Methods of estimation of spine density--are spines evenly distributed throughout the dendritic field?

Dendritic spines are small protrusions extending from the dendrites of nerve cells, which bear the majority of synapses. In the past, researchers quantified spine density as the number of visible spines per estimated micrometre of dendrite. This estimate ignores all those spines hidden from view due to their position on the dendrite. Dendrites vary in diameter and the underestimation in some will be greater than others. Estimation of dendritic length is also subjective and difficult in those which are tortuous. The Felman & Peters (1979) geometrical equation takes account of these criteria and provides a method of estimating 'true' spine numbers which does not involve slow and laborious reconstruction. This study compares ratios derived from both methods of estimation (spine density 2:1) at three loci in three experimental groups. Mean values of dendritic diameters and spine dimensions show the major cause for variation in the ratios between loci to be the shaft diameter of the dendrite. However, the greater ratio for apical as compared with basal and oblique dendrites is not as great as expected, bearing in mind that apical dendrites are approximately 2.5 times larger than oblique and basal dendrites. Therefore the spine distribution may not be the same throughout the dendritic field. Estimations of spine density based on visible spine counts are quicker, easier and sufficient for comparisons within the same locus. 'True' estimates (spine density 2) are more accurate and should be used when comparisons are being made between loci, cell types and species.