THE PLIOCENE PERIOD IN WESTERN ESSEX. 263
falls below 200 ft., at one or two spots (190-195), but not suffi-
ciently to invalidate the idea of a 200-ft. floor. However, in view
of the relations of the Essex valleys, presently to be noted, it is
probable that the stream had intrenched itself to some extent in
the 200-ft. floor before the advance of the ice. This leads us
to observe that the minor valleys entering the trough from north
and south are demonstrably pre-glacial in several instances.
Sherlock11 has noted a clear case of diversion by glacial plugging
at Codicote in the Maran valley. It has been shewn that the obse-
quent Colne has suffered glacial diversion at Welham Green,
causing it to vacate its former valley, now drift-filled, running
thence to Hatfield. Again, the valley of the consequent River
Rib was excavated in large part before the advent of glacial
conditions. This stream now enters the Lea at Hertford, flowing
southwards to Burleigh Heath and then making a marked bend
westwards through Wadesmill. The line of the upper part of
the valley is evidently continued beneath the Boulder Clay to
Ware, as shewn on the map (fig. 2), for a well-section reveals
the surface of the Chalk at 150 ft. O.D. just to the north of
the town, while close by, the sub-glacial plateau is at its normal
elevation of 220 ft.
The Lea Valley ice-lobe shews the same relations as that
in the valley of the Colne. The flat 200-ft. floor is present on
both sides of the main valley and in the valleys of the larger
tributaries. On the eastern side there is a well-developed
200-ft. bench at Sewardstone and Yardley Hill, to which the
writer has previously referred12 without, however, realising
its full significance. Small outliers of Boulder Clay rest on
its surface and since they are far removed from the larger
masses at Epping, Nazing, etc., they afford valuable evidence of
the extension of the ice down the valley. Southwards, the
continuity of the Sewardstone bench is broken by the high
ground of Pole Hill, Chingford, but on the further side of the
latter the feature reappears at Chingford Plain and Fairmead
Bottom, and constitutes a trough in the high ground leading
through into the Roding valley. It has been suggested13 that
the Roding formerly joined the Lea through this gap, a course
which appears feasible in the light of the numerous similar
11 Geology of the Country around Hertford (Mem. Geol. Surv.), 1924, p. 32.
12 Proc. Geol. Assoc., vol. xxxiv. (1923), p. 251.
13 S. H. Warren, Proc. Geol. Assoc, vol. xxi (1910), p. 454.
R