Kenneth Buck, University of Arkansas; Margaret Worthington, University of Arkansas; and Patrick Conner, University of Georgia


Rooting muscadines from hardwood cuttings is generally viewed as a difficult, if not impossible, task. The majority of the literature on the topic is from the first half of the 20th century, and even the more recent studies from the 80s and 90s had very little success using hardwood cuttings. Therefore, most muscadine breeders, nurseries, and germplasm repositories propagate muscadines by layering or rooting softwood cuttings under mist. In a breeding program, the selection of muscadine seedlings occurs at the end of the growing season in September. A reliance on softwood cuttings requires the postponement of propagation of selected seedlings until the next growing season, generally in June or July. This propagation schedule falls into the busiest time of the year for fruit breeding programs and delays the establishment of plots and evaluation of new selections by a full season. This yearlong delay adds considerable time to the already lengthy process of releasing a variety. The development of a reliable protocol for muscadine propagation by hardwood cuttings would allow propagation work to be conducted after the conclusion of the growing season at a time when work in the field is beginning to slow and would increase the speed of cultivar development. 

Despite all the literature claiming that muscadine propagation with hardwood cuttings is ineffective, the University of Arkansas System Division of Agriculture Fruit Research Station (UA-FRS) has been using hardwood cuttings to meet its modest muscadine propagation needs for the past 15 years. The methods used at the Fruit Research Station were adapted from the protocol used for hardwood propagation of bunch grapes. Using these methods, success rates are highly variable, depending on both cultivar and year, ranging from 10-70%. The goal of this SRSFC-funded study was to test the efficacy of the FRS propagation protocol and the impact of multiple factors frequently mentioned in previous studies that may affect rooting success.

There were five factors tested in this study:

  • Location- Vineyards in Clarksville, AR, Fayetteville, AR, and Ocilla, GA.
  • Cultivar- ‘Fry’, ‘Carlos’, and ‘Supreme.’
  • Collection Date- Cuttings were taken once at the beginning of November, December, January, and February during the winters of 2019/2020 and 2020/2021.
  • Storage- Half of all cuttings were given a month-long cold storage treatment at 4 °C prior to the rooting treatment.
  • Bottom Heat- Half of all cuttings were given continuous bottom heat at 26 °C.

Data on cutting length, cutting diameter, and number nodes per cutting were also collected 90 days after the cuttings were treated with rooting hormone and placed in perlite media under a mist system in a heated greenhouse.

Generally, cuttings taken in November outperformed cuttings taken at other dates. This trend is most notable at the Ocilla, GA vineyard, where rooting percentages were over 40% (Figure 1). We hypothesize that this is due to the incomplete dormancy of those vines. Located in southern Georgia, Ocilla had not experienced its first frost by the time cuttings were taken in November. However, cuttings taken in November also performed well in the other study locations. Supplying bottom heat early in the dormant season appears to have helped stimulate rooting, but did not seem to have an effect on cuttings taken in January or February. In addition, increased cutting diameter and cutting length positively affected rooting success. The rooting percentages found in this study were not high enough to justify commercial propagation of muscadines by hardwood cuttings. However, breeding programs, germplasm repositories, or growers with modest needs may find that transitioning to an off-season propagation protocol saves time and money.

Figure 1.  Percent of cuttings successfully rooted from ‘Carlos’, ‘Fry’, and ‘Supreme’ vines collected in early November, December, January, and February from vineyards in Clarksville, AR, Fayetteville, AR, and Ocilla, GA. Because of a significant three-way interaction, data is presented separately for each location. Means with different letter(s) in each panel are significantly different (a = 0.05) according to Tukey’s honestly significant difference.
Figure 2. Muscadine vines that have been successfully rooted from hardwood cuttings.
Posted in: