The publications in this collection do
not reflect current scientific knowledge
or recommendations. These texts
represent the historic publishing
record of the Institute for Food and
Agricultural Sciences and should be
used only to trace the historic work of
the Institute and its staff. Current IFAS
research may be found on the
Electronic Data Information Source
site maintained by the Florida
Cooperative Extension Service.
Copyright 2005, Board of Trustees, University
Fort Pierce ARC Research Report LP-1971-2
PERSISTENCE OF TROPICAL LEGIfxU.S PME. i' ETrr PASTURES
/ Albert E. Kretschmer, Jr.1/
Persistence of high populations of tropical legumes in permanent
pastures is of utmost importance if quality forage and significant
contributions of nitrogen to the pasture system are realized. Although
higher grass pasture yields can be obtained with high applications of
nitrogen fertilizer compared with legume-based pastures, the economics
do not always favor the use of nitrogen fertilizers. As a general rule,
the greater percentage of legume in a mixed pasture, the greater will be
the production of forage and beef per area. Optimum average content of
legume on a dry matter basis is probably between 25 and 50,..
Many factors are responsible for persistence of tropical legumes in
mixture with permanent grass pastures used for grazing. Several of the
important ones will be discussed in the following paragraphs.
EFFECT OF FERTILIZERS
There has been ample evidence to indicate the depressing effect of
continuous nitrogen fertilizer applications on tropical legume populations
in legume-grass pastures (6, 9, 22). I4ost of these results did not take
into consideration the effect of nitrogen fertilizers on legume persistence
when the nitrogen was applied infrequently for such purposes as increasing
pasture growth during or just prior to stress periods. Experimental
results at the ARC, Ft. Pierce (9) showed that an early fall nitrogen
application of about 50 !g/ha increased total yield of Siratro (Phaseolus
atropurpureus = l acroptilium atropurpureum)-Pangola (Digitaria deciubens)
mixtures but did n-ot effect Siratro plant populations during the following
summer. In the summer, the continued use of K severely reduced the
population of Siratro.
It is believed that perennial legumes would be able to compete with
grasses on soils higher in nitrogen than annual legumes. Populations of
both, however, would probably be less as soil nitrogen levels increase
(if other factors remained stable).
Many tropical soils are deficient in phosphorus. It is difficult to
generalize on the effect on legume persistence of phosphorus fertilization
to phosphorus deficient soils. If the applications benefit the grass more
than the legume it would certainly result in greater grass competition and
less legume persistence; and even if this should lead to greater dry matter
production, it probably would not outweigh the benefits of large legume
percentages. It has been shown that Guineagrass (Panicum maxim~n) re-
sponded more to phosphorus applications than Stylo (Stylosanthes guyanensis)
17 Agronomiiit, University of loridaE, Institute of Food and Agricultural
Sciences, Agricultural Research Center (ARC), P. O. Box 248, Fort
and Centro (Centrosema pubescens) (19); but in similar tests, Stylo
responded more than Paragrass (Brachiaria mutica). i;any tropical
legumes can survive on lot. levels of soil phosphorus which could
result in dominance of the legume over the accompanying grass. Some
soils, however, are so lou in available phosphorus and contain toxic
levels of aluminum and or manganese that many of the legumes cannot
survive. Under these conditions, adequate pasture production will
occur only by correcting the deficiency.
Species differences should also be considered. In one experiment
(1) Lotononis (Lotononis bainesii) responded to applications up to
about 10 Ik/ha of phosphorus while Stylo, Greenleaf desmodium (Desmodium
intortum) and Siratro yields increased up to a rate of 80 kg/ha.
Of the trace elements, legume responses to saall applications of
molybdenum have been outstanding in Australia. Little response to
this or other trace elements have been reported in the Western hemi-
shpere tropics. In Florida normal fall annual applications of fertilizer
for white clover production (300 to 500 1 /ha of 0-10-20 as N-P205 -
K20 with trace quantities of copper, manganese, zinc and boron) has
been adequate for insuring summer growth of tropical legumes. There
are many areas in the Latin American tropics where little or no response
to fertilizers would occur. Usually there is enough data on soil
nutrient status to ma-.e an informed guess of the type of fertilizer
CUTTING INTERVAL AMD CUTTIlG HEIGHT
Little information is available on the effect of grazing intensities
on persistence, but enough information has been published on cutting
interval (CI) and cutting height (CH) to speculate about grazing manage-
In Australia (7) Greenleaf desmodiua was found to persist well when
cut every four wee -s during the summer months. Increased yields occurred
as the CII vas increased from 3.3 to 7.5 cm and CI increased from 4 to 12
veeks. Lumber of plants per square meter at the end of the three year
test period at the four week CI 'were 7.2, 7.9, and S.5 when Ci! were 3.0,
7.5, and 15 cm, respectively. In a CI-Cl! test in Hawaii (22), 45 to
55j Greenleaf desmodiumi was maintained in ilkuyugrass (Pennisetum
clandestinum) and 40 to 58" in Pangola. Persistence was not affected by
CI of 5 or 10 veeks or CI of 5 and 13 cm. Also, in Australia (14)
Greenleaf desmodium in pure stand was not affected by cutting intervals
of 3 or 5 weeks or by cutting heights of 7.5 or 15 cm.
Close (5 cri) and frequent clipping (3 weeks) of perennial Glycine
(Glycine vightii = G. javanica) was detrimental to yields compared with
a CII of 15 cn and CI of6 or 12 weeks (15). Siratro production and
persistence also was found to be adversely affected by frequent CI (6).
Increasing the CI from 4 to 16 ueeks resulted in a linear increase in
yields and increased persistence.
Even within species differences are encountered. Two upright
and one decumbent Stylo cultivars were subjected to CI of 6, 12, and
18 weeks (5). The decumbent type survived more frequent cuttingjthan
did the Schofield or Endeavour" upright types. In Florida (10)
Greenleaf desmodium, 'esiodium heterocarpon, Siratro, and Townsville
stylo (Stylosanthes huilis) mi ed with Pangola, iandi setaria
(Setari aces~ -o Peinsacola bahiagrass (Paspalw~ notatum) were clipped
at six weei intervals from iay to December. Persistence of all vas
good after three years.
From the standpoint of grazing tropical legumes in mixtures with
grasses, a modifying influence is the selectivity of grazing. Trailing
legumes, i.e., Siratro, Glycine, Centro, tropical '-udzu, etc., are less
selectively eaten by cattle than many accompanying grasses. Other
modifying factors are growth habit and light. The combination of the
shade sensitivity and prostrate growth habit of Townsville stylo and
Lotononis, for example, mace it difficult for them to compete with tall
tropical grasses while t ing legumesa_~ahle to ascend the grass
to obtain- sunlight.
In one experiment (21) mixtures of legumes with Rhodesgrass
(Chloris gayana) were grazed at intervals of si:x weeks in the summer
and nine weels in the winter and cut bac; to 5 cm as sheep were moved
to another pasture. At the end of two years, Glycine, Siratro, and
Silverleaf desmodium (Desmodium uneinatum) survival was generally noor
with Glycine being somewhat superior. The legumes, however, were left
in a completely defoliated condition after each grazing. Plots not
cut during the two year period persisted well and eventually suppressed
the Rhodesgrass. These results are different from those reported for
moderate grazing by cattle, where leaves of these legumes normally
would be present at the end of the grazing cycle. Greenleaf and
Silverleaf desmodium were favored in mixtures with Mhodesgrass (13)
when continuously grazed by one beef animal per 0.3 ha compared with
one/O.4 ha. Constant, heavy grazing pressure is needed for Lotononis-
grass (4) or 'lounsville stylo-grass mixtures to permit penetration of
light. These legumes have growth habits that are adaptable to heavy
Survival and persistence of the annual, Town svllr-~tyOrin-th----
Northern Territory and northern Queensland, Australia have been good
under commercial grazing conditions. Tn Irrge part, the grass components
are native and are not as competitive as improved grasses. Survival
of the perennials, Centro, tropical Kudzu, Siratro, Glycine, and Stylo
in mixtures with improved grasses also has been good in many areas of
In Florida couoercial plantings, persistence of the annuals,
Aeschynomene americana, Townsville stylo, Hairy indigo (Indigofera
hirsute), and Alyce clover (Alysicarus vaginalis) has been variable.
The reasons for non-persistence are not fully understood but are
undoubtedly not a result of a single factor. Only small areas have
been planted to Siratro but its persistence appears to be fair to
good. Preliminary results indicate that persistence and adequate
plant populations can be maintained if there is rest period during
the summer after the normal heavy winter grazing period.
TEiPERATURES AND SOIL MOISTURE
Low temperatures can affect tropical legume growth. Iiost will not
persist as perennials where minimum temperatures frequently fall below
0 C. It is estimated that persistence of Siratro in Florida would be
poor when monthly minimum air temperatures fall below about 10C for more
than about two months. Under these conditions numerous frosts or freezing
days would be expected. In sub tropical areas where only a few frosts
are experienced, cold minimum temperatures can slow legume growth. Fortu-
nately for those living in the subtropics, growth of most of the grasses
being used is slowed by minimum air temperatures below about 150C. It
has been found (13) that growth of Greenleaf desmodium and Pangola slows
when temperatures fall below about 150c. Lonononis spring growth does
not commence until weekly minimum mean air temperatures reach 9C (4).
Silverleaf desmodium commenced spring growth when the weekly minimum air
temperatures exceeded about 9-100 (22). Optimum air temperatures in a
controlled greenhouse experiment (20) for Greenleaf and Silverleaf
desmodiums, Phasey bean (Phaseolus lathyr ides), Siratro, Townsville
stylo, and Clycine were fund to be 30/25-C (day/night). Dry matter
production was slowed when temperatures were 27/220C. The optimum
temperature for tropical legume growth was lower than that for tropical
grasses but higher than those for temperate grasses and legumes. Above
a 33/280C temperature regime the growth rate of Desmodium species declined
more than that for Siratro.
At the ARC, Ft. Pierce, production of D. heterocarpon, Siratp and
Greenleaf desmodium was found to be much greater than that for Pangola
or Pensacola bahiagrass during Ilarch through jay. Long-term mean
monthly day/night air temperatures for these respective months were
26/13, 28/16, and 30/1Oc (10, 12).
host of the trailing perennial legumes require a minimum of 750 to
1000 mm of rainfall annually. Siratro is more drought tolerant than the
Desmodiums, Centro and Glycine. Tolerance to high water tables generally
is good with Centro, Aeschynomene species, tropical Kudzu, and some Stylo
species. Almost all can survive periodic inundations of several days
duration. The tolerance to flooding is believed to increase with in-
creases in the sand content of the soil.
DISEASES AID II.SECTS
Damage to leaves of Centro, Siratro, and Desmodium species by the
bean leaf-roll caterpillar (Urbanus protus L.) has been severe at times
in south Florida. lo permanent damage to the legumes has resulted,
however. Other insects such as the Amnemus weevil that attack roots
and foliage of the Desmodiums and Glycne (2), nematodes, and insects
that reduce viable seed production may be important in the persistence
of the tropical legumes.
Aerial Rhizoctonia and rust (Uromyces phaseoli) damage to Siratro
and other legumes at the ARC, Ft. Pierce (9, 17) and-elsewhere has
been severe. Colletotrichum leaf spot and stem-canker disease of
Stylosanthes species (11, 16, 18) can be severe, as can legume little-
leaf virus. Fortunately, speciese- sully can be found that are resistant
to a particular insect or disease.
In certain instances insects or diseases may attack the grass component
of the mixture. The yellow sugarcane aphid (Sipha flava) and rust (Puccinia
oahuensis) on Pangola are examples. These attacks undoubtedly reduce grass
competition temporarily, and may help legume persistence.
ASSOCIATE GRASS VARIETY
Little work has been published on the effect of grass variety on
competition with tropical legumes. It is obvious that grazing selectivity,
growth patterns of above- and below-ground plant parts, responses to
fertilizers, resistance to drought and flooding, environmental factors,
etc., all enter into the ability of a legume to compete with a grass.
Selection of a legume that is compatible with a given grass may not be
too difficult; but the management of the established mixture may require
considerable attention. In the selection process it is easier to discard
the use of a particular legume than it is to select the best one. For
example, it would not be wise to plant Townsville stylo in an Elephant-
grass (Pennesitum purpureum) or Guineagrass pasture to be used for green-
chopping. Shading would prohibit successful growth of the legume. In
most areas Townsville stylo is not grown with Pangola because of competition.
In south Florida (8), however, this mixture has proved to be satisfactory.
Neither would Greenleaf or Silverleaf desmodium be selected in areas having
3 or 4 months without rainfall with low water tables. Most of the trailing
legumes have persisted well at least experimentally as mixtures with
Setaria (Setaria anceps), Pangola, Guineagrass, Jaraguagrass (Hyparrhenia
rufa), Bahiagrass and Rhodesgrass. Shading by the grass, even with trail-
ing legumes, may be more important than root competition. Under grazing
conditions, the spread between the grass's and legume's "Palatability"
(grazing selectivity) would be expected to exert an influence on legume
persistence. At low stocking rates, the wider the spread in favor of
the animal's preference to grazing the grass compared to the legume, the
less competition the grass would provide. As stocking rates increase,
however, the less effect selectivity would have on control of legume
persistence and the greater -effect-def olation-effect-would have on
Although there is limited information on persistence of tropical
legumes under the varied environmental conditions encountered in the
Latin American tropics and subtropics, there are several legumes that
have been sufficiently persistent to warrant preliminary use. Siratro
in drier areas and Centro and tropical Kudzu in wetter areas have been
competitive with a variety of grasses. Greenleaf and Silverleaf desmodiums
and the perennial Stylos appear to have narrower ranges of adaptability.
There is merit to the suggestion of some that mixtures of legumes with
the same (or possibly even different) growth habits should be seeded to the
same grass pasture. Unknown variations in the environment can be com-
pensated for by this method. This is particularly true until more
information is known of the effects of the local environmental conditions
and grazing management practices on legume persistence.
1. Blunt, C. G. and L. R. Humphreys. 1970. Phosphate response of mixed
swards at Mt. Cotton, south-eastern Queensland. Aust. J. Exp. Agr.
2. Braithwaite, B. M., and J. R. Rand. 1970. The pest status of Amnemus
spp. in tropical legume pastures in north coastal New South Wales.
Proc. XI Int. Grassld. Congr. 676-681.
3. Bryan, W. W. 1970. Changes in botanical composition in some sub-
tropical sown pastures. Proc. XI Int. Grassld. Congr. 636-639.
4. J. P. Sharpe and K. P. Haydock. 1971. Some factors
affecting the growth of lotononis (Lotononis bainesii). Aust. J.
Exp. Agr. Anim. Husb. 11:29-34.
5. Grof, B., Harding, W. A. T. and R. F. Woolcock. 1970. Effects of
cutting on three ecotypes of Stylosanthes guyanensis. Proc. XI Int.
Grassld. Congr. 226-230.
6. Jones, R. J. 1967. Effects of close cutting and nitrogen fertilizer
on growth of a Siratro (Phaseolus atropurpureus) at Samford, south-
eastern Queensland. Aust. J. Exp. Agr. Anim. Husb. 7:157-161.
7. 1973. The effect of frequency and severity
of cutting on yield and persistence of Desmodium intortum cv.
Greenleaf in a subtropical environment. Aust. J. Exp. Agr. Anim.
8. Kretschmer, Albert E. Jr. 1968. Stylosanthes humilis a summer-
growing, self-regenerating, annual legume for use in Florida pastures.
Florida Agr. Exp. Sta. Circ. S-184.
9. 1972. Siratro (Phaseolus atropurpureus DC.)
a summer-growing perennial pasture legume for central and south
Florida. Florida Agr. Exp. Sta. Circ. S-214.
10. J. B. Brolmann, G. H. Snyder and G. J.
Gascho. 1973. Production of six tropical legumes each in combination
with three tropical grasses in Florida. Agron. J. 65:890-892.
11. Ronald M. Sonoda, and John B. Brolmann.
1973. Morphologic, agronomic and disease susceptibility differences
among Stylosanthes humilis accessions in south Florida. Submitted
to Florida Soil Crop Sci. Soc. Proc.
12. G. H. Snyder, J. B. Brolmann, and G. J.
Gascho. 1974. Seasonal production of dry matter and crude protein
in tropical legume-grass mixtures in south Florida. Submitted for
publication in Proc. XII Int. Grass. Cong.
13. Plucknett, D. L. 1970. Productivity of tropical pastures in Hawaii.
Proc. XI Int. Grassld. Congr. 38-49.
14. Riveros, Fernando, and G. L. Wilson. 1970. Responses of a Setaria
sphacelata-Desmodium intortum mixture to height and frequency of
cutting. Proc. XI Int. Grassld. Congr. 666-668.
15. Santhirasegaram, K., J. E. Coaldrake and M. H. M. Salih. 1966. Yield
of a mixed subtropical pasture in relation to frequency and height of
cutting and leaf-area index. Proc. X Int. Grassld. Cong. 125-129.
16. Sonoda, Ronald M. 1973. Incidence of Colletotrichum leaf spot and
stem canker on introductions and selections of Stylosanthes humilis.
Plant Disease Rpt. 57 (9):747-749.
17. A. E. Kretschmer, Jr., and J. B. Brolmann. 1971.
Web-blight of introduced forage legumes in Florida. Trop. Grass. 5:
18. A. E. Kretschmer, Jr., and J. B. Brolmann. 1973.
Colletotrichum leaf spot and stem canker of Stylosanthes spp. in
Florida. Trop. Agr. (Trinidad) In press.
19. Teitzel, J. K. and R. C. Bruce. 1971. Fertility studies of pasture
soils in the wet tropical coast of Queensland. 2. Granitic soils.
Aust. J. Exp. Agr. Anim. Husb. 11:77-84.
20. Whiteman, P. C. 1968. The effects of temperature on the vegetative
growth of six tropical legume pastures. Aust. J. Exp. Agr. Anim.
21. 1969. The effects of close grazing and
cutting on the yield, persistence and nitrogen content of four
tropical legumes with Rhodes grass at Samford, south-eastern Queen-
sland. Aust. J. of Exp. Agr. Anim. Husb. 9:287-294.
22. and Ann Lulham. 1970. Seasonal changes
in growth and nodulation of perennial tropical pasture legumes in
the field. I. The influence of planting date and grazing and
cutting on Desmodium uncinatum and Phaseolus atropurpureus. Aust.
J. Agr. Res. 21:195-206.
23. Whitney, A. Sheldon. 1970. Effects of harvesting interval, height
of cut, and nitrogen fertilization on the performance of Desmodium
intortum mixtures in Hawaii. Proc. XI Int. Grassld. Congr. 632-636.