lines and columns more quickly than by hand; but as he was personally wholly unacquainted with mechanism, he desired someone to elaborate all the details of such a machine, and asked me if I would professionally study the subject for him, and prepare models to illustrate each proposition. The matter seemed a very difficult one at first sight, and I said that it would be impossible for me to devote more than a portion of each day to its consideration. It was then arranged that I should give as much thought to the subject as I could, consistent with due attention to my general business, and to these terms was attached a guinea per day as a consulting fee.

The general idea on which the machine was based was the arranging of the respective letters in long narrow boxes, from which a touch of the key referring to any particular letter would detach the type required; this, when set at liberty, was to slide down an inclined plane to a terminal point, where other mechanism was to divide the letters so received, into lines if required, and thus build up a page of matter, such as a column in a newspaper, etc.

It will be at once understood that this was not a very simple matter, in consequence of the many signs required. We have first the twenty-six small letters of the alphabet, and the double letters, such as fi, fl, ff; ffi, ffl; then we have the points, or punctuations, signs of reference, etc.; there are also the ten figures and the twenty-six capital letters and their respective double letters, as well as blank types, called "spaces," of different thicknesses, required to divide separate words from each other, etc, Now, as a primary necessity, these numerous letters, when wanted, must, of course, come from different places, and all must descend grooves in the inclined planes in precisely equal times. The time of the whole journey down the incline, say, 2 ft. long, must not occupy any one type more that one-hundredth of a second more or less than the one before or behind it, or its arrival will be too soon or too late, and the word will be wrongly spelt. Thus, suppose the word ACT is required, and the keys A, C, and T, are touched rapidly in succession. If the letter C should arrive first instead of A, the word would not be " ACT" but "CAT," and so for every word. A type that is less than 1 in. in length must never, on its journey, arrive its own length in advance or in the rear of the others that are simultaneously rushing down the inclined plane to the same terminus.

The difficulty that this fact presented was almost beyond belief. Many models were made and much study devoted to it. Thus, suppose a type detached at the point A in the accompanying diagram (Fig. 12) is required to slide down the inclined plane to C, and another one from the point B, is immediately to follow, it will be seen that not only is the road to be travelled by A much longer than that by B, but B also has the advantage of coming straight down the inclined surface, encountering friction only on the one surface on which it rests; while A has not only got a longer journey to perform, but it lays its whole weight on the inclined surface, and rubs also against the inclined side of its groove, thus causing additional friction, so lessening the speed of its descent, and resulting in the arrival of B at its destination before, instead of after, A.

The result of studying this part of the question forced on my mind the important fact that the grooves on the surface of the inclined plane would have to be all of precisely the same length, and every letter, in descending, would have to encounter exactly the same amount of sideway rubbing surface. This knotty point was at last settled in so simple and perfect a manner, that when I had accomplished it I felt half ashamed that it had so long eluded me. The form of grooved incline thus indicated ensured a perfect spelling of every word, and removed the greatest obstacle on the way to success.

The diagram, Fig. 13, represents a portion of the inclined plane, with its small shallow grooves so arranged that any one of the letters a, b, c, d, e, f, g, and h, at the top of the inclined plane would, if allowed to slide down this series of curved grooves, pass along precisely similar paths, and travel precisely equal distances, before arriving at the terminus C.

It will be readily understood that a simple extension of this system would allow any number of letters arranged along the upper line to reach the terminus in the same time; hence each one would arrive in the order of its departure and every word would be spelt correctly.