The Pyramids and Temples of Gizeh. Flinders Petrie

      l Rigid tripod stand, 30 inches high, octahedral.

      m Rigid tripod stand, 16 inches high, octahedral.

      n Rigid iron tripod, 12 inches high, octahedral.

      o 12 signals, with plumb bobs.

      The above were all used, most of them continually; a few other instruments were also taken out, but were not needed.

9. Several of these instruments were of new or unusual patterns, which—as well as various fittings adapted to them—require some explanation. The dimensions are all in inches.

      A. The steel standard and straight-edge was on a new principle, employing the stiffness of a tube to maintain the straightness of a strip. It was skilfully executed by Mr. Munroe, of King’s Cross. A steel tube, 102 inches long, 2·0 diam., and ·06 thick (see Fig. 1, Pl. xv.) was supported at the two neutral points, 20·8 per cent. from the ends, resting on two fect at one point and one at the other. This tube carried a series of 15 flat beds, all dressed exactly to a straight line when the tube rested on its supports. These beds supported the actual standard, which was formed of three independent strips of steel, each 34 inches long, 2·0 wide, and ·1 thick, butting end to end. These strips bore on the upper face, along the front edge, very fine graduations, the lines being about 1/1000 wide. To ascertain the mean temperature throughout the whole length of the standard, a rod of zinc was screwed tightly to one end of the standard, and bore a scale divided to 1/200ths at the other end; the scale rising through a slot in the standard. The value of the divisions for various temperatures was carefully ascertained. As this standard was also a straight-edge, the edges of the three strips were all true straight lines, with a mean error of 1/1200th inch; and the edges were brought into one continuous straight line by adjusting screws set in the supporting beds, at the ends of the back edge of each strip. The object of having three separate strips was that they could be dismounted for independent use in measuring or drawing, and for testing each other’s straightness; that unequal heating of one edge should not cause as much distortion, in length or straightness, as if it were in one continuous piece; and that the weight should not be too great for the rigidity, in handling it when detached from the supports. The principle of separating the stiff part from the actual scale was adopted in order to use the regular drawn weldless steel tube, which is the stiffest thing for its weight that can be had, and also to prevent any unequal heating warping the straightness, as the tube was boxed in by a thin wooden sheath, and so was sheltered far more than the scale could be. The minor details were that strips were held down by screws with countersunk heads, bearing on steel spring washers; and they were pressed home against each other’s ends, and also against the back adjusting screws, by diagonally acting springs. Along the front of the tube were projecting screws, nutted on and adjusted to form a right angle with the face of the strip; so that the standard could be applied to any surface exactly at right angles.

      The value of the divisions was ascertained by comparison with a brass standard scale. This scale was tested by Capt. Kater in 1820, 1824, 1830, and 1831; and by the Standards Department in 1875 (see a report on it in the Report of the Warden of the Standards, 1875, Appendix x., pp. 36–41): as the steel standard was sufficient for comparisons, this scale was not taken to Egypt for fear of injury. The form of this brass standard is a bar, 42·14 long, 1·58 wide, ·17 thick; bearing a scale of 41 inches in length, divided to ·1 inch, with a vernier of 1/1000ths, and also bearing a metre divided to millimetres. The steel standard was ascertained, by means of this brass standard, to be exact at 19·6° cent.; and the mean error of graduation and reading combined was ·0002, the greatest error being ·0005. By the intermediary of a steel tape, the steel standard was further compared with the public Trafalgar Square standard; and according to that it was 1 in 60,000 longer, or true length at 17·8° cent, or a difference of ·021 on the length of the public standard, after allowing for the published error of ·019 inch. This is a guarantee that the length of the tape, which was used to transfer from the steel standard to the public standard, has no greater error than this; and, on the whole, I should place as much, or rather more, confidence in the series of comparisons between the Imperial, the brass, the steel standard, and the steel tape, made under the best circumstances indoors, rather than in comparisons between the steel tape, the Trafalgar Square standard, and certain steel rod measures, made in the open air, with wind and varying temperature. The difference in any case is immaterial, in regard to any of the points measured, in the present inquiry.

      B. The steel tape was over 100 feet long, ·37 inch wide, and ·008 thick, and weighed just over a pound. It was coiled on an unusually large drum (4·2 diam.), to avoid any chance of permanent distortion. Etched divisions, in the ordinary style, being too ill-defined, I had an unmarked length of tape, and divided it by fine cut lines at every 50 inches; the position of each line was shown by heating the steel to brown oxidation, and marking the number out of the brown by acid. It was found on trial that such lines did not weaken a piece of tape, even when it was violently twisted and wrenched; and that the steel, being hard drawn and not tempered, nothing under red heat softened it. The cuts were not put on with any special care, as their exact value was to be ascertained; but the worst error throughout was ·0098, the mean error ·0039 inch, and the total length true at 19·8° cent. This comparison was made when the tape was lying unstretched, on a flat surface, as ascertained by measuring successive 100-inch lengths on the steel standard. It stretched ·0127 per lb. on the whole length of 1,200 inches.

C. The steel chain of 1,000 inches I made on an entirely new pattern; and it proved, both in Egypt, and, some years before, at Stonehenge, to be very handy in use. The links are each 20 inches long, made of wire ·092 diam., this being as thin as can be used with fair care. The eyes (see Fig. 3, Pl. xv.) are wide enough to fold up one in the other, without any intermediate rings. They are rhomboidal, so that they cannot hitch one on the other, but will always slip down when pulled; and the internal curvature of the end of the eye is only just greater than that of the section of the wire, so that the linkage is sure when in use to come to its maximum length.* The junction of the eye is made with a long lapping piece, cut one-third away, and tinned to the stem. The whole was tested with 100 lbs. pull, to bring it to its bearings, before marking the divisions. The exact length of the links is unimportant, as, after the chain was made and stretched, a narrow collar of sheet copper was soldered about the middle of each link, the collars being adjusted to exactly 20 inches apart. Besides this, each link bore its own number, marked by a broad collar of copper for each 100, and a narrow collar for each 20 inches or link; thus, at 340 inches there were three broad and two narrow collars by the side of the central dividing mark on the link. These collars were put towards one end of the link, apart from the dividing mark, and counted from each end up to the middle, as usual. The central eye of the chain was not tinned up, but was held by a slip clutch; thus the chain could be separated into two 500-inch lengths if needed, each complete in itself, as for base lines for offsets. The handles were kept separately, hooking into any link at which accurate readings under tension might be needed. They were of the same wire as the chain, with wooden cross-bars. One of them included an inverted spring (see Fig. 2, Pl. xv.), so that the pull compressed the spring. When the pull reached 10 lbs., a small catch (not shown in the Figure) sprang out from the stem, and caught the coils. This left only a very small amount of play; and hence, when using it, the regulation of the tension did not require to be looked at, but was felt by the finger when at 10 lbs. pull.

      The advantages of this pattern are: (1) Great lightness and compactness of the chain, as it only weighs 2 1/2 lbs., and forms a sheaf 1 1/2 inch diam.; (2) consequent small error by catenary curves, and ease of carrying it clear of the ground by its two ends; (3) accuracy of the divisions;