The tangled history of the Tradescantia Andersoniana Group
There are several closely-related Tradescantia species which are native to temperate North America. They form mounds of long grassy leaves, topped with small blue or pink flowers. The number and classification of these species has changed a lot over time, as botanists studied wild plants and learned about their differences and similarities.
At the same time, they have become popular in cultivation – in their native North America, as well as introduced to Europe. They are frost-hardy and grow as deciduous herbaceous perennials, perfect for temperate-climate gardens. These cultivars have been developed into many shapes and sizes, with all kinds of different flower styles and colours.
This group of plants – like so many tradescantias – have a history of confused identity. In this article I’ll cover their history over the last hundred years, as the name went from Tradescantia virginiana, to Tradescantia × andersoniana, and finally to the Tradescantia Andersoniana Group.
The most common and well-known temperate species is Tradescantia virginiana. Historically, almost all garden cultivars of Tradescantia were labelled as this species. Gardeners and nurseries don’t generally care much about being precise with species identities. T. virginiana is common and widespread, and at a glance it was the obvious choice for the wild ancestor of these garden plants.
Then early in the 20th century, botanists Anderson and Woodson (1935) wrote a book about the native temperate Tradescantia species. Their focus was on the plants as they occurred in the wild, but they made a passing mention of garden cultivars that threw their identity into question:
It is a notable fact that the senior author has never seen a Tradescantia virginiana in a European garden which could unhesitatingly be referred to that species. Most of the plants in cultivation under that name show unmistakable evidence of mixture with T. canaliculata [T. ohiensis] and T. subaspera.Andeson & Woodson, 1935, p. 30
There was not a single garden cultivar that actually belonged to the species T. virginiana. Instead the authors believed that every single garden cultivar had hybrid ancestry, and also included the related species T. ohiensis (known as T. canaliculata at the time) and T. subaspera.
About twenty years later in another book, Anderson (1952, Chapter 2) wrote in more detail about how they had come to that conclusion. During their initial research, they had collected samples of wild plants to grow them in an experimental garden. And they found that when cultivated like this, the species T. virginiana, T. ohiensis, and T. subaspera were all incredibly prone to hybridisation. These hybrid seedlings were more vigorous and attractive than their wild parents, so they would certainly be selected and maintained by gardeners – who might have no idea the new plants were not pure T. virginiana.
Anderson despaired that, although his study of tradescantias had led to more accurate identification of wild species, the garden cultivars were still universally mislabelled as T. virginiana. Gardeners had no other option, because there was no correct name they could use to refer to these hybrids.
Tradescantia × andersoniana
Two years after Anderson’s book, botanists Ludwig and Rohweder (1954) tried to rectify the problem. They published the botanical hybrid name Tradescantia × andersoniana, in honour of Anderson himself. They referred to Anderson’s conclusions that cultivated plants are hybrids rather than pure T. virginiana, and decided that a single name would make it much easier to refer to those hybrids.
This would seem to be the end of the story, and over the next several decades gardeners gradually started using T. × andersoniana instead of T. virginiana to label garden cultivars.
But in fact, it wasn’t the end of the story. In a book on the taxonomy of cultivated plants, Walters et al. (1989, p. 31) pointed out that the name T. × andersoniana had never been described and was therefore invalid.
The rules for creating scientific plant names – like naming species and species hybrids – are very strict. Much stricter than the rules for naming cultivars! One of those rules is that, to name a new species or hybrid, there must be a precise botanical description of the plant in question. Ludiwg and Rohweder never described their hybrid, and none of the sources they referred to contained a description either. So T. × andersoniana is invalid, and can’t be used as a botanical name.
Tradescantia Andersoniana Group
For decades botanists had been pointing out that garden Tradescantia cultivars were labelled wrong, but never providing a valid alternative. Finally the Royal Horticultural Society (RHS) found a true horticultural solution – a cultivar group.
A cultivar is a type of plant that has been created or selected through human activity. Generally a cultivar belongs to a botanical species – for example, Tradescantia cerinthoides ‘Nanouk’ originates from the wild species Tradescantia cerinthoides. The name ‘Nanouk’ refers to a specific clone which people grow for its unique characteristics.
Sometimes, a cultivar belongs to a group of plants which doesn’t fit a particular botanical classification. This could be a subset of a single wild species which all share a particular trait, like a certain flower colour. Or it could be a collection of hybrids of known or unknown origin, which have certain characteristcs in common.
A cultivar group is the ideal was to classify these hardy garden Tradescantia cultivars. We don’t know for sure how they originated or what exactly their ancestors are. We do know that they don’t belong to any single particular species. We could label them all as just “Tradescantia” and be done with – but these plants clearly make up a defined group which is very different from, say, Tradescantia cerinthoides ‘Nanouk’! So instead, we place them in a cultivar group, which functions like a species, but for cultivated rather than wild plants.
According to the cultivar naming rules, we can repurpose the invalid name “andersoniana” for the newly-defined cultivar group (Brickell et al., 2016, Art. 21.6). So it becomes the Tradescantia Andersoniana Group. When referring to a specific cultivar, we can include the group name in brackets in place of a species – for example, Tradescantia (Andersoniana Group) ‘Osprey’.
The RHS started listing garden Tradescantia hybrids with the name Andersoniana Group in their 2000 Plant Finder (Royal Horticultural Society, 2000, p. 717). I haven’t found a hardcopy publication which officially establishes the group name with a description, so it will be established as part of the ICRA checklist (Brickell et al., 2016, Art. 27.1).
Although the group name has been in use for over twenty years, it still hasn’t been universally accepted. Part of the reason for that is that many sources didn’t recognise that the older T. × andersoniana name was invalid. Up until a few weeks ago, Plants of the World Online (POWO – a taxonomic database run by the world-leading Royal Botanic Gardens, Kew) listed the name as valid.
This threw me off for some time, because POWO is a very reliable source that I tend to defer to. But the RHS is another reliable source, and the two were in conflict! I did a lot of research to track down the original source of the T. × andersoniana name as well as the references from that publication, and check the botanical naming rules, to try and figure out what was going on.
Once I had come to the conclusion that the RHS interpretation was correct and the name must be invalid, I gathered my evidence and contacted the team behind POWO. And they responded very quickly, confirmed my findings and said they’d update the name. It’s now listed as “unplaced”, which means invalid. I’m proud to have made a small contribution to the accuracy of global taxonomy!
Hopefully the old name being correctly marked as invalid, along with the eventual published ICRA checklist, will help everyone to finally use the correct name for these plants: the Tradescantia Andersoniana Group.
Anderson, E. & Woodson, R. E. (1935). The species of Tradescantia indigenous to the United States. Contributions from the Arnold Arboretum of Harvard University, 9. doi:10.5962/bhl.title.153160. Biodiversity Heritage Library link.
Anderson, E. (1952). Plants, Man & Life. University of California Press.
Ludwig, W. & Rohweder, O. (1954). Zur Nomenklatur zweier Commelinaceen [On the nomenclature of two Commelinaceae]. Feddes repertorium specierum novarum regni vegetabilis, 56(3), 282. doi:10.1002/fedr.19540560304. Open access link.
Walters, S. M., Brady, A., Brickell, C. D., Cullen, J., Green, P. S., Lewis, J., Matthews, V. A., Webb, D. A., Yeo, P. F. & Alexander, J. C. M. (1989). The European Garden Flora: Volume II. Cambridge University Press.
Royal Horticultural Society. (2000). RHS Plant Finder 2000.
Brickell, C. D., Alexander, C., Cubey, J. J., David, J. C., Hoffman, M. H. A., Leslie, A. C., Malécot, V., Jin, X. (2016). International Code of Nomenclature for Cultivated Plants. PDF link.