WITH NOTES ON OTHER SPECIES. 03 



31), probably many times repeated, but whether all the original nuclei are involved 

 in these fusions, or whether some of them degenerate and disappear like those of the 

 periplasm of Peronospora, as described by Wager ('89), is uncertain. When the 

 protoplasmic layer reaches its greatest density and regularity of arrangement, the 

 nuclei are in this indistinct condition (Fig. 33). Just Avhen and how they regain 

 the vegetative structure, I am not jet able to say, having unfortunately failed to 

 obtain sections of oogonia at the stage of the formation of the oosphere origins and 

 of the rounding off of the oospheres. It is certain, however, that, as Hartog has said 

 ('89), and contrary to the statements of Dangeard ('90), the young oospores contain 

 but a single i-ather large nucleus (Fig. 35). This is commonly true also of the fully 

 formed oospheres, but sometimes these contain two nuclei which have not yet fused, 

 though usually lying near together (Fig. 36). Hartog ('92) states that in Saproleg- 

 nia the reduction of the number of nuclei to that of the future oospheres is completed 

 as early as the begiuning of the formation of the origins, while in Aclilya it may be 

 delayed until the young oospore. My observations agree with these so far as A. 

 Americana is concerned, but I have not examined any species of Saprolegnia. The 

 single nucleus, or the two which are to form it, shows the structure and reactions of 

 the vegetative nuclei. 



Although its protoplasm is little denser than that of the vegetative threads, the 

 antheridium is plurinucleate (Fig. 34, a). When the fertilization tube is formed, 

 most of the protoplasm and usually all of the nuclei of the antheridium pass into it 

 (Fig. 34, a). Hartog states ('92) that the nuclei of the tubes are derived by divi- 

 sion from those of the antheridia ; but, so far as A. Americana is concerned, I have 

 seen no reason for supposing that nuclear divisions occur here more than in the other 

 reproductive oi-gans. The number of nuclei in different antheridia of this species 

 does not vary widely, and the number in a tube corresponds pretty closely, as a rule, 

 to the number in an antheridium (Fig. 34, a) ; and as the growth of the fertilization 

 tube is accomplished apparently by the migration of the protoplasmic contents of the 

 antheridium, and not by any increase in its amount, there is no a j)riori reason for 

 nuclear division, under the circumstances. After passing into the tubes, the nuclei 

 undergo no change. It is occasionally possible to find a fortunate section through 

 the sexual organs and oospores, like that figured in Fig. 34, a, which shows well their 

 relations and the fate of the tubes. And one always finds that, in whatever stage of 

 development the oospores maj" be, the tubes are completely closed, as was stated by 

 DeBary ('81), and show their walls sharply defined throughout ; and that their pro- 

 toplasm and nuclei are in essentially the same condition until they begin to degener- 

 ate after the complete ripening of the spores. One sometimes observes a nucleus in 



