The peculiar type of the curves has long been a subject of comment and speculation. Special properties have been attributed to the "slope" or inclination of the upper and approximately straight portion of the curve. According to the writer's investigations, the "slope" is largely dependent on the degree of vulcanisation, so that it is difficult to "place" as an index of the specific nature of a rubber.[41] Moreover, it has recently been shown that the peculiar type of curve given by vulcanised rubber is the result of plotting the load against the sectional area of the unstretched test piece, whereas this area decreases progressively as the test piece stretches. If this decrease be allowed for, the curve obtained is an equilateral hyperbola.[42] Preliminary experiments with rubber compounded with large proportions of finely divided mineral matter, such as carbon black, show that the load-stretch curves obtained autographically are likewise reducible to equilateral hyperbolæ. [41] Bulletin R.G.A., October, 1921, p. 397. [42] _Hatschek Journal Soc. Chem. Ind._ 1921; _Trans._, p. 251. CHAPTER XXIII _THE PROPERTIES OF RUBBER_ This section, like the last, is divisible into two subsections. The first deals with raw rubber, the second with vulcanised rubber. We have already explained that, until recently, rubber was not used in the unvulcanised condition, but that the excellent physical properties of plantation rubber have made this possible. It is interesting to compare the physical properties of raw rubber with that vulcanised with sulphur. A compact sample of crepe as received from the East will give breaking strain of over 30 kilos per sq. cm. and over 300 per cent. elongation. When mixed with sulphur and vulcanised, a breaking strain of 150 kilos and elongation of 1,000 per cent. are not unusual. It is possible that crepe rubber would give higher figures if it could be prepared in the form of a compact ring, as used for tests on vulcanised rubber. In any case, the figures for vulcanised rubber are much in excess of those for raw crepe rubber. It must also be remembered that a breaking strain of 150 kilos is not permanent with vulcanised rubber, for reasons which will be explained later.[43] To obtain a reasonably permanent vulcanised product, the vulcanisation would not be carried further than to give a figure of 100 kilos. On the other hand, raw rubber is remarkable on account of its great permanency, although subject to some physical changes at ordinary atmospheric temperatures. Tensile tests, although valuable, do not tell us all about the physical properties of a sample of rubber. Abrasion tests, or tests designed to measure resistance to wear and tear, would be more valuable, but, unfortunately, these properties do not lend themselves to simple tests. There are grounds for believing that raw rubber is superior in some respects to fully vulcanised rubber, if prepared without the addition of