NO. 2192. DRAOONFLIES, CALIFORNIA AND NEVADA— KENNEDY 



523 



length is unreliable, for some have swollen and some have shrunken 

 in alcohol. 



With these differences in size go much more marked differences in 

 structure. In the 1917 brood or smallest size, the wing pads are 

 minute triangular projections, which are erect 

 and do not overlap. In the 1916 brood they are 

 2 mm. long, and are no longer erect but overlap 

 tlu-ee-fourths of their length. In the 1915 brood 

 they are 5 mm. long and assume the adult shape 

 and position. In the mature nymph or 1914 

 brood they are 9 mm. long. The sex appears in 

 all sizes. In the smallest or 1917 brood the fe- 

 males are distinguished by a pair of minute tu- 

 bercles at the posterior edge of the ventrum of 

 segment 8. In the next size, 1916, these are dis- 

 tinct triangular projections, though very minute, 

 while in the next size, 1915, they assume the adult shape but are only 

 haff the length of segment 9, while in the adult nymph they are 

 shghtly longer than segment 9. The male organ on segment 2 is less 

 apparent, being distinct in the adult stage only. 



The nymplis of both dorsalis and diadema have been described by 

 Necdham^ (fig. 147 is copied from Needham's description and shows 

 the divided median tooth of the middle lobe of the labium in the two 

 species). 



147 



Fig. 147.— Divided median 



TOOTH OF middle LOBE OF 

 NYMPHAL L.^BIUM (FROM 

 NEEDILiM). A. CORDtJ- 

 LEGASTEE DORSALIS. B. 

 CORDULEGASTER DUDE- 

 MA. 



J Proc. U. S. Nat. Mus., vol. 27, 1904, pp. 



