In July of 1987, we planted eight 30-cm-tall sour orange tree seedlings in a field of Avondale loam at Phoenix, Arizona and enclosed them in pairs in clear-plastic-wall open-top chambers. Since 18 November of that year, we have continuously pumped ambient air of ≈400 ppmv [CO2] through two of these enclosures, while through the other two we have continuously pumped air of ≈700 ppmv [CO2]. By the end of the second year of the study, the trunk plus branch volume of the [CO2]-enriched trees was ≈2.75 times greater than that of the ambient-treatment trees. Three years later, this factor had dropped to ≈2.0; but the decline in the [CO2]-enriched/ambient-treatment ratio of trunk plus branch volume was nearly perfectly offset by the relative fruit production advantage enjoyed by the [CO2]-enriched trees over that period. In Years 6, 7 and 8, however, there was a moderate drop in total productivity enhancement. This decline may be a delayed acclimation response, or it could be due to enhanced self-shading in the [CO2]-enriched trees or to the fact that, starting early in Year 6, many branches of the [CO2]-enriched trees grew all the way to the walls of their enclosures, so that many blossoms and young fruit were destroyed by intermittent physical trauma produced by the action of wind against the taut plastic in that year and in all succeeding years. Hence, we will have to maintain our experiment for several more years for this lateral growth obstruction to occur to the same degree in the ambient-air chambers as it has in the [CO2]-enriched chambers, in order to determine the long-term equilibrium effects of atmospheric [CO2] enrichment in a spatially confined environment.