762 SCIENCE. 
example of graded river (Marshall sheet, 
Mo.); that is, ‘‘its course has been eroded to 
almost as low a stage as possible, and its 
slope has become very slight, so thatits cutting 
power is trifling.’’ Reference to the balanced 
relation of load and carrying power might have 
here been made to advantage. There is perhaps 
some objection to citing the Missouri as an ex- 
ample of a normally graded river, as it is 
probable that the present Missouri has estab- 
lished its flood plain by aggrading rather than 
by degrading the valley floor. The Platte 
river (Lexington sheet, Neb.) is instanced as 
an ‘overloaded’ stream, of which classit is cer- 
tainly a very striking example; but it is to be 
regretted that, as the term ‘graded’ was 
adopted for the Missouri, ‘aggraded’ or ‘ ag- 
grading’ was not employed for the Platte. 
The linear Appalachian ridges are beautifully 
shown where the Susquehanna cuts across them 
(Harrisburg sheet, Pa.), types of their class for 
the world. The text of this example is some- 
what less satisfactory than that of the others: 
tributaries are said to cut down their beds 
more rapidly than the main river ; the sinuosi- 
ties of the side streams are ascribed to the 
retarded erosion of the Susquehanna across the 
hard rocks, instead of to the pause in uplift, 
during which the inter-ridge lowlands were 
etched out ; indeed, no explicit mention is made 
of these lowlands as local weak-rock pene- 
plains, although the extraordinary enclosed 
meanders of Conedoguinet creek are referred to 
a time when the Susquehanna was ‘held at one 
level for a considerable period’ ; ‘subsequently, 
by some means, the river succeeded in lowering 
its bed’ and its tributary followed suit. ‘ By 
some means’ might be well replaced by ‘after 
an uplift.’ The diagram giving a section of some 
of the ridges does not properly represent the 
dip of their strata, and the Medina formation is 
labelled Potsdam. Finer examples than the 
alluvial fans (‘cones ’) of southern California and 
the huge basin of Crater lake in Oregon could 
not be selected for the closing numbers of this 
folio, which contains ten sheets in all. 
TERRACES FRONTING THE ROCKY MOUNTAINS. 
AT various points piedmont to the Front range 
of the Rocky mountains in Colorado, there are 
[N. S. Von. XIII. No. 332. 
sloping gravel-covered plains or mesas, into 
which the streams have cut their valleys. Lee 
describes some of these ( The origin of débris- 
covered mesas of Boulder, Colorado. Journ. 
Geol., VIII , 1900, 504-511, 4 figs.) and empha- 
sizes the contemporary date of the graded 
piedmont surface, beveled across the under- 
lying rocks, and the deposits of coarse waste 
with which it is covered; both are the prod- 
uct of the lateral shifting of streams, cutting 
on one side and filling on the other ; the waste, 
25 to 50 feet in thickness, being chiefly a flood 
deposit. The present action of the stream at 
Boulder in its channel and on its flood plain 
imitates the former action by which the inter- 
mediate mesa terrace and the higher mesa were 
produced. It is suggested that the three grade 
plains thus indicated ‘‘ do not seem necessarily 
to require the assumption of any change in the 
attitude of the land subsequent to the elevation 
of the mountains, but are the natural sequences 
of erosion as influenced by the local distribu- 
tion and difference in hardness of the forma- 
tions involved.’’ A gradual down-cutting with 
an active lateral shifting appears to account for 
the features described. 
THE FORMATION OF DESERTS. 
‘Das Gesetz der Wiistenbildung in Gegen- 
wart und Vorzeit’ (Berlin, 1900, 175 p., 
50 views) is a new work by Walther, already 
known for his studies of desert denudation. 
The book opens with a comparison of sea floors 
and desert surfaces ; each one becomes smoother 
by filling with waste from the enclosing high- 
lands, the coarser waste remaining near the 
margins ; but the deposits in one contain few 
records of life, while those of the other may 
teem with fossils. Special accounts are given 
of the process and results of dry weathering, of 
wind erosion (deflation) and of water action in 
arid regions. The forms assumed under these 
processes are well deseribed and illustrated. 
Rock ledges are left bare, frequently with a 
pitted surface; they are sometimes found worn 
down to a small relief and strewn with ascanty 
covering of waste. The popular impression 
that deserts are plains is held to be not so far 
from the truth as is sometimes taught, since so 
large a part of arid landscapes is degraded 
