The vocal membrane is present in a wide range of species including non-human primates. It has been a matter of great interest how this small tissue contributes to the animal vocalizations and its evolutionary loss facilitated the speech communication in humans. There has been a theoretical study (Mergell et al., 1999), which predicted that the vocal membranes make the animal vocalizations more efficient by lowering the phonation threshold pressure. To examine the theoretical prediction, excised larynx experiments have been carried out for rhesus macaques (Macaca mulatta). Comparison of the larynges before and after surgical removal of the vocal membranes revealed that the phonation threshold pressure was indeed lower and, consequently, the vocal efficiency was higher in the presence of the vocal membranes. Irregular chaotic oscillations were also observed for some larynges with the vocal membranes. 

From two fresh cadavers of adult female macaques (#1, #2), the laryngeal samples were extracted, flash-frozen in liquid nitrogen, and stored at -80 degrees. The frozen samples were thawed shortly before the excised larynx experiments, which were performed under two conditions: (A) in the presence of the vocal membranes and (B) in the absence of the vocal membranes.

Under both conditions, individual larynx was mounted on a vertical tracheal tube. The flow rate of humid air (about 37 degrees; 100 % relative humidity) that comes from an air pump (SilentAirCompressor Sc820, Hitachi Koki Co., Ltd., Tokyo, Japan) was controlled by a pressure regulator (10202U, Fairchild, Winston-Salem, NC) and a digital mass flow controller (CMQ-V, Azbil, Santa Clara, CA). To induce vocal fold vibrations, the glottal air space was narrowed by manually adducting the arytenoid cartilages. Once the adjustment was made, positions of the arytenoid cartilages were fixed by a surgical suture. This configuration was used commonly for the two conditions (A) and (B). The dynamics of the vocal folds and the vocal membranes was monitored using a borescope with a view angle of 70 degrees (BAL-72718HT, Shodensha, Osaka, Japan) attached to a high-speed video camera (Fastcam Nova S6, Photron, Tokyo, Japan) (Sampling: 10000 FPS, Shutter speed: 1/25000 s). The acoustic sound and the sound pressure level (SPL) were measured by an omnidirectional microphone (Type 4192, Nexus conditioning amplifier, Bruel and Kjaer, Tokyo, Japan) and a sound level meter (Type 2250-A, Bruel and Kjaer), respectively, both located 15 cm from the larynx. The subglottal pressure was monitored using a pressure transducer (Differential pressure transducer, PDS 70GA, Kyowa, Osaka, Japan; Signal conditioner, CDV 700A, Kyowa), which was mounted flush on the inner wall of the tracheal tube, 2 cm upstream of the excised larynx. All signals were stored into a digital recorder (Controller, PXIe-8840, National Instruments; Input/output card, BNC-2110, National Instruments; Software, Labview, National Instruments, Austin, TX, USA) with a sampling frequency of 12.5 kHz.

To measure the phonation threshold pressure, the flow-rate was slowly increased from 0 l/min to a maximal value in 3 sec (the maximal airflow was set to be about 1.5 times of the onset airflow). Self-sustained oscillations of the larynx were continued to be measured for 4 sec. Then, the flow-rate was decreased to 0 l/min in 3 sec.

In the experimental condition (A), two oscillation patterns were observed:
(A1) the vocal folds did not show a strong vibration and only the vocal membranes oscillated, 
(A2) both vocal folds and membranes oscillated simultaneously.
The pattern (A2) was observed when both upper and lower parts of the vocal folds were adducted. The pattern (A1) was observed when the lower parts of the vocal folds were abducted from the condition of observing the pattern (A2).

A total of three oscillation patterns (A1), (A2), (B) were examined. For each oscillation pattern, the phonation threshold pressures were measured for 10 times.