NO. i:?Gi. REVISION OF AMERICAN SIPHON I PTERA—BA KER. 371 



The Siphonaptcra seem to have suff^-ed especially from the disre- 

 gard of all laws of priority. RJiynchoprlon and Jlectop.si/lhi were dis- 

 reo-arded and new names given the same groups. Species names were 

 discarded for all sorts of reasons, often simply because they were 

 more or less inappropriate. In this paper the attempt has })een made 

 to apply th(^ rides of priorit}' strictly, though it has been impossible to 

 investigate fully such special cases as Rhynchoprirm, Monopayllns^ etc. 



No writer on the fleas in the past has made any attempt to designate 

 generic types, and this fact has given rise to the greatest difficulties. 

 I have tried to determine this matter for all the genera, and the results 

 are given below for the Pulicidai. With the other, later genera, there 

 is no difliculty. 



PuUx Linntvus 1695; type, Irrltans Linnaeus. 



Ctenocephalus Kolenati 1859; type, canis Curtis {canis = novem- 

 dentatas Kolenati), 



Cemt(>2)JiyUu)< Curtis 1^32; type, gaJUnai Schrauk. = Ctemmotus 

 Kolenati 1863; type, fasciahis Bosc {/((sciattis — octodechndentnttis 

 Kolenati). = 7/'/e//(9/j>.'?///'/r,r Kolenati 1863; t3'pe, jr?6;?c?7//r/^^/' Grube. 



Ctenopldhaliniis Kolenati 1863; type, hisoctodentatus Kolenati. 



CtenopsyUvs Kolenati 1863; type, muscuU Duges {musculi = qvad- 

 ridentdtus Kolenati) . 



Ce)\(fop,sylli(.'< Kolenati 1863; typQ^ j)e?itartem(''^ Kolcnuti = Tyj)hlo- 

 psylla Taschenberg 1880; t3^pe, octactemu Kolenati. 



Pen<'!ir!ge)\fasclatus^ and gaUindB are clearly congeneric. Tasch- 

 enberg does not especiall}^ indicate a type for TypMopsyllu^ but no 

 matter which is taken for the type the genus becomes s^^nonymous 

 with some other. I have indicated the first species under his genus as 

 the type, thus throwing- it into Cerato psyllium. The two others of the 

 three groups in Taschenberg's Typhlopsylla fall into Ctimop.syll ns 

 and ('f('?u>j>/it/u///nus, respectively. It is to be noted that Wagner and 

 Rothschild still use TyphJopxyUa for the same group to wliicli the 

 earlier name Cfenophthalmus was applied. CteiiophtJudnms and 

 some of the other genera may be artificial groups, but Tyjddopmylla 

 is still more so, founded as it was principally on the absence of eyes. 

 This character has proven of scarceh^ more than specific value, every 

 possible gradation occurring from Pulex to Ceimtojm/ll us. 



On the other hand, it is certain that Wagner's reseparation and 

 recharacterization of these genera will have to be much modified, due 

 to reasons which are dilated upon in the discussions of genera. 



The synopsis used herein has been adopted simply as a matttM-of 

 expedienc}' in the study of the American species, as it was iniixissibh^ 

 to separate them either according to the scheme of T!is»-luMib(>rg or of 

 ^^ agiuM-. So the two have been combined. 



One hundred and thirty-five species are listed in this papt r for the 

 world. I have not the least doubt but that many hundreds will e\ iMit- 

 ually be found, and that these will fall into at l(>ast twenty-fire or thirty 



