and cry. The angular hills and long slopes of talus are not softened by any arborescent veil. The infrequent villages nestle behind sheltering bluffs, and are rarely visible from without the harbors. In winter all the heights are wrapped in snow, and storms of terrific violence drive commerce from the sea about them. Once pass within the harbors during summer and the repellent features of the landscape seem to vanish. The mountain sides are clothed with soft yet vivid green, and brilliant with many flowers. The perfume of the spring blossoms is often heavy on the air. The lowlands are shoulder high with herbage, and the total absence of trees gives to the landscape an individuality all its own. No more fascinating prospect do I know than a view of the harbor of Unalashka from a hill top on a sunny day, with the curiously irregular, verdant islands Bet in a sea of celestial blue, the shore lines marked by creamy surf, the ravines by brooks and waterfalls, the occasional depressions by small lakes shining in the sun. The sea abounds with fish; the offshore rocks are the resort of sea lions, and formerly of sea otters; the streams afford the trout-fisher abundant sport, and about their mouths the red salmon leap and play. In October the hillsides offer store of berries, and in all this land thei-e is not a poisonous reptile or dangerous wild animal of any sort. The inhabitants of this land are an interesting and peculiar race."
The Underground Houses of Techin, Tunisia.—The curious underground houses of Techin, and other places near Gabes, in Tunisia, are described by M. Albert Tissandier, in La Nature, as being easily cut out of the clay-limestone rock. A square pit, twenty or twenty-five feet deep, is generally dug first to form a central court, from the lower part of which are made the grottoes that serve as sleeping and store rooms. A gently sloping gallery rising from the level of the bottom to the level outside, and closed by a modern door, is the means of communication with the country without, or forms a path to other houses. Niches are cut in the walls of the entrance court for storage of the agricultural implements and things of minor value, and silos are provided for the grain crops and oil jars. The rooms are lighted only through the doorway from the entrance court. They are furnished with mats and carpets. A bed and a wooden pedestal for the lamp, rudely carved and whitewashed like the walls of the cavern, form the principal ornaments, and a few primitive utensils of enameled earthenware are the principal articles of furniture. The size of the house and the number of rooms, etc., vary according to the wealth and station of the proprietor.
A Measure for Odors.—Very interesting studies have been made by M. Eugène Mesnard of the perfumes of flowers, valuable to science and to the perfumer's art. This art is still in a rather crude state because it has never found a practicable way of measuring an odor or of determining the strength of the several odors which it may seek to combine. M. Mesnard has observed, however, that though the absolute intensity of an odor can not be measured, its comparative strength can be estimated. A perfumer who has five or six hundred kinds of fragrance in his shop can readily distinguish the differences between them, although he can not tell how strong any of them are. So it is possible to detect by the smell the existence of a large number of chemical substances, although it is impossible to guess how much, if any, of them may be present in the air. Suppose, now, the author says, we pass in a given receiver air charged with a known perfume and air which has passed over a special essence—spirits of turpentine, for example. It is possible to obtain in this way a mixture of neutral odor, or such that a very slight variation of its constituents on either side will cause the special odor of the perfume or the smell of the turpentine to prevail. We can in such case regard the two odors as equivalent, and have only to seek a means of determining the intensity of the turpentine odor to have a measure of that of the perfume. A measure for the turpentine is obtained by means of the property which it has of extinguishing the phosphorescence of phosphorus. For that purpose a bit of common starch is employed which has been dipped in a sulphide-of-carbon solution of phosphorus. The sulphide evaporates and the starch remains, a homogeneous substance impregnated with phosphorus which shines in the air. The
