1904.] SPONGES OF THE GENUS LEUCOSOLENIA. 359 



fourth or gastral ray. In the monaxons a proximal end imbedded 

 in the wall of the sponge is to he distinguished from a distal end 

 projecting freely into the water. The shaft of the inonaxon is 

 generally thickest towards the proximal end, at which it tapers 

 rapidly to a blunt or moderately sharp point. The distal end 

 usually has a barb or " lance-head," frequently rudimentary or 

 absent. The barb is in reality a double bend in the axis of the 

 spicule, and is comparable to a very thick bayonet, rather than 

 to a spear-head*. In cases where it is absent, the distal ends of 

 the monaxons become excessively sharp and fine. 



A remarkable point with reference to the monaxons of Leuco- 

 solenia, which I have found to hold good, not only with regard to 

 the species described in this memoir, but also for all other species 

 that I have examined, is that the monaxons can be separated more 

 or less easily into two varieties, distinguished by the fact that 

 one kind appears very refringent, the other, by comparison, pale,, 

 under the microscope. The refringent monaxons are always scarcer 

 than the pale ones, but their peculiar optical property makes it very 

 easy to find them, especially under low powers (Zeiss, Oc II. Obj. B). 

 In form the two kinds of monaxons may not differ essentially,, 

 but the refringent ones always show certain characteristics which 

 may be summed up by saying that they tend to be straighter,. 

 moi'e slender, and sharper than the others, and their distal barb is 

 less distinct or absent. When examined by means of polarised 

 light, the conditions are reversed, since the pale monaxons light 

 up brightly between crossed prisms, while the refringent forms 

 remain dark or feebly illuminated. This is particularly well seen 

 in the small monaxons of L. variabilis (see below, p. 380), where the 

 curved forms light up most brilliantly with crossed prisms, while 

 the straight refringent forms remain quite dark or only slightly 

 illuminated in all positions, when the stage of the microscope is 

 rotated. This shows clearly that the difference between the two 

 types is due to a difference in the relation of the axis of crystal- 

 lisation to the form of the spicule. 



The triradiate systems which do not acquire gastral rays do not 

 differ in any other structural feature from those which, by doing 

 so, become quadriradiates. In some cases the triradiates and 

 quadriradiates may differ in size, and their relative abundance 

 varies greatly in different specimens. In each triradiate system 

 we have to distinguish, as has been said above, an unpaired or 

 posterior ray, which in the oscular tube points away from the 

 oscular opening, and two paired lateral lays; the latter make 

 with the posterior ray paired lateral angles less than 120° and 

 sometimes almost approaching 90° ; the lateral rays at their 

 junction enclose an unpaii-ed anterior angle, which is greater 

 than 120° in proportion as the lateral angles are less. The lateral 

 lays are neai-ly always distinctly curved ; the posterior ray is 

 normally straight. 



* This point is, unfortunately, not very well brought out in the drawings accom- 

 panying this memoir. 



