Abstract
A study was made of (110) fracture in LiF using birefringence and etch‐pit techniques. Fracture always occurred parallel to birefringence bands, and it is suggested that the stress of these bands aids (110) crack propagation. Surfaces formed by (110) fracture appear to be flat. If (100) steps are present, as suggested by Kuznetsov, they are less than 100 Å in height. By comparing etch‐pit density data with birefringence data, it is estimated that only about one in twenty etch pits was due to single, unpaired dislocations. All the rest were probably caused by dipoles.