Some of my earliest biology memories are as a 9 year old at Balclutha Primary School working through food web diagrams. The idea of the interconnectedness and interdependence of life was an extremely powerful idea and I recall the first afternoon we worked on food webs more vividly than most in my primary years (although the day Rolf Harris came, sang and painted a picture was pretty cool; and then there was the great pea-shooter battle one afternoon – but I digress). 
Of course at the time it was comprehending that the big things eat the little things who eat the planty things that really sank in. Suddenly the natural world around me started to make sense in a cool and interesting way. From what I see from my own sons’ passages through primary school, the whole food web activity still seems to exist, hopefully influencing future biologists. Which is great, as there is still much to be done on understanding how food webs work. Hannah Buckley (Lincoln University) and colleagues from around the USA have worked on a very simple food web that can be found all around North America, the fluid filled pitcher plant, Sarracenia purpurea. The long-lived (> 50 years) carnivorous plant grows in bogs, sand plains and pine savannahs across much of North America Within this carnivorous plant lives a small community that typically contains 6 arthropod species, 9 protozoans and 17 bacterial species. The beauty of the pitcher system is that the food web is about as simple as you can find, the links between the different species is reasonably well understood and, more importantly, the system is replicated in thousands of sites at a continental scale. That makes for a great natural experiment which they investigated earlier by looking at the variation between populations. In the present study they wanted to see what effect temperature, rainfall, atmospheric nitrogen availability as well as the abundance of the keystone species (the pitcher plant mosquito, Wyeomyia smithii) had on food webs.
The team collected data from all over North America. At the level of individual pitcher plants, food web complexity was proportional to the volume of liquid in the plant – the more liquid, the more complexity (meaning the pitchers act just like little islands). 
At the scale of sites over North America there was a less clear cut picture of the pitchers. The further north the site was (moving to higher latitudes) the less variation there was is the various traits of the pitcher food web but, overall, food web structure was not driven very much by climate variables like rainfall and temperature and it seemed that food webs at sites were often built simply by which species historically happen to have arrived in the site in which particular order. This finding is a bit unexpected as it suggests that random processes play an important role at the larger scales. So there’s plenty more work to do on this great little model system. There is something quite satisfying in finding that food webs are still as cool and interesting as that day long ago in South Otago.