| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Crop and Soil Sci. Dep., Georgia Station, Univ. of Georgia, Griffin, GA 30223-1797
* Corresponding author (rcarrow{at}gaes.griffin.peachnet.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
14% compared with LF 
29% for the commercial/RGP-1 pool. Persistence (percentage turf coverage) after 24 mo of repeated drought stress cycles revealed an average of 91 and 76% for the GA and commercial/RGP-1 pools, respectively, with Southeast and experimental 96-6 exhibiting the highest canopy coverage at 93%. Grasses demonstrating the best combined ranking of LF 14%; least change of turf quality (TQ) under repeated drought stress events with 
TQ < 1.4; highest average TQ at TQ 6.4; and within the lowest evapotranspiration (ET) group (ET 2.72 mm d-1) were all from the GA gene pool (Southeast, 96-5, 96-6, 96-7). This research provided strong documentation that significant enhancement of drought resistance and tall fescue persistence can be achieved by use of the GA breeding/selection protocol.
Abbreviations: TC, change in turf color • TQ, change in average TQ from lowest to highest rating • AS, Al saturation • ET, evapotranspiration • GA-5, ‘Georgia 5’ • LF, leaf firing • RGP-1, Rutgers germplasm pool • RLD, root length density • TQ, turf quality • TRL, total root length
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Intraspecific differences in drought avoidance and tolerance have been reported for tall fescues in studies involving two to six cultivars (White et al., 1992, 1993; Carrow 1996a,b; Huang and Fry, 1998; Huang et al., 1998; Huang and Gao, 1999, 2000). Two forage-types [‘Kentucky 31’, ‘Georgia 5’ (GA-5)], two dwarf-types (‘Bonsai’, ‘MIC18’), and seven nondwarf turf-types (‘Arid’, ‘Falcon II’, ‘Houndog V’, ‘Mustang’, ‘Phoenix’, ‘Rebel II’, ‘Rebel Jr.’) were included in these studies.
Compared with Kentucky 31, the dwarf-type tall fescues exhibited lower drought resistance (White et al., 1993; Carrow, 1996a; Huang and Fry, 1998; Huang et al., 1998; Huang and Gao, 1999, 2000). Drought resistance or characteristics [i.e., low leaf firing, greater rooting depth, higher root length density (RLD)] related to improved drought resistance were reported to be greater for Rebel II and Arid when compared with Kentucky 31 (Carrow, 1996a, b; White et al., 1993). Turf-type cultivars demonstrating lower drought resistance or characteristics associated with low drought resistance were Houndog, Phoenix, and Rebel Jr. (Huang and Gao, 1999, 2000).
These studies illustrated intraspecific differences in drought resistance within tall fescues. However, the number evaluated is small compared with the number of commercial cultivars available, currently >100.
Improved drought resistance is important for turfgrass persistence, especially under limited irrigation or rainfall, high ET conditions, and for water conservation. Duncan and Carrow (1997)(1999) and Carrow and Duncan (1996) presented a turfgrass breeding and selection protocol to enhance overall drought resistance by emphasizing (i) improved genetic-based tolerance to soil physical and chemical stresses that limit root growth and/or viability; and (ii) greater indirect high temperature tolerance for maintenance of carbohydrate allocation to roots when under climatic high temperature stress (Huang and Gao, 1999) or drought-induced high temperature stress. This unique approach had been successfully used in sorghum [Sorghum bicolor (L.) Moench] for improvement of drought resistance (Maranville et al., 1993; Duncan, 1994) but had not been used for turfgrasses until by Duncan and Carrow (1997).
The objectives of this research were (i) to evaluate the effectiveness of the breeding and selection protocol for enhancing drought resistance/persistence reported by Duncan and Carrow (1997)(1999) and Carrow and Duncan (1996) by comparing performance of six experimentals developed under this protocol, five commercial tall fescues varieties, and a Rutgers germplasm subpool population, and (ii) to identify the cultivars or ecotypes with the greatest degree of drought resistance and persistence in the southern U.S. transition region.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The specific breeding approach for tall fescue development includes three components (Duncan and Carrow, 1997). The first component was to identify and incorporate tall fescue germplasm prescreened under multiple abiotic stress situations. The second component involved crossing stress tolerant survivors followed by further exposure to varying degrees of multiple abiotic and edaphic stresses in concert with mower scalping stress (carbohydrate depletion), the third component of the grass development program.
For the experimental population in this study, germplasm was combined from three gene subpools. Each subpool included only survivors and their polycrosses from sequentially imposed severe stress conditions. The survivors had to cross pollinate and produce seed under stress; consequently, only tolerant x tolerant plants were polycrossing. Polycrossed progeny survivors were then subjected to additional exposure to multiple abiotic stresses (described later) and severe scalping pressure during high temperatures (>32°C) in the summer months. Plant mortality (i.e., selection index) was 95 to 99% in the initial screens. The 1 to 5% survivors from each original gene pool became the subpool germplasm for the establishment of several populations.
One gene subpool came from the GA heat tolerant types collected within 25 to 30 cm of the asphalt along highways in the Piedmont region of middle to northern Georgia. These selections generally did not posses high quality turf traits (high shoot density, dark green color). However, they were assumed to have indirect high temperature and soil compaction tolerances that would favor a positive carbohydrate balance for maintenance of roots and shoots under summer stresses due to the asphalt heating the adjacent soil and traffic causing compaction on the shoulder of the highway. Without the ability to produce adequate carbohydrates for root maintenance during prolonged summer months, tall fescue could not persist under hot, humid climatic conditions even if it had high genetic tolerance to southeastern U.S. edaphic stresses (Jiang and Huang, 2001b). The GA heat tolerant subpool consisted of survivors and their polycrossed progeny after screening at pH 4.3 [50% Al saturation (AS) of the soil cation exchange capacity sites] for three cycles with scalping. Low pH was created in plots previously used for sorghum screening by S addition with the low pH condition used to enhance drought stress, while simultaneously imposing Al toxicity stress on the root tissues. Excess Al prunes roots. A screening cycle was considered to be 1 yr under field conditions. This gene subpool was considered the foundational germplasm and was reselected for TQ traits in Georgia as well as seed production traits in Oregon.
A second GA-developed gene subpool was formulated by wide scale collection of many tall fescue plants primarily from the Piedmont region but stretching into the coastal Florida region. Emphasis was placed on old, mowed tall fescue pasture swards that were subjected to harsh environments. The subpool was composed of survivors and their polycrosses from populations screened initially at either pH 4.3 (three cycles, 50% AS) and then at pH 4.0 (one or two cycles, 75% AS). This gene subpool contained survivors with forage to turf-type characteristics. In general, the first cycle resulted in 
50% attrition and 75% survivability by the third cycle. The pH 4.0 screen resulted in another 50% attrition, but 1% of superior stress-tolerant clones with the best turf traits were selected for population development.
A third gene subpool was developed from survivors and their polycrosses of original Rutgers lines after screening at pH 4.6 for 1 yr and then with further screening at pH 4.3 (two cycles), and then at pH 4.0 (two cycles). The high attrition within the initial year (>90%) followed by further loss resulted in <1% surviving germplasm of the original lines. These survivors included turf types. Creation of the RGP-1 was part of a cooperative germplasm exchange effort by Dr. Reed Funk of Rutgers University and University of Georgia to enhance germplasm characteristics of both programs. To date, this subpool has not maintained more than 50% survivability regardless of the number of breeding cycles in the breeding plots.
A key condition of the breeding program required survivors from the two GA subpools to produce seed under a final pH 3.6 stress condition without irrigation (i.e., to complete the life cycle under stress). All stress screening was conducted in the field on nonexpanding kaolinitic clays (Typic Kanhapludult) with high soil strength and no supplemental irrigation after establishment. The three levels of acid soil stress were achieved by pH adjustment with S to pH 4.3 (50% AS), 4.0 (75% AS) or 3.6 (90% AS). Screening under these conditions incorporates multiple abiotic stresses that limit root development or persistence; namely, high soil strength, soil drought, indirect high temperature stress, and the acid soil complex (Al toxicity, potential nutrient deficiencies) (Duncan and Carrow, 1997, 1999).
The third component of the breeding/selection protocol was imposition of scalping on the unmowed grasses using a 1.90- to 2.5-cm stress mowing height several times in the June to August heat stress period to enhance carbohydrate depletion. Thus, survivors that were able to produce sufficient carbohydrates to maintain their root systems were considered to have a high degree of indirect high temperature stress tolerance. Scalping was practiced in all original and subpool survivors/crosses with the exception being one experimental (96-5) discussed below.
Grasses
Twelve tall fescue ecotypes were evaluated under turfgrass management conditions. These included five commercial types: (i) Rebel Jr., a dark green semidwarf turf-type with intermediate growth rate and similar drought resistance to Kentucky 31 and GA-5 under Georgia Piedmont region conditions (Carrow, 1996a), (ii) Kentucky 31, a forage type, (iii) ‘Jesup’ (Bouton et al., 1997), a forage type developed in Georgia with better stand survival than Kentucky 31 under high temperatures, (iv) ‘Bravo’, a dark green turf-type screened in Georgia at pH 4.3 for one cycle for better drought resistance (Lesco, Inc., Rocky River, OH), and (v) ‘Stetson’, a dark green turf-type screened in Georgia for improved drought resistance at pH 6.0 for one cycle (Lesco, Inc.). No polycrossing occurred within Bravo or Stetson germplasm under the stress conditions in Georgia.
The seven experimental populations were (i) RGP-1, which consisted 100% of Rutgers material; (ii) Southeast (evaluated under the code 96-2), which included >90% surviving germplasm from the two Georgia subpools with emphasis on multiple tiller ecotypes from these pools and the remainder from the RGP-1 subpool. The types selected from these subpool populations were polycrossed in the field at pH 4.5 (25% AS) with scalping two to three times at 5 cm following seed harvest for at least one cycle. Survivors (no more than 10% from RGP-1) were then planted in rows for a second year at pH 4.5 and seed harvested. Seed production and rust (Puccinia spp.) resistance traits were reselected in Oregon by field removal of weak plants or infected plants; (iii) experimental populations 96-3, 96-4, and 96-7 also contained combinations of ecotypes primarily out of the Georgia-based subpools; (iv) 96-5 was comprised only of material from the GA heat tolerant subpool; and (v) 96-6 contained selections only out of the two GA subpools. As with Southeast, the experimentals were screened through an additional two cycles.
Performance Trial
The 12 entries were seeded on 10 Oct. 1996 at 2.9 kg 100 m-1 pure live seed into four replicated plots each 4.6 by 4.6 m in a completely randomized block design. Plots were separated from each other by a 1.5-m grassed buffer strip to remove border effects. Each plot contained four quarter circle, popup spray heads at the corners for uniform irrigation. Plots were mowed twice per week during the spring, early summer, and fall seasons at 5 cm with clippings returned.
The experiment was conducted at Griffin, GA, from 1996 to 1998 on a Cecil sandy loam (fine, kaolinitic, thermic Typic Kanhaplaudults) with 67.4% sand, 18.3% silt, 14.3% clay, bulk density of 1.55 g cm-3, and pH 4.8 within the surface 0- to 30-cm A horizon. The B horizon at 30 to 60 cm was 44.5% sand, 18.0% silt, 37.5% clay, bulk density of 1.61 g cm-3, and pH 4.4. Organic matter content within the surface 10 cm was 0.90%. With kaolinite nonexpanding 1:1 clay as the predominate clay type, this soil does not crack during drying and produces very high soil strength as it dries.
Fertilization consisted of 24 and 4.9 kg N ha-1 as 10-4.4-8.3 (N-P-K) on 10 Oct. and 3 Dec. 1996, respectively. In 1997 and 1998, fertilization was at 4.9 kg N ha-1 on 12 March (10-4.4-8.3), 23 to 30 May (33-0-0), and 20 October (1997 only; 10-4.4-8.3). Preemergence annual grass herbicide was applied each spring as needed. Since brown patch (Rhizoctonia solani Kühn) injury could interfere with LF ratings as a measure of drought resistance, a preventative fungicide was applied in mid-June 1997, early July, and mid-July 1998 to minimize this possible variable.
Irrigation and Water Use
Time domain reflectometry probes were installed in the center of each plot to determine average soil water content across the depths of 1 to 10, 10 to 20, and 20 to 60 cm (maximum rooting depth observed at the site). Before dry-down periods, all plots received sufficient water by irrigation or rainfall to bring the 0- to 60-cm depth to field capacity. Measurements were taken throughout the dry-down periods. Nonadapted turf types normally begin exhibiting drought stress symptoms 4 to 5 d into a dry-down cycle. Soil water content was determined at field capacity and at the end of each dry-down period to determine spatial water extraction and ET (sum of water extraction across all depths since drainage was zero) according to the soil-water balance method (Sharma, 1985) as described by Carrow (1996a) for turfgrass situations. There were three dry-down cycles in 1997 and two in 1998.
Root Measurements
Root samples were obtained 11 July and 10 Sept. 1997, and 28 June and 15 Sept. 1998 for each grass at the 2.5- to 30-cm and 30- to 60-cm depths. The surface 0 to 2.5 cm was discarded. Three cores of 6.3-cm diam. were taken per plot on each sample date. After washing free of soil, the roots were subsampled and root length determined by an image analyzing system (Harris and Campbell, 1989). Root weights were taken as a subsample and a complete sample.
Shoot Measurements
Visual quality ratings were obtained using the scale of 9.0 = ideal shoot density, color, and uniformity for the particular species; 6.5 = minimum acceptable quality for good turf; and 1.0 = no live turf. Shoot color ratings were on the same days as visual quality and were based on the scale of 9.0 = dark green color, 1.0 = no green color. Shoot density ratings were based on 9.0 = ideal shoot density and 1.0 = no live turfgrass. Turfgrass coverage ratings were based on visual estimation of the percentage turfgrass cover. Leaf firing refers to leaf chlorosis starting at leaf tips and margins, and progressing down the leaf in response to progressive drought stress. Initial injury is a yellowing but often progresses into a tan/brown color with death of the tan/brown areas. Leaf firing under field conditions is an overall measure of drought resistance since all climatic–soil–plant factors influencing drought stress are active. While it measures when a plant undergoes drought stress, it does not indicate what aspects of drought avoidance or tolerance are involved (Carrow, 1996a, b). The LF ratings were based on percentage of the leaves exhibiting the above symptoms.
The experiment consisted of 12 tall fescues arranged in a randomized complete block design with four blocks. Data were subjected to analysis of variance according to the GLM procedure Statistical Analysis Systems (SAS Institute, Cary, NC). Since year-to-date interactions occurred, data are presented by each year. LSD0.05 values are presented for mean comparisons.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Leaf firing averaged 16% (range 11 to 23%) during 1997 and 1998 for the combined Georgia experimental populations (i.e., 96-2 to 96-7) (Table 1). Each of these were screened through the multiple abiotic stress approach used for enhancement of drought resistance (Duncan and Carrow, 1997, 1999). All other grasses had a combined averaged of 32% LF (range was 29 to 35%). The contrast between the GA experimental population and the commercial/RGP-1 population was significant at F test (P = 0.001). Grasses exhibiting the least LF averaged across the seven measurement dates were Southeast (11%), 96-5 and 96-7 (13%), 96-6 (14%), and 96-4 (20%). Each of these grasses ranked in the lowest (best) LF statistical group for all dry-down periods (Table 1) based on LSD0.05.
|
Shoot Responses
On the basis of the four dry-down periods when turfgrass quality ratings were made at the beginning and end of a period, the GA experimentals as a group averaged a 0.63 reduction in TQ (TQ) compared with the dry-downs with a range of 0.48 (Southeast) to 0.90 (96-3) (Table 2), with the contrast significantly different at F test P = 0.001. The average decrease in TQ for RGP-1 and the commercial cultivars was 1.08 with a range of 0.90 (Kentucky 31) to 1.33 (Bravo).
|
TQ or range in TQ from highest to lowest rating) (Table 3). The remaining grasses exhibited a 2.95 TQ range. Grasses demonstrating the least change in TQ (i.e., greatest stability) throughout the growing season were 96-7 (TQ of 1.1); and 96-5 and Southeast (TQ of 1.2). Stetson had the widest TQ at 3.7.
|
Under the repeated imposition of drought stress in this study, turfgrass color ratings varied for each grass, particularly under drought stress, due to LF causing chlorosis and/or leaf desiccation (Table 3). The GA experimentals had an average change in turf color (TC) of 1.2 (range was 1.1 to 1.3), while all other grasses averaged a TC of 1.6 (range of 1.1 to 2.0).
Highest seasonal average turf color occurred for RGP-1, Rebel Jr., Bravo, and 96-3 (Table 3). These grasses also demonstrated turf color ratings within the top statistical group > 86% of ratings. Kentucky 31 and Jesup had the lowest average turf color and both had 6% or less ratings within the top color group.
Shoot density, along with color, substantially influenced overall turfgrass quality. Grasses with shoot density ratings within the top statistical group most often were Bravo (93%) and RGP-1, Southeast, and 96-5 (all at 86%; Table 3).
Turfgrass Cover
In October 1997, at 12 mo after establishment, percentage turf cover averaged across the Georgia experimentals was 89% with a range from 83% (96-7) to 95% (Southeast; Table 3) while commercial cultivars and RGB-1 averaged 77% turf cover with a range of 65% (Jesup) to 91% (RGB-1; F test of contrast significant at P = 0.05). At 24 mo after establishment and two summers of repeated drought cycles, GA experimentals averaged 91% turf cover with the least cover for 96-3 (84%) and the highest at 93% for Southeast and 96-6. All other tall fescues exhibited a combined average of 76% with Kentucky 31 least at 61% and Stetson highest at 88%.
Root Responses
On 11 July 1997, no differences in RLD x depth or TRL (total root length) were observed among the grasses (Table 4). However, by the end of summer, significant differences were apparent for RLD x depth and for TRL. Tall fescues exhibiting the highest RLD at the 30- to 60-cm depth were 96-7, 96-4, Jesup, and Southeast. Grasses with the highest TRL in September 1997 were all of the Georgia experimentals, Stetson, and Jesup.
|
Water Relations
Evapotranspiration rates when averaged across four dry-down periods in 1997 and 1998 ranged from 2.18 mm d-1 (Stetson) to 3.78 mm d-1 (RGP-1) (Table 5). RGP-1 and Rebel Jr. exhibited the highest average ET rates while all other tall fescues were in the lowest ET ranking with ET values from 2.18 mm d-1 to 2.72 mm d-1. Grasses extracting the highest quantity of water from the 20- to 60-cm zone were RGP-1 and Rebel Jr.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reduced LF during a drought stress period and percentage turf coverage at the end of each summer provided reasonable comparisons of drought resistance as demonstrated by Carrow (1996a)(b). On the basis of the LF criteria, Georgia experimentals as a combined pool exhibited significantly (P < 0.05) better drought resistance than the commercial/RGP-1 germplasm pool with average LF of 16% (GA experimentals) vs. 32% (Table 1). All five of the grasses with the lowest average LF were in the GA germplasm pool with four demonstrating average LF of 14% or less. The lowest average LF for the commercial/RGP-1 pool was 29% for RGB-1, which had been screened under some cycles of the GA protocol.
Enhanced drought resistance exhibited by the entire GA pool average and especially by several of the individual grasses presented strong evidence for the effectiveness of the breeding and selection approach used in this study. Important concepts used for improving drought resistance were exposure of the germplasm incorporated into the grasses to multiple root-limiting stresses and imposing a high level of multiple stresses so that the genetic-based resistance mechanisms would be expressed. Several breeders and stress physiologists have proposed this approach as a means of achieving more rapid progress in improving resistance to drought and other soil-based stresses (Baligar and Duncan, 1990; Maranville et al., 1993; Duncan, 1994), but this approach had not been previously used for turf improvement.
From a turfgrass management standpoint, delayed onset of LF and the accompanying decline in TQ would increase the opportunity to use natural precipitation rather than irrigation. Additionally, repeated drought periods would be expected to cause less injury and loss of turf coverage for more drought-resistant germplasm.
Greater persistence was demonstrated by higher average turfgrass coverage at the end of each summer stress period at 89% vs. 77% (12 mo) and 91% vs. 76% (24 mo) for the GA pool vs. commercial/RGP-1 pool (Table 3). Contrast F test was significant at P = 0.05 for both years. All of the GA germplasm ranked in the top statistical group for cover at the end of 24 mo (10 Sept. 1998), while two commercial/RGP-1 pool grasses were in the top grouping. Greater persistence would reduce the need for overseeding and should aid in limiting weed encroachment and helping to minimize soil evaporation.
Further evidence of enhanced drought resistance is provided by significantly less decline in TQ by the GA germplasm when subjected to repeated drought stress cycles. Average TQ was 1.48 for the GA germplasm compared with TQ of 2.95 for the commercial/RGP-1 pool (Table 3). On the basis of consistently low LF, greater persistence, and the maintenance of TQ under repeated drought stress, experimental 96-2 (Southeast) was released as a tall fescue with superior drought resistance by the University of Georgia in 1999 for commercial development (Tables 1, 2, 3).
Improved root development and maintenance of roots are important drought-avoidance aspects (Beard, 1989). On the basis of TRL at the end of summer 1997, all GA pool germplasm were in the top statistical group along with Jesup and Stetson (Table 4). In terms of deep rooting, three GA ecotypes (96-7, 96-4, Southeast) and Jesup ranked in the top grouping. By late summer 1998, three GA ecotypes (96-4, 96-3, Southeast), Bravo, and Stetson exhibited the highest TRL (Table 4), with no significant difference among grasses for deep rooting. GA pool grasses demonstrated on average ranking within the top statistical grouping on dates when significant root responses were noted at 3.2 out of 5.0 times compared with 1.5 out of 5.0 for the commercial/RGP-1 pool (Table 4). Thus, improved root development and maintenance were noted for the GA germplasm and would be expected to contribute to their enhanced drought resistance.
Reduced inherent (genetic-based) ET also contributes to drought avoidance (Beard, 1989). However, care must be taken to compare ET on a pool or individual grass basis. Genetic-based factors that may limit ET and conserve water for later use should limit ET but still maintain physiological function. In contrast, grasses that rapidly express LF, thereby expressing loss of physiological function, during a dry-down can also show low ET. Thus, grasses with low ET plus the ability to maintain TQ are those with the appropriate set of characteristics.
Turfgrasses ranking in the best group for average ET (low), TQ of <1.4, average TQ (6.4), and LF (14%) were Southeast, 96-5, 96-6, and 96-7 (Tables 1, 3, 5). Interestingly, the two grasses (Rebel Jr., RGP-1) extracting the most water from the 20- to 60-cm zone also had the highest average ET (Table 5). However, average LF was 32 and 29%, respectively, for Rebel Jr. and RGP-1 while TQ was 2.6 and 2.1, respectively. (Table 1, 3). This suggested that some grasses can readily extract soil moisture deeper in the root zone but do not possess genetic-based shoot features for controlling excessively high ET (Beard, 1989).
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 10, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. E. Karcher, M. D. Richardson, K. Hignight, and D. Rush Drought Tolerance of Tall Fescue Populations Selected for High Root/Shoot Ratios and Summer Survival Crop Sci., March 19, 2008; 48(2): 771 - 777. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Jiang and R. N. Carrow Broadband Spectral Reflectance Models of Turfgrass Species and Cultivars to Drought Stress Crop Sci., July 30, 2007; 47(4): 1611 - 1618. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Zhang and E. H. Ervin Cytokinin-Containing Seaweed and Humic Acid Extracts Associated with Creeping Bentgrass Leaf Cytokinins and Drought Resistance Crop Sci., September 1, 2004; 44(5): 1737 - 1745. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
