4'2'2 C. PIAZZI SMYTH ON MICROMETRICAL MEASURES OF 



Rule 4. When lines of light in the spectrum are so faint or ill-defined that 

 they no longer give the appearance of solid or liquid light, and are no longer 

 sharp and smooth-edged like knife blades, or stretched silver wires, but look 

 more like faint, uncertain, granular, worsted threads, — the lines representing 

 them on the plates are not drawn in full ink or with parallel sides, but of a 

 conical shape, or in dots, or in wavy lines, and in extreme cases with 

 cross lines. 



Rule 5. Faint broad bands in the spectrum, and its occasional portions of 

 continuous spectrum light, are never indicated in these drawings by any kind or 

 arrangement or succession of vertical lines, but by either horizontal, or oblique 

 lines; and these may be crossed over and over again to produce the required 

 degree of intensity in any case. For the lines of such shadings can evidently 

 be never confounded with true spectral lines ; which, being images of the slit 

 of the spectroscope, must all be vertical and parallel, when the spectrum 

 range is horizontal. 



Rule 6. Shading by either inclined or horizontal lines being symbolic and 

 abstract only ; such lines may, for facility of execution, be of almost any 

 degree of coarseness or width apart ; provided only that the amount of ink 

 contained in them shall, if symbolically supposed to be smeared up and down 

 within the upper and lower limits of the horizontal spectrum strip, — indicate 

 only a grey shade, not a full degree of blackness. And exceeding refinements 

 of such shade or faint-light surfaces in the spectrum itself, may be indicated on 

 the drawings by making the shading lines there cover more or less of height or 

 depth in the spectrum strij), as already adopted for the easier representation of 

 different degrees of brightness in the spectral lines alluded to under Rule 3. 



Drawings of the CH Spectrum in Blow-pipe Flame. 



Now all the latter of these rules came into play at once with our first 

 subject, or the blue-grey blow-pipe flame of coal-gas and air; for there is not 

 light enough in it to give off any lines of real, liquid light, only haze of 

 different degrees of rarefaction. 



Considering indeed the proverbial faintness of the blue base of flame, as 

 when a candle burns low and blue, it is rather surprising to find that so very 

 small a portion of it as enters between the almost closed jaws of the 

 spectroscope's slit, can yet be distinguished as made up into upwards of 400 

 parcels, separated one from the other always by different and definite spectrum 

 place ; and sometimes also by instantly recognisable features of physiognomy 

 or gradated intensity. 



Yet such is the case, for the plates now exhibited of the coal-gas and air 

 blow-pipe spectrum (Plates XLVIII. to LI.), show 81 separate existences in the 



