99 



pair of legs (tav. XVI, fig. 21). Lei us now consider, if the above quoted definilion 

 liolds good, when applied to these three species. 



The céphalothorax is in G. n. sp. from Hawaii only a little longer than broad 

 and in G. n. sp. from Funafuti a trifle broader than long, and the front margin 

 of the latter is not truncated, but almost bilobate ou account of a median incision 

 (lig. 14, p. 90). The same species as well as G. n. sp. from Hawaii has no transverse 

 cephalothoracic stripe; and the stripe or stripes of G. nobilis n. sp. are very indi- 

 stinct, to say the least (PI. II, fig. 8 a). The longitudinal line of the tergiles is at 

 least wanting in G. n. sp. from Funafuti. The arolium is simple in the latter 

 species, but two-branched and extending beyond tips of claws in the two others. 

 If we exclude G. n. sp. from Funafuti we will get the four other species characterized 

 by two very good structural feature and two of less importance, the few omitted, 

 which are in common with most species of Olpiuni; if we do not exclude the 

 species from Funafuti only a single character remains, namely that found in the 

 femur of the two first pair of legs. I have in vain tried to find other characters, 

 which could possibly assist in the defining of this genus; both flagellum, galea and 

 serrula exterior have nothing in common in the three examined species, if not 

 such features, which are found in most species of Olpiiim L. K. The palps are 

 rather short and clumsy in all the species except G. diinkliatiis L. K., and their 

 femora do not seem to bear a dorsal "tactile" hair near to the base, which at 

 least is often present in species of Olpiiim L. K. The tarsi of the two first pair of 

 legs are, in the three species examined, of almost equal length to tibia, but in 

 G. dimidiahis L. K. much shorter; the second tarsal joint is in these four specimens 

 at least a little longer than the basal joint , but in G. Patagonicus Ell. a trifle 

 shorter. 



The question arises, if Garypiiuis Dad. is a natural genus; if this question is 

 answered in the affirmative, we must ask, whether a species like G. n. sp. from 

 Funafuti, which has the trochantin as long as the tibial femoral part and the 

 arolium unbranched, ought to be included or excluded. These questions ought 

 according to our present knowledge to be answered in the affirmative, and the 

 Garijpinus Dad. is to be defined by the structure of the femurs of the two first 

 l)air of legs, being provided with a trochantin, from on third to one half as long 

 as the whole femur and connected with the tibial part by no real hinge-joint, 

 allowing movements in the vertical plane. Species with the structures of the 

 femoral intermediate between that of Olpiiim L. K. and this genus may possibly be 

 discovered and consequently make the systematic position of these species uncertain, 

 but as the genus has at least a practical value and has once been established, it 

 ought to be accepted for the present. By excluding G. n. sp. from Funafuti nothing 

 was gained, for the main character of the femurs would lose a good deal of its 

 value, while the character, found in the two-branched arolium would be rather 

 doubtful, owing to the occurence of a two-branched arolium in Olpiuin liirmaiiiciim 

 n. sp. f. inst. 



13- 



