38 BULLETIN UNITED STATES GEOLOGICAL SURVEY. 



On San Miguel Island are two small perennial springs ; one is situated 

 several hundred yards below the adobe building at the bay, and. the- 

 other, with a little better quality of water, on the elevated northern 

 point. 



The vegetation consists of low bushes, cactus, and grass, but no 

 trees. At our visit, the island, which is a Government possession, was 

 " dried up", being overstocked with starving sheep. 



With much difficulty we moved our camp through the shifting sand 

 up the western shore of Cuyler Harbor, where the best boat-landing is^ 

 offered, and found there on shell-mounds room enough to erect our 

 tents. 



The kitchen-middiugs, or kjokkenmoddings, of a former people are 

 found all over the island where sandy ground is met with. This singu- 

 lar mixture of all kinds of shells, bones, rocks, and flint-chips, spread 

 usually over a space of about a hundred yards square, and to a depth 

 of five feet, although the extent and depth of such shell deposits vary 

 greatly, is found on both sides of the harbor, especially over its north- 

 ern point, also along the northwest side to the west end, and covers in 

 great masses the low, sandy, western extremity of the island. The 

 deposits of the kjokkenmoddings are much exposed to the strong 

 northwest wind, and, as they are located on loose sand, are laid bare by 

 its action, offering therefore good facilities for surface collections. But 

 of these the casual visitors to the island, mostly persons interested in 

 stock-raising, excursionists, and amateur curiosity-hunters, have picked 

 uj) or destroyed the best, and much of what was left had been col- 

 lected several months previously to our visit by Mr. Dall, of the United 

 States Coast Survey, during a short visit to this island. Of the small 

 surface collection made here, I consider an unfinished mortar the most 

 interesting article, showing in its partially rough and incomplete state" 

 the mode of manufacturing such a utensil by the aborigines. But my 

 attention was especially given to the finding and exhuming of the old 

 cemeteries, which, as my experience taught me, promises the richest 

 reward. About half-way, and almost in a line between the two springsr 

 near Cuyler Harbor, I found a grave-yard, and soon another close by, 

 which yielded about 250 skeletons, and many utensils, implements, and 

 ornaments of stone, bone, and shell. iTot far from the upper spring, 

 another burial-place was discovered, which hardly returned any results* 



The mode of burying was similar to that previously observed on the 

 mainland, on the coast of California, which I described in the Smith- 

 sonian Eeport of 1874. The bodies were buried in the kjokkenmoddings, 

 because the kitchen-refuse offered here the only ground which is firm 

 enough to resist caving, and also to prevent the winds from uncovering 

 the dead, as would occur with loose sand. The skeletons were found 

 from three to six feet under ground, and often from three to four resting 

 one above the other, separated, if at all, by the bones of the whale. The 

 bodies were deposited without any order as to position and direction of 



