Because coccolithophores are so small, I used scanning electron microscopy (SEM) to identify and count specimens. Even at the highest possible magnification, a light microscope cannot resolve many coccolithophore details, and the shallow depth of field at high power makes is impossible to focus on the whole cell. The magnification and depth of field available with SEM are about 100 times greater than with the highest power light microscope. An added benefit is that it was a straightforward matter to prepare the samples that had been filtered from water collected during research cruises.
To prepare the filters for SEM examination I first punched a piece out of each with a hole punch. Using double-sided sticky tape, I stuck the punched-out circle on an SEM stub, a solid aluminum cylinder about the diameter of a rivet on a pair of jeans. I promptly recorded each stub number and sample identification-a crucial step. Then, I sputter coated the stubs with gold-palladium, a conductive material that improves contrast in the SEM image and prevents buildup of electrons on the samples.
In a light microscope, glass lenses focus a beam of light on the sample. In an electron microscope, electromagnetic lenses focus a beam of electrons on the sample. The electrons are produced by passing a current of high voltage electricity through a tungsten filament.
As the beam of electrons scans the sample point by point, a detector collects some of the electrons that interact with and some that bounce off the sample. The number of electrons detected depends on the composition and topography of the sample. The captured electrons are converted to an electronic signal which in turn forms an image, point by point, on a monitor.
With the sample as close as possible to the electron beam and magnified 3500 times, the area displayed on the monitor screen was about 45 by 33 micrometers. The area of a filter is roughly 200,000 times that size. I once estimated that it would take about 30 years working full-time to count all the coccolithophores on all 169 filters from the two cruises.
I set a reasonable goal of counting at least 100 coccolithophores from each sample. I also limited the number of screens I would view to obtain the count. There were many more screens without coccolithophores than with them. Keeping track of the number of screens viewed enabled me to calculate how much of each filter I examined. Based on the species census from the area I viewed, I estimated the species concentration for each water sample.
Sitting alone in a darkened room counting objects that appear on a small black-and-white screen may seem tedious to some. To me, it is as exciting as exploring an alien planet. Once handling the controls and making tick marks on a data sheet becomes automatic, watching the screen feels like being there on the filter, rolling along, scanning a scene strewn not with boulders but with life forms, many exquisitely beautiful and as alien as anything we might imagine emerging from a UFO.
Acknowledgement: The scanning electron microscopy for this study was done at the Texas A&M Electron Microscopy Center. I wish to thank the staff there for the instruction and assistance they generously provided as I learned to use SEM instruments and techniques.
Scanning electron microscopy has the potential to create more questions than it answers. I was unable to identify these three specimens. Any ideas? (Photos by Vita Pariente)
