The invention of these instruments is due to physiological investigations of the pulse. The problem of studying the pulse by graphic, or at least experimental methods, begins with the investigations of Hales and Poiseuille. The first great success in this line was the construction of the "Kymographion" of Ludwig, but this instrument had the disadvantage that it could be applied only by scission of an artery. This circumstance, of course, confined the application of the instrument to the study of the pulse of animals. After several attempts by Hérisson, Chelius, and others, Vierordt succeeded in constructing his sphygmograph, by which curves of the normal human pulse could be obtained. Some years afterwards Marey constructed his much more sensitive instrument, which was made still handier by the use of air transmission. Buisson was the first to use air transmission for sphygmography, but Upham had used it before for similar purposes. A considerable number of sphygmographs has been constructed since, and though they may show some improvements in detail, the technique of the sphygmograph has made no marked progress since Marey, and his instrument has been found by experimental tests remarkably exact. The curves traced with the sphygmograph are extremely variable in shape and size. In almost every normal curve, however, a steep ascent may be seen; it is called the up-stroke or percussion stroke, and this part of the sphygmographic curve has the name of the anacrotic phase. This line of ascent ends abruptly and within the limits of the usual speed of the recording drum it goes over into the descent by a sharp angle. The descending part of the curve is called the catacrotic phase. The descent is not so abrupt and is not a more or less straight line, but is interrupted by secondary elevations. The first secondary elevation is the largest and is called the dicrotic.[38]