162 The American Geologist. September, L903 
necticut trough, South American arrivals by way of the In- 
diana basin, and slight invasions from the Dakota sea by way 
of Traverse straits. These three openings then remained in 
existence during the greater part of Upper Devonic time. 
Postscript. — Since the above was set up by the printer an 
important paper by H. S. Williams* has appeared. In this 
work the "shifting of fauna's" is discussed. In the main the 
author restricts his studies to> New York and Pennsylvania. 
His correlations and derivation o& the Devonic faunas are not 
at variance with, those df the present writer. 
THE DELTA-PLAIN AT ANDOVER, MASS. 
By F. S. Mills, Andover, Mass.f 
PLATES XXII-XXIV. 
The origin and structure of glacio-fluvitile plains, spread 
over the land in sub-aerial conditions, or formed in temporary 
lakes of the glacial epoch, have been well studied and fairly 
comprehended. 
The frontal plain and valley train melt into the surround- 
ing surface features, with no distinct external demarkation. 
In fact, their exact nature as plains would hardly attract atten- 
tion, except as a matter of geological study. The sandplain, 
on the contrary, rises with an abruptness above its surround- 
ings, that arrests the attention even of one not interested from 
the physiographers' standpoint. While it is true that the sand- 
plain is the result of ice streams plus glacial detritus, yet, the 
external conditions of deposition and the topographic result, 
are strikingly dissimilar to the other two types. The division 
line between the topographic forms they have left behind, as 
witnesses of former conditions, is very distinctly drawn. That 
such a plain, also, can only be of sub-aqueous deposition gives 
an individuality to it, not in conformity with the other forms 
of wash-plains. It would seem almost justifiable to give it an 
individual classification and not catalogue it with the others. 
The application of the term "sand-plain" genericallv to 
such sub-aqueous structures, is somewhat misleading. It is a 
* The Correlation of geological faunas. A contribution to Devonian pal- 
eontology. Bull. U. S. Geol. Surv., No. 210, June, 1903, pp. 1-147. 
t The writer is indebted to Prof. R. S. Tarr of Cornell University for advice 
and suggestions during the work on the Ballardvale delta-plain. 
