Although it is generally assumed that small arteries are truly isometric when studied in a myograph, these arteries significantly shorten during the normalisation procedure. We assessed the effects of this shortening on calculation of morphological variables in small arteries using vessel length before and after normalisation. Subcutaneous resistance arteries were dissected from gluteal biopsies obtained from normotensive subjects (n = 7, supine pressure 116/69 +/- 5/3(SE) mmHg, internal artery diameter 276 +/- 16 microM) and patients with essential hypertension on no treatment (n = 13, supine pressure 160/102 +/- 4/2 mmHg, internal diameter 256 +/- 9 microM) mounted on a small vessel myograph. Using the actual shortened 'normalised' length compared with measurement using the original artery length, media thickness was significantly greater both in normal subjects (24.0 +/- 2.7 microM vs. 22.7 +/- 2.7 microM, 2 P < 0.05, paired t-test) and in hypertensives (27.6 +/- 1.5 microM vs. 26.5 +/- 1.4 microM, 2 P < 0.001). Calculated from the actual shortened 'normalised' length vs. original artery length, media:lumen ratio was also increased in the normal subjects (8.6 +/- 0.6 vs. 8.2 +/- 0.6 2 P < 0.05) and in the hypertensives (11.2 +/- 0.7 vs. 10.7 +/- 0.7, 2 P < 0.001). For the media thickness in normal subjects this was an increase of 6% (95% CI: 1-12%) and in hypertensives of 4% (95% CI: 2-6%). Standard methods for study of small arteries underestimate media thickness and media:lumen ratio in normal subjects and in hypertensives.(ABSTRACT TRUNCATED AT 250 WORDS)