229

Euphytica 84: 229-235, 1995. (~) 1995 Kluwer Academic Publishers. Printed in the Netherlands.

Evaluating different bahiagrass cytotypes for heat tolerance and leaf epicuticular wax content C.R. Tischler & B.L. Burson United States Department of Agriculture, Agricultural Research Service, Grassland, Soil and Water Research Laboratory, 808 East Blackland Road, Temple, TX 76502 USA Received 17 May 1995; accepted 2 June 1995

Key words: Paspalum notatum Flugge, bahiagrass, heat tolerance, leaf epicuticular wax

Summary Bahiagrass (Paspalum notatum Flugge) is a polymorphic species indigenous to South America which has become naturalized in the southeastern U.S. The most common form in the United States is Pensacola bahiagrass, P. notatum var. saurae Parodi., which is a valuable forage. Pensacola is a sexual diploid, while most other bahiagrasses are apomictic tetraploids. Pensacola bahiagrass is considered to have greater heat tolerance (based on an in vitro heat tolerance test) than a number of other Paspalum species, but has less leaf epicuticular wax (a drought avoidance mechanism) than other species. Both heat tolerance and leaf epicuticular wax are desirable characteristics for species grown where periodic drought occurs. We measured both characteristics over two years in a collection of 23 bahiagrass accessions, many of which had been collected in South America near the center of origin of the species. The collection included various ploidy levels. For both years, no accessions ranked statistically lower in damage in the heat tolerance test than Pensacola, although eight had significantly higher damage. Two entries in addition to Pensacola had very low damage in the heat tolerance test. Pensacola was high in leaf wax, with 16 accessions rated significantly lower in wax. The accession having the lowest wax content was a triploid, which also exhibited considerable leaf death in the field in response to drought and heat. The diploid entries tended to be higher in leaf wax than the other ploidy levels. This study has identified additional bahiagrass germplasm which may be of use in a breeding program.

Introduction Paspalum is a large genus of the Gramineae family with nearly 400 species, most of which are native to the Western Hemisphere (Chase, 1929). It is one of the most economically important genera of the Paniceae tribe because many species are valuable forage grasses. One of the more important members is bahiagrass, P. notatum Flugge. This polymorphic species is indigenous to southern Brazil, Uruguay, northeastern Argentina, and Paraguay, and has become naturalized in many areas of South and Central America, West Indies, Mexico, and the southeastern U.S. It was officially introduced into the U.S. in 1913, but probably already had been accidently brought into the country.

The most common form in the United States is "Pensacola' bahiagrass, P. notatum var. saurae Parodi. It was inadvertently introduced into the Pensacola, Florida area prior to 1926 (Burton, 1946, 1967). The grass has spread throughout the southeastern U.S. where it is an important forage. Unlike other bahiagrass types, Pensacola is a sexual diploid with 20 chromosomes (Burton, 1946, 1955) and has been the subject of extensive breeding efforts (Burton, 1955, 1982). A recent survey of the heat tolerance and epicuticular wax content of several Paspalum species revealed that Pensacola bahiagrass had more capacity for heat tolerance than other species, but a relatively low amount of wax (Tischler et al., 1990). Heat tolerance and high levels of leaf epicuticular wax are desirable characteristics for species growing in areas where periodic drought occurs. This is espe-

230 cially true for bahiagrass because it usually grows on sandy soils, with low water-holding capacity. Without frequent precipitation, plants growing in these soils are often subjected to drought stress. The test which indicated the high heat tolerance for bahiagrass (Tischler et al., 1990) is a modification of the classical in vitro heat tolerance test reported by Sullivan (1972). It is based on the premise that leakage of solutes through the plasmalemma following a heat treatment represents a measure of relative damage. Sulivan (1972) and Bouslama & Schapaugh (1984) indicated that results from this test are correlated with the relative drought resistance of plants under field conditions. In nature, heat damage is a normal consequence of drought. When little or no water is available for transpirational cooling, the energy balance of the leaf changes greatly, and heat transfer must occur either through convection or radiation. As a consequence, leaf temperature increases and may be several degrees higher than air temperature. Leaf wax content partly controls the degree of desiccation a leaf experiences under water stress because when stomates are tightly closed, the major avenue of water loss from the leaf is the cuticle. High wax levels make the cuticle more impervious to water loss. Tischler & Voigt (1990) reported a highly significant relationship between epicuticular wax content and both cumulative water loss and rate of water loss for 19 lovegrass [Eragrostis curvula (Schrad) Nees.] selections, when water loss was expressed as a fraction of water contained. Other studies have also indicated that high levels of leaf epicuticular wax are strongly associated with the ability of a plant to tolerate water stress (Blum, 1975; Schonherr, 1976). We wondered whether other bahiagrass accessions also have high levels of heat tolerance as well as high levels of leaf epicuticular wax. Our objectives were to measure the heat tolerance and leaf epicuticular wax content of a range of bahiagrass germplasm collected from different geographical areas near its center of origin. Because Pensacola is a diploid and most bahiagrass types are tetraploids, a second objective was to determine if ploidy level was associated with heat tolerance.

Materials and methods Seed of 30 bahiagrass accessions, most of which were collected near the center of diversity of this species, were germinated during the spring of 1991. These

plants were transplanted into a field nursery on Houston black clay soil (fine montmorillonitic, thermic Udic Pellusterts) at Temple, TX. In the fall of 1991, entire plants of 23 accessions were collected, separated into ramets, and placed in pots in a greenhouse. In April 1992, a replicated nursery was established near the original nursery with three plants each of 23 accessions randomized in a row within each of six plots (replications). The identification number, collection site, and chromosome number of each accession are listed in Table 1. Plants within a row were spaced 0.6 m apart, while spacing between rows was 1 m, to allow cultivation with a small tractor. In April of both 1992 and 1993, the nursery was fertilized with 67 kg ha- i N in a single application. No herbicides were used in the nursery, and weeds were removed by hand. The nursery was irrigated regularly in 1992, but no irrigation water was applied in 1993.

Heat tolerance testing Heat tolerance was measured in July and August 1992 and in July 1993, by the technique of Martineau et al. (1979), as modified by Tischler et al. (1990). The major modifications were a 45 min heat treatment rather than 15 min, and the elimination of controls to correct for spurious ion leakage. Tests in our laboratory, using several different grass species, have indicated that controls are not necessary if the samples are incubated at room temperature. In all cases, results for an entry and sampling time are the mean of twelve values (six replications with two subsamples per replication). When the July 1993 sampling was performed, no rain had fallen for one month, and plants were showing visible signs of water stress.

Wax analysis Amounts of epicuticular wax were determined by the colorimetric method of Ebercon et al. (1977). Two subsamples from each entry in each replication were harvested before 0800 h, when the leaves were wet with dew. These samples were enclosed in moist paper towels, and taken to the laboratory. Leaf samples were harvested early in the morning and kept moist to prevent leaf rolling, which makes it difficult to measure leaf area. A photoelectric leaf area meter was used to measure leaf area. Approximately 60 cm 2 of total leaf surface was collected from each subsample. The leaves were cut into segments 2.5 cm long, and the wax was extracted by stirring the segments in 15 mL chloroform

231

Table 1. Place of collection and chromosome number of Bahiagrass accessions studied Entry

9

10 11 12 13 14

15 16 17 18 19 20 21 22 23

Collection Site

PI or collection No.

2n

I Km NW Manfredi, Cordoba Province, Argentina Arocena. Santa Fe Province, Argentina Igr. Maschwitz (50 Km NW Buenos Aires). Buenos Aires Province, Argentina Salto Encantado, 42 Km E of intersection of Routes 12 & 220. Misiones Province. Argentina Route 9, 87 Km NW Santiago del Estero. Tucuman Province, Argentina Route 34, 33 Km N Metan, Salta Province. Argentina Near Rio Uruguay at Concepcion del Uruguay, Entre Rios Province, Argentina Isla Santa Candida, near W bank of Rio Parana between Parana and Santa Fe, Santa Fe Province. Argentina Same as entry 8 Near Route BR 116 46 Km N Lages. Santa Catarina, Brazil 5 Km NW Vacaria, Rio Grande do Sul. Brazil Route BR 293.44 Km E Bage, Rio Grande do Sul. Brazil Route BR 290 30 Km E Alegrete, Rio Grande do Sul, Brazil 30 Km W Guaiba, Rio Grande do Sul, Brazil Chololo. 15 Km N Paraguari, Paraguay Desarrollo Ganadero-Estancia. 10 Km E Caacupe. Paraguay Heisecke Estancia. 12 Km S Bella Vista. Paraguay Heisecke Estancia, 50 Km W San Juan Bautista, Paraguay Rio Santa Lucia. 12 Km W Montevideo, Uruguay Route 26, 129 Km NE Paysandu, Uruguay Route 8, 8 Km SW Treinta y Tres, Uruguay 6 Km E Bradenton. Florida, USA Tifion, Georgia, USA (Pensacola bahiagrass, originally collected at Pensacola. Florida)

508828 508834

40 40

330-79

40

508839

40

508841 508842

40 40

508836

20

291-79 508848

20 20

508825 404472

50 40

404478

40

404481 404482 404665

40 40 40

404667 404670

40 40

404672 404861 404864 404867

40 40 40 40 30

for 15 sec. T h e c h l o r o f o r m w a s s u b s e q u e n t l y e v a p o -

20

Visual observations

rated f r o m the s a m p l e s , and 5 m L o f acidic K2Cr207 w a s a d d e d to the w a x residue. S a m p l e s w e r e h e a t e d 1 h

In e a r l y S e p t e m b e r , 1993 the field n u r s e r y w a s v i s u a l l y

in a b o i l i n g w a t e r bath to o x i d i z e the wax, and 12 m L

e v a l u a t e d to d e t e r m i n e if any e n t r i e s w e r e s h o w i n g

d i s t i l l e d w a t e r w a s a d d e d . A b s o r b e n c e at 590 n m w a s

s i g n s o f leaf firing or c r o w n death. T h i s w a s after a

m e a s u r e d w i t h a c o l o r i m e t e r . Wax values w e r e calcu-

p e r i o d o f a p p r o x i m a t e l y 2 m o n t h s w i t h o u t rain. Each

lated as d e s c r i b e d by T i s c h l e r et al. (1990).

e n t r y in e a c h r e p l i c a t i o n was o b s e r v e d .

232 Table 2. Heat tolerance values of bahiagrass accessions, 1992 and 1993

Entry

Damage (%) July-Aug 1992

July 1993

6

59.0a I

18.1 ab

13

54.6 ab

14.3 bcde

Data analysis Standard analysis of variance techniques were used to analyze the data. For the heat tolerance data, Spearman rank correlations were used to compare rankings between the various sampling dates.

I1

53.3 ab

10.7 cdefg

20

52.4 ab

18.7 ab

Results

15 8*

51.4 ab 49.7 abc

15.8 bc 9.8 efg

Heat tolerance

12 14

49. I abcd 48.4 abcd

15.5 bcd 15. I bcd

10

45.1 abcde

15. I bcd

21 1 22

43.0 bcde 41.9 bcde 41.7 bcde

14.4 bcde 21.0 a 12.2 cdef

2 16

41.0 bcdef 40.9 bcdef

12.0 cdefg 7.8 fg

3 4 5

39.9 bcdef 36.1 cdefg 34.5 defg

9.1 fg 10.9 cdefg 10.5 defg

17

32.5 efg

9.4 efg

23* 18 19

31.8 efg 30.7 efg 26.6 fg

7.1 g 9.3 efg 9.0 fg

9* 7*

24.2 g 21.4g

15.1 bcd 7.3 fg

* Denotes diploid entries. i Values in a column not followed by the same letter differ significantly at the 5% level, Duncan's Multiple Range Test.

Chromosome number determinations The chromosome number of each entry was determined by counting the somatic chromosomes in root tips. A small clone was taken from each entry and placed in a container of water inside a mist chamber in a glasshouse. After new root growth was initiated, the roots were collected (0800-0930 hr), placed without fixation in a saturated solution of bromonaphthalene for 2 h, and hydrolyzed in 1 N HCI for 10 min at 60 ~ C. Root tips were stained in Feulgen solution, macerated in a drop of aceto-carmine strain, coverslipped, heated, pressed firmly, and observed using phase contrast microscopy.

From the overall analysis of variance, the entry by sampling date interaction was significant between the 1993 sampling and both 1992 samplings. However, the interaction was not significant for the two 1992 samplings, and the 1992 data were combined for analysis. Heat damage in July 1993 was considerably lower than in 1992 (Table 2). In 1992, the nursery was kept moist, while in 1993, samples were collected approximately one month into a drought. Other studies have indicated that water or heat stress tend to lower the level of damage in the heat tolerance test (Blum & Ebercon, 1981). Also, the discrimination between entries in 1993 was not as stringent as in 1992, because maximum discrimination is achieved when mean damage is approximately 50% (Sullivan et al., 1972), and levels of damage were considerably lower in 1993. Thus, the statistical treatment of the 1993 data is more conservative, but significant differences are evident. Ranking of genotypes over the two years differed somewhat, although Spearman rank correlations between all sampling dates were significant (p = 0.05). Perhaps the most obvious ranking differences are the values for entry 9, which was very low for damage in 1992, but relatively high in 1993. Entry 1 also was somewhat inconsistent, in that it showed the highest damage in 1993, but only intermediate damage in 1992. However, with th~se exceptions, correlation was good between years. Pensacola bahiagrass (entry 23), which was the most heat tolerant Paspalum species in an earlier report (Tischler et al., 1990), was also among the most heat tolerant of the accessions studied here (Table 2). It was lowest in damage in 1993. Although it was not the lowest in 1992, it did not differ statistically from the entry with the lowest value. The diploid entries, with one exception, were more heat tolerant than many of their tetraploid counterparts, as three of the diploid entries are at the bottom of the range of damage values. The exception was entry 8, and it was inconsistent

233 Table 3. Leaf epicuticular wax content of bahiagrass accessions, combined over 1992 and 1993 Entry

Wax

Entry

(mg/dm2) 17 9* 19

Wax (mg/dm 2)

1.859 a

10

1.099 def

1.744 a

2

1.093 def 1.080 def

1.687 ab

18

8*

1.602 abc

4

1.067 ef

23*

1.564 abc

21

1.066 ef

3

1.407 bc

15

1.053 ef

7*

1.309 cde

12

1.052 ef

16

1.205 de

14

1.043 ef

I

I. 192 de

6

1.042 ef

20

1.137 def

13

0.832 fg

II

I. 125 def

22

0.694 g

5

I. 106 def

* Denotes diploid entries. I Values not followed by the same letter differ significantly at the 5% level, Duncan's Multiple Range Test.

across years, exhibiting good heat tolerance in 1993, but poor heat tolerance in 1992, when compared to the other entries. Entry 22, a triploid, was intermediate in damage, but field observations during the extended drought period of 1993 indicated that under such conditions, it was the least desirable of all genotypes in the experiment because of an almost complete loss of green leaves. The pentaploid entry, number 10, was also low in heat tolerance (Table 2).

Leaf wax Epicuticular wax amounts varied almost by a factor of three from the highest to the lowest, although the means of 13 entries were grouped between 1.0 and 1.2 mg/dm 2 (Table 3). The four diploid entries (including Pensacola) were near the top of the range; however, some tetraploid entries had similar values. The remaining tetraploid entries extended to the low end of the range. Entry 22, the triploid, had the least amount of leaf wax, and its wax load was significantly lower than those of the other 21 entries. Entries 22 and 13 were very low when compared to the nearly continuous range of values in the data set (Table 3). Entry 10, the pentaploid accession, was intermediate in wax content.

Visual observations The field nursery was visually ranked for stress symptoms in September 1993, after more than 60 days with no rainfall or supplemental irrigation, and daily maximum temperatures near 40o C. All entries appeared stressed, but entry 22 was conspicuously more stressed than the other 21 entries. Its leaves were desiccated, brown in color, and senesced to the stem. This appearance was observed in all plants of this accession in each of the 6 replications. None of the other entries exhibited these symptoms. However, after rainfall in September, entry 22 initiated new leaf growth and recovered. This qualitative difference in behaviour of entry 22 indicates that it is the least desirable forage grass of all the entries studied, as it would provide almost no forage to livestock during a drought.

Discussion Results from this study corroborate earlier findings (Tischler et al., 1990) in that Pensacola bahiagrass exhibits a high capacity for heat tolerance. The heat tolerance test (Table 2) demonstrated that of the bahiagrass accessions evaluated, Pensacola was in the group with the least amount of damage in both 1992 and 1993. Pensacola was the entry with the least amount of damfige in the 1993 tests (Table 2). Differences in heat tolerance between Pensacola and the other bahiagrass entries were not as pronounced as those between Pensacola and the different Paspalum species reported by Tischler et al. (1990). This may be because bahiagrasses as a group may have more tolerance to heat damage than other Paspalum species. The germplasm used for this investigation is unique from several perspectives. Many entries were collected in an area of South America which is considered to be the species' center of origin and is where the maximum diversity within the species should exist. The accessions used provide a good representation of the geographical region (Fig. 1). The collections also include considerable variability for chromosome number and reproductive behavior. Most naturally occurring bahiagrass types are 40 chromosome tetraploids which reproduce by apomixis. This is demonstrated by the high number of apomictic tetraploids in this study (Table 1). Besides Pensacola, three additional sexual diploid (2n = 2X = 20) accessions were included and all were collected in eastern Argentina.

BOL

234

,

4 L

~

I

1 \-

_

I

.

/

t t.I

O

/

\

76.1

% %

I

50/'

I

KM

P15Oo16

"-.

/

I

17o

/

06

\~..) I

I

/I

k

c / S

4 Q

[

I

I

O

1

,

O18 /

I_

I

i/

I4

I

t ,

,

F. N I';'T

i13

g

/

\

\ \

--~

__\

l

q

I!

\1

!

1

"x_ - _,_

/

/

~'~ o12

I"

~\~ 20[ 7m / ) 8"/-9 3o~ ,

\_

\

ATLANTIC OCEAN

~

020

U R U G U A Y O21

\ \

/ I I \

/

/ [.-

- -/

/ --

--

t N I

14o

o13

1

!/

" - -,.~ I

I I I

\

"v

11o

I

9

I

o19

Collection Sites m2n=2x=20 02n=4x=40 a2n=5x=50

Fig. I. Geographicaldistribution of the Bahiagrass accessions used in this study.

Entries 8 and 9 were collected on an island in the Parana River where the diploid cytotype is relatively common. This may be the same area where the original source of germplasm existed which resulted in Pensacola bahiagrass (Burton, 1967). Entry 7, the other diploid, was collected 235 Km to the southeast near the Uruguay River in Entre Rios Province (Fig. 1). Its diploidy was not too surprising, because essentially all diploid accessions collected in the wild have been from Entre Rios and eastern Santa Fe Province (Burton, 1967). The pentaploid, 50-chromosome accession was not expected and to our knowledge this is the first report of a bahiagrass with this chromosome number. The results were not conclusive as to whether there is a relationship between heat tolerance and chromosome number. Two diploid entries, 7 and 23, had low solute leakage values for 1992 and 1993. However, solute leakage of two other diploids, entries 8 and 9, were inconsistent across years. Because these four

accessions are sexual, variability among accessions for this trait would be expected. The triploid and pentaploid entries were intermediate for that trait. There was also considerable variability among the apomictic tetraploid accessions. Although these tetraploid accessions reproduce by apomixis, they are polymorphic and each acqession probably represents a different apomictic ecotype. These ecotypes are genetically different and would be expected to differ for the two traits measured because they probably originated independently. All diploid accessions were in the top one third of the entries for epicuticular wax content and three, including Pensacola bahiagrass, were statistically similar to the entry with maximum wax (Table 3). Wax content may be a more reliable indicator of drought tolerance because the triploid entry, number 22, had the least amount of wax and exhibited the most severe symptoms of drought stress in the field.

235 When both heat tolerance and wax content are considered for all entries, it appears that entries 17, 19, and 23 would be the most drought tolerant. However, these entries have nothing in common regarding chromosome number or collection site. Entry 17 was collected in northern Paraguay, probably the most tropical area explored, whereas entry 19 was obtained near Montevideo, Uruguay, the most temperate area. Entry 23 is Pensacola bahiagrass. Entry 6 was collected from a relatively dry location in the Andes in northwest Argentina at an elevation of about 900 m. It was expected to have low solute leakage rates and high epicuticular wax amounts but it was the highest for heat damage and near the lowest for leaf wax. Findings from this study confirm those reported by Tischler et al. (1990) regarding Pensacola bahiagrass. Also, three or four other bahiagrass accessions were identified with potentially similar or superior drought tolerance as Pensacola. Because the traits measured are heritable (Jordan et al., 1984; Sullivan & Ross, 1979) and some of these accessions are sexual, it may be possible to improve drought tolerance in future bahiagrass cultivars.

References Blum. A.. 1975. Effect of the Bm gene on epicuticular wax and the water relations of Sorghum bicolor. Israel J. Bot. 24: 50. Blum, A. & A. Ebercon, 1981. Cell membrane stability as a measure of drought and heat tolerance in wheat. Crop Sci. 2 I: 43-47. Bouslama, M. & W.T. Schapaugh, Jr., 1984. Stress tolerance in soybeans. I. Evaluation of three screening techniques for heat and drought tolerance. Crop Sci. 24: 933-937.

Chase, A., 1929. North American species of Paspalum. Conn. U.S. Natl. Herb. Vol. 28: Part 1. Burton, G.W., 1946. Bahiagrass types. J. Am. Soc. Agron. 38: 273281. Burton, G.W., 1955. Breeding Pensacola bahiagrass,Paspalum noraturn: 1. Method of reproduction. Agron. J. 47:311-314. Burton, G.W., 1967. A search for the origin of Pensacola bahia grass. Econ. Bot. 21: 379-382. Burton, G.W., 1982. Improved recurrent restricted phenotypic selection increases Bahia forage yields. Crop Sci. 22: 1058-1061. Ebercon, A.. A. Blum & W.R. Jordan, 1977. A rapid colorimetric method for epicuticular wax content of sorghum leaves. Crop Sci. 17: 179-180. Jordan, W.R.. P.J. Shouse. A. Blum. ER. Miller & R . L Monk, 1984. Environmental physiology of sorghum. 1I. Epicuticular wax load and cuticular transpiration. Crop Sci. 24: I 168-1173. Martineau, J.R., J.E. Specht, J.H. Williams & C.Y. Sullivan, 1979. Temperature tolerance in soybeans. 1. Evaluation of a technique for assessing cellular membrane thermostability. Crop Sci, 19: 75-78. Schonherr, J., 1976. Water permeability of isolated cuticular membranes: The effect of cuticular waxes on diffusion of water. Planta 131: 159-164. Sullivan. C.Y., 1972. Mechanisms of heat and drought resistance in grain sorghum and methods of measurement, p. 249-264. In: N.G.P. Rao & L.R. House (Eds). Sorghum in the Seventies. Oxford & IBH Publishing Co, New Delhi. India. Sullivan, C.Y. & W.M. Ross, 1979. Selecting for drought and heat resistance in grain sorghum, p. 263-281. In: H. Mussell & R. Staples (Eds). Stress Physiology in Crop Plants. John Wiley & Sons, New York. Tischler, C.R. & EW. Voigt, 1990, Variability in leaf characteristics and water loss in the Weeping Lovegrass complex. Crop Sci, 30: 11[-117. Tischler, C.R., EW. Voigt & B.L. Burson, 1990. Evaluation of Paspalum germplasm for variation in leaf wax and heat tolerance. Euphytica 50: 73-79.