20 . IRVING HARDESTY 



of the coil of three of the latter membranes were lost in their 

 removal, but in these cases the loss did not preclude the desired 

 measurements, the basal ends being saved. 



In explanation of table 1, it should be noted that, with the 

 different vertical sections employed, the measurements of the 

 width of the membrane could not be taken across the same identi- 

 cal regions of the coil. The celloidin blocks were not oriented 

 on the microtome so that the median vertical sections always 

 involved the same diameter of the cochleae and, therefore, the 

 measurements of the ends of the membrane had to be taken in 

 each case at vary in ji unknown distances from the actual tips 

 of these ends. Owmo to the fact that the basal end of the mem- 

 brane is not coiled, no median vertical section of the cochlea 

 can pass through this end transversely. Manifestly oblique 

 sections of either end could not be used for measurements of 

 width, and transverse sections of the basal end must of necessity 

 have passed considerable distances from its actual tip. With 

 the teased out membranes coiled on the slide, the micrometer 

 scale could be arranged as a radius involving the widest part of 

 the apical end and crossing the other turns at the specified in- 

 tervals. Afterward could be taken a separate measurement 

 of the basal end, transversely across its narrowest part. The 

 actual termination of each end of the membrane and of the apical 

 end especially, is bluntly rounded. Measurements of the ends 

 were taken near but not involving the rounded part. 



Further, it seemed manifest in comparing the teased preparations 

 with the sections that, in the sections the membrane had suffered 

 some shrinkage. Practically all the shrinkage in width occurs 

 in the thicker outspanning zone and especially toward the apical 

 end where this zone is more voluminous. This explains some 

 of the differences in the width of this zone shown in table 1 

 between the measurements from the sections and those of the 

 teased out membranes. In all the sections of the adult cochleae, 

 the first half turn of the membrane seemed to conform less in 

 shape to what the teased preparations indicated it should be than 

 in other regions of the coil. In the cochlea from which figures 

 1 to 4 were drawn, the outer edge of the outspanning zone in 



