From the above facts we are very safe in drawing the conclusion that during the great quaternary glacial period the conditions which now cause glaciation must have existed in an aggravated degree, and those which now give us temperate climates in regions once glaciated must have disappeared or been reversed. On the other hand, the warm climates which prevailed during the tertiary and other geological epochs, and permitted a temperate flora to flourish as far north as Grinnell Land and Spitzbergen, could only have occurred under conditions exactly the reverse of those which produced the cold. If high land in high latitudes is the principal cause of the present glaciation of Greenland, still higher land must have been so in causing the still greater glaciation of the former period. Scandinavia, Laurentia, the British Islands, the Alps, Apennines, Rocky Mountains, Sierra Nevada, and all great mountain ranges in the northern hemisphere must have stood at greater elevations. There must have been such an accumulation of ice and snow as to chill the air, cause fogs, and prevent the summer heat of the sun from melting off the water which fell in the solid form during winter; and on the other hand, there must have been hot summers and great expanses of ocean to the south to supply the abundant evaporation which became condensed by contact with the chilly mountains and uplands of the north. One supposition is that the Isthmus of Panama was then submerged, so that the Gulf Stream ran into the Pacific. But this wants geological confirmation, as the Isthmus shows no sign of such recent marine formations as must have been deposited if it had been submerged to a sufficient depth to let the Gulf Stream escape, and the extension of the ice-cap in North America to much lower latitudes than in Europe, points rather to the conclusion that the Gulf Stream must have run very much in its present course. The only geological evidence bearing on this question is the recent discovery of deep oceanic deposits such as the Globigerena ooze, above tertiary deposits in Barbadoes and Jamaica, leading to the inference that the whole West Indian area was a deep sea in comparatively modern times. This no doubt might affect both the temperature and the velocity of the Gulf Stream to a considerable extent.