126 The American Geologist. February, 1902. 
varying grain. 2. The time of cooling, other conditions being the 
same, varies as the square of the thickness of the dike.* From this 
last law it is assumed that the size of the crystals varies as the 
snuare of their distances from the nearest margin; then the square 
root of their area which can be measured varies directly as the 
distances from the margin. Thus we have a simple law of easy 
application. RicHarp E. Dopce 
Secretary pro tem. 
Notes ON THE SURFACE GEOLOGY oF Rio GraNnpe po Sut, Brazi.7 I 
cannot give you any definite information from my notes about Ter- 
tiary deposits in Brazil; because in Rio Grande do Sul, where I had 
opportunities for observations and study there are no deposits which 
I know certainly to be of Tertiary age. 
The rocks there are Archean, or, at least, low down in the series, 
and on their upturned edges are loose materials, cemented gravels 
etc. which are almost certainly Tertiary or probably Quaternary, with 
exceptions to be below noted, and with the further exceptions, per- 
haps, of the terraces near the coast. 
* I have had no opportunity to study the terraces along the sea 
coast. I noticed in Rio Grande do Sul that the coast is low and 
sandy. The city of Rio Grande do Sul is on a sand spit. From 
Pelotas inland toward Bagé are terraced sands and clays, the age of 
which I did not have an opportunity to inquire into. This terracing 
shows that there has been uplifting of a broad area there. The road 
passes off of these deposits and enters the hills at about four leagues 
west of Pelotas. 
Thence all the way inland to Bagé and Lavras, the rocks are very 
generally hidden by a deposit which I will call loess, and which I 
think is allied by origin as well as character to the loess of the Miss- 
issippi valley. It is of fine grain, clayey, of yellow or drab color 
where not darkened by organic matter, and shows but little variation 
in coarseness or fineness. However, streaks of+sand appear in the 
lower parts of it, and sometimes it passes downward into a sandy 
layer. This latter, however, may probably belong to the cascalho or 
gravel group to be below described. 
The loess is not derived from rocks in the immediate neighbor- 
hood where it occurs, for it preserves its identity or great similarity 
of character over wide areas where the rocks vary widely. 
Below the loess in places and lying between it and the “bed-rock” 
is cascalho, or gravel and sand. This is of material of the immediate 
neighborhood in which it is found. It consists principally of pebbles 
and of grains of yuartz and other hard materials, and at Lavras is 
confined to the immediate neighborhood of the quartz veins from 
which it originated. 
*RIEMANN, Partielle Differentiel Gleichungen. 
+These notes are from private letters by Mr. Mills in reply to inquiries 
about the geology of the region spoken of. The notes are quite brief, but they 
relate to a region about whose geology very little is known. Aside from a 
slight rearrangement and the omission of parts relating to other localities the 
letters are published just as he wrote them.—J. C. BRANNR, Stanford Univer- 
sity, California, December 3, 1901. 
ee 
