Living muscle-fibres from freshly killed mice were mounted in isotonic saline /protein media and examined with a Smith interference microscope, usually with a white light-source. When the A or I bands near the edge of a fibre were observed to match the colour of the background field, their refractive indices were close to that of the mounting medium; although it is extremely probable that diffraction at the edges of the adjacent unmatched bands affected their apparent match, so that they were not exactly of this refractive index. Matched A bands were distinguished from matched I bands by examining them in plane-polarized light, by rotating the preparation through a right angle under the interference microscope to display their birefringence, and from the colour of the unmatched bands when the fringe system of the microscope was left unaltered.

In any one fibre, the refractive indices of the 71-band regions were always higher than that of the Z-band regions. The H bands had lower refractive indices than the A bands, and the Z bands higher than the I bands, but these were both too narrow to be matched satisfactorily by this method. The refractive indices of the solutions in which matched I bands were found ranged from 1·358 to 1·363, and those in which matched A bands were found from 1·360 to 1·366. The mean refractive index of the A and I bands was very close to 1·363, which is equivalent to a total solid content of 16% w/v.

These findings are in good general agreement with those of H. Huxley and Hanson (1957) and Bennett (1955), who measured the distribution of solid material in isolated glycerinated mammalian myofibrils; but the difference between the refractive indices of the A- and Z-band regions of the living fibres appeared to be very much less. Only part of this discrepancy can be accounted for by the presence of non-fibrillar solid material, because the total amount of this is extremely unlikely to exceed 50% of the total myofibrillar fibrous protein. It therefore seems probable that, because of the diffraction from the unmatched bands, the true refractive indices of the A bands were higher than those of the solutions in which they appeared matched, and those of the Z bands were correspondingly lower than those of the solutions in which they appeared matched.

The maximum error involved here (when the sarcomere interval was approximately 2·5 μ) can be quantified from independent estimations of the non-fibrillar material in whole muscle (Szent-Gyôrgyi and others, 1955; Hanson and H. Huxley, 1957); and from this it seems highly probable that the refractive indices of the I bands were not lower than 1·350 (equivalent to a solid content of 9% w/v), and those of the A bands were not higher than 1·375 (equivalent to a solid content of 23% w/v).

The development, in the last 8 years, of interference microscopes capable of detecting and measuring very small phase differences, together with immersion techniques for the refractometry of living cells, has enabled very precise measurements to be made of the solid concentrations of the cytoplasm and inclusions in many living cells.

Hitherto, however, these techniques have not been applied to the problem of measuring individually the total solids in the different regions of living mammalian striated muscle-fibres, although A. Huxley and Niedergerke (1958) have obtained values for the refractive index of the A- and Z-band regions of living striated muscles in the frog in the course of their notable studies (by interference microscopy) of their contraction, and H. Huxley and Hanson (1957) have made similar measurements on the different regions of glycerinated mammalian myofibrils.

There are two major difficulties in an investigation of this kind, both arising from the nature of mammalian muscle. First, the tissue itself is very fragile, and it is extremely difficult to isolate and separate single fibres without damaging them at some point along their length, and it is of first importance to ensure that the segment of the muscle on which measurements are being made is in a completely viable condition. Secondly, the very close spacing of the striations introduces optical artifacts and errors in the measurement of phase differences by interference microscopy. The investigation described below was undertaken with the aim of meeting these difficulties and obtaining fairly precise values for the refractive indices (and hence the dry solid content) of the different regions that are distinguishable in visible light in the normal striated muscle of the mouse.

The measurements were made by the matching method of immersion refractometry developed by Barer and Ross in 1952, and fully described by Barer and Joseph (1954, 1955), except that the material was examined by interference microscopy and not by phase contrast. Isolated fibres were mounted in a range of buffered isotonic protein solutions of different refractive indices until one was found in which the I- or A-band regions of the fibre exactly matched the background field in colour if white light was used (fig. 1), or in intensity if nearly monochromatic mercury light was used (fig. 3). The refractive index of the matched region was then close to that of the mounting medium. These values could then be interpreted in terms of water and solid content, since, as was pointed out by Davies and Wilkins (1951, 1952) and by Barer (1952), the specific refraction increments of nearly all proteins and amino-acids, the principal constituents of cytoplasm and muscle-fibres, are very close to 0-0018. Thus, values for the dry solid content of the Z-band regions could be obtained to within 1 % of solid, and nearly as accurate values could be obtained for the weakly birefringent A-band regions.

FIG. 1.

(plate). Colour-photomicrographs of segments of living muscle-fibres under a Smith interference microscope with a tungsten light source without a filter. All the photomicrographs were taken with a 2-mm shearing objective and are to the same scale, indicated at the bottom of the page. The reference area in every case is on the left of the photograph and the fibres are all orientated with their long axes in the north-south direction in the field. A, a fibre mounted in a saline /protein solution with a refractive index of 1 · 360, with the analyser set to give a yellow background. The I bands near the edge of the fibre show the same colour as the background. B, the same preparation as in A, with the analyser set to give a red background. The I bands again match the background colour. c, a similar preparation to that in A and B, mounted in a saline I protein solution with a refractive index of 1 · 360. The analyser is set to give a blue background and the I bands at the edge of the fibre match this colour. The unmatched A bands show up in yellow-green, which indicates a shift up the Newtonian series. D, the same preparation as in c with the analyser set so that the background and matched I bands show up in red. The unmatched A bands are now blue, which indicates a similar colour shift. E, a preparation of a fibre mounted in a saline /protein solution with a refractive index of 1 · 365. The fringe adjustment of the microscope condenser was the same as in c and D. The analyser was set to give a blue background and the A bands match this colour. The unmatched I bands show up in indigo, which indicates a shift down the Newtonian series, and a phaseshift in the opposite direction to that of the unmatched bands in c. F, the same fibre as in E, with the analyser set so that the background and matched A bands show up in purple-red. The unmatched I bands are orange-yellow, which indicates a phase shift in the same direction as in E. G, a fibre mounted in a saline /protein solution with a refractive index of 1 · 360. The muscle-bands are exactly aligned with the optical axis of the microscope, so that the I bands match the red background colour in the centre as well as at the edge of the fibre. H, a much-stretched fibre, mounted in a saline /protein solution with a refractive index of 1 · 360. In the regions of greatest stretch the I bands do not match the red background colour but are yellow, which indicates that they have a lower refractive index than the mounting medium.

FIG. 1.

(plate). Colour-photomicrographs of segments of living muscle-fibres under a Smith interference microscope with a tungsten light source without a filter. All the photomicrographs were taken with a 2-mm shearing objective and are to the same scale, indicated at the bottom of the page. The reference area in every case is on the left of the photograph and the fibres are all orientated with their long axes in the north-south direction in the field. A, a fibre mounted in a saline /protein solution with a refractive index of 1 · 360, with the analyser set to give a yellow background. The I bands near the edge of the fibre show the same colour as the background. B, the same preparation as in A, with the analyser set to give a red background. The I bands again match the background colour. c, a similar preparation to that in A and B, mounted in a saline I protein solution with a refractive index of 1 · 360. The analyser is set to give a blue background and the I bands at the edge of the fibre match this colour. The unmatched A bands show up in yellow-green, which indicates a shift up the Newtonian series. D, the same preparation as in c with the analyser set so that the background and matched I bands show up in red. The unmatched A bands are now blue, which indicates a similar colour shift. E, a preparation of a fibre mounted in a saline /protein solution with a refractive index of 1 · 365. The fringe adjustment of the microscope condenser was the same as in c and D. The analyser was set to give a blue background and the A bands match this colour. The unmatched I bands show up in indigo, which indicates a shift down the Newtonian series, and a phaseshift in the opposite direction to that of the unmatched bands in c. F, the same fibre as in E, with the analyser set so that the background and matched A bands show up in purple-red. The unmatched I bands are orange-yellow, which indicates a phase shift in the same direction as in E. G, a fibre mounted in a saline /protein solution with a refractive index of 1 · 360. The muscle-bands are exactly aligned with the optical axis of the microscope, so that the I bands match the red background colour in the centre as well as at the edge of the fibre. H, a much-stretched fibre, mounted in a saline /protein solution with a refractive index of 1 · 360. In the regions of greatest stretch the I bands do not match the red background colour but are yellow, which indicates that they have a lower refractive index than the mounting medium.

FIG. 3.

(plate). Photomicrographs of segments of living muscle-fibres under a Smith interference microscope, with a tungsten light-source and an Ilford 807 mercury-green filter. A-c under a 4-mm shearing objective, scale as in B. D-E under a 2-mm double focus objective, scale as in D. A, a fibre, 75 μ in width along the line a-b, mounted in saline. The analyser set to give maximum background extinction. Four dark fringes are visible between the edge and middle of the fibre. This indicates a phase change of approximately 4 wavelengths in the light passing through its thickest part. B, a similar fibre mounted in a saline protein medium of refractive index 1-360 with the analyser set as before. One set of bands matches the background field. c, the same preparation as in B, with a different analyser setting. The same set of bands matches the background field, now bright. D, another fibre mounted in a saline /protein medium with a refractive index of 1 · 360 under a 2-mm double-focus objective, showing the narrow H bands. The I bands match the background intensity. The A bands show up dark, which, at the analyser setting here used, indicates that they have a higher refractive index than the I bands. The H bands are paler than the A bands. This indicates a lower refractive index than the A bands. E, a similar fibre under the same objective mounted in a saline I protein medium with a refractive index of 1 · 364. The A bands of the fibre, orientated with its long axis in the northsouth direction in the microscope field, match the background. This indicates that the lower of these two refractive indices was equal to that of the mounting medium. F, the same region of the fibre shown in E, orientated in the east-west direction in the microscope field. The A bands no longer match the background, which clearly demonstrates their anisotropic nature.

FIG. 3.

(plate). Photomicrographs of segments of living muscle-fibres under a Smith interference microscope, with a tungsten light-source and an Ilford 807 mercury-green filter. A-c under a 4-mm shearing objective, scale as in B. D-E under a 2-mm double focus objective, scale as in D. A, a fibre, 75 μ in width along the line a-b, mounted in saline. The analyser set to give maximum background extinction. Four dark fringes are visible between the edge and middle of the fibre. This indicates a phase change of approximately 4 wavelengths in the light passing through its thickest part. B, a similar fibre mounted in a saline protein medium of refractive index 1-360 with the analyser set as before. One set of bands matches the background field. c, the same preparation as in B, with a different analyser setting. The same set of bands matches the background field, now bright. D, another fibre mounted in a saline /protein medium with a refractive index of 1 · 360 under a 2-mm double-focus objective, showing the narrow H bands. The I bands match the background intensity. The A bands show up dark, which, at the analyser setting here used, indicates that they have a higher refractive index than the I bands. The H bands are paler than the A bands. This indicates a lower refractive index than the A bands. E, a similar fibre under the same objective mounted in a saline I protein medium with a refractive index of 1 · 364. The A bands of the fibre, orientated with its long axis in the northsouth direction in the microscope field, match the background. This indicates that the lower of these two refractive indices was equal to that of the mounting medium. F, the same region of the fibre shown in E, orientated in the east-west direction in the microscope field. The A bands no longer match the background, which clearly demonstrates their anisotropic nature.

An interference microscope was used instead of a phase contrast microscope for these measurements, for two reasons.

(1) With phase contrast the ‘halo’ artifact, inherent in the optical system, that surrounds the image of all unmatched objects, made it impossible, because of the close spacing of the striations, to obtain a reliable match for any of the individual regions, while with the interference microscope this particular optical artifact could be eliminated.

(2) By using the interference microscope with a white light-source, it was possible to obtain valuable supporting evidence for distinguishing the A- and /-band regions from the colour of the unmatched bands relative to the background.

Material

For nearly all the experiments described below, the material used consisted of fibres from the gastrocnemius muscle of a freshly killed mouse (Mus musculus), although in the first few experiments the pectoralis major muscle was also used with entirely comparable results. Both adult and immature mice were used in approximately equal numbers without any difference being found in their muscle-fibres, and fibres from mice of different laboratory stocks were similarly indistinguishable.

Preparation of the specimen

At first, considerable difficulty was found in isolating the fibres in a viable condition for microscopical examination. Attempts were made to emulate the extremely elegant and delicate preparations of single fibres dissected out along their entire length that A. Huxley and his collaborators have made from the semitendinosus muscle of the frog, Rana temporaria (Huxley and Niedergerke, 1954; Huxley and Taylor, 1955), but it was found that in the time necessary to do this (frequently several hours), the mammalian fibres at room temperature invariably deteriorated completely, and a quicker method had to be devised. However, when a portion of a muscle was very rapidly removed from a freshly killed mouse and mounted (in saline or a suitable protein medium) and examined within 5 min of the death of the animal, such preparations were found to contain, together with many obviously damaged fibres, some that appeared in every way intact over several millimetres of their length. Occasionally such fibres occurred at the edge of such a preparation in a position adjacent to a suitable background reference area, so that they could be measured interferometrically, and a few of the initial measurements recorded here were made on fibres prepared in this manner by Dr. Eva Cairns. It was, however, a chancy method, and a much more reliable modification of this technique was later devised by Dr. J. Logothetopoulos, which will now be described in detail.

No anaesthetic was used to kill the mice lest it should affect the tissues. Instead, they were first stunned by a blow on the head and immediately killed by separating the neck vertebrae. The whole gastrocnemius muscle of one leg was then rapidly removed and place in a Petri dish of ice-cold 0 · 9% saline.

A portion of the belly of the muscle about f cm long and 20 or 3 mm wide was then removed by means of a clean cut with fine dissecting scissors in a direction parallel to the line of the fibres (fig. 2, A). The ends of this portion were then cut off squarely (fig. 2, B), and the remaining rectangular piece was gently separated along the natural cleavage lines between the fibres into several smaller bundles about 12 mm in width by means of fine dissecting needles (fig. 2, c). Still smaller bundles, each containing only a very few fibres or single fibres, were then separated out in a similar manner at one end of each of these pieces of tissue and spread out fanwise with the needles (fig. 2, D). The pieces thus prepared were immediately transferred to an excess of their mounting medium in a watch-glass and from there to a drop of the mounting medium on a slide, after which they were covered with a coverslip and at once examined. With a certain amount of practice, such preparations could be made in less than 4 min from the time of the death of the mouse; they contained a very high proportion of apparently undamaged lengths of individual fibres. In fact, out of approximately 50 preparations so made, less than half a dozen contained no intact fibres at the edge of the preparation suitably placed for making interferometric measurements, and all of them contained intact fibres in some place. It was found that the important thing was to avoid excess teasing, because very thoroughly separated fibres were usually damaged, and it was frequently possible to make completely valid observations on an edge of a fibre that was still in contact with its neighbours (fig. 3, B, c).

FIG. 2.

Illustrations of the rapid dissection technique developed by Dr. J. Logothetopoulos for making viable preparations of muscle fibres. A, a portion being cut from the belly of the muscle. B, the same portion with its ends cut off, leaving an approximately cylindrical bundle of parallel fibres, c, this bundle being separated by dissecting needles into smaller bundles about } mm in diameter. D, the final preparation ready for mounting: one of the small bundles ‘fanned out’ by further gentle separation with dissecting needles into bundles of only a few fibres or single fibres. Stages A-c were performed in a Petri dish of ice-cold saline. Stage D was done on the microscope slide in a large drop of the protein /saline mounting medium. The processes illustrated took about 2 min to perform. (Redrawn from a rough sketch by Dr. J. Logothetopoulos.)

FIG. 2.

Illustrations of the rapid dissection technique developed by Dr. J. Logothetopoulos for making viable preparations of muscle fibres. A, a portion being cut from the belly of the muscle. B, the same portion with its ends cut off, leaving an approximately cylindrical bundle of parallel fibres, c, this bundle being separated by dissecting needles into smaller bundles about } mm in diameter. D, the final preparation ready for mounting: one of the small bundles ‘fanned out’ by further gentle separation with dissecting needles into bundles of only a few fibres or single fibres. Stages A-c were performed in a Petri dish of ice-cold saline. Stage D was done on the microscope slide in a large drop of the protein /saline mounting medium. The processes illustrated took about 2 min to perform. (Redrawn from a rough sketch by Dr. J. Logothetopoulos.)

It was, of course, most unlikely that any of the fibres so prepared were in fact intact over their entire length, but, for reasons discussed below (p. 234), it seems probable that the portions on which observations were made were in a viable physiological state at the time of their preparation and for a short time afterwards. All the measurements and photographs here recorded were made without delay, and never later than 6 min after the death of the mouse, although the fibres themselves frequently showed no visible signs of necrotic changes for 20 min or longer. The final deterioration of a fibre was often preceded by its undergoing a slow contraction or exuding blebs of homogeneous material on its surface, after which the striations usually became faint or disappeared, and irregular areas of different phase-change spread across the preparation.

Preparation of the mounting media

The protein solutions used as mounting media in order to match the refractive indices of the different regions of the muscle-fibres were made from Armour’s bovine plasma albumin, fraction V, dissolved in saline. The dried albumin powder contains a small amount of salt, and also itself ionizes in aqueous solution. Ross, in 1952, studied its effect on the shrinkage andswelling of spherical cells and concluded that its solutions exert an osmotic pressure equivalent to a dry salt content of 12 to 1%. Thus a 40% solution in distilled water had a tonicity approximately equal to that of a 0·3% NaCl solution, and a solution of this concentration isotonic with mammalian blood (0·9% NaCl) could be made by dissolving the powder in a 0·6% NaCl solution, and this could be diluted with 0·9% NaCl to any concentration required. All these solutions, however, were acid (about pH 5·0), and although this does not appear to affect many living cells when these media are used for immersion refractometry, it was found that they frequently caused the muscle-fibres to go into a state of tonic super-contraction. Consequently a solution of sodium bicarbonate was used as a buffer, and a 40% solution of the bovine plasma albumin in a 0·6% NaCl was diluted with a 1·3% NaHCO3 solution, which has a tonicity equivalent to that of a 0·9% NaCl solution. At dilutions below 20% of protein, such as were suitable for the immersion refractometry of the muscle material, these solutions had a pH between 6·8 and 7·2 (by Beckman meter), and in them the fibres nearly always appeared relaxed and susceptible to measurement. Thus the tonicity and pH of the immersion media were adjusted to approximate to physiological conditions. The refractive indices of the solutions, which ranged from 1·35 to 1·37, were all measured with a Bellingham and Stanley pocket refractometer, with a built-in yellow filter with a transmission equivalent to the mean of the two sodium lines (589 m/4).

Microscopic examination and orientation of the specimens

The fact that striated muscle-fibres contain A-band regions that are appreciably birefringent means that the matching of such regions with the background field depends not only on the refractive index of the mounting medium, but also on the plane of vibration of the light passing through them. This, at first sight, would suggest that an interference microscope that does not itself use a polarizing optical system, such as the Dyson interference microscope manufactured by Cooke, Troughton, and Simms of York, would be better for making measurements on a material of this kind, because it would be possible to measure both the refractive indices of these anisotropic regions by the introduction and orientation of a suitable polarizer. A Dyson microscope was in fact used with some success on one occasion. However, it is a difficult instrument to set up and adjust very rapidly, and the need for speed in getting the specimens ready for examination necessitated the use of an instrument that was simpler to operate, and the Smith double-refracting interference microscope was found to be considerably more suitable.

Two Smith interference microscopes were used in the course of the work, one manufactured by Messrs. Charles Baker of Holborn (the Baker interference microscope) and one manufactured, under licence, by the American Optical Co. of Buffalo (the A. O. Baker interference microscope). The latter had a rotating stage for the easy orientation of the specimen, while the former was capable of being rapidly converted into a plane-polarizing microscope for distinguishing the A and I bands. Although both instruments thus possessed different advantages, they were optically identical and quite capable of measuring the different refractive indices of the birefringent A-band regions when the fibres were correctly orientated in the microscope field. This is because their condensers and objectives are fixed so that the plane of vibration (electric vector) of the ‘ordinary’ object beam is in the ‘north-south’ direction of the microscope field (as normally viewed by an observer from behind the instrument) in the case of all the ‘shearing’ objectives and the 16 mm and 4 mm ‘double-focus’ objectives; and in the ‘east-west’ position in the case of the 2-mm ‘double-focus’ objective (Smith, 1958).

In striated muscle, the plane of vibration (electric vector) of the faster ordinary ray in the (positively birefringent) A-band regions is in the direction at right angles to the long axis of the fibre, so that if the fibre is orientated in the ‘north-south’ direction in the field and observed with any objective other than the 2-mm double-focus, it is possible to measure the higher of the two refractive indices of this region. Conversely, with the 2-mm double-focus objective, a fibre so orientated would enable the lower of the two refractive indices to be measured. In practice, it was hardly necessary to measure both refractive indices, since the birefringence of the H-band regions is actually very small, and the difference is barely within the limits of experimental accuracy when translated in terms of water and solid content (see p. 238). Nearly all the measurements were made with 2-mm and 4-mm shearing objectives, and this necessitated a north-south orientation of the fibres because the east-west direction of the shearing systems would otherwise give overlapping images. Thus, most of the measurements made on the A-band regions were of the higher of its two refractive indices. The Z-band regions, of course, presented none of these difficulties, and if matched to the background, remained so in whatever direction they were orientated.

Preliminary measurements

Approximate values for the refractive index of the striated muscle material were obtained from the examination, by interference microscopy, of isolated single fibres mounted in 0·9% NaCl, because these were of sufficient thickness to retard the light passing through them by several wavelengths when they were mounted in a medium with a refractive index close to that of pure water, and this retardation was visible in the form of a series of interference fringes. One such fibre which had a diameter (measured with an eyepiece micrometer) of 52 μ, when viewed in nearly monochromatic light (obtained by using an Ilford 807 mercury green gelatine filter with a tungsten light-source), showed a total of 3 dark longitudinal fringes across the width of the fibre in both the A- and Z-band regions between its edge and middle line. This indicated a phase retardation of approximately 3 wavelengths through its thickest part; and this meant that, if the thickness of the fibre in the direction of the optical axis of the microscope was equal to its width at right angles to this, an approximate value for the refractive indices of both the A- and Z-band regions, n, could be obtained from the formula
formula
where ϕ = the phase change in the middle of the fibre in wavelengths (about 3), λ = the mean wavelength of the light used (0 · 54 μ approximately), t = the measured width of the fibre (52 μ), and m = the refractive index of the saline mounting medium (1 -334). This gives an approximate value of 1 · 366 for the refractive index of the muscle material. Similarly, in another, partly contracted, fibre shown in fig. 3, A, a total of 4 dark bands (alternating with 4 bright bands) can be discerned between the edge and middle line of the fibre along the line A-B, where the width of the fibre is 75 μ. This gives an approximate value of 1 · 363 for its refractive index.

With these data as a guide, similar muscle-fibres were then mounted in a protein medium with a refractive index of 1 · 360, and it was found that, under the interference microscope, one set of bands did in fact match the background in colour at all settings of the analyser when white light was used, and in intensity when nearly monochromatic green light was used. Fig. 3, B, c shows one such fibre in green light at two different analyser settings; the latter with the intensity of the background field, and matched bands adjusted so as to give maximum extinction. In the course of just over 50 subsequent experiments, matched bands were visible in 40 preparations that had been mounted in media with refractive indices between 1 · 358 and 1 · 366. In the remaining preparations, which were mounted in media with refractive indices just above and below this range, no matched bands were seen. This meant that the refractive indices of the I- and A-band regions all lay close to this range of values; and it was obviously necessary that they should be distinguished from one another and, if possible, measured individually.

Distinguishing the I-and A-band regions

The certain recognition of the I- and A-band regions in the fibre was not quite as simple as might at first appear, since neither their relative widths (which were usually about equal) nor their appearance in ordinary unpolarized white light provided reliable criteria.

Three reliable methods, however, were found; and all of them were used, at different times, for distinguishing the bands.

The most obvious method was to use the appearance of the different regions in plane-polarized light. When the fibres were placed between crossed polaroids giving background extinction, their A bands showed up brilliantly. It was necessary first to find a matched band under the interference microscope that was in some way distinctive or individually recognizable by means of some sort of marker. The condenser of the microscope was then removed and replaced with an ordinary Abbe condenser, and the interference objective was replaced with a Watson 2-mm fluorite objective in another objectiveholder. The quarter-wave plate was removed and the polarizer and analyser of the microscope adjusted to give extinction. By this means, matched I bands were demonstrated in several preparations mounted in a medium with a refractive index of 1·360; and matched A bands were similarly recognized in a medium with a refractive index of 1·365. The disadvantages of this method were that it was not always easy to find a band that was clearly recognizable with both optical systems, and also the necessary delay of about 2 min in changing the lenses and readjusting the microscope involved some risk of the specimen deteriorating.

The second method depended on the fact that the /-band regions, being isotropic, will continue to appear matched to the background however the fibre is orientated in the microscope field, while the anisotropic A bands, if they should be matched when lying in one position, will not be so when they are turned at right angles to it. Fig. 3, E shows a fibre mounted in a medium with a refractive index of 1 · 364, and viewed in nearly monochromatic green light with a 2-mm double-focus interference objective. It is orientated with its axis in a ‘north-south’ direction in the microscope field; and the A bands match the intensity of the background field, which means that the lower of their two refractive indices was 1 · 364. Fig. 3, F shows the same region of the same fibre rotated through a right angle, so that its axis is now east-west; and the A bands that were matched are now appreciably darker than the background field.

The main disadvantage of this second method was that it was necessary to use double-focus interference objectives, since only these would permit the rotation of the preparation through a right angle without its being masked by a double image: and the use of these objectives for making measurements on rather large objects like muscle-fibres is open to serious criticism on account of the ‘halo’ artifact produced by the superimposed out-of-focus secondary image. It is therefore quite possible that matches obtained on the musclebands with these objectives are subject to error; although such errors cannot be very large, since all the measurements so made were within the limits of those obtained with the shearing objectives.

The two methods outlined above served to establish (if it were necessary) that the A-band regions consistently had slightly higher refractive indices than the I bands; and it was then possible to use a simpler, third method to determine which set of bands had been matched in any given medium—from the colour of the unmatched bands when a white light-source was used.

The Smith interference microscope is a flexible instrument and it is possible to adjust it so that, for a given rotation of the analyser, the background field will change colour either up the Newtonian series of interference colours, or down it, depending on the direction of tilt of the condenser surfaces relative to the objective. (This in no way affects the fact that it is always necessary to rotate the analyser in a certain direction in order to make a phase-change measurement, because a phase object in the field will always exhibit a colour shift in the corresponding direction.) Therefore the colour of an unmatched object relative to the background does not in itself give any indication as to whether the object is giving an advance or retardation in phase of the light passing through it. If, however, the initial adjustment of the microscope is left unaltered, and two different specimens are examined in succession, it is possible to distinguish between two phase-advancing or phase-retarding objects, or between objects giving phase changes of different sign, by means of the direction of the shift in colour that they exhibit. If, for example, the first object were known to be a phase-retarding object of higher refractive index than its mounting medium, and it appeared in a colour slightly higher in the Newtonian series than that of the background, and the second object, with the same background colour, had a colour lower in the Newtonian series, it would mean that the second object was giving an acceleration in phase and therefore had a lower refractive index than the mounting medium.

This can be seen in the colour photomicrographs of muscle-fibres shown in fig. i, C-F. c and D show a fibre mounted in a medium with a refractive index of 1·360, while E and F show a similar fibre mounted in a medium with a refractive index of 1·365, the microscope being left in exactly the same adjustment for all the photographs (and for A, B, G, and H as well). In C and E the analyser has been adjusted so that the background and the matched bands are of a similar light-blue colour, but the unmatched bands in c are yellowishgreen, a shift up the Newtonian series, and in E they are dark blue, a shift down the series. Similarly, in D and F the background and the matched bands are of approximately the same purplish-red colour, but in the first case the unmatched bands are blue and in the second case they are yellow. This can only mean that if, in the preparation mounted in the medium with a refractive index of 1 · 360 (c, D), the unmatched bands were giving a phase retardation and were therefore of higher refractive index, and were A bands, then the unmatched bands in the preparation mounted in the medium with a refractive index of 1 · 365 (E, F) must have been phase-advancing and therefore of lower refractive index, and hence were I bands. Therefore, provided that the A bands are invariably of higher refractive index than the I bands, which seems highly likely, if it was the I bands that were matched in the first medium of refractive index 1 · 360, it must have been the A bands that were matched in the other medium of refractive index 1 · 365.

It would, of course, have been equally possible to determine the sign of the phase-change given by the unmatched bands by using nearly monochromatic light and rotating the analyser in an appropriate direction from the position of background extinction and seeing if they darkened or brightened, but the method described above was simpler and quicker, since it depended on only one observation and the colour of the unmatched bands could be recognized at once.

Results

With the aid of these criteria for distinguishing the I- and A-band regions, it was possible to measure their refractive indices individually with the following results, based on the examination of just over 50 specimens.

Z-band regions were found matched in media having refractive indices ranging from 1· 358 and 1·363, but not outside this range. In these preparations, the A bands always appeared to give a phase retardation (figs. 1, A-D, G; 3; B-D).

A bands were similarly found matched in media with refractive indices ranging from 1·360 to 1·366. These measurements, however, were made with a 2-mm shearing objective with the fibre orientated north-south, and therefore represent the limits of range found for the higher of the two refractive indices of the 4-band material. A few measurements of the lower of the two refractive indices of the N-band regions were also made with a 2-mm ‘double focus’ objective with the fibres at the same orientation, but too much reliance was not put on them for reasons already discussed (p. 232). These gave values of 1·364 and 1·365 (figs. 1, E, F; 3, E, F).

The less extensive Z-band and ZZ-band regions, in the middle of the I and A bands respectively, were also frequently observed under the interference microscope, but they were too narrow to be matched with certainty. From their colour, however, it was clear that their refractive indices were close to those of the I- and A-band regions. The Z bands always appeared to have a slightly higher refractive index than the I bands, and the H bands slightly lower than the A bands.

The viability of the fibres measured

As already mentioned, measurements were only made on portions of fibres that appeared in every way to be absolutely undamaged, but it was certain that practically all of these had been cut at some point along their length. It could therefore be argued that, being no longer completely intact, none of the portions examined was in a physiological condition comparable to those in the living animal, and that the refractive index measurements made on them were not necessarily the same as for undissected muscle. This argument cannot be entirely refuted, but the results obtained strongly suggest that it is not in fact valid.

If the surface of a muscle-fibre is damaged so as to become permeable to the molecules of the mounting medium, one would expect an interchange of material to occur between the inside and outside, causing gradients of refractive index and phase change; and this is in fact what was found in the case of all the obviously damaged fibres. The cut ends of the apparently intact fibres also nearly always showed similar phase gradients, but it took many minutes for these to spread inwards over a distance of several hundred microns to the regions where the measurements were made. It was also very unlikely that a very rapid and completely uniform alteration in water content (causing an alteration in refractive index) had taken place at the moment that the fibres were dissected, because of the extremely narrow range (1·358 to 1·366) of refractive index found in all the apparently intact fibres measured. One would not expect any considerable alteration of refractive index, if it occurred, to be so uniform; and fibres whose refractive indices had been appreciably changed by handling could reasonably be expected to exhibit a greater degree of variation. The maximum range in refractive index of the material was probably very little greater than that measured, because although most of the fifty-two fibres that were examined were in media within these limits, 12 were in media with refractive indices just outside these limits, and none of these fibres appeared matched, although in other respects they seemed normal.

Occasionally, in a fibre which had been appreciably stretched, parts of otherwise matched Z-band regions showed a small negative phase change, indicative of a slightly lower refractive index (fig. 1, H). The significance of this observation will be discussed below (p. 236). Nearly all the fibres examined, however, were in a ‘resting’ condition (unstretched and uncontracted), with a sarcomere interval very close to 2-5 p.

Optical difficulties involved in matching the muscle-bands

Because the sarcomere interval (about 2·5 μ) was small compared to the total thickness of a fibre, it was only on rare occasions that one could find a preparation where the A and I bands were exactly aligned in the direction of the optical axis of the microscope through the whole thickness of the fibre: but cases of this can be seen in figs, 1, G and 3, B, C, where the matched bands stretch right across the fibre. It was, however, only necessary to obtain a match in the thinner regions 5 to 10 μ in from the edge, and it was of little importance if the bands in the centre were out of alignment.

The thickness of the fibre, however, made measurements even in this region open to one serious objection, for which the writers are indebted to Professor H. S. Bennett of the University of Washington, Seattle. This is that with a relatively large cylindrical object such as a muscle-fibre seen under a microscope objective of high numerical aperture, it is inevitable that some of the more oblique illuminating rays must pass through the under surfaces of the fibre to reach the region tangential to the optical axis on which the measurements were made, and this might introduce errors into the phase-change measurements. Although this was less likely to be important here, when it was in fact zero phase changes that were being measured, it was obviously important to devise some way of determining whether the fibre thickness did have an effect. The experiment described below was suggested by Dr. A. Szent-Gyôrgyi at Woods Hole.

The living fibres were never less than 50 μ in thickness, but fixed preparations of fibres could be cut into much thinner sections. Unstained preparations were therefore made of material fixed in Carnoy’s fluid and in formalin, which was embedded in paraffin wax and cut longitudinally into sections varying from 10 μ to 30 μ in thickness. It was then necessary to find a suitable mounting medium of the same refractive index as one of the band regions of this fixed material and, in the 10-μ sections, it was found that in methyl salicylate, with a refractive index of 1·536, one set of bands exactly matched the background in colour and intensity under the interference microscope. The 30μ sections were then similarly mounted and examined, and exactly the same match was obtained.

Although the maximum thickness of the sections here was only about half that of the living fibres, the experimental conditions were more closely comparable than would at first appear, since the sarcomere in the fixed material was also appreciably less. Therefore, although it proved nothing, this experiment quite strongly suggested that this particular factor was probably not important in the matching of the bands with the living material. Moreover, as was pointed out by A. Huxley (1958), living muscle-fibres are seldom exactly circular in cross-section, but are much more usually polygonal so as to fit snugly against adjacent fibres, and quite often have ‘sharp’ edges. It is certain that many of the measurements made were through such edges, where there was a lesser thickness of the material.

The small sarcomere interval of the resting fibres that were almost exclusively examined does, however, give rise to another very much more serious optical difficulty, for it means that the individual A-band and I-band regions were both only about 0·54 μ wide. This means that the centre of each A and I band was no more than 0·27 μ from the phase boundaries of the adjacent I and H, and A and Z bands, respectively. At this distance it is highly probable that any phase-change measurement made in an A- or Z-band region (including the zero phase change indicated by an apparent match) will be affected by diffraction from the images of the adjacent bands. H. Huxley and Hanson (1957), considering this in connexion with their measurements on glycerinated myofibrils, obtained densitometer traces of the diffraction gradients at the vertical edges of a uniform phase object (a myoglobin crystal) with the interference microscope, and concluded that no accurate phasechange measurement could be made less than 0-3 p away from a phase boundary.

It therefore seems extremely probable that the match obtained by the present method between a muscle-band and the background field did not indicate a zero phase change in this region, but that the I bands had somewhat lower refractive indices than those of the solutions in which they appeared matched, and the A bands had correspondingly higher refractive indices than the solutions in which they appeared matched. Confirmation of this was rather strikingly provided by the few instances in which stretched fibres were observed mounted in the protein media, because, as already mentioned (p. 234), the I bands then appeared to have a lower refractive index than when the muscle was in the resting condition (see fig. r, H). H. Huxley and Hanson (1957) used stretched myofibrils to obtain valid measurements for the phase changes in the Z- and ZZ-band regions, and they calculated the solid content of the A-band regions from the ratio of these measurements; but with the present technique it was not normally possible to stretch the musclefibres. Fortunately, however, it is possible to estimate the maximum errors involved in measuring the refractive indices of the A and I bands of the unstretched fibres from independent measurements of the non-fibrillar material in whole muscle; and this will be discussed in the final section of this paper.

The most striking components of the I- and A-band regions of striated muscle are the fibrous protein elements located in the myofibrils. These have recently been shown, mainly through the very careful and complete histochemical and electron-microscopic investigations of H. Huxley and Hanson, to be longitudinally orientated systems of submicroscopic rodlets of actin and myosin that interdigitate and overlap each other in the A-band regions (H. Huxley and Hanson, 1954,1957; H. Huxley, 1953,1957). In the Z-band regions only actin rodlets are found, and the ZZ-band regions are bridged by the myosin rodlets without actin. Our present results are in complete accord with these findings, because the A-band regions were consistently found to have higher refractive indices than the I bands or the ZZ bands. This is best illustrated in fig. 3, D, where the lighter regions of the fibre, the matched I bands and the pale H bands, represent the regions of lower refractive index.

The mean refractive index of the I and A bands found by the present method was 1·363. This is approximately equivalent to a total solid concentration of 16% w/v, or 88% of water assuming that the solid constituents of these regions have a specific volume of 0-75. This indicates a rather higher water content than the generally quoted figure of 80% of water in mammalian muscle, based on weighing and drying. Estimates based on bulk weighing, however, are often rather inaccurate (see, for example, Ross and Billing’s findings compared to those of Henry and Friedman with respect to the water content of bacterial spores (Ross and Billing, 1957); and not much reliance should be placed on them. Huxley and Niedergerke’s (1958) value of 1·378 for the mean refractive index of the A and I bands of whole muscle-fibres in the frog is, however, more nearly in agreement with this figure; and it remains a possibility that in the present investigations the refractive indices of the portion of fibre examined were somewhat lowered through some sort of leakage at the cut ends. This, however, seems rather unlikely, for the reasons already discussed on p. 234, and in the argument that follows we shall assume that our value is correct.

The refractive index of the solution in which matched I bands were most frequently found was 1·360, and that in which matched A bands were most frequently found was 1·365. If these really represented the refractive indices of the I and A bands, the A bands would contain only about 3 % more solid matter than the I bands. However, H. Huxley and Hanson (1957), in their very thorough quantitative investigation of the fibrous protein components in isolated glycerinated myofibrils from the psoas muscle of the rabbit, found that the H-band regions contained approximately 3 times as much solid material per unit length as the I bands: an enormous discrepancy. Some of this discrepancy might be accounted for by assuming that the intact fibres contained a considerable amount of uniformly distributed extra material in addition to the fibrous actin and myosin in the myofibrils (in the sarcolemma and other loci); but in order to account for the whole of this difference the amount of this non-fibrous material would have to exceed 10 times that of the fibrous protein in the myofibrils, which is exceedingly unlikely. It is therefore much more probable that most of the discrepancy is accounted for by the diffraction effect already discussed on pp. 235 and 236; and that the true refractive indices of the A bands are considerably higher than 1·365, and those of the I bands lower than 1·360.

Actually, Szent-Gyôrgyi, Mazia, and Szent-Gyôrgi (1955) and Hanson and Huxley (1957) have made estimations of the non-fibrous protein in whole muscle-fibres, and have found that this does not exceed 50% of the fibrous protein in the myofibrils. Taking Hanson and Huxley’s lower figure of 33%, if this non-fibrous protein were uniformly distributed along the length of a fibre it would have a dry mass per unit area of almost exactly o-6 times that of the fibrous protein (actin) in the Z-band regions. The A bands in the fibres would therefore have 2 · 25 times as much total solid material per unit length as the I bands. Assuming the mean solid content of the I and A bands to be 16%, if the true solid content of the I bands was 10% and the true solid content of the A bands 22%, the above A:I ratio of 2-25:1 would be correct. Because, in the resting fibres examined, the A- and I-band regions were almost equal in width, it is reasonable to expect that the low error in the measurement of the refractive index of the A bands will almost equal the high error in the measurement of the I bands. This means that, at its highest, the refractive index of the A bands was 1·375 rather than 1·365, or o-oi higher than that of the mounting medium in which they most frequently appeared matched; and, at the lowest, the refractive index of the I bands was 1·350 rather than 1·360, or 0 ·01 lower than that of the medium in which they most frequently appeared matched.

These represent plus and minus errors of just over 4% w/v for the solid content of the I and A bands respectively. They are, however, maximum errors based on the assumption that the non-fibrous solid material is uniformly distributed, when in fact it may be more concentrated in the I -band regions as suggested by A. Huxley and Niedergerke (1958). It is, however, almost certain that there is an error due to diffraction from the unmatched bands, and that the true refractive indices of the A and I bands lie somewhere between those of the solutions in which they appeared matched and the upper and lower limits just defined. (The above estimates for the A bands, however, are based on the higher of their two refractive indices, and as their birefringence can be taken as approximately 0 · 004, their mean refractive index can be reckoned as being 0-002 lower, and consequently their solid content 1% less than the figures given.)

It seems probable that the values obtained by Bennett (1955) for glycerinated myofibrils, and the values quoted by Huxley and Niedergerke (1958) for the refractive indices of the I -band regions in frog muscle when the sarcomere interval was less than 3 μ (and those of the A bands where the sarcomere interval was 3 μ or over), were subject to the same kind of diffraction error as that found here.

We should like first of all to acknowledge the extremely generous support of the Muscular Dystrophy Association of Canada. This body provided the salary of one of us (K. F. A. R.) as a Research Associate while working on this problem in the University of Toronto, and paid his passage across the Atlantic. It also made substantial additional travel grants to enable us to visit colleagues working in the same field both in North America and in England. As a result of this, we are very much indebted to the following people for some extremely helpful discussions: Mr. A. F. Huxley in Cambridge, England, and at Woods Hole, Mass.; Professor H. S. Bennett in Cambridge, Mass., and in Buffalo, N.Y.; Dr. A. Szent-Gyôrgyi in Woods Hole, Mass.; Dr. Jean Hanson in London; and Mr. F. H. Smith in Croydon.

We are very specially indebted to Dr. Eva Cairns and Dr. John Logothetopoulos of the Charles Best Institute, University of Toronto, for their skill in making the preparations of living muscle-fibres. The improved dissection techniques developed by Dr. Logothetopoulos contributed more than any other single factor to the completeness of the investigation, after Dr. Cairns had shown what could be done in this way.

We should also like to thank Dr. John Duckworth, Head of the Department of Anatomy in the University of Toronto, for providing laboratory space, apparatus, and encouragement; and Dr. J. R. Baker of the Department of Zoology, Oxford, and Mr. F. H. Smith of Messrs. Charles Baker for their valuable criticism of our manuscript. One of the interference microscopes used, the English Baker interference microscope, was provided out of a grant to one of us, K. F. A. R., from the London University Central Research Fund. The remaining microscopes and several other instruments were provided out of grants to W. G. B. C. from the Canadian Muscular Dystrophy Association.

The cost of the colour plate, fig. 1, was also paid for out of a special grant from the Muscular Dystrophy Association of Canada.

Barer
,
R.
,
1952
. ‘
Interference microscopy and mass determination.’
Nature
,
169
,
366
.
Barer
,
R.
and
Joseph
,
S.
,
1954
. ‘
Refractometry of living cells. Part I. Basic Principles.’
Quart. J. micr. Sci
.,
95
,
399
.
Barer
,
R.
1955a
. ‘
Refractometry of living cells. Part II. The immersion medium.’
Ibid
.,
96
,
I
.
Barer
,
R.
1955b
.
‘Refractometry of living cells. Part III. Technical and optical methods.’
Ibid
.,
96
,
423
.
Barer
,
R.
and
Ross
,
K. F. A.
,
1952
. ‘
Refractometry of living cells.’
J. Physiol
.
118
,
38
.
Bennett
,
H. S.
,
1955
. ‘
Values for the ratio of mass per unit length in the crossbands of striped muscle myofibrils, as measured with the interference microscope.’
Anat. Rec
.,
121
,
263
.
Davies
,
H. G.
, and
Wilkins
,
M. H. F.
,
1951
.
Physical aspects of cytochemical methods.Cytochemistry Commission for the Soc. for Cell Biol., Stockholm
.
Davies
,
H. G.
1952
. ‘
Physical aspects of cytochemical methods.’
Nature
,
169
,
541
.
Hanson
,
J.
, and
Huxley
,
H. E.
,
1957
. ‘
Quantitative studies on the structure of cross-striated myofibrils. II. Investigations by biochemical techniques.’
Biochim. biophys. Acta
,
23
,
250
.
Huxley
,
A. F.
1958
.
Personal communication
.
Huxley
,
A. F.
and
Niedergerke
,
R.
,
1954
. ‘
Structural changes in muscle during contraction. Inter- ference microscopy of living muscle fibres.’
Nature
,
173
,
971
.
Huxley
,
A. F.
and
Niedergerke
,
R.
1958
. ‘
Measurement of the striations of isolated muscle fibres with the interference microscope.’
J. Physiol
.,
144
,
403
.
Huxley
,
A. F.
and
Taylor
,
R. E.
,
1955
. ‘
Function of Krause’s membrane.’
Nature
,
176
,
1068
.
Huxley
,
H. E.
1953
. ‘
Electron microscope studies of the organisation of the filaments in striated muscle.’
Biochim. biophys. Acta
,
12
,
387
.
Huxley
,
H. E.
1957
. ‘
The double array of filaments in cross-striated muscle.’
J. biophys. biochem. Cytol
.,
3
,
631
.
Huxley
,
H. E.
and
Hanson
,
J.
,
1954
. ‘
Changes in the cross-striations of muscles during contraction and stretch, and their structural interpretation.’
Nature
,
173
,
972
.
Huxley
,
H. E.
1957
. ‘
Quantitative studies on the structure of cross-striated myofibrils. I. Investigations by interference microscopy.’
Biochim. biophys. Acta
,
23
,
229
.
Ross
,
K. F. A.
,
1952
.
Quoted by Barer, R., and Joseph, S., 1955a
.
Ross
,
K. F. A.
and
Billing
,
E.
,
1957
. ‘
The water and solid content of living bacterial spores and vegetative cells as indicated by refractive index measurements.’
J. gen. Microbiol., ï6, 418
.
Smith
,
F. H.
,
1958
.
Personal communications
.
Szent-Gyôrgyi
,
A. G.
,
Mazia
,
D.
, and
Szent-Gyôrgyi
,
A.
,
1955
. ‘
On the nature of the cross-striations of body muscle.’
Biochim. biophys. Acta
,
16
,
339
.