DECEMBER 31, 1897.] 
struction is called to the aid of simple ex- 
eavation. After showing that corry lakes 
are true rock basins and that local deforma- 
tion cannot account for them, Fugger ad- 
vances the idea that they lie in funnel- 
shaped cavities that once led down to sub- 
terranean channels opened by solution, but 
now obstructed ; even advocating this ex- 
planation in corries of crystalline rocks, 
and defending it by 2 rather elaborate phys- 
ico-ehemical argument. 
Apart from the general difficulty of be- 
lieving in the sufficiency of underground 
solution in resistant rocks, it seems impos- 
sible that erosion thus determined could ex- 
ceed that effected by the active streams that 
descend the steep slope on the open sides of 
eorries. As the problem is presented by 
the writers mentioned above and by various 
others, glacial erosion seems to be the most 
competent cause for corry lakes. 
THE 14000 MALDIVE ISLANDS. 
Tue rarely visited Maldive archipelago is 
described in an interesting article by Ros- 
- set (Mitth. Geogr. Gesell. Vienna, XX XIX., 
1896, 597-637). The islands are all of 
coral formation, seldom more than two 
meters above sea level, with much un- 
healthy swampy surface. They are seldom 
more than a few miles in diameter. More 
than a hundred islets may form the circum- 
ference of a single atoll, and sometimes the 
individual islets themselves have a ring- 
like, atoll form. The seaward submarine 
‘slopes are steep ; the shores are attacked by 
heavy surf, and the natives believe that 
the land area is decreasing. The islands 
are separated by deep passages through 
which strong currents run one way or the 
other according to the monsoon season. 
Many channels breach the reefs and give 
access to quiet anchorages in the lagoons. 
The colors about an atoll vary from the 
purple waters off shore to the green, shal- 
low water, the white coral strand, the olive 
SCIENCE. 
987 
brown reef with dark green vegetation, 
and the bright green lagoon. A descrip- 
tion of the people and their history fol- 
lows. 
W.M. Davis. 
HARVARD UNIVERSITY. 
CURRENT NOTES ON ANTHROPOLOGY. 
ELEMENTS OF MELANESIAN ART. 
Aw article of prime value to students of 
early art and to anthropologists in general 
is that by Dr. K. Th. Preuss, in the Zezt- 
schrift fiir Ethnologie, 1897, Heft III. and 
IV., on the artistic designs of the natives 
of Kaiser Wilhelms Land, New Guinea. 
The material he had at his command was 
a collection of over five thousand speci- 
mens now in the Museum of Ethnography, 
Berlin. He considers it practically com- 
plete, presenting the world of their art in 
line and figure. His article is illustrated 
with 199 figures in the text, yielding ample 
means for studying the leading motives of 
these savage artists. The analysis of their 
favorite forms is traced out with masterly 
precision, and as one follows the author in 
his unraveling the strange and intricate 
figures he copies, no doubt is left of the 
success of his undertaking. 
In some introductory pages he refers to 
the bearing of such studies on the question 
of transmission or independent origin, and 
on the tendency of primitive man to copy 
from nature and to conventionalize his 
copies. Several popular impressions are 
corrected and sounder methods of compari- 
son explained. 
THE EXTENSION OF THE ARAWACK STOCK. 
Tuts stock of South American languages 
has peculiar interest, as it is that which 
spread over the West Indian Archipelago 
and the Bahamas at some remote date; and 
if any of the native languages of our Gulf 
States had South American affinities, they 
should be looked for in the Arawack and not 
