The trials that afforded the two exceptions were with mortars containing a smaller proportion of cement than the six others. These facts seem to show that the addition of bitter-water, within certain limits, improves the cement, but that beyond these limits it is injurious; and that where the proportions of cement are great, an increased addition of bitter-water may be advantageous. These particular experiments were made in consequence of finding that the addition of a little bitter-water hastened the setting of cement A when immersed.

12th. Mortars of cement and sand are injured by any addition of lime whatever, within the range of the experiments; that is to say from sand 1, lime }, and cement; to sand 1, lime 1, and cement 2. No exceptions in 67 comparisons.

13th. Stone-lime, in the proportions tried, gives better mortar than shelllime, as 153 to 133: but some previous trials had afforded results slightly the best with shell-lime.

Table No. LXVIII.

Trials made in June, 1836, of mortars made in September, 1835. The results show the weights, in pounds, required to separate each inch square of surface of bricks joined by mortars. The object is to compare grout with mortar.

In order to compare the strength of grout with that of mortar, bricks were joined (as before described) with the mortar given in the table-there being four pairs to each kind of mortar. To obtain similar joints of grout, bricks were supported on their ends and edges, in a box large enough to contain all, in such a way as to admit the proper quantity of grout to flow in between each pair. The box was not disturbed until the grout had become quite stiff, when it was first laid on one side, and then taken to pieces. The excess of grout was carefully cleared away from the bricks, which were removed without injury to any of the pairs, and put away by the side of the bricks joined with mortar.

It will be seen that, in every case but one, the grout was much inferior

to the mortar. The average strength of all the mortars in the table is $1.78, and the average strength of all the grouts is 20.06

Changes of bulk on slaking lime—making mortar, grout, &c.

A great many measurements were made of the changes of bulk in the operations of slaking lime, making mortars, &c., and the results, as might be expected, varied with the qualities of the lime. The following condensation of the results may be useful.

1 lime air-slaked gave, as a mean,

1 of air slaked lime in powder and 0.50 water made, as a mean, 0.75 thin paste, 2 trials varying from .70 to .80.

1 of lime (quick) pounded to powder, made 0.90 of powder, 1 trial.

1 of lime slaked to powder, kept dry for 3 months, still measured 1.00,

2

0.65 to 0.67

0.76 thinner paste. 19

0.67 to 0.94

1.84 powder

202 of mortar with 87 of water made 290 of grout.

213

do.

87

do. do. 305

do.

CHAPTER XXIV.

Observations and experiments on Concrete, &c.

It was ascertained, by careful measurement, that the void spaces, in 1 bulk of sand No. 1, taken from the middle of the heap, amounted to 0.33: the cementing paste, whatever it may be, should not be less therefore, than onethird the bulk of this sand. Taking one bulk of cement A, measured in powder from the cask, and a little compacted by striking the sides of the vessel, water was added till the consistence was proper for mortar: 0.35 of water was required to do this, and the bulk of the stiff cement paste was 0.625. To obtain, at this rate, an amount of cement paste equal to the voids (0.33) in the sand, will require, therefore, 0.528 cement in powder, and 0.185 of water, or

Dry sand,
Cement in powder,
Water,

1.000

.528 making a bulk of 1.000 of mortar.

.185

It is by no means certain that a mortar composed on this principle will be the most tenacious that can be made on the contrary our experiments indicate that the mortar would be stronger with a smaller proportion of sand; but possessing the minimum quantity of cementing constituent, which is by far the most expensive ingredient, it affords the cheapest admissible mortar, made of cement and sand; and as it was probable, that it would shrink very little on drying, it was tried as a pointing for exposed joints, and also as stucco, and it answered very well for both purposesbecoming very hard, and never showing the slightest crack. An excess of cement, and a very slight excess of water, above the stated proportions, should be allowed for imperfect manipulation, because the proportions suppose every void to be accurately filled.

Extending the application of this principle to concrete-experiment showed that one bulk of stone fragments (nearly uniform in size, and weighing about 4 oz. each) contains 0.482 of void space. To convert this bulk of stones into concrete, we, in strictness, need use no more mortar than will fill this void space; and to compose this mortar we need use no more cement than is necessary to occupy, in the state of paste, the voids in 0.482 of sand. This concrete would therefore be composed as follows: Stone fragments about 4 oz. each, Sand No. 1

1.0001

.482

.255

.089

making a bulk of 1.000 of

concrete.

Obtaining thus a cubic yard of concrete by the use of one-fourth of a cubic yard of cement in powder, (about one and a half bbls.)

But the above fragments were of nearly equal size, and of a form approaching the spherical: affording more void space than if they had been more angular, and had varied in size from about six oz. to less than one oz. such as would commonly be used. We have found that clean gravel, quite uniform in the size of the pebbles, which were about half an inch in average diameter, afforded voids to the amount of 0.39. And Mr. Mary, a French Engineer, used pebbles, probably mixed of coarse and fine, of which the voids were 0.37. The above allowance of 0.482 for void space is therefore quite large.

In all cases of the composition of concrete, the quantities expressed above, should be ascertained by actual measurement of the particular cement, sand and fragments, or pebbles, that are to be used. No better mode of measuring the void spaces, will be found, probably, than measuring the quantity of water that can be poured into a vessel already filled with stone fragments, pebbles, or sand, as the case may be.

Although the hydraulic property of cement will be the cause, in all cases of its use in concrete, it may happen that the cement at hand is more energetic than is actually necessary, and that the concrete would fully accomplish the object in view, even if it should be two or three weeks in becoming hard and impervious to water. Under such circumstances lime may take the place of part of the cement, with great economy. The lime may

be added either in the state of powder that has been slaked some time, or in the state of paste: but in either case, the previous slaking must be complete.

The mortar is to be made first, and then the pebbles, or broken stones, may be mixed therewith by turning them over several times with the shovel.

When it is to be deposited under water, it is still a disputed point whether the concrete, prepared as above, should be used immediately, or be left in heaps to stiffen to such a degree as to require the use of pickaxes to break down the heaps: but, in works out of water, there can hardly be a case in which it will not be best to place it at once in its allotted space, where it should be compacted by ramming till none of the stone fragments project above the common surface. One or two trials will show how much mortar over and above the strict proportion is necessary in each case.

In circumstances where ramming cannot be applied, as when depositing concrete in deep water, the concrete should be more yielding and plastic -containing a larger proportion of mortar, and the mortar should be rammed before being deposited, in order thoroughly to imbed the larger constituents.

In many situations where concrete may be resorted to with great advantage, the economy need not stop at the above proportions. This substance may be rammed between, and upon, stones of considerable size-the only indispensable precaution being, to make sure that the stones are perfectly clean, are well imbeded in the concrete, and are far enough apart to permit the full action of the rammer between them.

The following case occurred at Fort Adams in October, 1836. The proportions adopted were, fragments of granite, of nearly uniform size, and about 5 oz. each,

Sand No. 1

Cement A, in powder,

Water rather more than

Bulk of

1.000

than 1.000.

Experiment gave 16.683 as the number of cubic feet of concrete made by 1 barrel of cement-187 barrels were consumed which afforded 115.52 cubic yards of concrete. There were also used, 11.29 struck Winchester bushels of sand, and 22.58 struck Winchester bushels of granite frag

Brought over,

There were 151 days labour, applied to making mortar-making concrete-depositing the concrete in its proper place, ramming it into a compact mass, and doing all other work required in the operation. 151 days at $ 0.92. Supervision

Cost of 115.52 cubic yards,

Cost of one cubic yard $ 6.40

$590.24

138.92

10.00

$739.16

Springs of water flowed over this work continually; and were allowed to cover each day's work. The next morning the concrete was always found hard and perfectly set.

Had we dispensed with one half of the cement used, and used in lieu thereof, as much paste of lime, as the cement dispensed with would have furnished of paste of cement, the cost would have been materially reduced, and the work have been still very hydraulic, and very strong. In that case, the bulk would not have been altered, but would have been as before, 115.52 cubic yards. We should have used 933 bbls. of cement less than we did: and, as cement, in passing to the state of paste, diminishes in bulk in the proportion of 1 to .625, we should have used 93.5×.625 equal to 58.43 barrels of paste of lime. Saving, thereby, the difference between the cost of 93.5 barrels of cement and 58.43 barrels of paste of lime.

93.5 barrels of cement at $ 2.45
58.43 do. of paste of lime at $ 0.60

Amount saved

$ 229.07

36.06

$193.01

$739.16, less $ 193.01, equal $ 546.15; the cost of 115.52 cub. yards. Cost of one cubic yard $ 4.73. Another Instance.

This was rammed into a mould of the capacity of 13.786 cubic feet.

Total cost of 13.786 cubic feet,

0.306 per cubic foot, or $ 8.26 per cubic yard.

cost $2.57

.22

.40

1.03

$ 4.22

This became very hard, and is a very good substitute for stone, in certain applications.

This was rammed into a mould of the capacity of 7.812 cubic feet; and the whole cost was $2.15, being $ 0.276 per cubic foot, or $ 7.45 per cubic yard.