1830] 



ill the Madras Light House. 



281 



material composing the mirror was specuUira metal of highly brilliant 

 lustre, and polished in the most exquisite manner, it would not be assign- 

 ing too little to mirrors of looking-glass of the common description, to 

 assume that they would only return 50 out of 100.000, or one half of the 

 incident rays. In this case, what would be the value of the nine re- 

 flectors ? In the most favourable situation it w^ould appear that they 

 would increase the brilliance of the 12 lamps so as to render them equal 

 to fourteen, and in less favourable ones, that their assistance is not much 

 superior to the addition of a single extra lamp. 



I am aware, however, that this estimate has been made to rest upon an 

 assumption, which may not, upon examination, be found to correspond 

 with the actual circumstances of the case. It is indeed highly probable 

 from the nature of the apparatu*!, that the duly regulated distribution, 

 necessary to produce an exact equality of light may not be attainable, 

 and that great irregularity in this respect may be found to exist. Admit- 

 ting such liowever to be the case, the above would still be a correct 

 measure of the arerarje eifect produ':-ed througho\it the whole extent of 

 the horizon-, and a fair statement of the benelit derived from the appara- 

 tus, as it is evident, that if, owing to such inequality, it should be found 

 that the effect produced in any particular point of the circumference 

 were much in excess of what is above assigned, such a circumstance 

 could only be occasioned, by a corresponding defect in some other quar- 

 ter, which would fall short in an equal degree. 



In order, however, to prevent this point remaining a subject of doubt, I 

 have thought it worth while to lay down on paper, in the accompanying 

 plans (Figs. 1G, 17 and 18), the whole of the diverging rays, in the situa- 

 tions in which they are actually reflected, making the angles of incidence 

 and reflection equal, in accordance with the law in optics before alluded 

 to, so as to exhibit in one view the whole of the spaces in which the re- 

 flected images of eveiy lamp, from the surface of each reflector are to be 

 found.* The manner in which this has been done is represented in 

 Fig. 14, where the positions of the rays ceand bd, have been ascertained 



* It will be observed that in the 17th and 18th figures one angular space on each side is 

 made to overlap, and in part to fall in rear of the line g f of Fig. 14, parallel to the surface 

 of the central mirror, and it is evident that not only these, but also a small portion of the 

 rays in front of them, towards A and C could not be visible, owing to their being inter- 

 cepted by the opaque sides A and C. I have thought it better however to leave them ia 

 Figures 17, 18, and 19, in the places to which they would be directed by reflection, in order 

 that each group of rays might be duly accounted for, but they have afterwards been de- 

 ducted in summing up the effects and arranging them in a tabular form. 



