30 
W. Pengelly on the Red Sandstones, 
in the same stratum, they cannot be due to inequahties in the 
weathering of alternate beds. Where first observable they are 
confined to a very limited vertical zone, but further eastward they 
are of frequent occurrence and extend from the base of the cliff 
far up towards its summit. 
Between Salcombe-Mouth and Weston-Mouth — about one 
mile — the marl is more mottled than elsewhere east of Sidmouth ; 
it differs, however, from that further west in being traversed by 
plates of selenite or transparent crystallized* gypsum. This sub- 
stance is first observed immediately east of Salcombe-Mouth, 
where it occupies a zone of about twenty feet in height, but which 
rapidly developes into much greater dimensions. Taken in large 
masses the plates vary much in thickness and are frequently 
somewhat curved ; each of them consists of two laminae, which 
were probably formed by the simultaneous deposition of films of 
sulphate of lime on the opposite walls of a joint, until by con- 
tinued accretion the interspace was completely filled. The fracture, 
however, is fibrous, the fibres being perpendicular to the surfaces 
of the plate. Each lamina is formed of a series of stellate crys- 
talline polygons measuring about two and a half inches by one 
and a half. The plates resolve themselves into three systems or 
series, one set being approximately horizontal and the other two 
sensibly vertical ; of the latter, one system takes what may be 
called an east-and-west, and the other a north-and-south direction. 
Bifurcations are of somewhat frequent occurrence. If the selenite 
be supposed to occupy what were previously joints, we must con- 
clude that it was not prepared to take possession of all that were 
in existence, since it sometimes occurs that the continuity of an 
open joint is interrupted, but not terminated, by one of the plates. 
Had such joint been formed subsequently to the deposition of the 
selenite, the latter, being remarkably brittle, would necessarily have 
been fractured as well as the marl surrounding it, This substance 
traverses the cliff from the point just named to Branscombe-Mouth, 
a distance of four miles. It varies considerably in amount, luul 
