(A) INTRODUCTION (cont.)
Compound Microscope.—The placing of two convex lenses in such relative positions that one should magnify an enlarged image of a small near object formed by the other naturally soon followed the invention of the telescope, and seems to have first occurred to Hans Zansz or his son Zacharias Zansz, spectacle-makers at Middelburg in Holland, about 1590. One of their compound microscopes, which they presented to Prince Maurice, was in the year 1617 in the possession of Cornelius Drebell of Alkmaar, who then resided ia London as mathematician to king James I. In order to make clear the successive stages by which the rude and imperfect microscope of that period has, after remaining for two centuries unimproved in any essential particular, been developed within the last half-century into one of the most important instruments of scientific research that the combination of theoretical acumen and manipulative skill has ever produced, it is necessary to explain the principle of its construction, and to show wherein lay the iniperfection of its earlier form.
In its simplest construction, as already stated, the compound microscope consists of only two lenses,—the "object-glass" CD, fig. 3, which receives the light-rays direct from the object AB placed near it, and forms an enlarged but reversed image A'B' at a greater distance on the other side, and the "eye-glass" LM, which receives the rays that diverge from the several points of this image as if they proceeded from the points of an actual object occupying the position and enlarged to the dimensions A'B', and bring these to the eye at E, so altering their course as to act as a simple microscope in magnifying that image to the observer. It was early found useful, however, to interpose another lens FF, fig. 4 (the "field-glass"), between the object-glass and the image formed by it, for the purpose of giving such a slight convergence to the pencil of rays as shall reduce the dimensions of the image, and thus allow a larger part of it to come within the range of the eye- glass, so that more of the object can be seen at once. And it was soon perceived that the eye-glass and the field-glass might be advantageously combined into an "eye-piece," in which a perforated diaphragm might be inserted at the focal plane of the image (i.e., in the focus of the eye-glass), so as, by cutting off the peripheral portion of the field of view, to limit it to what can be seen with tolerable distinctness.
It is obvious that the magnifying power of such an instrument would depend (1) on the proportion between the size of the image formed at BB and that of the actual object, and (2) upon the magnifying power of the eye-glass. And further the proportion which the size of the image bears to that of the object depends upon two factors,—(1) the focal length of the object-glass, and (2) the distance between the object-lass and the plane BB occupied by the image it forms. If we diminish the focal length of the object-glass, the object must be brought nearer to it, so that, while the distance of the image on the other side remains unchanged, that distance comes to bear a larger proportion to the distance of the object, and the size of the image is augmented in a corresponding ratio. On the other hand, the object-glass remaining unchanged, the distance at which it forms the image of the object can be increased by a lengthening of the tube of the microscope; and, as this involves a shortening of the distance between the object-glass and the object, the proportion which the former bears to the latter is augmented, and the image is correspondingly enlarged. Thus an increase in the magnifying power of the compound microscope may be gained in three modes, which may be used either separately or in double or triple combination,—viz., (1) shortening the focus of the object-glass, (2) lengthening the tube of the microscope, and (3) increasing the magnifying power of the eye-glass by shortening its focus. This, it may be remarked, also lengthens the distance of the image from the object-glass, by bringing the focal plane BB nearer the eye-glass. The second of these methods was not unfrequently used in the older microscopes, which were sometimes made to draw out like telescopes, so as to increase the amplifying power of their object-glasses. But, whilst very inconvenient to the observer, such a lengthening of the one distance involved such a shortening of the other as greatly impaired the distinctness of the image by increasing the aberrations of the object-glass, so that this method came to be generally abandoned for one of the other two.
When lenses of from 1 to 4 inches focus were used as object-glasses, and their apertures were restricted by a stop to the central part of each, tolerably distinct images were given of the larger structural arrangements of such objects as sections of wood or the more transparent wings of insects,—which images would bear a further moderate enlargement by the eye-glass without any serious deterioration either by want of definition or the introduction of colour-fringes. But when lenses of less than 1 inch focus were employed in order to obtain a higher magnifying power, the greater obliquity of the rays so greatly increased their aberrations that defective definition and the introduction of false colours went far to nullify any advantage obtainable from the higher amplification; while the limitation of the aperture required to keep these aberrations within even moderate limits occasioned such a loss of light as most seriously to detract from the value of the picture. On the other hand, the use of deeper eye-pieces to enlarge the images formed by the object-glasses not only brought out more strongly all the defects of those images, but introduced a new set of errors of their own, so that very little was gained by that mode of amplification. Hence many of the best of the older microscopists (notably LEEUWENHOEK, q.v.) made some of their most valuable discoveries by the use of the simple microscope; and the amount of excellent work thus done surprises every one who studies the history of microscopic inquiry. This was still more the case, as already stated, when the use of single lenses of very short focus was superseded by the introduction of the Wollaston doublet. And the substitution of these doublets for the single lenses of object-glasses, while the single lens of the eye-glass was replaced by a Herschel’s aplanatic doublet, and the field-glass was a convex lens whose two curves had the proportion of 1:6 (the form of least spherical aberration), constituted the greatest improvement of which the instrument seemed capable in pre-achromatic times. [Footnote 261-1]
It has been only within the last sixty years (1820-30) that the microscope has undergone the important improvement which had been worked out by Dollond in the refracting telescope more than sixty years previously,—namely, the correction of the chromatic aberration of its objectives by the combination of concave lenses of flint glass with convex lenses of crown, while their spherical aberration is corrected by the combination (as in Herschel’s aplanatic, doublet) of convex and concave surfaces of different curvatures. The minute size and high curvature of the lenses required as microscopic objectives were long considered as altogether precluding the possibility of succes in the production of such combinations, more especially as the conditions they would have to meet differ altogether from those under which telescopic object-glasses are employed. For the rays from distant objects fall upon the latter with virtual parallelism; and the higher the power required the longer is the focus given to them, and the smaller is the deflexion of the rays. In the microscope, on the other hand, the object is so closely approximated to the objective that the rays which proceed to it from the latter have always a very considerable divergence; and the deflexion to which they are subjected increases with that reduction of the focal length of the objective which is the necessary condition of the increase of its magnifying power. And thus, although the telescopic "triplet" worked out by, Dollond (consisting of a double-concave of flint glass, interposed between two double-convex lenses of crown) can be so constructed as to be not only completely aplanatic (or free from spherical aberration) but almost completely achromatic (or free from chromatic aberration), this construction is only suitable for microscopic objectives of long focus and small angular aperture, the rays falling on which have but a very moderate divergence. And though, as will presently appear, some of the early attempts at the achromatization of the microscope were made in this direction, it was soon abandoned for other plans of construction, which were Ifound to be alike theoretically and practically superior.
It seems to have been by Professor Amici, then of Modena, about 1812, that the first attempts were made at the achromatization of microscopic objectives; but, these attempts not proving successful, he turned his attention to the production of a reflecting microscope, which was a decided improvement upon the non-achromatized compound microscopes then in use. In the year 1820, however, the subject was taken up by Selligues and Chevalier of Paris, who adopted the plan of superposing three or four combinations, each consisting of a double-convex of crown cemented to a plano-concave of flint. The back combination (that nearest to the eye) was of somewhat lower power than those, placed in front of it, but these last were all of the same focus, and no attempt was made by these opticians to vary the construction of the several pairs thus united, so as to make them correct each others’ aberrations. Hence, although a considerable magnifying power could be thus obtained, with an almost complete extinction of chromatic aberration, the aperture of these objectives could not be greatly widened without the impairment of the distinctness of the image by a "coma" proceeding from uncorrected spherical aberration.
In ignorance, it would appear, of what was being done by the Paris opticians, and at the instigation of Dr Goring (a scientific amateur), Mr Tulley—well known in London as an able constructor of telescopic objectives—began, about the year 1824, to work object-glasses for the microscope on the telescopic plan. After many trials [Footnote 262-1] he succeeded, in 1825, in producing a triplet of 9/10 inch focus, admitting a pencil of 18º, which was so well corrected as to perform very satisfactorily with an eye-piece giving a magnifying power of 120 diameters. He afterwards made a similar triplet of shorter focus, which, when placed in front of the previous one, increased the angle ot the transmitted pencil to 38º, and bore an eye-piece giving a magnifying power of 300 diameters. These triplets are said by Mr Ross to have never been exceeded by any similar combinations for accurate correction throughout the field.
Having come into possession, at the end of 1826, of an objective of Chevalier’s construction, Mr J. J. Lister carefully examined its properties, and compared them with those of Tulley’s triplets; and this comparison having led him to institute further experiments he obtained results which were at first so conflicting that they must have proved utterly bewildering to a less acute mind, [Footnote 262-2] but which finally led him to the enunciation of the principle on which all the best microscopic objectives are now constructed. For he discovered that the performance of such composite objectives greatly depends upon the relative position of their component combinations,—the effect of the flint plano-coneave upon the spherical aberration produced by the double-convex of crown varying remarkably according to the distance of the luminous point from the front of tho objective. If the radiant is at a considerable distance, the rays proceeding from it have their spherical error undercorrected; but, as the source of light is brought nearer to the glass, the flint lens produces greater proportionate effect, and the under-correction diminishes, until at length a point is reached where it disappears entirely, the rays being all brought to one point at the conjugate focus of the lens. This, then, is one aplanatic focus. If, however, the luminous point, is brought still nearer to the glass, the influence of the flint continues for a time to increase, and the opposite condition of over-correction shows itself. But, on still further approximation of the radiant, the flint comes to operate with less effect, the excess of correction diminishes and at a point still nearer to the glass vanishes, and a second aplanatic focus appears. From this point onwards under-correction takes the place of over-correction, and increases till the object touches the surface of the glass. As every such doublet, therefore, has two aplanatic foci for all points between which it is over-corrected, while for all points beyond it is under-corrected, the optician is enabled to combine two or more doublets with perfect security against spherical error. This will be entirely avoided if the rays be received by the front glass from its shorter aplanatic focus, and transmitted through the back glass in the direction of its longer aplanatic pencil. By the approximation of the two doublets over-correction will be reduced, while their separation will produce under-correction; and thus, by merely varying the, distance between two such combinations, the correction of the spherical error may be either increased or diminished according to a definite rule. Slight defects in one glass may thus be remedied by simply altering its position in relation to the other,—an alteration which may be made with very little disturbance of the colour-correction. This important principle was developed and illustrated by Mr Lister in a memoir read to the Royal Society on January 21, 1830, On some Properties in Achromatic Object-glasses, aplicable to the improvement of the Microscope; and it was by working on the lines there laid down that the three London opticians Ross, [Footnote 262-3] Powell, and James Smith soon produced microscopic objectives that surpassed any then constructed on the Continent, while the subsequent adoption of the same principles by French and German opticians, as also by Professor Amici of Florence, soon raised their objectives to a corresponding level.
It has proved more advantageous in practice to make the several components of an achromatic objective correct each others’ aberrations than to attempt to render each perfect in itself; and the mode in which this is accomplished will vary with the focus and angular aperture given to each combination. Thus, while a single "telescopic triplet" answers very well for the lowest power usually made (4 inches focus), and the same plan may be used—though at the sacrifice of angular aperture—for objectives of 3 inches, 2 inches, and even 1 inch focus, the best performance of these powers requires the combination of two doublets. And, while this last system also serves for objectives, of 2/3 inch and 1/2 inch of low angle, a third component is required for giving to these objectives the aperture that renders them most serviceable, as well as for all higher powers. Instead of combining three achromatic doublets, however, many makers prefer placing in front a plano-convex of crown, and adding a third lens of crown to the doublet at the back, still using a doublet in the middle, the whole combination thus consisting of six lenses, four of crown and two of flint. Further, Mr Wenham has shown that the whole, colour-correction may be effected in the middle by interposing a double concave of dense flint between two double-convex lenses of crown,—the back lens, as well as the front, being then a plano-convex of crown, making five lenses in all. This plan of construction, though suitable to objectives of moderate angular aperture, and advantageous in regard to comparative simplicity and economy of construction, does not seem so well adapted for objectives to which the largest attainable aperture is to be given—these being usually constructed with a triplet in front, a doublet in the middle, and a triplet at the back, so as to consist of eight separate lenses. And the first-class constructors of achromatic objectives in the United States usually place in front of these, in their highest powers, a single plano-convex of crown, by the addition of which a greater working distance can be obtained. But, as every such addition increases the liability to error from imperfections in the centring and grinding of the lenses (as well as loss of light by the partial reflexion of oblique rays from their surfaces), it is obvious that the most exact workmanship, involving a proportionate costliness, is required to bring out the full effect of such complex construction. And where anorular aperture is regarded as the quality of primary importance it will be usually found preferable to have recourse to objectives constructed on either the "water" or the "oil" immersion system, to be presently described.
The great increase thus attained in the perfection of the corrections of microscopic objectives for both spherical and chromatic aberration of course rendered it possible to make a corresponding increase in their angular aperture. The minute scales of the wings of butterflies and other insects were naturally among the objects much examined; and it was soon perceived that certain lines and other markings became clearly discernible on these scales with objectives of what was then considered large angle which were utterly undistinguishable with non-achromatized microscopes (however high their magnifying power), and very imperfectly shown under achromatic objectives of small angle. Hence these scales came to be used as "test-objects," for judging of the "definition" and "resolving power" of microscopic objectives—the former property consisting in the clearness, sharpness, and freedom from false colour of the microscopic images of boundary lines, and depending on the accuracy with which the aberrations are corrected, while the latter term designates that power of separating very closely approximated markings which is now known to be a "function" of aperture. The insect-scales formerly most valued for these purposes were those of the Morpho menelaus (fig. 5) and the similarly lined scales of the Polyommatus argus (azure-blue), the "battledoor" scales of the same butterfly (fig. 6), the ribbed scales of the Lepisma saccharina (sugar-louse), and the minute and peculiarly marked scales of the Lepidocyrtus curvicollis (fig. 7), commonly known as the Podura. The writer recollects the time when the satisfactory "resolu-tion" of the first three of these tests was considered a sufficient proof of the goodness of even high-power objectives, and when the Podura-markings, if visible at all, could only be distinguished as striae. The further opening-out of the aperture, however, enabled these striae to be resolved into rows of "exclamation marks"; and, while there is still some uncertainty as to the precise structure of which these markings are the optical expression, practical opticians are generally agreed that the Podura-scale is very useful as a test for definition, with even the highest objectives, though it only serves as a test for a very moderate degree of re-solving power. For the latter purpose it has been completely superseded by the closely approximated markings of the silicified envelopes of certain diatoms (which, however, show themselves in very different aspects according to the conditions under which they are viewed, figs.8-11), and also by lines artificially ruled on glass, as in Nobert’s "test-plate," the number of lines in the nineteen bands of which is stated by M. Nober to range from 1000 to 10,000 to a Paris line, while Dr Royston Pigott gives the numbers in an English inch as 11,529 to the inch in the first band, and 112,595 in the nineteenth. This last dimension (as will afterwards appear) approaches the minimum distance at which such markings are theoretically separable by any magnifying power of the microscope.
The enlargement of the angle of aperture of microscopic objectives and the greater completeness of their corrections, which were obtained in the first stance by the adoption of Mr Lister’s principles, and were demonstrated by the resolution of the test-objects then in use, soon rendered sensible an imperfection in their performance under certain circumstances, which had previously passed unnoticed; and the important discovery was made by Mr Andrew Ross that a very decided difference exists in the precision of the image according as the object is viewed with or without a covering of thin glass, as also according as this cover is thin or thick. [Footnote 264-1] As this difference increases in proportion to the widening of the aperture, it would obviously be a source of great error and embarrassment if a means could not be found for its rectification. Its optical source, however, having been found by Mr Ross to lie in the "negative aberration" which is produced in the rays proceeding from the object to the front glass of the objective by the interposition of the plane-glass cover, and which increases with its thickness, his practical ability enabled him at the same time to indicate the remedy, which consists in under-correcting the front lens and overcorrecting the two posterior combinations, and in making the distance between the former and the latter capable of adjustment by means of a screw-collar, as shown in fig. 12. For when the front pair is approximated most nearly to the next, and its distance from the object is increased, its excess of positive aberration is more strogly exerted upon the other two pairs than it is in the contary conditions, and thus neutralizes the negative aberration produced by the interposition of the covering-glass. This correction is not needed for objectives of low or medium power and small angle of aperture; but it should always be provided when the angle exceeds 50º,—unless (as is now generally done in the case of objectives constructed for students’ use) the maker adjusts them originally, not for uncovered objects, but for objects covered with glass of a standard thickness, say 0·005 or 0·004 inch. A departure from that standard to the extent of one or two thousandths of an inch in either direction, though extremely injurious to the performance of objectives whose aperture is 125º or more, scarcely makes itself perceptible in those of 90º or 100º. And the same may be said in regard to the immersion-objectives next to be described, which are peculiarly suitable to the purposes of minute histological research.
Footnotes
261-1 This combination was made in the first microscope of which the writer became possessed, about the year 1830 ; and be well recollects the great superiority to any compound microscope of the old construction which was proved by its power of separating the lines on the Menelaus scale, and of bringing into view the details of the structure of animalcules, with a clearness that only an achromatized object-glass could surpass.
262-1 It is due to Air Joseph J. Lister to mention that Tulley’s filial success with this low power seems to have been attained by working on a suggestion given him by that gentleman. See Monthly Microscopical Journal, vol. iii. (1870), p. 134.
262-2 Thus he found that, while each of Chevalier’s doublet combinations, when used singly, presented a "bur" or "coma" outwards, this coma, instead of being exaggerated by the combination of two of these doublets, was much diminished. On the other hand, while two of Tulley’s triplets, each of which performed admirably by itself, were used together, the images of all objects not in the Centre presented a strong bur inwards with all under-correction of colour.
262-3 In 1837 Air Lister gave Air Ross a projection for an objective of 1/8 inch focus, in which a triple front was combined with two doublets. The great superiority of this lens, admirably executed by Air Ross, caused him to adopt its plait as the stalldard one for high powers; and it is still in general use—the back lens also being sometimes made as a triplet.
264-1 Trans. Soc. of Arts, vol. li.
Read the rest of this article:
Microscope - Table of Contents