vol. xx. (2) NOTES ON COTTESWOLD-MALVERN REGION in 
In a former paper (68, p. 16) I have suggested that the 
River Teme, instead of following a course from Knightsford 
Bridge to its confluence with the Severn below Worcester, 
formerly flowed through the valley on the west of the Malverns 
to a junction near Gloucester. Most of the deposits in this 
valley appear to be the rearranged remnants of the gravels and 
sands of the old river, and it is observed that the proportion 
of one of the constituents, the Quartzose pebbles, which are 
persistent in these gravels, increases towards the north. They 
were probably transported from Permian and Triassic rocks 
in West Worcestershire by an ancient tributary flowing across 
the present Severn Valley from the direction of the Stour, 
and joining the Teme on the north of the Malverns. This 
hypothesis, however, necessitates some modification of former 
ideas as to the position of the main streams of the region when 
Drift pebbles were being carried into the Teme Valley, for any 
river flowing in the present course of the Severn would separate 
the Teme from the Permian and Bunter areas unless they then 
extended farther to the west. Whether a former connexion 
was severed by the cutting back of a stream along the course 
of the Lower Severn, or whether the main stream in the Plain 
was, at the period in question, only a southerly continuation 
of the Avon and flowed farther to the east than at present, 
can only be conjectured. The changes above outlined were 
mere phases of the long Tertiary denudation that resulted in 
the carving out of the Severn Plain, of which our knowledge is 
too limited for the solution of the problem. The time-element, 
however, presents no difficulty, though the transport of the 
Quartzose Drift into the Malvern -area must have come to an 
end before the great ice sheets invaded the Midlands and 
scattered Arenig and Irish Sea erratics, and other Glacial 
Drift, over the Permian and Bunter areas of West Worcester- 
shire. Otherwise the two kinds of Drift would have been 
carried together into the Teme Valley, where no far-travelled 
Glacial Drift has yet been observed. Moreover, post-Pliocene 
time does not appear to have been sufficient for the lowering 
of the valley on the west of the Malverns to its present depth 
below the position of some of the gravels. Two small brooks, 
4 
