By ERNST GUNDLACH.
I wish to call attention to a few mistakes that are quite commonly made by microscopists and writers in stating the result of their optical tests of microscope objectives.
If the image of an object as seen in the microscope appears to be unusually distorted and indistinct toward the edge of the field, and satisfactory definition is limited to a small portion of the center, the cause is often attributed to the spherical aberration of the objective, while really this phenomenon has nothing to do with that optical defect of the objective, if any exists, but is caused by a lack of optical symmetry. If a perfectly symmetrical microscope objective could be constructed, then, with any good eye-piece, it would make no difference to the definition of the object were it placed either in the center or at the edge of the field, even if the objective had considerable spherical aberration. But, unfortunately, our most symmetrical objectives, the low powers, leave much to be desired in this respect, while our wide angle, high powers are very far from symmetrical perfection.
There are two causes of this defect in the latter objectives, one being the extreme wideness of their angular apertures, and the other the great difference in the distances of the object and the image from the optical center of the objectives.
Another mistake is often made in regard to the cause of certain prismatic colors that are sometimes, in a striking degree, produced by otherwise good objectives. According to the nature of these colors, whether yellow or blue, green or indigo, they are generally regarded as evidences of either chromatic over or under correction of the objective. Of course the presence of either of these defects is certainly and correctly indicated by the appearance of one or the other of the colors, under certain circumstances; but the simple visibility of prismatic color is by no means a reliable indication of over or under correction of color, and, indeed, to the honor of our opticians, it may be stated that very few objectives are made that cannot justly be called achromatic in the general sense of the term. By far the most common causes of prismatic color, in otherwise carefully constructed objectives, are the so-called chromatic aberrations of second or higher order. Every achromatic lens which is, as it should be, at its best at about two-thirds of its aperture, is inside of this ring or zone, toward the center slightly under and outside, toward the edge, slightly over corrected.
This defect is the greater, the less the difference of the dispersive powers of the two glasses used in the construction of the lens, for a given proportion of their refractive indexes, and therefore the degree of visibility of the colors of the aberrations of the second order depends greatly on the nature of the glass employed in the construction of the lens.
This defect may be corrected by a suitable combination of two or more lenses, though not without again having similarly, as in the correction of the first color, some faint remnants of color, the aberrations of third or still higher order. But even the correction of the third or still higher order may, if the angular aperture is very wide, leave quite visible and disturbing remnants of color.
Another and not uncommon explanation of the cause of this unwelcome color, though not so serious and damaging a charge to the maker of the objectives, is its attribution to the so-called "secondary spectrum." This error, like that previously mentioned, is certainly indicated by the appearance of certain colors under certain conditions, but being, as a rule, one of the least defects of even our best objectives in most cases, it is probably not the true source of the disturbance.
The secondary spectrum is very commonly confounded with the chromatic aberration of higher order. While the latter is produced by imperfections in the form of the lens, the former is due to an imperfection of the optical qualities of the material from which the lens is constructed, the crown and flint glass.
A glass prism of any angle will project upon a white surface a spectrum of any length, according to the arrangement of the light source, the screen, and the prism. So with two prisms of the same kind of glass, but of different angles, two spectra can be produced of exactly equal length, so that if one is brought over the other, with the corresponding colors in line, they will appear as one spectrum. But if one of the prisms is made of crown and the other of flint glass, then their spectra cannot be arranged so that all their corresponding colors would be in line, for the proportional distances of the different colors differ in the two spectra. If two colors of the spectra are, by suitable arrangement, brought exactly in line, then the others will be out. The two spectra do not coincide, and the result, if an achromatic lens be made of these glasses, must be a remnant of color which cannot be neutralized. This remnant is the secondary spectrum.
Although this peculiar disharmony in the dispersive powers of the two glasses, crown and flint, was discovered almost immediately after achromatism was invented, it was only recently that the first successful attempts were made to produce different glasses, which, possessing the other requirements for achromatic objectives, would produce coincident spectra, or nearer so than the ordinary crown and flint glass do. It was about twelve years ago, if my memory serves me, when I learned that a well-known English firm, engaged in the manufacture of optical glass, had brought out some new glass possessed of the desired qualities, and a little later I received a circular describing the glass. But at the same time I learned that the new glass was very soft and difficult to polish, and also that it had to be protected from the atmosphere, and further, that an English optician had failed to construct an improved telescope objective from it. I had ordered some samples of the glass, but never received any.
A few months ago, news from Europe reached this country that another and seemingly more successful attempt had been made to produce glass that would leave no secondary spectrum, and that Dr. Zeiss, the famous Jena optician, had constructed some new improved objectives from it. But the somewhat meager description of these objectives, as given by an English microscopist, did not seem fit to excite much enthusiasm here as to their superiority over what had already been done in this country. Besides this, the report said that the new objectives were five system, and also that extra eye-pieces had to be used with them. I confess I am much inclined to attribute the optical improvement, which, according to Dr. Abbe's own remark, is very little, more to the fact that the objectives are five system than to the new glass used in their construction.
After a close study of a descriptive list of the new glass, received a week or two ago from the manufacturers, I find, to my great regret, that this new glass seems to suffer from a similar weakness to that made by the English firm twelve years ago; as all the numbers of the list pointed out by the makers as having a greatly reduced secondary spectrum are accompanied with the special remark "to be protected." Furthermore, from a comparison of the dispersive and refractive powers of these glasses, as given in the list, I find that objectives constructed from them will leave so great aberrations of higher order, both spherical and chromatic, that the gain by the reduction of the secondary spectrum would be greatly overbalanced.
In conclusion, I wish to say that while I would beware of underestimating the great scientific and practical value of the endeavor of the new German glass makers to produce improved optical glass, and the great benefit accruing to opticians and all others interested in the use of optical instruments, I think it wise not to overestimate the real value of the defects of the common crown and flint glass, which I have sought to explain in this paper. And, for myself, I prefer to fight the more serious defects first, and when its time has come I will see what can be done with the secondary spectrum.
Read before the American Association, Buffalo, August, 1886.