ARCHER, ON STEPHANOSPH ERA PLUVIALIS. 121 
cells occasionally, though rarely, occur (and still more rarely 
families of five, six, or seven), through arrested subdivisions 
of the parent primordial cells; whilst I have once or twice 
met with families of sixteen cells, arising from the “ transition 
generations’ (Nag.) being carried one stage further—that is, 
the (normally) eight cells each dividing once again before 
arriving at the fully formed or “ mature stage,” which may in 
similar manner be called the ‘“ permanent generation” 
(Nag.). 
A second change connected with the propagation consists in 
the disassociation from the hyaline sphere, and the contraction 
of the primordial cells, as before, into rounded bodies ; but in 
this instance this is followed by the formation round each of 
a special cellulose wall, through which protrude two cilia, by 
the agency of which these Chlamydomonas-like cells swim 
vigorously about, at first within the old parent-globe, after- 
wards outside it, when they have made their escape by its 
destruction. Finally, these Chlamydomonas-like bodies come 
to rest, lose their cilia, and assume a Protococcus-like appear- 
ance, or these Protococcus-like resting cells may apparently 
be formed without an intermediate motile Chlamydomonas- 
like state. These Protococcus-like cells have the power of 
increasing in size after their formation, and before any 
further development. Cohn and Wichura’s researches have 
demonstrated the remarkable fact that not only are these 
capable of revival wpon being covered with water after com- 
plete and lengthened desiccation, but that such desiccation is 
absolutely necessary to induce further changes in the direc- 
tion of the renewal of the Stephanosphera-globe. A few 
hours after being remoistened the cell-contents of these 
resting-spores become divided into two, then into four 
(possibly sometimes eight) daughter-cells, the cellulose wall 
vanishing. These daughter-cells become biciliated, and 
presently one by one separate and swim freely away as so 
many pear-shaped zoospores. Afterwards, coming to rest, 
they acquire a membrane, which at first closely surrounds the 
contents, but presently expands, and, standing off, leaves the 
body of the primordial cell in the middle, which protrudes a 
pair of cilia through the wall, thus again assuming a Chlamy- 
dococcus-like appearance. ‘The contents often in this stage 
present a number of hyaline, frequently branched or forked 
projections in various directions from the outer protoplasmic 
layer (like those from the opposite ends of the primordial 
cells of the mature plant), and which touch the inner surface 
of the cellulose wall. But after an interval these Chlamy- 
dococcus-like structures begin to divide ; if the protoplasmic 
