764 Charles Paul Alexander 



flies. That this resemblance is presumably not fundamental is suggested 

 by the immature stages of these same generalized forms, which show 

 but few features that can be reconciled with those of a panorpid-like 

 ancestor. Unfortunately the fossil record helps but little. At the begin- 

 ning of the Tertiaries, crane-flies were already numerous both in species 

 and in individuals, but in most cases they are clearly referable to existing 

 genera. It is evident, then, that the investigator must go still further 

 back for his evidence, and the fossil crane-flies of the Upper Mesozoic 

 are, unfortunately, still largely unavailable. The phylogenetic develop- 

 ment of the group may be considered from two standpoints, the geological 

 record and comparative morphology. 



THE GEOLOGICAL RECORD 



The most important works on fossil crane-flies are those of Loew 

 (1850), Scudder (1894), Handlirsch (1906-08, and 1910, a and b), Meunier 

 (1906), and Cockerell (1910, 1913, 1916, and 1917, a and b). The writer 

 is indebted to Professor Cockerell for assistance in determining the age 

 of many of the fossil-bearing beds. Excellent accounts of the various 

 deposits may be found in Cockerell (1913) and in Tillyard (1917). 



The Tipuloidea, representing the superfamily of cratie-flies, is herein 

 considered as being constituted of six families, of which two, the 

 Eoptychopteridae and the Architipulidae, did not survive the Mesozoic 

 period. The Eoptychopteridae are known only from the Mecklenburg 

 Lias (lowest Jurassic), and include but three monotypic genera. The 

 Architipulidae are known from the same formation and include eight 

 species arranged in three genera. The other four families contain recent 

 forms and are included in this paper. 



The North American Eocene and Miocene, as represented by the White 

 River and Green River beds and the Florissant shales, respectively, give 

 evidence of having had a northern fauna, especially in the Eocene. This 

 is well shown by the great development of the Cylindrotominae, which 

 in the White River basin almost dominated the crane-fly fauna during 

 the Eocene. It seems probable, moreover, that the group was forced 

 into colder regions of the globe during the Oligocene, when the tropical 

 element reached far to the north. No group of crane-flies that can be 

 considered as being tropical has yet been found in the Florissant. On the 

 other hand, the European Oligocene, as shown by the Gurnet Bay beds 



