Mixing a Balanced Nutrient Solution©

N.J.J. Combrink
P.O. Box 3172, Matieland, 7602, South Africa
Email: [email protected]

INTRODUCTION
Plant growers are dependant on the process of photosynthesis, driven in green leaves with
energy from the sun, building carbohydrates with carbon dioxide (CO2), and water (H2O).
Thirteen essential plant nutrients affect this process by ensuring that the leaves are green,
filled with chlorophyll, and that the integrity of membranes and strength of cell walls are
maintained. Plant roots absorb these nutrients from fertile soil. But, these nutrients can
also be supplied by a well-balanced nutrient solution in soil-less culture. Arnon and
Hoagland (1940) established some of the first well-balanced nutrient solutions, enabling
soil-less crop production. Apart from the introduction of chelated iron, modern nutrient
solutions do not differ much from the older ones. In this paper, an extract form “Nutrient
Solution Management” (Combrink and Kempen, 2011), guidelines will be given to enable
the reader to “build” his/her own nutrient solution. However, before this can be done, it is
important to ensure that terms such as pH, salt concentration, and some basic chemical
principles are understood.

ACIDITY AND ALKALINITY

pH
pH is a value taken to represent the acidity or alkalinity of an aqueous solution with
values ranging from 1 to 14. At a reading of 7, the pH is considered to be neutral. At this
point equal concentrations of hydrogen ions (H+) and hydroxyl ions (OH-) occur. Values
lower than 7 indicate acidity and solutions with pH values higher than 7 are alkaline. The
pH scale is logarithmic, thus the pH value of water will not decrease linearly when acids
are added to it.

Total Alkalinity
Total alkalinity is the aggregate concentration of bases such as carbonates, bicarbonates,
and hydroxides. The concentration of these ions (CO32-, HCO3-, and OH-) is determined
with a titration, as measure of total alkalinity. It is usually expressed as HCO3-, but some
laboratories may express total alkalinity as CaCO3. The alkalinity can be neutralized by
adding an equivalent amount of acid (H+). Most soils are well-buffered, thus their pH
values do not change easily. This is not the case for soil-less conditions where special
precautions are needed to control the pH and the availability of nutrients in solution. Due
to this pH-sensitive availability of some nutrients, soil-less crops are grown in solutions
with pH values between 5.3 and 6.3. Compared to soil pH values, this pH range may
seem to be relatively low, but a look at the dissociation pattern of phosphate (Table 1)
shows why neutral to alkaline nutrient solutions should be avoided. In addition, micro-
nutrients such as Fe and Zn precipitate, forming insoluble salts at pH values higher than 7.

Table 1. Phosphate dissociation pattern (Steiner, 1984).

pH HPO42- H2PO4-
(Forming a soluble Ca salt) (%) (Forming an insoluble Ca salt) (%)
pH = 5.0 100 0
pH = 6.0 90 10
pH = 6.5 78 22
pH = 7.0 50 50
pH = 8.0 15 85

13
 Problems also arise with pH levels lower than 5, where the concentration of H+ is too
high, restricting the absorption of Ca2+. This problem is aggravated in the presence of
ammonium (NH4+). It is recommended that the pH limits for soil-less production systems
be set at 5.3 to 6.3, with an optimum pH at about 5.8, where most crops should grow well.

ELECTRICAL CONDUCTIVITY
Electrical conductivity (EC) is a well-known term amongst soil-less growers. The EC of
water is used as an indication of its salt content since higher concentrations of charged
particles (ions) increase the electrical conductivity of water or nutrient solutions. When
salts are added to water, the crystals dissolve or ionise. Using table salt or sodium chloride
(NaCl) as an example, one sodium cation (Na+) and one chloride anion (Cl-) will be
released per dissolved NaCl unit. These charged ions in the water enable the flow of an
electrical current. The higher the concentration of ions (charge) per unit volume of water,
the better the conductivity and the higher the EC will be. (But, one unit of CaCl2 releases
one Ca2+ cation and two Cl- anions with a potential conductivity of about double that of
NaCl, an aspect that will be dealt with later). At a very high EC, the water contains too
much dissolved salts, restricting the osmotic movement of water from the root zone
solution into the roots. Take note that non-polar molecules (sugar or urea) dissolve in
water without the release of charged particles. Thus, these substances dissolve but do not
increase the EC of a solution but do affect the osmolarity of the solution and may thus also
restrict the uptake of water.
The use of different EC units may confuse growers. Siemens (S) is the SI unit for
conductivity. With very low conductivity levels, a 1000 times’ smaller unit such as milli
Siemens (mS) can be used. The unit “mho” is equivalent to S and mmho to mS. Some
laboratories use the electrical resistance of soil as indication of its salt content. Due to an
inverse relationship between conductivity and resistance, higher resistance values
(measured in ohm) are indicative of low salt contents. A lower resistance implies that more
salts are in solution, the conductivity will thus be high. The direct relationship between
conductivity and salt content is easy to understand and an important managerial tool for
soil-less growers.
The distance between the two electrodes, while measuring the conductivity of a solution,
also affects the reading. The reading in mS·m-1 will be 100 times bigger than a reading in
mS·cm-1. Most growers use mS·cm-1 as unit to test the EC of standard nutrient solutions.
Depending on the crop and the season, these values may vary between 0.8 and 3.0 mS·
cm-1. The EC of good quality feeding water may be very low, explaining why most
laboratories prefer to express the conductivity of irrigation water in mS·m-1.
Relation of mS·cm-1 to other EC units: 1 mS·cm-1 = 1dS·m-1
= 1 mmho·cm-1
= 100 mS·m-1
= 1000 S·cm-1
Since temperatures affect EC readings, EC meters are calibrated at 25°C. Electrical
conductivity meters such as those used by growers are not temperature compensating. These
EC measurements can be corrected by subtracting 2% from the measured value per 1°C
higher than 25°C solution temperature. With solutions colder than 25°C, 2% can be added
per 1°C deviation.

Feeding Water Quality

The term “feeding water” is used to describe an untreated water source. Different factors
can be used to define feeding water quality but it’s chemical composition and the
presence of potential dangerous micro-organisms are the most important. The risk to find
toxic levels of micro nutrients in feeding water increases with increased EC levels.
Micronutrients are usually present at very low concentrations, but can be phytotoxic at
low levels; too low to even affect EC readings.

14
Feeding Water pH
Water pH cannot be used as sole parameter for pH adjustments. This must be done by
adjusting total alkalinity, the aggregate concentration of bases such as carbonates,
bicarbonates, and hydroxides. Laboratories measure the total concentration of these bases
(CO32-, HCO3- and OH-) with a titration. Since fertilizers may be slightly acidic, the
feeding water should be slightly alkaline to prevent the pH from dropping too low when
mixing.

UNWANTED IONS IN FEEDING WATER

Iron and Manganese

High levels of soluble (ferrous) iron may be found in water from the mountainous areas of
the Cape and along the Drakensberg. This iron is in a reduced state (Fe2+) and when
oxidized, ferric iron Fe3+ precipitate as an insoluble, red substance. Using this water to
irrigate with sprinklers, the red pigment can be seen on leaves and on garden walls.
Manganese is also soluble in its reduced state and precipitates as insoluble MnO2 when
oxidized. Should water with high Fe or Mn concentrations be used for drip irrigation, the
ions are oxidized (micro-organisms may also be involved). Insoluble residues then block
the drippers. Fe and Mn can be precipitated and removed by aeration of the feeding water,
before using it. The higher the pH, the quicker it will precipitate.

Sodium and Chloride as well as Nutrients at Potential Phytotoxic Levels

Sodium and Cl ions can only be efficiently removed with expensive water purification
systems such as reversed osmosis, currently used to purify sea water in Sedgefield.
Should water purification companies claim that they can remove these ions from feeding
water, samples should be obtained and the water should be chemically analysed to
measure the Na and Cl levels before and after the process to determine whether the
process lives up to its claim. The same expensive system of reversed osmosis is needed to
remove nutrients such as Mg, Cu, Zn, and B, when these are present at phytotoxic levels.

Ions Associated With Alkalinity

The alkalinity level in saline feeding water is usually high, due to high levels of CO32-,
HCO3- and OH-. These ions can easily be replaced by nitrate or phosphate, simply by
using nitric- or phosphoric acid to lower the alkalinity.

BASIC HYDROPONIC CHEMISTRY

Mole
Mole is Avogadro’s number with the value of 6.02×1023. The atomic mass, expressed in
grams, contains Avogadro’s number of atoms. For example, 12 g of carbon (C) and 23 g
of sodium (Na) contain the same number of atoms = 6.02×1023 = 1 mole. The atomic
mass of some elements are shown in Table 2.

Molecular Mass
One mole of a substance is a mass in grams numerically equal to the molecular mass.
With KCl as an example, 39.1 and 35.5 are added to reach 74.6, the molecular mass of
this salt. Thus, 74.6 g of KCl contains Avogadro’s number of KCl units: 1 mole (6×1023)
K and 1 mole Cl atoms.

15
Table 2. Chemical formula, ions of some elements and complex ions with their respective
atomic and molecular masses.

Element Symbol Ion Atomic Complex Chemical Molecular

mass ion formula mass
Hydrogen H H+ 1.0 Bicarbonate HCO3– 61
Boron B 10.8 Carbonate CO3– 60
Carbon C 12.0 Ammonium NH4+ 18
Nitrogen N 14.0 Nitrate NO3– 62
Oxygen O 16.0 phosphate H2PO4– 97
Sodium Na Na+ 23.0 Sulphate SO42- 96
Magnesium Mg Mg2+ 24.3
Silicon Si 28.1
Phosphorus P 31.0
Sulphur S 32.1
Chloride Cl Cl– 35.5
Potassium K K+ 39.1
Calcium Ca Ca2+ 40.1
Manganese Mn 54.9
Iron Fe 55.8
Copper Cu 63.5
Zinc Zn 65.4
Molybdenum Mo 95.9

Equivalent Mass (Mass per Charge)

The atomic mass of K is 39. When losing an electron it becomes a single charged cation
K+. Its mass (and equivalent mass) remains to be 39 (the mass of an electron is close to
zero, it can be ignored). The atomic mass of Ca is 40, but when ionised, it loses 2
electrons, becoming a doubled charged cation (Ca2+). Its atomic mass is 40, but its mass
per single charge (equivalent mass) is 20. By dividing atomic mass by the charge of its
ion, the equivalent mass is calculated. The molecular mass is calculated by adding the
mass of all the atoms in a compound. Its equivalent mass is the molecular mass, divided
by the charge of the ion (in the molecule) with the highest charge or valance. Thus, its
equivalent mass is the mass of material with mole positive and mole negative charges. As
an example: The mass of 1 mole CaCl2 is 111 g (40+35.5+35.5) but 1 mole of positively
(and negatively) charged units, or the equivalent mass is 111/2 or 55.5 g of CaCl2. The
equivalent mass of a salt is its molecular mass divided by the highest valence (charge) of
the ions that will be released when it is dissolved in water. With mole charged particles
per equivalent, the term molc is used, a unit 1000 times bigger than mmolc. The
concentration of macro-nutrients in nutrient solutions is expressed as mmolc·L-1.

Transforming ppm or mg·L-1 to mmolc·L-1

Chemical laboratories usually use mg·L-1 as unit for their water analyses reports. These
can be transformed to mmolc·L-1 by dividing it by the equivalent mass of the ion or the
complex ion, as shown in Table 3.

16
Table 3. Concentrations expressed as mg·L-1 (or ppm) must be divided by the equivalent
mass of the substance to transform it to mmolc·L-1.

Substance Eqv. mass Substance Eqv. mass

Ammoinium-N 14 Nitrate-N 14
Ammonium (NH4+) 18 Nitrate (NO3-) 62
Bicarbonate (HCO3–) 61 Phosphate–P 31
Calcium (Ca)2+ 20 Phosphorus 31
Carbonate (CO3)2- 30 Potassium (K+) 39
Chloride (Cl–) 35.5 Sodium (Na+) 23
Magnesium (Mg)2+ 12.1 Sulphate (SO4)2- 48

Molar (M)
By dissolving 1 mole KCl (74.6 g) into 1 litre of distilled water, the molarity of the
solution will be 1 M KCl. In this case 6×1023 positively charged ions (K+) and 6×1023
negatively charged ions (Cl-) would be present in the water. Due to these charged
particles, the solution will be able to conduct an electrical current, which is used as
indication of the ion concentration. By dissolving only 2% of 74.6 = 1.49 g KCl per litre
distilled water, the molarity of the solution will be 2% of 1 M = 0.02 M. This solution can
be used as standard EC test solution to calibrate your EC-meter, because it’s EC will be
2.77 mS·cm-1 at 25°C.

Normality (N)
By dissolving the equivalent mass of a substance into 1 litre of distilled water (1
mmolc·L-1) the solution’s normality will be one (1 N). The normality of acids or liquid
fertilizers can be calculated with the following formula:
N = (SG x C) / Eqv. mass
N = Normality
SG = Specific gravity of the solution
C = Concentration as g·kg-1 or (% × 10)
Eqv. mass = Equivalent mass of the dissolved chemical

ESSENTIAL NUTRIENTS

Micro-Nutrients
These are needed in extremely low concentrations. In European soil-less production
environments, micronutrient concentrations are expressed as μmol·L-1, a unit 1000 times
smaller than mmol·L-1. In South Africa, micronutrient concentrations are expressed as
mg·L-1 (or ppm).

Macro-Nutrients
These are known as N, P, K, Ca, Mg, and S. These are still seen as the only essential
macronutrients, although Si and other minerals may positively affect the growth of some
plants under specific conditions. Roots absorb macronutrients as ions: NH4+, K+, Ca2+,
Mg2+ (cations) and NO3-, H2PO4-, and SO42- (anions). Only N can be absorbed as a
positively charged cation (NH4+) as well as a negatively charged anion (NO3-). Several
chemicals can be used to release these ions into feeding water.

Non-Essential Na and Cl
1. Sodium. Sodium is not essential for crop growth but most South African water sources
contain some sodium. Most crop plants are salt excluders with limited uptake of Na+ by
roots. Some plants do not transport sodium ions from the roots to the shoots (Epstein and
Bloom, 2004). The plant’s ability to tolerate high Na+ levels may improve in the presence of
Ca2+. This was demonstrated with saline-sensitive beans that could be grown in the presence
of 50 mmolc·L-1 NaCl, but only with addition of relatively high Ca levels (La-Haye and

17
Epstein, 1971).
2. Chloride. Chloride is considered to be an essential micronutrient, but it is usually present
at macronutrient quantities. Please do not confuse the chloride ion (Cl -) with chlorine gas
(Cl2), used by municipalities and growers as an oxidant to sterilize water. Some saline water
sources contain high levels of chloride that may limit the uptake of nitrate (Kafkafi, Valoras
and Letey, 1982). Fortunately the opposite is also true. Weigel et al. (1973) used feeding
water with a high chloride level and by increasing the nitrate level in a nutrient solution from
1.25 to 7.0 mmolc·L-1, found that chloride concentrations in soybean leaves dropped.

BASICS BEFORE STARTING WITH A NUTRIENT SOLUTION

Nutrient Recipes
By combining and evaluating the information available in the literature, Steiner (1968) made
a huge contribution when he summarised basic nutritional principals for soil-less plant
production. He proposed a well-balanced “universal nutrient solution” by dealing with
cations- and anions separately and described safe areas for the K+ : Ca2+ : Mg2+ and NO3- :
H2PO4- : SO42- ratios. Using mmolc·L-1 as unit, he expressed the ideal cation ratio as
35:45:20 % and chose 60:5:35 % as anion ratio. Some deviations from these ratios are
allowed and he defined the outer limits beyond which deficiencies or toxicities may
develop (Steiner, 1968). Before starting with nutrient solutions, the role of NH4+ should first
be clarified.

Ammonium is Used as pH Regulator or N-Source for Acid-Tolerant Crops

Steiner did not include ammonium (NH4+) in his nutrient solution and used NO3- as the sole
nitrogen source. As in Europe, all South African soil-less growers use ammonium.
Ammonium is used in soil-less production systems to regulate the pH in the root zone. To
understand this, it should be remembered that a neutral charge is maintained in the root.
When a cation or anion is absorbed, the root secretes H+ or OH- respectively. Due to the fact
that NH4+ is a very small ion and is easily absorbed, NH4+ is exchanged by H+ at such a rate,
that the root zone is acidified. During decomposition of organic substrates, bicarbonate is
released, increasing the alkalinity and the substrate’s pH. Some ammonium may then be
useful to counteract this. Since too much NH4+ may acidify the rhizosphere to such an
extent, that the uptake of Ca may be restricted, it should be carefully managed, especially on
crops sensitive to blossom-end-rot, a Ca deficiency disorder (Combrink, 2005). Kafkaffi
(2000) claims that the danger of damage by ammonium increases at high root zone
temperatures. The difference in sensitivity of crops to ammonium toxicity is ascribed to
differences in sugar concentrations in the roots where ammonium is metabolised.
Ammonium-loving crops such as orchids and disas react positively to the presence of
ammonium.

Compiling Your Own Nutrient Solution

Publications by Arnon and Hoagland (1940), Steiner (1984), and Dutch research workers
(de Kreij et al., 1999; Straver et al., 1999) helped local fertilizer companies to develop
“ready mix” nutrient solutions. The cation and anion ratios of these commercial mixes fall
within the limits of Steiner’s “universal nutrient solution.” These products allow growers
to avoid the tedious job of weighing and mixing fertilizers from several different bags,
when compiling their own mixes. But, “ready mixes” can only be used where the feeding
water contains almost no dissolved salts (EC <0.3 mS·cm-1). Most manufacturers of “ready
mixes” make use of Cl-containing fertilizers. This is acceptable for growers with good
quality feeding water but may be a risk for growers with saline sensitive crops and high
levels of Cl- in their feeding water. In addition, these saline water sources may also contain
high levels of Mg and by simply adding the normal Mg level (using a “ready mix”), the
optimum Mg level can easily be exceeded and Mg at this higher level may suppress the
uptake of the other cations (Ca and K). This underlines the importance that nutrients,
already present in feeding water, should be taken into account when preparing a nutrient

18
mix.

Predicting the Electrical Conductivity of Solutions

Electrical conductivity is used as indication of the amount of dissolved salts in solution.
The unit mg·L-1 is commonly used by laboratories where feeding water is chemically
analysed. However, it is difficult to balance ion concentrations using this unit. This
problem is solved by using mmolc·L-1. In addition, the EC of the solution can then also be
predicted with relative accuracy. By dividing the cation (or anion) concentration
(mmolc·L-1) by 10, a fairly accurate EC value is calculated (mg·L-1). As an example, if
you add 20 mmolc·L-1 cations (and anions) to distilled water, the EC will be about 2
mS·cm-1. By multiplying an EC value with 10, you can thus estimate its ion concentration
in mmolc·L-1. Using feeding water with an EC of 0.3 mS·cm-1, you know that dissolved
salts are present at a concentration of about 3 mmolc·L-1.

Electrical Conductivity and Nutrient Uptake

The size of a root system as well as ion concentration in root zones affect the rate of ion
uptake. It is obvious that a well-developed root system is needed for optimum uptake and
that a specific root system is likely to absorb more nutrients at higher nutrient
concentrations (higher EC levels). The increase in ion uptake is linear with increased
nutrient concentrations, but only up to a certain level. As soon as the EC rises above a
certain level, the rate of ion uptake decreases (Bar-Yosef, 1995). Uptake of the massive
double-charged Ca ion declines first. This may induce a Ca-deficiency (Combrink, 2005).
This illustrates that the rate of ion uptake can be increased by an increased ion
concentration – but only up to a certain point. Even saline tolerant crops grow slower,
should the EC of the root zone solution be increased beyond the crop’s optimum need
(Ribeiro and Combrink, 2006). Benoit (2003) compared a salt sensitive to a salt tolerant
crop and showed that the yield potential of the sensitive crop declined at a rate of at least
10% for each 1 mS·cm-1 rise beyond its optimum EC of 2 mS·cm-1. The yield of the salt
tolerant crop only declined at EC levels higher than 4 mS·cm-1 at a rate of approximately
5% for each 1 mS·cm-1 higher than EC 4 mS·cm-1.
In an effort to improve the sugar content of ‘Daniela’, one of the first “long shelf life”
tomato cultivars grown in South Africa, the nutrient solution’s EC of 2 mS·cm-1 was
increased to 4 mS·cm-1 with addition of 2 mmol NaCl per litre. The fruit’s total soluble
solids (°Brix) did increase but the associated higher sugar percentage was due to the fact that
less water was absorbed and not due to a better sugar production. This was reflected by an
associated 30% reduction in yield and fruit size (Combrink, 1998).
PLANNING A MACRONUTRIENT SOLUTION
Steiner was a prominent figure on the soil-less culture scene for many years and it may be
useful to do our first exercise with his “universal nutrient solution.” At this stage we
ignore the composition of the feeding water, assuming that it is of good quality, not
needing pH adjustment.
The macronutrient percentages for cations in Steiner’s ‘Universal nutrient solution’ are:
35% K+, 45% Ca2+, and 20% Mg2+ (on a mmolc·L-1 basis). The recommended anion ratio
is 60% NO3-, 5% H2PO4-, and 35% SO42-. Say that the aim is to have a nutrient solution at
an EC of 2 mS·cm-1, then the cation as well as anion totals need to be 20 mmolc·L-1. The K+
level will thus be 35% of 20 or 7 mmolc·L-1, Ca2+ at 45% of 20 at 9 mmolc·L-1, and Mg2+ at
20% of 20 mmolc·L-1 at 4 mmolc·L-1. The same procedure will indicate that 12 mmolc·L-1
NO3-, 1 mmolc·L-1 H2PO4-, and 7 mmolc·L-1 SO42- are needed. Pure hydroponic-quality
salts are used as sources of these ions.

Procedure to Select Chemicals

1) Look for the smallest nutrient need; 1 mmolc·L-1 H2PO4- and choose a P source. Since
no ammonium is prescribed for this mix, this eliminates ammonium phosphate. Thus,
choose mono potassium phosphate and add 1 mmolc·L-1 potassium (K), as well as the

19
 associated 1 mmolc·L-1 phosphate at the anion side (Table 4).
2) Look for the next smallest need, 4 mmolc·L-1 Mg. Choose a Mg source. Mg-nitrate and
Mg-sulphate options are available. Choose the Mg salt with the anion with the smallest
need in your planned solution (nitrate vs. sulphate): The choice is MgSO4. Add 4
mmolc·L-1 Mg2+ at the cation side and 4 mmolc·L-1 SO42- at the anion side.
3) The next smallest need is 3 mmolc·L-1 SO42-. Since the Mg need is already satisfied, K-
sulphate is chosen as SO42- source. Apply the balance of 3 mmolc·L-1 SO42- and
remember also to add 3 mmolc·L-1 of K+ at the cation side.
4) The next smallest need is 3 mmolc·L-1 potassium (7 needed, but 1 K-phosphate and 3
K-sulphate were given). With only nitrate still needed at the anion side, the only option
is to use KNO3 at 3 mmolc·L-1.
5) The last action is to apply the Ca need as well as the balance of nitrate still needed with
9 mmolc·L-1 of ammonium-free Ca-nitrate (This procedure changes when using NH4+-
containing Ca-nitrate, Table 5).

Table 4. Mixing Steiner’s ammonium-free macro-nutrient solution (Steiner, 1984).

EXERCISE: Steiner's solution
Nutrients to be added (mmolc L-1): 7 9 4 12 1 7
mol. eqv. Application
Fertilizer mass mass (See y) K+ Ca2+ Mg2+ NO3- H2PO4- SO42-
KNO3 (4 th step) 101 101 3 x 101 = 303 3 3
K2 SO4 (3rd step) 174 87 3 x 87 = 261 3 3
st
K H2PO4 (1 step) 136 136 1 x 136 = 136 1 1
z
Ca(NO3)2.4H2O (Last) 236 118 9 x 118 = 1062 9 9
Mg SO4.7H2O (2nd step) 246 123 4 x 123 = 492 4 4
7 9 4 12 1 7

z Ammonium-free Ca-nitrate is used for this exercise.

y Nutrient concentration (mmolc L-1) x equivalent mass of fertilizer = Application (mg L-1 or g 1000 L-1)

Adjust Alkalinity and Add Nutrients

The aim of this exercise is to set your feeding water’s total alkalinity between a minimum
of 0.2, up to a maximum of 1.0 mmolc·L-1. Some fertilizers such as potassium sulphate
(K2SO4) and mono potassium phosphate (MKP) may contain acid residues. On average,
the H+ released from fertilizers used for a normal macronutrient mix varies from 0.2 to
0.4 mmolc·L-1. With only this limited acidity in fertilizers to neutralize, an alkalinity
buffer of 0.4 should be sufficient. The alkalinity of feeding water for organic substrates
(releasing HCO3- during decomposition) should be lowered to 0.2 to 0.4. Using inert
substrates, the input alkalinity may be set at 0.5 to 1.0 mmolc·L-1.
Add Alkalinity
Distilled water is an example of water with a neutral pH and alkalinity of zero, as would
also be found in pure rain water. Water from high rainfall areas may have no alkalinity, it
may even be slightly acidic. Due to the low solubility of CaCO3 (Table 16) calcitic lime
can be used to increase the total alkalinity of these water sources without the danger of an
over-dosage.
Water from the Cederberg area (low EC and pH) flows into the Clanwilliam dam. Water
is then fed to growers via cement canals. The exposure of the water to the cement ads
alkalinity to the water. It is possible that the alkalinity may increase as it moves to
growers further downstream. In winter with a low need for irrigation water, the flow will
be slow, thus allowing more time for the water to interact with the cement that may

20
further increase the alkalinity of the water. The effect is about the same as where
municipalities treat corrosive water by letting it flow through a lime pebble tank.
Where lime-pebble pre-treatment operations are impractical and the alkalinity of the
feeding water is too low, a soluble alkaline such as KOH can be added, at a levels
between 0.2 to 0.8 mmolc·L-1.

Decrease the Alkalinity

With addition of acid to alkaline feeding water, carbon dioxide gas as well as water is
released. The HCO3- is replaced with NO3-, H2PO4- or SO42- by using either nitric acid
(HNO3), phosphoric acid (H3PO4), or sulphuric acid (H2SO4) as source of H+.

Table 5. A mild lowering of the alkalinity of a low-EC feeding water (suitable for the use
of a “ready mix”), before adding nutrients.

Application EXERCISE: pH correction and Steiner's solution

Dry products Liquids
mol Eqv
-1
mmolc L x eqv mass
-1
mmolc L / N mmolc L-1 mmolc L-1
+ + + 2+ 2+ + - -
H2PO4- SO42- HCO3-
-1 -1
mass mass (g 1000 L ) (L 1000 L ) Na NH4 K Ca Mg H NO3 Cl

Water composition: 0.8 0.3 0.2 0.4 0.9

HNO3 (N = 12.7) 0.039 L 0.5 0.5
KOH 56 56
pH corrected water: 0.8 0.3 0.2 0 0.5 0.4 0.4
To be added: (Demonstrated in the table below)* 7 9 4 12 1 7

Final nutrient solution: EC ± 2.13 mS cm-1: 0.8 7 9.3 4.2 12.5 1 7 0.4 0.4

Table* only for calculation of added nutrients Na+ NH4+ K+ Ca2+ Mg2+ NO3 -
H2PO4 -
SO4 2-
Cl - HCO3-
KNO3 101 101 3 x 101 = 303 3 3
K2 SO4 174 87 3 x 87 = 261 3 3
K H2PO4 136 136 1 x 136 = 136 1 1
NH4 H2PO4 115 115
NH4NO3 80 80
Ca (NO3)2 . 4H2Oz 236 118 9 x 118 = 1062 9 9
Mg SO4.7H2O 246 123 4 x 123 = 492 4 4
Mg(NO3)2 .6H2O 256 128
Totals added * 7 9 4 12 1 7

z = Pure Ca (NO3)2.4H2O is used for this exercise. Most Ca-nitrate sources in S-Africa contain some ammonium.

Although the alkalinity is not very high in the example (Table 5), it is lowered from 0.9
to 0.4 because the plants will be grown in an organic substrate (adding alkalinity) and
Steiner’s solution contains no acidifying ammonium. The grower can thus expect a rise in
alkalinity and pH as the nutrient solution passes through the root zone. A higher alkalinity
would be needed for a nutrient solution containing a high ammonium level, especially
using an inert substrate. The EC of the feeding water, containing 1.3 mmolc·L-1 ions, can
be calculated as 1.3/10 = 0.13 mS·cm-1. This EC value remains unchanged after the acid
treatment. After addition of 20 mmolc·L-1 salts for Steiner’s solution, the EC should be
0.13 + 2.0 = 2.13 mS·cm-1. Due to relatively low levels of nutrients in the feeding water
and the fact that only a small increase in nitrate level results due to the addition of 0.5
mmolc·L-1 nitric acid, the end result does not differ much from Steiner’s recipe. This
feeding water is pure enough to allow the use of a “ready mix,” obtainable from most
fertilizer companies.

High EC-Feeding Water Requires Major Adjustments

In Table 6 the EC of the feeding water is 0.7 mS·cm-1, a level often found in South
African production areas. Should 20 mmolc·L-1 be added to this water, using a “ready
mix,” the expected EC of the nutrient solution will be about 2.7 mS·cm-1, a level where
the uptake of water may be slightly restricted. In addition, the levels of Ca, Mg and

21
sulphate (usually present in saline water) will probably be unnecessarily high. The
following example (Table 6) shows that only 17 mmolc·L-1 need to be added to equal
Steiner’s nutrient solution, ending with a final EC of 2.4 mS·cm-1 and a lower fertilizer
bill and better ion ratios in the root zone.

Table 6. Lowering of the alkalinity of a high EC feeding water before adding nutrients,
aiming to satisfy ratios as suggested by Steiner.

Application Salinity, pH correction and adapted Steiner

Dry products Liquids
mol Eqv
-1 -1
mmolc L x eqv mass mmolc L / N mmolc L-1 mmolc L-1
+ + ++ ++ +
mass mass
-1
(g 1000 L ) -1
(L 1000 L ) Na NH4 +
K Ca Mg H NO3 -
H2PO4- SO4= Cl - HCO3-

Water composition: EC of feeding water ± 0.7 mS cm-1 4 2 1 2 3 2

HNO3 (N = 12.7) 0.126 L 1.6 1.6
KOH 56 56
pH corrected water: (EC unaffected, still at ± 0.7 mS cm-1) 4 2 1 0 1.6 2 3 0.4
To be added: (Do it in the table below) * 7 7 3 10.4 1 5.6

Final nutrient solution: EC ± 2.4 mS cm-1 4 7 9 4 12 1 7.6 3 0.4

Table* only for calculation of added nutrients

+ + + ++ ++ - - = -
Na NH4 K Ca Mg NO3 H2PO4 SO4 Cl HCO3-

KNO3 101 101 3.4 x 101 = 343 3.4 3.4

K2 SO4 174 87 2.6 x 87 = 226 2.6 2.6
K H2PO4 136 136 1 x 136 = 136 1 1
NH4 H2PO4 115 115
NH4NO3 80 80
Ca (NO3)2 . 4H2Oz 236 118 7 x 118 = 826 7 7
Mg SO4.7H2O 246 123 3 x 123 = 369 3 3
Mg(NO3)2 .6H2O 256 128
Totals added * 7 7 3 10.4 1 5.6

z = Ammonium-free Ca-nitrate is used for this exercise.

The alkalinity in the example shown in Table 6 is 2 mmolc·L-1. Growers with high
alkalinity levels, as in this case, may choose to add phosphoric acid at the same level as
needed for H2PO4-. Since phosphoric acid will be used to supply 1 mmolc·L-1 H2PO4-,
Nitric acid (always added first) should only be applied at 0.6 mmolc·L-1 to lower the
alkalinity to 1.4 and then with phosphoric acid down to 0.4 mmolc·L-1. By adding
phosphoric acid directly to water with high alkalinity and Ca levels, Ca-phosphate may
precipitate. In the following example, nitric acid is used as sole acid to lower the
alkalinity from 2.0 to 0.4 by adding H+ and NO3- at 1.6 mmolc·L-1. In this case the added
nitrate cannot be ignored and will be taken into account when deciding how much nitrate
to add.
The planned added cations and anions must be equal. Start with the cations, adding
levels of K, Ca, and Mg to reach Steiner’s preset values (mmolc·L-1). With no K in the
water the K application will be 7 mmolc·L-1. However, the water contains some Ca and
Mg, thus top it up with 7 and 3 mmolc·L-1 respectively. At this stage, take note that the
total added cations will be 17 mmolc·L-1. Now plan to add exactly 17 mmolc·L-1 anions.
Top nitrate up by adding 10.4 mmolc·L-1 and add the total phosphate need of 1.0 mmolc·
L-1. Of all the macro nutrients, plants can adapt to sulphate at a very wide range. As long
as the sulphate level ends up at a level ±50% higher or lower than the target value, the
plants will still be happy. In this example the phosphate target is exceeded by about 10%
in order to balance the cations and anions.
Only now you can start using the lower table to add the ions, starting with the smallest
need first, usually phosphate. Select the ammonium-free MKP and by adding K and P at 1
mmolc·L-1, satisfying the P-need and lowering the K-need from 7 to 6 mmolc·L-1. With the
Mg-need at only 3 mmolc·L-1, this ion is added next. Now choose between the two Mg-
salts. Since Mg-sulphate will also supply sulphate, and the fact that the sulphate-need is
smaller than the nitrate need, this salt is chosen. By adding Mg-sulphate at 3 mmolc·L-1,

22
the sulphate-need is the smallest, thus the only other sulphate-containing salt is used.
After adding K- sulphate at 2.6 mmolc·L-1, the next smallest need is K at 3.4 mmolc·L-1.
The only anion still needing a top-up is nitrate. After adding K-nitrate at 3.4 mmolc·L-1,
ammonium-free Ca-nitrate can be used to top up both Ca and nitrate, using 7 mmolc·L-1.

Ammonium Affecting Root Zone pH

Organic substrates decompose with time, releasing HCO3- (Benoit, 2003). Coco peat
decomposes slower than pine sawdust and therefore its associated rise in alkalinity and
pH is also slower. The expected increase in root zone alkalinity (especially during the first
6-8 weeks after starting with a fresh organic substrate) can be counteracted with addition
of more ammonium to the nutrient solution. With inert substrates (rockwool, sand, or
gravel) or with NFT systems, lower ammonium levels may be used. Apart from its effect
on pH, orchid need relatively high ammonium levels (Pienaar, 2005).
Steiner’s ammonium-free solution was used as an example, but in practice, growers
have to use ammonium. Mono potassium phosphate (MKP) was used as K- and P-source
in our previous examples, but mono ammonium phosphate (MAP) is a popular
ammonium- and P-source. The ammonium-free Ca-nitrate [Ca(NO3).4H2O] that was also
used in previous examples, was only recently introduced to the Republic of South Africa
on commercial scale. Most South African growers buy Ca-nitrate, an ammonium-
containing product, from big fertilizer companies. Looking at its chemical composition,
Ca(NO3)2.0.2(NH4NO3.10H2O), one can see that 11/12 of the applied N is given as
nitrate-N and 1/12 as ammonium-N. By ignoring the ammonium content of Ca-nitrate and
by using MAP (NH4H2PO4) as P-source (Table 7), ammonium can reach an extremely
high level.

Table 7. Ammonium phosphate as P-source in combination with ammonium containing

Ca-nitrate, using Mg-sulphate, or Mg-nitrate as alternative Mg sources.

Concentra- Eqv Application

-1 -1 + + 2+ 2+
Fertilizer tion tank mass g 1000 L mmolc L NH4 K Ca Mg NO3- H2PO4- SO4-2
KNO3 A and/or B 101 404 4 4.0 4.0
K2SO4 A 87 174 2 2.0 2.0
NH4H2PO4 C 115 115 1 1.0 1.0
5Ca(NO3)2.NH4NO3.10H2O B 108 891 9 0.75 8.3 9.0
MgSO4.7H2O A 123 492 4 4.0 4.00

Three stock solution tanks needed 1.75 6.0 8.25 4.0 13.0 1.0 6.0

+
NH4 as % of N-Total: 1.75/(1.75+13.0)*100 = 11.9%

Concentra- Eqv Application

-1 -1 + + 2+ 2+
Fertilizer tion tank mass g 1000 L mmolc L NH4 K Ca Mg NO3- H2PO4- SO4-2
K2SO4 A 87 522 6 6.0 6.0
NH4H2PO4 A 115 115 1 1.0 1.0
5Ca(NO3)2.NH4NO3.10H2O B 108 891 9 0.75 8.25 9.0
Mg(NO3)2.6H2O B 128 512 4 4.0 4.0

Two stock solution tanks needed 1.75 6.0 8.25 4.0 13.0 1.0 6.0

The second option, shown in Table 7, shows that Mg-nitrate can be used as Mg source
to replace Mg-sulphate. This is only possible where the K-need is equal to, or higher than
the sulphate-need.

23
MIXING PROCEDURE, USING AMMONIUM-CONTAINING CA-NITRATE
The tomato nutrient solution is used in the following example, containing ammonium at 1
mmolc·L-1. The mixing procedure, as was previously demonstrated, was done with
ammonium-free Ca-nitrate. Using ammonium-containing Ca-nitrate, this procedure needs
adjustment, as discussed below and illustrated in Table 8.

Step 1
Deal with the nutrient at the lowest need first (ammonium). Using ammonium-containing
Ca-nitrate, this salt is dealt with first. Calculate the Ca-nitrate application (needed to
satisfy the Ca need) and then see how much ammonium was applied concurrently.
Calicum is needed at 8.5 mmolc·L-1. Multiply this Ca-need with 1.1 = 9.35 mmolc·L-1.
This is the level of nitrate in this application. With Ca2+ at 8.5 mmolc·L-1 the difference
(9.35-8.5) is ammonium (0.85 mmolc·L-1). Peppers need a low ammonium level, forcing
growers to use the ammonium-free Ca-nitrate product.

Table 8. Mixing procedure, using ammonium-containing Ca-nitrate.

-1
Nutriënt mix for tomatoes (mmolc L ), ignoring feeding water : 1 7 8.5 3.5 12.5 1.5 6

mol eqv. Application

FERTILIZER Adding mass mass (See y) NH4
+
K+ Ca++ Mg++ NO3
-
H2PO4
-
SO4
=

K NO3 Step 4 101 101 3.15 x 101 = 318 3.15 3.15

K2 SO 4 Step 5 174 87 2.5 x 87 = 218 2.50 2.50
K H2PO4 Step 3 136 136 1.35 x 136 = 184 1.35 1.35
NH4 H2PO 4 Step 2 115 115 0.15 x 115 = 17.3 0.15 0.15
X
5Ca(NO3)2.(NH4NO3.10H2O) Step 1 216 108 9.35 x 108 = 1010 0.85 8.50 9.35
Mg SO4.7H2O Step 6 246 123 3.5 x 123 = 431 3.5 3.5

1 7 8.5 3.5 12.5 1.5 6

X = Ammonium-containing Ca-nitrate
y = Nutrient concentration (mmolc L -1 ) x equivalent mass of fertilizer = Application (g 1000 L -1 )

Step 2
The next step is to add the smallest need, in this case ammonium. By adding 0.15 mmolc,
MAP, the NH4+-level is lifted from 0.85 to 1.0 mmolc.

Step 3
In the previous step, phosphate was lifted to 0.15, still needing 1.35 to reach a level of 1.5
mmolc. This can be done by using potassium phosphate as third step.

Step 4
The next smallest need is nitrate: Add K-nitrate at 3.15 mmolc.

Step 5
The need for K was lowered in the previous step. We already have nitrate and phosphate
at their required levels, thus 2.5 mmolc K-sulphate must be used.
Step 6
Both Mg and sulphate need 3.5 mmolc. This is applied using Mg-sulphate.

PLANNING A MICRONUTRIENT SOLUTION

Since high levels of Fe and Mn may block drippers, Fe and Mn should be removed from
the feeding water in advance. It is seldom possible to remove all the Fe and Mn but with
these nutrients at safer lower levels, Mn may be topped up to its required level. Due to its
quick oxidation and loss in irrigation pipes, the remaining Fe in the feeding water must be
ignored and the total Fe-need applied as chelated Fe (Deckers, 2002). As with

24
macronutrients, the other micronutrient levels in feeding water should also be topped up
to optimum levels. The procedure that should be followed when preparing nutrient
solutions (macro- and micronutrients) will be discussed after a closer look at the solubility
of salts.

SOLUBILITY OF SALTS
Some growers use partly shaded porta-pool reservoirs to mix their nutrient solutions. After
filling these reservoirs with water, fertilizers are dissolved into it, using pumps to circulate
the water. The problem is that algae grow in nutrient-rich water when it is exposed to light.
Even though the solution may not appear green, low concentrations of algae may still block
filters and drippers. The modern trend is to switch to smaller fertilizer tanks, containing 100x
concentrated stock solutions to be released into passing feeding water. However, some ion
combinations may form relatively insoluble salts that may precipitate in concentrated
solutions.

Why not Concentrate the Entire Nutrient Solution into One Tank? By concentrating a
standard nutrient solution with an EC of 2.2 mS·cm-1, you do not find a linear increase in
EC, as expected (Fig. 1).

Measured Expected
Electrical conductivity (mS/cm)

25

20

15

10

0
Std x2 x3 x4 x5 x6 x7 x8 x9 x10
Times concentrated

Fig. 1. Expected and measured electrical conductivity (EC) values of a standard nutrient
solution at increasing concentrations (up to 10x).

The lower than expected EC with the increase in concentration can be ascribed to ion
losses due to the precipitation of Ca-sulphate, Ca-phosphate, and other relatively insoluble
ion combinations.
Exactly the same problem exists for calcium and phosphate, especially at a relatively high
pH. Thus, Ca and phosphate should also be kept apart when solutions are concentrated.

25
 Another relatively insoluble ion combination is ammonium-magnesium-phosphate-
hexahydrate at pH = 7. This implies that not all three of these ions (NH4, Mg, and H2PO4)
may be together at high concentrations, should the pH of the concentration tank be higher
than 5.5.
When preparing stock solutions, ammonium-containing solutions should never be
exposed to high pH levels, where ammonium losses occur due to the formation of NH3
gas. The stock solution pH should always be below 6.5. Most growers with modern
greenhouses use concentrated stock solutions. They use special electronic or volumetric
equipment to control the release from the stock solution tanks into the feeding water, to
reach the greenhouse at the correct EC and pH.

Table 9. Nutrient solutions for different crops, grown with drain-to-waste production
systems. The EC values represent listed nutrients alone (Subscript x).
x
Macronutrients EC Micronutrients
-1 -1 -1
(mmolc.L ) (mS.cm ) (mg.L )
NH4 K Ca Mg NO3 H2PO4 SO4 Fe Mn Zn B Cu MO

Disa 1.7 1.9 1.4 1.0 3.3 0.7 2.0 0.60 0.45 0.55 0.25 0.20 0.03 0.04
Cymbidiium 1.0 2.8 2.0 1.5 4.0 0.8 2.5 0.73 0.45 0.55 0.25 0.20 0.03 0.04
Anthurium 0.3 3.8 2.5 1.9 6.1 0.8 1.6 0.85 0.85 0.15 0.25 0.30 0.04 0.05
Lettuce 0.7 5.5 5.8 1.0 10.0 1.0 2.0 1.30 1.00 0.55 0.25 0.30 0.05 0.05
Rose 0.7 4.3 6.5 2.5 10.4 1.1 2.5 Z 1.40 0.85 0.30 0.25 0.20 0.03 0.05
Gerbera 1.5 5.5 6.0 2.0 11.0 1.2 2.8 1.50 2.00 0.30 0.25 0.30 0.05 0.05
Gypsphila 0.7 4.0 7.3 3.0 12.0 1.0 2.0 1.50 0.85 0.55 0.25 0.30 0.05 0.05
Strawberry 1.0 5.0 6.5 2.5 10.0 1.0 3.0 1.50 1.15 0.75 0.45 0.10 0.05 0.05
Alstromeria 0.7 5.0 7.0 2.5 10.5 1.0 3.7 1.52 1.40 0.30 0.25 0.30 0.05 0.05
Amaryllis 1.2 6.3 6.0 2.0 12.0 1.0 2.5 1.55 0.55 0.55 0.30 0.30 0.04 0.05
Beans 1.0 5.5 7.0 2.5 12.0 1.0 3.0 1.60 0.85 0.30 0.25 0.20 0.03 0.05
Carnation 1.0 6.0 7.0 2.0 12.3 1.2 2.5 1.60 1.40 0.55 0.26 0.33 0.05 0.05
Crysanth 1.0 6.0 7.0 2.0 12.3 1.2 2.5 1.60 1.40 0.55 0.26 0.25 0.05 0.05
Brinjal 1.0 4.5 7.0 4.0 12.5 1.0 3.0 1.65 0.85 0.55 0.30 0.38 0.05 0.05
Courgette 1.0 5.5 7.0 3.0 13.0 1.0 2.5 Z 1.65 0.85 0.55 0.30 0.38 0.07 0.05
Cucumber 1.0 5.5 7.5 2.5 13.0 1.0 2.5 Z 1.65 0.85 0.55 0.30 0.30 0.05 0.05
Pepper 0.3 5.2 9.0 3.5 12.8 1.2 4.0 1.80 0.85 0.55 0.30 0.30 0.05 0.05
Tomato 1.0 7.0 8.5 3.5 12.5 1.5 6.0 y 2.00 0.85 0.55 0.30 0.30 0.05 0.05
Melon 1.2 7.0 9.8 4.0 13.0 1.5 7.5 y 2.20 0.85 0.55 0.25 0.35 0.05 0.05
Cherry toms 1.5 8.0 9.5 4.0 13.0 1.5 8.5 y 2.30 0.85 0.55 0.30 0.30 0.05 0.05
Swiss chard 1.0 7.0 10.8 4.2 13.0 1.5 8.5 y 2.30 0.85 0.55 0.25 0.35 0.05 0.05
Amaranthus 1.0 7.0 10.8 4.2 13.0 1.5 8.5 y 2.30 0.85 0.55 0.25 0.35 0.05 0.05
X
The crop recipes should produce these EC values with zero Na and Cl in the feeding water. These values may
be increased with 10% to 20% by concentrating the nutrient solutions during cold and low radiation
conditions. During sunny, hot and low humidity conditions (with poor climate control structures) the EC
values can be reduced by lowering the nutrient concentrations with 15 to 25%.
Y
The relatively high sulphate levels may be reduced with 2 up to 3 mmolc·L-1 and replaced with equivalent
levels of Cl, only with a drain-to-waste system and with the water Cl levels < 2 mmolc·L-1.
Z
Addition of silicon (Si) to these nutrient solutions may reduce the incidence of powdery mildew.

RECIPES FOR SPECIFIC CROPS

Standard (input) nutrient solutions should be changed when root zone solutions indicate that
imbalances develop. Recipes for drain-to-waste systems (Table 9) differ from those used in
closed systems but the root zone norms (not discussed in this paper) are exactly the same.
It is dangerous to use nutrient recipes from foreign countries, without understanding their
conditions. As an example, relatively high Mg levels occur in Israel’s feeding water. Thus,
hardly any Mg is used by Israeli growers. Dutch growers use inert media such as rock wool,
whereas most South African growers use organic substrates, such as pine sawdust. Side-

26
effects of decomposing organic material, such as lowered root zone N levels during the first
6 to 8 weeks after transplanting, associated with an increased root zone pH or alkalinity,
should be taken into account (Benoit, 2003).

Literature Cited
Arnon, D.I. and Hoagland, D.R. 1940. Crop production in artificial culture solutions and
in soils with special reference to factors influencing yields and absorption of inorganic
nutrients. Soil Sci. 50:443-476.
Bar-Yosef, B.T. 1995. Plant Nutrition Workshop. Association for Vegetables Under
Protection, Stellenbosch, November, 1995.
Benoit, F. 2003. Conductivity and pH. Chapter 4 of the ‘International Course for Soil-less
Crop production’, European Vegetable R&D Centre, Sint-Katelijne Waver, Belgium.
Combrink, N.J.J. and Kempen, E. 2011. Nutrient solution management.
www.greenhousehydroponics.co.za. Stellenbosch, South Africa.
Combrink, N.J.J. 2005. Nutrient solutions and greenhouse management.
www.greenhousehydroponics.co.za. Stellenbosch, South Africa.
Deckers, S. 2002. Sampling, interpretation and fertilization advice in hydroponics. AIPP-
Fertigation Workshop, Stellenbosch University, October, 2002.
De Kreij, C., Voogt, W., Van Den Bos, A.L. and Baas, R. 1999. Bemestings Adviesbasis
Substraten. Proefstation voor Bloemisterij en Glasgroente, Naaldwijk, The Netherlands.
Epstein, E. and Bloom, A.J. 2004. Mineral Nutrition of Plants: Principles and
Perspectives. Sinauer Associates, Inc. Publishers, Sunderland, Massachusetts.
LaHaye, P.A. and Epstein, E. 1971. Calcium and salt toleration by bean plants. Physiol.
Plant. 25:213-218.
Kafkafi, U., Valoras, N. and Letey, J. 1982. Chloride interaction with nitrate and
phosphate nutrition in tomato (Lycopersicon esculentum L.). J. Plant Nutr. 5:1369-1385.
Pienaar, D. 2005. A study of greenhouse production techniques for evergreen Disas. MSc
(Agric) thesis, University of Stellenbosch, South Africa.
Ribeiro, J. and Combrink, N.J.J. 2006. Amaranthus tricolor L leaf yields affected by
salinity, harvesting stage and harvesting methods. S. Afr. J. Plant Soil 23:29-37.
Steiner, A.A. 1968. Soilless culture. Proceedings of the 6th Colloquium of the
International Potash Institute, Florence, Italy.
Steiner, A.A. 1984. The universal nutrient solution. p.633-649. Proceedings of IWOSC
1984 6th International Congress on Soilless Culture.
Straver, N., De Kreij, C. and Verberkt, H. 1999. Bemestingsadviesbasis Potplanten.
Proefstation voor Bloemisterij en Glasgroente, Naaldwijk, The Netherlands.
Weigel, R.C., Schillinger, J.A., Mcgraw, B.A., Gauch, H.G. and Hsiao, E. 1973. Nutrient-
nitrate levels and the accumulation of chloride in leaves of snap beans and roots of
soybeans. Crop Sci. 13:411-412.

27
28