Ecosystem networks and flux

Any species that we observe under a microscope is part of an ecological network that depends on resources and interactions with other species. Think of all the ecosystems we are so used to seeing and reading about regularly: Serengeti, Yosemite, Kaziranga, Madagascar and Artic, each of these have completely different profiles of macro-species from the other. Relationships between species can be complex and includes: symbiosis, cooperation, competition and predator-prey. Can we pry out these relationships when we observe different species under a foldscope. can we distinguish these from chance co-occurrences?

Exactly about a year ago, I started sampling life in this pond near my place. I sampled water from the same  spot each time to see how many species I could find. For most of the year, the pond would be covered with duckweed and the samples in bottles would last stably for months (see posts Day 1, Day 6, Day 17, Day 27, Day 80 from a previous sampling).  Several species were observed at a boring regularity. A month back, I examined the pond and to my surprise there was no duckweed. A couple of weeks ago, I took a sample of pond water and in one week the water was smelling like rotten eggs. Clearly there has been a change in the ecosystem, and this suspicion was corroborated by the kinds of species I saw, most of which I hadn’t seen previously.

Note: Several videos follow. One thing that is really fun to see is the kinds of motion of the different organisms. I have also embedded my comments in the videos to reduce text.


1. Video 1. Paramecium. The more I see this ciliate, more questions I have. See the video for what was running in my head. In the last part of the video, I included a 600x compound microscope view. since I wanted to show the spiral bacteria trapped in the food vacuole.

2. Video 2. Ciliate-19176-2-LI. A very Paramecium-like ciliate, although differs from it in many ways.

3. Video 3. Ciliate-19176-3-LI. An Oxytricha-like ciliate with cilia/cirri under the oral apparatus. Black granules are seen in the cytoplasm.

4.  Video 4. Ciliate-19176-4-LI. My best guess is this is related to Epalxis.

DiatomsAt least 5 types of Diatoms were seen in the sample.

5. Diatom-19176-5-L1. My best guess is Pinnularia.  One can see the nice gliding motion. Those droplet like structure in the diatom might be oil droplets.

6-9. Other diatoms

Diatom-19176-6-LI: Fragilriopsis-like, Diatom-19176-7-LI: Amphipleura, Diatom-19176-6-LI: Square-shaped diatom, Diatom-19176-6-LI: Navicula?

Euglenozoans. These are basal eukaryote related to the more deadly kinetoplastids (Leishmania, Trypanosoma). I particularly like them as they have a unique modified DNA base called Base J and some of my own work has been intimately involved in the discovery of an enzyme involved in this modification.

10. Video of Euglena. Thank you Dr. John Hall of University of Maryland for the identification. No amount of computational knowledge of Euglena prepares you for the first time view.

11. Leaf-like Phacus, wonderful motion pattern.

12.  An unknown Euglenozoan with a very interesting gyrating motion pattern. Euglenoid-19176-12-LI

13. Another unknown Euglenozoan  with a Phacus like moving pattern. Euglenoid-19176-13-LI

14.  A flagellate with a wiggly motion. Flagellate-19176-14-LI

15. A chlorophyte. Chlorophyte-19176-16-LI. While this could be a diatom shell, I still think it looks more like a charophyte type chlorophyte.


17. Amoeboflagellate  (Amoeboflagellate-19176-17-LI )These transition between the flagellate and amoeboid forms. Although I did not see the transition, I have a strong suspicion based on previous observations that they belong to this group. Watch the video for a particular pattern of arrangement. Is this a chance observation or does it have a deeper signficance.


18. Nematodes (round worms). Nematode-19176-18-LI

The longest nematode I have seen.  Many nematodes attack plant roots. I wonder if this was the cause for the death of the duckweed?

19/20. Bacteria: Several different types  of bacteria were seen, but two that caught my attention were this Beggiatoa like long bacterium (Bacterium-19176-19-LI) and these long Spirllum like spiral bacterium (Bacterium-19176-20-LI). In Saad’s post on Paramecium, we see this long Beggiatoa-like bacterium along with Paramecium. Just like in the video below.


Now in this pond water derived system all the species were observed in the same cross-section of space.  Some might be chance associations, others might have  various levels of dependencies on the state of resources in the environments and other species. Over multiple such observations could we now create ecological networks based on co-occurrence of species, where the nodes are the species and the edges are co-occurrences. The thickness of the edges is proportional to the number of co-occurrences.  For example, if I see Paramecium and the long chained Beggiatoa like bacterium this will have a co-occurrence score of 1. Now if I see it again (as I think I do in Saad’s post), then it will have a co-occurrence score of 2 (a thicker edge). Over multiple observations, we will start noticing the strongest associations and eliminate the ones observed by chance. Thus we could build an ecological network. A very basic (and rather unaesthetic) version is below (Click on image for high-res). Its an all-vs-all network, note the thicker Paramecium-bacterium edge.


P.S. My aquarium lasted more than a week on this occasion. Paramecium was still seen after 3 days and Oxytricha after a week.

4 Comments Add yours

  1. Manu Prakash says:

    Incredible post @Laks.

    1) I absolutely love that you have tracked “organism ID”. I am thinking of this as a simple feature in a Foldscope app (whenever we make an observation; it stamps an organism unique ID number – with date of observation and some info from the user to make a unique identifier in video data.

    2) This will be a phenomenal way to have communities engage and go beyond individual identifiers. I will ask several people what tools they would suggest for mapping tasks; but I found a few open source ones after a little searching:
    Package foodweb in R –

    I know Eric Berlow and will ask him anout tools – but here is a recent paper from him:

    As a community – we should have specific posts on sharing online tools to understand the big picture from the “microscopy” data. Specially automated tools that can take our posts and map them in interesting network manners.

    3) Finally – i know you will enjoy reading this. Noble prize for Keystone species concepts – you will love reading this NY Times piece:

    Love the bacteria video – and the “jump” cell that seems to crawl. The sound effect is ire 🙂

    Finally, sorry for being away from posting for a few weeks. The twins have really broken my foldscope posting schedule – so I need to come up with some rules. But I have some fun surprises on the plate.


  2. laksiyer says:

    Dear @Manu. A foldscope app that would reduce posting time would be absolutely awesome. The id scheme is one we had planned in a previous post, but we could change it to a universal scheme once you have the app. Finding ways to use the foldscope data for meta analysis would be fantastic. I just saw a post by Wolfram on Wolfram language . If we can mold something like it for our problems that would be fantastic (Like how Mathematica can do). I am going to try all the network tools you recommend for my own work. I use cytoscape, but I am looking for something more aesthetic. I might have to turn to R after all. The Keystone species is a phenomenal idea. I wonder if it is the duckweed in my case.

    Get some sleep when you can.

  3. Janice says:

    Hi Laks. Great to see your diatom pictures! You are correct about oil droplets in the diatom. This is a unique feature that provides nutrition to the diatom, and enables buoyancy. Over geologic time diatom oil converted into the fossil fuel deposits we extract today (see:

    A very nice video of a pennate diatom! I don’t think I would call this a Pinnularia since this taxon is not very tolerant of sulfide-rich environments. Two diatom genera that are found in such conditions are Navicula and Fallacia. For example, take a look at: I think you will see the resemblance in shape to Pinnularia.

    The colonial form in the first picture is a number of diatoms on their sides (known as girdle view in contrast to being face up or in valve view). Can’t tell the shape of the valve, so it’s more difficult to give this diatom a name. I would guess that it might be Fragilariaforma virescens ( Same is true for the square diatom; it’s on it’s side, but might be the same. (Perhaps a Foldscope with an oil immersion lens furnishing 1200 to 1400X will be available some day to resolve these more difficult identifications!!)

    I can see why you think the pennate diatom pictured is an Amphipleura. I don’t think this is so for the same reason as Pinnularia. Amphipleura is found in oligotrophic conditions; it can’t survive in sulfide-rich environments. I would say that the taxon is probably a Navicula salinarum (see:

    The other Navicula you have identified is probably right. It could be another N. salinarum at a different stage in the vegetative reproduction cycle, or some other Navicula.

    Also, about the keystone species concept… Duckweed is not a keystone species. It’s considered to be an invasive in some areas and undergoes a “boom” or “bust” reproduction cycle, depending on the physical and chemical conditions of the ecosystem. Like diatoms, duckweed has been used as indicator taxa in toxicity tests to assess ecosystem health. One review of the literature was published in 1990, but be aware that the family for duckweed is the Araceae, not the Lemnaceae ( Also, diatoms have been found in the phyllosphere of duckweeds:*~hmac=89994a8f921efb8cb059f9ae43bb186477e1c0275a956f52a246f3db97e3b43a.

    One more thing… The keystone species concept is controversial. See: Ecologists debate the concept all the time. Another concept you might be interested in concerns trophic cascades in freshwater (e.g., Ecosystem ecology is a vast, interesting field!

    Great work, Laks! I look forward to more of your contributions!


  4. laksiyer says:

    Thank you so much for your comments @Janice. This is exactly why we need experts like you on the list. Morphology is only the first step, but you have to take into account so many other factors when you explore a ecosystem as you point out. I am going to leave my annotations in as people can compare my annotation to your nuanced comments to see how to approach this. I will of course incorporate the likely annotation in my database. One of my plans is to make large networks where one can start quantitating various network parameters. Thank you for the articles, am reading them now.

Leave a Reply