wonderful and useful machines ever used. The actual operations of
making a thread are however practically as left by Samuel Crompton over
a hundred years ago. It is only in details of mechanism involved in
making the various operations more perfectly automatic, and of greater
size and productiveness, that the long line of inventors since
Crompton's first mule was made, has been engaged.

To-day, such is the great size and wonderfully perfect automatic action
of these machines, that they are found 120 feet long, while in width,
over all, they may be 9 or 10 feet. Such a mule of this length would
contain over 1300 spindles, each spinning and winding 64 inches of
thread in about 15 seconds, and one man with two youths would be
sufficient to give all the attention such a machine required.

Independently of a vast number of inventors of smaller importance, there
are several names which stand out in greater prominence in the history
of the developments of the mule. Among these names must certainly be
placed, ahead of any others that might be named, that of Richard Roberts
of Manchester, who succeeded in 1830, after about five years'
application, in making the mule self-acting.

A good number of ingenious individuals had contributed more or less to
this result between the dates of Crompton's and Roberts' inventions, and
doubtless the results of the labours of these would be of great service
to Roberts in his great task.

Indeed, several inventors had previously brought out what might be
termed self-action mules, but it remained for Roberts to endow it with
that constant and automatic motion which obtains to-day in practically
the same form as left by him.

The special portion of mechanism with which his name is more especially
identified, is what is denominated the "Quadrant." This is practically
the fourth part of a large wheel, which is so arranged and connected
that it performs almost exactly the same functions on a mule that
Holdsworth's differential motion performs on the bobbin and fly frames.

To look at it, one would imagine it to be--what it really is--one of the
simplest pieces of mechanism possible, yet the actions performed by it
are complex and beautiful in the extreme. Later on, these actions of
the Quadrant will be carefully examined.

Image: FIG. 26.--Mule head showing quadrant.

The self-actor mule is an intermittent spinning machine, _i.e._, it is
not continuous in action, as are most machines used i