E. S. Snell on a Rainbow caused by reflected Light. 19 
vertex 4° above it. Its extremities came within 15° of the hori- 
zon, and if prolonged a little, would have intersected the pri- 
mary itself. This gorgeous display lasted some ten minutes, 
when the third bow began to fade at the top, and soon wholly dis- 
appeared. (See Fig. 1.) 
Mr. Adams rightly judged that this additional bow was the 
effect of the sun’s rays, reflected from the Connecticut river, which 
runs on the west side of the Seminary hill, and whose waters 
were then very placid. It was in fact a primary bow produced 
by an image of the sun in the river. As the vertex of the re- 
flected bow was 4° above that of the secondary, or 16° above 
that of the primary, the image of the sun must have been 16° 
below the sun itself, and the sun therefore 8° above the horizon. 
In first considering the case, I found it difficult to conceive, that 
the stratum of light, reflected so obliquely from a narrow belt of 
water, could have sufficient thickness to form a bow occupying 
between 160° and 170° of a circle. 
The accompanying vertical section through the axes of the 
bows, exhibits in their true proportions, the breadth of the river, 
the distance and elevation of the observer, and the thickness of 
the reflected sheet of light; and shows pretty accurately what 
Zp S 

n 
\ 

N 
5 G 
— 
ie 
must have been the situation of the drops concerned in producing 
the upper and lower parts of the bow. (See fig. 2.) The breadth 
the river, AB, is a little more than half a mile. (The observer 
at D, is elevated 100 feet above the level of AB, and his distance 
fram B, the nearest bank of the river, is three-fourths of a mile. 
SDH is the axis of the direct bows, inclined 8° to the horizon. 
The axis of the reflected bow, ID H’, intersects S DH at D, ma- 
king with it an angle of 16°.. ABEF is the reflected sheet of 
light which penetrates the shower and forms the image-bow. Its 
thickness in the section GB, measures about 370 feet. RD, 
D, making with DH the angles 40° 17’ and 42° 2’, are the 
@ and violet rays from the summit of the primary bow; vD, 
8° much elevated, as to cross the secondary, and extend at the 
