It will be observed that a second test was carried out in the case of Nos. 3 and 6; in the first test of these samples, 18 and 20 stones respectively made up the 4 lbs. necessary for the experiment, while 20 and 30 stones respectively were required to make up the required weight in the second test. The Lindifferon stone shows the best results for toughness by the physical test of abrasion, the wet test in the second trial showing a loss of only 4.97 per cent., while that of Newbigging similarly tested shows 20:41 per cent, of loss.

It is notable that in the case of the former stone the results obtained by the wet test are superior to those by the dry test by an amount equal to 5.37 per cent. This is a peculiarity, but the tabulated results obtained from the different tests of the stone are amply borne out in practice. It is an excellent material for surface repairs where the roadway is shaded or located in a damp situation. When steam-rolled it is very durable, and the detritus which is formed through wear is composed of fine sand; it is of a homogeneous nature, the grains being scarcely distinguishable. This material was only used to a limited extent previous to the introduction of steamrolling, as, when laid in patches and exposed to the wheels of passing vehicles, the metalling was pulverized in dry weather, and as much so during frost. The test of experience shows that this stone is of a brittle or hard nature when in a dry state, while toughness appears to be stimulated by moisture. The Newbigging stone, an extremely tough basalt, does not come out well in the wet test compared with the Lindifferon stone just mentioned. It is nevertheless a very durable class of macadam, and its fitness, under all conditions of traffic and weather, as a suitable and economical material for road maintenance, has been recognized for a long period of years. These two specimens of stone will be further dealt with under the chapters on Quarrying and Stone-breaking.

133. The Drop Test.—This is carried out by subjecting the specimen of stone under examination to repeated blows from a falling weight. A machine for testing the tenacity of stones by this means has been used for a number of years by the Surveyor of the Truro Rural District Council, Cornwall. The hardness of a specimen of stone is found by the usual mineralogical test. The tenacity of the stone by this method of testing depends on the resistance of the material to a hammer weighing 15 lbs. falling 10 inches on the specimen, which is placed on an anvil. Samples of rock to be tested weigh 3 lbs., broken to road metal capable of passing through, in all dimensions, a 21-inch ring. The macadam thus prepared is, by the repeated action of the weight falling, reduced sufficiently small for its passing through a sieve having a -inch mesh, which size is taken to represent exhausted road metal, and the number of blows necessary to accomplish this being automatically registered. A sample of greenstone from Penzance, No. 5 test, having a hardness. * Vide Paper read by Mr Thomas Clark, surveyor, Truro District Council, at the

*

16

according to the mineralogical test of 6.5, took 900 blows to reduce it. This represents in the drop test the tenacity of the stone, and the dust resulting from the experiment, under of an inch, is given as 12 ounces. A sample of stone from Clee Hill (basalt), which has the same degree of hardness as the previous stone, shows a tenacity of 724, with 12 ounces of dust less than inch. It is considered by this test that any sample of stone which is unable to withstand 200 blows is unfit for road-metalling purposes.

The County Surveyor of Cornwall, Mr. T. J. Hicks,* commenting on the drop test as carried out and described in the preceding paragraphs, states: "Some years ago the surveyor of a highway district here introduced the only stone-testing machine I have yet seen that is worth anything, and even that is not entirely to be depended on. Taking spar under this test and greenstone spar, our worst material, gives hardness, 7; specific gravity, 2.56; number of blows, 264 to crush to dust. Greenstone, our best material, gives hardness, 65; specific gravity, 3.07; blows, 799; dust, 11 ounces."

"To show that this test is not altogether to be depended on, however, I know a stone that shows very little difference to the greenstone, and that costs about one-third more for breaking, yet under traffic it cracks up readily into gravel, and makes mud quickly."

134. The Absorption Test.-The absorption test consists in determining the liability of a stone to absorb water. It is carried out by weighing the stone carefully after being thoroughly dried, and immersing the sample in water for twenty-four hours. It is then wiped dry and weighed, the difference between the two results obtained being due to the water absorbed by the stone. The stones of admitted suitability for road-metalling purposes absorb, as a general rule, the least quantity of water.

Granite absorbs water to the extent of one part of the latter to every 80 to 700 parts of the former; it is notable, however, to observe that clay slate absorbs water in similar proportions to granite, as will be seen from Table XIV., taken from Prof. Rankine's Civil Engineering.

TABLE XIV.

Showing Proportion of the Absorption of Water by Different Stones.

Granite absorbs one part of water in from 80 to 700 parts of stone.

Clay slate has, of course, no place as a material in connection with road construction and repairs.

'Surveyor' Exhibit of the Building Trades Exhibition held in London, Surveyor and Municipal and County Engineer, 26th March 1897.

* Vide the Quarry, January 1897.

135. The Weathering Test.—This test is effected in a similar manner to the preceding, and has for its object the determination of the ability of the stone, after absorbing water, to withstand the effects of the disintegrating action of frost. It may be carried out artificially by soaking the stone in sulphate of soda (Glauber's salt) and hanging the sample up for a few days to dry. This process must be repeatedly performed, when it is said that the stone will behave as if under the action of a thaw succeeding frost. To properly carry out this test, however, it is necessary to expose the specimen of rock to the actual disintegrating action of frost. As in the absorption test, the stone is immersed in water for twenty-four hours, and then placed outside when the thermometer registers 32° Fahr. or under. The frozen specimen is then thawed and the process alternately repeated, when the loss in weight by disintegration is ascertained. As this test can only be effected during the prevalence of frost, it is consequently not always easy of application.

136. The Crushing Test.—This test consists in determining by means of a hydraulic press the resistance of a sample of carefully dressed cubes of stone to crushing. The results obtained by this physical test are, so to speak, of a positive nature; yet they may be very misleading in so far as the selection of a suitable stone for road metal is concerned. Too much importance is generally attached to this property, especially by quarry-owners.

To show how delusive it is to attach undue weight to a high resistance to crushing, solely on account of a stone possessing this property, the following different kinds of stone may be compared. A specimen of rock from the Clee Hill quarries already mentioned, which is described as a tough, hard, fine-grained basalt, and "almost perfect as a road stone," is capable of sustaining a crushing test of 6 tons per square inch. Guernsey granite (syenite), also an excellent stone, and much used in London and the southern towns of England for road repairs, shows a crushing resistance of 12.75 tons per square inch.

Flints, on the other hand, are a very suitable material where the traffic is moderate, and are used to a considerable extent in many English counties where more suitable material cannot be obtained. This description of material, however, cannot be compared with either of the preceding stones for durability when subjected to the vagaries of traffic and weather; nevertheless, it has a high resistance to crushing, being slightly over 14 tons to the square inch. From this comparison it is evident that a high resistance is no indication of the wearing qualities of a road stone under the practical test of traffic.

A stone having a high resistance to crushing clearly shows that other qualities are necessary in order that it may be successfully used for metalling purposes. It has been shown that toughness alone is not enough for a good stone to possess, while one having a high resistance to crushing is equally deficient; so that of these two properties, it is apparent that the one is quite

as essential as the other for a stone to possess in order that it may be considered as having good wearing qualities.

137. In France the Administration des Ponts et Chaussées has adopted a comparative numerical value of the quality of the materials used on the national roads. This scale of quality, ranging from 0 to 20, has been formulated from the combined results of experiments, attrition and crushing tests; and the quantity of materials consumed by a certain amount of traffic. The following is a list of most of the varieties of rock used, with the quality of each kind expressed numerically :

TABLE XV.

(Ministère des Travaux Publics Routes Nationales, 1893.) *

138. As bearing upon this subject of physical tests with special reference to the mud-producing proclivities of limestone, the results of the following interesting experiments may be given of road-stone testing carried out by Professor Elliot, University College, Cardiff,† on behalf of the Western Mail of that borough.

It appears from statements in this journal that most of the streets in that borough are of macadam, the material being generally limestone, basalt and granite being only used in special cases. As a consequence, it is asserted that no less than 30,000 tons of mud were removed last year from the Cardiff streets. With a view to remedying this state of affairs the Western Mail commissioned Professor Elliot to make tests of the various stones employed as macadam, the experiments being specially directed to ascertain* Roads and Pavements in France, by Alfred P. Rockwell.

+ Engineering, 30th September 1898.