November i8, 1920] 



NATURE 



379 



immersion combinations, the refractive index of 

 the immersion fluid being 1-447. As these systems 

 are not homogeneous, the cover glasses are 

 optically worked fused quartz of uniform 

 thickness. 



The slides are also of fused quartz, fitted 

 into ;i carrier of a special type, which 

 ensures that the surface of the slide is a constant 

 distance from the objective, a point that in prac- 

 tice is of considerable importance. The quartz 

 oculars are five in number, and range from an 

 initial magnification of 5 to 20, giving camera 

 magnifications of from 200 to 3600 diameters. 

 The latter is a good deal too high for satisfactory 

 results with most objects — in fact, it is doubtful, 

 on theoretical grounds, whether such a magnifica- 

 tion is justified. The quartz sub-stage condenser 



Fic. !■— A, »park gap; B, quant c<md«niing Icr.i ; C* quart i DrUmt; D, bos contain- 

 ing rcAcctinK <iuar f pri.n ; C, fluoteKent ocular. The po»ii»on of tb« other parts 

 <l«*cribcd will la svidcat to any tnicro*copisl. 



is made with a duplex top, .so that a combination 

 is available for each objective to ensure that a 

 suitable cone of illumination is used in each case. 

 This is used as a glycerine immersion system with 

 the two higher-power objectives, and as a dry 

 system with the lowest one. 



The source of light is produced by a high- 

 fonsion discharge in air between melnl electrodes, 

 iisu;illy cadmium or magnesium, although other 

 metals may be .used if they produce a suitable line 

 ^p< ' trum. There are obvious limitations in this 

 r. spcct, as the tharactcr of the spectrum emitted 

 must be such that the principal lines in the ultra- 

 violet region are sufficiently .separated and of con- 

 siderable intrinsic brillianrv. The spectrum of 



NO. 2664, VOL. 106] 



iron, for instance, is excluded, as, although it is 

 rich in bright lines, these are so numerous and 

 therefore so close together that the isolation of 

 one Ime is impossible under the conditions realised 

 in this method. 



The spark is produced by means of an induction 

 coil of special design giving a heavy discharge of 

 relatively low potential, the equivalent spark-gap 

 being about 5 cm. This is further reduced by 

 placing a condenser immersed in oil in parallel 

 with the spark-gap. The interrupter may be 

 either an electrolytic one or a mercury break, 

 the latter appearing to be more satisfactory. 

 Special arrangements are made for accurately 

 adjusting both the length of the spark and its 

 position in relation to the optic axis of the micro- 

 scope. The image of the snark is projected by 

 means of a quartz lens, so that, after pass- 

 ing through a pair of quartz prisms of 

 opposite rotation, an image of the spark in 

 one wave-length is obtained approximately 

 at the position of the iris diaphragm below 

 the sub-stage condenser. To facilitate ad- 

 justment, a disc of uranium glass is placed 

 at the latter position so that an image of 

 the .spark can be observed and focussed as 

 required, after which the uranium glass in 

 its carrier is swung aside. The direction 

 of the illuminating beam is at right angles 

 to the optic axis of the microscope; it has, 

 therefore, to be reflected by a right-angled 

 quartz prism along the axis in the same 

 way that the mirror operates in an 

 ordinary microscope. 



The preparation being placed on the 

 stage, the light adjusted, and the con- 

 denser accurately focussed on the object, 

 the actual focussing of the image has to 

 be carried out. This is effected by means 

 of a fluorescent searcher eye-piece which 

 is mounted above the quartz ocular, and 

 by the use of which an image is seen on 

 a fluorescent screen and focussed by 

 means of an auxiliary magnifier. This 

 operation is one of considerable difiiculty, 

 and only after long practice can success 

 be assured. Its difiiculty varies, too, 

 according to the wave-length used ; in 

 some cases the fluorescent image is 

 bright, but in others it is much 

 more difficult to see. Some objects themselves 

 fluoresce, with the result that a sharp visual image 

 cannot be obtained. The method now largely 

 adopted by the writer is to observe the object by 

 monochromatic light as emitted by a quartz 

 mercury vapour lamp. This illuminant has bright 

 lines in the violet, blue, green, and orange 

 regions, and by means of suitable screens any one 

 of these can be transmitted. 



The image having been focussed visually in one 

 of these lines, the fine adjustment of the micro- 

 scope is moved bv a predetermined amount so that 

 the image is in focus for any desired wave-length 

 in the ultra-violet. This method is quite prac- 

 ticable provided that the fine adjustment of the 



