220 ME. S. H. WAREEN OlS" A LATE GLACIAL [Juiie I912, 



and Holocene ; but they fail, at least when applied upon the rough 

 and ready basis of two species of elephant, in respect of the difference 

 between pre-Glacial and inter-Glacial. 



Upon the average, Eleplias antiquiis may be earlier than E. 

 primigenius. but this rule is not of universal application. Occa- 

 sionally the two species occur together in contemporary association. 

 It is more essentially relative climate than relative age that these 

 two species represent. 



Ele])has antiquus may be taken as representative of the southern 

 fauna, and E. primigenius of the northern fauna. As has been 

 frequently emphasized by Dr. Boyd Dawkins. the boundary between 

 these two zoological provinces fluctuated northwards and south- 

 wards according to secular, and even seasonal, climatic changes. 

 Thus the same ground may be occupied by both northern and 

 southern animals within the circle of a single year. 



At the same time, it may be worth while to recall that Leith 

 Adams has shown that at Ilford, where Ele-phas antiquus and 

 E. pyrimigenius occur in contemporary association, the latter species 

 is not of the Arctic form.^ The Ilford variety of E. primigenius 

 is smaller than the Arctic form, and possesses certain peculiarities 

 in the structure of its teeth. How far this might apply to other 

 localities, I will make no attempt to discuss. 



In any case, it must be conceded that the two species mentioned 

 do not in this country represent a well-defined sequence in time. 

 But, when we come to the human industries, I know of no authenti- 

 cated instance in which these do not occur in the same invariable 

 order of stratigraphical succession. It is true that all the stages 

 that are found in Trance have not hitherto been recognized in this 

 country, where the material at our command is so much less ex- 

 tensive. In spite of this, and although there may be, as one would 

 expect, local differences, the general order of succession is the same 

 in the two areas. 



This conclusion will not pass without dispute, but my own 

 experience leads me to the belief that it is justified. It would not 

 be fitting to enter into a full discussion of these wide problems in 

 the present paper. But, as the question of correlation is of so much 

 importance, it may be desirable to give some brief outline of the 

 Palaeolithic succession of the Stoke Xewington district, which has 

 been so ably worked out by Mr. Worthington Smith,^ especially as 

 this is so nearly related to the lower part of the Lea Valley with 

 which we are dealing. In the river drifts which lie between the 

 surface-levels of 70 and 100 feet O.D., the earliest implements 

 found consist chiefly of rude ovate forms, although a smaller 

 number of pointed types also occur. These are typical of the series 

 coming in to the ' sequence-dates ' of 7.30 to 7.40,^ although their 



^ A, Leith Adams, ' The British Fossil Elephants ' pt. ii, p. 79 (Monogr. 

 Palaeont. Soc. for 1879). 



2 ' Man the Primeval Sarage ' 1894. 



3 The ages that these sequence-dates represent can [be gathered from the 

 table (p. 223). The use of this system possesses many practical advantages, 

 although one cannot enter into these in the present paper. See ' On the Value 



