A Review of a Masters Thesis by Noel lames Gagnon (Graduate School, University of Minnesota, Duluth Branch, July, 1963.)
Mr. Gagnon devised an experiment to test the hypothesis that: (1) bassoon reeds having identical lay contours tend to produce similar pitch, timbre, and response characteristics (providing the variables of cane elasticity and reed exterior measurements are constant); and (2) it may be possible to construct reeds from a single pattern that are consistently satisfactory to several players provided a design which satisfactorily facilitates the above characteristics can be determined.
Eleven reeds having various lay contours were constructed (see examples). Gagnon admits that his designing process entailed much hypothesizing since the exact relationship between performance characteristics and lay-contour design was not previously known. All exterior measurements were kept constant as was the attempt to maintain uniformity in the elasticity and quality of cane.
In order to determine which design gave the best qualities of pitch, tone quality, and response, a panel of experienced bassoonists evaluated each of the eleven reeds for these characteristics. These isolated designs were then combined into a single lay-contour design which served as a pattern in the construction of ten additional reeds having an identical lay-contour. The design used consisted of a composite of two lay contours judged as being best by the panel (see example 5-11). These reeds were submitted to several bassoonists for evaluation to determine the degree of consistency with which the selected design facilitated satisfactory playing and sounding characteristics .
The cane from which reeds in the samples were constructed was selected on the basis of its observable qualities of homogeneity and was selected from 62 pieces purchased from a single source. Selection was accomplished by soaking the cane in water and timing the point of saturation for each piece of cane. While this experiment was not refined enough to produce conclusive evidence, it gave some indication that the natural rate of free vibration of cane can be predicted to a certain extent by the rate of saturation of the cane. 30 pieces were needed for the experiment; all the cane which reached saturation in the period from four to eight hours was selected. This period was chosen because it was the shortest span within the distribution containing 30 pieces of cane.
Measurement during construction of the lay contours was accomplished with a dial indicator and was not allowed to deviate from the pattern more than .001 of an inch. Measurements for thickness were made at 1/8-inch intervals down the center and sides of each reed tongue.
In the listening test, each reed was played by Gagnon; then each panel member compared the tone quality of each reed to that of the reed previously presented. Panel members evaluated reeds on one of four possibilities: (1) the reed is superior to the last; (2) the reed is inferior to the last; (3) the reed sounds the same; and (4) the reed sounds different but is neither superior or inferior.
For the playing test, each panel member played each reed and offered comments on the responding characteristics of each; they were permitted to use any techniques deemed necessary in testing these reeds. After each reed had been played once, reeds showing similar characteristics were tried again to compare them more critically. Emphasis was placed on evaluating the responding characteristics of the reeds.
The ten reeds in the consistency sample, patterned after the "ideal" lay-contour design, were submitted to five bassoonists for trial. Each bassoonist was asked to classify these reeds according to three categories: (1) satisfactory for all types of playing; (2) usable for some types of playing; and (3) poor for all types of playing. In addition, bassoonists commented on the playing characteristics as a whole.
Reed #11 (lay-contour design) was rated satisfactory by the whole panel. Reeds #5 and # 11 were rated most highly on the listening test while reeds #5, 9, and 11 were rated most highly on the playing test. An analysis of the lay-contour design of these reeds showed that reeds #9 and #11 were patterned after the same design. An examination of the designs revealed that the area back from the tip in design #5 was considerably thinner than in the design for reeds #9 and #11. On the other hand, the design for reeds #9 and #l l indicated that its root area was thinner than that of the design for #5. Reed #5 was probably too thin in the heart area toward the root. By combining the good features of the two designs (#5 and #11), it was thought that the poor features of each would be eliminated. Thus, the ten reeds in the consistency sample were made from a design having the same tip-area thickness as #11 and the same root-area thickness as #5.
