Agronomy Journal 92:336-345 (2000)
© 2000 American Society of Agronomy
NEW CROPS
Critical Phosphorus Levels for Salicornia Growth
Abdullah H. Alsaeedia and
Adel M. Elprincea
a Water Studies Center, King Faisal Univ., Al-Hassa, Saudi Arabia
aprince{at}kfu.edu.sa
 |
ABSTRACT
|
|---|
Basic information is needed to estimate P status of Salicornia bigelovii Torr. with increased precision if progress is to be made in adapting this halophyte as a conventional oilseed crop for direct seawater irrigation in coastal areas of arid regions. The P nutrition of salicornia was studied by using the nutrient solution technique in the greenhouse. The culture solutions were seawater, enriched to 1/2-strength modified Hoagland's solution [pH = 6.2 ± 0.2; electrical conductivity (EC) = 54 ± 2 dSm-1], and P treatments ranging from 0 to adequate amounts of P for full growth. All stages of deficiency symptoms were observed at harvest, ranging from stunted growth, necroses, purplish coloration, chlorosis, and dark bluish-green coloration to no symptoms. Dry and fresh weight of the tops were plotted against soluble P and total P values to give plant nutrient calibration curves that were fitted to the composite growth model by the least squares procedure. On the basis of the curve sharpness coefficient and range parameter output values and ease of sampling, semiwoody internodes, matured internodes and soluble P concentration were selected from the five plant parts sampled and the two forms of P determined as the best combination to diagnose the P status of Salicornia. About 480 mg kg-1 of soluble P in semiwoody internode tissue was tentatively set as the critical concentration for growth of the salicornia. The corresponding value for total P was set at 780 mg kg-1.
Abbreviations: EC, electrical conductivity • SRS, sum of residual squares • WOM, Ware–Ohki–Moon model
|
INTRODUCTION
|
|---|
THE HALOPHYTE Salicornia bigelovii Torr. is a potentially valuable new high-yielding oilseed crop for direct seawater irrigation in coastal areas of the arid regions (Glenn et al., 1991). Coastal regions of particular opportunity are the sandy desert seacoasts along the Gulf of California, Red Sea, Arabian Gulf, Indian Ocean, and the sand plains of south coast of Australia (Glenn et al., 1992). Salicornia is a leafless plant with green, jointed, succulent stems (internodes) that ultimately form terminal fruiting spikes in which seeds are borne (Wiggins, 1980, p. 367–368). Salinity tolerance of salicornia may be explained by the general compartmentation theory of salinity tolerance in angiosperms (Flowers, 1985; Gorham et al., 1985; Ayala et al., 1996). Salicornia in the field requires P fertilizer for satisfactory growth. The need for P has been indicated by field experiments (Riley and Abdal, 1992) and pot experiments (Alsaeedi, 1997). No attempt has been made, however, to utilize critical P concentration levels in salicornia tissue for growth evaluation in the field. This approach involves the concept of the critical nutrient concentration, which has been defined as that nutrient concentration at which growth first begins to decrease, and which is located within the transition zone that separates the zone of deficiency from the zone of adequacy (Marcy, 1936; Ulrich, 1952; Ulrich and Hills, 1990). The critical level is best estimated through the use of solution cultures where all nutrients except the one under study are supplied in ample amounts and all other environmental factors are favorable for growth (Ulrich and Berry, 1961).
Certain parts of a plant and their forms of P reflect the internal status of P better than others do. According to Ulrich and Berry (1961), the plant part selected should have the following criteria: (A) it should be comparable for all plants at all sampling dates; (B) it must have a relatively sharp transition zone from deficiency to adequacy; (C) it should have a broad range of concentration between deficiency and abundance; (D) it should have a relatively constant critical concentration; and (E) it should be relatively easy to sample. The form of the nutrient determined should be the one with a relatively fast and accurate method of determination and result in a calibration curve that fulfills criteria (A), (B), and (C) (Ulrich and Hills, 1990).
Freehand graphics is often used to determine the critical level of an element in a given part of the plant. This graphical procedure has been criticized (Ware et al., 1982) due to its arbitrariness. Another approach is fitting a mathematical model to the experimental critical level curve (Ware et al., 1982; Alsaeedi and Elprince, 1999). In particular, the composite model outputs additional useful parameters. These parameters have been employed successfully for the selection of the best combination of plant tissue and nutrient form as indicators for diagnostic purposes (Alsaeedi and Elprince, 1999).
Basic information is needed to estimate P status of salicornia with higher precision if progress is to be made in adapting this halophyte as a conventional agricultural crop. Our study was designed to achieve this goal. Our specific objective was to determine the most reliable plant tissue and P form to indicate the P status in salicornia grown in solution culture in the greenhouse and, additionally, to determine the critical P deficiency levels for the crop.
|
Materials and methods
|
|---|
Experimental Procedure
The experiment was carried out in the greenhouse and facilities of the Water Studies Center in Al Hassa, Saudi Arabia, about 70 km from the Arabian Gulf.
Seed Germination
Salicornia seeds (Salicornia bigelovii Torr. SOS 10) were sown on 2 Mar. 1998 in a plastic flat lacking drainage and doubly lined with polyethylene. Approximately 1000 seeds from a single plant were sown in acid-washed quartz coarse sand in each flat (2.33 kg sand/flat, pH 6.9). The flat was positioned 5 cm from vertical banks of long cool white fluorescent lights, which provided a 14-h photoperiod under 650 UE m-2 s-1 photosynthetically active radiation (PAR; 425–700 nm). The flat was flooded to saturation with distilled water and within 24 hours, abundant germination was observed. Plants were watered as necessary to avoid drought. On Days 9 and 56 after sowing, the flat was watered with 500 mL of 1/2-strength Hoagland's nutrient solution less P, salinized to 2 (2 g NaCl per 1000 mL distilled water). For the remainder of the time that the seedlings were in the flat, they were watered with distilled water. Seedlings stayed 28 d in the plant growth chamber and were then moved to the greenhouse.
Transplanting
All plants in the experiment were transplanted to their respective grow-out pots after 72 d. Individuals selected for transplanting appeared uniform and healthy. Randomly selected seedlings were transplanted to 2.5-cm-thick polystyrene rings with a 6-cm outside diameter and 1.5-cm inside diameter. Nonabsorbent cotton supported the seedlings within the rings. After 13 d, dead plants were replaced with new transplants.
Three seedlings were grown with their roots submerged in 20-L plastic buckets (white outside, black inside). The root environment was maintained dark to prevent the growth of algae in the nutrient solution. The rings that encompassed the plant stem were inserted into a hole in the bucket lids, which were made of 2.5-cm-thick polystyrene insulation material. The nutrient solution was a modified 1/2-strength Hoagland's nutrient solution (Epstein, 1972, p. 39), salinized with seawater and containing different P treatments (Table 1)
. The seawater was artificial and contained less P (Table 1) for the pots belonging to the P treatments (0, 1/512 and 1/256 P levels); natural seawater (seawater from Salwa, on the Arabian Gulf, Saudi Arabia, diluted with tap water to 3.5%) was used for the rest of the pots. The solutions in the pots were under 7-h daily aeration. The treatments were replicated four times and were arranged in the greenhouse in a randomized complete block design.
Monitoring pH of Solution Culture
The pH of the culture solutions was monitored every week using a portable pH meter (Model EC10, Hach Europe, Belgium). Any pH change beyond the 6 to 6.4 range was corrected by adding appropriate amounts of H2SO4 and NaOH at 6 molc L-1 as computed by our program AdjpHEC. This computer program included a code for a seawater titration curve. Accordingly, the pH was kept at the 6.2 ± 0.2 values.
Monitoring EC of Solution Culture
The EC was monitored every 2 wk using a portable conductivity meter (Model CO150, Hach Europe, Belgium). Any salinity decrease below 53 dSm-1 was corrected by adding an appropriate amount of solid NaCl as computed by the program AdjpHEC. This computer program included a code of seawater salinity vs. volume of NaCl(s) titration curve. Accordingly, the EC was kept at 54 ± 2 dSm-1 values.
Distilled Water Addition
Distilled water was added to compensate for transpiration losses, which were insignificant initially. When the plants became sizable, the pots required daily additions of about 2 L.
Nutrient Addition
At 1/8th strength, nutrients exclusive of P were added to each pot at the beginning of each week. Macro–micro (less P) stock solution (Table 2)
was added every week at a rate of 12.5 mL per 20-L pot.
Harvesting
Harvesting was done on 5 Sept. 1998. At harvest the plants showed a gradation of deficiency symptoms from severe to nonexistent, and varied in height from a few centimeters for dwarf plants to 45 cm for normal plants. The inflorescent tissue (spikes) were at the milk stage.
Preparation of Samples
The roots and plant tops were separated, placed on 2-mm stainless steel sieves and washed with 0.2% by volume liquid soap, rinsed with distilled water and air-dried overnight. The plants were separated into (i) woody internodes (old internodes), (ii) semiwoody internodes (matured internodes), (iii) young internodes, (iv) spikes, and (v) roots. All plant parts were placed in paper bags, dried at 70°C in a forced-air oven and weighed after 48 h. Total top weight was recorded as the sum of plant parts exclusive of roots. All dried materials were ground in a mill equipped with stainless steel cutting blades to pass through a 40-mesh stainless steel sieve and were stored in plastic containers until analyzed.
Chemical Analysis
Ground plant samples were analyzed for P soluble in 2% acetic acid and for total P in HNO3–HClO4 digests by the vanadomolybdophosphoric acid method (Johnson and Ulrich, 1959; Greenberg et al., 1985).
Composite Model Fitting of the Calibration Curves
The composite model (Alsaeedi and Elprince, 1999) identifies responsive (linear) and nonresponsive (plateau) regions and also provides a clear indication of where the critical point lies (joint from nonlinear regression). In addition, it is a conceptual model based on hypotheses underlying the growth process.
This composite model (1990) was proposed for curve fitting the three functional forms
 | (1) |
 | (2) |
 | (3) |
where
, the sum was over the adequate–constant yield zone (Block II data), and nII was the number of data pairs (xi, yi) in Block II;
, where xmax was the nutrient concentration corresponding to ymax, which was estimated by taking the average within Block II;
, the sum was over the deficient–step yield zone (Block I data) and nI was the number of data pairs (xi, yi) in Block I;
, where yo is the yield corresponding to xo, which was estimated by taking the average within Block I; and a and b were constants related to c, Yo, and Xo by the relationships
 | (4) |
and
 | (5) |
due to the constraint that there were no discontinuities in the resulting model, i.e.,
. Subsequently, b was a measure of the sharpness of the dimensionless nutrient calibration curve and was used to calculate the nutrient calibration curve sharpness coefficient b'
 | (6) |
where
and was denoted as the curve sharpness coefficient. We redefined ymax as the average over the number of plant parts under investigation, np, so that
 | (7) |
This definition guaranteed the conversion of the nutrient calibration curves to a single ymax value irrespective of the plant tissue type.
In deriving the above model equations, Alsaeedi and Elprince (1999) assumed that the law of diminishing returns applied to the yield and nutrient uptake, so that
 | (8) |
and
 | (9) |
where Xs is a reduced nutrient concentration in the soil solution and 
and ß are constants. Equation [8] assumes that the rate of increase in yield with the increase in nutrient concentration in soil solution is proportional to the decrement from the maximum obtained so far. Equation [9] assumes that the rate of increase in tissue nutrient concentration is proportional to the decrement from the maximum already obtained. The rate of increase in yield as a function of tissue nutrient concentration then follows the equation
 | (10) |
where
is a rate constant. Equation [10] indicates that the rate of increase in yield with increase in tissue nutrient concentration is directly proportional to the ratio of the decrement from maximum yield, (1 - Y), to the decrement from nutrient adequacy, (1 - X). Solving Eq. [10] yields Eq. [4] after employing the conditions that Y is equal to Yo when the tissue nutrient reduced concentration is equal to Xo; and Y = 1 when the reduced nutrient concentration is equal to 1.
Solving Eq. [2] for the tissue nutrient concentration x yields
 | (11) |
Substituting Y = 0.8, 0.9, and 0.99 into Eq. [11] yields x0.8, x0.9, and x0.99, respectively, where x0.9 is the critical level and x0.8 to x0.99 is the transition zone range. Another useful parameter is the range parameter r defined by the equation
 | (12) |
where v is tissue extract volume (e.g., 50 L kg-1) and d is the minimum detection limit for the nutrient analytical method (0.2 mg L-1 for PO4–P by the vanadomolybdophosphoric acid method) (Greenberg et al., 1985).
Data Blocks
Block II and I data are required to compute xo, yo, xmax, and ymax. They can be easily identified (Alsaeedi and Elprince, 1999) by examining the yield–tissue nutrient concentration data, (xi, yi), and the calculation of the 
function defined by the equation
 | (13) |
where k is an arbitrary constant. In moving from the low to high treatment levels, the magnitude of experiences two sudden jumps. Block I consists of the (xi, yi) pairs before the first jump. Block II consists of the (xi, yi) pair of the second jump and the pairs after. Sometimes, only one sudden jump in the magnitude of may occur. In this case, Block I again consists of the (xi, yi) pairs before the sudden jump. Block II, on the other hand, consists of the (xi, yi) pairs after the sudden jump. In both cases, Block III data are the (xi, yi) pairs that satisfy the constraints Y > Yo and X < 1. Subsequently, Y can be predicted for Block III using Eq. [4], using Eq. [3] for data pairs that satisfy the constraints X < 1 and Y < Yo and using Eq. [5] for pairs that satisfy the constraint X > 1.
WOM Model Fitting of the Calibration Curves
The Ware–Ohki–Moon model (Ware et al., 1982) in a reduced form is
 | (14) |
where 
and 
are constants.
The Goodness-of-Fit and Test of Significance
The goodness-of-fit of the model equations was described in terms of 
2 as
 | (15) |
where Yi was the y-axis value of the ith datum point, and Ypredicted was the predicted y-axis value based on the x-axis value of the ith datum point, Xi. If 2 = 1, then the goodness-of-fit is perfect; 2 
1, always. In the linear regression setup, 2 coincided with r2, the square of the correlation coefficient (Schulthess and Dey, 1996). The corresponding population correlation coefficient differs from zero at the 0.05 significance level as long as
(Steel and Torrie, 1960).
|
Results and discussion
|
|---|
Visual Symptoms of Phosphorus Deficiency
Phosphorus deficiency was marked in the salicornia plant by stunted growth, abnormal coloration, and necroses. The tops of deficient plants became slightly chlorotic after passing an earlier stage of dark bluish-green color. At this slightly chlorotic stage, the lower half of the top seemed slightly yellowish compared with the upper half. Deficient plants had a purplish color, apparently anthocyanins (Ozanne, 1980), that developed first in old internodes and then progressed toward the tips, which became necrotic. Some spikes had the purplish color in their tops. Top growth was reduced before abnormal coloration or necrosis was observed (Table 3)
. This behavior contrasts to that of the root, which appeared to grow normally under low P supplies (Table 3). The roots of P-deficient plants had an abnormal purplish color. A purplish pigment could be expressed from the roots. Presence of the pigment in phloem and absence in xylem indicated internodes were the origin. No abnormal growth was observed with very high P solution concentrations. Phosphorus seemed readily mobilized in salicornia, and when a deficiency occurred, the element contained in the older internode tissues was transferred to the active meristematic regions. These visual symptoms of P deficiency in the halophyte salicornia seem, generally, not different from those reported for nonhalophyte plants (Ulrich and Berry, 1961; Hylton et al., 1965; Ozanne, 1980).
View this table:
[in this window]
[in a new window]
|
Table 3 Effect of P supply in the solution culture on visual deficiency symptoms of tops and on fresh and dry weights of salicornia
|
|
Distribution and Remobilization of Phosphorus in Salicornia
As seen in Table 4
, internodes of different physiological age vary in soluble P concentration. As the internodes matured, they decreased in concentration of soluble P under deficient (384, 246, and 179 mg kg-1 for young, semiwoody, and woody internodes, respectively), as well as under adequate conditions (1484, 1316, and 1216 mg kg-1 for young, semiwoody, and woody internodes, respectively) (Table 4). Furthermore, the reproductive tissues (spikes) appear to have the highest concentration of soluble P compared with the other plant parts under both deficient (728 mg P kg-1) and adequate (2601 mg P kg-1) conditions (Table 4). These results indicate that the internodes may remobilize P to the younger internodes and spikes at both low and high levels of rhizosphere P fertility. The remobilization of P from internodes seems to be a part of an organized senescence (Grabau et al., 1986; Hanway and Weber, 1971). The higher concentration of soluble P in the spikes relative to the internodes could also be related to a higher metabolic activity in portions of the plant where cell division and growth are more rapid (Hylton et al., 1965).
View this table:
[in this window]
[in a new window]
|
Table 4 Mean concentration of soluble P and mean dry weight of plant parts of salicornia under deficient and adequate P supply
|
|
Woody internodes may be regarded as an extension of the vascular tissue of the root. Under P deficiency, mean concentration of soluble P in the root was 212 mg P kg-1 and in its vascular tissue was approximately 179 mg P kg-1 (Table 5)
. If the mass of vascular tissue is assumed to be 50% of total root, then soluble P proportional to total root is 42% in the vascular tissue and 58% in the cortex. Under adequate P supply, similar calculation indicates that soluble P proportional to total root is 26% in the vascular tissue and 74% in the cortex. Thus, under both low and high P supply, the root cortical tissue seems to accumulate soluble P actively, while stele tissue may leak (Crafts and Broyer, 1938) and actively become loaded (Gorham et al., 1985). However, low-P plants seem to have low energy demands compared with high-P plants due to their relatively lower concentrations gradients.
View this table:
[in this window]
[in a new window]
|
Table 5 Means of plant parts total-P proportional to total plant for salicornia (%). Standard deviations are in parentheses
|
|
Under adequate P supply, if negligible P translocation from tops to root is assumed (Clarkson and Scattergood, 1982), then soluble P concentration in the root's phloem is near zero and the rate of P uptake is equal to 2376 x 0.26 = 618 mg P kg-1root. Under P deficiency, on the other hand, P translocation from tops to root is approximately equal to 212 x (0.42 - 0.26) = 34 mg P kg-1root through the phloem; from root to tops through the xylem, 212 x 0.26 = 55 mg P kg-1root. These movements result in a net uptake rate equal to 21 mg P kg-1root. Since these rates apply to mature plants, where mass of young internodes is only about 10% of total tops, most of the transported soluble P from the root ends in the spikes. This translocation contributes only 1.5% to the total soluble P in spikes under deficiency and 7% under adequate P supply. Thus, P remobilization from woody and semiwoody internodes seems to be the major source of soluble P in spikes under both low and high rhizosphere P fertility.
Total P distribution was significantly (P < 0.05) similar in low-P and high-P plants (Table 5). A large proportion of plant P was invested in spikes, as they contained about 1/2, the internodes about 1/3, and the roots about 1/6 of plant P (Fig. 1)
. The ability of salicornia to compartmentalize total P seems to be independent of the rhizosphere P fertility.
View larger version (121K):
[in this window]
[in a new window]
|
Fig. 1 Compartmentation of total P in salicornia plant parts (roots, woody internodes, semiwoody internodes, young internodes, and spikes at the milk stage)
|
|
Total P consists of free orthophosphate (soluble P) plus organically combined P. The percent of soluble P relative to total P was in general similar (66%) in low-P and high-P plant parts (Table 6) . An exception was a lower proportion of soluble P (30%) in roots of low-P plants. These results indicate that salicornia roots retain organic P when the rhizosphere's P levels are low. This behavior is in contrast to that of internodes, which remobilize P to the spikes. Thus, root systems grown on low-P rhizosphere solution may benefit at the expense of the rest of the plant by retaining much of the P absorbed. Phosphorus taken up by the root may be utilized first to meet local demands for P (e.g., for energy transformation, maintenance of membranes, and genetic molecules) and only the excess P translocated to distance sinks. Thus, low-P roots did not remobilize P and consequently continued to grow. This behavior contrasts that of the internodes, which decreased in mass at low-P fertility.
View this table:
[in this window]
[in a new window]
|
Table 6 Means of soluble-P proportional to total-P in different parts of salicornia plant (%). Standard deviations are between brackets
|
|
The above results of P distribution and remobilization in the halophyte salicornia seem generally not different from those reported for nonhalophyte plants (Snapp and Lynch, 1996).
Relation of Phosphorus Contents to Plant Growth in Salicornia
The relation of P contents to plant growth in salicornia has been employed for the selection of the best combination of plant part and P form for diagnostic purposes (Ulrich and Berry, 1961). The tissue P concentration, and subsequently the critical P concentration, are a function of the P supply and of the plant part physiological age (Table 7)
. By selecting internodes of approximately the same age at all sampling dates, growth becomes largely independent of plant part age and dependent on the P tissue concentration, as influenced by the balance between the P supply and the P requirements for growth by the plant. If the internode is always sampled at the same physiological age (e.g., semiwoody internode), the critical concentration will remain much the same throughout the year as required by criterion A from Ulrich and Berry (1961).
The nearly vertical portion of the curve (Fig. 2)
shows a zone of P deficiency characterized by increases in top weight but relatively constant soluble P concentration with a gradual increase in P supply. The horizontal portion of the curve shows a zone of P adequacy where soluble P concentrations increase and top weights are relatively constant with increase in P supply. The zone of transition (0.8 Y 0.99) for semiwoody internode tissue covers a range from 372 to 773 mg kg-1. The soluble P concentration in this zone of the curve at 10% from satisfactory top growth is 483 mg kg-1. This value is the critical soluble P concentration. The corresponding value for total P in semiwoody internodes is 781 mg kg-1.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 2 Composite model fitting a P deficiency calibration curve for Salicornia with one adjustable parameter c. The sum of residual squares are plotted against c in the inset figure
|
|
The calibration curves for the two P components, soluble P and total P, in semiwoody internodes (Fig. 3)
show the soluble P curve is a little sharper than the total P curve. For simplicity, the raw data (Table 7) have been deleted in Fig. 3. Either a soluble P or total P analysis of semiwoody internodes would serve as a basis for an estimation of the P status of the salicornia plant (Fig. 3 and Table 8)
. Since the method for acetic acid soluble P is very fast and accurate, this method is preferable to the total P analysis.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3 Curves of the relation among relative yield (unitless) of dry weight of tops and total P and soluble P concentrations in semiwoody internodes of salicornia
|
|
View this table:
[in this window]
[in a new window]
|
Table 8 Correlation coefficient (r) and linear regression (y = a1 + a2 x), where y is soluble P concentration and x is total P concentration in plant tissues in g kg-1
|
|
Figure 4
shows the relative yields of dry weight tops and fresh weight tops against soluble P in semiwoody internodes for all P treatments. The sharpness of the calibration curve based on dry weight tops is higher than that based on the fresh weight tops. Consequently, the curve based on dry weight tops appears more reliable for the determination of critical P deficiency levels.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 4 Curves of the relation among dry and fresh weights of tops and tissue soluble P concentrations in semiwoody internodes of salicornia
|
|
The calibration curves in Fig. 5
help the selection of the best plant part to serve as an indicator of the plant's overall P status. The conductive tissues (roots and woody internodes) both have a comparatively sharp transition zone (Fig. 5). Therefore, any one of the conductive tissues would fulfill the criteria for a sharp transition zone (criterion B). Furthermore, based on this criterion, semiwoody internode tissue is a better indicator than young internodes and spikes.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 5 Curves of the relation among dry weight tops and tissue soluble P concentrations in five parts of the salicornia plant
|
|
Along with this sharpness of the transition zone, it is also desirable to have the range of soluble P concentration between deficiency and adequacy as large as possible if we are to detect a difference readily (criterion C). The range of P concentration of the conductive–absorptive tissue (2200 mg kg-1) exceeds that of the conductive (1060 mg kg-1) and photosynthetic tissues (1210–1970 mg kg-1). Thus the root tissue meets criterion C more fully than the other plant parts and the semiwoody internodes are preferable to the woody internodes.
Criterion D, the constancy of the critical nutrient concentration, remains to be determined from more greenhouse experiments conducted for several seasons.
Criterion E for selecting the indicator part of the plant concerns the relative ease of sampling. Based on this criterion, the root tissue is eliminated and the semiwoody internodes become preferable to the woody internodes. (The mass of the woody internodes, compared with the mass of the semiwoody internodes, is relatively small.)
The young internodes are less desirable as the indicator part of the plant to sample due to the broader transition zone (Fig. 5). With regard to sharpness of transition zone and range of soluble P concentration, there was little choice among the woody and semiwoody internodes. The greater difficulty of sampling eliminates the woody internodes. Therefore, the semiwoody internodes are preferable to the other tissues because of their ease of sampling, wide range of soluble P concentration, and relatively narrow range of the transition zone.
Fitting the Composite Model to Data
The curves in Fig. 2 to 5 resulted from the fit obtained by the least squares procedure, using the composite model equations directly and considering c to be the only adjustable parameter. The sum of residual squares (SRS) obtained from the data with different trial values of c have been plotted in the inset figure, showing a minimum SRS at c = 6.1, which was the value of c used to plot the curve (Fig. 2). The agreement is very good with a goodness of fit (2) equal to 0.872. Estimates of the model input and output parameters for the curves in Fig. 2 to 5 are presented in Table 9
together with the corresponding total P response curves. Table 9 also contains model parameters for similar set of curves based on fresh weight tops instead of dry weight tops.
The reduced yield for deficient plants (Yo) seems approximately constant and equal to about 0.5 (Table 9). The corresponding reduced concentration of soluble P (Xo) was tissue-dependent with concentrations following the order roots < woody internodes < semiwoody internodes < young internodes < spikes. Roots of deficient plants had soluble P at only about 8% of their adequate concentration, the woody internodes at about 14%, the semiwoody internodes at about 16%, the young internodes at about 21%, and the spikes at about 26%. Furthermore, the roots of deficient plants are 92% depleted of their soluble P and only 80% depleted of their total P concentration (Table 9). These results are in agreement with our previous conclusions that salicornia roots grown under low-P rhizosphere solutions may benefit at the expense of the rest of the plant by retaining much of absorbed P, while the internodes remobilize P to the spikes.
The growth rate constant c varied from 2.7 to 30 and the curve sharpness coefficient b' from 0.65 to 8.80 depending on tissue type, tissue age, and P form (Table 9). Generally, the values of c and b' for the conductive tissues were higher than for the photosynthetic tissues. In addition, the values of c and b' increased as the internode's age increased. It seems that plant growth is more dependent on P concentration in the source tissues than in sink tissues.
The goodness-of-fit (2) values varied from 0.694 to 0.919 and seemed independent of tissue type and age, P form, and whether the calibration curves were based on fresh or dry weight tops (Table 9).
Criterion C for the selection of the indicator combination of tissue type and nutrient form requires a calibration curve that should have a broad range in concentration between deficiency and abundance if a difference is to be detected readily. A quantitative measure for this range is the range parameter r. The values of r for the roots are generally higher than for the other tissues (Table 9). In addition, the values of r for the total P-based curves were generally higher than for the soluble P-based curves. In all of these cases, however, the magnitude of r was high enough to guarantee the detection of a difference readily. Thus, any of the plant parts, P form, and fresh or dry tops selections fulfilled criterion C.
Criterion B, on the other hand, requires a calibration curve with a relatively sharp transition zone. A quantitative measure of the sharpness of the calibration curve is the curve sharpness coefficient b'. The values of b' for the calibration curves based on dry weight tops were higher than those based on fresh weight tops (Table 9). Consequently, the response curves based on dry weight tops appear more reliable for the determination of critical P deficiency levels for salicornia growth.
Furthermore, based on the values of b', soluble P is a better indicator nutrient form than total P (Table 9). In addition, the conductive tissues are better indicators than the photosynthetic tissues. Within the photosynthetic tissues, the semiwoody internodes are better indicators.
In summary, the computation of the model parameters r and b' lead to the same conclusions derived from the curve drawing in Fig. 2 to 5 regarding the selection of the best combination of P form and plant tissue for diagnostic purposes. However, the model parameters b' and r gave a better means of quantifying the selection procedure than was previously available.
Model Verification
It should be mentioned that the above curve-fitting was done using the composite model after verifying the finding by Alsaeedi and Elprince (1999) which was that the composite fitting procedure gave a better means of quantifying the nutrient deficiency calibration data than the WOM model. The verification process was done using the goodness-of-fit parameter 2 as a basis for comparing the degree of fit for the composite and WOM models. The goodness-of-fit by the composite model (0.82 ± 0.07) significantly (P < 0.05) exceeded the WOM model (0.75 ± 0.10) for the 20 calibration curves of salicornia at hand.
Phosphorus Critical Levels
The estimated critical P levels in the halophyte salicornia (Table 9) are generally within the range of values reported for nonhalophyte plants (Ulrich and Berry, 1961; Hylton et al., 1965; Ozanne, 1980). The estimated critical level for soluble P in the semiwoody internodes is 483 mg kg-1 (Table 9). Based on the interpretation by Ulrich and Berry (1961), the probability of a P response would be relatively high in any plant with soluble P values of 483 mg kg-1 or less. If soluble P in the plants were between 483 and 773 mg kg-1, a response from a P treatment would be less likely to occur. The probability of a response would be relatively small with a soluble P concentration of 773 mg kg-1 or more. For plants with a soluble P level below 483 mg kg-1, the response to P fertilization will be greater the longer comparable plants remain below 483 mg kg-1.
Phosphorus Status in Salicornia versus Glycophytes
Salinity tolerance of salicornia may be explained by the general compartmentation theory of salinity tolerance in angiosperms (Flowers, 1985; Gorham et al., 1985; Ayala et al., 1996). In this model, Na+ and Cl-, which overwhelmingly predominate as the inorganic ions in halophytes, are believed to be sequestered into the vacuole, while various specific enzymatically compatible organic solutes are produced and retained in the cytoplasm as a means of balancing the water potentials between the two compartments. The basis of the model is the discovery that in vitro, the enzymes of halophytes are no more salt-tolerant than those of salt-sensitive glycophytes. A subsequent implication of this discovery is the hypothesis that the enzymatically based P nutrition in halophytes may not be different from the range of values in glycophytes. The present results on P status in the halophyte salicornia are in support of this hypothesis. As shown above, P deficiency symptoms, distribution, remobilization, and critical levels in salicornia seem to have the behavior of and to be in the range of values of nonhalophyte plants.
|
ACKNOWLEDGMENTS
|
|---|
The Saudi Basic Industries Corp. (SABIC) supported this project under contract FR9704.
Received for publication May 13, 1999.
|
REFERENCES
|
|---|
- Alsaeedi A.H. The response of salicornia to fertilizers under irrigation with sea water: A pot experiment. Alexandria J. Agric. Res. 1997;42:187-193.
- Alsaeedi A.H., Elprince A.M. A composite model for plant nutrient calibration curves. Water Studies Bull. 1999;40:1-21.
- Ayala F., O'Leary J.W., Schumaker K.S. Increased vacuolar and plasma membrane H+–ATPase activities in Salicornia bigelovii Torr. in response to NaCl. J. Exp. Bot. 1996;47:25-32.
- Clarkson D.T., Scattergood C.B. Growth and phosphate transport in barley and tomato plants during development of, and recovery from, phosphate-stress. J. Exp. Bot. 1982;33:865-875.[Abstract/Free Full Text]
- Crafts A.S., Broyer T.C. Migration of salts and water into xylem of the roots of higher plants. Am. J. Bot. 1938;25:529-535.[Web of Science]
- Epstein E. Mineral nutrition of plants: Principles and perspectives. New York: John Wiley & Sons, 1972.
- Flowers T.J. Physiology of halophytes. Plant Soil 1985;89:41-56.
- Glenn E.P., Hodges C.N., Lieth H., Pielke R., Pitelka L. Growing halophytes to remove carbon from the atmosphere. Environment 1992;34:40-43.
- Glenn E.P., O'Leary J.W., Watson M.C., Thompson T.L., Kuehl R.O. Salicornia bigelovii Torr.: An oilseed halophyte for seawater irrigation. Science (Washington, DC) 1991;251:1065-1067.[Abstract/Free Full Text]
- Gorham J., Wyn Jones R.G., McDonnell E. Some mechanisms of salt tolerance in crop plants. Plant Soil 1985;89:15-40.
- Grabau L.J., Blevins D.G., Minor H.C. P nutrition during seed development: Leaf senescence, pod retention, and seed weight of soybean. Plant Physiol. 1986;82:1008-1012.[Abstract/Free Full Text]
- Greenberg, A.E., R.R. Trussell, and L.S. Clesceri (ed.). 1985. Standard methods for the examination of water and wastewater. 6th ed. Am. Public Health Assoc., Washington, DC.
- Hanway J.J., Weber C.R. Accumulation of N, P, and K by soybean [Glycine max (L.) Merrill] plant parts. Agron. J. 1971;63:406-408.[Abstract/Free Full Text]
- Hylton L.O., Ulrich A., Cornelius D.R., Okhi K. Phosphorus nutrition of Italian ryegrass relative to growth, moisture content, and mineral constituents. Agron. J. 1965;57:505-508.[Abstract/Free Full Text]
- Johnson C.M., Ulrich A. Analytical methods for use in plant analysis. Calif. Agric. Exp. Stn. Bull. 1959;766:25-78.
- Marcy P. The quantitative mineral nutrient requirements of plants. Plant Physiol. 1936;11:749-764.[Free Full Text]
- Ozanne P.G. Phosphate nutrition of plants: A general treatise. In: Khasawneh A., et al. , ed. The role of phosphorus in agriculture. Madison, WI: ASA, CSSA, and SSSA, 1980:559-585.
- Riley J.J., Abdal M. Preliminary evaluation of Salicornia production and utilization in Kuwait. In: Lieth H., Al Massoum A., eds. Towards rational use of high salinity tolerant plants. Netherlands: Kluwer Academic Publ, 1992:1-11.
- Riley J.P., Skirrow G. Chemical oceanography. Vol. 1, 2nd ed New York: Academic Press, 1975.
- Schulthess C.P., Dey D.K. Estimation of Langmuir using linear and nonlinear least squares regression analysis. Soil Sci. Soc. Am. J. 1996;60:433-442.[Abstract/Free Full Text]
- Snapp S.S., Lynch J.P. Phosphorus distribution and remobilization in bean plants as influenced by phosphorus nutrition. Crop Sci. 1996;36:929-935.[Abstract/Free Full Text]
- Steel R.G.D., Torrie J.H. Principles and procedures of statistics. New York: McGraw-Hill, 1960.
- Ulrich A. Physiological basis for assessing the nutritional requirements of plants. Annu. Rev. Plant Physiol. 1952;3:207-228.
- Ulrich A., Berry W.L. Critical phosphorus levels for lima bean growth. Plant Physiol. 1961;36:626-632.[Free Full Text]
- Ulrich A., Hills F.J. Plant analysis as an aid in fertilizing sugar beets. In: Westerman R.L., ed. Soil testing and plant analysis. Madison, WI: ASA, CSSA, and SSSA, 1990:430-448.
- Ware G.O., Ohki K., Moon L.C. The Mitscherlich plant growth model for determining critical nutrient deficiency levels. Agron. J. 1982;74:88-91.[Abstract/Free Full Text]
- Wiggins I. Flora of Baja, California. Palo Alto, CA: Stanford Univ, 1980.
TOP
ABSTRACT
INTRODUCTION
