514 
transference of yellow fever by means of a par- 
ticular kind of gnat by the recent American 
Commission is hardly intelligible for bacillary 
disease. Moreover, it does not seem to be 
borne out by their experiments, nor does it ap- 
pear to satisfy certain endemiological condi- 
tions. It is proposed to deal more fully with 
the endemiology and epidemiology of the dis- 
ease on a later occasion. 
12. We think that the evidence in favor of 
the etiological importance of the fine small 
bacillus is stronger than any that has yet been 
adduced for any other pretended ‘ yellow fever 
germ.’ At the same time thereis much further 
work to be done ere its final establishment can 
beclaimed. The acquisition of a new intestinal 
bacterium would explain the immunity of the 
‘acclimatised.’* 
THE NEW STAR IN PERSEUS. t¢ 
THE first news of Anderson’s discovery of a 
new star in Perseus was received at this Observ- 
atory on February 24th. An examination of the 
region near the star, made that evening with 
the 40-inch telescope, failed to show any evi- 
dence of nebulosity, but the bright moonlight 
would have rendered a faint nebula invisible. 
At that time the magnitude of the star appeared 
to be about 0.5. Its color was yellow, with a 
decided reddish cast, very similar to that of a 
Orionis. Very little time was spent in examin- 
ing the spectrum visually, as it was felt that 
photographs would be more valuable than 
drawings based on micrometer measures. We 
had fortunately just received a fresh supply of 
Erythro plates through the kindness of the 
International Color Photo Company of Chicago, 
and it was therefore possible to photograph the 
entire spectrum from Ha to Hr. Beyond this 
point in the ultra-violet the absorption of the 
40-inch objective greatly enfeebles the spec- 
trum, which is still further weakened by the 
lack of perfect achromatism in this region. 
* The completion of the interim report, of which this 
is an abstract, was interrupted by the onset of attacks 
of yellow fever in both of us. The loss of my much- 
lamented colleague renders it advisable to submit the 
shortened report only for the time being.—H. E. D. 
f From Bulletin No. 16 from the Yerkes Observatory 
of the University of Chicago. 
SCIENCE. 
(N.S. Vou. XIII. No. 326. 
Photographs of the spectrum were obtained 
by Mr. Ellerman on February 24, 25, 26, 27 
and 28, March 4, 6 and 11. The comparison 
spectra which appear on these plates are those 
of titanium, hydrogen and sodium. 
On February 24th and 25th Mr. Ritchey photo- 
graphed the region of the Nova with the 40-inch 
telescope and color screen. In order to obtain 
a sufficient number of comparison stars the 
plates were given an exposure of one hour. 
The light of the Nova was intercepted by a 
small movable occulting disk, with which four 
(for the second plate, five) very brief exposures 
were given at intervals of about fifteen min- 
utes. The total exposure for the Nova was 
probably about half a second. In the result- 
ing photographs, the images of the Nova and 
the neighboring stars (of which more than forty 
appear in a region 12’ square) are small and 
appear to be well adapted for measurement. 
Through the kindness of Director Rees, these 
plates will be measured at the Columbia Col- 
lege Observatory. The position of the Nova 
was measured micrometrically by Professor 
Burnham on March 34d. 
The wedge photometer used with the 40-inch 
telescope in the determination of standards of 
faint stellar magnitude has been employed by 
Mr. Parkhurst in measuring the brightness of 
the Nova. Hitherto objectives of one and two 
inches aperture have sufficed, but as the Nova 
decreases in brilliancy it will be followed with 
the 12-inch and 40-inch telescopes. A prelimi- 
nary reduction gives the following magnitudes: 
Date. Mag: 
1901, Feb. 25 1.0 
UG SOG 1.1 
a ES Pe 2.0 
iad “ 98 il: 9 
«Mar. 3 2.7 
“oe e 4 2.8 
“ce ay 5 OMe 
ut Ho 3 ell 
A photograph of the spectrum (G 440) taken 
with the one prism spectrograph on February 
28th has been measured by the writer. The 
resulting wave-lengths of the lines and bands, 
computed by the aid of Cornu-Hartmann for- 
mul, furnished data for attaching a scale to an 
enlargement of the photograph reproduced in 
the Bulletin. 
