12 PARASITES BELONGING TO THE GENUS GLOMERELLA. 



The first production of perithecia of Glomerella in pure cultures, so 

 far as records have been found, was by Atkinson, as reported by 

 Stoneman (89) in 1898. This ascogenous form was named Gnomo- 

 niopsis cingulata Stonem. (89) . The conidial form had been described 

 by Atkinson (2) as Gloeosporium cingulatum. Since Miss Stoneman's 

 work was done considerable study has been given by different in- 

 vestigators to the life histories and relationships of Gloeosporium 

 and Collet otrichum occurring upon different hosts. The system- 

 atic study and segregation of the species in this group of organisms 

 has usually been based upon the supposed fixity of their host rela- 

 tionships and on some slight but inconstant differences in their 

 morphological characters. A form occurring on a certain host was 

 generally assumed to be distinct from one occurring upon another 

 host, especially if there happened to be some slight difference in the 

 measurements of the few spores observed or in the appearance of the 

 affected portion of the host. Later, slight differences in the last two 

 particulars led to the segregation in some cases of several species from 

 the same host plant. Species were also sometimes based primarily 

 upon the occurrence of the fungus upon different parts of the host. 

 If the fungus at hand happened to occur upon the fruit it might be 

 regarded as distinct from a form which was found upon the foliage 

 or upon a shoot or branch of the same host. It is therefore necessary, 

 before any satisfactory understanding or designation of these organ- 

 isms can be obtained, to determine not only the ordinary range of 

 variability of their morphological characters and their complete life 

 histories, but also their host relationships. Where no fairly constant 

 morphological characters can be found to separate the forms growing 

 upon different portions of the same host plant, they should without 

 question be referred to the same species. Where they occur upon 

 different hosts, but still show no reasonably constant characters for 

 identification and separation, they should still be regarded as one 

 species. If it be held that these latter should be separated as so-called 

 physiological species, the burden of proof falls upon those who 

 take that position. Such a position can only be sustained by a suffi- 

 cient number of successful cross-inoculation experiments to demon- 

 strate that the organism will not pass from one host to another. 



The present paper covers the investigation of members of this group 

 of organisms obtained from 45 different host plants. 



PREVIOUS INVESTIGATIONS. 



Previous work with these organisms may be divided into two 

 parts: That primarily of a systematic or taxonomic character and 

 that chiefly concerned with cultural, cross-inoculation, and life-history 

 studies. 



252 



