Livestock Research for Rural Development 22 (5) 2010 Notes to Authors LRRD Newsletter

Citation of this paper

Coral vine consumption by St. Croix White hair lambs

S A Weiss, J P Muir* and R W Godfrey

University of the Virgin Islands, Agriculture Experiment Station,
RR 1 Box 10,000 Kingshill 00850, St. Croix U.S.V.I. USA
* Texas AgriLife Research, Texas A&M System,
1229 North U.S. Highway 281 Stephenville TX 76401 USA   ;   ;


Coral vine (COV; Antigonon leptopus), an invasive ornamental vine found throughout the tropics, may be a useful forage for ruminants and may be suppressed by small ruminants where it is not wanted. A pen trial supplemented 20 and 33% of a total mixed ration (TMR) with COV (4% TMR and 1% COV of body weight (BW); 4% TMR and 2% COV of BW, respectively) and compared these diets with a control that supplemented 25% of the TMR with guineagrass (Panicum maximum) fed to Saint Croix White hair lambs (4% TMR and 1% grass of BW).


The COV forage had an average 160 g/kg crude protein, 389 g/kg neutral detergent fiber, and 174 g/kg total condensed tannins. Four-month-old lambs had lower (P < 0.05) COV refusal rates and lower average daily gains (ADG) than animals two months older that rejected approximately 10% of the COV when fed at 20% and 50% when fed at 33% of the diet.  Six-month-old lambs fed COV at 20% of the diet had 141 g ADG compared to 179 g for those fed the control diet (P < 0.05).


This study shows that Saint Croix White hair lambs will consume COV although they benefit from early habituation to this forage which lowers weight gains compared to low quality guineagrass.

Keywords: Antigonon leptopus, average daily gain, guineagrass, Panicum maximum


Coral vine (COV; Antigonon leptopus) is a showy ornamental vine originating in Mexico that is now widely cultivated throughout the humid tropics (Jacob et al 2001; Plantoftheweek 2010) with documented medicinal (Lans 2006; Chistokhodova et al 2002) and apicultural (Abrol 2003) applications. It has, however, escaped to become a pan-tropical invasive weed that aggressively occupies abandoned lots, forests and fields (Tassin and Riviere 1999; NRCS USDA 2010). Control using herbicides is difficult and largely ineffective (Harkness and Byrd 1971) such that use of small ruminants to selectively suppress this vine is proposed. Its potential as a forage for goats and sheep is untested and its palatability to small ruminants used to control vegetation is also unknown.


A preliminary experiment conducted on private land on St. Croix, U.S. Virgin Islands indicated that St Croix White hair sheep readily consumed COV when it was the primary plant available in confined areas. However, regrowth was not as readily consumed by sheep, indicating possible low palatability.  Effect on weight gains was also not measured. St. Croix White hair sheep are adapted to warm, humid climates and research indicates good weight gains when fed tropical forages under tropical conditions (Godfrey and Weis 2005) and similar weight gains were reported under  temperate conditions (Wildeus et al 2005). This study, however, took place in Virginia USA, not under tropical conditions.


The use of St. Croix White hair sheep to control invading COV is a proposed biological management tool if it can be determined that lambs fed this perennial vine can thrive. In order to determine if COV fed to confined St. Croix White hair lambs was palatable, toxic or resulted in weight gains, a cut-and-carry feedlot experiment was designed. Fresh cut vines were substituted for a diet of guineagrass (Panicum maximum), a perennial grass that is widely used throughout the tropics as a forage (Muir and Jank 2004) and is readily consumed by St. Croix White hair sheep on the Island of St. Croix. Our objective was to measure laboratory nutritive values, refusal rates and average daily gains (ADG) of St. Croix White hair sheep fed two levels of COV in confinement.


Materials and methods 

St. Croix, U.S. Virgin Islands lies in the eastern Caribbean at 17 43’ N latitude and 64 48’ W longitude and is 218 square kilometers.  It is characterized as a tropical/subtropical environment with a bimodal rainfall climate with 20-yr annual rainfall averages of 1100 mm and mean monthly temperatures ranges from 22C to 32.8C. The annual mean temperature on St. Croix is 28.3C with a mean high temperature of 32.8C (Godfrey and Hansen 1996). The average annual relative humidity is 78% and average monthly relative humidity ranges from 74% in March to 81% in September. The soil of St. Croix is primarily mildly alkaline Fredensborg clay (fine carbonatic, isohyperthermic, Typic Rendolls, Mollisol) which is of relatively high inherent fertility compared to the soil found on most other Caribbean nations (Valencia et. al., 2001).


At the University of the U.S. Virgin Islands Agricultural Experiment Station, Saint Croix White hair lambs were tested to determine the effect of feeding fresh cut COV on animal performance and forage refusals.  The trial was carried out from June 6 through July 26, 2006 and was repeated February 4 through April 13, 2009. Feeding trials examined forage nutritive value, refusal rates, and ADG.  Data included observations on grazing behavior and consumption patterns in penned lambs using a cut-and-carry feeding system. 


Nine feed pens were designed to encompass three diets replicated three times (pens) fed to 36 lambs.  Experimental units consisted of 4 lambs per pen approximately 4-5 months of age in 2006 and 7-8 months of age in 2009. Pens consisted of mixed sex groups in 2006, but were separated into male and female groups in 2009 due to sexual maturity of intact male lambs.The forage ration was comprised of either fresh cut COV (Photos 1 and 2) or guineagrass hay and the concentrate ration was a pelleted total mixed ration (TMR) with 16% crude protein (CP). Treatments were as follows (percentages of lamb BW, dry matter (DM) basis):

1.                  Control TMR fed at 4% and guineagrass hay fed at 1%

2.                  4% TMR and 1% COV

3.                  4% TMR and 2% COV


Coral vine was harvested fresh daily in a cut-and-carry system from invaded sites along road sides, fence lines, and in over-grown fields. All forage feed rates were fed on a DM basis and feed values were calculated from live weights taken every 14 days. Prior to the beginning of the feeding trial, lambs were transitioned from native pasture to their respective ration for 14 days. Daily feed refusals were measured to determine percent rejected and to compare nutritive value of rejected material with that of fed forage.  Guinea grass hay, fresh COV, and COV refusals were sub-sampled for subsequent nutritive value evaluation.


Photo 1. Coral vine foliage offered to the lambs Photo 2.  Coral vine climbing on Leucaena leucocephala trees


Each year representative herbage sub-samples from material batched from each pen were dried in forced-air ovens at 55o C, ground through a 1-mm screen and analyzed at the Texas AgriLife Research, Stephenville Texas USA herbage lab for neutral detergent acid (NDF), acid detergent fiber (ADF), condensed tannin (CT) fractions, and CP concentrations. Fiber concentrations were determined according to Van Soest and Robertson (1980) using an Ankom 200 Fiber Analyzer (Ankom Technologies, Macedon NY). Nitrogen was estimated using an Elementar Vario Macro C-N Analyzer (Mt. Laurel, NJ).  Extractable (E), fiber-bound (FB), and protein-bound (PB) CT concentrations were determined by methods described by Terrill et al (1992) using CT from each entry to establish that entry’s standard curve (Wolfe et al 2008). 


Four lambs in each pen were considered one experimental unit.  No differences were measured among pens (replications = 3) so sex did not affect results. Year (n = 2) and diet (n = 3) were independent variables in the model; ADG and refusal rates, averaged over the entire experiment, were considered dependent variables.  These were considered significant at P ≤ 0.05, and multiple mean separations were conducted where appropriate at the same level of probability using a least significant difference (LSD).  All herbage component concentrations are presented as means +/- standard deviations because multiple, independent samples were taken. 


Results and discussion 

Acceptability and toxicity


Lambs readily accepted coral vine forage after a few days hesitation during the adaptation period, with the older animals in 2009 taking longer to adapt. No signs of toxicity were observed during the trials although sub-clinical and long-term effects were not measured.


Crude protein values were identical in the fed COV and the TMR while fiber values were low and CP high in the COV material compared to the guineagrass (Table 1).

Table 1.  Composition of concentrates and forages fed to St. Croix White hair lambs.

Diet component,  g/kg







Total CT

Total mixed ration
















Coral vine








Coral vine refusals








*CP, crude protein; NDF, neutral detergent fiber; ADF, acid detergent fiber; E, extractible; CT, condensed tannins; PB, protein-bound; FB, fiber-bound.

Total CT concentration was not measureable in the grass but was apparent in the COV, with E CT comprising 50% of the total.  These values compare to those measured in Sericea lespedeza, a legume that has been documented to suppress gastro-intestinal nematodes in small ruminants (Terrill et al 2007) as well as increase rumen-bypass (escape) protein in ruminants (Waghorn 2008). Protein-bound CT was particularly high relative to total CT compared to most plants with CT which have much larger proportions of E CT compared to PB CT (Muir et al 2009).




Despite the fact that COV CP values were well above those of the grass (Table 1), there were no measureable refusals of guineagrass compared to values from 4 to 50% for COV (Table 2) indicating poor palatability of the vine relative to guineagrass.

Table 2.   Average daily refusal of coral vine by Saint Croix White hair sheep fed at 1 or 2% of body weight, dry matter basis (year X diet interaction P = 0.001) compared to a control fed guineagrass at 1% of body weight.  All animals were fed a total mixed ration at 4% body weight on a dry matter basis.


% Refusals





0 a B

0 a C


1% Coral Vine

3.8 b B

10.1 a B


2% Coral Vine

10.9 b A

54.4 a A


Refusal rates within the same line followed by lower case letters and within the same column followed by upper case letters differ according to an LSD mean separation at P ≤ 0.05; 36% coefficient of variation.

Although leaf:stem ratios were not measured, visual observations indicated that lambs selected COV leaves over stems. A 22% decrease in CP, 32% decrease in T CT and 24% increase in fiber concentration of the COV refusals compared to fed material support this conclusion since leaves tend to have greater nutritive values than those of stems (Muir et al 2008).


Younger lambs in 2006 had lower refusal rates, 65% lower at the 1% COV rate and 80% lower at the 2% COV rate, compared to older lambs in 2009 (Table 2). This would indicate that lambs used to control invasive COV should be exposed to the plant as early in life as possible. It also indicates that lower amounts of COV or greater grazing pressure in pasture situations will result in greater use efficiency of this invasive plant.


Average daily gains


The differences in ADG patterns were similar between years (Table 2).  Weight gains in the control diet averaged 184 g/day, 114% greater than ADG reported for St. Croix White hair lambs raised on pastures with or without supplements (Wildeus et al 2005) and 70% greater than lambs pen-fed concentrate diets without forage supplement (Godfrey and Weis 2005). These differences may have been a result of a synergistic effect of guineagrass forage with the TMR.


Despite greater CP and lower fiber concentrations in the COV compared to guineagrass (Table 1), replacing the first with the latter at the 1% BW rate resulted in 56% lower ADG with younger animals (4 months of age) and 21% lower ADG with older animals (6 months of age) (Table 3).  

Table 3. Average daily gain (ADG) of Saint Croix White hair sheep supplemented fed coral vine at 1 or 2% of body weight (year X diet interaction P = 0.001) compared to a control fed guineagrass at 1% of body weight.  All animals were fed a total mixed ration at 4% body weight on a dry matter basis.










1% Coral Vine

103b B

138b A


2% Coral Vine

79c B

141b A


Values in the same column followed by different lower case letters and in the same line followed by different upper case letters differ (P < 0.05) according to an LSD multiple range separation; 12% coefficient of variance.

Replacing an additional 1% BW (2% total BW) of the TMR with COV in younger animals resulted in an additional 23% decrease in ADG in younger lambs but had no measureable weight change in lambs 2 months older.  The improved performance of older lambs on COV relative to younger animals was likely a result of lower metabolic rates in older animals (NRC 2007) in combination with lower COV consumption (Table 2) relative to the TMR feed.




Abrol D P 2003 Antigonon leptopus Bee World 84:130-131


Chistokhodova N, Nguyen C, Kachirskaia I, Cunningham G and Miles D H 2002 Antithrombin activity of medicinal plants from central Florida. Journal of Ethnopharmacology 81:277-280


Godfrey R W and Hansen P J 1996 Reproduction and milk yield of Holstein cows in the US Virgin Islands as influenced by time of year and coat color. Archivos Latinoamericanos de Produccion Animal 4:31-44


Godfrey R W and Weis A J 2005 Post-weaning growth and carcass traits of St. Croix White and Dorper X St. Croix White lambs fed a concentrate diet in the U.S. Virgin Islands. Sheep and Goat Research Journal 20:32-36


Harkness R W and Byrd C D 1971 Weed control in lime, avocado and mango groves. Proceedings of the Florida State Horticultural Society 84:285-290


Jacob A, Raju A J S, Raju V K, Victor P and Naidu S A 2001 Floral ecology, breeding system and pollination in Antigonon leptopus L. Plant Species Biology 16:159-164


Lans C A 2006 Ethnomedicines used in Trinidad and Tobago for urinary problems and diabetes mellitus. Journal of Ethnobiology and Ethnomedicine 45:313-318


Muir J P, Butler T J, Wolfe R and Bow J R 2008 Harvest techniques change annual warm-season legume forage yield and nutritive value. Agronomy Journal 100:765-770


Muir J P and Jank L 2004 Guineagrass (Panicum maximum Jacq). In Moser L, B Burson and L Sollenberger (editors) Warm-Season Grasses.  ASA Monograph 45. ASA, Madison, WS, USA


Muir J P, Terrill T, Valencia E, Weiss S, Jones P D, Mosjidis J and Wolfe R 2009 The wide range of condensed tannins in Caribbean Basin plants and their applicability to ruminant production systems. Proceedings Caribbean Food Crops Society 43rd Annual Meeting, St. Kitts and Nevis


NRC 2007 Nutrient requirements of small ruminants. National Research Council, Washington. DC, USA


NRCS USDA 2010 Plants profile. USDA


Plantoftheweek 2010 Coral vine.


Tassin J and Riviere J N 1999 Invasive plants of Reunion. Courrier de la Nature 177:28-33


Terrill T H, Mosjidis J A, Moore D A, Shaik S A, Miller J E, Burke J M, Muir J P and Wolfe R   2007  Effect of pelleting on efficacy of sericea lespedeza hay as a natural dewormer in goats.  Veterinary Parasitology 146:117-122


Terrill T H,  Rowan A M, Douglas G B and Barry T N 1992  Determination of extractable and bound condensed tannin concentrations in forage plants, protein concentrate meals, and cereal grains. Journal of Science of Food and Agriculture 58:321-329


Valencia E, Adjei M and Martin J 2001 Aquaculture effluent as a water and nutrient source for hay production in the seasonally dry tropics. Communications in Soil Science and Plant Analysis 32:1293-1301


Van Soest P J and Robertson J B 1980 Systems of analysis for evaluating fibrous feeds. p. 49-60.  In W J Pigden et al (editors) Standardization of Analytical Methodology for Feeds: Proceedings of an International Workshop, Ottawa, ON. 12-14 Mar. 1979. Rep. IDRC-134e. International Devevelopment Research Center, Ottawa, ON, Canada, and Unipub, New York USA


Waghorn G 2008 Beneficial and detrimental effects of dietary condensed tannins for sustainable sheep and goat production—Progress and challenges. Animal Feed Science and Technology 147:116-139


Wildeus S, Turner K E and Collins J R 2005 Growth performance of Barbados Blackbelly, Katahdin and St. Croix hair sheep lambs fed pasture- or hay-based diets.  Sheep and Goat Research Journal 20:37-41


Wolfe R M, Terrill T H and Muir J P 2008 Drying method and origin of standard affect condensed tannin (CT) concentrations in perennial herbaceous legumes using simplified butanol-HCL CT analysis. Journal of Science of Food and Agriculture 88:1060-1067

Received 23 March 2010; Accepted 30 March 2010; Published 1 May 2010

Go to top