FIGURE 3. Diagrammatic representation of the essential parts of the apparatus. These are all shown in vertical section, except the rheostat, which is shown in horizontal projection. Near the middle of the rectangular area, bounded by the arms of the "needle-bar" (B, B) are shown in vertical section the nozzle (N) with its slot (T) and the platinum wire (W); and immediately above this a horizontal section of the same at the level of the letter N of the vertical section.

B, B. The two vertical arms of the "needle-bar"; D, glass tube terminating in the bottle, which also receives the copper tube; the rubber suction tube is attached to the glass tube at D; H, portion of the "head" of the sewing machine, which receives the "needle-bar"; L, link by which the "needle-bar" is attached to the lever; N, nozzle of the copper tube; O, orifice of tube for the escape of steam from the hot-water tank; P, metal plug to fill the slot in N; R, rheostat ; S, brass spring to keep the platinum wire taut when hot; T, slot in one side of copper nozzle; V, lever connected with crank wheel; W, platinum wire.

of the terminal orifice of the nozzle and to emerge below from the bottom of the slot. The slot is then carefully stopped with a thin metal plug P (slight projections from its surface prevent its being forced inward too far), so that air can enter the tube through the terminal orifice only. If this orifice is kept close to the under surface of the wax plate, the melted wax will be completely withdrawn and will run down into the glass bottle; but if the orifice drops only a few millimetres below the under surface of the wax plate, the melted wax will not be withdrawn and will soon congeal, leaving the cut edges firmly reunited. This fact has been taken advantage of to produce at will either a temporary or a permanent cut. As it would be inconvenient to raise and lower the nozzle with its attached water reservoir, the slate platform which supports the wax plate during cutting is made movable in a nearly vertical direction. The front edge (that next the operator) is supported on two round-headed screws, one seen distinctly near the dotted line a, Figure 2, the other faintly, close to the detached "front plate-slide," further to the right. The height of the front edge of the platform can thus be regulated by turning these screws in or out, and accurate adjustment to the height of the fixed nozzle thus secured. The middle of the far edge of the platform rests on a square block (a, Figure 2) screwed to a long horizontal arm turning on a horizontal pivot at the left. Another horizontal arm (b) turning on a vertical pivot engages the slanting under side of the block, and when moved in a certain direction raises the block some 6 or 8 mm. This second horizontal arm is actuated by levers, not shown in the figures, which are moved by the operator's knee. Thus the far edge of the platform may be quickly raised or lowered at will, so that the middle of the platform, where the heated wire is melting the wax, will also be raised or lowered about half as much as the distant edge.

To prevent the slate platform from becoming heated by the hotwater tank below it, a felt lining is attached to its under surface and a removable screen of the same material is placed over the tank. This is seen at the left in Figure 2 a square sheet with a square notch cut out of one corner to accommodate the platform-block.

As thus arranged, the wire may be readily heated to the desired temperature, and, by operating the pedal as in sewing, it may be made to make rapid vertical excursions. Since the wire is central to the orifice in the copper tube, the wax plate may be moved in any direction, the melted wax being withdrawn with equal facility, whatever the direction of the cutting. The fine sharp cut, exactly perpendicular to the plane of the wax, which is produced by this machine, seems to meet all the requirements for cutting wax plates rapidly and accurately.

Proceedings of the American Academy of Arts and Sciences.

VOL. XLII. No. 24.- MARCH, 1907.

CONTRIBUTIONS FROM THE CHEMICAL LABORATORY OF HARVARD COLLEGE.

CONCERNING POSITION ISOMERISM AND HEATS OF COMBUSTION.

BY LAWRENCE J. HENderson.

INVESTIGATIONS ON LIGHT And Heat madE AND PUBLISHED, WHOLLY OR IN PART, WITH APPROPRIATION FROM THE RUMFORD FUND.

CONTRIBUTIONS FROM THE CHEMICAL LABORATORY OF
HARVARD COLLEGE.

CONCERNING POSITION ISOMERISM AND HEATS.
OF COMBUSTION.

BY LAWRENCE J. HENderson.

Presented by T. W. Richards, January 9, 1907. Received December 20, 1906.

CURRENT ideas of valence consist of two distinct conceptions, quantivalence and valence energy, which rest upon two bodies of fact of different sorts. Of these conceptions quantivalence has played by far the greater rôle in the development of organic chemistry and of stereochemistry, though the consideration of valence energy is present or implied in Le Bel's discussion of the asymmetric carbon atom, in von Baeyer's "Spannungstheorie," in Thiele's theory of partial valence, and in Werner's stereochemical theories and recent publications. In Richards's recent discussions of the compressible atom the conception of valence energy has shown itself more pertinent than the conception of quantivalence, and in the theoretical discussion of thermochemical data it has been, and of course is, of the greatest moment.

Conclusions regarding valence energy which are based upon heats of combustion are open to the criticism that from measures of the magnitude of the total-energy change it is sought to determine the magnitude of a quantity which depends perhaps entirely upon the free-energy change. Yet even to-day the determination of heats of combustion remains the one way possible of gaining quantitative data regarding the magnitude of valence energy in organic compounds, and it is probable that such information, properly interpreted, leads to not inaccurate conclusions. On the contrary, these conclusions may be very accurate when differences between the heats of combustion of similar substances are considered, and when such differences in very similar cases are compared, as in this paper, there is good reason to believe that changes in bound energy have been almost wholly eliminated.

Of these two ideas concerning valence, that of valence energy is the less clearly defined. J. Thomsen, it is true, has sought to show that