820 MR. E. T. BHOWNE ON BRITISH MEDXTS.?:. [Xov. 16, 



filiform cirrus was situated on the outer side of a tentacle-bulb, or 

 upon the " spur." The cause of this abnormal position I was 

 able to trace from the early stages. The tentacles do not all 

 occupy exactly the same position on the margin ; some develop 

 very close to the iuuer side of the margin. The filiform cirri are 

 close to the outer side of the margin, and, consequently, when a 

 tentacle-bulb makes its appearance close to the inner margin and 

 opposite a cirrus, in its subsequent growth it carries the cirrus in 

 an outward direction. 



AH the Valencia specimens were quite coloiu'less to the naked 

 eye, but under the microscope a narrow band of a yellowish-brown 

 colour is seen at the base of the ring-canal, and a narrow streak 

 of the same colour is also visible in the basal bulbs of the tentacles. 



DiPLEUROsoMA HEMisPH^RicuM (Allman). (Plate XLVIII. 

 figs. 3, 3 a.) 



Ametrangia htmisphcerica, Allman (1873). 



Dipleurosoma irregulare, Haeckel (1879); Haddon (1885). 



During my visits to Valencia in 1895 and 1896 I found a few 

 specimens of this species. 



Allman, in the description of the species, states that there are 

 three main radial canals with branches ; some of the branches enter 

 the ring-canal and others termhiate blindly. 



Haeckel found some specimens at Brighton, one of which agreed 

 with Allman's description, and the others showed considerable 

 variation in the arrangement of the canal-system. 



The Valencia specimens also show that the canal-system, which 

 I now only intend to describe, is very variable. 



The specimen which approaches nearest to Allman's descrip- 

 tion has three large radial canals, with lateral branches, and also 

 two smaller canals leading from the stomach to the ring- 

 canal (fig. 10, p. 827). The umbrella of this specimen measures 

 8 mm. in diameter, and the ova are just visible upon four of the 

 canals. 



Another specimen, of which I have given a figure (Plate XLVIII. 

 figs.3,3rt) shows another type of canal-system. There are four large 

 canals, which divide the umbrella into four equal parts, and four 

 smaller canals, interradially situated, one of which has reached the 

 ring-canal and the other three tenninate not far off. The gonads 

 ha\ e not yet commenced to develop. The umbrella is about 8 mm. 

 in width and 4 mm. in length. 



A very irregular system of canals is shown in another specimen 

 (fig. 11, p. 827). This has eight radial canals leaving the 

 stomach and joining the ring- canal, and some of these canals have 

 short lateral branches,whichprobably by further growth finally reach 

 the ring-canal. In a damaged specimen I have counted as many 

 as eleven canals entering the ring-canal. A slightly damaged 

 specimen of a male shows that the generative cells form a large 

 oval mass, on, at least, five of the radial canals, and that the 



