A genuinely rare diatom genus?

About ten months ago I speculated on the practicalities of producing a “red list” of British diatoms (see “A red list of endangered British diatoms”).   A couple of weeks ago, I reported on progress towards this goal (see “Why do you look for the living amongst the dead?”) and, today, I can show the first evidence for a genuinely rare freshwater diatom in Great Britain.   Great Britain, alas, rather than the United Kingdom, as we struggled to get the Irish grid references to plot on our graphs, but it is a step in the right direction.

In my post last year, I commented that one of the problems we faced was the taxonomic uncertainty, particularly as there have been so many new species described, often as a result of “splitting” taxa that older books regarded as a single entity.   One of the criteria my intern, Susannah, and I used was a stable taxonomic concept, in order to ensure that older records could be merged with our more recent records.   This does not have to be at the level of species and, indeed, the first subject of our scrutiny was a genus, Tetracyclus. It fits our bill perfectly as it is quite recognizable, despite not being very common, so we can be fairly sure that the chance of an identification error is low.


The left hand map shows all hectads (10 km2) squares from which we have at least one freshwater diatom sample in Great Britain; the right hand map shows those hectads with records of Tetracyclus spp.

As you can see from the map, there are a cluster of locations in the Scottish highlands, three in the Cheviots in Northumberland and a few in Snowdonia. Then there is a record in Pembrokeshire and another in Wiltshire.   That is 23 hectads throughout the UK for all three species, none of which, individually, is found in more than 15 hectads, and therefore each qualifies as “nationally rare” according to JNCC guidelines.   There may be a few more records out there but, at the same time, there is also a risk of false negatives – a single valve, remember, does not necessarily mean that a viable population of the organism is present at the site.   The Wiltshire record worries me, and I need to check that there has not been a transcription error as the record was transferred from notebook to database, and to ensure that the grid reference was recorded correctly.

The literature offers little help when it comes to understanding the habitat of Tetracyclus: West and Fritsch tell us that Tetracyclus lacustris “… prefers hilly districts and is often found in the plankton of mountain lakes” whilst T. rupestris “… occurs on dripping rocks in mountain areas.” Other writers also hint that it might be partly sub-aerial in distribution, which may explain why the records in our dataset only ever record it in small quantities.   Is it possible that our samples, which are mostly from submerged rocks in streams and lake littoral zones, are just picking up a few stray cells that have been washed away from their preferred habitat?   We can interpret each dot, perhaps, as indicating that the species was “present in the vicinity”.

We should also point out that one of the Snowdonian locations on the map is Llyn Perfeddau, the lake from which the first Tetracyclus lacustris specimens were described, back in 1843 by John Ralfs.   Allan Pentecost re-visited Llyn Perfeddau in 2014 but was unable to find it in his samples, which adds to the mystique that this genus and species exert.

So are these three Tetracyclus species the first bona fide candidates for a UK (or GB) red list of diatoms?   Each fulfils the criterion of “nationally rare”, being found in less than 15 hectads, albeit with the provisos set out above. However, rarity alone is not sufficient to place a species on a red list; we also need to demonstrate that it is vulnerable or endangered.   This implies knowledge of trends over time, not just patterns in space.   The default under such circumstances is to consider whether the locations where it is found are fragmented and are, themselves, threatened or vulnerable.   The restricted distribution in low nutrient waters in mountainous and northern areas does suggest that changes in land management or climate change could affect the small isolated populations that we do have, so a designation of “vulnerable” is probably appropriate.   Though I doubt that WWF will be replacing their panda logo with a diatom any time soon.


Pentecost A. (2014). In search of the Welsh Tetracyclus. The Phycologist 88: 42-43.



The Really Rare Diatom Show

Having set out the limitations of my exercise to define nationally-scarce or rare diatoms, I have drastically reduced my list of candidates from 377 species down to eight.   I suspect that gathering some more data (see point 1 in my previous post) will mean that I can reinstate a few more species to the list, but that will have to wait for another day.

Six of the ten species on my list belong the genus Gomphonema or near relatives.   One of these is Didymosphenia geminata (see “A journey to the headwaters of the River Coquet”); of the others, the most intriguing is Gomphonema tergestinum, a species that occurred in 81 hectads but which seems to be particularly common in north-west England and south-west Scotland, for reasons that I do not fully understand. This needs further investigation but it could be another species that has a distinct biogeography that is not explained solely in terms of a particular chemical environment.   All six of these species qualify as “nationally scarce” rather than “rare” and do remember that my analyses are, at this stage, very preliminary.


Nationally scarce Gomphonema species? a. Gomphonema clavatum; b. Gomphonema insigne; c. Gomphonema ventricosum; d. Gomphosphenia (Gomphonema) grovei; e. Gomphonema turgestinum. Scale bars: 10 micrometres (1/100th of a millimetre). All images from http://craticula.ncl.ac.uk/EADiatomKey/html/Craticula.html or http://rbg-web2.rbge.org.uk/ADIAC/db/adiacdb.htm.

One species that may qualify as genuinely rare is Tetracyclus emarginatus, for which we have just two records.   The genus itself is rare, and known mostly from the fossil record but is also sufficiently distinctive that it would not be misidentified or overlooked by analysts.   Another representative of the genus, T. rupestris, has been recorded from Britain but does not feature at all in our database.   When it has been recorded, it is from rock surfaces and damp mosses rather than submerged in streams, so it could have been overlooked in my preliminary analysis. A third representative of the genus, T. lacustris is, as the name suggests, likely to be under-represented in a database composed of samples from rivers, so I will reserve judgement on the rarity of this species though I suspect that it is another candidate for the red list.


Tetracyclus emarginatus. Scale bar: 10 micrometres (1/100th of a millimetre). Image from http://craticula.ncl.ac.uk/EADiatomKey/html/Craticula.html (photographer: Bernie ní Chatháin).

Another candidate is Cymbellonitzschia diluviana. Though I have tried not to comment on the distribution of species found in lakes, I will make an exception for this species because the habitat is quite well understood, thanks to the work of David Jewson and colleagues at the University of Ulster.   They found it to be most abundant on sand grains exposed to wave action in the littoral zone of Lough Neagh and a few other loughs and lochs with hard water and high pH.   As this combination of characteristics is relatively rare in the UK, it is reasonable to assume that C. diluviana will also be very limited in its distribution.


Cymbellonitzschia diluviana (left: valve view; right: girdle view of two recently-divided cells. Scale bar: 10 micrometres (1/100th of a millimetre). Image from http://craticula.ncl.ac.uk/EADiatomKey/html/Craticula.html (photographer: David Mann).

Finally, Chris Carter has suggested Entomoneis ornata as a candidate for the diatom “red list”, pointing out that it has not only rare in this country, but is also already on the Red List of Plants of Germany and is also considered to be rare in The Netherlands. Cells of Entomoneis are characteristically twisted around the apical axis, which tests even Chris’ photographic skills, and the genus is more common in brackish and marine waters than in freshwaters. However, it is certainly a species that should be on our preliminary list, and deserves further investigation.

In the next post I’ll consider the pros and cons of a “red list” of British diatoms.


Entomoneis ornata from the Oxford Canal, England, photographed by Chris Carter. Scale bars: 10 micrometres (1/100th of a millimetre)


Carter, C.F. & Belcher, H. (2010). A UK record of Entomoneis ornata (J.W. Bailey) Reimer in Patrick & Reimer 1975. Diatom Research 25: 217-222.

Jewson, D.H. & Lowry, S. (1993). Cymbellonitzschia diluviana Hustedt (Bacillariophyeae): habitat and auxosporulation.   Hydrobiologia 269/270: 87-96.

Ludwig G., Schnittler M. (1996) Rote Liste Gefahrdeter Pflanzen Deutschlands. (12 volumes but available as the list only from www.bfn.de/fileadmin/MDB/documents/RoteListePflanzen.pdf)

A “red list” of endangered British diatoms?

I have had two conversations about rare algae over the past two weeks. The first was an invitation to contribute to an exercise to develop a list of diatoms that might form the basis of a “red list” of endangered algae. The second was a retort from a colleague that such an act would be meaningless as algae don’t have the same biogeographical restrictions on their distributions as higher organisms, and that all algae will grow anywhere so long as the environment is suitable.   The argument that algae don’t have biogeographical restrictions is an old one (summarised as “everything is everywhere, the environment selects”) but several recent papers have shown this to be wrong. Some species do appear to be cosmopolitan, as my previous post shows, but others do seem to be restricted to particular regions. Even if the local environment does play a large role in determining the algae that are found at a location, that does not seem to obviate the need for a list of endangered algae. On the contrary, it might even help focus attention on locations where efforts to restore a site might make a real difference.

The problem, in my opinion, is more basic: phycologists working in freshwaters do not have a strong tradition of systematic recording of the distribution of organisms. You only need to look at the Freshwater Algal Flora of the British Isles, and to see how many species are described as “probably cosmopolitan”, to realise the scale of the problem.

Because of their widespread use in ecological assessment, the diatoms are one group of freshwater algae where there may be enough data to start making some sensible judgements about the rarity, or otherwise, of individual species. I had a look at a dataset compiled for a project that I was involved with a few years ago in order to see what might be possible. This dataset comprises 6500 samples from 3305 sites spread across Great Britain and Ireland, most of which also have location information. The basis for conservation assessments is the distribution in 10 km squares, termed “hectads”, of which 1111 were represented in my dataset, out of a total of about 3000 in Britain.   The two criteria I am using are “nationally rare” for species that occur in 15 hectads or fewer, and “nationally scarce”, for those which are only found in 16 to 100 hectads. Using these criteria, I produced a “long list” of 150 diatom taxa that are “nationally scarce” and a further 226 which may be “nationally rare”. This, however, is where the real work starts.

Scanning down this list, I see several problems that need to be addressed before we can make serious judgements about the rarity, or otherwise, of particular taxa. However, I do also see a number of taxa on this list that I do believe to be genuinely rare or scarce and which are, at least, worthy of more study.   The problems are many and will spill into the next post but let’s make a start:

  1. The dataset I’m using is for rivers, and I will need to merge it with some additional datasets to get good coverage of lakes and also of soft water and acid habitats.   I would not trust this provisional analysis to give an accurate overview of the distribution of acid-loving Eunotia species, for example, nor of planktonic diatoms such as Asterionella formosa;
  2. I also noticed some species typical of brackish taxa which have been recorded occasionally in freshwaters (e.g. Bacillaria paxillifer). More comprehensive coverage of coastal and estuarine environments would probably show many of these to be quite common.   The same reasoning applies to those diatom species associated with terrestrial habitats (e.g. Hantzschia amphioxys).
  3. Most of the samples in our database come from rock or plant surfaces and it is likely that diatoms that prefer other substrata have been under-recorded, which will complicate interpretation of their distribution. Absence of evidence is not evidence of absence.
  4. Many of the taxa that are rarely recorded belong to taxa that have been subject to taxonomic uncertainty over the past few decades, leading to variations in how they have been recorded. Some of the rare diatoms are “varieties” of common species but as these often are (or were) poorly described in the literature, many analysts have not tried to distinguish them.
  5. Finally, we have to be sure that the records actually represent living populations. Because diatomists usually work from the empty silica frustules, we cannot tell whether a cell was alive at the time it was collected.   If you find a number of frustules of the same species, then it is reasonable to assume that some of these were alive, but if a species is represented by a single frustule, we have to consider the possibility that this was washed into the site from elsewhere, and never actually grew there.

The positives from this process are that I think we can start to make some judgements about the rarity (or otherwise) of diatoms that are reasonably well circumscribed in the literature (i.e. a low chance of misidentification) and where the underlying taxonomy has been relatively stable. A further criterion at this stage is that the candidate taxa must be common in streams and rivers and, ideally, associated with hard surfaces rather than soft sediments.   That’s quite a lot of caveats, but in the next post I’ll start to sort through the list and see if we have some genuine candidates for “scarce” or “rare” diatoms.


The dataset referred to was developed for:

Kelly, M.G., Juggins, S., Guthrie, R., Pritchard, S., Jamieson, J., Rippey, B, Hirst, H. and Yallop, M (2008). Assessment of ecological status in U.K. rivers using diatoms. Freshwater Biology 53: 403-422.