Hcbard — Dcniiaptcra and Orthoptcra of Hawaii 371 



second gToup has the Hmbs relatively elongate and immaculate. 

 To it belongs nigroliiieafiis Perkins, but the validity of the other 

 species, referable to this type, is as uncertain as in the first 

 group. The third is represented by a single comparatively stout 

 species, with limbs relatively short and showing very weak traces 

 of annuli. 



As some smaller immaculate individuals have shorter limbs, 

 these groups are by no means sharply distinguished one from 

 the other, and recession of coloration in the annulate type may 

 cause such annuli almost to disappear, as might be expected. Pro- 

 ceeding further we find that individuals of the first two types 

 have tegmina varying from minute, lateral, and scarcely project- 

 ing lobes, to small, but overlapping, lobes, which wholly cover 

 the metanotum. Though each series shows a large proportion of 

 the specimens runing constant to one or the other of these ex- 

 tremes, yet certain individuals are intermediate. 



The metanotum of adult males, in which this area is ex- 

 posed, shows a slight, twin convexity, each side with a median 

 impression. In males with the metanotum nearly or wholly 

 covered by the tegmina, however, we find much higher specializa- 

 tion, as shown on Plate xxvii. 11. This might be considered 

 most important in determining the number of species repre- 

 sented, were it not for the fact that we know tegminal size to be 

 often (though not always) attributable to individual variation 

 within a species, whereas the disappearance of glandular special- 

 ization may result solely from tegminal reduction, leaving the 

 otherwise specialized area unprotected. 



In length of ovipositor many of the Gryllidae show very great 

 individual variation and in the present genus the extremes to 

 be found in each species are probably decided. 



In the immature stages minute, lateral tegminal lobes are pres- 

 ent, even at a time when no more than half the adult size has been 

 attained. This fact adds the further dit^culty that some of the 

 males, possessing them, though apparently adult, may not be mature 

 and might have had the larger overlapping lobes when adult. 



We, therefore, record the material before us as representing 



[69] 



