As I was looking through the recently published papers in PLoS ONE (all open-access!), I came across an interesting article on the preservation of phytoliths. It is an interesting and well-written paper, and is quite accessible—both in terms of copyright and of science content.
Plants often have little bits of rock in them, called phytoliths (phyto- plant, -lith rock). Phytoliths are formed within the plant by precipitating SiO2 in a non-crystalline form (opal). These microscopic stones can help maintain the structure of the plant, perhaps among other functions. They also preserve well, because SiO2 (glass, essentially) generally doesn’t react chemically with much in the environment.
Just like with fossilized bones or impressions of leaves, the size and shape of phytoliths can be used to identify the plant (or family of plants) which is producing them. If phytoliths are found in the geologic or archaeologic record, they can be used to determine what kinds of plants were in the area, or were being eaten. They also contain small traces of carbon, which can be used for radiocarbon dating (back to ~40 ka) or 13C isotope analysis.
This paper is looking at what happens to various phytoliths in the archaeologic or geologic record, and whether there are preservation biases (some phytoliths being destroyed more easily than others).
The authors took samples of four different types of modern, living plants. These samples were then burned away in a 500°C furnace, leaving just ash and the microscopic rocky bits. With some further, relatively gentle treatment, they were able to isolate the phytoliths. Some of these phytoliths were mounted on microscope slides and counted to determine the relative abundance of different sizes and shapes.
Isolated phytoliths were partially dissolved for six weeks, and the Si content of the liquid was measured. The partially dissolved phytoliths were dried, mounted on microscope slides, and they too were counted to determine relative abundance of the different sizes and shapes after treatment.
Phytoliths which were small, and had a large surface-area-to-volume ratio, tended to be preferentially dissolved—this is not an unexpected result, but is important. The authors argue that based on the Si solubility, the degree of preservation can be assessed (high Si solubility means better preservation); in situations where the Si solubility is low, some of the more delicate phytoliths are likely to be missing, and a count of phytoliths under those circumstances would yield biased results.
But don’t take my word for it! Read the paper. It’s better written than my short explanation, and a fine example of scientific scholarship.
 Parr, J.F.; Lentfer, C.J. & Boyd, W.E. 2001, ‘A comparative analysis of wet and dry ashing techniques for the extraction of phytoliths from plant material’, Journal of Archaeological Science, vol. 28, no. 8, pp. 875-886. DOI: 10.1006/jasc.2000.0623
 Cabanes D. & Shahack-Gross R. (2015) Understanding Fossil Phytolith Preservation: The Role of Partial Dissolution in Paleoecology and Archaeology. PLoS ONE 10(5): e0125532. DOI:10.1371/journal.pone.0125532
 Looy, C.V.; Kirchholtes, R.P.J.; Mack, G.H.; Van Hoof, T.B. & Tabor, N.J. 2011, ‘“Ochoan” Quartermaster Formation of North Texas, U.S.A., Part III: First Sign of Plant Life‘ Geological Society of America Abstracts with Programs, Vol. 43, No. 5, p. 383.