Mr. W. J. Sollas on the Oenus Cataffina. 355 



to' 



distinguished bj their great dilference iu size, the circles in 

 question often attaining a diameter of 0*001 inch. As the 

 sections through the cylindrical rods thus indicated vary from 

 longitudinal to transverse, so different forms are produced. 

 Next, one meets with a number of triradiate forms, sometimes 

 with the three rays inclined to one another at an angle of 120°, 

 sometimes less equally inclined. 



The ends of the rays are frequently seen, and prove to have 

 had an elongated conical form. The central space, from 

 which the arms radiate, frequently shows a cut surface, 

 eminently suggestive of the former existence of a fourth ray 

 which the section has removed. Thus, in PI. XIV. fig. 6, we 

 have the triradiate remains of a spicule with a cut surface in the 

 centre, circular in shape, and certainly due to the cutting away 

 of an arm which projected at right angles to the plane of the sec- 

 tion. In PI. XIV. fig. 14 the part cut away includes the upper 

 part of the three remaining rays ; and in this case we cannot 

 say whether a fourth arm was originally present or not. So, 

 again, the form of fig. 7 is inconclusive ; the cut surface looks 

 rather as if taken from one of the rays (?•) still remaining than 

 from one once at right angles to them. Even if this were the 

 case, there is still a possibility of a fourth arm being given off 

 on the opposite side, which should be visible on turning the 

 section upside down. Unfortunately my slice is mounted on 

 too thick a glass to permit of examination under a sufficiently 

 high magnification to decide this point. Thus, though sections 

 having a triradiate form abound in slices of the sponges under 

 examination, it is only in certain cases that they indicate 

 truly triradiate spicules. As the mounted slices of our 

 sponges are not mathematical planes, but possess a sensible 

 thickness, so it is possible to see something more than a mere 

 section across a spicule ; and thus we can very definitely make 

 out the existence of numerous quadriradiate forms in the fibre 

 (PI. XIV. fig. 1 7). These very much resemble the spicules of a 

 Pachastrella (PL XIV. fig. 13), one longer simple shaft dividing 

 into three shorter simple arms at one end. 



Irregular forms with apparently bifurcated rays are not un- 

 common ; and one instance of a five-radiate spicule has been 

 observed. 



The shaft is not always straight, but sometimes becomes 

 curved or even almost undulating (PI. XIV. fig. 18) ; the rays 

 likewise are sometimes curved (PI. XIV. fig. 11). In size and 

 in the relative length of the rays and shaft these spicules vary 

 greatly. In some the rays are scarcely of larger diameter 

 than the filiform spicules ; in others they are several times as 

 large, and appear giants by comparison. 



24* 



