129 



of giMirra tlii' cvstncjiriis nrr not foiiiud liy diii-cl, hut l>y iudircct, outi;in\vtlis 

 from the li'ifh<ij)ii(>re. I'dr I'xiiiiiplr, in ( '(illilhiuiiuioii, tlii' iVrtilizing substance 

 passes lid Ml tlic lricli(i<r.vnc, if ;it all, tliniuyli the Irirlidplinrc and .lometlmes nrreritl 

 cells beloir. hrfini' <'(,TtMiii lateral eidls ai'e ri'aeiied, wliicli lieeonie spoi-e hearins. 

 In Dudre.-tiuuja the trieliophore nf one procai'pie lilanieiit i^ives rise to si-veral 

 lateral tiilies, itself heeominn' non-spore liearinn', which eonvev the fertilizinjj im- 

 pulse to certain c(dls of other pi-ot'arpic lilaniciits which have no trichosivne in 

 other parts of the frond. Thus, cyslocarps are formed at great distances from 

 the trieiiouyne. In Pnhiidrs a similar arrangement ohtains, but the cystocarps 

 are not the auxiliary cells of the procarpic iilamiMits; they an- lateral expansions 

 of the tricho[)hoi-ic tubes. In other genera, the evidences :ire ample to disprove 

 the act of fertilization. In Pliiola xerrala, as far as observations went. 1 found the 

 antheridial plants very rare. .Not a single anlheridial plant or an antheroz<iid 

 was found in contact with the tridiogy lie. A very peculiar phas^' in the develoi)- 

 ment of tlii' cystocarp was noted a tricliophore with live trichogyns ; the anthero- 

 zoids would have had to pass through two, in some cases three, ci-lls to ha\e ler- 

 tilized tile lateral c-arpogenic cells. 



In I>atr((cliospentuim, the carpogiuiiiim develojis cystot-ar])s without any con- 

 nection with tile trichogyiie— an entirely uon-se.xual process. A cellulose plug 

 separates the trieiiogyne from the tricliophore. In three species of this genus 

 cellulose plugs wt're constant, and two nuclei in the trichogyne. Only aboizt 10 

 per cent, of specimens examined showed evident fusion of antherzoids with the 

 carpogonium. The oolilastenia tihuiients are not outgrowths of the carpogonium, 

 but from cells lielow, whiidi is in ojjposition to Thnret and Bornet. Physiologic- 

 ally, tlu'u, two great types of re]iroduction seem to oi'i'ur : one in which eysto- 

 oarps develop from the carpogonium ; the other in which cystocarps develop from 

 the cell below. 



To Slim up (lie sex phdsi' of Florldeir: Antherozoids very rare, non-motile and 

 in some c-ases wanting. Only a very few eases of actual fertilization recorded by 

 algologists. Not definitely known wiiether antherozoids fuse with cari)ogonia or 

 whether apogamy is the rule. The commingling of the nucleus of the antherozoid 

 with the nucleus of the tricdiogyne and the contents separated from the carpo- 

 gonium by a cellulose plug — a hint, no doubt, toward an old hereditary act of 

 ancestral forms. 



Therefore, the strongest point in ilie investigation of the Florideit is the sepa- 

 ration of the trichogyne from the trichoiiore. and fertilization not accomjilished. 



