American Palaeozoic Bryozoa. 147 



* 



On the other hand, to set against the very important points of re- 

 semblance above noted, we have — (L) to chronicle the discoveiy of 

 delicate radiating spines in Heteropora neozelanica, which character is 

 not developed in the Monticuliporidce; nor have I been able to detect 

 such structures in the five species of Heteropora examined by me. 

 (2.) The connecting foramina are more generally developed, or at 

 least more readily recognized, in the Heteroporoz than in the Monticu- 

 liporidce. As before remarked, I do not attach any weight to the fact 

 that in H. neozelanica, radiating spines intersect the tube cavity, 

 since if it were a character of real importance, such as the " septa" of 

 the Ccelenterata, it would be developed in all the species, which our 

 present knowledge of these forms justifies us in saying, is not the case. 

 A somewhat analogous instance is met with in Ptilodictya maculata, 

 Ulrich, in which the tubes usually are distinctly, as we may loose!} 7 

 term it, septate. Still there is no reason to doubt that the species is 

 intermediate between P. falciformis, Nicholson, and P. pavonia 

 D'Orbigny. In these that condition is only rarely met with, and then 

 it is not nearly so distinct as in P. maculata. On the whole, it is not 

 necessary to discuss the subject any further, since the points of 

 difference are so much outweighed by the points of resemblance, that 

 they can not possibly be considered of greater value than would con- 

 stitute a family distinction. Besides, we must bear in mind that I 

 have not attempted to demonstrate a generic identity, but only to 

 establish the relationship which I am convinced exists between the 

 Monticuliporidoz and Heteropora. 



Comparisons affecting the zoological position of the M onticuliporidai 

 and allied types might be carried on almost indefinitely, and the vari- 

 ous genera and families which I have, to a greater or less extent, re- 

 viewed in the preceding pages, are so inextricably interwoven, that by 

 separating them a positive injury is done to natural classification; and 

 in fact it would be preferrable to remove the whole assemblage from 

 the Bryozoa. But as this step would be as inadmissable as the first, 

 the only course left open for systematists is to leave them where the}* 

 really belong, with the Bryozoa. However, after extended stud} T of 

 these forms, it becomes evident that they differ widely from the typical 

 Cyclostomata (e. g. Diastoporidw, Idmoneidw, and Tubuliporidce), and 

 just as widely from nearl} r all of the Cheilostomata. On the whole I 

 have come to the conclusion that there are good reasons for the estab- 

 lishment of a fifth sub-order of the G-ymnolyEmata, which would include 

 the original Bryozoa, from which later, types of the sub-orders Cyclos 



