THE FLEAS OF NORTH AMERICA 67 



head with 2 lateral median bristles. Pronotal coinb with 24 or more long, black 

 spines. Femur I with several lateral setae ; coxa III without a row or patch of 

 spinelets on inner surface; last segment of posterior tarsus with 5 pairs of 

 lateral plantar bristles, the first of which is not shifted ventrally. Abdomen 

 with a small number of apical spines which never extend down to level of 

 spiracles. Tergal plate VII without posterior median process. Antepygidial 

 bristles. 1 long and 2 minute in male. 1 long and 1 or 2 much shorter but 

 not minute in female. Female with 1 receptaculum seminis, the head of which 

 is cylindrical, concave dorsally, much longer than broad, and much longer 

 than tail. In male, sternal plate VIII long, slender, with stout apical bristles 

 and apical membranous flap; sternal plate IX with a pair of prominent, bent, 

 internal vertical processes, an anterior median apophysis, and a pair of distal 

 lobes each of which is divided by a deep ventral incision ; movable finger of 

 clasper broad, flattened, and with 1 or more enlarged bristles or pigmented 

 spines on or near posterior margin. 



This genus is probably fully justified as here used because of the 

 host relationships of its species, all of which are probably true bird 

 parasites. Morphologically the generic group has unity and as a 

 unified group can easily be described. The chief trouble in using the 

 genus as restricted to bird-parasitizing species is that its characters 

 almost completely dovetail with those of other genera. It is particu- 

 larly to be noted that the male terminal] a of species of Ceratophyllus 

 are of the same type as those found in some of the species of Trichop- 

 sylla. This is true even when the shape and chaetotaxy of the mov- 

 able finger are considered. In such a species as C. riparius Jordan 

 and Rothschild the movable finger is broad, flattened, almost vertical, 

 and possesses a row of 5 stout bristles along the posterior margin. 

 In Trichopsylla the same condition exists in such species as T. 

 (Pleochaetls) sibynus (Jordan). The number of spines in the pro- 

 notal comb in Ceratophyllus is 24 or more, yet in Trichopsylla (T.) 

 atrox (Jordan) there are more than 24 spines in the pronotal comb. 

 One species of Ceratophyllus, C. adustus Jordan, may not have a 

 bird as a true host. 



The genus Ceratophyllus, even in its restricted sense, is a large one. 

 It is confined entirely to the Holarctic Region. 



Ceratophyllus adustus Jordan 

 Ceratophyllus adustus Jordan, 1932, Novitates Zool. 38: 253, figs. 10-11. 



Type host. — Erethizon epixanthum Brandt (American porcupine). 



Type locality. — Atlin, British Columbia. 



Range. — Known only from British Columbia. 



This species is known only from a single female which may have 

 been taken from an unnatural host. It is a typical bird flea except 

 for the reduction in the number of spinelike setae on the tibiae and 

 the loss of the upper one of the three antepygidial bristles. Jordan 

 gives two detail drawings of adustus accompanying the original de- 

 scription and adds: 



The Porcupine probably is the true host of this modified bird-flea, the 

 peculiarities of the new species presumably being due to a change of host from 

 Bird to Mammal. 



Ceratophyllus borealis Rothschild 



Ceratophyllus lorealis Rothschild, 1907, Ent. Monthly Mag. 18: 11, 1 fig. 



Type host. — "* * * probably came from the nest of a gannet 

 {Sula bassana)" 



