Документ взят из кэша поисковой машины. Адрес оригинального документа : http://herba.msu.ru/shipunov/author/shipunov2005b.pdf
Дата изменения: Fri Apr 29 20:58:40 2005
Дата индексирования: Mon Oct 1 22:15:09 2012
Кодировка:

Blackwell Science, LtdOxford, UKBIJBiological Journal of the Linnean Society0024-4066The Linnean Society of London, 2005? 2005 85? 112 Original Article DACTYLORHIZA DIVERSITY IN EUROPEAN RUSSIA A. B. SHIPUNOV and R. M. BATEMAN

Biological Journal of the Linnean Society, 2005, 85, 1 12. With 9 figures

Geometric morphometrics as a tool for understanding Dactylorhiza (Orchidaceae) diversity in European Russia
ALEXEY B. SHIPUNOV1 and RICHARD M. BATEMAN2*
1 2

Jodrell Laboratory, Royal Botanic Gardens Kew, Richmond, Surrey, TW9 3DS, UK Department of Botany, Natural History Museum, Cromwell Road, London, SW7 5BD, UK

Received 18 December 2003; accepted for publication 28 June 2004

Geometric morphometric techniques were employed to assess the diversity of lip shapes (305 samples from 83 populations) in flowers of European Russian Dactylorhiza (Orchidaceae: Orchidinae). We found significant agreement between the results from geometric morphometrics, classic morphometrics and the distribution of certain nuclear DNA markers. The lip shapes from Arctic Dactylorhiza samples occupied an intermediate position between D. maculata and D. fuchsii samples from Central Russia, supporting a hybrid origin of `northern tetraploids'. Lip shapes of the taxonomically controversial allotetraploid D. baltica were found to form a distinct group, with members having definite relationships with diploid D. incarnata samples from the same localities, indicating either their local origin or introgression with D. incarnata. In addition to demonstrating the value of geometric morphometric methods in studies of plant taxonomy and hybridization, we suggest future applications designed to explore pollinator-driven directional selection, developmental constraints and fluctuating asymmetry. © 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112.

ADDITIONAL KEYWORDS: developmental constraints diploid directional selection fluctuating
asymmetry Internal Transcribed Spacer landmarks tetraploid.

INTRODUCTION
Geometric morphometry (e.g. Bookstein, 1991; Pavlinov, 2001) is a relatively new technique that has generated valuable results in many fi elds of classic morphometry. The major difference from classic methods is the ability to understand the form of an object directly, as a cohesive whole, rather than indirectly via fragmentary measurements . Thus, the chosen geometric morphometric variables allow complete reconstruction of the shape . Two different kinds of geometric morphometry are most widely used: Fourier analysis of shape curves, and landmarkbased methods such a Thin Plate Splines (TPS) analysis (Adams, Rohlf & Slice, 2002; Jensen, Ciofani & Miramontes, 2002). Fourier analysis calculates several `shape curves' from the object outline and then derives `Fourier coefficients' that represent these curves; the coefficients

*Corresponding author. E-mail: r.bateman@nhm.ac.uk

can be used as variables for multivariate analysis of the objects investigated ( Jensen et al., 2002). The landmark approach is based on placing on the shape image several so-called `landmarks' pinpointing the most important locations on the object. These points are assumed to be homologous, at least in a geometric sense, because landmark-based methods operate only with the coordinates of these reference points , so the objects studied should be directly comparable (Pavlinov, 2001). TPS reveals the degree of `bending energy' necessary to transform a rectangular grid superimposed on one shape to fi t another shape (Rohlf & Slice, 1990, developing much earlier qualitative work by Thompson, 1917). Multivariate methods can also be applied to the results of TPS (Rohlf, Loy & Corti, 1996; Baylac, Villemant & Simbolotti, 2003). There are many taxonomic applications of TPS and other geometric morphometry methods (e .g. Ray, 1992; Kores, Molvray & Darwin, 1993; Pavlinov, 2000; Rueber & Adams, 2001; Jensen et al., 2002; Guill, Heins & Hood, 2003), but most of them have hitherto been performed on zoological and/or palaeontological objects

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

1

2

A. B. SHIPUNOV and R. M. BATEMAN before choosing a nine-landmark confi guration for the whole investigation (Fig. 1). This set of landmarks was based on points commonly used as the basis for multiple linear metric measurements in `classic' morphometric analyses (e.g. Bateman & Denholm, 1983, 1985; Tyteca & Gathoye, 2000). Most of the landmarks chosen were homologous in a biological sense , though landmarks 2 and 8 had only geometric support. Following the procedure, we obtained a large TPS-file with 18 (9 landmarks Ґ 2 dimensions) landmark coordinates for each specimen. The next step used two approac hes: (1) the first, group-based approach aimed to describe all the data simultaneously, so for each population we used a group average configuration revealed from the consensus configurations in the separate population analyses (Rohlf et al., 1996). These average configurations were returned by the tpsRelw program ( Rohlf, 2003b) and were based on the generalized orthogonal leastsquares Procrustes (GPA) procedures described by Rohlf & Slice (1990). (2) The second approach consisted of specifying reasonable subsets of data that could shed light on selected problems in Russian Dactylorhiza taxonomy; in this case, the data matrix based on individual plants was used.

(MacLeod & Forey, 2002). Botanical investigations using TPS (or other geometric morphometry methods) are sparse, despite the fact that many botanical features of taxonomic importance (e.g. leaves, sepals, petals) fit well with typical geometric morphometry conditions. There are many groups of plants that still require more intensive taxonomic treatments , especially those that apply new methods. One such group is Dactylorhiza Necker ex Nevski, an orchid genus that is common and widespread in temperate Eurasia. The level of taxonomic investigation of western European dactylorc hids has been relatively high (Nelson, 1976; Pedersen, 1998; Bateman, 2001), but nevertheless many questions remain unresolved. Furthermore, the eastern Dactylorhiza species, especially those from European Russia and Siberia, have been little investigated. Many unanswered questions concern allotetraploid species aggregates suc h as D. majalis, D. traunsteineri and D. baltica (HedrИn, 2002). Another problem is elucidating the relationships within the D. fuchsiiD. maculata species complex, which includes both diploids and autotetraploids (e.g. Bateman & Denholm, 2003). Earlier morphometric analyses of these problems ha ve often pointed to the form of the lip (= the labellum: the insect landing stage in the flower) as a particularly valuable source of diagnostic characters (e.g. Heslop-Harrison, 1948, 1954; Bateman & Denholm, 1985; Reinhard, 1990; Tyteca & Gathoye, 2000). In Dactylorhiza, unlike some other orchids, this structure is relatively fl at (and can be fully flattened without serious damage), contains no holes and is of macroscopic size (usually around 1 cm maximum in diameter). Thus, the dactylorchid lip is potentially an excellent model for investigations using geometric morphometry.

MATERIAL AND METHODS
Putative species analysed from Russia were the diploids D. incarnata and D. fuchsii, the autotetraploid D. maculata and the allotetraploid D. baltica. Also included were placeholders for allotetraploids from western Europe: D. purpurella, D. praetermissa, D. majalis s.s., D. traunsteineri s.s., D. lapponica, D. russowii and D. sphagnicola, together with the diploid D. euxina. The more distantly related Russian species D. flavescens was also measured. We used the flattened and dried flowers from 305 plants representing 83 populations of Dactylorhiza species and hybrids from European Russia and the British Isles (for sampling details see Shipunov et al., 2004). The lips were scanned on a table scanner (HP Scanjet 7400c) into digital bitmap images with a resolution of 200 dpi, and then the computer digitizer program tpsDig (Rohlf, 2003a) was used to plot appropriate landmarks. We tested several landmark sets

9

1

8 2 6 4

7

5

3

Figure 1. Relative positions on the lip of the nine selected landmark points for geometric morphometric analysis of Dactylorhiza species.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

DACTYLORHIZA DIVERSITY IN RUSSIA The tpsRelw program was then employed to calculate principal warps the orthogonal vectors that reflect all possible transformations from the consensus configuration to each individual configuration (Fig. 2). The partial warps were calculated as coordinates of projections of specimens on eac h of the vectors, and the resulting matrix (`weight matrix' of Rohlf et al., 1996) was used for principal components analysis (PCA) and multidimensional scaling (MDS). All `traditional' statistic calculations used the R program, version 1.71 for Windows (Venables et al., 2002). In order to obtain taxonomic `anchors', illustrations of mounted flowers in Reinhard (1990) were digitized and included in the analysis, in which they were denoted by upper-case letters. Sadly, the many hundreds of flowers mounted by Bateman & Denholm (e.g. 1985, 1989) could not be used satisfactorily due to the dissection of the spur away from the labellum during mounting, which marginally excised landmarks 1 and 9.

3

RESULTS
GROUP-BASED
ANALYSIS

PCA analysis of the group-based data matrix revealed three most important principal components that

encompassed 75.9% of the total variance . Although the second and third components accounted for similar amounts of variance, the ordination of PC1 and PC3 returned more taxonomic structure (Fig. 3) than did PC1 vs PC2, which served only to separate D. fuchsii samples. In general, clustering occurred between putative species according to a priori taxonomic assumptions , though gaps were not discernible between any putative species. There was also some overlap of taxa in the centre of the graph, where several samples of D. incarnata, D. fuchsii, D. baltica and D. purpurella were indistinguishable. At same time, the discriminant analysis of this data matrix fully supported species assignment (c2 = 1269.3, P < < 0.05). Also of interest was the extensive dispersion of D. incarnata specimens, which overlapped both with D. praetermissa samples and with the D. russowii, D. lapponica and D. praetermissa `anchors'. Four D. baltica populations formed a group located within D. fuchsii but far from D. praetermissa. In contrast, the D. euxina sample was placed close to D. praetermissa. The D. maculata and D. fuchsii populations from the Russian Arctic occupied a position intermediate between the centres of these two species. The hybrids (presumably triploids) between the diploid D. fuchsii and the tetraploid

Figure 2. Overall consensus configuration showing all possible transformations given our dataset of Dactylorhiza species.
© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

4

A. B. SHIPUNOV and R. M. BATEMAN

i

0.10
bb i

fl i P

i

i fc S prp b fc i m i i

i

0.05
PC3 (11.6% of variance)

fc fc fc fc b fc fc fc b fc fc fc fc fc fc fc fc fc fc fc fc fc fc F fc fc b

I

L prt prt R

fc

b

fc

0.00

e prt prt fx T m m m m m m m M m m m i

i

m fc

mm m mm

0.05

m m

0.10
m

0.15
fc

0. 2

0. 1

0 .0

0. 1

0.2

PC1 (48.8% of variance)

Figure 3. Principal components analysis ordination in Dactylorhiza species of all lip shapes generated via relative warps returned from principal warps analysis. b, D. baltica; e, D. euxina; fc and F, D. fuchsii; fl, D. flavescens; fx, hybrids D. fuchsii Ґ D. praetermissa; i and I, D. incarnata; L, D. lapponica; m and M, D. maculata; prp, D. purpurella; prt and P, D. praetermissa; R, D. russowii; S, D. sphagnicola; T, D. traunsteineri.

D. praetermissa were close to the second parent and to the D. traunsteineri `anchor'. The D. sphagnicola `anchor' was located between the D. fuchsii, D. maculata and D. incarnata groups. Lastly, the single D. flavescens sample was peripherally isolated, reflecting its greater phylogenetic disparity (cf. Bateman et al., 2003). Each point on the PCA graph corresponded with specific transformations of base consensus configuration (Fig. 4, outline 3); thus, the differences between species could be represented in terms of the transformations of the coordinate grids ( Fig. 4). The distribution of points revealed from the plot of PC1 and PC3 corresponded well with the variation in some molecular markers, especially ITS alleles (Shipunov et al., 2004) and allowed appreciable grouping on the PCA ordination (Fig. 5). There was also good agreement (Mantel test z-statistics = 75144.31, P < 0.05) between distance matrices, based on the scaled geometric and classic morphometric datasets, respectively. The 3D graph of PCA analysis proved useful for representing the results because the second and third components yielded nearly equal variance and were taxonomically discriminatory. The distinction between species increased, but still showed some overlap (Fig. 6). Each component could be characterized by

specific shape changes, corresponding with given a deviation from the zero point for all axes. Positive deviation in PC1 was characterized by changes in the positions of landmarks 1, 9, 4 and 6 towards the centre of the lip (the lip becoming more rectangular), and landmarks 3 and 7 laterally (the lateral lobes becoming more prominent). These transformations corresponded with typical descriptions of D. fuchsii labella. The positive deviation in PC2 refl ected narrowing and lengthening of the lip whic h corresponded well to the typical D. incarnata lip shape. PC3 represented decreasing size of the lateral lobes, a feature peculiar to D. incarnata and D. baltica.

THE `NORTHERN

TETRAPLOIDS'

Dactylorhiza plants from the European Russian Arctic are tetraploids that often have morphologies intermediate between the two spotted-orc hids, D. maculata and D. fuchsii; moreover, molecular markers support the putative hybrid origin of suc h plants (Shipunov et al., 2004). We therefore analysed a subset of individuals from several populations of D. incarnata, D. fuchsii and D. maculata from Central Russia (Tver ў province) and from northern Karelia (near the Arctic Circle).

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

DACTYLORHIZA DIVERSITY IN RUSSIA

5

1

2

3

4

5

6

7

Figure 4. Cartesian transformation grids illustrating the average landmark configuration for the ordinated Dactylorhiza species: 1, D. baltica; 2, D. fuchsii; 3, consensus configuration; 4, D. incarnata; 5, D. maculata; 6, D. baltica from the second subset analysis; 7, D. fuchsii from the second subset analysis.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

6

A. B. SHIPUNOV and R. M. BATEMAN

100

0.10
0 13

0 100 0 6 72

100

100

0.05 PC3 (11.6% of variance)

0 0 50 0 0 60 0 54 9 3 0 13 2 0 0 0 0 5 10 0

7

1

41 0 100 0

0.00

100 100

100

4

0 11 14 00 22 0 0 2

0.05

0

0 0 0

0.10

0.15
4

0. 2

0. 1

0 .0 0. 1 PC1 (48.8% of variance)

0.2

Figure 5. Principal components analysis ordination in Dactylorhiza species of all lip shapes, highlighting the percentage of the D. incarnata ITS allele observed in each sampled individual.

Both PCA and MDS gave similar results (Fig. 7). Northern plants definitely had an intermediate position between the diploid D. fuchsii and tetraploid D. maculata, far from the D. traunsteineri `anchor'. Three of nine samples of northern tetraploid spottedorchids, each from a different population, and the D. sphagnicola `anchor' also fell close to D. incarnata. Interestingly, the northern specimens often had a lip outline that was close to the overall consensus confi guration for all sampled taxa ( Fig. 4).

DACTYLORHIZA

BALTICA

Dactylorhiza baltica (= D. longifolia) represents the allotetraploid group of Dactylorhiza, whose members are believed to have originated from D. incarnata-like and D. fuchsii-like diploid parents (Heslop-Harrison, 1954, 1968; HedrИn, Fay & Chase, 2001). Most members of this group are apparently stable species , but some (e.g. D. sphagnicola and probably D. baltica: HedrИn, 2003) have features that suggest very recent origins. Our second subanalysis was therefore more focused, considering D. baltica, D. incarnata and D. fuchsii populations from western European Russia (Pskov and St. Petersburg regions). The PCA and MDS analyses revealed similar structures (Fig. 8). Most D. baltica specimens were located between D. incarnata and D. fuchsii and far from the

D. praetermissa `anchor', whereas the D. purpurella sample was closer to D. baltica. Most D. baltica specimens were located close to D. incarnata samples from the same region. Lip transformations are summarized as a 3D graph of the three fi rst components in Figure 9. The D. baltica samples formed a distinct group when projected onto the PC1/PC3 plane . The underlying positive deviation of PC3 transformations were lengthening and narrowing of the lip (landmarks 2 and 8 moving inward, and landmark 5 outward), characteristic of D. incarnata. The D. baltica `condition' was achieved by shortening and widening the lip (negative deviation in PC3), and at the same time increasing the relative size of the lateral lobes (positive deviation in PC1) without signifi cantly narrowing them (negative deviation in PC2).

DISCUSSION
IMPLICATIONS
FOR THE SYSTEMATICS AND EVOLUTION OF

DACTYLORHIZA

The results from classic morphometrics, geometric morphometrics and molecular markers indicates that all three techniques are in broad accordance and hence could be used jointly to further explore the systematics and evolution of Dactylorhiza. Geometric morphometry alone could not reveal signifi cant inter-

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

DACTYLORHIZA DIVERSITY IN RUSSIA

7

9

1

8 6 4 3 5

2

0.20 0.15 0.10 0.05 PC3 0.00 0.05 0.10 0.15 0.20 0.4

7

9

1

baltica fuchsii incarnata maculata praetermissa

8 6 7 5 4 3

2

0.2 0.1 0.0 0.1 PC2 0.2 0.3 0.3 0.2 0.1 PC1
8 6 7 5 4 3 2

0.0
9

0.1
1

0.2

Figure 6. 3D graph of principal components analysis ordination in Dactylorhiza species for all lip shapes; the three insets show the directions of shape change for positive deviations of each of the first three components.

species gaps, so this method needs support from other approaches when employed at the infrageneric level. Most of the outlines in the PCA ordination for all populations belonged to northern dactylorc hids that are believed to represent those lineages with on a verage the highest levels of introgression in their populations. The wide dispersion of D. incarnata samples could reflect the existing diversity of fl ower shapes. However, it could also be in part a methodological artefact, specifically a consequence of several diffi culties in choosing the location for landmarks 3, 4, 6 and 7 on their lips due to the shallowness (and in some smaller-lipped individuals the complete absence) of the sinuses separating the median and lateral lobes . The location of D. flavescens (taxonomically close to the better-known D. romana) supports the taxonomic position of this species, which was phylogenetically distant from all other species inc luded in this study. The D. euxina sample was placed among the allotetraploids, suggesting a closer similarity than previously believed. The positions of most anchors corresp-

onded well with known relationships among species (Bateman et al., 2003; HedrИn, 2003; Shipunov et al., 2004; Pillon et al., in press). `Northern tetraploids' occupied a position intermediate between typical D. maculata and D. fuchsii, offering additional support to the inferred hybrid origin of these plants. Some also resembled D. incarnata, perhaps showing evidence of further introgression, or perhaps indicating that in these cases D. incarnata was, atypically for allotetraploid dactylorc hids, the maternal parent of the hybrid. There is growing evidence that in orchids the maternal parent reliably contributes considerably more than does the paternal parent to the morphology of the progeny ( Bateman & Hollingsworth, 2004). The close relationship of D. praetermissa and D. baltica advocated by some observers (e.g. Delforge, 1995) received no support from our analysis. The samples from four populations of D. praetermissa were located between D. incarnata and D. maculata, so the lip morphology of D. praetermissa (typically rather

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

8

A. B. SHIPUNOV and R. M. BATEMAN
0.2

0.1

MAC n TRA n n n IN C SP H n n n FC H n

Dim 2

0.0

n

0.1

macu lata incarnata fuchsii

0.2 0. 1 0.0 Dim 1 0. 1 0 .2

Figure 7. Multidimensional scaling ordination in Dactylorhiza species for lip shapes from the subset of northern plants. FCH, D. fuchsii; INC, D. incarnata; MAC, D. maculata; n, `northern tetraploids'; SPH, D. sphagnicola; TRA, D. traunsteineri.

rounded and only shallowly three-lobed) w as closer to that of D. maculata than to that of D. fuchsii. This interpretation also receives support from some molecular markers; for example, some D. praetermissa plants bear the D. maculata ITS allele (Pillon et al., in press). The similarities observed between D. purpurella and D. baltica labella accord with the opinions of SoС (1980) and Averyanov (1990) that there is a relatively close relationship between these two species. Local resemblances between D. baltica and D. incarnata could support the hypothesis of a recent and local origin of D. baltica, but they could also be evidence of continuing introgression from D. incarnata. The separation of D. baltica from its supposed parental species by PC1 and PC3 refl ects the specific features of its lip, which is relatively short and wide, but has only shallow notches separating the middle and lateral lobes (Fig. 4). The ellipses of variation surrounding the mean landmark positions (Fig. 2) demonstrate that variation in dactylorchid lip shape is not merely allometric . As overall lip size increases, lip width increases on average proportionately more than does lip length (Bateman & Denholm, 1983: fig. 1), yet the majority of the ellipses of variation evident in Figure 2 were elongated longitudinally. In contrast, landmarks 3 and 7,

located at the apices of the lateral lobes, showed transversely elongated ellipses, suggesting that transformations between lip shapes are more complex than simple uniform expansion analogous to that evident within a single dactylorchid inflorescence during its ontogeny (cf. Alberch et al., 1979).

BROADER

IMPLICATIONS OF GEOMETRIC

MORPHOMETRIC TECHNIQUES

The transformations among populations represented by distant points on the PCA axes show the main directions of change from one species-specifi c form to another. Thus, the transition between different shapes could be described readily in terms of landmark shifts , as has been done above for D. baltica. These shifts both visualize and summarize taxonomically valuable features, especially in the cases of suspected hybridization. There exists a wide range of possible applications of geometric morphometry, from relatively simple two-dimensional organs suc h as undivided leaves to complex anatomical features suc h as transverse sections of petioles ( Volkova, Jufrjakov & Shipunov, in press). The labella configurations for different species revealed by the PCA ordinations coincide well with tra-

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

DACTYLORHIZA DIVERSITY IN RUSSIA

9

0.2

ip

0.1
ip bp PRT

bp bp prp

fs fs fs bp bp bs bs is bs bp bs fs fs bp FCH fs

Dim 2

0.0
is

INC is is

0.1
is is

is

0.2

0. 3

0. 2

0. 1 Dim 1

0 .0

0. 1

0 .2

Figure 8. Multidimensional scaling ordination in Dactylorhiza species for lip shapes from the subset of D. baltica and its potential parents. bs, D. baltica from Saint Petersburg region; bp, D. baltica from Pskov region; fs, D. fuchsii from Saint Petersburg region; is, D. incarnata from Saint Petersburg region; ip, D. incarnata from Pskov region; prp, D. purpurella; FCH, D. fuchsii `anchor'; INC, D. incarnata `anchor'; PRT, D. praetermissa `anchor'.

ditional descriptions of fl ower forms used in diagnostic keys and morphometric works (e.g. Stace, 1997). Such ordinations can be viewed as a `morphological space' wherein many different fl ower shapes could, in theory, be realized. The most densely occupied regions of this graph might be the forms that are most highly selected (most probably via pollinator preference), whereas empty regions might represent either forms eliminated by selection or morphologically impossible forms precluded by developmental constraints. Both of these possibilities merit further discussion and laboratory-based experimental researc h. There is strong evidence that many aspects of fl oral morphology in orchids are under strong selection pressure and/ or prone to drift, and so show relatively high levels of homoplasy in phylogenetic analyses ( Chase, 1999; Bateman et al., 2003; Tremblay et al., 2005). The species of Dactylorhiza discussed here each attract a wide range of pollinating insects by deception through mimicry of other (mostly non-orc hidaceous) species of fl owering plants, so there is little doubt that the combination of spur size and lip size , shape and markings is crucial to reproductive success in the group . This raises the likelihood of frequent convergences upon similar optimal labellum designs of different

lineages within Dactylorhiza. The recent development of a strong and detailed molecular phylogenetic framework for both the diploids and the tetraploids (Bateman et al., 2003; Pillon et al., in press) allows the identification of sister species that could provide a valuable model system for exploring, via morphometric techniques, the growth patterns that permit such convergence. This work could then be followed by evolutionarydevelopmental genetic studies (e .g. Cronk, Bateman & Hawkins, 2002) to elucidate the underlying genetic processes. Equally, these flowers could also constitute a useful model system for defining the nature and causes of developmental constraints. Continuing our theme of using landmark data to explore labellum shape , it would be interesting to use the labella as a rare example of a botanical study of fl uctuating asymmetry (cf. Rudall et al., 2002). If landmarks could be established with sufficient precision (say, within a confidence interval of 0.5%), the strongly bilaterally symmetrical orchid labellum could easily be divided vertically into two portions to explore the relative degree of somatic expression of identical genomes in the aggregates of cells on either side of the mirror plane separating the right and left halves.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

10

A. B. SHIPUNOV and R. M. BATEMAN

9

1

8 6 4 3 5 7

2

0.20

0.15

9

1

0.10 PC3

incarnata baltica fuchsii

8 6 7 5 4 3

2

0.05

0.00

0.05

0.15 0.10 0.05 0.00 0.05 PC2 0.10 0.15 0.3 0.2 0.1 0.0 PC1
8 6 7 5 4 3 2

0.10 0.4

0.1

0.2
9

0.3
1

0.4

Figure 9. 3D graph of principal components analysis ordination in Dactylorhiza species for lip shapes from the subset including D. baltica; the three insets show the directions of shape c hange for positive deviations of each of the first three components.

It should be possible to develop additional landmark-related techniques for monitoring the ontogeny of the discrete anthocyanin markings of the labellum, and thereby relating precise attractant patterns to degree of pollinator success in these orc hids. It should also prove feasible to explore the ontogeny of these markings, which in D. fuchsii and D. maculata have been said to develop outward from the vicinity of the spur entrance in parallel with development of the venation in the labellum, analogous to the ontogeny of butterfly wings (Heslop-Harrison, 1948, 1951). Although determining homologous landmark points for these complex markings would undoubtedly be challenging, the insights gained into fl oral evolution could justify the required effort.

Abramova, Zh. Altshuler, P. Buntman, V. Chernovol, G. Konechnaja, E. Kost, I. Kucherov, S. Nazarova, K. Markvicheva, A. Pegova, E. Peskova, N. Saitanova, D. Suhova, T. Vinogradova and P. Volkova. We also thank D. Kirkup and P. Rudall for valuable comments, and I. Pavlinov for our first acquaintance with geometric morphometrics techniques. This research was supported by the Royal Society of London, and sponsored by NATO and the British Foreign and Commonwealth Office.

REFERENCES
Adams DC, Rohlf FJ, Slice DE. 2002. Geometric morphometrics: ten years of progress following the `revolution'. Ecology and Evolution, SUNY, Stony Brook. Alberch P, Gould SJ, Oster GF, Wake DB. 1979. Size and shape in ontogeny and phylogeny. Paleobiology 5: 296317. Averyanov LV. 1990. A review of genus Dactylorhiza Neck. ex

ACKNOWLEDGEMENTS
We are grateful to all the colleagues and friends who helped us by collecting Dactylorhiza plant material: L.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

DACTYLORHIZA DIVERSITY IN RUSSIA
Nevski (Orchidaceae), 3 [in Russian]. Novosti Sistematiki Vysshikh Rastenij 27: 3262. Bateman RM. 2001. Evolution and classification of European orchids: insights from molecular and morphological characters. Journal EuropДischer Orchideen 33: 33119. Bateman RM, Denholm I. 1983. A reappraisal of the British and Irish dactylorchids, 1. The tetraploid marsh-orchids. Watsonia 14: 347376. Bateman RM, Denholm I. 1985. A reappraisal of the British and Irish dactylorchids, 2. The diploid marsh-orchids. Watsonia 15: 321355. Bateman RM, Denholm I. 1989. Morphometric procedure, taxonomic objectivity and marsh-orchid systematics. Watsonia 17: 449455. Bateman RM, Denholm I. 2003. The Heath Spotted-orchid (Dactylorhiza maculata (L.) SoС) in the British Isles: a cautionary case-study in delimiting infraspecific taxa and inferring their evolutionary relationships. Journal EuropДischer Orchideen 35: 336. Bateman RM, Hollingsworth PM. 2004. Morphological and molecular investigation of the parentage and maternity of Anacamptis Ґ albuferensis (A. fragrans Ґ A. robusta), a new hybrid orchid from Mallorca, Spain. Taxon 53: 4354. Bateman RM, Hollingsworth PM, Preston J, Luo Y-B, Pridgeon AM, Chase MW. 2003. Molecular phylogenetics and evolution of Orchidinae and selected Habenariinae (Orchidaceae). Botanical Journal of the Linnean Society 142: 140. Baylac M, Villemant C, Simbolotti G. 2003. Combining geometric morphometrics with pattern recognition for the investigation of species complexes. Biological Journal of the Linnean Society 80: 8998. Bookstein FL. 1991. Morphometric tools for landmark data. New York: Cambridge University Press. Chase MW. 1999. Molecular systematics, parsimony, and orchid classification. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN, eds. Genera Orchidacearum. 1. General introduction, Apostasioideae, Cypripedioideae. Oxford: Oxford University Press, 8188. Cronk QCB, Bateman RM, Hawkins JA, eds. 2002. Developmental genetics and plant evolution. London: Taylor & Francis. Delforge P. 1995. Orchids of Britain and Europe. London: Harper-Collins. Guill JM, Heins DC, Hood CG. 2003. The effect of phylogeny on interspecific body shape variation in darters (Pisces: Percidae). Systematic Biology 52: 488500. HedrИn M. 2002. Speciation patterns in the Dactylorhiza incarnata/maculata polyploid complex (Orchidaceae): evidence from molecular markers. Journal EuropДischer Orchideen 34: 707731. HedrИn M. 2003. Plastid DNA variation in the Dactylorhiza incarnata/maculata polyploid complex and the origin of allotetraploid D. sphagnicola (Orchidaceae). Molecular Ecology 12: 26692680. HedrИn M, Fay MF, Chase MW. 2001. Amplified fragment length polymorphisms (AFLP) reveal details of polyploid

11

evolution in Dactylorhiza (Orchidaceae). American Journal of Botany 88: 18681880. Heslop-Harrison J. 1948. Field studies in Orchis L., I. The structure of dactylorchid populations on certain islands in the Inner and Outer Hebrides. Transactions and Proceedings of the Botanical Society of Edinburgh 35: 2666. Heslop-Harrison J. 1951. A comparison of some Swedish and British forms of Orchis maculata L. sens lat. Svensk Botanisker Tidsskrift 45: 608635. Heslop-Harrison J. 1954. A synopsis of the dactylorchids of the British Isles. Beroff Geobotanische Forschung Institute RЭbel 1953: 5372. Heslop-Harrison J. 1968. Genetic system and ecological habit as factors in dactylorchid variation. Jahresberichte Naturwissenschaftlichen Vereins in Wuppertal 2122: 2027. Jensen RJ, Ciofani KM, Miramontes LC. 2002. Lines, outlines and landmarks: morphometric analyses of leaves of Acer rubrum, Acer saccharinum (Aceraceae) and their hybrid. Taxon 51: 475492. Kores PJ, Molvray M, Darwin SP. 1993. Morphometric variation in three species of Cyrtostylis (Orchidaceae). Systematic Botany 18: 274282. MacLeod N, Forey P, eds. 2002. Morphology, shape and phylogenetics. Systematics Association Special Vol. 64. London: Taylor & Francis. Nelson E. 1976. Monographie und Ikonographie der Orchidaceen-gattung, III. Dactylorhiza. ZЭrich: Published by the author. Pavlinov IY. 2000. Geometric morphometrics of skull shape in some muroid rodents (Mammalia: Rodentia): relation of the skull shape to trophic specialization [in Russian, with English summary]. Journal of General Biology 61: 583600. Pavlinov IY. 2001. Geometric morphometrics, a new analytical approach to comparison of digitized images. Information Technology in Biodiversity Research: Abstracts of the 2nd International Symposium. St. Petersburg: Russian Academy of Science, 4464. Pedersen HA. 1998. Species concept and guidelines for infraspecific taxonomic ranking in Dactylorhiza (Orchidaceae). Nordic Journal of Botany 18: 289310. Pillon Y, Fay MF, HedrИn M, Bateman RM, Devey D, van der Bank M, Chase MW. in press. Evolution and biogeography of European species complexes in Dactylorhiza (Orchidaceae). Taxon in press. Ray TS. 1992. Landmark eigenshape analysis: homologous contours: leaf shape in Syngonium (Araceae). American Journal of Botany 79: 6976. Reinhard HR. 1990. Kritische anmerkungen zu einigen Dactylorhiza-arten (Orchidaceae) Europas. Mitteilungsblatt Arbeitskreise Heimische Orchideen Baden-WЭrttemberg 22: 172. Rohlf FJ. 2003a. tpsDig v. 1.39. Ecology and Evolution, SUNY, Stony Brook. Rohlf FJ. 2003b. tpsRelw v. 1.35. Ecology and Evolution, SUNY, Stony Brook. Rohlf FJ, Loy A, Corti M. 1996. Morphometric analysis of Old World Talpidae (Mammalia, Insectivora) using partialwarp scores. Systematic Biology 45: 344362.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112

12

A. B. SHIPUNOV and R. M. BATEMAN
Stace CA. 1997. New flora of the British Isles, 2nd edn. Cambridge: Cambridge University Press. Thompson DW. 1917. On growth and form. Cambridge: Cambridge University Press. Tremblay RL, Ackerman JD, Zimmerman JK, Calvo RN. 2005. Variation in sexual reproduction in orchids and its evolutionary consequences: a spasmodic journey to diversification. Biological Journal of the Linnean Society 84: 154. Tyteca D, Gathoye JL. 2000. Morphometric analysis of the Dactylorhiza majalis group in France and western Europe, with a description of Dactylorhiza parvimajalis Tyteca et Gathoye, spec. nov. Journal EuropДischer Orchideen 32: 471511. Venables WN, Smith DN, the R Development Core Team. 2002. An Introduction to R. Bristol: Network Theory Ltd. Volkova PA, Jufrjakov IS, Shipunov AB. in press. The variability of different Drosera (Droseraceae) species revealed by classic and geometric morphometry. Bulletin of the Moscow Society of Naturalists, Biological Series in press.

Rohlf FJ, Slice DE. 1990. Extensions of the procrustes method for the optimal superimposition of landmarks. Systematic Zoology 39: 4059. Rudall PJ, Bateman RM, Fay MF, Eastman A. 2002. Floral anatomy and systematics of Alliaceae with particular reference to Gilliesia, a presumed insect mimic with strongly zygomorphic flowers. American Journal of Botany 89: 1867 1883. Rueber L, Adams DC. 2001. Evolutionary convergence of body shape and trophic morphology in cichlids from Lake Tanganyika. Journal of Evolutionary Biology 14: 325 332. Shipunov AB, Fay MF, Pillon Y, Bateman RM, Chase MW. 2004. Dactylorhiza (Orchidaceae) in European Russia: combined molecular and morphological analysis. American Journal of Botany 91: 14191426. SoС R. 1980. Dactylorhiza Necker ex Nevski. In: Tutin TG, Heywood VH, Burges NA, Moore DM, Valentine DH, Walters SM, Webb DA, eds. Flora Europaea, Vol. 5. Cambridge: Cambridge University Press, 333337.

© 2005 The Linnean Society of London, Biological Journal of the Linnean Society, 2005, 85, 112