Research projects

Most plants require a common set of resources that include water, light, N, P, K and a dozen micronutrients. Interspecific competition for resources is the norm in plant communities, so how can 20 species coexist with one another in a square metre of meadow, or 25 species in 1/16th of this area of chalk grassland?

I have been searching for answers (there is almost certainly more than one answer) to this question for more than 25 years and it is one of the themes of my book Demons in Eden: the paradox of plant diversity. Although at first I rejected the idea that plants have niches (Silvertown & Law 1987) because they were so difficult to find, we made a breakthrough in 1999 (Silvertown et al. 1999) that completely changed my mind (Silvertown 2004). Using methods of niche measurement for plants in wet meadows devised by my colleague David Gowing, we discovered that species segregate along hydrological gradients. An example for buttercups (Ranunculus spp.) is illustrated below.

Now that we have a way to measure plant niches, there is a host of questions to answer.

1. Foremost is the question, 'how widespread among different plant communities is niche separation on hydrological gradients?'. We are currently investigating hydrological niche separation in fynbos in the Cape Floristic Region. This was successfully begun under a

   Visit our fynbos website for the latest details

   research grant from the Leverhulme Trust and is now being extended to cover the conservation of fynbos under a Darwin Initiative grant. Next, we want to know how hydrological niches evolve. The evidence, so far as it goes at the moment (see Silvertown et al. 2006), is that hydrological niches are evolutionarily labile. We are investigating this in the fynbos too.

The Park Grass Experiment (PGE) at Rothamsted, UK is the longest running ecological experiment in the world.

It was set up by Lawes and Gilbert in 1856 and consists of a hay meadow to which a series of fertilizer treatments have been applied annually. The original aim was to look at the effect on yield of inorganic fertilizers especially phosphate and compare this with the effect of organic manures. However, it quickly became obvious that there were also large effects on species composition and it was decided to measure the percentage of each species in the hay. This has been repeated at irregular intervals to the present day. I have worked on the PGE since 1978.

Recently, we have been studying the molecular ecology of sweet vernal grass (Anthoxanthum odoratum) in the PGE using choloroplast microsatellite and nuclear markers to look at genetic diversity, differentiation, gene flow and reinforcement. See

These three figures illustrate some of the most important community results from Park Grass.

They show how:

1. The application of different nutrients influences the relative ratios of three plant guilds: grasses (Poaceae), legumes (Fabaceae) and other broad-leaves.

2. Precise G:L:O ratios vary from year to year under the influence of climate.

3. Despite the perturbations caused by climate, the ratio remains around an equilibrium value that is characteristic for each treatment. This means that, at the guild level, the PGE communities maintain a dynamic equilibrium. How the community does this is an interesting question because the species within each guild change over time. Hence, the communities are hierarchically organized.

See the right hand panel for recent references and the publications page for a complete list.