ORIGIN OF AGARICALES 655 



from the Gasteromyceteae at several different places. In this viewpoint 

 the simplest organisms of this group are supposed to have originated from 

 simple Thelephoraceae, which already had their spores perched asym- 

 metrically on the tips of the sterigmata from which they are discharged. 

 Several cases are known where near relatives of the Tremellales have 

 adopted the angiocarpous structure and with it have lost this peculiarity 

 of sterigmatal discharge. It is difficult however to understand how so 

 complex a structure could have been lost at the beginning of the Gastero- 

 myceteae and regained several times independently as some Agaricaceae 

 developed from them. 



Singer's Hypothesis No. 3 



Agaricales Derived Both from Polyporales and from Gasteromyceteae. 



This is really a sort of compromise. Its chief objection is that we have to 

 find a line of separation within the Agaricaceae between those descended 

 from the Polyporales and from the Gasteromyceteae. This is perhaps 

 even more difficult than the first or second hypothesis. 



The relationships within the Gasteromyceteae as well as their origin 

 are very uncertain. To be sure Eduard Fischer (1933 and earlier) proposed 

 a system that is fairly logical, progressing from simpler to more complex 

 structures, and this has been the basis for much of the work in this group 

 for the last fifty years. The characters that have been considered as the 

 more important are the modes of development of the gleba, the progres- 

 sive development of the columella (and its downward extension, the stipe) 

 and the tendency to progress from a coralloid to a multipilar and eventu- 

 ally unipilar structure. Other important characters are the development 

 of the capillitium, the formation of definite hymenial cavities or their 

 obliteration by ingrowing hyphae (plectobasidial structure), the auto- 

 digestion of the gleba to an evil smelling m9,ss attractive to insects, 

 angiocarpous or pseudoangiocarpous development, etc. 



One very important point, in the author's opinion, that has been sub- 

 ordinated to a secondary position is the character of the spores. It appears 

 to him that there are several types of spores that cut across the family 

 and ordinal boundaries as customarily recognized. These, perhaps, are of 

 much greater importance and it may be that they indicate closer relation- 

 ships than have been recognized. Besides spore structure a much more 

 extensive study of the very early stages of the formation of the spore 

 fruits is absolutely essential before a well-grounded system can be 

 established. 



In the genera G aster ella, Hymenog aster, and Gasterellopsis we have, 

 respectively, very small, unicameral spore fruits, a moderate sized organ- 

 ism with very numerous hymenial chambers with a more or less coralloid 

 arrangement, and a small, at first unicameral, structure with a percurrent 



