510 PEOr. T. W. BEIDGE ON THE 



question separates the closely related inner walls of the two series 

 of ventral diverticula, and, as regards their relations to the caudal 

 skeleton elements, these correspond in position with the dorsal 

 series. Wherever these ventral vacuities existed, corresponding 

 perforations in the opposed inner walls of the diverticula were 

 also present, and each of the latter seemed to communicate with 

 its fellow of the opposite side of the tail. In some instances it 

 was possible to determine the continuity of the inner wall of a 

 diverticulum with the corresponding wall of its fellow at the 

 edges of the vacuity ; and for this reason I am inclined to 

 believe that the caudal caeca are really in communication with each 

 other ventrally, through the fusion and subsequent perforation 

 of the inner walls of their ventral diverticula. Nevertheless, it 

 is perhaps desirable that the caution needful in the case of the 

 apparent dorsal perforations should also be observed here, at 

 all events until fresh or well-preserved specimens of Notopterus 

 have been submitted to examination. 



A somewhat interesting numerical and regional correspondence 

 is to be observed in the anterior half of the tail between the 

 various caudal skeletal elements and the arrangement of the 

 caecal outgrowths from the caudal divisions of the air-bladder. 



Thus, the ventral diverticula are situated exactly opposite the 

 intervals between the series of haemal spines, and, at the same 

 time, occupy the interspaces between successive pairs of radial 

 elements ; that is, each diverticulum fills up the interval between 

 two contiguous pairs of radial elements which, dorsaily, are 

 attached to the haemal spines of two successive caudal vertebrae. 

 Hence, therefore, the diverticula closely agree in number with 

 the pairs of radial elements, and in position may be said to be 

 intervertebral. Again, each of the series of fibrous ridges 

 w^hich separate the orifices of communication between the 

 ventral diverticula and the caudal caeca, is attached to the 

 opposed surfaces of a pair of radial elements as well as to the 

 fibrous membrane between them ; and as each ridge coincides 

 dorsaily with the haemal spine to which the two radial elements 

 belong, it is vertebral in its relations to the axial skeleton. 

 La-tly, the filiform caeca agree in number with the radial 

 elements, the posterior caecum of one diverticulum, and the 

 anterior caecum of the next succeeding diverticulum, being 

 situated immediately external to the corresponding factors of 

 two contiguuus pairs of radial elements. 



