map. The meeting point of the lines
thus drawn marked the location of the
observer's instrument on the map.
From this location, which was deter-
mined in the daytime, sights were
taken at night with the alidade on
the map. The telescope of the alidade
swings in a vertical as well as in a
horizontal plane and can therefore be
used for measuring vertical angles
along the lines of sight. The dip-
needle compass is so arranged that
the needle swings in a vertical in-
stead of in a horizontal plane. It is
used to detect differences in magnetic
attraction.

Three stations were occupied--one
In the knoll by the cottage formerly
occupied by Rev. C. E. Gregory, near
Loven's Hotel; one in a field on the
east slope of Gingercake Mountain;
and one on the terrace in front of
the summer residence of Miss Cannon,
at Blowing Rock. These stations are
marked on the map (fig. 1) by the
letters A, B, and C, respectively.
Two nights were spent on Brown Moun-
tain, but the conditions were so un-
favorable that no station was occu-
pied there. At each station at which
observations were made, vertical
angles to parts of Brown Mountain
were noted, dip-needle readings were
made, and photographs were taken.
Vertical angles were also measured
when practicable from each station to
each light seen. The procedure adopt-
ed was first to get a line of sight
to the light and then to note its
time of appearance and measure its
verticlal angle, but occasionally a
light remained visible for so short a
time that it had disappeared before
the telescope could be trained upon
it and a line drawn to fix its direc-
tion. Few records were kept of lights
for which lines of directions were
not drawn, but the total number seen
may have been nearly twice the number
recorded. The atmosphere proved too

The observations obtained in the
field were afterward adjusted on the
map. Many profiles along lines of
sight were constructed, the vertical
angles were plotted, and corrections
for the curvature of the earth's sur-
face and for refraction were made. In
this way the sources of some of the
lights were approximately determined.
The inset profile (fig. 1) illustrates
the method of locating a source of
light by means of a profile and ver-
tical angle drawn from station B
along line 15.

Space does not permit a detailed
statement of the individual observa-
tions made and of the inference drawn
from them. The geographic positions
of the sources of light as determined
by instrumental observations are only
approximate because of the diffi-
culties attending the use of the
instruments in darkness. The stations
were from 1,000 to 1,500 feet higher
than the summit of Brown Mountain, so
that the lines of sight to the lights
seen all passed several hundred feet
above the top of the mountain as
shown in inset profile (fig. I). This
fact caused the lights to appear over
the mountain rather than on or below
its crest, a feature noted both in
the first published description of
the lights, in the Charlotte Observer,
and in Professor Perry's description,
already quoted. The appearance of the
lights as described in these two
accounts, especially in that given by
Professor Perry, agrees so closely
with their appearance as observed by
the writer that no additional de-
scription of them need be given here.