( 174 ) 



si)ecies. At onr present state of knowledge of tlie existing forms of Sii)hona]itora 

 any detailed classification of the fleas mnst be prcmatnre. Wc know only such a 

 small {lercentage of the species actually in existence that to try to divide the Order 

 into a series of families and subfamilies must sooner or later prove itself to be but 

 an abortive attempt. 



Dr. Baker, however, in differentiating his JUiOjjalopsyllas from Euderlein's 

 Parapsi/llus, does so only by the number of the segments of the labial palpus. Wo 

 are, in fact, presented with a puzzle, botli nomenclatorially and morphologically. 

 Rhopalopsi/llm is said to be characterised by a "four-jointed" labial palpus, and 

 Parripxi/H/is by a five-jointed one. Now, the type of lUiajjalopsi/Uatt, namely lutzi) 

 is described by l?aker as having ap])areutly a six-jointed labial palpus ! Therefore 

 the position is this : (1) The American Pulex are certainly generically different from 

 in-itans : (2) The character by which Rhopalojisi/llns is said to be distinguished 

 from the nomenclatorially older genus ['arapsi/llus does not exist in the species 

 specified as ty])e. 



The confusion has further been intensified by treating, as Baker tloes, all 

 American Pulex without genal and thoracical combs as having a " fonr-joiuted " 

 labial palpns, regardless of the descriptions. Now, of the nine species enumerated 

 liy Raker under Kliopalop»i/Ui(s only one single one has a four-jointed labial palpus. 

 In our descriptions of coajti, australis, etc., it has been expressly stated that the 

 labial palpus consists of five segments. We did not mention the rostrum of 

 cleophofitis in our description. We now add that the labial palpus of this species 

 has six segments, the uuderlip itself being very short. In one of the specimens the 

 fourth segment is further divided on the hinder side only, indicating tliat there 

 may exist specimens of cleophontis with seven segments in the labial palpus. 



As we do not know Baker's lutxi, we are not in a position to give a rectified 

 diagnosis of Baker's Rhopalopsijllus, We also abstain from proposing a new 

 generic term for any of the eight American Pulex we have before us. The erection of 

 new genera in this order of insects should be limited as far as possible ; at any rate, 

 the characterisation of new genera should not be attempted without comparison of 

 the various allied species from difterent faunistic regions. In this regard we are 

 in perfect agreement with what Dr. Baker says on p. 123 of his recent paper.* 



Weyenbergh described one species of the group of Pulicidae under discussion. 

 AVe have twofcmales of this Pulex cacicoln, which, on examination, prove to belong to 

 the species which wc have described in 1904 as Pulex concitus from [.wo females 

 obtained at Sucre, in Bolivia, the synonymy being as follows : 



3. Pulex cavicola. 



Pulex cuvicola Weycnborgli, Period. Zmil. iii. p. 274 (1881). 



Pulej; condtus Kothschild, Nov. Zool. xi. p. 615. n. 10. t. 10. fig. 38. 40 (1904). 



Weyenbergh's specimens were obtained off Cadia leucopijya Bl., while ours 

 were found on Ilerodon boliviensis. We take the ojjportuuity of supplementing and 

 correcting the descriptions. The rostrum reaches well beyond tlic trochanter of the 

 foreleg, the labial palpus consisting of five segmeuts. The third segment of 

 the maxillary pal [Mis is about one-third the length of the fourth, or a little over 

 one-third, the fourth being a little longer than the second. The frons of the head 

 bears a groove near the oral angle, appearing as an incrassation of the skeleton in 



• I'm; U.a. Nat. A/us. xxix. p. VS.i (1906). 



