106 M. Fraunhofer on the Laws 



these angles, may have the proportion under consideration. 

 According to the theory, which will be treated of hereafter, 

 with the light falling vertically upon the system of wires, 

 the sines of the angles will have the said proportion. Partly 

 in order to be able to confirm this directly by experiments, 

 partly because the laws of this theory of the modification of 

 light can be better inferred from larger spectra, it appeared 

 to me, that, if it were possible to produce them, much finer 

 systems of wires than those which I had employed in my earlier 

 experiments would be desirable ; but in such fine systems the 

 spaces between the threads or wires must be equal, and a high 

 degree of accuracy is necessary, in order that the fixed lines 

 of the spectra should appear, without which the distances of 

 the colours from the axis could not be measured. 



To produce a considerably finer screw than that of which 

 I made use in my earlier experiments will not easily be 

 thought possible by those who are acquainted with the diffi- 

 culties of this kind of work. By means of a peculiar arrange- 

 ment I was enabled to trace upon a plate of glass, thinly 

 coated with leaf gold, parallel lines at such intervals, that 

 c = 0,00114} of an inch. If it is required to trace lines at very 

 small distances from each other, no gold remains on the glass, 

 and consequently no intervals appear. The spectra obtained 

 through such a system of lines, where s =0,00114 are, how- 

 ever, considerably larger than those formerly obtained, and 

 the fixed lines are very distinctly seen in them ; but the results 

 which could be obtained by them were still far from affording 

 any conclusion on the subject under inquiry. 



In a system of lines, which is employed in these experi- 

 ments, it is immaterial whether the lines of which it consists 

 are opaque, pellucid, or transparent. A system of spun glass 

 threads, for instance, produces the same phenomena as one of 

 metallic wires. I therefore laid on one side of a good plate of 

 glass such a thin coat of fat that it could with difficulty be 

 discovered with the naked eye. In this substance I traced pa- 

 rallel lines, which had intervals of only half the size of those 

 that were traced on the finest leaf gold. Through these systems 

 of lines spectra were produced in which the fixed lines could 

 be very distinctly perceived, and which are therefore well 

 adapted for accurately measuring their distances from the axis. 



