340 MARINE BIOLOGY OF THE SUDANESE RED SEA. 
(iii.) Toxa. (Text-fig. 18, C.) 
The toxa are very large, but not numerous. The amount of curvature 
varies considerably, and also the abruptness of the curve. For almost its _ 
whole length the toxon is very slender, so slender that the diameter cannot 
be accurately measured, but at the centre of the bow it swells out into com- 
parative stoutness. The ends of the spicule taper off to so delicate a point 
that it is almost impossible, when examining the spicules in sitw, to ascertain 
the exact point at which they end ; and in boiled-out preparations they are 
invariably broken. For this reason the lengths given here are only to be 
considered approximate. The length varies very considerably, from 0°31 mm. 
to 0°21 mm. The diameter at the centre is 0°001 mm., while the greater part! 
of the spicule does not exceed 0:00025 mm. in diameter. 
Locality. Suez. 
Distribution. Red Sea. 
ESPERELLA ERYTHRHANA, n. sp. (Text-fig. 19.) 
The present species is represented by four specimens all growing over 
masses of calcareous Polyzoan tubes. In Mr. Crossland’s notes they are 
described as follows :—‘‘ Patches of a very delicate branched Polyzoan, and 
with this Sponges, Compound Ascidians, &c.” 
Neither of the specimens is large, the largest measuring 65 mm. long, 
52 mm. wide, and 25 mm. high. They are unattached to any foreign body, 
but apparently are loosely embedded in mud. The whole mass is quite 
amorphous, and no definite external form can be assigned to it. 
The surface is smooth, but irregular, for branches of the Polyzoan project 
here and there all over the sponge, and in other places branches just beneath 
the dermal membrane cause projections and ridges to appear on the sponge- 
surface. Where the polyzoan is absent the surface is perfectly smooth 
and even. 
The oscula are very small, not exceeding 0°5 mm. in diameter. They are 
very inconspicuous, and can only be distinguished on the surface by means 
of alens. They are very numerous and are scattered fairly evenly over the 
whole sponge surface. 
The pores are large, measuring from 0:15 mm. to 0:2 mm. in diameter. 
They occur in large numbers everywhere on the outside of the sponge, 
usually scattered singly, but occasionally in small groups. These groups, 
however, never form true pore-areas. 
The colour is dark grey-brown in spirit. 
The texture of the sponge is lax, and where it is not held together by the 
polyzoan it is very easily torn. The whole mass, however, is firm and fairly 
resistant, since the polyzoan tubes serve as a supporting framework. 
