The correct selection and sampling of suitable wood is essential for accurate dating results.
The following has to be taken into consideration when sampling:

• Samples should not be taken from sections where the wood shows signs of tension or compression (reaction) e.g. root or branch sections
• Single samples should have at least 80 annual rings.
• As many samples as possible should be available, should analyses of one particular construction period be required. Only a well established chronological series of annual ring patterns can guarantee exact crossdating.

To determine the date a tree was felled ("the cutting year") the terminal ring of the last year of growth, must be there. If the terminal ring is missing, but sapwood is still available, we can estimate the "felling date" through experience. Adding the number of missing rings is known as sapwood dating.

Sampling methods used in dendrochronological dating:

• Sampling using an increment borer. A core sample of between 5 to 25 mm in diameter, according to the type of bore used, is removed from the wood.
• Sampling using a magnifying measuring lens: the measurements of the annual ring width pattern are taken directly from the sample. The annual ring widths are then copied on a 1:1 scale onto transparent film, or a blueprint is taken.
• The annual ring sequence of the sample is photographed and subsequent measurements taken from the photographs.
• X-ray photography (computertomography) can be used for samples where the annual ring sequence cannot be seen eg. in paintings or sculptures.