'672 PROF. MixcHix AND DR. REiD ON THE [June 16, 



magnifications. The very large triradiates, however, tend to 

 •collapse when decalcified, like the monaxons of C. contorta, so 

 that it is difficult to obtain satisfactory preparations of them. 

 In the minute triradiates also the filaments are clearly seen, and 

 present no other difficulties to the photographer than such as 

 are caused by their minute size, and by the fact that the rays 

 usually lie in different planes. A study of the filaments in this 

 sponge brings out a point of some morphological interest. The 

 minute triradiates, as is well known, are of two types of form, 

 regular and sagittal. An examination of the filaments shows, 

 however, that in both forms alike the filaments meet at the centre 

 at angles of 120° (fig. 22, PI. XXXIY., fig. 23, PI. XXXVI.)*. 

 'Thus the spicules of Heterojoegma are distinctly Clathrinid in type, 

 and the sagittal forms occurring in this sponge are to be regarded 

 as arising simply by secondary curvature of the rays of a primi- 

 tively regular triradiate ; they may be termed pseudo-sagittal. 

 It has already been pointed out by Bidder (1898) and Minchin 

 (1900, p. 109), that Heteropeg'ina is a sponge which stands apart 

 from other Heteroccela and approaches the Clathi-inidse in its 

 characters ; and the study of its axial filaments certainly supports 

 these conclusions. 



The Nature of the Axial Filament. 



If we compare one of the photographs given here (figs. 2, 6-8) 

 of the axial filaments of the triradiate systems of Clathrina 

 contorta, magnified 1000 diameters, with the figure of a young 

 spicule of this sponge given by Minchin (1898, pi. 42. fig. 49) at 

 the same magnification, we are at once struck by the fact that 

 the central triangle, as we have termed it, formed by the axial 

 filaments, is neai-ly as large as the whole spicide at this early 

 stage ; much larger, in fact, than the earliest stages of the 

 spicules that can be found. 



It was further shown by Minchin {I. c. pp. 572-579) that 

 the spicules in early stages of development, when examined 

 between crossed nicols, light up scarcely or not at all. Hence in 

 early stages the spicules must contain very little, if any, crystal- 

 line su.bstance, that is to say calcite, but must consist chiefiy of 

 non- crystalline substances, perhaps both organic and inorganic. 

 We have referred above to Ebner's statements as regards the 

 inorganic impurities in the spicvile. It must be supposed that 

 the first portions of the spicule formed consist chiefiy of these 

 " impurities," and that the axis of the spicule, as it grows in 

 length, is also formed of substance containing very little calcite. 

 On the other hand, the portion of the spicule formed later is 



_ * In all the preparations tlie axial filaments are very liable to become displaced, 

 since tliej' are entirely unsupported after decalcification of the spicule. In the 

 spicule photographed in fig. 23, it can be seen that the filament of the left-hand ray 

 is displaced, but that of the right-hand ray shows the typical regular angle. 



