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 RL-1974-7 June 1974
MORPHOLOGIC, AGRONOMIC AND DISEASE SUSCEPTIBILITY DIFFER CES
AMONG STYLOSANTHES HUMIL-I ACCESSIONS IN SOUTH FLORIDA.
Albert E. Kretschmer, Jr., Ronald M. Sonoda,. nd John B. Brolmann2/
Stylosanthes humilis HBK., an annual summer-growing legime, is being
used for grazing in south Florida pastures. Commercial seed s'"uices are from
Australia, and generally, genetic differences.within seed sources of unnamed
cultivars are large. This results in variations -in initiarlflowerig, and in
morphological characteristics in commercial pastures. Many' 6otype. fail to
flower in south, and particularly in north.Florida jfore the onset of',rosts.
To assess the genetic variability of the species,. 0,accessi6) were /
studied to determine flowering dates, plant growth rate, m!orphological chracter-
istics, and resistance to Colletotrichum leaf spot disease 4iflZield an green-
house experiments. .
Except for two hybrids not responsive to changes in photoperiod, initial
flowering dates were generally too late in the fall for use north of about
290 N latitude. 'Early :season accessions began flowering the last week.in
September, and mid-.and late-season types flowered during the first two weeks
of October and November, respectively. Relative time of flowering in south
Florida was-.similar to that r4portedfor respective accessions in Australia.
Increasing Colletotrichum leaf spot incidence was positively correlated
with larger plant growth and- longer terminal leaflets, although resistance was
not always consistent among accessions in separate experiments.
Additional Index Words: Colletotrichum, Flowering dates, Plant introduction,
Stylosanthes humilis HBK. is an annual summer-growing legume used for
grazing in south Florida (8). Although self-pollinated, there is a wide range
of variability (1,, 2, 3, 4, 5, 6, 7, 9) in the naturalized plant population in
Australia where the'legume has been used extensively for grazing. Variability
was also.noted in Florida. commercial plantings made with unnamed Australian
commercial seed lots (senior author's personal observation).
Natural reseeding in north and west Florida has been unsuccessful, primarily
due to poor seed production associated with early frosts'and late flowering
1/ University of Florida Agricultural Experiment Station Journal Series 5219_.
2/ Agronomist, Assistant-Plant Pathologist, and Assistant Legume Breeder,
respectively, University of Florida, IFAS, ARC, Fort Pierce, Florida 33450.
characteristics of plants from imported Australian seed. Cameron-(4) found
that daylength is the main factor controlling flowering, although both higher
night and lower day temperatures delayed or inhibited flowering in early and
midseason types. Even though differences exist between temperature and day-
length of north and west Florida compared with south Florida, initial flowering
and other data obtained in south Florida can be utilized to select cultivars
or breeding lines that would be satisfactory in areas located below about 32o
A field experiment was established to morphologically and agronomically
characterize accessions of S. humilis to select early flowering and vigorous
growing types. Also, because of recent findings (10, 11) that Colletotrichum
leaf spot disease might be of economic importance on this legume, plant
characteristics common to less susceptible accessions were examined. A green-
house experiment was designed to supplement data obtained'from the field
Materials and Methods
For the field experiment, seeds of 39 S. humilis accessions (Table l)
were germinated on filter paper in petri dishes. Seedlings were transferred
in the greenhouse to 5.7 by 5.7 cm pressed-peat pots filled with limed and
fertilized Oldsmar fine sand soil.. After establishment of one Rhizobia-
inoculated plant per pot; plants were transplanted to limed and fertilized
Oldsmar fine sand at 1.5 m intervals in a randomized block design field
experiment with four replications and one plant per plot. An excess of Dixie
cowpea inoculant was applied to each pot after transplanting the seedlings.
Transplanting to the field was completed for all but a few lines prior to
September 1971. In June 1972, after germination of seed produced by plants
grown in 1971, all but.one plant was removed from each plot.
Flowering data were obtained weekly'beginning on August 31, 1971 and
on September 26, 1972. Because of..warm weather in both years, it was possible
to obtain flowering data even on the late-season types which normally would
not have flowered. On Octbber 3, 1972, three measurements from each plant
were obtained of lateral growth from one edge to the other plant edge (passing
over the crown). Data for each replication were averaged and are called plant
diameters in the text, since almost all plants formed circular-shaped, pro-
strate growth patterns. These measurements were considered an indication of
productivity. On November 3, the largest trifoliate leaf from each of three
10-cm terminal shoots per replication was obtained for terminal leaflet
measurements. On September 28,'1972, the severity of Colletotrichum leaf
spot was determined by counting the number of spots within four 325 cm2
quadrants of each plant. On December 4, 1972, the severity of Colletotrichum
stem cankers was determined by counting cankers from five randomly selected
20-cm. stcm ceameirts per plant.
In the greenhouse experiment, seeds of 42 accessions were planted on
April 3, 1973 in lmed and fertilized Oldsmar soil-filled'styrofoam cups having
a volume of 500 cm Beginning on May 17, the third mature leaf was obtained
from each plant of the four replications that had germinated six weeks pre-
viously. Leaves from other, slower-germinating accessions were also obtained
six weeks after initial germination. Beginning on June 14, plants that had
germinated 10 weeks previously were harvested. On July 9, regrowth of plants
were inoculated by spraying with a Colletotrichum-water suspension. The
organism was isolated previously from a Mexican accession of S. subsericea
(IRFL 1329) extremely susceptible to the disease. On July 16-and July 23,
lesions on leaves and stems were counted. In an extension of this experiment,
six accessions were seeded on July 26 into soil-filled styrofoam cups having
a volume of 275 cm Seedlings were transplanted to the field on September 10
on 30 by 60-cm centers in a randomized block design to provide 15 replicates of
each accession. In addition, vegetative cuttings were established from two
plants growing in the field in close proximity (Clones A and B). The clonal
parents were a result of natural regeneration over a ten year period from
originally seeded commercial Australian seeds. Fifteen plants of each clone
were included in the replicated test. On October 6, at 5-6 p.m. plants were
inoculated with a Colletotrichum-water suspension. A hand atomizer was used.
Each plant was sprayed twice, once from each direction. Light rains had
fallen before inoculation and a light rain fell on the.night following inocula-
tion. Lesions and numbers of trifoliate leaves were counted on October 17.
Analysis of variance and covariance, simple correlations, and variances
were calculated according to standard methods.
Results and Discussion
Initial flowering dates in the fall of 1971 and 1972 are presented in
Table 1 in conjunction with a brief description of the seed origin, plant
diameter, terminal leaflet length, and L.S.D. values at 5 and 1% levels of
The date of transplanting to the field was not correlated with flowering
dates in 1971. Even though initial flowering was later in 1971 than in 1972,
a highly significant correlation (r = +0.90) existed for individual accessions
between flowering dates of 1971 and 1972. The correlation may have been
greater had 1972 data collection been started earlier. This would have led
to greater differences among early flowering types that were flowering at the
first observation date. High consistency of initial flowering dates was
confirmed by data obtained from the selections made from accession 850. Acces-
sions 850A, 850D, and 850E were selected for plant vigor and flowering dates
were not different from the parent, 850, while the selection 102H, selected
for early flowering, was about one month earlier than the parent.
In Table 1, the accessions are grouped into flowering categories based on
initial flowering dates in 1972. In Group I, progeny of hybrids (called hybrids
throughout the text) 1409 and 1410, were non-sensitive to day length, flowering
about five to six weeks after germination regardless of daylength.
In Group II, were early accessions that flowered prior to October 4, 1972.
Included were two named Australian cultivars, 'Patterson' and 'Lawson'.
In Group III, mid-season types flowered between October 4 and October 19.
Included were 13 accessions, three each from Costa Rica and Brazil, and the
rest from Australia.
In Group IV, including 16 mid-late accessions, flowering began between
October 20 and November 11 and included three from Costa Rica and six from
Brazil, and 'Gordon', an Australian cultivar.
In Group V is the single late accession from Australia obtained from Rodd's
bay. It was significantly later flowering than all others at Ft. Pierce and
was considered the latest flowering accession in Australia. Range in 1971 and
1972 flowering dates among Costa Rican accessions were November 15-27 and
October 10-26, respectively, showing no significant variation within years
except for the earlier 1143 accession. Brazilian accessions, obtained over
a greater area and latitudinal range, had respective flowering ranges of
November 8 to December 3 and October 3 to November 4.
Cameron (5) in Australia showed that initial flowering among lines varied
much more when seeded in December than in March. Comparable months at the
Agricultural Research Center (ARC), Ft. Pierce would be June and September,
respectively. Cameron also found tkat initial flowering was generally delayed,
the more southerly (in the southern hemisphere) the seeding location. Flowering
data for two accessions (1332 and 849) could be compared at the ARC, Ft. Pierce
(27.4N latitude), Landsdown, Queensland, Australia (10.70S) and Stamford, Queen-
sland, Australia (27.30S). Ft. Pierce and Stamford flowering dates were later
than those for Landsdown; and there was a tendency for Ft. Pierce flowering
dates to be later (four and seven days) than those for Stamford. An additional
nine accessions, 852, 853, 1333, 1337, 1338, 1339, 1340, 1331, and 1336, were
available for comparison with those of Landsdown. Based on Cameron's (6)
flowering data, the first three listed were mid-season, the next four mid-late,
and the last two late-season. Initial flowering of mid-, mid-late-, and late-
season accessions were 2 to 6, 2 to 3, and 1 to 2 weeks later in Florida than
in Landsdown. These results agree with Cameron's (5). He found, when comparing
flowering dates of S. humilis growing in lower and higher latitudes, that differ-
ences in latitude had a greater effect on initial flowering dates for early
flowering accessions than for late-season types.
Plant Diameter and Morphological Characteristics
Average plant diameter varied from 41 to 135 cm (Table 1). Selections
850A, 850D, and 850E made for vigor from Accession 850, had 128, 135, and
124 cm diameters respectively, compared to 91 for the 850 parent. This in-
dicates success in simple selection for increasing plant size. Although the
hybrid, 1409, was one of the earliest flowering, as well as the smallest in
plant diameter, there was no correlation between plant diameter and initial
Terminal leaflet length ranged from 1.14 to 2.60 cm. Ranges for other
morphological characteristics were as follows: width of terminal leaflets -
0.34 to 0.73 mm (avg. 0.47); average number of pairs of veins per terminal
leaflet 4,00 to 6.25; accessions with petiole hairs absent to sparse 850,
1025, 1332, and 1338; accessions with petiole bristles absent to sparse 852,
1335, and 1341. None of the accessions were classified as having stipules
with none to sparse hairs or bristles; while 850D, 850E, 1336, 1334, and 850A,
850E, 1136, 1336, and 1340 had stipules with abundant hairs and bristles,
respectively. Hair distribution on stems of 21 accessions was limited to one
side while hair density generally was rated as moderate except for 1332 where
abundant hairs were found. Absence of stem bristles was almost limited to
those accessions having hairs on one side of the stem. Black seed coat color
was found in 852, 1136, 43, 1335, 1411, and 1412. Other's color ranged from
light to dark brown. Seed pods of accessions 850, 1143, 1330, 1331, 1333, 1334,
1335, 1338, 1339, 1340, 1409, 1410, and 1412 were without hairs. Seed pod
beaks (hooks) of accessions 850, 1209, 1210, 1211, 1330, 1335, 1409, 1410 and
1412 were longer than 3mm while others were less.
Positive relationships are shown by correlations between plant diameter
and both terminal leaflet length (r = 0.76**) and leaf spot incidence (r =
0.74**). The significant relationship between stem canker and leaf spot
incidence (r = 0.35**) may have been greater if the dispiftby between rating
dates had been less. The implication of the positive correlation between plant
diameter and leaf spot incidence is unfortunate since breeding or selection
programs would include the search for more rapidly growing plants. Although
selections 850A, 850D, and 850E from the later flowering accession, 850, pro-
duced larger plants, leaf spot2disease was also more severe. Number of spots
were 75, 50 and 52 per 1300 cm respectively, compared with 21 for 850.
Contrariwise, successful selections made for earlier flowering, 102H and 105H,
resulted in plants of size and disease incidence equal to the respective
parents, 850 and 852.
There is the possibility that plant diameter was closely related to some
other factor (shading, higher moisture, etc.) that was more accommodating for
disease development. However, there did not appear to be a visual reIationship
between leaf density and plant diameter. Since the plants were prostrate and
clean fallow conditions were maintained between plants, differences in humidity
caused by denser foilage were minimized.
The greenhouse experiment was conducted to further study relationships of
size, diameter, and length of terminal leaflet to disease incidence. Data
presented in Table 2 were obtained from accessions grown in the greenhouse.
Yields ranged from 0.06 to 1.34 grams per plant, terminal leaflet length from
15.3 to 31.5 mm, terminal leaflet width from 3.3 to 7.6 mm, and numbers of
leaf spots on July 16 from zero to 43 per plant with an average of 15.
Simple correlation coefficients (r) among various relationships are as
follows: (1) number of lesions per plant and leaflet length 0.60**, (2)
number of lesions per plant and yield per plant 0.75**, and (3) yield and
leaflet length 0.55**.
Repeatability of recognizing and counting spots was good as evidenced by
the highly significant correlation (r = 0.89) of numbers of lesions per plant
on July 16 and July 23.
The variation of disease susceptibility within several accessions were
compared with two clones. Based on previous work, three disease resistant
accessions (886, 1410, and 1335) and three susceptible accessions (1409, 1547,
and 850A) were compared. Two hybrids, 1409 and 1410, were included since
genetic variability in their respective F6 and F5 generations was presumed
more than those of the other seed sources because of possible continued segre-
gation. The greatest genetic uniformity would be expected within the two
The average number of leaf spots per plant, numbers of trifoliate leaflets
per plant, number of trifoliate leaflets per lesion, and variances of the number
of lesions per plant are presented in Table 3.
There were large within-accession variations. After using the covariance
technique, it was possible to group the accessions based on residual (within-
accession) variation. Results indicated that the largest variation occurred
for Clone B, 1547 and 1410 plants; the next largest variation for group 1335,
Clone A, and 850 followed by 1409 which had the least within-accession variation
of diseased leaves. Since the group having the greatest within-accession
variation of disease incidence included clonal B plants, the conclusion is that
environmental factors and method of inoculation were associated with the large
variation within accessions. Furthermore, these data show the necessity of
having precise experimental conditions when determining disease susceptibility
under artificial conditions.
Consistent susceptibility for some accessions (886, 850A, and 1547) but not
for others (1409 and 1335) was shown between the two experiments. Vegetatively
propagated clones A and B, although obtained from nearby growing parent plants,
differed in their resistance. It was not determined how plant maturity differences
affected disease susceptibility.
Although differences in initial flowering dates of accessions varied con-
siderably, the early flowering types except for two hybrids, probably were not
early enough for areas north of about 290N latitude. Two hybrids tested showed
no photoperiodic response but were not considered sufficiently vigorous or long-
lived to be utilized commercially. Relative flowering time of accessions in
south Florida was similar to respective data from Australia; but differences in
latitudes had a greater influence on initial flowering of early- than that on
mid- or late-season types. It appears that plant breeding would be the surest
method for obtaining early flowering lines since natural occurring types tested
here and in Australia were much later.
It is not possible from present studies to assess the significance of
Colletotrichum leaf spot and stem canker disease. No economic damage caused
by this disease has been reported in Florida commercial plantings. Data indicate
that problems could arise if susceptible cultivars are used. Studies of disease
inoculation techniques show that the method of inoculation and, or environmental
or plant physiological factors are responsible for large variations among plants
of the same genotype.
1. Burt, R. L., L. A. Edye, W. T. Williams, B. Grof, and C. H. L. Nicholson
1971. Numerical analysis of variation patterns in the Genus
Stylosanthes as an aid to plant introduction and assessment. Australian
J. Agr. Res. 22: 737-757.
2. Cameron, D. F. 1965. Variation in flowering time and in some growth character-
istics of Townsville lucerne (Stylosanthes humilis). Austre 1in. J. Exp.
Agr. Anim. Husb. 5: 49-51.
Literature Cited (cont'd)
3. Cameron, D. F. 1967. Flowering time and the natural distribution and dry
matter production of Townsville lucerne (Stylosanthes humilis)
populations. Australian J. Exp. Agr. Anim. Husb. 7: 501-508.
4. 1967. Flowering in Townsville lucerne (Stylosanthes
humilis). 1. Studies in controlled environments. Australian J.
Exp. Agr. Anim. Husb. 7: 489-494.
5. 1967. Flowering in Townsville lucerne (Stylosanthes
humilis). 2. The effect of latitude and time of sowing on the
flowering time of single plants. Australian J. Exp. Agr. Anim. Husb.
6. 1970. Townsville lucerne (Stylosanthes humilis): A
comparison of introductions from Brazil and Mexico with naturalized
Australian ecotypes. Proc. llth Int. Grsld. Cong., Surfers Paradise,
Queensland, Australia, pp. 184-187.
7. Downes, R. W., I. B. Staples, P. A. Colman, and P. Pedley. 1967. The
effect of daylength and temperature on the growth and reproduction of
six strains of Townsville lucerne (Stylosanthes humilis). Australian
J. Exp. Agr. Anim. Husb. 7: 351-356.
8. Kretschmer, Albert E., Jr. 1968. Stylosanthes humilis, a summer growing,
self-regenerating, annual legume for use in Florida pastures. Florida
Ag. Exp. Sta. Circ. S-184. 21 pp.
9. 't Mannetje, L. 1965. The effect of photoperiod on flowering,growth habit,
and dry matter production in four species of the genus Stylosanthes.
Australian J. Ag. Res. 16: 767-771.
10. Sonoda, Ronald M. 1973. Incidence of Colletotrichum leaf spot and stem
canker on various introductions of Stylosanthes humilis HBK. Plant
Dis. Reptr. 57: 747-749.
11. 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).
Table 1. Origin, description, initial flowering dates, plant diameters and terminal
leaflet lengths of Stylosanthes humilis accessions.
PI. no;. dates
Cross #102, F4
Cross #136, F5
Cape York, P61-3
Sel. from 850
Sel. from 852
Sel. from 849
Sel. #3 commercial
Sel. #3 commercial
Sel. # commercial
Sel. #3 commercial
Sel. #1 commercial
El Capulin #1
El Capulin #2
Sel. from 850
Table 1. (Cont.)
Initial flowering (, Plant Leaflet
IRFL* Origin and PI. no. ? dates Maturity diameter length
no. description (Country no.) 1971 1972 group cm cm
850A A Sel. from 850 1-0.3(12/3) 6.5(11/4) IV 128 2.54
850D A Sel. from 850 10.3 6.5( ) IV 135 2.42
1336 B Pernambuco 358384(40267) t0.3 6.8(11/6) IV 132 2.60
1331 A 'Gordon' 358379 10.5(12/5) 6.5(11/4) IV 107 2.24
1334 A Rodd's bay 358350 14.5(1/2) 9.0(11/21)** V 65 1.61
LSD '.05 0.9 1.6 5 0.40
.01 1.2 2.1 33 0.52
* Agricultural Research Center, Ft. Pierce accession number.
" Seed originally collected in: A=Australia, B=Brazil, C=Costa Rica, M=Mexico and U=unknown.
4- PI numbers are USDA plant introduction numbers while 4 and 5 place numbers in parenthesis are
Brazilian and Australian accessions, respectively.
Weekly observations beginning on August 31, 1971 and September 26, 1972. Numbers are in weeks with
number 1 = flowering prior to the first observation date. Numbers in parenthesis represent approximate month/day
initial flowering date.
SI=hybrids, II=flowering before Oct. 4, III=flowering from Oct. 5 to Oct. 19, IV=flowering from Oct. 20
to Nov. 11, V-flowering after Nov. 11, 1972.
** Data not included in statistical analysis because of less than four replications.
Table 2. Dry.weight yields, terminal.leaflet length and width, and
Collectotrichum incidence of-S. humilis accessions growing
Sin the greenhouse.
Accession Terminal leaflet No.
no. Yield length width lesions
-' gr/plant mm mm ..- per plant
* Not included in statistical
analysis because there were less than four
Table 3. Number of Colletotrichum lesions, number of trifoliate leaflets,
number of leaflets per lesion, and variances-of S. humilis
L.S.D. .05 17.7 54 6 --
.01 23.5 72 N.S. --
* Not included in statistical analyses because there were less (9) than