496 Annals of the South African Museum. 



(Fig, 3, d and h), it takes up this stain easily and, ultimately in such 

 stained material, appears as a more or less deep blue diffuse mass in 

 which the lower end of the cell is embedded. The thickening of the 

 other pole is in general not as pronounced, and appears to mark a 

 region of gelatinisation of the membrane in which rupture subse- 

 quently takes place. 



The development of the colony from this initial stage takes place 

 as follows. The original cell gradually grows to quite considerable 

 dimensions (as much as 48 x 20 ,<), although division of the contents 

 may set in before the maximum size is reached. The contents apparently 

 contract slightly away from the membrane and thereupon undergo an 

 oblique division into two, a new membrane appearing around the 

 products (Fig. 3, h, d, and e). In many cases division does not 

 proceed further ; the membrane of the mother-cell ruptures at the 

 apex, and one (Fig. 3, c), or both (Fig. 3, d), of the daughter-cells 

 shift up to the neighbourhood of the opening and become attached 

 there to the mother-cell membrane by a- similar polar thickening, to 

 that which effects attachment in the case of the initial cell. It does 

 not appear, however, that these secondary points of attachment become 

 as markedly mucilaginous as the primary one. The free ends of the 

 ruptured mother-cell membrane are often more or less reflexed 

 (Fig. 3, c and h). The daughter-cells grow and sooner or later divide 

 again in a similar manner (Fig. 3, d), and in this way a repeatedly 

 branched colony may arise. 



It is not at all uncommon, however, for the initial cell to undergo 

 successive division into four parts (Fig. 3, b, g, h,j) before rupture of 

 the mother-cell membrane takes place. Some growth of the products 

 of the first division appears to occur, as the mother-cell membrane in 

 such cases generally appears more or less irregularly bulged out at 

 various points (Fig. 3, g), a phenomenon which may even be observed 

 in cases where division into but two parts has taken place (Fig. 3, e). 

 Of the four products of division, one or two usually move up to the 

 aperture of the ruptured mother-cell (Fig. 3, h,j), whilst the remainder 

 tend to become disposed in a more or less linear series within the 

 original mother-cell membrane (Fig. 3, h,j, k), one generally remaining 

 at the base. Cases where three cells have moved up to the opening 

 (as in the lower right hand part of Fig. 3, h) are rare.* It seems 



* The colony shown in Fig. 3, h is somewhat unusual, and I am not certain 

 how to interpret it. It may be the result of a number of zoospores settling 

 down and germinating at the same spot (i. e. a collection of colonies), or it may 

 be due to a considerable number of daughter-cells having shifted up to the 

 opening of the mother-cell. 



