There are three things in Iceland that have never been counted:—The islands in _Breiðifjörðr_, Broad-Fiord, the lakes of _Arnavatnsheiði_, Eagle-Lake-Heath, and the conical hills of _Vatnsdalr_, Water-Dale. Our stopping place, _Hnausar_, which signifies rough ground, is in the midst of these peculiar hills and in the center of the valley. We spent three days among the hills and found them of marked interest to the geologist. Hundreds of acres are covered with the cones rising from the plain to an elevation of from twenty-five to over one hundred feet. Oftentimes they are so near together that their bases are confluent and thus seem to be double peaked in a few instances. Geologists have given different reasons for this queer formation. One states that they are of glacial origin and were left when the ice melted in the form of moraines; another is of the opinion that they are the results of great avalanches upon the glacier, which in melting left them here. Another states that they are merely the weathered fragments of a local lava flow. I spent a day in their examination and so will give my reasons for rejecting the causes assigned by these gentlemen and substitute my own conclusions in order that future scientists interested in the geology of Iceland may confirm or refute according as they weigh the evidence. They can not be glacial moraine as there is no evidence of any glacial action in any way upon any of the fragments and it must be remembered that as compared with glaciated areas in other lands Icelandic glaciation is as if it occurred yesterday. In fact glaciers are still covering many square miles of the table land. There is no evidence of any water erosion on any of the stones. They could not have been avalanches upon the ice sheet for there are no mountains near at hand from which such masses of material could have come. And if it is argued that the avalanches were at a distance it turns the problem once more into that of the moraine. The character of the valley and its low mountains will not permit our reason to accept either the glacial or the avalanche theory.