116 ARTHUR: IMPORTANCE OF THE SPERMOGONIUM 
duced in chains, and those of the uredo being produced singly on 
evident pedicels, except in case of the genera of the Co/cosporieae 
and Chrysomyxeae, where they as well as the aecidiospores are in 
chains. In all cases the aecidium is developed before the uredo 
in point of time. 
This brings us to a consideration of the succession of spore- 
forms in the cycle of development. In the first place it may be 
said that the order is invariable. The germinating teleutospore 
gives rise to the sporidium, from the mycelium of which is pro- 
duced first the spermogonium, then the aecidium, and later comes 
the uredo, and then the teleutospore completing the cycle. Either 
the aecidium or the uredo may be omitted from the cycle of de- 
velopment, or both may be omitted, as occurs in many species, but 
the spermogonium is rarely omitted, and the teleutospore with its 
sporidium never. The developmental order and the suppression 
of spore-forms may be represented in four series. By using the 
Roman numerals I, II and III for aecidia, uredo and teleutospore 
respectively, the cipher for the spermogonium, and the letter y for 
the sporidium, the series may be shown in tabular form, or they 
may be represented diagrammatically. By far the largest number 
SUCCESSION OF SPORE-FORMS 
A. Complete series. B. Uredo wanting. | C. Aecidia wanting | D. No uredo or aecidia- 
| 
x y y y 
oO | fe) oO Oo 
I | 1 | | = 
II ~ ll | = 
III | III | Ill | Ill 
J = ee 
of species are illustrated by the first series, having all spore-forms 
present ; a much smaller number by the second series, having the 
uredo suppressed ; and a still less number by the third series, 
having the aecidium suppressed. The fourth series, with both 
conidial forms suppressed, claims many species, however, and 
possibly more than either the second or third. In each of the 
four series the spermogonium is sometimes suppressed. Very 
little attention has been given to the presence or absence of the 
spermogonium, however, and the recorded data bearing upon the 
matter are extremely meager. At present it is unwise to venture 
an opinion whether the non-appearance of this structure in any 
