40 NINTH REPORT. 



tions of organism, and the physiological relations of one kind of organism to 

 another, forms a series of the most fascinating as well as the most difficult 

 of biological problems. The small size of the bacteria and-the rapidity with 

 which they multiply make them very favorable subjects for experiment 

 along the line of the fundamental biological processes. An organism that 

 requires several hundred years to complete its life cycle is obviously not a 

 favorable subject for an experiment that requires the study of several gen- 

 erations. But if, as in the case of some of the bacteria, a new generation 

 may be produced every fifteen minutes, it is possible to obtain within a few 

 hours hundreds of generations and millions of individuals. 



There is another group of organisms about which I wish to speak, not so 

 simple as the bacteria in structure, but far inferior in that respect to the high- 

 est plants. I refer to the filamentous fungi, and I wish to call your attention 

 to some facts that again have to do with the question: What is a species? 

 As in the case of higher plants, the first attem])ts to classify these organ- 

 isms were upon a purely morphological or structural basis. But a deeper 

 knowledge of their life histories and physiological variations make it more 

 and more apparent that here, as among the bacteria, it is necessary to use 

 physiological means of distinguishing, shall we say species? For the present 

 we can avoid making the decision, and say forms or races, yet at the same 

 time we can hold our minds open to evidence as to whether these forms or 

 races are not after all incipient species. Two groups of these fungi especially 

 force themselves upon our attention from the point of view Ave are consider- 

 ing. One of the groups has been called the Uredineae or rust fungi, and the 

 ordinary rust of cultivated cereals is a typical example. The other group 

 is known commonly as the mildews, or more technically the family Erisy- 

 phaceae. The rose mildew and the grape mildew are common examples. 

 In both of these groups it has been found necessary to distinguish what have 

 been called biologic forms or races, distinguished from each other only by 

 the fact that they differ in capacity to infect different species or genera of 

 the host plant. Working with the wheat rust, which was formerly supposed 

 to be the same on many of the cultivated cereals and wild grasses, Erikson 

 has foimd that there are numerous races adapted more or less closely to the 

 species of single genera, and they are able to infect species of other genera 

 either with difficulty or not at all. Their forms cannot be distinguished 

 morphologically, and yet the infection experiment shows that physiologically 

 they are decidedly different from each other. In trying to conceive the 

 origin of these forms, there seems to be three possibilities. First, These bio- 

 logic forms may have had an origin from different species growing on a nar- 

 rowly limited group of host plants. There seems to be little evidence for 

 this view. Second, They may have been derived from one species, by sud- 

 den physiological changes in the fungus alone, without any influence of the 

 host. This would be similar to the origin of elementary species by mutation, 

 as found by de Vries among the evening primroses. There seems to be no 

 direct evidence for this view. Third, a group of biological forms which can- 

 not be distinguished morphologically may have originated from one species 

 which at first grew on a wide range of host plants, but when a strain or race 

 is propagated continuously on the same species of host, there is a special 

 adaptation of the fungus to that species of host, and it becomes able to in- 

 fect that one more readily, and others less readily, and at last not at all. 

 For this view there is some direct evidence. A form of rust which was capa- 

 ble of growing on four genera of host plants, was propagated for ten years 

 continuously on only one of the four. At the end of the ten year period it 



