150 FREDERICK W. SARDESON 



find, most of them good zone markers. They are not especially 

 difficult to recognize in the field, their more highly specialized 

 shapes of frond and of cell pattern making them rather easier 

 than Trepostomata in that respect. Their separation into species 

 and arrangement in taxonomic divisions is facilitated by reliable 

 macroscopic external characters as a rule. They afford also a 

 great variety of neat cabinet specimens, and, in short, may be 

 recommended as worthy of close acquaintance. 



The student will find a ready knowledge of the Treposto- 

 mata a great aid to the understanding of Cryptostomata. In 

 the former group one can select a simple supposed primitive 

 skeletal structure, from which the others can be traced with 

 increasing complexity or differentiation. The simplest Cryptos- 

 tomata compare with the most complex Trepostomata, and, 

 while serial arrangement of differentiated types can be made, 

 the series appear not to begin within this group but in the Tre- 

 postomata. It is, therefore, further convenient to begin with 

 species of this group which most resemble those of the other, 

 avoiding also for the present the taxonomic definitions until 

 after representative fossil species have been studied. The new- 

 est taxonomic arrangement 1 may conveniently be referred to, 

 however, and this one divides the sub-order into eight families. 

 They may well be arranged in three divisions or series : 



Bifoliate Cylindrical Fenestrate 



i. Ptilodictyonidae 4. Arthrostylidse 6. Fenestellida? 



2. Rhinidictyonidas 5. Rhabdomesodontidas 7. Acanthocladiidas 



3. Cystodictyonidae 8. Phylloporinidae 



The relation of the three groups will be discussed later, but any 

 one might be taken first, since they are coordinate, not succes- 

 sive, but related each to Trepostomata. 



Our knowledge of the Cryptostomata may be considered 

 fairly complete, although species and perhaps a family remain 

 to be discovered, while others might be eliminated. This task 

 of completing the knowledge of the several species and genera, 

 their fixed and variable characters, is one worthy of attention 



1 Eastman, op. cit., p. 278. 



