88 THE SAPKOLEGNIACE.^ OF THE UNITED STATES, 



In case of the androgynous species, the antheiidial branches begin to appear soon 

 after the oogonial ones (Fig. 21) ; but the antheridia have usually been formed and 

 come into contact with the oogonium before the basal wall of the latter has appeared. 

 Commonly they are applied by their sides to the oiigonial wall, but in a few species 

 (Figs. 54, 94) they present their ends, so that their longer axes are nearly at right 

 angles with the wall, instead of parallel with it, as in most species. 



When a certain stage in the differentiation of the contents of the oogonium has 

 been reached, as will be described later, the antheridia give rise, in most cases, to 

 slender tubular outgrowths from the sides applied to the oogonial wall. The tubes pene- 

 trate this wall and grow into the cavity within^ remaining simple or branching. These 

 structures are, morphologically at least, fertilization tubes. In species with pitted 

 oogonia the antheridia are often, but not always, applied to the thin places, and the fer- 

 tilization tubes can thus penetrate more easily. But the old view that the pits are 

 perforations of the membrane for the admission of these tubes is untenable fi'om any 

 point of view. 



The sexual organs are, then, specialized branches ; but their special character 

 does not prevent their showing occasional reversional features, recalling their primi- 

 tive nature. The oogonial branch may subdivide and bear an oogonium on each di- 

 vision (Fig. 22). It is not uncommon to see a young oogonium which has ceased its 

 normal development and produced one or more smaller oogonia by proliferation from 

 its surface (Fig. 23) ; and I have seen in A. Americana a fully formed oogonium, 

 which, after the formation of its basal wall, had reverted to the vegetative condition, 

 so to speak, and had given rise to an oogonial and an antheridial branch which had 

 reached their full normal development (Fig. 24). The production of an antheridial 

 branch from the very body of an oogonium occurs so commonly as to be normal in 

 A. racemosa (Fig. 96), in which the branch arises as often above the basal wall of the 

 oogonium as below it ; but the antheridium is probably cut off from its branch before 

 the oogonium is cut off. Further proof that there is no fundamental difference 

 between the two kinds of sexual branches may be found in the fact that antheridial 

 branches may produce at their ends small, though abortive, oogonium-like swellings, 

 even after giving off branchlets with normal antheridia (Fig. 25). I believe I have 

 also seen the formation of a spore-like body in a similar swelling, as observed by Zopf 

 ('90) in Peronospora calotheca, but have not been able to feel certain on this point. 



Having now traced the origin of the sexual branches and the formation of the 

 sexual organs in general, we pass to the detailed examination of the fate of their pro- 

 toplasmic contents. The dense mass of protoplasm which fills the oogonium when it 

 has attained its final form, contains, like that of the sporangium, very numerous 



