98 PARASITES BELONGING TO THE GENUS GLOMERELLA. 



(6) In most of the forms studied neither morphological nor physio- 

 logical differences sufficient for the segregation of species have been 

 found. All the material from the 36 hosts is referred to three species 

 of Glomerella, G. cingulata, which occurs on 34 of the hosts, G. gossypii 

 on one, and G. lindemuthianum on one. 



(7) Glomerella cingulata is exceedingly variable in all its characters 

 so far as they have been studied. The cause of this variability is not 

 yet clear. No constant or definite relation has been established 

 between the cultural and other environmental conditions and the 

 most important variations observed. 



(8) The fungus is found to be present in many cases in apparently 

 normal and healthy foliage, fruits, and sometimes in the stems of 

 its hosts, as shown by its development and fructification on such 

 portions of plants after they have been thoroughly washed in a corro- 

 sive-sublimate solution which has been shown to kill not only aseo- 

 spores and conidia but also the chlamydospores or appressoria of the 

 fungus. The chlamydospores or appressoria evidently send a germ 

 tube through the epidermis of the host as shown by Hasselbring, 

 and this remains in a quiescent condition until opportunity for 

 further development occurs. 



(9) Inoculation experiments with fruits have shown that most of 

 the forms from different hosts will produce the characteristic Glom- 

 erella rot on fruits of other hosts. It is also shown that there is great 

 variability in the virility of different races or strains of the fungus 

 from the same host. In one experiment races from the lemon, 

 grape, and fig produced more serious cases of bitter-rot of apple than 

 a race of the fungus derived from apples. These facts are of great 

 importance in connection with the selection and production of 

 disease-resistant varieties of plants. 



(10) The production or nonproduction of the perithecial stage of 

 Glomerella appears to be a fairly well-fixed hereditary race character. 

 Where a race of the fungus has been obtained by repeated trial of 

 spores from different sources and races which develops both conidial 

 and ascogenous fructifications in cultures it continues to produce 

 them for many generations. An ascogenous race from Persea 

 was grown for 23 generations from conidia. The last generation 

 produced perithecia about as abundantly as the first. No evidence 

 has been obtained to indicate that the production of perithecia is 

 controlled by any of the ordinary conditions of nutriment or environ- 

 ment. 



(11) Glomerella is a parasite which has apparently developed 

 special features, the most important of which is its method of infec- 

 tion by means of chlamydospores, or appressoria, and its ability to 

 remain in a dormant or quiescent condition until the host plant 

 becomes weakened or injured in some way or until specially favorable 



252 



