VISIBLE CONSTRUCTIVE ACTIVITY IN PROTOPLASM. 577 



exactly the shape which the parent had possessed. These elegant desmids claim 

 our interest because although their wall is composed principally of cellulose, and 

 is comparatively thick, it has a determinate outline, and in this, and in its pro- 

 tuberances, and, generally, in its entire shape, it is governed by the living cell-body 

 which has formed it. If such a desmid-cell extends in length or breadth, if it 

 bulges out in one place and remains constricted in another, this is caused only by 

 the activity of the protoplasm, which shapes and transforms the body in accordance 

 with the constructive plan of the species. 



The continued connection of the cell-couples produced by division, and the 

 origination of extensive cell-aggregates by the repeated formation of partition 

 walls, is much more usual than their separation. No less than five different 

 modifications may be distinguished of this process, which is connected with the 

 construction of so many plants. 



In the green aquatic filaments, of which two species (Zygnema pectinatwm 

 and Sjnror/yra arcta) are illustrated on Plate I. figs, vi and I, a wall may be de- 

 veloped by the protoplasm of each cell, which is first formed as a ring-like band 

 on the already existing cell-wall, and resembles the diaphragm in the tube of a 

 microscope. Gradually from this circular band a completely closed partition-wall 

 is produced, and the single cell becomes divided into two. In each of these cells 

 this process may be repeated, and thus in a very short time may arise a row of 

 four, eight, sixteen, &c., cells. These remain connected with one another, and the 

 whole row constitutes a cylindi-ical tube divided up by numerous transverse walls. 

 If the single cells are much swollen at the sides, the row of cells has the appearance 

 of a string of pearls. The intercalated partition-walls in these plants are all 

 developed parallel to one another, and are placed at right angles to the axis of the 

 cell-filament. 



The fact of these intercalated partition-walls being parallel distinguishes this 

 process from another, which is characterized by the fact that the insertion of 

 partition-walls occurs in two dimensions of space. In this case neither partitioned 

 tubes nor strings of pearls arise, but groups of cells arranged in one plane, which 

 are plate-like in appearance, and, to the naked eye, look like membranes or leafy 

 structures. This kind of structure is often shown by marine algee which grow on 

 stones. If all the cells adhere to the substratum, as in Hildenbrandtia, the out- 

 line of the plate is more or less circular, and green or red patches are to be seen on 

 the stone, which continually increase in size without altering their general form. 

 In this case there is no obstacle which could restrict the circular shape of the cell- 

 plate. If, on the other hand, only some of the cells adhei-e to the substratum, 

 while the others rise up from the stone, so that the whole floats in the water as 

 a thin film (only attached to the substratum at one point), then the further 

 development is unsymmetrical. It is suppressed towards the substratum, and the 

 whole layer usually has a fan-like appearance. 



If the arrangement of partition-walls in a cell occurs in three dimensions of 

 space, a tissue is then formed. The tissue-body developing most regularlj^ in this 



Vol I. 37 



