In this species & in a new Gerris {G. perseus, presentir to be described), 

 I have encountered an araount of individuai variation in the proportions 

 of antennae, tarsi etc, for which I was entirely unprepared. 



Champion (1) separates his two species as follows : 



« Pronotum with a large triangular ochraceous patch on the anterior 

 « lobe ; antennae with joints » [segments] « 2 and 3 subequal in length; 

 « first genital segment unidentate on the right side at the apex beneath 

 « in the d. Unidentaius, n. sp. 



e Pronotum with a narrow ochraceous median line; antennae with 

 « joint » [segment] « 2 longer than 3; first genital segment bidentate on 

 « the right side at the apex in the d. Bidentatus, n. sp. » 



The colour - characters, at least, bere given are far from Constant, 

 and the species are evidently remarkably variable, so much so that until 

 an opportunity of examining a large series of both sexes occurs, I must 

 decline to do more than identify 3 of the 7 as P. bidentatus. The fol- 

 ìowing table may, however, be of interest: 



antennal proportions 



40: 14: 14: 20 



antennae broken 



37 : 14 : ? ? 



26: 11 : 8 : 16 



28 : 12 : 8 : 16 



30: 12: 8 : 19 



32: 11 : 11 : 19 

 7 have a large triangular ochraceous patch on the 

 pronotum ; Nos 4-6 have a somewhat narrow centrai stripe of the same 

 colour, on the pronotum. Nos 1-2 have 3 narrow, longitudinal, more or 

 iess entire ochraceous stripes on the mesonotum, 4 & 5 have a short, 

 narrow stripe about the centre of the mesonotum, while 6 & 7 are im- 

 maculate. 



Therefore, according to (a), 1 & 7 ought to be unidentaius, 4-6 bi- 

 dentatus; according to (8), 1-3 & 7 ought to be unidentaius, 4-6 biden- 

 tatus. 



Nos 1 & 2, however, are unmistakably bidentatus, judging by the 

 structure of the genital segments; & I believe 3 to be the female of 

 the same species. Nos 4-7 I have labelled « Potamobates sp.?, probably 

 not bidentatus ». 



In the 5 o, the 6 segment of th^ connexivurn is produced on both 

 sides considerably beyond the apex of the abdomen ; this prolongation 

 is rounded & blunt, and densely hairy, Ifc is noteworthy that an appen- 



w 



no 1 cf fi 



'oni Gualaqniza. 





» 2cT 



do 





» 3 $ 



do 





» 4 9 



Rio Peripa 





» 5 9 



do 





» 6 o 



do 





» 7 j 



do 



w 



Nos 1, 2 



3 & 7 have a 



(l) Biologia Centr. Amer., Zool., Hemipt. Heteropt. II, pp. 154-5. PI. ix, 

 ss 20-2. 



