ORIGIN OF ASCOMYCETEAE 641 



the gametangia or from one of the latter the long ascus develops, tapering 

 toward its apex. In it the zygote nucleus divides repeatedly and finally 

 around each nucleus is formed an ascospore. Numerous nuclei degenerate 

 without acting as centers for spore formation. It is assumed that these 

 were the supernumerary nuclei of the gametangia which did not become 

 the privileged nuclei. The interpretation by Dangeard (1907), Atkinson, 

 Gaumann (1926), and others is that the gametangia are homologous to 

 those of the Mucorales and that the ascus represents the zygospore of the 

 latter and the sporangium which it usually produces upon germination 

 after a period of rest. In Dipodascus, however, there is no resting period 

 and the zygote develops immediately to become the ascus. In D. uni- 

 nucleatus Biggs, the mycelium is cellular, consisting of uninucleate cells. 

 The gametangia are also uninucleate and there are no supernumerary 

 nuclei. The ascus and ascospores develop as in D. albidus, the number of 

 ascospores being typically large but often small in poorly developed asci. 

 This species forms a close connection to Endomyces and Eremascus in the 

 Endomycetaceae in which single asci, of eight ascospores or less, arise as 

 the result of the sexual union of the two uninucleate gametangia. By in- 

 crease of the tendency toward the budding habit of vegetative growth we 

 can derive the Saccharomycetaceae. It is assumed that in some forms 

 similar to Dipodascus the elongated multinucleate ascus gave rise to a 

 series of branching hyphae containing the as yet unfused gamete nuclei 

 and their products, and producing terminal cells where the pairs of nuclei 

 unite and undergo meiotic division. Thus is interpreted the origin of the 

 dicaryon ascogenous hyphae and asci. By the growth of loose protective 

 vegetative hyphae between and around the ascogenous hyphae it is 

 assumed that such a loose perithecium is developed as we find in the 

 Gymnoascaceae. By the development of a firmer cortex the Aspergillaceae 

 were derived. From these organisms one can imagine the development of 

 Erysiphales, Myriangiales, Pseudosphaeriales, Sphaeriales, Pezizales, etc., 

 a complete reversal of the phylogenetic tree based upon the idea of the 

 origin of the Ascomyceteae from Florideae. (Fig. 207.) 



Whichever of the two phylogenetic series mentioned above is con- 

 sidered to be nearer the truth there are certain observed facts or interpre- 

 tations of facts that disagree. In both hypotheses the ascus is postulated 

 as being derived from a sporangium: for the Floridean ancestry it must 

 come from a tetrasporangium and for the Phycomycetous derivation from 

 a sporangium more or less on the plan of that of the Alucorales. In either 

 case the spores arise by the complete segmentation of the protoplasm, 

 with no enucleate epiplasm left over. Yet, in the ascus, part of the cyto- 

 plasm gathers around the nuclei and becomes encysted, leaving a portion 

 without nuclei, the epiplasm, surrounding the ascospores. Whether or not 

 this difference in manner of spore formation is so fundamental as many 



