1908.] NATURAL SCIENCES OF PHILADELPHIA. 461 



rests on the unproven assumption that the adults do not wander 

 far from the waters in which they have passed their earUer stages or 

 in which their offspring are capable of surviving. This assumption 

 is one of the weaknesses in the following attempted generalizations; 

 another is the real scantiness of our knowledge of the distribution of 

 even the winged individuals. How scanty this is may be seen by a 

 glance over the list of localities in Honduras, Nicaragua, etc., in Table 

 A and in the columns for these countries in Table B of the Introduction 

 to the Biologia volume quoted and a study of Tables 6-8 of the 

 present paper. 



It must be distinctly understood, therefore, that all which follows 

 is subject to future correction in these two important particulars. 

 In spite of these disadvantages, however, some generalization has 

 been deliberately attempted, in the belief that by so doing progress in 

 investigation wdll be hastened much more than if no such summary 

 were ventured. 



The Chief Odonatological Features of Mexico and Central 

 America. 



These are : the practical absence of the subfamily Cordulinse,^ some 

 species of which have been recorded from corresponding latitudes in 

 the Old World. 



Absence of the following genera, conspicuous or well developed 

 in other parts of America: (a) in Northern America,^ Ophiogomphus, 

 Gomphus, Dromogomphus, Odogomphus, Celithemis, Leucorhinia; (h) 

 in South America, Lais, Thore, Euthore, Microstigma, Telagrion, 

 Leptagrion, Diastatops, Potamothemis ; (c) in the West Indies, 

 Scapanea. 



The small number of genera, seven out of seventy-one, which are 

 restricted to this area. They are Pseudostigma, Thaumatoneura, 

 Paraphlehia, Hesperagrion, Anisagrion, Oplonceschna and Pseudoleon. 

 Three of these (Hesperagrion, Oplonceschna, Pseudoleon) embrace only 

 one species each. Oplonceschna and Pseudoleon should be good fliers 

 and, therefore, one would not expect their hmited distribution. 



The unity of the district, in that only one genus {Hesperagrion) is 

 restricted to Mexico north of the Isthmus of Tehuantepec, none to the 



* Already pointed out by Carpenter, Scient. Proc. Roy. Dublin Soc. (n. s.), 

 VIII, p. 450 (1897). 



^ Throughout this memoir, as in the Biologia volume on Odonata, by "North- 

 ern America" is meant all north of central Cahfornia, Arizona, New Mexico, 

 Texas and (east of this last) of the 30th parallel of north latitude. 

 30 



