Murder on the Barcode Express …

A long time ago, Agatha Christie imagined a train coming to a halt in a snowdrift somewhere in Croatia.  By the morning, one of the passengers was dead.   Eighty years later, a group, only slightly larger than Hercule Poirot’s pool of suspects, gathered in a room in modern Zagreb to plot another fiendish murder.   The victim, this time, would be  …. traditional diatom taxonomy.

“Murder” is far too strong a term for this particular whodunit; maybe I should say “aiding and abetting” rather than actually committing the crime, but I think the outcome might be the same.  The conspirators in Zagreb are all involved in developing methods that use molecular barcoding to identify diatoms and have been busily collecting sequences of the many diatom species in order to establish the libraries that we need to link these barcodes to the appropriate Linnaean binomial.   Some years into this, we still have no more than about 15% of freshwater diatom species matched to barcodes.  We are starting to think about ways of filling in the gaps more quickly than is possible using the conventional approach of isolating a diatom, growing it in culture and then sequencing the appropriate marker genes.

The most radical of these alternatives is to by-pass Linnaean binomials altogether and classify diatoms by their barcodes alone – as “operational taxonomic units” or OTUs.   Most of us have spent most of our careers using morphology-based taxonomy and any move away seems like an act of treachery towards a fundamental tenet of our craft.  But the time has come to take a dispassionate view and ask what a species name brings to ecology.   At a very practical level, the use of Linnaean binomials makes it much easier for us to compare data with colleagues and with records in the literature.    Taxonomists would argue that their work helps us to understand the relationships between species but, unfortunately, in this particular branch of science, we make little use of these relationships, and the role of taxonomy is primarily to give us a consistent means of organising the myriad tiny pieces of silica which we find in our samples.

That business of consistent naming could, in theory, be performed for barcodes just as efficiently using digital tags as OTUs and this would also work for the 85% of species where the link between traditional morphology-based taxonomy and marker genes has not yet been established.   So what about the link that Linnaean binomials give us to established knowledge?   Here, again, we need to be brutally frank: ecological information for most freshwater diatoms is limited to information about preferences for hardness/alkalinity, inorganic nutrients, organic pollution, acidity and salinity and that information can be replicated very easily by linking files of metabarcoding and environmental data.  There are very few experimental studies that offer insights into the ecology of freshwater benthic diatoms beyond that gained from looking for associations between diatom distribution and a few common variables.

The plotters plotting …  DNAqua-net workshop in Zagreb, November 2017.  The top photograph shows Zagreb cathedral against the skyline.

The problem is not that we do not see the merits of traditional Linnaean taxonomy, it is that we cannot make a strong case for the funding necessary to collect barcodes for all species.   The final downward thrust of the dagger will, in other words, be inflicted by the bureaucrats whose budgets will not stretch to cataloguing the enormous breadth of algal diversity.   Diatoms sit in the awkward middle ground between larger organisms such as fish where any suggestion of not using traditional taxonomy would be greeted with derision and the microbial world where the idea of applying Linnaean binomials to the enormous diversity uncovered by molecular techniques is equally risible.   Diatom names mean little to the bureaucrats who manage our environmental agencies and, given the choice between a spreadsheet of incomprehensible Latin names or one of equally incomprehensible OTUs, all else being equal, they will choose the cheapest.

“All else being equal” is the key phrase.   I think that there is growing awareness now that one downside of barcoding is that it risks sidestepping the need for trained biologists at all: samples will be collected by technicians, processed in high-throughput laboratories and results churned out through black box computer programs.   The situation for diatoms is worse than for most groups of organisms used for ecological assessment because so much attention is given to the laboratory stages of producing a list of taxa and relative abundances.  We are, however, now approaching the point when DNA sequencers can produce data of equivalent sensitivity to that produced by light microscopy.   The message that barcoding has the potential to be a good friend but a poor master could be lost as our paymasters recognise the potential for reducing costs.   What we need to do now is use those “little grey cells” to ensure that good biological insight is not the victim of a heinous crime.

Advertisements

As if through a glass darkly …

Life used to be so easy: I stared down my microscope, named the diatoms I could see, counted them and, from these data, made an evaluation of the quality of the ecosystem that I was studying.   Along with the majority of my fellow diatomists, I conveniently ignored the fact that I was looking at dead cell walls rather than living organisms.   My work on molecular barcodes as an alternative to traditional microscopy has been revelatory as I try to reconcile these two types of data.   At one level, what I see down the microscope is a benchmark for what I should expect to see in my barcode output.  Yet, at the same time, the differences between the two types of data show up the limitations of traditional data – and the assumptions that underpin the ways that we work.

Take a look at the plate below which shows two of the most common diatoms in UK rivers: Ulnaria ulna is one of the largest that I encounter regularly whilst Achnanthidium minutissimum is often one of the most abundant in my samples, particularly when the level of human pressure is relatively low.  When we analyse samples with the light microscope, we record individuals, so both of these score “1” in my data book despite the fact that U. ulna is about 100x larger (by volume) than A. minutissimum.

Specimens of Ulnaria ulna (top) and Achnanthidium minutissimum (bottom).  Both are from cultures used for obtaining sequences for the reference library for our molecular barcoding project.   Scale bar: 10 µm.   Photographs: Shinya Sato, Royal Botanic Gardens, Edinburgh.

When we analyse a sample using Next Generation Sequencing (NGS), we count not cell walls but copies of the rbcL gene, which provides the blueprint for Rubisco, a key photosynthetic enzyme.   As I write, there is no clear understanding of how the number of rbcL copies relates to the number of individuals.  We know that each chloroplast within a cell will have at least one copy of this gene, and usually several. There is also some evidence that larger chloroplasts have more copies of the gene than smaller ones and there is also likely to be a measure of environmental control.  The key message that I try to get across in my talks is that NGS data are different to the data we are used to gathering using microscopy.  These differences do not mean that it is wrong, just that we need to leave some of our preconceptions before starting to interpret this new type of data.

However, we could also argue that counting the number of copies of the gene for an important photosynthesis enzyme should be giving us a better insight into the contribution of a species to primary productivity than counting the number of cell walls.  In other words (whisper this …), rbcL might not just be different, it might be better, especially if our purpose is to understand the contribution the various species in the biofilm make to primary productivity in stream ecosystem.  At the moment there are plenty of problems with the NGS-based method, not least the fact that we often cannot assign half the copies of the rbcL gene in a sample to a species, but the situation is improving all the time …

Some recent work pushes this a little further.   Jodi Young and colleagues at Princeton University have demonstrated large variation in the kinetics of Rubisco in diatoms, and in their carbon-concentrating mechanisms (see “Concentrating on carbon …” for more about these).  Although their work is focussed on marine phytoplankton, the variation within Rubisco and carbonic anhydrases could go some way to explaining the sensitivity of diatoms to inorganic carbon (see “Ecology in the Hard Rock Café …”).   In other words, rbcL is not an irrelevant DNA sequence, as the term “barcode” may imply (in contrast to barcodes based on the ITS region, for example), it is deeply implicated in the reasons why a species lives in particular place.

And yet, and yet, and yet …  The same could be argued for morphology, up to a point at least.   The shape of a Gomphonema or a Navicula also helps us to understand the organism’s relationship with its environment.   The problem is that modern taxonomists tend to focus on a much finer level of detail – on the arrangement and structure of the various pores on the silica frustule, for example – and offer few insights into what these minute differences mean in terms of the ecophysiology of the organisms.  Even at the whole-cell scale, information on habit, which is linked to form (Gomphonema tending to live on stalks or short mucilage pads secreted from their foot poles for at least part of their life-cycle, for example) is rarely incorporated into assessment systems.   The move from using light microscopy to using NGS, in other words, means replacing an imperfect system with which we are familiar with one that we are still learning to understand.  Both offer unique information and the gains from using one approach rather than the other, will be offset by losses of insight.

That leaves us with two big challenges over the couple of years, as UK diatom-based assessments move from light microscopy to NGS.  The first is to work harder to understand what NGS outputs are actually telling us about the environment over and above the minimalist ecological status indices that spew out of our “black box” computer programs.   The second is to maintain an understanding of the properties of whole organisms and how these interact with one another and with their environments.   I guess I should add a third challenge to this pair: persuading middle managers who have at best a sketchy understanding of diatoms and phytobenthos and already-stretched budgets that any of this matters …

References

Badger, M.R. & Price, G.D. (2003).  The role of carbonic anhydrase in photosynthesis.  Annual Review of Plant Biology 45: 369-392.

Young, J.N. & Hopkinson, B.M.M. (2017).  The potential for co-evolution of CO2-concentrating mechanisms and Rubisco in diatoms.  Journal of Experimental Botany doi: 10.1093/jxb/erx130.

Young, J.N., Heureux, A.M.C., Sharwood, R.E., Rickaby, R.E.M., Morel, F.M.M. & Whitney, S.M. (2016).  Large variations in the Rubisco kinetics of diatoms reveals diversity among their carbon-concentrating mechanisms.  Journal of Experimental Botany 67: 3445-3456.

What’s a pretty diatom like you doing in a place like this?

Whilst looking at some samples from an experiment conducted on mesocosms beside a chalk stream, Candover Brook in Hampshire for Mark Ledger and colleagues, I came across a diatom that I had not seen before and which, at first glance, was out of place.   As the images above show, it is a diatom whose cells join together to form chains which, in turn, means that they typically present their sides to the viewer rather than the valve face, which is the way that the writers of identification guides generally assume that we can see. It took some time to track down a couple of cells that were lying face-upwards so that I could try to name the species and some of the few that were lying this way were damaged (see left hand image), perhaps itself a consequence of the naturally strong links between the cells.

Naming the genus was relatively straightforward: the valve shape, fine striae and very narrow axial area (the gap along the median line of the valve face between the two rows of striae) coupled with the tendency to form chains all pointed to Fragilariforma.   However, most of the Fragilariforma that  I encounter are in soft water, often acid habitats whilst this sample was from a flume beside a chalk stream in southern England.   After scratching my head a little more, and sending images to my friend Lydia in Germany, I eventually decided that Fragilariforma nitzschioides was the most likely name for this diatom.  Searching through my records, I found only one other record for this species: from the River Itchen (into which Candover Brook drains) in the mid-1990s.  That must be more than coincidence.   Interestingly, Hoffman et al. (2011) describe the species as “rare” and say that its ecological preferences are “difficult to define”.

The limited records that we have show that this species does not behave in the same way as most other representatives of the genus.   The weighted average of pH for the genus is 6.6 (see graph below), but there are plenty of records extending into more acid waters.  By contrast, the River Itchen population was recorded at pH 8.1 and the pH in Candover Brook will be very similar.   Most of the records for the genus came from relatively soft water, in contrast to the very hard water found in a chalk stream.  The scarcity of records of a species that is well described in the literature also suggests that this might be a genuinely rare diatom (see “A “red list” of endangered British diatoms”).

One other peculiarity of this species is the name itself.   Fragilariforma was one of a number of genera split away from Fragilaria by Dave Williams and Frank Round in 1986, originally as “Neofragilaria”.  Fragilaria nitzschoides, was not formally transferred at the time, presumably because the authors did not have access to the type material.  They presented good evidence for this new genus but a few people – notably Horst Lange-Bertalot – have continued to group these species under Fragilaria.   This is the situation in Diatomeen im Süsswasser-Benthos von Mitteleuropa but, curiously, for Fragilaria nitzschoides, he created the new combination of “Fragilariforma nitzschoides” purely as a synonym (see p. 268).   The good news is that the next version of this book (see “Tales of Hoffman”) does use these new names.

The relationship between Fragilariforma spp and pH (left) and alkalinity (right) in UK rivers, based on the mid-1990s dataset described in “The challenging ecology of a freshwater diatom”.  Vertical lines show the boundaries for high (blue), good (green), moderate (orange) and poor (red) status classes based on current UK standards and the arrows show the location of the River Itchen population of Fragilariforma nitzschoides along these gradients. 

References

Hofmann, G., Werum, M. & Lange-Bertalot, H. (2011).   Diatomeen im Süßwasser-Benthos von Mitteleuropa. A.R.G. Gantner Verlag K.G., Rugell.

Williams, D.M. & Round, F.R. (1987).  Revision of the genus Fragilaria.  Diatom Research 2: 267-288.

Williams, D.M. & Round, F.R. (1988).  Fragilariforma nom nov., a new generic name for Neofragilaria Williams & Round.  Diatom Research 3: 265-267.

The way things were …

Writing the previous post led me to contemplate how much things had changed over the time that I have been working in this field.  Back in the early 1990s when I first set out to look at the response of diatoms to nutrients in streams, few in the National Rivers Authority (NRA, predecessor to the Environment Agency) regarded phosphorus as a serious pollutant in rivers, and most biologists thought about ecological quality solely in terms of organic pollution and invertebrates.   In order to investigate the effect of nutrients, I wanted to visit sites where organic pollution was not a problem.

I was helped in this task by the work done by biologists at the then Institute for Freshwater Ecology (now Centre for Ecology and Hydrology) who had just developed the early versions of RIVPACS (“River Invertebrate Prediction and Classification System”) which established the principle of expressing ecological quality as the observed quality / expected quality.  This, in turn, required an ability to predict the “expected” condition for any stream.   The work that had developed these equations started from a dataset of invertebrate and environmental data collected from a wide range of “unpolluted” running water sites which, in those far off days, was compiled by asking biologists working for the Regional Water Authorities (predecessors to the NRA) for their recommendations of sites that were of “good” or “fairly good” quality.  Nowadays, screening sites to be used for calibrating ecological methods is a much more rigorous procedure but this was the first tentative step on a long journey and “expert judgement” was as good a place to start as any.

The paper that emerged from this exercise (see reference below) analysed data from these “unpolluted” sites and classified them into eight groups.  Each of these groups consisted of sites that shared similar invertebrate assemblages which reflected similarities in the habitat, from upland, fast flowing becks to deep, wide slow-flowing rivers in the lowlands.  The authors included a useful table that listed the physical and chemical characteristics of each of these groups and I noticed that the phosphorus concentrations reported for these spanned a very wide range.   This meant that I could use these as the basis for putting together a sampling program that spanned a long gradient of nutrient pressure without the complications of organic pollution.   The outcome of that work was the first of the two papers referenced in my previous post.

Time has moved on and I thought it would be interesting to revisit these “unpolluted” sites to see how they would be classified using the UK’s current standards for phosphorus.  This highlights a striking difference between the prevailing idea of “unpolluted” in the early 1980s and the present day, as all of these groups had average concentrations that equate to substantial enrichment by modern standards; in half the groups this average concentration would be classified as “poor status” whilst the maximum concentrations in three groups equates to “bad status”.   Whatever way you look at it now, these sites were far from “unpolluted”.

Classification of TWINSPAN end-groups of unpolluted river sites in Great Britain based on Armitage et al. (1984) along with average and maximum phosphorus concentrations recorded in each group and the phosphorus status based on current environmental standards.  M = moderate status; P = poor status; B = bad status.

I am not being critical of the approach taken by Patrick Armitage and colleagues.  In many ways, I regard the work of this group as one of the most significant contributions to the science of ecological assessment in my lifetime.   I am just intrigued to see how the thinking of ecologists and regulators has moved on in the thirty years or so since this paper was published.  I know from my own early conversations with NRA biologists that inorganic nutrients were not perceived as a problem in rivers until the early 1990s.   It was probably the European Community’s Urban Wastewater Treatment Directive (UWWTD) that started to draw the attention of biologists in the UK to these problems, and which led to the development of stricter environmental standards for nutrients, though not without opposition from several quarters.

This, then is a situation where good legislation provided the impetus needed to start the process.  There were places in the UK – rivers in the Norfolk Broads, for example – where nutrients were already being regulated, but these were special circumstances and nutrient problems in most rivers were largely ignored. Indeed, as I said in my previous post, phosphorus was not even measured routinely in many rivers.   I heard via my professional grapevine that it was the Netherlands who had made the case for the clauses in the UWWTD concerning regulating nutrients, as their stretches of the lower Rhine were subject to numerous problems caused by unregulated inputs of nutrients from countries upstream.   I do not know if this is true, but it is certainly plausible.   However, once the need to control eutrophication in rivers was codified in UK law, then the debate about how to evaluate it started, one of the outcomes of which was more funding for me to develop the Trophic Diatom Index (referenced in the previous post).  And, gradually, over time, concentrations in rivers really did start to fall (see “The state of things, part 2”).   I’d like to think the TDI played a small part in this; though this might also mean that I am partially responsible for the steep increase in water charges that everyone endured in order to pay for better water quality …

Reference

Armitage, P.D., Moss, D., Wright, J.F. & Furse, M.T. (1984).  The performance of a new biological water quality score system based on macroinvertebrates over a wide range of unpolluted running-water sites.  Water Research 17: 333-347.

The challenging ecology of a freshwater diatom?

amp_pedi_pollybrook

Amphora pediculus from Polly Brook, Devon, December 2016. Scale bar: 10 micrometres (= 1/100th of a millimetre).

The images above show one of the commonest diatoms that I find in UK waters.  It is a tiny organism, often less than 1/100th of a millimetre long, which means that it tests the limits of the camera on my microscope.  In recent months, however, it is not just the details on Amphora pediculus’ cell wall that I am struggling to resolve: I also find myself wondering how well we really understand its ecology.

The received wisdom is that Amphora pediculus favours hard water, does not like organic pollution and is relatively tolerant of elevated concentrations of inorganic nutrients.  This made it a very useful indicator species in a period of my career when we were using diatoms to identify sewage work s where investment in nutrient-removal technology might yield ecological benefits.  There were many nutrient-rich rivers, particularly in the lowlands, where any sample scraped from the upper surface of a stone was dominated by these tiny orange-segment-shaped diatom valves.   Unfortunately, twenty years on, many of those same rivers have much lower concentrations of nutrients (see “The state of things, part 2”) but still have plenty of Amphora pediculus.   Did I get the ecology of this species wrong?

The graph below shows some data from the early- and mid- 1990s showing how the abundance of Amphora pediculus was related to phosphorus.   The vertical lines on this graph show the average position of the boundaries between phosphorus classes based on current UK standards.   Records for A. pediculus are clustered in the “moderate” and “poor” classes, supporting my initial assertion that this species is a good indicator of nutrient-enriched conditions, but there are also samples outside this range where it is also abundant, so A. pediculus is only really useful when it is one of a number of strands of evidence.

aped_v_p

The relationship between Amphora pediculus and reactive phosphorus in UK rivers, based on data collected in the early-mid 1990s.  Vertical lines show the average boundaries between high and good (blue), good and moderate (green), moderate and poor (orange) and poor and bad (red) status classes based on current UK standards and the two arrows show the optima based on this dataset (right) and data collected in the mid-2000s (left).

If we weight each phosphorus measurement in the dataset by the proportion of Amphora pediculus at the same site (i.e. so that sites where A. pediculus is abundant are given greater weight), we get an idea of the point on the phosphorus gradient where A. pediculus is most abundant.   We can then infer that this is the point at which conditions are most suitable for the species to thrive.  In ecologist’s shorthand, this is called the “optimum” and, based on these data, we can conclude that the optimum for A. pediculus is 154 ug L-1 phosphorus.  The right hand arrow indicates this point on the graph below. However, I then repeated this exercise using another, larger, dataset, collected in the mid-2000s.   This yielded an optimum of 57 ug L-1 phosphorus (the left hand arrow on the graph), less than half of that suggested by the 1990s dataset.   There are, I think, two possible explanations:

First, the 1990s phosphorus gradient was based on single phosphorus samples collected at the same time that the diatom sample was collected (mostly spring, summer and autumn) whilst the mid-2000s phosphorus gradient was based (mostly) on the average of 12 monthly samples.  As phosphorus concentrations, particularly in lowland rivers, tend to be higher in summer than at other times of the year, it is possible that part of the difference between the two arrows is a result of different approaches.  (For context, in the 1990s, when I first started looking at the effect of nutrients in rivers, phosphorus was not routinely measured in many rivers, so we had no option but to do the analyses ourselves, and certainly did not have the budget or time to collect monthly samples).

However, another possibility is that the widespread introduction of phosphorus stripping in lowland rivers in the period between the mid-1990s and mid-2000s means that the average concentration of phosphorus in the rivers where conditions favour Amphora pediculus have fallen.   In other words, A. pediculus is tolerant of high nutrient conditions but is not that bothered about the actual concentration.   My guess is that it thrives under nutrient-rich conditions so long as the water is well-oxygenated and, as biochemical oxygen demand is generally falling, and dissolved oxygen concentrations rising (see “The state of things, part 1”), this criterion, too is widely fulfilled.   I suspect that both factors probably contribute to the change in optima.

But the second point in particular raises a different challenge:  We often slip into casual use of language that implies a causal relationship between a pressure such as phosphorus and biological variables whereas, in truth, we are looking at correlations between two variables.   Causal relationships are, in any case, quite hard to establish and the effect that we call “eutrophication” is really the result of interactions between a number of factors acting on the biology.   All of these simplifications mean that it is useful, from time to time, to look back to see if assumptions made in the past still hold.   In this case, I suspect that some of our indices might need a little fine-tuning.  There is no disgrace in this: the evidence we had in the 1990s led us to both to a conclusion about the relative sensitivity of Amphora pediculus to nutrients but also fed into a large-scale “natural experiment” in which nutrient levels in UK rivers were steadily reduced.   When we evaluate the results of that natural experiment we see we need to adjust our hypotheses.  That’s the nature of science.  As the sign on the door of a friend who is a parasitologist reads: “if we knew what we were doing, it wouldn’t be research”.

References

The 1990s dataset (89 records) is mostly based on data used in:

Kelly M.G. & Whitton B.A. (1995).   A new diatom index for monitoring eutrophication in rivers.   Journal of Applied Phycology 7: 433-444.

The mid-2000s dataset (1145 records) comes from:

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

Tales of Hofmann …

freshwater_benthic_diatoms_

For the past five years or so the constant companion on my desk whilst I stare down my microscope has been a thick tome (2.8 kg) entitled Diatomeen im Süßwasser-Benthos von Mitteleuropa by Gabi Hofmann and colleagues.  It serves as my aide-mémoire when I am analysing freshwater diatoms, jogging my memory when I see a diatom that I recognise but whose name I have forgotten.  Before this was published, I used a French publication Guide Méthodologique pour la mise en oeuvre de l’Indice Biologique Diatomées which was free to download (I cannot find a link on the web any longer, unfortunately).   Neither of these is the last word in diatom taxonomy, but that was not the point: a lot of the time, I just need a gentle reminder of the right name for the species I am looking at, and I don’t want to have to pore through a pile of books in order to find this.

One of the strong points of both books is that they are copiously illustrated, and the plates are arranged very logically so that similar-shaped diatoms are together, making it easy to pick out differences.   For most routine identification, this is exactly what is needed: we may pretend that we are logical people but, in truth, pattern matching beats using a key nine times out of ten.   The 133 plates in Diatomeen im Süßwasser … act as a visual index and, to make life even easier, the species descriptions are arranged alphabetically and cross-referenced in the plates.  Having found an image that resembles the diatom I am trying to identify, it is straightforward to flick to the description to check the details.

There is just one problem: Diatomeen im Süßwasser-Benthos von Mitteleuropa is in German, and quite technical German at that.   I tell people not to worry because all the images are in English but, in truth, I worry that I may lose some of the nuances due to my linguistic limitations.   I was delighted, then, to be asked by Marco Cantonati to help produce an English version of the book.  Marco is half-German so reads and speaks the language fluently, and I was able to work on his first drafts in English to produce the final text.   Conscious that translating a German book into English is only a partial solution for the almost 70% of the EU who have neither as their first language, we also unpicked the prose in order to put the information about each species into a series of “bullet points” so that it was more accessible and we also took the opportunity to update some of the taxonomy.   A large part of last weekend was spent poring over the proofs so it should not be long now before it is available to purchase.

The great irony for me is that I am putting the finishing touches to this book at the same time as I am helping the Environment Agency to move away from using the light microscope to identify diatoms altogether.   I am just finalising the last of the regular competency tests that I organise in which, Environment Agency staff will participate, after which routine samples will be sent off for Next Generation Sequencing rather than being analysed by light microscope.  I’ve written about the pros and cons of this before (see “Primed for the unexpected …”) but there is a funny side.   After over a decade of struggling with identification literature in a language that almost none of them spoke my dedicated band of Environment Agency analysts get the book they dreamed about two months after their last diatom slide is packed away.   My sense of timing is, as ever, impeccable …

Hofmann, G., Werum, M. & Lange-Bertalot, H. (2011).   Diatomeen im Süßwasser-Benthos von Mitteleuropa. A.R.G. Gantner Verlag K.G., Rugell.

Prygiel, J.  & Coste, M. (2000).   Guide Méthodologique pour la mise en oeuvre de l’Indice Biologique Diatomées.   NF T 90-354.  Cemagref, Bordeaux.