Ana səhifə

Supplementary information Experimental Sites and Vegetation


Yüklə 213.5 Kb.
tarix27.06.2016
ölçüsü213.5 Kb.
Supplementary information

1. Experimental Sites and Vegetation

The experimental sites were in the habitats of Holcoglossum amesianum located in a mountain area of Yangmachang Village (22°46’N, 100°36’E, alt 1600-1900m) in Longtan Town, Simao City, Yunnan Province of China, where November to April is the drought season (Yan, H. M., Wang, H. Q. & Yang, H. M. Species diversity of spiders in rice fields in China. Advances in Biodiversity Research 440-445 (China Science & Technology Press, Beijing, 1995)). We carried out the studies on pollination biology of H. amesianum there from February to April each year in 2003, 2004 and 2005. During observation, ambient air temperature was 24.3±3.7℃, relative humidity 31.5±3.0% and it was usually windless (n=60).



H. amesianum grows on the tree trunks in evergreen forests of broad-leaved trees or mixes of broad- and narrow-leaved trees in the mountain area. The canopy density of the plant population is 0.5-0.8, the slope ca. 45°, and the average height of forest stand 3-5m. Species in the forests include: Castanopsis delavayi Fr.,Ficus semicordata Buch.-Ham. ex J. E. Sm., Engelhasdtia colebrookiana Lindl. ex Wall.,Premna bracteata Wall.,Alnus nepalensis D.Don,Syzygium cuminii (Linn.) Skeels,Bauhinia variegata Linn. var. candida (Roxb.) Voigt,Salix tetrasperma Roxb.,Glochidion daltonii (Muell. Arq.) Kurz,Colon floribunda (Wall.) Crail.,Casearia graveolens Dalz.,Callicarpa arborea Roxb.,Meliosma wallichii Planch. ex Hook. f.,Cyclobalanopsis kerrii (Craib) Hu,Pinus khasya Roylect Gord.. H. amesianum is mainly epiphytic on the trunks of E. colebrookianaA. nepalensis and G. daltonii.

2. Description of Biological Characteristics of H. amesianum

H. amesianum is epiphytic, perennial herb, with thick fleshy roots. Stem usually 3-5cm long, 2cm thick, unbranched, enclosed in sheathing leaf bases. Leaves 4-7, fleshy, linear-tapering, 9-30cm long, 5-10mm broad, usually conduplicate with channels, sheathing at base. Inflorescences suberect, 23-25cm long, usually unbranched, with 1-10 flowers. Flowers 2.5cm across, thin-textured, white to pale pink, with pink mid-rib on sepals and petals, and purplish-red veins and ridges on lip. Dorsal sepal elliptic, 1.5cm long, 1cm broad; lateral sepals ovate-elliptic, similar in size to dorsal sepal. Petals obovate, 1.4cm long, 9mm broad; lip 3-lobed, 1.5cm long; side-lobes ovate-deltoid, 4mm long, 5mm broad; mid-lobe kiney-shaped, 1cm long, 1.4cm broad, retuse apically, denticulate-margined, with a sitee appendage at base near the mouth of the spur; disc with 3 ridges; spur sacculated, 6mm long, slightly backwards; column 6mm long; anther cap single-locular; pollinia 2, cleft.

3. Materials and Methods

3.1 Observation of floral morphology and structure and flowering phenology during blooming periods

3.1.1 Observation of flower morphological and structural characters

In each of the 10 populations of H. amesianum studied in nature, we selected more than 20 blooming plants randomly, observed and recorded the following parameters: number of flowers (buds) of each inflorescence, length of inflorescence, number of inflorescences of each plant, length and changes in color and size of petals.



3.1.2 Phenological observation in blooming period

In the populations of H. amesianum studied in nature, we observed and recorded the phenological states of all plants in blooming period from Feb to Apr of 2003, 2004 and 2005, including the early blooming period, peak blooming period and late blooming period. We randomly selected and marked 20 unbloomed inflorescences in each population, observed and recorded every 2 hours from 6:00 to 18:00 the blooming peculiarities of each single flower and single plant, including the time of opening, closing and (corolla) fading, the time of anther cap opening, the morphological changes of labellums and petals, the number of opening flowers on each inflorescence of each plant every day, and the opening sequence of the flowers on each inflorescence.



3.1.3 Observation of morphological structure, position and movement of floral parts.

At selected time points, we picked the opening flowers of H. amesianum, dissected (when necessary) and examined the floral parts under microscope, observed and recorded their morphological characters and the relative positions of anther caps, stipes, pollinia and stigmas, and took photos of them. Meanwhile, we carefully observed the pollination process of the flowers in the field sites, and took photos of them to record the morphology and movement of floral parts, especially the anther cap, stipe and pollinia, such as those shown in Fig. 1.



3.2 Observation of the pollination process

Each year from 2003 to 2005, we randomly selected 18 observation sites in 10 different populations of H. amesianum. There were 48 valid sites in total in three years, and about 40 flowers at each site. Beginning with the first open flower, we observed from 6:00 to 18:00 every day the pollination states of each flower during its entire blooming period, until all flowers of the sites had opened and gone through the pollination process. A total of 1911 flowers were observed in three years.



3.2.1 Observation of the blooming state and pollination behaviors

We recorded the blooming state, pollination state and fruiting state of the flowers in each site every day, and continued the observation until all flowers of the sites have opened.



3.2.2 Observation of species and behaviors of visiting insects or other animals

We observed continuously and kept track of any insects or other animals visiting the flowers of each inflorescence at each site. If any visitor came, we took photos of its visiting behavior and recorded the following: the counts, description, and species of the visitor to the flowers on each inflorescence; the number of visits, duration of stay, and the number of flowers per plant and of plants that the same visitor had visited. Finally the insects were caught and made into specimens as evidence.



3.2.3 Measurement of secreted nectar and detection of released odor

From 6:00-19:30 of every day, we checked the volume of nectar in the spur of flowers in the sites. The nectar volume of flowers, both unbagged or bagged before blooming, was measured with the 5-10μl micropipette. Odor released from the floral spur was detected by careful smelling.



3.2.4 Observation and quantification of fruit set and seeds in fruits

Successful fruit set following pollination is easily identified by morphology of the non-fallen floral organs and enlargement of the fruit. The rate of natural fruit set was determined based on the observation data from the 1911 flowers. And each year 10 fruits were randomly selected and the seeds from each fruit were observed and counted under microscope.



3.3 Test of breeding system

3.3.1 Experiments of manual self-pollination and manual cross-pollination

A total of 18 sites (9 pairs) for controlled manual self-pollination and manual cross-pollination were randomly selected from year 2003 to 2005 (6 sites, 3 pairs each year). There were 40-60 flowers in each site.



Manual self-pollination: The flower was bagged before blooming. After blooming but before fertilizing, the bag was opened temporarily, the pollinarium (pollinia and stipe) of the flower was peeled off and put into its own stigma cavity, and the flower was bagged again. Changes in the flower and the state of fruit setting were recorded.

Manual cross-pollination: Flowers on the paired plants in the same site were bagged before blooming. After blooming but before fertilizing, the bags were opened temporarily, pollinarium of one flower was peeled off and put into the stigma cavity of a different flower on the other plant, and vice versa, and then the flowers were bagged again. Changes in the flower and the state of fruit setting were recorded.

3.3.2 Experiments of natural self-pollination and bagged self-pollination

A total of 36 sites (18 pairs) for controlled natural self-pollination and bagged self-pollination sites were randomly selected from 2003 to 2005 (12 sites, 6 pairs each year). There were about 40 flowers in each site. Natural self-pollination: the flower was not manipulated; Bagged self-pollination: the soon-to-bloom flower was enclosed with a transparent bag to avoid insect entering. State of pollination and fruit setting of both samples were observed and recorded.



3.3.3 Experiments of natural and bagged asexual reproduction

A total of 24 (12 pairs) controlled sites in total (8 sites, 4 pairs each year) were selected randomly (2 of the sites were later discarded). There were about 40 flowers in each site, and all of them were bagged before blooming. After blooming but before fertilizing, the pollinarium of each flower in both of the paired sites was peeled off and discarded; then the flowers in one site were bagged, while those in the other site were left untreated. State of pollination and fruit setting were observed and recorded.



3.3.4 Experiments to assess apomixis

Identification of antipodal cells. Five alabastrums were randomly collected from each quadrat, and put into holly oil to make the ovules transparent, then stained lightly with hematine crystal and examined under microscope to determine whether antipodal cells were present in the embryo of ovary (Ma, S.M. et al. The methods for identification of apomixis in plant. Acta Bot. Boreal.-Occident. Sin. 22, 985-993 (2002)).

Karyotype determination. In each site, root tips of 5 individuals each were obtained from mature plants, seedlings and, in the second and third year, seedlings germinated from the seeds produced in the previous year by natural self-pollination. The tissue samples were treated and chromosome number and morphology were analyzed as described (Akpan, G.A. & Hossain, M.G. Karyotypes and evolutionary relations of Hibiscus asper Hook., H. Cannabinus L. and H. surattensis L. (Malvaceae). Botanical Journal of the Linnean Society 126, 207-216 (1998)).

4. Data Analysis

Statistic analysis of all the collected data was conducted using SPSS Software, Version 10.0.



4.1 Phenological and morphological characters of the blooming flowers

Usually on an inflorescence only one flower opens every day, and it lasts for 6-8 days. Following successful pollination the corolla turns pale and wilts but remains on the young fruit; without pollination the whole flower withers and falls off from the inflorescence. The fruit is long spindle-shaped, 4.5-6.5 cm long, 0.5-1.2 cm in diameter.



4.2 Pollination mechanism and statistic results of fruit set and seeds

During the 3 seasons of observation from 2003 to 2005, 957 out of 1911 flowers completed fertilizing and fruiting at 48 sites in 10 populations, and the rest 954 flowers faded. During the whole flowering periods, we did not witness any insects or other animals that would touch the anthers and stigmas, and did not find any of the flowers secrete nectar or odor. The quantitative results from observation are shown in Table 1.

In the 957 flowers that eventually succeeded in fertilizing and fruiting, we observed only one common mode of pollination accomplished by the same movement of floral parts. After a flower opened fully, the stigma secreted mucus, the anther cap fell off from the column, the stipe with the pollinia on its tip rose up from the clinandrium, the flexible stipe curved front-downwards, taking the pollinia across the edge of the restellum and curved back-upwards and insert it into the stigma cavity whose receptive surface is facing downwards but above the rostellum, and the stipe curved further to form a ring (1.5mm in diameter) to fasten the pollinia in the stigma cavity. Then the rostellum on both sides of stigma grew up quickly and sealed up the stigma to allow the flower to fertilize and fruit. From the clinandrium to the stigma cavity, it typically took about 20 minutes for the pollinarium to complete the circling process

The pollinia subovate, about 1.2mm long, 0.7mm broad; stipe thin lorate, united on clinandrium, about 2mm long, 0.7mm broad, apical junction with pollinia semi-circular. Rostellum about 1mm long; receptive surface of stigma cavity on the upper-back side of rostellum, about 1.5mm deep. After the anther cap fell off, the pollinia-carrying stipe revolves 360°to insert and secure the pollinia in the stigma cavity.

Among the 954 flowers that failed to pollinate, 469 flowers’ anther cap detached from clinandrium but not from the pollinia, the stuck bulky cap prevented the stipe from completing the curling or the pollinia from entering the stigma cavity; the pollinia, stipe and anther cap would wilt soon. In the rest 485 flowers, the anther cap fell off completely, but the lower half of the stipe was unable to separate from the clinandrium, so the stipe collapsed in the middle while rising up and could not continue moving, and the pollinia were left on the upper side of the rostellum and unable to reach the stigma cavity. The pollination could not be achieved if the pollinia-carrying stipe revolved less than 360°.

Besides the afore-described, we did not find any other pollination behavior among the flowers in the 48 sites representing 10 populations of H. amesianum.


Table 1. Observation of pollination biology in 48 sites

Site No.

Number of flower

Volume of nectar

Number of visiting agent

Odor

Number of fruit

Rate of fruit set

(%)


Unfruited quantity

Stuck anther cap

Folded stipe

Total

1

40

0

0

0

19

47.50

9

12

21

2

41

0

0

0

21

51.22

10

10

20

3

40

0

0

0

19

47.50

12

9

21

4

39

0

0

0

20

51.28

8

11

19

5

40

0

0

0

24

60.00

8

8

16

6

39

0

0

0

25

64.10

6

8

14

7

40

0

0

0

18

45.00

13

9

22

8

39

0

0

0

18

46.15

11

10

21

9

42

0

0

0

21

50.00

10

11

21

10

37

0

0

0

19

51.35

7

11

18

11

40

0

0

0

19

47.50

12

9

21

12

40

0

0

0

20

50.00

8

12

20

13

41

0

0

0

23

56.10

8

10

18

14

41

0

0

0

19

46.34

13

9

22

15

39

0

0

0

18

46.15

11

10

21

16

40

0

0

0

20

50.00

8

12

20

17

39

0

0

0

18

46.15

12

9

21

18

40

0

0

0

19

47.50

8

13

21

19

41

0

0

0

20

48.78

17

4

21

20

39

0

0

0

19

48.72

11

9

20

21

37

0

0

0

18

48.64

5

14

19

22

40

0

0

0

22

55.00

8

10

18

23

41

0

0

0

20

48.78

12

9

21

24

39

0

0

0

19

48.72

9

11

20

25

40

0

0

0

18

45.00

9

13

22

26

40

0

0

0

23

57.50

10

7

17

27

42

0

0

0

23

54.76

9

10

19

28

39

0

0

0

18

46.15

13

8

21

29

40

0

0

0

20

50.00

8

12

20

30

40

0

0

0

19

47.50

7

14

21

31

41

0

0

0

20

48.78

11

10

21

32

38

0

0

0

20

52.63

9

9

18

33

40

0

0

0

20

55.00

13

7

20

34

41

0

0

0

19

46.34

7

15

22

35

40

0

0

0

21

52.50

7

12

19

36

39

0

0

0

18

46.15

16

5

21

37

41

0

0

0

20

48.78

9

12

21

38

39

0

0

0

24

61.53

4

11

15

39

40

0

0

0

19

47.50

13

8

21

40

37

0

0

0

19

51.35

9

9

18

41

39

0

0

0

19

48.72

12

8

20

42

40

0

0

0

20

50.00

7

13

20

43

40

0

0

0

18

45.00

8

14

22

44

41

0

0

0

20

48.78

10

11

21

45

42

0

0

0

21

50.00

8

13

21

46

40

0

0

0

23

54.76

11

6

17

47

39

0

0

0

18

46.15

8

13

21

48

39

0

0

0

19

48.72

15

5

20

Total

1911

0

0

0

957

50.08

469

485

954

Minimum

37.00










18.00

45.00










Maximum

42.00










25.00

64.10










Mean

39.8125

0

0

0

19.9375

50.1267

9.7

10.1042

19.8750

Std. Deviation

1.16064










1.81491

4.33630









The rates of fruit set and errors (descriptive statistics of sites) are summarized in Table 1. The results show that the rate of fruit set from self-pollination in nature was 50.13±4.34% (n=48). The total rate of non-fruiting (failed pollination) was calculated to be 49.87% (n=48), of which 24.53±1.56%, (n=48) was due to the anther cap sticking with the pollinia, 25.38±6.31% (n=48) due to the stipe folding in the middle.

The seed counts are listed in Table 2.
Table 2. Seed quantity of fruit


Year

1

2

3

4

5

6

7

8

9

10

Total

2003

356481

293461

297833

370621

553689

427611

277689

336857

288652

306316

3509210

2004

537869

304281

279682

283652

380617

291268

312893

400082

309834

283452

3383630

2005

373416

278054

304153

565143

369840

283463

308981

358452

314983

390672

3547157

There were 280000-570000 seeds in each fruit. Under microscope, seeds were long-ovate, yellowish-brown, 0.18-0.20mm long, about 0.067mm broad.



4.3 Breeding system

4.3.1 Manual self-pollination and Manual cross-pollination

In the experiments of manual self-pollination and manual cross-pollination, after the pollinia entered the stigma cavity, the rostellum started to grow; 4-5 days later, petals turned pale and wilted, the base of the petals turned green and enlarged and which made the petals persist. The rostellum grew up and sealed the stigma cavity completely, and the flowers were fertilized and fruited. This post-pollination process was identical to that in the natural self-pollination. The observation and statistic results are listed in Table 3.


Table 3. Pollination experiments



Natural

self-pollination



Bagged self-pollination

Manual self-pollination

Manual cross-pollination

Site No.

Number of flower

Number of fruit

Rate of fruit set(%)

Number of flower

Number of fruit

Rate of fruit set(%)

Number of flower

Number of fruit

Rate of fruit set(%)

Number of flower

Number of fruit

Rate of fruit s et(%)

1

40

19

47.50

40

18

45

37

34

91.85

63

57

90.48

2

41

21

51.22

39

18

46.15

41

37

90.24

60

55

91.67

3

40

19

47.50

40

20

50

39

36

92.31

62

56

90.32

4

39

20

51.28

37

19

51.35

51

46

90.20

66

59

89.39

5

40

24

60

40

20

50

49

45

91.84

50

46

92.00

6

39

25

64.1

41

23

56.1

48

44

91.67

55

50

90.91

7

39

18

46.15

41

20

48.78

41

37

90.24

62

56

90.32

8

40

19

47.50

39

19

48.72

38

35

92.11

61

55

90.16

9

41

20

48.78

40

18

45

43

39

90.70

58

53

91.37

10

39

19

48.72

40

23

57.50



















11

37

18

48.64

42

23

54.76



















12

40

22

55

39

18

46.15



















13

40

22

55

40

19

47.5



















14

41

19

46.34

37

19

51.35



















15

40

21

52.50

39

19

48.72



















16

39

18

46.15

40

20

50



















17

41

20

48.78

40

18

45



















18

39

24

61.53

41

20

48.78



















Total

715

368

926.69

715

354

890.86

387

353

821.16

537

487

816.62

Mean

39.7222

20.4444

51.4828

39.7222

19.6667

49.4922

43

39.2222

91.24

59.6667

54.1111

90.7356

Std. Deviation







5.52964







3.69794







0.87977







0.82594

Std. Error Mean







1.30335







0.87161







0.29326







0.27531

Independent Sites Test results are summarized in Table 4, section A.


Table 4. Independent Sites Test




Levene’s Test for Equality

of Variances



t-test for Equality of Means

F

Sig.

t

df

Sig.

(2-tailed)



Mean Difference

Std. Error Difference

95% Confidence

interval of the Difference



Lower

Upper

A


Rate of fruit set, Equal

variances assumed



.602

.449

1.254

16

.228

.50444

.40224

-.34854

1.35716

Rate of fruit set, Equal

variances not assumed









1.254

15.937

.228

.50444

.40224

-.34854

1.35743

B

Rate of fruit set, Equal

variances assumed



2.783

.104

-1.271

34

.212

-1.99222

1.56794

-5.17866

1.19421

Rate of fruit set, Equal

variances not assumed









-1.271

29.671

.214

-1.99222

1.56794

-5.19587

1.21142

C

Rate of fruit set, Equal

variances assumed



10.687

.003

21.231

25

.000

39.75722

1.87261

35.90051

43.61393

Rate of fruit set, Equal

variances not assumed









29.760

18.663

.000

39.75722

1.33593

36.95766

42.55678

The rate of fruit set in manual self-pollination is 91.24±0.88% (n=9); and in manual cross-pollination 90.74±0.83% (n=9).

According to the Independent Sites Test (F=0.602; df=16, P>0.05), the rates of fruit set from manual self-pollination and manual cross-pollination did not have significant difference. The ANOVA-test results (not shown) also proved that.

4.3.2 Natural self-pollination and bagged self-pollination

The floral development process of self-pollination in bagged flowers was the same as in nature, as described above. The observation and statistic results are listed in Table 3. The rate of fruit set of natural pollination is 51.48±5.53% (n=18); of bagged self-pollination is 49.49±3.70% (n=18).

Independent Sites Test (F=2.783, df=34, P>0.05) results of natural self-pollination and bagged self-pollination, as summarized in Table 4, section B, show that the rates of fruit set of natural pollination and bagged self-pollination had no significant difference. The ANOVA-test results (not shown) also proved that.

4.3.3 Asexual reproduction with natural and bagged treatments

In the controlled emasculation experiments in paired sites, the rate of fruit set of emasculated flowers was zero (n=22), whether bagged or not.



4.3.4 Assessment of apomixis

Examination of H. amesianum flower buds showed the presence of antipodal cells in the embryo of ovary, which indicates that the embryo develops by normal meiosis, a process that is absent in apomixis.

Karyotype analysis of H. amesianum root tip tissues revealed that the parents and their progenies from self-copulation all have 2n=38 chromosomes with normal morphology (picture not shown) while no haploid, polyploidy or aneuploidy was detected, which indicates that the self-copulation enables reproduction of H. amesianum through normal fertilization, i.e. the fusion of male and female gametes.

4.3.5 Manual self-pollination vs natural self-pollination

Analysis was performed to compare the observation data of natural self-pollination in 4.3.2 (Table 3) (which is virtually the same as in 4.2, Table 1) with those of manual self-pollination in 4.3.1 (Table 3).



Independent Sites Test results of natural and manual self-pollination, as summarized in Table 4 section C, show that the rate of fruit set from natural self-pollination is significantly lower (F=10.687, df=25, P>0.05) than manual self-pollination, indicating that the natural self-copulating pollination was affected by certain factors and did not always succeed. As we have observed, unsuccessful self-copulation resulted from two types of mishaps as described above in 4.2.





Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©atelim.com 2016
rəhbərliyinə müraciət