The basis of all life on earth is primary production, or the growth of plant material as a result of the capture of sunlight. Natural landscapes have higher above ground production when they have higher rainfall, temperature, and concentration of nutrients in the soil. However, these landscapes very tremendously in how much of this production gets eaten while still living by animals such as caterpillars or cattle versus how much gets decomposed after plants die by animals such as earthworm or by bacteria in the soil. For example, in ecosystems like the Serengeti in East Africa, more than 50% of production is eaten and the rest decomposed, transferring energy and nutrients to the most abundant and diverse collection of large mammals in the world. In contrast, in other areas with similar plant production, such as oak forests and savannas in the Midwest, less than 5% of production is eaten and nearly 20% is not even decomposed. Large mammals are ten time less abundant and are represented by 90% fewer species. Scientists still don't understand how and why such "fates" of production vary across environments with different water availability, soil fertility and temperature. My work at NCEAS is to devise theoretical explanations for how plant nutritional characteristics, which heavily determine animals' willingness to eat or decompose plant leaves and stems, might vary with water, soils, and temperature. I will then review large available sets of data and published studies in the scientific literature to test the predictions of different explanations. The possibly vital linkage between the environment, plant nutrition, and the ability of animals or even bacteria to consume production may be the key to understanding why the earth's ecosystems work so differently and contain different types and numbers of species around the world.