458 CLARK: BATHYMETRICAL DISTRIBUTION OF CRINOIDS 



.Considering the paucity of our records and our exceedingly 

 imperfect knowledge of the true maximum range of the species 

 involved we appear to be justified in calling attention to the 

 curious and striking approximation between our hypothetical 

 nodes calculated according to the ascending powers of 2 and the 

 nodes as we actually find them. 



It is also evident that there is a similar regularity in the dis- 

 tribution of the nodes in the Caribbean Sea. If we should take 

 50 fathoms as our unit (a) and arrange hypothetical nodes accord- 

 ing to the ascending powers of 2, plus 2 itself, we should find 

 nodes at: 



50 (a): 100 (2a): 200 (2-a) : 400 (2 ;i a) : 800 (2 4 a). 



It will be noticed that this arrangement is not greatly different 

 from what we actually find. 



Taking the whole ocean into consideration, the difference in 

 temperature between 200 and 400 fathoms is 8?3 Fahrenheit; 

 between 400 and 800 fathoms 4?5; and between 800 and 1600 

 fathoms 2?1. In other words, the decrease in temperature is 

 roughly in inverse proportion to the distances between the nodes. 

 As temperature is one of the chief factors governing the distri- 

 bution of marine animals it is quite probable that the spacing 

 of the nodes is dependent upon it. 



But the temperature decrease in the sea is represented by 

 a regular curve, so that it cannot account for the original exist- 

 ence of the nodes. Their origin is probably to be sought in the 

 answer to the problem concerning the origin of the deep fauna. 



The fauna of the central East Indian or Malayan region is 

 a heterogeneous mixture of species which in their distribution 

 fall into five more or less distinct groups. This fauna in its 

 batfrymetrical distribution shows no nodes. 



