122 THE CRINOIDEA CAMERATA OF NORTH AMERICA. 



throughout, by means of which the surrounding water is brought in direct 

 contact with any part of the body. 



There are good reasons for believing that in the Camerata the water for 

 respiration was introduced near the arm bases through small openings, de- 

 scribed by us as respiratory pores, and then followed the canals and passages 

 along the test. Such openings have been observed not only among the Actino- 

 crinidae and Batocrinidso, but also among the Melocrinidse and Ehodocrinidse. 

 In the genus Dolatocrinus, they are large and slit-like as in Ophiurids, in Bato- 

 crinus and Actinocrinus round, and in Gilbertsocrinus at the outer end of long 

 tubes. The openings are always located between the rays and their main 

 divisions, a little above the arm regions. Some species of Dolatocrinus have 

 from four to six to each interradius, and two to four to each interdistichal 

 space, all arranged horizontally. In Dolatocrinus this vascular system prob- 

 ably extended only over the peripheral portions of the disk, for the inner 

 floor at the middle portion is perfectly smooth in the specimens. In Bato- 

 crmus and Teleiocrinus it probably extended to the outer margins of the orals 

 (Plate y. Figs. 16 and 17) ; while in Physetocrinus, when the orals are un- 

 represented, it apparently occupied the whole tegmen. 



We now come to the ventral structure of Siphonocrmus. It will be 

 remembered that in S. armosus not only the ambulacra, but large portions 

 of the anal tube, are subtegminal, and that the latter lies across the mouth 

 and covers portions of the ambulacra. The tube, however, in two other 

 species of this genus opens out centrally, thus showing that the subtegminal 

 condition of the tube had no important bearing upon the general structure 

 of the disk. As we understand the case, the anal tube, which is actually 

 the outer end of the hind gut, in place of becoming free and piercing the 

 central part of the disk as in the other two species, was roofed over in 

 >S'. armosus by the interambulacral pieces in a somewhat similar manner to 

 the calyx ambulacra of Megistocrinus. 



Now if it is true that in forms like Physetocrinus, Batocri?ius, Actinocrinus, 

 and Siphonocrinus, there is no second integument, it may be considered as 

 proved that a " vault," as an independent structure, did not exist in any of 

 the Camerata, nor, in fact, in any of the other groups, and that the structure 

 to which the term has been applied in these forms was evolved phylogenet- 

 ically from the disk of the primitive types, of probably Pre-Silurian time. 



The disk of the Fistulata also experienced notable changes in its palaBon- 

 tological development, but these took place on different lines. The plates 



