The severe limitations imposed upon nearly all astronomical observations by "seeing" the effects resulting from passage of light rays through the turbulent atmosphere of the earth are familiar to every observer. With a small instrument the effect may appear largely as a continual shifting and scintillation of the image of a star, but with a large telescope poor seeing usually manifests itself mainly as an enlargement and blurring of the image. The reason for this difference becomes apparent when one makes a knife-edge test with a large telescope on a bright star. The turbulent elements of the atmosphere appear as shifting dark and bright areas on the image of the objective. In general these turbulent elements or waves have dimensions of the order of a foot or two small compared to the aperture of a large reflector. Thus the integrated effect of a number of turbulent elements leads to an unsteady enlargement of the image, perhaps with some irregular shifting in position. The seriousness of this is evident when one realizes that ideally the 200-inch Hale telescope is capable of giving diffraction images of stars about %0 oi a second of arc in diameter, yet the size of the "seeing image" produced is in the range j4 second to perhaps 5 or 10 seconds, being about 2 seconds in diameter on the average. Only rarely are images as small as yi second observed, and one may consider himself fortunate to experience one hour out of 1000 of the finest seeing, even at the best locations.