The Astrophysical journal . ig. 2 shows the appearance of a plate (enlarged two diam-eters ; the fine lines have been lost in the process) when canal rays inhelium (a) and a helium-spectrum tube (b) were sources. If any Dop-pler effect was present, it would be made evident by a shift of the twohalves relativelv to one another. There might easilv be some doubt 4 G. F. HULL regarding the interpretation to be placed upon this shift. For,confining our attention to one radiation, there would be in generalon the plate two maxima corresponding to (say) the 8000 and 8001orders. If the radiation were s
The Astrophysical journal . ig. 2 shows the appearance of a plate (enlarged two diam-eters ; the fine lines have been lost in the process) when canal rays inhelium (a) and a helium-spectrum tube (b) were sources. If any Dop-pler effect was present, it would be made evident by a shift of the twohalves relativelv to one another. There might easilv be some doubt 4 G. F. HULL regarding the interpretation to be placed upon this shift. For,confining our attention to one radiation, there would be in generalon the plate two maxima corresponding to (say) the 8000 and 8001orders. If the radiation were shortened in wave-length by about o. 5tenth-meter, the maximum toward the shorter side of the echelonwould move over and take the place of the other maximum, while anew order would follow it and take its place. If the radiation werelengthened, the shift would be in the opposite direction. In bothcases the two maxima would look like the original two , though a variation in wave-length of o. 1 tenth-meter would pro-. 1 in 1 vO 1 1 100 CM 1 1 1CM \0 c CM CM CO i-» 0) M - O f*5 rf ON O t*J Tj- -* -* ^ *fr ?* IO Fig. 2 duce a very marked shift in the lines, one of o. 5 might not cause anoticeable displacement. But shifts as great as tenth-metercould be detected and measured by the single prism i7, the echelonbeing removed. The method therefore used for determining theorder of magnitude and nature of the shift was to remove the echelonand take a photograph of the spectrum through the prism F. It isevident that when the echelon is in use the width of the spectral linesused cannot be as great as o. 5 tenth-meter, otherwise the two chieforders would overlap. By width of a line is here meant the wave-length interval between two radiations whose intensities are one-halfthat of the maximum. When the echelon and prism were arranged as in Fig. 1, a shiftof the lines toward the red end of the spectrum denoted a shorteningof those lines. Adjustment for focus.—The slit and the
Size: 3187px × 784px
Photo credit: © The Reading Room / Alamy / Afripics
License: Licensed
Model Released: No
Keywords: ., bookcentury1800, bookdecade1890, booksubjectspectru, bookyear1895
