The Polyploid Marsh Orchids

After the Spotted Orchids and the Early Marsh Orchids (and excluding the Frog Orchid), the remaining British and Irish Dactylorhiza species are all part of a complex of polyploid plants. Whether these are different species, or a bunch of sub-species and varieties is open to debate. The concept of polyploidy was covered in the Spotted Orchids section. There it was the autopolyploid Heath Spotted Orchid under review. The polyploidy Marsh Orchids are all allopolyploids, meaning that they came about by hybridisation involving diploid zygotes from different species. In most cases the maternal parent is one of the Spotted Orchids and the pollen parent one of the Early Marsh Orchids. This group includes (possibly – may be more or fewer!):

            Southern Marsh Orchid – Dactlyorhiza praetissima

            Northern Marsh Orchid – D. purpurella

            Narrow-leaved Marsh Orchid – D. traunsteineroides

            Hebridean Marsh Orchid – D. ebudensis

            Western Marsh Orchid – D. occidentalis

There are other species in continental Europe and authors in the past have tried to shoe-horn some of the British species into the D. majalis complex, either as related species or as sub-species. A personal opinion is that grouping these on morphological grounds makes not a lot of taxonomic sense, but could be an identification aid. There is plenty of evidence that each of the five species mentioned above arose from different autopolyploid hybridisations. Some experts will argue that different events is all that is needed to allow these to definitely classified as separate species. Other experts counter that with the argument that they are just one species because they have genes inherited from the same ancestral parent species. Personally I prefer the separate species point of view. Imagine a scenario where just one polyploidy hybrid arose and was given species status, before it died out. Then years pass and another similar hybridisation event occurred, giving us a very similar looking plant. Can we be justified in saying that the species has occurred twice? Morphological similarities should not be the only deciding factor. Obviously there will be marked similarities, but taxonomy has to depend upon evolutionary pathways. Looking back over what species and sub-species (and varieties) have been claimed in the past (even the recent past) I find an almost bewildering array of conflicting information. Localised populations that have unique, or at least noticeable, characteristics add to the confusion. Add to that the common names that have been applied and then changed in some cases. I hope I can make some sense, based on recent molecular analyses and geographical distribution, of the polyploidy Dacts based initially on the five listed above. I will only give brief mention to physical characteristics as these are covered quite adequately by modern field guides. This may seem like heresy to some, but there is logic in my reasoning and I hope my text reveals this. The main problem for me is that the biometrics overlap considerably and this used to be the basics for taxonomy. What for one species may be within the upper quartile will be slap bang on the median for another. Leaf spotting may be absent to marked. Within a population of a single site there can be variation. Hybridisation may be responsible for some of this with successive introgression reducing the impact of the outsider’s influence. If that is the case then biometrics can provide false data. So can molecular data, especially when sample numbers are small, but if a ‘typical’ example is chosen then one is closer to the ‘truth’.

Each of these British polyploids have a heritage of a ‘maternal’ Spotted Orchid and a ‘paternal’ Early Marsh Orchid, supplying a zygote with an unreduced chromosome number. The mother was D. fuschsii, except for D. ebudensis where it was D. maculata. The sub-species of D. incarnata involved is unknown though some experts have made reasoned proposals. These parental species will not necessarily be the same populations we see today. They will be similar, but the polyploid events occurred thousands of years in the past and the strains involved may have died out or elvolved in the meantimetime. The questions that must be asked are:

1. If these, and other allopolyploids across Europe. have similar ‘parentage’, how come they look different? The answer lies in one of two theories.
The Post-differentiation theory holds that once polyploidy had occurred the populations undergo some genetic modification and selection for a particular habitat; one that is different to either parent species. Additionally there will be secondary gene flow through hybridisation with either of the parent species. And it should not be forgotten that polyploidy permits increased chances for random mutation leading to adaptive changes.
The Pre-differentiation model states that it is the variation within the parent populations that defines what the polyploid looks like. Some polyploids are quite recent, while others are more ancient. The parent species will have changed over that time.
Of course, both may occur as they are not mutually exclusive.

2. Did autopolyploidy occur just once for each species? Difficult to answer. D. ebudensis almost certainly has occurred just once based on its extreme localisation, and that its habitat and locale is less than 3,000 years old. D. purpurella is older but is still thought to to have arisen once and spread from that point. D. occidentalis is restricted to Ireland so again a single event. For the other two species the situation is less clear. They arose in southern Europe while the ice sheets covered Britain and their origin is further back. But it is likely that they arose from one event.
Does this imply that the creation of a new allotetraploid is an extremely rare event, or is it that most new polyploids do not find a habitat that favours them over their parent species and so die out as an entity?