SoLLAS — On Sponge Spicules. 383 



lies in the median plane of symmetry of the unpaired ray, it crosses 

 transversely the median plane of symmetry of the paired rays, which 

 make an angle of from 90° to 120" with the unpaired ray. 



I may now allude to a curious difference between the refractive 

 index of the triradiate and the acerate spicules of Grrantia. The 

 latter, as we have already stated, have a refractive index of 1'485 ; 

 the former, on the other hand, disappear in balsam (r. i. = say 1*520 

 to 1*525) when viewed by the extraordinary rays which vibrate in 

 the principal section of the analyser. Hence the axis of maximum 

 elasticity, which lies at right angles to the optic axis in the same 

 plane, cannot be vertical to the plane of the spicule, but is inclined 

 from it at an angle of 55° to 60°, or at 30° to 35° from the common 

 axis of the nicols : and the optic axis consequently makes an angle 

 of 30^ to 35° with the plane of the spicule, or of 55° to 60° with 

 a vertical ray of light (fig. 3). 



Although the triradiates and quadriradiates extinguish when one 

 ray is parallel to either diagonal of the nicols ; yet the parallelism 

 is not always exact, the difference often amounting to 8° or 10°. 



The acerates of Grantia extinguish at about 20° ; the angle 

 varying about 10"^, but usually much less, on each side of this. As 

 they become invisible at this angle, when mounted in balsam and 

 viewed by the extraordinary rays which vibrate in a plane at right 

 angles to the principal section of the analyzer, it follows that the 

 plane of the principal section of the analyzer then cuts the spicule 

 transversely to its length at an angle of 70° (fig. 5) ; and next, that 

 an axis of maximum elasticity at right angles to the optic axis lies 

 vertically in the spicule, and hence that the optic axis is horizontal. 

 The club, or retort-shaped, spicules of Grrantia extinguish when 

 the long ray makes an angle of 30° with a diagonal of the nicols ; 

 the thick, short ray then extinguishes, according to the angle it 

 makes with the long ray, at from 10° to 30° (fig. 6). 



A comparison of the relations of the spicules to each other, as 

 shown in figs. 4 to 6, will show that the acerate spicules are not 

 homologous, as one might have suspected, with the unpaired ray of 

 a sagittal spicule, but with one or both of the paired rays, probably 

 with both ; since in many of the sagittal spicules these rays both make 

 an angle of almost 90° with the unpaired ray, when they are both 

 nearly in the same straight line. The club-shaped spicule also 

 probably corresponds to the paired rays of the sagittal. It may be 

 that we are not justified in thus determining homologies by means 



