Several such axes, with their figure wheels, might be placed around a system of central wheels, with which the wheels of any one or more axes might at times be made to gear. Thus the figures on any one axis might, by means of those central wheels, be added to the figure wheels of any other axis.
But it may fairly be expected, and it is indeed of great importance that calculations made by machinery should not merely be exact, but that they should be done in a much shorter time than those performed by the human mind. Suppose there were no tens to carry, as in the first of the two cases; then, if we possessed mechanism capable of adding any one digit to any other in the units’ place of figures, a similar mechanism might be placed above it to add the tens’ figures, and so on for as many figures as might be required.
But in this case, since there are no carriages, each digit might be added to its corresponding digit at the same time. Thus, the time of adding by means of mechanism, any two numbers, however many figures they might consist of, would {60} not exceed that of adding a single digit to another digit. If this could be accomplished it would render additions and subtractions with numbers having ten, twenty, fifty, or any number of figures, as rapid as those operations are with single figures.
〈SUCCESSIVE CARRIAGE.〉
Let us now examine the case in which there were several carriages. Its successive stages may be better explained, thus—
| Stages. | |
|---|---|
| 2648 | |
| 4584 | |
| 1 Add units’ figure = 4 | 2642 |
| 2 Carry | 1 |
| 2652 | |
| 3 Add tens’ figure = 8 | 8 |
| 2632 | |
| 4 Carry | 1 |
| 2732 | |
| 5 Add hundreds’ figure = 5 | 5 |
| 2232 | |
| 6 Carry | 1 |
| 3232 | |
| 7 Add thousands’ figure = 4 | 4 |
| 7232 | |
| 8 Carry 0. There is no carr. |
Now if, as in this case, all the carriages were known, it would then be possible to make all the additions of digits at the same time, provided we could also record each carriage as it became due. We might then complete the addition by adding, at the same instant, each carriage in its proper place. The process would then stand thus:— {61}
| Stages | ||
|---|---|---|
| 2648 | ||
| 4564 | ||
| 1 | 6102 | Add each digit to the digit above. |
| 111 | Record the carriages. | |
| 2 | 7212 | Add the above carriages. |
Now, whatever mechanism is contrived for adding any one digit to any other must, of course, be able to add the largest digit, nine, to that other digit. Supposing, therefore, one unit of number to be passed over in one second of time, it is evident that any number of pairs of digits may be added together in nine seconds, and that, when all the consequent carriages are known, as in the above case, it will cost one second more to make those carriages. Thus, addition and carriage would be completed in ten seconds, even though the numbers consisted each of a hundred figures.
But, unfortunately, there are multitudes of cases in which the carriages that become due are only known in successive periods of time. As an example, add together the two following numbers:—
| Stages | |
|---|---|
| 8473 | |
| 1528 | |
| 1 Add all the digits | 9991 |
| 2 Carry on tens and warn next car. | 1 |
| 9901 | |
| 3 Carry on hundreds, and ditto | 1 |
| 9001 | |
| 4 Carry on thousands, and ditto | 1 |
| 00001 | |
| 5 Carry on ten thousands | 1 |
| 10001 |
{62}
In this case the carriages only become known successively, and they amount to the number of figures to be added; consequently, the mere addition of two numbers, each of fifty places of figures, would require only nine seconds of time, whilst the possible carriages would consume fifty seconds.
The mechanical means I employed to make these carriages bears some slight analogy to the operation of the faculty of memory. A toothed wheel had the ten digits marked upon its edge; between the nine and the zero a projecting tooth was placed. Whenever any wheel, in receiving addition, passed from nine to zero, the projecting tooth pushed over a certain lever. Thus, as soon as the nine seconds of time required for addition were ended, every carriage which had become due was indicated by the altered position of its lever. An arm now went round, which was so contrived that the act of replacing that lever caused the carriage which its position indicated to be made to the next figure above. But this figure might be a nine, in which case, in passing to zero, it would put over its lever, and so on. By placing the arms spirally round an axis, these successive carriages were accomplished.
Multitudes of contrivances were designed, and almost endless drawings made, for the purpose of economizing the time and simplifying the mechanism of carriage. In that portion of the Difference Engine in the Exhibition of 1862 the time of carriage has been reduced to about one-fourth part of what was at first required.
〈ANTICIPATING CARRIAGE.〉
At last having exhausted, during years of labour, the principle of successive carriages, it occurred to me that it might be possible to teach mechanism to accomplish another mental process, namely—to foresee. This idea occurred to me in October, 1834. It cost me much thought, but the {63} principle was arrived at in a short time. As soon as that was attained, the next step was to teach the mechanism which could foresee to act upon that foresight. This was not so difficult: certain mechanical means were soon devised which, although very far from simple, were yet sufficient to demonstrate the possibility of constructing such machinery.
The process of simplifying this form of carriage occupied me, at intervals, during a long series of years. The demands of the Analytical Engine, for the mechanical execution of arithmetical operations, were of the most extensive kind. The multitude of similar parts required by the Analytical Engine, amounting in some instances to upwards of fifty thousand, rendered any, even the simplest, improvement of each part a matter of the highest importance, more especially as regarded the