Siliceous and Delicious

I could easily have spent the remainder of the semester fretting about the details of my plans for the three projects on my platter for the academic year (diversity. rads, and morphospace, for short). It has been tremendously helpful to set a guillotine for the planning, and to put an emphasis on the recurrent, ongoing nature of the planning process—meaning it’s felt OK for me to put down project overviews that are distinctly lacking in details and subject to much change and adjustment. This way I’ve only wasted two whole working days on planning, rather than a whole month (see this time last year).

Here, the fruits of my labors (to be discussed in detail on Wednesday).

Diatom Diversity

Pre-Cenozoic Radiolaria

Diatom Morphospace

The major challenge in putting together the plan for the diatom morphospace project is that, as of yet, I have no idea what method of data collection I will actually use. In the scenario I discussed with Charles before I left for Australia, I would be making measurements on existing prepared slides ordered from the ODP archives using the CLSM. This was an entirely abstract conversation, and I have no idea how exactly nor whether this will actually work. If it does, it would allow me to capture 3-D co-ordinates describing the overall shape of diatom frustules, to be subsequently analyzed in some sort of dimensionality-reduced “morphospace”.

If this doesn’t work, an alternative is to use SEM images of some sort, but these are strictly 2-D images, and I have no vision of how I would capture overall 3-D information using that method. FIB-SEM allows physical slicing of specimens to obtain cross-sections of specimens, but this is incredibly time-consuming and expensive, and probably not an option for obtaining 3-D data of a large number of specimens in a large number of samples. It is, potentially, a good way of obtaining clear 2-D data from clean cross-sections through diatom frustules, which could be used to put together a dataset of 2-D metrics (shell thickness, pore size, etc), which is an altogether different sort of “morphospace” than (a) I envisaged initially, and (b) would be yielded by a successful implementation of the CLSM method described above.

As there is no way to predict which method, if any, will work, the only choice I have is to begin to try working with samples. Hence my gratitude for the planning deadline: the agonizing uncertainty about whether any of the methods I have thought up will actually work drastically sap confidence. The only thing that will actually push me forward in this regard is trial and error, samples in hand. So the sooner I can get samples and begin to mess around with them, the better.