Don Reinfeld, Bow Maker

Because of the problems of reproducing photographs from an old, poor quality xerox copy of magazine pages, there was no point in including figures 1a, 1b, 2, 3, and 4 here. The explanations of these figures, taken along with Prof. Gray's 'diagrammatic summary,' of fig. 1, still show the full value of his observations. If anyone can provide the correct citation to this article, which first appeared in The Strad (1980's perhaps?), please contact me via email.

As far as I know, the mechanism of the friction between bow hair and string has never been investigated with the electron microscope. The conventional light microscope has useful magnification of only x1,000, whilst the electron microscope can magnify x40,000 or more. In this investigation I have compared the surface of the individual unused horse-tail bow hair (unrosined) with that of the well used (several weeks) rosined bow hair. What follows is a theory to explain the mechanism of hair-against-string friction and the paramount role played by the rosin.

In order to establish a perspective in which to view horse-tail bow hair it is worth comparing the relative thickness of each: the human scalp hair is about 0.2mm thick, and horse-tail bow hair is about twice this thickness. The structure of the human hair is well documented in medical textbooks and my observations confirm that the structure of horse-tail hair is basically similar apart from its thickness. Both have a surface of scales which are tough and brittle, but the interior of the hair (see Fig. 1) is of softer and more plastic material. Incidentally, the whole hair is made up from dead cells, which have been converted into a protein called keratin. The surface scales (each about 0.3mm across) always overlap away from the 'horse-end' just as the human head hairs have scales that overlap always away from the scalp. This, of course, is the ideal arrangement for draining off excess liquid (i.e. rain).

Like most investigations, one has preconceived ideas and in this case I assumed that the friction between the string and newly-rosined bow hair would, after playing for some time, develop a surface of partly or wholly detached scales, the jagged edges of which would scrape against and vibrate the string. In other words playing with new hair would involve an initial period for the scales to be affected.

Fig. 2a shows the typical undetached layer of scales on an unused bow hair seen with the electron microscope (x12,000). Fig. 2b shows the same but with the hair rotated at and angled to the electron beam, giving the appearance of teeth on the scales. The function of these is, in fact, to grip the string in the initial 'playing-in' of the new hair so that the scales become detached in a row down the hair where the contact is made with the string.

What is extremely interesting is that the electron microscope revealed that the detachment of the scales leaves a groove down the well-used hair. Furthermore, within the groove itself exist numerous particles of rosin. The groove (g in Fig. 3) is shown at low magnification (x14,000) and loose scales (l) occur on the hair. However, it is necessary to turn to much higher magnification (x30,000) to investigate the interior of the groove (Fig. 4). The edges of the groove are marked by broken scales (s) and the floor of the groove is studded with numerous rosin particles (rp) of various sizes, some only a few thousandths of a mm in diameter. The particles (of fractured rosin) are of various shapes, some of which appear quite jagged.

Fig. 1 is a diagrammatic summary of both the unused (1a) and the used (1b) hair. In 1a the surface scales (s) are all attached: in 1b the bow-hair has developed a groove by removing a linear array of scales which has exposed the soft, plastic keratin along the floor of the groove. Rubbing the hair on the rosin block produces an extremely fine powder of rosin particles (rp). These stick into the soft keratin on the floor of the groove[, a process]...analagous to pushing a pea a little way into plasticene. Thus a band of rosin particles is produced projecting from the hair, particles small, even submicroscopic; yet in combination with all the other rosin particles on all the other hairs of the bow, they function in unison to produce smooth and intense vibration of the string.

Such an arrangement could only have been elucidated with the electron microscope. Thus it is not the superficial scales that the rosin adheres to, but the exposed soft part of the hair within the groove. During playing, the particles slowly get pulled back out of the plastic bed - hence the need to apply more rosin to the bow-hair.

With long playing, the groove widens with the accumulation of more rosin particles so that the instrument can be made to play slightly louder. However a groove which is too wide represents a thinning of the hair, which can cause it to snap (much to the consternation of the concert performer!).

The author would like to thank Charles Beare for his advice and for providing the unused bow hair. .