12 FORMATION OF SPOKES OF RHIZOPUS AND PHYCOMYCES. 



full size, divides rapidly to form a vast number of smaller daughter 

 nuclei. This multinucleated mass of protoplasm is then divided into 

 comparatively large blocks by narrow furrows, cutting progressively 

 inward from the periphery. These furrows cut inward at nearly right 

 angles to the periphery, but, as seen in surface sections, they intersect 

 each other at almost every angle. They are so narrow that they 

 appear in section as single lines which push aside the vacuoles, arrang- 

 ing them in a row on either side. In case the sporangium is .slightly 

 shrunken in tixing, however, they appear as slightly separated sur- 

 faces. As these cleavage furrows grow deeper they branch, curve, 

 and intersect each other until the whole mass is divided into multi- 

 nucleated pieces. These are then divided into uninucleated pieces by 

 furrows cutting inward from their surfaces. 



The nuclei then divide until there are usually from 8 to 12 in each 

 piece. Without further cleavage these multinucleated protoplasmic 

 masses then enlarge somewhat, secrete a protective wall, and become 

 the spores. They then go into a resting condition until germination. 



In PHohohix, Harper traces the entire development of the sporan- 

 gium. He iinds that when it has reached a considerable size its con- 

 tents are divided into three parts — a central vesicle of cell sap, which, 

 from the absence of a smooth, rounded surface, can not be consid- 

 ered as a central vacuole; outside this, a thin layer of spongy proto- 

 plasm with numerous nuclei; and outside this layer, extending to the 

 sporangium wall, a nuich denser mass of protoplasm, also containing 

 many nuclei and a few rounded vacuoles. In the spongy protoplasm, 

 and running parallel to the sporangium wall except at the lower side 

 where it extends to the periphery, a dome-shaped layer of vacuoles 

 then appears. These vacuoles are at first round, but later they be- 

 come flattened parallel to the surface of the sporangium until they are 

 disk-shaped. They finally fuse, edge to edge, to form a cleft, which, 

 with the aid of a circular furrow cutting upward through the spongy 

 protoplasm until it meets the lowest vacuoles in the series, cuts out 

 the columella. This columella is bounded at first by only a plasma- 

 membrane, outside of which is a more or less open cleft. Later the 

 columella wall is formed in this cleft. It has its dome-shaped outline 

 from the first, and does not begin as a cross wall at the base of the 

 sporangium, being rounded upward later by pressure of turgor from 

 below, as is described for Mucor in most standard text-books. (See 

 Bessey's text-book, p. 236.) 



The spore plasm is then invaded by surface furrows cutting pro- 

 gressively inward. These are much like those in Synchitrium^ but 

 wider, owing to the more shrunken condition of the protoplasm dur- 

 ing the process. While this is going on, the vacuoles in the spore 

 plasm become sharply angular, and these angles, continuing outward 

 as furrows, cut into each other and into the furrows from the surface, 



