Market growth is fundamental to marketing. Frank Bass's seminal diffusion theory explains growth in new product markets. We develop an analogous theory for established markets exhibiting sporadic growth or intermittent declines. Our theory suggests that market participants repeatedly take successful and unsuccessful actions that cause them to change or to mutate in myriad and often unpredictable ways. The environment sorts these mutations, determining winners and losers. Abundant mutations often cause different market participants to become winners, displacing past winners. Abundant mutations also often cause market growth because the natural selection mechanism leaves more surviving favorable mutations. So one nonobvious falsifiable implication of our theory is that displacement precedes growth and stability precedes decline. Another is that risk taking, diversity of opinions, and experimentation should precede growth. We develop a metric for measuring displacement. Using multiple publicly available data sets (one including sales for top firms for 55 years and another including sales for all automobile models for 25 years), we find that our metric provides a practical way to measure the rate of mutation and confirm our theory's predictions. Our easily replicated tests show that our displacement metric can predict intermittent market growth or decline in very different contexts without the need for exogenous idiosyncratic explanations. Moreover, other alternative covariates (trends, lagged growth, new product entry, macroeconomic indicators, etc.) are unable to predict growth or decline.