Maiicm 16, 



3.] 



SCIENCE. 



161 



ever, that splievical aberration has little, if an_y 

 thing, to do with it, as, in lenses constructed 

 on the modern curves, this defect has been 

 practically reduced to zero. If now we take 

 a perfectl3' corrected, wide-angled lens, and 

 focus it on the centre of the plate, we shall 

 find that the objects near the edges are some- 

 what indistinct, and by no possible combina- 

 tion of curves can this difficulty be whoUj' 

 remedied ; it is, however, reduced proportion- 

 ally to the size of the stop employed. It has 

 been shown, by Prof. E. C. Pickering and Dr. 

 C. H. Williams {Proc. Amer. acad. 1875, 

 300), that, with a single lens, a series of con- 

 centric circles would be focused on a spheri- 

 cal plate whose radius of curvature was 0.7 

 the focus of the lens. On the other hand, the 

 diameters of these circles could onlj' be ac- 

 curately focused on a spherical plate whose 

 radius of curvature was 0.3 of this focus. As 

 far as the writer is aware, no name has e^-er 

 been given to this optical defect ; but for con- 

 venience' sake it might be called the field 

 aberration. 



If the central object on the plate is of the 

 most interest, we shall focus on it, and then 

 push in the ])late as far as possible without 

 injuring the central definition, to obtain the 

 best possible result at the edges. Supposing 

 now we insert a smaller stop, the definition 

 over the whole plate will be improved certain- 

 I3' ; but, that at the centre having been suffi- 

 ciently sharp before, we can now afford to 

 push in the plate a little farther still, and obtain 

 better definition at the edges, without percep- 

 tiblj- injuring that at the centre. Therefore, 

 on theoretical considerations merel}', we should 

 always focus with the stop we are going to use. 

 But, on the other hand, for lenses of less than 

 45° angle, or wlien the illumination is very 

 faint, the practical advantage of a bright image 

 for focusing would more than compensate for 

 the advantages of using the other stop. In 

 practice, for accurate work, the best waj^ 

 would be to determine once for all the differ- 

 ence of focus required bj' each stop, and then 

 focus with the largest, and appl}' the proper 

 correction, depending on the stop used. 



W. H. Pickering. 



HISTORY OF THE APPLICATION OF THE 

 ELECTRIC LIGHT TO LIGHTING THE 

 COASTS OF FRANCE.^ 



II. 



The Serrin regulator, arranged ' for alter- 

 nating currents, has been adopted as the stand- 



^ Continued from No. 5. 



ard lamp. No other apparatus has given 

 better results. Especially with alternating 

 currents, its working is excellent, because the 

 armature of the electro-magnet detaches itself 

 very easilj' ; and besides, as the consumption 

 of iDOth carbons is uniform, the arc remains 

 absolutely fixed. 



The machines fof generating the current 

 have been of late j-ears the subject of attentive 

 study, which has been unfortunatel}' confined 

 to three types, — the Alliance, Gramme, and 

 de iSIeritens. The luminous intensities of each 

 of these machines have been measured under 

 carefully arranged conditions. Photometric 

 measurements in such cases are rather delicate. 

 To make them, since the intensity varies in 

 the vertical direction with different heights, a 

 movable mirror is used, which is placed at 

 different heights in the same vertical plane, 

 and which, in each position, throws the rays on 

 the photometer ; and thus the average intensi- 

 ties could be obtained. But, as the intensity 

 of the electric light constantly- varies, it was 

 necessary to make the observations at one- 

 minute intervals for each position of the mir- 

 ror. It is not necessary here to go into the 

 details of construction of the different ma- 

 chines ; the table below gives the results 

 obtained. 



It will be seen, that the Alliance machine 

 gives a far less intensity per horse-power than 

 the two others, which are approximately equal. 

 The de Meritens has certain characteristics of 

 stabilitj' and soliditj' which the Gramme ma- 

 chine does not possess ; it was, besides, pre- 

 ferred to use alternating currents. For these 

 reasons it has been adopted, and will be in- 

 stalled in all the new lighthouses. 



The figures giving the intensity in the pre- 

 ceding table refer to the naked light. When 

 this is placed in a fixed-light apparatus, these 

 intensities become, in round numbers, 12,000 

 carcels with the Alliance, and 20,000 carcels 

 with the Gramme No. 2. The flashes increase 



