Unfractionated heparin (UF-heparin) has been shown to prevent antigen-induced airway hyperresponsiveness (AHR), but it is ineffective when administered after the antigen challenge. We hypothesized that the failure of UF-heparin to modify postantigen AHR might depend on molecular weight. We therefore studied the effects of UF-heparin and three low-molecular-weight heparin fractions (medium-molecular-weight heparin [MMWH]; low-molecular-weight heparin [LMWH]; and ultralow-molecular-weight heparin [ULMWH]) on antigen-induced AHR and histamine release in bronchoalveolar lavage fluid (BALF). Specific lung resistance (SRL) was measured in 20 allergic sheep before, immediately after, and up to 2 h after challenge with Ascaris suum antigen. Airway responsiveness was expressed as the cumulative provocative dose of carbachol, in breath units, that increased SRL by 400% (PD400). PD400 was determined before and 2 h after antigen, both without and after treatment with aerosolized UF-heparin (1,000 U/kg) and various heparin fractions (0.04 mg/kg to 5 mg/kg) administered after the antigen challenge. Inhaled UF-heparin (n = 4), MMWH (n = 4), and LMWH (n = 6) failed to modify postantigen AHR when administered after the challenge. Only ULMWH (n = 6) inhibited postantigen AHR in a dose-dependent manner (percent protection ranged from 31% to 139%). In eight additional sheep, histamine in BALF was measured with a radioimmunoassay (RIA) before and after the segmental antigen challenge, without and after pretreatment with inhaled UF-heparin, LMWH, or ULMWH. Inhaled UF-heparin and LMWH inhibited antigen-induced histamine release as measured in BALF by 81% and 75%, respectively; whereas ULMWH was ineffective in this respect. We conclude that: (1) modification of antigen-induced AHR by fractionated heparins is molecular-weight dependent; and (2) only ULMWH attenuates AHR when administered after antigen challenge, via an unknown mast-cell-independent action.