This paper reports the use of thin sections to improve precision and accuracy when ageing seedlings of Nothofagus pumilio. We compare 2 methods for determining the number of rings on 85 basal disks from seedlings: (1) ring counts of wood disks viewed with reflected light and (2) ring counts of thin sections viewed with transmitted light. Samples included 14 to 53 rings. Comparison of ring counts from the 2 methods revealed discrepancies of 1 to 12 y for 85% of the seedlings. Two sources of error were identified. In 70 of 85 samples, up to 12 incomplete rings explained differences between ring counts of 2 radii on the same wood disk. Secondly, the small radii, large number of rings, and diffuse porous nature of the wood resulted in frequent errors when visually detecting rings on wood disks. One to 3 false rings were detected in 15 seedlings. Narrow and suppressed rings in 57 samples resulted in under-estimates of 1 to 12 y for ring counts on wood disks relative to thin sections. High variation in the incidence of narrow rings limited our attempt to visually crossdate ring-width patterns among seedlings. However, counts of thin sections allowed us to identify false and incomplete rings, improving precision and accuracy when estimating seedling age. For studies of slow-growing seedlings that require high-resolution age estimates, we recommend the preparation and dating of thin sections as outlined in this paper.

Nomenclature: Muñoz-Schick, 1980.