specialized representative of the group. The elongated, mem- 

 branous type of peridium and the occurrence of both telia and aecia 

 on galls fits in with this disposition. 



The intimation that some of the functions of the uredinial 

 stage have been taken over by the aecial and telial stages may be 

 briefly explained. The chief function of the uredinia is to insure 

 distribution by a rapid and extensive dissemination of the fungus. 

 The long tubular peridium of the roestelia-type contains many 

 more spores than the short cupulate-type and in this way larger 

 dissemination is made possible. In many of the species of Gym- 

 no sporangium the telial stage has become perennial, and in this 

 way continuity is insured, even though dissemination might be 

 less rapid. The writer does not wish to convey the understanding 

 that the assumption of the ordinary functions of the uredinial 

 stage by the other stages is to be regarded as the influence which 

 has suppressed the uredinia. The partial fulfillment of these 

 functions by other stages may, on the other hand, be the result and 

 not the cause of the disappearance of the uredinia. We are ignorant 

 of the cause of suppression or reduction, but the disappearance 

 of the uredinial stage and the development of new devices to 

 perform its ordinary functions seem in some way to be correlated. 



3. Experimental Investigation of Life-Histories 



History 

 The species of Gymnosporangium were among the first heteroe- 

 cious rusts in which a positive connection between the two alter- 

 nating phases was experimentally proved by controlled inocula- 

 tions. In the spring of 1865 Oersted (1865) showed by means of 

 cultures carried out in the Botanical Garden in Copenhagen 

 that the two fungi, Gymnosporangium Sabinae, inhabiting red 

 cedar, and Roestelia cancellata, inhabiting the pear, were really 

 but two phases in the life-history of a single organism. Oersted 

 states that he had suspected this relationship in 1862 but was 

 unable to actually carry out the experiment until 1865. It is 

 interesting to note that he also suspected a similar connection 

 between Puccinia graminis and Aecidium Berberidis, and in his 

 communication describing his successful cultures with the cedar 

 and pear rusts, June 10, 1865, he suggested that the same relation 

 doubtless existed between the cereal and barberry rusts. On 



