It has been pointed out recently* by one of us that the development of the remarkable chessman-spicule or discorhabd in the genus
Latrunculia
is a somewhat complicated process depending upon several factors. The protorhabd or axial thread appears first as a slender rod capable of independent growth. With these protorhabds two kinds of silica-secreting cells appear to be associated, viz., formative cells which are responsible for the actual deposition of the silica upon the protorhabd, and accessory silicoblasts which are supposed to collect supplies of silica and bring them to the formative cells to be used in the process of spicule-formation. The spicule in this case consists of an elongated axis with whorls of flattened lobes arranged at more or less definite intervals along its length, and it was suggested that the position of these whorls is determined by the fact that the spicule, at the time of their commencement, is in a state of vibration, due to the water currents flowing through the sponge, the whorls corresponding to the nodes or positions of comparative rest. The special accumulation of silica on the nodes appears to be due, not directly to the vibrations of the spicule, but to the fact that the formative cells exhibit a kind of tropism which induces them to settle down and perform their work in the positions where they are least disturbed by the vibrations. The whorls in this case are not sharply defined at the moment of their first appearance, so that it is impossible to obtain accurate measurements for mathematical analysis ; nevertheless, there are certain facts connected with their arrangement which, in our opinion, afford a fairly conclusive demonstration of the view that they are deposited approximately upon the nodes of a vibrating rod. Two species were investigated,
Latrunculia apicalis
and
L. bocagei
. In both species the spicule, at a certain stage of its development, consists of a straight rod with four thickenings, representing a basal manubrium and three incipient whorls. There is a basal thickening at one end, an apical thickening at the other, a median thickening at or near the centre, and a subsidiary thickening, usually between the median and apical thickenings, but occasionally between the median and basal thickenings. If these thickenings correspond to nodes, we have to account for the fact that a subsidiary thickening is developed only on one side of the median thickening. The solution of this difficulty is to be found in the arrangement of the formative cells (observed in
Latrunculia bocagei
only, though doubtless occurring in the other species also), for while there is a ring of formative cells round the median thickening and a similar ring round the subsidiary thickening, there is none around the part of the spicule where a second subsidiary thickening might be looked for, and hence no whorl is developed in this situation, in spite of its being a nodal point. No formative cells have yet been observed in relation to the basal and apical thickenings.
On the influence of vibrations upon the form of certain sponge-spicules