[ 144 ] 



XX. Notices respecting Neiv Books. 



Madeira Spectroscopic, 1881-82. By 0. Piazzi Smyth, Astronomer 

 Boy al for Scotland. W. and A. K. Johnston : Edinburgh. 



f'FHIS volume, read in connexion with ' Madeira Meteorologic ' 

 ■*■ (D. Douglas, Edinburgh) by the same author, forms a valu- 

 able contribution to the moist-climate history of that island ; but 

 its more immediate object (as we read on the title-page) was " the 

 revision of 21 places in the red half of the solar visible spectrum 

 with a Eutherf ord diffraction-grating at Madeira in the summer of 

 1881." 



This is effected in 32 closely printed large quarto pages, and 17 

 photo-lithographed plates. At first sight the latter rather shock 

 the eye by certain dense black columns of irregular heights and 

 masses of apparently coarse shading ; but by and by we ascertain 

 how this arises. 



Part I. opens by explaining the local disadvantages presented by 

 the Eoyal Observatory at Edinburgh for such work, and the need 

 of a station where the sun was nearer the zenith. Madeira, though 

 offering one day only in ten of pure blue sky (but that one of 

 splendid quality), was selected, and at the Quinto da Corvalho, 

 near Eunchal, a heliostat and spectroscope were erected. The 

 sun's beam from the heliostat was transmitted through a 6-inch 

 achromatic lens of long focus to a collimator of 31 inches focal 

 length, then reflected from a Eutherf ord grating 1*6 inch square 

 with 17,296 lines to the inch, and finally viewed with a telescope 

 of 52 inches focal length. The lenses of the collimator and tele- 

 scope were respectively of quartz, 2-25 inches in diameter ; and eye- 

 pieces were used magnifying 30, 45, and 64 times upon the second 

 or third order of spectrum. 



In front of the slit a white flint-prism separated the colours for 

 the particular portions of the spectrum examined. This fine piece 

 of apparatus secured the author his Madeiran spectra, drawn on a 

 nearly uniform scale of wave-numbers to six places of figures. 



The earlier Solar Maps with which these are compared are thus 

 enumerated : — 



(1) Sir David Brewster and Dr. Gladstone's in Phil. Trans. 

 1860. A careful copperplate, but on too small a scale, except as 

 to certain magnified groups : a prismatic spectrum. 



(2) Prof. Angstrom's normal (grating) solar spectrum, 1868. 

 Most exemplary, on stone with "incised" lines. 



(3) Kirchhoff and Hofmann's prismatic spectrum as presented 

 in Eoscoe's ' Spectrum Analysis ' (1869). Defective, owing to errors 

 of " registration " in the four tint stones used in the printing. 



(4) Erom Dr. Schellen's ' Spectrum Analysis :' a print of the last 

 map from wood-engraving, liable to similar errors. 



(5) Eutherford's single stone lithograph (about 1874), from his 

 fine photograph of the blue and violet half of a prismatic spectrum ; 



