130 PHAEOPHYCEAE 



(3) protection against intense illumination, 



(4) protection against epiphytes, 



(5) protection against covering by sand or silt, 



(6) mucilage organs. 



None of the evidence for any of these suggestions is entirely 

 satisfactory, and the whole problem demands further investi- 

 gation. 



Vegetative reproduction may take place by splitting of the thallus 

 or else by the development of special propagules (Sphacelaria). 

 Asexual reproduction is commonly secured by means of uni- or 

 biflagellate zoospores which are normally produced in speciaUzed 

 cells or sporangia. In one group (Dictyotales) tetraspores replace 

 the zoospores, these bodies being produced in groups of four in 

 each sporangium on plants which do not bear sexual organs. In 

 yet another group (Tilopteridales) asexual reproduction is by means 

 of uni- to quadrinucleate monospores. The homologies of these 

 monospores have been subject to much speculation and they have 

 been variously regarded as equivalent to 



(a) propagules of Sphacelaria, 



(b) simple forerunners of tetraspores, 



(c) degenerate tetraspores, 



(d) parthenogenetic ova. 



The second suggestion is perhaps the most satisfactory in our 

 present state of knowledge, especially when considered in relation 

 to the vegetative characters. Sexual reproduction ranges from 

 isogamy, with both gametes motile and characteristically bearing 

 two flagellae inserted laterally, through a series in which differentia- 

 tion first to anisogamy and finally to oogamy can be traced. Only 

 one species (cf. p. 184) is known in which the ova are retained on 

 the parent plant, so that, apart from this exception, fertiUzation 

 always takes place in the water. The change from isogamy to 

 anisogamy is also accompanied by a corresponding differentiation 

 of the gametangia. 



Both unilocular and plurilocular sporangia are commonly found, 

 but the fate of their products varies considerably (cf. p. 247). Most 

 species show an alternation of generations, but this is by no means 

 regular as there may be considerable modifications. Indeed, the 

 alternation in the Ectocarpales is so irregular that it has been 



