The Typical Form of the Cochlea and its Variations. 411 



different animals from the same relative angle, I made what I think is a 



close approximation to the actual length of the basilar membrane. In this 



way I constructed a scale drawing of the basilar membrane as seen from 



above on the axis of the modiolus, and I measured its length with a rotary 



map measurer. The reader who is not familiar with Gray's photographs 



will find that they are much more transparent, 



and that, consequently, much more detail is 



visible in them, than he might be inclined to 



expect. Fig. 1 gives, as an example, a difficult 



case in which Dr. Gray's photograph presents a 



view of the cochlea from a point on a line at 



right angles to the axis of the modiolus. 



Doubtless there is a variable error, which 



probably never exceeded 2 or 3 mm., and was 



usually much less. And even that maximum is Fig. i. — (See A. A. Gray ■ 



fairly small in comparison with most of the 'The Labyrinth of Animals,' 



lengths recorded. The small error caused by the ™}' \ " Pla f? ' ? . T e 

 ° - Tiger.) x (about). 



rising of the cochlea to an apex I neglected 



entirely. For a basilar membrane of 52 - 4 mm. (No. 12) the error neglected 

 is not more than - 16 mm. For No. 52 the error is 01 mm. for a length of 

 16"3 mm. The results obtained confirm my estimate of the reliability of 

 the measurement of the basilar membrane. 



In order to trace the connection between the dimensions of the cochlea 

 and those of the body as a whole, I sought for a measurement which would 

 represent this. I found that one had been recorded by E. Lydekker for 

 most of the species represented in Gray's work, namely, the length of the 

 head and body, not including the tail. For the aquatic mammals I took the 

 length of the whole body. The method is rough, but no other was 

 available, and, under the circumstances, it is probably good enough. 



A. The Typical Form of the Cochlea. — Table I gives the values of the 

 coefficient of correlation between the various series of measurements, and of 

 the probable error, derived with the use of Karl Pearson's formula?. I have 

 neglected here any consideration of " the slant height of the cochlea," as 

 that is from its definition much less a matter of the height of the cochlea 

 than of the diameter of the lowest whorl, which is given independently. 



1. There is an absolutely certain positive correlation between the 

 diameter of the lowest whorl, on the one hand, and, on the other hand, 

 the diameter of the second whorl, the length of the basilar membrane in the 

 first two whorls, the total length of the basilar membrane, the major axis of 

 the oval window, and the diameter of the tube of the cochlea. These two 



