1 3 
Davis. — Spore Formation in Derbesia. 
believing that the asters associated with mitosis are such dynamic centres, 
and their radiating fibres either the paths of delicate protoplasmic streams 
or developments from such streams. According to this view the aster 
stands as the morphological expression of dynamic activities, and the 
centrosome and centrosphere are rather the result of these activities, 
associated with the rays converging to a common point, than the cause 
of the structure. 
The structure of the cilia-forming apparatus in epithelial cells of 
animals is of interest in relation to this study of Derbesia. Earlier views, 
which held the granules below the cilia to be centrosomal in nature, have 
been discredited by later work. The subject is discussed in a paper of 
Wallengren (’05) on Anodonta. Wallengren has followed the mitotic 
division of the nucleus in the epithelial cells from the gills of this mussel. 
A centrosome (diplosome) is present, lying among the granules at the bases 
of the cilia, and accompanies the nucleus in its mitosis. During the mitosis 
the cilia first disappear, followed by the basal granules and the fibrils that 
extend from them into the cytoplasm. The cell divides by constriction 
after which a thick protoplasmic layer is formed under the new cuticula. 
Granules arise in this layer of protoplasm without genetic relation to the 
mitotic figure. From each of these granules a fibril is first developed 
extending into the cytoplasm, after which a cilium grows out through the 
cuticula. The granules, fibrils, and cilia form, all together, the ciliated 
apparatus corresponding to the blepharoplast of plants with their cilia and 
internal fibrils, when present. 
The significant feature of this history of the development of the cilia- 
forming apparatus in epithelial cells, in relation to the origin of the 
blepharoplast in plants, is its independence of centrosomes or other struc- 
tures of the mitotic figure. The fibrils penetrating the interior of the cell, 
however, probably hold important physiological relations to the nucleus. 
Such fibrils are well known in epithelial cells, and have been described and 
figured as penetrating to the protoplasm immediately investing the nucleus ; 
see discussion of Gurwitsch (’04), p. 67 . Their presence indicates that the 
granules bearing cilia have intimate dynamic connexions with the proto- 
plasm and nucleus of the cell. The blepharoplast of Derbesia during its 
development shows clearly a similar physiological dependence upon the 
nucleus and cytoplasm, a dependence which probably lasts as long as the 
cilia are motile, although conspicuous protoplasmic strands connecting 
the nucleus with the blepharoplast are not evident after the blepharoplast 
is fully formed. 
Studies upon Infusoria have established the presence of granules 
at the bases of the cilia, which in some forms are, of course, very numerous. 
Maier (’03) believes that these can hold no possible relation to centrosomes, 
which are unknown in the Infusoria. These conclusions are thus in accord 
