*
i
John w. SHIVE
. . . Dissertation submitted to the Board of University
Studies of the Johns Hopkins University in
, - conformity with the requirements
for the degree of Doctor of
Philosophy * . . .
PHYSIOLOGICAL RESEARCHES
Station N
Baltimore, Maryland ---
U. S. A.














TITUTITUTyº
# *\
Ém.º.
sº
.
%
:
univERSITYºuth
TNuruu tº
N
-*. - “. . . .
~" – . - 2”
º- ( , ººº-º-º: .
:SJ R ( { } zºº. - -
*...!!!" -
fºr
... : E & --- ‘. .
sº ----. … . --- º: -
't Hºnºut Atº: * - -
- • Tº Tº v.r:- .
*Y*\{}º."
- º
s:
-- Cº-º-º:
... } TºmTºmmºnſtrummitml T
4-
#hiſmiſſiºſimmiſſiſſiſſiſſiſſil;
ºl-
º:
lſº

Science . . "3
* º
J2% 1
t , Jé &
*.…? * +4.
A STUDY OF PHYSIOLOGICAL BALANCE
IN NUTRIENT MEDIA
* x
By
JOHN W. SHIVE
*
~gº ‘i. -
Dissertation submitted to the Board of University
Studies of the Johns Hopkins University in
conformity with the requirements
for the degree of Doctor of
Philosophy
{l? EPRINTED FROM PHY sloi,o GICAL RESEARCHES, Vol. 1, No. 7, Nov 12 MBER, 1915]
{
>
* A STUDY OF PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA
"…S. - JOHN W. SHIVE
º ABSTRACT?
s This publication presents the results of an experimental study of the
physiological balance of the salts in the medium and to the total concentration
W. of the latter. Three different total concentrations were employed, of 0.1,
1.75, and 4.00 atmospheres, in terms of possible osmotic pressure. For
each of these three total concentrations 36 different proportions of the three
component nutrient salts, Ca(NO3)2, MgSO4, and KH2PO4, were simul-
taneously tested. All possible sets of proportions of these three salts were
thus included, for increments of change equal to one-tenth of the total
possible osmotic pressure or diffusion tension. A small amount of ferrio
phosphate was added to each culture solution, to supply iron. There were
36 cultures of 6 plants each, in each concentration series and all three of
these series were carried out simultaneously, and then repeated. The
culture vessels held 250 cc. and the solutions were renewed every three days,
for the total growth period of 23 days which was the same for both triple
series. The plants were measured by five quantitative criteria and by two
more qualitative ones. The quantitative criteria were: (1) dry yield of tops,
(2) dry yield of roots, (3) total water loss by transpiration and guttation dur-
ing the growth period, (4) water requirement per gram of dry tops, and (5)
water requirement per gram of roots. The two more qualitative criteria
were: (1) apparent condition of tops, especially as to two forms of leaf in-
jury, and (2) apparent condition of roots, especially as to lateral branching.
The main results of these tests follow.
1. For young wheat plants within the first four weeks of their growth,
the three salts here employed form a nutrient medium well suited to the
development of the plants, when the total osmotic concentration is about
1.75 atmospheres and when the salts are present in either one of two some-
what similar sets of proportions. One of these two good solutions contains
the salts in the proportions: KHAPO, 0.0180 m; Ca (NO3)3, 0.0052 m; and
MgSO, 0.0150 m. The other has the proportions: KHAPO, 0.0108 m;
Ca(NO3)3, 0.0078 m; and MgSO, 0.0020 m. In both cases about 0.0044
gram of ferric phosphate is added per liter of solution. These solutions
N growth of young winter wheat plants in water cultures, with respect to
1 Botanical contribution from the Johns Hopkins University, No. 46.
2 The manuscript of this paper was received July 1, 1915. This abstract was preprinted, without change,
from those types and was issued as Physiological Researches Preliminary Abstracts, vol. 1, no. 7, Octol er, 1915.
327
PHYSIOLOGICAL RESEARCHES, vol. 1, No. 7
SERIAL No. 7, NOVEMBER, 1915 - ". I () -
328 JOHN W. SHIVE
appear to be nearly equal physiologically; but the first is probably a little
better than the second for these plants. The second produced a yield of tops
2 per cent less than the yield given by the first, and a yield of roots 1 per
cent less.
2. Judging from the yields of tops and of roots, these solutions are equal
to any and are superior to most of the nutrient media previously described
by other investigators and now in general use for water culture work, at
least when these media are prepared with a total concentration of 1.75
atmospheres.
3. For any given total concentration the best physiological balance of
Salt proportions for tops is not the same as that for roots.
4. For any given total concentration one set of salt proportions may be
found which produces a higher yield of wheat tops than does any other set,
and the same is true regarding root yields, but these two sets of salt pro-
portions are different.
5. For any given set of salt proportions the total concentration of the
medium determines the growth of the plants. It is quite impossible to
Select any single set of salt proportions which always produces either very
low or very high relative yields throughout the three total concentrations
here employed. The relative physiological value of any given set of salt
proportions unquestionably varies, in general, with the total concentration
of the medium, but no general rule can yet be formulated to express the man-
ner of this variation.
6. The actual values of the top yields, for each set of salt proportions,
vary with the total concentrations here employed, but not in the same order;
the optimal concentration (1.75 atm.) always gives the highest yield, the
Supra-optimal (4.00 atm.) gives the medium yield, and the sub-optimal
(0.01 atm.) gives the lowest yield.
7. The values of the cation atomic ratios #. *. and ‘. appear to
determine the growth of the plants in many cases, but this relation is not al-
ways clear for any single ratio; it generally requires two of these ratio values
to determine the physiological properties of the nutrient solutions with any
given total concentration. The cation atomic proportions characterizing
the best physiological balance, as here brought out, vary markedly, and in
no simple manner, with the total concentration of the solution.
8. For total concentrations of 1.75 and 4.0 atmospheres serious disturbance
of physiological balance manifests itself in more or less pronounced injury
to the leaves, the conditions determining this injury being apparently re-
lated to the relative proportions of the three salts rather than to the osmotic
partial concentration of any one salt.
9. Characteristic pathological conditions in the roots are also observed
in those cultures where leaf injury is severe.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 329
*º
10. With total concentrations of 1.75 and 4.00 atmospheres the highest
yields of tops were produced by cultures showing slight leaf injury and these
Same plants usually produced apparently more vigorous root systems than
any others of their respective series, but this apparent correlation does not
hold, either directly or inversely, for dry yields of roots.
11. The amount of water lost by the plants during the period of their
growth appears, in a general way, to give a valuable indication of the top
yields, low transpiration generally corresponding to low yields of tops and
high transpiration to high yields. This generalization is especially true for
the optimal and supra-optimal total concentrations of the medium (1.75 and
4.00 atm.).
12. While no general relation is manifest between the actual amounts
of transpiration and the sub-optimal and optimal total concentrations, the
culture of the supra-optimal series all agree in giving lower transpiration
values than are given by any of the cultures of either the sub-optimal or opti-
mal series.
13, For each set of salt proportions, the water requirement per gram of
dry tops varies with the total concentration of the medium, in the reverse
order; the water requirement for tops is highest for the sub-optimal concen-
tration and lowest for the supra-optimal.
14. There appears to be no clear general relation between water require-
ment per gram of roots and total concentration, for the sub-optimal or opti-
mal total concentrations. For almost every set of Salt proportions this
water requirement is markedly lower, however, for the supra-optimal total
concentration than for either the optimal or sub-optimal.
INTRODUCTION
A culture solution may influence a plant growing in it in two ways, cor-
responding to the two entirely different sets of properties possessed by Solu-
tions in general. The chemical properties of the solution may affect the
plant in one way, the physical properties in a very different way, and both
sets of properties may have an active influence upon the growth of the
plant at the same time. Perhaps the physical property of a culture solution
most apt to influence plant growth is its osmotic concentration, or diffusion
tension, by virtue of which it may tend to resist water absorption by plant
rootS.
The chemical properties of complex nutrient solutions vary widely accord-
ing to the nature and proportions of the solutes present. On the other hand,
the osmotic properties of a solution containing a given set of Solutes in
given proportions, in so far as these properties are known to affect plants,
are dependent upon the total concentration and upon the extent to which
the component solutes are ionized in the solution. The chemical properties
and the physical properties may produce the same effect upon an organ-
330 JoHN W. SHIVE
ism, but in very different ways. For example, if the roots of a healthy
Seedling are plunged into a dilute solution of potassium cyanide, death
results almost immediately. It is clear that this is a chemical effect,
and that Osmotic action can have no influence in bringing about the
death of the plant; the solution is of very low osmotic concentration
but the solute is highly poisonous. On the other hand, if the roots
of such a seedling as that just mentioned be plunged into a molecular
solution of cane sugar or glucose, death also results. Here it appears that
the injurious action is primarily due to osmotic extraction of water from the
roots; it is not due to the entrance of a poison into the organism. When such
a non-poisonous solution is not sufficiently concentrated to bring about
death through water removal, it may still retard growth and other forms of
plant activity, and even bring about death, by hindering water absorption.
In such a case the osmotic force resists water entrance instead of causing
actual water exit. It is a fact of observation that plants grow less rapidly
in more concentrated culture solutions (without toxic effects) than they do
in solutions less concentrated but composed of the same salts in the same
relative proportions. Changes in growth resulting from alterations in the
total concentration of the surrounding medium seem to be dependent upon
the ease with which water may be taken into the cells;” this is primarily an
osmotic relation. A culture solution of high total concentration offers
greater resistance to water absorption by plant roots than does a similar
solution of lower concentration, and if the medium be sufficiently concen-
trated to overcome completely the absorbing power of the roots, death of the
plant must soon follow from lack of water. Loeb' and others have shown
that the osmotic extraction of water from an organism by the solution
surrounding it may sometimes bring about the same effect as does a chemical
change, but chemical and osmotic changes of the medium are not nearly
always followed by similar responses.
In studying the chemical influence upon the plant, of the nutrient solu-
tion in which it is grown, it is therefore desirable to control, as far as possible,
the osmotic influence of the medium as well as its chemical nature. This
has been emphasized in the recent work of Tottingham." To vary the
chemical nature of the medium in which a plant is rooted, and infer that
the resulting physiological reactions are due to the chemical changes in the
surroundings, is not logically possible unless the osmotic properties of the
8 See.in this connection:
Livingston, B. E., On the nature of the stimulus which causes the change of form in polymorphic green
Algae. Bot. Gaz. 30: 289–317. 1900.
The rôle of diffusion and osmotic pressure in plants. Chicago, 1903. Pages 124–144.
Osmotic pressure and related forces as environmental factors. Plant World 16: 165–176. 1913.
* Loeb, J., On the nature of the process of fertilization and the artificial production of normal larvae (Plutei)
from the unfertilized egg of the sea urchin. Amer. Jour. Physiol. 3: 135–138. 1899.
* Tottingham, W. E. A quantitative chemical and physiological study of nutrient solutions for plant cul-
tures. Physiol. Res. 1: 133–245. 1914.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 331
f
solution are known not to have been seriously changed when the chemical
properties were altered.
The general problem of the salt requirements of plants still offers many
opportunities for constructive research, and nutrient solutions will surely
have to be employed in experiments bearing on this problem, for salts and
salt combinations can not be supplied to growing plants excepting in aque-
ous solution. It therefore follows that nutrient solutions and their effect
upon plant growth must be well understood before very satisfactory advances
in our knowledge of salt requirements may be expected. Now, in testing
the effect of culture solutions upon plants growing in them, it is of course
necessary to compare the growth-rate of these plants with that of others
grown in some other solution used as a standard of comparison. The use
of distilled water as such a standard is generally possible only with very short
periods of growth, for plants in distilled water soon become unhealthy and
eventually die. Thus while the growth of seedlings in distilled water may be
logically compared with their growth in a nutrient solution for the first few
days of their growth, it is clearly useless to compare the amounts of growth
obtained in these two media during a period of, say, a month; plants in dis-
tilled water are frequently dead within two or three weeks and for the re-
mainder of the observation period the dead cultures can furnish no better
criterion for growth comparison than can an empty bottle. Even for short
growth periods, the use of distilled water as a standard for comparisons has
not remained unchallenged. That it is not suitable for such use is obvious,
in the first place, because it removes salts from the plants. This phase of
the matter has been dealt with by True and Bartlett” and by True.” Fur-
thermore, as has been shown by many writers,” distilled water, unless pre-
pared with exceptional care, appears always to contain small quantities of
dissolved poisonous substances, and the resulting toxicity varies in a way
not to be readily measured or controlled. :
It thus appears that the standard medium required for studies of the salt-
nutrition of ordinary plants must be a more or less complex Salt solution.
This has been somewhat tacitly recognized by many authors, and nutrient
solutions for plant cultures have been the subject of many papers in the
literature of plant physiology. The results of these studies have been so
well brought together and digested in the first part of Tottingham's valuable
paper on culture media that only very brief historical consideration is here
needed.
° True, R. H., and Bartlett, H. H. Absorption and excretion of salts by roots, as influenced by concentration
and composition of culture solutions. U. S. Dept. Agric. Bur. Plant Ind. Bull. 231. 1912.
7 True, R. H., Harmful action of distilled water, Amer. Jour. Bot. 1: 255–273. 1914.
8 For literature in this connection, see:
Livingston, B. E., Further studies on the properties of an unproductive soil. U. S. Dept. Agric. Bur. Soils
Bull, 36. 1907,
Hoyt, W. D., Some toxic and antitoxic effects in cultures of Spirogyra. Bull. Torr. Bot. Club 40: 333–360.
1913.
332 JOHN W. SHIVE
The first standard nutrient solution for plants was proposed by Sachs”
in 1860. Since then a large number of different solutions have been recom-
mended as producing good growth of various kinds of plants. Of these, the
one most generally employed at the present time is perhaps that of Knop.”
The medium now generally known as Knop's solution contains by weight
four parts of calcium nitrate and one part each of potassium nitrate, magne-
sium sulphate, and either mono- or di-potassium phosphate.” A trace of
iron is added.
Tottingham has demonstrated that Knop's solution may be markedly
improved for the growth of young wheat plants by altering the proportions
of the four main component salts, and it seems probable, or at least possible,
that many or all the standard nutrient solutions thus far proposed might be
greatly improved by similar modifications. The modified Knop's solution
devised by Tottingham gives a considerably greater growth of wheat seed-
lings than does Knop's solution itself. Certainly, the plan adopted by Tot-
tingham in working out the salt proportions of his nutrient solution, is the
most logical and complete method which has appeared in the literature up to
the present time, and the solution which he recommends for young wheat
plants, based upon such thoroughgoing experimentation as it is, must be
regarded as the nearest approach to a standard nutrient medium so far
described.
Tottingham employed eighty-four different solutions, all of approximately
the same total osmotic concentration, but each solution differing from all
the others in its proportions of the four main salts, mono-potassium phosphate,
potassium nitrate, calcium nitrate, and magnesium sulphate. Each solu-
tion also contained the usual trace of iron, as ferric phosphate, the amount
of this salt being the same in all cases. Tests of these eighty-four different
sets of proportions of the main nutrient salts, with total osmotic concen-
trations of 2.50 atmospheres (about the optimum total concentration for
young wheat plants) showed that the best solution for the growth of tops
during the first four weeks after germination contained the salts in the
following volume-molecular proportions: mono-potassium phosphate, 0.0130
m; potassium nitrate, 0.0049 m; calcium nitrate, 0.0144 m; magnesium sul-
phate, 0.0145 m. This solution showed an improvement of 11 per cent over
Rnop's Solution of the same osmotic concentration, the percentage being
calculated on the basis of dry weight of tops grown in the latter solution.
The present writer has repeated the tests of Tottingham's optimal series of
* Sachs, J., Vegetationsversuche mit Ausschluss des Bodens tiber die Nährstoffe und sonstigen Ernahrungs.
bedingungen von Mais, Bohnen und anderen Pflanzen. Landw. Versuchsst. 2: 219–268. 1860.
* Knop, W., Quantitative-analytische Arbeiten über den Ernährungsprocess der Pflanzen. II. Landw. Ver.
suchsst. 4: 173–187. 1862.
* The di-potassium salt seems to be more frequently met with in the literature than is the mono-potassium
Salt, in connection with Knop's solution, but the latter is by far the more satisfactory of the two, on both chemical
and physiological grounds, as has been clearly and convincingly shown by Tottingham.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 333
eighty-four Solutions as compared to Knop's solution (all of them having
the same Osmotic concentration, 2.50 atmospheres), with wheat, and the
results are in very satisfactory agreement with those obtained by the earlier
author. The best growth of tops (measured in terms of dry weight) occurred
with the same proportions of salts as those giving the best growth in the
earlier work. While Tottingham reports, for his best solution, an improve-
ment of 11 per cent over Knop's proportions, the corresponding improvement
in the repeated experiment was 12 per cent.
From a consideration of Tottingham's work, as well as from earlier publi-
cations, it is clear that nutrient solutions prepared according to the formulas
of Knop and Tottingham are very suitable for supporting plant growth, and
Tottingham's Solution may safely be employed as a standard for culture
comparisons. However, in all experimentation bearing on the growth of
plants in nutrient solutions, and related to the general problems of salt
requirements, physiological balance, and salt antagonism, it is quite essential
that the standard nutrient solution used as a basis of comparison be as simple
as possible, besides being capable of producing excellent growth. What-
ever salts the standard solution may contain, it should be characterized by
approximately optimum Salt proportions and, at the same time, should have
approximately optimum osmotic concentration for the particular kind of
plant upon which the experimentation is to be carried out. Since it may be
supposed that different plant forms may differ in their requirements for a
physiologically balanced solution, it is clear that such a chemically and
physically optimum solution can not be satisfactorily determined by the
usual hit-or-miss procedure in such cases (calling that optimum which has
been shown only to be good), but must rest upon a more or less thorough
testing of a rather large number of possibilities. Tottingham's general
method for finding the optimum solution for a given set of salts and for a
given plant is the only one available, and, although it may appear somewhat
elaborate and tedious in operation, yet its employment is quite essential
if the physiological properties of the medium in which growth is to be taken
as standard are to be at all well understood. It thus appears that studies
on salt nutrition will frequently have to involve some such series of tests as
those employed by Tottingham. In such work, the simpler the mixture
may be, the more readily may its optimum, as regards salt proportions and
osmotic concentration, be determined. If a simple series of salts may be so
brought together as to give as good plant growth as does the optimal mixture
of a more complex series, the former is much to be preferred.
The experimentation here to be set forth aimed primarily to find out
whether a simpler solution than that used by Tottingham might not be pos-
sible, without loss of ability to support plant growth. Possible simplification
lies here only in the employment of a smaller number of salts. Tottingham
used four main salts in eighty-four different combinations for each osmotic
334 JOHN W. SHIVE
concentration tested. If the number of salts might be reduced to three,
a similar Series of tests (increments of partial concentrations being, as in
Tottingham's work, one-tenth of the total concentration), would involve
only thirty-six different combinations, which would entail a saving of work
greatly to the advantage of rapid progress. If a two-salt solution might be
employed, such a series of tests would include but nine combinations, but
such a solution is clearly impossible in the present case, since six different
chemical elements (K, Ca, Mg, N, S, and P) are required in every adequate
nutrient medium, and only two of these can be carried by a single salt. It
appears, then, that a three-salt mixture (neglecting the trace of iron, which is
of course always to be added) is logically possible as a standard culture solu-
tion, and that such a standard solution containing but two salts is impossible
on a priori grounds.
The omission of potassium nitrate from the Knop-Tottingham formula
leaves the three Salts, mono-potassium phosphate, calcium nitrate, and
magnesium sulphate, which, with the trace of ferric phosphate also present,
contain all the essential elements required in a nutrient solution. These
dissociate in solution to form all the ions present in the Knop-Tottingham
four-salt solution. They do not precipitate too readily when mixed in
solution, and they permit total concentrations suitable for plant growth.
Such a three-salt solution seems to be chemically as well adapted to the
growth of plants in water culture as is the Knop-Tottingham four-salt
solution. -
It was the purpose of this investigation to study the properties and possi-
bilities of the three-salt combination just suggested, and to determine, as
nearly as possible, the salt proportions needed to produce approximately
optimal growth of wheat and of buckwheat, during early stages of their
development. The results obtained with wheat will alone be presented
in the present paper; those for buckwheat will be published later.”
The investigation was carried out at the Laboratory of Plant Physiology
of the Johns Hopkins University. It is a pleasure to acknowledge indebted-
ness to Professor Burton E. Livingston, not only for suggesting the problem,
but also for helpful advice during the progress of the work, and for valued
aid and criticism in the preparation of the manuscript.
MATERIALS AND METHODS
THE STOCK SOLUTIONS OF SINGLE SALTS
The salts used in the present work were Merck’s “highest purity” mono-
potassium phosphate and calcium nitrate, and Merck’s “blue label” mag-
nesium sulphate, the last containing the usual seven molecules of water.
*** A preliminary report of some of the results of the present study has already appeared:—Shive, J. W., A
three-salt nutrient solution for plants. Amer. Jour. Bot. 2 : 157-160. 1915.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 335
It was found necessary to dry the calcium nitrate, because of varying amounts
of water in the lumps of salt. This was accomplished by heating the salt
in a large crystallizing dish in an electric oven at 150°C., until it formed a
solid mass. It was then pulverized in a mortar, returned to the oven, and
dried to constant weight at 150°C. Any free moisture contained in the stock
supply of the other salts was eliminated by drying them at 105°C. and allow-
ing them to cool in a desiccator, from which they were taken as needed.
In preparing the original stock solutions, from which, after further dilution,
the culture solutions were to be made up, gram-molecular weights of the salts
(weighed to one milligram) were dissolved separately in Jena glass flasks,
and each solution was then made up to liter volume at 15°C. These solutions
were always stored in the flasks in which they were prepared, but were never
kept for a long time; in no case longer than six days. The distilled water
used throughout was obtained from a “Barnstead” still, with tin-lined con-
denser, and was stored in a twenty-gallon stone-ware jar, from which it was
drawn through a block tin tube into a glass carboy for immediate use.
Throughout this work, special precautions were taken to clean thoroughly
(with potassium bichromate in sulphuric acid followed by thorough rinsing)
all vessels employed as containers of distilled water or of Solutions.
For immediate use the volume-molecular single salt solutions were diluted
to concentrations better suited to the preparation of the culture media
themselves. These diluted stock solutions were kept in liter flasks. Each
of these flasks was connected with the upper end of a 50 cc. burette, graduated
to 0.1 cc. and provided with a glass stopcock below. The connection was
made by means of a glass tube bent into the form of an inverted U, one end
extending to the bottom of the flask, the other passing through a tightly
fitting stopper in the top of the burette and extending downward just to the
zero mark on the burette scale. A second glass tube was also passed through
the stopper in the top of the burette, and ended just below the lower surface
of this stopper. Suction applied to this tube caused the solution to flow
from the flask into the burette, thus filling the latter. Since the longer arm
of the U-tube extended to the bottom of the flask, and since the zero of the
burette scale was always at a higher level than the surface of the liquid in the
flask, any excess of solution in the burette siphoned back into the flask
when suction was discontinued, leaving the burette filled to the Zero mark.
In this way the solutions came into contact with glass surfaces only, and
this device for automatically obtaining the proper level in the burette after
each measurement was found to be a valuable saver of time. The con-
centrations of these diluted stock solutions were so calculated that no less
than 2 cc. of any one was ever drawn from the burette at one time, in the
preparation of the culture solutions; errors that might result from slight
inaccuracies in reading burettes graduated only to 0.1 cc. were thus largely
obviated.
336 JOHN W. SHIVE
THE CULTURE SOLUTIONS
In a series of solutions having three given constituent salts, the relative
proportions of these three salts may, of course, be varied so as to give a very
large number of solutions, all of the same total concentration; and the total
concentration of a culture solution with any given set of proportions of the
three salts may likewise be varied to give another large number of solutions,
differing neither in the kind of salts nor in their relative proportions, but
only in the total concentration.
Throughout this paper, concentrations will be expressed in terms of diffu-
sion tension or maximum osmotic pressure (osmotic concentration), in
atmospheres, or in terms of gram-molecules per liter of solution (volume-
molecular concentration).
Three series of nutrient solutions were employed, each solution contain-
ing the three salts, mono-potassium phosphate, calcium nitrate, and mag-
nesium sulphate, in varying proportions. These three series differed widely
in total concentration, and were so chosen that one possessed a total con-
centration about optimal for wheat as here grown, one of the others being
below, and one above this total concentration. All solutions of the same
series had approximately the same osmotic concentration. Tottingham
called attention to the fact that the concentration employed for his optimal
series (2.50 atmospheres of diffusion tension) may be slightly above that
required for optimal growth of wheat, and the results of preliminary experi-
ments with three-salt mixtures, in series such as are to be described below,
showed that a total salt concentration of 1.75 atmospheres is well within the
range required for optimal growth of wheat. These tests showed further
that a concentration or 4.0 atmospheres is well above the optimal range for
wheat. The following experiments were therefore carried out with three
series of nutrient solutions having concentrations of 0.1, 1.75, and 4.0 atmos-
pheres, the first being chosen as having been shown by preliminary tests
to lie well below the optimal range of total concentration for the plant here
employed. These three series will hereafter be termed sub-optimal (0.1 atm.),
optimal (1.75 atm.), and supra-optimal (4.0 atm.).
The method here adopted to control the osmotic concentrations in a series
of solutions differing in the proportions of the three salts used was the same
as that employed by Tottingham. In varying the relative proportions of
the component salts apportionments were made in such a way that a de-
crease in the partial concentration of one salt was balanced by increases in
the partial concentrations of the remaining salts, sufficient to keep the total
osmotic concentration of the solution constant. The total Osmotic con-
centration to be employed for any series was considered to be divided into
ten equal parts and these parts were distributed among the three salts in all
possible proportions. Thus each salt produced, in the different solutions
of the same series, from one to eight-tenths of the total osmotic concentration.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 337
*
A detailed discussion of the methods of calculation by which the partial
diffusion tension and volume-molecular concentration of each salt, for such
complex mixtures with a fixed total Osmotic concentration, may be approxi-
mately determined, has been presented by Tottingham (pp. 177–182 and 192)
and need not be repeated here. The calculations were carried out in such a
manner as to allow every three-salt Solution to contain that particular amount
of each salt which would have been required to produce the desired partial
Osmotic concentration if that salt had been alone in the solution. Each
Solution contained an amount of each of the three salts such that the total
osmotic concentration should possess the fixed magnitude required by the
particular series to which the solution belonged, if it were assumed that the
presence of the remaining two salts exerted no influence upon its degree of
ionization. Expressed in another way, each of the three salts may be con-
ceived as placed in the solution of the other two as though that solution were
pure water, so as to give the requisite Osmotic concentration for the first salt
(one, two, etc., tenths of the total diffusion tension of the complete solution)
without any reference to the presence of the other two.
The lowering of the freezing-point was determined for all of the three-salt
solutions here employed, and the osmotic concentration was calculated from
the lowering in each case. The method here used was the same as that
described by the writer in connection with his study of Tottingham's solu-
tions.” From these determinations it appeared that the lowerings were
approximately the same for all of the complete solutions of the sub-optimal
and optimal series. It therefore follows, in these cases, from the relation
holding between the lowering of the freezing-point and maximum possible
Osmotic pressure, that the amount of salt required to produce one-tenth of
the total osmotic concentration of the complete solution was also constant
for the sub-optimal and for the optimal series. This amount may be termed
the osmotic unit for the series in question, following Tottingham's similar
usage. In the supra-optimal series, however, the value of this osmotic unit
was variable, becoming greater or less in magnitude as the partial osmotic
concentration of the salt in question was increased or decreased. This, of
course, is to be expected in such solutions as these, where the concentration
of the salt is too high to permit approximately complete dissociation.
The volume-molecular partial concentrations of the solutions in the supra-
optimal series were calculated to give a total Osmotic concentration of 4.50
atmospheres, but the cryoscopic tests gave depressions of the freezing-point
indicating a total osmotic concentration of only about 4.0 atmospheres in
every case, throughout the entire series. This seems to indicate that the
decrease in the degree of dissociation caused by the influence of one salt
upon another is considerable in these more concentrated mixtures. With the
12 Shive, J. W., The freezing points of Tottingham's nutrient solutions. Plant World 17: 345-353. 1914.
In table I, page 349, of this paper, the second number in the fifth column should be 2.53 instead of 2.40 as given.
338 t JOHN W. SHIVE
lower concentrations, sub-optimal and optimal, this influence is lacking, or
is too slight to be detected by the methods here employed. Tottingham's
supra-optimal series of nutrient solutions showed a similar depression in
total Osmotic concentration when this was determined by the cryoscopic
method (Shive [1914]). The average diffusion tension of that entire series
was 7.22 atmospheres, while Tottingham's calculation placed this value at
8.15 atmospheres.
The subject of dissociation equilibrium in solutions of two salts possessing
a common ion is treated in text-books of physical chemistry, but the influence
of one salt upon the dissociation of others in complex solutions where no com-
mon ion is present has not yet been worked out satisfactorily. Mellor” points
out that there is no interchange of ions in mixed solutions of strongly dis-
sociated electrolytes, but that the solutions contain the same ions as were
previously present in the unmixed solutions. With feebly dissociated electro-
lytes the phenomenon is much more complex, and is not yet understood well
enough to allow satisfactory calculation.
The results of the cryoscopic tests justify the supposition that all the
solutions of the sub-optimal series possessed a total osmotic concentration
of 0.1 atmosphere, and, similarly, that those of the optimal series had a total
concentration of 1.75 atmospheres, within the limits of accuracy imposed by
the cryoscopic method as here employed. Since the solutions of the supra-
optimal series agreed in showing a lowering of the freezing-point correspond-
ing to a diffusion tension of 4.0 atmospheres, instead of 4.50 as calculated,
it is safely to be supposed that these solutions actually exhibited an osmotic
concentration very close to 4.0 atmospheres. In default of more precise
knowledge, these suppositions will be regarded as true in the following dis-
cussions. Whatever may be the magnitude of the errors thus involved, it is
certain that such variations from the assumed osmotic concentration can not
be great enough in any case to produce any sensible influence upon the plants;
such organisms are much less sensitive to osmotic conditions of the medium
than is the cryoscopic method of determination as here used.
Table I gives the volume-molecular partial concentrations of each salt
required to produce from one-tenth to eight-tenths of the total osmotic
concentration, for each of the three series of solutions here considered. To
determine the volume-molecular partial concentration of any salt in any
solution of the three series, it is necessary simply to find in the first column
the number indicating how many tenths of the total osmotic concentration
are to be apportioned to the particular salt in question, and then to read the
required partial volume-molecular concentration, given opposite this number
on the same line, in the proper column for the particular series and salt
involved.
18 Mellor, J. W., Chemical statics and dynamics. - London, 1909. Pages 202–205.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 339
For convenience in the preparation of the three series of complete nutrient
Solutions to be described later, the concentrations of the stock solutions from
which the culture solutions were prepared differed widely in the three series,
but the three separate stock solutions used in the preparation of each single
Series were equal in molecular concentration. The concentrations of the
stock solutions used in the preparation of the sub-optimal, optimal, and supra-
optimal series were one-fortieth volume-molecular (m/40), one-fourth volume-
molecular (m/4), and volume-molecular (m), respectively.
In preparing a culture Solution, the required amount of stock solution of
mono-potassium phosphate was first drawn from the proper burette into a
250 cc. volumetric flask about three-fourths filled with water. To this was
now added, in a similar manner, the required amounts of the stock solutions
of calcium nitrate and of magnesium sulphate, in the order here given. As
TABLE I
Volume-molecular partial concentrations of mono-potassium phosphate, calcium nitrate,
and magnesium sulphate required to produce from one to eight tenths of the total osmotic
concentration, for each of the three series of solutions
Š § 3 SUB-OPTIMAL SERIES OPTIMAL SERIES SUPRA-OPTIMAL SERIES
Ö E. (ToTAL concLNTRATION (TOTAL CONCENTRATION (TOTAL concLNTRATION
* - 3 0.1 ATM.) 1.75 ATM.) 4.0 ATM.)
#: ;
: #é KH:PO, Ca(NO), MgSO, KHPO, Ca(NOs). MgSO, KH:PO, Ca(NO3), MgSO.
1 0.0002 || 0.00015 0.0003 || 0.0036 0.0026 0.0050 0.0090 0.0066 0.0142
2 0.0004 || 0.00030|0.0006 || 0.0072 0.0052 0.0100 0.0180 0.0132 || 0 .0258
3 0.0006 || 0 00045| 0.0009 || 0.0108 0.0078 0.0150 || 0.0270 || 0.0189 || 0.0428
4 0.0008 || 0.00059| 0.0011 || 0.0144 || 0.0104 || 0.0200 || 0.0368 0.0265 || 0.0570
5 0.0010 || 0.00074 0.0014 || 0.0180 || 0.0130 || 0.0250 || 0.0460 0.0331 || 0.0713
6 0.0012 || 0.000.89 0.0017 | 0.0216 || 0.0156 || 0.0300 0.0580 || 0.0417 | 0.0890
7 0.0014 || 0.00104 0.0020 | 0.0252 || 0.0182 || 0.0350 | 0.0690 || 0.0496 || 0.1070
8 0.0017 | 0.00119) 0.0023 0.0288 || 0.0208 || 0.0400 || 0.0800 0.0576 || 0.1240


the solutions were added, the flask was shaken to promote mixing. The
volumetric flask was finally filled to the mark with water.
To each complete solution, as it was prepared, was added three drops of a
uniform suspension of ferric phosphate in distilled water. Each cubic
centimeter of this suspension contained approximately 0.0022 g. of ferric
phosphate, or 0.00082 g. of iron. The ferric phosphate was prepared by
precipitation from a solution of ferric nitrate with mono-potassium phos-
phate, and the precipitate was subsequently thoroughly washed.
Variations in the proportions of the three salts here employed, by incre-
ments of one-tenth of the total osmotic concentration, produce a series of
thirty-six solutions, as has been remarked. Similar variations in the four
salts used by Tottingham yield a series of eighty-four solutions. The Smaller
number of solutions is a distinct advantage in favor of a three-salt medium,_
340 JoHN W. SHIVE
an advantage that can scarcely be overestimated, considering the large
number of cultures to be dealt with in studies of this kind.
For the concrete expression of the relative osmotic proportions of the three
salts in each solution of each series, for facility in designating individual
cultures, and for ease of reference in discussion, the thirty-six solutions may
be conveniently arranged on a triangular diagram, as shown in figure 1.
Graphic schemes similar to this have been used extensively in physical chem-
istry, for considerations involving relative proportions of three component
parts. The diagram as here used was employed by Schreiner and Skinner'4
in a study somewhat similar to the present one. Each side of an equilateral
triangle is divided into seven equal parts, and lines are drawn through each
of the points of division thus obtained, parallel to each of the other two sides
R2 /Cl C2 C3 04 , \/C5 \/C6 C7
Rl 4\/\/\/\ C5 \/G6 \\
One part KH2PO4
FIG. 1. Diagram showing solution numbers and osmotic proportions of the three
salts.
of the triangle. Each of the original points and each of the intersections of
the lines represents a culture Solution or a culture. The diagram shows eight
horizontal rows of cultures, one above the other. The lowest row contains
eight cultures, the next seven, etc., and the eighth row contains but a single
one. All the Solutions represented in the lowest row have approximately
one-tenth of their total Osmotic concentration due to mono-potassium
phosphate, those in the second row have approximately two-tenths due to
this salt, and so on, until the apex of the triangle is reached. This culture,
representing the eighth row, has approximately eight-tenths of its total
* Schreiner, O. and Skinner, J. J., Ratio of phosphate, nitrate, and potassium on absorption and growth,
Bot. Gaz. 50: 1-30. 1910.

PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 341
Osmotic concentration due to mono-potassium phosphate. The first culture
at the left of each row has one-tenth of its total osmotic concentration due to
calcium nitrate, the second culture of each row has two-tenths derived from
this salt, and so on to the opposite side of the triangle. Thus the eight
cultures of the first row have, respectively, from one-tenth to eight-tenths of
their total osmotic concentration due to the calcium salt. The seven cul-
tures of the second row have from one-tenth to seven-tenths of their osmotic
concentration derived from the calcium salt, and so on to the apex of the
triangle. The last culture of each row has one-tenth of its total osmotic
concentration derived from magnesium sulphate. Proceeding from the
right to the left side of the triangle, the partial osmotic concentration of
magnesium sulphate in each solution increases in a manner similar to that
followed for calcium nitrate passing in the opposite direction. Thus the
eight cultures of the first row have from one-tenth to eight-tenths of their
osmotic concentration due to calcium nitrate, and from eight-tenths to one-
tenth produced by magnesium sulphate, and in the cultures of each suc-
ceeding row, the osmotic proportions of these two salts are similarly dis-
tributed. Each side of the diagram may be designated by the name of one
of the three salts; thus the base (as shown in fig. 1) may be considered as the
mono-potassium side, while the left and right sides are, respectively, the
calcium nitrate and the magnesium sulphate sides.
The method adopted by Tottingham (p. 194) to designate individual
cultures is again employed here. The rows of cultures are numbered R1 to
R8, from base to apex of the triangle, and the individual cultures are numbered
in each row from left to right, C1 to C8. Thus the cultures of the first row
are numbered R1C1 to R1C8, the cultures of the second row, R2C1 to R2C7,
etc., until the single culture of the eighth row is numbered R8C1. To de-
termine the osmotic proportions of the salts in any culture Solution, it is only
necessary to observe the row in which it occurs and its number in the row.
Thus, culture R4C2 has four-tenths of its osmotic concentration due to
mono-potassium phosphate, since it is in row four; two-tenths due to calcium
nitrate, since it is culture number two from the left side; and four-tenths
derived from magnesium sulphate, since it is the fourth culture from the
right side of the triangle. -
In table II are given the actual partial volume-molecular concentrations
of each solution of the three series used in the present study. The Solu-
tion numbers in the table correspond to the numbers in the diagram of
figure 1, already explained.
It will be observed that solutions R1C3 to R1C8, also R2C4, R2C5, and
R2C7, of the supra-optimal series, are omitted from this table. These Solu-
tions formed slight precipitates after standing from one to three days, and
were rejected on that account, as not suitable for a study of this kind.
342 JOHN W. SHIVE
TABLE II
Partial volume-molecular concentrations of mono-potassium phosphate, calcium nitrate,
and magnesium sulphate in the solutions employed in the sub-optimal, optimal, and
supra-optimal series.
PARTIAL CONCENTRATION, VOLUME-MOI, ECULAR
SOLUTION
NUMBER Supra-optimal series (4.00 atm.)*
Sub-optimal series (0.1 atm.) Optimal series (1.75 atm.)

KH2PO4 || Ca(NO3)2 KH2PO4 Ca(NO3)2 MgSO4 || KH2PO4 || Ca(NO3)2 | MgSO4
MgSO4
C6
R4C1
C2
C3
C4
C5
R5C1
C2
C3
C4
R6C1
C2
C3
R7C1
C2
R8C1
0.00021
0.00021
0.00021
0.00021
0.00021
0.00021
0.00021
0.00021
0.00041
0.00041
0.00041
0.00041
0.00041
0.00041
0.00041
0.00062
0.00062
0.00062
0.00062
0.00062
0.00062
0.00082
0.000.82
0.000.82
0.000.82
0.00082
0.00103
0 00103
(). 00103
0.00103
0.001.23
0.00123
0.001.23
0.001.44
0.001.44
0.0016.5
0.00015
O .00030
0.00045
0.00059
0,00074
0.00089
0.00104
0.00119
0.00015
0.00030
0.00045
0.00059
0.00074
0.00089
0.00104
0.00015
0.00030
0.00045
0.00059
0.00074
0,00089
0.00015
0.00030
0.00045
0.00059
0.00074
0.00015
0.00030
0.00045
0.00059
0 00015
0.00030
().00045
0.00015
0.00030
0.00015
0.00228
0.00199
0.00171
0 00124
0.00114
0,000.86
0.00057
0.00028
0.00199
0.00171
0.00142
0.00114
0.00086
0.00057
0.00028
0.00171
(). 00142
0.00114
0.00086
0.00057
0.00028
0.00142
0.00114
0.00080
0.00057
0.00028
0.00114
0.00086
0.00057
0.00028
0.000.86
0.00057
0.00028
0.00057
0.00028
0.00028
0.0036
0.0036
0.0036
0.0036
0.0036
0.0036
0.0036
00.036
0.0072
0.0072
0.0072
O.0072
0.0072
0.0072
0.0072
0.0.108
0.0108
0.0108
0.0108
0.0108
0.0108
0.0144
0.0144
0.0144
0.0144
0.0144
0.0180
0.0180
0.0180
0.0180
(). 0216
0.0216
0.0216
0.0252
().0252
(), ()288
0.0026
0.0052
0.0078
0.0104
0.0130
0.0156
0.0182
0.0208
0.0026
0.0052
0.0078
0.0104
0.0130
0.0156
0.0182
0.0026
0.0052
0.0078
0.0104
0.0130
0.0156
().0026
0.0052
0.0078
0.0104
0.0130
0.0026
0.0052
0.0078
(), 0.104
0 0026
().0052
0.0078
0 0026
(). 00:52
0.0026
0.0400
0.0350
0.0300
0.0250
0.0200
0.0150
0.0.100
0.0050
0.0350
0.0300
0.0250
0.0200
0.0150
0.0100
0.0050
0.0300
0.0250
0.0200
0.0150
0.0100
0.0050
0.0250
0.0200
0.0150
0 0100
0.0050
(). 0200
(), ()150
0.0100
0.0050
0.0150
0.0100
0.0050
(), ()10()
(), ()()5()
(). 00:50
& e º a 9 s tº s
• * * * * * * *
• * = < * * * *
0.0580
0.0580
().058()
().0600
0.0600
0.0800
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * g º s # 8 º'
s s e g º a *
& e º s e < * *
s e º a s a # *
0.0265
0.0331
0.04.17
0.0066
0.0132
(), ()198
().0265
0.0331
0.0066
0.0132
0.019.8
0.0265
0.0066
0.01.32
() ()108
0.0066
(), ()132
().0066
s e º 'º e º tº º
& g g g g g g tº
tº º $ tº $ tº e e
0.0428
0.0285
0.0142
0.0428
().0285
(). 0.142
().0285
(). ()142
0.0142
* Eight solutions of the supra-optimal series formed precipitates after standing from one to three days and
are for this reason omitted from the table.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 343
THE PLANT'S AND THEIR TREATMENT
The wheat used was of the variety known as Fulcaster, from the same
lot as was used by Tottingham in 1913, which was supplied by the United
States Department of Agriculture. All the seed used was fairly uniform
in size, and germinated well.
Germination was brought about by a modification of the methods employed
by Livingston,” by Schreiner and Skinner,” by other workers in the United
States Bureau of Soils, and by Tottingham. The seeds were soaked in dis-
tilled water for a period of from six to eight hours and were then placed on
wet blotting paper in a moist chamber for from twelve to fifteen hours, or
until they germinated. Uniform and vigorously germinated seeds were then
Selected, one by one, from the lot, and were placed upon the germination net,
to grow till large enough for the culture bottles.
This net is constructed as follows: A glass rod, 5 mm. in diameter, is bent
to form a square frame, with one end of the rod continued to form a diagonal
across the Square. This frame is placed between two layers of paraffined
mosquito netting, and the whole rests upon the top of a circular “granite-
ware” pan (35 cm. in diameter and 7 cm. deep) with the corners of the frame
resting on the slightly projecting rim. The two pieces of netting are firmly
stretched over the top of the pan and over the frame by binding beneath
the marginal rim of the pan with a heavy cord. The top layer of netting
upon which the germinating seeds are to be placed, is thus raised about 5 mm.
above the top of the pan, while the lower net is nearly parallel to the upper
and about 5 mm. below it. While water is allowed to flow into the pan and
overflow, the water surface comes into contact with the upper net, but does
not completely flood the seeds, which are always in contact with water,
and still freely exposed to the atmosphere.
The pan with the germinatingseeds stoodin agreenhouseroom, and a stream
of tap water was allowed to flow into the pan continuously, through a small-
bore glass tube that penetrated the double net. After three or four days,
this procedure resulted in very uniform and vigorous seedlings about 5.5
cm. tall and entirely free from fungus. Flowing tap water seemed to yield
better results than could be obtained by the use of distilled water, even with
frequent changes, as was, indeed, to be expected, on account of the extraction
or leaching effect always produced by distilled water. The double net is
necessary for buckwheat, but the lower one may be omitted for wheat.
Selected seedlings, from 5 to 6 cm. tall, were mounted in thoroughly
paraffined, flat cork stoppers (in a manner similar to that described by
Tottingham, pages 173–175), which were then placed in the culture
15 Livingston, B. E., A simple method for experiments with water cultures. Plant World 9: 13–16. 1906.
10 Schreiner, O. and Skinner, J. J., Some effects of a harmful organic soil constituent. U. S. Dept. Agric. Bur.
Soils Bull, 70. 1910.
PHYSIOLOGICAL RESEARCHES, VOL. 1, No. 7
SERIAL No. 7, Nov EMBER, 1915
344 JOHN W. SHIVE
bottles. The latter were wide-mouth flint-glass bottles of 250 cc. capacity.
These had been in use as culture vessels for several years, so that the absence
of excessive amounts of soluble matter at the inner surface of the glass was
assured. r
In water culture work, the glass culture vessels are sometimes covered with
black paper or other black material, to exclude light from the roots of the
plants. This practice is particularly objectionable when cultures are ex-
posed to direct sunlight, since the rate of heat absorption by the black mate-
rial is often sufficient to raise the temperature of the culture solutions con-
siderably above that of the surroundings—a complication to be avoided as
far as possible. In the present work, the culture vessels were covered with
shells or jackets of Bristol board, which was dark brown on one side and
nearly white on the other. To make these shells, rectangular pieces of
Bristol board were cut sufficiently long to wrap around the culture bottle,
with some overlapping; the pieces were about 1.5 cm. greater in width than
the height of the bottle. One long edge of this rectangle was folded over
So as to form a projection about 1.5 cm. wide, perpendicular to the remain-
ing part. This projecting portion was slashed across from its free edge,
at intervals of about 1.5 cm., so as to allow the cut edges to overlap when the
main portion was wrapped about the culture bottle. When this cylindrical
shell, which was dark within and almost white without, was in place, the
horizontal upper rim slightly overlapped the edge of the cork stopper in the
mouth of the bottle. Light was thus almost completely excluded from the
roots of the seedlings, and undue absorption of heat was avoided. These
shells were held in place by a cord fastened around the outside; they could
be easily and quickly removed from the culture bottles at any time, and
replaced again without disturbing the cultures. With proper care, such
shells may be used repeatedly.
The seedlings were mounted in the cork stoppers and placed in the culture
bottles as rapidly and as carefully as possible, so that effects incident to hand-
ling should be approximately the same on all the seedlings, and the shock of
transfer should be reduced to a minimum.
In studying the physiological effects of nutrient solutions it is always
recognized as desirable to control all conditions affecting the organisms
excepting those directly dependent upon the properties of the solutions, but
this has never yet been satisfactorily accomplished. If aerial conditions
can not be kept constant, it is clearly essential that all the cultures of a series
be exposed to these conditions in an approximately similar way, so that all
of the cultures shall experience the same variations in the surroundings.
The cultures of a series are usually exposed by being placed side by side in
Some definite arrangement, upon a table or bench, the arrangement being
sometimes changed from time to time. The aerial conditions affecting the
plants may vary widely, however, within the area occupied by such an
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 345
arrangement of cultures, and they do not all experience the same environ-
mental changes. For example, cultures are almost certain to shade each
other, Some thus receiving more light than others, and cultures entirely sur-
rounded by other cultures are undoubtedly subjected to a lower degree of
atmospheric evaporating power than those not so surrounded. These
undesirable inequalities of treatment were avoided in the present work by
placing all the cultures upon two rotating tables, which stood side by side
in the greenhouse.
The rotating tables here employed are constructed in the following manner.
A motor-cycle wheel is mounted in a horizontal position upon a heavy, iron
tripod-base, especially constructed for the purpose and having a radius of
53 cm. The wheel is provided with a special axle, extended at one end
beyond the cone, to form a projecting piece 5 cm. in length and 2.2 cm. in
diameter. This projection is tightly fitted into a suitable opening in the
center of the tripod. Upon the wheel thus mounted, rests a circular wooden
platform 1.2 meters in diameter, which is firmly secured by screws through
the rim of the wheel. The platform itself is three-ply, with the grain of the
wood crossed, and it is further prevented from warping by radial cleats
bolted to its lower surface, and by its attachment to the rim. Such tables
are easily capable of supporting a weight of 150 pounds, distributed uniformly
about the center.
The tables were rotated by a small electric motor, belted to a reducing
gear, which, in turn, was belted to the table. The belt connecting the first
table to the reducing gear and also that joining the two tables were applied
directly to the rims of the motor-cycle wheels. Each table made one com-
plete revolution in about four minutes, thus exposing all the cultures to
approximately similar changes of light, heat, and moisture conditions. The
tables containing the cultures were rotated continuously during the entire
time of an experiment, excepting that they were stopped for a short period
every three days to allow the solutions to be changed.
The three series of wheat cultures here considered were conducted simul-
taneously for a period of twenty-three days after the plants were placed
in the nutrient solutions. The nutrient solutions were renewed every third
day. Each cork stopper, with its plants, was removed from its bottle and
was transferred, as quickly and as carefully as possible, to another bottle
containing the new solution. The used culture solution was discarded,
after its volume had been determined. The original amount of solution
(250 cc.) minus the amount remaining in a culture bottle when the solution
was changed was, of course, a measure of the absorption (and approximately
of the transpirational loss) for the culture in question and for the given time
period. Data of the total transpirational water loss for each culture for the
entire growth period, as well as for the partial periods between each two
changes of solutions, were thus obtained. The used culture bottles were
346 JOHN W. SHIVE
thoroughly cleaned, being rinsed first with tap water and then with distilled
water, before each new filling.
During the period of growth, the general character of the development of
tops and roots, as well as any unusual phenomena occurring in the plants,
were noted. At the end of an experiment, the tops of the plants were severed
from the roots just above the remains of the seed. The tops and roots were
then dried separately in an electric oven at a temperature of from 75° to
80°C., for a period of about twenty-four hours, after which they were dried
to constant weight at a temperature of from 102° to 104°C. The weighing
bottles were transferred from the oven to a desiccator and were allowed to
cool before weighing.
Records of temperature and of the evaporating power of the air in the
greenhouse room where the cultures were carried out, were kept through-
out the experiments. Temperature changes were recorded by means of a
thermograph protected from direct sunlight and standing beside the rotating
tables. The evaporating power of the air was measured by means of stand-
ardized porous cup atmometers;” several of these instruments were placed
among the cultures on the rotating tables and daily readings were taken.
Readings were corrected by multiplying by the coefficient of correction of the
cup used.* The cups were frequently restandardized, but no marked
alteration in the coefficients was detected, although the cups were not cleaned
throughout the entire period of operation.
The first series of cultures was conducted in September, 1914, and this was
repeated in November.
RESULTS AND SPECIAL DISCUSSION
1. INTRODUCTORY
Each of the two triple series of cultures here considered extended over a
time period of twenty-three days. Series A, sub-optimal, optimal, and supra-
optimal concentrations, was carried out between September 8 and October 1,
1914. Series B was exactly like series A but was carried out between Novem-
ber 13 and December 6. During the period of series A, the maximum tem-
perature experienced by the cultures was 32°C. (on September 23) and the
minimum was 12°C. (on September 29). The water loss from the porous cup
atmometer, indicating the evaporating power of the air, gave a daily mean of
7.7 cc., a maximum daily rate of 14.3 cc. (on September 16), a minimum daily
rate of 3.4 cc. (on September 12), and a total loss from the instrument of
177 cc. for the entire time period. During the growth period of series B a
17 Livingston, B. E., Atmometry and the porous cup atmometer. Plant World 18: 21–30, 51–74, 95–111, 143–
149. 1915. Also reprinted, Tucson, 1915. (This paper gives numerous references to the earlier literature of the
instrument.)
18 The cups used were obtained from the Plant World, Tucson, Arizona.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 347
maximum temperature of 28°C. occurred on November 23, and a minimum
of 13°C. was reached on November 18. The evaporation rate from the at-
mometer gave a daily mean of 10.8 cc., a maximum daily rate of 14.2 cc. (on
December 4), a minimum daily rate of 10.4 cc. (on November 30), and a
total loss from the instrument of 248 cc. for the entire time period. These
atmometer readings are all corrected to the Livingston cylindrical standard.
The present section will deal with the growth of the plants in the various
cultures, as showing the physiological effects produced by the different total
concentrations and the different salt proportions. The responses of the
plants to the solution in which they grew will here be studied by means of
three kinds of plant measurements, two of them strictly quantitative and
the other more qualitative. The most useful quantitative measurements
are in terms of the dry weights obtained from the cultures at the end of the
period of growth. The more qualitative comparisons are made in terms of
the apparent condition of the plants at the end of the culture period, and
they deal mainly with unusual or pathological conditions as these might be
detected by general observation. Finally, the second quantitative method
of comparing the plants of the various cultures is by means of the relative
amounts of water lost by transpiration during the entire culture period.
These transpiration values will be considered in connection with the corre-
sponding dry weights, to give the water requirements of the different cultures.
The results obtained by these three methods of study will be set forth in
order.
2. DRY WEIGHTS
A. Presentation of dry weight data for tops and roots
Since the tops and roots were weighed separately, there are two sets of
dry weight measurements available for each culture, both of which are of
great interest. Also, by summing these two weights the total dry weight of
any culture (six plants) may be obtained. Since, however, these sums are
so markedly dominated by the dry weights of tops, they are of comparatively
little interest here, and need not be further considered.
The numerical data of the yields of tops and of roots are presented in
tables III, IV, and V. Each of these tables gives the dry weights of tops
and of roots for series A and B, for a single one of the three total concentrations
here employed. The first column gives the culture numbers, corresponding
to their positions on the triangular diagram showing the variations in Salt
proportions (fig. 1). Then follow three columns referring to tops and three
referring to roots. The first two columns in each of these groups of three
present the data for series A and B, respectively, and the third column in
each group gives the numbers obtained by averaging the corresponding data
from series A and B. Each of these averages thus represents the two series
combined. Each of the data is expressed in terms of the corresponding datum
348 JOHN W. SHIVE
TABLE III
Relative dry weights of tops and roots of wheat grown 28 days in three-salt solutions having
sub-optimal total concentration (0.1 atm.); series A conducted from September 8 to Octo-
ber 1, series B, from November 18 to December 6, 1914.

ToPs (6 PLANTs) Roots (6 PLANTs)
CULTURE
NUMBER
Series A Series B Average Series A Series B Average
R1C1 1.00% 1.00 1.00L 1.00% 1.00 1.00
- (0.2454) (0.2748) (0.2601) (0.0948) (0.1088) (0.1036)
C2 1.32 1, 11 1,22 1.13 0.97 1.05
C3 1.30 1.20 1.25 1.11 0.87 0.99
C4 1.25 1.34 1.30 0.87 0.95 0.91
C5 1.37 1.10 1.24 1.01 0.70 0.86L
C6 1.42 1.33 1. 38 1.09 0.87 0.98L
C7 1.27 1.22 1.25 0.87 0.69 0.78L
C8 1.38 1.08 1.23 0.94 0.66 0.80L
R2C1 1,09 0.97 1.03L 1.27 1. 11 1.19
C2 1.24 1.16 1.20 1.05 1.03 1.04
C3 1.47 1.31 1.39 1.00 0.85 0.93
C4 1.58 1.38 1.48 1.05 0.82 0.94
C5 1.38 1.42 1.40 0.80 0.73 0.77L
C6 1.40 1.46 1.43 0.78 0.73 0.76L
C7 1.47 1.30 1.39 0.97 0.62 0.80L
R3C1 1.27 0.94 1.11 L 1.60 1.17 1.39H
C2 1.34 1.22 1.28 1.23 1. 14 1.19
C3 1.47 1.37 1.42 0.96 1.04 1.00
C4 1.63 1.51 1.57H 0.95 0.92 0.94
C5 1.54 1.49 1,52H 0.93 0.70 0.82L
C6 1.60 1, 10 1.35 0.97 0.85 0.91
R4C1 1.04 0.95 1,00L 1.45 1.17 1.31H
C2 1.33 1.09 1.21 1.35 1.09 1.22H
C3 1.49 1.33 1.41 1.23 0.88 1.06
C4 1.52 1.62 1.57H 0.82 0.95 0.89 L
C5 1.66 1.63 1.65H 1.01 0.76 0.89 L
R5C1 1.19 0.99 1.09L 1.36 1.23 1.30H
C2 1.35 1.58 1.47 1.25 1, 11 1.18
C3 1.50 1.31 1.41 1.04 0.76 0.90
C4 1.32 1.31 1.32 0.75 0.68 0.72L
R6C1 1.25 0.95 1. 10L 1.48 1.13 1.31FH
C2 1.45 1.13 1 .29 1.07 1.07 1.07
C3 1.52 1.36 1.44 1. 19 0.85 1.02
R7C1 1.26 0.96 1.1.1L 1.52 1. 19 1.36H
C2 1.38 1.20 1.29 1.40 1.16 1.28H
R8C1 1,24 0.96 1. 10L 1.53 1.17 1.35H
K++ 1.24 1.46 1.35 0.98 0.78 0.88
Tººk 1.62 1.61 1.62 0.92 0.91 0.92
* The dry weight of culture R1C1 is always taken as unity and the other weights are expressed in terms of this.
The actual dry weight of culture R1C1 is given in parenthesis, in grams.
** K and T represent Knop's solution and Tottingham's best solution for wheat tops, respectively. These
data are introduced for comparison.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA
349
TABLE IV
Relative dry weights of tops and roots of wheat grown 23 days in three salt solutions having
optimal total concentration (1.75 atm.); series A conducted from September 8 to October
1, series B, from November 18 to December 6, 1914.
Tops (6 PLANTs)
ROOTS (6 PLANTs)
CULTURE;
NUMBER
Series A Series B Average Series A Series B Average
R1C1 1.00% 1.00 1.00L 1.00% 1.00 1.00
(0.4038) (0.4170) , (0.4104) (0.1016) (0.1099) (0.1058)
C2 1.26 1. 12 1.19 1.26 0.95 1. 11H
C3 1. 10 1.29 1.20 0.90 0.96 0.93L
C4 1.21 1. 12 1, 17 1. 11 1.02 1.07H
C5 1.24 1.27 1.26 1.02 0.95 0.99
C6 1.27 1.05 1. 16 1. 10 0.96 1.03
C7 1, 17 1.04 1. 11 1.04 0.98 1.01
C8 1.23 1. 10 1.17 1.02 0.88 0.95
R2C1 1.02 1.03 1.03L 0.97 0.94 0.96
C2 1.23 1.05 1. 14 1.16 0.94 1.05H
C3 1.28 1.22 1.25 1. 11 0.75 0.93L
C4 1.34 1.19 1.27 1.07 0.83 0.95
C5 1.15 1.20 1, 18 0.95 0.85 0.90L
C6 1.26 1.19 1 .22 1.00 0.96 0.98
C7 1.33 1.13 1.23 1.17 0.90 1.04
R3C1 1.18 1.11 1.15 1.07 0.96 1.02
C2 1.29 1.19 1.24 1.21 0.93 1.07H
C3 1.38 1.34 1.36H 1. 18 0.96 1.07H
C4 1.27 1.28 1.28 0.95 0.94 0.95
C5 1.23 1.27 1.25 0.93 0.90 0.92L
C6 1.23 1.31 1.27 0.94 0.91 0.93L
R4C1 1. 13 1, 11 1. 12 , 1.02 1.05 1.04
C2 1.36 1, 19 1.28 1.15 1.04 1.10H
C3 1, 32 1. 20 1.26 0.96 0.85 0.91L
C4 1.33 1.21 1.27 0.97 0.84 0.91 L
C5 1,38 1, 21 1,30H 1.08 1.00 1.04
R5C1 1, 18 1, 20 1, 19 1.07 1.06 1.07H
C2 1.41 1.36 1.39H 1. 14 1.01 1.08H
C3 1.30 1.17 1.24 0.97 0.85 0.91 L
C4 1.37 1.18 1.28 1.10 0.95 1.03
R6C1 1.21 1.13 1.17 1. 14 0.98 1.06H
C2 1.27 1.10 1. 19 0.97 0.85 0.91 L
C3 1 .30 1. 12 1.21 0.93 0.81 0.87L
R7C1 1 .22 1.09 1.16 1. 10 0.95 1.03
C2 1.33 1.28 1.31FI 1.04 0.91 0.98
R8C1 1.23 | . . . . . . . . . . 1.23 1.21 | . . . . . . . . . . . . 1.21
K++ 1, 16 1.02 1.09 1.01 0.88 0.95
Tºkº 1 .22 1.32 1.27 1.08 0.96 1.02

* The weight of culture R1C1 is always taken as unity and the other weights are expressed in terms of this.
The actual dry weight of culture R1C1 is given in parenthesis, in grams.
** K. and T represent Knop's solution and Tottingham's best solution for wheat tops, respectively. These
data are introduced for comparison.
350
JOHN W. SHIVE
TABLE V
Relative dry weights of tops and roots of wheat grown 28 days in three-salt solutions having
supra-optimal total concentration (4.0 atm.); series A conducted from September 8
to October 1, series B, from November 18 to December 6, 1914. -
TOPS (6 PLANTs)
ROOTS (6 PLANTS)
CULTURE
NUMBER
Series A Series B Average Series A Series B Average
R1C1 1.00% 1.00 1.00L 1.00% 1.00 1.00L
(0.2668) (0.3628) (0.3598) (0.0938) (0.0798) (0.0868)
C2 1.13 1. 12 1.13 0.94 1.14 1.04L
C3 |. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C4 |. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C5 |. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C7 | . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C8 0.97 1.29 1.13 0.78 1.45 1.12
R2C1 0.88 1.06 0.97L 0.99 1.21 1.10
C2 1. 14 1.13 1. 14 1.06 1.20 1. 13
C3 1.33 1.25 1.29H 1.19 1.16 1. 18
C4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C6 1.27 1.33 1.30H 1.08 1.33 1.21
C7 | . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
R3C1 0.99 1. 10 1.05 1.01 1.21 1. 11
C2 1.03 1.16 1.10 1.08 1.24 1. 16
C3 1.24 1.28 1.26H 1.09 1.27 1.18
C4 1.20 1.28 1.24 1.01 1.19 1.10
C5 1.26 1.25 1.26H 1.11 1.25 1.18
C6 1.18 1.09 1. 14 1.19 1.29 1.24
R4C1 1.02 1. 11 1.07 1. 16 1.40 1.28H
C2 1.10 1.36 1.23 1. 12 1.39 1.26H
C3 1.34 1.42 1.38H 1.23 1 .45 1.34H.
C4 1. 16 1.27 1.22 1, 18 1.38 1.28H
C5 1.16 1.30 1.23 1.18 1.30 1.24
R5C1 0.91 1.05 0.98L 1.14 1.42 1.28H
C2 1. 14 1.27 1.21 1, 18 1.30 1.24
C3 1.15 1.29 1.22 1.05 1.25 1.15
C4 1.23 1.18 1.21 1.18 1.28 1.23
R6C1 0.98 1. 12 1.05 1.17 1.40 1,29H
C2 1. 14 1.20 1.17 1.19 1.24 1, 22
C3 1.01 1.15 1.08 1.08 1.18 1.13
R7C1 1.00 1.02 1.01 L 1.22 1.25 1.24
C2 1.02 1.06 1.04 1.19 1.24 1.22
R8C1 0.94 0.86 0.90L 1 .22 1. 14 1, 18
K** 1,09 1.00 1.05 0.97 1.02 1.00
Tºº 1.23 1.22 1.23 1.24 1.25 1.25

* The dry weight of culture R1C1 is always taken as unity and the other weights are expressed in terms of this.
The actual dry weight of culture R1C1 is given in parenthesis, in grams.
** K and T represent Knop's solution and Tottingham's best solution, respectively. These data are intro-
duced for comparison.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 351
for culture R1C1, considered as unity, but the actual dry weight for this cul-
ture is given in parentheses, in grams. The actual weight, in any case,
may be obtained by multiplying the relative weight by the actual weight of
culture R1C1 as given in the same column. The maximum yield in each
series is here indicated by black-face type. The last two items in each column
give the data obtained from cultures in Knop's solution and in Tottingham's
best solution for wheat tops, with total concentrations equal to that of the
three-salt solution considered in the same table. As has been remarked, the
Supra-optimal series was not complete, on account of the instability of certain
solutions. Also, culture R8C1 of optimal series B suffered an early death,
for some reason not related to the properties of this particular solution; its
data are therefore omitted.
In the discussion of these data the dry yields of tops will be treated first,
followed by a similar treatment of root yields. For facility in making com-
parisons, the dry weights of tops and of roots relative to that of culture R1C1
for each actual series and for the three series of averages were plotted on
triangular diagrams corresponding to that shown in figure 1. Only the
three diagrams for average yields will be presented here, however (figs. 2–4
and 6–8), and only these averages will be discussed in detail. While con-
siderable difference occurred between the two series (A and B) of each pair,
these differences must be related to unknown conditions, and the most
prominent features of both individual series of each pair are clearly brought
out by the corresponding series of averages. (The diagrams of these figures
also show the ion ratio values for the different culture solutions, which will
be considered below. For the present discussion the narrow lines crossing
the triangles may be ignored altogether, as well as the marginal numbers
designating these lines, and also the ratio formulas.)
At the intersections showing the culture locations on the diagrams of
figures 2–4 and 6–8, were placed numbers representing the relative average
dry weights, taken directly from the proper columns of tables III, IV, and V.
Thus each diagram represented graphically the distribution of average dry-
weight magnitudes or yields in its particular Series. To facilitate the study
of this distribution, the total range of yields was divided into a lower one-
fourth, a medium one-half, and an upper one-fourth, and three regions were
outlined on each triangle, corresponding to these three partial ranges. (In
tables III, IV, and V, the average yields are marked with an L if they lie
within the lower one-fourth of the total range for the series, and with an H if
they lie within the upper one-fourth.) The low and the high regions were
then separated from the medium region by broken lines, each drawn through
points showing equal values. These lines are shown in figures 2–4 and 6–8,
but the yield values are omitted from the diagrams, to avoid complication.
Also, the culture numbers are omitted from these triangles; they are of course
the same as those shown in figure 1. One of the broken lines thus drawn
352 - JOHN W. SHIVE
upon the triangular diagram represents the upper limit of the area of low
average dry weight (thus separating the lower one-fourth from the medium
one-half) and the other indicates the lower limit of the area of high average
weights (separating the upper one-fourth from the medium one-half). High
areas (very good growth) are indicated by crosses and low areas (very poor
growth) by small circles. The position on each diagram of the culture giv-
ing the greatest average dry weight is shown by a larger circle.
B Dry weights of tops
Inspection of the columns referring to tops in tables III, IV and V, and of
the triangular diagrams of figures 2–4, brings out the fact that the yield of
0.18 (). 09
0, 12
(Mg/Ca)
FIG. 2. Diagram showing relative yields of wheat tops in sub-optimal total con-
centration (0.1 atm.), averages from series A and B. Area of low yields (1.00 to 1,16)
indicated by small circles; area of high yields (1.49 to 1.65) indicated by crosses. The
culture giving the highest yield is marked by a larger circle.
tops is directly related to the total osmotic concentration and to the osmotic
proportions of the salts in the culture solutions, as was of course to be expected.
The following paragraphs will deal with these relations. The points to be
brought out will be presented in three groups, referring to the three total
concentrations employed. The discussion will proceed with reference to the
triangular diagrams of figures 2–4, and will deal especially with the ranges of
the high and low average dry weights, or the extents of the corresponding
high and low areas as marked on the diagrams. In this study it is to be
remembered that the position of any culture on the triangular diagram, and
the range of any area, is a graphic representation of the osmotic proportions

Physiological, BALANCE IN NUTRIENT MEDIA 353
of the three salts as they occur in that culture solution, or of the range of
these proportions in the cultures giving high or low average dry weights of
tops. To interpret these positions upon the diagrams in terms of osmotic
proportions of salts reference is to be had to figure 1. . -
(1) Sub-optimal total concentration (0.1 atm.) (fig. 2). The average dry
weights from sub-optimal series A and B (plotted on the diagram of figure 2)
indicate an area of low average weights (1.00–1.16) including the entire row
of cultures on the left margin of the triangle, while the area of high average
weights (1.49–1.65) occupies the right central region.
The highest average dry yield of tops occurs with culture R4C5 and is 65
per cent higher than the corresponding average for culture R1C1. The
0 - 18 0.09
- (Mg/Ca )
FIG. 3. Diagram showing relative yields of wheat tops in optimal total concen-
tration (1.75 atm.), averages from series A and B. Area of low yields (1.00 to 1.10)
indicated by small circles; area of high yields (1.29 to 1.39) indicated by crosses. The
culture giving the highest yield is marked by a larger circle. ,
same culture gave the highest actual yield in both series A and series B, the
increase over that of culture R1C1 being 66 per cent for the first and 63 per
cent for the second series. From this agreement it appears that the greatest
production of dry weight of tops in this three-salt solution with 0.1 atmo-
Sphere of diffusion tension is to be expected with the salt proportions of
culture R4C5. This solution is characterized by having five-tenths of its
total diffusion tension due to calcium nitrate, one-tenth due to magnesium
sulphate, and four-tenths due to mono-potassium phosphate.
(2) Optimal total concentration (1.75 atm.) (fig. 3). The diagram repre-
senting the average yields for the optimal total concentration shows an area

354 & JOHN W. SHIVE
of low yields (1.00–1.10) at the extreme lower left. It shows two main
areas of high yields (1.29–1.39) in the left central region, which are nearly
joined at cultures R4C2 and R4C3. Secondary areas of high yield are
also indicated for cultures R4C5 and R7C2, but these yields are not relatively
very high.
Optimal series A and B agree in showing the highest dry weight of tops for
the same culture, R5C2. In series A the yield for this culture was 41 per
cent, and in series B 36 per cent higher than the corresponding dry yield
for culture R1C1. The highest average yield of tops, of course, also occurs
for culture R5C2, and is 39 per cent higher than the average yield for culture
R1C1. It thus appears that the greatest production of dry weight of tops
0, 18 0.09
é
(Mg/K) 1 .. 3 &$º (Ca/K,
Aº"
/>
0.34
º
ŻºłżºSWA 1.2
£º
Sº
%3A% WVNA
8 U.P , º
"/º/>
3. 2
13.8, r—r—ſ I y w * A-6.4
18.8 7 - 4 3, 7 2.0
1 1 - 0 5. 5 2,9 1 .. 5 0 - 74 0.24
(Mg/Ca)
FIG. 4. Diagram showing relative yields of wheat tops in supra-optimal total con-
centration (4.0 atm.), averages from series A and B. Area of low yields (0.90 to 1.02)
indicated by small circles; area of high yields (1.26 to 1.38) indicated by crosses. The
culture giving the highest yield is marked by a larger circle.
in this three-salt solution with 1.75 atmospheres of total diffusion tension
is to be expected with the salt proportions of solution R5C2. This solution
is characterized by having two-tenths of its total diffusion tension due to
calcium nitrate, three-tenths due to magnesium sulphate, and five-tenths
due to mono-potassium phosphate.
(3) Supra-optimal total concentration (4.0 atm.) (fig. 4). The incomplete-
ness of the supra-optimal series, on account of the instability of some of the
solutions, renders the location of the high and low areas on the diagram of
figure 4 somewhat unsatisfactory. It appears highly probable that the
main regions of high and low dry weights of tops are here rightly indicated,
however.





















PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 355
The area of low average yields of tops (0.90–1.02) occupies three regions
at the left margin of the diagram (fig. 4), and the main area of high average
yields (1.26–1.38) occupies the center of the triangle.
The highest dry weight of supra-optimal series A occurred with culture
R4C3 and was 34 per cent higher than that given by culture R1C1 of the
same Series. Similarly, the highest dry weight of series B occurred with the
same culture and was 42 per cent higher than the corresponding weight of
culture R1C1. The highest average yield (culture R4C3) is 38 per cent
higher than that for culture R1C1. It thus appears that the greatest pro-
duction of dry weight of tops in this three-salt solution with 4.0 atmospheres
of total diffusion tension is to be expected with the salt proportions of culture
R4C3. This solution is characterized by having three-tenths of its total
diffusion tension due to calcium nitrate, three-tenths due to magnesium
sulphate, and four-tenths due to mono-potassium phosphate.
(4) Relation of total concentration to the physiological effects of the various
osmotic proportions of the salts. Consideration of relative dry weights of
tops (figs. 2–4). It is important to appreciate the effect of the total con-
centration of the medium in determining the relative physiological values of
the various salt proportions represented in each series. This may be done by
referring to the three triangular diagrams representing average dry weights
of tops (figs. 2-4). The positions and ranges, on these diagrams, of the areas
of low and high average dry weights, and the positions of the solutions show-
ing maximum yields, as these positions and ranges are altered relative to the
total concentration, will now receive attention.
In passing from the sub-optimal to the optimal series (from fig. 2 to fig. 3),
it is seen at once that the extent of the area of low top yields becomes greatly
reduced. While this area extends along the entire left margin of the diagram
representing the sub-optimal series (fig. 2), it is restricted to a small region at
the lower left in the diagram representing the optimal (fig. 3). While eight
solutions lie within this area on the sub-optimal diagram, only two (R1C1
and R2C1) are thus included on the other. Nevertheless, it is to be noted
that these two solutions, the dry weights for which lie within the lower one-
fourth of the total range of these weights for the optimal series, are the same
solutions (characterized by the same salt proportions) that show the lowest
two dry weights in the sub-optimal series. It is to be noted further, that the
remaining left-marginal solutions of the optimal diagram (fig. 3) are all
characterized as giving very low medium weights. These two total con-
centrations thus agree in indicating solutions R1C1 and R2C1 as very poorly
adapted to the production of dry weights of tops of wheat as here grown.
They also agree in showing the entire left marginal row of solutions on the
triangular diagram as physiologically poor.
Turning to the supra-optimal diagram (fig. 4), the area of low top yields
appears here in three separate portions lying at the left margin of the triangle,
356 JoHN W. SHIVE
and the marginal solutions that lie between these separate portions of the
area of low yields show very low medium weights. Thus the solutions of
the left marginal row are indicated as of poor physiological balance for all
three concentrations. In the supra-optimal series, solutions R1C1 and R2C1
(which show the two lowest yields in both of the other two series, as has been
noted) do not show the lowest yields, but the next to the lowest is here
given for one of these two, R2C1 (0.97). •
It is apparent that some differences are manifest between the physiological
properties of the left marginal solutions, according to the total concentrations,
but these differences are not greatly marked. These eight solutions are
characterized by low osmotic partial concentrations of calcium nitrate, but
they include the whole range of partial concentrations in the case of both of
the other salts. The worst two solutions (R1C1 and R2C1) are characterized
by low partial concentrations of mono-potassium phosphate and high partial
concentrations of magnesium sulphate. It thus appears that physiological
balance is greatly disturbed by low partial concentration of the calcium salt,
and this disturbance tends to be aggravated by relatively high partial con-
centrations of magnesium sulphate accompanied by low ones of mono-
potassium phosphate.
Turning now to the areas of high relative dry weights, as represented on
the diagrams of figures 2–4, if we pass from the sub-optimal to the optimal
triangle (from fig. 2 to fig. 3), it appears as though the high area of the
former drew away from the right margin and became more centralized, also
dividing into two separate portions. These two portions of the high area of
the optimal diagram (fig. 3) lie mainly to the left of the central vertical axis
of the triangle and do not extend to the margin at any point. Passing to the
supra-optimal diagram (fig. 4), these two regions of high yields appear to
become combined in a more restricted region about the center of the diagram,
but the supra-optimal series is not perfectly satisfactory, on account of
incompleteness due to the chemistry of the salts.
If we compare the three solutions that show highest average yields of tops
in the three series, respectively, we find that the best proportions of salts
are not at all the same for the three different total concentrations here tested.
With the sub-optimal total concentration (fig. 2) the best physiological balance
is shown for solution R4C5, which is characterized by having only one-tenth
of its total diffusion tension due to MgSO4 while the remaining nine-tenths of
the total diffusion tension is almost cqually divided between the other two
salts, five-tenths being due to Ca(NO3), and four-tenths to KH2PO4. With
the optimal total concentration (fig. 3), solution R5C2 shows the best yield,
and this solution occupies an entirely different position on the diagram from
that occupied by R4C5. This solution has a relatively low partial diffusion
tension due to Ca(NO3)2 (two-tenths), and a relatively high osmotic content of
MgSO4 (three-tenths). With the supra-optimal total concentration (fig. 4)
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 357
the solution showing the highest dry weight of tops occupies still another
position on the diagram. This solution has equal osmotic proportions of
Ca(NO3)2 and MgSO4, three-tenths of its total diffusion tension being due to
each of these salts, and four-tenths are due to KH2PO4.
From these considerations it may be safely concluded, as has already been
pointed out by Gile,” McCool,” and Tottingham, that physiological salt-
balance in such nutrient media as these is markedly dependent upon the
total concentration of the solution. It is therefore quite out of the question
to state any given set of proportions of the three salts here used, that may be
expected to give any particular kind of growth (as optimum, good, poor, etc.),
unless the total concentration of the Solution to be employed is also specified.
Gms
Optimal *-*-*.
Supra-optimal —
Sub-optimal * - * * * *- :-
. 60 —

. 55 - N- ---
.45- ^, - y *s, ..”
|^)/ -N/º- M \,…, TITN
35 — \ h f - - º
º \ v t W | f \ f v’ w ,” , \ / \
tº 3O - \{i . \
- ,\, ' "---|z w f *
\
. 25 —
T R5 R3 R1 R7 R4 R4 R3 R3 R5 R5 R4 R2 R2 R4 Rl R3 R3 R5 R2 R2 R6 Rl R6 R2 Rl R6 R1 Rl R7R3 R2 R2 R4 Rl R8 Rl
C2 C3 C3 C2 C5 C2 C4 C6 C4 Cl C4 C4 C3 C3 C5 C5 C2 C3 C7 C6 C3 C2 C2 C5 C5 cl C4 c6 cl Cl C2 Cl Cl C7 Cl Cl
–––––––––––––––––––1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–
FIG. 5. Average actual yields (grams) of wheat tops (series A and B) for sub-opti-
mal, optimal, and supra-optimal total concentrations.
Consideration of absolute dry weights of tops. The effect of the total con-
centration of the medium upon the influence of various salt proportions, as
these determine physiological properties, may be brought out in another way,
by considering the actual dry weights instead of the relative ones treated
above. To bring out the relations in question the average yields of tops for
the optimum concentration were arranged in the order of their magnitudes,
beginning with the highest. These form a rather uniformly decreasing Series
of numbers, which were next plotted to form the graph shown as the upper
one in figure 5. Here the abscissas are taken arbitrarily to represent the
19 Gile, P. L., Lime-magnesia ratio as influenced by concentration. Porto Rico Agric. Exp. Sta. Bull. 12.
1912. -
20 McCool, M. H., The action of certain nutrient and non-nutrient bases on plant growth. Cornell Agric.
Exp. Sta. Mem. 2: 121-170.
358 JOHN W. SHIVE
different cultures, the numbers of which are placed below, and the ordinates
represent the average dry weight values. With the same abscissas, the cor-
responding dry weight values for the sub-optimal and supra-optimal con-
centrations were then plotted on the same sheet, using the same scale for the
ordinates. As is shown in figure 5, there result three graphs, which lie one
above the other. The upper One, representing the actual average yields of
the optimal series, exhibits a uniform downward slope to the right, as was
previously arranged. Next below this comes the supra-optimal graph, and
that for the sub-optimal series is lowest.
From figure 5 it is at once clear that the supra-optimal and sub-optimal
graphs, while showing a general downward slope to the right, are very irregu-
lar and not at all parallel to each other or to the optimal graph. This indi-
cates again that the same set of salt proportions (denoted by the same cul-
ture number) has very different physiological properties with the three differ-
ent total concentrations.
Furthermore, the optimal series definitely exhibits the greatest average
absolute dry weight for every set of salt proportions throughout the entire
series. There is no crossing of the graphs; every set of salt proportions in the
supra-optimal series gave a lower dry weight average than did the same pro-
portions in the optimal, and every one in the sub-optimal gave a lower aver-
age weight than did the same set in the supra-optimal. The fact that each
series has its own particular set of optimal proportions of the three salts is
again apparent in this series of graphs. It may be added that the highest
yield for the optimal concentration is not equalled by the highest one for the
supra-optimal, and that the highest for the supra-optimal is not equalled by
that for the sub-optimal. º
Relation of yields of tops to the proportions of the chemical ions. a. Ion ratios.
In connection with the problems of salt antagonism and the physiological
balance of nutrient solutions, considerable attention has been given by vari-
ous writers to the ratios of partial Salt (ion, element, etc.) concentrations.
Such ratios are frequently stated in such a way as to show the number of
atoms of one element occurring in the solution, for each single atom of another
element, as in the case of Loew’s” ratio of magnesium to calcium. Of course
these ratios, like the partial concentrations from which they are derived,
may also be expressed directly in terms of weight, thus giving the number of
grams of one element present in the Solution for a single gram of another
element. But this sort of expression is practically without either chemical
or physiological meaning, and therefore the weight method will not be here
employed. It is obvious that the cations here dealt with are atoms, while
the corresponding anions are atomic groups. It therefore follows that ratios
* Loew, O. and May, D. W., The relation of lime and magnesia to plant growth. U. S. Dept. Agric. Bur.
Plant Ind. Bull. 1. 1901.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 359
of atoms and also those of atomic groups may be employed to characterize
a given nutrient solution. Since ions are to be considered as the chemically
and metabolically active units in such solutions as those here dealt with, a
detailed study of ionic ratios has been carried out, with the aim of finding
out whether certain ones of these various ratio values may be important in
determining the physiological properties of the various salt combinations.
The present section will be devoted to such a study.
In the solutions here dealt with there are three cations and three anions,
neglecting the very small amount of iron which is uniformly present. It
follows that there are three possible cation ratios, Mg/Ca, Mg/K, and Ca/K,
and their reciprocals, the latter not requiring special attention; also, there
are three anion ratios, SO4/NO3, SO4/PO4, and NO3/PO4, and their recipro-
cals. This consideration neglects the H-ions that arise from KH2PO4,
and assumes complete ionization of all the salts. From the nature of the
salts used it is clear that the ratio SOA/NO3 always has a value just one-
half as great as that of the ratio Mg/Ca for the same solution. Simi-
larly, the ratio NOa/PO, has twice the magnitude of the ratio Ca/POs, and
the ratio SO4/PO4 is identical in value with the corresponding ratio Mg/K
in the same solution. These relations follow from the fact that for every ion of
Cathere are two ions of NO3, while the ions of Mg and K are accompanied
by single ions of SO, and PO4, respectively. From the way in which these
three-salt nutrient solutions are prepared, if the values of two out of the three
cation ratios are known the value of the third may be found, since the total
salt concentration of the solution is also known and only three salts ever
enter into its composition. A similar relation holds for the magnitudes of
the three anion ratios, and since these are definitely related to the cation
ratio values, it follows that if the values of any two cation ratios or of any two
anion ratios be given (of course not including both of any pair of reciprocals,
which are really the same ratio), all the other ionic ratios may be directly
calculated, including those formed from one cation and one anion.
The chemical nature of each of the solutions used in this study being thus
determined by any two ion ratio values, taken together with the total con-
centration, it becomes of considerable importance to determine the ratios
for all of the different solutions of each of the three series and to study these
values with reference to the physiological properties of the various solutions.
On account of the known relations holding between the different ratio
values, as noted above, it becomes unnecessary to determine any but the three
different cation ratios for each solution. These determinations are accom-
plished by calculation directly from the partial molecular salt concentrations
given in table II. Thus, in the case of solution R1C1 with optimal total con-
centration, for example, the partial molecular concentration of MgSO4 is
0.0400, that of Ca(NO3), is 0.0026, and that of KHAPO, is 0.0036, from which
is appears that this solution contains, per liter, 0.0400 gram-ions (or gram-
pHYSIOLOGICAL RESEARCHES, VOL. 1, NO, 7
SERIAL NO. 7, NOVEMBER, 1915
360 JOHN W. SHIVE
atoms) of Mg; 0.0026 gram-ions of Ca, and 0.0036 gram-ions of K. The
three cation ratio values characterizing this solution are then obtained by
* tº e º tº & Mg 0.0400
wº- * t l - : — = — =
performing the operations indicated by the ratios themselves Ca 0.0026
Mg 0.0400 tº tº 3 &
40; † = ± = 11.10, etc. Th * 2 of
15.40; K 0.0036 0, etc. These operations are of course justified
by the chemical principle (based on the atomic theory of matter) that the
number of atoms of an element present in a given mass of it is proportional
to the number of gram-atoms contained in the mass, a gram-atom (similar
to a gram-molecule) being the number of grams of the element in question
that is represented by its atomic weight.
b. Presentation of ion ratio values. The ratio values, Mg/Ca, Mg/K, and
Ca/K, for all the thirty-six sets of salt proportions and for all three total
concentrations here studied, are set forth in table VI. It will be noted that
the ratio values are the same for the sub-optimal and optimal concentrations,
but that the supra-optimal concentration has a series of its own. This is
on account of the apparent depression of ionization occurring in the stronger
solutions, as has been noted above. The first column of table VI gives the
solution numbers, denoting the salt proportions on the basis of partial
diffusion tensions. Columns 2, 3 and 4 present the three ratio values for
the sub-optimal and optimal concentrations, and columns 5, 6 and 7 give
those for the supra-optimal. The solutions giving the greatest yields of
tops and of roots are designated in the table by letters, which refer to notes
below to this effect.
The data of table VI were plotted upon the triangular diagrams, in a man-
ner similar to that in which the dry weights were plotted, after which lines
were drawn through approximately equal ratio values, thus dividing the
triangular area into strips with reference to each ratio. The diagrams
thus obtained have been represented by the narrow lines in figures 2–4, and
will be similarly employed in figures 6–8 and in figures 10 and 11, as bases on
which the other data are plotted. It will be noted, from the figures just
mentioned, that the cation ratio diagram for the sub-optimal and that for
the optimal total concentration (figs. 2, 3, 6, 7) are identical, while the cor-
responding diagram for the supra-optimal concentration (figs. 4 and 8) is
somewhat different, corresponding to the fact that each cation ratio of table
VI has a different value for the sub-optimal and optimal series from that given
for the supra-optimal series. As has been remarked, this is due to the de-
pressed ionization in the more concentrated series of solutions. In these
figures the narrow lines approximately represent the respective ratio values,
while the broad lines show by their intersections the different osmotic salt
proportions and the locations of the different cultures or solutions. Since
there are three ratios to be considered, there are three sets of ratio lines shown
on each triangle. Each side of the triangle is intersected by all of the lines
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA
361
TABLE VI
Values of the three cation ratios for three-salt solutions with sub-optimal, optimal, and
supra-optimal total osmotic concentration.
SOLUTION
$U B-OPTIMAL AND OPTIMAL CONCENTRATION
SUPRA-OPTIMAL CONCENTRATION
NUMBER
Mg/Ca Mg/IK Ca/K Mg/Ca Mg/K Ca/K
RIC1 15.40 11 : 10 0.72 18.80 13.78 0.73
C2a 6.74 9.72 1.44 8.10 11.89 1.47
C3 3.85 8.34 2.16 1. . . . . . . . . . . .". . . . . . . . . . . . . . . . . . . . . . . .
C4 2.40 6.95 2.88 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C5 1.54 5. 55 3.60 l. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C6 0.96 4.17 4.32 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C7 0. 55 2.78 5.04 |. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C8 0.24 1.39 5.76 0.24 1.58 6.40
R2C1 13.46 4.86 0.36 16.21 5.99 0.37
C2 5.77 4.17 O .72 6.73 4.95 0.75
C3 3.21 3.47 1.08 3.60 3.95 1. 10
C4 1.92 2,77 1.44 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C5 1.15 2.08 1.80 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C6 0.64 1.39 2. 16 0.68 1.58 2.31
C7 0.27 0.69 2.52 |. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
R3C1b 11. 55 2.78 0.24 13.46 3.30 0.24
C2 4.81 2.32 0.48 5.40 2.64 0.49
C3 2.56 1.85 0.72 2.88 2.11 0.73
C4 1.44 1.39 0.96 1.61 1.59 0.98
C5 0.77 0.93 1.20 0.86 1.05 1.22
C6 0.32 0.46 1.44 0.34 0.52 1.54
R4C1 9.61 1.74 0.18 10.89 1.94 0.18
C2 3.85 1. 39 0.36 4.32 1, 55 0.36
C3° 1.92 1.04 0.54 2, 16 1, 16 0.54
C4 0.96 0.69 0.72 1.07 0.77 0.72
C50 0.38 ().35 0.90 0.43 0.39 0.90
R5C1 7.70 1. 11 0.14 8.63 1.24 0.14
C2° 2.88 0.83 0.29 3.24 0.93 0.29
C3 1.28 0.56 0.43 1.44 0.62 0.43
C4 0.48 0.28 0.58 0.54 0.31 0.58
R6C1 5.77 0.69 0.12 6.49 0.74 0.11
C2 1.92 0.46 0.24 2. 16 0.49 0.23
C3 0.64 0.23 0.36 0.72 0.24 0.34
R7C1 3.85 0.40 0.10 4.32 0.41 0.09
C2 0.96 0.20 0.20 1.07 0.20 0.19
R8C1 1.92 0.18 0.09 2, 15 0.18 0.09
* Best solution for yield of roots, optimal concentration.
Best solution for yield of roots, sub-optimal concentration.
9 Best solution for yield of tops and roots, supra-optimal concentration.
d Best solution for yield of tops, sub-optimal concentration.
* Best solution for yield of tops, optimal concentration.
362 John W. Shive
for just one ratio and by some of the lines for each of the other two ratios.
The ratio values represented by the lines of each set are shown along that
side of the triangle which is intersected by all of that particular set, and that
side may be considered as the triangle base for the ratio in question. The
three ratio formulas are placed just outside of the triangle, on the three sides
that are their respective bases, Mg/Ca being below, Mg/K at the left, and
Ca/K at the right. It will be noticed at once that the ratio values do not
alter at a constant rate, proceeding across the triangular diagram, so that
the scale represented by the spacing of the narrow lines is much larger at
one end of each series of values than at the other. In this respect the ratio
diagrams are fundamentally different from those of Osmotic Salt proportions,
represented by the broad lines of the same figures. º
c. Relation of ion ratio values to yields of tops. A study of the relations
between the various ion ratio values and the magnitudes of the corresponding
yields, shown in figures 2–4, brings out some interesting points. In the
diagram for the sub-optimal series (fig. 2) the area of low top yields, extending
along the left side of the triangle embraces the full range of the value of the
ratio Mg/K, from 0.18 to 11.10. The magnitude of this ratio is therefore
not at all related to the production of low top yields in this series. Each of
the other two ion ratios, however, shows a definite relation to the position and
extent of this low area, which embraces ratio values of Mg/Ca varying
from 1.62 to 15.40, and of Ca/K varying from 0.09 to 1.21. The total range
of each of the last two ratios is considerably greater than the range embraced
by the low area of the diagram, so that there occurs a range of ratio values,
in each case, that are characterized as not producing low top yields. It there-
fore follows that low top yields are to be expected, with this sort of 3-salt
solution having a total concentration of 0.1 atmosphere, when the value of
Mg/Ca lies between 1.62 and 15.40 while the value of Ca/K lies between
0.09 and 1.21. The cultures giving low yields of tops are characterized as
including all but the very lowest values of Mg/Ca here tested and as re-
stricted to the lowest values of Ca/K.
Similarly, for the area of high dry weights of tops, with the sub-optimal
total concentration (fig. 2), this area is seen to be confined to ratio ranges as
follows: Mg/Ca, 0.38–2.36; Mg/K, 0.31–2.58; and Ca/K, 0.65–1.43. These
ranges are not very great in any case, and all lie near—but do not include—
the lowest values of the respective ratios occurring in the series as here ar-
ranged. Since any two ratio values are sufficient to determine the position
of a point on the triangular diagram, it may be said that high top yields may
be expected in this series of solutions (0.1 atm. total concentration), when the
value of Mg/Ca lies between 0.38 and 2.36, and that of Ca/K lies between
0.65 and 1.43. As the series is here arranged the similar range of the third
ratio value (Mg/K) is also determined by these data.
The highest average yield of tops here occurred (culture R4C5) with the
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 363
ratio values: Mg/Ca, 0.38; Mg/K, 0.35; and Ca/K, 1.08. This means that
this best solution for tops contains 0.38 atom of Mg for 1 atom of Ca, 1.08
atom of Ca for 1 atom of K, and 0.35 atom of Mg for 1 atom of K. It
thus contains the atoms Mg, Ca, and K in the proportions 0.35 : 1.08 : 1.00.
The remaining two diagrams of top yields (figs. 3 and 4) may be treated in
a similar manner, thus bringing out the ranges of the various ratio values
embraced by the low and high areas, respectively. The results of such com-
parisons are brought together in table VII. This table is divided into four
parts. In the first part are given the minimum and maximum ratio values
TABLE VII
Minimum and maximum values for the three cation ratios, and ranges of these values,
for each entire series and for high and low yields of wheat tops, with each total concen-
tration tested. Also, these ratio values for the three best solutions for yields of tops.
Mg/Ca RATIO Mg/K RATIO Ca/K RATIo
Tº jº Tº Tº g Tº Ts g "…
~ - 8 -> ~ -: E - ~ — ~
#5 | = | #5 || 5 || 5 | #5 | #5 || 5 | #5
# | #5 | # #3 | #5 | # | #3 | #5 | #
| Minimum 0.24 0.24 0.24 0.18 0.18 0.18 0.09| 0.09| 0.09
Entire series | Maximum | 15.40 15.40 18.80| 11.10| 11.10| 13.78 5.76|| 5.76|| 6.40
Range 15.16|| 15, 16| 18.56|| 10.92| 10.92. 13.60 5.67| 5.67| 6.31
Minimum 1.62 7.36|| 1 , 23| 0.18| 3.56|| 0 , 18| 0.09| 0.29| 0.09
Low yields Maximum | 15.40| 15.40| 18.80 11.10| 11.10| 13.78] 1.21| 1.08| 0.85
Range 13.78 8.04 17.57 10.93| 7.54|| 13.61| 1.12 0.79| 0.76
Minimum 0.38 1.58 0.74 0.31] 0.53 0.74 0.65 0.19| 0.35
High yields | Maximum 2.36 5.16|| 4.07| 2.58| 2.84 4.45| 1.43| 1.45| 2.31
Range 1.98 3.58| 3.33. 2.27 2.31] 3.71 0.78 1.26 1.96
Highest
yields 0.38 2.88| 2.16|| 0.35| 0.83| 1.16|| 1.08 0.29| 0.54
and the ranges for these, for the entire series. In the second part are given
the ratio values and ranges for low yields of tops (lowest one-fourth of the
total range of yields). In the third part are given the corresponding numbers
for high yields (highest one-fourth of the total range of yields). In the fourth
part are given, for ready comparison, the ratio values characterizing the
solution that produced the highest yield for each of the three different total
concentrations. -
The results shown in table VII are expressed graphically in figures 2–4,
the first of which has been discussed. Reference to these figures and to table
VII shows that the areas of low and high yields of tops are generally limited
364 JOHN W. SHIVE
to certain ranges of the cation ratio values, these ranges being less extensive
than the corresponding total ranges for the series in question.
There are just two exceptions to this last statement; as has been remarked,
the magnitude of the ratio Mg/K is without correlation to the yield of tops
in the sub-optimal series (fig. 2), and the same is true for the supra-optimal
series (fig. 4). Also, the range of Mg/K that defines the area of low top yields
in the optimal series (fig. 3) embraces somewhat more than the upper one-
half of the total range of this value in this case, so that here also, this ratio
does not appear to furnish a very definite criterion by which to judge the
physiological properties of the medium. It is also to be noted that the ranges
of the ratio Mg/Ca that correspond to the areas of low top yields in the sub-
optimal (fig. 2) and supra-optimal (fig. 4) series are almost coextensive with
the corresponding total ranges for the respective series and that the range, of
this same ratio value, that limits the low yields in the optimal series (fig. 3),
embraces almost the whole of the upper one-half of the corresponding total
range. From these observations it appears that the solutions giving low dry
weights of tops are not very satisfactorily defined in terms of either the Mg/Ca
or the Mg/K ratio values, though these values offer somewhat better criteria.
for judging the physiological balance of the solution in the case of the optimal
total concentration than in either of the others.
The areas representing low yields are well defined in all three diagrams
(figs. 2–4) by the ranges of the ratio-value Ca/K, this range embracing only
the lower one-fifth (or less) of the corresponding total range. With all
three total concentrations the solutions giving high top yields are all clearly
limited to low values of all three cation ratios, but in no case are any of these
solutions characterized by the very lowest ratio values here considered. In
general, no single ratio and no pair of ratios appear to determine the pro-
duction of high top yields any more than that of low ones.
It is instructive to observe the marked differences between the atomic
proportions of Mg, Ca, and K characterizing the solutions giving the highest
yields of tops in the three series, respectively. While, as has been remarked,
the most perfectly balanced solution (for dry weight of wheat tops) of the
sub-optimal series (total concentration 0.1 atm.) contains 0.35 and 1.08
atoms of magnesium and of calcium, respectively, for each single atom of
potassium, yet the most perfectly balanced solutions of the optimal and of the
supra-optimal series (total concentrations 1.75 and 4.0 atm.) are very
different in this respect. The greatest yield of tops was obtained in a
solution having the atomic proportions, Mg, 0.83; Ca, 0.29; K, 1.00, with
optimal total concentration, while the best solution for top growth with
supra-optimal total concentration possessed the proportions, Mg, 1.16;
Ca, 0.54; K, 1.00. It is thus obvious that the cation atomic proportions
characterizing the best physiological balance, as here brought out, vary mark-
edly, and in no simple manner, with the total concentration of the solution.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 365
C. Dry weights of roots
Turning now to the dry weights of roots, as given in tables III, IV, and V,
it appears that these, as well as the yields of tops, show marked variations
according to the Osmotic proportions and total Osmotic concentrations of the
solutions in which they grew. The following consideration of the average
root yields will follow the same general lines as were followed in considering
the average weights of tops, but will not be as detailed as was the former
discussion. It will have reference to the triangular diagrams of figures 6–8.
(1) Sub-optimal total concentration (0.1 atm.) (fig. 6). The average dry
weight of roots for the sub-optimal series ranges from 0.72 to 1.39, relative
to the average yield for solution R1C1. The diagram for these root yields
0, 18 0.09
O. 12
0.27 /*
* N 0.14
+
0.44
* * } +/º
0.24
+
0.62
f
+ () , 47
(Mg/K) 1.0 A'ſ.".[. ~. (Ca/K)
+ No
+ © o 0.72
+/ +/3
1 .. 8 + 1 + * 1.08
+ + // J
+ + o
2 7 + 1 + L^ o |
- © o 1 .. 8
4.
+ o
*
4. 4 Ø o 2.4
.* o o o o o O O
8.9 oz ov o
11 . 1 / 21 D 7°N/º NZ * > *S*NZººs’ \ 5.8
-7—r r I I —T- —V
11 . 5 - 5 2.9 1 .. 5 0.74 0.24
1 5.4 6. 8 3 - 7 2. 2 1 - 1 0, 37
(MS/Ca)

FIG. 6. Diagram showing relative yields of wheat roots in sub-optimal total con-
centration (0.1 atm.), averages from series A and B. Area of low yields (0.72 to 0.89)
indicated by small circles; area of high yields (1.22 to 1.39) indicated by crosses. The
culture giving the highest yield is marked by a larger circle.
(fig. 6) shows an area of low relative values (0.72–0.89) occupying about a
third of the entire triangle, about its lower right angle. The area of high
root yields (1.22–1.39), on the other hand, occupies nearly the whole of the
left marginal region and includes solutions R8C1 and R7C2 of the right margin.
The greatest dry weight of roots for this total concentration is shown for
solution R3C1, near the lower limit of the area of high root yields. This
solution is characterized by having one-tenth of its total diffusion tension
due to Ca(NO3)2, six-tenths due to MgSO4, and three-tenths due to KH2PO4.
(2) Optimal total concentration (1.75 atm.) (fig. 7). The range of average
relative root yields for the optimal series is from 0.87 to 1.11. Comparing
366 * - JOHN W. Shive
the diagram for this series (fig. 7) with that for the sub-optimal series (fig. 6)
it appears as though the area of low relative yields had here migrated upward
and to the left, while the area of high yields had correspondingly migrated
downward and to the right. The main region of low dry weights of roots
(0.87–0.93, fig. 7) lies mainly to the right of the central vertical axis of the
triangle, but includes two solutions on that axis. The main region of high
root yields (1.05—1.11) lies to the left of the central vertical axis, the greatest
root yield being shown for solution R1C2,on the lower margin of the triangle.
This solution is characterized by having two-tenths of its total diffusion
tension due to Ca(NO3)3, seven-tenths due to MgSO4, and one-tenth due
to KH2PO4.
§. 0. 18., 0.09
0, 12
0.27
0.14
(Mg/Ca )
FIG. 7. Diagram showing relative yields of wheat roots in optimal total concen-
tration (1.75 atm.), averages from series A and B. Area of low yields (0.87 to 0.93).
indicated by small circles; area of high yields (1.05 to 1.11) indicated by crosses. The
culture giving the highest yield is marked by a larger circle.
(3) Supra-optimal total concentration (4.0 atm.) (fig. 8). The total range
of average relative dry weights of roots for the supra-optimal series extends
from 1.00 to 1.34. On the diagram for this series (fig. 8) the area of low
average root yields (1.00–1.08) occupies the extreme lower left region of
the triangle. The corresponding area of high average yields (1.26-1.34)
here occupies the center of the triangle and is extended to the right, but not
as far as the right margin, and also to the left, where it widens to include
three solutions on the left margin. This triangle is somewhat unsatisfactory
on account of its incompleteness (due to chemical relations, as has been
stated), but there is no suggestion of an area of low yields at the right of the
central axis (as in figs. 6 and 7), and it seems fair to suppose that the effect

PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 367
of this high concentration has been to shift the region of poor growth as is
here indicated. The highest average root yield is shown for solution R4C3,
near the center of the diagram. This solution is characterized by having
three-tenths of its total diffusion tension due to Ca(NO3)2, three-tenths due
to MgSO4, and four-tenths due to KH2PO4.
(4) Relation of total concentration to the physiological effects of the various
osmotic proportions of the salts. Consideration of relative dry weights of roots
(figs. 6–8). The effect of total concentration upon the positions and ranges of
the areas of low and high average yields of roots, as shown on the triangular
diagrams (figs. 6–8), will now receive attention.
0, 18 0.09
(Mg/Ca )
FIG. 8. Diagram showing relative yields of wheat roots in supra-optimal total
concentration (4.0 atm.), averages from series A and B. Area of low yields (1.00 to
1.08) indicated by small circles, area of high yields (1.26 to 1.34) indicated by crosses.
The culture giving the highest yield is marked by a larger circle.
Comparison of the diagrams of figures 6–8 makes it clear enough that total
concentration has a very pronounced effect in determining the relative in-
fluence of the different salt proportions upon growth of roots as here measured.
This effect is so pronounced that it is obviously impossible to select any single
set of salt proportions which might agree, throughout the three total con-
centrations, in producing either a very low or a very high dry weight of roots.
The study of the growth of roots in relation to their chemical and physical
environment is in such an early stage, and the data here presented appear
to be so complicated, that further generalization need not be undertaken.
The point to be emphasized in this connection, as in the case of tops, is that
a culture solution is not to be defined, as physiologically balanced for root

368 JOHN W. SHIVE
growth, in terms of its salt proportions alone; its total concentration must
also be stated.
Consideration of absolute dry weights of roots. The actual average dry
weights of roots obtained in a period of twenty-three days, with the various
concentrations and salt proportions here studied, will now be taken up, as
was done for the corresponding average yields of tops. The average dry
weights of roots for the optimum concentration were arranged in the order
of their magnitudes, beginning with the highest, and, as in the case of tops,
these, numbers form a rather uniformly decreasing series, which is shown
graphically as the uniformly sloping, nearly straight line of figure 9. This
graph, and the remaining two graphs of that figure, are plotted in the same
Optimal
ſ Supra-optimal.
- Sub-optimal — — — — — — —
• 14-
• 13 -
•l2 +
•ll-
• 10-
• O9–
.08 -
.07:
1 R1 R4R5 R3
C2 C2 C2 C3
| | |
Rl R3 R5 R6 R2 R4 R4 R2 R5 R7 Rl R8 R3 Rl R
C4 ºf Cl º de gº Gl C7 c; Cl g; * º * C
| | | |
3 Rl Rl R3 R2 R6 R
C4 C6 C3 C6 C3 C2 C
| | | | | | | | | | | | | | | |
5 R3 R4 R4 R2 R6
3 C5 C3 C4 C5 C3
| |
FIG. 9. Average actual yields (grams) of wheat roots for sub-optimal, optimal,
and supra-optimal total concentrations.
manner as was employed for tops (fig. 5). Unlike those for top yields,
there is little or no tendency for these three graphs to show parallelism;
frequent crossing is manifest. There is not even any marked tendency for
the sub-optimal and Supra-optimal graphs to slope downward to the right.
Practically the only tendency toward any agreement is exhibited between the
sub-optimal and supra-optimal graphs for the first nine cultures as arranged
in figure 9. These two graphs appear, in this region, to rise and fall simul-
taneously, but the fifth culture (R1C4) of the supra-optimal series is omitted,
so that even this amount of agreement is uncertain. It should be emphasized
that the order in which the cultures are arranged for the graphs of figure 9,
is not at all the same as their order in figure 5, so that these two sets of graphs

PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 369
are not directly comparable excepting in the general way just pointed out.
(For a direct comparison see figure 15.)
It therefore appears, that the physiological value, for the production of
root yields, of any given set of salt proportions unquestionably varies as the
total concentration of the medium is altered, which is quite similar to the
general conclusion reached from the study of top yields. -
Relation of yields of roots to the proportions of the chemical ions. The dis-
cussion of root yields with reference to the proportions of the three cations
in the solutions will follow the same general lines as were followed in the
TABLE VIII
Minimum and mazimum values for the three cation ratios, and ranges of these values,
for each entire series and for high and low yields of wheat roots with each total concen-
tration tested. Also, these ratio values for the three best solutions for yields of roots.
Mg/Ca RATIO Mg/K RATIO Ca/K RATIO
Tº || 3 || 3 | "Gº || 3 | "… Tº || 3 | "…
.5- -: 3– .5 -> -: .5- .5 ~ = .5-
#5 *4 3. 8 #5 pºt #5 3. E p-4 3. 8
o : or ~ ‘p 3 o ºg ow 2- P : © . 3 -> P .
A - || 5 3 šč | A - || 5 s §e A- || 5 § #é
se #3 gºt | Es | #3 gºt as 53 gºt
(ſ) O CO Cſ) O (ſ) C/D O Cſ)
Minimum 0.24|| 0 , 24 0.24|| 0 , 18 0.18 0.18 0.09 0.09| 0.09
Entire series | Maximum | 15.40 15.40 18.80| 11.10| 11.10| 13.78, 5.76 5.76 6.40
Range 15.16|| 15, 16| 18.56|| 10.92 10.92| 13.60 5.67| 5.67| 6.31
Minimum 0.24| 0.32 7.52 0.25 0.21| 8, 18 0.46 22 0.44
Low yields | Maximum 2.06| 2, 21 | 18.80| 6.37| 3.56|| 13.78 5.76|| 3.59| 1.46
Range 1.82| 1.89| 11 . 28 6. 12| 3.35 5.60 5.30 3.37| 1.02
Minimum 0.89| 2.29 0.75 0.18 0.53 0.54|| 0.09| 0.11| 0.10
High yields | Maximum | 12.23 9.58 11.64 4.26 10.33 2.33 0.45| 1.45 0.86
Range .11 .34| 7.29| 10.89| 4.08; 9.80| 1.79| 0.36|| 1.34 0.76
Highest
yields 11.55| 6.74 2, 16| 2.78 9.72| 1 , 16|| 0 , 24| 1.44 (). 54.

corresponding discussion of top yields, but it will be less detailed. The
maximum and minimum ratio values for the regions of high and low root
weights, as shown on the diagrams of figures 6, 7, and 8 are set forth in table
VIII, which conforms in every way with table VII.
As in the case of the top yields, the areas of high and of low dry weights of
roots are generally limited to certain ranges of the three cation ratio values
here considered, these ranges being always less extensive than the corre-
sponding total ranges. In two cases the limiting range of ratio values em-
braces almost all of the corresponding total range: (1) in the sub-optimal
series the area of low root yields (fig. 6) embraces all but the very lowest
370 JOHN W. SHIVE
values of the ratio Ca/K, and (2) in the optimal series the area of high yields
(fig. 7) embraces all but the very lowest and very highest values of the ratio
Mg/K. - -
For all three total concentrations, the solutions giving high root yields are
clearly defined as having very low values of the Ca/K ratio and rather ex-
tensive medium ranges of the ratios Mg/Ca and Mg/K. Further details
may be discovered from a study of table VIII and of the diagrams of figures
6–8, but their statement here would add nothing leading to valuable general-
izations at present. It is clear that the ratio values characterizing solutions
giving low or high yields vary in a pronounced way with the alteration of the
total concentration.
The atomic proportions of Mg, Ca, and K that characterize the three solu-
tions giving the highest yields of roots in each series deserve special attention
here, as in the case of top yields. For the sub-optimal total concentration
(0.1 atm.) the most perfectly balanced solution for root yields contains 2.78
atoms of magnesium and 0.24 atoms of calcium for each single atom of potas-
sium. For the optimal series (total concentration 1.75 atm.), these atomic
proportions are Mg, 9.72: Ca, 1.44: K, 1.00, and for the supra-optimal
series (total concentration 4.0 atm.), they are Mg, 1.16: Ca, 0.54: K, 1.00.
The last of these sets of proportions is the only one of the three that is the
same for the highest yields of both roots and tops. The cation atomic
proportions characterizing the best physiological balance for root yields,
as here brought out, vary as markedly and in as recondite a way, with vari-
ation in total concentration, as they do in the case of yields of tops.
3. APPARENT CONDITION OF PLANTS
Condition of tops
No apparent differences (aside from those of size) in the tops of the wheat
plants of any series occurred during the first ten days of their growth, and
none appeared at all in the sub-optimal series. In the optimal and supra-
optimal concentrations, however, evidences of disturbed growth appeared
in the tops of a number of cultures, usually about the tenth day after the
young plants had been placed in the solutions. This disturbance was practi.
cally identical with that described by Tottingham and called by him magne-
sium injury. It manifested itself in two forms.
The more pronounced of these forms of injury consisted in a lateral rolling
backward of the entire length of the injured leaves. Leaves so injured
showed the effect at a very early stage and never attained the mature form
and size. The young leaves, just emerging from the sheath, rolled back
laterally throughout their entire length, frequently also coiling into a close
spiral. The first few leaves produced by the plants usually escaped injury
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 371
altogether or suffered only from the less pronounced form to be described
below, but the later leaves (in cultures where pronounced injury occurred)
were attacked at a very early stage of their development, even while they
were emerging from the sheath.
The pronounced form of injury did not occur on leaves that had already
developed to the usual size, though such leaves were frequently affected by
the less pronounced form. This form of injury manifested itself, in either
young or old leaves that had previously appeared normal, by a yellowing
and shrivelling of a small region several centimeters from the tip. The
distal portion, above the first lesion, remained green and apparently in healthy
condition for some time, but finally died. This less pronounced form of
injury usually began with the third leaf formed, mainly after the leaf had
attained a considerable size. It did not appear to retard growth perceptibly;
indeed, the largest and most vigorous cultures, in both optimal and supra-
optimal series, always showed this less pronounced form of leaf injury.
Where both forms of injury occurred on the same plant, the less pronounced
form was the first to make itself evident.
The cultures of the optimal and supra-optimal series may be divided,
according to the injury they sustained, into three classes: (1) those that
were without injury, (2) those that exhibited only the less pronounced form
of injury, and (3) those that exhibited the more pronounced form. For the
optimal total concentration of the solutions the observations on leaf injury,
from both series A and B, are summarized in the triangular diagram of
figure 10, and a similar presentation of these observations for the supra-
optimal concentration of the solutions is given in the diagram of figure 11.
In these diagrams the cultures suffering from the pronounced form of injury
occupy the area denoted by crosses. These cultures were also affected by
the less pronounced form of injury. Those exhibiting only the less pro-
nounced form occupy the area denoted by small circles. The area represent-
ing uninjured cultures is unshaded. The three areas thus shown for each
diagram are separated, as heretofore in similar cases, by broken lines.
For the optimal series, the area of slight injury (marked by small circles)
occupies the entire left central portion of the triangle (fig. 10), extending from
base to apex and reaching the left margin at culture R8C1. The area of
severe injury (marked by crosses) lies along the left margin of the triangle,
including all the cultures on this margin excepting R8C1, and extending to the
right, at the lower left angle of the triangle, to include culture R1C2. On
the supra-optimal triangle (fig. 11), the area of slight injury occupies a posi-
tion corresponding to that of the same area on the optimal triangle. The
area of severe injury in this series occupies the same region as in the optimal
diagram, but extends farther to the right in its lower portion, so as to include
cultures R1C2, R2C2, and R3C2.
372 JOHN W. SHIVE
(Mg/Ca )
FIG, 10, Triangular diagram for optimal total concentration (1.75 atm.), showing
the solutions producing no leaf injury (unshaded), those giving only slight leaf injury
(indicated by small circles), and those giving severe leaf injury (indicated by crosses).
(Ca/K)
f : * N N
$3. s/b/ S/N-X 0.85
&
º
Hº: *%
</
šº
C
º 4. & - NA
1 1 - 0 5.5 2.9 e 1 .. 5
(Mg/Ca)
FIG. 11. Triangular diagram for supra-optimal total concentration (4.00 atm.)
showing relation of salt proportions to leaf injury, as in figure 10.













PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 373
A study of the diagrams of figures 10 and 11 brings out the fact that the
severely injured cultures have osmotic partial concentrations of Ca(NO3),
varying from 0.1 (R1C1) to 0.2 (R1C2), but culture R8C1, in which this
concentration has also a value of 0.1, was not severely injured. It is thus
clear that the partial Osmotic concentration of Ca(NO3)2 is not a clearly
limiting factor in the production of this injury. Similarly, the injured cul-
tures have osmotic partial concentrations of MgSO4 varying from 0.2 (R7C1)
to 0.8 (R1C1), while culture R8C1, not seriously injured, exhibits this partial
concentration as 0.1. From this it appears that an osmotic partial con-
centration of MgSO4 as low as 0.1 does not produce severe injury. Never-
theless, it is clear that values of this partial concentration from 0.2 to 0.6
may or may not be severely injurious, according to other conditions. Finally,
the osmotic partial concentration of KH2PO4 may have any value from 0.1
to 0.7 without producing severe leaf injury, and also, for the same range,
it may produce injury. The only value of this partial concentration not
ever accompanied by pronounced injury is 0.8 (R8C1). From these con-
siderations it may be concluded that severe injury is definitely absent only
when the osmotic partial concentration of Ca(NO3)2 is higher than 0.2,
when that of MgSO4 is lower than 0.2, and when that of KH2PO4 is higher
than 0.7. It is thus obvious that the conditions determining severe injury
must be related to the relative proportions of the three salts rather than to
Osmotic partial concentration of any one salt, a conclusion already reached
by Tottingham for his similar series of cultures.
Considering the ratios of the osmotic partial concentrations, as was done
by Tottingham, it appears that severe injury accompanies the ratio
MgSO4/Ca(NO3)2 when this ratio has a value between 2.0 (R7C1) and 8.0
R1C1. For the remainder of the series, not severely injured the value of this
ratio varies from 1.0 (R8C1, R6C2) to 3.0 (R2C2, optimal). This makes it
clear that no severe injury is produced with these ratio values below 2,0, and
that severe injury always occurs when these values are greater than 3.0. Os-
motic ratio values of MgSO4/Ca(NO3)2 from 2.0 to 3.0 are thus questionable,
and may or may not produce severe injury, apparently according to other con-
ditions. The highest value of this ratio with which no injury occurred is
1.33 (culture R3C4). If therefore seems that no injury is to be expected,
in 3-salt solutions such as were here employed, with osmotic ratios of
MgSO4/Ca (NO3), whose values are less than 1.33. Tottingham (page 203)
has placed the limiting value of this ratio at 0.25, and notes that osmotic
ratios of MgSO4/Ca(NO3)2 with values of 2.0 or above may be expected to
produce marked injury in his 4-salt solution. .
A consideration of the cation ratio values with reference to leaf injury
brings out some interesting features. The ranges of injury indicated on the
diagrams of figures 10 and 11 are seen at once to bear no definite relation to
either of the ratios Mg/K or Ca/K; these regions embrace all or nearly
374 JOHN W. SHIVE
all of the total ranges of the values of both these ratios. On the other hand,
it is at once obvious that solutions characterized by values of the ratio Mg/Ca
below 1.5 give plants without injury in the optimal series, and those where
this ratio value is below 2.2 give uninjured plants in the supra-optimal
series. From these considerations it appears that the limit of slight injury
is practically set by the magnitude of the ratio of magnesium to calcium.
The limit of severe injury is not so readily defined; it surely is related to other
properties of the solution than the Mg/Ca ratio.
Condition of roots
No great differences between the roots of the various cultures of the sub-
optimal series were apparent. The roots of the optimal and supra-optimal
series, however, showed marked differences even during the first ten days of
their growth. During the first twelve days after they were placed in the
nutrient solutions, the roots of cultures R1C1, R1C2, R2C1, R3C1, R4C1,
R5C1, R6C1, and R7C1, were distinguished from the remaining cultures of
these series by the absence of secondary roots and by short, thickened pri-
mary roots. During the later stages of growth numerous lateral roots
appeared in these cultures, but took the form of short, stubby outgrowths.
In cultures R1C1, R2C1, R3C1, and R4C1, these numerous lateral roots
assumed the form of small, rounded tubercles. This latter condition of the
roots was coincident with the most severe injury to the leaves, and injured
roots were confined to the areas of severe leaf injury on the diagrams of figures
10 and 11. The root systems of cultures characterized by slight leaf injury
appeared to be particularly well developed, although the lateral roots of cul-
tures R1C3 and R2C2 were somewhat shorter than those of the other cultures
with slight leaf injury. In cultures where leaf injury was not apparent
at all, root development did not appear to be so vigorous, in general, as in
those suffering slight leaf injury. It thus appears that the most vigorous
root growth (for optimal and supra-optimal total concentrations) may be
expected to occur in Solutions whose osmotic ratio MgSO4/Ca(NO3), is
somewhat above 1.33. The partial concentration of KH2PO4 appears to
have no influence upon the plants, as far as either leaf or root injury is
concerned.
4. TRANSPIRATION DATA
Transpiration as a measure of growth
Introductory. Transpiration being a process that is going on at all times
in the living plant, one to which many, other plant processes are surely
related, and also one that is somewhat readily measured, it has seemed worth
while to study the relative amounts of water lost from the various cultures
during their growth and to compare these transpiration data with the results
of the other plant measurements.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 375
The relative rate of water loss has been emphasized by several writers as
an approximate measure of plant growth. Thus Whitney and Cameron”
and other workers with plant cultures in the U. S. Bureau of Soils,” have taken
the total amount of transpirational water loss during the growth of a culture
as being more or less directly proportional to the growth of the plant. The
relation between the amount of transpiration, dry weight, leaf surface, etc.,
for cultures of wheat during the first few weeks of their growth, was studied by
Livingston,” who came to the conclusion that in many instances the amount
of water lost by plants of the same species and of the same age, grown in the
same locality but in different soils or solutions, is roughly proportional to the
extent of leaf surface, and therefore, to the size of the plants. Of course such
a relation can be expected to hold only where the relative amounts of water
lost by transpiration are determined only by the size of the plants, not by
internal physiological conditions. If the plants differ internally in some
manner that affects transpiration otherwise than by differences in size, then
transpiration should not be a measure of the size, although it might still
be a valuable indicator of the general health of the plants. The general
complexity of the problem thus suggested, as to whether transpiration may
be reasonably employed to indicate relative growth and health of plants,
apparently furnishes the explanation for Harris's" unexplained remark that
Livingston's data do not justify his conclusions in this regard.
The transpiration diagrams. As has been remarked, the amount of water
lost from each culture during the period of its growth was determined by
summing the various losses recorded during that period. The total loss
thus obtained for each culture was then expressed in terms of the total loss
for culture R1C1 of the same series, considered as unity. The relative loss
thus obtained for each culture of series A and that for the corresponding
culture of series B, were averaged in all cases.
Table IX presents the transpiration data in three sections, each section
referring to a single total concentration of the medium. The first column
of the table gives, as heretofore, the culture or solution numbers referring
to the triangular diagrams. In each of the three sections of the table (sub-
optimal, optimal, and supra-optimal), the first two columns give the losses,
relative to the loss from culture R1C1 in each case, for series A and B, re-
spectively. The third column of each section gives the average losses for
series A and B together, also relative to the average loss for culture R1C1.
* Whitney, M. and Cameron, F. K., Investigations in soil fertility. U. S. Dept. Agric. Bur. Soils Bull. 23.
1904.
28 Livingston, B. E., Britton, J. C., and Reid, F. R., Studies on the properties of an unproductive soil. U. S.
Dept. Agric. Bur. Soils Bull. 28. 1905.
Livingston, B. E., Further studies on the properties of unproductive soils. U. S. Dept. Agric. Bur. Soils
Bull. 36. 1907.
24 Livingston, B. E., Relation of transpiration to growth in wheat. Bot. Gaz. 40: 178–195. 1905.
25 Harris, F. S., Effects of variations in moisture content on certain properties of a soil and on the growth of
wheat. Cornell Univ. Agr. Exp. Sta. Bull. 352. 1914.
PHYSIoloGICAL RESEARCHES, vol. 1, No. 7
SERIAI, No. 7, Nov EMBER, 1915
376 JOHN W. SHIVE
TABLE IX
Transpiration data for wheat grown 83 days in 3-salt solutions; series A, conducted from
September 8 to October 1, series B, from November 18 to December 6, 1914.

SUB-OPTIMAL TOTAL CONCEN- OPTIMAL TOTAL CONCENTRATION | SUPRA-OPTIMAL TOTAL CONCEN-
TRATION (0.1 ATM.) (1.75 ATM.) TRATION (4.0 ATM.)
CULTURE
NUMBER
Series A Series B | Average | Series A Series B | Average | Series A Series B | Average
R1C1 1.00% 1.00 | 1.00L 1.00 1.00 | 1.00L 1.00 1.00 1.00L
g (131) (160) (145) (170) (213) | (192) (126) (135) (131)
C2 | 1.42 1.21 1.32 1.21 1.08 || 1.15 1.08 1.16 1. 12
C3 1.41 1.41 | 1.41 1.17 1.12 | 1.15 1. . . . . . . . . . . . . . . . . . . . . . . .
C4 1.56 1.55 | 1.56 1.16 1.19 1.18 l. . . . . . . . . . . . . . . . . . . . . . . .
C5 1.78 1.27 1.53 1.19 1.14 | 1.17 | . . . . . . . . . . . . . . . . . . . . . . . .
C6 1.91 1.57 | 1.74.H 1.20 1.07 | 1.14 | . . . . . . . . . . . . . . . . . . . . . . . .
C7 1.61 1.47 | 1.54 1.18 1.14 | 1.16 |. . . . . . . . . . . . . . . . . . . . . . . .
C8 1.78 || 1.44 | 1.61H 1.15 1. 14 | 1.15 1.06 1.35 | 1.21
R2C1 1.22 1.16 | 1.19 L 1.07 1.08 || 1.08 0.97 1.07 || 1 ,02L
C2 | 1.38 1.35 | 1.37 1. 14 1. 11 1.13 1. 11 1. 16 || 1 , 14
C3 1.58 1.49 | 1.54 1.21 1.09 1.15 1.07 1.40 | 1.24H.
C4 1.79 | 1.51 | 1.65H 1.15 1.08 | 1.12 |. . . . . . . . . . . . . . . . . . . . . . . .
C5 1.59 1.58 || 1.59H 1.02 1.13 | 1.08 |. . . . . . . . . . . . . . . . . . . . . . . .
C6 1.57 1.66 1.61.H 1.15 1.17 1.16 1.15 1.33 1.24H
C7 1.91 1.46 | 1.69H 1.12 || 1.08 | 1.10 | . . . . . . . . . . . . . . . . . . . . . . . . .
R3C1 1.48 1.18 || 1.33 1.17 0.99 || 1.08 L 0.86 1.15 | 1.01 L
C2 || 1.69 1.50 1.60H 1.31 1.09 1.20BI 1.09 1.27 1.18
C3 || 1.66 1.54 | 1.60H 1.35 1.19 1.27H 1.15 1.36 | 1.26H
C4 1.86 || 1.67 1.77H 1. 14 1.15 1.15 1. 14 1.27 1.21
C5 1.78 1.51 1.65H 1, 16 1.14 1.15 1.13 1.27 1.20
C6 1,80 1.64 | 1.72H 1.15 1, 16 || 1 , 16 1.15 1.31 1.23H
R4C1 1.36 1.19 1.28 1.15 1.07 1. 11 1.06 1. 16 || 1 , 11
C2 1.74 1.34 1.54 1.21 1.08 1.15 1.08 1.31 1.20
C3 1.75 1.48 1.62H 1.18 1.09 1. 14 1.23 1.38 1.31H
C4 | 1.66 | 1.50 1.58H 1.24 1.09 1, 17 1, 15 1.34 1.25H
C5 1.86 1.59 1.73H 1.24 1.27 1.26H 1.22 1.34 1.28H
R5C1 1.37 1. 19 1.28 1. 14 1. 16 || 1 , 15 1.00 1. 13 | 1.07L
C2 1.58 1.36 1.47 1.34 1.20 1.27H 1, 14 1.30 1 .22
C3 | 1.66 1.37 1, 52 1, 24 1. 14 1, 19H 1.07 1.36 1.17
C4 1.53 1.34 1.44 1.28 1,06 1.17 1.19 1.26 1.23H
R6C1 1.45 | 1.20 1.33 1.23 1.13 1.18 1.05 1. 18 1. 12
C2 1.70 1.38 1.54 1.21 1.13 1, 17 1.05 1.24 1.15
C3 1.74 1.44 | 1.59H 1.23 1.08 1.16 1.05 1 .22 1. 14
R7C1 1,50 1.21 1.36 1. 19 1.07 1.13 0.98 1. 10 1.04L
C2 1.71 1.38 1.55 1.28 1.21 1.25H 1.03 1, 16 || 1.10
R8C1 || 1 , 47 1.17 1.32 1.24 1.00 1. 12 1.09 0.96 1.03L
K** 1.84 | 1.57 1.71 1, 16 1.07 1. 12 0.99 1.04 1.02
Tº: 1.54 1.51 1.53 1.23 1.22 1.23 1, 18 1.19 1. 19


* The loss for culture R1C1 is always taken as unity, and the other losses are expressed in terms of this.
actual loss for culture R1C1 is given in parenthesis, in each case, in cubic centimeters.
** K and T represent Knop's solution and Tottingham's best solution, respectively: these data are intro-
duced for comparison.
The
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 377
*
The letters L and H, following the average data as in the tables III, IV and V,
represent numbers lying, respectively, within the lower and within the upper
one-fourth of the total range of the series of averages. -
The average data of table IX were plotted on the triangular diagrams, as
were those of the dry weight yields, and the diagrams thus obtained were
Compared with the yield diagrams already given (figs. 2–4 and 6–8). Since
the notations L and H of table IX indicate the low and high regions of these
transpiration diagrams, the latter need not be here presented. The following
observations obtained by comparing them with the corresponding yield
diagrams may be given, however.
On the sub-optimal diagram the region of low transpiration embraces
only the extreme lower portion (culture R1C1 and R2C1) of the region of
low yields of tops (fig. 2). The region of high top yields lies entirely within
that of high transpiration, but the latter region is considerably larger than the
former. These two diagrams agree, in general, in showing the lower top
yields and lower transpirations as occupying the left side of the triangle,
while the higher yields and transpirations occupy the middle and lower right.
About as good agreement is manifest between the diagrams of root yields
(fig. 6) and that of transpiration, in the sub-optimal series. It is to be empha-
sized here, however, that low transpiration corresponds to high yields of
roots, while high transpiration corresponds to low yields of roots.
In the optimal series, there is an exceedingly close agreement between
transpiration and yield of tops (fig. 3), both for the regions of low and for
those of high values. Here the amount of water transpired during the
growth of the plants appears to be as good a criterion as is the final dry
weight, for judging the comparative growth obtained in the different solutions.
Comparison of the transpiration diagram with that of root yields for the
optimal series (fig. 7), shows that there is here no detailed agreement. All
that may be said is that the low transpiration data agree, in a general way, with
the data of high root yields, by lying to the left of the region of high transpi-
ration and and of low root yields.
The supra-optimal diagram of transpiration agrees very closely, but not
quite exactly, with the diagram of top yields (fig. 4); the agreement is exact
in the case of the regions of low values. No general agreement is to be noted
between the diagram of transpiration and that of root yields (fig. 8) in the
supra-optimal series, though the culture giving highest root yields also gave
highest transpiration.
From the last three paragraphs it appears that the amount of water lost
by the plants is generally a valuable indication of the top yields, under the
conditions of these cultures, low transpiration corresponding to low yield of
tops and high transpiration to high yield. This is especially true of the opti-
mal and supra-optimal total concentrations. It also appears that there is
no marked correlation between yield of roots and transpiration, though the
378 JOHN W. SHIVE
sub-optimal and optimal series agree in showing a general, but inverse, cor-
respondence in this regard. For the sub-optimal series it appears that high
transpiration is correlated with low root yields.
Relation of transpiration to total concentration of the medium. To bring
out the relations between the amounts of water lost by transpiration and the
total concentration of the nutrient solution, graphs of the average tran-
spiration data for each of the three series were prepared, in a manner quite
similar to that employed for the study of top and root yields (figs. 5 and 9).
These three transpiration graphs are shown in figure 12. The actual losses
for the optimal series are here arranged in the order of their magnitudes,
from the highest to the lowest, and are then plotted to form the nearly
straight line of the figure. With the various cultures arranged in the order
oc.
sº Optimal *====s*mºsºme
Sub-optimal — — — — — — —
Supra-optimal
A -
260 — ^ ^ / \ A
f -
- \
2OO – * |! \ , - \ \| | | \
O V | \ ||
•º Nº. ...--> J._*~
150 — TN2. “J/N---- NJ--r' J/\
N--
* 1 RB R3 RAR, R3 RB Rl Ré R2 Ré Rl Rl R5 R3 R4 R3 R5 R6 Rl Rl R2 R3 R4 Rl R4 R2 R7R2 R4 R8 R2 R
C2 C3 C5 C2 C2 C3 C4 C1 C6 C2 c5 C7 C4 C6 C4 C5 Cl C3 C3 C8 C3 C4 G2 C2 C3 C2 Cl C4 ClCl C7 C
| 1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–
FIG. 12. Amounts of water lost by transpiration (cubic centimeters) from wheat
plants grown with sub-optimal, optimal, and supra-optimal total concentrations.
determined by this graph the two other series are plotted on the same scale.
These three graphs show clearly that there is no general tendency for transpi-
ration to be either higher or lower with the sub-optimal than with the optimal
total concentration. On the other hand, it is plain that none of the tran-
spiration data for the supra-optimal series is nearly as great as are the cor-
responding data for the sub-optimal and optimal series. This latter relation
was brought out in the study of the relation of top yields to total concen-
tration (fig. 5), but in that case a pronounced and general difference is ob-
served between the top yields of the sub-optimal and optimal series. Since
no such difference is here manifest, it is clear that the criterion of transpi-
ration, while valuable as indicating large differences in growth (without the
destruction of the plants), is not always as definite in its indications as is the
criterion of dry weight of tops.

tº tº
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 379
Water requirement
Introductory. The relations between the amounts of water lost by tran-
spiration and the dry weight yields of the various cultures may be studied by
means of the ratios of transpiration to yield. Such ratios represent the
water requirements of the plants, being the amount of water required,
in each case, to produce a single gram of yield.
Apparently the first worker to give attention to water-requirement as a
measurable physiological character of plants was Woodward,” whose work
was carried out over 215 years ago. This author found that plants grown in
water with low salt content had a much higher water requirement (per unit
of green weight) than those grown in water containing more dissolved salt.
Many other experimenters have studied the water requirements of plants
grown under various conditions, although water cultures have been em-
ployed in but little of this work. Physiologists do not seem generally to
have realized the importance of this criterion of plant growth, and most of
the experiments reviewed in Briggs and Schantz's” excellent and exceedingly
useful summary of the literature of this subject were carried out by agricul-
turists, and with only approximate control of the various effective conditions.
Sorauer’s experiments with water cultures may, however, be mentioned
here, and also those of Heinrich.
Sorauer” studied rye, barley, wheat and oats, grown 53 days in “normal”
([1885], page 87) nutrient solutions of different total concentrations (from
0.05 to 1.00 “per cent”) and found that the water requirement (per unit of
dry yield) decreased with increase in the total concentration of the medium.
Thus, according to Briggs and Schantz's ([1913), page 55) computations of
Sorauer's results, the water requirement for wheat in these tests was 768=40
when the total concentration of the solution was 0.05 per cent, and only
469 == 18 when the total concentration was 0.50 per cent.
Heinrich.” grew oats to maturity in nutrient solutions containing 4H2KPO4
+CaCl2+5Ca(NO3)2+2MgSO4+2Fe (Briggs and Schantz [1913], page 55),
with total concentrations varying from 0.1 g. to 3.0 g. per liter. The weakest
solution showed a water-requirement (per unit of dry yield) of 642 and the
next to the weakest (0.25 g. per liter) gave a corresponding value of 688.
Throughout the remainder of the series the water-requirement decreased as
the total concentration of the medium increased, and the value obtained
for the plants grown in the solution containing 3.0 g. per liter was 515. Thus,
* Woodward, J., Some thoughts and experiments concerning vegetation. Phil. Trans. Roy. Soc. London 21:
193–227. 1699.
27 Briggs, L. J., and Shantz, H. L., The water requirement of plants. II. A review of the literature. U. S.
Dept. Agric. Bur. Plant Ind. Bull. 285. 1913.
28 Sorauer, P., Nachtrag zu den “Studien über Verdunstung.” Forsch. Geb. Agric. Phys. 6: 79–96. 1883.
29 Heinrich, R., Ueber die Wassermengen, welche die Haferpflanze aus verschiedenen Nährstoff-Concentra-
tionen während ihrer Vegetationszeit verbraucht. Zweiter Bericht öber die Verhältnisse und Wirksamkeit der
Landwirtschaftlichen Versuchs-Station, Rostock. Pages 179–174. 1894.
380 - join W. SHIVE
with the exception of the very weakest solutions (where the plants were
obviously unhealthy), the water requirement is less with more concentrated
solutions and greater with weaker ones, which is in general agreement, with
the conclusion reached by Sorauer.
In the present case the average amount of water lost by transpiration from
each culture in series A and B was divided by the corresponding yield of
tops and also by the corresponding yield of roots. The water requirement
ratios thus obtained, for tops and for roots, expressed in terms of the cor-
responding ratio for culture R1C1, are set forth in table X. The actual
average water requirements (grams of water required to produce a single
gram of dry material) are given for culture R1C1, in each case, this number
being placed in parenthesis below the relative value (unity) of culture R1C1
in the comparative series. The letters L and H indicate, as in previous
tables, that the numbers opposite which they are placed fall within the lower
one-fourth and within the upper one-fourth, respectively, of the total range
of their respective series. º:
Relative water requirements. A study of the six triangular diagrams pre-
pared from the data of table X brings out several interesting suggestions, but
no precise generalizations can be made in this connection, and the diagrams
will not be given here. It may, however, be worthy of mention that, in the
optimal series, low water requirement of tops corresponds, in a general way,
with high top yield, while in the sub-optimal series low water requirement of
roots corresponds with low top yield. These diagrams emphasize the fact
that optimum physiological balance, or the best salt proportions, of the
nutrient medium, to give either low or high water requirement ratios, does
not occur in the same solutions in the three different series. How a given
set of salt proportions affects the plant in this respect is determined by the
total concentration of the mixture.
Relation of total concentration to the actual magnitudes of the water re-
quirement ratios. As in the cases of yields of tops and roots and of tran-
spiration (figs. 5, 9 and 12), the actual water requirements of tops and of roots
have been graphically represented in a linear way, and these graphs are given
in figures 13 and 14. In both cases the data for the optimal series were
arranged in a decreasing order and then plotted as ordinates, with arbitrary,
equal abscissa increments representing the different solutions. The order
of the solutions or cultures is thus not the same in figures 13 and 14. After
this order was determined as the descending series for the optimal con-
centration, the corresponding data for the sub-optimal and supra-optimal
concentrations were plotted on the same chart, in each case. These charts
bring out the general relations between total concentration and water
requirements of tops and of roots.
a. Water requirement of tops. For the water requirement of tops (fig. 13)
it is seen at once that the total concentration of the medium determines the
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 381
TABLE X
A mount of water required for the production of each gram of yield of tops and of roots
(water requirement), relative to that of culture R1C1. The data are averages from
series A and B. a’
SUB-OPTIMAL TOTAL CONCEN- OPTIMAL TOTAL CONCEN- SUPRA-OPTIMAL TOTAL CON-
CULTURE TRATION (0.1 ATM.) TRATION (1.75 ATM.) CENTRATION (4.0 ATM.)
NUMBER
Tops Roots Tops Roots Tops Roots
R1C1 1.00L* 1.00L 1.00 1.00L 1.00L 1.00
(559) (1407) (466) (1802) (363) (1506)
C2 1.08L 1.23L 0.96 1.03L 1.00L 1.09EI
C3 1.13 1.34 0.95 1.22 |. . . . . . . . . . . . . . . . . * . . . . . a
C4 1.20 1.71 1.02H 1.10 …].…..
C5 1.20 1.74 0.92L 1.17 l. . . . . . . . . . . . . . . . . . . . . . . .
C6 1.25H 1.76 0.94 1.15 . . . . . . . . . . . . . . . . . . . . . . . .
C7 1.23H (?) | 1.98H 1.06H 1.15 1. . . . . . . . . . . . . . . . . . . . . . . .
C8 1.28H 1.99H 0.99 1.21 1.07 1. 11H
R2C1 1.16 1.00L 1.00 1.07L 1.05 0.93
C2 1.10 1.31 1.00 1.07L . 1.00L 1.01
C3 1. 11 1.65 0.92L 1.24H 0.97L 1.05.H
C4 1.11 1.76 0.89 L 1.18 |. . . . . . . . . . . . . . . . . . . . . . . .
C5 1.13 2.07H 0.91L 1.21 . . . . . . . . . . . . . . . . . . . . . . .
C6 1, 12 2. 13H 0.96 1. 19 0.96L 1.05H
C7 1.22 2.09H 0.91L 1.07L |. . . . . . . . . . . . . . . . . . . . . . . .
R3C1 1.21 0.95L 0.93L 1.06L . 0.96L 0.92
C2 1, 24H 1.34 0.96 1. 12 1.05 1.02
C3 1.13 1.51 0.93L 1, 18 1.00L 1.07H
C4 1.12 1.86H 0.90L 1.21 0.97L 1.10H
C5 1.06L 1.94H. 0.92L 1.25H 0.96L I.02
C6 1.31H 1.90H 1.00 1.25H 1.10H 1.00
R4C1 1.28H 0.97L 0.99 1.07L 1.04 0.87L
C2 1.26H 1.24L 0.90L 1.05L 0.97 L 0.92
C3 1.13 1.53 0.90L 1.25H 0.95L 0.98
C4 1.05L 1.85H 0.91L 1.29H 1.02 0.98
C5 1.06L 1.97H 0.98 1.21 1.02 1.03
R5C1 1.21 1,00L 0.96 1,08I. 1,08BI 0.84L
C2 1.03L 1.26 0.91L 1.18 1.01 0.99
C3 1.08L 1.69 0.96 1.29H 1.00L 1.07H
C4 1. 12 2.05H 0.92L 1. 14 1.02 1.00
B6C1 1.22 1.01 L 1.01 1.11 1.07 0.87L
C2 1.03L 1.41 0.99 1.25H 0.981, 0.94
C3 1. 11 1.64 0.96 1.32H 1.05 1.01
R7C1 1.26H 0.98L 0.98 1. 10 1.03 0.84L
C2 1.18 1.18 0.95 1.23 1.05 0.90L
R8C1 1. 19 0.96L 1.03H 1. 10 1. 13H 0.86L
K++ 1.29 1.94 1.02 1, 18 0.98 1.02
Tº: 1.02 1.66 0.96 1.20 0.97 0.95

* The water requirement for culture R1C1 is always taken as unity, and the other water requirements are
expressed in terms of this. The actual water requirement for culture R1C1 is given in parenthesis, in each case.
** K and T represent Knop's solution and Tottingham's best solution respectively; these are introduced for
comparison.
382 JoHN W. SHIVE
CO Optimal assºmºmºsºms
per tº Sub-optimal — — — — — — —
grm A Supra-optimal
sº * * F-- A
7OO N A \ l \ N /\ N f \
\----" " | \ ! N f \ / \ FV f, ſ \
" ſº | | w ! \ /N, \, / \ ſ \ /
l \ | \ | \ | \ A \, v/ \ ~~\ I V \
| ſ \, i \ | NS J \ A \ / * * *
600 – * M \ | \! s= sº v’ \ |
\ } V \ ^
g \!
500 -
400 -
* | RIRs R1 RS R3 Ri R8 R2 R&R. R4 RAR7Rs R2 55 R1 R& RS FZR1 Ri R3 R3 Rl 52 R3 RSR2R3 RA R2 R4 RAR3 Re
C 701 º: Ol C6 G1 Cl C2 C2C8 C1 C5 Cl Cl C6 C3 C2 C2 C3 C2 ca C6 Cl C3 C5 C3 C5 C4 C5 gº C4 C7 C3 C2 C4 C4
1–1–4–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1– | | | | | |
FIG. 13. Water requirement (cubic centimeters) per gram of wheat tops grown in
sub-optimal, optimal, and supra-optimal total concentrations.
C C e
per Optimal *===
grim Sub-optimal - - - - - - -
Supra-optimal —
3OOO = ,”
\ ,” |
/ 2^ |
** ~ * \ ,’ | 11
IS-, z l P \ iſ'
( \ º l f ! | l
A l \ I \ f | , t
2500 º / \ | \ | \ 1-----' | |
22’ I \ I \ l l | |
-T-I-Hi M. A | w ſ | \ t
t i l 7-- l \ | ! | |
d \ | \ || | | | I
| || \ } \|| \ } = ! I | |
2000 - | W W! \ | = +- | |
- I | | |
! \ | | | M | º /
- | f* ~. f
l -“TN | | | | | | | | | A \!
1500 - ^ A-Z \----|-- -k | \ i A\ /#, Iſ
N J \ | * ~, / V
. . \-l. \Aſe
* as - - -
1OOO - - g
R6 R5 R4 R7 R6 R3 R3 R4 R2 Rl R3 Rl R4 R2 R2 R3 R5 R2 Rl Rl Rl R5 R6 Rl R7 R8 R5 R2 R4 R2 R2 R3 R4 Rl Rl R3
gº ca gº ºf gº C5 C6 º C3 C3 C4 C8 C5 C5 C6 C3 C2 C4 C5 C6 C7 C4 Clcá C1 Cicl Cz cl & Cº cl ca cºol &
- L. l. I |—|—|—|−1–1—1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1–1
FIG. 14. Water requirement (cubic centimeters) per gram of wheat roots grown in
sub-optimal, optimal, and supra-optimal total concentrations.

PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 383
general position of the graph. There is here no crossing of the graphs; the
highest water requirements of tops are all shown as occurring in the sub-
optimal series, the lowest requirements are those of the supra-optimal series,
and the optimal series exhibits water requirements of tops that occupy an
intermediate position. A clear generalization is here indicated: the higher
the total concentration the lower is the amount of water required for the
production of a gram of dry tops. This conclusion is perfectly definite, for
the conditions of the present study at least, and appears to agree with what
might be expected from an a priori consideration of the problem in hand.
It appears that water loss by transpiration is retarded more by high concen-
tration of the nutrient medium than is the production of yield of tops, and
it seems quite clear that this is due to a physical (rather than to a chemical)
property of the solution, being probably related directly to osmotic phenomena.
With more concentrated solutions the resistance to water absorption by the
roots is increased, and this resistance, in turn, decreases the transpiring
power” of the plants.
The arrangement of these three graphs of water requirements of tops is
entirely different from that of the graphs of dry weights of tops (fig. 5).
The optimal total concentration gives highest yields of tops and medium
water requirements of tops, the supra-optimal concentration gives medium
yields of tops and lowest water requirements, and the sub-optimal con-
centration gives lowest yields and highest water requirements. It appears
that a very important generalization (as far as the water relations of plants
are concerned) is here touched upon, one that will surely repay more thorough
study.
b. Water requirement of roots. The graphs of figure 14 indicate that
there is no clear relation between water requirement of roots and total
concentration of the medium, for the sub-optimal and optimal total con-
centrations. On the other hand, these graphs do show a very clear relation,
as to the water requirements of roots, between the optimal and supra-optimal
total concentrations. The root water requirement is almost always markedly
lower, for any set of salt proportions, in the supra-optimal solutions than in
the optimal ones. The relations are here about the same as for the water
requirements of tops, as far as these two higher total concentrations are
concerned, and the same condition holds in comparison with transpiration
(fig. 12). In a general way it may be said that the root water requirements
for the sub-optimal concentration are lower and more nearly like those for
the optimal concentration than is the case with the water requirements of
tops.
30 Livingston, B. E., and Hawkins, Lon. A., The water-relation between plant and soil. Carnegie Inst. Wash.
Pub. 204: 3–48. 1915.
Pulling, H. E., and Livingston, B. E., The water-supplying power of the soil as indicated by osmometers.
Carnegie Inst. Wash. Pub. 204: 49–84. 1915.
384 JOHN W. SHIVE
Comparison of figures 5, 12, 13 and 14, shows that the relation between
the optimal and supra-optimal total concentrations is much the same, whether
this relation is judged by the criterion of dry weight of tops, by that of
transpiration, by that of water requirement of tops, or by that of water
requirement of roots.
5. GENERAL COMPARISON OF QUANTITATIVE PLANT DATA
Relation between the areas of low, medium, and high relative plant values, as
shown on the various triangular diagrams
The quantitative plant data that have been presented in the preceding
sections (those of yield of tops, yield of roots, transpiration, water require-
ment of tops and water requirement of roots) have been studied with refer-
ence to their distribution into areas of low, medium, and high values on the
triangular diagrams. A range of total concentration—the physical property
of the nutrient medium—from 0.1 atmosphere to 4.0 atmospheres has been
studied, this total range being represented in the experiments by the limiting
concentrations just mentioned and by an optimal concentration of 1.75
atmospheres. For each kind of plant measurement three triangular diagrams
have thus been required, each diagram corresponding to one of these three
total concentrations. The chemical character of the nutrient medium,_
the proportions of the three main component salts, has been studied by
varying the salt proportions, in each of the three total concentrations, from
osmotic proportions of 1 : 1 : 8 to those of 1 : 8 : 1 and to those of 8 : 1 : 1,
the increments of variation being always one-tenth of the total diffusion
tension or possible osmotic pressure of the solution as a whole. The total
range of salt proportions has thus been represented by 36 different sets of
proportions in each case, each set of proportions having its particular lo-
cation on the triangular diagram. The position of a given solution on the
diagram thus characterizes it as having a definite set of salt proportions (ion
ratio values, etc.), or as having a certain complex of chemical properties, and
the particular one of the three diagrams on which a given solution is located
characterizes it as having a certain physical character determined by its
total concentration.
It is of course clear that salt proportions other than those actually employed
might have been tested, and that each of these would have had a definite
location on the triangle. It is also obvious that total concentrations other
than the three here employed might have been included in the experiments,
and each one of these would have called for a separate triangle of its own.
The whole system of possible variations in the chemical and physical prop-
erties of the nutrient solutions here dealt with may be pictured by means
of a solid diagram, a right prism with an equilateral triangle as base. Passing
from below upward, each successive horizontal (triangular) section of this
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 385
prism may represent a different total concentration of the nutrient solution,
and each vertical line passing through the prism (thus cutting all the hori-
Zontal triangular sections) may represent a different set of salt proportions.
The various average values obtained for the five different plant measure-
ments considered have been studied by plotting them on their respective
diagrams, always expressing each value in terms of that for solution R1C1
of the same diagram, and the area of each diagram has been separated into
three portions, designated as the areas of low, medium, and high values.
Values of any plant measurement that lie within the lower one-fourth of the
total range of that particular measurement, for that particular total con-
centration, have been termed low values; those lying within the upper one-
fourth have been termed high values; and all other values for that kind of
measurement and for that total concentration have been termed medium
values. This notation simplifies the problem of comparing a large number
of measurements, and the grouping of the thirty-six different values into but
three classes should tend to avoid many complications arising from smaller
variations in the values, brought about by the influence of unknown condi-
tions upon the growth and development of the plants.
To compare all the different plant measurements and to relate them to the
chemical and physical conditions of the solutions in which the plants were
grown presents a somewhat complicated problem, which has been attacked
only in part in the foregoing discussions. For each kind of plant measure-
ment it is necessary to determine how the value alters with its position on the
triangle and also how it alters with the position of the triangle in the prismatic
diagram. The sub-optimal diagram of these studies may be considered as
the lower base of the prism in question, the supra-optimal triangle being the
upper base, while the optimal triangle is a horizontal section of the prism
between the two bases. Each kind of plant measurement requires a prism
of its own and a not unimportant part of the interpretation of the results
involves comparisons between similar areas or points in the different pris-
matic diagrams.
Some of these various and complicated relations have been brought out
in the foregoing discussions and still others will be evident from the tabular
summary of all quantitative plant measurements presented in table XI.
This table shows the relative value (low, medium, or high, denoted by the
letters L, M and H) of each average plant measurement (series A and B
combined) for each of the three different total concentrations and for each of
the thirty-six different sets of salt proportions. The salt proportions are
denoted by the solution numbers, in the first column, and the total con-
centrations are shown by the words sub-optimal, optimal, and supra-optimal,
as heretofore. Table XI is virtually a summary of all the triangular diagrams,
and of the columns of averages given in tables III, IV, V, IX, and X, in terms
of low, medium and high values. From it may be determined the relations
386
John W. SHIVE
TABLE XI
Summary of growth measurements of plants grown in 8-salt solutions of 86 different sets
of salt proportions and of three different total concentrations, in terms of low (L),
medium (M), and high (H) values, from tables III, IV, V, IX, and X.
-ao"№s| = |E | | | || + |x = * || ſr ||======| =====| ====|===| ==|=| - - -
#ŠſeurſądO========| =======| ======| =====| ====|===| ==>| - = •
ſą
>
}-ndoſës！\ , Q × × × × țE ȘI:=======| ======| =====| ====|===| ==|=| ===
3Țeurſ,�
:-doºfhäus | H H | | | | | > | > − − | | − ||======| =====| ====|===| ==|=| ===
ſzą
£ €∞
šģl'œuĪĻĢdO ! >: 's > №r; , q ≥ ± − ×=======| ===========| ====|===| ==|=| ===
-maaºſs | – — ± = = = = =× × × × × × × | == = = = = | = = = = = | ====| ===| == |* |* a *
3 | leurſąđo eidns | - = | | | || z | = = = || ſr ||======| =====| ====| ===|== | -|• 3: oo
£{
<
șeando|º=======| =======| ======| =====| ====|===| ==|=| • • •
Z
<
ğ | Ieurņđo-qns= = = = = = = = | = = = = = = => №r; ſı; ſı; ſı; ſı:=====| ====|===| ==|=| - ==
{[eu：Iņdo e IdnŞ | , q | q | | | || = | = = = || > | | × × × × × ×=====| ====|===| ==|=| • • •
-}
ſą
{mundo | x = = = = = = = | = = = = = = =======| =====| ====|===| ==| |s še •
O
O
* | ſºuņđo-qns| = = = = = = = = | = = = = = = == = = = = = | = = = = =====| ===| ==ķī: | ）; № co
Ieuaņđo eIdnS | , q = | | | | | = | , } = {I} | | ||B |= = = = = = | = = = = =====| ===| ==~ | ~ ~ ~
£ſeurſąđO= = = = = = = = | = = = = = = => > ķIĘ > > > | > > > > ķE====| ===| ==| |cº g} +
►
ſ！
! | retando quș| − × × × × × × × | − × × × × × × | - = = = F = | — z = H H----|---|--|-oo ş; +
£
wae∞ſºsșaecºrovaeſ eNQvºº cºq co\ſ)！=）cºro！ cºq caelvae eºſș）„3 × þr:
„, „g|ğ § § č Š Š 5 & | ğ § § § 355 | ğ § § § 35 | ğ $ $ $ $ | § § § 5 || ğ § 3 | $ $ | ğ | ±±
eròÞ<-oqcº O
ſåſåpåſåſåpåpå£，§ 6
€）. №


PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 387
holding between the different kinds of plant measurements for the same total
concentration and also the relation holding between the same kind of plant
measurements for different total concentrations and for different sets of salt
proportions.
Table XII presents a correlation summary of all the quantitative plant
data and shows the distribution of the high, medium, and low values obtained
for each single criterion (yield, transpiration, and water requirement) as
compared with those for the other criteria, the data being derived from those
of table XI. The table as a whole is divided into an upper, middle and
lower Section, each of these sections referring to one of the three total con-
centrations here studied. It is divided vertically into four sections. The
first of these presents the comparison between the yields of tops, on the one
hand, and yields of roots, transpiration, and water requirements on the other.
The second section gives the comparison between the root yields, on the one
hand, and transpiration and water requirements on the other. The third
section gives the comparison between transpiration, on the one hand, and the
water requirement of tops and that of roots, on the other. The fourth and
last section gives the comparison between the water requirement for tops and
that for roots. The letters H, M and L, stand for high, medium, and low,
as these terms have been used in table XI and in the preceding discussion.
The meaning of the numerical data of table XII is best brought out by
examples. For the sub-optimal concentration, out of a total of four cul-
tures giving high yields of tops, one gave medium, and three gave low yields
of roots, all four gave high transpiration values and high water requirement
for roots, and one gave medium and three low water requirement for tops.
Similarly, of 24 cultures giving medium yields of tops, two gave high, 14 gave
medium, and eight gave low yields of roots. All of the comparisons are
stated in this manner. It will be seen that table XII may also be read in the
vertical direction thus stating the comparison in the reverse order. For
example, out of eight cultures with sub-optimal total concentration giving
high root yields, two gave medium, and six gave low top yields. In order
to bring out the dominant agreements and disagreements encountered in
these comparisons the individual numerical data are marked in four different
ways. Those numbers are printed in italics which represent more than half
of the total number of cultures belonging in the group indicated by the letter
(H, M, L) on the same line and at the left of that vertical section of the table.
Thus, for the sub-optimal total concentration, the three cultures giving low
root yields comprise more that one-half of the four cultures giving high top
yields, and so this figure 3 appears in italics. Similarly, the 14 cultures
giving medium root yields comprise more than half of the 24 giving medium
top yields and the figure 14 is italicized. Reading the comparisons verti-
cally, all those numbers that represent more than half of the total
number of cultures belonging to the group indicated by the letter at the

TABLE XII,
Relative distribution of high, low and medium dry weights, transpiration, and water-requirements for each of the three total concentrations
here used, being a summary of the average data given in tables III, IV, V, IX, X and XI.
§
COMPARISON OF TOP YIELDS WITH ROOT YIELDS, TRANS- COMPARISON OF ROOT YIELDS WITH TRANS- COMPARISON OF TRANSPIRATION jº. OF
PIRATION, AND WATER REQUIREMENT PIRATION AND WATER REQUIREMENT WITH WATER REQUIREMENT º: º
OF TOPS AND ROOTS OF TOPS AND ROOTS OK TOPS AND ROOTS £
- - AND ROOTS
TOTAL -
CONCENTRATION º e g
Water-requirement Water-requirement Water-requirement W Wat,
Yield Yield Transpira- Yield | Transpira- Trans- Yºr }: er
tops TOOtS tion TOOtS tion piration º Kºłº
Tops Roots Tops Roots Tops Roots Op
H M L L | H M L | H M L H M L | H M L | H M L H M L | H M L H M L
Sub timal 0.10 * — 1 3 — — – 1 3 || 4 — — H — 8 — 3 5 — | – 1 7 H 4 & 3 || 9 6 — H 3 2 3
º & M 2 14 8 || 11 13 — 6 1 4 4 || 7 15 2 | M 7 8 2 || 2 10 5 || 2 12 3 || M. 4 II 4 || 2 9 8 || M | 5 10 15
8t,DOl. L 6 2 — | – 6 2 5 1 | — — 8 || L 8 3 — | 3 5 3 || 9 2 — | L | – 1 1 || – – 2 | L 3 3 2
H 2 2 — 4 — — . — 2 — 4 — H 3 6 — 1 5 3 | — 5 4 EI — 4 2 1 5 — H — 3 —
Optimal 1.75 atm. M 7 12 10 || 2 26 1 || 2 15 12 || 8 15 7 || M 2 12 2 9 6 — I I 5 || M 3 14 11 || 7 14 7 || M 4 9 6
L | – — — 1 1 — ? — I — — 2 | L 1 9 — — 5 5 || 8 2 — 1 1 — — 2 | L 4 7
Supra-optimal H 1 4 — || 4 1 — | — 1 5 || 3 2 H 2 3 1 || 1 3 2 | – 3 3 || H 1 3 4 || 3 5 EI — 1 9
* M | 4 13 1 || 4 13 1 || 1 1 0 7 || 4 I I 3 || M | 6 10 4 || 2 9 9 || 6 || 1 3 || M | – 7 7 || 4 7 3 || M. 1 7 4
. UU at In. L 1 3 — — 5 2 2 1 | – 2 L | — 1 1 | — — ? | 1 1 — | L 2 2 2 | — 3 L 6 7 —

PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 389
head of the column in which they occur are shown in black face type. Thus,
for example, the six cultures giving low top yields comprise more than half
of the eight giving high root yields, so that the figure 6 is in black type,
but this same figure becomes also italicized because six is more than half of
the eight cultures giving low root yields. Similarly, the figure 14 mentioned
above as italicized is also in black face type, because 14 is more than half
of the 17 cultures giving medium root yields. By this notation it is clear
that the best correlations are indicated by black italics, where the agreement
is good in both directions. Good agreement in the horizontal direction of
comparison is indicated by italics and good agreement in the vertical direction
is indicated by black type, which may be either italic or roman. Many of
the points thus brought out have already been mentioned in the preceding
discussions and space need not be here devoted to further consideration
of these interesting agreements and disagreements.
Relation between the actual values for the various plant measurements
Finally, figure 15 is here presented to bring out the relations that hold
between the actual values of the various plant measurements, which are
represented graphically in much the same manner as that employed for
figures 5, 9, 12, 13, and 14. In this case, however, the various different salt
proportions (indicated at the base by the usual solution numbers) are all
arranged in the decreasing order of their magnitudes for top yields of the
optimal Series. As in the previous graphs, a narrow line represents the
optimal total concentration, a broad line represents the supra-optimal, and
a narrow broken line represents the sub-optimak concentration. It will
be seen that the figure shows five groups, each one of which comprises three
graphs. The lower group (A, yield of roots) and the second group (B, yield
of tops) are all plotted on the same scale and are directly comparable, as is
indicated by the numbers at the left. The graphs of the third group (C,
transpiration) are plotted on a different scale, which is indicated on the left
margin. The fourth and fifth groups (D, water requirement per gram of
tops, and E, water requirement per gram of roots) are all plotted on the
same scale but the group E as a whole has been displaced downward in order
to save space, as is clear from the ordinate values given at the left.
Inspection of figure 15 will bring out many of the relations between the
various plant measurements, on the one hand, and the salt proportions and
total concentration of the solution on the other, also between the different
plant measurements for the same total concentration, the same set of pro-
portions, etc. These graphs again emphasize the points already brought
out, that the total concentration of the medium clearly determines the yield
of tops (group B) and also the water requirement of tops (group D), for all
salt proportions here tested; the order of magnitude with reference to total
concentration is the same for all salt proportions in the case of either one of these
C. C.A'ſ AP
GAP/7
3000.
27:30.
2%0.
/300.
l
l
2/0. º
|
W l
t . ſ L- l
l | l
fº ºn l t | A t
/OL/L/T * \ l f * * t
| z l | ! \ l
u,” z'\ || || t \ |
\| 1 l
ſ º
f t
I !
700- ,">|----- |- - - - - - "* 22°-----,
* +-- * • * * ~1 - N
600. e ---------' * gº **
300.
46%. _* ---------|--~|_----it TN
JOOH- &
C. C.
* ** -
23OH ,’ \ z ^{< - * - , TV
27NZV * z w z * , ‘ *-i- * 2 ” * - - - , ... * - - - ,’ \,
e ~2 \ ~|~~ *N* 's -7s \\ 7NXS, *
º & "- - -]. - " w *
2 * *
~|~~ *
* - - -
421-Tº-y/S-TN/~ `i–Tºrºſs || ---H--S_ 's,
~~
2^
P--
**-
GA/M \
a: N
l B
Q45. V 2^ ,”
(240. n / * N
f \ - - - * ~ * *
\ * --~~" \ M * = --~~ /*
(255. \,' \ . ,"|\, 21, |\, .
"L’ \ ſ \ A. f \
(230– ` / \ A. \ f \
- - \ | f | \
\,, f `s,
025F - LEGEMD *
OPT////70 – 5 U8-OA 7/MAL ---- 50PRA-OP7/MAL–
(220+ A-Roor Yeza C-Trawserarrow.
- B- Top yield D-Warer/reg Toes.
E- Warer Reg Roors.
Q/5-H
Aſ At, N
sº-TS ,” * * ". t
a/2+\}^*= *N4–f
Ağājā//?7/º/ā75757.5/574727.3777/75/37&/º3/3/737.377577777757-757777737)
C2 C3 C3 C2C5 (2 C-466 (4 C/ C4 (4C3 C3 C3 (5 C2 C3 C7 C6 C3 C2C2 (5 (8 C/ C4 c6 6/ C/C2C/C/C7C/ C/
FIG. 15. Graphic summary of plant measurements.
390




PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA 391
two plant measurements, but the order for yield of tops is not at all the same
as for water requirement of tops. The water requirement of tops is less as
the total concentration of the medium increases, as has been found by other
writers. The yield of tops, on the other hand, is greatest for the optimal,
less for the supra-optimal, and least for the sub-optimal total concentration.
Also, for transpiration (group C), the supra-optimal total concentration ap-
pears always to give the lowest value, for all the different sets of salt pro-
portions, and the same generalization is true for nearly all the sets of Salt
proportions in the case of the water-requirement of roots (group E).
For yield of roots (group A) there is no general relation between the dry
weights and the total concentration of the medium, but each particular set
of salt proportions seems to have its own relations in this regard. Thus,
solution R3C3 gives the highest root yield with the optimal total concentration,
a medium yield with the sub-optimal, and the lowest with the Supra-optimal.
The same is true of solutions R5C2, R4C5, R1C4 and R2C2, but this is not
true for any of the other sets of salt proportions. A similar shifting of the
order of magnitude of the plant measurements, as related to the sub-optimal
and optimal total concentrations, for different sets of Salt proportions, is
seen in the case of transpiration (group C) and water requirement of roots
(group E), as has already been mentioned.
On the whole, the graphical summary of figure 15 shows clearly enough
that the effect of any set of salt proportions is not, in general, relatively .
the same for different total concentrations, at least with the exception of the
water requirement of tops, where there is considerable parallelism between
the graphs for the optimal and supra-optimal series. It also shows that the
effect of any total concentration is not generally the same for different sets
of salt proportions. This summary shows, further, that root characters and
top characters, of wheat grown as in these experiments, do not generally vary
in the same way with reference to the conditions of the medium. Finally,
it shows that different criteria for plant measurement may generally
be expected to exhibit more or less pronounced differences in the relations
between growth and the properties of the solution in which the plants are
grown. Some of the details and exceptions to these generalizations are
best brought out by study of the triangular diagrams previously given.
PHYSIological RESEARCHES, vol. 1, No. 7
sERIAL No. 7, Nov EMBER, 1915
392 * JoHN W. SHIvE
COMPARISON OF DRY WEIGHT YIELDS OF TOPS GROWN IN
THE BEST SALT PROPORTIONS OF THE THREE-SALT SOLU-
TION, WITH THE CORRESPONDING YIELDS FROM SOLU-
TIONS EMPLOYED BY OTHER AUTHORS
As previously stated, Knop's solution and Tottingham's best solution for
tops (T3R1C4) were included in each of the three experimental series here
carried out, for purposes of comparison.
For the sub-optimal series (0.1 atm.) the highest average yield of tops
(culture R4C5) was 1.65 times the top yield obtained from culture R1C1.
The average yields of tops obtained from Knop's solution and from Totting-
ham's best Solution, both having the same total concentration as culture
R1C1 (0.1 atm.), were 1.35 and 1.62, respectively. Thus the three-salt
Solution with proportions giving the highest yield of wheat tops, shows an
improvement of 22 per cent over Knop's solution, while Tottingham's best
Solution shows an improvement over Knop's solution of 20 per cent, all for
a total concentration of 0.1 atmosphere.
For the optimal series the highest average yield of tops (culture R5C2)
was 1.39 relative to the yield from culture R1C1. The corresponding yields
obtained from Knop's solution and from Tottingham's best solution for tops
were 1.09 and 1.27, respectively, relative to the same control. For this total
concentration (1.75 atm.) the best salt proportions of the three-salt solution
produced an average dry weight of wheat tops 28 per cent higher than the
yield obtained from Knop's solution with the same total concentration, while
Tottingham's best Solution for wheat tops produced a corresponding average
yield 17 per cent above that obtained from Knop's solution.
For the supra-optimal series (4.0 atm.), the highest average dry weight
of tops (culture R4C3) was 1.38 relative to the yield obtained from culture
R1C1. Knop's solution and Tottingham's best solution for tops, with the
same total concentration, yielded corresponding dry weights of 1.05 and 1.23,
respectively. The best salt proportions of the 3-salt solution with this
total concentration (4.0 atm.) showed an improvement of 31 per cent over
Knop's solution, while Tottingham's best solution showed a corresponding
improvement of 17 per cent.
In order to compare 3-salt solution R5C2, which produced the highest
yield of wheat tops in the optimal series of the present studies, with Totting-
ham's best solution for wheat tops and with Knop's solution, all with the same
total concentration as was employed by Tottingham (2.50 atm.) a series of
three cultures of each of these three solutions, all having a total concentra-
tion of 2.50 atmospheres, was carried out during the 23-day period from
February 25 to March 19, 1915. The average dry weight of tops produced
by three-salt solution R5C2, relative to the average yield produced by
Knop's solution was 1.22, and the corresponding average yield from Totting-
i
M
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA
393
ham's best solution for tops was 1.07. Therefore, with a total concentration
of 2.50 atmospheres, Tottingham's best solution produced a yield of tops
7 per cent higher than that obtained from Knop's solution, while the three-
salt solution with the best salt proportions for wheat tops produced a cor-
responding yield 22 per cent higher than that produced by Knop's medium.
The data for all these comparisons are presented in table XIII, which gives
the volumé-molecular partial concentrations of the solutions and the average
dry weights of tops, relative to the yields obtained from the same total
concentration of Knop's solution taken as unity. The actual values ob-
tained with Knop's solution are given in parenthesis below the assumed
relative values of unity.
TABLE XIII
Comparison of yields of wheat tops obtained from various total concentrations of Knop’s
solution, Tottingham's solution No. T3R1C4, and 8-salt solution No. R5C2.

# YIEI,D OF
# Z. PARTIAL MOLECULAR CONCENTRATIONS tº-
# É NAME OF SOLUTION THAT FROM
§ : KNO3 KH2PO4 Ca(NO3)2 MgSO4 sººn
atm g.-mol. per l.g.-mol, per l.g.-mol, per l.g.-mol. per l.
Knop's solution. . . . . . . . . . . . . 0.00028 || 0.00022 || 0.00070 0.00024 1.00
0.10 Tottingham’s solution & (0.3527)
& No. T3R1C4. . . . . . . . . . . . . . 0.00062 || 0.00019 || 0.00058 0.00046 1.20
3-salt solution No. R5C2 . . . . . . . . . . . . . 0.00082 || 0.00074 || 0.00028 1.22
Knop's solution. . . . . . . . . . . . . 0.0059 || 0.0044 || 0, 0145 || 0.0050 1.00
1.75 Tottingham’s solution. . . . . . . (0.4485)
gº No. T3R1C4. . . . . . . . . . . . . . 0.0034 || 0.0108 || 0.0101 || 0.0081 1.17
3-salt solution No. R5C2. . . . . . . . . . . . . 0.0180 || 0.0052 0.0150 1.28
Knop's solution. . . . . . . . . . . . . 0.0083 || 0.0063 || 0.0210 0.0070 1.00
2.50 Tottingham’s solution : - (0.6807)
& No. T3R1C4 . . . . . . . . . . . . . . 0.0049 || 0.0156 0.014.4 0.01.16 1.07
3-salt solution No. R5C2 . . . . . . . . . . . . . 0.0257 || 0.0074 || 0.0214 1.22
Knop's solution. . . . . . . . . . . . . 0.0198 || 0.0147 0.0488 || 0.0166 1.00
4.00 Tottingham's solution (0.3774)
& No. T3R1C4. . . . . . . . . . . . . . 0.0114 0.0354 || 0.0336 || 0.0278 1.17
3-salt solution No. R5C2. . . . . . . . . . . . . 0.0368 0.0198 0.0428 1.31
The comparative efficiencies of the 3-salt solutions R5C2 and R3C3 of the
optimal series, with the best and second best salt proportions for dry
weight yields of wheat tops, respectively, and of various other nutrient
solutions commonly employed by workers in physiology, were determined
by a special experiment. This experiment consisted in a duplicate series of
cultures, all having a total concentration of 1.75 atmospheres, which was
conducted during the 24-day period from December 17, 1914, to January
10, 1915.
The following eleven solutions devised by earlier writers were included
394 JOHN W. SHIVE
in this comparison. The formulas are given in the list below, in terms of
gram-molecules per liter (m). A trace of ferric phosphate was added to the
Solutions for which iron is not mentioned in the formula. As thus made up
these solutions all agree with the two 3-salt solutions with which they are
compared in having a total concentration of 1.75 atmospheres.
Birner and Lucanus” solution: Ca(NOs), 0.0133 m; KH,POs, 0.0108 m;
MgSO4, 0.0061 m; Fea(PO4)2, 0.0043 m.
Crone’s” Solution: KNO, 0.0266 m; Cas(PO.), 0.0022 m; Fea(PO.),
0.0019 m; MgSO4, 0.0056 m; CaSO4, 0.0049 m.
Detmer's" solution: Ca(NOs), 0.0149 m; KHAPO, 0.0035 m; MgSO,
0.0051 m; KCl, 0.0083 m. e
Hartwell,” Wheeler and Pember's solution: Ca(NO,), 0.0136 m; CaFI,(PO.),
0.0027 m; MgSO4, 0.0077 m; KCI, 0.0075 m.
Knop's [1862] solution: KNO, 0.0059 m; KHAPO, 0.0044 m; Ca(NO3),
0.0145 m; MgSO4, 0.0050 m.
Pfeffer's" solution: Ca(NO3)3, 0.0136 m; KNO3, 0.0056 m; KH, PO,
0.0041 m; MgSO4, 0.0046 m; KCl, 0.0037 m.
Sachs’ [1860] solution: KNO3, 0.0155 m; Cas(PO.), 0.0025 m; MgSO4,
0.0065 m; CaSO4, 0.0075 m; NaCl, 0.0067 m.
Schimper’s” solution: Ca(NO3)3, 0.0118 m; KNOs, 0.0048 m; KH, PO,
0.0028 m; MgSO4, 0.0040 m; NaCl, 0.0083 m.
Schreiner and Skinner's solution: NaNO3, 0.0278 m; CaFIA (PO4)2, 0.00066 m;
K2SO4, 0.0030 m.
Tollens” solution: Ca(NO3)3, 0.0128 m; KNOa, 0.0052 m; KH, PO4,
0.0038 m; MgSO4, 0.0049 m; NaCl, 0.0054 m.
Tottingham's [1914] solution: KNO,0.0034 m; KH2PO4,0.0108 m; Ca(NO3),
0.0101 m; MgSO4, 0.0081 m.
Table XIV presents the results of this comparative study. The notation
of this table is self-explanatory. The solutions are arranged in the order of
the total dry weight produced.
From table XIV it is clear that the two 3-salt solutions here tested (R5C2
81 Birner, H., and Lucanus, B., Wasserkulturversuche mit Hafer in der Agric.-Chem. Versuchsstation zu
Regenwalde. Landw. Versuchsst. 8: 128–177. 1866. Note that this solution contains the same main salts as
do the 3-salt Solutions that are the subject of the present study, and that its proportions are nearly the
same as those of 3-salt solution R3C0. The iron of Birner and Lucanus' solution is supplied as ferrows
phosphate, however.
3? Crone, G., Ergebnisse von Untersuchungen über die Wirkung der Phosphorsäure auf die höhere Pflanzen
und eine neue Nährlösung. Sitzungsber Neiderrhein. Ges. Nat-und Heilk. Bonn. 1902. Pages 167–173.
* Detmer, W., Practical plant physiology, translated by S. A. Moor. London, 1898. Page 2.
* Hartwell, B. L., Wheeler, H. J., and Pember, F. R., The effect of the addition of sodium to deficient amounts
of potassium upon the growth of plants in both water and Sand cultures. Ann. Rep. Rhode Island Agric. Exp.
Sta. 20: 299-357, 1907.
35 Pfeffer, W., The physiology of plants, translated by A. J. Ewart, 1: 420. Oxford, 1900.
3° Schimper, A. F. W., Zur Trage der Assimilation der Mineralsalze durch die grüne Pflanze. Flora 73: 207–
261, 1890.
*7 Tollens, B., Ueber einige Erleichterungen bei der Cultur von Pflanzen in wässerigen Lösungen. Jour.
Landw. 30: 537–540. 1882.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA - 395
and R3C3, total concentration 1.75 atm.) proved to be as good as or better
than any of the other solutions mentioned, when employed with this total
concentration. These two 3-salt solutions may be considered as practically
equal in the production of top yields and of total dry weight, but it appears
from the data just given that solution R5C2 is considerably better suited to
the production of root yields than is solution R3C3. This latter indication
is not clearly borne out by the other tests of these solutions, however. From
table IV it is seen that solution R3C3 (1.75 atm.) gave in one instance (series
A) a somewhat higher yield of roots than did solution R5C2 in the same
test. On the other hand, in another test (series B) the relation between
TABLE XIV
Relative dry weights of tops and of roots of wheat grown 24 days (Dec. 17, 1914, to Jan. 10,
1915) in various nutrient solutions, all having a total osmotic concentration of 1.75
atmospheres.
Sachs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.00% 1.00 1.00
(0.362) (0.134) - (0.497)
Schimper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.95 1. 14 1.00
Detmer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.95 1.22 1.03
Tollens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.28 1. 10 1.23
Schreiner and Skinner. . . . . . . . . . . . . . . . . 1.42 0.83 1.26
Hartwell, Wheeler and Pember. . . . . . . . 1.55 0.96 1.39
Pfeffer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.52 1.13 1.41
Knop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.60 1.07 1.45
Tottingham. . . . . . . . . . . . . . . . . . . . . . . . . . 1.77 1, 13 1.60
Crone... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.95 1.05 1.70
Birner and Lucanus. . . . . . . . . . . . . . . . . . 1.93 1.17 1.72
Shive, R3C3. . . . . . . . . . . . . . . . . . . . . . . . . . 1.93 1. 16 1.73
Shive, R5C2. . . . . . . . . . . . . . . . . . . . . . . . . . 1.93 1.22 1.74
* The data are each expressed in terms of the corresponding value obtained from Sachs' solution,considered as
unity. The actual values for Sachs' solution in grams, are given in parenthesis.
these two solutions, as to root yield, is reversed, and is similar to that brought
out in table XIV. The averages of series A and B (table IV) indicate that
these two solutions are practically equal in the production of root yields.
These same averages indicate that solution R5C2 gives top yields a little
higher than those obtained from solution R3C3. On the whole, it seems
advisable to employ solution R5C2 (1.75 atm.) as the best solution, for
wheat, resulting from the present study.
The only instance in which a yield of 3-salt solution R5C2 (1.75 atm.)
is surpassed by one of the other solutions with 1.75 atmospheres of total
concentration occurs with Crone's solution, which gave a slightly higher
yield of tops than did solution R5C2 (1.95 as compared to 1.93). But
396 - * John W. Shive
Crone's solution here gave a much lower yield-of roots and a slightly lower
total yield than did solution R5C2. -
Two instances are recorded in table XIV showing other solutions as giving
yields equal to the corresponding one from solution R5C2. (1) Birner and
Lucanus' solution gave as high a top yield, but it fell far below solution
R5C2 in root yield and slightly below it in total yield. Solution R3C6 of
the present study, which has the same main Salts in nearly the same pro-
portions as has Birner and Lucanus' solution, appears to be markedly in-
ferior to solution R5C2 in the production of both tops and roots, according
to the data of table IV (page 349). Put the Birner and Lucanus solution
contains ferrous phosphate (Fe3(PO4)2) instead of the ferric salt employed
with the 3-salt solutions of the present paper. (2) Detmer's solution equals
solution R5C2 in root yield but is obviously very poor by the criterion of
top yield and by that of total dry weight.
It should be emphasized in this connection that the conclusions here
reached, as to the relative values of these various solutions, are to be taken
as established only for the total concentration here employed (1.75 atm.),
and for the kind of plant here used (winter wheat), during the first four weeks
of its development from the seed. Whether some of the other solutions
listed in table XIV may or may not surpass 3-salt solution R5C2, when used
with other total concentrations, or with other plants, or other developmental
stages of the same plant, or even under other climatic conditions cannot
now be stated.
LITERATURE CITED
Numbers in brackets throughout the preceding pages refer to the year of publication
and to the corresponding numbers that follow authors' names in this list. Where more
than one reference to the same year occurs under a given name, these are serially num-
bered in italics, within the brackets.
BARTLETT, R. H. See True and Bartlett.
BIRNER, H., AND LUCANUs, B. Wasserkulturversuche mit Hafer in der Agric.-Chem. Versuchsstation zu Regen-
walde. Landw. Versuchsstat. 8: 128–177. 1866.
BRITTon, J. C. See Livingston, Britton and Reid.
BRLGGs, L. J., AND SHANTz, H. L. The water requirement of plants. II. A review of the literature. U. S.
Dept. Agric. Bur. Plant Ind. Bull. 285. 1913.
CAMERON, F. K. See Whitney and Cameron.
CRoNE, G. Ergebnisse von Untersuchungen über die Wirkung der Phosphorsäure auf die höhere Pflanzen und
eine neue Nährlösung. Sitzungsber. Neiderrhein. Ges. Nat.—und Heilk. Bonn, 1902. Pages 167–173.
DETMER, W. Practical plant physiology, translated by S. A. Moor. London, 1898.
GILE, P. L. Lime-magnesia ratio as influenced by concentration. Porto Rico Agric. Exp. Sta. Bull. 12.
HARRIs, F. S. Effects of variations in moisture content on certain properties of a soil and on the growth of wheat.
Cornell Univ. Agric. Exp. Sta. Bull. 352. 1914.
HARTwelI., B. L., WHEELER, H. J., and PEMBER, F. T. The effect of the addition of sodium to deficient
amounts of potassium upon the growth of plants in both water and sand cultures. Ann. Rep.
Rhode Island Agric. Exp. Sta. 20: 299–357. 1907.
HAwkINs, Lon A. See Livingston and Hawkins.
HEINRICH, R. Ueber die Wassermengen, welche die Haferoflanze aus verschiedenen Nährstoff-concentrationen
während ihrer Vegetationszeit verbraucht. Zweiter Bericht tiber die Verhältnesse und Wirksamkeit
der Landwirtschaftlichen Versuchs-Station, Rostock. Pages 170–174. 1894.
PHYSIOLOGICAL BALANCE IN NUTRIENT MEDIA • 397
2.
Hoyt, W. D. Some toxic and antitoxic effects in cultures of Spirogyra. Bull. Torrey Bot. Club. 40: 333–360.
1913.
KNOP, W. Quantitative-analytische Arbeiten über den Ernährungsprocess der Pflanzen. II. Landw.
Versuchsst. 4: 173–187. 1862. -
LIVINGSTON, B. E. |1900]. On the nature of the stimulus which causes the change of form in polymorphic green
algae. Bot. Gaz. 30: 289–317. 1900. e
[1903] The rôle of diffusion and osmotic pressure in plants. Chicago, 1903.
[1905] Relation of transpiration to growth in wheat. Bot. Gaz. 40: 178–195. 1905.
[1906] A simple method for experiments with water cultures. Plant World 9:13–16. 1906.
[1907. Further studies on the properties of unproductive soils. U. S. Dept. Agric. Bur. Soils. Bull.
36. 1907.
[1913) Osmotic pressure and related forces as environmental factors. Plant World 16: 165–176. 1913.
[1915] Atmometry and the porous cup atmometer. Plant World 18:21-30, 51–74, 95–111, 143–149.
1915, Also reprinted, Tucson, 1915. -
See Pulling and Livingston.
LivingSTON, B. E., BRITTON, J. C., AND REID, F. R. Studies on the properties of an unproductive soil. U. S.
Dept. Agric. Bur. Soils. Bull. 28. 1905.
LIVINGSTON, B. E., AND HAWKINS, LON A. The water-relation between plant and soil. Carnegie Inst. Wash.
Pub. 204: 3–48. 1915.
LoBB, J. On the nature of the process of fertilization and the artificial production of normal larvae (Plutei)
from the unfertilized egg of the sea urchin. Amer. Jour. Physiol. 3: 135–138. 1899.
LoLw, O. AND MAY, D. W. The relation of lime and magnesia to plant growth. U. S. Dept. Agric. Bur. Plant
Ind. Bull. 1. 1901. -
LUcANUs, B. See Birner and Lucanus.
MAY, D. W. See Loew and May.
McCool, M. H. The action of certain nutrient and non-nutrient bases on plant growth. Cornell Agric. Exp.
Sta. Mem. 2: 121–170. 1913.
MELLOR, J. W. Chemical statics and dynamics. London, 1909.
PEMBER, F. R. See Hartwell, Wheeler and Pember,
PFIEFFER, W. The physiology of plants, translated by A. J. Ewart, 1: 420. Oxford, 1900.
PULLING, H. E. AND LIVINGSTON, B. E. The water-supplying power of the soil as indicated by osmometers.
Carnegie Inst. Wash. Pub. 204: 49–84. 1915. .
SACHS, J. Vegetationsversuche mit Ausschluss des Bodens (iber die Nährstoffe und sonstigen Ernährungs-
bedingungen von Mais, Bohnen und anderen Pflanzen. Landw. Versuchsst. 2: 219–268. 1860.
Scar MPER, A. F. W. Zur Frage der Assimilation der Mineralsalze durch die grüne Pflanze. Flora 73: 207–261.
1890.
Sch. REINER, O. AND SKINNER, J. J. |1910, 1], Ratio of phosphate, nitrate and potassium on absorption and
growth. Bot. Gaz. 50: 1–30. 1910.
*s (1910, 2), some effects of a harmful organic soil constituent. U. S. Dept. Agric. Bur. Soils Bull. 70.
1910.
SHANTz, H. L. See Briggs and Shantz.
SHIvE, J. W. (1914] The freezing points of Tottingham's nutrient solutions. Plant World, 17:345-353. 1914.
[1915] A three-salt nutrient solution for plants. Amer. Jour. Bot. 2: 157-160. 1915.
SKINNER, J. J. See Schreiner and Skinner.
Sor AUER, P. Nachtrag zu den “Studien über Verdunstung.” Forsch. Geb. Agric. Phys. 6: 79–96. 1883.
Toll ENs, B. Ueber einige Erleichterungen bei der Cultur von Pflanzen in Wässerigen Lösungen. Jour. Landw.
30: 537–540. 1882.
TOTTINGHAM, W. E. A quantitative chemical and physiological study of nutrient solutions for plant cultures.
Physiol. Res. 1: 133–245. 1914.
TRUE, R. H. Harmful action of distilled water. Amer. Jour. Bot., 1: 255–273. 1914.
TRUE, R. H. AND BARTLETT, II, H. Absorption and excretion of salts by roots, as influenced by concentration
and composition of culture solutions. U. S. Dept. Agric. Bul. Plant Ind. Bull. 231. 1912.
WHEELER, H. J. See Hartwell, Wheeler and Pember.
WHITNEY, M., AND CAMERON, F. IX. Investigations in soil fertility. U. S. Dept. Agric. Bur. Soils Bull. 23.
1904.
WooDWARD, J. Some thoughts and experiments concerning vegetation. Phil. Trans. Roy. Soc. London 21:
193–227. 1699.
=
JoHN W. SHIVE , ,
UNIVER
Hill
iſſil
VITA
The writer was born at Halifax, Pennsylvania, February 13, 1877. He
attended the Cumberland Valley State Normal School, Shippensburg,
Pennsylvania, from 1896 to 1898, and in 1906 graduated from Dickinson
College, Carlisle, Pennsylvania, with the degree of Bachelor of Philosophy.
He taught in the public schools of Pennsylvania from 1898 to 1903, and from
1906 to 1911 he was teacher of science in Perkiomen Seminary, Pennsburg,
Pennsylvania. During the years from 1911 to 1915 he attended the Johns
Hopkins University as a graduate student in Plant Physiology, Botany and
Chemistry. He was University Scholar in Botany for the year 1911–1912,
University Fellow in Plant Physiology for the years 1913–1914 and 1914–
1915. During the summer of 1913, he was engaged in research at the
Desert Laboratory of the Carnegie Institution of Washington, at Tucson,
Arizona. - -

§§§§§،******** ：،- * ** ！!！ ** - *~--~~~~ -….…….№aeae， §§§***********-
^- ،-§§Ķ：-#ț¢
¿**ş，•--★ →
****§§2:3，3:-
·§§§;-- - - --: §§§§），|-#-###$$$$
ĢĒ####- |--###########žā####§§§§§§§
r ≡≡: №- …*… * * · · · *---- 。 、、、、、、、z šºšºšº，----Źr:::*şț¢, ſåſºſ
ſae§§§§§;&###################
- * & ******* · -####################ſae
·###########§§
§§§§§§§§§
¿??¿ $ ******<!！！！！
#####ſº:*********