Nov. II, 1886J 



NATURE 



43 



It is essential to the value and significance of the photomeiry 

 that simultaneous electrical measurements should be made. 



The possible variations in the coupling of the magneto- 

 electric machines, in the rate of running, anl in the nature, 

 form, and adjustment of the carbons, present a wide field of 

 experimenting. 



The continuance of the experimental working for many months 

 will serve for a trial of the trustworthiness of the De Meriten's 

 apparatus for lighthouse service. 



(4) The cost of maintaining a lighthouse supplied with gas has 

 been very variously estimated. It must vary from place to place, 

 especially with the price of coal. The actual working expenses 

 of oil lighthouses on the English coast, and gas lighthouses on 

 the Irish coast, with allowance for the price of coal and labour, 

 should furnish trustworthy data for a comparison. But to com- 

 plete these data the quantity of light produced and utilised in 

 each case needs also to be known, and as both the oil and gas 

 burners tried hitherto at the South Foreland are of the service 

 kind, the photometry now in progress will supply this knowledge. 

 Some information may also be gathered from the expenditure on 

 each illuminant at the outset and during the course of the 

 experiment. It should be possible to state, if it were desired 

 to maintain on the South Foreland a light of 50 or 100 or 

 200 thousand candles, what its annual cost cost would be with 

 each illuminant. 



(5) Although the action of lenses is mainly calculable, and, so 

 far, does not require trial, it is modified by two quantities which 

 vary slightly, namely, the reflection and absorption of light by 

 glass, and is aft'ected to some extent by errors of workmanship. 

 It will, therefore, be of interest to obtain an exact comparison 

 between the light emitted by a naked flame, and that from the 

 same flame concentrated by different types of lighthouse lens. 

 The prediction of the effect of a lens is less possible when the 

 illuminant is of large size ; and the failure of lenses, constructed 

 for use as a revolving light with gas, to utilise (except by 

 broadening the beam) the light produced at a distance of more 

 than two or three inches from the focus of the lens, if they are 

 found to fail so far, may be worth demonstrating. 



With the electric light a very close correspondence should be 

 found between the calcidated effect of the cylindrical lens and of 

 the condensing prisms and the results of photometry. 



(6) The measurements which have been made of multiform 

 gas and oil may be taken to show that any number of lights at a 

 given distance cause so many times the illumination which one 

 light causes ; or assuming that the above must be the case, and is 

 involved in the conception of comparative illumination, the pro- 

 portional variation of the photometric results with the changes 

 from uniform to biform, cS:c., on clear nights, gives evidence of 

 the trustworthiness of the photometric methods. 



(7 and 8) The two questions, which I have numbered thus, can 

 hardly be treated separately, since clearness difters only in degree 

 from slight haze, and slight haze from fog. Together they con- 

 stitute the chief object of this inquiry. 



The ob.'ervations of the experimental lights which have been 

 made from a distance, may be expected to yield, when they are 

 collected and compared, much information as to the distances at 

 • which the several lights have been seen in various weather. But 

 the changes which are necessary when photometric testings are to 

 be made, or when an extensive programme is to be exhibited, 

 must to some extent have interfered with the observation of the 

 changes due to variations in the transparency of the atmosphere. 

 It might be well for at least one month, to show the same lights 

 nightly, and to inform the distant observers that this was about 

 to be done, in order that their observations might have the more 

 value. A single light of each kind shown through the cylindrical 

 lenses would serv'e as well as multiform lights, and it would be 

 best to use that size of gas-burner which had through the lens 

 equal illuminating power with the six-wick oil lamp. Unless, 

 indeed, it is assumed, as I should be inclined to assume, that 

 equal lights from gas and oil have the same power of penetrating 

 haze ; in which case it would be more instructive to show from 

 lo>er A a single electric light supplied from one machine, and 

 from B and C either gas or oil also single, and either oil or gas 

 of such size and number as to have at close quarters on a clear 

 night an illuminating power equal to that of the electric light. A 

 sufficient series of distant observations of these lights would show 

 (I ) whether the electric light maintained its equality with the 

 larger hydrocarbon flame through slight haze, or became more 

 nearly equal to a flame of much less initial brightness ; and (2) 

 whether the taller beam of multiform oil or gas had much 



advantage over tlie beam sent forth from a single lens. I believe 

 it will be found that the relative brightness of two, or more, to 

 one, will be maintained at any distance and through any haze 

 which permits of photometry, but that, when the single light is 

 lost at 5 miles or 500 yards, the triple light will be invisible at 6. 

 The actual figures corresponding to these conjectural figures must 

 be found, and the Trinity House Committee will then be able to 

 judge in what cases such an extension of range is worth the 

 increased expenditure. 



In the case of the electric light, the observations which have 

 already been made show that it loses in haze a larger proportion 

 than the hydrocarbon flames. Further observations on this 

 point will be of much interest and importance. The most 

 valuable are observations of the distances at which an electric 

 and a gas or oil light, whose relation in clear weather is known, 

 cease to be visible. Such observations are strictly photometric 

 observations, in which the lights observed are brought to an 

 equality of minimum appreciable brightness, and the distances 

 at which their brightness is equal are measured. These are 

 dependent upon the weather, and may be practicable on only a 

 few days in each month. Still more rarely will the opportunity 

 offer of measuring the lights in hut No. i through a slight uni- 

 form mist ; but such measurements ought to be made. I 

 would suggest the possibility of testing in the photometric 

 shed through an artificial mist produced by blowing steam from 

 the boiler in the adjoining engine-house into the middle of 

 the shed. 



It is said that Faraday proposed at first the use of a very small 

 lens with the electric light. Unless conclusive experiments have 

 been made on this point, it may be well to place the experi- 

 mental electric light in the focus of a larger and of a smaller 

 annular lens, each subtending the same angle, and to note 

 whether the effect differs. 



It might also be worth trying whether biform gas, with a small 

 enough number of jets to have the same illuminating power as 

 single oil, would be better seen through slight haze. The trial 

 would not be between gas and oil, but between placing a strong 

 light behind one lens, and placing half the light behind each of 

 two superposed lenses. 



(9) Owing to the nearness of the three lanterns, the illumi- 

 nation or halo which spreads round each of them in a fog seems 

 almost to blend. That which surrounds the gas lantern is not 

 much greater than that around its neighbours on either side. I 

 do not think that, on the one occasion on which I have seen the 

 lights in a foj, the ex-focal light was of much service. 



If a lighthouse lantern was surrounded by a mist or cloud ex- 

 tending far enough laterally to extinguish its principal beam, but 

 so little above it as to allow the scattered light to fall upon 

 a higher stratum of cloud, the position of the lighthouse might 

 only be seen from the illumination of the cloud above it. But 

 this state of things would happen rarely in most places, and a 

 better plan of turning it to account than the addition to a burner 

 of rings of ex-focal flame would be to employ the upper prisms 

 to send a second beam skyward. Whether the general illumina- 

 tion about the experimental lanterns has been visible when the 

 three centres of light were not visible to an observer towards 

 whom the beams were directed, may perhaps be gathered from the 

 record of observations. 



(10) When engaged on a clear night in judging of the experi- 

 mental lights, the eye of the observer is continually caught by 

 the sudden flashes of the Calais light. The revolving light at 

 Grisnez is equally visible, but does not catch the eye in the 

 same manaer. It might be well to try on some rather hazy 

 night, whether, if one lantern alone were lighted, and during 

 successive quarters of an hour the light were alternately kept 

 steady and flashed in sDme such groups of flashes as the Calais 

 light, the observers patrolling the line of posts became aware of 

 the light at a greater distance when it was flashed than when it 

 was steady or revolving. Even a slight mist is a great leveller 

 of distinctions, but it seems possible that the use of flashing may 

 increase the range of a light as much as an addition to its inten-ity 

 or size. 



Some of the questions raised in the latter part of this report 

 might perhaps have been omitted, as having already received an 

 answer, if, while thinking the matter over, I had been able to 

 consult some of the experienced members of the Trinity House 

 Committee who are charged with the conduct of this inquiry. I 

 have ventured here in writing, as at other times by word of 

 mouth, to make the suggestions which have occurred to me, 

 knowing that they will receive friendly attention if they are 



