Humboldt State University ® Department of Chemistry

Robert A. Paselk Scientific Instrument Museum

Abbe's Refractometers. 3rd. ed. Carl Zeiss Co. Jena (1907)
© Richard A. Paselk 1999




The daylight which falls upon the mirror R passes through the double prism A B, closed for the purpose of measurement, into the telescopeF: the arrows indicate the direction of the circulation of hot water round the prisms: in order to save space, the thermometer is only shown in part. The magnifier L is fitted with a plaster reflector, the latter not shown in the figure.


E. ABBE, Jena, 1874 ("Neue Apparate &c."), published by Mauke.

E ABBE, "Carl's Repertorium der Physik", Vol. 15, 1879, p. 643.

C. PULFRICH, "Zeitschrift für Instrumentenkunde", Vol. 18, 1898, p 107.

The above three papers deal with the theory and the construction of the refractometers. The voluminous literary matter on the various methods of refractometrical investigation has been collectred in the form of a special refractometrical bibliography, copies of which we shall be glad to supply gratis on application.



 The ABBE Refractometers described in the following pages are designed for the measurement of the refractive indices and the mean dispersion of fluid plastic and solid bodies.
By virtue of their extraordinarily simple manipulation (the refractive index nD being, read of directly on a graduated circle after a single movement requiring no particular skill) and their extensive range of measurements, embracing refractive indices from nD = 1.3 to nD = 1.7, they have been found to answer all practical requirements in physical and chemico-technical laboratories, as well as in dispensaries and institutions for the analysis of food products; for distinguishing a number of substances, for testing their purity, and for rapidly determining the concentration of solutions. The more important substances which particularly lend themselves to refractometrical investigation are enumerated on page 8.
The method of measurement is based upon the observation of the position of the border line of the total rejection in relation to the faces of a prism of flint-glass, into which the light from the substance under investigation enter by the action of refraction.

The refractometer is mainly composed of the following parts:
1. The double ABBE prism A B, which contains the fluid and can be rotated on a horizontal axis by, means of an alidade (j in Figs. 1 and 3).
2. A telescope for observing the border line of the total reflection formed by the prism.
3. A sector S, rigidly connected with the telescope, on which divisions (representing refractive indices) are engraved*.

The double prism (A B, Fig. 1) consists of two similar prisms of flint-glass, each cemented into a metal mount and having a refractive index ;nD= 1.75 the fluid to be investigated (a few drops) is deposited between the two adjoining inner faces of the prisms in the form of a thin stratum (about 0.15 mm thick). The former of the two prisms, that farther from the telescope (which can be folded up or be removed), serves solely for the purpose of illumination, while the border line is formed by the second flint prism. Further details on this point, as also in regard to the examination of plastic and solid bodies, are given on page 9 and those following.

The border line is brought within the field of the telescope by rotating the double prism by means of the alidade in the following manner. Holding the sector, the alidade is moved from the initial position, at which the index points to nD = 1.3, in the ascending scale of the refractive indices until the originally entirely illuminated field of view is encroached upon, from the direction of its lower half, by a dark portion; the line dividing the bright and the dark half of the field then is the

* With regard to refractometers not having a sector, but provided instead with a scale in the eye-piece of the telescope which is rigidly connected with the prisms (butter, milk-fat and immersion refractometers), see special prospectus in each case.


Manner of


Appearance of the Border Line. Reading of nD. Influence of Temperature. Three Models.
"border line". When daylight or lamp light being employed, the border line, owing to the total reflection and the refraction caused by the second prism, assumes at first the appearance of a band of colour, which is quite unsuitable for any exact process of adjustment. The conversion of this band of colour into a colourless line, sharply dividing the bright and dark portions of the field, is the work of the compensator.
The compensator, which finds its place in the prolongation of the telescope tube beyond the objective, i. e., at a point between the objective and the double prism, consists of two similar AMICI prisms, of direct vision for the D-line and rotated simultaneously, though in opposite directions, round the axis of the telescope by means of the screw head M. In this process of rotation the dispersion of the compensator passes through every value from zero (when the refracting edges of the two Amici prisms are parallel and on different sides of the optical axis) up to double the amount of the dispersion of a single Amici prism (the refracting edges of the two Amici prisms being parallel and on the same side of the axis). The above-mentioned dispersion of the border line, which appears in the telescope as a band of colour, can thus be annulled by rotating the screw head M, thereby giving the compensator an equal, but opposite, dispersion. The opposite equal dispersions will then neutralism each other, with the result that the border line appears colourless and sharply defined.
The border line is now adjusted upon the point of intersection of the reticle by slightly inclining the double prism to the telescope by means of the alidade. The position of the pointer on the graduation of the sector is then read off by the aid of the magnifier attached to the alidade. The reading supplies the refractive index nD of the substance under investigation itself, without any computation, and with a degree of exactness approaching to within about 2 units of the 4th decimal. Simultaneously the reading of the scale on the drum of the compensator (T in Fig. 1) enables the mean dispersion nF - nC being arrived at by the aid of a special table and a short process of computation. The accuracy of the measurement of dispersion is increased by taking the mean of two readings of the drum varying by 180°.
As the refractive index of fluids varies with their temperature, it is of importance to know the temperature of the fluid contained in the double prism during the process of measurement; the temperature, of solid bodies during refractometrical investigation in a room is, on the other hand, of no consequence.
If, therefore, it be desired to measure a fluid with the highest degree of accuracy attainable with the Abbe refractometer (to wothin 1 or 2 units of the 4th decimal of nD), it is absolutely necessary to bring the fluid, or rather the double prism containing it, to a definite known temperature and to be able to control this temperature so as to keep it constant within some tenths of a degree for a period of several hours; hence a refractometer principally required for the investigation of fluids must be provided with heatable prisms.
We construct three models of the ABBE Refractometer, viz:
I. The Refractometer with Heatable Prisms, which is primarily designed for the investigation of fluids.
II. The Refractometer with Non-heatable Prisms.
III. The Refractometer for purposes of Demonstration and Practice.



The double prism (A B, Fig. 1) is opened out and closed up by means of a screw head v, which acts in the manner of a bayonet catch. In order to apply a small quantity of fluid to the prisms, without opening the casing, the screw v is slackened and a few drops of fluid poured into the funnel-shaped mouth of a narrow passage (not seen in the figure). On again tightening the screw, the fluid is distributed by capillary action over the entire space between the two prisms. This arrangement facilitates the investigation of even rapidly evaporating fluids, such as the etheric solution of milk fat employed in the refractometrical determination of the proportion of fat in milk. - In the case of viscous fluids (resins, &c.) a drop of moderate size is applied with a glass rod to the dull prism surface, the double prism being opened out for the purpose. The prisms are then closed up again and before the measurement is proceeded with, the refractometer is left standing for a few minutes in order to compensate any cooling or heating of the prisms which occurred while they were separated.

The fitting for heating the prisms of the refractometer is constructed in its essential parts on Dr. R. WOLLNY'S plan of enclosing the prisms in a metal casing with double walls through which water of a given temperature is circulated. A similar heating appliance is fitted to our butter and our milk-fat refractometers. The thermometer provided registers the temperature of the water circulating through the prism casing at its point of exit.

The attainment of a constant temperature in the current of water is facilitated by the use of

a Spiral Heater and a Water pressure Regulator.

The spiral heater is shown in Fig. 2. The water impelled by the direct or indirect pressure of the source of supply passes at a uniform rate of speed through a heated long copper pipe. The spiral heater, about 3 1/2 metres in length, is housed in the space between two telescoping metal cylinders. The inner cylinder is provided with a copper bottom, through which the heated air generated by a Bunsen burner, or a petroleum or spirit lamp, is evenly distributed and conducted to the copper pipe. The upper end of the apparatus bears a coarse wire grating by which the heated air escapes. The objects to be examined (fats, &c.) may be placed upon this grating for the purpose of melting or preliminary warming.

The top of the spiral heater is connected with the tube joint D of the refractometer by a short length of tightly stretched tubing, which should not run





horizontally, but should rather incline upwards in the direction of D. In this way the accumulation of air bubbles, which would obstruct the uniform flow of water, is checked. In order to be able to watch the presence of air bubbles, as well as observe the speed of the current, it is a good plan to fit a piece of glass tubing between E and the cistern B. This, too, should tend in an upward direction and be connected at E and B by means of very short joints of flexible tubing. If the fall from cistern B to the sink is slight, the use of a glass or metal pipe again commends itself. As already mentioned, slack or kinked tubes increase the already great resistance encountered by the current of water in its long passage, and should therefore be absolutely avoided. To facilitate an immediate suspension of work, it is advisable to fit a stopcock at an easily accessible point (a screw stop .without spring action, by preference). On screwing up the stopcock and turning off both water and gas, everything will come to a complete stand within a minute; similarly, the same temperature will be restored within from 5 to 10 minutes on resuming work, provided the gas tap is regulated exactly as it was before, to facilitate which the gas tap should be provided with a long key.

. It is of advantage generally not to have too sluggish a flow of water, to first obtain, a certain approximate temperature by appropriate manipulation of the lamp burner, and finally to regulate the heat to the cxact degree required by varying the difference in elevation between cisterns A and B of the water-pressure regulator. The required degree of temperature having thus been obtained, the tap of the supply pipe is turned off so as just to allow a thin, barely continuous, stream of water to issue from the overflow pipe of cistern A.

The difference in elevation between the two cisterns A and B may be varied in two ways; cistern A is either suspended from a cord running on a roller, the free end of which is made fast as with roller blinds, or recourse is had to the contrivance shown in Fig. 2. Cistern A is hung by two hooks to a board, the latter being made to slide up and down in a frame formed of a board about 1 metre in length, fitted with two strips along its sides, the whole suspended against the wall. Holes are bored through this board, arranged in zig-zag lines and each about 1 cm higher than the next lower one, the board bearing cistern A being kept at the desired elevation by inserting a peg into one of the holes. It is an easy matter to find experimentally the number of holes by which the board requires moving in order to cause the temperature to vary by 1°. As the cost of packing and carriage of such a frame would be quite out of proportion to its commercial value, we do not undertake to supply it, but any, carpenter would readily produce one at short notice.

If the requirement of a very high temperature (60° C, or more) should necessitate the employment of two spiral heaters, these should not be arranged one behind the other, but side by side.





It is comparatively easy to keep the temperature of the water circulation constant to within I or 2 tenths of a degree for a prolonged period. When using gas, a gas-pressure regulator is not an absolute necessity, as variations of pressure, such as may be caused by the opening of a neighbouring gas tap, make no appreciable impression on the temperature of the circulating water. More pronounced varitions, however (for instance, those occurring towards nightfall, or at the beginning and suspension of work in a large factory), make themselves felt in a more disturbing manner, and may necessitate the use of a gaspressure regulator.

The Investigation of solid bodies is restricted to the method of grazing incidence, as explained on page 10 and by Fig. 7. The experimental process there described should be employed in, for instance, the testing of the adjustment of the refractometer by the aid of the testing plate.

Latterly the ABBE refractometer with heatable prisms has been extensively applied also to the purposes for which WOLLNY's butter-refractometer and our milk-fat refractometer are primarily designed. To assist in this case the owner of an ABBE refractometer in the conversion of the nD values ascertained into equivalents of the scales of the butter or milk-fat refractometers, or vice versa, we have compiled a set of conversion tables, which are supplied with each specimen of the ABBE refractometer.

For the above-mentioned two purposes we also fit this refractometer, if desired, with the two special thermometers which are, as a rule, supplied only with the butter or milk-fat refractometers. The pair of thermometers may, moreover, be added to any ABBE refractometer, Model I, previously supplied by us. In the latter case it is necessary that the removable screw joint, either with or without a thermometer, be sent to us for adaptation (see Specification B, p. 15).

Scope of Applicability*:

The ABBE refractometer with heatable prisms is applied to tests as to purity, or the determination of the proportion, of known components in a mixture.

A. In industrial chemical laboratories and in institutions for the analysis of food products it is applied to the examination of butter, cheese, margarine, cocoa fat, lard and other comestible fats; of salad oils, cod-liver oil, lubricants, alkalies, linseed oil, varnish, turpentine, petroleum, paraffins ceresin and other kinds of wax; glycerine, aniline; aqueous, alcoholic and ethereal solutions as, for instance, the solution of ether and milk-fit adopted in NAUMANN's method of determining the percentage of fat in milk; milk serum.

B. In clinical laboratories it is employed, i. a., for determining the quantity of albumen in blood serum.

* See our Refractometrical Bibliography, copies of which are supplied gratis on application.




For the examination of solid and plastic bodies one of the prisms is removed by pressure on a nickelled spring (f, Fig. 6), as it is only required for the measurement of fluids. The measurement is carried out by reflected or grazing incident light.

As the method in which reflected light is employed imposes the least restrictions on the shape of objects for measurement, it is most generally adopted and shall therefore be described first.

A. The Measurement of the Refractive
Index of solid and plastic objects by reflected light. 

Solid bodies are ground and polished so as to produce a good plane surface. By means of the alidade and the sector the telescope of the refractometer is inclined to a sufficient degree to bring the rectangular polished face of the prism into a horizontal position.

The polished surface on the object is now moistened with a few drops of mono-bromonaphthalene and placed against the horizontal surface of the prism. The flap K (Fig. 3) is then hinged back so as to come in contact with the object, when, aided by the capillary effect due to the presence of the mono-bromonaphthalene between the two surfaces, the flap will keep the object attached to the prism, even after the latter has been restored, by means of the alidade, to the position required for measurement,



  Grazing incidence. Model II. Model I.

By removal of the flap the ground surface of the prism has become exposed. The light striking this surface is thence directed into the telescope by reflection on the polished surface, thus also on the object. The appearance of the border line now becomes less clearly defined than in the case of a stratum of fluid between two prisms, because the darker portion of the field of view is illuminated by the partially, the brighter portion by the totally, reflected rays. The most favourable condition of illumination must be ascertained empirically by revolving the entire sector. Carful achromatisation of the border line is most essential in this method. In other respects the rules given on page 4, regarding the reading on the sector, and on the drum of the compensator, are applicable. With small objects or badly polished surfaces a high degree of accuracy demands a series of independent measurements.

In the case of viscous plastic bodies (resin, &c,) a piece about the size of a pea is applied to the prism and a glance through the square end of the prism will show whether the substance adheres closely to the surface without air blisters intervening. A slight preparatory warming of the substance may sometimes be of advantage. The great adhesion of such substances renders the use of the flap K superfluous.

After each measurement the surface of the prism must be most carefully cleaned xvith soft linen saturated with a suitable solvent (alcohol, ether, benzol), according to the nature of the substance examined. Want of cleanliness induces an indistinct appearance of the border line, and may even result in false readings of the refractoinoter.

B. Measurement by Ganzing Incident Light
(for solid objects only).

If the body to be measured has two polished surfaces at right angles to each other and intersecting in a sharp edge, it can be examined by grazing, incident light (Fig. 5).

In regulating the illumination it must be observed that the light reflected by the mirror enters the object parallel to the polished surface of the prism. The appearance of the border line is the same as in the measurement of fluids by transmitted light (p. 4), namely, considerably clearer than in the method described under A.

The method of grazing incidence is also applicable to the refractometer with heatable prisms.




As shown in Fig. 7, the double prism is opened out and the object joined to the polished surface of the prism by means of a small quantity of monobromonaphthalene. The movable prism is hinged down, so far that the light from a clear sky which strikes the bright metal surface F enters the object in the manner indicated. Care should be taken not to mistake the image of the edge of the beam of a window frame or of the ridge of a roof for the border line. In cases of doubt a piece of white paper spread over the surface F will slightly obscure the field of view, when the border line will be recognized without fail.

C. The Measurement of Fluids.

The double prism is turned by means of the alidade into the position shown in Fig. 6, the inner prism surfaces horizontal, the removable prism undermost. After having previously removed the lower prism (see p. 9), a few drops of the fluid are applied to its ground surface. It is then reinstated, sliding the ground surface along the other one until the nickelled spring f is clearly heard to snap into its notch. Having ascertained by a glance into the square polished surface of the prism that the fluid fills up the entire space between the two prisms, the measurement proceeds in the usual manner. The double prism is manipulated as described on page 9 and the specimen applied to the horizontal polished prism surface only on occasions when extremely liquid or volatile substances require to be dealt with.

Testing the Adjustment of the Refractometer.

We supply with each refractometer a testing plate made of glass of known refractive index, the value of which is marked on the upper ground surface of the plate. The plate has two plane polished surfaces which form a sharp edge. For the purpose of testing the refractometer the refractive index of the plate P is very accurately measured by applying it several times in succession, the prism surface being wiped clean after each measurement. The mean of the several measurements is compared with the value stated on the plate, and if the 4th decimal, the adjustment of the refractometer may be considered satis-





factory. More pronounced discrepancies suggest the probability of some component part of the refractometer having become deranged, and in such a case it is always advisable to forward the instrument to our works for readjustment.

Owners are cautioned against any attempt on their own part at correcting the instrument or entrusting it to any mechanician, however skilful, for repairs instead of applying to us.

Experience has taught that, with careful treatment, derangement of the refractonieter is improbable, even after years of wear.


This refractometer embodies only the parts required to explain the nature of the ABBE refractometer, viz: the ABBE double prism, the telescope with reticle, and a sector graduated in 1/2 ° and provided with a nonius, while the compensator and the heatable prism casing are omitted.

As the border line cannot be rendered colourless it is necessary to employ monochromatic light (LOWE's sodium burner, or SIEDENTOPF'S mercury arc lamp). This refractometcr is peculiarly suitable for demonstration from the reason that the optical constants of the apparatus required for the measurement of refractive indices - the angle of refraction and the refractive index N of the flint prism - have to be determined on the apparatus itself, and because the three methods described under A, B, C on pages 9 - 11 and the differences in the appearance of the border line peculiar to them - more particularly the quality of the colour band under white light, and the distribution of light on both side of the border line - are immediately demonstrable.

The refractometer facilitates the determination of the refractive index of solid and fluid bodies, ranging from refractive index 1 up to nearly 1.7, that being the index of the flint prism.

* See C. PULFRICH in "Zeitschr. f. Instrumentenkunde". Vol. 18, 1898, pp. 114 et seq.




Fig. 10-12 indicate the path of the rays in the observation of total reflection on air used for the measurement of the refractive index of the flint prisms.

Manipulation of the Refractometer.

The right hand grasps the telescope, which is rigidly joined to the graduated sector, between the two bands. The left hand moves the revolving arm A, which bears the nonius and on the axis of which the double prism is secured. The nonius gives readings of 1'. The measurement of the angle of the prism requires that the telescope be adjusted vertically to the two prism surfaces in succession. For that purpose the reflected image of the lines of the reticle, which are illuminated by the prism a, must be brought into coincidence with the reticle itself (see Fig. 9), to do which it is advisable to direct the opening a towards the clear sky, a well-lighted white wall, or an extensive source of light.

Computation of the Refractive Index N of the Prism.

It is required to determine the angle i, under which the boundary ray of total reflection on air emerges from the prism. The position of the instrument in relation to the source of light for the three methods of illumination - for, here also all three are equally available - is explained by the path of the rays outlined in Fig. 10, 11, 12.

The formula for the computation of N

is derived from the fundamental equations

The logarithmical calculation of N in accordance with the formula given is facilitated by substituting for the right hand portion of the formula

representing an auxiliary angle. The value of N will then at once be found to be





Computation of the Refractive Index n of a Substance.

It is required to determine the angle i (by magnitude and sign) under which the boundary ray of total reflection on a substance emerges from the prism. The angle i (as also r) should be designated with the "+" or "-" sign, according as the emerging ray points towards, or iway from, the refractive face of the prism. The refractive index n of the object will then be deduced from i, N and by the following equations:

In the original brochure a price list follows at this point.


Index of Instruments

Museum Home

Instrument Literature


HSTC (1921-34)
HSC (1935-1953)
HSC (1954-1973)
© R. Paselk
Last modified 22 July 2000