-^-'//--AV( ./.\ .\"./ /TA-./Z./.s /-. . xxxvill. 



cells of the more complicated ccLMio.^amete^ ,oo^o„ia aiul anthe- 

 ndia) are also terminal cells. All .,f thc^c >>cxiial or-an^ are 

 multinucleate. In the Siphonales (X'auchcria cxcci.ted) alTof 



M.on-s ot the protoplasm. The same o.,uhtinn. of sexual (le<;en- 

 cratKrn are also found in the .,o...nia of \ 'aucheria (Davis, : 04) 

 and Saprok-nia (Davis, :o3). The a<aeement of all of the 



structures mentioned above 



l)lasmic behav 



This xiew 



lor seems to establish beyond questir>n their common homology. 



The problems of the ori-in of the C(XMio<.ametes are verv 

 clifficult with the meager evidence at hand. 'll- author believes 

 y the snnple.t types have probably been derived from struc- 

 tiues like the sexual organs of the isogamous Siphonales, which 

 stnictuich gave up the habits of forming uninucleate gametes 

 and actmg as c^^nocytic units became multinucleate sexual ele- 

 ments. A physiological development very similar to such 1 

 change must have taken place in Per()nosi)()ra and' som > ■ ) - 

 of Pythium when their conidia ce:ised f '.n,!!^!!'^-"^!'!!l''''''?,!l 



took the habit of , ... ---- 



regards the C(i'no<'-. 



- ' >>'>nuna, etc.) have developed direcllv from the 

 illiNt r u"" ' ''''''''''' ^'^"^ h^^terogamous algTc, such a 

 not iJI 1!- ; , ;^"'"'^'''''^'^''^'"^'very comi)licated prol)lems that 

 1-c ij;^;::;' -^^^^^^^^^^ nu-nti<.neda 

 ; 04^', : 04/.). ( UMv.guinl'rHre'pn^^^^^^^ t 1 ^ ^^T''* 



field is likely' to p, 



