Photographs of Broken Diatom Valves. By Dr. J. D. Cox. 855 



have been made with light strictly central ; for I have sought to 

 secure dioptric images and to avoid diflfraction ones as far as 

 possible. From the list of the photographs it will be seen that the 

 objects have been mounted, some of them dry, some in balsam, and 

 some in the very dense medium of Prof. H. L. Smith. Whilst 

 both the denser medium and the dry mounts have their advantages 

 for purposes such as the resolution of lined surfaces by oblique hght, 

 it should be remembered that by the exaggerated contrast in the 

 refractive power of the object and the medium, the prismatic effect 

 of a broken edge is increased. This exaggerated refraction and 

 accompanying diffraction interferes more or less with the true 

 presentation of such a marginal broken line. At first sight it 

 seems much stronger and bolder, and what is coarsely presented is 

 much more easily photographed ; but my experience leads me to 

 the opinion that the most truthful pres'entation of the details is 

 given when the difference in the refractive indices of object and 

 medium is as small as is consistent with the discrimination of the 

 object. The image will, of course, be much fainter, but I find it 

 also more delicately exact and of a better quality. Up to the limit 

 of good definition in balsam, therefore, I prefer to use mounts in 

 this medium, and the result seems to be worth the extra care and 

 nicety requu-ed in all parts of the manipulation. 



It will be seen from the description of the plates that the 

 exposure of the sensitive film was not always in proportion to the 

 amplification. _ The greater the difference in refractive index, the 

 more quickly is a negative taken, for reasons already hinted at. 

 The dry mounts and those in Prof. Smith's medium were therefore 

 photographed more quickly than the balsam mounts. Within 

 moderate limits, however, the amplification was varied whilst the 

 time of exposure and the medium remained the same. The plates 

 so taken were not of the same density, but the only important 

 resulting difference was that the denser plates printed more slowly. 

 Again, it is difficult, if not impossible, to manipulate the light so as 

 to make it entirely uniform on different evenings. The lamp may 

 be trimmed a little differently, the state of the atmosphere may 

 affect it, or the plates themselves may not be exactly alike. There 

 will therefore be, at last, room for the exercise of judgment based 

 upon experience in determining the exposure to be given, and one 

 must expect to spoil a plate occasionally. 



I will only add that it has been a fixed rule with me to leave 

 the photograph untouched. Any "stopping out," stippling, or 

 retouching in any of the forms known to practical photographers, 

 must, in my opinion, greatly diminish the value of a photograph 

 for scientific purposes by introducing more or less of personal 

 interpretation. 



