28 NEW YORK STATE MUSEUM 



ff Gill lamellae double, at least on some of the anterior ab- 

 dominal segments 

 g Antennae shorter than the body; tracheae of gill lamellae 



pinnately branched Callibaetis 



</(/ Antennae longer than the body; tracheae of gill lamellae 



palmately branched Gloeon 



eee Unknown Coloburus 



Jn the preparation of the foregoing keys I have used freely 

 Eaton's Monograph of Recent Ephemeridae, that great storehouse 

 of information concerning the structure of mayhies. Although 

 in a few minor details I have not been able to accept the classifi- 

 cation therein given, I wish to acknowledge my obligation at 

 every turn to its great wealth of illustration, and to express my 

 admiration for the spirit in which its classification is set forth: 

 "It is only by taking cognizance of points of difference and agree- 

 ment in many details, in the anatomy and the mode of develop- 

 ment and the habit of leading representatives of the various 

 alliances of genera, at different periods of their lives, before and 

 after their exclusion from the egg, that the mutual affinities of 

 the several associations of genera to one another can be demon- 

 strated adequately. Until such comparisons can be and shall 

 have been carried out, the whole question of their arrangement 

 can only be dealt with in a tentative and experimental manner; 

 and it will be fortunate if error be avoided in the necessary 

 grouping of the genera into provisional alliances of apparently 

 kindred forms, preparatory to the study of their affinities. It 

 is far more easy to demonstrate defects in proposed methods of 

 classification than to devise a trustworthy system in their 

 stead." 



I have correlated nymphal and adult structures, and have ex- 

 pressed that correlation in the foregoing keys, wherein all the 

 major divisions are strictly parallel for the two stages. That 

 this is now possible is a sign of progress toward a natural sys- 

 tem of classification. The one serious incongruity in Eaton's 

 system — the interpolation of J o 1 i a in the subfamily E p h e- 

 merinae; an incohgTuity that grew out of a previous error, 

 inherited from Jolv — the breeding of C h i r o t e n e t e s has 

 enabled nie to remove. The nymph " J o 1 i a r o e s e 1 i " is 

 doubtless that of the sole European species of C h i r o t e n- 

 etes, Ch. ignotus Walker. A comparison of the figures 



