Diatom hunting in the Pirin mountains

I started 2018 peering down my microscope at a sample that I collected whilst in Bulgaria back in the summer.   I have written about my trip to the Pirin mountains before (see “Desmids from the Pirin mountains”) but the diatom sample that I collected from Lake Popovo had remained unexamined since I got back.

I had waded into the littoral zone of this steep-sided corrie lake and picked up a few of the smaller stones, which I had then scrubbed with the toothbrush stowed in my rucksack to remove the thin film of diatoms.  These, like most of the algae that I collect on my travels, get treated to a bath in local spirits to ease the journey back to the UK.  This is not an ideal preservative for soft-bodied algae but is not a problem when your primary interest is diatoms with their tough silica cell walls.  Once I got back, I had them prepared and mounted ready for inspection, but then got distracted by other things and have only just got around to having a proper look.

The two most abundant taxa were the Achnanthidium minutissimum complex (probably at least three species) and Cymbella excisiformis.  Together, these constituted over eighty per cent of all the diatoms in the sample.  Ten years ago, I would have called the organism I was calling Cymbella excisiformis by a different name, Cymbella affinis, but opinions have shifted more than once.  The original Diatomeen im Süsswasser-Benthos von Mitteleuropa has images of C. affinis that are actually C. tumidula, and also describes C. excisa as a separate species.   However, the most recent view is that C. affinis and C. excisa are two names for the same species, with C. affinis taking precedence.   To confuse matters yet further, the population illustrated below shows a gradation of features from “C. affinis” to “C. excisiformis”, suggesting that the use of length:width as a discriminating factor is over simplistic.  Krammer tried to explain his rationale for distinguishing between these species in his 2002 monograph but he uses the name “C. excisa” for the organism called “C. affinis” in our 2017 English edition.  Confused?  You will be ….

Cymbella excisiformis” from Lake Popovo, Pirin Mountains, Bulgaria, August 2017.   Based on Lange-Bertalot et al. (2017)’s criteria of length:breadth 4.2-5.3 in C. excisiformis compared to 3.1 – 3.8 in C. affinis, images a., b. and c. are C. excisiformis whilst d., e., f. and g. are C. affinis.   Scale bar: 10 micrometres (= 1/100th of a millimetre).

This is another good example of points that I have made several times before: that we should always try to identify populations rather than single cells, and that we should treat dimensions stated in the literature as indicative rather than definitive (see “More about Gomphonema vibrio”).    Length:width, in particular, can change a lot during the life-cycle of the diatom.

Species of Gomphonema were also present in the sample.  Though not numerically abundant (none constituted more than one per cent of the total count), they included some large cells which, in addition, have extensive mucilaginous stalks, so their contribution to total biomass is greater than their low abundance suggests.   I’ll write more about the ecology of these species in the next post.   Finally, I also found other Cymbella species, as well as some Encyonema and Encyonopsis, and a few valves of Eucocconeis flexella, a relative of Achnanthidium and Cocconeis which has a distinctive diagonal raphe.

Gomphonema spp. from Lake Popovo, Pirin mountains, Bulgaria, August 2017.   h., i.: G. acuminatum; j.: G. truncatum; k.: unidentified girdle view; l.: G. pumilum.  Scale bar: 10 micrometres (= 1/100th of a millimetre).

There is, at this point in time, no official Bulgarian method for assessing the ecological status of lakes using diatoms so I have evaluated Lake Popovo as if it were a low alkalinity lake in the UK instead.  Using the method we developed, this one sample has an Ecological Quality Ratio of 0.92, which puts it on the border between high and good status.   Looking around the lake, I see no reason why it should not be firmly in high status but, at the same time, I am using an evaluation method that was designed for lakes 2000 kilometres away, so maybe we should not expect perfect results.    However, I have performed similar exercises at other lakes far from the UK and also got similar results (see “Lago di Maggiore under the microscope”) which points to a basic robustness in this approach.

The outflow of Lake Popovo leads into a cascade that ends in the first of a series of lakes, the “Fish Popovski” lakes.   I wrote about the desmids in this lake back in September (see “”Desmids from the Pirin mountains”) and will return to this sample in order to describe the diatoms in another post.   But, meanwhile, the assemblage at Popovo reminded me of the littoral algae in another lake that I really should tell you about …

Miscellaneous diatoms from Lake Popovo, Pirin mountains, Bulgaria, August 2017.   m.: Cymbella sp.; n.: Encyonema neogracile; o. and p.: Eucocconeis flexella (raphe valve and girdle view respectively).  Scale bar: 10 micrometres (= 1/100th of a millimetre).

Lake Popovo, photographed from close to the location from which my sample was collected.  The brass plate on the rock at the right hand side gives the altitude as 2234 metres above sea level.  The photograph at the top of the post shows Lake Popovo against a backdrop of the Pirin mountains.


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

Krammer, K. (2002).  Diatoms of Europe volume 3: Cymbella.   A.R.G. Gantner Verlag K.G., Ruggell, Germany.

Lange-Bertalot, H., Hofmann, G., Werum, M. & Cantonati, M. (2017).   Freshwater Benthic Diatoms of Central Europe: Over 800 Common Species Used In Ecological Assessment (edited by M. Cantonati, M.G. Kelly & H. Lange-Bertalot).   Koeltz Botanical Books, Schmitten-Oberreifenberg.

The UK lake diatom assessment method is described in:

Bennion, H., Kelly, M.G., Juggins, S., Yallop, M.L., Burgess, A., Jamieson, J. & Krokowski, J. (2014).  Assessment of ecological status in UK lakes using benthic diatoms.  Freshwater Science 33: 639-654.

Details of the calculation can be found in the UK TAG method statement.


In pursuit of Bulgarian diatoms …

Our visit to Rila Monastery was one of the cultural highlights of our trip to Bulgaria in the summer (see “The art of icons …”) with the added bonus that it is set amidst some spectacular scenery.   The monastery was founded by St Ivan Rilski (“St John of Rila”) in the 10th century, though the present structures date from the 19th century.   St Ivan, I gleaned from my reading, shared a love of nature with St Francis and St Cuthbert and Rila’s remote location reflects his desire to live the life of a hermit.

Beyond the monastery the road twists and turns amidst the forest that lines the valley.   Occasionally, the canopy is broken by meadows and orchards of apricot and plum trees.  The apricots were ripe for harvest and at an easy height for foraging as we passed by.   We were too late in the year to find many wild flowers in these meadows but the absence of any of the paraphernalia associated with intensive agriculture made us suspect that these would have been a riot of colour in the spring.    Eventually, we came to a small hamlet set amidst a wider area of meadows, where we refuelled at a small café serving the delicious local bean soup (“bob chorba”) before making our way across the meadow to the stream.

A haystack in a meadow near Rila Monestery, August 2017.

The Rila stream itself flowed through a densely-wooded channel, with a variety of substrates from coarse sand to moss-covered boulders.   The larger stones were mostly granite, reflecting the underlying geology of the region, which almost certainly means that the water was very soft.  The surrounding vegetation and low population density also mean that the water is probably as pure as we are likely to find anywhere in Europe.   I had a toothbrush and sample bottle in the bottom of my rucksack and scrubbed a few cobble-sized stones to remove the thin surface film and stowed this for the journey back down the valley.

Sampling Rila stream at Kirilova meadows, a few kilometres upstream from Rila Monestery in August 2017.   The photograph at the top of the post shows the scenery around Kirilova meadows.

It has taken until now for me to get around to looking at this sample, which turned out to have a large population of Gomphonema rhombicum, along with quite a lot of Achnanthidium minutissimum and relatives, confirming my suspicion of a circumneutral, low nutrient environment.   G. rhombicium is a relatively uncommon diatom, so I was intrigued to have a closer look.  It has a similar outline to G. pumilum, which is very common, but is larger and has a distinct broad lanceolate axial area and, consequently, relatively short striae.   The axial area broadens out a little further at the central area, where there is also a single stigmoid.  I wish now that I had had a look at the sample before digestion as many of the larger Gomphonema species have long mucilaginous stalks, whereas G. pumilum and relatives tend to be attached to the substrate by short mucilaginous pads.   I can find nothing in the literature that alludes to the habit of the living cell and, on this particular occasion, am in no position to judge the shortcomings of my peers.

By coincidence, the type location for Gomphonema rhombicum is given as “Appleby, Westmoreland”, which is just over an hour’s drive from where I live.   The most obvious place to hunt for G. rhombicum would be the River Eden; however, the scant details on the ecology of G. rhombicum that I can find suggest a preference for softer water than found here.  This is a geologically-complex area so there is a possibility of suitable habitat existing in a stream in the vicinity.  If the Eden was the location from which the original population of G. rhombicum was collected, then I suspect that it may have been a casualty of the agricultural intensification that has taken place in this area since it was first described.

Cleaned valves of Gomphonema rhombicum from Rila stream at Kirilova meadows, August 2017.  a. – g. show valve views; h. shows a girdle view.  The scale bar is 10 micrometres (= 100th of a millimetre).

That leaves me with just one option: a return to the Rila National Park.  I think I can just about live with that.   We stayed at Hotel Pchelina, a few kilometres down the valley from Rila Monestery and our dinner that evening was the most delicious grilled trout that I have ever tasted.  If a return visit was needed to solve the mystery of Gomphonema rhombicum’s habit then I think that is a hardship with which I can just about cope …


Iserentant, R. & Ector, L. (1996).  Gomphonema rhombicum M. Schmidt (Bacillariophyta): typification et description en microscopie optique.   Bulletin Français de la Pêche et de la Pisciculture 341/342: 115-124.

Desmids from the Pirin mountains

Our travels in southern Bulgaria took us south from Rila Monastery to Bansko, on the edge of the Pirin Mountains and, from here, via a chair lift and a rather more strenuous walk than we had expected to Popovo Lake  situated in a corrie overlooked by rugged mountains soaring up to over 2800 metres.

The outflow from the lake cascades over the lip of the corrie and down the hillside to a series of smaller lakes, before merging with some other small streams to form the Mesta river which flows south from Bulgaria and through northern Greece to the Aegean Sea.   The footpath that took us back from Popovo took a gentler route down the hillside but brought us close enough to the first of this series of lakes for the bright green areas at the margins to pique our interest.  Getting closer, we found ourselves on soft, yielding Sphagnum bog, more familiar to us from the moorlands of northern England (see, amongst other posts, “Back to the bog“) than in southern Europe.   A lot of the rock that we had passed on our hike up from the chair lift terminus had been granite, so the water around us would clearly have been soft enough for Sphagnum and, I guess, the marshy land was testimony to a higher level of precipitation than the cloudless skies that we encountered suggested.

The first of the “Fish Popovski” lakes below Popovo Lake in the Pirin Mountains, southern Bulgaria, with the marshland area clearly visible in the foreground.

One of the pools in this bog attracted my attention: a mat-like portion of the substratum, had floated to the surface whilst still being loosely attached at one corner.   This is a good clue that the substratum is jam-packed with algae, doing a double job of binding the silty particles together into a cohesive whole and, at the same time, pumping out oxygen as a by-product of photosynthesis in order to make the mat buoyant.   I last wrote about this in 2013 when I found some mats of Oscillatoria limosa in my local river (see “More from the River Wear”).   The same phenomena seem to be at play here although, on closer investigation, it was desmids rather than Cyanobacteria which were responsible for the mat.   There were, in fact, far fewer filamentous algae in this particular mat, something that I quickly noticed as there was very little physical integrity to the mat, and it dissolved into a suspension of fine particles as soon as I tried to remove a piece.   Had I stayed until nightfall, I expect the mat would have gradually sunk again as the rate of photosynthesis declined and oxygen production ceased.

The desmid mat floating up from the bottom of the bog pool beside the first of the Fish Popovski lakes in the Pirin Mountains.

The most abundant desmids in the mat were Hyalotheca dissilens and Closterium baillyanum.  The former is a filamentous desmid, whose chains of cells are enclosed in a broad mucilage sheath and, whilst there were many fewer of these filaments than I have seen in more cohesive mats, I suspect that these played a role in trapping the other algae, plus organic and inorganic particulate matter to form the structure that I saw.  Closterium baillyanum, by contrast, has large, robust cells and, in this case, the cell wall has a distinct brown colour.  Other desmids that I found in my brief examination included Tetmemorus granulatus, which has cylindrical cells with a narrow incision in the broadly rounded apex (most clearly visible at the left-hand side of the illustration) and two species of Euastrum: E. humerosum and E. ansatum.  There were also a few large cells of Eremosphaera, a green alga though not a desmid (see “More from Loughrigg Fell”), and various assorted unicellular algae.  I’ll write about the diatoms in a separate post.

Some common desmids from the Pirin mountains: a. Hyalotheca dissilens; b. Closterium baillyanum prox.; c. Tetmemorus granulatus.  Scale bar: 25 mm (= 1/40th of a millimetre).

Unfortunately, the chloroplasts in these illustrations are not at their best.   My preference is always to keep algal samples fresh for as long as possible but, as I was moving around southern Europe in August I adopted my usual practice when travelling and added some vodka to the sample as a temporary preservative.   Nonetheless, Dave John, who identified the species, could find enough in the general morphology and cell wall characteristics to guide him.   Desmid species are considered to be cosmopolitan so he was able to use identification literature from Northwest and Central Europe in order to do this.   The search terms “desmid” and “Bulgaria” yielded no papers when I looked on Web of Science, so this looks like a relatively unexplored corner of Europe, as far as this group is concerned.   Having said that, all five of the species illustrated here are listed in the checklist of Romanian algae, and it is quite likely that there are local publications that have not made it onto the major bibliographic databases.

Incidentally, it was only after I had bought a miniature of vodka that I realised that I should have used this as an opportunity to buy a bottle of rakija, the local spirit.  For some reason, I had assumed that this was an ouzo-type spirit and that it would give the water an opaque milky-white appearance (caused by the lower solubility of essential oil of anise in water compared to alcohol).   Experimental studies later that same evening showed that the local rakija was closer to the Romanian ţuică, prepared from grapes, plums or apricots (the latter is especially good), and would have made a fine preservative.   I am older and wiser although, in the immediate aftermath of my experiment, that wisdom may not have been immediately apparent.

More desmids from the Pirin mountains: d. Euastrum humerosum; e. E. ansatum. Scale bar: 25 mm (= 1/40th of a millimetre).


Cărăuş, I. (2017). Algae of Romania.  A Distributional Checklist of Actual Algae.  Version 2.4.  Original print edition published by University of Bacău, Romania.  Latest version available online [https://www.researchgate.net/publication/285888477_The_algae_of_Romania]

The art of icons …

A week off from algae, as I travel around Bulgaria on holiday.  In between exploring mountains (and, I admit, pulling a toothbrush from my knapsack on a couple of occasions for a sneaky diatom sample), I have been learning about the intricacies of Eastern Orthodox icon painting, as a break from my normal scientific and artistic routines.  My interest was piqued by a visit to the superb icon gallery at the National Museum of Art of Romania in Bucharest last year, though this mostly served to demonstrate how little I knew, either about icons or their context in Orthodox worship.

Context is important because, in our secular age, we are most likely to encounter religious art in a gallery rather than a church.  My initial response to an icon, such as that in the image below, is to place it into a Western art historical context.   I note the relatively simple modelling of the features, depicting archetypes of religious figures and the flat background.  There is no attempt to place the figure in three-dimensional space, as most religious painters from the Renaissance onwards would have tried to do.  They were trying to draw the viewers in, creating space inside the picture that encouraged them to engage with the subject matter.  Painters of the Counter-Reformation, such as Rubens, went further, painting the protagonists in their religious paintings life size and dressing them in contemporary clothes to encourage viewers sense of participation.

An icon of Christ Pantocrator from the Bankso school of icon painters (late 18th / early 19th century) in southern Bulgaria.  The image at the top of the post shows the iconostasis at Mānāstirea Stavropoleos, Bucharest, Romania.

By contrast, by flattening everything but the subject’s physiognomy, the Orthodox icon painter projects his subjects into our space, encouraging a different type of engagement.   Orthodox Christianity has a strong tradition of contemplative prayer, in which knowledge of God is attained through meditative practices such as repetition of a meaningful word or short phrase.  In this context, icons can serve as objects that help viewers to concentrate their minds while they step away from the everyday world and towards the divine realm.  One manifestation of this is that there is typically more activity in an Orthodox Church, compared to a Catholic or Protestant church, outside of organised services, as worshippers make their own private devotions in front of icons.

This use of repeated phrases suggests parallels with eastern religions – the Hindu incantation “Om mani padme hum” being the best-known example.  Look, too, at the right hand of Christ in the icon below.  That, too, resembles the symbolic hand gestures – mudras – found in Hindu and Buddhist contemplative practices.   Whether there is more than a superficial resemblance, in this particular instance, is a moot point.  Christ’s hand is raised to confer a blessing on the viewer and the position of the fingers is related to this.  They spell out “ICXC” –  IhcoyC XpictoC, or “Jesus Christ”.   The confusion with eastern practices arises, I suspect, from the way that the fourth finger is bent over to touch the thumb, similar to the Chin Mudrā.

On the other hand, there would have been ample opportunity for exchange of ideas along the Silk Road.  Early Christianity extended much further east, and Buddhism further to the west before the rise of Islam. Diarmaid MacCulloch has suggested that the principle of monasticism, for example, may have been brought into the church by early missionaries returning from the east and, if this is the case, then it is possible that practices associated with monasticism would also have flowed east.  And, equally, there is no reason to assume that the movement was entirely one-way or solely between Christianity and Buddhism.  Our first reaction on walking into Rila monastery in southern Bulgaria was to notice the physical similarities with the huge Madrassas that we saw in Uzbekistan earlier this year (see “Reaching for the stars …“).

What we can see an Orthodox icon, in other words, is a product of time and place, only if we also recognise that time and place are continua, that ideas can flow and that there is a ‘natural selection’, of sorts, that selects and shapes these to fit local circumstances.  Traveling broadens the mind, without a doubt, but sometimes you need to unload your preconceptions in order to free up the mind to see the world through fresh eyes.

The courtyard of Rila monastery in southern Bulgaria with the Church of the Nativity on the right.