118 ARTHUR: IMPORTANCE OF THE SPERMOGONIUM 
that with a few exceptions the spermogonia are the constant ac- 
companiment and precursors of the aecidia, and that this points to 
a sexual relation between them (page 169). Inthe second edition 
of the work in 1865 this statement is elaborated, and has been 
made familiar to English-speaking botanists through the transla- 
tion of 1887 (page 276). As exceptions DeBary recorded in his 
earlier work that in cultures of Endophyllum Sempervivt, which 
usually produces spermogonia, repeated generations of aecidia 
occurred without a trace of spermogonia. In 1879 Schroter 
observed that aecidia of Uromyces on Ervum and of Puccinia on 
Galium Aparine are produced throughout the warm months, but 
that only the first generation in springtime is accompanied by 
spermogonia. In 1891 Barclay made observations upon a 
Uromyces on Jasminunt sempervivum in India which produced 
aecidia accompanied by spermogonia when sporidia from germin- 
ating teleutospores were sown, but which produced aecidia unac- 
companied by spermogonia when aecidiospores were sown. In 
1895, in an article on rust-fungi with repeated formation of aecidia, 
Dietel added to the number of such species, coming to the con- 
clusion that, in those species of Uromyces and Puccinia which 
form aecidia and teleutospores but no uredospores, the aecidio- 
spores have the power again to bring forth aecidia, provided the 
mycelium is not perennial in the host. He called the aecidia 
arising directly from germination of teleutospores ‘“ primary 
aecidia,’’ and those arising from germination of aecidiospores 
‘secondary aecidia,” and noted that spermogonia were usually 
absent from the latter. The same descriptive method has since 
been employed with the uredo; the primary uredospores being 
those which arise first accompanied by spermogonia; and the 
secondary uredospores, often of a different size and appearance, 
being those which come later without spermogonia. 
Going back to the statement of DeBary, which is also the 
accepted view of every subsequent writer, that usually the spermo- 
gonia are the accompaniment and precursors of the aecidia, or in 
a few cases, we may add, of the uredo, we are now in a position 
to point out that the statement is only true in a superficial way, 
and in reality is misleading. Every one who has made cultures of | 
the rusts knows that, in about a week after sowing the germinat- 
i 
