Correspondence. 391 
given, which, in itself, is a great service to the petrographer. These 
references of course are to all sorts of petrographical literature, but the 
book itself, in its treatment, involves only those characters which can 
be apprehended without the microscope. The glossary, embracing 
mining and petrological terms, also gives the names of many rocks that 
have been detined in America. n. h. w. 
Determinative Mineralogy and Blowpipe Analysin (Brush). Revised 
and enlarged by S. L. Penfield. New York, John Wiley and Sons 
1896. Oct., pp. 108. Price S3.50. Every American chemist and min- 
eralogist knows well the excellencies of '"Brush's Determative Mineral- 
ogy," for it has been in their laboratories for more than twenty years, 
and in constant use. This revision by Penfield, an expert and enthusi- 
astic mineralogist, has, inirabile dictu, increased its value. The chief 
change consists in the infusion of more chemistrj' into the treatment of 
the subject. There is a short introductory chapter which concisely 
sets forth the fundamental jjrinciples of the chemical nomenclature and 
combinations which are involved in the methods and in the tables. It 
is also noticeable that throughout the work, excepting in the tables, 
which are the same as in the last edition, there is considerable enlarge- 
ment. N. H. w. 
CORPxESPONDENCE. 
The Ice-Sheet in Glacial Narraganseii' Bay. Owing to my not 
seeing proof of my paper on "The Retreat of the Ice-Sheet in the Nar- 
ragansett Bay region," the following important features of the Barring- 
ton stage were not discussed in the proper place (see pages 150-168). 
Evidence was cited on p. 156 to show that in the case of the Green- 
wich Cove stage there was a change of water-level before the ice disap- 
peared from the head of the cove. To be exact, the water-level fell off 
from a hight at least 50 ft. above present sea-level to or below the exist- 
ing marine limit. 
A renewed study of the Barrington sand-plain made with my col- 
league, Mr. T. A. Jaggar, Jr., shows that it, too, has a drainage crease 
"springing out of the air," on the northern border as if from the van- 
ished ice-sheet, and terminating on the southern margin in a fan head- 
ing between two lobes of the delta front. This fan spreads out in the 
swampy foreground of the clay field and is of the subaerial type. As in 
the Greenwich Cove stage, we have here again evidence of the dejjosition 
of a delta-like sand-plain with the water as high as .50 ft. above the 
present sea-level; the next recorded event is the formation of a subaerial 
fan whose base is near or at the present sea-level. This fan is known 
to be contempcraneous with the ice remnant at the head of the com- 
pleted sand-plain, because the crease marks the path of drainage out of 
the ice, across the plain, and on to the fan. 
I have already given reasons for considering the Barrington stage 
later than the Greenwich Cove stage. If these stages were coutempo- 
