This post continues a theme that I’ve touched upon before: that those of us who study diatoms are Not Like Other Biologists (see “Diatoms and Dinosaurs”). Inside every diatomist, there is a deep-seated yearning for silica that overrides the love affair with carbon that drives most biologists. Hand a diatomist a sample of living organisms and he (or she) will drop it into a pot of strong oxidising agents and then smile contentedly until the last vestiges of life have fizzed out of the cells.
I have commented before on the problems that this causes us when we are trying to understand their ecology (see “The meaning of … nothing” and “All things bright and beautiful”) and this post extends that theme, looking at problems of mapping the distribution of diatoms. It is based on work done by my intern, Susannah Collings, who has been funded by the British Phycological Society over the summer to look into the prospects for a UK “red list” of rare and threatened diatoms (see “A “red list” of endangered British diatoms?”). We decided to use Achnanthes oblongella, a species that is not particularly rare, is straightforward to identify and which has a fairly well-understood ecology to illustrate some of the challenges.
The database that we used for these analyses contains records from 6308 samples from over 3000 sites throughout the UK (and some sites in the Republic of Ireland too), within which there were 2180 records of Achnanthes oblongella. In many of these, only a few valves of Achnanthes oblongella were recorded, but there were 384 samples where it constituted a third or more of all the diatoms recorded. In sixteen samples it constituted 80 per cent of all diatoms, with two samples having 98% A. oblongella.
Achnanthes oblongella: top left: three examples of the rapheless valve; bottom left: three examples of the raphe valve (photographs: M. Bayer, RBGE; scale bar: 10 micrometres, or 1/100th of a millimetre). Right: histogram showing the relative abundance of A. oblongella in samples.
When these records are plotted onto a map using DMAP software, we see the records clustered in those parts of the country that are associated with soft, often peaty and acidic waters: Cornwall and Devon, North Wales, parts of the Pennines and Lake District, south-west Scotland and the highlands. There are also a number of records in the New Forest and around London, the latter probably reflecting patches of heathland on the Greensand. But, wait a minute, remember that predilection for silica that I mentioned above; how can we be sure that each of these dots on the map represents a living population of Achnanthes oblongella at that location? The analyst never saw a living cell of the species, only parts of a dead cell. In situations where the relative abundance is very low, can we eliminate the possibility that a few dead cells had been washed downstream and ended up in our samples?
The distribution of Achnanthes oblongella in Great Britain: a. all records; b. only records where A. oblongella constitutes > 1% of all diatoms; c. only records where A. oblongella constitutes > 2% of all diatoms.
It may seem surprising that no-one has studied how well samples of cleaned diatoms represent living populations but the problems associated with naming diatoms from live material (real or perceived … that’s a subject for another day) mean that the literature on this topic is sparse. We compared the map produced with all data with maps produced if we just used records with more than one or two per cent of Achnanthes oblongella and, in this particular case, the broad patterns are the same, even though we are trimming away almost half of the data. I suspect that this is partly because A. oblongella is quite straightforward to identify, so the number of spurious records is relatively low. The distinctive ecology means that even where low abundances are due to washed-in cells, there is an inoculum a short distance upstream. The dots, therefore, may indicate proximity to viable populations, rather than direct evidence of their presence in situations where the percentage in the sample is low.
All this points up the great irony of compiling a red list for diatoms: red lists provide an inventory of the conservation status of species. It is supposed to catalyse action for biodiversity conservation. Yet to gather the data we need, we first have to kill the organisms that we are trying to protect, most of which have never been observed in the live state at all. I did a quick search amongst my own books and via Google and could not find a single image of a living cell of Achannthes oblongella. The silica shell is, I am ashamed to say, the only reality that many of my fellow diatomists recognise.