Does Scots pine seed colour affect its germination?

3 downloads 0 Views 45KB Size Report
We also observed that dark seed had dark wings, and this was related to tree age, .... twice as fast in dark seeds than light seeds: mean germination per day was.

SCOTS PINE SEED COLOUR AND GERMINATION

Mukassabi, T.A., Polwart, A., Coleshaw, T. and Thomas, P.A. (2012), Seed Sci. & Technol., 40, 155-162

Does Scots pine seed colour affect its germination? T.A. MUKASSABI1, A. POLWART1, T. COLESHAW2 AND P.A. THOMAS1 1

2

School of Life Sciences, Huxley Building, Keele University, Staffordshire, ST5 5BG, UK (E-mail: [email protected]; [email protected]; [email protected]) Natural Nature Reserves, Team West, Natural England, Parkside Court, Telford, TF3 4LR, UK (E-mail: [email protected])

(Accepted March 2012)

Summary Little is still known about the effect of seed colour on the germination of Scots pine. Using Scots pine seed from the English Midlands, we found that seeds were readily divided into dark and light categories and that light seeds could be further divided into 4 sub-groups. Dark seeds exhibited high values of mass and viability, whereas all shades of light seeds showed lower mass and poorer germination. Commercial seed lots contained very few light seeds. We also observed that dark seed had dark wings, and this was related to tree age, since younger trees produced more dark winged seeds than older trees. The mass of dark seeds could be predicted from the mass of random samples of 100 mixed seeds, and the mean mass of 100 dark seeds can be used to predict the number of dark seeds within the sample.

Introduction In some plant species, germination can vary between seeds of different colour, ascribable to a number of reasons including seed maturity and differences in timing or technique of seed collection (Murdoch, 2007). In Scots pine (Pinus sylvestris L.), there has long been a debate about the effect of seed colour on germination. Some workers have found that dark seeds showed better germination than light seeds (Baldwin, 1942; Steven and Carlisle, 1959; Stone, 1958) while others have found no difference (Grazywacz and Roschacka, 1980). Baldwin (1942) classified Scots pine seeds into four colour classes: light, brown, dark and mottled. A possible reason for the different colour of Scots pine seed used in forestry may be that they are affected by abrasion during the commercial extraction operation, abraded seeds being paler (Gordon and Faulkner, 1992). However, Steven and Carlisle (1959) observed that Scots pine seeds are naturally variable in colour. They estimated the percentage of the different colour classes in all principal native woodlands as 10.2% light coloured, 49.1% brown, 36.7% dark and 4.0% mottled. Since colour variation is widespread, it remains an important question as to how this affects germination. In this study, we investigated this question and whether or not colour can contribute to a rapid prediction of the germinability of Scots pine seeds. 155

T.A. MUKASSABI, A. POLWART, T. COLESHAW AND P.A. THOMAS

Materials and methods Seed source Seeds used in this study were collected from Wybunbury Moss in Cheshire, England (53°02'N and 2°25'W, 71 m a.s.l.) in 2008 and 2009. Seed were manually extracted from cones collected from ten trees which were cut and gently lowered to prevent loss of cones. The age of sampled trees was determined from ring counts. Cones were air-dried for three weeks at temperatures not exceeding 30°C, and then dried in an oven at 30-50°C for 24 to 48 hours to open the cones. After drying, seeds were shaken from the cones. Mean seed mass was determined for all seed sources using replicate batches of 100 seeds, randomly selected regardless of colour. Overall seed mass and germination was compared with three commercial seed sources of Scots pine: from a plantation at Carrbridge, Scotland (57°17'N, 3°48'W, 252 m a.s.l.), from a native woodland at Glenbeg, Scotland (57°19'N, 3°38'W, 330 m a.s.l.) and from a plantation in Sierra de Guadarrama, Spain (40°40'N, 4°09'W, 1329 m a.s.l.). Overall mass and germination of seeds Seeds from 14 different cones were randomly sampled from the crops of 2008 and 2009 at Wybunbury Moss. Seeds from commercial sources were sampled randomly from purchased supplies. Mean mass of 100 seeds, regardless of colour, from all sources of seed was measured using 20 replicates from each. Mean germination of each seed source was assessed by selecting 20 seeds randomly, irrespective of colour, and germinating them in 9 cm diameter Petri dishes on a double layer of Whatman filter paper grade 2. There were ten replicates for each seed source. Each dish was watered with 5 ml of deionized water every five days and kept at room temperature and 16 hours daily light. Germination was monitored daily for 21 days and speed of germination was calculated (Maguire, 1962) as the number of new germinates each week, divided by 7 to give a daily mean, summed over a three week period. For the Wybunbury Moss seeds, overall viability was also tested using a 1% solution of 2,3,4-Triphenyl tetrazolium chloride (TTC) (1 g TTC dissolved in 100 ml deionized water at pH 7.0). Five seeds of each colour, replicated 20 times, were cut longitudinally into two halves (attached at the base) through the embryo and part of the endosperm. Seeds were placed in a test tube with 5 ml of 1% TTC, and all test tubes were incubated in a water bath at 30°C for 24 hours. Viability was assessed under a binocular microscope by change in embryo colour, red indicating viability (Lakon, 1949). Seed colour Three groups of 100 winged seeds were randomly sampled from each of ten trees at Wybunbury Moss in 2009 and used to determine seed colour, mass or germination. Seed colour was visually ranked into two main groups of (1) dark and (2) light. There was very little overlap in colour and seeds were readily ranked but, if in doubt, colour 7.5YR 4/4 (Munsell, 2011; Colour Chart for Plant Tissues) was used as the dividing colour. Light seeds were then divided into four sub-groups by colour: (a) light brown covered entirely with small spots of black; (b) light brown with a large spot of black on one or 156

SCOTS PINE SEED COLOUR AND GERMINATION

both sides; (c) light brown with two large dark spots around both acute ends; and (d) light brown with only one large dark spot around one end. Seeds from each colour group were weighed, and germination was tested in Petri dishes as above using 20 randomly selected seeds replicated 10 times. The effect of seed colour on germination in the commercial supply (Scotland plantation) was determined by dividing the seeds into dark and light (using the same dividing colour) and repeating the germination trials using ten replicates of 20 seeds for each colour. This trial was repeated using seeds extracted from 14 large cones, randomly selected, from the crops of 2008 and 2009. Here seeds were divided into dark and light, counted and their germination assessed as above. Wing colour The colour of seed wings was determined for the seeds from Wybunbury Moss used used in the seed colour experiment (above), in order to clarify the ability of using it as an indicator of seed colour. Wing colour was ranked using Munsell (2011) from a very pale yellow-brown to very dark brown in five classes: 1 = Buff (7.5YR 8/6), 2 = Earth yellow (5YR 7/10), 3 = Bronze (10R 6/10), 4 = Copper (10R 5/10), 5 = Brown (10R 3/4) and 6 = Russet (10R 3/6). Statistical analysis One-way ANOVA was used to assess germination differences across seed colour ranks and also between wing colour and seed colour and viability. For Wybunbury data, linear regression analysis was used to investigate the relationship between various parameters of the seed samples and the mass of dark seeds in a sample. Minitab© Version16 software was used for analyses.

Results Overall seed mass and germination For seed selected at random, regardless of colour, overall mean seed mass from Wybunbury was 0.35 ± 0.005 g per 100 seeds (s.e., n = 20), much lower than in the commercial seed sources used in the study (table 1). Consequently, the number of seeds per kg is also markedly different. Germination of seed was variable with the large seeds from Spain giving notably lower germination (40%; n = 10) than seed from the UK. The highest germination (82%) was observed for seed from the Scottish native woodland (table 1). The overall viability of Wybunbury Moss seeds assessed using the TTC method was 78% (n = 20) compared with 64% (n = 10) from the germination test. Seed colour For commercial seed from the Scotland plantation, one-way ANOVA showed no significant differences in germination between dark and light seeds (75 and 80%, respectively, n = 10). Germination of seeds collected from Wybunbury was, however, influenced by colour. In the ten trees sampled at Wybunbury, dark seeds made up 22-82% of the number of 157

T.A. MUKASSABI, A. POLWART, T. COLESHAW AND P.A. THOMAS

Table 1. Mean mass and germination percentage of Scots pine seeds selected at random, regardless of colour, from all sources of seed used in the investigation. Wybunbury Moss samples included seeds from two consecutive years. Number of seeds per kg was calculated based on the seed mass obtained. Characteristic

Scotland Native*

Scotland Plantation*

Spain*

Wybunbury Moss

Mass of 100 seed (g) (mean ± s.e., n = 20)

0.64 ± 0.003

0.52 ± 0.003

1.22 ± 0.007

0.35 ± 0.005

Number of seeds per kg

157,356

192,159

82,311

287,191

Germination percentage (%)

82

71

40

64

*Seeds from commercial sources.

seed with a mean of 68 ± 5.4% (table 2). The dark seeds were heavier than the light ones, so the dark seeds formed 66-95% of the total mean mass with an overall mean of 88 ± 2.6%. Germination trials in Petri dishes showed that across the ten trees, the germination of dark seeds ranged from 88.7-95.0%, with an overall mean of 91.8%. Germination of light seeds, however, varied from 0-12% between trees with an overall mean germination of 7.6% (table 2). When the light seeds were further separated into four sub-groups, germination was low, ranging from 0% in 2b to 9.5% in 2c (table 3). Speed of germination was more than twice as fast in dark seeds than light seeds: mean germination per day was 1.4 in dark seeds, 0.1-0.6 in light (table 3). Sampling all the seeds in 14 large cones from Wybunbury Moss produced similar results. Overall the cones contained a significantly larger number of dark seeds than light seeds (82 ± 2.2 and 18 ± 2.2% for dark and light seeds, respectively, s.e., n = 14: F1,27 = 321.56; P < 0.001), and the germination of dark seeds was significantly higher (91.5 and 5.3%, for dark and light seeds, respectively, n = 14: F1,27 = 478.11; P < 0.001). A linear regression in this study showed a positive relationship between the total seed mass of the sample of 100 seeds and the mass of dark seed it contains (R2 = 0.94; P < 0.001). Thus, at Wybunbury Moss, the mean seed mass of random samples can be used to predict the mass of dark seeds: Dark seed mass (g) = -0.8839 + 1.036 Total seed mass (g) Moreover, another linear regression showed a positive relationship between the mean dark seed mass and mean number of dark seeds (R2 = 0.60; P < 0.05). As a result, for this study site the mean mass of a dark seeds sample can be used to predict the number of dark seeds it contains: Number of dark seeds = 12.59 + 102.4 Dark seed mass (g) It is likely that a similar relationship will exist for other Scots pine sources.

158

112

106

54

55

112

54

53

58

50

2

3

4

5

6

7

8

9

10

4

4

5

5

3

2

3

6

4

1

Wing colour

32 ± 5.4

27

21

37

25

25

34

26

18

33

78

Proportion of sample by number (%)

0.062 ± 0.0100

0.077

0.023

0.064

0.032

0.046

0.080

0.080

0.052

0.077

0.135

Mass (g) /sample

Light seed

7.6

11.0

9.5

6.5

10.1

0.0

9.2

12.0

8.0

8.9

1.1

Germination (%)*

* For the germination test n = 10 × 20 seeds randomly sampled.

Mean ± s.e.

105

1

Tree number and age (years)

68 ± 5.4

71

79

63

75

75

66

74

82

68

22

Proportion of sample by number (%)

0.518 ± 0.0406

0.711

0.453

0.506

0.502

0.548

0.553

0.591

0.710

0.547

0.262

Mass (g) /sample

91.8

91.2

94.5

88.7

93.0

92.9

92.4

95.0

90.8

90.1

89.0

Germination (%)*

Dark seed

88.3 ± 2.62

90.2

95.1

88.7

94.1

92.2

87.4

88.1

93.1

87.7

66.0

Mass (%) /sample

64.4

69.5

76.6

58.5

72.6

69.8

63.9

73.1

76.2

63.7

20.6

Total germination*

Table 2. The number, mass and germination of groups of 100 winged Scots pine seeds, randomly sampled from ten trees in 2009. Wing colour was ranked from a very pale yellow-brown to very dark brown in five classes (Munsell, 2011): 1 = Buff (7.5YR 8/6), 2 = Earth yellow (5YR 7/10), 3 = Bronze (10R 6/10), 4 = Copper (10R 5/10), 5 = Brown (10R 3/4) and 6 = Russet (10R 3/6).

SCOTS PINE SEED COLOUR AND GERMINATION

159

T.A. MUKASSABI, A. POLWART, T. COLESHAW AND P.A. THOMAS

Wing colour Wing colour varied from dark brown to very pale yellow between trees, but within trees was mostly dominated by just one colour (table 2). Using samples of 100 seeds, the relationship between wing colour and seed colour and viability was examined. Means of several groups of 100 winged seeds from 10 trees were included (table 2). Oneway ANOVA showed that there were significant differences between the wing colour associated with dark and light seeds. Darker wings contained dark seeds and light seeds mostly had very light wings (F5,9 = 14.20; P < 0.05). Moreover, the same application showed that light wing colour was significantly associated with low seed viability (F5,9 = 9.62; P < 0.05). There was also a significant difference between tree age and the proportion of dark wings in a sample: younger trees (< 60 years old) produce a greater proportion of darker seed wings (F7,9 = 24.17; P < 0.05). Table 3. The percentage of dark and light (n = 10 × 100 seeds) Scots pine seeds and their germination ability (n = 10 × 20 seeds). Seed colour

Percentage of seed (n = 10 × 100 seeds)

Seed germination (%) (n = 10 × 20 seeds)

Speed of germination

1

Dark seed

67.7

91.8

1.4

2a

Light brown seed covered entirely with small spots of dark black colour*.

10.0

9.4

0.1

2b

Light brown seed covered with a large spot of black on one or both sides.

2.5

0.0

0.0

2c

Light brown seed with two large dark spots around the both acute ends.

8.8

9.5

0.6

2d

Light brown seed with only one large dark spot around one of acute ends.

11.0

15.0

0.3

* Light seeds are divided into 4 sub ranks. Germination tests were made using 20 seeds replicated 10 times. Speed of germination was calculated as mean germination per day over the three weeks.

Discussion This study gave unequivocal results of a difference between dark and light seeds in viability and germination. Dark seeds collected from Wybunbury Moss were about ten times more viable than light seeds. Commercial seeds, however, did not show the same difference, undoubtedly because the very light, empty seeds were removed during seed processing, since only 2-3 per cent of empty seeds is normally acceptable in the final product (Gordon, 1992a). The remaining difference in seed colour within the commercial supplies was attributed to the mechanical procedures used in crop collection and cleaning (Steven and Carlisle, 1959). This, we feel, goes a long way to explaining the previous contradictory results. In Wybunbury seeds, collected by hand, light seeds formed up to 30% of the whole crop. 160

SCOTS PINE SEED COLOUR AND GERMINATION

Spanish seeds, from a commercial supplier, showed high mass compared with seeds collected manually from a similar site in Spain (Castro et al., 2008) which ranged between 0.007 and 0.014 g per 100 seeds. However, Castro et al. (2008) pointed out that seed mass in Scots pine is strongly determined by maternal tree and differences in seed mass between years is very common. However, many trees of wetlands or other extreme environments tend to produce large crops of light seeds (Streng et al., 1989). The light coloured seeds in all Wybunbury samples had low mass and often low germination, suggesting that the majority of light coloured seeds were empty. This was supported by direct observations during the study since most light seeds shattered under slight pressure and were indeed empty. Scots pine seeds do not normally have any dormancy and have the ability to germinate immediately upon releases from the cone (Castro et al., 2005). But in this study, viability, as judged by TTC, was 10% higher than in the germination trials. It is difficult practically to achieve perfect germination conditions (Gordon, 1992b). Further, discrepancy between the two methods for Scots pine seeds from Wybunbury could be because of immaturity or lack of nutrient reserves in the seeds (Gordon, 1992b). Nonetheless, results of this study suggest some dormancy in Scots pine and further work on seeds obtained naturally is needed. Since the dark seeds showed higher speed of germination and germination percentage, results of this study suggest that some other maternal and ecological factors such as tree age, availability of light or nutrients can be the key factor of increasing the proportion of viable (dark) seeds in Scots pine. In many plant species, the seed position in different organs on the mother plants can affect seed colour, size and germination (Gutterman, 2000). This study shows that seed and wing colour can be used to visually give a rapid prediction of likely viability. Mukassabi et al. (2012) found no significant relationship between the percentage of dark, full seeds and characteristics such as cone length, number of seeds per cone, tree height, trunk diameter, canopy shading, water table and location of trees sampled at Wybunbury Moss. Observational data during this study suggested that darker coloured cones showed signs of producing a greater proportion of dark seeds which requires further investigation. However, it was found that the mass of dark seeds can be accurately predicted from the mass of random samples of 100 mixed seeds. Also, the number of dark seeds can be predicted from the mean mass of 100 dark seed sample. These relationships, are likely to hold for other seed sources, and are of direct practical value in giving a rapid method of assessing seed viability. This study also shows that a greater proportion of dark, viable seed can be obtained from younger trees.

References Baldwin, H.I. (1942). Forest Tree Seed of the North Temperate Regions with Special Reference to North America. Chronica Botanica, Waltham, Massachusetts. Castro, J., Zamora, R., Hodar, J.A. and Gomez, J.M. (2005). Ecology of seed germination of Pinus sylvestris L. at its southern Mediterranean distribution range. Investigacion Agraria: Sistemasy Recursos Forestales, 14, 143-152.

161

T.A. MUKASSABI, A. POLWART, T. COLESHAW AND P.A. THOMAS

Castro, J., Reich, P.B., Sanchez-Miranda, A. and Guerrero, J.D. (2008). Evidence that the negative relationship between seed mass and relative growth rate is not physiological but linked to species identity: a withinfamily analysis of Scots pine. Tree Physiology, 28, 1077-1082. Gutterman, Y. (2000). Maternal effects on seeds during development. In Seeds: The Ecology of Regeneration in Plant Communities, (ed. M. Fenner), pp. 59-84, CABI Publishing, Oxon, UK. Gordon, A.G. (1992a). The processing of cones and seeds. In Seed Manual for Forest Trees, (ed. A.G. Gordon) Bulletin 83, pp. 86-97, Forestry Commission, London. Gordon, A.G. (1992b). Seed testing. In Seed Manual for Forest Trees, (ed. A.G. Gordon) Bulletin 83, pp. 105115, Forestry Commission, London. Gordon, A.G. and Faulkner, R. (1992). Identification and assessment of cone and seed crops. In Seed Manual for Forest Trees, (ed. A.G. Gordon) Bulletin 83, pp. 71-79, Forestry Commission, London. Grazywacz, A.P. and Roschaka, J. (1980). The colour of Pinus sylvestris L. seed and their susceptibility to damping-off. European Journal of Forest Pathology, 10, 138-144. Lakon, G. (1949). The topographical tetrazolium method for determining the germination capacity of the seed. Plant Physiology, 24, 389-394. Maguire, J.D. (1962). Speed of germination – Aid in selection and evaluation for seedling emergence and vigo. Crop Science, 2, 176-177. Munsell (2011). Color Chart for Plant Tissues. Munsell Color, Michigan, USA. Murdoch, F.A. (2007). Seed colour indicates germinability of buloke (Allocasuarina luehmannii). Ecological Management & Restoration, 8, 234-237. Mukassabi, T.A., Polwart, A., Coleshaw, T. and Thomas, P.A. (2012). Scots pine seed dynamics on a waterlogged site. Trees, in press. Steven, H.M. and Carlisle, A. (1959). The Native Pinewoods of Scotland. Oliver & Boyd, Edinburgh. Stone, E.C. (1958). The seed dormancy mechanism in pine. In The Physiology of Forest Trees, (eds. K.V. Thimann). pp 611-628, Ronald Press, New York. Streng, D.R., Glitzenstein, J.S. and Harcombe, P.A. (1989). Woody seedling dynamics in an east Texas floodplain forest. Ecological Monographs, 59, 177-204.

162