Children and young people ranging in age from 5 months to 21 years, with burns covering greater than or equal to 30 per cent of the body surface area, were entered into a prospective study comparing the resuscitative efficacy of hypertonic lactated saline (HLS) and Ringer's lactate-colloid (RL-colloid). The 24- and 48-h fluid requirements of children resuscitated with the RL-colloid were significantly greater than those resuscitated with HLS (P less than 0.02); 2.37 +/- 0.91 ml/kg/per cent burn (HLS) v. 3.43 +/- 1.51 ml/kg/per cent burn (RL-colloid) at 24 h post-burn and 4.18 +/- 1.37 ml/kg/per cent burn (HLS) v. 6.32 +/- ml/kg/per cent burn (RL-colloid) at 48 h (P less than 0.01). The urine output between the two groups was not significantly different, and the haematocrits were equally maintained. Patients in the RL-colloid group gained significantly more weight at 48 h (P less than 0.05). The sodium requirements and resulting sodium balances were not significantly different. The desired and significant elevation (P less than 0.001) of serum sodium in the HLS group was maintained for the 5 days of the study. Colloid oncotic pressures, serum albumin and albumin/globulin (A/G) ratios were not significantly different between the two groups until 48 h post-burn. The RL-colloid group received plasmanate during the second 24-h period. The significant elevation in serum albumin, A/G ratio and colloid oncotic pressure persisted for only 2 days--by 96 h post-burn these values were no longer significantly different.(ABSTRACT TRUNCATED AT 250 WORDS)