39^ BOTANICAL GAZETTE 



[JUNE 



did not form in pustules. On apple it took but slowly, though 

 finally a few small pink pustules of spores were formed. On steri- 

 lized bean pods the growth was abundant, in a few days becoming 

 black from the crust of perithccia formed over the surface. But 

 here, as on the bean agar, the perithecia did not seem to develop asci 

 and spores. 



Such a large variation may seem incredible, and it may be attrib- 

 uted to a contamination, for which there is always a chance. If 

 this was a contamination and not a variation, either the two forms 

 were present on the original apple and were transferred together 

 unnoticed from thp dilution plate to the tube, or the second form must 

 have entered the tube after the transfer was made. As the apples 



autumn 



until February, it does not seem possible that the two form.s could 

 have been on the apple together. If they had been, they would have 

 been noticed in the cultures, as they were constantlv worked with 

 durmg the fall and winter. The second possibility seems no more 

 likely. The variation came in the middle of the winter, when spoies 

 of Gloeosporium would not be liable to be free in the laboratory; 

 and as the new form was totally unlike anv of the other species 

 of Gloeosporium in culture in the laboratory, it could not have 

 come from them. The only explanation of the phenomenon is that 

 one or more indinduals of the original form changed quite suddenly 

 their course of development under cultural conditions; but whether 

 that IS admitted or not, we have at present this new form. It is 

 undoubtedly a Gloeosporium of the Glomerella type, with the develop- 

 ment of the perithecia considerably different from other known 

 lorms. 



Mutations, so far as is known bv the writer, have not previously 

 been recorded among fungi, but the form just described seems to 

 be one without question. Whether aU of the variations that have 

 appeared m the study of this group should be classed as such is ques- 



minor 



tionabJe; 



should not be so considered. 



^uuuia not be so considered. No two individuals of any species are 

 aiike, and the variations are perhaps no greater than we could expect 

 among individuals of almost any group. The effect of variation on 



group 



