top of page




The inspiration for this project is totally unrelated to Grasshopper, Kangaroo, or even design in general. Our final assessment for 12.007 (Geobiology), which I took in my Freshman Spring, consisted of writing a blog post about an interesting, current topic in geobiology. I was unsure of what I should do until my advisor asked me what it was about the class that had most interested me. I had an answer for her almost immediately. While the class was very interesting overall, I was absolutely fascinated by some pictures they had shown us briefly in one of the lecture presentations. 

The organisms in question are called coccolithophores. These are tiny little organisms with a profound impact on the global ecosystem. What made them fun for me was that they are gorgeous — in contrast to the general bacterium, they make plates (coccoliths) out of calcium carbonate (CaCO3), which aggregate in a random, three-dimensional pattern to form what looks like a suit of armour. The coccoliths that are shed by the organisms are actually what make up the carbonate oozes covering 35% of the oceans, which can get shifted above the waterline to make remarkable chalk formations like the white cliffs of Dover. 

In short, I got very distracted by the geometry of the coccolithophores, and spent as much time attempting to model them as I did writing the actual blog post.


Coccolithophores come in a range of shapes and sizes. The most common, by far, is Emiliania huxleyi. Coinciedentally, it is also my favourite, and that is the species I ultimately modelled.

Emiliana huxleyi.png
Types of Coccolithophores.png
Coccolithophore Bloom.jpg

Packing Arrangement

I could have simply distributed points on a mesh and built coccoliths around those, but that would not have replicated the beautiful self-organising structure of a true coccosphere. Since the coccoliths are produced one by one, and pushed out from the interior of the cell in arbitrary locations, the scales have a three-dimensional aspect to them.

3D Arrangement.jpg

I resolved this issue using Kangaroo's solid-packing feature. First, I created two concentric mesh spheres, distributing points on the outer one. I used these as centres for virtual spheres, which were then attracted towards the surface of the inner mesh. Since the points can't come within a certain radius of each other, they arranged themselves randomly. I then built oblongs around each of these points, representing the coccoliths. This arrangement can be reset every time the program is run. Additionally, by determining how many points are generated, coccoliths, I can create countless variations of the packing, thereby emulating nature.

Screenshot 2019-04-26 at 9.43.11 PM.png
Screenshot 2019-04-26 at 9.43.52 PM.png
Screenshot 2019-04-26 at 9.43.29 PM.png

Cross-sections showing effects of increasing the number of points used in generating packing arrangement.

One aspect of this program which is not wholly accurate is that the coccoliths intersect with one another. This does not happen in real life, but was necessary if I intended to 3D print them.

Screenshot 2019-04-20 at 3.30.58 PM.png
Screenshot 2019-04-20 at 3.31.22 PM.png

Initial attempt, with single coccolith selected to demonstrate modularity


The coccoliths were also built in Grasshopper, but without Kangaroo. My initial versions featured a solid disc, while the final edition had ribbing both on the lower and upper shields. My modelling process was based off descriptions of the production of real-life coccoliths, which was also quite fun to do.

Screenshot 2019-04-26 at 9.06.34 PM.png

Modular unit

Changing variables to affect output geometry

Screenshot 2019-04-26 at 9.08.28 PM.png

Final Arrangement

190426_Rendered Coccoliths.jpg


Due to the nature of the powder 3d print, the shape crumbles when placed under pressure. It returns to dust in the same way true coccoliths do, which leave behind chalky powder when they die.

Bonus Kangaroo

Since I was working with cells, I reckoned I could attempt to model the mitosis of coccolithophores as well as the static shape. I developed a way to emulate mitosis, a video of which is shown below. It would be relatively simple to replace the spherical mesh used for these models with the dynamic mesh of the mitosis, but my computer couldn't handle processing it. Thus, I leave it as an exercise to the reader to imagine what it could look like.

bottom of page