A. Tylor on Changes of the Sea-Level, &§c. 21 
I think, that the western limit of the shower was more or less 
convex towards the point of observation. 
4, If the river bends eastward at all on the North and South 
of the observer’s position, this circumstance would virtually add 
so much to its width, since the rays forming the branches of the 
bow would then fall lower than BF in the vertical projection. 
Whether this is so, I am not informed. 
It is obvious from an inspection of the figure, that most of the 
reflected bow would disappear sooner than the direct bows, inas- 
much as the angular thickness of the luminous stratum, at its in- 
tersection with the shower, would rapidly diminish as the shower 
retiréd. And, furthermore, as the vertex was produced by the 
nearest drops, this part must have vanished first, as was observed 
to be the fact. 
Had the sun been about 6° above the horizon, the vertex of 
the image-bow would have coincided with that of the secondary ; 
and by its opposite arrangement of colors, would have partially 
neutralized its tints, and made a white segment common to the 
wo, It was in this aspect that the direct and reflected bows 
presented themselves to the view of Dr. Halley, in 1698, on the 
bank of the river Dee, of which he published an account in the 
Transactions of the Royal Society. 
0 the inquiry, why do not those, who live near a sheet of 
water, more frequently witness the reflected bow, it may be re- 
plied, that, if the water is not of greater width than a fraction of 
a mile, the favorable circumstances already enumerated will very 
rarely concur to produce the phenomenon with much distinct- 
hess; and if there are several miles of water, so as to reflect a 
am of large vertical thickness, yet the surface would very 
rarely be smooth enough, directly after a shower, to form a single 
and well-defined image of the sun. And it may be added, that there 
are few persons, except such as have made the theory of the rain- 
bow a matter of careful study, who would consider a bow as par- 
ticularly noticeable and worthy of description, simply because it 
happened to intersect the others, especially if, as must ordinarily be 
the case, the intersecting bow was only a short and indistinct arch. 

Arr. TIL—On Changes of the Sea-Level effected by existing 
Physical Causes during stated periods of time ; by ALERED 
* Tytor, F.G.S.* 
Introduction. 
Tue First Part of the ensuing paper is occupied with the de- 
tails of the probable amount of the solid matter annually brought 
into the ocean by rivers and other agents, in suspension and solu- 
tion; and the conclusion is arrived at, that the quantity of detri- 
tus thus distributed on the sea-bottom would displace enough 
® From the Philosophical Magazine for April, 1853. 
