262 Proceedings of the Royal Irish Academy. 



2. The cause for the diminution in number is not far to seek. The 

 number of polyps is smaller in the younger parts, and consequently the 

 number of canals communicating with these is smaller. 



.3. The number of canals in the outer series bears no proportion to the 

 number in the inner series in the different parts, although the number 

 diminishes in both cases from the base upwards. It diminishes more rapidly 

 in the inner series. 



Let us consider the different series, and see to what extent these may be 

 considered of taxonomic value. 



1. Tran.9rerse series. — This series, as has been pointed out, ser\-es to connect 

 the outer longitudinal series with the inner longitudinal series, and, as might 

 be expected, is of no specific importance. 



2. Outer longitudinal series. — The canals of this series communicate 

 directly with the polyps, are all of equal size, have no inOuence on external 

 or internal form, are constant in all specimens, and cannot therefore be 

 taken into account in specific determination. 



3. Inner longitudinal series. — We have here to deal with a series which 

 has the following characteristics : — 



(a) The canals are not all of equal value. 



[h) They exert an influence on the external form of the colony. 



(f ) They produce an effect on the surface of the axis. 



The .superficial results produced by this series of canals have been 

 used by different authors as a liasis of classification ; but no systematic 

 examination has ever been attempted, nor has any causal explanation ever 

 been given, so that it may serve some useful purpose to study the actual 

 influence exertetl and the constancy of the results. 



Studer (xxxviii) in 1901 makes the following note: — "A transverse 

 sec;tion of a colony of Scirjtearia flagellum (PL IX, fig. 11) shows that the 

 polyps arise on two sides of the axis ; there are two large longitudinal canals 

 in the plane perpendicular to that of the polyps." 



Thomson and Henderson (XXXIX , p. 315, in describing Jvncella trilineata 

 say : — " Polyps arise in three different bands, leaving three narrow bare 

 strips, each of which has in its centre a slight rib or keel. Under each bare 

 strip lies a lar^ge longitudinal canal. The axis shows longitudinal grooves." 



These are practically the only two references to the phenomena under 

 investigation. 



Let us consider each in detail : — 



(a) The cnnnls are not all of equal value. — A transverse section of any 

 Juncellid colony reveals the fact that there is a certain number of the 



