Vol. 56.] SHALES OF THE WELSH BORDERLAND. 371 



lithological character, and were naturally mapped as one formation, 

 while the Upper Ludlow is decidedly different in this respect, so 

 the main line of separation was drawn at its base. Nevertheless, 

 the Lower Ludlows may be at the present day readily distinguished 

 from the Wenlock Shales palaeontologically, for they contain a 

 perfectly distinct graptolitic fauna. 



My own researches have led me to the conclusion that the best 

 palaeontological division of these two rock-groups is one which 



(a) A group characterized by graptolites of the priodon-Flemingii-ty^Q 



(that is, Wenlock Shales) from 



(b) A group characterized by graptolites of the colonus-tyipe ( = Lower 



Ludlow). 



In this opinion I am supported by Miss E. M. E-. Wood, who has 

 been working for some time on the fauna of the Lower Ludlow 

 rocks. This classification is certainly of a practical working value ; 

 for, as a result of our field-work in many localities, we have both of us 

 always found that the Flemingii-fauna, dies away where the colonus- 

 forms become abundant. I would therefore propose to restrict the 

 term Wenlock Shale to the rocks containing the first of these 

 faunal groups, and the term Lower Ludlow to the rocks con- 

 taining the second. 



The present paper is concerned only with the Wenlock Shales and 

 their graptolitic fauna. My work on these Wenlock Shales has 

 extended over a considerable part of the Welsh Borderland, but I 

 have chosen three special areas for detailed work. 



The most important district appears to be that which lies round 

 the town of Builth. There the Wenlock Shales can be studied 

 from base to summit ; this area therefore is taken as the typical one. 

 There is no other district along the Welsh Borderland where the 

 rock-succession is so complete and so fossiliferous throughout, 

 though excellent confirmatory sections occur in the two other areas 

 that I have studied, namely, in the Long Mountain and in the 

 Dee Valley. In the Long Mountain, the basal beds of the Wenlock 

 Shales are not seen, while in the Dee Valley the upper part of the 

 series contains but few fossils, and the succession is complicated 

 by the coming in of the grits, which attain a great development 

 farther north. 



I have purposely avoided, for the present, those districts where 

 limestones or grits enter largely into the sequence, as my object has 

 been to study the succession of life where the Wenlock Series was 

 most typically developed. Now, the Wenlock is essentially a 

 shale-formation, though locally, as is well known, some of the 

 beds of shale are replaced at different horizons by beds of limestone 

 or grits. Also, in the localities where limestones or grits form an 

 important feature in the succession, the graptolites are fewer in 

 number both as regards genera and species, and a zonal classification 

 will prove to be a matter of greater difficulty. 



