Born in Lawrence, New York, Stebbins studied biology at Harvard where he obtained his PhD in 1931. After working at Colgate University he moved to Berkeley in 1935 and to Davis in 1950, where he established the department of genetics, holding the chair until his retirement in 1973.

In his Variation and Evolution in Plants (1950) Stebbins was the first to apply the modern synthesis of evolution, as expounded by Julian Huxley, Ernst Mayr, and others to plants. In collaboration with Ernest Babcock, Stebbins also studied polyploidy – the occurrence of three or more times the basic (haploid) number of chromosomes. When an artificial means of inducing polyploidy was developed Stebbins applied it to wild grasses and in 1944 managed to establish an artificially created species in a natural environment. He also used the technique to double the chromosome number of sterile interspecific hybrids and in so doing created fertile polyploid hybrids. Fertility tends to be restored in polyploid hybrids because the two different sets of chromosomes from the parent species will each have an identical set to pair with at meiosis and so the formation of gametes is not disturbed. Polyploids have proved extremely useful in plant-breeding work. Knowledge of naturally occurring polyploid systems has also helped greatly in understanding the relationships and consequently in classifying difficult genera such as Taraxacum (the dandelions).

Stebbins also studied gene action and proposed that mutations that result in a change in morphology act by regulating the rate of cell division in specific areas of the plant.