. 
a 
"quently have a petroleum-like odor. 
APRIL 4, 1884.] 
ealling it a ‘ jacobi,’ as all derived units should 
be related to the fundamental units of length, 
mass, and time, through simple, decimal ratios. 
There is danger, in fact, of the simple ele- 
SCIENCE. 
421 
gance of the absolute system being destroyed 
by excessive ornamentation; and it is well 
enough to make haste slowly in adding to what 
has already been done. 
INTELLIGENCE FROM AMERICAN SCIENTIFIC STATIONS. 
GOVERNMENT ORGANIZATIONS. 
Geological survey, 
Paleontology. — Prof. H. 8. Williams, from his pre- 
liminary study of the specimens he collected during 
last season in Genesee and Wyoming counties, N.Y., 
from the Genesee slate and Portage formations, 
reaches some interesting conclusions. He is con- 
vinced that the black shales which appear in the 
lower Portage of this region, and continue to appear 
as thin zones up to a point just below the Portage 
sandstones, represent merely an interrupted continu- 
ation of the deposits called the Genesee slates. After 
the typical Genesee slate was deposited to some con- 
siderable thickness, the Portage fauna made its ap- 
pearance in the soft, blue, argillaceous shales. A 
hundred feet higher another black shale appears of 
several feet thickness, and then olive shales come in; 
and for several hundred feet this alternation con- 
tinues, the black shales becoming thinner with each 
repetition, and containing an increasing amount of im- 
purity, siliceous and argillaceous, so that in the upper 
part there are only dark-gray bands or streaks of the 
olive shales, with fine paper-like layers of black. The 
earlier Portage black slates bear the same fauna as 
the Genesee, but the specimens are fewer. Although 
these black slates are interstratified with the olive 
shales, they do not contain the Portage fauna. It is 
confined to the olive layers, and, higher up, to the blu- 
ish argillaceous shales. Near the top of the Portage 
series the sandstones come in. They are of a light- 
gray color, and are generally calcareous. They fre- 
With them the 
Chemung fauna is associated. The lowest observed 
appearance of that fauna was in Java township, 
Wyoming county, in the first of the gray sands lying 
just above the last-observed black zone, which was 
bituminous. 
Professor Williams also says that some interesting 
features have been revealed by the study of a large 
series of specimens of Spirifera mesocostalis Hall. 
In the representatives of the species from the upper 
Devonian, there is a well-developed median septum 
in the ventral valve, asin the genus Spiriferina; but 
the punctate character of the shell of that genus has 
not been observed in any of the specimens. The lower 
forms, at its first appearance in the Ithaca group, very 
rarely show any trace of the septum. As far as Pro- 
fessor Williams’s examination has gone, he finds that 
the median septum is more fully developed and more 
generally present, the higher up the specimens are 
found. In harmony with this observation is the ref- 
erence, by Mr. Whitfield, of a similar specimen from 
Wisconsin (Geol. Wisconsin, iv. 332) to Spiriferina 
under the name of Spiriferina (?) ziczac. 
STATE INSTITUTIONS. 
New-York state survey. 
Rainfall of western New York. — To ascertain how 
much water is likely in different seasons to flow off 
of the surrounding watershed into Oak-Orchard 
Swamp, it was necessary to study with great care the 
rainfall of the western part of the state for the past 
fifty years. A careful analysis was therefore made of 
observations taken at Rochester university since 1880, _ 
and by the U.S. signal-service at Buffalo and Roches- 
ter since 1870. The result of this discussion of the 
Rochester rainfall is quite remarkable. It is shown, 
that from 1830 to 1880, during the very period when 
the woods were being cut off from the western part 
of the state, the rainfall steadily increased from a 
mean annual precipitation of 27.7 inches to 38 inches. 
The average was 34 inches. From 1868 to 1881 in- 
clusive, there was the greatest average rainfall known 
for a similar period in that locality: it was 388.73 
inches. The greatest recorded monthly, daily, and 
spring rainfalls occurred between 1870 and 1880. 
This decennial period is therefore a safe one from 
which to estimate maximum amounts of water likely 
to be discharged from watersheds in the western 
parts of the state; but 4owns whose future water- 
supply is estimated from the amounts received into 
lakes or streams since 1868 may find themselves very 
short of water, if the mean annual precipitation 
should decrease to that of the period from 1830 to 
1840. Long periods of small average rainfall will 
doubtless recur in the region near Lake Ontario. 
The city of Rochester should be prepared for a time 
when, for ten years, the average yield of water from 
its present source of supply, the basin of Hemlock 
Lake, may amount to only three-quarters of the 
average flow from 1868 to 1881. 
Quantity of water evaporated from various water- 
sheds. — While the mean rainfall of this region has 
increased during the past fifty years, the summer flow 
of the streams has greatly diminished. This is due 
partly to the loss of retaining-power in the ground, 
owing to the removal of the soft forest mould, which 
in former times readily absorbed the rain and melt- 
ing snows, and so prevented these invaluable waters 
from rushing off and wasting themselves in destruc- 
tive floods, and partly to the enormous increase 
in evaporation. The proportion of rainfall, which, 
owing to evaporation, is lost for use in springs, lakes, 
and streams, is known to but few. In the special 
