The role of light and temperature in the germination of twenty herbaceous species from the highlands of Ethiopia

The role of light and temperature in the germination of twenty herbaceous species from the highlands of Ethiopia

Flora (1998) 193 411-423 © by Gustav Fischer Verlag The role of light and temperature in the germination of twenty herbaceous species from the highl...

2MB Sizes 0 Downloads 4 Views

Flora (1998) 193 411-423

© by Gustav Fischer Verlag

The role of light and temperature in the germination of twenty herbaceous species from the highlands of Ethiopia DEMEL T'EKETAY Swedish University of Agricultural Sciences, Faculty of Forestry, Department of Forest Vegetation Ecology, S. 901 83 Umea, Sweden (Permanent address: Forestry Research, Ethiopian Agricultural Research Organisation, P.O. Box 2003, Addis Abeba, Ethiopia) Accepted: October 15,1997

Summary The effects of different light qualities, i.e. white light, darkness and light filtered through green leaves, as well as constant (10-30 0c) and diurnally alternating temperatures (20 versus 20/12,30/12 and 35/12 °C) on seed germination of 20 herbaceous species from the highlands of Ethiopia were tested in a laboratory and a glasshouse. Species differed considerably with respect to the capacity of fresh seeds to germinate at 20 °C in light. Only six of the species showed more than 80% germination whereas in seven of the species germination ranged from 1-50%, and in seven of them none of the fresh seeds germinated. In the laboratory test of fresh seeds, 11 of the species had significantly higher percent germination in light than in darkness. In the glasshouse, percent germination from seeds incubated in daylight was significantly higher than those incubated in darkness or under leaf shade for 15 out of 17 species tested, indicating the induction of secondary dormancy during incubation in either darkness or leaf shade. The species showed diverse responses to the various constant temperature treatments. For six of the species germination was more than 50% at all of the constant temperatures whereas for seven of the species germination was less than 50% at the different constant temperatures. Germination was more than 50% at the constant temperatures between 15 and 30 °C but lower than 43% at 10 °C for four of the species. For one of the species, germination was more than 50% at the constant temperatures between 10 and 25 but germination declined to 18% at 30 °C. Of the 16 species tested, 11 showed significantly higher germination from at least one of the alternating temperatures than the control. The results provide evidence that germination of almost all of the species studied is inhibited or greatly suppressed by either darkness or leaf shade or both, suggesting that germination and establishment of the species would be prevented if the seeds are buried deep into the soil or dispersed under vegetation canopy. The positive germination responses of seeds to alternating temperatures indicates that the species require disturbances resulting in the formation of gaps and/or in exposing seeds to the soil surface if they are buried deep into the soil or dispersed under vegetation canopy. The results agree well with the ecology of the species studied, since almost all of them are known to form persistent soil seed banks in the highlands of Ethiopia. Key words: soil seed bank, alternating temperature, leaf shade, constant temperature, light quality

1. Introduction In many tropical forests there is a persistent soil seed bank dominated by the dormant seeds of a small fraction of the species present in the community (WHITMORE 1983, GARWOOD 1989). VAZQUEZ-YANES & OROZCOSEGOVIA (1994) summarized the endogenous and external factors that have been implicated in the endurance by seeds of the deleterious forest soil environment as: (a) the presence of a dormancy mechanism that prevents fast germination, allowing seeds to be buried by litter or soil shortly after dissemination, (b) interruption of respiratory metabolism and other cellular functions of seeds, (c) the presence of a hard and/or impermeable coat that

prevents fast dehydration of seeds and diminish predation, (d) abundant seed production that may allow some seeds to survive the attacks of parasites and predators, and (e) the presence of strong chemical defences in the seeds against parasitism and predation. A number of factors such as light, temperature, moisture, oxygen, carbon dioxide, ethylene, nitrates and allelochemicals can affect germination in the field (BEWLEY & BLACK 1994, EGLEY 1995). Some species produce seeds with environmental sensors that delay germination until specific conditions of light and temperature or a combination of both occur in the environment (VAZQUEZ-YANES & OROZCO-SEGOVIA 1993, 1994). This feature is characteristic of early colonizers FLORA (1998) 193

411

r i

of gaps, and the environmental sensors help the seeds to detect changes in their surroundings that correspond to the appearance of favourable conditions for germination and establishment. For instance, in dense forests, the appearance of wide alternation of temperatures in the soil and changes in the photon flux density and quality of the light arriving at the ground are signals of canopy destruction, disappearance of litter layer, exhumation of seeds due to soil disturbance or a combination of these factors (WHITMORE 1983, GARWOOD 1989, PROBERT 1992, V AZQUEZ-YANES & OROZCO-SEGOVIA 1993, 1994, FENNER 1995). Results from recent works in the highlands of Ethiopia show that many herbaceous and a few woody species form persistent soil seed banks in forests (DEMEL & GRANSTROM 1995), cultivated fields (DEMEL 1997) and sites abandoned after cultivation (DEMEL 1998). However, except information on the species diversity and seed density in the soil, very little is known about the environmental factors that either prevent or trigger germination of seeds in the soil. Knowledge about the responses of seeds in the soil to the various environmental factors can provide a background for predicting possible outcomes in the event of disturbances and for managing the vegetation and the surrounding landscape. The objective of this study was to investigate the germination responses of twenty herbaceous species to light and temperature. These species are known to colonize disturbed forests, cultivated fields and pastures in the highlands of Ethiopia and elsewhere (Appendix 1), and most of them form persistent soil seed banks (DEMEL & GRANSTROM 1995, DEMEL 1997,1998). The germination responses of seeds to different constant and diurnally alternating temperatures as well as to various light qualities were studied in the laboratory and glasshouse by simulating the conditions under intact vegetation cover, in deep soil and in gaps.

annual rainfall is about 1 200 mm with the highest rainfall extending between June and September and a less pronounced wet period occurring in March and April. The mean annual temperature is in the range of IS-20°C (Anonymous 1988). The forest covers an area of ca 500 hectares and was originally dominated by the conifers Juniperus procera and Podocarpus falcatus, but most of the larger trees of Juniperus procera have been removed through selective cutting since the 1950s (DEMEL 1992). The soil samples were spread to a thickness of ca 1 cm on cotton cloth in plastic trays. They were incubated in a greenhouse set at a daily fluctuating temperature of 20/12 °C, which is a typical temperature range between day and night at the soil surface within the Gara Ades forest (DEMEL unpub. data). On sunny days in the summer, temperature in the greenhouse reached up to 30°C. The samples received 12 h of light each day from Osram HQIT 400 W fluorescent tubes except during the summer when there were long hours of daylight. The species selected for the study represent both dicotyledonous and monocotyledonous herbs with seed dry mass ranging between 0.02 and 1.58 mg (Table 1). Information on their habit, habitat/ecology and geographical distribution is presented in Appendix 1. For clarity, the names of the species will be mentioned only once, and thereafter the generic name will be used except the two species of Dichrocephala and Eragrostis. The germination tests were carried out in Sweden where the soil seed bank studies were carried out. Except the fresh seeds tested for germination soon after collection, all other seeds were air-dried immediately upon collection and then stored in tightly closed plastic bottles at room temperature (20°C) until they were used in the different tests (Table 1).

2.2. Germination tests The responses of seeds to various light qualities as well as different constant and diurnally alternating temperatures were investigated in different tests. The age of seeds at the time of each test is shown in Table 1.

2.2.1. Response to light

2. Materials and methods 2.1. Seed collection In this paper, the term 'seed' refers to the germinating structure that in some cases would be more accurately described as a fruit. Seeds of all species, except Verbascum sinaiticum which was collected at Menagesha State Forest (ca. 50 km west of Addis Abeba), were collected from plants grown in a glasshouse at Umea (Sweden) from soil samples collected for a seed bank analysis from a forest, cultivated fields and fields abandoned after cultivation (DEMEL & GRANSTROM 1995, DEMEL 1997, 1998) at GaraAdes. Gara Ades is located within the dry Afromontane forest belt, about 400 km east of Addis Abeba (9° 20' N, 41 ° 15' E). The bedrock in the area is volcanic and the soil is deep, stony and dark grey loam with a relatively high content of organic matter. The altitude ranges from 2 200 m to 2775 m. The mean 412

FLORA (1998) 193

The effect of light on germination was studied by incubating seeds under different light qualities in two different tests which were carried out in a laboratory and glasshouse. In the laboratory, fresh seeds (Table 1) were tested in a germination incubator at 20°C in continuous light from fluorescent tubes and in darkness. The temperature (20°C) was used to incubate the seeds since the mean daily temperature of the soil surface at Gara Ades forest is around 20°C (DEMEL unpubl. data). In the glasshouse, seeds were tested under three different light environments: in natural daylight, in light filtered through one layer of green leaves (hereafter referred to as leaf shade) and in darkness employing the method developed for a similar study (DEMEL & GRANSTROM 1997). Simultaneously, seeds (of the same age as those incubated in the glasshouse) were also incubated in the laboratory at 20°C in light and darkness to check if germination in light and dark conditions is

." 1

Table 1. List of species tested for germination with their families, mean seed dry mass and age of seeds (in months) at the time of each germination test [nomenclature follows HEDBERG & EDWARDS (1989, 1995)]. Species

Family

Seed dry Light Alternating Light Constant mass (mg) andlDark Temperatures quality Temperatures (in Laboratory) (in Glasshouse)

Asteraceae Rosaceae

0.19 0.20

Fresh Fresh

2 2

10 10

12 12

Asteraceae

0.15

Fresh

2

13

13

Asteraceae

0.13

Fresh

2

13

14

Crassulaceae

0.03

Fresh

2

11

10

Asteraceae

0.12

Fresh

6

9

Asteraceae

0.02

Fresh

2

5

Asteraceae

0.04

Fresh

3

10

10

Lobeliaceae

0.02

Fresh

7

15

17

Primulaceae

0.15

Fresh

5

Lobeliaceae

0.07

Fresh

3

11

Polygonaceae

1.58

2

8

Lamiaceae

0.74

Fresh

Asteraceae 0.41 Scrophulariaceae 0.15

Fresh Fresh

Cyperaceae

0.28

Poaceae

Dicotyledons

Acmella caulirhiza DEL. Alchemilla cryptantha A. RICH. Conyza abyssinica SCH. BIP. ex A. RICH. Conyza hochstetteri SCH. BIP. ex A. RICH. Crassula alsinoides (HoOK. f.) ENGL. Dichrocephala chrysanthemifolia (BL.) DC. Dichrocephala integrifoUa (L.f.) KUNTZE HeUchrysum foetidum (L.) MOENCH. Lobelia neumannii T. C. E. FRIES Lysimachia ruhmeriana VATKE Monopsis stellarioides (PRESL) URB. Persicaria nepalense (MEISEN) MIYABE Salvia nilotica Juss. ex JACQ. Sonchus oleraceus L. Verbascum sinaiticum BENTH.

5 8

11

4

6

14

10

Fresh

3

11

13

0.26

Fresh

5

13

15

Poaceae

0.08

Fresh

2

10

12

Poaceae

0.16

Fresh

2

Poaceae

0.22

Fresh

2

10

12

4

Monocotyledons

Cyperus sesquijlorus (TORR.) MATTF. & KUK. subsp. appendiculatus (K. SCHUM.) K. LYE Digitaria velutina (FORSSK.) P. BEAUV. Eragrostis schweinfurthii CHIOV. Eragrostis tenuifoUa (A. RICH.) STEUD. Poa leptoclada HOCHST. ex A. RICH.

affected by temperature. In the dark treatments, both in the laboratory and glasshouse, the petri dishes were wrapped with aluminium foil immediately after sowing to avoid exposure of seeds to light. In the test involving light filtered through green leaves, seeds were sown in petri dishes which were immediately placed on a table under leaves of Bergenia crassifolia (L.) FRITSCH which were arranged between two glass sheets of

60 cm x 40 cm x 3 mm. To avoid exposure of seeds to light from the different sides, the petri dishes were enclosed within a frame formed by four 3.5 cm high metal bars on which the glass sheets were placed. The leaves were replaced every second day. Lysimachia, Sonchus and E. tenuifolia were not tested due to shortage of seeds. A quantaspectrometer (QSM-2500, Techtum Instruments, FLORA (1998) 193

413

Umea, Sweden) was used to detennine the red: far-red ratio of light reaching the seeds. Photon flux densities were measured at 660 and 730 nm wavelength. The red: far-red ration of daylight was 1.12 and that ofleaf-filtered light 0.08. The average daily maximum and minimum temperatures in the glasshouse during the experiment were 28°C and 13 °C, respectively.

2.2.2. Response to temperature

dark), i.e. both in the laboratory and glasshouse, were subjected to two-way ANOVA using the computer programme SYSTAT (ANONYMOUS 1992). Following the ANOVA, Tukey's HSD multiple comparison was used to test for significant differences among treatment means (ANONYMOUS 1992). Germination results from seeds exposed to light after incubation in dark and under leaf-shade for 21 days were analysed separately. The ANOVA were performed separately for each species.

2.2.2.1. Constant temperatures Germination responses of the different species to constant temperatures were studied by incubating seeds on a thermogradient at 10, 15, 20, 25 and 30°C in continuous light from fluorescent tubes. Before the test, the seeds were stored for various periods (Table 1) to overcome or reduce the degree of dormancy that was observed during the gennination test of fresh seeds at 20°C in light. Persicaria and E. tenuifolia were not tested on the thermogradient because of shortage of seeds.

2.2.2.2. Alternating temperatures Seeds that were stored for two to seven months (Table 1) were also tested in gennination incubators at 20°C and diurnally alternating temperatures of 20112, 30/12 and 35/12°C (12 hours each) in continuous light from fluorescent tubes. The two species of Conyza were not tested at 35/12 °C due to shortage of seeds. Similarly, the two species of Dichrocephala, Salvia and Sonchus were not tested in any of the alternating temperatures because of shortage of seeds. Except Persicaria and Sonchus, all other species had four replicates of 25 seeds each in all the germination tests. In the case of Persicaria and Sonchus, five replicates of 10 seeds each were used in all the gennination tests. Seeds of Lysimachia incubated on the thermogradient were replicated five times with 10 seeds in each replicate. Except in the test involving light and dark treatments in the laboratory, all other germination tests carried out with seeds of Salvia had five replicates of 10 seeds each. In all germination tests, seeds were placed in 5.5 cm petri dishes on cotton which was kept moist with distilled water.

2.3. Assessment and statistical analysis Seeds incubated in light were inspected every day and the germinated seeds were discarded. Seeds were considered to have genninated when the radicle penetrated the seed coat. Seeds incubated under leaf shade and in the dark were inspected after 21 days of incubation and the genninated seeds were discarded. After 21 days, the seeds which were incubated in darkness and leaf shade were exposed to full daylight and the test was continued for 21 more days. Before the statistical analyses, all percentage data were arcsine transformed to approximate normality (ZAR 1984). The germination data from tests involving different constant and alternating temperatures, light and dark treatments (in the laboratory) as well as daylight, light filtered through green leaves and dark treatments (in the glasshouse) were analysed by one-way ANOVA. Germination data from tests involving both temperature (20 versus 28/13 0c) and light (light versus 414

FLORA (1998) 193

3. Results 3.1. Germination of fresh seeds Species differed considerably with respect to the capacity offresh seeds to germinate at 20°C in light (Table 2). Only six of the species, i.e. the two species of Conyza, the Table 2. Mean percent gennination (standard error) of fresh seeds incubated in light and darkness at 20°C. The seeds incubated in darkness were exposed to light after 21 days of incubation, and the test was continued for additional 21 days. The final germination data were arcsine transformed and subjected to one-way ANOVA. Significant differences of treatment means were tested by Thkey's HSD multiple comparison. Treatment means on the same row with different letters were significantly different at P ~ 0.05. The gennination data within and after 21 days of incubation were analysed separately. Species

Gennination within 21 days

Germination after 21 days

Light

Light

Dark

Dark

Dicotyledons 0 0 10 (6)a Oa C. abyssinica 98 (1.2)a 3(1.9)b C. hochstetteri 82 (3.8)a 6 (6)b C. alsinoides 0 0 D. chrysanthe- 96 (2.8)a 33 (l2.4)b mifolia D. integrifolia 87 (6.6)a Ob 11 (3.4)a Ob H.foetidum L. neumannii 0 0 L. ruhmeriana 33 (5.3)a Ob M. stellarioides 0 0 0 P. nepalense 0 84 (6.9)b S. nilotica 100 a 30 (7)b S. oleraceus lOOa 37 (18.5)a Ob V. sinaiticum

A. caulirhiza A. cryptantha

0 0 10 (6)a Oa 98 (1.2)a 92 (1.6)b 82 (3.8)a 78 (9)a 0 0 96 (2.8)a 81 (2.5)b 97 (1.9)a 94 (3.8)a 11 (3.4)a Ob 0 0 33 (5.3)a 3 (l)b 0 0 0 0 88 (8.2) a 100 a 92 (5.8)a 100 a 37 (18.5)a Ob

Monocotyledons

C. sesquiflorus 0 1 (1)a D. velutina 11 (l)a E. schweinfurthii E. tenuifolia 0 50 (2)a P. leptoclada

0 Oa Ob

0 1 (1)a 11 (l)a

0 3 (3)a Ob

0 Ob

0 50 (2)a

0 Ob

1

------~---------------------------------oj

two species of Dichrocephala, Salvia and Sonchus, showed more than 80% germination. In seven of the species germination was between one and 50%, and in seven of them, none of the fresh seeds germinated (Table 2).

3.2. Response to light In the laboratory test of fresh seeds, 11 of the species had significantly higher percent germination in light than in darkness (Table 2). Although there was statistically significant difference in the light and dark germination, fresh seeds of Salvia germinated to exceptionally high percentage (84%) in darkness (Table 2). Germination was 50% or more in light and 0-35% in darkness for the two species of Conyza, the two species of Dichrocephala, Sonchus, Poa. For Alchemilla, Helichrysum, Lysimachia, Verbascum, Digitaria and E. schwein-

furthii, germination was less than 40% in light and nil in darkness. None of the fresh seeds of Acmella, Crassula, Lobelia, Monopsis, Persicaria, Cyperus and E. tenuifolia incubated in light and darkness germinated. When the seeds from all the species initially incubated in darkness were exposed to light, germination was greatly increased in the two species of Conyza, the two species of Dichrocephala and Sonchus while very little or no increase was observed in all the other species (Table 2). This may suggest the induction of secondary dormancy for Lysimachia, Verbascum, E. schweinfurthii andPoa. In the glasshouse, germination percentages from seeds incubated in daylight, leaf shade and darkness differed significantly for each species, except Poa and Persicaria for which germination was generally low « 20%) and nil, respectively, from all treatments (Table 3). For all the species, except Poa and Persicaria,

Table 3. Mean percent germination (standard error) of seeds incubated in light and darkness at 20°C (in laboratory) and in daylight, darkness and under leaf shade at 28/13 °C (in glasshouse). The seeds incubated in darkness and under leaf shade were exposed to light after 21 days and the test was continued for additional 21 days. The final germination data were arcsine transformed and subjected to ANOVA. For the glasshouse tests involving daylight, darkness and leaf shade one-way ANOVA was used to analyse the data, while the tests involving light treatments (light versus dark) and temperature treatments (20 versus 28/13 0c) were analysed by two-way ANOVA. Following the ANOVAs, Tukey's HSD multiple comparison was used to test for significant differences between treatment means. Treatment means on the same row with different letters or asterisk marks were significantly different at P ~ 0.05. The germination data within and after 21 days of incubation were analysed separately. For Persicaria, final percent germination after scarification of seeds is given after a slash mark. Species

Germination within 21 days

Germination after 21 days

20°C

20°C

28/13°C Leaf shade

Light

Dark

Daylight

Dark

Leaf shade

1 (l)a 6 (2)a 38 (5)b* 13 (7.2)a** 23 (4.1)a Ob 51 (1.9)c* 9 (3.4)d** 96 (1.6)a 15 (4.4)b 99 (l)a* 66 (8.1)c** 96 (0) a 9 (5.8)b 93 (4.7)a* 42 (11.8)c** 38 (6.2)a 1 (1)b 69 (16.2)c* 1 (l)b** 96 (2.8)a 33 (l2.4)b 77 (4.4)a* 1 (1)d**

3 (1)** 0*** 25 (15.9)*** 0*** 0** 0**

6 (2)a 33 (3)a 96 (1.6)a 96 (O)a 40 (5.9)a 96 (2.8)a

4 (1.6)a 44 (5.4)a 96 (1.6)a 93 (5.8)a 3 (1.9)b 81 (5.2)b*

42 (3.8)b* 71 (3)b* 99 (1)a* 94 (3.8)a 99 (l)c* 84 (5.2)ab*

91 (3)c** 65 (1.9) b* 99 (l)a* 93 (3)a* 96 (4)c* 74 (7.4)b*

79 (4.1)** 82 (2.6)** 97(1.9)* 95 (3.8)* 71 (7.4)** 82 (2.5)*

87 (6.6)a 100 a 99 (1)a 84 (5.9)a

Ob Ob 24 (5.4)b Ob

60 (3.7)c* 68 (2.5)c* 56 (7.3)c* 99 (1)c*

Ob** 19 (11)b** 15 (15)b** 3 (1.9)b**

0** 0** 0** 3 (1.9)b**

97 (1.9)a 100 a 99 (1)a 84 (5.9)a

94 25 24 65

69 (2.5)b* 81 (6.6)a* 84 (3.3)* 97 (1.9)a* 80 (12)* 100a* 57 (8.1)c* 58 (3.5)c* 58 (6.8)* 95 (3)* 100b* l00b*

0 100 a 85 (lO)a

0 92 (2)b Ob

0 0 06 (2.5)ab* 100a* 93 (1.9)a* 15 (9)b**

0 10 (3.2)** 0***

0/94 l00a 85 (10)a

0/88 94 (2.5)a Ob

Light

Dark

Daylight

Dark

28/13 °C

Dicotyledons

A. caulirhiza A. cryptantha C. abyssinica C. hochstetteri C. alsinoides D. chrysanthemifolia D. integrifolia H.foetidum L. neumannii M. stellarioides P. nepalense S. nilotica V. sinaiticum

(3.8)a (8.5)b (5.4)b (15.1)a

0/90 l00a* 93 (1.9)a*

4 (2.5)/82 6 (4)/94 96 (2.5)a* 94 (2.5)* 89 (7.2)a* 77 (10.3)*

Monocotyledons C. sesquiflorus 47 28 D. velutina E. schwein82 furthii P. leptoclada 13

(5.8)a Ob (3.7)a 1 (l)b (2.6)ab 69 (8)b (3.8)a

10 (3.5)a

84 (3.7)c* Ob** 0** 3 (1.9)b** 6 (2.6)** 100c* 97 (1.9)a* 67 (8.2)b** 69 (7.6)**

53 (5)a 80 (5)a 28 (3.7)a 31 (9.3)a 82 (2.6)ab 69 (8)b

87 (4.4)a* 90 (1.9)a* 92 (2.3)* 91 (5.3)b* 60 (5.9)** 100b* 97 (1.9)a* 72 (9.9)b* 70 (6.8)*

16 (5.4)a* 14 (2.6)a*

13 (3.8)a 11 (4.4)a

16 (5.4)a*

19 (6.8)*

15 (3)a*

FLORA (1998) 193

22 (8.1)* 415

Fig. 1. Gennination of seeds at the different constant temperatures. Mean final gennination (e) and speed of germination (A) expressed as the number of days to reach 50% of final gennination. Vertical bars represent one standard error of the mean, and treatment means in a graph with different letters were significantly different at P ~ 0.05 (Turkey's HSD multiple comparison following one-way ANOVA).

percent germination from seeds incubated in daylight was significantly higher than those incubated under leaf shade and in darkness (Table 3). These results indicate the induction of secondary dormancy in seeds incubated 416

FLORA (1998) 193

under leaf shade and in darkness. Germination percentages from seeds incubated in darkness were significantly higher than those incubated under leaf shade for Alchemilla, the two Conyza species, Salvia and Verbas-

~

I

C. sbyssht;8

AK:I1emII/s 100

a

b

c

C. I10cIJstettsn

a

a b

eo

a

a

c

eo

a b

20

a

a CrsSSIM

b

b

100

Lysimac1i8

LobeIJs

Ht;/icIJrysun

b

b

eo

c

a

80

c c

c

o

to

c

E "m

C)

a d

:p

It1Kb8scun

b

100

a

b

a

d CyP9IVS

a

a

b

b

eo eo

a

c a Dlgtsr18

a

a

E sclJwelntr.rtIJII b

100

a

b

a Etentlfoll8

Poe

b

b

b

eo

b

b

b

a

eo

a a

a

a

a

b

Temperature (°0) Fig. 2. Mean percent germination of seeds at a constant (control) and diurnally alternating temperatures. Vertical bars represent one standard error of the mean, and bars in a graph with different letters were significantly different at P::; 0.05 (Turkey's HSD multiple comparison following one-way ANOVA).

cum. For all the other species, there were no significant differences in germination from seeds incubated under leaf shade and in darkness (Table 3). The species can be grouped into four categories based on the results obtained from the two-way ANOVA of germination data (Table 3) from tests involving light and dark treatments at 20 (in the laboratory) and 28/13 °C (in

the glasshouse); (a) species that showed significant differences in percent germination from both light and temperature treatments as well as from the interaction between light and temperature treatments: C. abyssinica, C. hochstetteri, Crassula, D. integrifolia, Helichrysum, Monopsis, Cyperus and Digitaria; (b) species that showed significant differences in percent germinaFLORA (1998) 193

417

'f'"

tion from both light and temperature treatments but not from the interaction between light and temperature treatments: Acmella, Alchemilla, D. chrysanthemifolia and Lobelia; (c) species that showed significant differences in percent germination only from the light treatment but not from temperature and interaction between light and temperature treatments: Salvia, Verbascum and E. schweinfurthii; (d) species that showed no significant differences from any of the treatments: Poa and Persicaria. Percent germination of seeds of Alchemilla, the two species of Conyza and Helichrysum incubated in darkness at 28/13 DC was significantly higher than those incubated in darkness at 20 DC. When the seeds from all the species incubated under leaf shade and in darkness were exposed to light, germination was either as high as those initially incubated in light or substantially increased for Acmeila, Alchemilla, the two species of Conyza and Dichrocephala, Monopsis, Salvia, Cyperus and Digitaria (Table 3). Seeds of Crassula, Helichrysum, Lobelia and Verbascum incubated initially in darkness in the laboratory at 20 D showed little or no increase in germination when exposed to light. However, seeds of the same species initially incubated under leaf shade and in darkness at 28/13 DC exhibited great increase in percent germination or as high germination percent as those initially incubated in daylight (Table 3).

3.3. Response to temperature 3.3.1. Constant temperatures The species responded differently to the various constant temperatures on the thermo gradient. Seeds of the two species of Conyza, Lobelia, Monopsis, Sonchus and Verbascum germinated to more than 50% at all of the constant temperatures whereas for Acmella, Alchemilla, Crassula, Lysimachia, Cyperus, Digitaria and Poa germination was less than 50% at the different constant temperatures (Fig. 1). Germination was more than 50% at the constant temperatures between 15 and 30 DC but lower than 43% at 10 °C for D. integrifolia, Helichrysum, Salvia and E. schweinfurthii (Fig. 1). Seeds of D. chrysanthemifolia germinated to more than 50% at the constant temperatures between 10 and 25 but germination declined to 18% at 30°C (Table 3). Speed of germination, defined here as the number of days elapsed to reach 50% of final germination at the different constant temperatures, increased as the temperature increased for 12 of the species (Fig. 1). For C. hochstetteri, D. integrifolia, Helichrysum, Lysimachia and Salvia speed of germination increased as the temperature increased up to 25°C and then started to decline. In the case of D. chrysanthemifolia, speed of germination increased up to 20°C and declined thereafter. 418

FLORA (1998) 193

3.3.2. Alternating temperatures The responses of species to the constant and alternating temperature treatments were diverse (Fig. 2). Germination was similar at the constant and alternating temperatures for C. hochstetteri and Verbascum. In the case of C. abyssinica germination was 98% both at 20 and 20/12 °C but declined at 30/12 DC. Germination was less than 15% and 20°C but increased with increasing alternation in temperature for Digitaria and the two species of Eragrostis. For Acmella and Monopsis, germination increased with increase in alternation of temperature up to 30/12 DC but was completely inhibited or nearly so at 35/12 DC. Germination was nil or less than 45% at 20°C, increased to more than 85% at 20/12°C and declined thereafter for Alchemilla, Crassula, Helichrysum, Lobelia, Lysimachia and Cyperus. For Poa, germination was almost completely inhibited at 35/12 while it was similar at 20, 20/12 and 30/12 °C (50% or less). Persicaria did not respond to any of the treatments.

4. Discussion The results in this study should be interpreted coutiously since the experimental conditions used were simple compared with the complex interactions of light and temperature regimes that the seeds experience in their natural habitats. However, the results can provide useful clues concerning the important roles played by light and temperature in the germination and regeneration ecology of the species.

4.1. Capacity to immediate germination and response to dry storage The results showed that in 70% of the species tested, fresh seeds were dormant with less than 50% germination in light at 20°C while 30% of the species had more than 80% germination (Table 2). The possession of dormancy mechanisms at the time of dispersal facilitates burial of seeds into the soil. The species that had dormant fresh seeds, a feature of species with persistent soil seed banks, required alternating temperature to break the dormancy in the seeds (Table 3). However, after dry storage of eight to fifteen months (Table 1), germination reached more than 80% in five of the species that had dormant fresh seeds, i.e. Helichrysum, Lobelia, Monopsis, Verbascum, E. schweinfurthii (Table 3 and Fig. 1), indicating that seeds were released from their dormancy during dry storage. The species which showed the capacity for immediate germination had light and temperature requirements which suggests that germination of freshly-dispersed seeds may be prevented by the intervention of limiting factors operating in the

T i.

field. This is evident from the fact that at least three of the species that exhibited high initial germinability are known to accumulate persistent seeds in the field (DEMEL & GRANSTROM 1995, DEMEL 1996, 1997, 1998). Comparison of freshly-collected and stored seeds of most of the species with respect to final germination percentage reveals a marked increase in germinability with storage. Notable examples are Helichrysum, Lobelia, Monopsis, Verbascum and Cyperus, for which storage was associated with progressive increase in germination percentage (compare Table 2 with Table 3 and Fig. 1).

4.2. Response to light Germination of seeds incubated in darkness was prevented completely or significantly reduced both at 20°C and 28113 °C in almost all of the species tested (Tables 2 and 3). Exceptions were Salvia (Tables 2 and 3) and E. schweinfurthii (Table 3) for which germination in darkness was more than 80 and 65%, respectively. Similarly, leaf shade inhibited or significantly suppressed germination in all the species tested except Poa. These results indicate that germination of these species would be prevented if their seeds are buried in deeper soil or dispersed under the canopy of vegetation, characteristic features of species forming persistent soil seed banks (THOMPSON & GRIME 1979, GRIME et al. 1981, WHITMORE 1983, GARWOOD 1989, PONS 1992, FENNER 1995). Germination from deeper soil where emergence would be difficult or under the canopy of vegetation where competition with established plants is great could be fatal to the developing seedlings (PONS 1992). Dependence on light for germination is believed to be correlated with size of seeds, and germination in darkness declines progressively as the size of seeds decreases (GRIME et al. 1981, FENNER 1995, DEMEL & GRANSTROM 1997). The requirement of light by small seeded species would avoid germination too deep in soil for the seedlings to reach the surface on the available nutrient reserves (PONS 1992). The results from the present study support this hypothesis; 18 out of 20 species, having seeds ranging between 0.02-1.6 mg, showed a light-requirement for germination (cf. Tables 2 and 3). With the exception of Verbascum, all of the other species are known to form soil seed banks in forests, cultivated fields and sites abandoned after cultivation (DEMEL & GRANSTROM 1995, DEMEL 1996, 1997, 1998). Although no information is available concerning the soil seed bank characteristics of Verbascum, the response of its seeds incubated in darkness and under leaf shade (Tables 2 and 3) suggest that the species is capable of accumulating long-lived seeds in the soil. After burial into the soil, the successful germination and

establishment of these species depends on the appearance of disturbances which result in the removal of vegetation canopy and/or exhumation of the seeds to the soil surface. In a previous study (DEMEL & GRANSTROM 1997), we have shown that 10 out of 12 of the forest species tested from the highlands of Ethiopia exhibited leaf shade induced dormancy, and this feature is quite common in many tropical species (AMMINUDDIN & NG 1982, FENNER 1980a, 1980b OROZCO-SEGOVIA et al. 1987, OROZCO-SEGOVIA & VAZOUEZ-YANES 1989, RAICH & GONG 1990, VAZQUEZ-YANES & OROZCOSEGOVIA 1982, 1990 VAZQUEZ-YANES & SMITH 1982). Since seeds of Persicaria did not respond to any of the light treatments, they were mechanically scarified by removing about 2 mm of the seed coat by the sharp edge of a pincer. Seeds from the different treatments germinated between 82-94% (Table 3) within 15 days after scarification, indicating that the hard seed coat blocked the entrance of water to the seeds, thereby preventing germination. In the field, an impermeable seed coat has been implicated to delay germination, facilitate incorporation of seeds into the soil and extend the longevity of seeds in the soil (FENNER 1985, BASKIN & BASKIN 1989, BEWLEY & BLACK 1994).

4.3. Response to temperature Due to the diversity of germination responses to the different constant temperature regimes (Fig. 1) it was difficult to establish optimum temperatures for germination which are common to all or groups of species. Of the 18 species tested at the five constant temperature regimes on the thermogradient (10-30°C), 11 germinated to more than 50% at four of the constant temperatures (Fig. 1). GRIME et al. (1981) found that the capability of germinating over a wide range of temperature was evident in species from dry habitats, and they considered the relative insensitivity of seeds to temperature as characteristic of species in which water supply acts as the primary determinant of the timing of germination in the field. The results from the present study seem to support this hypothesis, since several of the species that germinated in a wide range of temperatures grow in areas that experience a dry season of up to seven months annually. Germination was faster for most of the species incubated at the different constant temperatures on the thermogradient (Fig. 1). This agrees with the ecology of the species. For gap colonizing species, success in seedling establishment, survival and reproduction during later stages of the life-cycle depends on rapid exploitation of temporarily favourable conditions (GRIME et al. 1981). Therefore, speed of germination and the rate of seedling development play important roles in the regeneration FLORA (1998) 193

419

T ecology of the species studied here, since they are gap colonizers. Except for five of the 16 species, at least one of the alternating temperatures resulted in significantly higher percent germination than the constant temperature (Fig. 2). Verbascum was exceptional among the species tested in that it had similar percent germination at 20°C and at the three alternating temperatures. For nine of the species, germination was either inhibited or severely reduced at 35/12 °C (Fig. 2), suggesting that either the amplitude or 35°C is too high for germination. The positive responses of species to alternating temperatures indicates that the occurrence of alternating temperatures, characteristic of gaps, is a cue for germination of the seeds, and this feature conforms with the ecology of the species (Appendix 1). All of the species tested here are adapted to colonizing gaps or disturbed areas where the amplitude of diurnal alternation in temperature becomes greater near the soil surface following removal of the insulating canopy or litter. Alternating temperatures play an important role in helping seeds to detect the appearance of a gap and to sense their depth in the soil (THOMPSON et al. 1977, GRIME et al. 1981, THOMPSON & GRIME 1983, VAZQUEZ-YANES & OROZCO-SEGOVIA 1993, 1994). The requirement of alternating temperatures, which is also considered to represent an adaptation of small-seeded species, ensures that germination occurs close to the soil surface in vegetation gaps. The results provide evidence that the germination of seeds of the species tested is controlled by the light quality and temperature conditions to which they are exposed. This implies that for the successful germination and establishment of seedlings in the field, the appearance of gaps free of vegetation cover and with marked amplitude of alternating temperatures are of critical importance.

5. Acknowledgements I am grateful to my wife, MEKDES MULU, for assisting me in the laboratory and glasshouse. I also thank Dr. MESFIN TADESSE, Dr. M. THULIN and Dr. O. RYDING for allowing me to use their unpublished works, and for confirming the identity of some of the species used in the study.

6. References AMMINUDDIN, B. M., & NG, F. S. P. (1982): Influence of light on germination of Pinus caribaea, Gmelina arborea, Sapium baccatum and Vitex pinnata. Malay. For. 45: 62-68. Anonymous (1988): National atlas of Ethiopia. Ethiopian Mapping Authority, Addis Ababa. - (1992): SYSTAT for Windows: Statistics, Version 5 Edition. SYSTAT, Inc., Evanston, IL. 420

FLORA (1998) 193

BASKIN, 1. M., & BASKIN, c. c. (1989): Physiology of dormancy and germination in relation to seed bank ecology. In: LECK, M. A., PARKER, V. L., & SIMPSON, R. L. (eds.): Ecology of soil seed banks. Academic Press, San Diego, California, 53-66. BEWLEY,1. D., & BLACK, M. (1994): Seeds: physiology of development and germination. Plenum Press, New York & London. BOULOS, L. (1976). Sonchus. In: TuTIN, T. G., HEYWOOD, V. H., BURGES, N. A., MOORE, D. M., VALENTINE, D. H., WALTERS, S. M., & WEBB, D. A. (eds.): Flora Europaea. Vol. 4. Cambridge University Press, Cambridge, 327-328. CUFODONTIS, G. (1952-1973): Enumeration plantarum Aethiopiae spermatophyta. Bull. Jard. Bot. Etat Brux. Bruxelles, Vols. 23-42. DEMEL ThKETAY (1992): Human impact on a natural montane forest in south-eastern Ethiopia. Mountain Res. Developm. 12: 393-400. - (1996): Seed ecology and regeneration in dry Afromontane forests of Ethiopia. PhD thesis, Swedish University of Agricultural Sciences, Umea. - (1997): The impact of clearing and conversion of dry Afromontane forests into arable land on the composition and density of soil seed banks. Acta Oecol. 18: 00-00 (in press). - (1998): Soil seed bank at an abandoned Afromontane arable site. Feddes Repertorium 109: 161-174. -, & GRANSTROM, A. (1995): Soil seed banks in dry Afromontane forests of Ethiopia. J. Veg. Sci. 6: 777-786. - & GRANSTROM, A. (1997): Germination ecology of forest species from the highlands of Ethiopia. J. Trop. Ecol. 14: 793-803. EGLEY, G. H. (1995): Seed germination in soil: dormancy cycles. In: KIGEL, 1., & GALILI, G. (eds.): Seed development and germination. Marcel Dekker, New York, 529-543. FENNER, M. (1980a): Germination tests on thirty-two East African weed species. Weed Res. 20: 135-138. - (1980 b): The inhibition of germination of Bidens pilosa seeds by leaf canopy shade in some natural vegetation types. New Phytol. 84: 95-101. - (1985): Seed ecology. Chapman and Hall, London and New York. - (1995): Ecology of seed banks In: KIGEL, J., & GALILI, G. (eds.): Seed development and germination. Marcel Dekker, New York, 507-528. FICHTL, R., & ADMASU ADI (1994): Honeybee flora of Ethiopia. Margraf Verlag, Weikersheim. GARWOOD, N. C. (1989): Tropical soil seed banks: a review. In: LECK, M. A., PARKER, V. L., & SIMPSON, R. L. (eds.): Ecology of soil seed banks. Academic Press, San Diego, California, 149-209. GRIME, J. P., MASON, G., CURTIS, A. V., RODMAN, J., BAND, S. R., MOWFORTH, M. A. G., NEAL, A. M., & SHAW, S. (1981): A comparative study of germination characteristics in a local flora. J. Ecol. 69: 10 17 -1059. HAINES, R. w., & LYE, K. A. (1983): The sedges and rushes of East Africa. East African Natural History Society, Nairobi.

---rI

HEDBERG, I., & EDWARDS, S. (eds.) (1989): Flora of Ethiopia. Vol. 3 Addis Ababa University, Addis Ababa and Uppsala University, Uppsala. - (1995): Flora of Ethiopia and Eritrea. Vol. 7 Addis Ababa University, Addis Ababa and Uppsala University, Uppsala. MESFIN TADESSE, & REILLY, T. (1995): A contribution to studies on Helichrysum (Compositae-Gnaphaliae) - a revision of species of north-east tropical Africa. In: HIND, D. J. N., JEFFREY, C., & POPE, G. V. (eds.): Advances in Compo sitae systematics. Royal Botanic Gardens, Kew,

379-450. OROZCO-SEGOVIA, A., & VAZQUEZ-YANES, C. (1989): Light effect on seed germination in Piper L. Acta Oecol. Oecol. Plant. 10: 123-146. -, - COATES-EsTRADA, R., & PEREZ-NASSER, N. (1987): Ecophysiological characteristics of the seed of the tropical forest pioneer Urera caracasana (Urticaceae). Tree Physiol. 3: 375-386. PHILLIPS, S. (1995): Poaceae (Gramineae). In: HEDBERG, I., & EDWARDS, S. (eds.): Flora of Ethiopia and Eritrea. Vol. 7. Addis Abeba University, Addis Abeba and Uppsala University, Uppsala. PONS, T. L. (1992): Seed responses to light. In: FENNER, M. (ed.): Seeds: the ecology of regeneration in plant communities. C.A.B International, Wallingford, 259-284. PROBERT, R. J. (1992): The role of temperature in germination ecophysiology. In: FENNER, M. (ed.): Seeds: the ecology of regeneration in plant communities. C.A.B International, Wallingford, 285-325. RAICH, 1. & GONG, W. K. (1990): Effect of canopy openings on tree seed germination in a Malaysian dipterocarp forest. J. Trop. Ecol. 6: 203 - 217.

w.,

THOMPSON, K., & GRIME, J. P. (1979): Seasonal variation in the seed bank of herbaceous species in ten contrasting habitats. J. Ecol. 67: 893-921. -, - (1983): A comparative study of germination responses to diurnal fluctuating temperatures. 1. Appl. Ecol. 20:

141-156. -, -, & MASON, G. (1977): Seed germination in response to diurnal fluctuations of temperature. Nature, London 267:

147-149. THULIN, M. (1979): Monopsis (Lobeliaceae) in tropical Africa. Bot. Notiser 132: 131-137. VAZQUEZ-YANES, c., & OROZCO-SEGOVIA, A. (1982): Germination of the seeds of a tropical rain forest shrub, Piper hispidum Sw. (Piperaceae) under different light qualities. Phyton 42: 143-149. -, - (1990): Ecological significance of light controlled seed germination in two contrasting tropical habitats. Oecologia 83: 171-175. -, - (1993): Patterns of seed longevity and germination in the tropical rain forest. Ann. Rev. Ecol. and Syst. 24: 69-87. -, - (1994): Signals for seeds to sense and respond to gaps. In: CALDWELL, M. M., & PEARCY, R. W. (eds.): Exploitation of environmental heterogeneity by plants: ecophysiological processes above and below ground. Academic Press, New York, 209-236. -, & SMITH, H. (1982): Phytochrome control of seed germination in the tropical rain forest pioneer trees Cecropia obtusifolia and Piper auritum and its ecological significance. New Phytol. 92: 447-485. WHITMORE, T. C. (1983): Secondary succession from seed in tropical rain forests. For. Abstr. 44: 767-779. ZAR, J. H. (1984): Biostatistical analysis. Prentice-Hall, Englewood Cliffs, New Jersey.

Appendix 1. Habit, habitat/ecology and geographical distribution of the species tested.

Species

Habit and HabitaUEcology

Geographical Distribution

- Perennial stoloniferous herb, 20-50 cm high; lower parts of stem creeping and upper parts ascending and reaching up to 40 cm; - A common weed of usually wet places, grassy slopes, forest floors, stream banks; occasionally also found in cultivated fields and in evergreen scrub; 550-2600 m. - Weak annual herb, rooting at lower nodes; - On moist shaded ground in montane forests and along streams, sometimes as weed in lawns; 1500-3200 m. - Bushy perennial herb or shrub, 0.75-1.5 m high; - Montane slopes with Hypericum, Helichrysum, Kniphofia, etc. ; dry hillside with dense shrubs and scattered tress; 1850-3300 m . - Perennial herb with usually 2-6 erect or ascending stems from the base, 50 - 80 cm high; - Open evergreen bushland, grazed pasture, grassy or rocky hillsides, open Combretum woodland, Acacia woodland, margins of cultivation; 1660-3050 m.

- Widespread in many tropical and subtropical countries.

Dicotyledons A. caulirhiza

A. cryptantha

C. abyssinica

C. hochstetteri

- Cameroon, Ethiopia, Kenya, Madagascar, Sudan, Tanzania, Uganda and south to Transvaal - Only known from Ethiopia.

- Cameroon, Ethiopia, Kenya, Malawi, Nigeria, Rwanda, Socotra, Somalia, Sudan, Tanzania, Uganda, Zaire, Zambia, Zimbabwe, S. Arabia and Yemen.

FLORA (1998) 193

421

Appendix 1. (continued)

Species

Habit and UabitatJEcology

Geographical Distribution

C. alsinoides

Creeping perennial herb, rooting from lower nodes; - Usually in upland forest with Juniperus/Hagenia, sometimes on boulders or fallen logs, mostly in shade; 1500-3100 m. - Erect stiff and widely branching annual or shortlived perennial herb, 30-75 cm high; - Damp places in grassland, forests, amongst grass along stream and river banks, occasionally weed of cultivated and fallow fields or waste ground; 1830-3750 m. - Erect or ascending annual herb, 30-75 cm high; - Weed of arable land, fallow and waste land, normally growing in wet spots and depressions, swampy ground in forests by stream banks and along water courses, rarely in open rocky places; 1750-3500 m. - Erect annual or short-lived perennial herb, 0.5-1.2 m high: - Disturbed places, roadsides, along forest paths, open Acacia scrub, wooded grassland, margins of Erica bushland, Margins of mixed evergreen bushland and cultivation; (200-)1700-3890 m. - Annual erect to decumbent herb; stems 3-30 cm high, rooting at lower nodes; - Upland grassland and scrub, usually on bare or rocky ground, or in cultivation; 1 650- 2900 m. - Erect annual herb, about 15 cm high; - Disturbed areas; ca. 2400 m.

- Cameroon, Ethiopia, Fernando Po, Madagascar, Sudan, south-east to east South Africa, Yemen and Zaire.

D. chrysanthemifolia

D. integrifolia

H. foetidum

L. neumannii

L. ruhmeriana

M. stellarioides

- AnnuallPerennial decumbent herb, often stoloniferous; stems 5-60 cm high, usually rooting at lower nodes; - Upland grassland and forest margins;

P. nepalense

- Weak-stemmed straggling or erect annual herb growing to 30 cm high; - A common weed in highland crops, sometimes dominant, also on fallow land; 1350-3200 m.

s. nilotica

- perennial rhizomatous herb; - Very common in moist and semi-arid parts of Ethiopia; in grassland or bushland, usually on somewhat moist or shaded places; usually on disturbed ground, on meadows and along roadsides; 1 300-3 800 m - Erect, annual herb, up to 1 m high; - Road side margins, grassy fields, cultivated ground and waste places; 1650-2440 m. - Erect woolly herb to 2 m high; - On disturbed ground, rocky places, along roads and as a weed in fallow ground; 900-3000 m.

- In Africa, west to Guinea and south to Malawi; also in Yemen, S. Arabia, India and Java.

- Widespread in Africa, Arabia and Asia.

- Burundi, Cameroon, Ethiopia, Fernando Po, Ghana, Kenya, Malawi, Nigeria, Rwanda, Saudi Arabia, Somalia, South Africa, Spain, Sudan, Tanzania, Uganda, Yemen, Zaire and Zambia. - Cameroon, Ethiopia, Kenya and Sudan.

- Angola, Cameroon, Congo, Ethiopia, Kenya, Madagascar, Malawi, Mozambique, South Africa, Sudan, Tanzania, Uganda, Zimbabwe and Afghanistan. - Burundi, Cameroon, Comoro Islands, Ethiopia, Fernando Po, Kenya, Malawi, Rwanda, Tanzania, Uganda, Zaire and Zambia.

1 200-3 200 m.

s. oleraceus V. sinaiticum

- Congo, Ethiopia, Fernando Po, Kenya, Madagascar, Malawi, RwandalBurundi, South Africa, Tanzania, Uganda, Afghanistan, China, India, Indonesia, Japan, New Guinea, Philippines and Sri Lanka. - Ethiopia, East and Central Africa, Malawi, Zambia and Zimbabwe.

- A cosmopolitan weed native to Eurasia and N. Africa; also found in Sudan, Somalia and Socotra. - Egypt, Ethiopia, Kenya, Sudan and the Middle east.

Monocotyledons C. sesquiflorus

422

- Perennial herb with mostly crowded culms, 10-80 cm long, growing from a creeping rhizome;

FLORA (1998) 193

- Ethiopia, Kenya, Tanzania, Uganda, ...

Appendix 1. (continued)

Species

Habit and HabitatJEcology

C. sesquiflorus

- In damp shady woodland-margins, along forestroads, in upland forest, including bamboo, where light-intensities are high; 1200-2550 m. - Loosly tufted to straggling annual with slender culm, 20-60 cm high and rooting at lower nodes; - A weed of disturbed and arable land, usually on light soils in shade; 700- 2000 m. - Slender annual or short-lived perennial with culms up to 35(-55) cm high; - Short grassland on shallow soils often along drainage channels, and in clearings in upland forest; 1800-3600m. - Tufted perennial with strongly flattened basal leaf-sheaths; culm erect, 20- 90 cm high; - Disturbed ground; roadsides, field margins and overgrazed pastureland; 1400-2500 m. - Slender, erect or weak and ascending tufted perennial; culms 12-80 cm high; - Disturbed ground, woodland, among rocks on mountains, preferring moist shady situations; 2300-4050 m.

D. velutina

E. schweinfurthii

E. tenuifolia

P. leptociada

Geographical Distribution

- Ethiopia, Egypt and Yemen to South Africa.

- East Africa, including Ethiopia and Yemen.

- Tropical Africa, also in India, Australia and South America.

- Cameroon, Ethiopia, southwards through East Africa to Zimbabwe, Sudan and Yemen.

Sources: CUFODONTIS (1953-1972); HAINES & LYE (1983); MESFIN & REILLY (1995); MESFIN (unpubl.); PHILLIPS (1995); RYDING (unpubl.); THULIN (1979, unpubl.); BOULOS (1976).

FLORA (1998) 193

423