224 Transactions of the Society. 



principle for explaining them : " any two pencils of light (i.e. two 

 spectra, the direct image of the flame there included) entering the 

 objective, give origin to a set of lines, the direction of which forms 

 a right angle witli the connecting line of the pencils, or spectra, 

 and the distance of which is (exactly) inverse to the distance of 

 the spectra (i.e. of corresponding points in them). 



For instance, you will get diagonal lines on the quadratical 

 gratings, the distance of which is in the ratio oi 1 : yj'l narrower 

 as the distance of the real lines, by excluding (iig. 5) all the spectra 

 besides [except] a and h, or besides c and d, etc. 



On the rhombical field (fig. 6) you will see a set of lines 

 parallel to the short diagonal, in the same distance as the real 

 lines, by the spectra (fig. 7) c and h (or a and e, or d and /) — and 

 another set of lines, parallel to the long diagonal and narrower in 

 the ratio 1 ; s/o, by the spectra a and d or e and/ — a, s, o. 



You will rememljer, that I made the experiments with the 

 open flame of a petroleum lamp. I took the light from the narrow 

 side of. the flame, with the concave mirror, in observing the simple 

 grating, to show the duplication of the lines in part B, with the 

 two- or three-hole diaphragm, and to show the disappearance of all 

 the lines, by a single-hole diaphragm which only one pencil admits. 

 For the experiments with the crossed grating I took the light from 

 the broad face of the flame (in order to get a nearly circular image) 

 with the plane mirror, adjusting the distance of the lamp in that 

 way, that the different spectra appear separated. But 1 observe 

 that in all these experiments the different aspects of the object will 

 appear in the same way, if a broader beam of light is applied and 

 the spectra are superposed — provided that the diaphragms, i.e. the 

 holes in them, are exactly confoiiu to the distances of correspond- 

 ing points in the superposed spectra. 



The preparation I am about to send you fits best to Zeiss' 

 Objective aa (ca 1^ in.) which you have got from Mr. Zeiss, or, if 

 you want, will find at Mr. Baker's shop, 244 High Holborn. But 

 anv similar lens of 1 in. to lA in. foCus will do as well, if you 

 adapt the diaphragms to the dimensions of the spectrum produced, 

 and bring it as near as possible to the back-lens of the system. 

 For settling the diaphragm, I send a ring with the society-screw, 

 which is to be screwed bet^^■een the objective and the nose-piece of 

 your microscope. Within this ring there is another ring, turning 

 from without by a prominent pin. On the under face of this ring, 

 which is just above the back-lens, you fix the diaphragms (made 

 in card-board) by some wax. Turning the diaphragm without 

 altering the focus will allow to pass at once from the artificial 

 image to the normal image (and vice versa) which will appear as 

 soon as the hole (or holes) get such a position as to admit consecu- 

 tive pencils or spectra, proving in that way that the abnormal 

 images are not caused by the diaphragms, but only depend on 



