±1 ABSENCE OF ASCOGONIUM 



(female) cells takes place just before the formation of the 

 secidium or its representative ; in the reduced micro- and other 

 forms, such as /'. Adoxae and Uromyces Scillarum, it is probable 

 that the conjugation of the two (vegetative) cells takes place 

 at some more or less indefinite period before the formation of 

 teleutospores. 



There is a general agreement among investigators that a 

 structure resembling an ascogonium (from which the mass of 

 basal cells may be supposed to originate) does not exist in the 

 Uredinales, notwithstanding the suggestions to that effect by 

 Massee (1888) and Richards (1896). If it did exist, or had 

 existed, it would do something towards accounting for the 

 definiteness in form usually presented by an secidium ; if it is 

 totally unrepresented, the secidium cannot be regarded as a 

 morphological unit, but only as a collection of female cells. It 

 is possible that traces of its existence are shown by the large 

 multinucleate cells, containing 12 — 15 nuclei, which have been 

 described by Olive (1908) in the mycelium at the base of the 

 young secidium of Puccinia Cirsii-ianceolati and by others at 

 the base of teleuto-sori, but this question must remain open till 

 further investigations are made. The existence of an asco- 

 gonium of that kind would, of course, be inconsistent with the 

 trichogyne-interpretation of the sterile tip-cell in the secidium. 



/Ecidiospore-mother-cells and secidiospores with three el- 

 even more nuclei are frequently met with : these represent the 

 result of a fusion of three or more cells of the spore-bed. But 

 since uredospores and teleutospores with cells containing more 

 than two nuclei seem to be unknown, it is probable that these 

 abnormal secidiospores undergo no further development. 



The Nature of the Spermatia. 



There are two and only two possible interpretations of the 

 spermatia — either they are male gametes, or they are conidia, 

 i.e. merely additional multiplicative spores like the uredo- 

 spores. In favour of the former view the following arguments 

 can be adduced : 



(1) The time of their appearance, just before the formation 



