830 



SCIENCE 



[N. S. Vol. XXXI. No. 804 



3. The existence of a continental mass twice 

 greater in elevation than any other continent. 



4. Geological data showing the structure of 

 Antarctica. 



5. Evidences of a former mild climate and 

 extensive vegetation in the vicinity of the 

 pole. 



6. Meteorological data elucidating the at- 

 mospheric circulation of the southern hemi- 

 sphere. 



7. The highest tableland of the world, with 

 the location of the south geographic pole on 

 an unbroken ice-cap. 



It thus appears that Shackleton has solved 

 the difficult problem of equally satisfying by 

 his expedition the demands of science and the 

 expectations of the public. 



A. W. Greelt 

 U. S. Aemt 



SPECIAL ARTICLES 



PREDICTION OF RELATIONSHIPS AMONG SOME 

 PARASITIC FUNGI 



A FLOWERING plant which would produce 

 two separate and dissimilar sorts of fruit 

 would indeed be a curiosity, and yet there are 

 some of the common parasitic fungi which 

 exhibit two, three and even four kinds of 

 fruiting bodies or spores. In addition to the 

 variability displayed by many species of fungi 

 in the production of different sorts of spores, 

 a large number of the rust-fungi present a 

 still greater complexity of existence by hav- 

 ing the life-cycle divided into two distinct 

 alternating phases, which inhabit wholly dif- 

 ferent and unlike host plants, such as a sedge 

 and a composite, or a broad-leaved deciduous 

 tree and an evergreen. 



In these species which are known to change 

 hosts and on that account are termed 

 hetercecious, the one phase consists usually of 

 seeia, accompanied by one other spore-struc- 

 ture, the pycnium, and the other phase of 

 telia, either alone, or accompanied by ure- 

 dinia. 



The combination in one species of these 

 pleomorphic and hetercecious characters may 

 make the working out of the life-history a 

 very difficult problem. The connection or re- 



lation between two alternating phases is best 

 shown by means of cultures. A culture in 

 which a spore from one phase on one host is 

 sown upon another host, and subsequently 

 gives rise to a spore-form of the alternating 

 phase, is the only conclusive evidence that the 

 two phases are related and merely represent 

 different forms of the same parasite. Cul- 

 tures, therefore, must play an important role 

 in the study and investigation of the rust- 

 fungi, especially of those forma which are not 

 only pleomorphic but also heteroBcious. 



In order that the culture work may be car- 

 ried on in an expeditious manner, entailing 

 as small an amount of unprofitable labor as 

 possible, it is essential that the experimenter 

 should be guided by some ideas of probable 

 relationships between alternating phases. It 

 often happens that there is nothing in the 

 form or habit of either fungus or host which 

 will give the slightest hint regarding the al- 

 ternate host. In such instances a notion of 

 relations can be gained only by field obser- 

 vations. The finding of spore-structures of 

 two alternating phases in close proximity in 

 the field is usually the only obtainable factor 

 indicative of a connection between them. 

 This is the case with many of the species of 

 the genera Puccinia and Uromyces, the com- 

 mon rusts of grasses and sedges. The associa- 

 tion of telial and secial stages is, to be sure, 

 not proof of their affinity, but only a bit of 

 prima facie evidence. The closeness of the 

 association, the abundance of the infection, 

 and the occurrence of stages of other species 

 must all be taken into account. A great deal 

 has already been written' emphasizing the 

 value of these observations of association in 

 the field and it seems unnecessary to make 

 further explanation here, suffice it to say that 

 this method of gaining clues to relationships 

 is largely a deductive one. From the fact that 

 related alternating phases are often found 

 associated together, we infer that other asso- 

 ciated phases may be related. Association, in 



' See " Clues to Relationship among Heteroscious 

 Rusts," Bot. Gaz., 33 : 62-66, 1902, and " A Search 

 for Rusts in Colorado," Plant World, 11: 69-77, 

 1908. 



