Despite some well-documented drawbacks, the acetylene reduction assay (ARA) remains one of the most widespread methods for measuring biological nitrogen (N₂) fixation (BNF) in symbiotic and free-living niches due to its low cost, simplicity, and high throughput potential. Because ARA measures a proxy reaction (the reduction of acetylene to ethylene by the nitrogenase enzyme), a conversion ratio (‘R ratio’) is required to estimate equivalent fixation of N₂. Based on the biochemistry of the reactions, the theoretical ratio is usually taken to be 3:1. However, ¹⁵N₂ calibrations often generate ratios that deviate considerably from this value. We synthesized calibrated R ratios for terrestrial BNF studies, asking whether values converge on the theoretical ratio and vary across N-fixing niches. From 253 mean values (n = 2,072 samples), we find that some niches (legumes, soil, litter) do center on 3:1, while others fall significantly above (wood, lichen) or below (biocrusts). Moss in particular shows a bimodal distribution that may indicate contributions from alternative nitrogenases. However, almost all niches have very wide distributions (up to 2 orders of magnitude); applying ratio values spanning even the 25th -75th percentile cause BNF rates to vary by a factor of 1.5–2.5, and up to > 8. Despite this, only a minority of studies (~ 30% of 345) perform calibrations, and this proportion has not increased over time. We conclude that high variability precludes the use of theoretical values to obtain accurate BNF estimates via ARA, and that historical data should be considered with appropriate caution. Values should be calibrated directly when the goal is to generate accurate rates or cross-condition comparisons.

尽管存在一些有据可查的缺点，但乙炔还原测定法（ARA）成本低，操作简便且价格高昂，因此仍然是共生和自由生境中生物氮（N ₂）固定（BNF）的最广泛测量方法之一。吞吐量潜力。由于ARA衡量的是一种代用反应（通过固氮酶将乙炔还原为乙烯），因此需要转化率（“ R比率”）来估算N _2的等效固定率。基于反应的生物化学，理论比例通常取为3∶1。但是，¹⁵ N ₂校准通常会产生与该值明显不同的比率。我们合成了用于地面BNF研究的校准R比值，询问这些值是否收敛于理论比值并且在固定N的小生境之间变化。来自253个平均值（n = 2,072个样本），我们发现某些生态位（豆类，土壤，垃圾）以3：1居中，而其他生态位则明显高于（木材，地衣）或低于（生物结皮）。苔藓尤其显示出双峰分布，这可能表明其他固氮酶的贡献。但是，几乎所有壁ni都有非常广泛的分布（最多2个数量级）。应用甚至跨越25％-75％百分率的比率值，都会导致BNF率变化1.5-2.5倍，并且大于>8。尽管如此，只有少数研究（约345％的30％）执行校准，并且该比例并没有随着时间增加。我们得出的结论是，高变异性排除了使用理论值通过ARA获得准确的BNF估计值的可能性，并且应谨慎考虑历史数据。