growth of trees and smaller plants. The fir-forests here meet the snow-banks in actual mechanical conflict, and the front ranks of trees, though of good size, are weighed down by the snow and grow prone and interlaced upon the ground. The snow-fields rise three thousand to four thousand feet above the snow-line, and there are miniature glaciers at the heads of the valleys-representatives of the great glaciers that once filled these valleys to their mouths. The precipitation remains, the snow-fall remains, but the glaciers are gone. Here we have just the conditions most favorable to the formation of glaciers according to the theory of those who regard glaciers as thermal phenomena, but no glaciers-because of the high annual temperature. With an increase of the average annual temperature, even with increased evaporation and precipitation, it is evident that no glaciers would form ; but with a depression of temperature which should cause the rain-bearing winds from the Pacific to do all the year what they now do only in winter; viz., heap up snow on the highlands; and some of this snow-fall should accumulate year after year, the mountainslopes and draining valleys would soon be occupied by glaciers, as they were in former times. So if winter conditions could be made permanent on the great water-shed of the Canadian highlands, and the water of the St. Lawrence, the Mississippi and Red Rivers were retained in the form of snow and ice, glaciers would fill again the lakebasins, override the highest summits, and cover with an ice-sheet all the old glaciated areas.

Even if the evaporation from adjacent seas were somewhat diminished by the cold, that would not change the result, though it would prolong the time. The evaporation from the ice-cold oceans in the regions surrounding the north and south poles is now sufficient to produce continental glaciers in Greenland and on the Antarctic Continent, and it requires no argument to show that like conditions would produce like results in what is now the temperate zone.

On the theory that increased evaporation and precipitation would cause an extension of glaciers, and as an illustration of their local origin, it has been suggested that the water with which the dry regions of our Western Territories were once abundantly supplied produced the glaciers of which we find traces on the adjacent mountains, and the exhaustion of the water caused the disappearance of the glaciers. We shall see, however, that this is a speculation which is as yet sustained by no proof. It is well known to all geologists that the interior of the North American Continent has been occupied by a succession of great fresh-water lakes, extending in time from the early Eocene to and through the Quaternary. The history of these lakes has been admirably worked out by King, Gilbert, and Russell. Those of the Tertiary were numerous and broad, providing ample evaporating surfaces, but so far as we know they contributed nothing to the formation of glaciers-which could not have existed, indeed, under the warm sun

of the Tertiary ages, except on mountains higher than any the continent now bears. In the Tertiary the climate was sub-tropical over all the area of the United States south of the British line, as is shown by the fact that palms and cinnamon-trees grew as far north as Vancouver's Island and the falls of the Missouri.

The relations which the great Quaternary lakes, Bonneville, La Hontan, etc., bore to the former glaciation of the adjacent mountains is an interesting subject of inquiry. As I have mentioned, it has been suggested that it is the relation of cause and effect, but this is supported by no proof, and opposed by strong circumstantial evidence. The lakes and the glaciers may have been synchronous, and, to some extent, co-operative phenomena; but the relationship was rather fraternal than filial, as they had probably a common parentage.

The cause of the former wide spread of water-surfaces in the undrained portions of the Great Basin was either more copious precipitation or less rapid evaporation than at present. It is well known that the supply of moisture of this region is derived from the rain-bearing winds which blow steadily on to the land from the Pacific, and "the testimony of the rocks" is conclusive to the effect that there has been no change in the outline or elevation of the land, or the relations of land to sea since the Tertiary age, which could have materially increased or diminished the precipitation.

So in regard to the topography of the interior. Since the end of the Tertiary it has remained essentially the same. The hydrographical basins have been filled and emptied, but the old beach-lines which mark their sides prove that the country has remained substantially undisturbed. It is apparent, therefore, that the causes of any variation in the amount of precipitated or accumulated moisture must be climatic and not topographical. King, Gilbert, and Russell have shown that there have been several alternations of wet and dry climate in the Great Basin, and they are substantially in agreement that there have been two wet and two dry periods, of which the last is the present.

It would seem easy to determine by observation the relationship between the lakes and glaciers of that region, since some of the glaciers descended far below the highest water-level, as was the case with the Little Cottonwood glacier, to which reference has already been made, but the actual contact of the glaciated surface and the lake sediments is there covered and concealed by modern débris. The observations made elsewhere by Gilbert and Russell will, when published, probably demonstrate that which can now only be conjectured. We can confidently predict, however, that it will be found that the same climatic condition which produced the accumulation of water in the lake-basins also caused the accumulation of congealed water on the highlands. A greatly increased rainfall might produce lakes without forming glaciers, but we appeal in vain to the facts or the imagination for a probable cause of an increased oceanic evaporation, with a more abundant

precipitation on the land. Hence we seem driven to the acceptance of the other alternative-diminished evaporation-for the filling of the reservoirs of the Great Basin. And when we search for a cause of diminished evaporation only one presents itself, but that offers an easy and natural solution of the problem. A depression of temperature would certainly reduce evaporation (since the power of air to absorb moisture varies directly with the temperature), and at the same time form lakes in the valleys, and glaciers on the mountains. To prove this, we have only to cite the phenomena presented by summer and winter in the Western Territories. In winter the snow-fall on the highlands is heavy, and the accumulation of moisture in this form is large; the skies are cloudy, and the evaporation is small. In summer the sky is cloudless, the heat intense, evaporation and desiccation rapid. In the spring the snows melt, flood the valleys and form temporary lakes, which in midsummer dry up to playas. A climatic change which would perpetuate the conditions of winter and spring would inevitably produce glaciers and lakes, and these would be in the main. synchronous; and thus all we find recorded in the past history of this region would be repeated. But to intensify and prolong the summer would not produce either lakes or glaciers.

From the facts which have been enumerated above, it will be seen that from all sides we get evidence confirmatory of the theory that a certain period in the history of this continent was marked by the spread of ice and snow over a very much larger portion of the surface than they now occupy; and that we are fully justified in designating this time as an ice or glacial period; also that this was a period during which, from some extraneous cause, the climate was made colder, and the conditions which now prevail on Alpine summits perpetually, and in winter elsewhere temporarily, were more wide-spread and continuous.

That the Ice period was cold and not warm is also proved by the presence of the remains of an arctic flora and fauna in all regions near the old glaciers; the arctic shells of the Champlain, the arctic plants in the Quaternary clays, the reindeer, the musk-ox, the woolly elephant, and woolly rhinoceros, all tell the same story.

On the preceding pages the Ice period is spoken of as a single geological epoch of the Quaternary age: and so it must be reckoned in any general division of geological time. But the evidence is conclusive that the Ice period was double; that is, there were two maxima of cold separated by a long interval in which the climate was ameliorated, and over large areas which had been for ages occupied by glaciers and snow-fields, the ice and snow were withdrawn, and the surface was covered with vegetation, again to be partially taken possession of by glaciers.

Just how far north the glaciers retreated during the interglacial warm period we do not yet know, but probably not far beyond the Great Lakes; since the vegetation which covered Southern Ohio, dur

ing the interval represented by peat-beds between the first and second bowlder clays, was that of a cool climate, and the interglacial beds have not been traced beyond Scarboro Heights, on Lake Ontario.*

Facts similar to those from which we have sketched the history of the Ice period in North America, observed in Europe and Asia, afford abundant evidence that the conditions which existed here prevailed over all the northern hemisphere. In South America also similar phenomena have been observed and reported by many geologists. Hence, any explanation offered of the records of the glacial period found here must be comprehensive enough to include the whole great field; and the difficulties which here oppose the acceptance of a theory that is only local in its scope, grow until they become insurmountable. That the conditions which prevailed simultaneously in different parts of the northern hemisphere during the Ice age were synchronous with similar conditions in the southern hemisphere is not proved, nor is it probable that it is susceptible of proof. By many, perhaps most geologists these conditions are supposed to have alternated at the north and south. This much, however, we are justified in asserting, that at an epoch holding the same relative position in geological history north and south of the equator, either simultaneously or alternately, cold climatic conditions prevailed in both hemispheres and left records that are alike in character and import.

An inquiry into the nature of the cosmical influence which we must credit with the phenomena of the Ice period would lead beyond the scope of this paper, and open questions too broad and suggestive to be settled or even adequately discussed in the space at our command. I shall, however, have accomplished the end I had proposed to myself if I have shown-1. That the Ice period was a cold period. 2. That the record of the Ice period on our continent is more complete and impressive than it has been represented to be. 3. That it is the product of general and not of local causes. 4. That these causes were not topographical or even telluric, but extraneous and cosmical.

The question here passes rather into the hands of the astronomer and physicist. The work of the geologist is done when he has shown that the complete solution of the problem does not lie within his domain ; that no telluric agency is adequate to produce the phenomena; and that some cosmical cause, such as a variation in the heat radiated by the sun, as suggested by Newcomb, changes in the eccentricity of the earth's orbit, as advocated by Croll, or some other general and all-pow. erful influence, must be credited with effects as wide-spread and stupendous as those the Ice period has left behind it.

Although this paper is limited in its scope to a consideration of the glacial phenom ena of North America in the Quaternary age, and to certain erroneous notions which are entertained in regard to it, it may not be out of place to say that it is believed by many geologists that there have been several ice periods, and one at least as far back as the Permian epoch.