INTRODUCTION 17 



identity. Here again the actual number of sesquipedalian names is not large, 

 but such examples as Chaetohasidiella vennicularioidcs, Pseiidopcrisporium 

 erigeronicola, and V erticilliodochiiim tubercidarioides permit no question of 

 the desirability of such abbreviation. Such improvements can properly be 

 made without change of citation, but this has seemed to be impossible in 

 the case of personal hybrids, such as Raciborskiomyces. Both brevity and 

 uniformity have also been promoted by rendering consistent in the shorter 

 form the divergent practices with respect to imparasyllabic neuters in -at, 

 as in Lophiostomaceae, Phomalcs, etc., the doublet, rr, before h, the double ii 

 in personal genitives, etc. 



The time and energy demanded by the present treatment have been so 

 far in excess of what was anticipated that the original plan of revising the 

 terminology of mycology to render it more consistent and definite has been 

 relinquished for the present. However, a few steps in this direction have 

 been taken by employing Arthur's terms for the rusts, definitizing the use of 

 paraphysis, paraphysoid and pseudoparaphysis, and bringing the terms 

 among lichens into closer harmony with those used in the other fungi. 



The Dichotomous Key 



Saccardo long ago emphasized the point that many a fungus must be 

 diligently sought by the tyro in more than one place (Sylloge Fungorum 1 :VI 

 1882). This is still true today, but an endeavor has been made to mini- 

 mize this difficulty by inserting certain genera in two or even three places 

 in the key. The latter has been made as definite and consistent as possible, 

 and its use facilitated by employing the dichotomous method throughout. 

 It is a much simpler and easier task to construct keys after the pattern of 

 Hoehnel (1923), in which there are regularly several and frequently ten or 

 more choices under one heading, few of them worded in the same terms 

 and almost none of them strictly comparable. In such a key, the user is 

 obliged to do most of the work that the author should have done for him, 

 but under much more difficult conditions. 



The sequence of criteria in the various families and sections has been as 

 strict as possible. In the Ascomycetes and Phomales in particular, the great 

 majority of genera differ from their immediate neighbors in but one essen- 

 tial, and in consequence both sequence and definition are as a rule exact. 

 On the other hand, as in Phycomycetes and Gasteromycetes, the contrast is 

 usually much less definite and the key necessarily partakes of the same 

 character. However, in both cases the sequence in the key serves as a 

 diagnosis for each genus, and one in which the salient criteria stand out 

 much more clearly than in the usual description with its attention to trivial 

 features. Furthermore, the dichotomous key provides a very useful test of 

 proposed genera, since it renders it impossible to take refuge in vague state- 

 ments as to validity and the differences from related ones. It can be safely 

 said that the number of new genera would be greatly reduced if every 

 author would subject his proposals to the test of such a key. 



