STOEMS 405 highway between Europe and America. The charts of the ocean in the neighborhood of Cape Horn, published by the London meteo- rological office (1871), do not give the per- centages of storms ; but for the cape of Good Hope, the charts of Cornelissen (1874), of the meteorological institute of the Nether- lands, show that during the southern summer the storms in this region are comparatively few and feeble. In winter severe westerly gales are exceedingly abundant to the south of the cape. In the immediate neighborhood of the continent, and on either side, the influ- ence of the land is shown in the frequency of storm winds blowing off shore during winter, and on shore during summer. Storm Warn- ings. Suggestions for storm warnings were broached even in the 18th century, before the introduction of the electric telegraph had ren- dered the undertaking practicable. The Euro- peans were the first to engage systematically in storm warning ; but in the early demonstration of the feasibility of the idea, the merchants and the individual meteorologists of the United States took the lead. If we pass by that which Franklin, Espy, Bache, Redfield, Loomis, Mau- ry, and others did to advance our knowledge of atmospheric phenomena, and consider merely the steps taken to establish national systems of storm warnings, we shall note that in 1835 the. joint meteorological committee of the Franklin institute and the American phil- osophical society at Philadelphia appointed, at the suggestion of Bache and Espy, a sub-com- mittee to memorialize congress upon the sub- ject of a national weather bureau for the study and prediction of storms. When the success of the Morse telegraph was beyond all question, Redfield in 1846, and Loomis in 1847, urged its systematic application to the problem in hand. This idea was greatly furthered by Prof. Joseph Henry, who as secretary of the Smith- sonian institution had just removed to Wash- ington. In 1847, in behalf of that institution, he organized a system of volunteer meteoro- logical observations and reports. Through the liberality of the National telegraph line, Prof. Henry was in 1857 able to begin the publica- tion of a telegraphic weather bulletin, and to make successful weather predictions. Mean- while, the appointment of Espy as meteorolo- gist successively to the war and navy depart- ments, and the publication (1850 and 1857) of his famous reports on meteorology, had awa- kened a universal conviction that storm pre- dictions were practicable. Already many mer- chants were habitually obtaining at their own expense weather reports from distant sections. The Smithsonian weather bulletin was of ne- cessity discontinued in 1861, and an effort to revive it in 1864 was frustrated only by a con- flagration which destroyed a portion of the Smithsonian building early in the following year. It does not appear that anything more was done in America in the prosecution f the subject of weather predictions until in 1868, in his inaugural report as director of the Cincinnati observatory, the writer proposed this as a work proper for one branch of the in- stitution which he proposed to build up in that city. A few months after this date the Cin- cinnati chamber of commerce authorized him to obtain at its expense, for three months, the necessary telegrams, and to publish daily weath- er predictions. This system went into opera- tion on Sept. 1, 1869. In a modified form, and pending further negotiations, the work was continued at the joint expense of the wri- ter and the Western Union telegraph company from December, 1869, to May, 1870, and after- ward entirely at the expense of that company, whose manifold weather maps were in much demand until the commencement in Novem- ber, 1870, of the great work of the army signal office. (See SIGNAL SERVICE.) Simultaneous- ly with the spread of the telegraph in Europe began the publication of weather bulletins, and their collation and study. Kamtz says that even in 1835 he had begun to study the weather reports in the Berlin papers, but it required the excitement of the Crimean war, in 1854, to force the importance of the subject upon the attention of European governments. In that year Leverrier as director of the Paris observatory took up the subject ; his tele- graphic reports began in 1855, and his inter- national bulletin in 1858. Weather probabili- ties were not begun till 1863 ; these were dis- continued in 1865, and only lately have been revived, but storm warning signals have been uninterruptedly displayed since 1860. In 1861 Admiral Fitzroy, of the meteorological depart- ment of the board of trade, began the display of storm signals in England, and in the latter part of the year the publication of weather fore- casts ; his system of warnings consisted rather of a series of signals announcing the presence of storms, than of any real prediction of their advent. The Fitzroy system ceased in 1866, shortly after his death, but was renewed in 1867. Since that time the British ports have regularly received storm warnings, but the dis- play of the storm signal and storm drum was only revived in March, 1875. The French and English systems of storm warnings were in some respects preceded by the system organ- ized in Holland by Buys-Ballot, who in 1854 had announced his famous rule for that coun- try in regard to the direction of the wind, as depending on barometric disturbance. In 1860 he began the communication to the ship- ping ports of storm warning despatches, and was in fact by his signals the first to utilize the despatches contained in the telegraphic weather bulletins of Leverrier. The organ- ization of the French, English, and Dutch systems suffices to furnish for other European nations such storm predictions as are needed for their respective ports. Thus Spain, Italy, Sweden, Hanover, Russia, Austria, and Tur- key receive regularly from Paris and London announcements of the condition of the weath-
