A study of the precision of LAL and GC-MS analyses for endotoxin measurement was conducted by comparing the levels of endotoxin in pure LPS (Escherichia coli serotype 055:B5) solutions. The objectives of the study were to establish and validate the GC-MS technique analyzing specific 3-hydroxy fatty acids (3-OH 14:0 and 3-OH 16:0) as chemical markers of endotoxins, and to evaluate the correlation between the GC-MS technique and the LAL technique using samples of pure LPS (E. coli), pure Gram-negative bacteria (E. agglomerans), and a cotton dust which contains different types of Gram-negative bacteria and physical states of endotoxin. Calibration curves for the 3-OH 14:0 and 3-OH 16:0 (with 3-OH 13:0 as the internal standard) were constructed with concentrations injected into the GC-MS ranging from 2 to 46 ng (r²=0.9993 in both cases). Sample recovery of the 3-OH 13:0, 3-OH 14:0, and 3-OH 16:0 was determined to be an average of 80%, 82%, and 82% respectively. The ratios (GC-MS:LAL) of the levels of endotoxin found in the E. coli, E. agglomerans, and cotton dust samples were 2:1, 20:1, and 10:1, respectively. The limit of detection (LOD) of the GC-MS analyses was an average of 0.300 ng/ul which was approximately one thousand times greater than the average LOD of the LAL assays.