10 
THE LOWER LIAS OF KEYNSHAM. 
all the lower beds of the Inferior Oolite are missing, as in the 
classical section at Midford. 
Of the passage from the lowest beds of the Lower Lias down 
into the B^hetic series beneath, we have obtained no direct 
evidence ; the section north of Bitton Station which exhibits 
the junction is no longer accessible. The Gotham marble has 
been recorded in the Gr.W.B. cutting at Saltford, ^ and it is 
probable that this characteristic bed is present at the base 
of the Lias throughout the district. 
Before proceeding tp give a detailed account of the Lower 
Lias in the Keynsham district, it will be necessary to explain 
the system of zoning which we consider most useful for 
explaining the faunal sequence here and in the neighbouring 
areas to north and south. 
It is immediately obvious that, however limited the area 
considered, no series of species could ever be discovered whose 
ranges are exactly conterminous, so that no two are ever 
found together, and so that no gap exists between the last 
appearance of one and the first entrance of the next. The 
very exactness of the requirements is sufficient to demon- 
strate the impossibility of their satisfaction. The most we 
can hope for is, that we may be able to find a series of species 
or groups, of common and general occurrence, whose periods 
of greatest abundance (“maxima ”) always succeed each other 
in the same definite order and such that there is neither a 
large interval between the times of occurrence of two succes- 
sive members of the series, nor any considerable period of 
overlap. As an example of a bad selection of a zonal series 
which is apt to lead to erroneous conclusions, we take the 
following abstract case. Suppose three species. A, B and 0, 
to be three selected zonal indices which, whenever they all 
three occur in the same locality, always have their maxima in 
1 Q.J.G.S., vol. xvi., p. 399, and Wright, Lias Ammonites, p. 36. 
