The old cellulose wall shows signs of great tension, its conical extremity rounding out under the slowly increasing pressure from within.  Suddenly it gives way at the apex.  At the same instant, the inclosed gonidium (for it is now seen to be fully formed) acquires a rotary motion, at first slow, but gradually increasing until it has gained considerable velocity.  Its upper portion is slowly twisted through the opening in the apex of the parent wall, the granular contents of the lower end flowing into the extruded portion in a manner reminding one of the flow of protoplasm in a living amoeba.  The old cell wall seems to offer considerable resistance to the escape of the gonidium, for the latter, which displays remarkable elasticity, is pinched nearly in two while forcing its way through, assuming an hour glass shape when about half out.  The rapid rotation of the spore continues during the process of emerging, and after about a minute it has fully freed itself (Fig 1, a).  It immediately assumes the form of an ellipse or oval, and darts off with great speed, revolving on its major axis as it does so.  Its contents are nearly all massed in the posterior half, the comparatively clear portion invariably pointing in advance.  When it meets an obstacle, it partially flattens itself against it, then turns aside and spins off in a new direction.  This erratic motion is continued for usually seven or eight minutes.  The longest duration I have yet observed was a little over nine and one-half minutes.  Hassall records a case where it continued for nineteen minutes.  The time, however, varies greatly, as in some cases the motion ceases almost as soon as the spore is liberated, while in open water, unretarded by the cover glass or other obstacles, its movements have been seen to continue for over two hours.

The motile force is imparted to the gonidium by dense rows of waving cilia with which it is completely surrounded.  Owing to their rapid vibration, it is almost impossible to distinguish them while the spore is in active motion, but their effect is very plainly seen on adding colored pigment particles to the water.  By subjecting the cilia to the action of iodine, their motion is arrested, they are stained brown, and become very plainly visible.

After the gonidium comes gradually to a rest its cilia soon disappear, it becomes perfectly globular in shape, the inclosed granules distribute themselves evenly throughout its interior, and after a few hours it germinates by throwing out one, two, or sometimes three tubular prolongations, which become precisely like the parent filament (Fig 2).