| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Agronomy, 1575 Linden Drive, Univ. of Wisconsin-Madison, Madison, WI 53706
* Corresponding author (ndeleongatti{at}students.wisc.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A single-eared morphology may not be the most efficient design under current agronomic practices. For example, after the switch from hand to machine harvest, starting in the late 1930s, it has been repeatedly shown that prolificacy is highly and positively correlated with grain yield (Russell, 1984; Carlone and Russell, 1987; Subandi, 1990; Duvick, 1997), although prolificacy can also be associated with poor stalk strength and standability (Carena et al., 1998; Jampatong et al., 2000; Thomison and Jordan, 1995).
There is a general agreement that prolificacy in maize is a relatively simply inherited trait; however, no consensus exist about the number of genes and the nature of their expression (Duvick, 1974; Harris et al., 1976; Sorrells et al., 1979; and Doebley and Stec, 1993). The relationship between ears per plant and grain yield has been also investigated in the Golden Glow maize population (Coors and Mardones, 1989; Maita and Coors, 1996). Maita and Coors (1996) evaluated selection response at four planting densities: 28889, 44444, 57778, and 73333 plants ha-1. After 20 cycles of selection for prolificacy in Golden Glow [GG(MP)], ear number increased 0.03 ears plant-1 cycle-1 while grain yield per plant increased 1.3 g cycle-1, averaged over all four densities. The correlation between ear number and grain yield per plant was r = 0.71*. Furthermore, the more advanced cycles had progressively greater yields at the higher planting densities than the earlier cycles, suggesting that increased prolificacy conferred greater advantage under more competitive conditions. Over the 20 cycles of selection, for each increase of 0.1 ears plant-1 in the GG(MP), the optimum planting density (that which produced the highest grain yield) increased by approximately 3 100 plants ha-1. The most important advantage of prolificacy, however, may not come from producing more ears per se, but rather from the avoidance of barrenness under intense competition (Tollenaar et al., 1992).
The morphology of many of the prolific plants from advanced cycles of GG(MP) was distinct from that of a typical maize plant. Prolific plants tended to have more tillers, and the lateral branches subtending ears (shanks) on the main stalk were longer than normal.
The process of crop domestication in maize involved an increase in apical dominance, which is associated with increased concentration of resources in the main stalk and subsequent suppression of lateral branches (Doebley et al., 1997). It has been shown that five genomic regions may control the major traits differentiating cultivated maize from teosinte [Z. mays subsp. mexicana (Schrader) Iltis] (Beadle, 1980; Doebley and Stec, 1993). Among these genomic regions there is one of distinctly large effect that has been identified as the teosinte branched1 (tb1) gene. Recently, molecular cloning and analysis of tb1 suggests that its protein functions as a repressor of organ growth (Doebley et al., 1997). Changes in expression have been suggested as the most likely explanation for the evolutionary differences observed between maize and teosinte (Doebley and Lukens, 1998).
In recent cycles, the most extreme plants in the GG(MP), when grown under optimal conditions and low planting densities, had a teosinte-like morphology where many of the nodes of the main stalk produced very long primary lateral branches that may occasionally produce secondary lateral branches at several nodes terminating in multiple female inflorescences. To better characterize changes in morphology due to selection in GG(MP), the objective of our study was to estimate direct and correlated responses of a number of morphological and plant-architectural traits in GG(MP) after 24 cycles of mass selection for prolificacy.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
At harvest, the uppermost ears from approximately 100 to 300 plants were selected on the basis of prolificacy and a visual assessment of plant health, root and stalk strength, and grain moisture. A balanced composite was made from the selected ears for the next cycle of selection. On the basis of the size of isolation and planting density, selection intensity was approximately 2.5 to 5.0% for the first 12 cycles, and 0.5 to 1.0% thereafter. In preparation for this study, seed stocks from Cycles 6, 12, 18 and Cycles 0, 23, 24 were renewed in 1992 and 1995 respectively, by making balanced composites of at least 100 plant-to plant crosses for each cycle. All seed was stored at approximately 40°F and 400 g kg-1 humidity until the initiation of this experiment.
Experimental Design
The evaluation experiment was conducted at two locations, Madison and Arlington, WI, in 1997. The soils at both sites are Plano silt loam (fine-silty, mixed, mesic Type Arguidoll). At the Madison location, nitrogen credits from the prior year's crop of alfalfa (Medicago sativa L.), together with fall and winter applications of manure were sufficient to supply all needed nutrients in 1997. At Arlington, 168 kg ha-1 N was applied in the form of anhydrous ammonia prior to planting. Treatments were restricted in split plots in a randomized complete block experimental design with four replications in each location. At each location, main plots were plant densities (15 070 and 73 197 plants ha-1). The entries in the subplots were Cycles 0, 6, 12, 18, 23, and 24 of prolificacy selection from the GG(MP) [GG(MP) C0, C6, C12, C18, C23, and C24]. Subplots consisted of two 6.1-m rows with 0.8 m between rows. The Madison location was hand-planted on 5 May 1997, and the Arlington location was hand-planted on 17 May 1997. Subplots were over-planted and thinned to 7 plants row-1 for the lower density (15 070 plants ha-1) and 34 plants row-1 for the higher planting density (73 197 plants ha-1). Trials were fertilized according to soil recommendations, and weeds were chemically and manually controlled. Ten competitive plants were measured in each subplot at both densities and at both locations for all traits evaluated before plant senescence. After senescence, 10 plants were measured for traits at 15 070 plants ha-1, while 15 plants were measured for traits at 73 197 plants ha-1. The average of the 10 (or 15) measurements in each subplot was the experimental unit for statistical analysis.
Approximately 2 wk after pollen shed, the number of visible ear shoots was counted. The heights of the uppermost and the lowermost visible ear shoot on the main stalk were determined as the distance from the first visible node above the ground to the node of ear attachment. The position (node) of those ear shoots was obtained as the number of nodes from the first visible node above the ground to the node under consideration. The total length of the main tassel was obtained as the distance from the lowermost primary tassel branch to the tip of the central branch. The number of primary tassel branches was also counted. The length of the main stalk was measured as the distance from the first visible node above the ground to the uppermost node under the flag leaf. The number of nodes on the main stalk was counted, which allowed the estimation of average internode length of the main stalk.
Tillers (the set of lateral branches originating from a main stalk node located below ground level) were counted, and the inflorescence type at the tip of the lowermost tiller (the tiller originating from the lowermost node) was classified as female, male, or mixed. Tiller traits were evaluated only at the planting density of 15 070 plants ha-1.
After senescence, the total length of the lowermost tiller and the uppermost primary lateral branch on the main stalk were measured from the node of attachment on the main stalk to the node at the base of the terminal inflorescence. Node number of tillers and the uppermost lateral branches were recorded, and the average internode length below the ear was calculated.
The direct response to selection, number of ears per plant, was estimated as the total number of ears hand-harvested from the main stalk and lateral branches of evaluated plants. Ears were collected and classified according to their position on the plant, relative to the uppermost active node on the main stalk. All ears from a given class were placed on a scaled table to measure length and maximum width and obtain an overall average for each specific ear class and for each plot evaluated. The number of kernel rows per ear was also counted for each ear, and the mean over all ears of a given class was calculated for each plot. Ears were shelled, and the cob width was measured.
The kernels obtained from each ear were counted and weighed. The moisture content was determined from a representative sample obtained from each ear class, and kernel dry weight (0 g kg-1 H2O) was calculated for all ears of a given class from each plot. The average kernel depth was estimated as one half the difference between the average ear width and cob width for each ear class from each plot.
Statistical Analysis
Independent analyses were performed for each of the phenotypic variables. The analysis included traits described as direct selection response (i.e., ears per plant) and correlated responses to selection (i.e., traits related to plant and ear morphology) and was performed by the MIXED procedure of SAS System (Littell et al., 1996) for a split plot design (Milliken and Johnson, 1994). Locations and replications were considered random effects, and density and cycles of selection fixed effects.
The restricted maximum likelihood method (REML) was used for variance component estimation, and F tests were performed to test the fixed effects of cycle, density, and their interactions. The error terms for testing the density, cycle, and density x cycle interaction effects were the location x density, location x cycle, and location x density x cycle interactions, respectively. Whenever the F test for the effect of cycles was significant, the orthogonal polynomials methodology proposed by Carmer and Seif (1963) was used to study the linear and quadratic effects of cycles on each dependent variable in each of the two densities. Linear and quadratic regression coefficients were calculated for each trait depending on the significance of the F test performed. If the linear contrast was significant, the response per cycle was the linear regression coefficient, and it was referred to as a least-squares response. If the quadratic contrast was significant, the response was referred to as the average response, and was calculated as (C24–C0)/24. Protected LSD tests (Snedecor and Cochran, 1989) were also performed for comparing means of cycles. Phenotypic correlations were used to examine associations between traits.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There were significant cycle x planting density interactions for four (ears per plant, height and node position of the lowermost ear shoot, and number of kernels per plant) of the 21 traits analyzed at both planting densities (data not provided). For most traits, location, cycles x location, plant density x location, and plant density x replication interactions were not significant (data not provided).
Number of ears per plant, which represents the direct response to selection, had a significant cycle x planting density interaction, and, therefore, results are provided for each planting density. For other traits, changes over the 24 cycles of selection were similar in direction when measured at either the higher or the lower planting densities. However, the magnitude of change was usually greater at the lower planting density, and, therefore, for ease of illustration, most emphasis will be focused on response at the lower planting density.
Direct Response—Ears per Plant
The number of ears per plant had a quadratic response at the lower planting density, whereas the response was linear at the higher density (Table 1). Selection response for ears per plant was 0.14 ears cycle-1 (8.7% per cycle) at the lower planting density, and 0.03 ears cycle-1 (4.7% cycle-1, Table 2) at the higher planting density. The quadratic response at the lower density was due to a change in response beginning at C18. The response from C0 to C18 was similar to the 3.3% cycle-1 response reported by Maita and Coors (1996) after 20 cycles of selection in GG(MP). In the current study, the number of ears per plant after C18 seemed to increase at a faster rate. For example, the increase of 2.24 ears plant-1 from C18 to C24 was more than twice the change of 0.90 ears plant-1 that occurred from C0 to C12.
|
|
Correlated Responses
Main Stalk
The length of the main stalk showed a quadratic response to selection at both planting densities that reflected an initial reduction in stalk length and an increase after C12 similar to the C0 (Tables 1, 2). The number of nodes and the length of the main stalk internode, however, showed linear responses at the lower planting density (0.04 nodes cycle-1 and -0.04 cm cycle-1, respectively), and C0 and C24 differed significantly for both traits at the lower density. The increase in nodes was compensated by the decrease in internode length, and the overall length of the main stalk therefore remained unchanged.
Uppermost Lateral Branch
The total length of the uppermost lateral branch had a quadratic response to selection, decreasing initially, and then increasing to a length significantly greater than the length at C0. At the lower planting density, the average selection response was 0.28 cm cycle-1. The number of nodes of the uppermost lateral branch decreased linearly at a rate of -0.10 nodes cycle-1. Internode length had a quadratic response similar to total length, and the response at the lower planting density was 0.04 cm cycle-1. The elongation of the internodes, rather than the development of new nodes, produced the increase in total length of the uppermost lateral branch. Similar trends were observed at the higher planting density (Table 2).
While total length of both the uppermost lateral branch and the main stalk showed quadratic selection responses, the nature of the responses differed (Table 1). The length of the uppermost branch increased significantly from C0 to C24, while the length of the main stalk at C0 and C24 did not differ (Table 2). Also, whereas the number of nodes increased and the internode length decreased in the main stalk, the opposite occurred for the uppermost lateral branch.
The selection response for the position of the uppermost lateral branch was quadratic, when significant (Table 1). At the lower planting density, the height and position of the uppermost lateral branch increased by 0.51 cm cycle-1 and 0.04 nodes cycle-1. The change in height and position of the lowermost lateral branch was opposite in direction and of greater magnitude. At the lower planting density, the lowermost lateral branch decreased by 1.57 cm cycle-1 and 0.08 nodes cycle-1, respectively. There were significant cycle x planting density interactions for height and position of the lowermost lateral branch, and this interaction was due primarily to the larger changes occurring at the lower planting density. For example, the position of the lower lateral branch decreased two nodes from C0 to C24 at the lower planting density, but, no change was evident at the higher density (Table 2).
Tillers
Number of tillers per plant increased by 0.08 tillers cycle-1 (9% cycle-1) from 0.86 tillers plant-1 at C0 to 2.86 tillers plant-1 at C24 (Table 2). The length of the lowermost tiller had a quadratic response with an initial decrease from 161.6 cm at C0 to 119.5 cm at C12, followed by an increase to 155.9 cm at C24 (Table 1). Tiller length at C0 and C24 did not differ significantly (Table 2), and the number of nodes on the lowermost tiller did not change throughout the 24 cycles of selection. The internode length of tillers, therefore, followed the pattern established by total length, i.e., a quadratic response with an initial decrease from C0 to C12, followed by a significant increase from C12 to C24, resulting in no difference between C0 and C24.
Similar to total tiller length and tiller internode length, the frequency of female inflorescence terminating the lowermost tiller had a quadratic response (Table 1). Initially, there was a higher frequency of male inflorescences, changing gradually to a more female type by C12 then increasing back to a more male type by C24. There was no significant difference between C0 and C24 (Table 2).
Tassels
There was an overall reduction in size of the tassel from C0 to C24. For example, at the lower planting density, tassel length decreased at a linear rate of -0.18 cm cycle-1, and the number of branches on the male inflorescence also decreased linearly by -0.19 branches cycle-1. The trends were similar at the higher planting density (Table 2).
Ear Morphology
Individual ear size and weight were reduced over the 24 cycles of selection, but the reductions were compensated by the increase in number of ears per plant. At the lower density, the number of kernels per ear decreased by -4.6 kernels cycle-1. The number of kernels per plant at C24, however, was more than twice that at C0. Similar results were obtained at the higher planting density (Table 2).
Individual kernels became smaller from C0 to C24. Kernel dry weight per ear decreased at a linear rate of -1.44 g cycle-1 at the lower planting density. The overall kernel weight of the plant, however, had a positive linear response of 5.56 g cycle-1. Once again, as the number of ears per plant increased, the weight of individual ears decreased, but the total weight of kernels per plant increased significantly. The patterns of response at the two planting densities were similar (Table 2).
There were linear decreases in number of kernel rows (-0.11 kernel rows cycle-1), ear length (-0.10 cm cycle-1), ear width (-0.03 cm cycle-1) and kernel depth (-0.01 cm cycle-1) at the lower planting density. Responses at the higher planting density were more erratic, generally in the same direction but of lesser magnitude. There were no significant differences between C0 and C24 for ear length and kernel depth (Table 2).
Relationships among Traits
The potential number of nodes in maize is determined very early in plant development. By the time of tassel elongation, the numbers of nodes, leaves on the main stalk, tillers, ear shoots, and root primordia have usually been determined (Siemer et al., 1969; Kiesselbach, 1999). The internodes of the main stalk begin to elongate after tassel differentiation (and after axillary meristem differentiation) through the formation of new cells by an intercalary meristem located near the base of each internode above the first (Kiesselbach, 1999). It would be expected that traits related to differentiation of axillary buds and node number would be under different morphogenetic control than those related to internode elongation. To an extent, this pattern is reflected in the relationships among traits in GG(MP). Vegetative traits reflecting bud initiation and node number tended to be associated with each other (Table 3). For example, number of nodes on the main stalk, number of nodes on the uppermost lateral branch, position of the uppermost and lowermost lateral branches, number of tillers, and number of tassel branches tended to be significantly correlated with each other as well as with the number of ears per plant. In some cases, however, the correlations were negative, suggesting a compensatory reduction in activity in one tissue to support another tissue. For example, at the lower planting density, the correlation between number of tassel branches with number of nodes on the main stalk was significant (r = -0.90, P 
0.05), as was the correlation between number of nodes on the uppermost lateral branch with number of nodes on the main stalk (r = -0.84, P 0.05).
|
The selection protocol, however, was confounded with the unintentional selection for earlier flowering in the first 12 cycles, and selection for earlier flowering may have included genes that influenced the rapidity and extent of internode elongation. Maita and Coors (1996) reported that the change in planting density at C12 (from about 17 000–57 000 plants ha-1) had a correlated effect on maturity and plant height. From C0 to C12, the expression of ear shoots was favored by the lower planting density used in the selection plot. One to two passes through the isolated selection plot were usually adequate to identify the desired number of prolific plants prior to significant pollen shed. After C12, the increase in planting density delayed the development of ear shoots and made identification of plants with multiple ear shoots more difficult. More passes through the selection plot were necessary to identify the desired number of prolific plants, and the selected prolific plants spanned a greater range in flowering dates. The selection methodology, therefore, indirectly favored earlier flowering, shorter genotypes prior to C12. After C12, selection for flowering date was relaxed. Perhaps, traits that reflect growth rates would change more as a result of unintentional selection for maturity rather than from pleiotropic effects of prolificacy. Plant height, and tiller and internode length may also fit in this category.
The number of tillers per plant was the only tiller trait that was correlated with any other trait on the main stalk or the lateral branches, including reproductive characteristics. This was not expected because tillers are considered basal reiterations of the main stalk of the maize plant (Moulia et al., 1999). The essential morphology of tillers and main stalks should closely parallel each other. Lateral branches, however, are distinct from the main stalk, not only in the sex of the terminal inflorescence but also in several physiological traits. Moulia et al. (1999) found the rates of leaf development of lateral branches were significantly (58%) greater than for either main stalks or tillers. We studied traits associated across cycles of selection, and selection was for increased activity of axillary buds at nodes on the upper portion of the main stalk. Associations measured in our study, therefore, represent morphogenetic correlations due to selection for upper meristem activation rather than levels of coordination in physiological pathways within a single genotype.
The frequency of female inflorescences terminating the lowermost tiller was correlated strongly with length of the lowermost tiller (r = -0.93, P 0.01, data not provided). The longer the tiller, the more masculine the terminating inflorescence, which may be related to a sexual threshold zone, because sexual differentiation is determined by hormonal regulation (Iltis, 2000).
Prolificacy has been associated positively with grain yield in maize. In this study, as the number of ears per plant increased, an overall decrease in ear size was observed. The number of kernels per ear and dry kernel weight per ear decreased with selection for prolificacy, but number of kernels per ear (r = -0.82, P 0.05) and dry kernel weight per ear (r = -0.93, P 0.01) were correlated negatively with ears per plant at the lower planting density (Table 3). An overall increase in grain yield was associated with the increase in ears per plant, which, under the lower planting density, was represented by highly significant correlations between ears per plant and kernel number per plant (r = 0.98, P 0.01) and between ears per plant and kernel dry weight per plant (r = 0.95, P 0.01).
Selection for prolificacy in the GG(MP) increased grain production per plant at both planting densities even though the kernel size, weight, and number per ear were reduced. It is not evident whether the morphological changes induced by this type of selection represent a more efficient plant type for grain production, although many studies have shown that prolificacy per se is correlated with grain yield in maize. The morphology of the GG(MP) population at C24 is not typical of the traditional ideotype of modern maize.
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
As the number of ear per plant increased significantly over cycles of selection in the GG(MP), a general activation of meristematic tissue was observed throughout the plant. This activation was reflected on the complete development of more ears and tiller buds per plant and the elongation of the uppermost lateral branches due to the elongation of internodes. In maize, the number of active nodes in the main stalk, tillers, and roots of the plant are determined very early in development, leaving the elongation of the different plant parts for later stages when new cells are formed by the intercalary meristem (Kiesselbach, 1999).
The domestication of maize is an intriguing phenomenon. Even though there are great differences in morphology between modern maize and its ancestor teosinte, it has been suggested that a relatively small number of loci with large effects were involved in the early evolution of the key traits that distinguish them (Doebley et al., 1995). Some of the differences between maize and teosinte may involve the regulation of axillary meristematic growth, and Doebley and Wang (1997) suggested that this regulation in maize produces the arrest in growth of some organs or particular plant parts, such as the arrested growth of internode elongation and some axillary buds. The changes accompanying selection for prolificacy in the GG(MP) may be due to similar genetic phenomena. Repression and activation at different stages of development may be related to a single locus teosinte banched1 (tb1). Doebley and Stec (1994) and Doebley et al. (1995) demonstrated that tb1 had a pivotal role on the divergence of the current maize from its ancestral form. The mutant tb1 is a recessive mutant of maize that affects plant architecture and maps to chromosome arm 1L (Burnham, 1959). Plants homozygous for this allele have long lateral branches tipped by ears at some upper nodes of the main stalk and tillers at the basal nodes (Doebley et al., 1995). The allele tb1 arose as a spontaneous mutant in a maize population that is highly influenced by the genetic background and varies in its response to different environments (Doebley et al., 1995). Thus, it may be possible that this same common factor is acting slightly different in distinct backgrounds, such as GG(MP) and the F2 produced between teosinte and maize used by Doebley and colleagues (Doebley and Stec, 1991, 1993).
As selection for prolificacy proceeded in the GG(MP), the number of nodes increased and length of the internodes decreased on the main stalk, whereas the number of nodes decreased and the internode length increased in the uppermost lateral branch. The overall increased activation of nodes on the main stalk and the elongation of the uppermost lateral branches due to the elongation of internodes rather than the increase in node number observed in the GG(MP) are similar to the pattern of development of tb1 in the primary lateral branches of the F2 studied by Doebley (Doebley and Stec, 1991 and 1993). The male and female inflorescences became smaller but, number of kernels and weight per plant increased significantly across cycles, as has been reported in the cross between maize and teosinte (Doebley and Stec, 1991 and 1993).
The average length of internodes in the primary lateral branch and the percentage of female spikelets in the primary lateral inflorescence are strongly associated with one another and they probably represent different adult manifestations of a common developmental program (Doebley and Stec, 1993). In the GG(MP), a high phenotypic correlation (r = -0.93, P 0.01, data not provided) was also observed between the total length and the terminal inflorescence type of the lowermost tiller, a key feature in the architectural divergence observed between maize and teosinte. During crop plant domestication, an increase in apical dominance (i.e., an increase in the concentration of resources on the main stalk and the subsequent suppression of lateral branches) has been observed in maize (Doebley et al., 1997). Our data suggest a common factor could be acting at different times in development producing different phenotypic effects in the different plant parts analyzed over cycles of selection in the Golden Glow maize population.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Received for publication March 12, 2001.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. de Leon, J. G. Coors, S. M. Kaeppler, and G. J. M. Rosa Genetic Control of Prolificacy and Related Traits in the Golden Glow Maize Population: I. Phenotypic Evaluation Crop Sci., May 27, 2005; 45(4): 1361 - 1369. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. de Leon, J. G. Coors, and S. M. Kaeppler Genetic Control of Prolificacy and Related Traits in the Golden Glow Maize Population: II. Genotypic Analysis Crop Sci., May 27, 2005; 45(4): 1370 - 1378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. V. Butruille, H. D. Silva, S. M. Kaeppler, and J. G. Coors Response to Selection and Genetic Drift in Three Populations Derived from the Golden Glow Maize Population Crop Sci., September 1, 2004; 44(5): 1527 - 1534. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Plant Registrations | Soil Science Society of America Journal | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
