458 Gas HE Gilbert—Special Processes of Research. 
covered marked out the path of the rainstorm. When the. 
times of the loudest thunder, and the times of certain other 
events of the storm had been similarly tested, a comparison of 
the results showed that the observations on the times of first 
rain produced the most orderly system of lines, and the rain 
front was therefore selected as one of the bases for the next 
classification. 
It will be observed that the second classification, by place 
and time, was superposed on the first classification, by storms. 
The third classification was likewise superposed on the first, 
but not on the second, except in the sense that it used its 
results; that is to say, the third classification was applied to 
the individual storm, and substituted new bases in lieu of place 
and time. These bases, which may be called the horizontal 
axes of the storm, were the middle path and the rain front. 
The classification was once more graphic, and was accomplished 
SS ee ee 
TEN t 
\ \ 
‘ ) S 
N.Y, / 4 FEN UO ES 
‘y i Ni a 
i] \ NO \ SSG } 
{ CONNEC|TICUR | 
- 
i K ome wy 
k n us f rw 
/ Ae Is 
\\ ily a reg 
\ puev~d g eG 
/ aX oes i Xe 
uate y vd 
Fig. 1. Lines of synchronous rain-front, storm of July 21, 1881. 
in essentially the following manner. A straight line drawn on 
a sheet of paper represented for any and every instant of the 
history of the storm, the portion of the middle line of its path 
momentarily included in the storm area. A curved line trans- 
verse to this represented with its proper form the line of first 
rain or the rain-front (fig. 2). Then observations of thunder, 
of heaviest rain, of temperature, of lightning strokes, etc., were 
platted on the same sheet, the position of each observation be- 
ing determined by measurement from the axes. Its distance 
from the middle path axis was made equal to its distance from 
that path on the map showing the storm track; its distance 
from the rain-front axis was determined by a computation 
based on the velocity of the storm and the time elapsed before 
or after the rain-front passed the same station. If for a given 
instant of time a chart were prepared, showing, in their proper 
place relations, all the features of the storm observed at that 
‘ 
