334 



NATURE 



\Aug. 26, 187 



was made to deal with this - question, excepting the simple 

 expedient of placing a spherical mirror on the landward side, 

 where no light was wanted, and thus the rays intercepted by the 

 mirror were reflected back again through the flame, so as to be 

 ultimately aced on by the apparatus at the seaward side. But 

 this device did not in any way fulfil the condition of allocating 

 the rays proportionally to the varying distances at which the 

 light had to be seen in the different azimuths, nor to the ampli- 

 tude of the arcs. What was required wrs a system by which the 

 ■whole light from the lamp should be spread horizontally and 7t'ith 

 strict equality over any ^iven arc in azimuth ; and at a light of 

 umqual range, which must be sen at dij^erent distances in dtj^cr- 

 ent az'muths, the rays should he allocatid to each of such arcs in 

 the compound ratio of the number tf Jegrees and the distances 

 from which the I'^ht has to be seen in such arcs. 



It is unnecessary to give a description of the vnrious methods 

 of solving this problem for fixed and revolving lights, which I 

 have elsewhere published under the name of " Azimuthal Con- 

 densing Lights."^ All that is here required is to indicate gene- 

 rally the mode of dealirg with fixed condensing lights, which 

 was first employed for .some narrow navigable channels on the 

 west coast of Scotland in 1857, and which is now adopted in 

 many different countries. 



We shall take a case of the simplest kind in order to illustrate 

 the principle : — 



Let a lamp be surrounded b) the'fixed light apparatus of Fresnel, 



which allows the rays to pass through it unaltered in direction 

 in azimuth, so as thus to show a light of equal intensity all round 

 the horizon, but which operates on the rays in altitude by bend- 

 ing upwards, to the horizontal, those rays which would fall on 

 the lightroom floor and be lost, and also by bending downwards 

 to the horizontal those rays which would naturally pnss up to the , 

 clouds. This instrument then strengthens the light passing to 

 the sailor's eye, by bending upwards and downwards those rays 

 which would otherwise be entirely lost. Suppose, however, that 

 the light docs not require to be seen at all in an arc in the 

 direction of the land, and that there are two other sectors in azi- 

 muth in which the light has to be seen at greater distances than 

 any others. If we place outside of Fresnel's apparatus In the 

 azimuths towards the land (which may therefore be made dark) 

 straiglit prisms which have each the property of spreading the 

 light that falls on them over the sectors that require most 

 strength, and if we proportion the number of these stta'ght 

 prismi to the required distances and to thp number of degrees 

 which have to be illuminaterl, we shall then fulfil this simplest case 

 of the problem, viz., the due strengthening of the light in tlie 

 directions of the longest ranges, and its uniform distribution in 

 azimuth over those sectors. 



The diagram represents in horizontal middle section the design 

 for a new light which is about to be erected and which requires to 

 illuminate different arcs with light of different intensity. A is the 

 lamp encircled in front by B, which represents part of Fresnel's 



'/'-' 



'>f 



1^ / . Bar ^ ^ ' K*' * J , r * A/ y^ft ■/ ^'^ ^'^ 



vwwWW 







fixed light apparatus, outside of which are shown straight vertical 

 piisms numbered I to 14 for condensing the rays over the arcs in 

 azimuth that have to be s'rengthei.ed, and which arcs have cor- 

 responding numbers I to 14. The novelty in this arrangement 

 is the mode adopted for reducing the space which would other- 

 wise be occupied by the condensing prisms but for which there 

 is no room in a lantern of ordinary size. 



My friend Piof. Swan, of St. Andiews, among other ingenious 

 devices in a paper read to the Royal Scottish Society of Arts, pro- 

 posed, in order to reduce the space occupied by the apparatus, to 

 place prisms behind others and to project the rays from the prisms 

 behind, forwards through spaces left between the prisms in fiont. 

 In the piescnt design I have availed myself of this pioposal. 

 The prisms 10 to 14 throw their light between the pritms 3 to 8. 

 Owing, however, to the natural divergence due to the size of the 

 flame, much useful light would be lost by impinging agaibst the 

 edges of the outer prisms unless those prisms were separated 

 farther from each other in order to afford wider spaces for the 

 conts of light to pass through. But this again would in ci ease 

 the space occupied by the apparatus. The difhculiy may be 

 overcome by cutting out the apex of the outer prisms as shown 

 in piisms 4 to 9. This would, as in Buffon's annular lens, also 

 maieiially reduce the absorption of the light which passes 

 through them. For facility of construction, however, instead of 

 one prism cut in this manner, two small separate prisms arranged 



^ Edin. New Phil. Journal, 1855, p. 273. " Lighthouse Illumination," Edin- 

 burgh, 1871 ; second edition, p. 79. 



symmetrieally with the la:ger one have Lecn subbti.uted.^ Dy 

 means of these groups of i7o/«/rw»w the apparatus is xeduced 

 within prac ieable dimensions, while the quantity of glass em- 

 ployed in the apparatus is materially lessened, and the loss from 

 absorption is reduced by about one-tenth. 



It is only necessary to add that while the cost of the first appa- 

 ratus of the kind will be increased by the greater nuu ber of 

 moulds required for casting the prisms, it will on the other hand 

 be reduced by the smaller quantity of glass required. The 

 amount of glass surface which has to be grour.d and polished is 

 ol course the same for each pair of twin prisms as for one sinj;le 

 large prism. 



The new apparatus will, in addition to what has been de- 

 scribed, require at the back of the flame the Dioptric spherical 

 mirror whieh I pn posed in 1849^ with the improvements sujj- 

 gested by Mr. J. T. Chance in ib62, atjd above the flarm, 

 prisiiiS of the new forms suggested by Mr. Brebner and m>SLl , 

 and also independently by Prof Swan, and which were first iniro- 

 duced at Loehindaal in Argyllshire, in 1869.^ The apparatus 

 will therefore embrace in all six different optical agent.s, and will 

 compress into one sector of 82° light wliieh would liaturally 

 diverge uselessly over 278'. This condensing apparatus is, how- 

 ever, uot nearly so powenul as others now in use. In the two 



' If ihe .'piibrn be of large size, more than two prisms may of course 1,? 

 substituted, 

 * Trans. Roy. Scott. Soc. of Arts, 1850. 

 3 "Lighthouse Illumination," p. 75- 



