284 TJniversitij of California Publications in Botany [Vol.6 



the genera of the Tuberales which they resemble in general character, 

 liabit, and habitat, more closely than genera of any other order. 



This arrangement, like all preceding arrangements, has been based 

 principally upon a study of mature ascocarps. In only one case, that 

 of Tuber candidum, were young plants available for investigation, but 

 these served rather to strengthen than weaken the probability of the 

 relationships indicated in plate 26. 



In the following explanation of the table, it is hoped that the link- 

 ing of the Tuberales with the Pezizales, and the further theory of re- 

 lationships indicated, will be justified so far as justification of any ar- 

 rangement is possible with our present imperfect knowledge of devel- 

 opment in these plants. 



Except for the absence of localized mycelium in Hydnocystis, there 

 is little difference between this genus and Peziza, either in superficial 

 characters or fundamental structure. The latter exists first as a com- 

 pletely closed body which opens during development as a result of 

 more rapid internal than external growth, the hymenium for the first 

 time becoming exposed. Peziza, then, is of angiocarpous origin. Both 

 Hydnocystis and Genea also appear sometimes as closed bodies, and 

 while the earliest stages are not known, it is not unreasonable to as- 

 sume that these genera, too, are originally angiocarpous, the open 

 forms representing cases of greater inequality of growth in cortical 

 and subcortical tissues. An interesting point here, which may or may 

 not be significant, is the presence in many genera of the Pezizales of 

 branched paraphyses similar to those found in several species of 

 Genea; and even the secondary cortex of the latter is, to an extent, 

 duplicated in Gonida, Caldesia, Heteropatella, and other members of 

 the above-named order. 



In plate 26, two main lines of development labeled respectively 

 A and B are indicated, the ascocarps of the former differing from 

 those of the latter in the possession of external openings from the 

 hymenium. This condition is first observed in Hydnocystis in which, 

 apparently, the subcortical growth has exceeded the cortical in certain 

 forms. Others remain completely closed and give rise to the genera 

 of line B. In Geopora the comparatively simple ascocarp form of 

 Hydnocystis has become complicated by greater disparity of growth 

 in the two tissues, causing infoldings of the entire wall, and inward- 

 extending projections from the inner surface. In a few cases here 

 also is found such great disparity of tissue development that the cortex 

 has yielded to the pressure from within, and openings are formed. As 



