447 



but the thorax and abdomen of the fossil suborder are variable and 

 resemble both families to some degree. 



The true Zygoptera make their appearance in the Jurassic period. 

 The oldest of these, comprising the families Epallagidae and Stele- 

 opteridae, have been found in the lithographic quarries of Bavaria. 

 The majority of species from this source belong to the Epallagidae 

 and are fortunately in a good state of preservation. The wings are 

 not petiolate, the nodus and stigma are present, the nodus being situ- 

 ated near the middle of the wing and the stigma being long and nar- 

 row. There is an oblique arculus and a more or less oblique triangle ; 

 the radial sector and the second median vein arise far distad of the 

 nodus ; and the costal field contains more than ten cross-veins proxi- 

 mad of the nodus. The abdomen is not greatly lengthened and the 

 legs are also normal in this regard. In the Steleopteridae the wing- 

 is distinctly petiolate; there are about five antenodal cross-veins; and 

 the veins M 3 and the radial sector arise proximad of the nodus. The 

 arculus and quadrangle are similar to those of the Agrionidae (Cal- 

 opterygidae). The family Steleopteridae is considered to be the 

 forerunner of the Coenagrionidae. 



The Tertiary deposits furnish us with the next oldest representa- 

 tives of the group. True Zygoptera, Anisoptera, and a single Jfamily 

 of Anisozygoptera have been found in the Florissant of Colorado 

 and in the Tertiary deposits of Baden, Germany. Many of the species 

 are referable to extant genera. More than eleven genera of Zy- 

 goptera have been found in these strata. 



The first nymphs to appear in the geological record are described 

 by Hagen from the Baltic amber and from the Tertiary of Rhein- 

 land and Baden, Germany. Many of these forms had caudal tracheal 

 gills and were apparently true Zygoptera. Scudder ('90) has also 

 figured and described a nymph from the Florissant which doubtless 

 belongs to the Zygoptera. 



The following tabular summary gives the characters which have 

 been developed successively in the past, beginning with the family 

 Dictyoneuridae of the Paleodictyoptera from which the Protodonata 

 are thought to have been derived. 



