We now come to some remarkable data in a rather difficult department of super-geography. Vast fields of aerial ice. There's a lesson to me in the treachery of the imaginable. Most of our opposition is in the clearness with which the conventional, but impossible, becomes the imaginable, and then the resistant to modifications. After it had become the conventional with me, I conceived clearly of vast sheets of ice, a few miles above this earth--then the shining of the sun, and the ice partly melting--that note upon the ice that fell at Derby--water trickling and forming icicles upon the lower surface of the ice sheet. I seemed to look up and so clearly visualized those icicles hanging like stalactites from a flat-roofed cave, in white calcite. Or I looked up at the under side of an aerial ice-lump, and seemed to see a papillation similar to that observed by a calf at times. But then--but then--if icicles should form upon the under side of a sheet of aerial ice, that would be by the falling of water toward this earth; an icicle is of course an expression of gravitation--and, if water melting from ice should fall toward this earth, why not the ice itself fall before an icicle could have time to form? Of course, in quasi-existence, where everything is a paradox, one might argue that the water falls, but the ice does not, because the ice is heavier--that is, in masses. That notion, I think, belongs in a more advanced course than we are taking at present. Our expression upon icicles: A vast field of aerial ice--it is inert to this earth's gravitation--but by universal flux and variation, part of it sags closer to this earth, and is susceptible to gravitation--by cohesion with the main mass, this part does not fall, but water melting from it does fall, and forms icicles--then, by various disturbances, this part sometimes falls in fragments that are protrusive with icicles. Of the ice that fell, some of it enclosing living frogs, at Dubuque, Iowa, June 16, 1882, it is said (_Monthly Weather Review_, June, 1882) that there were pieces from one to seventeen inches in circumference, the largest weighing one pound and three-quarters--that upon some of them were icicles half an inch in length. We emphasize that these objects were not hailstones. The only merger is that of knobby hailstones, or of large hailstones with protuberances wrought by crystallization: but that is no merger with terrestrial phenomena, and such formations are unaccountable to orthodoxy; or it is incredible that hail could so crystallize--not forming by accretion--in the fall of a few seconds. For an account of such hailstones, see _Nature_, 61-594. Note the size--"some of them the size of turkeys' eggs." It is our expression that sometimes the icicles themselves have fallen, as if by concussion, or as if something had swept against the under side of an aerial ice floe, detaching its papillations. _Monthly Weather Review_, June, 1889: