerhartz 
noted a tendency to low leukocyte count. Jansen found the red cell count 
1.5-4 millions; and leukopenia (below 5,000) in 60 per cent of the cases, with 
relative lymphocytosis (30-55 per cent). Schittenhelm and Schlecht ('19) also 
noted a relative lymphocytosis. Burger observed variable counts; the red cells 
usually slightly decreased (4 millions) and the leukocytes usually normal. 
Enright found a typical count of erythrocytes 3.5 millions (hemoglobin 60); 
leukocytes 4,000; polymorphonuclears 36, lymphocytes 48, and mononuclears 
16 per cent; eosinophiles normal. Tallquist claimed that the increase in 
mononuclears is not a specific symptom, although lymphocytosis is charac- 
teristic. Lubarsch found a marked destruction of erythrocytes, with hemor- 
rhages into the connective tissues, resembling scurvy. In famine edema in 
Russia, Abel ('23) described a slight decrease in erythrocyte count, and moderate 
leukopenia with decrease in neutrophiles and occasional lymphocytosis. 

Hydremic anemia has also been noted in various animals on protein-poor 
diets; by Friedberger and Frohner ('08) in draft oxen and horses; by Frohner 
and Zwick ('15) in sheep and cattle; by Hoare ('15) in sheep, cattle and pigs; and 
by Hutyra and Marek in various domestic animals. Kohman ('20) produced 
edema with anemia in rats by an aqueous diet poor in protein and fats; and her 
results were confirmed by Maver ('20) in rats, dogs and guinea pigs. 

In pellagra, Marie ('08, '10) concluded that secondary anemia is apparently 
frequent, though not constant. Leukocytosis is infrequent, probably due to 
complications. The differential count is uncertain, but there is probably an 
increase of large mononuclears. Findlay ('20) stated that although observers 
in Italy and Roumania have reported an increase in large mononuclears, this 
has not been generally confirmed. Thus Bardin ('13) and Hillman ('13) in 
America noted a definite increase in the small lymphocytes. Woodcock ('18) 
found some lymphocytosis in Turkish war prisoners, but Paton ('18) obtained 
a normal differential leukocyte count. In Armenian refugees, Findlay noted an 
increased total leukocyte count, with a relative increase in lymphocytes and 
decrease in neutrophiles. Huck ('23) has recently given an extensive review of 
the literature on the subject (including several references in addition to those 
above mentioned). He concludes that in all stages of uncomplicated pellagra 
the blood picture shows a secondary type of anemia, with normal leukocyte and 
platelet counts. The differential count shows an increase in lymphoid elements 


in severe pellagra, with increase of polymorphonuclear eosinophiles during 

The effects of a dietetic deficiency of iron upon the blood structure have 
attracted much attention, on account of the iron content of hemoglobin. Since 
the loss of iron from the body is slight, however, but little is needed in the food 
except during growth. Von Hosslin ('82) in young, growing dogs on diets very 
deficient in iron found at first a decrease in hemoglobin with but slight loss in 
the total mass of blood; but later a decrease also in the blood volume. Similar 
experiments with variable success in the production of anemia in animals on iron- 
poor diets were made by Hall ('94, '96); by Kunkel ('95), Tartakowsky ('04) 
and Stoeltzner ('09a) in puppies; by Hausermann ('97), in man, rats, rabbits 
and kittens; by Abderhalden ('00) in rats, rabbits and guinea pigs; by Schmidt 
('12) in white mice; and by Brinchmann ('21) in guinea pigs. Lazarus ('13) 
opposed Immermann's doctrine that inanition causes the purest type of hypo- 
plastic anemia. He concluded that the anemic appearance (pale skin and 
mucosae) is often deceptive, and that a qualitative dietetic deficiency, espe- 
cially of iron, is more important than a quantitative deficiency. Happ ('22) 
found that well balanced diets, deficient in iron, do not produce anemia in the 
rat in the first generation, although slight anemia may occur in the second 

In human rickets, Comby ('01) stated that the blood is usually found nor- 
mal, though sometimes simple or splenic anemia occurs (with enlarged spleen). 
According to Cheadle and Poynton ('07), there is usually a simple anemia, 
proportionate in general to the severity of the rickets. The leukoctyes are 
usually increased, and nucleated red corpuscles occur. The extreme anemia 
associated with an enlarged spleen is probably due to a syphilitic complication. 
Wohlauer ('11) reviewed the literature, indicating usually alteration of the blood 
in rickets, but the data are variable. Heubner says the erythrocytes may sink 
to 2 or 3 millions, with leukocytosis up to 20 or 30 thousand; the changes being 
somewhat proportional to the severity of the rickets. Poikilocytosis, nucleated 
red cells and megalocytes may occur. Kuttner's findings were similar. Schiff 
and Widowitz, however, in rachitis with severe digestive disturbances 
found a condition of chlorosis, with marked decrease of hemoglobin without 
corresponding decrease in the red cell count. 

In adult osteoporosis, due to chronic dietetic deficiency in protein, calcium 
and phosphorus, Alwens ('19) found the hemoglobin below 80 in 16 cases, with 
the red and total white cell count within physiological limits. There was 
lymphocytosis in 13 cases, and eosinophilia in 7. Happ ('22) concluded that 
diets producing rachitoid changes in the rat may also produce anemia, if the 
diet is prolonged, and also in the second generation. 

The effects of oxygen deficiency are of interest in this connection. Albitzki 
('84) reviewed the earlier work on thjs topic. The blood changes include 
deformity and destruction of the erythrocytes, which are decreased in number. 
Granules (probably from degenerated red cells) are found in the blood and urine; 
hemoglobinuria, nasal hemorrhages and bloody diarrhea occur. Askanazy 
('13) mentioned blood stasis and ecchymoses as a result of oxygen-deficit, with 


anemia in chronic asphyxia. Martin, Loewenhart and Bunting ('18) obtained a 
progressive hyperplasia in the red marrow of the long bones in rabbits, which 
probably represents a compensatory hypertrophy of the hemopoietic tissue. 
The literature on oxygen deficiency is reviewed by Morgulis ('23). 

Vitamin Deficiency. — The effects of a deficiency in vitamin A upon the blood 
appear slight. Hess and Unger ('19) found no anemia in 5 infants fed 8 or 9 
months on a diet deficient in vitmin A. Cramer, Drew and Mottram ('21a) 
and Happ ('22) likewise observed no anemia in rats with diets thus deficient. 
Cramer, Drew and Mottram ('22), however, noted a progressive decrease 
in the number of blood platelets (thrombopenia) in the rat. Bedson and 
Zilva ('23, '23a) failed to confirm this, finding no marked decrease in the 
platelet count. 

Upon diets deficient in vitamin B, the results are more striking. In human 
beriberi, Takasu ('03) found in infants a decreased erythrocyte count (range of 
2,400,000-4,800,000 in 17 cases), but nearly always an increase in leukocytes 
(range of 9,000-34,000 in 15 cases). In chronic cases there is an increase in the 
mononuclear leukocytes. Chun ('17) observed an increase of about 100 per cent 
in the leukocyte count. Findlay ('20), however, claimed that in beriberi there is 
a decrease in the total leukocytes, though not in the lymphocytes or neutro- 
philes. Nagayo ('23) claims that human beriberi differs from experimental 
polyneuritis in that anemia is not present and lymphocytosis is frequent. 

The work of Breaudat ('10) was inaccessible. 

In 200 chicks and 150 pigeons with experimental polyneuritis, Tasawa ('15) 
noted general anemia as a striking symptom. Weill, Arloing and Dufourt ('22), 
in pigeons on polished rice diet, likewise found a rapid and progressive decrease 
in the red cell count and hemoglobin, with a tendency to considerable leukocytosis. 
Hoffman ('22) found that even in normal pigeons there is a marked individual 
variation in the blood cell counts (erythrocytes 2.4-4.0 millions; leukocytes 
40,000-320,000), so conclusions as to changes must be guarded. In 13 normal 
pigeons the erythrocytes averaged 3 . 1 millions per cumm. , the leukocytes 1 70,000 
in 7 latent polyneuritic, erythrocytes 2.5 millions, leukocytes 220,000; in 16 
severe polyneuritic, erythrocytes 2.2 millions, leukocytes 260,000; in cured 
polyneuritic pigeons, erythrocytes 3.4 millions; leukocytes 390,000. These 
results appear to agree with those of Weill, Arloing and Dufourt, and also to 
those (previously cited) for general inanition in the pigeon. 

Suski ('23) obtained somewhat different results with adult pigeons (300- 
380 g.). On vitamin-free diet (autoclaved rice, wheat protein, lard and 
salt mixture) there was marked loss in body weight, slight loss in average 
erythrocyte count, and some irregularity in differential leukocyte count (abso- 
lute leukocyte counts not given). With the addition of butter and orange juice 
to the diet (deficient in vitamin B only), there was simDar decrease in body 
weight, increase in average erythrocyte count (from 4,140,500 to 4,780,000), and 
in relative polymorphonuclear count (from 48 to 50 per cent) with correspond- 
ing decrease in lymphocytes and transitional forms. 

In a dog on polished rice diet with symptoms of polyneuritis, Brucco ('20) 
observed a fall in the hemoglobin content and the erythrocyte count. 


In rats on diets deficient in vitamin B, Drummond ('18) found (in the 
"black variety of Mus norvegicus") no significant change in total or differential 
blood counts. Cramer, Drew and Mottram ('21, '21a), however, observed a 
marked decrease in lymphocytes (lymphopenia) in the circulating blood of 
rats, associated with a general atrophy of lymphoid tissue throughout the body. 
Prompt recovery occurred upon administration of vitamin B. Happ ('22) 
concluded that although deficiency in vitamin B does not produce anemia in the 
rat, "Diets so deficient in water soluble B as to produce polyneuritis diminish 
leucopoietic activity and cause a severe leucopenia with a shift to the right in the 
Arneth formula." Weitbrecht ('22) found in young rats on vitamin-free diets 
(also deficient in iron) a tendency to anemia of chlorotic type. In all his 
experiments there was a reduction in the number of leukocytes (especially 
lymphocytes) , with nuclear changes in form. This lymphopenia is ascribed to a 
general atrophy of the lymphatic apparatus. 

The blood changes in scurvy have been studied frequently, on account of the 
hemorrhagic tendency characteristic of this disorder. In infantile scurvy, 
Fraenkel ('06) found a simple anemia, with decreased hemoglobin and erythro- 
cyte count, leukocytosis and appearance of nucleated red blood cells. Nobe- 
court, Tixier and Maillet ('13-15) concluded that the anemia arises from 
myeloid lesions, and recovery is sometimes long delayed. Hess and Fish ('14) 
in infantile scurvy found the hemoglobin very low (35-70), although the red 
cell count may be normal or slightly below. There is usually a leukocytosis 
(10,000-40,000). The blood platelets vary within normal limits. Brandt ('19) 
found the hemoglobin nearly normal, the erythrocyte and platelet count some- 
times increased, the total and differential leukocyte count nearly normal 
(sometimes lymphocytosis). 

In human adult scurvy, Sato and Nambu ('08) found a marked hydremic 
anemia. In 54 cases, the hemoglobin averaged 31.8 per cent; the red cell count 
2,409,323; leukocytes 6,856. In 19 convalescent, the hemoglobin averaged 
50.7; erythrocytes 3,539,947; leukocytes 7,405. The differential count was 
nearly normal; blood-platelets somewhat increased. In convalescent scorbutics, 
Wassermann ('18) noted cases where the erythrocyte count rose to 6 or 7 
millions; the hemoglobin to no or 120 per cent. Aschoff and Koch ('19) and 
Bierich ('19) described the hemorrhages and histological appearances during 
absorption of the extravasations. Comrie ('20) found secondary anemia and 
lymphocytosis, with the following average in 50 cases; hemoglobin 55 per cent; 
erythrocytes 4,080,000; total leukocytes, 7,510; polymorphonuclears 45 per 
cent; large lymphocytes 20 per cent; small lymphocytes 29 per cent; mononu- 
clears, 2 per cent; eosinophiles 4 per cent. Hausmann ('22) also made differ- 
ential counts in scurvy, noting a reduction in the neutrophiles. Hess ('20) con- 
cludes that in general the blood-picture in scurvy resembles that of chlorosis, with 
hemoglobin decreased proportionately more than the number of erythrocytes. 
The variations reported in total and differential leukocyte count may depend 
upon various stages, degrees of severity, or complications in the cases studied. 

In experimental scurvy of the guinea pig, no marked blood changes have been 
observed. Jackson and Moore ('16) found no appreciable leukocytosis. 


Herzog ('21) obtained negative results as to erythrocytes, total and differential 
leukocyte count, and blood platelets. Findlay ('21a) found only a slight 
decrease in the erythrocyte count. Bedson (' 2 1 ) in guinea pigs and monkeys (also 
one human) found but slight variation from the normal in the red cells, leuko- 
cytes (total and differential) and platelets. He reviews the literature showing 
contradictory results on the platelets. Only Mouriquand ('21) obtained more 
positive results, finding a decrease in the red cell count from 5,518,000 (hemo- 
globin 80 per cent) to 3,250,000 (hemoglobin 65) in 24 days, with anisocytosis 
and poikilocytosis. On giving orange juice, the red cells increased to 5,201,000 
(hemoglobin 60) on the 28th day; with complete recovery and 5,406,000 red 
cells (hemoglobin 90) on the 37th day. 

Effects of Water Deficiency. — According to Lorenzen ('87), a decrease in 
drink to reduce body fat was recommended by Plinius, and in the 19th century 
by the French physician, Dancel. The method was rediscovered by Oertel, who 
ascribed obesity to excessive water consumption. He explained the effect of 
thirst through decrease of the water-content in the blood, producing a concen- 
tration of the red corpuscles which was supposed to increase the oxidation of the 
body fat. Denning ('99), however, found but slight increase in the human red 
cell count or hemoglobin during thirst (up to 1 week). Naegeli ('12) stated that 
the blood may become concentrated, with abnormally high red cell count, as a 
result of thirst, or of loss of body fluid through diarrhea, etc. Rubow ('20) con- 
cluded that in dry diet cures the effects are partly due to the resultant concen- 
tration of the blood, the water content of which may be reduced 8-12.5 per 
cent. Marriott ('20) holds that diarrhea in infants may give rise to a toxic 
condition, with anhydremia, largely due to the general desiccation of the body 
and measured by the concentration of the blood. In a recent review, Marriott 
('23) concludes that thirst causes anhydremia with impairment of the circula- 
tion, resulting in marked functional disturbances throughout the body. Through 
loss of water from the blood plasma, the blood is greatly concentrated, and the 
red cell count may be doubled. Destruction of red cells also occurs in severe 
cases, with resultant delay in recovery after water is administered. 

Among animals, the effects of thirst upon the blood have often been studied. 
Falck and Scheffer ('54), in a dog fed dry bread found the water content of the 
blood decreasing from 86.11 to 82.83 per cent. Bowin ('80) in dogs and rabbits 
on a relatively dry diet (meat or vegetables) found the red blood cell count 
nearly normal for 4 or 5 days, then becoming progressively higher, reaching 
nearly double (8,544,000) on the 13th day. 

Giirber ('89) noted that while in a frog kept in a moist place the erythrocyte 
count decreased from 836,000 to 516,000 in 6 days, in another frog kept in a 
dry place it increased from 865,000 to 1,352,000 in 5 days. Durig ('01) found 
a smaller increase of 3-16 per cent in the red cell count of frogs losing 
13-35 per cent in body weight by desiccation. 

Pernice and Scagliosi ('95a) found in a dog fed dry bread the red cell count 
increased somewhat during the first 4 days, from 5,177,000 to 7,409,000, with 
corresponding hemoglobin increase from 65 to 105. Later the red cell count 
decreased to 4,712,000 and the hemoglobin to 55, on the nth day, shortly 


before death. The leukocyte count, on the contrary, decreased from 12,400 to 
9,300 on the 4th clay, then steadily increased to 41,850 shortly before death. 
In 3 young chickens fed dry maize, there was found a progressive erythrocyte 
count during the 8 or 9 days up to death, the increase in the 3 cases being 
from 3,131,000 to 3,596,000, from 3,069,000 to 4,092,000, and from 3,007,000 to 
4,185,000, respectively. The hemoglobin values were practically doubled 
(45-55 up to 90-100). The leukocyte count was variable, showing a final 
decrease in the first case (12,400 to 4,650) and increases in the other 2 (4,65c to 
17,050 and 10,850 to 13,950). 

In 2 dogs fed dry meat powder mixed with fat, Straub ('99) found the dry 
content of the blood increased from 22.03 P er cent (average) to 24.49 P er cent. 
Blix ('16) obtained a reduction of n per cent (maximum) in the water content 
of the blood in rabbits by fasting and thirst. Keith ('22) studied the dilution 
of the blood and the changes in hemoglobin and red cell counts in dogs which were 
given water after total inanition for periods of 2-4 weeks. 



Changes in the lymphoid tissue of the bone marrow have already been 
described in Chapter VII, with the skeleton. The lymphoid tissue of the 
alimentary canal, spleen and thymus will be considered along with these organs 
in later chapters. The present chapter deals with the lymph and lymph glands 
in general. Although the characteristic involution of the lymphoid tissue during 
malnutrition usually results in atrophy of the lymph nodes, they often appear 
swollen (especially in rickets, beriberi and scurvy), perhaps chiefly through 
secondary infections in conditions of lowered resistance. After a brief summary, 
the details will be presented under (.4) effects of total inanition, and (B) effects 
of partial inanition. 

Summary of Effects on the Lymph and Lymphatic Glands 

During total inanition the quantity of lymph apparently increases during the 
first third of the fasting period, but gradually decreases later, with changes in 

The lymphatic glands during inanition appear variable in size. In ema- 
ciated human adults they are in many cases extremely atrophic, but in others 
they appear normal in size or even enlarged (probably from secondary infections). 
Enlargement appears more frequent in atrophic infants. In fasting animals, 
the results are also variable, although marked atrophy of the lymph glands 
appears characteristic. During partial inanition, changes in the size of the 
lymph glands likewise appear somewhat inconstant, but enlargement appears 
characteristic in rickets, beriberi and scurvy, especially in the mesenteric 
nodes, often probably due to secondary infection. 

Microscopically, the lymphatic glands during inanition usually show a very 
characteristic atrophy of the lymphoid tissue, even in cases where a decrease 
in the size of the gland as a whole may be offset by a distension of the blood 
vessels and lymph sinuses. In general, there is a marked diminution in the 
number of lymphocytes (by emigration), which renders the less affected 
stroma (reticulum) and trabeculae very prominent. The lymphoid nodules 
and cords are reduced in size, and mitoses are decreased in number or absent. 
Numerous phagocytic cells are found, often containing pigment derived from 
excessive destruction of erythrocytes (especially in regions of hemorrhage in 
scurvy). An increased number of phagocytes and plasma cells has been noted 
during hibernation. Retterer's claim that lymphatic glands may be trans- 
formed into hemolymph glands by inanition lacks confirmation. Secondary 
infections may occasion inflammation, however, and occasionally even suppura- 
tion of the lymphatic glands, especially in scurvy. 



Cirrhosis of the lymph glands has been noted in pellagra. The lymphoid 
tissue appears especially sensitive to a dietary deficiency of fat, while in rickets 
a general lymphoid hyperplasia appears characteristic. A deficiency in vitamins 
(especially of vitamin B) tends to cause a general atrophy of the lymphoid 
tissue, associated with lymphopenia in the circulating blood. During chronic 
thirst, the changes in the lymph nodes resemble those typical for inanition in 
general, with hyperemia and lymphoid atrophy. 

Although there are numerous variations, the changes in the structure of the 
lymphatic glands during inanition in general resemble those found in the other 
lymphoid organs, including the bursa of Fabricius ("cloacal thymus" in birds), 
bone marrow (considered with the skeleton) and the thymus, spleen and intes- 
tinal lymphoid structures (to be considered in later chapters). 

Upon adequate refeeding after inanition, the lymphatic glands in general 
recuperate promptly, showing rapid increase of weight, associated with active 
mitosis and recovery of normal structure in the lymphoid tissue. 

(A) Effects of Total Inanition, or on Water Only 

The effects will be considered first in man, adult and infant, and then in 
the lower animals. 

The observations upon the effects of inanition on the lymphatic system of 
the human adult appear rather scarce. In autopsies upon emaciated victims 
of the Madras famine (226 men, 155 women and 78 children), Porter ('89) found 
the mesenteric lymph glands nearly normal in size in about a third of the adults 
and five-sixths of the children (57 men, 37 women and 51 children). They were 
enlarged and swollen (probably chiefly through infections) in 36 men, 6 women 
and 11 children, and were extremely atrophied in 73 men, 74 women and 2 
children. Pigmentation of the glands was noted in 12 men, 8 women and 
2 children. In 12 cases (4 men, 4 women, 4 children) of extreme emaciation 
without evident complications, the mesenteric glands appeared about normal 
in size in 1 man, 3 women and all 4 children; much enlarged in 1 man; and 
atrophied in 2 men and 1 woman. 

In an adult man who died of starvation (60 days on water only) with loss of 
about 40 per cent in body weight, Meyer ('17) carefully studied the lymph 
nodes, 18 hours post mortem. "The inguinal lymph nodes were barely palpable 
on both sides, and all except the medial nodes, which were slightly reddish, were 
pale. The mesenteric nodes were small and pale, but numerous. The pre- 
vertebral nodes formed a chain of soft, flat, pale bodies, and the only very red 
specimens were a pair of iliac nodes, one on each side. That on the right was 
1 X 0.5 centimeters and only a few millimeters thick. That on the left was very 
much smaller. There was nothing noteworthy about the rest of the nodes of 
the entire body. Some of the right bronchial nodes were calcified. The 
cysterna chyli contained a little yellowish fluid." 

Microscopic examination revealed the following: 

"Iliac Nodes. — The parenchyma especially of these lymph nodes is very 
much depleted and some portions of the nodes are comprised of the collapsed 


coarser framework merely. Germinal centers are absent and large trabeculae 
and large sinuses are especially evident. Some of the latter contain granular 
detritus. Few degenerated cells are found, however, although some yellow 
pigment is present. Only a few polymorphonuclear leukocytes are seen, but a 
good many acidophile cells with finely granular protoplasm and round vesicular 
nuclei are present. The parenchyma is so depleted that one can actually 
count the cells in sections ten microns thick in most portions of the sections. 
The nuclei of the lymphocytes contain few chromatin granules, stain lightly, 
and look more transparent than usual, some of them appearing as empty 

"Prevertebral Nodes. — The prevertebral nodes seem somewhat better 
preserved, but show some polychromatophilia. They, too, contain no germinal 
centers, phagocytes, or giant cells. Large cells with a large, oval, vesicular 
nucleus, which look as though they might have an endothelial origin, are 
usually numerous. These cells are contained in both sinuses and parenchyma. 
These portions contain large acidophile cells and masses of degenerated 

"Bronchial Nodes. — The bronchial nodes are not so depleted as might be 
expected, but show considerable pneumonokoniosis. They contain almost 
nothing but small lymphocytes. Some portions of the abdominal lymph nodes 
are wholly depleted, being represented merely by a folded mass of trabeculae and 
connective tissue." 

It may be noted, however, that Retterer ('02b) found that lymph glands of 
human adults examined 24 hours post mortem may show as an artefact the 
rarefaction of the lymphoid tissue and other changes resembling those produced 
in the lymph glands of animals by inanition. 

The lymph glands during infantile malnutrition have been frequently 
studied. Baginsky ('84, '84a) and Fede ('98) noted atrophy of the intestinal 
follicles and lymphoid tissue; but Mattei ('14), on the contrary, found evidence 
of hyperactivity, as will be shown later, in the chapter on the alimentary canal. 
Moldenhauer ('99) noted hyperemia and increase of stroma in the mesenteric 
nodes of athreptic infants. Thiercelin ('04) described the axillary and inguinal 
lymph glands in athreptic infants as swollen (secondary infection ?) and the 
mesenteric glands as slightly hypertrophied. 

Among 1,000 New York primary school children, 6-12 years of age, of whom 
40 per cent were malnourished, Sill ('09) observed that 90 per cent had "ade- 
noids," 40 per cent hypertrophied tonsils, and 4.5 per cent tubercular cervical 
lymph nodes. In another series of 210 markedly malnourished children, 
75 per cent had enlarged cervical glands, and of 101 tested for tuberculosis, 55 
gave a positive (von Pirquet) reaction. Schelble ('10), however, found no 
significant histological changes in the mesenteric glands from 17 cases of peda- 
trophy, 9 of which were fixed shortly after death. 

Among the lower animals, data concerning the effects of inanition upon the 
lymphatic system are more numerous. From experiments on several dogs and 
rabbits of various ages and sizes, which were subjected to absolute inanition for 
variable periods up to death, Collard de Martigny (1828) concluded that: 


"Durant le premier tiers environ du temps de l'abstinence, la quantite 
de la lymphe est tres considerable, et augmente d'autant plus, que l'animal est a 
jeun depuis plus longtemps. Dans les deux autres tiers du temps de l'absti- 
nence, la quantite de la lymphe diminue graduellement. Quelques heures 
avant la mort, le canal thoracique n'en contient que tres-peu. Les vaisseaux 
lymphatiques des diverses regions du corps se vident de lymphe d'autant plus 
tard comparativement qu'ils s'en etaient remplis moins lentement. Generale- 
ment la vitesse avec laquelle la lymphe parcourt des vaisseux est tres-peu con- 
siderable. Elle augment d'autant plus, que la quantite de ce fluide devient 
plus forte. Elle diminue graduellement, a mesure que la lymphe est en moindre 
proportion dans le systeme lymphatique. Durant la periode de son augmenta- 
tion en quantite, la lymphe devient graduellement plus riche en matiere color- 
ante, en caillot et en fibrine. La lymphe est d'autant moins coagulable, coloree 
et fibrineuse, dans la reste de la duree de l'abstinence, que la mort et moins 

Tiedemann ('36) found that with the decrease in blood volume during inani- 
tion, there is an increased activity of the absorbent system, so that the fluids of 
the serous and joint cavities are diminished. 

Bourgeois ('70) stated that the results of Collard de Martigny as to the 
lymph during inanition were confirmed by Magendie and Bouchardat. In 
numerous fasting mammals, Bourgeois found "Les ganglions lymphatiques sont 
tres-developpes, quelquefois injectes, surtout dans l'abdomen." 

The observations of Cunningham ('80), Hofmeister ('87), Erdely ('05) and 
Holthusen ('10) on the atrophic changes in the intestinal lymphoid tissue during 
inanition will be considered in the chapter on the alimentary canal. 

In 6 severely starved young and adult rabbits, Morpurgo ('88, '89, '89a) 
noted atrophic changes in the intestinal lymphoid tissue (to be mentioned later) 
and also in the lymphatic glands. The cervical and mesenteric glands appear 
reduced in size and consistency. The lymphoid cells are greatly decreased in 
number, in both the superficial and the deeper lymph spaces. The medullary 
cords are very slender and poor in lymphoid cells. Mitoses persist, however, 
both in cortex and medulla, in the places where mitoses normally occur. "Leur 
nombre est certainement diminue; et il me semble aussi que la substance chroma- 
tophile etait devenue plus rare." The intercellular "tingible bodies" of Flem- 
ming were observed. 

In similar rabbits refed 4 or 5 days after inanition Morpurgo ('90) found the 
number of lymphoid cells in the mesenteric lymph glands, especially in the 
medullary cords, more numerous. The lymph vessels still appeared to contain 
relatively few cells, however. Numerous mitoses were found in both cortex 
and medulla, and they appeared larger in diameter than during starvation. 
Phagocytes and " tingible bodies " of Flemming appeared less numerous. 

Retterer ('02) subjected adult guinea pigs to total inanition, and during 
the first 5 days found the lymphatic ganglia grey in color, with dilated lymph 
sinuses free from red cells. In animals starved to death in 7 days, however, 
the small peripheral ganglia appeared reddish; the central ganglia appeared grey 
on the surface, but the medulla was reddish, with sinuses containing numerous 


red corpuscles. He also found all the ganglia with blood-filled sinuses in a dog 
emaciated by repeated hemorrhage and inanition, although this was not 
observed in a cat after 12 days of total inanition. Retterer concluded that the 
ordinary lymph glands may be changed into hemolymph glands by inanition. 

In a more detailed histological study of the lymphatic glands from the 
fasting animals, Retterer ('02a) found the sections appearing spongy and 
rarefied. The medullary cords and lymphoid follicles appear less distinct 
than normally. The reticulum loses its affinity for hematoxylin, but stains 
deeply with the acidophile stains. In the lymphoid masses, the lymphoid 
cells have become separated by the cytoplasmic atrophy, and the stroma, 
though also atrophic, becomes distinct as granular filaments. The atrophy 
of the lymphoid tissue is accompanied by enlargement of the peripheral and 
central lymph sinuses, which contain free cells of various types: (1) numerous 
chromatic nuclei of 4 to 7^; (2) leukocytes; (3) cells with clear, unstained cyto- 
plasm and irregular margins; (4) cytoplasmic masses with several nuclei. In 
many of the free cells, the cytoplasm has undergone " degenerescence hemo- 
globique" (phagocytosis of red corpuscles?). The nuclei of the lymphoid cells 
are poor in nucleoplasm, and the chromatin may become fragmented so as to 
simulate mitosis (as claimed by Morpurgo). In summary: 

"En un mot, l'atrophie qui suit l'abstinence prolongee se traduit dans le 
ganglion lymphatique par la rarefaction du tissue et la transformation du 
protoplasma commun et continu en cellules libres ou leucocytes. La mac- 
eration, les agents mecaniques ou chimiques conduisent au meme resultat." 

According to Hammar ('09), unpublished observations by Hellman show that 
the lymphoid tissue in general undergoes "accidental involution" during inani- 
tion, though not so marked as in the thymus. The changes during inanition 
in the weight of the bursa of Fabricius (a lymphoidal appendage of the cloaca in 
birds) were observed by Jolly and Levin ('11) for the pigeon, chicken and duck. 
With a loss of 30-37 per cent in body weight, the bursa lost 48-77 per cent, 
or about the same as the thymus and spleen. On refeeding pigeons 8-15 days 
after starvation for 8 days, with increase of 28 per cent in body weight the bursa 
of Fabricius increased 102 per cent, which was relatively greater than the 
increase in the spleen (53 per cent), but less than that in the thymus (246 per 

The distribution of fat in the lymph nodes during inanition was studied by 
Holthusen ('10). Holmstrom ('11) noted an increased deposit of lipoidal 
granules in the lymph nodes, chiefly in the sinus reticulum, of fasting rabbits. 
Normal conditions were restored after a week of refeeding. 

Jolly ('n) studied the histological changes in the bursa of Fabricius (or 
"cloacal thymus") in pigeons during inanition. As in the involution of the 
thymus, there is an atrophy, especially in the cortex, due to emigration of 
lymphocytes. "Les modifications histologiques que nous venons de decrire 
consistent done essentiellement en une disparition graduelle des lymphocytes 
avec conservation du bourgeon epithelial qui forme la trame de la substance 
medullaire. Cette involution rappelle celle qui est due a l'age, mais elle n'est pas 
definitive. Si on laisse mourir l'animal, elle n'a pas le temps d'aboutir a 


l'atrophie sclereuse; si on renourrit l'animal, le follicule se repeuple en lympho- 
cytes en peu temps et se reconstitue." 

In the lymphatic glands, Jolly ('14) found the changes during inanition less 
marked than in the spleen and bursa of Fabricius (above noted). Two puppies 
1 month old were given water only for periods of 6 and 8 days, respectively, 
with 2 normal controls. The loss in body weight was 27.1 per cent; in the 
cervical and popliteal ganglia, 41.3 per cent; mesenteric ganglia, 56.9 per cent; 
thymus, 68.1 per cent; spleen, 73.5 per cent. In a rabbit of 6^ months, starved 
7 days, the loss in body weight was 23.5 per cent; popliteal ganglia, 38.8 per cent; 
vermiform appendix, 43.2 per cent; mesenteric ganglia, 52.5 per cent; spleen, 
62.8 per cent; and thymus, 87.9 per cent. 

The histological changes correspond to the degree of loss in weight. The 
lymphocytes become scarce, the reticulum distinct. The cortex of the lympha- 
tic glands is less affected than the medulla. The follicles become smaller, and 
persist long (as in the spleen), but finally disappear. The terminal lesion is a 
sclerous atrophy. Cells containing blood-pigment, etc. occur in both lymph- 
glands and spleen. Mitoses progressively diminish in number, but disappear 
in only extreme stages. Other lymphoid organs, s ach as the vermiform appendix, 
tonsils and bone marrow, are similarly affected. The effects are more marked 
in the lympho-epithelial organs (thymus, bursa of Fabricius) than the lympho- 
lymphatic (ganglia) or hemo-lymphatic (spleen, bone marrow). The effects 
appear progressively greater in the following order: peripheral ganglia; tonsils; 
mesenteric ganglia; bone marrow, appendix; spleen; bursa of Fabricius; thymus. 
The lymphoid nuclei supply nitrogenous and phosphorized materials for the 
starving organism (Jolly '24). 

Howell ('14) found that in dogs starved 48 hours the lymph shows but little 
evidence of chyle-fat, and also a great decrease in the number of lymphocytes, in 
comparison with the milky lymph an hour or two after feeding. 

Ikeda ('21, '22) studied rabbits in various stages of acute and chronic inani- 
tion, and upon refeeding. During starvation the mesenteric and peripheral 
lymph nodes, in addition to atrophy and fatty changes, show degenerative 
changes in the parenchyma. The lymph follicles of the vermiform appendix, 
besides the degenerative process, may show also regenerative activity, with 
mitosis of the lymphocytes. In the mesenteric glands during inanition, there is 
fat in the parenchyma, but not in the sinuses as extracellular fat. Animals with 
well developed lymphatic system and richly lymphocytic blood are much more 
resistant to inanition. Ikeda considers it probable that at a certain period of 
inanition (related to the consumption of fat?) there is a vigorous demand for 
lymphocytes, emphasized by an increased transport of lymphocytes from the 
hemopoietic system into the blood, with compensatory regeneration in the 
hemopoietic organs. 

The observations of Lefholz ('23) as to the relations between the diet and the 
amount of lymphoid tissue will be mentioned later, under partial inanition. 

Hibernation. — The lymphatic glands in the hibernating hedgehog were 
studied by Carlier ('92). The germinal centers become inactive, with but 
few mitoses. The phagocytes become very numerous, occurring in the lymph- 
sinuses and throughout the gland, excepting the germinal centers and the 


fibrous tissue. They are large and occasionally multinucleated. The cyto- 
plasm contains numerous yellow or brownish pigment granules (iron-containing), 
also fragments of degenerated tissue cells and of erythrocytes, etc. The 
macrophages become less numerous after hibernation and are distinct from the 
phagocytic giant cells of the spleen. During hibernation there is also an 
increase in plasma cells of the lymph nodes, which nearly disappear elsewhere, 
excepting in the tongue. 

(B) Effects of Partial Inanition 

A few observations have been recorded upon changes in the lymphatic 
glands during deficiencies of protein, fats and carbohydrates, salts, vitamins and 

During pellagra (classified as a protein deficiency), Kozowsky ('12) noted 
hyalin changes in the blood vessels of the lymph-nodes. Harris ('19) noted 
cirrhosis of the lymph-nodes, with increased pigment and decreased parenchy- 
matous elements. 

In human malnutritional edema, Schittenhelm and Schlecht ('19) found the 
mesenteric lymph glands somewhat swollen. 

In rats with edema produced by diets deficient in protein and fats, Kohman 
('20) found usually a congestion of the lymph glands, especially the cervical 

Lefholz ('23) continued the work of Settle, who noted an apparent hyper- 
trophy of the lymphoid organs of kittens on a diet rich in fat and calories. The 
palatine and pharyngeal tonsils and especially the intestinal (aggregated) 
follicles were found to become nearly twice as large on diets high in protein or 
sugar and also in calories; while if the excess calories are given in the form of fat, 
these organs become nearly 3 times the normal size. The mesenteric lymph 
glands also appear larger on a diet rich in fat than on one rich in sugar or 
protein. Other lymphoid organs (spleen, thymus and cervical lymph glands) 
show no constant response to variations in diet. There is some evidence 
indicating a reciprocal relationship in the size of these organs, so that if one is 
unusually small, the other will be large, thus making a tendency to uniform 
total amount of the lymphoid tissue. 

In human rickets, enlargement of the mesenteric lymph glands was observed 
by Whistler (1645) an d (slightly) by Glisson (1650). Seibold (1827) noted 
hardening of the lymph glands at autopsy. Dickinson ('69), Comby '(01) 
and Cheadle and Poynton ('07) found the lymph glands often swollen, and 
Jenner ('95) found "albuminoid infiltration" of these and other organs. Woh- 
lauer ('11) concluded that the lymph glands are usually swollen (especially the 
cervical, axillary and inguinal) ; although Frolich ascribes the swelling to infec- 
tious complications. Pfaundler ('22) states that in rickets the lymphoid tissue 
is markedly hypertrophied throughout the body. No data were found concern- 
ing the lymphatic glands in experimental rickets of animals. 

Beriberi. — According to Cyr ('69), Tiedemann produced in animals by faulty 
diet a disease resembling beriberi, in which the lymphatic glands appeared swol- 


len. McCarrison ('21) noted atrophy of the lymphoid structures in the intes- 
tines of polyneuritic pigeons. In monkeys on autoclaved rice diet, although the 
lymphoid nodules in the colon were frequently swollen, the lymphoid cells of 
the intestinal mucosa in general were greatly reduced in number. The mesen- 
teric glands, especially those of the colon were invariably much enlarged (from 
toxic absorption). 

Cramer, Drew and Mottram ('21, '21a) found that diets deficient in vitamins 
(especially vitamin B) produce in rats and mice a marked atrophy of lymphoid 
tissue thoughout the body, associated with lymphopenia in the circulating blood. 
Peyer's patches become very atrophic, and the ordinary lymph glands, even 
though not macroscopically decreased in size, are found histologically almost 
free from lymphocytes. "They consist almost entirely of endothelial cells and 
large empty lymph-spaces." They hold that there is a specific relation between 
vitamin B and the nutrition of lymphoid tissue. 

In human beriberi, Strong and Crowell ('12) found no enlargement of the 
superficial, cervical or mesenteric lymph of glands. The mesenteric glands were 
small. Tasawa ('15), however, mentioned hypertrophy of the gastrointestinal 
lymphatic apparatus as characteristic of human beriberi. Nagayo ('23) states 
that lymphatic hypertrophy is a characteristic distinguishing human beriberi 
from experimental polyneuritis, in which lymphatic atrophy occurs. 

In human scurvy, Lind (1772) frequently observed swollen and purulent 
axillary and mesenteric lymph glands. Sato and Nambu ('c8) found the mesen- 
teric glands often hemorrhagic, especially when intestinal ulcerations were 
present; also an increase of intestinal lymphoid tissue. Aschoff and Koch ('19) 
likewise noted that the inguinal lymph glands frequently showed blood in the 
peripheral lymph sinus, or hematogenous pigment in the stroma. "Weiter- 
hin ist ein haufiger Befund der Sinuskatarrh der Lymphdrtisen; die Sinus sind 
erweitert und vorwiegend mit abgestossenen und verfetteten Sinusepithelien 
prall ausgefiillt, doch tritt die Beteiligung von Leukozyten ganz zuriick. Auch 
Oedem der Lymphknoten wird ofters beobachtet." 

Jackson and Moore ('16) observed swollen axillary and inguinal lymph 
nodes in experimental scurvy of guinea pigs. Hess ('20), however, who has 
recently reviewed the subject, concludes that this enlargement probably occurs 
only in advanced cases complicated by general infection. As a rule, the 
enlargement is confined to nodes draining areas where hemorrhage has occurred. 
On section such nodes appear reddish or brownish on account of the contained 
blood pigment; sometimes the peripheral sinus is distended with pigment- 
laden cells. Where secondary infection has occurred, extensive necrosis of the 
glands may result, especially in the mesenteric nodes when severe intestinal 
lesions are present. Hojer ('24) describes lymphoid atrophy in advanced cases. 

Thirst. — In a dog which died after 11 days on a diet of dry bread, Pernice 
and Scagliosi ('95a) found the lymph nodes all hyperemic, with distended blood 
vessels. The lymphoid cells are less numerous, especially in the medulla, so 
the septa appear thicker and the reticulum more distinct. The intestinal 
lymphoid areas are also hyperemic. Similar changes were noted in the lymph 
glands of 3 young chickens on dry maize diet. 



The spleen, like the lymphatic system in general, usually undergoes atrophy 
during conditions of inanition and malnutrition. The splenic enlargement 
sometimes found is in most cases to be ascribed to complications, such as 
tuberculosis, syphilis or other infections. After a brief summary, the effects of 
inanition upon the spleen will be considered in detail under (A) effects of total 
inanition and (B) effects of partial inanition. 

Summary or Effects on the Spleen 

As to the changes in the weight of the spleen during inanition, the great 
normal variability makes it often difficult to reach conclusions. In general, 
however, it is clear that during total inanition (likewise on water only) in human 
and animal adults the spleen shares in the marked atrophy of the lymphoid 
organs. The relative loss in the spleen weight is usually greater than the loss 
in body weight, often twice as great; but the spleen loss may be slight in the 
early stages of inanition (guinea pig) . In emaciated human adults, the spleen is 
sometimes enlarged on account of complicating infections, such as tuberculosis 
or syphilis. 

In the young, the results are more variable. In some cases there is a definite 
decrease in the weight of the spleen, in others the inherent growth impulse may 
overcome the tendency to atrophy during inanition. In atrophic human in- 
fants, syphilis or other infections also may produce marked splenic enlargements. 

During hibernation and subsequent inanition, the changes in the weight of 
the spleen appear very irregular. 

Upon ample refeeding after inanition, the normal weight of the spleen is 
promptly restored, sometimes apparently with a transient over-compensatory 

During the various forms of partial inanition, the changes in the weight of 
the spleen are variable. In human pellagra and famine edema, the spleen usu- 
ally appears atrophic, but sometimes hypertrophied. In human rickets the 
spleen usually presents an enlargement of doubtful significance (often due to 
complications) . In experimental rickets of rats, the spleen is irregular in weight. 
In human beriberi, the spleen is usually enlarged; whereas in animals on diets 
deficient in vitamin B there is a very marked and constant splenic atrophy. In 
human and animal scurvy, the spleen is variable in weight ; usually enlarged, but 
often within normal range, or even subnormal in the early stages. During 
thirst, there is a very marked loss in the weight of the spleen, comparable to that 
during total inanition. 

During inanition in general, the structural changes of the spleen, though 
varied in degree and character, usually present certain characteristic features. 



In the somewhat exceptional cases of splenic enlargement, there is often conges- 
tion and general hyperplasia. In the more characteristic decrease in the size 
and weight of the spleen, the atrophy takes place in the parenchyma, but not in 
the stroma. The capsule and trabeculae thus become relatively more promi- 
nent, giving a variable degree of fibrosis and sclerosis. 

The atrophy affects especially the lymphoid tissue, so that the Malpighian 
nodules and pulp-cords become variably reduced in size and indistinct. The 
lvmphoid cells decrease in number and size. The nuclei tend to be slightly 
decreased in size and the number of mitoses is greatly diminished. In the 
sinuses and red pulp, the erythrocytes are variable in amount, giving in some 
cases an appearance of congestion; in others, of anemia. Correlated with the 
increased destruction of erythrocytes, there is a variable degree of hemosiderosis. 
Granular masses of pigment occur both extracellular and intracellular (in the 
endothelium and macrophages of the pulp-cords). Nucleated erythroblasts 
may also appear. 

During the various types of partial inanition, the structural changes in the 
spleen in general resemble those noted for total inanition. Where splenic enlarge- 
ment occurs, there is usually congestion and a variable degree of general hyper- 
plasia. In the more frequent diminution in size of the spleen, there is a relative 
fibrosis, with marked atrophy of the pulp (especially of the lymphoid structures) 
and a variable increase in pigmentation and hemosiderosis. Sometimes the 
changes appear inflammatory in character, and hemorrhages rarely occur 
(infantile beriberi, scurvy, thirst). It may furthermore be noted that these 
changes are not peculiar to inanition, but occur also in many other conditions 
affecting the spleen. 

Upon adequate refeeding after inanition of any type, the spleen in general 
makes a prompt recovery in normal structure as well as in size. Abundant 
mitoses accompany the proliferation in the lymphoid tissue during recuperation 
from its atrophic condition. 

(^4) Effects of Total Inanition, or on Water Only 

The changes in the weight of the spleen in man and animals will be reviewed 
first, followed by the changes in structure. 

Changes in Weight. — The normal weight of the spleen is so exceedingly 
variable that conclusions as to changes are often difficult. From a review of the 
literature, Willien ('36) concluded that the spleen is small and firm after inani- 
tion. At the autopsy of a 19 year old girl who had died from starvation, Schult- 
zen ('62, '63) noted a very small spleen — "Lien perparvus, ^Yi" longus, 2^" 
latus, i/-£" crassus." Curran ('74) found the weight of the spleen only 1 ounce 
in a greatly emaciated, starved old woman. Bright (et al.) ('77) stated that the 
spleen in a case of starvation was considerably below normal weight; but at 4^ 
ounces (about 130 g.) it was not much subnormal for an initial body weight 
of 121 pounds (final weight 74 pounds). 

Casper-Liman ('82) found a very small spleen at death from starvation, but 
opposed Tardieu's view that this is a characteristic sign of medicolegal impor- 



tance. In a large man, who died from voluntary starvation, Voelkel ('86) 
described the spleen as "auffallend klein (10 cm. lang, 5 cm. breit, noch nicht 
ganz 2 cm. dick)." Cohnheim ('89) held that during starvation the relative 
loss in the weight of the spleen is next to that of the adipose tissue. 

Porter ('89) recorded the weight of the spleen among native victims of the 
Madras famine, as shown in the accompanying table. 

Weight of the Spleen in Victims of the Madras Famine (Porter '89) 

No. and sex 

Average weight (and 
range) in ounces 

Average ratio to corre- 
sponding body weight 

16 men 

Spleen hypertrophied 
11. 4 (7-20 %) 
8.7 (6-12 Ji) 
Spleen of normal size 
4-8 (4-7) 
4 • (3-6) 
Spleen atrophied 
2.3 (all under 4) 
1.83 (all under 3) 

1: 125 

9 women 



1 7 women 



The normal weight of the spleen in native Indians was not known, but Porter 
concluded that its loss in weight was relatively greater than that in any 
other organ. 

Marked atrophy of the spleen was likewise noted by Formad and Birney 
('91) in two cases of death from starvation. 

Dlinschmann ('00) also held that the spleen is very small in starvation, 
reduced one-half or more; and Aschoff ('13) stated that the spleen loses in 
relative as well as absolute weight. Meyer ('17) found the spleen weight only 
53 g. in a man who died after 60 days on water only. 

The data by Krieger ('20) for the weights of the spleen in emaciated adults 
(infections excluded) are summarized in the accompanying table. 

Weight of the Spleen in Emaciated Adults (Krieger '20' 

Condition in group 

No. of 

Normal weight 



weight aver- 
age, grams 

loss in weight 

I. Insane, without chronic organic 





! 150 
1 123 1 








r 4 8.o 
I36.5 1 

1 Making allowance for normal decrease in weight of the spleen after age of so. 


It will be noted that the estimated loss in spleen weight varied from 27.6-48 
per cent in the various groups, which is approximately the same as the aver- 
age loss in corresponding body weight. 

Weber ('21) compared the weight of the spleen as found in 1,257 autopsy 
records at Kiel for the years 1914-1918. Unfortunately no body weights were 
available. Comparing the period of good nutrition (1914-1915) with that of 
subnutrition (1916-1918), the average weight of the spleen shows a slight 
decrease in the males from 156.5 g. to 137 g. ; and in the female from 141. 5 g. to 
128 g. From 50 necropsies in cases of death from starvation, Stefko ('23) 
concludes that there is a loss in both relative and absolute weight of the spleen. 

n — 1 — 1 — 1 — r 



— c 


••{*»?;•. •*. • - .'? 

Body Length in Cenfimctera 

1 ■»?«•?<»<»• T r " 1 1 1 1 1 lJj r 1 1 1 1 1 l_j 1 1 1 1 1 1 1 

Fig. 75. — Graph showing the individual weights of the spleen, according to body length, 
in atrophic human cases, newborn to adult, from various sources. The curve of normal spleen 
weight is from data compiled by Prof. R. E. Scammon. Great individual variation is evident, 
but in most cases the weight is below normal. 

Sison ('20) in 4 adult males on voluntary fasting observed a decrease in 
the splenic area of dulness on percussion, which he ascribed partly to a decrease 
in the size of the spleen, and partly to increased tympany of the overlying lung 

Bean and Baker ('19) from a study of a large series of organ weights at 
autopsy concluded that the weight of the spleen varies directly with the degree 
of general nutrition of the body (body weights not available). Pearl and Bacon 
('22), however, from a statistical study of the ratios between the weights of 
various organs at autopsy found an indication of an increased absolute weight 
of the spleen in fatal tuberculosis, which usually produces marked emaciation 
of the body. 

In Fig. 75, representing a field graph of the spleen weight in emaciated 
individuals (from various sources), under 20 years of age, arranged according to 



body length, it will be noted that in those above 120 cm. in body length, there is 
no very marked tendency to subnormality in the weight of the spleen. Many 
of the enlarged spleens are probably due to syphilis, tuberculosis, or other 
complications, however. Among the children (Fig. 76), the splenic atrophy is 
more striking, but here also the weight is often above normal. 

• 100 • 70 

•40 «7Q 
** •54*«' ''30 »3S •36 •7T»3S #J} 

mi Mil' 


, I ; 

CO • 


26 1 ' 

LO j ! • ! ! 1 1 1 

ZA 1 ' '» 

4- i x 

X 4- X 

??x x xx X 

cc I . 

X -1- -H 

X X 

?d V ■ 

tu lt * 


■ — r ASl > • 


^-"nJc.W^ ■ 

ifl £ 

J-j-we! n 

u-"3 S0^ , ' 

<^rf<^V ' . . 

1$ * 4- X ^ 

5^— r + t^ -"- " 

-(►-»■ it 

y • " 

*" jl : * : "*" ' zz 

~i~ XX 

• • ■• Ik • atr 1 1 


4 ■___, ,_._, " • BQdy.Lerid 

ttp j cni, 

4c 44 * 46 45 50 5c 54 56 

58 60 6c 64 * 3 66 68 70 

Fig. 76. — Graph showing the individual weights of the spleen, according to body length, 
in atrophic infants from various sources. The larger dots represent original Minnesota cases. 
The curve of normal spleen weight is from data compiled by Prof. R. E. Scammon. Most 
of the cases are clearly below normal, although there is great individual variation. 

Children. — Ohlmiiller ('82) observed a weight of 6.8 g. in the spleen of an 
atrophic infant of 8 weeks (body weight 2,381.2 g.), while in a ''normal" 
infant of the same age (body weight 4,149.5 g.) the spleen weighed 20.2 g. 
Unfortunately the previous weight of the atrophic infant is not stated. De 
Tommasi ('94) and Thiercelin ('04) concluded that in atrophic infants the spleen 
is small. Bovaird and Nicoll ('06) from weights at 571 autopsies on children of 
various ages (newborn to 5 years) at the N. Y. Foundling Hospital concluded 
that the spleen "has no apparent relation to the size or state of nutrition of the 
child." Mattei ('14) likewise concluded that the spleen in athreptic infants is 
very irregular in weight, though often small. In 7 cases below 3 months of age 
he found the spleen ranging between 5 and 10 g., body weight 2,260-3,100 g. 
(previous body weight not given). 

Lesage ('14) found a weight of 8 g. in the spleen of an atrophic infant of 4 
months (normal 15 g.). Nobecourt ('16) and Marfan ('21) concluded that in 
general the spleen is relatively small in malnourished infants. In famine- 
stricken children of various ages, Nicolaeff ('23) found the spleen 48-51 per 
cent subnormal in weight according to age. 


From the original data in Table 3, it will be observed that the weight of the 
spleen in atrophic infants is exceedingly variable. Of the cases above normal, 
only No. 1 1 can be excluded on account of syphilis. All the cases in which the 
age, body length, maximum and final body weights and weight of the spleen in 
atrophic infants were available are summarized in Table 2. Known syphilitic 
cases have been excluded. Thus it is apparent that, compared with the normal 
for final body weight, the spleen averaged 9.5 per cent above normal (or 5.6 
below normal in the Minnesota series alone). Compared with the normal for 
the maximum body weight during life, the spleen averaged 14.9 per cent below 
normal; and for height, 20.5 per cent below normal. For age, the spleen aver- 
aged 31.3 per cent below normal (18.1 per cent for the Minnesota series alone). 
Thus in this series the average loss in weight of the spleen appeared somewhat 
less than that in the body as a whole. 

On the other hand, from the field graph of all cases of spleen weight according 
to body length in atrophic infants, as shown in Figs. 75 and 76, it appears that in 
most cases the spleen weight is decidedly subnormal, according to body length. 
Doubtless if all complications (especially infections) tending to produce splenic 
enlargement could be excluded, the average weight of the human spleen during 
inanition would appear still lower. 

Among animals, a more uniform loss of weight in the spleen has frequently 
been observed during inanition. A remarkably small size of the spleen in dogs 
and rabbits on total inanition was noted by Collard de Martigny (1828). 
In pigeons on total inanition with average loss of about 40 per cent in body 
weight, Chossat ('43) found an average loss of 71 per cent in the weight of the 
spleen. Similarly, Bidder and Schmidt ('52) found an apparent loss of 72 per 
cent in a cat with loss of about 50 per cent in body weight. 

Manassein ('68, '69) also noted profound atrophy of the spleen in animals 
starved with or without water. Thus in 47 adult rabbits with average loss of 
39 per cent in body weight, the spleen lost 65 per cent. In 8 young adult rabbits 
(3?^ months old) with average loss of about 33 per cent in body weight, the 
spleen lost 52 per cent. In 3 young rabbits (23-25 days old), with body loss 
of 35 per cent, the spleen lost 38 per cent. An apparent loss of 74 per cent was 
observed in the spleen of 2 starved cats, and of 79 per cent in 2 starved crows. 
On fully refeeding 5 rabbits after a severe fasting period, the spleen had 
recovered normal weight. 

Bourgeois ('70) found in general the spleen atrophied to less than half its 
normal size in various starved mammals and birds, with or without water. 
A small spleen in dogs on total inanition or water only was noted by Carville 
and Bochefontaine ('75), Luciani and Bufalini ('82) and Mankowski ('82). 
In 6 fasting rabbits, with or without water, Skoritschenko ('83) found with 
average loss of 43.7 per cent in body weight an apparent loss of 26.5 per cent in 
spleen weight. 

In 20 pigeons on absolute inanition, compared with 20 controls, Lukianow 
('89) found an average loss of 34 per cent in body weight and of 58 per cent in 
the spleen, with no appreciable change in water content. In a dog fasting 22 
days with loss of 32 per cent in body weight, Voit ('94) found the spleen slightly 


heavier than in a normal control of the same litter (probably an individual 
variation). In newborn kittens on acute inanition (water only), Bechterew 
('95) noted a relative increase in the weight of the spleen in kittens, but a decrease 
in puppies. 

In 4 groups of guinea pigs (10 in each group) starved without water by 
Lazareff ('95), with average losses of 10, 20, 30, and 36 per cent in body weight, 
respectively, the corresponding losses in the weight of the spleen were o, 31, 
37 and 44 per cent (Table 5). Thus apparently the spleen is at first resistant, 
but loses greatly in the second period. Kusmin ('96) noted that, of the various 
organs, the spleen, liver and intestine suffered the greatest relative loss in 
rabbits and guinea pigs, with or without water, at ordinary or elevated tempera- 
tures. In 3 rabbits on water only with loss of 35-41 per cent in body weight, 
Weiske found a loss of about two-thirds in the fresh weight of the spleen, and 
slightly more in the dry weight. A similar loss in spleen weight was observed 
by Quattrochi ('01) in underfed puppies with loss of about 25 per cent in body 

Miescher ('97) noted during the migratory fast of the Rhine salmon an 
enormous decrease in the size of the spleen, which becomes reduced to stroma, 
with loss of blood and atrophy of the lymphoid cells. 

Beeli ('08) found apparent losses up to 83 per cent in the weight of the 
spleen in 4 cats starved (water only) with losses up to 51 per cent in body weight. 
Jolly and Levin ('12a) in birds (pigeon, fowl, duck) starved 4-9 days without 
water found an apparent average loss of 60 per cent in spleen weight; body loss 
30 per cent. In the guinea pig, they found the spleen loss 52 per cent, body loss 
36 per cent; and in the rat, spleen loss 46 per cent, body loss 26 per cent. Jolly 
('14) in 2 puppies, 1 month old, on water 6-8 days, with body loss of 27 per 
cent, noted an apparent loss of 73.5 per cent in the spleen weight, which was 
relatively greater than the loss in the thymus (68.1 per cent), or lymphatic 
glands (41-57 per cent). In 2 rabbits, 6}^ months old, with loss of 23.5 percent 
in body weight, the spleen lost 63 per cent, which was less than the thymus loss 
(88 per cent) but greater than that of the lymphatic glands (39-53 per cent). 

Giannelli ('16) reported a marked decrease in the size of the spleen and other 
viscera in the teleost, Tinea vulgaris, after 5 months without food. 

In adult albino rats on acute inanition (water only), Jackson ('15) found, 
with body loss of 34 per cent, an apparent average loss of 51 per cent in the 
weight of the spleen; while in a series on chronic underfeeding, with loss of 36 
per cent in body weight, the spleen lost only 29 per cent. The great individual 
variability in the size of the spleen was emphasized. 

In underfeeding experiments upon younger, growing albino rats, the effects 
upon the spleen weight appear variable (Table 4). In experiments on rats 
beginning at 3 weeks of age, Jackson ('15a) found an average loss of 42 per cent 
in the spleen of those underfed (at constant body weight) to 10 weeks of age, 
whereas in longer experiments the spleen showed but slight apparent loss. 
Stewart ('18) in rats severely underfed from birth found at 3 weeks an apparent 
loss of 49 per cent in the spleen weight; while in those underfed up to 10 weeks 
there was an apparent increase of 24 per cent in the spleen. In another series 


held at birth weight for 16 days, there was an average increase of 38 per cent in 
the spleen weight. Barry ('20, '21) likewise found the spleen 34 per cent above 
normal weight in the newborn (stunted) offspring of severely underfed pregnant 
albino rats. 

Upon refeeding albino rats after underfeeding from 3 to 12 weeks of age, 
Stewart ('16) found rapid recovery of normal spleen weight within a few days. 
There was apparently even an excessive (over-compensatory) growth of the 
spleen during the first 2 weeks of refeeding, but after 4 weeks the spleen weight 
became normal. The results of Jackson and Stewart ('19) likewise indicated 
an over-compensatory growth in the spleen after a short period of ample refeed- 
ing, with a tendency to lag behind later (Table 7). In such rats refed to adult 
stages after early periods of underfeeding, Jackson and Stewart ('20) found the 
spleen varying in different groups from 12.6 per cent below normal to 11.8 per 
cent above normal weight (Table 8). These differences are probably insignifi- 
cant, in view of the great normal variability in the weight of the spleen. 

In young steers held on various planes of nutrition, including those markedly 
retarded in growth by underfeeding, Moulton, Trowbridge and Haigh ('22a) 
found the spleen weight in general nearly proportional to the body weight in all 

In pigeons subjected to acute inanition (water only) with loss of 38.7 per 
cent in body weight, Findlay ('21) found an average loss of 60 per cent in the 
spleen weight (Table 13). In chronic underfeeding, the spleen lost 71 per cent. 

Inlow ('22) in 2 fasting dogs ("fed a half-day's ration every third day") 
with losses in body weight of 31.9 and 46.2 per cent, respectively, noted a corre- 
sponding (estimated) shrinkage of 60 and 86 per cent in the spleen. The initial 
size of the spleen was measured directly by means of a laparotomy at the 
beginning of the experiment and compared with the findings at necropsy. 
Inlow concluded that similar losses in spleen and body weight following ligation 
of the pancreatic ducts are likewise due to the resultant inanition. 

In frogs (species?) which had lost two-thirds of their body weight by starva- 
tion, Blumenthal ('04) found the spleen reduced to a diameter of 1 millimeter. 
Gerhartz ('06) found in female Rana esculenta a spleen weight of 0.028 g. in a 
normal frog with body weight of 47.5 g. In one starved 3 months in a warm 
room, with body weight of 35 g., the spleen weighed only 0.001 g. ; and in 
another starved 4^ months, body weight 40 g., the spleen was extremely 
small (pin-head size). Ott ('24) in an extensive study on Rana pipiens dur- 
ing torpidity and at various stages of inanition, up to a loss of 60 per cent in 
body weight, found the average spleen weight very irregular in the various 
groups and in the sexes (Table 6), so that he was unable to draw any general 

A. L. Gaule ('93) and J. Gaule ('01) noted seasonal changes in the weight of 
the frog's spleen. Valentin ('57) found only a slight apparent loss (about 10 
per cent) in the spleen of the hibernating marmot. Pappenheim ('01) even 
claimed that in spermophiles the spleen during hibernation becomes greatly 
enlarged. Thus during the seasonal changes, including hibernation, the 
weight of the spleen appears very irregular. 


Changes in Structure.- — The effects of total inanition (or on water only) upon 
the structure of the spleen will be reviewed first in man, adult and infant; later 
in the lower animals. 

In the victims of the Madras famine, Porter ('89) noted in the usually 
atrophic spleen a variable thickening of the capsule and usually a deficiency in 
the pulp, with pigmentation in a few cases. In a man of 30 years who died from 
voluntary starvation, Stschastny ('98) observed in the spleen a pronounced 
atrophy, with scarcely recognizable Malpighian bodies, hyalinization of vessels 
and trabeculae, numerous eosinophile cells, hemoblasts in mitosis, and con- 
siderable pigmentation with hemosiderin. In a similar case, Meyer ('17) found 
the Malpighian nodules nearly obliterated, only a few barely distinguishable. 
The splenic sinuses were not evident, but erythrocytes (no nucleated forms) 
were abundant in the parenchyma. Only a few giant cells and phagocytes 
were found, although pigment was relatively abundant. The staining reactions 
of the cells were normal, though somewhat faint. 

In extensive material from cases of starvation (age 1-63 years) Stefko 
('23) finds hyperemia of the spleen, with atrophy of the lymphoid follicles, but 
increase in the reticulum and trabeculae. 

As to the changes in the spleen of malnourished infants, Moldenhauer ('99) 
noted a relative increase in the stroma. Lucien ('08) found the spleen in 
athreptic infants firm and sclerotic. The Malpighian nodules were apparent, 
though small. In general, the histological changes were less marked than in 
the 'thymus and other organs. Helmholz ('09) in atrophic infants noted 
extensive pigmentation in the spleen, though no relation was found between 
the intensity of the pigmentation and the severity of the disease. The (iron- 
containing) pigment usually occurred in large intercellular masses as well 
as in fine intracellular granules. The Malpighian nodules were usually free 
from pigment. In general, the spleen presented relatively few erythrocytes, 
and relatively conspicuous fibrous tissue. Schelble ('10), on the other hand, 
found no significant histological changes in the spleen of atrophic infants. 

In cases of congenital spasmodic atrophy of infants, Lesage and Cleret 
('14) described perivascular sclerosis, with secondary fibrous trabeculae per- 
vading the parenchyma. Mattei ('14) during athrepsia found the splenic 
capsule and trabeculae somewhat thickened. The white pulp of the spleen 
appears more abundant at first; later the red pulp gradually replaces the white, 
which persists only in the Malpighian nodules. Siderosis was found constant, 
with active red cell destruction in the macrophages of Billroth's cords. Nobe- 
court ('16) reviewed the literature, emphasizing the splenic siderosis, which 
accompanies the hepatic siderosis, as found by Triboulet, Ribadeau -Dumas 
and Harvier ('10), Helmholz and others. "La pulpe rouge de la rate est 
incrustee de blocs pigmentaires plus ou moins volumineux. Les sinus et les 
cordons de Billroth sont la siege d'une macrophagie active avec nombreuses 
figures de destruction globulaire, de sorte qu'il parait evident qui tout le fer mis 
en liberte provient de la destruction des globules rouges." The large vessels 
appear congested, but the venous sinuses appear nearly normal. Marfan ('21) 
stated that the splenic and hepatic siderosis found in athreptic infants is a 


macrophagic reaction to the increased erythrocytic destruction in the anemia 
associated with the condition. 

In famine-stricken children of various ages, Nicolaeff ('23) described the 
spleen as firm in consistence, with atrophy of the pulp and increase of fibrous 
trabeculae. Some edematous cases showed hyperemia or hemorrhagic infiltra- 
tion, with atrophy of the lymphoid tissue, early hyalin degeneration and 
increased amount of brown pigment. Splenic and hepatic hemosiderosis was 
likewise observed by Stephani ('23). 

In starved animals, Tiedemann ('36) found the blood vessels of the viscera 
in general, including the spleen, contracted and empty, hence giving the organs 
a pale or greyish tint. Falck ('75) in starved dogs found the spleen "braun- 
roth, glanzend, schlaff , fein gerunzelt und blutleer." 

Morpurgo ('88, '89), in rabbits starved 5-13 days, described the spleen as 
atrophic and anemic, with relative enlargement of the trabeculae and reduction 
of the lymphoid tissue in the Malpighian nodules and pulp cords. Mitoses 
occur in the normal adult rabbits, but become rare during inanition. Giant cells 
were observed in the spleen of a starved rabbit 15 days old. In rabbits refed 
5 days after starvation for 10 days, Morpurgo ('90) found the spleen somewhat 
increased in size but still subnormal in weight. The pulp appeared relatively 
increased (compared with the starved condition), and the venous spaces and 
trabeculae correspondingly decreased. The histological structure resembled 
the normal. Mitoses were abundant in the pulp and especially in the center of 
the lymphoid nodules. 

Coen ('90) noted the changes during starvation (with or without water) 
in 3 rabbits and 1 kitten. The spleen shows marked atrophy of the splenic 
pulp, rendering the connective tissue trabeculae very prominent. The Malpi- 
ghian nodules are poor in lymphoid cells. The veins appear distended. Some 
small hemorrhagic foci occur, with both extracellular and intracellular hematic 
pigment, occurring chiefly just beneath the capsule, more sparsely in the 
splenic pulp. Blumenthal ('04) found the histological structure fairly well 
preserved in the spleen of starved frogs. Opie ('04) observed a marked decrease 
in the number of eosinophiles in the spleen of fasting guinea pigs, as stated in 
Chapter XV. Cesaris-Demel ('06) noted pigmented granules in the cells and 
connective tissue of the spleen in rabbits with marasmus produced by bacterial 

In 4 starved cats, Beeli ('08) found that the atrophy of the spleen is caused 
by decrease in the pulp. The trabeculae remain unchanged in size, and there- 
fore become relatively very prominent and closely arranged, though inconspicu- 
ous in the normal spleen. The Malpighian nodules are also reduced in size 
and closer together. In some places, the blood capillaries (sinuses) are dilated. 
The nuclear diameters of the lymphoid cells in the Malpighian nodules were 
measured and tabulated according to size and frequency distribution, showing 
a tendency to progressive decrease in nuclear size during inanition. 

In birds (pigeon, fowl, duck), guinea pigs and rats fasting for various 
periods, Jolly and Levin ('12a) noted atrophy of the lymphoid tissue, especially 
in the Malpighian nodules, the pulp otherwise being unaffected. Refeeding 



Fig. 77. — Photograph of a portion of a section of the spleen from a normal albino rat 
(Mo. 9), female, age 4 months. M, a splenic nodule (Malpighian corpuscle) with central 
arteriole, cut longitudinally. P, red pulp, with intermingled lymphoid pulp-cords. C, 
capsule. Trabeculae are scanty. Zenker fixation; hematoxylin-eosin stain. X90. 

Fig. 78. — Photograph of a portion of a section of the spleen of a male albino rat (S. 18), 
body weight reduced from 196 to 121 g. (loss of 38 per cent) in 8 days on water only. 
Spleen greatly atrophied (weight 0.115 £•)• Splenic pulp (P) reduced in amount. M, 
rudimentary splenic nodule (Malpighian corpuscle). Capsule (C) and trabeculae (T) appear 
relatively hypertrophied. There is a somewhat diffuse pigmentation, which is not evident in 
the photograph. Zenker fixation; hematoxylin-eosin stain. X90. 


restores the normal structure in about 15 days, probably by mitoses of the 
remaining lymphocytes. Jolly ('14) concluded that in fasting puppies and rab- 
bits the effects in the spleen are greater than those in the lymphoid organs gener- 
ally, excepting the thymus (and the bursa of Fabricius in birds). In guinea 
pigs subjected to acute (complete) or chronic (incomplete) total inanition, 
Rondoni and Montagnani ('15) found the spleen greatly reduced in size by 
atrophy of the lymphoid tissue in the splenic pulp and Malpighian nodules. 
The trabeculae and supporting tissue become more prominent, being relatively, 
if not absolutely, increased in amount. 

In 3 albino rats starved on water only, Sundwall ('17) found in the spleen: 
" prominent capsule ; trabeculae appear very much thickened ; extreme congestion ; 
reduction in areas to complete disappearance of Malpighian follicles as a result 
of the congestion, and hyperplasia of endothelial cells; the latter are filled with 
red blood cells and pigment hemoglobin and hemosiderin." Asada ('19) found 
congestion of the blood vessels and striking pigmentation of the spleen in rabbits 
after 10 days of total inanition. 

Inlow ('22) described the capsule of the atrophic spleen in starved dogs as 
shrunken and thinned, with compact structure, hyalin appearance and few 
nuclei. The trabeculae become more conspicuous through comparatively 
greater atrophy of the parenchyma. The cytoplasm of the pulp cells and sup- 
porting tissue has largely disappeared, so the pulp nuclei are closely packed. The 
number of lymphocytes outside the Malpighian nodules is markedly decreased; 
cells of large size predominate. The Malpighian nodules and their germinal 
centers appear more definitely delimited. The larger blood vessels appear 
normal; the splenic sinuses are smaller, but distinct. The free blood pigment 
in the pulp is increased in amount. The atrophic changes in the spleen appear 
roughly proportional to the length of the inanition and the loss in body weight. 

In the fasting rabbit, Okuneff ('23) described a diminution in the number of 
the lymphoid splenic cells; but they remain unchanged in size and mitochondrial 

Figures 77 and 78 represent the histological changes in the spleen of the 
albino rat during acute inanition. 

Papers dealing with chemical changes (especially water-content) of the 
spleen during general inanition include those of Lukianow ('88) in pigeons, 
Tonninga ('93) in rats and rabbits, and Roger ('07) in rabbits. Those concern- 
ing changes in fats and lipoid content include Terroine ('20) and Okuneff ('22). 

During hibernation, Aeby ('75) studied the chemical changes in the spleen 
of the marnot. Mann ('16) noted marked congestion of the spleen in the 
hibernating gopher, Spermophilus tridecemlineatus. According to Mann and 
Drips ('17), this congestion reaches a maximum in a few days, persists about 40 
days, then gradually subsides. 

(B) Effects of Partial Inanition 

The effects of partial inanition upon the spleen will be considered under defi- 
ciencies of protein (including pellagra and famine edema), of salts (including 
rickets), of vitamins (including beriberi and scurvy), and of water (thirst). 


Effects of Protein Deficiency. — The reasons for including pellagra and 
malnutritional edema under protein deficiency were given in Chapter V. 

Pellagra. — According to Fraenkel ('69-' 70) the spleen appeared subnormal 
in weight in 23 out of 30 cases. Lombroso ('92) found it atrophic in 41 and 
hypertrophied in 12. Babes and Sion ('00) likewise found the spleen usually 
atrophic, except in malarial cases. Nicholls ('12, '13) in 8 negro pellagrins 
noted an average weight of only 3 ounces for the spleen. Raubitschek ('15) 
and Harris ('19), from a comprehensive review of the literature on the pathology 
of pellagra, concluded that the spleen is usually decreased in size, though some- 
times enlarged. 

As to the structural changes in the spleen during pellagra, Fraenkel ('69-'7o) 
noted hyperemia, more rarely pigmentation. Tuczek ('93) found atrophic 
changes and abnormal pigmentation, which are also included in the review by 
Marie ('08, '10). Kozowsky ('12) described thickening of the trabeculae, 
venous congestion and hyalinization of the small arteries. An increase in the 
fibrous tissue (sclerosis), especially of the blood vessels, and marked hematoge- 
nous pigmentation of the spleen are mentioned by Nicholls ('12/13), Raubit- 
schek ('15) and Harris ('19). 

The weight of the spleen in human famine edema appears somewhat vari- 
able. Paltauf ('17) observed a range of 70-235 g. Atrophy of the spleen 
was reported by Hiilse ('17) and Oberndorfer ('18), the latter finding weights 
as low as 50 g. Enright ('20) found the spleen enlarged only in cases com- 
plicated by malaria; otherwise small, sometimes weighing only 3 ounces. Mann, 
Helm and Brown ('20) reported the spleen normal or shrunken in 200 necropsies. 
Prince ('21) found the spleen usually slightly increased in volume. 

As to the structure of the spleen in famine edema, Oberndorfer ('18) de- 
scribed it as sclerotic; the pulp anemic and atrophic; color dark reddish brown; 
follicles small but distinct. Enright found the consistency of the spleen normal 
except in the malaria cases, when it appeared hard and pigmented. Fracassi 
('22) noted splenic fibrosis. 

In guinea pigs on maize diet (mixed deficiency of protein, vitamins and salts), 
Rondoni and Montagnani ('15) obtained a sclerosis and sometimes hyperemia 
of the spleen. ■ Rondoni ('19) found that the spleen formed 0.14 per cent of 
the body in maize-fed guinea pigs (normal 0.12 per cent), indicating a slight 
increase in relative weight. 

In monkeys, pigs and albino rats on corn-oil cake, maize and similar diets 
(mixed deficiencies of protein, etc.) producing emaciation, Sundwall ('17) found 
the spleen intensely congested, showing hemosiderosis, and sometimes amyloido- 
sis and hyalin changes; proliferation of endothelial cells of the pulp, containing 
red blood cells and pigment; and reduction in size of the Malpighian bodies. 

In pigeons and monkeys on diets with mixed deficiency of protein, vitamins, 
etc., McCarrison ('19b, '21) found marked atrophy of the spleen, especially in 
the pigeons. 

Lefholz ('23) found that in kittens the spleen shows no consistent response 
to variations in the protein, fat or sugar content of the diet (thus, like the thymus 


and lymph glands, differing from the tonsils and other lymphoid structures of 
the alimentary canal). 

In human rickets, enlargement of the spleen was noted by Whistler (1645), 
Seibold (1827) and many subsequent observers; but its significance has been 
much disputed {cj. reviews by von Starck '96 and Sasuchin '00), being frequently 
attributed to syphilis or other complications. Von Starck ('96) found at Kiel 
a splenic enlargement (" Milztumor ") in 53 out of 113 autopsies of rachitic 
children (57 per cent), but it also occurred in 77 out of 148 non-rachitic (52 per 
cent). A palpable "Milztumor" was also found in 68 out of no living rachitic 
children, which is said to agree approximately with the findings by Rehn at 
Frankfort (64.8 per cent), and by Kuttner at Berlin (73.3 per cent). Never- 
theless, von Starck concluded that the splenic enlargement is merely a frequent, 
complication, and not due to the same cause as the rickets. Sasuchin ('00) 
found splenic enlargement in 12 out of 16 cases, the exceptions showing general 
emaciation. Comby ('01) found the spleen usually enlarged in rickets. Vincent 
('04) claimed that it is moderately enlarged in the progressive stage, but not 
permanently. Cheadle and Poynton ('07) stated that the rachitic spleen is 
sometimes enlarged; while Stoltzner ('03, '09) found no significant change (aside 
from complications). Pfaundler ('22) states that in rickets the spleen is often 

Wohlauer ('n), from original observations and a review of the literature on 
human rickets, concluded: 

" Ein besonderes Verhalten zeigt die Milz. Sie ist stets vergrossert manch- 
mal in excessivem Grade, bis zum Sechs und Achtfachen ihres normalen Masses 
(Heubner). In leichten Fallen lasst sie sich am Lebenden nicht nachweisen, 
da auch die aufgetriebenen Darme die Untersuchung erschweren. Wir fiirchten 
uns zu wiederholen, es muss aber gesagt werden, dass auch hier wieder die 
Meinungen geteilt sind (v. Starck, Miiller, Kuttner, Fox, Ball, M. Cohn, 
Stoltzner, Monti). Mit dem Ablauf der floriden Erscheinungen verschwindet 
der Milztumor, der sich meistenteils bei Kindern im zweiten Lebensjahr findet." 

In rats with experimental rickets, Shipley, Park, McCollum and Simmonds 
('21) found the spleen apparently atrophic in some rats, more or less enlarged in 
others. McCollum, Simmonds, Shipley and Park ('21) noted that the spleen is 
frequently greatly enlarged; while McCollum, Simmonds, Kinney, Shipley and 
Park ('22) state that the spleen is enlarged in some cases, but usually not above 
normal. These conclusions were apparently not based on actual weights. 
Jackson and Carleton ('23) found the weight of the spleen so irregular and 
variable in rachitic rats that no conclusion could be drawn, although in most 
cases the average was below that in the controls (Table n). 

But few observations upon the histological changes in the spleen during 
rickets are available, and these are all in the human species. Sasuchin ('00) 
concluded that the changes are very characteristic, appearing early and in many 
cases proportional to the intensity of the rachitic symptoms. The most marked 
changes are proliferation of the connective tissue, constriction of the arterial 
lumina, inflammatory appearances and atrophy of the Malpighian bodies. The 
lesions resemble those in lues and tuberculosis, but these diseases were carefully 


excluded. These changes may affect the hemopoietic function of the spleen 
and be related to the anemia occurring in rickets, as was also emphasized by 
Saricinelli ('03). He found at first a hyperplasia of the splenic pulp followed 
by a progressive proliferation of the connective tissue stroma, which gradually 
replaces the parenchyma. Pfaundler ('22) holds that during rickets there is a 
marked swelling of the lymphoid organs throughout the body. The spleen 
is often enlarged, due to hyperplasia of pulp and follicles. 

Vitamin Deficiency.- — In some cases previously cited, such as infantile 
atrophy, the effects are probably due in part to vitamin deficiencies. We may 
consider these under the three well known vitamins — A, B and C. 

Deficiency of vitamin A alone has apparently no marked effect upon the 
spleen, according to experiments on rats by Emmett and Allen ('20) and Cramer, 
Drew and Mottram ('21a). Davis and Outhouse ('21) noted frequent conges- 
tion of the splenic sinuses, but no degenerative changes. Meyerstein ('22) 
made a few observations on the spleen in young white rats on diets deficient in 
vitamins A and B. 

Vitamin B. — In 125 cases of human beriberi, Ellis ('98) found the average 
weight of the spleen 9.27 ounces, in comparison with 6.28 ounces in 204 dying 
from other causes, thus indicating a hypertrophy of the spleen in beriberi. 
Strong and Crowell ('12) noted some congestion of the spleen, with relatively 
small follicles. Andrews ('12), in 18 necropsies of infantile beriberi, likewise 
found the spleen very hyperemic, and even hemorrhagic; but with no increase 
of splenic tissue. Nagayo ('23) claimed that splenic passive congestion dis- 
tinguishes human beriberi from experimental polyneuritis. 

In experimental beriberi (or polyneuritis) produced in pigeons by polished 
rice diet (deficient in vitamin B), Funk and Douglas ('14) mentioned atrophy 
and degenerative changes in the spleen, among other organs. . Tasawa ('15) 
studied the effects in about 150 pigeons and 200 fowls, finding the body emaci- 
ated and the spleen always atrophic; capsule wrinkled and thickened, and 
trabecular conspicuous; the pulp atrophic and the follicles scarcely visible. 
"Ein Bild wie die gewohnliche Stauungsmilz ist iiberhaupt nicht zu sehen." 
Emaciation of the body and atrophy of the spleen were likewise noted by Mack- 
enzie (' 1 5) in pigeons and by Drummond (' 1 6) in young chickens. 

Voegtlin and Lake ('19), in cats, dogs and rats with polyneuritis produced by 
deficient diets, noted in the spleen degenerative changes similar to those (above 
mentioned) found by Sundwall ('17), but less extensive. 

As previously mentioned, McCarrison ('19, '19a, '21) found that a diet of 
milled and autoclaved rice gives rise to an atrophic degeneration of the spleen 
and other organs in pigeons and monkeys. Atrophy of the lymphoid tissues 
in general was observed. Brucco ('20) noted evidences of regenerative activity 
in the spleen and bone marrow of dogs on a polished rice diet. Findlay ('21) 
found a loss of 65-67 per cent in the spleen of fowls and pigeons with beriberi 
(Table 13). Lopez-Lomba ('23) noted a brief increase preceding the marked 
decrease in the weight of the spleen in adult pigeons on vitamin-free diet. 

Cramer, Drew and Mottram ('21) found in mice and rats that the spleen, 
like the lymphoid tissue in general, undergoes a specific and profound atrophy 


upon dietary deficiency of vitamins (especially vitamin B), the result being 
similar to that produced by the Roentgen rays and radium. 

In chickens with polyneuritis from dietary deficiency of vitamin B, Souba 
('23) found the spleen markedly subnormal in weight, the loss being relatively 
exceeded only by the testes. In experimental avian beriberi, Korenchevsky 
('23a) also found splenic atrophy, but persistence and even hypertrophy and 
hyperplasia in the germ centers of the nodules. 

Vitamin C. — In human scurvy, Sato and Nambu ('08) found the spleen 
enlarged in tuberculous cases; otherwise normal in size. Much granular pigment 
was noted, and subcapsular hemorrhage in 1 of 13 autopsies. Aschoff and Koch 
('19) found the spleen generally normal in size, not enlarged in uncomplicated 
cases. Histologically they found manifold changes (especially pigmentary), 
generally secondary in character. Congestion was observed, but no hemorrhages. 

Jackson and Moore ('16) found the spleen frequently enlarged in scorbutic 
guinea pigs. Hess ('20) has reviewed in detail the literature on human and 
animal scurvy, indicating that the spleen is usually enlarged and congested, 
showing pigmentation, and sometimes hemorrhage and hyperplasia. Hojer ('24) 
describes lymphoid atrophy, siderosis, occasional hemorrhages and necrosis. 

Bessesen ('23) found the weight of the spleen variable at different stages of 
scurvy in the guinea pig. In the early stages, the spleen showed a loss in weight ; 
but after the appearance of definite scorbutic symptoms, it appeared hyper- 
trophic, being 35-56 per cent above normal weight (Table 12). 

Water Deficiency. — In a dog 76 days old, on dry diet with loss of 21 per cent in 
body weight, Falck and Scheffer ('54) noted an apparent gain of 9 per cent in the 
spleen (probably an individual variation), in comparison with a litter-mate con- 
trol. Bowin ('80) found that in dogs and rabbits on dry diet, with loss of about 
50 per cent in body weight, the losses in the various organs were similar to those 
in total inanition. The spleen lost relatively more than the body as a whole. 

Pernice and Scagliosi ('95a) studied the effects of a dry diet upon the spleen 
in 1 dog and 3 young chickens. In the dog, the spleen at autopsy appeared 
small and dry, with wrinkled, thickened capsule and hypertrophied trabeculae. 
The pulp was atrophic, dark brown in color, with sharply demarcated follicles. 
Many cells were in mitosis. Small, subcapsular hemorrhages occurred. In the 
chickens, the spleen was likewise atrophic. The thickened capsule in many 
places showed round cell infiltration. In the pulp just beneath the capsule 
were numerous hemorrhages of various size, around the Malpighian bodies, 
which appeared very atrophic. The arteries showed inflammatory changes. 
The pulp was scanty, with many pulp cells in mitosis. 

In adult albino rats on a relatively dry diet, Kudo ('21) found in acute thirst 
experiments, with average loss of 36 per cent in body weight, a loss of 66 per 
cent in the spleen; in chronic thirst experiments, with loss of 52 per cent in body 
weight, a loss of 73 per cent in the spleen; and in one rat with neither food nor 
water, with loss of 47 per cent in body weight, a loss of 63 per cent in the spleen 
(Table 9). In young albino rats (1 month old) held at constant body weight 
by relatively dry diets for various periods the spleen shows a marked loss 
(36-48 per cent) in all but one group, in which one exceptionally large spleen 
reduced the average loss (Kudo '21a) (Table 10). 



The thymus is an organ characteristic of the growth period, and is of especial 
interest in pediatrics on account of its pronounced tendency to atrophy in all 
conditions of inanition or malnutrition. Following a brief summary, the effects 
of inanition upon the thymus will be considered in detail under (A) total inani- 
tion, and (B) partial inanition. 

Summary of the Effects on the Thymus 

The weight of the thymus responds so promptly and extensively to conditions 
of malnutrition as to justify fully Simon's designation of it as a "barometer of 
nutrition." In this respect, it is rivalled only by the adipose tissue. The 
loss is relatively far greater than that in the whole body. In both man and 
lower animals the loss in thymus weight usually reaches at least 75 per cent 
before death from acute inanition (total complete, or on water only) ; and in 
chronic forms (incomplete total) may reach 90 to nearly ico per cent, though it 
is doubtful whether it ever completely disappears as sometimes stated. 

The thymic atrophy during inanition occurs at every age, but is especially 
notable is the young, where the thymus is both relatively and absolutely larger 
than at later ages. The persistent growth tendency found during inanition in 
many of the organs during infancy does not appear in the thymus, although its 
loss in the newborn rat appears relatively less than at later ages. 

The characteristic thymic atrophy appears not only in the various degrees of 
total inanition, but also in most forms of partial inanition, as well as in many 
other conditions (exhaustion, disease, etc.), and is designated by Hammar as 
"accidental involution," in contrast with the normal "age involution" of the 

In most cases the degree of thymic atrophy appears closely correlated with 
the general malnutrition of the body; although certain exceptions occur, in 
which the thymus may be large in spite of general emaciation. This is found in 
some cases of infantile atrophy, and has been claimed in human rickets, but is of 
doubtful significance. 

Recovery of the normal thymic weight usually occurs promptly upon refeed- 
ing, unless the inanition has been extremely severe. There is some evidence 
indicating persistent subnormality in the thymus of rats permanently stunted 
by underfeeding, but this is still somewhat uncertain. 

The histological changes in the thymus during inanition are equally striking 
and characteristic. The resultant involution affects all parts of the thymus ; but 
especially the lymphoid tissue, which here (as elsewhere throughout the body) 
undergoes a pronounced atrophy. Although there are numerous variations in 



the details, the typical process appears very similar in all types of inanition, 
including hibernation and various diseases involving malnutrition. There is a 
marked decrease in the number of mitoses, but not complete cessation, except 
perhaps in extreme stages. The number of lymphocytes becomes very greatly 
reduced, chiefly through emigration into the adjacent tissues and vessels, but 
partly (especially in later stages) by degeneration of lymphocytes, with phagocy- 
tosis by the reticulum cells. The reticulum is much more resistant, and assumes 
a somewhat epithelioid, embryonal appearance, recalling the "reduction" 
phenomena among certain invertebrates. The reticulum presents a progressive 
increase in lipoidal content, with degenerative changes in the later stages. 

The atrophy first and foremost affects the cortex, which becomes rarefied 
by the " delymphoidisation," and lighter than the medulla in stained sections, 
producing the so-called thymic "inversion." Later the differentiation between 
cortex and medulla disappears. The Hassall's corpuscles are relatively resis- 
tant, but also ultimately undergo atrophy and cystic degeneration, and in 
extreme stages may disappear entirely. 

In contrast with the striking atrophy of the parenchyma, the fibrous stroma 
is relatively more resistant, and becomes more prominent, giving an appearance 
of fibrosis or sclerosis in the capsule and interlobular septa. There is usually a 
prompt disappearance of the ordinary adipose content, thus differing from the 
typical age involution of the thymus. Although relatively more resistant, the 
fibrous stroma also decreases progressively in absolute volume, so that in extreme 
stages the thymus is reduced to a small mass of vascular fibrous tissue, contain- 
ing only indistinct remnants of the parenchyma. 

Unless the inanition has been extremely severe, there is prompt regeneration 
of the thymus upon ample refeeding. The remaining cells undergo rapid mito- 
sis; the reticulum cells regenerate new Hassall's corpuscles in the medulla, and 
become infiltrated with lymphocytes in the cortical region (as during normal 
development). Thus the normal structure is ultimately restored, excepting 
extreme cases where permanent injury may result. 

(A) Effects of Total Inanition, or on Water Only 

The literature upon the "accidental involution" of the thymus has been 
thoroughly reviewed by Hammar ('o6, '10, '21) and is also included by Biedl 
('16, '22). For convenience, the effects of total inanition will be discussed 
first in the human species; later in the lower forms. 

The earliest observation upon human thymic atrophy was apparently by 
Verheyen (17 10) who noted its occurrence in persons subjected to strenuous 
activity ("qui corpus vehementer exercent"). As in the case of most organs, 
however, the earlier observations upon the human thymus during inanition con- 
cerned chiefly the weight or size, together with the gross appearance, in various 
cachectic conditions. Thus Meckel (1810, 1820) observed that the general 
nutritive condition of the body greatly affects the rapidity with which the 
thymus atrophies. In a weak, malnourished child of 2 years, he found the 
thymus "fast ganz geschwunden, saftlos, weich, viel kleiner als bei einem 


wohlgeniihrten 6 jahrigen." Similar observations were made by Haugsted 


Simon ('45) stated: "I think it extremely probable that the thymus may 
within a few days, if not hours, vary remarkably in the same individual, accord- 
ing to the immediate state of the general nutrition. Its size seems to be, caeteris 
paribus, if I may venture to use the phrase, a barometer of nutrition and a very 
delicate one." 

Herard ('47) found great variation in the weight of the thymus, even aside 
from the nature of the disease causing death. "La constitution de l'enfant, 
son etat de maigreur ou d'embonpoint, semblent etre les principales conditions 
qui influencent ces variations." Ecker ('53) likewise noted a decreased size in 
the thymus during malnutrition. 

Friedleben ('58) made extensive and careful observations on the weight of 
the thymus in the well-nourished and malnourished, and concluded: "Das 
relative Gewicht der Thymus stellt sich im Sauglingsalter bei akuten Krank- 
heiten um viermal, in chronischen um zwolfmal niederer als der normalen 
Thymus; in der ersten Kindheit bei akuten Prozessen funfmal, bei chronischen 
sechsmal; in dem Knabenalter bei akuten Krankheiten viermal, bei chronischen 
neunmal niederer als in gesunden Individuen." Friedleben's animal experiments 
will be mentioned later. 

Thaon ('72) similarly found that at death from traumatism or acute diseases 
the thymus is always large; while in chronic disorders with malnutrition the 
thymus is largely consumed. 

Seydel ('94) emphasized the medico-legal importance of the thymus weight, 
claiming that thymus atrophy, accompanied by extreme emaciation and with- 
out signs of organic disease, is a sure sign of death from inanition. Hansen 
('94) weighed the thymus in 108 cases, with results similar to those of Thaon 
('72). He opposed Seydel's claim that the thymus may totally disappear in 
conditions of extreme exhaustion. Filomusi-Guelfi ('95) questioned the im- 
portance of thymic atrophy as a sign of death from starvation, since it occurs in 
all chronic diseases or conditions involving malnutrition. 

Farret ('96) confirmed Thaon's finding of thymus atrophy in athreptic 
infants, with 27 tabulated cases. He concluded that the thymus is functionally 
related to the nutrition and development of the organism. He cited the case of 
Durante, who found in a cachectic infant an atrophic thymus (1.5 g.) showing 
an intense sclerosis which he thought might have caused the fatal condition. 

The atrophy of the thymus during inanition, especially in its medico-legal 
aspect, is discussed by Dwornitschenko ('97), von Mettenheimer ('98) and Diin- 
schmann ('00). Von Mettenheimer (like Simon) considered the weight of the 
thymus as the best index of nutrition of the body, its atrophy being comparable 
to that of the adipose tissue. He described the occurrence of fibrosis and degen- 
eration of Hassall's corpuscles, and held that thymic atrophy is the cause of 
pedatrophy. Ghika ('01), however, maintained that the thymic atrophy is the 
effect, rather than the cause, of athrepsia. 

Stokes, Rurah and Rohrer ('02) and Rurah ('03), like previous observers, 
found a marked and constant atrophy of the thymus (average weight 2.2 g.) in a 


series of 18 malnourished infants. They also studied the histological changes in 
the thymus, finding a general fibrosis, with thickening of the capsule and inter- 
lobular connective tissue. Hassall's corpuscles increase in size and undergo 
hyalin degeneration. A later stage shows "almost entire disappearance of the 
lymphoid structure of the lobule and an increase in the reticular endothelium." 
Dudgeon ('05) studied the weight and structure of the thymus in 15 cases of 
primary infantile atrophy and in 41 cases of secondary atrophy. He concluded 
that the atrophy of the thymus is closely associated with the general wasting of 
tissues. Histologically, the most characteristic changes in the atrophic thymus 
include: fibrosis, usually well marked, with thickening of the outer coat of the 
blood vessels; atrophy of the lymphoid corpuscles, which are replaced by endo- 
thelial cells, connective tissue cells and small "giant cells;" all varieties of degen- 
eration in Hassall's corpuscles. The average weights found by Dudgeon are as 
shown in the accompanying table. 

Weight of the Thymus in Infants with Various Conditions of Nutrition (Dudgeon'os) 


No. of cases 

Average wt. of 
thymus, giams 










6.53 (or 4.575) 



Primary atrophy 

Secondary atrophy (tuberculous) .... 
Secondary atrophy (non-tuberculous) 

Acute diseases 

Fetal specimens 

"Sudden death." "Found dead". . . 

Somewhat similar thymus weights were recorded by Fortescue-Brickdale 
('05), who found the average in 12 marantic infants to be 2.45 (1.3-4.8) g.; in 9 
tuberculous children, 1.9 to 10 g. ; in 9 chronic emaciated (non-tuberculous), 
3.14 (1-4.7) g-5 m 2 ° acute diseases, 5.7 g. 

Hammar ('05, '06) made an extensive and thorough study of the involution 
of the thymus, a continuation of his earlier animal experiments (to be mentioned 
later). Hammar introduced the term "accidental involution" to indicate the 
atrophy caused by malnutrition or other abnormal factors, in contrast with the 
normal "age involution" of the thymus. The establishment of a reliable norm 
of growth in weight made it possible to measure precisely the degree of atrophy at 
any stage of postnatal development. The amounts of parenchyma and stroma 
in both cortex and medulla were also measured, making possible a quantitative 
measurement of the histological changes. 

In general, Hammar found that the process of atrophy in the "accidental" 
involution is characterized by a subnormal amount of parenchyma, which, in 
the cortex, may disappear entirely. The stroma is also reduced, but to a lesser 
degree; it therefore becomes relatively more abundant, but there is no true 
fibrosis or sclerosis. There may be a "paradoxical" adipose deposit in the 
stroma of the athreptic thymus, even in infants where normally no fat occurs. 
In the parenchyma, the lymphocytes becomes variably decreased in number; 


sometimes nearly all disappear. " Die Retikulumzellen legen sich dabei dichter 
an und bilden haufig sogar Komplexe, die ihr urspriingliches epitheliales Aus- 
sehen mehr oder weniger wiedergewonnen haben. Auch sie fallen einer Degen- 
eration allmahlich anheim, welche allerdings von dem normalen durch Hyper- 
trophic der Zellen eingeleiteten und zur Bildung Hassal'scher Korperchen 
fiihrenden Prozess deutlich zu trennen ist." Hassall's corpuscles persist in the 
earlier stages of inanition, becoming relatively prominent; later they are vari- 
able, sometimes disappearing. 

Bovaird and Nicoll ('06) studied the weight and gross structure of the 
thymus in 571 autopsies on children up to 5 years of age. In general, the loss 
in thymus weight (80 or 90 per cent, in extreme cases) is relatively greater 
than in body weight; but there are exceptional cases indicating that factors 
other than the extent of malnutrition may influence the weight of the organ. 
The greatest loss of thymus weight occurs in marasmus, diarrheas, and similar 
chronic exhausting disorders. The atrophic involution includes actual absorp- 
tion of thymic tissue, which becomes fibrous and fatty, as during the normal age 

Thompson ('07) in 20 cases of infantile marasmus found the average weight 
of the thymus 2.472 g. (range 1.0-4.75). There is marked histological 
atrophy of the lymphoid tissue, and increase of interlobular connective tissue, as 
described by Dudgeon. Hassall's corpuscles become prominent and the differ- 
entiation between cortex and medulla usually disappears. 

Naegeli ('08) held that the disappearance of the medullary cells of the 
thymus during inanition indicates that these are large lymphocytes, rather than 
epithelial in character. 

Lucien ('08, '08c), in fatal cases of athrepsia, found an average thymus 
weight of only 0.97 g. (range 0.05-2.50), indicating an average loss of over 80 
per cent. Grossly the thymus appears reddish, with a firm, fibrous consistence. 
The histological changes are variable, with 4 progressive stages: (1) Early 
involution, with retention of the distinction between cortex and medulla; little 
change in structure, aside from increased perilobular connective tissue and dila- 
tion of blood vessels. (2) The cortico-medullary distinction disappears; 
lymphocytes equally distributed throughout, but the medulla is more vascular 
and contains more Hassall's corpuscles. (3) The peripheral (cortical) zone 
becomes clear, filled with irregular "epithelioid" cells, and the lymphocytes 
become more concentrated in the central (medullary) zone, making an inver- 
sion of the lobular structure; Hassall's corpuscles numerous throughout almost 
the entire lobule, appearing simple or compound in structure, and in various 
stages of evolution or degeneration, sometimes hyalin, cystic or calcified. (4) 
The most extreme stages show the thymus converted into a fibrous tract, with 
numerous thick- walled vessels ; vestiges of the thymic lobules appear as scattered 
small lymphoid nodules; sometimes a few degenerating Hassall's corpuscles are 
still visible. 

Feldzer ('10) and Tixier and Feldzer ('10) concluded that the thymic 
atrophy in athrepsia is probably secondary, rather than primary in character. 
The regressive process is a sclerosis, distinct from the normal adipose age 


involution, and presents 5 stages: (1) Moderate fibrous proliferation, especially 
interlobular; blood vessels strongly congested; differentiation of darker cortex 
and lighter medulla not well marked; cells show mitoses; Hassall's corpuscles 
polymorphic, but not abnormal. (2) The congestion and cellular hyperplasia 
become decreased; interlobular connective tissue increased; lobular reticulum 
more apparent, with sclerosis extending from vessels and septa; distinction 
between cortex and medulla has disappeared; the various cell types tend to 
homogeneity; lymphocytes, macrophages and eosinophile myelocytes become 
rare; Hassall's corpuscles numerous and polymorphic. (3) The thymic lobules 
become fragmented by sclerotic bands, proceeding especially from the 
perivascular connective tissue; sometimes the normal lobular topography 
appears inverted, the cortical zone through rarefaction and sclerosis becoming 
clearer than the medulla; lymphocytes with opaque nuclei predominate; a 
thin zone of mononuclears appears around the Hassall's corpuscles, which are 
still numerous, but sometimes cystic. (4) The thymus is now largely 
fibrous, with small, irregular lobules; homogeneity of cell types still 
more evident; cell activity restricted to a small zone around Hassall's 
corpuscles, which are chiefly cystic and sometimes united into irregular masses. 
(5) Thymus now presents a fibrous (collaginous) mass, poor in cells, with a few 
cell islands as remnants of the thymic lobules ; in the thickened reticulum appear a 
few pycnotic lymphocytes; usually no traces of the Hassall's corpuscles remain. 

Tixier and Feldzer hold that the thymic sclerosis is comparable to that 
occurring during inanition in other (especially the lymphoid) organs. 

The effects of inanition upon the thymus in atrophic infants are reviewed by 
Lesage ('n) and Monckeberg ('12), who considered the thymic involution as an 
effect, rather than a cause. Schridde ('13, '21), however, contends that the 
atrophy of the thymus is a real factor in infantile malnutrition, or a cir cuius 
vitiosus, inanition causing a sclerous atrophy of the thymus, which in turn 
intensifies the malnutrition. Schridde also described an increasing amount 
of fatty granules in the reticulum cells of the cortex; and a disappearance of 
the lymphocytes and eosinophile leukocytes during the progressive sclerosis. 
Plasma cells rarely occur (contrary to some authors). 

Cremieu ('12) claimed that the involution of the thymus by inanition, 
X-rays, etc., is fundamentally similar to the normal age involution. "II ne 
s'agit que dedeux modalites differents d'un processus uniforme: l'envahissement 
conjunctif. D'un cote, le tissu conjonctif reste sec; de l'autre, il se charge de 
vesicules adipeuses." 

Matti ('13) gave an extensive review of the work on thymic involution in 
man and animals. 

Hornowski ('13) found the weight of the thymus usually varying directly 
with the general nutritive condition in the newborn. Lesage and Cleret ('14) 
emphasized sclerosis during athrepsia as the characteristic change, occur- 
ring to a variable degree in all organs, especially in the thymus. 

Mattei ('14) gave a careful description of the changes in the atrophic thymus 
of athreptic infants, his findings being in general agreement with those of Lucien 
and Feldzer. He concluded that "Le tissu propre de l'organe presente 


deux modifications importantes: /' inversion thymique et la sclerose marquee, 
les deux alterations se rencontrent dans le meme organe a, differents degres, 
suivant les cas." He described epithelioid cells and giant mononuclear cells 
as abundant in the cortex during thymic inversion, with peripheral invasion 
of the lobule by adipose tissue. 

Nobecourt ('16) reviewed the changes in the thymic atrophy of malnourished 
infants, following in general the findings of Lucien, Feldzer and Mattei. He 
noted that the nature of the thymus atrophy has been explained in three ways: 
(1) as a primary cause. of the general malnutrition (Farret, Durante, v. 
Mettenheimer) ; (2) as a simple, secondary atrophy, comparable to that occur- 
ring in other organs during inanition (Friedleben, Clark, Ghika, Seydel, Sokoloff) ; 
and (3) as due to toxic or infectious causes (Tixier, Feldzer, Martel, Marfan). 

Hart ('17) described the thymus in a starved child of 3 years, complicated 
with " eine floride Rachitis leichten Grades." The thymus weighed 5 g. Frozen 
sections stained with Sudan showed no sclerosis, but numerous peripheral 
cells with vesicular nuclei filled with fatty droplets of variable size, and some 
cells with complete fatty degeneration. In the center of the lobules, Hassall's 
corpuscles contained fatty detritus, but cells with fatty degeneration were rare. 
The cells in the septa were mostly fat-free. The ordinary stains (hematoxylin, 
etc.) showed the typical inversion of the lobule, with lighter cortex and darker 

Hammar ('21) has recently reviewed the problem of thymus involution, 
including the accidental involution caused by disease, pregnancy, X-rays, and 
especially inanition. In congenital pyloric stenosis and in infantile pyloro- 
spasm, accurate measurements showed the thymic parenchyma reduced 80 per 
cent. In general, the cortex is more labile, being rapidly reduced by wholesale 
emigration of lymphocytes, which infiltrate the adjacent interstitium and 
vessels. The medulla also becomes richer in lymphocytes. The number of 
mitoses is greatly reduced; but a few persist, especially in the reticulum cells. 
The distinction between cortex and medulla apparently disappears, although 
fat stains show abundant fat droplets appearing in the cortical reticulum. The 
medulla may become relatively richer in nuclei, producing the "inversion" 
of the French authors, which appears oftener in illness than in starvation. 

During inanition-involution, according to Hammar, the medulla remains 
unchanged longer than the cortex, but later decreases in size by general atrophy 
(and some degeneration) of the individual medullary cells. Hassall's corpuscles 
are more resistant, hence become relatively more numerous, although there is a 
slow decrease in absolute numbers. In extreme cases the parenchyma may 
decrease to 1 or 2 per cent of the original amount, and the thymus presents only 
narrow strips, poor in lymphocytes, and in which Hassall's corpuscles may disap- 
pear entirely. During Rontgen involution, the lymphocytes do not emigrate, 
but disintegrate in situ. There are also some variations in the process of involu- 
tion in different diseases, but in general the process resembles that of inanition- 
involution. (Hammar '17, '18, '20, '21.) 

In famine-stricken children of various ages, Nicolaeff ('23) found the thymus 
90-98 per cent subnormal in weight, compared with Hammar's norm for age. 



The atrophied thymus was found reduced to fibrous cords, with small islets of 
gland tissue. Similar data are reported by Stefko ('23a). Stephani ('23) 
noted fat in the thymus cells of atrophic infants. 

Keilmann ('23) studied the thymus in 86 nurslings and 73 children of the 
second year (no individual data). Unlike most previous investigators, he was 
unable to find any regular relation between thymus weight and the character 
of the disease causing death. He found both high and low thymus weights in 
all conditions, and therefore doubts whether the thymus weight can be accepted 
as an index of nutrition. He cites the observations of the Italians, Oliari and 
Spolverini (cf. abstracts in Arch. f. Kinderh., 65:124-130) in support of this 
conclusion. Keilmann's observations upon the histological changes in the 

1 -L 

1 1 


! ' 


: 1 M 





1 II' 


M I 




j ' ' 




1 I 1 I i 





1 t 

1 MM 1 


III hi 

1 11 ; 

1 1 ' 

- ^V(7i<5.W" 

< — "m 


1 ■ ! i * i 1 

ill M . 


11 ;il M 


J : ! j 


' | 

■ | j 

U 1 II i 


b~ M ! 


; | 1 1 

i ! i 



1 i.j , | 

£U 1 


1 " 

1 ! ! 

ii 1 1 i 



1 ■ 1 1 ! ; 





y 1 


i i 1 

1 I iiil 

t1 < 




M f 

1 i ' 

1 1 i l ' 1 




B' M ! 

!/T ' 

'I 'III 

1 ! 

II ' ; 1 | 

I /\\ 1 III 

1 ! 






y \\\ Mil 




Pi ^ 


M M 


i r 



. • T 

1 I 

. . 




1 1 


" l 

ML r 

; III 




1 " III! 

1 M 

! i 

1 1 

- r 


1 i 1 j 

1 1, 

X !• 

| 1 1 

! 1 ' Ml 


y . ' 

i ' 1 ' '■ 1 

' ! 

1 in 

| j 1 

' • * 

mI • 

• t 

. * 

1 M ' 


1 1 1 1 1 T 1 

t | 


1 1 

I 1 mi 

TIT t 

S • ' 

I ill 

; . ' 

I i M • 

! i ! 

. M • 

III li! 1 ! 

1 1 ! 

it i I 

r M 



I ] 


"• 1 


ft • 



'Ill " 

• _[ (i 


t •• 

_ <TT ~ *1 

» j 11 1 1 4 

f • 

i ! i 

• 1 : 

■ r*" 




J li 


• ! n«J *. 

Si i* i 1 — . 

Jiit M 


• til ,r 1 : & ". 

■ it 1 ** 

V . 

, . 



"• i 

| « • • 

• •• ♦ * 

- »•! ♦ - 



• • • 

»*• >!*!•• 

»• ., I* 1 A 


. • 

I * 


* * 

-Hi • 

• A ,1 « 

« *i 1 

, f , » 

' , 

m W 

• ' 1 *! 

I *•'':<■ 

: | , 

• t 

• : -• ■ • i ■ 


• % n 

■ 4 • 1 • • 


■ n . 

j 1 . • 





~4l AA 46 46 50 51 54 56 56- 60 62. 64 6b 60 70 11 74. 76 
Fig. 79. — Graph showing the individual weights of the thymus, according to body length, 
in atrophic infants from various sources. The larger dots represent original Minnesota cases. 
The normal curve is from data compiled by Prof. R. E. Scammon. The profound atrophy of 
the thymus during malnutrition is evident, the weight being above normal in only 2 out of 
about 300 cases. 

thymus are likewise in disagreement with those of previous investigators. He 
noted the so-called "inversion" of cortex and medulla in 4 cases, but did not 
find a disappearance of the lymphocytes accompanying the process of fibrosis. 
He claims that Hassall's corpuscles are usually greatly increased. He admits 
that low thymus weights are in part due to emaciation, but holds that they are 
not always pathological, since the norm is usually put too high. 

My own observations upon the weight of the thymus in atrophic infants, in 
opposition to the conclusions of Keilmann, show clearly the marked effect of 
malnutrition upon the thymus weight, confirming the findings of most previous 
observers. In the 15 cases shown in Table 3, the thymus weight ranged from 
0.1-5.8 g. (the latter in a slightly rachitic infant of 6 months). The normal 
for the newborn is about 13 g., increasing with age (see Fig. 79). 


In a series of about 40 cases (about half of which represent original observa- 
tions upon Minnesota cases), in which complete data were available, the average 
loss in thymus weight in atrophic infants has been calculated upon various 
bases, as shown in Table 2. Thus in comparison with Scammon's norm for the^ 
corresponding final body weight, the thymus averages 71.8 per cent, below nor- 
mal for the entire series, or 75.8 per cent below for the Minnesota data alone. 
Compared with the normal for the maximum body weight reached during life, 
the thymus weight averages 80.7 per cent subnormal. Compared with the norm 
for corresponding height, the thymus loss is 80.6 (80.4) per cent; while for age 
the thymus appears 82.6 (84.4) per cent subnormal. 

The marked depression of the thymus weight in malnourished infants also 
appears clearly in Fig. 79, a field graph representing all available cases in which 
the body weight was 20 per cent or more subnormal according to body length, 
irrespective of the cause of death. The large dots represent original Minnesota 
data. Out of nearly 300 cases, only 2 appear above normal, and the profound 
atrophy in nearly all cases is clearly evident. 

Among the animals, the effect of unfavorable environment upon the thymus 
has long been known. Wharton (1659) found that the strenuous labor to which 
young oxen were subjected when yoked causes a marked atrophy of the thymus. 
Gulliver (1842) similarly observed that "in overdriven lambs the thymus will 
soon shrink remarkably and be nearly drained of its contents, but will become 
as quickly distended again during rest and plentiful nourishment." 

Friedleben ('58), in addition to the above mentioned observations on the 
human thymus, starved young puppies for 12 hours, 40 hours and 14 days, 
respectively, and demonstrated that the thymus atrophied relatively more than 
the liver, spleen or the entire body. In the longest experiment, the thymus was 
reduced to a mere trace, while the body lost 45.7 per cent in weight. Manassein 
('69) included some data indicating a marked loss in the weight of the thymus 
in fasting young rabbits, but he makes no comment thereon. 

Hofmeister ('92) pointed out the similarity between the hunger involution 
of the thymus and that in other lymphoid tissues, and thought there is a relation 
between this and the blood lymphocytes. Voit ('94) noted, in dog starved 22 
days, almost complete disappearance of the thymus, which weighed 31 g. in 
a normal litter control. 

During fasting or hibernation in frogs, Ver Eecke ('99) found a loss of 75 
per cent or more in the weight of the thymus. The atrophy involves both cortex 
and medulla. There is a rarefaction of the lymphoepithelial tissue (including 
the concentric corpuscles), which is more marked in the medulla. Persistence 
of the connective tissue gives an appearance of pseudosclerosis. 

We come now to the work of Hammar and his co-workers, who introduced 
more accurate methods of histological analysis, which have contributed largely 
to the rapid recent progress in our knowledge of the thymus problem in man and 
animals. Hammar ('05a) first described the changes as found in rabbits and 
frogs subjected to inanition, and also in pathological cases in animals and man. 
The rabbit's thymus may lose half its weight in 3 days of fasting. The mitoses 
decrease rapidly in number (especially in the cortex) and soon disappear. The 


lymphocytes of the cortex become greatly reduced in number through migration 
into the surrounding connective tissue, especially around the blood vessels and 
lymph vessels. There may also be migration into the medulla. The emigration 
of the lymphocytes leaves the cortical reticulum more prominent and the surface 
cells may become epithelioid in appearance (as in embryonal stages). The dis- 
tinction between cortex and medulla is lost, and degenerative changes appear in 
the remaining cells. There is at least a relative increase in the interlobular 
connective tissue and the vascular stroma, but it is uncertain whether there is an 
actual sclerosis, as described especially by the French authors. The myoid cells 
and Hassall's corpuscles undergo variable, regressive changes, sometimes disap- 
pearing. In the teleost fish, Labrns rupestris, starved 23-31 days, Hammar 
('09) described a reduction to one-sixth in the weight of the thymus, and histo- 
logical changes similar to those found in higher vertebrates. 

Hammar's pupil, Jonson ('08, '09), made a series of accurately controlled 
experiments upon young rabbits on acute inanition (water only) or chronic 
underfeeding (maintenance of constant body weight) ; also a series refed after 
inanition. In 4 weeks of chronic inanition, the thymus weight decreased to about 
^0 and the parenchyma to about ^'65 of the normal. In 9 days of acute inani- 
tion, the thymus was reduced to 34 an d the parenchyma to 34 normal weight. 

Of the parenchyma, the cortex suffered the greatest reduction — to 342 in 2 
weeks of underfeeding and to % in 5 days of acute inanition. In later stages the 
cortex has usually disappeared. The cortex atrophies chiefly by emigration 
of the lymphocytes into the lymph spaces (and veins?). The persistent reticu- 
lum cells gradually assume an epithelioid appearance, undergoing degenerative 
changes only in the later stages. The number of mitoses in the entire thymus 
was reduced from 28,500,000 to 6,500,000 in 4 days of acute inanition, and from 
10,500,000 to 3,100 in 4 weeks of chronic inanition. The persistent mitoses 
were chiefly in the reticulum cells. Hassall's corpuscles showed a varied resis- 
tance. The unicellular forms disappeared (by simple atrophy?) even in the 
second week of underfeeding, while the multicellular corpuscles were reduced 
(partly by degeneration) from 139,200 to 16,100 in 4 weeks. During acute 
inanition, the unicellular forms decreased from 170,000 to 44,000, and the 
multicellular from about 741,500 to 352,700. The interstitial (fibrous) tissue 
was reduced to about % normal weight in 4 weeks of underfeeding; with a 
smaller loss in acute inanition. The interstitial fat is completely resorbed, 
and the interstitial fibrous tissue assumes a loose, edematous appearance. 

Jonson's refed rabbits made a rapid recovery, showing a distinct increase in 
the weight of thymus and parenchyma even in 2 days, and nearly normal condi- 
tions in 3 weeks. "Die Zunahme des Parenchyms wird in erster Linie durch 
Einwanderung von Lymphocyten aus den Lymph- (und Blut-) Wegen bedingt, 
wozu friihzeitig eine nicht geringe Zunahme von Mitosen in Lymphocyten und 
Retikulumzellen hinzukommt." The cortex regenerates first. Differentiation 
of the typical structure in the medulla, with regeneration of Hassall's corpuscles 
(by hypertrophy of the reticulum cells), begins only after 2 weeks of refeeding. 
The interstitial fibrous stroma recuperates rapidly, and fat cell groups begin to 
appear even after 2 days of refeeding. 


Kallmark ('11), in connection with his study of the effects of inanition 
upon the blood of young rabbits, also made some observations upon the 
lymphoid organs, including the thymus. In an initial control of 3 months, 
the thymus weight was 3.15 g. ; after 19 days of underfeeding (at mainte- 
nance), 0.3 g. ; after 28 days of underfeeding, plus 2 days on water only, 
0.17 g. ; after 31 days of underfeeding, 0.20 g. ; and in a final, full fed control, 
4 months old, 3.45 g. The blood lymphocyte count does not show much 
change, but there is a transient decrease at the beginning of inanition, which is 
ascribed to the atrophy of the lymphoid tissues of the body in general. There 
is also a transient increase in the blood lymphocyte count upon refeeding, 
indicating an over-compensatory regeneration in the lymphoid organs. 

Jolly and Levin ('n) studied the effects of inanition upon the weight of 
the lymphoid organs in the pigeon, chick, duck and guinea pig. In the birds, 
with loss in body weight of 30-37 per cent, the thymus lost 51-80 percent, which 
is relatively greater than the loss in the bursa of Fabricius (48-77 per cent), or 
spleen (53-67 per cent). In young fasting guinea pigs, the thymus and spleen 
showed similar losses. In pigeons, the thymus also made the most rapid 
recovery in weight. In a later paper ('11a), they showed that in all these 
animals the histological process of involution in the thymus during inanition is 
similar to that described by Hammar and Jonson. The diminution in weight 
is due chiefly to loss of the cortex, which is caused primarily by emigration of 
the lymphocytes, with decrease in mitosis and some loss by phagocytosis. 
Degenerative changes are described in the cortex and medulla; also the regenera- 
tive process upon refeeding (in pigeon and guinea pig). The degenerative 
and regenerative changes, especially in Hassall's corpuscles, are more fully 
considered in a later paper ('12). 

Holmstrom ('n, '12) found lipoidal granules stainable with scarlet red in 
the reticulum cells of normal newborn rabbits. These granules, which are 
independent of the ordinary interstitial fat, become more numerous with age, 
and also during malnutrition. Cremieu ('12) noted in 2 malnourished kittens 
a sclerous atrophy of the thymus, like that produced by the X-rays, etc. Levin 
('12) described the typical process of involution in the thymus during inanition, 
and the recovery upon refeeding. In underfed tadpoles of Rana fusca, Dustin 
('13a) observed an atrophy of the thymus, with cessation of mitosis and trans- 
formation into small cells with pycnotic nuclei. 

The extensive monograph of Salkind ('15) includes several inanition experi- 
ments on the thymus. Puppies of various ages and underfed to various degrees 
showed an atrophic thymus, with the cortex reduced to a narrow zone of 
lymphocytes and the medulla large and vascular. Mitoses are greatly decreased 
in number. The stroma cells show evidences of phagocytosis. Mast cells 
are numerous. Thus underfeeding depresses the secretory activity of the 
thymus (production of lymphocytes); but increases the phagocytic activity, 
and ultimately causes sclerosis. In rats one month old without food or water, 
signs of "delymphoidisation" appear in 12 hours, with pycnosis in the follicular 
centers and phagocytosis of lymphocytes by the reticulum cells. This process 
continues and at 3 days there is an inversion, the cortex becoming light and the 



medulla darker in appearance. The formation of HassaH's corpuscles from 
the epithelioid cells continues. At 4 days, "delymphoidisation" is complete 
and there is but slight distinction between cortex and medulla. Mitoses have 
entirely disappeared; Hassall's corpuscles become cystic and are destroyed by 
eosinophiles. Recovery of the rat upon refeeding is impossible at this stage, 
but occurs after 2 days of starvation. The process of "relymphoidisation" is 
somewhat slow, mitoses appearing rare after 48 hours of refeeding. At least 
a week is required to re-establish the normal lymphoid structure. A chick 

Fig. 80. Fig. 81. 

Fig. 80. — Photograph of a cross section of the thymus in a normal albino rat (S. 2) at 3 
weeks of age. The gland is mostly cortex, the few light areas representing the medulla. The 
capsule and interlobular septa are thin. A, perithymic adipose tissue. Zenker fixation; 
alum-hematoxylin stain. X22. 

Fig. 81. — Photograph of a cross section of the thymus in an albino rat (S. 11) held at 
constant body weight by underfeeding from 3 to 10 weeks of age. The thymus is greatly 
reduced in size, and the distinction between cortex and medulla obliterated. The capsule and 
interlobular septa are relatively thickened. Some masses of atrophic, perithymic adipose 
tissue (A) are included in the section. Zenker fixation; alum-hematoxylin stain. X22. 

(a few days old) after 12 hours' starvation showed beginning "delymphoidisa- 
tion," similar to that in mammals. Lizards starved 1-5 weeks also showed 
the typical thymic involution, but a teleost fish ("Chat de Mer") showed no 
appreciable change in 6 days. 

Some of the changes observed in the weight of the thymus of the albino rat 
during inanition by Jackson and his co-workers are shown in Table 4. In rats 
underfed from the age of 3 weeks to 10 weeks, Jackson ('15a) noted a loss of 
about 90 per cent in the thymus. In rats underfed from birth for various 


periods, Stewart ('18, '19) found a smaller loss (30-80 per cent); and in the 
stunted offspring of severely underfed pregnant rats, Barry ('20, '21) found 
the thymus only 21 per cent subnormal in weight. 

The reduction in the thymus of underfed young albino rats is shown by 
Figs. 80 and 81. 

Upon refeeding rats underfed from 3 to 12 weeks of age, Stewart ('16) 
found the thymus still somewhat subnormal after 2 weeks; but after 4 weeks 
it appeared 50-70 per cent above normal, possibly indicating an over-com- 
pensatory growth. In rats underfed from birth to 3, 6 or 10 weeks of age, 
and subsequently refed to 25, 50 or 75 g. body weight, Jackson and Stewart 
('19) found the thymus variable, but in most cases still slightly subnormal in 
weight (Table 7). In rats similarly refed to maximum (adult) weight, after 
earlier periods of underfeeding, Jackson and Stewart ('20) found the thymus 
still subnormal in weight in 3 of the 4 groups (Table 8). The evidence would 
seem to indicate a permanent dwarfing of the thymus in most cases, although 
the great normal variability renders the results uncertain. 

Trowbridge, Moulton and Haigh ('18) and Moulton, Trowbridge and Haigh 
('22a), in steers on different planes of nutrition, found that the thymus ("heart 
sweetbread") averages higher in weight in fat animals, but the reduction in 
the poorly nourished is much less than might be expected from the results of 
of inanition in other species. 

McCarrison ('19b, '21) gave organ weights indicating a great loss in the 
weight of the thymus in pigeons during starvation as well as on various deficient 
diets. Findlay ('21) found total disappearance of the thymus in starved 
pigeons and fowls (Table 13). 

Ikeda ('22) in fasting rabbits found that the thymus atrophies more rapidly 
than the other viscera, due chiefly to emigration of lymphocytes, but also to 
degeneration of the parenchyma. Okuneff ('22) observed large amounts of 
isotropic, and especially of anisotropic, lipoids in the thymus of rabbits starved 
until the ordinary reserve fat was assumed to be exhausted. 

Several observations indicate that the thymus undergoes a typical involution 
during hibernation. A winter-atrophy of the thymus was noted by Ver Eecke 
('99) in the frog (species?) ; and by Schaffer ('08) in the mole. A similar process, 
with the typical involution changes during hibernation, was described by Aime 
('12, '12a) for chelonians {Emys vittata, Clemnys leprosa, Testudo mauritanica) . 
The annual regeneration is apparently by a process of budding, the epithelioid 
buds being later invaded secondarily by small thymic cells, which form the 
cortex (as in normal development). Dustin ('13) observed a seasonal 
involution in adult amphibia, Rana fusca and Bufo vulgaris. Mann ('16), 
however, found apparently no uniform change in the thymus of hibernating 
spermophiles (Spermophilus tridecemlineatus) . 

(B) Effects of Partial Inanition 

These will include changes in the thymus during various partial dietary 
deficiencies, especially in rickets, beriberi, scurvy, and thirst (aqueous inanition). 

2 g8 inanition and malnutrition 

Lefholz ('23) found that the thymus, spleen and cervical lymph glands of 
kittens showed no consistent response to dietary variations in the amounts of 
fat, sugar and protein, although the tonsils and lymphoid structures of the 
alimentary canal generally were markedly affected. 

In human rickets, Seibold (1827) noted: "Die Glandula Thymus und das 
Mediastinum fand man meistens mit einzelnen Verhartungen." 

In 14 non-rachitic infants, du Castel ('08) found the thymus weight 
above 8 g. (which was considered normal) in only 1, the average being 4.50 g. 
But in 13 rachitic infants below 3 years of age, the thymus exceeded 8 g. 
in 8 cases, the maximum being 15 g. and the average 9.42 g. "On 
peu done conclure que dans la majeure partie des cas le thymus rachitique est 
hypertrophic; l'augmentation du poids de l'organe est due a la proliferation du 
tissu lymphoide; on trouve assez souvent des myelocytes en grande nombre; 
les corpuscles de Hassall sont egalement plus nombreux que dans les thymus 
non rachitique; enfin, le plus souvent, on n'y constate plus la sclerose, si fre- 
quente chez ces derniers." It may be noted, however, that du Castel's thymus 
weights are all actually subnormal (c/. Fig. 79) ; whether the thymus is larger 
than would be expected in these cases, according to the general nutrition of 
the body, is uncertain on account of the lack of data concerning body weight, etc. 

Sweet ('21) advanced the theory that rickets may be due to a deficient 
secretion of the thymus; but this fails to account for the apparent antagonism 
between rickets and starvation, which has repeatedly been noted in man and 
animals, in spite of the profound atrophy of the thymus during starvation. 

Marfan ('22) confirms du Castel's claim that the thymus is enlarged in 
human rickets, accompanying a general hypertrophy of the lymphoid organs 

In experimental rickets in rats, Shipley, Park, McCollum and Simmonds 
('21) found the thymus atrophic (no weights given) in animals on diets deficient 
in fat soluble A, with or without deficiency also in phosphorus. A similar 
condition was mentioned by McCollum, Simmonds, Shipley and Park ('21) 
in rats on low calcium diets, with "undersized" bodies. 

From a study of the weights of the thymus in an extensive series of albino 
rats, Jackson and Carleton ('23) found that the thymus averages slightly above 
Donaldson's norm in the "normal control" group, but shows a progressive 
decrease in weight in the test groups, reaching 70 per cent below normal in the 
severely rachitic group (Table n). The weight-length ratio of the body was 
nearly normal, but nearly all of the test rats were subnormal in weight according 
to age, indicating a retardation in growth. This would tend to depress the 
thymus weight, irrespective of any specific effect of the rickets. 

Several observations are available upon the thymus in beriberi (vitamin 
B deficiency). Andrews ('12) in 18 necropsies of infantile beriberi found only 
slight apparent changes in the thymus, aside from congestion. 

In animal experiments, however, a profound atrophy of the thymus in beri- 
beri or polyneuritis was found by Funk and Douglas ('14) in pigeons; by Wil- 
liams and Crowell ('15) in pigeons and chickens; by Douglas ('15) in pigeons; 
by Drummond ('16) in chickens; by McCarrison ('19, '19a, '19b, '19c, '19c, 


'20a, '21) in pigeons and monkeys; by Emmett and Allen ('20) in rats; by 
Cramer, Drew and Mottram ('21, '21a) in mice and rats; and by Findlay ('21) 
and Korenchevsky ('23a) in pigeons and fowls. Findlay found total disappear- 
ance of the thymus in the rice-fed as well as the starved animals (Table 13). 
Lopez-Lomba ('23) noted a transient hypertrophy in the second period (9th 
to 14th days), preceding the final atrophy of the thymus in adult pigeons on 
vitamin-free diet. 

It may be noted, however, that during beriberi there is usually a general 
emaciation, with marked loss in body weight, which might account for the 
thymic atrophy, independent of any specific effect of the vitamin deficiency. 
Thus Williams and Crowell ('15) concluded: "The experimental evidence 
indicates that there is no apparent fundamental connection between beriberi 
and the atrophy of the thymus; when the latter occurs in birds fed on polished 
rice, as it frequently does, it is due to some other cause." On the other hand, 
the atrophy of the thymus in beriberi is in agreement with the doctrine of 
Cramer, Drew and Mottram ('21, '21a), according to which a deficiency of 
vitamin B causes a specific atrophy of lymphoid tissue throughout the body. 

In scurvy, but few data on the thymus are available, according to Hess 
('20). Jacobsthal ('00) in a case of infantile scurvy found no apparent gross or 
histological changes in the structure of the thymus. Bierich ('19) mentioned 
an apparent enlargement of the thymus in 1 out of 8 (adult) cases. Aschoff 
and Koch ('19) noted no abnormalities of the thymus in adult scurvy. 

During thirst (aqueous inanition), Falck and Scheffer ('54) observed an 
apparent loss of 63 per cent in the thymus of a dog on dry diet, with loss of 
20 per cent in body weight. In adult albino rats on acute thirst experiments, 
Kudo ('21) found an apparent loss of 78 per cent in the thymus weight; while in 
the chronic thirst series the loss averaged 90 per cent (Table 9). In similar 
experiments on young albino rats, held at constant body weight by a relatively 
dry diet, Kudo ('21a) found a loss of 68.9-91.3 per cent in the weight of the 
thymus in the various test groups, being relatively greater than the loss in 
any other organ (Table 10). 



The effects of inanition upon the alimentary canal are widespread and sig- 
nificant. A knowledge of the changes during the earlier stages of inanition is 
necessary to understand the normal processes of digestion and absorption. 
The effects during the later stages are associated with the marked disturbances 
of the digestive system which are characteristic, not only in total or partial 
inanition in the narrower sense of the term, but also in the state of malnutrition 
which is associated with so large a proportion of diseases in general. In both 
children and adults, there is often established a "vicious circle," the primary 
involvement of inanition causing an atrophy of the alimentary tract, which in 
turn serves to intensify the general state of malnutrition. The process of 
recuperation is also conditioned by the possibility of regeneration in the atro- 
phied and degenerated tissues of the digestive system, a consideration of great 
practical importance to the physician. Following a somewhat brief summary 
of the effects upon the entire canal, the changes will be reviewed in detail under 
(i) the mouth, pharynx and esophagus, (2) the stomach, and (3) the intestines. 

Summary of the Effects on the Alimentary Canal 

The changes in the alimentary canal during inanition vary greatly in the 
different regions. 

In the mouth, atrophic and degenerative changes have been noted in the 
epithelium of the oral mucosa, but mitoses persist in the deeper layers. The 
tongue is very resistant to loss in weight (frog), but during extreme thirst 
degenerative changes occur in the lingual muscle fibers (fowl). Lesions of the 
oral mucosa occur in pellagra, and especially in the gingival regions during 

In the pharynx, the size of the tonsils is greatly affected by the character 
of the diet as to calories, protein, carbohydrates and especially the fat content. 
The weight of the pharynx and esophagus is found reduced nearly in proportion 
to the body weight during inanition (pigeons and steers). The lesions in 
these segments are usually slight. 

The stomach presents extremely variable changes during inanition. In 
human starvation, it usually appears contracted and small in adults, but fre- 
quently above normal weight in malnourished infants. In animals, the loss 
in weight is usually relatively less (rarely more) than in the body as a whole, 
and there may even be a persistent gain in gastric weight in the young (rat and 
human infant) during chronic underfeeding. 



Structurally, the stomach during inanition presents a variable degree of 
atrophy, with degenerative changes in extreme cases. The mucosa may appear 
normal, aside from simple atrophy; but in advanced stages of inanition there is 
frequently congestion and often localized degenerative changes — erosion of 
the superficial epithelium, ulceration, hemorrhages, regressive changes in the 
gastric glands; and edema, leukocytic infiltration or fibrosis in the stroma of 
the mucosa and submucosa. The smooth muscle tissue likewise shows a vari- 
able degree of atrophy. 

The localized degenerative gastric lesions in the advanced stages are prob- 
ably due largely to gastritis, which appears to be more frequent during human 
starvation than in experimental inanition among animals. Infantile atrophy 
is usually secondary to gastroenteritis (or other disorders, such as syphilis or 
tuberculosis), which interferes with nutrition and establishes a "vicious 
circle." The resultant inanition causes atrophy of the alimentary canal and 
associated glands, which in turn intensifies the general state of malnutrition. 

Changes similar to those described for inanition occur in the alimentary 
canal of the fasting salmon; and also (to a variable extent) in various animals 
during hibernation, during which a characteristic leukocytic infiltration of the 
stroma has been observed. 

Very similar atrophic and degenerative gastric changes occur to a variable 
degree also in the different types of partial inanition, including protein and 
vitamin deficiencies, thirst (aqueous inanition), etc. 

In the intestines, the changes during inanition in general resemble those 
in the stomach. There is a variable loss in weight with shortening and attenua- 
tion of the intestinal wall. In both man and animals there is a general and 
progressive atrophy in the earlier stages of inanition, following certain minor 
changes in the mucosa associated with the normal process of digestion and 
absorption. The epithelium (surface and glandular) contains mitochondria 
and lipoidal granules which are somewhat resistant to inanition. The atrophy 
affects all the tissues, but especially the intestinal glands and lymphoid struc- 
tures. In the later stages, as in the stomach, more profound and degenerative 
lesions may occur, with enteritis, ulceration, hemorrhages, and disintegration 
of surface epithelium, glands, and even entire villi. 

In the young, similar intestinal changes may occur; but the process is in 
some cases modified by the persistent growth impulse. 

In various types of partial inanition, a variable degree of intestinal atrophy 
and, in more extreme cases, degenerative and inflammatory lesions occur, 
which are essentially similar to those found during total inanition. General 
atrophy and weakness of the intestinal walls, and especially of the tunica muscu- 
laris and associated sympathetic ganglia, may result in distension and meteor- 
ism. The lymphoid structures appear variable, perhaps because the primary 
atrophic effect of the inanition is sometimes opposed by the toxic or inflamma- 
tory conditions. The tendency to hemorrhages is especially marked in scurvy 
and thirst (aqueous inanition). 

Upon refeeding after a period of inanition, mitosis (which is depressed during 
inanition) is greatly accelerated in the cells of the various atrophic intestinal 


tissues (especially the lymphoid and epithelial), and the normal weight and 
structure may be promptly restored. If the inanition has been severe or 
prolonged, however, recuperation is usually slow and difficult. In adult starva- 
tion or infantile atrophy and similar conditions in experimental inanition, the 
degenerative lesions in extreme cases may render recovery impossible. 


The data as to the effects of inanition upon these segments of the alimentary 
canal are scanty. The teeth have already been discussed in Chapter VIII. 

The Mouth. — Rabl ('85) found many mitoses in the tongue glands of Sala- 
mandra atra starved 5-7 months. Morpurgo ('88, '89) noted persistent 
mitoses in the epithelium of the lingual mucosa and palate in both young and 
adult rabbits during fasting. The mitoses occur chiefly in the deeper layer 
of cells. No abnormality in the chromatin was noted. Porter ('89) noted 
furring of the human tongue as characteristic in chronic famine, though it may 
become raw and denuded when diarrhea and dysentery set in. "The epithelial 
tissue was very oily, and a true fatty transformation seemed to affect the deeper 
layers of epithelium in greater or less extent, leading to an imperfect develop- 
ment of the growing cells and ready detachment of those already formed." 

Reese ('13, '13a) found no significant change in the digestive tract (includ- 
ing tongue, palate and esophagus) of the hibernating alligator. 

Ott ('24) noted that during hibernation and subsequent inanition in the 
frog (Rana pipiens), with progressive loss of body weight up to 60 per cent, the 
tongue (unlike the skeletal musculature) shows no definite change in weight 
(Table 6), and therefore becomes relatively much larger. 

Lesions of the oral mucosa, with stomatitis, etc., were noted by Chittenden 
and Underhill ('17) in a dog with experimental pellagra, and similar effects are 
mentioned by Harris^ ('19) in human pellagra. Inflammation, hemorrhages 
and ulceration of the gingival mucosa, especially in the peridental region, are 
well known symptoms of human scurvy, as noted, for example, by Sato and 
Nambu ('08), Bierich ('19) and Comrie ('20). The changes are reviewed in 
detail by Hess ('20). Beach ('23) found pustules in the mouth, pharynx and 
esophagus of fowls on diets deficient in vitamin A . 

Tiedemann ('36) cited observations indicating dryness of the mouth and 
throat during thirst in man and animals. Falck and Scheffer ('54), in a dog on 
aqueous inanition (dry diet) for 4 weeks with loss of 20 per cent in body weight, 
observed an apparent loss of only about 8 per cent in the tongue (with hyoid) 
and the same in the esophagus. In fowls on a dry diet, Pernice and Scagliosi 
('95a) found the tongue showing marked passive hyperemia. The most super- 
ficial muscles appeared pale, with the muscle fibers cloudy and desiccated in 
appearance; cross-striation partly absent. The cells of the lingual cartilage 
appeared atrophic and shrunken, with some cells degenerated into amorphic 
granular masses. The crop (a dilation of the esophagus) likewise appeared 
hyperemic and hemorrhagic, with variably atrophic gland cells. Small cell 
infiltration occurred in the mucous, submucous and muscular tunics. The 
muscle cells were pale and cloudy, with poorly stained nuclei. 


Marriott ('23) has recently reviewed the effects of thirst in man, including 
the cessation of salivary secretion, with dryness of the mucous membranes and 
shrivelling of the tongue and lips. 

The pharynx and esophagus (including crop) were found by Chossat ('43) 
to average 34 per cent subnormal in weight in pigeons starved with loss of 40 
per cent in body weight. Bourgeois ('70) from fasting experiments on various 
animals (mammals and birds) concluded that there are no marked changes in 
the mouth, pharynx or esophagus. 

Nothwang ('91) noted that in pigeons dead from thirst the esophagus and 
interior of the crop appeared dry. 

Thiercelin ('04) found congestion and sometimes small, superficial erosions 
in the esophagus of athreptic infants. 

Moulton, Trowbridge and Haigh ('22a) in steers on various planes of nutri- 
tion, including some greatly retarded in growth, found the esophagus (gullet) 
nearly proportional to the body weight in all cases. 

The experiments of Lefholz ('23) upon kittens indicate that the palatine 
and pharyngeal tonsils are greatly affected by the character of the diet. With a 
diet high in sugar and protein, and also in calories, these organs are nearly 
doubled in size; while if the excess calories are provided in the form of fats, 
they are nearly trebled in size. 


Under this section will be included some general observations upon the 
gastrointestinal tract, as well as those referring to the stomach alone. The 
data on total inanition (or water only) will be considered first, followed by 
the various forms of partial inanition. 

(A) Effects of Total Inanition, or on Water Only 

The effects upon the human stomach (adult and infant) will be considered 
first; followed by the data for the infrahuman species. 

Human Adults. — Lucas (1826) cited an observation by Ballin of a contracted 
stomach with thickened walls in a case of starvation (religious dementia). 
From a review of the literature, Willien ('36) concluded that during inanition 
the walls of the digestive tract become atrophic, usually without inflammation. 
Tiedemann ('36) noted that in starved adults the stomach appears narrow and 
constricted, sometimes containing bile. Schultzen ('62, '63) likewise observed 
a contracted stomach in a girl 19 years old, who died from starvation. Curran 
('74), on the contrary, found the stomach distended with gas in an old woman. 
Jewett ('7 5) found no gastric lesion excepting a disappearance of the mucosa 
in a man of 74 years who died from chronic starvation. Bright et al. ('77) 
in the Harriet Staunton case found the stomach small, with very thin walls, 
through which the undigested contents were clearly visible. Falck ('81) 
stated that the stomach in starvation is usually found contracted and nearly 
empty, with the mucosa markedly folded and white, sometimes reddish in 
places (as found by Schultzen). Voelkel ('86) noted very thin and pale walls 
in the stomach and intestines of a starved man. 


In 459 autopsies upon victims of the Madras famine (including 226 men, 
155 women and 78 children), Porter ('89) classified the stomach as "large" 
in about ^3 of the men, x /b of the women and only 3 children; and as "small" 
in }/2 of the men, % of the women, and % of the children. It was found empty 
in % of the adults and ^3 of the children. The gastric mucosa appeared con- 
gested in 36 men, 14 women and 7 children, with variable pigmentation in 12 
men, 1 woman and 1 child. It was usually soft, with an anemic, catarrhal 

In a starved man, Stschastny C98) noted marked gastric hyperemia. The 
lining epithelium was lacking, but the peptic glands were preserved and a few 
scattered "hyaline Kugeln" (parietal cells ?) were observed. 

Meyer ('17) recorded a weight of 112 g. in the stomach of a man who 
died from starvation. The mucosa was pale and yellow, but presented no 
gross lesions. The parietal cells of the gastric mucosa were better preserved 
than the chief cells, which were disintegrated. The post mortem changes (18 
hours) could not be excluded. 

According to Ivanovsky ('23), Oppel noted that during the Russian famine 
perforating gastric ulcers became frequent on account of the use of indigestible 

The gastric changes have frequently been described in atrophic infants, 
which represent a variable group with mixed nutritional deficiencies, frequently 
complicated by infections, etc. 

Parrot ('77) described two types of gastric lesions (ulcerous and "diph- 
theroid") as characteristic and of primary importance, in infantile athrepsia. 

Baginsky ('84, '84a) claimed that there is in athreptic infants a primary 
atrophy of the gastrointestinal mucosa, which will be considered later, under 
the intestine. 

Marfan ('94) found the surface gastric epithelium usually lacking in dys- 
peptic infants, possibly due to post mortem changes. The gastric mucosa is 
replaced by a fibrous layer, infiltrated with leukocytes and containing multi- 
nucleated cells, desquamated epithelium, remnants of the gastric glands, etc. 

Fede ('97, '98, '01a) found chiefly atrophic changes, similar to those in 
other organs, in the gastric mucosa of athreptic infants. Instead of the more 
extensive lesions described by previous authors, he finds merely a marked 
thinning of the gastrointestinal wall, with occasional ecchymoses, but no 
ulcerations or destructions of glands or villi in uncomplicated cases, aside from 
post mortem changes. 

Thiercelin ('04) described in athrepsia the various gastric lesions, gross and 
microscopic, similar to those found by Parrot and Marfan. The changes in 
the mucosa are variable. There are often punctiform hemorrhages, with intense 
capillary congestion and hemorrhagic exudate between the gastric glands. 
Rarely ulcerations occur, with localized lesions of interstitial and parenchyma- 
tous gastritis, and frequently sclerosis. The other tunics may also be involved, 
and the interstices infiltrated with leukocytes. 

Schelble ('10) made careful histological study of the stomach and intestines 
in 37 atrophic infants, but could find no evidence to support the theory of 


gastrointestinal atrophy as the cause of pedatrophy. Vigor ('n) demonstrated 
that the distension of the abdomen ("ectasie abdominale" of Variot) which 
occurs in malnourished infants is due to distension of the stomach and colon by 
gas (air). 

Mattei ('14) found microscopic gastric lesions in about one-third of his cases 
of athreptic infants. The gastric glands appear " decapitated/' with vacuolated 
nuclei and cloudy swelling of the cytoplasm. The stroma is infiltrated with 
leukocytes, which also distend the lymph-vessels, and occur in the submucous 
and muscle coats. The submucous lymphoid follicles are hypertrophied. The 
literature on this subject is reviewed fully by Nobecourt ('16). 

In necropsies of famine-stricken children, Nicolaeff ('23) found evidence of 
gastritis, with punctiform extravasations of the gastric mucosa. 

The data recorded in Table 3, for atrophic Minnesota infants, would indicate 
that in spite of decreased body weight the weight of the stomach is in all cases 
above normal. The exact amount of the increase is difficult to estimate, on 
account of the lack of an accurate norm for the postnatal growth in weight of 
the stomach. 

Among the animals, the data upon the gastric effects will be reviewed in 
chronological order, excepting the early changes noted chiefly in studies of 
digestion and resorption, the phenomena in the migrating salmon, and the 
effects of hibernation, which will be considered separately. 

Collard de Martigny (1828) and de Pommer (1828) observed that in dogs, 
cats and rabbits, starved without food or water, the gastrointestinal mucosa in 
general appeared pale, without signs of inflammation. The gastric and intestinal 
secretions are decreased. 

Chossat ('43), in pigeons on total inanition with loss of 40 per cent in body 
weight, found an average loss of 33.4 per cent in the weight of the stomach as a 
whole. The thick gastric muscle (of the gizzard) alone lost 39.7 per cent (com- 
parable to the loss of 42.3 per cent in the skeletal musculature); but the thick- 
ened, cornified epithelial lining actually increased in weight from 1. 09-1. 23 
g., due to absorption of water, rendering the epithelium soft and pulpy. Schu- 
chardt (47) similarly obtained a loss of 34 per cent in the gastric muscle of 
starved pigeons. 

In a cat starved 18 days with loss of 50 per cent in body weight, Bidder and 
Schmidt ('52) found an apparent loss (compared with a control) of 30.9 per cent 
in the weight of the esophagus, stomach and intestines. 

In various animals, chiefly rabbits, starved with or without water, Manas- 
sein ('68, '69) found usually no gross changes in the stomach; although sometimes 
the mucosa appeared easily detachable, and petechial hemorrhages of varied 
size were occasionally observed. The average loss of gastric weight in 47 adult 
rabbits (body loss 39 per cent) was 34 per cent; in 8 rabbits 3%$ months old 
(body loss 33 per cent), the gastric loss was 16 per cent; in 3 rabbits 23-25 days 
old (body loss 35 per cent), the gastric loss was 25 per cent. In 5 rabbits refed 
after a fasting period, the stomach still appeared 11 per cent subnormal in weight. 
In 2 crows, starved with loss of 36 per cent in body weight, the loss in gastric 
weight was 30 per cent. 


Bourgeois C70) from fasting experiments on various animals (mammals and 
birds) and an extensive review of the previous literature concluded that the 
stomach becomes markedly small and constricted; the mucosa greatly folded, 
pale and not inflammed, although thickened near the cardiac and pyloric ends. 
The average loss in gastric weight is 33 per cent. 

In fasting dogs of various ages, Falck ('75) found the fundic region of the 
stomach containing air and some liquid; the mucosa white and greatly wrinkled. 
The pars pylorica was strongly contracted. 

In tritons starved 8-14 days, Schmidt ('82) noted that the gastric nuclei 
appeared small and dark, with very few mitoses. Upon refeeding, no mitoses 
were found in 1% hours; a few appeared in the fundus in 3^—6 hours; but 
none in the pyloric region until 7 hours after feeding. 

In the stomach of frogs starved for long periods (up to i^ years), Gaglio 
('84a) found the stomach waxy white in color, with mucous contents. The 
serosa and subserosa appear slightly thickened. The muscle tunic in part shows 
atrophic changes, with loss of striation in the muscle fibers, and degenerative 
changes in the nuclei. The submucosa is not much affected, but the muscularis 
mucosae may show changes similar to those in the muscle tunic. The muscle 
fibers sometimes appear vacuolated, with enlarged nuclei. The greatest changes 
occur in the tunica mucosa, which shows pronounced atrophy of the gastric 
glands, with progressive replacement by connective tissue (fibrosis or cirrhosis). 

Bizzozero and Vassale ('87) found no decrease in the number of mitoses of 
the gastric and intestinal glands in dogs fasting (up to 17 hours only). 

Morpurgo ('88, '89, '89a) in fasting rabbits of various ages usually found no 
marked changes in the stomach. The lymphoid follicles become atrophic, and 
sometimes the gastric glands appeared likewise. In the glands, mitoses persist 
in reduced number, but rarely occurred elsewhere in the stomach. In rabbits 
refed after a fasting period Morpurgo '(90) observed an increased number of 
mitoses in the gastric gland cells, but none in the lymphoid or connective tissues. 
The mast cells had apparently disappeared. 

Coen ('90) noted in the stomach of a rabbit starved 84 hours (without water) 
an exudate of the gastric and intestinal mucosa, resembling that of gastroenteritis. 
The exudate, composed of amorphous material and leukocytes, also infiltrated 
the mucosa and glands, down to the submucosa. Similar, but less marked, 
appearances were observed in 2 rabbits on water only, and in a kitten on total 

In dogs fasting several days, Nikolaides and Savas ('95) found, in sections 
prepared by Altmann's chrome-osmic method, numerous black (fatty?) granules 
in the epithelial cells of the pyloric and Brunner's glands, but not in the mucous 
glands. These granules disappeared slowly upon refeeding. 

Lazareff ('95), in guinea pigs starved with average losses in body weight of 
10, 20, 30 and 36 per cent, found corresponding losses of only 1.9, 6.5, 6.5 and 
1 1.8 per cent in the gastric weight (Table 5), which are relatively less than the 
loss in the intestines. Kusmin ('96) likewise found the loss in gastric weight 
relatively less than that in the intestines of fasting dogs, rabbits and guinea 
pigs; and Weiske ('97) obtained similar results in fasting rabbits (on water only). 


Mann ('98) found the chief and parietal cells well preserved in the stomach 
of a rabbit after 22 days of inanition (on water only?). The connective tissue 
stroma appeared scanty. 

Sedlmair ('99) noted but little change in the relative weight of the empty 
stomach and intestines in starved cats, the loss being nearly proportional to that 
of the whole body (50-55 per cent). 

Fede ('00a) and Quattrochi ('01) by chronic underfeeding of puppies 
obtained atrophy in the gastrointestinal wall, but no ulcerations, hemorrhages 
or destruction of glands, etc., as had been described in atrophic human infants. 

Swirski ('02) measured the gastric and intestinal contents in fasting rabbits 
and guinea pigs, and demonstrated that the feces are swallowed unless precau- 
tions are taken. 

Smallwood and Rogers ('n) described thinning of the gastric wall in Nectu- 
rus starved 4-16 months. The epithelial cells become vacuolated; the nuclei 
poor in chromatin. 

Greene and Skaer ('13) found in the basal portion of the gastric epithelial 
cells in fasting kittens and puppies a certain amount of persistent fat granules 
(liposomes), with no relation to absorption fat. 

In several dogs and a fox which had died from protracted inanition, Morgulis, 
Howe and Hawk ('15) found no striking changes in the stomach and intestines. 
The parietal cells of the gastric glands remain conspicuous. The glandular 
and surface epithelial cells in general stain poorly, with absence of cytoplasmic 
granulation: the nuclei migrate toward the basement membrane. Leukocytes 
often invade the mucous and submucous layers, either diffusely or in masses 
like solitary nodes. 

In adult albino rats, underfed or subjected to acute inanition (water only), 
with loss of 33-36 per cent in body weight, Jackson ('15) found an average loss 
of 57 per cent in the weight of the stomach and intestines (with mesentery) 
(Table 4). In younger rats underfed for various periods, the loss in gastro- 
intestinal weight was much less, and in those held at constant body weight by 
underfeeding from 3 to 10 weeks of age, there was even an increase of 28 percent. 
Stewart ('18, '19) discovered that in albino rats underfed from birth the increase 
in gastrointestinal weight may be even greater, reaching a maximum of 100 
per cent (Table 4) . In the offspring of severely underfed pregnant female albino 
rats, however, Barry ('20, '21) found the stomach and intestines nearly normal 
in weight. 

In albino rats refed after underfeeding from 3 to 12 weeks of age, Stewart 
('18) found that the weight of the empty alimentary tract became nearly normal 
within 4 weeks. Similar results were obtained by Jackson and Stewart ('19) 
in rats underfed from birth for various periods and then refed to body weights 
of 25, 50 and 75 g. (Table 7). In another series of rats refed to adult condi- 
tion, Jackson and Stewart ('20) found the alimentary canal even above normal 
weight, excepting the severely stunted group, in which it was practically normal 
(Table 8). 

Sundwall ('17) in the stomach of a starved albino rat (on water only) 
described intense congestion of the cardiac region, with complete loss of the sur- 

3 o8 


face epithelium and gastric glands. The persistent connective tissue stroma 
was covered with fibrin, and necrotic areas were noted. 

Moehl ('22) observed frequent gastrointestinal disorders in underfed horses. 






(.' a 

- $-■ 



• ■ 



: '■ 

mm ~~ t * 

cm- 1 '"- 

5: — -' 





, 5.m. 



R5. 64 


Figs. 82 to 84. — From sections through the pyloric region of albino rats. Bouin's fixation; 
hematoxylin-eosin stain. X50. l.m., longitudinal muscle layer; cm., circular muscle; s.m., 
submucosa; m.m., muscularis mucosae; m.p., membrana propria; gl.g., gastric glands; p.c, 
parietal cell; su.e., surface epithelium; a, necrotic surface membrane; ed., edemic spaces. 
(Miller '23.) 

Fig. 82 shows a section of the pyloric region in a normal initial control rat 11 days old; 
body weight, 9.5 g; stomach weight, 0.0569 g. Fig. 83 represents the stomach struc- 
ture in a final (age) control, full-fed to 37 days of age; body weight, 68 grams; stomach weight, 
0.541 g. Fig. 84 shows the pyloric stomach in a rat underfed from birth to 39 days of age; 
body weight 9.66 g; stomach weight, 0.151 g. Note the necrotic surface membrane 
(a), the atrophic glands and edemic interglandular spaces (ed), and the pycnotic condition of 
the nuclei in the mucosa. 

Moulton, Trowbridge and Haigh ('22a) found subnormal weight of the 
stomachs in steers on low planes of nutrition. 


Miller ('22) noted that starvation must be severe to produce marked changes 
in the gastrointestinal epithelium of albino rats. In cells with atrophic degenera- 
tion, the mitochondria may be transformed from rod-like to spherical forms, 
with reduction in number or even total absence. 

Miller ('23) has recently made a careful and extensive study of the gastro- 
intestinal tract in albino rats severely underfed from birth for various periods 
(up to 43 days). The tract shows a marked increase in weight (in accordance 
with the observations of Jackson and Stewart), the increase being relatively 
greater in the stomach than in the intestine. Measurements also show an 
increase in the thickness of the mucous and muscular tunics of the stomach. 
Histologically, there appears a variable (usually slight) edema and regressive 
structural changes in the mucosa. There is atrophy and degeneration of the 
surface epithelial cells in restricted areas. The cells show nuclear degeneration, 
cytoplasmic shrinkage, vacuolation, loss of secretory granules, etc. In extreme 
cases, the tunica mucosa may be almost completely necrotic. The gastric tela 
submucosa shows no distinct changes, except occasional hemorrhages. The 
tunica muscularis shows a variable degree of atrophy with degeneration of the 
muscle fibers in certain restricted regions only. The various changes are shown 
by Figs. 82, 83 and 84. 

Ott ('24), in the leopard frog (Rana pipiens) during hibernation and subse- 
quent fasting with loss in body weight up to 60 per cent, found marked varia- 
bility in the weight of the empty gastrointestinal canal. The average loss in 
weight, however, usually appeared relatively much less than that of the whole 
body, reaching a maximum of 53 per cent in the males and 30 per cent in the 
females (Table 6). 

Changes during Digestion and Absorption. — Numerous studies have been 
made of this process, usually involving more or less incidental observations upon 
the changes in the early stages of inanition (a few hours up to a few days), and 
subsequent refeeding. 

Ebstein ('70) in dogs fasting 1-4 days found the pyloric glands composed of 
cylindrical cells with clear (slightly granular) cytoplasm and elliptical nuclei 
basally placed. Changes are apparent 1 or 2 hours after refeeding, reaching 
their maximum in 4 or 5 hours. The gland cells become cubical in form and 
deeply staining; the nuclei become spherical and centrally placed. Heidenhain 
('70) also described changes in the gastric gland cells (chief and parietal) in the 
cat and dog, up to 5 days of fasting. The changes in the herbivora (rabbit, 
guinea pig, sheep) were found less distinct. Similar results were obtained by 
Bentkowsky ('76). 

Theohari ('99) studied the changes in the gastric mucosa of the cat, rabbit, 
guinea pig and especially the dog, in various stages up to 4 or 8 days after feed- 
ing. The parietal cells during fasting lose the vacuoles found in the early stages 
of digestion and the fuchsinophile granules are less numerous. Nuclear and 
granular changes also occur in the chief cells. Stinzing ('99) followed the 
changes in the chief and parietal cells of the dog, up to 1 1 days of inanition. At 
the height of digestion (4 hours after feeding) the chief cells are large, the parietal 
cells small. At the end of digestion (12 hours), the converse is true; the change 


in nuclear size is especially striking. After n days of inanition, both chief and 
parietal cells are enlarged. The parietal cells now often are adjacent to the 
lumen; their nuclei are chromatin-poor and the cytoplasm vacuolated. 

Paira-Mall ('oo) found that in birds the gastric gland cells after fasting 2 
days usually contain more secretory granules (in agreement with Langley). 
The lumina of the compound glands appear distended with secretion during 
digestion, but are smaller and collapsed in the fasting animals. 

Beguin ('02) described the epithelium of the gastric and intestinal mucosa of 
reptiles, in relation to digestion and hunger. Kahle ('13) investigated the 
changes in the gastric glands of the tortoise (Testudo graeca) after various 
periods of fasting and refeeding. There are changes in the cell size, nuclear 
position, and cytoplasmic granules. He cited observations upon the frog by 
Langley and Sewall. 

Jacquet and Jourdanet ('12) described the changes (especially in the ergasto- 
plasm) of the gastric glands in the dog during fasting (2 days) and refeeding. 

Changes in the Migrating Salmon. — These are of especial interest because 
the salmon apparently takes little or no food during its migration up the rivers to 
the breeding places. Miescher-Ruesch ('80) found the stomach and esophagus 
folded and contracted in fish taken at Basel, on the Rhine, in contrast with the 
distended condition in those taken from the sea. A slimy substance with shed 
epithelial cells was noted on the surface of the mucosa. Miescher ('97) 
described the histological changes in the salmon during this period. Stone 
('97) stated that when the Quinnat salmon enters the fresh water, the appetite 
weakens, the throat gradually becomes narrowed and the stomach shrunken, 
so as to become entirely incapacitated for receiving food. Gulland ('98) and 
Paton ('98) described a desquamative catarrh of the gastrointestinal mucosa in 
the migrating salmon. Brown ('98) concluded that this was chiefly a postmor- 
tem change, but described certain other histological changes in the gastric and 
intestinal mucosa. Greene ('10) confirmed the observation of Rutter (Bull. 
U. S. Bureau of Fisheries, 1902, 22:122-) that the digestive tract of the Pacific 
salmon decreases markedly in size during the migratory fasting period. The 
whole question of fasting in the river salmon has recently been reviewed exten- 
sively by Heitz ('18). 

Changes during Hibernation. — Valentin ('57) found an apparent loss of 14.6 
per cent in the weight of the stomach in the marmot after 44 days of hibernation 
(body loss of 8.3 per cent); while the average loss after 166 days was 47 per cent 
(body loss 35.5 per cent). 

In the hibernating hedgehog (Erinaceus europaeus), Carlier ('92) found but 
few changes in the tongue, although the tissues (as elsewhere) appeared to 
stain less intensely and the glands appear inactive. In the stomach, the cardiac 
gland cells appear smaller and more granular. The cells lining the ducts become 
swollen, apparently through accumulation of mucinogen. There is a thin layer 
of mucus, and epithelial debris on the surface of the gastric mucosa. Wandering 
cells (migrated leukocytes) in large numbers infiltrate the stroma of the mucosa 
and submucosa, corresponding to the decrease in blood leukocytes. These 


migrated leukocytes degenerate and are removed by macrophages. Plasma 
cells are scarce in the canal below the esophagus. 

Changes in the gastric glands during hibernation were described also by R. 
and A. Monti ('03) in the marmot, and by Corti (,'03) in bats. Reese ('13, 
'13a) found no significant changes in the alimentary canal (tongue, palate, 
esophagus, stomach, small intestine, rectum) of the alligator after 4 or 5 months 
of hibernation. 

(B) Effects of Partial Inanition on the Stomach 

Comparatively few data are available concerning changes in the stomach 
during the various types of partial inanition. 

McCarrison ('19, '21) has emphasized the atrophic and degenerative changes 
in the alimentary canal of monkeys on autoclaved rice diet (mixed deficiency of 
protein, vitamins, etc.). These include: 

" (a) Congestive, necrotic and inflammatory changes in the mucous 
membrane; sometimes involving the entire tract, sometimes limited to 
certain areas. 

" (b) Degenerative changes in the neuromuscular mechanism of the tract, 
tending to dilatation of the stomach, ballooning of areas of small and large 
bowel, and probably also to intussusception. 

" (c) Degenerative changes in the secretory elements of the tract of the 
gastric glands, the pyloric glands, the glands of Brunner, the glands of Lieber- 
kiihn, and of the mucous glands of the colon. These changes are such as must 
cause grave derangement of digestive and assimilative processes. 

" (d) Toxic absorption of the diseased bowel, as evidenced by the changes in 
the mesenteric glands. 

"(e) Impairment of the protective resources of the gastrointestinal mucosa 
against infecting agents, due to hemorrhagic infiltration, to atrophy of the 
lymphoid cells, and to imperfect production of gastrointestinal juices." 

In persons malnourished through war famine (deficiency in calories and 
vitamins), Reiss ('21) found the stomach usually atonic with hypoacidity. 
(Curschmann found hyperacidity, however.) In autopsies on cases of edema 
disease (due chiefly to protein deficiency), Schittenhelm and Schlecht ('18) 
twice observed marked congestion of the blood vessels in the gastrointestinal 
mucosa, with numerous hemorrhages, especially in the small intestine. 

Sundwall ('17) in monkeys and rats on corn-meal and similar diets (mixed 
deficiency of protein, etc.) found congestion of all layers in the gastrointestinal 
tract, atrophy of the muscle coat, hemorrhages, ulcerations and superficial 
erosions of the epithelium. 

In pellagra, digestive disturbances are prominent, with the usual changes in 
the gastrointestinal canal (Roberts '12; Boyd '20). In lamziekte in cattle, 
Hedinger ('15) described hyperemia of the gastrointestinal mucosa, with enter- 
itis and hemorrhages. 

During chlorine hunger in dogs, gastric hemorrhages were observed by 
Cahn ('86). 


In human rickets, Comby ('01) and others have reported that the stomach 
and intestines are usually dilated. The same was found by Jackson and Carle- 
ton ('22) in albino rats with experimental rickets. The stomach and intestines 
without contents were found subnormal in weight, however (Table 11). 

In human beriberi (vitamin B deficiency), Ellis ('98) noted congestion of the 
stomach in 31 out of 57 autopsies; with blood clots in 4. Strong and Crowell 
('12) found acute gastroduodenitis. Andrews ('12), in infantile beriberi 
(18 necropsies), found the gastric mucosa anemic; duodenum congested; intes- 
tines otherwise normal, or a few petechiae. In experimental polyneuritis of 
fowls and pigeons, Tasawa ('15) observed mild catarrhal inflammation of the 
crop, and distinct atrophy of the stomach (gizzard) muscle. Lumiere ('20a) 
concluded that pigeons on polished rice diet are subjected to starvation, with 
stagnation of the ingested rice in the alimentary canal due to lack of digestive 
secretions. Findlay ('21) noted a loss of 19 per cent in the stomach weight 
in rice-fed fowls, and of 23 per cent in rice-fed pigeons, with slightly larger losses 
during simple inanition (Table 13). 

In human scurvy, Sato and Nambu ('08) observed that the esophageal 
mucosa was intact; the gastric mucosa variable, anemic or congested, with pete- 
chiae in one case. Hess ('20) stated that the scorbutic gastric lesions are usually 
unimportant, with occasional hemorrhages or ulcerations. In experimental 
scurvy in guinea pigs, Bessesen ('23) found the weight of the stomach somewhat 
subnormal in the early stages, but distinctly above normal in the later stages (Table 
12). The intestines appeared similar in weight changes. The gastrointestinal 
contents, however, appeared greatly increased in the early stages of scurvy. 

In a dog on dry diet with loss of 20 per cent in body weight, Falck and Schef- 
fer ('54) noted an apparent loss of 17.9 per cent in the weight of the stomach, and 
the same in the intestine. 

In a similar dog, Pernice and Scagliosi ('95a) found the stomach contracted 
with strongly folded mucosa; pale near the cardia, reddish near the pylorus. 
The gastric mucosa showed numerous small, superficial, rounded erosions or 
ulcerations, containing brownish, hemorrhagic masses. Microscopically the 
stomach showed hyperemia and hemorrhagic infiltrations, with atrophy of 
the glands in some places. Some mitoses (normal or abnormal) were seen in the 
fundus glands. The superficial epithelial cells appeared reduced in size, vacuo- 
lated and poorly stained. Round cell infiltration occurred in the interglandular 
stroma, in the submucosa and in the muscular coat. 

Kudo ('21) noted a loss of about 30 per cent in the empty stomach and 
intestines of albino rats on a dry diet with loss of 30-52 per cent in body weight 
(Table 9). In young rats held at constant body weight for various periods by 
dry diets, Kudo ('21a) found a progressive increase in the weight of the stomach- 
intestines, both empty and with contents (Table 10). This indicates a con- 
tinued growth of the tract similar to that found by Jackson and Stewart in 
young rats during underfeeding. 


Some data concerning changes in the weight and structure of the intestine 
in general have already been mentioned in the foregoing pages in connection 


with the stomach. These include observations by Andrews, Barry, Beguin, 
Bessesen, Bidder and Schmidt, Bizzozero and Vassale, Boyd, Brown, Coen, 
Collard de Martigny, Comby, Falck and Scheffer, Fede, Gulland, Hedinger, 
Heitz, Jackson, Jackson and Carleton, Jackson and Stewart, Kudo, Kusmin, 
Lazareff, McCarrison, Miller, Morgulis, Howe and Hawk, Ott, Quattrochi, 
Reese, Schittenhelm and Schlecht, Sedlmaier, Stewart, Strong and Crowell, 
Sundwall, and Weiske. Further details concerning the changes in the intestines 
will now be considered under (A) effects of total inanition (or on water only), 
and (B) effects of partial inanition. 

(A) Effects of Total Inanition, or on Water Only 

The effects upon the human intestine (adult and infant) will be presented 
first, followed by the data for the infrahuman species. 

Human Adult. — Donovan ('48) in famine victims did not find the intestinal 
inflammation and ulceration described by Duncan, Collard de Martigny and 
others. He observed: "total disappearance of the omentum, and a peculiarly 
thin condition of the small intestines which (in such cases) were so transparent 
that if the deceased had taken any food immediately before death, the contents 
could be seen through the coats of the bowel . . . This condition I look 
upon as the strongest proof of starvation." Fowler ('70), however, maintained 
that in acute starvation there may be no attenuation of the intestinal walls. 
Curran (74) found the intestines thin and transparent, somewhat distended 
with' gas; while Bright ('77) found them empty and collapsed, with greatly 
atrophic walls. In both cases, the omentum was atrophic and fatless. Schult- 
zen ('62, '63) described the large intestine as contracted and nearly empty; the 
mucosa reddish in places. 

Falck ('81) noted that in starvation the intestines are usually found con- 
tracted and nearly empty, containing a slight amount of bile. The mucosa in 
both large and small intestines is usually normal. Casper-Liman ('82) found the 
"Darmtract stellenweise verengert, ganz leer oder hochstens einzelne verhartete 
Kothreste enthaltend, die Haute des Darmkanals bis zur Durchsichtigkeit 

Porter ('89) in numerous autopsies upon victims of the Indian famine noted 
that the peritoneum usually appeared healthy, excepting a variable amount of 
serous effusion in nearly % of the adults and L ^ of the children. The mesentery 
showed more or less fat in nearly half of the adults, but in only \'i of the children. 
A few cases presented edema, thickening and congestion. The duodenum and 
jejunum showed in general an anemic, catarrhal appearance, in some places 
showing loss of the surface epithelium, with congestion, atrophy and pigmenta- 
tion of the stroma. In the ileum, the changes were more extensive; the mucosa 
atrophic, with indistinct villi; the epithelial layer usually absent, and the stroma 
markedly altered, with fatty and granular pigmentary degeneration, probably 
due to extravasated blood. The lymphoid tissue, including solitary glands and 
Peyer's patches, appeared very atrophic and absent over large areas, in children 
as well as in adults. The mucosa of the large intestine likewise usually appeared 


anemic, sometimes showing inflammatory thickening or ulcerated areas. The 
lymph nodes were pale and inconspicuous. The changes characteristic of true 
dysentery rarely occur in "famine dysentery." 

Meyer ('17) found the stomach and intestines only slightly filled with gas, 
and no ulcerations as reported by Formad and Birney ('91). The ascending 
and transverse colons were relaxed; the descending colon firmly contracted; 
the iliac and pelvic colon nearly empty. The appendices coli were completely 
absent and only very small masses of omental fat remained in the intervascular 
areas. Histologically the intestine appeared congested in places, with complete 
disintegration of the mucosa over extensive areas in both small and large intes- 
tines (partly due to postmortem change). Depletion of the solitary and aggre- 
gated lymph nodules, and congestion of the submucous plexuses were also noted. 

Rubner ('19), Determann ('19), Ivanovsky ('23) and others have reported 
an increased occurrence of hernia and intussusception as a result of the malnutri- 
tion and famine during the war. The hernia may be due partly to the intestinal 
atrophy and partly to general weakness of the abdominal walls. Sison ('20) 
could sometimes see and feel the peristaltic movements of the intestines through 
the abdominal wall during voluntary starvation. Hehir ('22) noted severe 
intestinal disorders, resembling dysentery, during the chronic starvation in the 
siege of Kut. 

In malnourished infants, the condition of the intestines has frequently been 
studied. Parrot ('77) held that athrepsia is a condition of malnutrition second- 
ary to gastroenteritis, with "diphtheroid" or ulcerated conditions in the intes- 
tines, similar to those mentioned for the stomach. As a sign of death from 
inanition in the newborn, Tardieu ('80) noted: " Verdauungstractus atrophisch, 
durchscheinend, leer." Ohlmuller ('82) recorded an intestinal weight of 140 g. 
in an atrophic infant of 56 days, with body weight of 2,381 g. (previous weight 
not stated). In a well nourished control of the same age, the intestinal 
weight was 183 g., body weight 4,150 g. 

Blaschko ('83) found no degeneration of the sympathetic plexuses of the 
intestinal wall in atrophic children. Baginsky ('84, '84a) described a marked 
distension and atrophy of the intestinal wall in athreptic infants, including 
partial atrophy of the muscle fibers, degeneration of the plexuses of Auerbach 
and Meissner, atrophy of the lymphoid follicles, and almost complete disappear- 
ance of glands and villi. Baginsky considered these atrophic changes as the 
primary cause of pedatrophy, a conclusion which was opposed by many subse- 
quent investigators. 

Cantalamassa ('92) claimed that in starved infants the colon shows the 
greatest decrease in diameter and in thickness of the walls, in comparison with 
other parts of the intestine. 

Gerlach ('96) demonstrated that the characteristic atrophic appearance of 
the intestinal wall described by previous authors can be produced by simple 
mechanical distension with gases, etc. 

Fede ('97, '98, '00, '01) likewise opposed Baginsky's doctrine of intestinal 
atrophy as the cause of athrepsia. He found a certain degree of thinning in the 
intestinal wall, with other atrophic changes in the glands, etc., but no destruc- 


tion of glands or villi. Later ('01a) Fede stated that the athrepsia of Parrot is 
due to defective alimentation caused by gastrointestinal intoxication, while in 
other cases infantile atrophy with profound malnutrition is secondary to syphilis, 
tuberculosis, etc. 

De Lange ('oo) described marked atrophic intestinal changes in two cases 
of pedatrophy. 

Heubner ('01, '01a) opposed the doctrine of a primary intestinal atrophy 
as the cause of infantile atrophy, ascribing the appearance to physical distension 
of the gut, and to postmortem changes. Nothnagel ('03) found the typical 
atrophic changes in the intestinal wall in 80 per cent of all autopsies, irrespective 
of the causes of death, and interpreted them as due to intestinal catarrh (even 
without clinical symptoms). 

Bloch ('03, '04, '05, '06) described the usual variably atrophic changes in 
the intestine of atrophic infants, and agreed with the previous investigators who 
ascribed the changes primarily to intestinal distension or inflammatory compli- 
cations. In addition, however, he noted a decreased number of Paneth cells 
in the intestinal glands of Lieberkuhn, and thought this might be significant. 

Tugendreich ('04), however, could not confirm Bloch's observation of a 
deficiency in the Paneth cells. After reviewing the various theories as to the 
cause of infantile atrophy, Tugendreich concluded that "Mit der Ablehnung 
der Darmatrophie ist das Wesen der Sauglingsatrophie wieder in tieferes Dunkel 

Thiercelin ('04) described the intestinal lesions at various stages of infantile 
atrophy. The intestine is increased in length, as found by Marfan in all 
infantile chronic dyspepsias. There are progressive but variable atrophic 
changes in the intestinal mucosa, and also in the tunica muscularis, with 
degenerative changes in the plexuses of Auerbach and Meissner. 

Lucien ('08) found that in athrepsia the intestinal tract shows no constant 
changes, and does not often present gross lesions. He doubted that the athrep- 
tic condition is due to preceding gastrointestinal disorder, and ascribed it rather 
to lesions in the hemopoietic system, the kidney and especially the endocrine 
glands. Herter ('08) described a form of growth retardation due to chronic 
intestinal infection and imperfect absorption of food in the intestine. 

Helmholz ('09), like Tugendreich (versus Bloch) found no deficiency in the 
Paneth cells in atrophic infants. He concluded that postmortem changes 
proceed more rapidly than in normal conditions. 

Stickel ('10) described the changes in a malnourished infant and starved 
puppies. Schelble ('10) could find no structural changes in the intestine which 
would account for the condition of pedatrophy, and ascribed it rather to a 
general disturbance in the intermediary metabolism. 

Lesage ('11) described an atrophy in all the elements of the alimentary canal 
of dystrophic infants. Eosinophile cells are said to occur in considerable 
quantities, and to show some peculiarities. The muscle fibers are much 
diminished in diameter (Variot and Ferrand). According to Lesage and Cleret 
('14), in congenital spasmodic atrophy the intestines sometimes show prolifera- 
tion of the submucous connective tissue, but a leukocytic infiltration predomi- 


nates at this level, causing dissociation and atrophy of the glandular cul-de-sacs. 
Sometimes the remnants of the cul-de-sacs are surrounded by abundant 
lymphoid tissue. 

Aschoff ('13, '21) stated: "Die Atrophie der Darmschleimhaut hat friiher, 
besonders in der Padiatrie, eine grosse Rolle gespielt. Seitdem Heubner 
nachgewiesen, dass es sich dabei um einfache postmortale Dehnungsvorgiinge 
handelt, ist die Atrophie in den Hintergrund getreten." 

Mattei ('14) found the intestinal epithelium normal in 14 cases of athrepsia. 
Leukocytic infiltration occurred in the villi and interglandular spaces of the 
mucosa and submucosa. The solitary follicles of the small intestine sometimes 
showed signs of hyperactivity. The literature on the intestinal changes was 
fully reviewed by Nobecourt ('16). In general, he concluded that the changes 
appear to be consequent upon chronic inflammatory processes, especially in 
the small intestine and colon. 

Finkelstein and Meyer ('22) state: "Formerly it was believed that inanition 
was the cause of severe 'atrophy' and was due to interference with food absorp- 
tion in consequence of a chronic inflammation and destruction of the secretory 
mechanism. The foundations of this teaching are today, however, overthrown; 
for the concurrent reports of all observers show that the intestine of the atrophic 
child is anatomically normal. It is clearly a question of functional disturbances, 
leading to a reversive metabolism, recognized by Parrot many decades ago." 

In famine-stricken children of various ages, Nicolaeff ('23) found the small 
intestines increased in length (due to meteorism?), with hyperemia, atrophy of 
the mucosa, and edema of the submucosa. The large intestine showed colitis, 
with frequent necrotic erosions. There was complete disappearance of mesen- 
teric and subperitoneal fat, with large amount of peritoneal fluid in some cases. 

The weights of the empty intestines in atrophic Minnesota infants appear 
relatively large, as shown in Table 3, but the increase is uncertain on account 
of no adequate norm for comparison. 

In animals, the effects of experimental inanition upon the intestine have 
frequently been observed. The general results of the more severe stages of 
inanition will be presented first, followed by a summary of the changes observed 
during the earlier periods following digestion and absorption. Finally a few 
observations during hibernation will be mentioned. 

Tiedemann ('36) noted that in starved animals the entire intestinal canal is 
contracted and contains only a slight amount of bile and mucus. Chossat 
('43) observed in starved pigeons a shortening of the intestine with loss of 42 
per cent in weight, nearly proportional to the loss in body weight. Manassein 
('66, '68, '69) observed that the intestines, especially the small intestine, become 
considerably shortened in rabbits and pigeons during starvation. He observed 
an average loss of 28 per cent in the intestinal weight of 47 adult rabbits (body 
weight loss 39 per cent), but usually less in younger rabbits. Upon refeeding, 
nearly normal length and weight of the intestine were restored. 

Voit ('66) noted an apparent loss of 18 per cent in the empty intestines of a 
starved cat, with loss of 33 per cent in body weight. Bourgeois ('70) found a 
reduction of about one-fourth in length and also a decreased diameter in the 


intestines of various starved animals. The mucosa appeared pale, otherwise 
normal; as likewise noted by Falck ('75) in fasting dogs. 

In the fasting larvae of Rana tigrina and Bufo melanosticus, Cunningham 
('80) found general intestinal atrophy, the lymphoid nuclei being replaced by 
masses of fatty and pigment granules. The changes begin in the upper intes- 
tinal segments. The lining epithelium degenerates first in the middle portion, 
and finally disappears completely. 

Isaew ('87) noted granular degeneration and vacuolation in the sympa- 
thetic ganglion cells of the alimentary canal in starved dogs. Paneth ('88) 
observed persistence of the goblet cells, and also the characteristic Paneth cells, 
in the intestine of the mouse during fasting experiments. 

Hofmeister ('87) found a gradual decrease in the number of mitoses in the 
intestinal lymphoid cells of cats