Dr Shanta Kakri of IRRI supports the plan for two celled C4 rice. Maize already got an HI of 50% higher than rice but it is disliked for food in Asia. Journal of Experimental Botany, Page 1 of 10 doi:10.1093/jxb/err022

Crops can be divided into two broad categories in terms of the way in which they use photosynthesis to convert sunlight and water into carbohydrates.  In C3 plants carbon dioxide is first fixed into a compound containing three carbon atoms while in C4 plants carbon dioxide is initially fixed into a compound containing four carbons atoms.This seemingly minor variation in photosynthesis makes an important difference: C4 plants are around 50% more efficient than C3 plants, and despite accounting for just 3% of plant species, C4 plants contribute 30% to terrestrial productivity.

The world faces pressure from a growing population, and productive land is increasingly at a premium. One way to improve yields without cultivating more land is to engineer crops to use C4 photosynthesis. To do this, scientists must understand the evolutionary steps that lead from C3 to C4 photosynthesis.Most C4 plants are native to tropical and sub-tropical regions.  They include maize, sorghum, millet and sugarcane.  Rice, a crop vital to the nutrition of huge numbers of people across the world, is a C3 plant, which explains why there is a big push to develop a C4 rice with enhanced yield.All C4 plants evolved from C3 plants. Scientists think that this process took place over many millions of years. No one knows exactly what causes the sequence of changes that makes it possible for plants to learn this trick, and although the C4 pathway is considered highly complex, this system has evolved independently in many groups of plants.

Williams assessed the presence or absence of 16 traits known to be important for the C4 pathway in 73 different plants, some using C4 photosynthesis, some using the C3 pathway, and others that seem to use a blend of both C3 and C4. Johnston then developed Bayesian modelling techniques, to produce a model that predicts the steps associated with this highly complex evolutionary process. The model was underpinned by data occupying a 16-dimensional space with 65,536 nodes within that space...Some interesting facts..

For more information on this story contact Alex Buxton, Office of Communications, University of Cambridge, [email protected] 01223 761673