86 



UINTACRIiSrUS : ITS STRUCTURE AND RELATIONS, 



is no enclosed plate, and 51 per cent have more than nine plates in one or 

 more areas. 



By combining the foregoing data into a few classes according to the 

 maturity of the specimens, we are enabled to obtain an average whereby 

 the influence of the misleading irregularity in individual specimens is to 

 a large extent eliminated. I have accordingly taken two types of inter- 

 brachial areas, viz. : one containing not more than seven plates in any area, 

 which might be said to be the U. loestfalicus type : and the other contain- 

 ing nine plates or more in some of the areas, which would represent the 

 U. socialis type, according to Clark. These I have an-anged in four groups 

 of minimum, medium, large, and maximum sizes. This process gives the 

 following interesting and significant result : — 



RECAPITULATION AS TO INTERBRACHIALS AND BASE. 



From this table we can see the general tendency of the individual 

 variations and modifications by growth of the supplementary plates, to 

 a degree of certainty which gives to the conclusion of Wachsmuth 

 and Springer the force of a fixed law. In the smallest class 73 per 

 cent of the specimens have but few interbrachials, and only 10 per 

 cent have more than nine ; while in the class of maximum size the.se 

 proportions are reversed, 84 per cent of the specimens having more than 

 nine. In the intermediate classes the gradual progress of these differences 

 is well marked. 



The development of the intersecundibrach spaces and interpinnulars is 

 considerably more irregular, and yet it follows the same general rule. The 

 smaller number prevails among the smaller specimens, and is the exception 

 among the larger; while the proportion of the larger numbers increases 



