1913.] NATURAL SCIENCES OF PHILADELPHIA, 231 



differ in the following characters, the difference shown by Dijsagrion 

 being stated: 



I. Nodus at .37 of the wing-length. 



6. The veins generally, posterior to Mi, have a slighter caudal 



curvature as they approach the hind margin. 



7. Between Mi and Mia are two rows of cells beginning under the 



stigma (D. fredericii, not D. packardii). 



8. Between Mia'and M2 is one row of cells increasing to two, three, 



and finally five rows. 



9. M2 separates from Mi at .15^ of the distance from nodus to stigma. 



12. Between Rs and Ms is one row of cells increasing to two, three, 



and eventually eight rows. 



13. Arculus apparently at the second antenodal. 



15a. Between Ms and"M4 is apparently only one cross-vein proximal 

 to the level of the subnodus. 



17. Between Mt and Cui are (a) 8-10 antenodal cells (the quadri- 



lateral is much shorter than in Phenacolestes) and (6) at the 

 wing-margin onl}^ one row of cells. 



18. Cubito-anal cross-veins two in D. fredericii, one of them proximal, 



the other distal, to the arculus; none showTi in D. packardii. 



19. Between Cui and Cu2 at wing-margin one {D. packardii) or three 



(D. fredericii) rows of cells. 



Comparison of the fossil jNIelanagrion (umbratum) with 

 Phenacolestes. 



According to the figures and description of this Miocene genus 

 from Florissant, Colo., given by Scudder,!" Melanagrion and Phena- 

 colestes agree in the characters above numbered 5, 7, S,^^ 9, 10, ^^ 

 13, 14, 15a, 16a and b, 18, and differ in the following characters, 

 the difference shown by Melanagrion being stated : 



1. Nodus at .3 (?) of the Aving-length. 



2. Only two antenodals. 



3. More than 26 postnodals (27, Scudder). 



4. Stigma having both proximal and distal ends much less oblique. 

 6. The veins generally, posterior to Mi, have a slighter caudal 



curvature as they approach the hind margin. 



II. Rs separates from'Mi+2 slightly proximal to the subnodus. 



» This is taken by measurement from Scudder's figure, although he says that 

 the '"nodal sector arises from the principal .... scarcely more than one-fifth 

 way to the pterostigma" (p. 129), and in another place (p. 128) "at scarcely 

 one-fifth the distance from the nodus to the pterostigma." Whether figure oi- 

 text be correct, both differ from the condition in Phenacolestes. 



10 Scudder, Tertiary Insects of North Aynerica, 1890, pi. 13, figs. 12, 14, p. 136, 

 described this form as a Lithagrion. It was made the type of a new genus, 

 Melanagrion, by Prof. Cockerell in Bull. Amer. Mus. Nat. Hist., XXXIII, p. 138, 

 1907. . ,. , 



" At the extreme margin there are four rows m Melanagrion. 



16 



