58 HENSHAW : SEMI-FOSSIL SHELLS IN HAMAKUA DISTRICT, HAWAII. 



» 



That this forest was of the usual island type is certain, and it consisted 

 for the most part of ohia and koa together with numerous smaller trees like 

 the pua, kopiko, ahakea, tree lobelias and many other shrubs and berry 

 bearing trees, with the usual tangle of ieie vines and ferns. 



The forest, proper, probably never extended in this region much, if any 

 above 3,000 or 3,500 feet. Above this altitude the slopes are steeper and 

 the soil more scanty and rocky. Here the maniani begins to be numerous, 

 a tree which indicates a thin and i)o()r soil, a scanty rain supply and a con- 

 siderable altitude. 



The region of the W^aimea plains appears never to have had a large (as 

 compared with some other parts) rainfall. In the absence of definite data 

 it may be assumed to be not far from 40 inches at Mana on their upper 

 border, that figure being the average for several years in the town of Waimea 

 as given by Professor C. J. Lyons. A small rainfall would seem to be 

 indicated also from the fact that nowhere on the plains appear marked evi- 

 dences of erosion. The deep gulches which gash the windward side of the 

 island at short intervals are on the plains conspicuous by their abence, 

 although nowhere are they deeper and more numerous than a few^ miles to 

 the north east in the rainy Kohala mountain district. The Kohala moun- 

 tains in fact, seem to intercept and rob the trades of their moisture before 

 they reach the plains. The surface of the plains is by no means flat, but is 

 gently and in places quite steeply rolling. On their upper edge and probably 

 on the very edge of the former forest, at an altitude of about 3,000 feet, occur 

 the semi-fossil deposits which form the subject of this paper. The fossils 

 have been found in two distinct localities, viz., at Mana and at Palihouk- 

 apapa. The two places, however, are only about four miles apart, and 

 though the latter is several hundred feet higher than the former, to all 

 intents they may be considered conchologically to be one and the same. 

 There is, however, some difference in the character of the deposits at the 

 two places. At Mana the shells occur in the horizontal strata, two or three 

 inches thick, and under a deposit of about a foot of humus. The very prim- 

 itive digging implements at the writer's disposal prevented anything like a 

 thorough examination of the extent of the deposits, but the evidence all 

 goes to show in that in no one spot are they extensive. All that were found 

 were included within an area of a few hundred square yards, the shell-bearing 

 strata in some spots occupying only a few square feet, in others a few 

 squar yards. 



At Palihoukapapa the deposits are, or seem to be, even less extensive, 

 and instead of being in horizontal strata are in the nature of pockets, some- 

 times containing a bushel or more of shells. Over how large an area here 

 the deposits occur there are no present means of telling. 



I see no reason to doubt that the shells in both localities are entirely 

 local in origin, and that they were swept into their present position by water 

 resulting from local freshets. In certain favoured localities in the islands, shells 



