Evolution of Spur Length in a Moth-pollinated Orchid

Full text

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8) .

(9)

(10) !.

(11)

(12)

(13) . "#"$ %&%##$$

(14) '(')'

(15)

(16) ' *+&".

(17) .

(18)

(19)

(20)

(21) !"#" #"$""%

(22)

(23) %

(24)

(25)

(26) %&

(27)

(28) ' (

(29) )

(30) * ' +

(31) *'!"#"'*

(32)

(33)

(34) %, -

(35) .

(36) / '0 . '

(37)

(38)

(39) 1#!'22' '3,+415.#. 6.11#6.#' (

(40)

(41)

(42) % %% %

(43) )

(44)

(45)

(46) )

(47)

(48) )

(49)

(50) %

(51) %

(52) .

(53)

(54)

(55) %

(56) )

(57)

(58) 7

(59) ) ' 3 3

(60)

(61)

(62)

(63)

(64)

(65)

(66) %%

(67)

(68) .

(69)

(70) '

(71) % 3 8#9

(72) ) %

(73) )

(74)

(75)

(76)

(77)

(78) 8!9

(79)

(80)

(81)

(82)

(83)

(84)

(85)

(86)

(87)

(88)

(89)

(90) 829

(91)

(92)

(93) % :

(94)

(95)

(96)

(97)

(98)

(99)

(100) 869 %

(101) %

(102) % )

(103)

(104) %

(105) ' 0

(106) %,

(107) %

(108)

(109) )

(110)

(111) )

(112)

(113)

(114) %

(115)

(116)

(117) )

(118) %

(119)

(120)

(121)

(122)

(123)

(124)

(125) .

(126) ' 0

(127)

(128)

(129) ; ,*,)

(130)

(131)

(132)

(133)

(134) %

(135)

(136)

(137) % .

(138)

(139) %

(140)

(141)

(142)

(143)

(144) %

(145) )

(146)

(147) )

(148)

(149) . '

(150)

(151)

(152)

(153) )

(154)

(155)

(156) ;

(157) %%

(158) %

(159) )

(160)

(161)

(162)

(163)

(164) .

(165)

(166)

(167) ' ( :

(168)

(169)

(170)

(171)

(172)

(173)

(174) %

(175)

(176) %%

(177) %

(178) ' 0%

(179)

(180)

(181) : %%

(182)

(183)

(184)

(185) '(

(186) .

(187)

(188) %%

(189) :

(190) %

(191)

(192) ) ' ( 7

(193)

(194) %

(195)

(196) .

(197) .

(198)

(199)

(200) % %%

(201)

(202) ' !& %

(203) %

(204)

(205)

(206)

(207)

(208)

(209)

(210)

(211) %%

(212)

(213)

(214)

(215)

(216)

(217) "# $

(218) " " $Plant Ecology, & '($

(219)

(220) University, ")*+-./

(221)

(222) $ <* +

(223) !"#" 3,,4#= #.=!#6 3,+415.#. 6.11#6.# $ $$$ .##2"5=8. $>> '7'>

(224) ? @ $ $$$ .##2"5=9.

(225) “Writing a book is an adventure. To begin with, it is a toy and an amusement; then it becomes a mistress, and then it becomes a master, and then a tyrant. The last phase is that just as you are about to be reconciled to your servitude, you kill the monster, and fling him out to the public” Winston Churchill.

(226)

(227) List of Papers. This thesis is based on the following papers, which are referred to in the text by their Roman numerals. I. Boberg, E., Alexandersson, R., Jonsson, M., Maad J., Ågren, J. and Nilsson, L.A. Pollinator shifts and the evolution of spur length in the moth-pollinated orchid Platanthera bifolia. (Manuscript). II. Boberg, E., Xu, L. and Ågren. J. Phenotypic selection on floral traits in divergent populations of the moth-pollinated orchid Platanthera bifolia. (Manuscript). III. Boberg, E. and Ågren, J. Reproductive isolation among divergent populations of the moth-pollinated orchid Platanthera bifolia. (Manuscript). IV. Boberg, E. and Ågren, J. (2009) Despite their apparent integration, spur length but not perianth size affects reproductive success in the moth-pollinated orchid Platanthera bifolia. Functional Ecology, 23:1022–1028. Paper IV is printed with permission from the publisher..

(228)

(229) Contents. Introduction.....................................................................................................9 The mechanical fit......................................................................................9 Divergent and disruptive selection ...........................................................10 Reproductive isolation..............................................................................10 Selection on correlated floral traits ..........................................................11 Aims of this thesis ....................................................................................12 Methods ........................................................................................................13 Study species ............................................................................................13 Pollination mechanism .............................................................................13 Study populations and site descriptions ...................................................15 Population differentiation and pollinators (I) ...........................................16 Reciprocal translocation experiment (I) ...................................................17 Phenotypic selection study (II).................................................................17 Estimating the strength of reproductive isolation (III) .............................17 Experimental manipulation of flower morphology (IV) ..........................18 Results and discussion ..................................................................................19 Population differentiation and pollinator shifts (I)...................................19 Disruptive and divergent selection (II).....................................................21 Intraspecific reproductive isolation (III) ..................................................22 The adaptive significance of spur length and perianth size (IV)..............23 Conclusions ..............................................................................................23 Summary in Swedish ....................................................................................25 Långa sporrar och matchande tungor .......................................................25 Acknowledgements.......................................................................................30 References.....................................................................................................31.

(230)

(231) Introduction. Flower morphology and phenology of flowering vary extensively in animalpollinated plants (Harder and Barrett 2006), and the relative importance of pollinator-mediated selection and random processes for this variation remains a central question in plant evolutionary biology. Animal pollinators have, like all organisms, restricted ranges limited by abiotic or biotic factors. Spatial and temporal variation in pollinator availability may result in divergent selection between plant populations and differentiation of floral traits that affect pollinator attraction and efficiency (Herrera et al. 2006; Johnson 2006). A number of studies of pollinator-driven floral differentiation in plants have focused on large-scale patterns of relationships between pollinators and closely related plant species (Schemske and Bradshaw 1999; Castellanos et al. 2003; Whittall and Hodges 2007). Plant species with large intraspecific variation in floral traits may be in the initial stages of divergence. As Grant and Grant (1965) stated, it is in these plant systems we can study floral evolution as an “ongoing process rather than a historical event”. In this thesis I explore the evolutionary mechanisms responsible for the evolution and maintenance of intraspecific differentiation in floral traits.. The mechanical fit Flower parts may be subject to pollinator-mediated selection due to their effects on the mechanical fit between the plants reproductive organs and the pollinator’s body (Darwin 1862). Some plant species produce floral spurs, a tubular structure that contains nectar as a reward to animal pollinators. The length of the spur is expected to affect the mechanical fit to pollinators and is thus essential for effective pollination (Darwin 1862; Nilsson 1988; Johnson and Steiner 1997; Pauw et al. 2009). If the spur is too short to match the length of the proboscis of the pollinator, the pollinator will not come into contact with the flower’s reproductive organs. A long spur will force the pollinator to probe deeper into the flower to reach the nectar at the bottom of the spur and hence cause more effective removal and deposition of pollen. Consequently, selection will favour plant individuals that have a spur, which is longer than the proboscis of the primary pollinator (Nilsson 1988).. 9.

(232) Divergent and disruptive selection Divergent and disruptive selection may explain the evolution and maintenance of intraspecific floral variation (Schluter 2000). The relationship between phenotypic traits and relative fitness within a population can be visualized as a surface, an adaptive landscape, where the horizontal axes represent trait variation and the height of the surface represents the fitness of a given trait combination. The adaptive landscape can include valleys and peaks, which indicate trait combinations associated with fitness minima and optima, respectively. Divergent selection is the process by which natural selection pulls trait means of populations towards different adaptive peaks. Divergent selection among habitats may be the result of stabilizing selection for different optima (Benkman 2003) or linear selection in opposite directions (Caruso et al. 2003; Hall and Willis 2006). In a contact zone between two divergent populations, floral traits may have a bimodal distribution. Bimodal traits may be subject to disruptive selection that results from the inferior fitness of individuals with an intermediate phenotype (Schluter 2000; Hendry et al. 2009). Disruptive selection may maintain high levels of genetic variation and contribute to the evolution of reproductive isolation and speciation (Coyne and Orr 2004; Rueffler et al. 2006). There is a lack of studies exploring the pattern of current selection in relation to population differentiation in floral traits, and surprisingly few studies have attempted to investigate disruptive selection on these traits in contact zones between divergent populations.. Reproductive isolation The evolution of reproductive isolation between divergent plant populations may contribute to the maintenance of floral variation (Grant 1949, 1994; Coyne and Orr 2004). Reproductive isolation barriers reduce gene flow among populations, and may arise as a side-effect of divergent selection or by reinforcement, i.e., selection against hybridization (Coyne and Orr 2004; Rueffler et al. 2006). In plants, prezygotic isolating barriers, which prevents mating and fertilization between species, are generally stronger than postzygotic isolating barriers, which result from the inviability or sterility of hybrids (Lowry et al. 2008a). Populations that differ in their spatial distribution, flowering time or in their primary pollinators may be reproductively isolated due to limited pollen transfer (Grant 1994; Coyne and Orr 2004). Differences in spatial distribution and flowering time may enforce reproductive isolation due to the reduction in among-population encounters (Rice and Salt 1990; Johnson et al. 1996; Levin 2006). Differences in flower morphology may enforce reproductive isolation via pollinator preferences 10.

(233) (Chittka et al. 1999; Kennedy et al. 2006), or when pollen transfer among morphologically different flowers are physically impossible, for example when pollen is deposited on different parts of the pollinator’s body (Nilsson 1983; Ramsey et al. 2003; Maad and Nilsson 2004). Few studies have investigated the importance and strength of these different components in the maintenance of species boundaries, and only one previous study has attempted to quantify reproductive isolation among divergent plant populations (Lowry et al. 2008b).. Selection on correlated floral traits Variation in floral traits tends to be structured such that the sizes of different flower parts are positively correlated within and among populations (Armbruster and Schwaegerle 1996; Davis 2001; Ordano et al. 2008). Flower depth and perianth size represent two different aspects of floral morphology that sometimes are phenotypically correlated and that both may influence pollination success. While perianth size may affect the overall attractiveness to pollinators (Galen and Newport 1987; Schemske and Ågren 1995; Conner and Rush 1996), spur length may affect the mechanical fit between flower and pollinator (Nilsson 1988; Johnson and Steiner 1997). The extent to which variation in perianth size and spur length is shaped by pollinator-mediated selection on one or several characters is poorly known. If pollinators exert selection on floral morphology, trait correlations could evolve as a result of correlational selection, i.e., when selection acts on a specific combination of traits, or alternatively as a result of genetic and developmental correlations between one or a few traits subject to selection (Conner 2002).. 11.

(234) Aims of this thesis The general objective of this thesis was to explore the evolution and maintenance of population differentiation in floral traits. I studied the mothpollinated orchid Platanthera bifolia that vary widely in morphology and phenology among Scandinavian populations. I addressed the following questions: 1. Does among-population variation in spur length reflect adaptive evolution associated with shifts in primary pollinators? (I) 2. Does current selection via female and male pollination success contribute to the maintenance of differences in morphology and phenology among populations? (II) 3. To what extent do different components of prezygotic reproductive isolation contribute to the maintenance of population differentiation in floral morphology? (III) 4. What are the functional and adaptive significance of perianth size and spur length, two traits that covary among populations? (IV). 12.

(235) Methods. Study species Platanthera bifolia (L.) L.C. Rich. is a terrestrial orchid with a Eurasian distribution and occurs in a variety of grassland and woodland habitats (Hultén and Fries 1986). It flowers in June-July in northern Europe. The plant produces two oval leaves at the base of a spike-like inflorescence. The inflorescence contains 10-20 white flowers, which open sequentially, basically to apically. Nectar is secreted from unicellular hairs that cover the inside walls of the spur (Stpiczynska 1997) and the flowers emit a strong scent at night (Vogel 1990). Each flower has two pollinaria, which are situated on each side of the spur entrance. The pollinium contains hundreds of pollen massulae (Nazarov and Gerlach 1997). Platanthera bifolia varies extensively in floral traits among Scandinavian populations (paper I). On the island of Öland, off the southeast coast of Sweden, populations are longspurred in woodland habitats and short-spurred in grassland habitats (Fig. 1; paper I).. Pollination mechanism Platanthera bifolia is pollinated by several species of moths (Nilsson 1983). When a moth inserts its proboscis into the spur to feed on nectar it touches the viscidia that fold around the proboscis and the pollinaria are removed when the moth withdraws from the flower (Darwin 1862). When a moth carrying pollinia visits a flower, massulae attaches to the sticky surface of the stigma that is situated in three lobes around the spur entrance (Nilsson 1983). The fruit set of P. bifolia is pollen limited in Swedish mainland populations (Mattila and Kuitunen 2000; Maad and Alexandersson 2004).. 13.

(236) Figure 1. The two morphologically and phenologically divergent forms of Platanthera bifolia on Öland, SE Sweden. A. Side view of a flower from a woodland plant (above) and a grassland plant (below). B. Inflorescence of a grassland plant. C. Front view of a flower from a grassland plant (left) and a woodland plant (right). D. Inflorescence of a woodland plant. Scale bars = 1cm.. 14.

(237) Study populations and site descriptions My main study populations were located in grassland and woodland habitats on the Baltic island Öland, off the Swedish southeast coast (Fig. 2). The study of population differentiation in spur length included additional populations in Scandinavia (paper I). The climate on Öland is characterized by summers with low precipitation and high temperatures and a long vegetation period. Humans have kept livestock on Öland since 3900–3300 B.C., and the impact of cattle, mostly cows, sheep and horses, has formed the landscape throughout the centuries (Forslund 2001). Most grassland study populations of P. bifolia on Öland were located on the Great Alvar in the southern part of the island, except for the Melösa population, which was located in a coastal dry grassland situated on the northeast shore of Öland. The alvar are semi-natural grasslands characterized by thin, nutrient-poor sandy soils on limestone bedrock with full exposure to the sun and often strong winds. The coastal habitat is very similar to the alvar habitat, but has a slightly thicker layer of soil. The woodland habitat is characterized by deep, moraine and clay soils on limestone bedrock. The open deciduous woodlands on Öland have been affected by centuries of human activities like mowing, clearing and grazing by cattle. The woodland is dominated by oak (Quercus robur) and ash (Fraxinus excelsior), and various shrubs such as Corylus avellana, Euonymus europeus and Lonicera xylosteum. Woodland habitats represent a shadier and less windy environment that is less subject to drought than alvar habitats. The woodland study populations of P. bifolia are situated in the Mittlandskogen forest in the middle of the island. Apart from the 5 grassland and 4 woodland populations, I also studied a P. bifolia population at Jordtorp, which represents a mixed habitat. The Jordtorp site is 80m2 large, and located within a closed deciduous forest. It is characterized by a mosaic of thin, sandy soils and deeper moraine patches. The site is a semi-natural grassland with many deciduous trees and shrubs, and has occasionally been cleared and grazed by cattle.. 15.

(238) Figure 2. Map of the central parts of Öland, SE Sweden, and the position of the long-spurred and short-spurred study populations and one population bimodal for spur length.. Population differentiation and pollinators (I) The geographical variation in spur length of P. bifolia was determined in 52 populations of P. bifolia (n = 10-200 plants per population) and the proboscis length of pollinating moths was determined in 12 of the P. bifolia populations (n = 1-96 moths per population) across Sweden and Norway. Pollinating moths were observed and caught in natural populations of P. bifolia (1 – 96 moths per population) during 1973, 1980, 1984, 1985, 2002 and 2005. The moths that were caught either carried pollinia or were observed pollinating flowers. In addition, to examine floral differentiation on a more local scale, I documented variation in plant height, flower length, flower width, spur length, stem length and the number of flowers by recording the phenotype of 531 individuals in 12 populations in 7 grassland and 5 woodland populations on the island Öland, SE Sweden.. 16.

(239) Reciprocal translocation experiment (I) To determine the adaptive significance of population differentiation in flower morphology, I conducted a translocation experiment at the sites of one grassland and one woodland population on the island Öland. I used 184 cut inflorescences (one inflorescence per individual plant) from two grassland (Melösa and Bårby) and two woodland populations (Ismantorp and Gråborg) of P. bifolia. At each site, plants were randomly distributed at nodes within a quadratic grid containing 400 nodes in total, each separated by 0.5 m. Plants were left in the field for three subsequent nights. For each plant, I recorded the morphology and quantified pollination success as the proportion plants with deposited pollen massulae and the proportion plants with pollinia removed.. Phenotypic selection study (II) I investigated phenotypic selection in one grassland population (Melösa) and one woodland population (Ismantorp) during three years and in one population in a mixed habitat (Jordtorp) during one year. The effects of standardized flowering start, spur length, plant height and number of flowers on relative female and male fitness (seed output and pollen removal) were estimated with multiple regression analyses following Lande and Arnold (1983). In the population in the mixed habitat, I found that spur length was the only trait with a clearly defined bimodal distribution. Since cubic splines across the whole dataset suggested that female and male fitness was low in plants with very short spurs and with intermediate spur lengths (Fig. 4) I performed the analyses on a subset of individuals in the population. By excluding individuals with very short and very long spurs I could test the hypothesis of disruptive selection on spur length.. Estimating the strength of reproductive isolation (III) I investigated prezygotic barriers caused by geographical separation, differences in flowering time, pollinator morphology and behaviour, and among-population incompatibility. The strength of each individual barrier to gene flow was calculated as the proportional decrease in gene flow between different spur-length categories (long-spurred vs. short-spurred), relative to gene flow within spur-length categories (cf. Ramsey et al. 2003; Coyne and Orr 2004). Spatial isolation was investigated by estimating the number of plants within a certain survey area from five grassland populations and four woodland populations (Fig. 2). I assumed random mating within each survey 17.

(240) area and no pollination from outside the survey area. To estimate the probability of within and between spur-length category matings, I calculated the frequency of long-spurred and short-spurred plants in each survey area. The probability of mating within spur-length categories was calculated as the sum of the squared frequencies of each spur type, and the probability of mating between spur-length categories was calculated as two times the frequency of long-spurred plants multiplied by the frequency of short-spurred plants. I investigated temporal isolation by surveying 100 plants in each of two short-spurred populations (Bårby and Melösa) and two long-spurred populations (Gråborg and Ismantorp). I assumed that half of the fruits produced by flowers open during the period of flowering overlap were the product of mating between spur-length categories and that half were the product of mating within spur-length categories, while all fruits produced by flowers receptive outside the period of flowering overlap were the product of within-category matings. Pollinator isolation was estimated in a reciprocal translocation experiment conducted in two arrays at each of one grassland site (Ölands Skogsby) and one woodland site (Ullevi). In each array, 8 long-spurred and 8 short-spurred flowering plants were placed in pots and randomly assigned to nodes separated by 0.5 mm in a 5 x 5 m grid (one plant per node). All massulae produced by the experimental plants were stained with rhodamine B or fast green. After three nights, removal and deposition of stained massulae were recorded. The proportion of stained massulae transferred between short-spurred and longspurred plants relative to the proportion transferred between plants of the same spur category was determined. I investigated the degree of postpollination isolation by performing controlled crosses within and between one short-spurred grassland population (Bårby) and one long-spurred woodland population (Gråborg; n = 46 plants per population). The proportion of flowers forming a fruit after crosses between short-spurred and long-spurred populations relative to the proportion of flowers forming a fruit after crosses with another population of the same spur-length category was determined.. Experimental manipulation of flower morphology (IV) To examine the combined and independent effects of spur length and perianth size on pollination success and seed output I manipulated spur length (short vs. long) and perianth size (small vs. large) in a factorial design in a long-spurred woodland population. At the end of the flowering season, I recorded number of pollinia removed, number of pollinated flowers, number of fruits produced, mean fruit volume, and total fruit volume as an estimate of seed output. The number of flowers per inflorescence was recorded for each plant and included as a covariate in the analysis. 18.

(241) Results and discussion. Population differentiation and pollinator shifts (I) The results demonstrated extensive variation in spur length among habitats and showed that the primary pollinators varied across the Scandinavian distributional range of P. bifolia (Fig. 3). The mean spur length of Scandinavian populations of P. bifolia was strongly correlated with the mean proboscis length of local pollinators. Populations in the subalpine habitat were pollinated by the small geometrid moth Entephria caesiata, which has a very short proboscis, populations in grassland habitats were primarily pollinated by the short-proboscid hawkmoth Deilephila porcellus, populations in deciduous woodland habitats were primarily pollinated by the long-proboscid pollinator Sphinx ligustri and populations in coniferous woodland habitats were primarily pollinated by the medium-proboscid Hyloicus pinastri (Fig. 3). On the island Öland, I documented marked morphological differences between P. bifolia populations in dry grasslands and deciduous woodlands, respectively. Plants were long-spurred, tall and large-flowered in woodland populations and short-spurred, short and small-flowered in grassland populations. Furthermore, a reciprocal transplant experiment conducted between a woodland and grassland population of P. bifolia on Öland suggested that these differences are genetically based (paper IV). The proboscis length of the primary pollinators of woodland and grassland populations differed by nearly 20 mm. In the translocation experiment, long-spurred plants had higher pollination success than short-spurred plants both at the site of the longproboscid pollinator and at the site with a short-proboscid pollinator. Taken together, the results suggest that intraspecific variation in spur length reflects adaptive evolution in response to selection exerted by the locally most important pollinator, but also indicate that costs associated with producing a large inflorescence with long spurs may influence the evolution of the optimal floral phenotype in P. bifolia.. 19.

(242) Figure 3. Mean spur length and proboscis length of pollinating moths in populations of Platanthera bifolia in different regions and habitats (± SE).. 20.

(243) Disruptive and divergent selection (II) I detected disruptive selection on spur length in the population bimodal for spur length in a mixed habitat. In one of three years, there was evidence of divergent selection on plant height between the grassland and woodland population. In that year, relative female fitness increased with plant height in the population where plants were tall-statured, but tended to decrease with plant height in the population where plants were short-statured. However, in the other two years, selection tended to favour taller plants in both populations. In all three years, there was selection for longer nectar spurs in the short-spurred grassland population and a tendency of selection for long spurs also in the long-spurred woodland population. There was directional selection for earlier flowering and more flowers through female function in most years in both the woodland and grassland population. Stabilizing selection was not detected in any population. In the population in the mixed habitat, there was a significant positive quadratic coefficient for spur length through female function, which was consistent with disruptive selection (Fig. 4).. B) 6. Relative fitness. 5. Frequency. 20. 4 15. 3. 10. 2. 5. 1. 0. 0 -2. -1. 0. 1. Spur length. 2. 3. Relative female fitness. A) 25. 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -1.5. -1.0. -0.5. 0.0. 0.5. 1.0. 1.5. Spur length. Figure 4. A) The frequency distribution of standardized values for spur length in the population in a mixed habitat (n = 176) and the relationship between spur length and relative female fitness (seed output; straight line) and male fitness (pollen removal; dashed line), analysed with cubic splines. B) The relationship between relative female fitness (seed output) and standardized spur length estimated with a quadratic regression analysis confined to plants with intermediate spur lengths (n = 39).. The results gave partial support for the hypothesis that current selection is responsible for the maintenance of population differentiation of P. bifolia. The lack of difference in optimal spur length, plant stature and flowering phenology might be due to the limitations of the fitness measure that only consider seed output in a single year. Furthermore, limited spur length variation might have precluded the detection of stabilizing selection in the grassland and woodland population.. 21.

(244) The results support the hypothesis disruptive selection contributes to the maintenance of a bimodal distribution of spur length in P. bifolia in mixed habitats, but provide very limited support for the hypothesis that divergent selection contributes to the maintenance of population differentiation in morphology and flowering phenology.. Intraspecific reproductive isolation (III) There was evidence of strong reproductive isolation between the shortspurred grassland populations and the long-spurred woodland populations of P. bifolia on Öland. Spatial isolation, differences in flowering time, pollinator morphology and behaviour were responsible for impeding a large proportion of potential pollen flow between morphologically divergent populations. Spatial separation impeded 78 % of potential gene flow between long-spurred and short-spurred plants in the woodland habitat, and 98 % of potential gene flow in the grassland habitat. Reproductive isolation due to pollinators and flowering time was asymmetric. The difference in flowering time between long-spurred and short-spurred populations was responsible for impeding 86 % of potential pollen flow from short-spurred to long-spurred plants and 28 % was impeded in the opposite direction. In sympatric experimental populations, pollinator isolation was responsible for excluding 88 % of potential gene flow from short-spurred to long-spurred plants at the woodland site. At this site, short-spurred plants had very low pollen deposition, but the few plants that were pollinated were not reproductively isolated from the long-spurred plants. At the grassland site, 58 % of the pollen flow from long-spurred to short-spurred plants was impeded by the pollinator barrier. Only 17 % of the gene flow was impeded in the opposite direction. Furthermore, I documented a postpollination isolation mechanism manifested as a 49 % reduction in fruit formation in short-spurred plants when pollinated with pollen from long-spurred plants compared to pollen from short-spurred plants. In contrast, no reproductive isolation was detected when long-spurred plants were pollinated with pollen from short-spurred plants. Thus, the post-pollination barrier was working in the opposite direction to the pollinator and temporal barrier. The results suggest that several components of reproductive isolation contribute to the maintenance of population differentiation in P. bifolia.. 22.

(245) The adaptive significance of spur length and perianth size (IV) The results of the manipulation experiment suggest that spur length is critical for reproductive success in P. bifolia (Fig. 5). However, results do not support the hypothesis that pollination success is affected by perianth size, or that the correlation between spur length and perianth size observed within and among populations of P. bifolia is maintained by correlational selection. Plants with long spurs had more flowers pollinated, more pollen removed and produced more and larger fruits compared to plants with short spurs. In contrast, perianth size did not affect the pollination success or fruit production of P. bifolia. Morphological divergence among populations often involves several correlated characters. Covariation of traits among populations may, besides from correlational selection, be due to genetic correlations among traits and divergent selection on one or a few of these traits (Armbruster and Schwaegerle 1996; Ashman and Majetic 2006). Population differentiation in perianth size within P. bifolia may be driven by pollinator-mediated selection on spur length and genetic correlations between spur length and perianth size. 1200. number of pollinia removed. 20. total fruit volume. 1000 800 600 400 200 0. 16 12. small perianth large perianth. 8 4 0. short spurs. long spurs. short spurs. long spurs. Figure 5. The effects of spur length (short vs. long) and perianth size (small vs. large) on total fruit volume and the number of pollinia removed in a long-spurred, large-flowered population of Platanthera bifolia (mean ± SE).. Conclusions In this thesis, I have explored the evolutionary mechanisms responsible for the evolution and maintenance of population differentiation in floral traits. The results suggest that population differentiation in spur length reflects adaptive evolution in response to selection exerted by the locally most effective pollinator, and demonstrate how variation in local selection 23.

(246) regimes and reproductive isolation contribute to the maintenance of population differentiation in the moth-pollinated orchid Platanthera bifolia. The substantial among-population variation in spur length was related to variation in pollinator morphology, which suggests adaptive evolution in response to pollinator shifts. The results suggest that disruptive selection contributes to the maintenance of a bimodal distribution of spur length in mixed habitats. Remarkably few studies have tested the hypothesis of disruptive selection on divergent plant traits. The results indicate the potential for detecting this diversifying form of selection in contact zones between differentiated populations. More studies in other plant-pollinator systems are needed to evaluate the importance of disruptive selection in floral evolution. I have demonstrated experimentally that spatial isolation, differences in flowering time, pollinator-mediated assortative mating and postpollination incompatibility result in strong reproductive isolation between morphologically divergent populations. To the best of my knowledge, only one previous study has quantified the individual strength of intraspecific reproductive isolation barriers in plants (Lowry et al. 2008b). The results suggest that pollinator isolation and temporal barriers result in asymmetric pollen transfer between divergent populations, which is counterbalanced by a crossing barrier. The compensatory effect of the crossing barrier suggests a role of reinforcement in the evolution of postpollination isolation between populations, which would be interesting to investigate in further studies. In the translocation experiment, the tall-statured, long-spurred form of P. bifolia had a higher pollination success in both the woodland and the grassland habitat. Furthermore, in the phenotypic selection study, selection for longer spurs was detected in both the grassland and woodland population. Although the results in many aspects support the hypothesis that population differentiation reflects adaptive evolution in response to pollinator-mediated selection, a full understanding of the mechanisms underlying floral evolution in P. bifolia will only be possible when demographic costs associated with producing tall inflorescences and longspurred flowers have been investigated. In this thesis, I have explored the association between population differentiation in floral traits and pollinator-mediated selection. Taken together, the results suggest that pollinator-mediated selection can shape the evolution of intraspecific floral variation.. 24.

(247) Summary in Swedish. Långa sporrar och matchande tungor - hur du blir en framgångsrik nattviol Varför finns det så många blommor? Blomväxternas till synes oändliga variationsrikedom utgör fortfarande ett högst aktuellt forskningsområde inom evolutionsbiologin. Nya arter och varianter av arter kan uppstå genom de val blombesökare som bin och fjärilar gör när de söker föda. Här presenterar jag ny forskning som tyder på att samspelet mellan orkidéer och pollinerande nattfjärilar styrt evolutionen av olika varianter av nattviol. Det är en varm sommarkväll i början av juni och en ligustersvärmare vaknar upp och börjar spinna med sina vingar för att få upp värmen. En söt kaneldoft sprider sig över Ölands skogsängar och når slutligen ligustersvärmarens förgrenade antenner. Nattfjärilen följer doftspåret genom den snåriga Mittlandsskogen tills den når en glänta. Där, omgiven av sin starka doft, lyser den som en fackla i dunklet, nattviolen. Två varianter av nattviol upptäckta på Öland På Öland finns det två varianter av nattviol (Platanthera bifolia). En storväxt variant som växer i skogsmark och en mindre som växer på alvarmark. Hur har dessa varianter uppstått? Och vad är det som håller dem isär? Dessa frågor har jag arbetat med i mitt doktorandprojekt. Nattviolen är i motsats till vad namnet antyder en orkidé. De vita blommorna doftar starkt i skymningen för att locka till sig pollinerande nattfjärilar. Den vita färgen står i klar kontrast till den gröna växtligheten, också det en anpassning till nattfjärilarnas sena kvällsvanor. Som belöning för nattfjärilarnas arbete erbjuder blommorna energirik nektar som bildas i sporren, en slangliknande utväxt vid blommans bas.. 25.

(248) En nattviolsblomma med sin långa sporre (till vänster), samt en skogsnattviol som besöks av en ligustersvärmare (till höger). Foto: Steve Johnson.. Jag har visat att de öländska varianterna av nattviol skiljer sig från varandra på många sätt. Skogsnattviolen börjar blomma ungefär två veckor tidigare än alvarvarianten. Skogsnattviolen är högväxt med ungefär 4 centimeter långa sporrar och har stora blommor medan alvarnattviolen är lågväxt med ungefär hälften så långa sporrar och små blommor. En förklaring till varför nattviolerna är större i skogen än på alvaret skulle kunna vara att skogen erbjuder en mer skyddad och näringsrik miljö. Jag flyttade plantor mellan skog och alvar för att ta reda på om skillnaderna mellan dem har en genetisk grund eller beror på miljön de lever i. Ännu två år efter utplanteringen kvarstod alla mätbara skillnader vilket talar för att de har en genetisk grund. Olika sorters svärmare pollinerar nattviolerna. Svärmare är en grupp nattfjärilar som inte landar utan ”svävar” framför blomman när de suger nektar. Ligustersvärmaren som är den största svärmaren i Sverige besöker ofta vanliga trädgårdsväxter som syren och kaprifol under försommarkvällarna. Det händer att den misstas för en kolibri av förvånade trädgårdsägare! När svärmaren besöker en nattviolsblomma sticker den ner sin sugsnabel i sporren på jakt efter nektar. Snabeln kommer då åt två klibbiga plattor som genast viker sig runt snabelfästet. När fjärilen drar ur sugsnabeln följer pollenklubborna med, fastklistrade på snabelns bas och vid nästa blombesök kan pollenet fastna på pistillens märke. När blomman pollinerats börjar blommans fruktämne så småningom svälla till en kapsel fylld med tusentals små frön som sedan sprids med vinden.. 26.

(249) Ju längre desto bättre Evolutionen av blommornas utseende styrs i hög grad av beteendet hos de insekter som pollinerar dem. När Charles Darwin 1862 fick se orkidén madagaskarstjärna med sina 30 centimeter långa sporrar fick han ytterliggare en pusselbit till sitt livsverk om evolutionsteorin. Darwin insåg att evolutionen av växternas långa sporrar drivs av längden på sugsnabeln hos de insekter som pollinerar dem. Bara en fjäril med en sugsnabel som matchar orkidéns sporre skulle kunna suga nektar och samtidigt ombesörja att blomman pollineras. En viktig skillnad mellan skog och alvar på Öland är deras nattfjärilsfauna. På alvaret är en mindre svärmarart som kallas liten snabelsvärmare vanlig. Den har en kort sugsnabel. I skogen är däremot ligustersvärmaren vanligare och den har en mycket lång sugsnabel. Jag har visat att det finns en slående överensstämmelse mellan längden på sporren hos de öländska nattviolerna och längden på sugsnabeln hos de lokalt vanligaste svärmarna. Jag fann samma mönster även över nattviolens hela utbredningsområde i Skandinavien där också andra nattfjärilsarter pollinerar nattviolerna. Sambandet mellan snablarnas och sporrarnas längd tyder på att nattfjärilarna styrt evolutionen av sporrens längd hos nattviolen. Men för att verkligen kunna bevisa det krävs experiment som undersöker sambandet mellan nattviolernas sporrelängd och deras fortplantningsframgång i olika miljöer med varierande nattfjärilsfauna. Vad händer om en skogsplanta sätts ut på alvaret och en alvarplanta sätts ut i skogen? Missgynnas de gentemot de lokala nattviolerna? Jag fann att den långsporrade varianten var mer framgångsrik än den kortsporrade i skogen, där de flesta fjärilarna har långa snablar, men också på alvaret, där de flesta fjärilarna har korta snablar. Detta tyder på att det finns andra faktorer än pollinatörer som påverkar framgången hos de två varianterna. En hypotes är att kostar det på att producera en stor planta med långa sporrar i en sådan resursfattig miljö som alvaret. Skillnaden i resurstillgång mellan miljöerna i kombination med det urval som pollinatörerna utövar kan vara nyckeln till den variationsrikedom vi finner hos de Öländska nattviolerna. Manipulerade nattvioler avslöjar det naturliga urvalet Darwins teori om evolutionen av långa sporrar har bekräftats genom experiment där sporrens längd förkortats vilket lett till en sämre fortplantning hos växten. Jag var intresserad av att på ett liknande sätt ta reda på om sporrens längd och blommans storlek, som ju skiljer sig avsevärt mellan alvar- och skogsnattvioler, var en evolutionär anpassning. För att ta reda på detta använde jag mig av en klipp och klistra-teknik för att skapa blommor med olika utseende och sedan mäta deras fortplantningsframgång. Jag klippte av blomkronor och förkortade sporrar (genom att vika dem bakåt 27.

(250) och fästa dem med tejp) i en population av storblommig, långsporrad skogsnattviol. Jag fann att en förkortad sporre ledde till sämre fortplantning medan en förminskning av blommans kronblad inte hade någon effekt. Skillnader i blomstorlek mellan alvar- och skogsnattvioler verkar inte vara en evolutionär anpassning utan kan vara resultatet av att denna egenskap ”åkt snålskjuts” på evolutionen av sporren. Ofta är gener sammankopplade på så vis att det naturliga urvalet på en viss egenskap indirekt ger effekter även på andra egenskaper. Lagom är inte alltid bäst Det naturliga urvalet är inte bara en historisk företeelse som bidragit till den artrikedom vi idag ser omkring oss, utan är en ständigt pågående och mätbar process. En het debatt inom artbildningsforskningen rör huruvida nya arter kan uppstå genom naturligt urval inom en och samma population. I en del populationer av darwinfinkar på Galápagosöarna har forskarna funnit två olika näbbformer som utvecklats för att kunna utnyttja olika föda. I samma population finner man också finkar med en intermediär näbbform som har svårt att utnyttja någon av födokällorna och som därför missgynnas av det naturliga urvalet. Ett liknande samband har jag funnit hos nattviolerna på Öland. I en population där alvar- och skogsnattvioler växer tillsammans fann jag individer med halvlånga sporrar, mitt emellan alvar- och skogsnattviolernas sporrelängd. Genom att mäta sporrens längd och mängden frö som producerats för varje individ fann jag ett samband som liknade det för darwinfinkarna. Individer med sporrar som varken är långa eller korta producerar färre avkommor än de individer som har antingen långa eller korta sporrar. Det naturliga urvalet bidrar på så sätt till att filtrera bort korsningar mellan skogs- och alvarnattvioler på Öland, något som i sinom tid hade kunnat leda till att de blev två skilda arter. Vem har sex med vem? Idag vet vi att alvar- och skogsnattvioler ibland växer sida vid sida och därför rent hypotetiskt skulle kunna fortplanta sig med varandra. Fortplantning mellan varianterna skulle innebära att de genetiska skillnaderna mellan dem suddas ut och efter hand försvinner. Jag har ägnat en del av min forskning åt att undersöka vem som egentligen har sex med vem. Jag har kommit fram till att fortplantningen i huvudsak sker inom respektive variant. Det finns en rad faktorer som hindrar varianterna från att korsa sig med varandra. Med hjälp av olika experiment har jag räknat ut vilka av dessa faktorer som är viktiga. Skillnader i deras rumsliga fördelning i landskapet, blomningstiden, pollinationen och en fysiologisk korsningsbarriär som gör att få frukter bildas från korsningar mellan varianterna bidrar alla till att varianterna sällan korsar sig med varandra.. 28.

(251) Detta kan förklara hur skillnaden mellan alvar- och skogsnattvioler bevaras trots det ofta mycket korta avståndet mellan dem. Skymningen har övergått i natt och mörkret sänker sig över Mittlandsskogen. Nattfjärilen har under kvällen ihärdigt besökt ett stort antal nattvioler för att tillgodose sitt behov av energirik nektar och pollenklubborna sitter som en krona på snabelfästet. Men nya doftämnen av honlig karaktär studsar nu mot doftreceptorerna på hans antenner och hans sökande tar sig en annan vändning…. Mina viktigaste resultat: •. Det finns ett geografiskt samband mellan längden på nattviolernas sporre och de pollinerande nattfjärilarnas snabel.. •. Mellanformer av skog- och alvarnattviol missgynnas av det naturliga urvalet.. •. Parning mellan varianterna motverkas av den rumsliga uppdelningen i landskapet, skillnaden i blomningstid, pollinatörernas morfologi och beteende samt en fysiologisk korsningsbarriär.. •. Längre sporrar hos skogsvarianten är sannolikt en evolutionär anpassning medan stora blomkronor troligtvis är en egenskap som ”åkt snålskjuts” på evolutionen av sporren.. 29.

(252) Acknowledgements. First, I want to thank Steve Johnson and Anders Nilsson who first attracted me to the field of orchid evolution and has inspired me immensely throughout the years. Thanks to Anders Nilsson for confiding much of his data with me, I hope I did his vision some justice. I am very grateful to my supervisor Jon Ågren, for his patience and for teaching me the art of science and thoroughness. Thanks to Ronny Alexandersson and Johanne Maad for always being there when I needed scientific advice or pep-talks. Thanks to Anna-Karin Borg-Karlsson, Mikael Hedren and Pernilla Ellneskog-Stamm for inviting me to their respective lab, Per Toräng, Nina Sletvold, Steve Johnson, Ronny Alexandersson, Johanne Maad for valuable comments on the manuscripts and Håkan Rydin, Per Toräng and Bengt Carlsson for comments on the Swedish summary, and my dear friend and statistical advisor Saskia Sandring. Thanks to all the students and field assistants who helped me throughout the years, and the people of Öland who opened their land and their homes to me, Eje Rosén and all my friends at Station Linné, Kristin and Mats Bertilius for friendship, housing and field work, and to all staff at the Plant Ecology department and former and new PhD students of VäxtBio. What would a day at work be like without the little things? Fågelsången and morning lattes with Sylvia, myspys, lyxlunch and fruktstund with Norbert and Per and the morning update from Lollo and Sandra? Thanks to my family in Helsingborg and Göteborg for the love and support, and my brother Carl Boberg, for the illustration on the cover. I want to thank my son Alvin for giving me perspective on life. And last, thank you Jonas for all your hard work, in the field and at home. You helped me kill the monster.. 30.

(253) References. Armbruster, W. S., and K. E. Schwaegerle. 1996. Causes of covariation of phenotypic traits among populations. Journal of Evolutionary Biology 9:261-276. Ashman, T. L., and C. J. Majetic. 2006. Genetic constraints on floral evolution: a review and evaluation of patterns. Heredity 96:343-352. Benkman, C. W. 2003. Divergent selection drives the adaptive radiation of crossbills. Evolution 57:1176-1181. Caruso, C. M., S. B. Peterson, and C. E. Ridley. 2003. Natural selection on floral traits of Lobelia (Lobeliaceae): Spatial and temporal variation. American Journal of Botany 90:1333-1340. Castellanos, M. C., P. Wilson, and J. D. Thomson. 2003. Pollen transfer by hummingbirds and bumblebees, and the divergence of pollination modes in Penstemon. Evolution 57:2742-2752. Chittka, L., J. D. Thomson, and N. M. Waser. 1999. Flower constancy, insect psychology, and plant evolution. Naturwissenschaften 86:361377. Conner, J. K. 2002. Genetic mechanisms of floral trait correlations in a natural population. Nature 420:407-410. Conner, J. K., and S. Rush. 1996. Effects of flower size and number on pollinator visitation to wild radish, Raphanus raphanistrum. Oecologia 105:509-516. Coyne, J. A., and H. A. Orr. 2004. Speciation. Sinauer, Sunderland, MA. Darwin, C. 1862. On the various contrivances by which British and foreign orchids are fertilised by insects, and on the good effects of intercrossing. J. Murray, London. Davis, S. L. 2001. Phenotypic and genetic correlations among floral traits in two species of Thalictrum. Journal of Heredity 92:361-366. Forslund, M. 2001. Natur och Kultur på Öland. Länstyrelsen i Kalmar län, Kalmar, Sweden. Galen, C., and M. E. A. Newport. 1987. Bumble bee behavior and selection on flower size in the Sky Pilot, Polemonium viscosum. Oecologia 74:20-23. Grant, V. 1949. Pollination systems as isolating mechanisms in angiosperms. Evolution 3:82-97. Grant, V. 1994. Modes and origins of mechanical and ethological isolation in angiosperms. Proc. Natl. Acad. Sci. U. S. A. 91:3-10. Grant, V., and K. A. Grant. 1965. Flower pollination in the Phlox family. Columbia Univ. Press, New York. 31.

(254) Hall, M. C., and J. H. Willis. 2006. Divergent selection on flowering time contributes to local adaptation in Mimulus guttatus populations. Evolution 60:2466-2477. Harder, L. D., and S. C. H. Barrett, eds. 2006. Ecology and evolution of flowers. Oxford University Press, UK. Hendry, A. P., S. K. Huber, L. F. De Leon, A. Herrel, and J. Podos. 2009. Disruptive selection in a bimodal population of Darwin's finches. Proceedings of the Royal Society B-Biological Sciences 276:753759. Herrera, C. M., M. C. Castellanos, and M. Medrano. 2006. Geographical context of floral evolution Pp. 278-294 in L. D. Harder, and S. C. H. Barrett, eds. Ecology and evolution of flowers. Oxford Univ. Press, Oxford. Hultén, E., and M. Fries. 1986. Atlas of North European vascular plants: north of the Tropic of Cancer. Koeltz, Königstein. Johnson, P. A., F. C. Hoppensteadt, J. J. Smith, and G. L. Bush. 1996. Conditions for sympatric speciation: A diploid model incorporating habitat fidelity and non-habitat assortative mating. Evolutionary Ecology 10:187-205. Johnson, S. D. 2006. Pollinator-driven speciation in plants. Pp. 295-310 in L. D. Harder, and S. C. H. Barrett, eds. Ecology and evolution of flowers. Oxford Univ. Press, Oxford. Johnson, S. D., and K. E. Steiner. 1997. Long-tongued fly pollination and evolution of floral spur length in the Disa draconis complex (Orchidaceae). Evolution 51:45-53. Kennedy, B. F., H. A. Sabara, D. Haydon, and B. C. Husband. 2006. Pollinator-mediated assortative mating in mixed ploidy populations of Chamerion angustifolium (Onagraceae). Oecologia 150:398-408. Lande, R., and S. J. Arnold. 1983. The measurement of selection on correlated characters. Evolution 37:1210-1226. Levin, D. A. 2006. Flowering phenology in relation to adaptive radiation. Syst. Bot. 31:239-246. Lowry, D. B., J. L. Modliszewski, K. M. Wright, C. A. Wu, and J. H. Willis. 2008a. The strength and genetic basis of reproductive isolating barriers in flowering plants. Philosophical Transactions of the Royal Society B-Biological Sciences 363:3009-3021. Lowry, D. B., R. C. Rockwood, and J. H. Willis. 2008b. Ecological reproductive isolation of coast and inland races of Mimulus guttatus. Evolution 62:2196-2214. Maad, J., and R. Alexandersson. 2004. Variable selection in Platanthera bifolia (Orchidaceae): phenotypic selection differed between sex functions in a drought year. Journal of Evolutionary Biology 17:642650. Maad, J., and L. A. Nilsson. 2004. On the mechanism of floral shifts in speciation: gained pollination efficiency from tongue- to eyeattachment of pollinia in Platanthera (Orchidaceae). Biological Journal of the Linnean Society 83:481-495. 32.

(255) Mattila, E., and M. T. Kuitunen. 2000. Nutrient versus pollination limitation in Platanthera bifolia and Dactylorhiza incarnata (Orchidaceae). Oikos 89:360-366. Nazarov, V. V., and G. Gerlach. 1997. The potential seed productivity of orchid flowers and pecularities of their pollination systems. Lindleyana 12:188-204. Nilsson, L. A. 1983. Processes of isolation and introgressive interplay between Platanthera bifolia (L) Rich and P. chlorantha (Custer) Reichb.(Orchidaceae). Botanical Journal of the Linnean Society 87:325-350. Nilsson, L. A. 1988. The evolution of flowers with deep corolla tubes. Nature 334:147-149. Ordano, M., J. Fornoni, K. Boege, and C. A. Dominguez. 2008. The adaptive value of phenotypic floral integration. New Phytologist 179:11831192. Pauw, A., J. Stofberg, and R. J. Waterman. 2009. Flies and flowers in Darwin's Race. Evolution 63:268-279. Ramsey, J., H. D. Bradshaw, and D. W. Schemske. 2003. Components of reproductive isolation between the monkeyflowers Mimulus lewisii and M. cardinalis (Phrymaceae). Evolution 57:1520-1534. Rice, W. R., and G. W. Salt. 1990. The evolution of reproductive isolation as a correlated character under sympatric conditions - experimentalevidence. Evolution 44:1140-1152. Rueffler, C., T. J. M. Van Dooren, O. Leimar, and P. A. Abrams. 2006. Disruptive selection and then what? Trends in Ecology & Evolution 21:238-245. Schemske, D. W., and H. D. Bradshaw. 1999. Pollinator preference and the evolution of floral traits in monkeyflowers (Mimulus). Proceedings of the National Academy of Sciences of the United States of America 96:11910-11915. Schemske, D. W., and J. Ågren. 1995. Deceit pollination and selection on female flower size in Begonia involucrata - an experimental approach. Evolution 49:207-214. Schluter, D. 2000. The ecology of adaptive radiation. Oxford University Press. Stpiczynska, M. 1997. The structure of nectary of Platanthera bifolia L. Orchidaceae. Acta Societatis Botanicorum Poloniae 66:5-11. Whittall, J. B., and S. A. Hodges. 2007. Pollinator shifts drive increasingly long nectar spurs in columbine flowers. Nature 447:706-U712. Vogel, S. 1990. The role of scent glands in pollination. Smithsonian Institution Libraries, Washington.. 33.

(256) , - ) )-) )).

(257)

(258)

(259)

(260)

(261)

(262)

(263) './ 0

(264) 1 .. 1 , )) 2 0

(265) 1 .. 13-) - ) 13)

(266) )

(267) 1)

(268) 221

(269) 2( )4,5 ) 2 )) 6 27 5)) ()35 )

(270) 221 )) (

(271) 1

(272) ))/ 8 2)

(273) 22) -) /)) ) 2 0

(274) 1 .. 14 9: ;

(275) 13++#3 ))5)

(276) ( )

(277) <8 2)

(278) 22) -) /)) ) 2 0

(279) 1 .. 1=4>. /) (

(280) '

(281) ( )4

(282)

(283) 4)

(284) '(')'

(285)

(286) ' *+&".

(287)

(288)

(289) .

(290)

No results found