12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IO3 



collina (Ohia IchuaJ, a tree towering 15 meters or more above the ground. The 

 cloud zone is characterized by low, dwarfed shrubs in more sheltered parts and 

 by mat-forming Panicum, Paspahnn, Isachne, and mosses in exposed windswept 

 areas. 



The maritime, Haole Koa, and Guava zones are mostly made up of 

 introduced plants, according to Hosaka; the Ohia and Cloud zones 

 contain the largest number of endemic species. The Ohia zone has 

 the largest number of indigenous plants. 



Achatinella apcxfulva is almost entirely confined to the Ohia zone 

 and the upper portion of the Koa zone. The width of the Ohia zone 

 obviously varies in different parts of the range. In the Nuuanu 

 Valley region the Ohia zone is comparatively narrow; to the west, 

 near the Poamoho Stream, it is undoubtedly wider than at Kipapa 

 Gulch. 



MATERIAL AND METHODS 



Over 11,302 shells were studied, the majority of which are from 

 287 localities. Out of this number 6,707 adult shells were measured 

 from lots containing 5 or more shells and having a length range of 

 at least 3 length classes. The total number of adults from plotted 

 localities is 3,687, of which all but 283 were collected by W. Meinecke. 

 Mr. Meinecke is the first person to make a large collection of 

 Achatinella from the Koolau Range with the localities plotted on a 

 good map such as the United States Geological Survey topographic 

 advance sheets of the map of Oahu. All his localities prior to 1933 

 were plotted from memory; after that date, up to 1937, he again 

 collected many of his former localities and many new ones, so that 

 many errors made in plotting from memory are checked and cor- 

 rected. Mr. Meinecke is such an unusually careful worker, with 

 such an extraordinary ability to work with maps, that I have the 

 greatest confidence in the exact plotting of his localities. This opinion 

 is further strengthened by a check of some of his localities, all of 

 which were found to be correctly plotted. A few of his localities 

 collected before July 1932 are not dependable — in fact, this comment 

 applies to all localities collected by anyone prior to 1932, with the 

 exception of the localities of Dr. C. M. Cooke, Jr., in Nuuanu 

 Valley which were mapped by sketch maps made in the field. 



Other collectors who have supplied shells with locality data plotted 

 on the United States Geological Survey topographic advance sheets 

 and collected since 1932 are G. W. Russ, H. Lemke, and H. Lemke, Jr. 



From localities that are not plotted on any map come 3,002 adult 

 shells; they are mostly localized from the Meinecke and other care- 



