armatures in the HymenoiJteroits genus CoUetes. 47 



in ligat'iis is somewhat transitional, in tlie others it is 

 impossible not to recognize it instantly.) 



The armature of ligatus is not vinlike that of the pre- 

 vious species (balteatus, etc.), but ditlers in the much 

 greater (broader and longer) triangular chitinous dilatation 

 of the sagittae at their point of deflexion. This occupies 

 a much larger part of their dorsal aspect, and looks nearer 

 their bases than the smaller (apical-looking) triangles in 

 the other species. 



The colour of the seventh ventral plate is very dark, 

 making the broad hyaline space at the apex of each lobe 

 particularly conspicuous. All my specimens have the 

 same bold oblique (downwards and inwards) curl or roll 

 of the lobes. It seems hardly possible to flatten them 

 out completely without splitting them by the pressure. 

 So they must be somewhat rigidly chitinized in parts. 



7. Hylmformis, Evr. PI. VI, 7, 7«, 7/^ PI. IX, 40. 



8. Casijicus, Morawitz. PI. VI, 8, 8a, 8/), 8f. PI. IX, 47. 



I shall treat of these two species together, as they seem 

 near allies, and their synonymy is at present somewhat 

 entangled. 



The questions whether and how they difler are best 

 determined by consvilting the writings of Morawitz who 

 was the original describer of caspicus, and who also wrote 

 a careful re-description of liyheiformis from examination 

 of Eversmann's type-specimens, the original description 

 being, he tells us, altogether inadequate. 



The conclusions to which a study of these descriptions 

 brings me are supported by the evidence of specimens 

 received under the names respectively of hyLviforniis 

 (from Prof. Perez and Herr Kohl) and caspicus (from Prof. 

 Perez only). Hylxiformis ^ is at once separable from 

 caspicus ^ by its thoroughly squamose style of pilosity, 

 and hylmformis $ from caspicus $, both by the external 

 characters given already in my Tables and (at a glance) by 

 the armature, in which each apex of the unnotched stipes 

 is narrow and elongate in Jiylieiformis, while in caspicus 

 it is a broad, blunt, almost equilateral triangle. (See PI. IX, 

 46 and 47.) 



The species, however, which Radoszkowski calls hylxi- 



formis (and also his floralis, probably not = the floralis of 



Eversmann !) is figured with an evidently blunt and short 



