654 THE PHYLOGENY OF THE FUNGI 



The chief merit of this suggested system is that the Agaricales appear 

 to be an extension, as it were, of the Polyporales with a gradual increase 

 in complexity of ontogeny from strict gymnocarpy to pseudoangiocarpy 

 and finally to complete angiocarpy. From the latter it would seem logical 

 to extend this development to the Gasteromyceteae. Here, where the 

 spores are set free from the basidia before the spore fruit opens we find 

 that they are symmetrically attached to the sterigmata or even practically 

 sessile on the surface of the basidium. 



There are two obstacles to this system. In the first place the type of 

 the trama and character of the basidia and cystidia are in the main quite 

 different in the Agaricales and Polyporales. So that even where, exter- 

 nally, similarities and apparent transitions seem to occur careful study of 

 the anatomy and chemical reaction of the hymenium and trama show that 

 these are perhaps better explained as morphological convergences but not 

 true relationships. Furthermore it would seem more reasonable to expect 

 that the Agaricaceae which possess a well-developed universal veil, as 

 revealed by the production of volva, annulus, etc., and which have an 

 angiocarpic mode of development, would be far more likely to have arisen 

 from the mostly strictly angiocarpic Gasteromyceteae and that the reduc- 

 tion and final disappearance of a universal veil and the appearance of 

 pseudoangiocarpy and finally of gymnocarpy would indicate a gradual 

 degeneration from the complex to simplified forms, now that the firm 

 universal veil (peridium) had become no longer necessary. A third and 

 very strong objection is that the Gasteromyceteae that show the closest 

 similarity to the Agaricaceae lead by gradually simplified structures to 

 the forms assumed by almost all mycologists to be the simplest ones, the 

 forms that well may have developed from organisms related to Aleuro- 

 discus, Corticium, Tomentella, etc. 



Singer^s Hypothesis No. 2 



Agaricales Derived Wholly from Gasteromyceteae. This is the hypoth- 

 esis, favored by Singer and at least in part by Bucholtz (1903), that the 

 Polyporales do not give rise to the Agaricales; the latter arise from the 

 Gasteromyceteae and at some points, mostly by convergent evolution, 

 assume forms externally similar to some of those of the Polyporales. Its 

 chief virtue is that it explains the presence of the universal veil and of 

 angiocarpy in some of the Agaricales and allows a logical explanation of 

 their disappearance as the rather definite affinities to the Gasteromyceteae 

 become more distant. The undeniable close similarities of some Agaricales 

 to some Polyporales is difficult to explain away. To the author the most 

 outstanding objection to this hypothesis is the necessity of assuming the 

 evolution anew of the sterigmatal function of spore discharge at several 

 points, for if this hypothesis is adopted the Agaricaceae probably arose 



