VISIBLE CONSTRUCTIVE ACTIVITY IN PROTOPLASM. 575 



thus become connected into a tiny disc. This disc consists of as many cells as there 

 are connected clumps of protoplasm, and presents almost the appearance of a honey- 

 comb. Out of this combination each cell can again separate itself from its com- 

 panions, its protoplasm can divide up afresh, and generally the whole process 

 described above may be repeated. 



The Water-net and the discs of Pediastrum develop young nets and discs 

 accordingly, from the divided protoplasm in the individual cells. These escape as 

 small colonies of cells from the space in which they were formed, and here a 

 definite isolation of the young cell-colony occurs. In Glceocapsa, on the contrary, 

 of which a species (Oloeoca'psa sanguinea) is represented in Plate I., figs, n and o, 

 the young cell-groups remain joined together. Each cell always divides up, two 

 and two, into protoplasmic clumps, which surround tliemselves innnediately with a 

 thick cell-wall. The old cell- wall, however, does not disintegrate nor rupture ; it 

 does not allow the young cell-colony to escape, but it stretches, and the young and 

 old cell-walls ai-e now seen layered one above another. If this pi-ocess is repeated 

 many times, protoplasmic balls arranged in pairs are to be seen inserted within a 

 whole system of concentrically stratified cell-walls. A process similar to that just 

 described is observed in the ovules of seed-plants, and has been called, though not 

 very happily, "free cell-formation". 



Gemmation is essentially different from the process just described. It is 

 observed in plants both with and without chlorophyll, but is not really frequent 

 in the vegetable kingdom. Its characteristic feature is that the protoplasm at a 

 certain point of the cii'cumference of a cell pushes outwards, and in this way a wart 

 or bud-like elevation of the cell-wall, an actual protuberance, arises which, though 

 at first not very prominent, soon increases in area, and in the end assumes the size 

 and shape of the body from which it was produced. We may distinguish two 

 kinds of gemmation. Either an open communication is maintained between the 

 outgi'owth and the structure from which it was produced, and no separation occurs 

 at the place of origin ; or, the parent cell is shut off" from the outgrowth by a cell- 

 wall which subsequently splits, and the outgrowth is detached from the cell-body 

 from which it arose. Very pretty examples of the first kind are exhibited by tJie 

 Siphonese, especially in Vaucheria, illustrated in Plate 1, fig. a. The tubular cells 

 appear branched, each branch consisting of a tube ending blindly, and all these 

 branched tubes are in free communication with one another. The entire Vaucheria 

 is really only a single, much-branched cell — of course a cell which must be called 

 gigantic in comparison with ordinary plant-cells. Species of the genus Bryopsis 

 shape themselves similarly, but in these the outgrowths are much more regular 

 than in Vaucheria, the whole cell, branched and thus pouched, almost resem- 

 bling a moss with axes, leaves, and rhizoids. In the genus Caulerpa the cell 

 also produces outgi-owths, some of which resemble roots, whilst others imitate the 

 shapes of leaves, reminding one, in some species, of small fem-fronds. 



Of the second kind of gemmation yeast may be taken as a type. The shape 

 of individual yeast-cells is ellipsoidal. When the yeast-cell grows, the elliptical 



