156 &. WILLIS—DRIFT PHENOMENA OF PUGET SOUND. 
An unconformity by erosion is recognized and is used to distinguish 
formations of the Vashon glacial epoch, above, from those of an intergla- 
cial epoch, called Puyallup, below. 
The formations of the Puyallup interglacial epoch are stratified gravels 
and sands, being deltas and relatively deeper water deposits. They ex- 
hibit synclinal and faulted structures which are attributed to the melt- 
ing of included ice-masses. They are also correlated with widespread 
and thick lignite beds. From these facts it is inferred that the gravels 
-were superglacial and bore a forest growth over stagnant ice, as is now 
the case with the marginal moraines of the Malaspina glacier. The sands 
overlying the gravels accumulated in deep waters, probably ponded by 
the northern ice: 
Beneath the city of Tacoma is a complex mass of gravels exhibiting 
plunge structure and faults. The deposit is attributed to the outflow 
from a lake which covered Steilacoom plains as the ice retreated from the 
immediate vicinity. The delta formed during the Puyallup interglacial 
epoch and was added to during the withdrawal of the Vashon ice. 
The lowest deposits observed in the southeastern part of the Sound 
basin are blue clays, in some places stratified, elsewhere structureless and 
containing many rounded and angular stones. These are both consid- 
ered to be formations due to the spread of an ice-sheet earlier than the 
Vashon, from which it is separated by the Puyallup interglacial epoch. 
This earlier glacial epoch is named from Admiralty inlet, and its forma- 
tions are designated the Admiralty clays, stratified, and the Admiralty 
till. The Admiralty epoch was one of general glaciation in the Sound 
basin. Itis possible, if not probable, that it was preceded by other epochs 
of interglacial and glacial conditions,which further study may distinguish. 
The classification of epochs and formations herein described is pro- 
visional and may be modified by further detailed observation. The 
paper is published to afford a working hypothesis for those who may 
enter this field of complex drift phenomena. Its basis in observation is 
not broad enough to give it conclusive value. 
