CELLULAR AND NON-CELLULAR PLANTS 53 



cally this union often takes place at a very early stage, the asexual 

 germs becoming associated together to form a multicellular bod)-. The 

 multicellular coenobia of certain Protococcales, such as Pediastrum 

 and Hydrodietyon, arise in this way ; since those members of this 

 group of Algae which lead a solitary existence (" eremobic " forms) are 

 undoubtedly homologous with single cells, say of a Pediastrum colon}', 

 they may properly be designated unicellular organisms. In other 

 cases the asexual reproductive cells may never display any tendency 

 to disperse, but remain united from the moment of their origin (by 

 division) onwards. Thus Niigeli assumes that the Nostocaceae, in which 

 the plant-body consists of multicellular filaments, are phylogenetically 

 derived from unicellular Chroococcaceae ; he supposes that in the case 

 of some of the Chroococcoid ancestors of the Nostocaceae the several 

 individuals produced by division of a single cell ceased to separate 

 from one another and remained permanently united, thus giving rise 

 to a single multicellular organism. Nageli, in fact, goes so far as to 

 suggest that this case illustrates a " principle which underlies the 

 formation of tissues throughout the vegetable kingdom." 23 



There can be no doubt, however, that the multicellular condition 

 has also arisen in other ways. In the genus Gaulerpa, for example, 

 the plant-body, though containing a single undivided protoplast as in 

 all Siphoneae, is nevertheless differentiated into organs resembling 

 rhizomes, roots and leaves, and altogether closely approximates to the 

 habit of a (terrestrial) Higher Plant. Such a plant cannot properly be 

 termed " unicellular " ; on the contrary the Gaulerpa individual is 

 clearly comparable to an entire multicellular plant, and its continuous 

 but multinucleate protoplast corresponds to the totality of the separate 

 uninucleate protoplasts of the multicellular organism. If this view is 

 the correct one, it follows, as was first recognised by Sachs, 24 that not 

 only Caulerpa, but also the rest of the Siphoneae and all the Phyco- 

 mycetes, instead of being included among unicellular organisms, ought 

 rather to be classed as non-cellular (or coenocytic 24 ") plants, and thus 

 contrasted with the ordinary cellular forms. 



Since very few non-cellular plants are highly differentiated, the 

 acquisition of cellular structure must have been a powerful factor in 

 the evolution of the higher forms of plant-life. The multicellular con- 

 dition does in fact involve such undoubted advantages, that it is 

 improbable that it owed its origin in every case to the modification of a 

 reproductive process, as conjectured by Nageli : in certain instances it 

 must have arisen by a repeated septation and consequent subdivision of 

 the primarily continuous protoplast of a non-cellular plant-bod}'. The 

 advantages attendant upon cellular structure are of course quite 

 independent of the mode of origin of that structure; in other words, 



