— 229 — 



cannot be correct, especially as one of the "species" (H. floweri) has 

 been t'ound in Abyssinia as well as at Tanga, while the three others 

 are recorded t'rom tbe country between tbese localities whicb, as far as 

 we know, represent tbe most northern and the most soutliern limits for 

 the distribution of tbe whole group. 



Some other differences whicb, to judge from the descriptions, should 

 be charaeteristic to the «species» mentioned are of still less value than 

 those, already discussed, because they are rather unimportant, and at 

 the same time very irregulär in their occurrence. For instance, if two 

 forms differ in one small point or the other, they niay agree in most 

 others, and as far as I can see these small differences contirm my 

 opinion, viz. that the forms in question ought to be regarded only as 

 individuäl variatious of one and the same species. — Already in the 

 most important charaeteristic of the species, the lepidosis of the back, 

 we find some Variation whicb just was Tor^iee's main reason for 

 establishing bis species, H. squamulatus. In MoCQTJARD's discription 

 of II. tropidolepis the dorsal scales are said to be «plus ou moins fortement 

 carenees, legerement imbriquees et de dimensions tres inegales». In 

 Tornier's speeimen the scales are much more regularly disposed: 

 «Zwischen grossen gekielten Schuppen liegen nämlich wesentlich kleinere 

 kiellose Schuppen, die einander dachziegelartig decken und so ange- 

 ordnet sind, dass ein Kreis von ihnen jede einzelne Kielschuppe einschliesst.» 

 The great difference from H. tropidolepis which Turnier believes to 

 express by this description of the lepidosis I am not able to see. 

 According to my opinion there is in this nothing of specific value which 

 contradicts what Mocquard says. Tbe only rather important difference, 

 as far as I can find, is that the small scales are «kiellos» in Tornier's 

 speeimen, «moins carenees» in that of Mocquard. In other respects the 

 difference with regard to the scales appear to be about the same as 

 that between the scales of the speeimens from Tanga and Njoro. In 

 the latter the lepidosis is very regulär, almost as Tornier says, but in 

 the Tanga speeimen the small scales which even in this one are smooth 

 vary in size, and the difference between the larger of them, and 

 the large, keeled, tubercle-like scales is not so prominent as in the 

 Njoro speeimens. In the latter we Und as a rule two small, 

 smooth scales between every large, distinctly keeled tubercle-like scale, 

 but in the Tanga speeimen only one. ßesides, the very small scales 

 which form a median row along the back are much more distinet in 



