366 HYATT'S CLASSIFICATION, ETC. 



section is found, one species in Oahu and one in Lanai. The 

 most a.bundant species is L. gravida. This form connects the 

 whole series as stated above, with Amastra. This inference, 

 based upon the form, periostracum, and development, is also 

 not inconsistent with the habitat of L. gravida, which oc- 

 cupies semi-arboreal positions on bushes, and also arboreal 

 stations, but only on leaves of certain species of short trees 

 especially the olona. L. concirina is found only on Lanai 

 and is perhaps a dwarfish descendant of the L. gravida. It 

 has a similar stout spiral, the apex has a suffused pinkish 

 tone in most shells, and the colors are superficial. 



L. gravida leads into a. very distinct variety, L. suffusa, 

 and in another direction to the albinic shell L. straminea. 

 The turritelloidal character of the spire is most pronounced 

 in certain varieties of sanguinea, reported also as occurring 

 upon bushes and ferns. The nearest affine of this shell on 

 Oahu is L. gravida passing into suffusa, which sometimes has 

 a pink shell with a dark periostracum and dark apex. But 

 the pattern and aspect of L. sanguinea is not that of a denizen 

 of Oahu. Its zigzag pattern and narrow spire would be more 

 suitable in Molokai or Maui than in this island. Under these 

 circumstances it is not practicable to decide whether san- 

 guinea arose on Oahu, or is descended from some form that 

 migrated from Molokai or Maui. Information is also scanty 

 with reference to the direction of migrations on Oahu, but 

 so far as the evidence goes it would seem to have been in the 

 usual direction, from the southern valleys to the north and 

 then to the west. 



L. sanguinea is certainly the most highly specialized form 

 and thus occurs only in the more northern valleys of the 

 Konahuanui range, while gravida stops short with Nuunau, 

 leaving a gap from there to Ahonui of numerous valleys in 

 which neither species has been observed. This gap may pos- 

 sibly account for the want of obvious connection between 

 sanguinea and gravida. At any rate the distribution and re- 

 lations of the more Amastra-like gravida shows that the mi- 

 grations of the group were from the south to the north. 

 Since gravida is alone represented and only by a rare and 



