NORTH AMERICAN NEUROPTERA. 251 



to 9, or on some of them, or entirely wanting, and then some of the 

 segments 5 to 8, or less divided in two halves by an impressed longitu- 

 dinal middle line. 



Of the 21 subgenera accepted by De Selys, only 4 are known by 

 raised nymphge, and 3 more by supposition; of the 113 species in De 

 Selys last list, to which now are to be added about 40 more, only 12 

 are known by raised nymphae, and 17 only by non-raised nymphae, for 

 which, in a few cases, supposition is tried. It is to be presumed, of 

 course, that the four-fifths of the species not yet known may modify the 

 given characters and necessitate, perhaps, for some of the subgenera a 

 higher standing. 



In 1872 Mr. L. Cabot described 8 species, of which only 2 were 

 raised, all of which were contained in the collection of the Museum. 



Of the now described 20 species 21 are from the West Indies and 

 Mexico, 4 from Brazil. 4 from I]ast India and 1 from Europe, of which 

 12 were raised. There are known now 61 species from the United States 

 and Mexico, so that the known nymphae represent one-third of the known 

 imagos. 



The very important question whether the nymphae corroborate the 

 adopted subgenera or groups cannot yet be considered seriously, because 

 only 4 of the 21 subgenera are represented by raised nymphse. After 

 all some remarks seem not to be out of place. 



Two species may be excluded directly from the others by some very 

 heterogeneous characters. 



1. The species referred to as possibly belonging to Cyclogomphus has 

 the lobe on the middle third of the front margin of the mask deeply 

 cleft. This character is quite unique for Gomphus and combined with 

 the lack of the apical hooks of the anterior tibiae seems to indicate a 

 higher position than a subgenus. It is the most aberrant nympha among 

 those known of the genus Gromphus. 



2. The species from Brazil supposed to belong to Epigomphus. The 

 unusual length of the 1 0th segment of the abdomen and the also unusual 

 occurrence of two teeth on the lobe of the front border of the mask seem 

 to indicate a higher position than a subgenus. The length of the 10th 

 segment corresponds well with Epigomphus, and was indeed the reason 

 for my supposition. 



The known nymphae, the two just spoken of excepted, are divided in 

 two large groups : 



1. The palpus is without terminal hook, not dentated ; the middle third 

 of the front margin of the mask produced in a rounded lobe ; ten specimens. 



