This paper looked at 63 sagittal split ramus osteotomy sites. The type of split was classified according to the Plooij paper and bone measurements were taken at four sites. Of these sites, the thickness of the bone in one point, 1.5mm above the lingula, was the only site where there were statistically significant split patterns. The mean thickness for a bad split was 5.09 +/- 1.61mm compared to a good split of 6.33 +/- 1.04mm. They concluded that the percentage of ideal split patterns is high and that mandibles with thinner rami are more prone to bad splits. They acknowledge that a multifactorial mechanism is responsible for undesirable split patterns.