The Typical Form of the Cochlea and its Variations. 413 



significance. There is, then, no correlation between the number of whorls 

 and any other dimension of the cochlea. 



4. Fig. 2 gives the distribution of the frequency of the different numbers 

 of whorls, i.e. there were amongst the 52 organs of mammals measured by 



io times fO 



o. : t .o 



\w/?or/s 2 3 4 



Fig. 2. — Number of Whorls — Distribution of Frequency of each Number. 



Dr. Gray, three organs having \\ whorls, five having If whorls, and so on. 

 One organ had 4| whorls. The figure shows a striking evenness of 

 distribution about a point between 1\ and 2f whorls. The average number 

 of whorls of all 52 organs is 2 - 56 whorls, or a little more than 2J turns. 

 Thus, the average and the highest frequency agree. We may therefore put 

 down the typical number of whorls of the mammalian cochlea as two and a 

 half. 



5. The absence of correlation between the number of whorls and the 

 diameter of the second whorl is interesting. It means that the addition of a 

 third whorl (or of more than one) does not necessitate an expansion of the 

 second whorl to make more room for it. Generally the other whorl is merely 

 added on where the second one stopped without any change in the other 

 dimensions of the cochlea. 



6. The two negative correlations of the " ratio of the diameter of the second 

 whorl to that of the first " seem to indicate that the bigger the cochlea is, the 

 greater is its rate of curvature, i.e. the smaller is the diameter of the second 

 whorl relatively to that of the first. This correlation, especially as relating 

 to the diameter of the first whorl, approaches towards being a practical 

 certainty. The coefficient is almost 5£ times the amount of the probable 

 error, 6 times yielding practical certainty of correlation. 



7. In this connection reference may be made to the only other description 

 applied to the cochlea by Dr. Gray, namely, the distinction of the " flat " from 

 the " sharp " cochlea. In the latter the whorls seem to be piled on the top 

 of one another ; in the former the second whorl lies more or less in the coil 

 of the first. This difference is connected with the ratio of the diameter of the 

 second whorl to that of the first. If an arbitrary numerical value be given 

 to Dr. Gray's degree^ of flatness and sharpness: very sharp or sharp and 



