About Us | Help Videos | Contact Us | Subscriptions

Agronomy Journal - Biofuels

Vegetative Propagation of Napiergrass and Energycane for Biomass Production in the Southeastern United States


This article in AJ

  1. Vol. 104 No. 2, p. 518-522
    unlockOPEN ACCESS
    Received: Sept 16, 2011

    * Corresponding author(s): joe_knoll@yahoo.com
Request Permissions

  1. Joseph E. Knoll *a and
  2. William F. Andersona
  1. a USDA-ARS Crop Genetics and Breeding Research Unit, P.O. Box 748, Tifton, GA 31793. Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture


With renewed interest in planting large areas to bioenergy crops, improvements in establishment methods are needed. Our objectives were to evaluate propagation methods with stem cuttings of napiergrass (Pennisetum purpureum Schum.), and to determine the optimum planting date for napiergrass and energycane (Saccharum hybrid) in the southeastern Coastal Plain of the United States. A greenhouse study with ‘Merkeron’ napiergrass showed only minor differences between horizontal buried planting and vertical planting with one node exposed. However, cuttings taken from the lower portion of the parent stem were superior to younger material from the upper portion. The rooting hormone indole-3-butyric acid (IBA) did not affect propagation success. Nine napiergrass genotypes were compared for response to cutting length in the field (1, 2, 5, or 10 nodes cutting−1, with 10 nodes plot−1). Single-node cuttings tended to produce more shoots plot−1 initially in the fall, but many of these did not survive the winter. Generally, higher emergence percentages were achieved with 5- and 10-node cuttings. Seven genotypes of napiergrass and two of energycane were compared for response to planting date. Five biweekly plantings were made beginning 17 Sept. 2009, and six plantings beginning 2 Sept. 2010. In 2009 initial fall emergence was greater in earlier plantings, but the following spring, both early and late plantings had similar numbers of shoots cutting−1, while intermediate plantings had the least. In 2010 there was less variation in fall emergence among planting dates, but the following spring the earliest plantings had more shoots cutting−1 than later plantings.


    IBA; indole-3-butyric acid

Napiergrass and energycane are high-yielding, perennial bunchgrasses that have potential as bioenergy crops in the lower southeastern United States (Anderson et al., 2008). Seeds produced by these grasses have low germination and result in seedlings that are not true-to-type. Thus, vegetative propagation is the preferred method of establishment for these crops (Hanna et al., 2004). Crown divisions can be used for propagation but this method is laborious and expensive. Stem cuttings, sometimes referred to as billets, are generally used to establish both napiergrass and energycane.

Some napiergrass genotypes have survived for many years under good management (Hanna et al., 2004) in the mild temperate climate of southern Georgia (USDA Plant Hardiness Zone 8a; average minimum temperature: –12 to –9°C). Also, Roach (1978) has shown that energycane is more cold tolerant than ordinary sugarcane (S. officinarum L). Vegetative propagation of new cold-tolerant genotypes of napiergrass and energycane has not been extensively studied, and the optimal planting time for these crops has not been determined for locations north of Florida with colder winter temperatures. Thus, stand establishment from stem cuttings can be problematic in this region. With renewed interest in planting large acreages to bioenergy crops, improvements in stand establishment of napiergrass and energycane are imperative.

Three methods of planting napiergrass cuttings are reported in the literature (Woodard et al., 1985; Ssekabembe, 1998). Short cuttings containing two to three nodes can be planted in soil vertically or at an angle, with one or two nodes below the surface and one above. Cuttings of varying length can also be completely buried horizontally, about 10 cm below the soil surface for tall varieties (Woodard et al., 1985; Sladden et al., 1991). Woodard et al. (1985) compared these three planting methods in the field and found no significant effects on establishment or winter survival. Ssekabembe (1998) noted only a slight improvement in sprouting from angled planting vs. horizontal burial, and reported that very wet conditions may cause buried cuttings to rot more often than angled cuttings. Conversely, horizontal burial may be more advantageous under drier conditions. Woodard et al. (1985) reported that cutting length and planting date also affected propagation success in napiergrass. Planting date has also been reported to affect first-year yields of sugarcane in Louisiana (Viator et al., 2005; White et al., 2010).

The purpose of this research is to determine the best propagation method for napiergrass in the southeastern Coastal Plain of the United States, closer to the northern limit of adaptation for these crops. First a greenhouse study was conducted to examine the response of Merkeron napiergrass cuttings to vertical vs. horizontal planting and to application of rooting hormone. Two field studies were also conducted, each over 2 yr. The objective of the first field study was to determine the optimum cutting length for nine napiergrass genotypes, while the objective of the second field study was to determine the optimum planting date for seven genotypes of napiergrass and two of energycane.


Greenhouse Experiment: Planting Methods and Rooting Hormone

Mature canes of Merkeron napiergrass (Burton, 1989) were harvested from field-grown nursery plots at Tifton, GA, in early December 2009, before killing freeze. Leaves were removed from the canes, and two-node cuttings (approximately 20 cm) were cut from either the lower portion or upper portion of the stems. Lower-node cuttings were selected from the first through sixth nodes just above the brace roots. Cuttings with roots already initiated were avoided. Upper-node cuttings were selected from the portion of the cane just below where the leaves were beginning to senesce. Cuttings with shoots already growing were also avoided. Cuttings were immersed for 10 s in aqueous solution of IBA (Hortus USA Corp., New York) at varying concentrations (0, 10, 25, 50, or 100 mg IBA L−1). They were then planted in rectangular 30 L Romana planters (Akro Mils, Akron, OH) either vertically with one node below the surface or horizontally buried about 10-cm deep. The pots were filled with a mixture of Pro-Mix BX with Biofungicide (Premier Horticulture, Inc., Quakertown, PA) and clean sand at a volumetric ratio of 3:1. Each pot contained eight cuttings, and all treatment combinations were replicated twice in a randomized complete block design. Cuttings were maintained in a greenhouse at approximately 26°C, and were watered once daily for 15 min by mist to maintain high humidity. The total number of shoots (both primary and secondary) per pot was counted at 7, 10, and 14 d after planting. Cuttings were then carefully harvested. The roots were carefully washed, and then the fresh shoots and roots were immediately weighed. The lengths of the longest root and shoot of each cutting were measured, and the total number of shoots, both above and below the soil surface, for each cutting was recorded. The number of viable cuttings in each pot was counted. Cuttings producing roots and at least one shoot, and which showed no symptoms of decay, were scored as viable.

Field Experiment 1—Stem Cutting Length

Both field tests were conducted at Tifton, GA, on an Alapaha loamy sand (loamy, siliceous, subactive, thermic Arenic Plinthic Paleaquults). Nine napiergrass genotypes were included in the cutting length test: Merkeron, 04-1-62, 07-29, 07-83, 07-110, 07-148, 07-219, 07-239, and 07-353. Merkeron is a high-biomass cultivar (Burton, 1989), while the other genotypes were developed in the USDA-ARS Tifton breeding program, and are under evaluation for use in biomass production. Cuttings were taken from the lower portion of the stem, and all leaves were removed. Cuttings contained 1, 2, 5, or 10 nodes each, and each plot contained 10 total nodes. All cuttings were planted horizontally, approximately 10-cm deep. This experiment was repeated three times with plantings on 29 Oct. 2009, 17 Sept. 2010, and 15 Nov. 2010; each planting contained three replications. For the first two plantings, emergence of primary shoots was counted about 4 wk after planting. Fall emergence was not observed for the 15 Nov. 2010 planting. The following spring (25 May 2010 and 6 Apr. 2011) the number of primary shoots (hills) was again counted for each plot. Soil and air temperatures at the research site were warmer in the 2010–2011 season, while precipitation was greater in the 2009–2010 season (Fig. 1). This hastened growth and heat accumulation, which concluded the experiment earlier in 2011. No irrigation or fertilizer was applied during the experiment in either year.

Fig. 1.
Fig. 1.

Monthly precipitation and average 10-cm soil temperatures for the duration of the field experiments at Tifton, GA. Data were obtained from the Georgia Automated Environmental Monitoring Network (University of Georgia, 2011).


Field Experiment 2—Planting Date

A second experiment was conducted to assess planting date effects on propagation from stem cuttings. This test included seven napiergrass genotypes (Merkeron, 04-1-62, 07-83, 07-110, 07-219, 07-239, and 07-353), and two energycane genotypes (Ho-02-144 and L 79-1002; Bischoff et al., 2008). In 2009, the first planting was made on 17 September, and successive plantings were made every 2 wk for five total plantings. In 2010, the first planting was made on 3 September, and then every 2 wk for six total plantings. Each planting consisted of four replications. Cuttings were taken from the lower portion of the stem, and contained 10 nodes each. As before, all leaves were removed from the cuttings, and all cuttings were planted horizontally, approximately 10-cm deep. Emergence of primary shoots was counted biweekly until mid-winter. The following spring (25 May 2010 and 6 Apr. 2011) the number of shoots (hills) was recorded for each plot. No irrigation or fertilizer was applied during the experiment.

Data Analysis

Data were analyzed using the GLIMMIX procedure in SAS v. 9.2 (SAS Institute, Cary, NC). Tukey's studentized range test was used to determine significant differences between treatment LS means at α = 0.05. The CORR procedure was used to test for associations between dependant variables.


Greenhouse Experiment: Planting Methods and Rooting Hormone

Shoots emerged from the cuttings within a week of planting. As expected, shoots were first observed growing from vertically planted stems, compared to horizontal buried stems. By the end of the test (14 d) the mean number of primary shoots cutting−1 was similar for all treatments except horizontally planted upper stems, which averaged fewer primary shoots (Table 1). More total shoots were counted on vertically planted cuttings than horizontal cuttings (Table 1), reflecting earlier growth of secondary shoots. Vertically planted cuttings tended to have more root growth (greater root mass and greater root length) than horizontally planted cuttings from the same portion of the cane. This is probably also a result of the fact that their shoots sprouted sooner and were thus able to produce more photosynthate for the developing roots. Across all treatments, root mass was highly correlated with shoot mass (R = 0.726). The strongest effect on overall cutting performance was the maturity of the cuttings. In general, cuttings from the lower cane (older, more mature material) were more prolific than the younger material from the upper portion of the cane. Cuttings from the lower stem had greater root mass, root length, shoot mass, and shoot length than cuttings from the upper stem (Table 1). Interactions between planting method and cutting maturity were detected. Cuttings from the lower stem planted vertically had the greatest root mass and highest number of total shoots cutting−1, while cuttings from the upper stem planted horizontally had the least root mass and root length. Only 57.5% of horizontally planted cuttings from the upper stem remained viable (Table 1). Vertically planted upper stems showed 85% viability, which was not significantly different from that of the lower stems (97.5–100%). Overall, the rooting hormone IBA did not affect rooting or shoot growth across treatments.

View Full Table | Close Full ViewTable 1.

Measurements of shoot and root components at 14 d after planting two-node stem cuttings taken from upper or lower positions on Merkeron napiergrass stalks and planted in the soil horizontally (buried) or vertically.

Cane position Planting method Root mass Shoot mass Root length Shoot length Primary shoots cutting−1 Total shoots cutting−1 Viable cuttings
g cm %
Upper horizontal 0.34d† 10.6b 16.3c 40.9b 1.6b 2.0c 57.5b
Upper vertical 1.01c 9.5b 26.4b 43.8b 1.8a 2.5b 85.0a
Lower horizontal 2.13b 27.2a 37.6a 64.8a 1.9a 2.0c 100.0a
Lower vertical 2.84a 21.8a 42.0a 61.8a 1.9a 3.1a 97.5a
Within columns, means followed by the same letter are not significantly different.

The results of the greenhouse test corroborate the results of several previous studies which have shown that the quality and maturity of the propagation material is important in vegetative establishment of napiergrass from stem cuttings. Woodard et al. (1985) observed that harvesting and planting cuttings too early in the season (4 July) resulted in fewer surviving hills per plot than later plantings because the stems were too immature. Woodard and Prine (1990) later observed that increased fertilization of napiergrass nurseries led to enhanced vigor and greater number of primary tillers in cuttings taken from those nurseries. A similar observation was reported by Sollenberger et al. (1991). Higher rates of fertilization hasten maturity of the stems, making them more amenable to vegetative propagation. Similarly, cuttings taken from the lower portion of the stem are more mature than those from the upper, younger portion, and this affects the success of propagation (Woodard et al., 1985). This trend was also observed in a related species, tender purple fountaingrass (P. setaceum ‘Rubrum’), in which a higher percentage of rooted cuttings was achieved from lower nodes than from more distal nodes (Cunliffe et al., 2001). In typical grasses most of the root system arises from adventitious rooting at the basal nodes (Jones, 1985), thus cuttings from the lower nodes would be more apt to sprout roots than distal nodes. In sugarcane cuttings with multiple nodes, the lower nodes also produce roots more readily; however, shoots tend to be produced by the upper nodes (Jones, 1985).

Previous studies have shown little effect of planting method on establishment of napiergrass (Woodard et al., 1985; Ssekabembe, 1998); however, these results indicate that planting method can have an effect on propagation success when younger material must be used, particularly early in the growing season or when planting material is limited. These results suggest that vertical planting of younger material will result in higher establishment than horizontal planting, though it is still preferable to use more mature material for propagation.

Field Experiment 1—Stem Cutting Length

A significant response to stem cutting length was observed in the three field plantings. In the earlier plantings (29 Oct. 2009 and 17 Sept. 2010) the number of primary shoots plot−1 in the fall was greatest for plots containing 10 single-node cuttings and least for plots containing one 10-node cutting (Table 2). As a percentage of cuttings producing a shoot, however, nearly 100% of the 5- and 10-node cuttings grew, while only 28.5 to 48.1% of single-node cuttings produced a shoot. A reversed trend was observed for these two plantings in the spring. The plots containing 10 single-node cuttings had significantly fewer shoots plot−1 than the other treatments. Thus only 28.6 to 51.5% of hills from single-node cuttings that grew in the fall survived to regrow in the spring. There was a small but nonsignificant decrease in surviving hills from two-node cuttings, while the 5- and 10-node cuttings generally had greater than 100% survival (Table 2), indicating that most hills survived and additional buds also sprouted after the winter. A late planting of this test was made on 15 Nov. 2010 and no fall growth was observed, so only spring shoot counts were recorded. Two-node cuttings produced more shoots plot−1 than 10-node cuttings, but other differences were not significant. However, as a percentage of cuttings producing a shoot, only 27.0% of single-node cuttings survived the winter, while 65.9, 85.2, and 96.3% of 2-, 5-, and 10-node cuttings, respectively, produced at least one shoot (Table 2). Some genotypic interactions were observed in terms of scale, but the general trends observed for emergence vs. cutting size did not vary. A previous study (Woodard et al., 1985) demonstrated a similar effect of cutting length on propagation success in napiergrass accession PI-300086. With the same number of total nodes per plot, shorter cuttings of two nodes produced more primary shoots per plot in the fall than longer cuttings, though the percentage of successful cuttings was also lower among the shorter cuttings.

View Full Table | Close Full ViewTable 2.

Average fall and spring emergence of primary shoots (hills), percentage of cuttings producing at least one shoot, and percentage of emerged hills surviving the winter for napiergrass cuttings of varying node numbers, with 10 total nodes in each plot.

Planting date Nodes cutting−1 Fall Spring Fall vs. spring Fall Spring Surviving hills
shoots plot−1 %
29 Oct. 2009 1 2.9a† 0.8b *** 28.5c 8.1c 28.6
2 2.2ab 2.0a ns‡ 43.7b 40.0b 91.5
5 1.7bc 2.4a ** 63.0a 83.3a 144.4
10 1.0c 1.8a ** 48.1ab 81.5a 175.0
17 Sept. 2010 1 4.8a 2.5b *** 48.1c 24.8c 51.5
2 3.9b 3.4a ns 69.6b 60.7b 89.4
5 3.4bc 3.9a ns 90.7a 98.1a 113.0
10 2.7c 3.8a *** 100.0a 100.0a 139.2
15 Nov. 2010 1 2.7ab 27.0c
2 3.5a 65.9b
5 2.9ab 85.2a
10 2.3b 96.3a
**Significant at the 0.01 probability level.
***Significant at the 0.001 probability level.
Within columns and planting dates, means with the same letter are not significantly different.
ns, not significant at the 0.05 probability level.

Field Experiment 2—Planting Date

Planting date had significant effects on fall emergence counts. In fall 2009 only 13.9% of cuttings emerged after the 2 November planting, and no fall emergence was observed after the latest (16 November) planting. Also in fall 2010 only 22.2% of cuttings germinated from the 29 October planting, and none from the 15 November planting emerged in the fall (Table 3). In the subsequent springs the effects of planting date were still apparent, but the trend differed between years. In spring 2010 the lowest numbers of shoots cutting−1 were recorded for the October plantings, while the 17 September and 2 November plantings had significantly more shoots cutting−1. The last planting was intermediate, and not significantly different from the other dates (Table 3). In spring 2011 earlier planting resulted in more shoots cutting−1 even though overall germination percentage did not vary to the same degree (Table 3).

View Full Table | Close Full ViewTable 3.

Average fall and spring emergence of primary shoots (hills) and percentage of cuttings producing at least one shoot for napiergrass and energycane cuttings at varying planting dates. Each cutting contained 10 nodes from the lower half of the stem.

Year Planting date Fall emergence
Spring emergence
shoots plot−1 % shoots plot−1 %
2009 17 Sept. 2.2a† 94.4a 2.1a 94.4a
1 Oct. 1.2b 72.2b 1.4b 72.2b
15 Oct 0.8bc 55.6b 1.3b 91.7ab
2 Nov. 0.3c 13.9c 2.1a 100.0a
16 Nov. 1.6ab 94.4a
2010 3 Sept. 2.4a 80.6a 4.5a 100.0a
17 Sept. 1.9a 69.4ab 3.4b 100.0a
1 Oct. 1.1b 61.1b 2.4c 97.2a
15 Oct 2.2a 80.6a 2.5c 100.0a
29 Oct. 0.4c 22.2c 1.6 d 88.9ab
15 Nov. 1.5 d 75.0b
Within columns and years, means with the same letter are not significantly different.

In northern and central Florida, Woodard et al. (1985) concluded that earlier planting allows plants of PI-300086 napiergrass to fully establish before winter, while very late planting allows the plant to use reserves stored in the original cane cutting for subsequent growth the following spring. Intermediate planting dates were found to result in greater stand losses over the winter, likely due to a lack of stored reserves in the young roots and rhizomes. In this experiment, a similar pattern was observed in the 2009–2010 season. However in the following year early plantings appeared to have an advantage over later ones. The average soil temperature in December 2010 was 2.3°C colder than the same month in 2009 and less rain was received (Fig. 1). Perhaps the earlier onset of winter along with drier conditions in 2010 led to reduced vigor in the later plantings compared to the previous year. It appears that in colder locations, early planting is highly advantageous for establishment of both napiergrass and energycane, while very late plantings are more risky. Some genotypic differences in emergence were observed. For both fall and spring shoot counts, napiergrass Genotypes 04-1-62 and 07-219 were consistently lower than the others, while Merkeron and 07-83 showed consistently high spring shoot counts (Table 4). The differences measured in field propagation success of cuttings from different napiergrass genotypes clearly demonstrate that there is significant genetic variation among napiergrass for the ability to propagate from cuttings. An interaction of genotype and year was observed for the energycanes. In fall 2009 emergence counts (shoots cutting−1) for both energycanes were similar to most of the napiergrasses, but in fall 2010 the energycanes had very low fall emergence. However, in spring 2011 emergence of the energycanes was again similar to that of the better napiergrasses (Table 4).

View Full Table | Close Full ViewTable 4.

Fall and spring emergence of seven napiergrass genotypes and two energycane genotypes, averaged across planting dates. Fall emergence means do not include the latest plantings (16 Nov. 2009 and 15 Nov. 2010), for which no emergence was observed until the following spring.

Genotype Fall Spring Fall Spring
shoots cutting−1
Merkeron 1.8a† 1.9abcd 2.5b 3.2ab
04-1-62 0.8b 1.2cd 1.2cd 2.2bc
07-110 1.1ab 1.6abcd 2.2bc 2.8ab
07-219 0.4b 1.1d 0.5de 1.3c
07-239 1.3ab 1.5bcd 3.7a 3.8a
07-353 0.9ab 1.5bcd 2.2bc 2.8ab
07-83 1.3ab 2.0abc 2.0bc 3.0ab
Ho-02-144‡ 1.1ab 2.3ab 0.1 e 2.3b
L79-1002‡ 1.3ab 2.4a 0.4 de 2.5b
Means within columns with the same letter are not significantly different.


Based on these results, it is recommended that napiergrass and energycane cuttings be planted no later than mid-September in the Georgia Coastal Plain, or about 90 d before the first freeze. Earlier planting is advantageous, provided that the propagation material is sufficiently mature. Napiergrass cuttings should be 2 to 10 nodes in length. Single-node cuttings can also be used but they may need to be planted more densely than larger cuttings. This is important as smaller billets may be more amenable to mechanized planting than longer stem pieces. Maturity of the cutting material is critically important for napiergrass propagation; the upper half of the stem should be avoided if possible. Though horizontal burial of the cuttings may be the most practical approach, vertical planting may be used if irrigation is available. If young material must be used, it should be planted vertically. Under greenhouse conditions, rooting hormone treatment with IBA did not significantly improve propagation success in napiergrass.


The authors would like to thank Freddie Cheek and Tony Howell for assistance with this project in the field. We would also like to thank Dr. Brian Scully and Prof. R. Dewey Lee for critical reviews of this manuscript.




Be the first to comment.

Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.