An Investigation of Factors Affecting the Rooting Ability of Hardwood Muscadine Cuttings

in HortScience
Authors:
Kenneth BuckDepartment of Horticulture, University of Arkansas System Division of Agriculture, Fayetteville, AR 72701

Search for other papers by Kenneth Buck in
ASHS
Google Scholar
PubMed
Close
,
Margaret WorthingtonDepartment of Horticulture, University of Arkansas System Division of Agriculture, Fayetteville, AR 72701

Search for other papers by Margaret Worthington in
ASHS
Google Scholar
PubMed
Close
, and
Patrick J. ConnerDepartment of Horticulture, University of Georgia, 2360 Rainwater Road, Horticulture, Tifton, GA 31793

Search for other papers by Patrick J. Conner in
ASHS
Google Scholar
PubMed
Close

Rooting hardwood cuttings from muscadine (Vitis rotundifolia Michx. syn. Muscadinia rotundifolia) vines has traditionally been considered an exceptionally difficult task. Many previous studies observed almost no root formation, leading to a general consensus that muscadines should either be propagated by softwood cuttings or vegetative layering. However, the University of Arkansas System Division of Agriculture Fruit Breeding Program has been using a hardwood rooting protocol for muscadines with moderate success for the past 10 years. The application of this protocol to meet the modest propagation needs of the breeding program has significantly shortened the time required to advance selections. The goal of this research was to more adequately describe the factors affecting the rooting ability of hardwood muscadine cuttings. This research investigated the effects of cultivar, bottom heat, cold storage, vineyard location, and cutting collection date on the outcome of muscadine hardwood cuttings. The study was conducted during the dormant seasons of 2019–20 and 2020–21, and an overall rooting percentage of 16% was observed. There were multiple higher-order interactions affecting rooting efficacy. Cuttings taken in November generally rooted at higher rates, although interactions with vineyard location and cultivar played a significant role in those results. The Ocilla, GA, location performed exceptionally well in November with rooting percentages greater than 40%. The effects of supplying bottom heat and/or a cold storage treatment on rooting success declined as the dormant season progressed. Other variables such as increased cutting length and diameter were associated with increased rooting success. A second statistical analysis using only data from November showed that when cuttings were not given a cold storage treatment that rooting percentages were greater than 27%. Ultimately, this research shows that institutions with modest muscadine propagation needs can successfully propagate plants from hardwood cuttings.

Abstract

Rooting hardwood cuttings from muscadine (Vitis rotundifolia Michx. syn. Muscadinia rotundifolia) vines has traditionally been considered an exceptionally difficult task. Many previous studies observed almost no root formation, leading to a general consensus that muscadines should either be propagated by softwood cuttings or vegetative layering. However, the University of Arkansas System Division of Agriculture Fruit Breeding Program has been using a hardwood rooting protocol for muscadines with moderate success for the past 10 years. The application of this protocol to meet the modest propagation needs of the breeding program has significantly shortened the time required to advance selections. The goal of this research was to more adequately describe the factors affecting the rooting ability of hardwood muscadine cuttings. This research investigated the effects of cultivar, bottom heat, cold storage, vineyard location, and cutting collection date on the outcome of muscadine hardwood cuttings. The study was conducted during the dormant seasons of 2019–20 and 2020–21, and an overall rooting percentage of 16% was observed. There were multiple higher-order interactions affecting rooting efficacy. Cuttings taken in November generally rooted at higher rates, although interactions with vineyard location and cultivar played a significant role in those results. The Ocilla, GA, location performed exceptionally well in November with rooting percentages greater than 40%. The effects of supplying bottom heat and/or a cold storage treatment on rooting success declined as the dormant season progressed. Other variables such as increased cutting length and diameter were associated with increased rooting success. A second statistical analysis using only data from November showed that when cuttings were not given a cold storage treatment that rooting percentages were greater than 27%. Ultimately, this research shows that institutions with modest muscadine propagation needs can successfully propagate plants from hardwood cuttings.

The muscadine (Vitis rotundifolia Michx. syn. Muscadinia rotundifolia) is a perennial woody liana native to the southeastern United States that has been cultivated by European settlers for centuries (Basiouny and Himelrick, 2001; Munson, 1909). Although a member of the grape family Vitaceae, there are significant morphological and genetic differences that make this species distinct from the more widely cultivated species such as Vitis vinifera L. or Vitis labrusca L., including differences in chromosome number and leaf shape (Comeaux et al., 1987; Liu et al., 2016). This has led taxonomists to divide Vitis further into two subgenera: Euvitis, or bunch grapes, and Muscadinia, which includes muscadines and two tropical grape species with similar morphologies (Basiouny and Himelrick, 2001).

One of the distinctions between muscadines and other grape species is the recommended method of vegetative propagation. Although layering was a standard method of propagation in the 19th century and earlier for nurseries working with perennial plant species, the development of novel protocols for propagation by hardwood and softwood cuttings began to be studied extensively in the 20th century (Hartmann et al., 1997). Hardwood and softwood cuttings are both taken from the current season’s growth. Softwood cuttings are taken from actively growing tissues in the early summer, whereas hardwood cuttings are taken when periderm has formed and shoot growth has ceased during the dormant season. Concurrent research conducted on propagation by hardwood cuttings in both V. vinifera and V. rotundifolia revealed highly effective methods for V. vinifera and persistent inconsistencies in the rooting ability of V. rotundifolia (Cowart and Savage, 1944; Doelle and Mitchell, 1964; Goode et al., 1982). Hardwood cuttings soon became the main propagation method for bunch grapes. Meanwhile, other research revealed that using mist systems in a greenhouse setting were an effective way to propagate V. rotundifolia by softwood cuttings (Sharpe, 1954).

Past research into the hardwood propagation of muscadine cuttings evaluated many factors potentially affecting rooting success. Rooting studies in muscadines have been conducted at vineyards across the southeastern United States, but the variation in rooting success attributable to the study location has not been investigated. Cutting collection date is mentioned extensively in the literature, although it has not actually been tested as a factor in most studies. November is the most frequent month recommended to collect muscadine hardwood cuttings, although success varied from 0% to 80% among the studies using cuttings taken in November depending on the treatment combination used (Cowart and Savage, 1944; Newman, 1907; Niven, 1918; Woodroof, 1935). It has also been reported that muscadines can root readily when taken in December and that taking cuttings later than December could lead to injury to the vine and poor rooting (Goode et al., 1982; Niven, 1918; Whatley, 1974).

Multiple studies have used cold storage treatments as part of their rooting protocols (e.g., Whatley, 1974). However, the only two studies that specifically compared cuttings that received cold storage and cuttings that did not observed no significant effect (Cowart and Savage, 1944; Newman, 1907). Bottom heat also appears to play a significant role in promoting rooting in dormant cuttings. Goode et al. (1982) had 0% rooting success when muscadine cuttings were not given bottom heat compared with a high of 9% rooting when cuttings were taken in November and given bottom heat. Newman (1907) noted that continuously supplied bottom heat was essential to achieving 80% rooting. The application of various synthetic rooting hormones, although not studied extensively, does not appear to have an effect on rooting success in hardwood muscadine cuttings compared with no hormone application (Cowart and Savage, 1944; Goode et al., 1982). Differences in rooting success attributable to genotype are another factor of interest in previous studies. Dearing (1947) noted that some staminate cultivars are not difficult to propagate by hardwood cuttings compared with the historically important pistillate cultivar Scuppernong. Goode et al. (1982) found that ‘Hunt’ outperformed ‘Cowart’ vines taken from the same location. Most studies conducted on hardwood propagation in muscadines are at least several decades old, and most of the cultivars used in these early studies have since been replaced by newer, improved varieties.

The position on the vine from which a cutting is taken may also have a role in promoting the rooting of dormant cuttings in muscadines. Cane position has been shown to have an effect on rooting efficacy in other Vitis species. Cuttings taken from basal cane segments rooted at a higher rate than cuttings taken from apical segments in both hardwood and softwood cuttings of Vitis vinifera and Vitis aestivalis Michx. (Daskalakis et al., 2018; Keeley et al., 2004). To date, a single study has directly investigated the effect of cane position on rooting in muscadines and observed moderately increased rooting success in cuttings taken from basal positions on the cane (Castro et al., 1994). Although not directly comparable to cane position, Goode et al. (1982) saw increased rooting in cuttings with larger diameters.

By the end of the 20th century, significant advances had been made in propagation technology, and the nursery industry was adopting newer, more advanced techniques for propagation. However, despite the labor involved and space required, layering was still a recommended propagation method for muscadines (Hartmann et al., 1997). Nurseries with the ability to maintain the high humidity necessary for softwood cuttings to succeed expect to have rooting success of 90% or more (Basiouny and Himelrick, 2001). The lack of a scientific consensus on the effectiveness of rooting from hardwood cuttings and the relative ease with which muscadine can be propagated by softwood cuttings led to the abandonment of hardwood propagation of muscadines.

Neither of the two widely accepted methods of vegetatively propagating muscadines are ideal for breeding programs. Softwood cuttings are generally taken in June and must be maintained throughout the summer, which adds to the workload of the fruit breeding personnel during an already busy season. Propagation by layering requires a significant amount of field space and labor to be effective (Hartmann et al., 1997), which directly limits the number of seedlings that can be evaluated by the breeding program. The most significant reason that these methods are not ideal, however, is the amount of time required from when propagation is initiated until plants are ready to be transplanted into the field.

The breeding cycle for muscadines is similar to that of other perennial crops: from the time of making the cross to releasing a cultivar generally requires at least 10 to 15 years. A muscadine vine will usually not produce fruit until its third year. If a seedling is determined to be worthy of selection, it will be vegetatively propagated and tested at multiple locations and observed for multiple years (Basiouny and Himelrick, 2001; Fehr, 1991). The time required to release a cultivar is a significant impediment in the field of crop improvement, and significant amounts of time and money have been spent determining new plant breeding methods that could accelerate the plant breeding cycle. Propagation by hardwood cuttings instead of softwood cuttings or layering is one way a muscadine breeding program can shorten its cycle relatively easily.

Muscadines flower from mid-May to mid-June and require ≈100 d to fully ripen (Goldy, 1992). For areas in the northern extent of its native distribution, this can be as late as September or October. Plant breeders therefore taste fruit and make selections well after the recommended time to take softwood cuttings (Basiouny and Himelrick, 2001; Sharpe, 1954). Breeding programs must, therefore, wait until the following June to propagate the selected seedling vine for replicated trials, and these softwood cuttings will not be ready for transplanting into the field until too late in the growing season. The plants must be overwintered in a greenhouse and planted the following spring after the frost free date. In summary, an entire growing season is taken up by the propagation process when using softwood cuttings in the context of a muscadine breeding program. Hardwood cuttings, on the other hand, are taken during the dormant season in the months after the breeder has made selections in August or September. By the following spring planting season, the hardwood cuttings have had months to establish and can be planted without delay. Therefore, an effective and reliable hardwood propagation protocol is able to reduce the muscadine breeding cycle by an entire year.

The staff of the University of Arkansas System Division of Agriculture (UA) Fruit Breeding Program applied a hardwood rooting protocol developed for bunch grapes to muscadine grapes and discovered that hardwood muscadine cuttings rooted at low rates but still produced enough viable plants to meet the program’s modest needs. These cuttings were taken in early December and placed into cold storage until early January, when they were dipped in rooting hormone and placed into a 100% perlite rooting medium under a mist system. This protocol has been the main propagation method for the UA muscadine breeding program for more than 10 years, despite the scientific consensus that hardwood propagation of muscadines could not be done or was too difficult to be effective. Conservative estimates of rooting success were ≈10% year-to-year, with significant differences between the success of different genotypes (D. Gilmore, personal communication).

This study was designed to accomplish two goals. The first was to evaluate the efficacy of a bunch grape hardwood rooting protocol that has been applied to muscadines over the past 10 years in the UA Fruit Breeding Program. The second was to evaluate five factors identified within the literature that likely affect the success of rooting muscadines from hardwood cuttings.

Materials and Methods

Collection of hardwood cuttings.

The study was conducted over 2 years in a greenhouse at the Milo J. Shult Agricultural Research and Extension Center in Fayetteville, AR, from Nov. 2019 to June 2020, and Nov. 2020 to June 2021. Hardwood cuttings were taken from three different vineyards to represent diverse muscadine growing regions: the aforementioned Milo J. Shult Research and Extension Center (lat. 36.0991 N, long. −94.1722 W), the UA Fruit Research Station in Clarksville, AR (lat. 35.5332 N, long. –93.4037 W), and Paulk Vineyards near Ocilla, GA (lat. 31.5746 N, long. –83.0806 W). Three cultivars were selected for this study: Fry, Carlos, and Supreme.

Cuttings were collected from mature vines at each vineyard location. Cuttings were taken from various cane segments and were ≈15 to 20 cm long and 5 to 10 mm wide, with a minimum of three nodes. If leaves were present, they were removed at the time cuttings were collected. The cuttings were cut perpendicular at the base and at a 45° angle at the top to ensure polarity was maintained and facilitate water runoff from the mist system. Hardwood cuttings were collected at the beginning of each month during November, December, January, and February. Collection dates for the 2019–20 season were 4 Nov., 4 Dec., 6 Jan., and 4 Feb., and collection dates for the 2020–21 season were 2 Nov., 1 Dec., 4 Jan., and 2 Feb. At the time of collection, half of all cuttings received a cold storage treatment of 4 °C for 30 d before planting. A summary of factors and treatments evaluated in this experiment is provided in Table 1.

Table 1.

Treatments included as fixed effects in the experiment conducted during the 2019–20 and 2020–21 seasons.

Table 1.

Mist bed and greenhouse conditions.

The rooting containers used for this study were SureRoots Deep Cell 50-cell plug trays (T.O. Plastics, Clearwater, MN) with 12.7 cm deep cells. Trays were cut into 10-cell experimental units to facilitate replication and randomization within the study. The rooting media was 100% perlite, following the rooting protocol used by FRS employees for V. vinifera cuttings. Cuttings were dipped in 0.1% Indole-3-butyric acid powder (Bonide Products Inc., Oriskany, NY) before being inserted into the rooting media such that one node was fully submerged in the rooting media and a second node was level with the surface of the rooting media. One cutting was planted into each cell.

Ambient temperatures in the greenhouse were measured by a centrally located thermostat and maintained between 18 and 24 °C through the course of the study. Two greenhouse benches measuring 1.5 m × 3.0 m were used as mist beds. A 1.9 cm inline sprinkler valve (Rain Bird Corp., Azusa, CA) was connected to a standard hose valve at native city water pressure. A Galcon 8056S AC-6S (Galcon USA LTD., Simi Valley, CA) programmable irrigation controller was wired to the valve. The valve was programmed to run the mist system for 15 s every 10 min with an irrigation window of 6:00 am to 6:00 pm. The irrigation line was 0.64 cm in diameter and suspended ≈0.61 m above the mist benches. Three Coolnet Pro foggers (Netafim Irrigation Inc., Fresno, CA) were spaced evenly lengthwise across each bench. These foggers were a four-nozzle system and each nozzle flowed at 7.6 L·h−1. In addition, an internal check valve inside each fogger ensured that shut off happened quickly after valve closure to maintain a consistent 15-s mist interval. The media was also hand watered to field capacity approximately twice a week during the study as well as immediately after cuttings were placed in the media to ensure adequate moisture availability. Half of all cuttings received continuously supplied bottom heat at 26 °C. The bottom heat treatment was applied using 1.5 m × 53 cm Redi-Heat heavy-duty propagation mats (Phytotronics Inc., Earth City, MO) programmed to maintain an average temperature of 26 °C with a Redi-Heat digital thermostat. The attached soil probe was inserted ≈5 cm into the perlite rooting media.

Data collection.

Hardwood cuttings were removed from the media and growth characteristics were assessed by hand using Pittsburgh digital calipers (Harbor Freight Tools, Camarillo, CA) 90 d after they were placed in the mist beds. For each year of the study, the first date of data collection fell in the first week of February (cuttings that were taken in November and not given the storage treatment) and the last date of data collection occurred during the first week of June (cuttings that were taken in February and given the storage treatment). Rooting occurrence was recorded as both a binary outcome and a quantitative measure of the number of roots and the length of the longest root in cm for each cutting in the 10-cell experimental units. Achieving perfect uniformity of the cuttings was not possible. Therefore, three additional measurements were taken: cutting diameter, cutting length, and number of nodes. Cutting diameter was recorded in mm at the top and bottom of the cutting. The overall length of the cutting in centimeters and the number of nodes per cutting were also recorded. The average of these two measurements was calculated for each experimental unit for the final analysis.

Experimental design and analysis.

Cuttings were placed in 10-cell experimental units with one unit for each of the two blocks in this study. In this study, each block corresponded to a bench within the greenhouse. The proportion of cuttings that rooted in each experimental unit was analyzed using PROC GLIMMIX in SAS 9.4 (SAS Institute Inc., Cary, NC) as a mixed model with five fixed effect treatments, two random effects, and a negative binomial distribution. Bottom heat was the main plot effect and the other four factors (collection date, cold storage, cultivar, and location) were completely randomized within the split plot. Block and year were analyzed as random effects. Four and five-way interactions were dropped from the statistical model to only include main effects and interaction effects that were assumed to have any biological significance as proposed by Harrell (2015). To interpret higher-order interactions, the results of mean separations tests were sliced by location when applicable. Mean separations were performed with Tukey’s honestly significant difference. Graphical representations of data were constructed using the ggplot2 package in RStudio (Wickham, 2016). After analyzing the initial model, a second model was created to distill the results of this study into concise recommendations for breeders rooting hardwood muscadine cuttings. This new model used only data from the November harvest date and treated heat, storage, and their interaction as fixed effects while cultivar and location were treated as random effects.

Results

First frost dates and chill accumulation.

The three vineyard locations selected for this study experienced different climatic conditions across the 2-year study period. For both years, the first frost occurred at the Fayetteville location in mid-October and at the Ocilla location in the first week of December. At Clarksville, the first frost was 31 Oct. 2019 and 30 Nov. 2020 (Table 2). Chill hours were calculated using the Below 45 °F model starting 1 Oct. and the cumulative chilling amount by the first of each month during the trial period was reported for each location. At the first collection date in November, the Clarksville, AR; Fayetteville, AR; and Ocilla, GA locations received an average of 41, 193, and 0 chill hours, respectively (Table 2). By the end of the study in February, the Clarksville, AR; Fayetteville, AR, and Ocilla, GA locations had received an average of 1160, 1541, and 550 chill hours, respectively (Table 2).

Table 2.

First frost dates and chill hours accumulated by the first of each mo. cuttings were taken for each vineyard location during 2019–20 (year 1) and 2020–21 (year 2).

Table 2.

Rooting percentages and correlations among response variables.

The percentage of cuttings that formed roots for the 2 years of this study were 17.1% and 15.0%, respectively, for an overall rooting percentage of 16.0%. There were four higher order interactions that significantly affected the proportion of hardwood muscadine cuttings that rooted: Location × Cultivar × Date, Location × Heat × Date, Heat × Storage × Date, and Location × Cultivar × Storage (Table 3). There were multiple significant positive correlations between rooting success and length and diameter of the cutting (Table 4). Overall length of the cutting was positively correlated with rooting success (r = 0.22, P < 0.001). The number of roots per experimental unit was also positively correlated with cutting length (r = 0.11, P < 0.05). Cutting diameter was significantly correlated with all three rooting variables: rooting success (r = 0.25, P < 0.001), root number (r = 0.15, P < 0.01), and length of the longest root (r = 0.11, P < 0.05). The number of nodes per cutting was not significantly correlated with any of the rooting variables (Table 4).

Table 3.

Analysis of variance results of the main and interaction effects for the five factors used in this study, including collection date, vineyard location, cultivar, cold storage treatment, and bottom heat treatment.

Table 3.
Table 4.

Pearson’s correlation coefficients for variables measuring rooting success and cutting attributes. Rooting success was measured as the proportion of cutting per experimental unit that formed roots by the end of the 90-d rooting period. The root number and longest root variables were the sum of the number of roots in each experimental unit and the sum of the longest root per cutting in each experimental unit, respectively.

Table 4.

Location × Cultivar × Storage interaction.

Because the interaction effect of Location × Cultivar × Storage was significant (P = 0.004, Table 3), means separation was performed after slicing results by location. Only the Fayetteville, AR, location had significant differences between the Cultivar × Storage treatment combinations (Fig. 1). ‘Carlos’ was the only cultivar that rooted significantly better after receiving a cold storage treatment in the Fayetteville location; 18.9% of stored cuttings rooted compared with 6.2% of cuttings that did not receive the cold storage treatment. There was no difference in the rooting ability of the three cultivars when cuttings were not stored. However, stored cuttings of ‘Carlos’ and ‘Fry’ from Fayetteville rooted significantly better (18.9% and 18.7%, respectively) than stored cuttings of ‘Supreme’ (6.5%).

Fig. 1.
Fig. 1.

The effect of Location × Cultivar × Storage treatment combinations on rooting success. Results were sliced by location. Means followed by different letters within each location are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

Citation: HortScience 57, 5; 10.21273/HORTSCI16393-21

Heat × Storage × Date interaction.

The interaction of Heat × Storage × Date was significant (P = 0.004, Table 3). Therefore, means separation was performed after slicing results by date. For cuttings taken in November, 37.8% of cuttings that were not stored and received bottom heat rooted. This treatment combination performed significantly better than the November treatments that did not receive bottom heat, but it was not significantly different from the treatment that received both bottom heat and cold storage (Fig. 2). Furthermore, cuttings that received bottom heat and cold storage rooted significantly better (27.5%) than cuttings that were cold stored but did not receive bottom heat (11.7%). There was no difference between the rooting percentage of cuttings that received storage and those that did not when bottom heat was not supplied (Fig. 2).

Fig. 2.
Fig. 2.

The effects of the Heat × Storage × Date treatment combinations on rooting success. Results were sliced by collection date. Means followed by different letters within each collection date are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

Citation: HortScience 57, 5; 10.21273/HORTSCI16393-21

In December, the experimental units that received no cold storage and no bottom heat had significantly lower rooting percentages (1.0%) than all other treatment combinations (Fig. 2). Cuttings that were stored and received bottom heat performed significantly better (21.7%) than cuttings that were stored but did not receive bottom heat (9.9%). There was no difference in rooting ability between the treatment that received bottom heat but not storage and the treatment that was stored but did not receive bottom heat. In January, for cuttings that were not stored, supplying bottom heat significantly increased rooting percentages (15.3% vs. 5.3%). There were no other differences between treatment combinations. In February, there were no significant differences between treatment combinations. The four treatment combinations ranged from 6% to 11% during this month, the worst overall rooting percentages observed for any of the months in this interaction (Fig. 2).

Location × Cultivar × Date interaction.

For the Clarksville, AR location, there were significant differences in rooting percentages between cultivars within each collection date for all months except for November (Fig. 3). In December, ‘Fry’ had lower rooting success than ‘Carlos’ and ‘Supreme’ (5.6% vs. 10.6% and 15.3%, respectively). However, in January ‘Fry’ rooted at more than double the rate of ‘Supreme’ and ‘Carlos’. ‘Supreme’ had significantly fewer cuttings with roots in February (3.2%) than it did in any other month in Clarksville (13.4% average). ‘Carlos’ rooted equally well in Clarksville across all four dates tested (Fig. 3).

Fig. 3.
Fig. 3.

The effect of the Location × Cultivar × Date treatment combinations on rooting success. Results were sliced by location. Means followed by different letters within each location are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

Citation: HortScience 57, 5; 10.21273/HORTSCI16393-21

For the Fayetteville, AR, vineyard location, there were no significant differences between cultivar rooting efficacy within each month. The only significant differences were across dates and cultivars (Fig. 3). ‘Carlos’ rooted significantly better in November (26.7%) than in December (7.1%), January (10.2%), or February (7.1%). ‘Supreme’ rooted significantly better in November (18.3%) than in December (4.7%) and February (7.4%) but not in January (8.8%). There were no differences in rooting ability for ‘Fry’ across the four collection dates in Fayetteville.

The Ocilla, GA, location in November had the highest rooting percentages observed across the entire study (Fig. 3). ‘Fry’ cuttings collected in November rooted at 48.8%, more than four times the average rooting percentage of the other three collection dates (11.7%). Similarly, 42.9% of ‘Supreme’ cuttings rooted, also more than four times the average of the other collection dates (9.6%). ‘Carlos’ rooted at 29.7%, which is more than four times the average of the other collection dates (7.1%) but only slightly more than the rooting percentage of the ‘Carlos’ from Fayetteville in November (26.7%). Unlike ‘Carlos’, ‘Fry’ and ‘Supreme’ rooted at much higher rates in Ocilla, GA when cuttings were taken in November compared with the other two locations (Fig. 3). Rooting success for cuttings taken from Ocilla, GA in December, January, and February was comparable to the rates observed in cuttings collected in the same month at Fayetteville, AR and Clarksville, AR.

Location × Date × Heat interaction.

For the Clarksville, AR, location there were no significant differences in rooting percentages for cuttings with and without bottom heat within each month of collection. Nor were there any differences in rooting success across months for cuttings that were not supplied bottom heat (Fig. 4). However, supplying bottom heat significantly increased the rooting ability of cuttings in November (24.8%) compared with February (6.2%). Overall, there were fewer significant comparisons between cuttings taken in Clarksville than the other two study sites.

Fig. 4.
Fig. 4.

The effect of the Location × Date × Heat treatment combinations on rooting success. Results were sliced by location. Means followed by different letters within each location are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

Citation: HortScience 57, 5; 10.21273/HORTSCI16393-21

For the Fayetteville location, cuttings taken in November that were supplied bottom heat (30.4%) outperformed cuttings that received bottom heat in December (9.8%) and February (7.0%). That treatment combination also outperformed cuttings in every month but November that were not supplied bottom heat. Within each month, there were no differences between cuttings supplied bottom heat and those without (Fig. 4).

As seen in the results of the Location × Cultivar × Date interaction, the cuttings collected in the Ocilla, GA site in November significantly outperformed most other treatment combinations (Fig. 4). Cuttings taken in November and supplied with bottom heat rooted at 47.4%, although that is not significantly different from cuttings taken in November that did not receive the heat treatment (33.1%). Among cuttings taken in Ocilla, GA, the November cuttings supplied with bottom heat outperformed cuttings not supplied with bottom heat in every other month. For cuttings that were supplied bottom heat, November was a significantly better date to attempt to collect cuttings than January and February. Ocilla, GA, also had the only significant within-month difference for the heat treatment. In December, cuttings supplied with bottom heat rooted at 22.6%, while those without bottom heat rooted at only 2.9%.

For the second model specifically testing the effect of heat and storage on cuttings taken in November across all locations and cultivars, there was no significant effect of supplying bottom heat on the rooting ability of hardwood muscadine cuttings (P = 0.38). The interaction effect between the cold storage and bottom heat factors was also nonsignificant (P = 0.64). However, a higher percentage of cuttings taken in November that were placed in the mist bed within 24 h of collection rooted (27.2%) compared with those given a 1-month cold-storage treatment (19.4%).

Discussion

This study represents the first investigation into the rooting ability of hardwood muscadine cuttings in more than 25 years. Past research used single vineyard locations to evaluate rooting success across a narrow range of collection dates. The combination of multiple important factors within this study (collection date, bottom heat, cold storage, cultivar, and vineyard location) allowed for a comprehensive investigation and analysis of the interactions between these variables and the treatment combinations most likely to yield a higher percentage of rooted cuttings. The 2-year average rooting percentage of 16% reported here is higher than many of the less successful rooting studies of the 20th century (Goode et al., 1982; Woodroof, 1935) but far lower than the few papers that concluded hardwood propagation in muscadines is relatively easy (Newman, 1907; Whatley, 1974). The multiple significant higher-order interactions add considerable complexity to the interpretation of the results, they also provide possible explanations as to why past studies came to radically different conclusions from one another.

The Location × Cultivar × Storage effect was significant, but sliced results by location showed that only the Fayetteville vineyard had significant differences between treatment means. Furthermore, the only cultivar that performed differently based on storage in that interaction was ‘Carlos’ (Fig. 1). Fayetteville is the northernmost location and receives more chill hours than the other two vineyard sites (Table 3). It would be expected that a crop with low chilling requirements such as muscadines, generally only requiring 200 to 300 chill hours (Basiouny and Himelrick, 2001), would not see significant differences in rooting due to an added cold storage treatment. However, a high number of chill hours (>2000) have been shown to increase the number of roots per cutting but not overall rooting percentage in Euvitis hardwood cuttings (Keeley et al., 2004). Regardless of the physiological effects of chill hours on rooting ability, cold storage treatments have been used previously in muscadine rooting protocols. Whatley (1974), based in Alabama, reported that muscadines root readily from hardwood cuttings taken in December when cold stored for 60 to 90 d, significantly longer than the 30-d storage treatment applied in this study.

The effects of heat and storage on rooting success varied depending on the collection date of the cuttings (Fig. 2). Bottom heat and cold storage had a larger effect on rooting outcome in November and December than in January and February. The attenuated effects of these treatments in January and February are possibly a result of climatic conditions cuttings experienced during the study. If the number of chilling hours accrued by Vitis vines do in fact have an effect on rooting success (Keeley et al., 2004; Smith and Wareing, 1972), it is possible that those requirements have been fulfilled by January or February at the vineyard locations used in this study. In that case, a month-long cold storage treatment (or lack thereof) would not affect the physiology of the cuttings in a way that promotes rooting.

In addition, conditions in the greenhouse were warmer during the 90-d rooting period for cuttings taken relatively late in the winter. Cuttings taken in February and given the cold storage treatment were not removed from rooting media for data collection until the beginning of June, when supplemental bottom heat may not be as necessary for stimulating rooting due to high ambient temperatures in the greenhouse. Cuttings taken in November that did not receive the storage treatment but were supplied bottom heat rooted at 37.8%, significantly higher than 12 of the 15 other treatment combinations in the comparison of Heat × Storage × Date treatment effects (Fig. 2). Of the three treatment combinations within the Heat × Storage × Date interaction that were not significantly different from one another, two were from November and one from December. The effect of date and bottom heat on rooting ability is mentioned extensively throughout the available literature. Newman (1907) and Dearing (1947), located in South Carolina and North Carolina, respectively, reported rooting success as high as 80% when cuttings were taken in November and supplied with bottom heat while Goode et al. (1982), located in Georgia, observed 0% rooting in cuttings taken from late November to February and not supplied bottom heat. Hypothetically, supplying bottom heat early in the dormant season may prevent these cuttings from achieving full dormancy and therefore help stimulate rooting.

It is expected that rooting success is dependent to some extent on the cultivar being propagated, although in this study no cultivar consistently outperformed the others. Cultivars such as Thomas and Hunt have been shown to root better than ‘Scuppernong,’ although these cultivars are no longer widely planted (Dearing, 1947; Niven, 1918; Woodroof, 1935). None of the cultivars used in this study have been used in previous research on propagation. It was observed in our research that the rooting success of muscadine cuttings depended not only on the cultivar, but also on the location and date in which the cutting was taken (Fig. 3). Notably, ‘Fry’ outperformed ‘Supreme’ at the Ocilla, GA, and Clarksville, AR, locations.

Previous studies rarely mention the effect of location on rooting success, although Dearing (1947) felt that some observed differences between sites could be attributed to soil types and overall climatic differences. To date, no other study has specifically tested the effect of vineyard location on the rooting ability of hardwood muscadine cuttings. Fluctuations in rooting ability through the winter season have been well-documented in other perennial fruit crops and are generally linked to physiological factors related to dormancy (Bassuk and Howard, 1981; Guerriero and Loreti, 1975; Smith and Wareing, 1972). The native range of muscadines covers various climatic regions, resulting in variations between when vines enter dormancy. Thus, the prescriptive recommendations of previous studies regarding when hardwood cuttings should be taken may not be applicable to all locations where muscadines are grown. In this study, cuttings taken from Ocilla, GA (USDA Hardiness Zone 8b) performed exceptionally well in early November, well before the first frost date at that location for both years of the study and when no chilling hours had been accumulated (Table 3, Fig. 3). We hypothesize that the high rooting percentages from cuttings taken in November in Ocilla, GA, could be due in part to the incomplete dormancy of the vines at this location. The precipitous drop in rooting percentages in Georgia after November would be explained by the vines fully entering dormancy (Figs. 3 and 4). At the Clarksville and Fayetteville locations, the more consistent rooting percentages from month-to-month would be explained by the fact that by early November, these vines were already fully dormant (Table 3). The UA Fruit Breeding Program intends to mimic the results observed in this study from cuttings taken in November from Ocilla by taking hardwood cuttings from Clarksville in early October this year (Margaret Worthington, Personal Communication). The goal of changing this date to October from December is to potentially take advantage of the vines in Clarksville not being completely dormant while still keeping the propagation schedule outside the busy growing season.

Vegetative buds play an important role in rooting Vitis species from cuttings by serving as endogenous sources of auxin hormones that promote adventitious rooting (Kawai, 1996; Thomas and Schiefelbein, 2004). In this study cuttings were collected with at least three nodes. There was no observed correlation between the number of nodes per cutting and rooting success (Table 2), although the exogenous rooting hormone application may have affected these findings by supplying a large amount of synthetic auxin.

Significant positive correlations with increased rooting percentage were found for both increased length of the cutting and thicker cutting diameter. The average recommended length of hardwood cuttings is often mentioned throughout the literature as part of rooting protocols but has not been studied as a factor in rooting success of muscadines (Dearing, 1947; Newman, 1907; Woodroof, 1935). Increased cutting length has been observed to affect rooting in other fruit crops (Aljane and Nahdi, 2014; Wainwright and Hawkes, 1988), although the correlation between cutting length and rooting success is not always positive as it was in this study (Exadaktylou et al., 2009). Our results also support previous findings on the association of increased cutting diameter and higher rooting success (Castro et al., 1994; Goode et al., 1982). The differentiation of rooting ability based on cane position is well documented in both other Vitis species and other woody plant species (Hartmann et al., 1997; Keeley et al., 2004). In this study, cane position was not recorded at the time of cutting collection, and cuttings were taken from a variety of positions along the cane. Cutting diameter may function as an estimate for cane position and might explain the observed correlation. The differences in rooting ability were possibly due to a gradient in moisture, nitrogen levels, and chemical composition that has been observed from proximal portions of the cane to the growing point of the plant (Geny et al., 2002; Hartmann et al., 1997; Tukey and Green, 1934).

A second statistical model was conducted after initial data analysis to provide better broad recommendations for those attempting to root muscadines from hardwood cuttings. This second dataset included only cuttings taken in November because those cuttings generally outperformed cuttings taken during other months. The model was simplified to only test the effects of the heat and storage treatments. Location and cultivar were made random factors to make broad recommendations for breeders across the country propagating a broad range of cultivars and selections. The negative effect on rooting ability observed when cuttings taken in November were given a cold storage treatment supports our hypothesis that taking cuttings before vines are completely dormant is the ideal hardwood propagation method. Therefore, we recommend that breeders hoping to save a year advancing new selections compared with softwood cuttings should take cuttings from vines after they have fruited, but before they are fully dormant. These cuttings should be placed in the mist bed without cold storage within a day of collection. Previous researchers (e.g., Goode et al., 1982; Newman, 1907) found that bottom heat improved rooting percentages. The average percent rooting of cuttings supplied bottom heat was 30%, compared with 17% of cuttings which were not given supplemental bottom heat, yet the effect of bottom heat was nonsignificant in this study (P = 0.38). Bottom heat was assigned to main plots in this split plot design experiment because of practical constraints related to the size of the heat mats. Main plot treatments in a split plot experiment are measured with less precision than they would be in a randomized complete block design. Therefore, the nonsignificant effect of bottom heat in this study can be partially attributed to its assignment to main plots. Another possibility for the inconsistency between our results and previous research could be differences in the ability of researchers to maintain stable ambient greenhouse temperatures during the winter months. Thus, providing supplemental bottom heat may be a helpful option for anyone attempting this protocol if it is difficult to consistently heat the greenhouse during the propagation period.

Conclusion

This research provides significant evidence that rooting muscadines from hardwood cuttings is a more effective method of propagation than many previous studies have concluded. The complex experimental design of the study allowed for the higher-order interactions affecting the rooting success of muscadine hardwood cuttings to be elucidated. Cuttings taken from Georgia in early November formed roots more than 40% of the time, perhaps indicating that an incomplete transition to dormancy may have played a role in the success of this study and others. When cuttings from November were analyzed alone with location and cultivar considered random effects, applying a cold storage treatment reduced rooting percentages from 27% to 19%, and the effect of bottom heat was nonsignificant. Increased cutting length and cutting diameter were significantly correlated with increased rooting success while an increased number of nodes was not. Thus, although the rooting percentages reported in this study may not allow for commercially successful propagation of muscadines by hardwood cuttings, breeding programs or germplasm repositories with modest needs may find that transitioning to an off-season propagation protocol may save time and money.

Literature Cited

  • Aljane, F. & Nahdi, S. 2014 Propagation of some local fig (Ficus carica L.) cultivars by hardwood cuttings under the field conditions in Tunisia Int. Sch. Res. Notices 809450 https://doi.org/10.1155/2014/809450

    • Search Google Scholar
    • Export Citation
  • Basiouny, F.M. & Himelrick, D.G. 2001 Muscadine grapes ASHS Press Alexandria, VA

  • Bassuk, N.L. & Howard, B.H. 1981 A positive correlation between endogenous root-inducing cofactor activity in vacuum-extracted sap and seasonal changes in rooting of M.26 winter apple cuttings J. Hort. Sci. 56 301 312 https://doi.org/10.1080/00221589.1981.11515006

    • Search Google Scholar
    • Export Citation
  • Castro, P.R.C., Melotto, E., Soares, F.C., Passos, I.R.S. & Pommer, C.V. 1994 Rooting stimulation in muscadine grape cuttings Sci. Agr. 51 436 440 https://doi.org/10.1590/S0103-90161994000300009

    • Search Google Scholar
    • Export Citation
  • Comeaux, B.L., Nesbitt, W.B. & Fants, P.R. 1987 Taxonomy of the native grapes of North Carolina Castanea 52 197 215

  • Cowart, F.F. & Savage, E.F. 1944 The effect of various treatments and methods of handling upon rooting of muscadine grape cuttings Proc. Amer. Soc. Hort. Sci. 44 312 314

    • Search Google Scholar
    • Export Citation
  • Daskalakis, I., Biniari, K., Bouza, D. & Stavrakaki, M. 2018 The effect that indolebutyric acid (IBA) and position of cane segment have on the rooting of cuttings from grapevine rootstocks and from Cabernet franc (Vitis vinifera L.) under conditions of a hydroponic culture system Scientia Hort. 227 79 84 https://doi.org/10.1016/j.scienta.2017.09.024

    • Search Google Scholar
    • Export Citation
  • Dearing, C. 1947 Muscadine grapes USDA Farmers’ Bulletin 1785

  • Doelle, H.W. & Mitchell, T.C. 1964 Mist propagation method for rooting hardwood cuttings of the grape variety ‘Black Shiraz’ Amer. J. Enol. Viticult. 15 17 22

    • Search Google Scholar
    • Export Citation
  • Exadaktylou, E., Thomidis, T., Grout, B., Zakynthinos, G. & Tsipouridis, C. 2009 Methods to improve the rooting of hardwood cuttings of the ‘Gisela 5’ cherry rootstock HortTechnology 19 254 259 https://doi.org/10.21273/HORTSCI.19.2.254

    • Search Google Scholar
    • Export Citation
  • Fehr, W. 1991 Principles of cultivar development: Theory and technique Macmillan Publishing Company New York, NY

  • Geny, L., Dalmasso, R. & Broquedis, M. 2002 Polyamines and adventitious root formation in Vitis vinifera L OENO One 36 97 102 https://doi.org/10.20870/oeno-one.2002.36.2.1689

    • Search Google Scholar
    • Export Citation
  • Goldy, R.G. 1992 Breeding muscadine grapes Hort. Rev. 14 357 405

  • Goode, D.Z. Jr, Krewer, G.W., Lane, R.P., Daniell, J.W. & Couvillon, G.A. 1982 Rooting studies of dormant muscadine grape cuttings HortScience 17 644 645

    • Search Google Scholar
    • Export Citation
  • Guerriero, R. & Loreti, F. 1975 Relationships between bud dormancy and rooting ability in peach hardwood cuttings Acta Hort. 54 51 58 https://doi.org/10.17660/ActaHortic.1975.54.6

    • Search Google Scholar
    • Export Citation
  • Harrell, F.E. 2015 General aspects of fitting regression models 13 44 Regression Modeling Strategies. Springer Series in Statistics. Springer Cham, Switzerland https://doi.org/10.1007/978-3-319-19425-7_2

    • Search Google Scholar
    • Export Citation
  • Hartmann, H.T., Kester, D.E. & Davies, F.T. Jr 1997 Plant propagation principles and practices 6th ed. Prentice Hall Englewood Cliffs, NJ

  • Kawai, Y. 1996 Changes in endogenous IAA during rooting of hardwood cuttings of grape, ‘Muscat Bailey A’ with and without a bud J. Jpn. Soc. Hort. Sci. 65 33 39

    • Search Google Scholar
    • Export Citation
  • Keeley, K., Preece, J.E., Taylor, B.H. & Dami, I.E. 2004 Effects of high auxin concentrations, cold storage, and cane position on improved rooting of Vitis aestivalis Michx. Norton cuttings Amer. J. Enol. Viticult. 55 265 268

    • Search Google Scholar
    • Export Citation
  • Liu, X.Q., Ickert-Bond, S.M., Nie, Z.L., Zhou, Z., Chen, L.Q. & Wen, J. 2016 Phylogeny of the Ampelocissus-Vitis clade in Vitaceae supports the new world origin of the grape genus Mol. Phylogenet. Evol. 95 217 228 https://doi.org/10.1016/j.ympev.2015.10.013

    • Search Google Scholar
    • Export Citation
  • Munson, T.V. 1909 Foundations of American grape culture T.V. Munson and Son Denison, TX

  • Newman, C.C. 1907 Rotundifolia Grapes South Carolina Agr. Exp. Sta. Bul. 132

  • Niven, L.A. 1918 How to propagate muscadine grapes The Progressive Farmer 33 20

  • Sharpe, R.H. 1954 Rooting of muscadine grapes under mist Proc. Amer. Soc. Hort. Sci. 63 88 90

  • Smith, N.G. & Wareing, P.F. 1972 Rooting of hardwood cuttings in relation to bud dormancy and the auxin content of the excised stems New Phytol. 1972 63 80 https://doi.org/10.1111/j.1469-8137.1972.tb04811.x

    • Search Google Scholar
    • Export Citation
  • Thomas, P. & Schiefelbein, J.W. 2004 Roles of leaf in regulation of root and shoot growth from single node softwood cuttings of grape (Vitis vinifera) Ann. Appl. Biol. 144 27 37 https://doi.org/10.1111/j.1744-7348.2004.tb00313.x

    • Search Google Scholar
    • Export Citation
  • Tukey, H.B. & Green, E.L. 1934 Gradient composition of rose shoots from tip to base Plant Physiol. 9 157 163

  • Wainwright, W. & Hawkes, H.Y. 1988 The influence of the length of hardwood cuttings on the propagation of blackcurrant (Ribes nigrum) Acta Hort. 227 266 268 https://doi.org/10.17660/ActaHortic.1988.227.46

    • Search Google Scholar
    • Export Citation
  • Whatley, B.T. 1974 Vegetative production of Vitis rotundifolia Combined Proc. Intl. Plant Prop. Soc. 24 376 377

  • Woodroof, J.G. 1935 Developments in growing muscadine grapes in the south Proc. Amer. Soc. Hort. Sci. 33 447 449

  • Wickham, H. 2016 ggplot2: Elegant graphics for data analysis Springer-Verlag New York, NY

Contributor Notes

We thank Jackie Lee, David Gilmore, Michael Brown, and the staff of Paulk Vineyards for assistance in plot maintenance. Thanks to Lacy Nelson, Carmen Johns, Mason Chizk, Carly Godwin, Autumn Brown, and Hannah Mather for assistance in data collection. We also thank Garry McDonald for use of greenhouse space and consultation on mist bed construction. David Gilmore and Scott Loving provided useful information about the hardwood rooting protocol used at the University of Arkansas System Division of Agriculture Fruit Research Station, and Jeff Bloodworth provided numerous suggestions about factors to investigate in this study. This research was funded by a grant from the Southern Region Small Fruit Consortium and Hatch Project ARK02599.

M.W. is the corresponding author. E-mail: mlworthi@uark.edu.

  • Collapse
  • Expand
  • View in gallery
    Fig. 1.

    The effect of Location × Cultivar × Storage treatment combinations on rooting success. Results were sliced by location. Means followed by different letters within each location are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

  • View in gallery
    Fig. 2.

    The effects of the Heat × Storage × Date treatment combinations on rooting success. Results were sliced by collection date. Means followed by different letters within each collection date are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

  • View in gallery
    Fig. 3.

    The effect of the Location × Cultivar × Date treatment combinations on rooting success. Results were sliced by location. Means followed by different letters within each location are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

  • View in gallery
    Fig. 4.

    The effect of the Location × Date × Heat treatment combinations on rooting success. Results were sliced by location. Means followed by different letters within each location are significantly different using Tukey’s honestly significant difference at α = 0.05. Error bars represent the standard error of each measurement.

  • Aljane, F. & Nahdi, S. 2014 Propagation of some local fig (Ficus carica L.) cultivars by hardwood cuttings under the field conditions in Tunisia Int. Sch. Res. Notices 809450 https://doi.org/10.1155/2014/809450

    • Search Google Scholar
    • Export Citation
  • Basiouny, F.M. & Himelrick, D.G. 2001 Muscadine grapes ASHS Press Alexandria, VA

  • Bassuk, N.L. & Howard, B.H. 1981 A positive correlation between endogenous root-inducing cofactor activity in vacuum-extracted sap and seasonal changes in rooting of M.26 winter apple cuttings J. Hort. Sci. 56 301 312 https://doi.org/10.1080/00221589.1981.11515006

    • Search Google Scholar
    • Export Citation
  • Castro, P.R.C., Melotto, E., Soares, F.C., Passos, I.R.S. & Pommer, C.V. 1994 Rooting stimulation in muscadine grape cuttings Sci. Agr. 51 436 440 https://doi.org/10.1590/S0103-90161994000300009

    • Search Google Scholar
    • Export Citation
  • Comeaux, B.L., Nesbitt, W.B. & Fants, P.R. 1987 Taxonomy of the native grapes of North Carolina Castanea 52 197 215

  • Cowart, F.F. & Savage, E.F. 1944 The effect of various treatments and methods of handling upon rooting of muscadine grape cuttings Proc. Amer. Soc. Hort. Sci. 44 312 314

    • Search Google Scholar
    • Export Citation
  • Daskalakis, I., Biniari, K., Bouza, D. & Stavrakaki, M. 2018 The effect that indolebutyric acid (IBA) and position of cane segment have on the rooting of cuttings from grapevine rootstocks and from Cabernet franc (Vitis vinifera L.) under conditions of a hydroponic culture system Scientia Hort. 227 79 84 https://doi.org/10.1016/j.scienta.2017.09.024

    • Search Google Scholar
    • Export Citation
  • Dearing, C. 1947 Muscadine grapes USDA Farmers’ Bulletin 1785

  • Doelle, H.W. & Mitchell, T.C. 1964 Mist propagation method for rooting hardwood cuttings of the grape variety ‘Black Shiraz’ Amer. J. Enol. Viticult. 15 17 22

    • Search Google Scholar
    • Export Citation
  • Exadaktylou, E., Thomidis, T., Grout, B., Zakynthinos, G. & Tsipouridis, C. 2009 Methods to improve the rooting of hardwood cuttings of the ‘Gisela 5’ cherry rootstock HortTechnology 19 254 259 https://doi.org/10.21273/HORTSCI.19.2.254

    • Search Google Scholar
    • Export Citation
  • Fehr, W. 1991 Principles of cultivar development: Theory and technique Macmillan Publishing Company New York, NY

  • Geny, L., Dalmasso, R. & Broquedis, M. 2002 Polyamines and adventitious root formation in Vitis vinifera L OENO One 36 97 102 https://doi.org/10.20870/oeno-one.2002.36.2.1689

    • Search Google Scholar
    • Export Citation
  • Goldy, R.G. 1992 Breeding muscadine grapes Hort. Rev. 14 357 405

  • Goode, D.Z. Jr, Krewer, G.W., Lane, R.P., Daniell, J.W. & Couvillon, G.A. 1982 Rooting studies of dormant muscadine grape cuttings HortScience 17 644 645

    • Search Google Scholar
    • Export Citation
  • Guerriero, R. & Loreti, F. 1975 Relationships between bud dormancy and rooting ability in peach hardwood cuttings Acta Hort. 54 51 58 https://doi.org/10.17660/ActaHortic.1975.54.6

    • Search Google Scholar
    • Export Citation
  • Harrell, F.E. 2015 General aspects of fitting regression models 13 44 Regression Modeling Strategies. Springer Series in Statistics. Springer Cham, Switzerland https://doi.org/10.1007/978-3-319-19425-7_2

    • Search Google Scholar
    • Export Citation
  • Hartmann, H.T., Kester, D.E. & Davies, F.T. Jr 1997 Plant propagation principles and practices 6th ed. Prentice Hall Englewood Cliffs, NJ

  • Kawai, Y. 1996 Changes in endogenous IAA during rooting of hardwood cuttings of grape, ‘Muscat Bailey A’ with and without a bud J. Jpn. Soc. Hort. Sci. 65 33 39

    • Search Google Scholar
    • Export Citation
  • Keeley, K., Preece, J.E., Taylor, B.H. & Dami, I.E. 2004 Effects of high auxin concentrations, cold storage, and cane position on improved rooting of Vitis aestivalis Michx. Norton cuttings Amer. J. Enol. Viticult. 55 265 268

    • Search Google Scholar
    • Export Citation
  • Liu, X.Q., Ickert-Bond, S.M., Nie, Z.L., Zhou, Z., Chen, L.Q. & Wen, J. 2016 Phylogeny of the Ampelocissus-Vitis clade in Vitaceae supports the new world origin of the grape genus Mol. Phylogenet. Evol. 95 217 228 https://doi.org/10.1016/j.ympev.2015.10.013

    • Search Google Scholar
    • Export Citation
  • Munson, T.V. 1909 Foundations of American grape culture T.V. Munson and Son Denison, TX

  • Newman, C.C. 1907 Rotundifolia Grapes South Carolina Agr. Exp. Sta. Bul. 132

  • Niven, L.A. 1918 How to propagate muscadine grapes The Progressive Farmer 33 20

  • Sharpe, R.H. 1954 Rooting of muscadine grapes under mist Proc. Amer. Soc. Hort. Sci. 63 88 90

  • Smith, N.G. & Wareing, P.F. 1972 Rooting of hardwood cuttings in relation to bud dormancy and the auxin content of the excised stems New Phytol. 1972 63 80 https://doi.org/10.1111/j.1469-8137.1972.tb04811.x

    • Search Google Scholar
    • Export Citation
  • Thomas, P. & Schiefelbein, J.W. 2004 Roles of leaf in regulation of root and shoot growth from single node softwood cuttings of grape (Vitis vinifera) Ann. Appl. Biol. 144 27 37 https://doi.org/10.1111/j.1744-7348.2004.tb00313.x

    • Search Google Scholar
    • Export Citation
  • Tukey, H.B. & Green, E.L. 1934 Gradient composition of rose shoots from tip to base Plant Physiol. 9 157 163

  • Wainwright, W. & Hawkes, H.Y. 1988 The influence of the length of hardwood cuttings on the propagation of blackcurrant (Ribes nigrum) Acta Hort. 227 266 268 https://doi.org/10.17660/ActaHortic.1988.227.46

    • Search Google Scholar
    • Export Citation
  • Whatley, B.T. 1974 Vegetative production of Vitis rotundifolia Combined Proc. Intl. Plant Prop. Soc. 24 376 377

  • Woodroof, J.G. 1935 Developments in growing muscadine grapes in the south Proc. Amer. Soc. Hort. Sci. 33 447 449

  • Wickham, H. 2016 ggplot2: Elegant graphics for data analysis Springer-Verlag New York, NY

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 258 258 6
PDF Downloads 290 290 6