Saddle-splay coefficient

The droplets shown in Figure 5.3 present in fact a double twist rather than a simple unidirectional twist. Double twist is discussed below in relation to the saddle-splay coefficient. 

One might wonder why the most usually observed focal domains are not those where the singularities are still further reduced, i.e. one focal sheet reduced to one point, the other one being sent at infinity (in such a case the layers are concentric spheres). These spherulites are indeed observed, but only when the saddle-splay coefficient is favourable,... 

However, there is no physical reason why the coefficient 24 itself should be exactly zero, and therefore in special geometries the saddle-splay contribution must be taken into consideration. The 24 term is nonzero for director configurations containing stable point defects [52]. For example, it influences the director fields in cylindrical [43, 44, 178, 194, 201-203] and spherical [47,204,205] cavities, in nematics confined between concentric cylinders in bend geom-... 

---