Difference between indices of refraction for H, at 15°:

a

a-Bromonaphthalene. The a-bromonaphthalene employed was obtained from Kahlbaum, and was purified only by fractional distillation. Boiling point, 284-286° (corr.). Density, d 15°/15° = 1·49340; d 15°/4° = 1.49214.

=

Mixture of equal volumes of a-Bromonaphthalene and Heptane at 15°. Density, d 15°/15° -1.10062. Average density, unmixed, d 15°/15° = 1·09096, showing a difference, due to contraction on mixing, of 0.00966. Composition of mixture, 1 mol. C1H,Br+0.954 mol. CH16. Mol. wt., 302-443. Density, d 15°/4° = 1·09965.

Reduction in dispersion for H, -Ha: Calc. ....

H, 0·333

9.079

Mixture of equal volumes of Phenetole and Benzyl Cyanide at 15°.

=

Density, d 15°/15° 0.99653; average density, unmixed, 1.00888, showing expansion of 0.01235. Composition, 1 mol., CH100+ 1.0975 mol., CH,N. Mol. wt., 250-415. Density, d 15°/4° 0·99569.

=

Reduction in dispersion for H, -H: Calc............ 6·514

a

Found ......... 6·522

α

-0.05.

Difference between indices of refraction for Ha at 15°:

In the case of this mixture, where the refractive power of the two substances differ very little, it is seen that the numbers obtained are only very slightly lower than those calculated, and that there is no appreciable difference in the dispersion.

Phenetole.-Boiling point, 170.3° (corr.). Density, d 14.7°/4° =

Benzyl Cyanide.-Boiling point, 233.5° (corr.). Density, d 15.2°/4° :

1.02025.

=

III. IMPROVED SPECTROMETER SCALE READER.

The usual method of reading the scale and vernier of a spectrometer by means of a lens is far from convenient, as it requires the observer to change his position from that occupied when using the observing telescope for the spectrum lines each time he has to register the result; there is often a difficulty also in getting a satisfactory illumination of the scale, especially in a darkened room. The loss of time involved is very undesirable during a determination of the refractive indices of fluids in a hollow prism, particularly when the coefficients of expansion are large and readings are made for several lines of reference on both

sides of the prism, because accurate results can only be obtained if the temperature of the substance is kept constant.

To overcome these inconveniences an arrangement has been devised by which the scale and vernier are read by means of a long, narrow telescope, so arranged that its eye-piece is situated just below that of the observing telescope. The apparatus is shown in the figure and has been found to be convenient and to work very satisfactorily.

In this instrument, the support carrying the observing telescope has a slot through which the reading telescope passes, and this slot is

made wide enough to allow the latter to move to the right or left over rather more than the entire length of the vernier by about 5° on each side; both telescopes work from the central column which supports the circle. The reading telescope is provided at one end with a right-angle prism, which is not fixed, but supported by two screws, one on each side, so that its position can be changed by hand until the most convenient view of the scale and vernier is obtained; this prism is not placed directly over the scale, but sufficiently far back to avoid casting a shadow upon it. The illumination of the scale

and vernier is effected by means of a small gas jet fixed on the side of the observing telescope, and moving with it, the gas being supplied by means of a piece of rubber tubing sufficiently long to give freedom of movement; this gas supply is also provided with a by-pass so that it can be easily reduced and only sufficient light left for the small mirror used to illuminate the bright pointer in the eye piece of the observing telescope. A piece of ground glass is also fixed above the observing telescope to diffuse the light falling upon the scale and vernier, and thus facilitate the readings. The reading telescope can be drawn back out of the slot, moved to any part of the circle, and after being pushed back into position, used for readings in relation to the vernier of the table supporting the prism; when this is done it is well to hold a piece of ground glass in such a position that the light may pass through it on to the scale. Unless it is necessary to keep the observing telescope fixed in position, after reading the scale it is more convenient and simpler to loosen the clamp and move both telescopes to the position of this second vernier. The instrument is arranged to read to 30", which appears to be as small an amount as is useful in determining refractive indices.

A few remarks may be made in reference to other parts of the instrument shown in the figure. In this, the position of the copper cell (Trans., 1892, 61, 287) is shown with its lid removed, and supported on a little shelf (which is generally turned up out of the way when not in use), ready to receive the prism, also the elastic tubes through which the water is circulated so as to keep the cell and prism at any desired temperature. The table supporting this cell and the prism can be rotated by hand to any desired position; it is, however, also provided with a fine adjustment passing up through the central column of the stand, and worked by a mill head at the foot of the stand. The arrangement for supporting the vacuum tube has also been found to be very convenient, especially when alternately using both the vacuum tube and Bunsen flame. The vacuum tube is not only held in position by the bar and clamp connected with the collimator tube, but the clamp itself slides on a small bar at right angles to the apparatus, having a spring on one side, and an adjusting screw on the other, so that its position in relation to the slit can be adjusted with precision; it can also be pressed back out of position and kept there by means of a milled head screw, whilst the Bunsen flame, coloured by potassium, lithium, sodium, or thallium salts, is used, and when liberated is pushed back into position without any further adjustment. Two cylinder lenses, crossing each other, are also used, as seen in the figure, and are found to be very useful, especially for upright vacuum tubes. In making the observations of the lines of reference, a rather wide slit is used, because it is less trying to the eyes, and has been found

quite as accurate as a fine slit, provided all the measurements are made to the fixed side or jaw of the slit, and not to the centre of it, of course taking care to use this for both sides of the prism. The thermometer employed is divided to 0.1°, but this is none too fine. If the temperature is kept constant, independent determinations of molecular refractions are found to differ only very slightly in the second place of decimals.

XXVII.-The Condensation of Formaldehyde with Ethyl Malonate, and Synthesis of Pentamethylene-1: 2: 4tricarboxylic Acid.

By J. FRANK BOTTOMLEY and W. H. PERKIN, jun.

In a previous communication on the condensation of formaldehyde and ethyl malonate (Haworth and Perkin, Trans., 1898, 73, 330), it was shown that the following substances were formed either directly or indirectly during this process:

(1) Ethyl methylenemalonate, CH2:C(CO,Et), and two polymerides called respectively ethyl paramethylenemalonate and ethyl metamethylenemalonate.

(2) Ethyl propanetetracarboxylate, (CO2Et),CH·CH2 CH(CO2Et)2. (3) Ethyl tetramethylenetetracarboxylate,

(CO,Et),C<H2>C(CO2Et)2.
CH2

The further investigation of this condensation has again given interesting results, since it has been found that, under conditions. different from those used in the first instance, the reaction takes place in such a way that 2 mols. of formaldehyde condense with 3 mols. of ethyl malonate and ethyl pentanehexacarboxylate is formed, thus:

(CO,Et,),CH,+CH,O+CH,(CO,Et),+CH,O+CH,(CO,Et),

(CO2Et),CH·CH2 C(CO2Et)2 CH2• CH(CO2Et)2 + 2H2O.

This new ester, which is solid and melts at 53-55°, is decomposed by boiling with baryta water, with formation of a syrupy polybasic acid which, when heated at 240°, loses carbon dioxide and yields. glutaric acid.

This curious result may be explained on the assumption that, under the conditions employed, the hexabasic ester, or the hexa basic acid produced from it by hydrolysis, is decomposed into propanetetracarboxylic acid together with methylenemalonic acid or its decomposition products, thus: